Confronting the Past: Archaeological and Historical Essays on Ancient Israel in Honor of William G. Dever 9781575065717

William G. Dever is recognized as the doyen of North American archaeologist-historians who work in the field of the anci

183 99 8MB

English Pages 400 [396] Year 2006

Report DMCA / Copyright

DOWNLOAD FILE

Polecaj historie

Confronting the Past: Archaeological and Historical Essays on Ancient Israel in Honor of William G. Dever
 9781575065717

Citation preview

Confronting the Past

William G. Dever

Confronting the Past Archaeological and Historical Essays on Ancient Israel in Honor of

William G. Dever

Edited by

Seymour Gitin, J. Edward Wright, and J. P. Dessel

Winona Lake, Indiana Eisenbrauns 2006

ç Copyright 2006 by Eisenbrauns. All rights reserved. Printed in the United States of America.

www.eisenbrauns.com

Library of Congress Cataloging-in-Publication Data Confronting the past : archaeological and historical essays on ancient Israel in honor of William G. Dever / edited by Seymour Gitin, J. Edward Wright, and J. P. Dessel. p. cm. Includes bibliographical references. ISBN-13: 978-1-57506-117-7 (hardback : alk. paper) 1. Palestine—Antiquities. 2. Israel—Antiquities. 3. Middle East— Antiquities. 4. Palestine—History—To 70 a.d. 5. Middle East—History—To 622. 6. Bronze age—Palestine. 7. Iron age—Palestine. 8. Bible. O.T.— Antiquities. 9. Bible. O.T.—Criticism, interpretation, etc. 10. Excavations— Archaeology—Middle East. I. Dever, William G. II. Gitin, Seymour. III. Wright, J. Edward. IV. Dessel, J. P. DS111.C66 2006 933—dc22 2006026894

The paper used in this publication meets the minimum requirements of the American National Standard for Information Sciences—Permanence of Paper for Printed Library Materials, ANSI Z39.48-1984. †‘

Contents Acknowledgments . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . viii Preface . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ix William G. Dever: Bibliography . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . xiii Prepared by Wade Kotter

Part 1 Archaeology The Sad Fate of Statues and the Mutilated Statues of Hazor . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Amnon Ben-Tor

3

An Archaeologist on Mars . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 17 Giorgio Buccellati Of Pots and Paradigms: Interpreting the Intermediate Bronze Age in Israel/Palestine . . . . . . . . . . . . . . . . . . . 23 Shlomo Bunimovitz and Raphael Greenberg A Decorated Ivory Lid from Tel Miqne–Ekron . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 33 Trude Dothan Har Resisim 126: An Ephemeral Early Bronze Age Site . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 41 Barry M. Gittlen Gezer Rectified: The Dating of the South Gate Complex . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 51 James W. Hardin and Joe D. Seger An Early Iron Age I House with a Cultic Corner at Tall al-ºUmayri, Jordan . . . . . . . . . . . . . . . . . . . . . . . . . . . 61 Larry G. Herr Aphrodite/Astarte on Horseback . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 75 Vassos Karageorghis Planning a Paleolithic Picnic . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 81 Albert Leonard, Jr. The Archaic Palace at Ebla: A Royal Building between Early Bronze Age IVB and Middle Bronze Age I . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 85 Paolo Matthiae Tel Beth-Shean and the Fate of Mounds in the Intermediate Bronze Age . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 105 Amihai Mazar Early Bronze Age IV Transitions: An Archaeometallurgical Study . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 119 Suzanne Richard The Chipped Stone Assemblage from Beªer Resisim in the Negev Highlands: A Preliminary Study . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 133 Steven A. Rosen, Sorin Hermon, Jacob Vardi, and Yael Abadi Podium Structures with Lateral Access: Authority Ploys in Royal Architecture in the Iron Age Levant . . . . . 145 Ilan Sharon and Anabel Zarzecki-Peleg Chariot Fittings from Philistine Ashkelon . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 169 Lawrence E. Stager

Contents

vi

Goddesses and Cults at Tel Dor . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 177 Ephraim Stern Cypriot Anthropomorphic Figurines of a Certain Type . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 181 Stuart Swiny

Part 2 Bible and Ancient Israel Women and the Worship of Yahweh in Ancient Israel . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 189 Susan Ackerman Purity Strategies and Political Interests in the Policy of Nehemiah . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 199 Rainer Albertz Proto-Globalization and Proto-Secularization in Ancient Israel . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 207 Norman K. Gottwald The Miraculous Wine of Cana in Its Galilean Ceramic Context . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 215 Baruch Halpern The Archaeology of Memory: King Solomon, Chronology, and Biblical Representation . . . . . . . . . . . . . . . . 219 Ronald Hendel Khirbet el-Qôm and Hebrew and Aramaic Epigraphy . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 231 André Lemaire The Ecological Consequences of a Siege: A Marginal Note on Deuteronomy 20:19–20 . . . . . . . . . . . . . . . . 239 Aren M. Maier, Oren Ackermann, and Hendrik J. Bruins Hierarchy or Heterarchy? Archaeology and the Theorizing of Israelite Society . . . . . . . . . . . . . . . . . . . . . . . 245 Carol Meyers Israel and Its Neighbors Then and Now: Revisionist History and the Quest for History in the Middle East Today . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 255 Eric M. Meyers Ostracon No. 7 from Arad Reconsidered . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 265 Nadav Naªaman Looking for Bethel: An Exercise in Historical Geography . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 269 Anson F. Rainey In Solomon’s Temple (1 Kings 6–7): Between Text and Archaeology . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 275 Mark S. Smith

Part 3 Ancient Near East The Predecessors of the Hyksos . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 285 Manfred Bietak “His Seed Is Not”: 13th-Century bce Israel . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 295 David Noel Freedman and David Miano Solomon and Gezer . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 303 Volkmar Fritz Hezekiah’s Tribute, Long-Distance Trade, and the Wealth of Nations ca. 1,000–600 bc: A New Perspective . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 309 John S. Holladay, Jr. Gezer and Circumcision . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 333 Philip J. King

Contents

vii

Covenant and Blood Rituals: Understanding Exodus 24:3–8 in Its Ancient Near Eastern Context . . . . . . . . . 341 Theodore J. Lewis Trade Relations between Mari and Hazor (State of Research, 2002) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 351 Abraham Malamat Implicit Population Figures and Historical Sense: What Happened to 200,150 Judahites in 701 bce? . . . . . . 357 Ziony Zevit Indexes Index of Authors . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 367 Index of Scripture . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 372 Index of Other Ancient Sources . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 375

Acknowledgments The publication of this volume was made possible by the generous support of Richard Scheuer, Shirley D. Curson, and Joan Kaye Cauthorn. The Editors thank Edna Sachar for her skillful preparation of the manuscript in Jerusalem and Jim Eisenbraun and Beverly McCoy for their guidance through the entire publication process.

Preface

The title of this volume honoring W. G. Dever, Confronting the Past: Archaeological and Historical Essays on Ancient Israel, reflects the broad scope of Bill’s scholarship. Dealing with a wide range of subjects relating to the Bronze and Iron Ages, Bill confronts the historical implications of archaeology—as he has so often remarked, “If archaeology is not history, it is nothing.” The essays presented herein also include those dealing with data from earlier and later periods originating within the geographical area of ancient Israel, considered part of the wider biblical world as understood by Bill’s mentor, G. Ernest Wright, and his teacher, W. F. Albright. In addition to celebrating Bill’s contribution to the world of scholarship, this volume demonstrates the deep appreciation, respect, and affection in which he is held by the community of archaeologists and other scholars in the wider field of ancient Near Eastern studies. This is our tribute to our teacher, colleague, and friend, and to his scholarship and unswerving dedication to excellence. William G. Dever was born on November 27, 1933, in Louisville, Kentucky, USA. Because he was the son of a Christian Church minister, it was only natural that early on he showed an interest in biblical studies, which he was to pursue throughout his academic career. In 1955, Bill earned a B.A. in Religious Studies at Milligan College, Tennessee, where he had met Norma Elizabeth Spangler; Bill and Norma were married in 1953. Continuing his studies, he received both a B.D. in Hebrew and Greek from Christian Theological Seminary and an M.A. in Semitics from Butler College in Indianapolis, Indiana, in 1959. Subsequently, he entered the doctoral program in Biblical Theology at Harvard University under the supervision of G. Ernest Wright. At Wright’s suggestion, he changed his major to archaeology, and in 1962 and 1964, he participated in Wright’s excavations at Shechem. Under Wright’s and Frank Moore Cross’s tutelage, Bill earned his Ph.D. from Harvard University’s Department of Near Eastern Languages and Civilizations in 1966, with his dissertation on The EB IV–MB I Period in Syria–Palestine, ca. 2150–1850. In 1964, Bill joined the staff of the long-term excavations at Tell Gezer initiated by Wright and Nelson Glueck. The project was sponsored by the Hebrew Union College Biblical and Archaeological School in Jerusalem (HUCBASJ) (renamed the Nelson Glueck School of Biblical Ar-

chaeology in 1972), and was also supported by the Semitic Museum of Harvard University. Bill remained with Wright in Jerusalem as the Archaeological Fellow at HUCBASJ from 1964 to 1965, and returned to Harvard in 1966 to finish writing his dissertation. From 1966 to 1971, he served as the Director of the Gezer excavations, with H. Darrell Lance as Associate Director. In 1967, Bill was appointed Assistant Professor of Archaeology at HUCBASJ, and from 1968 to 1971, he served as its permanent Archaeological Director. While at HUCBASJ, he also directed the excavations at Khirbet el-Qôm and Jebel Qaºaqir (1969– 1971). In 1971, Bill became the fourth long-term Director of the W. F. Albright Institute of Archaeological Research in Jerusalem (AIAR), formerly the American School of Oriental Research, and in 1973, he was also appointed AIAR Professor of Near Eastern History and Archaeology. In 1972 and 1973, he conducted salvage excavations at Shechem. In 1969, while Bill and Norma were living in Jerusalem, their son, Sean William, was born. In 1975, Bill and his family returned to the United States, where he was instrumental in creating a graduate program in Syro-Palestinian Archaeology at the University of Arizona in Tucson. He held appointments as Professor of Near Eastern Archaeology in the departments of Oriental Studies (now Near Eastern Studies) and Anthropology and he led the Near Eastern Studies Department during two terms as Department Head, and also became the first faculty member of the Committee on Judaic Studies. During his 28 years at the University of Arizona, Bill served as the major adviser to 24 doctoral students in archaeology. He also co-directed the Central Negev Highlands Project with Rudolph Cohen of the Israel Department of Antiquities (now the Israel Antiquities Authority), which involved three seasons of excavation at Beªer Resisim from 1978 to 1980. Bill also served as the Principal Investigator for two excavations led by his students, one directed by Steven Falconer and Bonnie MagnessGardiner at Tell el-Hayyat in Jordan from 1981 to 1985 and the other directed by J. P. Dessel, Beth Alpert Nakhai, and Bonnie Wisthoff at Tell el-Wawiyat in Israel from 1986 to 1987. He returned to Gezer for two seasons of excavation in 1984 and 1990. In 1981–1982, Bill was the John Simon Guggenheim Memorial Fellow in Israel, and in the spring of that year, he

x was awarded the Israel Museum’s Percia Schimmel Prize for distinguished contributions to the Archaeology of Eretz-Israel and the Land of the Bible. Bill has also held a number of distinguished lectureships, including the Winslow Lectureship at Seabury-Western Theological Seminary in Evanston, Illinois, in 1972, the Samuel and Althea Stroum Lectureship at the University of Washington in Seattle in 1985 and again in 1990–1991, and the Archaeological Institute of America’s Charles Eliot Norton Lectureship in 2003–2004. Since the early 1970s, Bill has been a visiting professor at several major universities and has been invited to lecture at institutions and conferences throughout the world. In 1997, the American Schools of Oriental Research (ASOR) honored Bill with the P. E. MacAllister Field Archaeology Award for distinguished contributions to Near Eastern archaeology. Also in 1997, the Hebrew Union College–Jewish Institute of Religion in Cincinnati awarded him the degree of Doctor of Humane Letters honoris causa for his scholarly contributions to the field of Biblical Archaeology. From 1978 to 1984, Bill served as the editor of the Bulletin of the American Schools of Oriental Research, and from 1989 to 1997 as the editor of the Annual of the American Schools of Oriental Research series. As a long-term ASOR trustee, he served as its vice-president from 1982 to 1988 and also as Vice-President of the Cyprus American Archaeological Research Institute from 1994 to 1996. Since 1989, he has also been an AIAR trustee. Bill retired in 2003, and now holds the position of Professor Emeritus of Syro-Palestinian Archaeology at the University of Arizona. In 1989, after Bill and Norma divorced, he and Pamela Jo Gaber were married, and Bill converted to Judaism. Bill, Pamela, Pamela’s daughters/Bill’s stepdaughters Jordana Lee Gaber Saletan and Hannah Susan Gaber Saletan, and Bill and Pamela’s son, Evan Jacob Gaber Dever, born in 1990, now live in Westchester County, New York. Tragically, Bill and Norma’s son, Sean William Dever, after successfully overcoming cancer at an early age, died of an aneurism in 2001. In his memory, Bill and Norma, with the help of their friends, established the annual Sean W. Dever Memorial Prize awarded by AIAR for the best published article or paper presented at a conference by a Ph.D. candidate in Syro-Palestinian and Biblical Archaeology. One of Bill’s most significant and enduring contributions to the field of archaeology—and perhaps the most defining characteristic that emanates from all of his professional endeavors—is his concern and aspiration for increased rigor and professionalism in the discipline as practiced by Americans. In the four primary components of his professional career—fieldwork, scholarship, editing, and teaching—Bill constantly worked to raise the level of

Preface professionalism within “Syro-Palestinian Archaeology” (for many years the term he championed over the more cloistered “Biblical Archaeology,” to which he returned later in his career). His goal in this pursuit was clear: to bring the archaeology of the southern Levant more in line with the anthropologically oriented archaeology that was practiced in the New World. Under Bill’s leadership at Gezer, beginning in the mid-1960s, emphasis was placed on excavation methodology and a multidisciplinary approach to fieldwork that included the systematic collection of animal bones, plant remains, and soil samples and the integration of a geologist, Rueben Bullard, as an essential staff member of the excavation team. While all of this was not entirely innovative, the incorporation of these elements into the explicit conceptual framework of the excavation and its methodology was revolutionary. Some 40 years later, it is now clear that Gezer led the way: these practices soon became standard operating procedures at the vast majority of excavations in Israel and Jordan. The Gezer project was also dedicated to training the next generation of American archaeologists, fostered through its pioneering program of a large-scale archaeological field school. The staff that the project nurtured went on to direct excavations at Ashkelon, Gush Halav, Tell Halif, Meiron, Tel Miqne–Ekron, Nabratein, Sepphoris, and Shema in Israel, at Khirbet Iskander and Tall al-ºUmayri in Jordan, at Idalion on Cyprus, at Carthage in Tunisia, and at Tell Maskuta in Egypt. In his scholarship, Bill emphasized the same set of ideals and aspirations for increased rigor and professionalism, and this had a major impact on the subjects and issues that he addressed. Based on the research he conducted in the 1970s as a follow-up to his Ph.D. dissertation on EB IV (MB I) pottery, he published a number of landmark articles that came to define the character of this period and its material culture. This was also true of his analysis of Iron Age material culture and especially his work on cultic practices in ancient Israelite religion, initiated with his excavation and publication of the Khirbet el-Qôm inscription, which mentions the name of the goddess Asherah, later defined as the consort of Yahweh. Bill also produced expeditious interim field reports, including reports on his excavations at Gezer, Khirbet el-Qôm, Jebel Qaºaqir, Shechem, and Beªer Resisim, in which he continually stressed the integrity of the data and especially the importance of stratigraphic relationships. Among his most important publications are his multivolume final reports on the Gezer excavations. Bill also worked assiduously to decouple archaeological chronology from the biblical narratives, for example, in the publication of his 1972 Winslow lectures, Archaeology and Biblical Studies: Retrospects and Prospects (1972).

Preface But perhaps most importantly, he devoted much effort and thought to explicating the intellectual and philosophical underpinnings of the practice of archaeology in a way that was quite rare for the discipline. For many years after the conclusion of the Gezer excavations, in articles such as “Impact of the ‘New Archaeology’ ” (1988), he wrote forcefully about the importance of processual archaeology. With a distinctive clarity and sense of purpose, Bill sought an understanding of the archaeological data devoid of any biblical reference, which could wait until after the data had been properly excavated, analyzed, contextualized, and thoroughly studied. Only then would it be proper to reconsider the texts (usually the Hebrew Bible) and work at reintegrating them with archaeological and historical evidence. More recently, in articles such as “Archaeology, Texts, and History-Writing: Toward an Epistemology” (1994), Bill has devoted himself to reintegrating a more rigorous style of historical analysis and reshaping the study of Bronze and Iron Age archaeology into a form of historical archaeology more resonant with the wider global archaeological perspective. In his later scholarly work, he has unabashedly achieved this objective, and thus in a way has come full circle to a reformulated and unapologetic form of “Biblical Archaeology.” In this more recent role, Bill has taken the lead in responding to the “minimalists” among biblical scholars and to the few archaeologists who unjustifiably seek to alter the traditional Iron Age chronology and with it the early history of ancient Israel. His passion for these issues and for teaching and sharing his knowledge and interests has had an enormous impact on his students and colleagues, as well as on the lay community—so much so that his more than 300 publications, many of which have become basic reference works, have helped to shape definitively the discipline of the archaeology of ancient Israel. His most recent books, What Did the Biblical Writers Know and When Did They Know It? What Archaeology Can Tell Us about the Reality of Ancient Israel (2001), Who Were the Early Israelites and Where Did They Come From? (2003), and Did God Have a Wife? Archaeology and Folk Religion in Ancient Israel (2005) were written for the serious student, as well as for the general public and have been critically acclaimed by both groups. Bill brought the same approach and dedication that he demonstrated in his scholarship to his editorial work, one

xi of his most important achievements. After the conclusion of fieldwork at Gezer, Bill’s 17-year stewardship as editor of the Bulletin of the American Schools of Oriental Research and the Annual of the American Schools of Oriental Research provided a strong guiding hand in bringing a greater degree of professionalism to these ASOR publications. In his role as editor, he again was able to articulate his all-abiding concern for the integrity of the data and attention to scholarly rigor and in this way had a powerful impact on how ASOR presents itself to its constituency and the public at large. In tandem with his editorial work, Bill endeavored to reshape the pedagogical core of the field of Biblical Archaeology in the United States by initiating a new graduate program in Syro-Palestinian Archaeology at the University of Arizona in 1975. It is rare for an individual scholar single-handedly to build a complete academic graduate program, but this is precisely what he did. In this endeavor as well he maintained the guiding principles that had brought him so much success in his excavation programs and scholarship: professionalism and rigor. Working with the Department of Anthropology, he conceptualized the study of Syro-Palestinian archaeology as part and parcel of the study of archaeological anthropology. He fully appreciated the deep roots archaeology has in anthropology, as well as in history, and encouraged his students to study not only texts but also archaeological method and theory and comparative method. While this has now become standard practice in most graduate programs focusing on the southern Levant, it was highly innovative in 1975. Few scholars are able to contribute to shaping a field of scholarship and to influence a generation of scholars as much as William G. Dever has for the past 40 years. On behalf of all of Bill’s colleagues, students, and friends who have experienced his infectious enthusiasm and passion for scholarship, we express our abiding friendship and our deep sense of gratitude by dedicating this festschrift to him. May he go from strength to strength ºad meªah veºesrim. Seymour Gitin J. Edward Wright J. P. Dessel

William G. Dever: Bibliography Prepared by Wade Kotter Weber State University

Books 1970 1974a 1974b

1984 1987 1990 1994 1998 2001 2003 2005

Gezer I: Preliminary Report of the 1964–66 Seasons. Annual of the Hebrew Union College–Nelson Glueck School of Biblical Archaeology 1. Jerusalem: Hebrew Union College (with H. D. Lance and G. E. Wright). Archaeology and Biblical Studies: Retrospects and Prospects. William C. Winslow Lectures of SeaburyWestern Theological Seminary, 1972. Evanston, IL: Seabury-Wester Theological Seminary. Gezer II: Report of the 1967–70 Seasons in Fields I and II. Annual of the Hebrew Union College–Nelson Glueck School of Biblical Archaeology 2. Jerusalem: Hebrew Union College (editor; co-authored with H. D. Lance, R. G. Bullard, D. P. Cole, and J. D. Seger). The Relationship between Bible, Oriental Studies, and Archaeology from the Perspective of an Archaeologist. Occasional Papers of the Horn Archaeological Museum. Berrien Springs, MI: Andrews University Press. Gezer IV: The 1969–71 Seasons in Field VI, the “Acropolis.” Annual of the Nelson Glueck School of Biblical Archaeology 4. Jerusalem: Hebrew Union College (principal author and editor). Recent Archaeological Discoveries and Biblical Research. Samuel and Althea Stroum Lectures in Jewish Studies. Seattle, WA: University of Washington Press. Scarab Typology and Archaeological Context: An Essay on Middle Bronze Age Chronology. Studies on Scarab Seals 3. San Antonio, TX: Van Siclen (with W. A. Ward). Gezer: At the Crossroads of Ancient Israel. Tel Aviv: Hakibbutz Hameuchad (Hebrew). What Did the Biblical Writers Know and When Did They Know It? What Archaeology Can Tell Us about the Reality of Ancient Israel. Grand Rapids, MI: Eerdmans. Who Were the Early Israelites and Where Did They Come From? Grand Rapids, MI: Eerdmans. Did God Have a Wife? Archaeology and Folk Religion in Ancient Israel. Grand Rapids, MI: Eerdmans.

Edited Volumes 1973 1978 1989 1994 1995

1996 1997 2003

Biblical Archaeology. Jerusalem: Keter (with S. M. Paul). A Manual of Field Excavation: Handbook for Field Archaeologists. Cincinnati, OH: Hebrew Union College (with H. D. Lance). Recent Excavations in Israel: Studies in Iron Age Archaeology. Annual of the American Schools of Oriental Research 49. Winona Lake, IN: Eisenbrauns (with S. Gitin). Preliminary Excavation Reports: Sardis, Paphos, Caesarea Maritima, Shiqmim, ºAin Ghazal. Annual of the American Schools of Oriental Research 51. Baltimore, MD: American Schools of Oriental Research. Preliminary Excavation Reports: Sardis, Bir Umm Fawakhir, Tell el-ºUmeiri, the Combined Caesarea Expeditions, and Tell Dothan. Annual of the American Schools of Oriental Research 52. Ann Arbor, MI: American Schools of Oriental Research. Preliminary Excavation Reports: Sardis, Idalion, and Tell el-Handaquq North. Annual of the American Schools of Oriental Research 53. Boston, MA: American Schools of Oriental Research. The Echoes of Many Texts: Reflections on Jewish and Christian Traditions: Essays in Honor of Lou H. Silberman. Brown Judaic Studies 313. Atlanta, GA: Scholars Press (with J. E. Wright). Symbiosis, Symbolism, and the Power of the Past: Canaan, Ancient Israel and Their Neighbors from the Late Bronze Age through Roman Palaestina. Winona Lake, IN: Eisenbrauns (with S. Gitin).

William G. Dever: Bibliography

xiv

Articles 1967a 1967b 1968a 1968b 1968c 1969a 1969b 1970a 1970b 1970c 1970d 1971a 1971b 1971c 1971d 1972a 1972b 1972c 1972d 1973a 1973b 1973c 1973d 1973e 1974a 1974b

1975a 1975b 1976a

1976b 1977a 1977b

Archaeology and the Six Day War. Biblical Archaeologist 30/2: 73, 102–7. Excavations at Gezer. Biblical Archaeologist 30/2: 47–62. Ethnic Movements in East Central Europe and the Near East, ca. 2300–1800 b.c. Year Book of the American Philosophical Society 1967: 500–503. Excavations at Gezer, 1964–1967. Pp. 26*–32* in Jerusalem through the Ages: The Twenty-Fifth Archaeological Convention, ed. J. Aviram. Jerusalem: Israel Exploration Society. Gezer: A Palestinian Mound Re-excavated. Zeitschrift für Kunstgeschichte und Archäologie 8: 65–74. Gezer, se remet à vivre. Bible et Terre Sainte 116: 8–16. The Water Systems at Hazor and Gezer. Biblical Archaeologist 32/3: 71–78. Gezer: A City Coming to Life. Qadmoniot 3/2: 57–62 (Hebrew). Iron Age Epigraphic Material from the Area of Khirbet el-Kôm. Hebrew Union College Annual 40–41: 139–204. The ‘Middle Bronze I Period’ in Syria and Palestine. Pp. 132–63 in Near Eastern Archaeology in the Twentieth Century: Essays in Honor of Nelson Glueck, ed. J. A. Sanders. Garden City, NY: Doubleday. Vestigial Features in MB I: An Illustration of Some Principles of Ceramic Typology. Bulletin of the American Schools of Oriental Research 200: 19–30. Further Excavations at Gezer, 1967–1971. Biblical Archaeologist 34/4: 94–132 (with H. D. Lance, R. G. Bullard, D. P. Cole, A. M. Furshpan, J. S. Holladay, J. D. Seger, and R. B. Wright). Inscriptions from Israelite Tombs in the Hebron Hills. Qadmoniot 4/3: 90–92 (Hebrew). An MB I Tomb Group from Sinjil. Bulletin of the American Schools of Oriental Research 204: 31–37. The Peoples of Palestine in the Middle Bronze I Period. Harvard Theological Review 64/2–3 (Paul W. Lapp Memorial Issue): 197–226. ‘Biblical Archaeology’—or ‘The Archaeology of Syria Palestine’? Christian News from Israel 22/1: 21–22. Middle Bronze Age I Cemeteries at Mirzbâneh and ºAin-Sâmiya. Israel Exploration Journal 22/2–3: 95–112. A Middle Bronze I Site on the West Bank of the Jordan. Archaeology 25/3: 231–33. William Foxwell Albright: Orientalist Extraordinaire. Christian News from Israel 22/3–4: 172–73. The EB IV–MB I Horizon in Transjordan and Southern Palestine. Bulletin of the American Schools of Oriental Research 210: 37–63. Excavations at Shechem and Mt. Gerazim. Pp. ix–xi in Eretz Shomron: Proceedings of the Thirtieth Archaeological Convention, ed. J. Aviram. Jerusalem: Israel Exploration Society. The Gezer Fortifications and the ‘High Place’: An Illustration of Stratigraphic Methods and Problems. Palestine Exploration Quarterly 105/1: 61–70. Tower 5017 at Gezer: A Rejoinder. Israel Exploration Journal 23/1: 23–26. Two Approaches to Archaeological Method: The Architectural and the Stratigraphic. Eretz-Israel 11 (Dunayevski volume): 1*–8*. The MB IIC Stratification in the Northwest Gate Area at Shechem. Bulletin of the American Schools of Oriental Research 216: 31–52. The Middle Bronze Occupation and Pottery of ºAraq en-Naªsâneh (Cave II). Pp. 33–48 in Discoveries in the Wâdi ed-Dâliyeh, ed. P. W. Lapp and N. L. Lapp. Annual of the American Schools of Oriental Research 41. Cambridge, MA: American Schools of Oriental Research. MB IIA Cemeteries at ºAin es-Sâmiyeh and Sinjil. Bulletin of the American Schools of Oriental Research 217: 23–36. A Middle Bronze I Cemetery at Khribet el-Kirmil. Eretz-Israel 12 (Glueck volume): 18*–33*. The Beginning of the Middle Bronze Age in Syria-Palestine. Pp. 3–38 in Magnalia Dei: The Mighty Acts of God. Essays on the Bible and Archaeology in Memory of G. Ernest Wright, ed. F. M. Cross, W. E. Lemke, and P. D. Miller. Garden City, NY: Doubleday. A Copper Hoard of the Middle Bronze Age I. Israel Exploration Journal 26/4: 163–69 (with M. Tadmor). Palestine in the Second Millennium b.c.e.: The Archaeological Picture. Pp. 70–120 in Israelite and Judean History, ed. J. H. Hayes and J. M. Miller. Philadelphia, PA: Westminster. A Reevaluation of Tell Beit Mirsim Stratum J. Bulletin of the American Schools of Oriental Research 226: 1– 14 (with S. Richard).

William G. Dever: Bibliography 1978a 1978b 1979 1980a 1980b 1980c 1981a 1981b 1981c 1981d 1982a

1982b 1982c 1983a 1983b 1983c

1984a 1984b 1984c 1985a

1985b 1985c 1985d 1985e 1986a

1986b 1986c

xv

Kathleen Kenyon (1906–1978): A Tribute. Bulletin of the American Schools of Oriental Research 232: 3–4. Preliminary Report of the Pilot Season of the “Central Negev Highlands Project.” Bulletin of the American Schools of Oriental Research 232: 29–46 (with R. Cohen). Preliminary Report of the Second Season of the “Central Negev Highlands Project.” Bulletin of the American Schools of Oriental Research 236: 41–60 (with R. Cohen). Archaeological Method in Israel: A Continuing Revolution. Biblical Archaeologist 43/1: 41–48. Cave G26 at Jebel Qaºaqir: A Domestic Assemblage of Middle Bronze I. Eretz-Israel 15 (Aharoni volume): 22*–32*. New Vistas on the EB IV (“MB I”) Horizon in Syria-Palestine. Bulletin of the American Schools of Oriental Research 237: 35–64. Archaeological Method in Israel: A Continuing Revolution. Qadmoniot 14/1: 15–21 (Hebrew). Biblical Theology and Biblical Archaeology: An Appreciation of G. Ernest Wright. Harvard Theological Review 73/1: 1–15. The Impact of the ‘New Archaeology’ on Syro-Palestinian Archaeology. Bulletin of the American Schools of Oriental Research 242: 15–29. Preliminary Report of the Third and Final Season of the “Central Negev Highlands Project.” Bulletin of the American Schools of Oriental Research 243: 57–78 (with R. Cohen). Monumental Architecture in Ancient Israel in the Period of the United Monarchy. Pp. 269–306 in Studies in the Period of David and Solomon and Other Essays, ed. T. Ishida. Tokyo: Yamakawa-Shuppansha / Winona Lake, IN: Eisenbrauns. Recent Archaeological Confirmation of the Cult of Asherah in Ancient Israel. Hebrew Studies 23 (M. Mansour festschrift): 37–43. Retrospects and Prospects in Biblical and Syro-Palestinian Archaeology. Biblical Archaeologist 45/2: 103–7. Beªer Resisim: A Late Third Millennium b.c.e. Settlement. Qadmoniot 16/2–3: 52–57 (Hebrew). The Late Bronze, Iron Age, and Hellenistic Defences at Gezer. Pp. 19–34 in Essays in Honour of Yigael Yadin, ed. G. Vermes and J. Neusner. Totawah, NJ: Allanheld, Osmun. Material Remains and the Cult in Ancient Israel: An Essay in Archeological Systematics. Pp. 571–87 in The Word of the Lord Shall Go Forth: Essays in Honor of David Noel Freedman in Celebration of His Sixtieth Birthday, ed. C. L. Meyers and M. O’Connor. Winona Lake, IN: Eisenbrauns for the American Schools of Oriental Research. Asherah, Consort of Yahweh? New Evidence from Kuntillet ºAjrud. Bulletin of the American Schools of Oriental Research 255: 21–37. Gezer Revisited: New Excavations of the Solomonic and Assyrian Period Defenses. Biblical Archaeologist 47/ 4: 206–18. Yigael Yadin: In Memoriam. Bulletin of the American Schools of Oriental Research 256: 3–5. From the End of the Early Bronze to the Beginning of the Middle Bronze. Pp. 113–35 in Biblical Archaeology Today: Proceedings of the International Congress on Biblical Archaeology, Jerusalem, April 1984, ed. J. Amitai. Jerusalem: Israel Exploration Society. Relations between Syria-Palestine and Egypt in the ‘Hyksos’ Period. Pp. 69–87 in Palestine in the Bronze and Iron Ages: Papers in Honour of Olga Tufnell, ed. J. N. Tubb. London: Institute of Archaeology. Solomonic and Assyrian Period ‘Palaces’ at Gezer. Israel Exploration Journal 35/4: 217–30. Syro-Palestinian and Biblical Archaeology. Pp. 31–79 in The Hebrew Bible and Its Modern Interpreters, ed. D. A. Knight and G. M. Tucker. Philadelphia, PA: Fortress. Village Planning at Beªer Resisim and Socio-Economic Structure in Early Bronze Age IV Palestine. EretzIsrael 18 (Avigad volume): 18*–28*. ªAbel-beth-maºacah: “Northern Gateway of Ancient Israel.” Pp. 207–22 in The Archaeology of Jordan and Other Studies: Essays in Honor of Sigfried F. Horn, ed. L. T. Geraty and L. G. Herr. Berrien Springs, MI: Andrews University Press. The Holy Land: An Archaeologist’s Introduction. Pp. 13–18 in Treasures from the Holy Land: Ancient Art from the Israel Museum. New York, NY: Metropolitan Museum of Art. Late Bronze Age and Solomonic Defenses at Gezer: New Evidence. Bulletin of the American Schools of Oriental Research 262: 9–34.

xvi

William G. Dever: Bibliography

The Contribution of Archaeology to the Study of Canaanite and Early Israelite Religion. Pp. 209–47 in Ancient Israelite Religion: Essays in Honor of Frank Moore Cross, ed. P. D. Miller, P. D. Hanson, and S. D. McBride. Philadelphia: Fortress. 1987b Funerary Practices in EBIV (MBI) Palestine: A Study in Cultural Discontinuity. Pp. 9–19 in Love and Death in the Ancient Near East: Essays in Honor of Marvin H. Pope, ed. J. H. Marks and R. M. Good. Guilford, CT: Four Quarters. 1987c The Middle Bronze Age: The Zenith of the Urban Canaanite Era. Biblical Archaeologist 50/3: 148–77. 1988a Cave I.3 A: The Pottery. Pp. 21–33 in Gezer V: The Caves in Field I, by J. D. Seger. Annual of the Nelson Glueck School of Biblical Archaeology 5. Jerusalem: Hebrew Union College. 1988b Impact of the ‘New Archaeology’. Pp. 337–52 in Benchmarks in Time and Culture: An Introduction to Palestinian Archaeology, ed. J. F. Drinkard, G. L. Mattingly, and J. M. Miller. Atlanta, GA: Scholars Press. 1989a Archaeology in Israel Today: A Summation and Critique. Pp. 143–52 in Recent Excavations in Israel: Studies in Iron Age Archaeology, ed. S. Gitin and W. G. Dever. Annual of the American Schools of Oriental Research 49. Winona Lake, IN: Eisenbrauns. 1989b The Collapse of the Urban Early Bronze Age in Palestine: Toward a Systemic Analysis. Pp. 225–46 in L’urbanisation de la Palestine à L’âge du Bronze ancien: Bilan et perspectives des recherches actuelles, ed. P. de Miroschedji. British Archaeological Reports International Series 527. Oxford: British Archaeological Reports. 1989c In Memoriam: Yigal Shiloh 1937–1987. Bulletin of the American Schools of Oriental Research 274: 1–2. 1989d Metalworking Technology at the End of the Early Bronze Age in the Southern Levant. Bulletin of the Institute for Archaeo-Metallurgical Studies 14: 1–4 (with J. F. Merkel). 1989e Yigael Yadin: Prototypical Biblical Archaeologist. Eretz-Israel 20 (Yadin volume): 44*–51*. 1989–1990 Religion and Cult in Ancient Israel: Social and Economic Implications. Pp. 175–80 in Atti del Convegno Internazionale Anathema: Regime delle offerte e vita dei santuari nel Mediterraneo antico, 15–18 giugno 1989, Roma, ed. G. Bartoloni, G. Colonna, and C. Grottanelli. Scienze dell’Antichità 3–4. Rome: Università degli Studi di Roma “La Sapienza.” 1990 Of Myths and Methods. Bulletin of the American Schools of Oriental Research 277–78: 121–30. 1991a Archaeology, Material Culture and the Early Monarchical Period in Israel. Pp. 103–15 in The Fabric of History: Text, Artifact and Israel’s Past, ed. D. V. Edelman. Journal for the Study of the Old Testament Supplement Series 127. Sheffield: Sheffield Academic Press. 1991b Tell el-Dabºa and Levantine Middle Bronze Age Chronology: A Rejoinder to Manfred Bietak. Bulletin of the American Schools of Oriental Research 281: 73–79. 1991c Unresolved Issues in the Early History of Israel: Toward a Synthesis of Archaeological and Textual Reconstructions. Pp. 195–208 in The Bible and the Politics of Exegesis: Essays in Honor of Norman K. Gottwald on His Sixty-Fifth Birthday, ed. D. Jobling, P. L. Day, and G. T. Sheppard. Cleveland, IL: Pilgrim. 1992a A Case-Study in Biblical Archaeology: The Earthquake of ca. 760 bce. Eretz-Israel 23 (Biran volume): 27*–35*. 1992b The Chronology of Syria-Palestine in the Second Millennium b.c. Ägypten und Levante 3: 39–51. 1992c The Chronology of Syria-Palestine in the Second Millennium b.c.e.: A Review of Current Issues. Bulletin of the American Schools of Oriental Research 288: 1–25. 1992d Dever’s Reply to Finkelstein, Gottwald and Zertal. Pp. 79–85 in The Rise of Ancient Israel, ed. H. Shanks. Washington, DC: Biblical Archaeology Society. 1992e How to Tell an Israelite from a Canaanite. Pp. 27–56 in The Rise of Ancient Israel, ed. H. Shanks. Washington, DC: Biblical Archaeology Society. 1992f The Late Bronze–Early Iron I Horizon in Syria-Palestine: Egyptians, Canaanites, ‘Sea Peoples,’ and ‘ProtoIsraelites.’ Pp. 99–110 in The Crisis Years: The 12th Century b.c.: From Beyond the Danube to the Tigris, ed. W. A. Ward and M. S. Joukowsky. Dubuque, IA: Kendall/Hunt. 1992g Pastoralism and the End of the Early Bronze Age in Palestine. Pp. 83–93 in Pastoralism in the Levant: Archaeological Materials in Anthropological Perspectives, ed. O. Bar-Yosef and A. Khazanov. Monographs in World Archaeology 10. Madison, WI: Prehistory Press. 1993a Biblical Archaeology: Death or Rebirth? Pp. 706–22 in Biblical Archaeology Today, 1990: Proceedings of the Second International Congress on Biblical Archaeology, Jerusalem, June–July 1990, ed. A. Biran and J. Aviram. Jerusalem: Israel Exploration Society. 1987a

William G. Dever: Bibliography 1993b 1993c

1993d 1994a

1994b 1994c 1994d

1995a 1995b 1995c

1995d 1995e 1995f 1995g 1995h 1996a 1995b

1995c 1995d

1996a 1996b 1997a 1997b

1997c

xvii

Further Evidence on the Date of the Outer Wall at Gezer. Bulletin of the American Schools of Oriental Research 289: 33–54. The Rise of Complexity in the Land of Israel in the Early Second Millennium b.c. Pp. 98–109 in Biblical Archaeology Today, 1990: Proceedings of the Second International Congress on Biblical Archaeology, Jerusalem, June–July 1990, ed. A. Biran and J. Aviram. Jerusalem: Israel Exploration Society. What Remains of the House Albright Built? Biblical Archaeologist 56/1: 25–35. Ancient Israelite Religion: How to Reconcile the Differing Textual and Archaeological Portraits. Pp. 105–25 in Ein Gott allein? JHWH—Verehrung und biblischer Monotheismus im Kontext der israelitischen und altorientalischen Religionsgeschichte, ed. W. Dietrich and M. Klopfenstein. Orbis biblicus et orientalis 139. Freiburg: Universitätsverlag. Archaeology, Texts, and History-Writing: Toward an Epistemology. Pp. 105–17 in Uncovering Ancient Stones: Essays in Memory of H. Neil Richardson, ed. L. M. Hopfe. Winona Lake, IN: Eisenbrauns. From Tribe to Nation: A Critique of State Formation Processes in Ancient Israel. Pp. 213–38 in Nouve fondazioni nel Vicino Oriente Antico: Realta ed ideologia, ed. S. Mazzoni. Pisa: University of Pisa. The Silence of the Text: An Archaeological Commentary on 2 Kings 23. Pp. 143–68 in Scripture and Other Artifacts: Essays on the Bible and Archaeology in Honor of Philip J. King, ed. M. D. Coogan, J. C. Exum, and L. E. Stager. Louisville, KY: Westminster/John Knox. Ceramics, Ethnicity and the Question of Israel’s Origins. Biblical Archaeologist 58/4: 200–13. The Death of a Discipline. Biblical Archaeology Review 21/5: 50–55. Orienting the Study of Trade in Near Eastern Archaeology. Pp. 111–19 in Recent Excavations in Israel: A View to the West. Reports on Kabri, Nami, Miqne-Ekron, Dor, and Ashkelon, ed. S. Gitin. Archaeological Institute of America Colloquia and Conference Papers 1. Dubuque, IA: Kendall/Hunt. Palaces and Temples in Canaan and Ancient Israel. Pp. 605–14 in vol. 1 of Civilizations of the Ancient Near East, ed. J. M. Sasson. New York, NY: Scribner. Social Structure in Palestine in the Iron II Period on the Eve of Destruction. Pp. 416–31 in The Archaeology of Society in the Holy Land, ed. T. E. Levy. London: Leicester University Press. Social Structure in the Early Bronze IV Period in Palestine. Pp. 282–96 in The Archaeology of Society in the Holy Land, ed. T. E. Levy. London: Leicester University Press. Will the Real Israel Please Stand Up? Part I: Ideology and Israelite Historiography. Bulletin of the American Schools of Oriental Research 297: 61–80. Will the Real Israel Please Stand Up? Part II: Archaeology and the Religions of Ancient Israel. Bulletin of the American Schools of Oriental Research 298: 37–58. Archaeology and the Current Crisis in Israelite Historiography. Eretz-Israel 25 (Aviram volume): 18*–27*. Idalion, Cyprus: Conquest and Continuity. Pp. 85–113 in Preliminary Excavation Reports: Sardis, Idalion, and Tell el-Handaquq North, ed. W. G. Dever. Annual of the American Schools of Oriental Research 53. Boston, MA: American Schools of Oriental Research (with P. Gaber). The Importance of Research Design. Pp. 37–48 in Archaeology’s Publication Problem, ed. H. Shanks. Washington, DC: Biblical Archaeology Society. The Tell: Microcosm of the Cultural Process. Pp. 37–45 in Retrieving the Past: Essays on Archaeological Research and Methodology in Honor of Gus W. Van Beek, ed. J. D. Seger. Winona Lake, IN: Eisenbrauns for the Cobb Institute of Archaeology. The Identity of Early Israel: A Rejoinder to Keith W. Whitelam. Journal for the Study of the Old Testament 72: 3–24. Revisionist Israel Revisited: A Rejoinder to Niels Peter Lemche. Currents in Research: Biblical Studies 4: 35–52. Archaeology and the ‘Age of Solomon’: A Case-Study in Archaeology and Historiography. Pp. 217–51 in The Age of Solomon: Scholarship at the Turn of the Millennium, ed. L. K. Handy. Leiden: Brill. Archaeology, Urbanism, and the Rise of the Israelite State. Pp. 172–93 in Aspects of Urbanism in Antiquity: From Mesopotamia to Crete, ed. W. E. Aufrecht, N. A. Mirau, and S. W. Gauley. Sheffield: Sheffield Academic Press. Chronology and Settlement Patterns of Palestine in the Middle Bronze Age. Pp. 285–301 in The Hyksos: New Historical and Archaeological Perspectives, ed. E. D. Oren. Philadelphia, PA: University Museum.

xviii 1997d 1997e

1997f 1997g 1997h

1997i

1998a 1998b

1998c 1999a

1999b 1999c

2000a 2000b 2000c 2000d

2000e 2001a

2001b

2002 2003a 2003b 2003c

William G. Dever: Bibliography Folk Religion in Early Israel: Did Yahweh Have a Consort? Pp. 27–56 in Aspects of Monotheism: How God Is One, ed. H. Shanks. Washington, DC: Biblical Archaeology Society. Hurrian Incursions and the End of the Middle Bronze Age in Syria-Palestine: A Rejoinder to Nadav Naªaman. Pp. 91–100 in Ancient Egyptian and Mediterranean Studies in Memory of William A. Ward, ed. L. H. Lesko. Providence, RI: Brown University Press. Is There Any Archaeological Evidence for the Exodus? Pp. 67–86 in Exodus: The Egyptian Evidence, ed. E. S. Frerichs and L. H. Lesko. Winona Lake, IN: Eisenbrauns. New Archaeological Data and Hypotheses on the Origins of Ancient Israel. Pp. 20–50 in Archaeology and Biblical Interpretation, ed. J. R. Bartlett. London: Routledge. On Listening to the Texts—and the Artifacts. Pp. 1–23 in The Echoes of Many Texts: Reflections on Jewish and Christian Traditions: Essays in Honor of Lou H. Silberman, ed. W. G. Dever and J. E. Wright. Brown Judaic Studies 313. Atlanta, GA: Scholars Press. Philology, Theology, and Archaeology: What Kind of History Do We Want, and What Is Possible? Pp. 290– 310 in The Archaeology of Israel: Constructing the Past, Interpreting the Present, ed. N. A. Silberman and D. Small. Sheffield: Sheffield Academic Press. Archaeology, Ideology, and the Quest for an “Ancient” or “Biblical” Israel. Near Eastern Archaeology 61/1: 39–52. Israelite Origins and the ‘Nomadic Ideal’: Can Archaeology Separate Fact from Fiction? Pp. 220–37 in Mediterranean Peoples in Transition: Thirteenth to Early Tenth Centuries bce, ed. S. Gitin, A. Mazar, and E. Stern. Jerusalem: Israel Exploration Society. What Did the Biblical Writers Know and When Did They Know It? Pp. 241–57 in Hesed ve-Emet: Studies in Honor of Ernest S. Frerichs, ed. J. Magness and S. Gitin. Atlanta, GA: Scholars Press. American, Palestinian and Biblical Archaeology: End of an Era? Pp. 91–102 in Assembling the Past: Studies in the Professionalization of Archaeology, ed. A. B. Kehoe and M. B. Emmerichs. Albuquerque, NM: University of New Mexico Press. Archaeology and the Israelite Cult: How the Kh. el-Qôm and Kuntillet ºAjrûd ‘Asherah’ Texts Have Changed the Picture. Eretz-Israel 26 (Cross volume): 8*–15*. Can ‘Biblical Archaeology’ Be an Academic and Professional Discipline? Pp. 11–22 in Archaeology, History and Culture in Palestine and the Near East: Essays in Memory of Albert E. Glock, ed. T. Kapitan. Atlanta, GA: Scholars Press. Biblical and Syro-Palestinian Archaeology: A State-of-the-Art Assessment at the Turn of the Millennium. Currents in Research: Biblical Studies 8: 91–116. The Case of Ancient Israel: Unique or Not? Pp. 63–67 in The Breakout: The Origins of Civilization, ed. C. C. Lamberg-Karlovsky and M. Lamberg-Karlovsky. Cambridge, MA: Harvard University Press. Marginalia: Reply to Davies. Currents in Research: Biblical Studies 8: 124–33. Nelson Glueck and the Other Half of the Holy Land. Pp. 114–21 in The Archaeology of Jordan and Beyond: Essays in Honor of James A. Sauer, ed. L. E. Stager, J. A. Greene, and M. D. Coogan. Harvard Semitic Museum Studies in the Archaeology and History of the Levant 1. Winona Lake, IN: Eisenbrauns. Save Us from Postmodern Malarkey. Biblical Archaeology Review 26/2: 28–37. Iron Age Kernoi and the Israelite Cult. Pp. 119–33 in Studies in the Archaeology of Israel and Neighboring Lands in Memory of Douglas L. Esse, ed. S. R. Wolff. Studies in Ancient Oriental Civilizations 59. Chicago, IL: Oriental Institute. The Silver Train: Response to the Papers of E. Stern and S. Gitin. Pp. 49–51 in Hacksilber to Coinage: New Insights into the Monetary History of the Near East and Greece, ed. M. S. Balmuth. New York, NY: American Numismatic Society. Theology, Philology, and Archaeology: In the Pursuit of Ancient Israelite Religion. Pp. 11–33 in Sacred Time, Sacred Place: Archaeology and the Religion of Israel, ed. B. M. Gittlen. Winona Lake, IN: Eisenbrauns. Chronology of the Southern Levant. Pp. 82–87 in Near Eastern Archaeology: A Reader, ed. S. Richard. Winona Lake, IN: Eisenbrauns. An EB IV Tomb Group from Tell Beit Mirsim. Eretz-Israel 27 (H. Tadmor and M. Tadmor volume): 29*–36*. Foreword. Pp. ix–xi in Near Eastern Archaeology: A Reader, ed. S. Richard. Winona Lake, IN: Eisenbrauns.

William G. Dever: Bibliography 2003d

2003e 2003f

2003g

2003h

2003i 2004a 2004b 2004c 2005

2006

xix

The Patriarchs and Matriarchs of Ancient Israel: Myth or History? Pp. 39–56 in One Hundred Years of American Archaeology in the Middle East: Proceedings of the American Schools of Oriental Research Centennial Celebration, Washington, DC, April 2000, ed. D. R. Clark and V. H. Matthews. Boston, MA: American Schools of Oriental Research. Religion and Cult in the Levant: The Archaeological Data. Pp. 383–90 in Near Eastern Archaeology: A Reader, ed. S. Richard. Winona Lake, IN: Eisenbrauns. The Rural Landscape of Palestine in the Early Bronze IV Period. Pp. 43–59 in The Rural Landscape of Ancient Israel, ed. A. M. Maeir, S. Dar, and Z. Safrai. British Archaeological Reports International Series 1121. Oxford: Archeopress. Some Observations on Comparative Method and Ancient Music in Near Eastern Archaeology. Pp. 1–21 in Sources of Music in Ancient Israel/Palestine, ed. R. Randhofer. Orientwissenschaftliche Hefte 8. Halle: MartinLuther-Universität. Syro-Palestinian and Biblical Archaeology: Into the Next Millennium. Pp. 513–27 in Symbiosis, Symbolism, and the Power of the Past: Canaan, Ancient Israel, and Their Neighbors from the Late Bronze Age through Roman Palaestina, ed. W. G. Dever and S. Gitin. Winona Lake, IN: Eisenbrauns. Whatchamacalit: Why It’s So Hard to Name Our Field. Biblical Archaeology Review 29/4: 56–61. Histories and Non-Histories of Ancient Israel: The Question of the United Monarchy. Pp. 65–94 in In Search of Pre-Exilic Israel, ed. J. Day. New York, NY: Continuum. Kathleen Kenyon (1906–1978). Pp. 525–53 in Breaking Ground: Pioneering Women Archaeologists, ed. G. M. Cohen and M. S. Joukowsky. Ann Arbor, MI: University of Michigan Press. Walter E. Rast, 1930–2003. Biblical Archaeology Review 30/1: 15–16. Some Methodological Reflections on Chronology and History-Writing. Pp. 413–21 in Radiocarbon Dating and the Iron Age of the Southern Levant: The Bible and Archaeology Today, ed. T. E. Levy and T. Higham. London: Equinox. Archaeology and Ancient Israelite Iconography: Did Yahweh Have a Face? Pp. 461–76 in “I Will Speak the Riddles of Ancient Times”: Archaeological and Historical Studies in Honor of Amihai Mazar on the Occasion of His Sixtieth Birthday, ed. A. M. Maeir and P. de Miroschedji. Winona Lake, IN: Eisenbrauns.

In Press 1. 2.

3. 4.

5. 6. 7.

Histories and Non-Histories of Ancient Israel: What Archaeology Can Contribute. In Recenti tendenze nella ricostruzione della stora di Israel, ed. M. Liverani. Rome: Academia dei Lincei. Archaeology and the Fall of the Northern Kingdom: What Really Happened. In the “Up to the Gates of Ekron” (1 Samuel 17:52): Essays on the Archaeology and History of the Eastern Mediterranean in Honor of Seymour Gitin, ed. S. W. Crawford et al. Jerusalem: Israel Exploration Society. Archaeology and the Question of Sources in Kings. In The Book of Kings: Sources, Composition, Historiography and Reception, ed. A. Lemaire and B. Halpern. Leiden: Brill. Ethnicity and the Archaeological Record: The Case of Early Israel. In The Archaeology of Difference: Gender, Ethnicity, Class, and the “Other” in Antiquity: Essays in Honor of Eric M. Meyers, ed. D. R. Edwards and C. T. McCollough. 2 vols. Annual of the American Schools of Oriental Research 60–61. Atlanta, GA: American Schools of Oriental Research. Ahab and Archaeology: A Commentary on 1 Kings 16–22. In the Shalom Paul Jubilee Volume, ed. C. Cohen, B. Schwartz, and V. Hurowitz. Winona Lake, IN: Eisenbrauns. The Middle Bronze Age I. In The Ancient Pottery of Israel and Its Neighbors from the Neolithic through the Hellenistic Period, ed. S. Gitin. Jerusalem: Israel Exploration Society. Religion and Cult in Early Bronze Palestine. In the Walter E. Rast Memorial Volume, ed. M. Chesson. Winona Lake, IN: Eisenbrauns.

Review Articles 1971

Archaeological Methods and Results: A Review of Two Recent Publications (O. R. Sellers et al., The 1957 Excavations at Beth Zur, Cambridge: American Schools of Oriental Research, 1968; W. F. Albright and J. L.

William G. Dever: Bibliography

xx

1990

1991

1993

1996

1999

2001

Kelso, The Excavations of Bethel 1934–1960, Cambridge: American Schools of Oriental Research, 1968). Orientalia 40/4: 459–71. Archaeology and Israelite Origins: Review Article (V. Fritz and A. Kempinski, Ergebnisse der Ausgrabungen auf der Hirbet él-Msas (ˇel Msas) 1972–1975, Pts. I–III, Wiesbaden: Harrassowitz, 1983). Bulletin of the American Schools of Oriental Research 279: 89–95. Archaeological Data on the Israelite Settlement: A Review of Two Recent Works (I. Finkelstein, ºIsbet Íar†ah: An Early Iron Age Site near Rosh Haºayin, Israel, Oxford: British Archaeological Reports, 1986; I. Finkelstein The Archaeology of the Israelite Settlement, Jerusalem: Israel Exploration Society, 1988). Bulletin of the American Schools of Oriental Research 284: 77–90. Syro-Palestinian Archaeology “Comes of Age”: The Inaugural Volume of the Hesban Series—A Review Article (O. S. LaBianca, Hesban I: Sedentarization and Nomadization: Food System Cycles at Hesban and Vicinity in Transjordan, Berrien Springs, MI: Andrews University Press, 1990). Bulletin of the American Schools of Oriental Research 290–91: 127–30. Archaeology and the Religions of Israel (R. Albertz, A History of Israelite Religion in the Old Testament Period I: From the Beginnings to the End of the Monarchy, Louisville, KY: Westminster/John Knox, 1994). Bulletin of the American Schools of Oriental Research 301: 83–90. Histories and Non-Histories of Ancient Israel (L. L. Grabbe, ed., Can a “History of Israel” Be Written? Sheffield: Sheffield Academic Press, 1997; K. W. Whitelam, The Invention of Ancient Israel: The Silencing of Palestinian History, New York, NY: Routledge, 1996). Bulletin of the American Schools of Oriental Research 316: 89–105. Excavating the Hebrew Bible, or Burying It Again? (I. Finkelstein and N. A. Silberman, The Bible Unearthed: Archaeology’s New Vision of Ancient Israel and the Origin of Its Sacred Texts, New York, NY: Free Press, 2000). Bulletin of the American Schools of Oriental Research 322: 67–77.

Reports 1966 1967 1968 1969a 1969b 1969c 1970a 1970b 1971a 1971b 1971c 1972a 1972b 1972c 1972d 1973a 1973b 1973c 1973d 1974 1975 1978 1979a 1979b 1980a

Gezer (Notes and News). Israel Exploration Journal 16/4: 277–78. Gezer (Notes and News). Israel Exploration Journal 17/4: 274–75. Gezer (Chronique archéologique). Revue Biblique 75/3: 572–76. Gezer (Chronique archéologique). Revue Biblique 76/4: 563–67. Gezer (Notes and News). Israel Exploration Journal 19/4: 241–43. Khalit el-Fûl (Chronique archéologique). Revue Biblique 76/4: 572–76. Gezer (Chronique archéologique). Revue Biblique 77/3: 394–98. Gezer (Notes and News). Israel Exploration Journal 20/3–4: 226–27. Djebel Qaºaqir (Chronique archéologique). Revue Biblique 78/4: 595–97. Gezer (Chronique archéologique). Revue Biblique 78/3: 425–28. Jebel Qaºaqir (Notes and News). Israel Exploration Journal 21/4: 229–30. Gezer (Chronique archéologique). Revue Biblique 79/3: 413–18. Shechem (Notes and News). Israel Exploration Journal 22/2–3: 156–57. Shechem (Notes and News). Israel Exploration Journal 22/4: 239–40. Tel Gezer (Notes and News). Israel Exploration Journal 22/2–3: 158–60. Report of the Director of the Albright Institute, Jerusalem. Bulletin of the American Schools of Oriental Research 212: 2–4. Report of the Director of the Jerusalem School. Bulletin of the American Schools of Oriental Research 209: 5–8. Shechem (Notes and News). Israel Exploration Journal 23/4: 243–45. Sichem (Chronique archéologique). Revue Biblique 80/4: 567–70. Report of the Director of the Albright Institute. Bulletin of the American Schools of Oriental Research 216: 1–3. Sichem (Chronique archéologique). Revue Biblique 82/1: 81–83. Ó. Beªer Resisim (Notes and News). Israel Exploration Journal 28/4: 263–64 (with R. Cohen). Beªer Resisim, 1979 (Notes and News). Israel Exploration Journal 29/3–4: 254–55 (with R. Cohen). Exploration de Haut Negev Central (1978) (Chronique archéologique). Revue Biblique 86/3: 466–67. Beªer Resisim, 1980 (Notes and News). Israel Exploration Journal 30/3–4: 228–31 (with R. Cohen).

William G. Dever: Bibliography 1980b 1981a 1981b 1981c 1983 1985a 1985b 1985c 1986 1987a 1987b 1991 1994

xxi

The Second Season of the Central Negev Highlands Project (Israel). American Journal of Archaeology 84/2: 204. Beªer Resisim (1980) (Chronique archéologique). Revue Biblique 88/4: 571–73 (with R. Cohen). Ó. Beªer Resisim: 1980. Óadashot Arkheologiyot 76: 36–37 (with R. Cohen) (Hebrew). The 1980 Season of the Central Negev Highlands Project (Israel). American Journal of Archaeology 85/2: 192. Bronze Age Village Agriculture in the Jordan Valley: Tell el-Hayyat, 1982. American Journal of Archaeology 87/2: 231–32. Tel Gezer: 1984. Excavations and Surveys in Israel 3: 30–31. Tel Gezer (1984) (Chronique archéologique). Revue Biblique 92/3: 412–19. Tel Gezer, 1984 (Notes and News). Israel Exploration Journal 35/1: 64–65. The 1984 Season at Gezer: Late Bronze and Iron Age City Defenses. American Journal of Archaeology 90/2: 223. Recent Archaeological Discoveries Bearing on the Early Bronze IV Period in Jordan. American Journal of Archaeology 91/2: 275–76. Typological and Technological Comparisons of Late 3rd-Millennium bc Metal Implements from Palestine. American Journal of Archaeology 91/2: 307 (with J. F. Merkel). Tell Gezer, 1990 (Notes and News). Israel Exploration Journal 41/4: 282–86 (with R. Younker). Tel Gezer 1990. Excavations and Surveys in Israel 12: 48–49 (with R. Younker).

Encyclopedia and Dictionary Entries 1970 1976a 1976b 1985 1990

1992

1993

1995 1996a 1996b 1997

Gezer. Pp. 125–26 in Dictionnaire archéologique de la Bible, ed. A. Negev. Paris: Hazan. Archaeology (44–52); Aroer (55); Gezer (361–63) in The Interpreter’s Dictionary of the Bible: Supplementary Volume, ed. K. R. Crim. Nashville, TN: Abingdon. Gezer (2: 428–43); El-Qôm, Khirbet (4: 976–77) in The Encyclopedia of Archaeological Excavations in the Holy Land, ed. M. Avi-Yonah. Jerusalem: Israel Exploration Society. Amaziah (26); Archaeology, Methods of (53–59); Battering Ram (97–98); Gezer (343–45); Weights and Measures (1126–31) in Harper’s Bible Dictionary, ed. P. J. Achtemeier. New York, NY: Harper and Row. Ezion-Geber (284–85); Gezer (328–29); Hazor (359–60); Heshbon (378–79); Hittites (382–83); Jericho (438– 40); Pottery Chronology (702–4); Shechem (816–17) in The Mercer Dictionary of the Bible, ed. W. E. Mills. Macon, GA: Mercer University Press. ºAffula, el- (1: 87–88); Archaeology, Syro-Palestinian and Biblical (1: 354–67); Areini, Tell el- (1: 651–52); Beer-Resisim (1: 640–41); Beit Mirsim, Tell (1: 648–49); Beitin (1: 651–52); Dhahr Mirzbâneh (2: 184–85); Dothan (2: 226–27); Gezer (2: 998–1003); Hazor (3: 87–88); Israel, History of (Archaeology and the Israelite Conquest) (3: 545–58); Nahariyeh (4: 995–96); Palace (5: 56–58); Palestine, Archaeology of (Bronze and Iron Ages) (5: 109–14); Qaºaqir, Jebel (5: 572–73); Temples and Sanctuaries (Syria-Palestine) (6: 376–80) in The Anchor Bible Dictionary, ed. D. N. Freedman. New York, NY: Doubleday. Beªer Resisim (1: 158–60); Gezer (2: 496–506); Jebel Qaºaqir (2: 665–67); Qôm, Khirbet el- (4: 1233–35) in The New Encyclopedia of Archaeological Excavations in the Holy Land, ed. E. Stern. Jerusalem: Israel Exploration Society. Israele: Musei e Istituzioni. Pp. 133–34 in vol. 4 of Enciclopedia dell’Arte Antica, Classica e Orientale. Rome: Treccani. Archaeology, Methods of (pp. 59–66); Gezer (pp. 374–75) in HarperCollins Bible Dictionary, ed. P. J. Achtemeier. San Francisco, CA: HarperSan Francisco. Palestina. Pp. 281–25 in Enciclopedia dell’Arte Antica, Classica e Orientale: Secondo Supplemento, 1971– 1994. Rome: Treccani. Abu Hawam, Tell (1: 9); ºAin es-Samiyeh (1: 35–36); Akko (1: 54–55); Ashdod (1: 219–20); Beªer Resisim (1: 285–87); Bethel (1: 300–301); Beth-Shemesh (1: 311–12); Biblical Archaeology (1: 315–19); Ceramics: Syro-Palestinian Ceramics of the Neolithic, Bronze, and Iron Ages (1: 459–65); Coele-Syria (2: 41); Daliyeh, Wadi ed- (2: 101–2); Gezer (2: 396–400); Hermon, Mount (3: 15–16); Jebel Qaºaqir (3: 210–11); Kabri, Tel (3: 261); Keisan, Tel (3: 278–79); Levant (3: 350–51); Macalister, Robert Alexander Stewart (3: 390–91); Mor, Tel (4: 49–50); Qôm, Khirbet el- (4: 391–92); Shaft Tombs (5: 14–15); Syria-Palestine (5: 147); Timnaº

William G. Dever: Bibliography

xxii

1999a

1999b 2000

(5: 217–18) in The Oxford Encyclopedia of Archaeology in the Near East, ed. E. M. Meyers. New York, NY: Oxford University Press. Aharoni, Yohanan (1: 18–19); Glueck, Nelson (1: 450–51); Mazar, Benjamin (2: 141); Sukenik, Eliezer (2: 515); Wright, George E (2: 661–62); Yadin, Yigael (2: 664) in The Dictionary of Biblical Interpretation, ed. J. H. Hayes. Nashville, TN: Abingdon. Israelites. Pp. 382–84 in The Encyclopedia of the Archaeology of Ancient Egypt, ed. K. A. Bard. New York, NY: Routledge. Abel Beth-Maacah (pp. 3–4); Gezer (pp. 499–500); Makkedah (pp. 847–48) in Eerdmans Dictionary of the Bible, ed. D. N. Freedman. Grand Rapids, MI: Eerdmans.

Reviews 1968 1971 1973 1975 1976 1977a 1977b 1979 1981a 1981a

1981c 1982a 1982b 1983a 1983b

1983c 1985 1986 1989a 1989b

Y. Aharoni, The Land of the Bible: A Historical Geography. Philadelphia: Westminster, 1967. Journal of Biblical Literature 87/2: 229–31. R. Amiran, Ancient Pottery of the Holy Land. New Brunswick, NJ: Rutgers University Press, 1970. Israel Exploration Journal 21/1: 69–72. Excavation Reports on Palestinian Archaeology: Some Review Notices. Bulletin of the American Schools of Oriental Research 212: 32–36. J. Mallet, Tell el-Farºah (Région de Naplouse): L’installation du Moyen Bronze antérieure au rampart. Paris: Gabalda, 1973. Journal of Biblical Literature 94/4: 607–8. K. M. Kenyon, Digging Up Jerusalem. London: Benn, 1974. American Anthropologist 78/3: 711–12. R. S. Boraas and S. H. Horn, Heshbon 1973: The Third Campaign at Tell Hesban. A Preliminary Report. Berrien Springs, MI: Andrews University Press, 1975. Journal of Biblical Literature 96/4: 578–80. J. A. Sauer, Heshbon Pottery 1971, Berrien Springs, MI: Andrews University Press, 1973. Bulletin of the American Schools of Oriental Research 227: 78–79. O. Negbi, Canaanite Gods in Metal: An Archaeological Study of Ancient Syro-Palestinian Figurines. Tel Aviv: Tel Aviv University, 1976. Journal of Biblical Literature 98/1: 101–3. J. A. Blakely and L. E. Toombs, The Tell el-Hesi Field Manual. Cambridge, MA: American Schools of Oriental Research, 1980. Bulletin of the American Schools of Oriental Research 242: 86–87. M. S. Joukowsky, A Complete Manual of Field Archaeology: Tools and Techniques of Field Work for Archaeologists. Englewood Cliffs, NJ: Prentice-Hall, 1980. Bulletin of the American Schools of Oriental Research 242: 86. A. Kempinski, The Rise of an Urban Culture: The Urbanization of Palestine in the Early Bronze Age, 3000– 2150 b.c. Jerusalem: Israel Ethnographic Society, 1978. Israel Exploration Journal 31/4: 254–57. R. Amiran, Early Arad: The Chalcolithic Settlement and Early Bronze City. First–Fifth Seasons of Excavation, 1962–1966. Jerusalem: Israel Exploration Society, 1978. Israel Exploration Journal 32/2–3: 170–75. E. Stern, Excavations at Tel Mevorakh (1973–1976), Pt. 1: From the Iron Age to the Roman Period. Jerusalem: Hebrew University, 1978. Journal of the American Oriental Society 102/2: 399–400. Y. Aharoni, The Land of the Bible: An Historical Geography (rev. ed.). Philadelphia: Westminster, 1979. Journal of Biblical Literature 102/1: 119–20. D. N. Freedman, ed., Archaeological Reports from the Tabqa Dam Project—Euphrates Valley, Syria. Cambridge, MA: American Schools of Oriental Research, 1979. Bulletin of the American Schools of Oriental Research 249: 94–95. P. R. S. Moorey and P. J. Parr, eds., Archaeology in the Levant: Essays for Kathleen Kenyon. Warminster: Aris & Phillips, 1978. Bulletin of the American Schools of Oriental Research 251: 70–71. B. Mazar, ed., Eretz-Israel 15 (Aharoni volume). Jerusalem: Israel Exploration Society, 1980. Journal of Near Eastern Studies 44/1: 71–72. Megiddo: City of Destruction (documentary film). Shevat Films, 1982. Archaeology 39/1: 66, 72. D. P. Cole, Shechem I: The Middle Bronze IIB Pottery. Winona Lake, IN: Eisenbrauns, 1984. Bulletin of the American Schools of Oriental Research 276: 86–88. M. S. Drower, Flinders Petrie: A Life in Archaeology. London: Gollancz, 1985. Biblical Archaeologist 52/3: 146.

William G. Dever: Bibliography

xxiii

J. Mallet, Tell el-Farºah II/1. 2: Le Bronze Moyen. Stratigraphie des vestiges du Bronze moyen II (1ere moitié du IIe millénaire av. J.C.) dans les chantiers principaux II nord et IV. Paris: Éditions Recherche sur les Civilisations, 1987. Paléorient 15: 154–58. 1989d P. de Miroschedji, Yarmouth 1: Rapport sur les trois premières campagnes de fouilles à Tel Yarmouth, Israël (1980–1982). Paris: Éditions Recherche sur les Civilisations, 1988. Paléorient 15: 153–54. 1989e L. G. Perdue, L. E. Toombs, and G. L. Johnson, eds., Archaeology and Biblical Interpretation: Essays in Memory of D. Glenn Rose. Atlanta: John Knox, 1987. Shofar 7/4: 66–67. 1989f A. D. Tushingham, Excavations in Jerusalem 1961–1967, vol. 1. Toronto: Royal Ontario Museum, 1985. American Journal of Archaeology 93/4: 610–12. 1990 A. Kempinski, Syrien und Palästina (Kanaan) in der letzten Phase der Mittlebronze IIB Zeit (1650–1470 v. Chr.). Wiesbaden: Harrassowitz, 1983. Journal of the American Oriental Society 110/2: 345–47. 1991a H. E. Eilberg-Schwartz, The Savage in Judaism: Anthropology of Israelite Religion and Ancient Judaism. Bloomington, IN: Indiana University Press, 1990. Shofar 9/43: 135–37. 1991b A. Mazar, Archaeology in the Land of the Bible, 10,000–586 b.c.e. New York, NY: Doubleday, 1990. American Journal of Archaeology 95/3: 546. 1991c P. R. S. Moorey, A Century of Biblical Archaeology. Cambridge: Lutterworth, 1991. Bulletin of the AngloIsrael Archaeological Society 11: 29–30. 1991d W. H. Stiebing, Out of the Desert: Archaeology and the Exodus/Conquest Narratives. Buffalo, NY: Prometheus, 1989. Journal of Religion 71/4: 571–72. 1991–1992 H. Weippert, Palästina in vorhellenistischer Zeit. Munich: Beck, 1988. Journal of the American Oriental Society 111/3: 645–46. 1992a D. T. Ariel, Excavations in the City of David, 1978–1985 II: Imported Stamped Amphorae Handles, Coins, Worked Bone, and Ivory and Glass. Qedem 30. Jerusalem: Hebrew University, 1990. Journal of Biblical Literature 111/3: 509–10. 1992b W. J. Bennett and J. A. Blakely, Tell el-Hesi: The Persian Period (Stratum V). Winona Lake, IN: Eisenbrauns, 1989. Journal of the American Oriental Society 112/4: 684–85. 1992c D. L. Esse, Subsistence, Trade, and Social Change in Early Bronze Age Palestine. Chicago: University of Chicago Press, 1991. Journal of the American Oriental Society 112/3: 495–96. 1993a A. Kempinski and R. Reich, eds., The Architecture of Ancient Israel from Prehistory to the Persian Period. Jerusalem: Israel Exploration Society, 1992. Journal of the American Oriental Society 113/3: 495–96. 1993b G. Palumbo, The Early Bronze Age IV in the Southern Levant: Settlement Patterns, Economy, and Material Culture in a “Dark Age.” Rome: University of Rome, 1991. Bulletin of the American Schools of Oriental Research 290–291: 144–45. 1993c R. T. Schaub and W. E. Rast, Bâb edh-Dhrâº: Excavations in the Cemetery Directed by Paul W. Lapp (1965– 67). Winona Lake, IN: Eisenbrauns, 1989. Israel Exploration Journal 43/4: 281–83. 1993d N. A. Silberman, A Prophet from amongst You. The Life of Yigael Yadin: Soldier, Scholar, and Mythmaker of Modern Israel. New York, NY: Addison-Wesley, 1993. Archaeology 46/2: 59–61. 1994a P. M. M. Daviau, Houses and Their Furnishings in Bronze Age Palestine: Domestic Activity Areas and Artifact Distribution in the Middle and Late Bronze Ages, 1993. Journal of the American Oriental Society 114/4: 667–68. 1994b T. Dothan and M. Dothan, People of the Sea: The Search for the Philistines. New York, NY: Macmillan, 1992. Journal of the American Oriental Society 114/4: 668–69. 1994c E. Miron, Axes and Adzes from Canaan. Stuttgart: Steiner, 1992. Bulletin of the American Schools of Oriental Research 296: 81. 1995a A. Ben-Tor, ed., The Archaeology of Ancient Israel. New Haven: Yale University Press, 1992. Journal of Biblical Literature 114/1: 121–22. 1995b P. Bienkowski, ed., Early Edom and Moab: The Beginning of the Iron Age in Southern Jordan. Sheffield: Sheffield Academic Press, 1992. Bulletin of the American Schools of Oriental Research 297: 91–93. 1995c I. Finkelstein, Shiloh: The Archaeology of a Biblical Site. Tel Aviv: Tel Aviv University Press. Biblical Archaeology Review 21/6: 6–10. 1995d R. Gonen, Burial Patterns and Cultural Diversity in Late Bronze Age Canaan. Winona Lake, IN: Eisenbrauns, 1992. Israel Exploration Journal 45/4: 299–300. 1989c

xxiv 1995e 1995f 1996a

1996b 1996c

1996d 1996e 1996f 1998a

1998b

1998c 1998d 1999a

1999b 1999c 2001a 2001b 2002a

William G. Dever: Bibliography Z. Gal, Lower Galilee during the Iron Age. Winona Lake, IN: Eisenbrauns, 1992. Israel Exploration Journal 45/4: 202–5. G. D. Pratico, Nelson Glueck’s 1938–1940 Excavations at Tell el-Kheleifeh: A Reappraisal. Atlanta, GA: Scholars Press, 1993. Bulletin of the American Schools of Oriental Research 298: 95–96. D. Barag et al., Masada: The Yigael Yadin Excavations, 1963–1965. Final Reports, vol. 4: Lamps; Textiles; Basketry, Cordage and Related Artifacts; Wood Remains; Ballista Balls. Jerusalem: Israel Exploration Society, 1995. Journal of the American Oriental Society 116/3: 539–40. E. Bloch-Smith, Judahite Burial Practices and Beliefs about the Dead. Sheffield: JSOT Press, 1992. Israel Exploration Journal 46/1–2: 131–32. S. Bourke and J.-C. Descoeudres, eds., Trade, Contact, and the Movement of Peoples in the Eastern Mediterranean: Studies in Honour of J. Basil Hennessy. Sydney: University of Sydney, 1995. American Journal of Archaeology 100/1: 197. V. Fritz, An Introduction to Biblical Archaeology. Sheffield: JSOT Press, 1994. Journal of the American Oriental Society 116/3: 591. D. R. Harris, ed., The Archaeology of V. Gordon Childe: Contemporary Perspectives. Chicago: University of Chicago Press, 1994. Journal of the American Oriental Society 116/1: 133–34. T. N. D. Mettinger, No Graven Images? Israelite Aniconism in Its Ancient Near Eastern Context. Stockholm: Almquist & Wiksel, 1995. Bulletin of the American Schools of Oriental Research 302: 93–94. A. Biran, D. Ilan, and R. Greenberg, Dan I: A Chronicle of the Excavations, the Pottery Neolithic, the Early Bronze Age and the Middle Bronze Age Tombs. Jerusalem: Nelson Glueck School of Biblical Archaeology, 1996. Bulletin of the American Schools of Oriental Research 309: 82–84. I. Finkelstein, Living on the Fringe: The Archaeology and History of the Negev, Sinai and Neighboring Regions in the Bronze and Iron Ages. Sheffield: Sheffield Academic Press, 1995. Journal of the American Oriental Society 118/4: 567–68. B. O. Long, Planting and Reaping Albright: Politics, Ideology, and Interpreting the Bible. University Park, PA: Pennsylvania State University Press, 1997. Near Eastern Archaeology 61/2: 131. S. W. Manning, The Absolute Chronology of the Aegean Early Bronze Age: Archaeology, Radiocarbon and History. Sheffield: Sheffield Academic Press, 1995. Journal of the American Oriental Society 118/3: 440–42. I. Finkelstein, Z. Lederman, and S. Bunimovitz, Highlands of Many Cultures—The Southern Samaria Survey: The Sites. Tel Aviv: Tel Aviv University, 1997. Bulletin of the American Schools of Oriental Research 313: 87– 88. B. S. J. Isserlin, The Israelites. New York, NY: Thames & Hudson, 1998. Biblical Archaeology Review 25/3: 57. T. L. Thompson, The Mythic Past: Biblical Archaeology and the Myth of Israel. New York, NY: Basic Books, 1999. Biblical Archaeology Review 25/5: 64–66. R. Cohen and R. Westbrook, eds., Amarna Diplomacy: The Beginnings of International Relations. Baltimore, MD: Johns Hopkins University Press, 2000. Middle East Studies Association Bulletin 35/1: 61–62. I. Finkelstein and N. A. Silberman, The Bible Unearthed: Archaeology’s New Vision of Ancient Israel and the Origin of Its Sacred Texts. New York, NY: Free Press, 2000. Biblical Archaeology Review 27/2: 60–62. E. Fox, Sacred Geography: A Tale of Murder and Archaeology in the Holy Land. New York, NY: Prometheus, 2001. Christian Century 119/20: 45–46.

Part 1

Archaeology

The Sad Fate of Statues and the Mutilated Statues of Hazor Amnon Ben-Tor Institute of Archaeology, The Hebrew University of Jerusalem

In the image [of the emperor] there is the idea (eidos) and form (morphe) of the emperor. . . . The emperor’s likeness is unchanged in the image, so that who sees the image sees the emperor in it, and again who sees the emperor, recognizes him to be in the image. . . . The image might well say: “I and the emperor are one.” Who therefore adores the image adores in it also the emperor. For the image is the form of the latter and his idea. (Athanasius of Alexandria [Freedberg 1989: 392])

Introduction In the well-known Sumerian myth describing the goddess Inanna’s descent to the netherworld, we read: Let not your good metal be covered with dust of the netherworld, Let not your good lapis lazuli be broken up into the stone of the stoneworker, Let not your boxwood be cut up into the wood of the woodworker. (Kramer 1951: 3)

This passage probably refers to the fate of Inanna’s statue (Jacobsen 1976: 57; Buccellati 1982: 5), which apparently had been made of a combination of these three materials. An Egyptian text of New Kingdom dating describes the sun god Re thus: “His bones being silver, his flesh gold, his hair from true lapis lazuli” (Lichtheim 1976: 198), much like the above description of Inanna’s statue. Indeed, one such Egyptian statue of a falcon-headed deity, made of silver, gold, and lapis lazuli, has miraculously survived (Roehrig 1996: 4–7). A large number of statues were made of wood, and among these, mesu wood was the preferred type. The exact botanical definition of this type of wood, sometimes referred to as “flesh of the gods” or “bone of the gods” (Foster 1993: 779–80; Hurowitz 1995: 343), is unknown. The different materials from which statues were fashioned can also be gleaned from descriptions in the Old Testament: “The iron-smith fashions it, and works it over the coals. . . . The carpenter . . . fashions it Author’s note: The manuscript of this essay was completed in 1999.

with planes . . .” (Isa 44:12–13); “Alas for you who say to the wood ‘wake up!’ to silent stone ‘Rouse yourself!’ . . . See, it is gold and silver plated . . .” (Hab 2:19); and “Their idols are silver and gold, the work of men’s hands” (Ps 115:4). Statues of deities and of mortals, therefore, were apparently made of a variety of materials, such as clay; wood; and different types of stone, including semiprecious; various metals, such as iron and bronze; and precious metals, such as gold and silver. A number of different materials may have been used for the same statue, for example, a wood, stone, or metal statue clothed in gold or silver garments (Oppenheim 1949). Evidence of statues made of all of the above-mentioned materials and combinations thereof also comes from the archaeological record (see, among many others, Strommenger 1964: pls. 6, 10–12, 50–61, 91, XXII–XXIII, 152– 53, 162–63, 178, 184, 215, etc.; Spycket 1968: 7–27; 1981: 260–86, 326–48, 402–30; Negbi 1976; Weiss, ed. 1985: cat. nos. 133–36, 170). Apparently the most common material—wood—is rarely represented among the archaeological finds, even from the later Classical period, due to its fast rate of decay, even in the relatively dry climate of the Near East. Nevertheless, a relatively large number of wooden statues, some of them gilded, have been preserved in the extremely dry climate of Egypt (Reeves 1990: 129, 131, 133). It is impossible to estimate the number of statues that were produced in any part of the ancient Near East at any given time: there must have been hundreds, if not thousands. The vast majority of statues portrayed deities, kings, and dignitaries; for the purposes of this study, however, no

Amnon Ben-Tor

4 distinction is made—indeed, none is necessary—between these three types of statues. It can be stated with certainty that only a very small fraction of the original number of statues has survived and an even smaller number has been recovered; it goes without saying that, since not all sites have been excavated, there are certainly many more statues to be discovered. Furthermore, the total number of statues recovered at the numerous sites thus far excavated throughout the ancient Near East, many of them of major importance in ancient times, must represent only a small minority of those that were originally produced. There is no reason to assume that the situation would be different at the unexcavated sites. The question therefore arises: excluding the wooden statues that may not have survived due to decay and climatic conditions, where have all the statues gone? Written records and archaeological evidence provide an answer to what must have happened to many of the statues, documenting three main causes for their disappearance: exile, burial, and mutilation/demolition (Keel 1973; Brandes 1980). Evidence of all three phenomena has been found in the archaeological record.

The Hazor Statues The number of Bronze Age statues already discovered at Hazor, while the excavations are still in progress, 1 exceeds the number found at all other sites in Israel combined, demonstrating yet again the site’s great importance in the second millennium bce. The significance of these artifacts as works of art and as indicators of Hazor’s cultural milieu as well as chronological matters such as the dating of their manufacture are beyond the scope of this essay. In the present context, I shall discuss the state of preservation of each object and its archaeological context, factors that may eventually also have a bearing on the question of the identity of the people who brought about the destruction of Canaanite Hazor (Ben-Tor 1998). Nine artifacts of this sort were found at Hazor in the excavations directed by Yadin in the 1950s, and another nine in the renewed excavations conducted since 1990, 2 not counting the torso of a basalt statue and part of a basalt seated statue recently found on the surface of the site and two broken-off heads of basalt statues discovered in Iron 1. Among the numerous summary reports on the renewed excavations at Hazor, see Ben-Tor 1995; 1996a; 1996b; 1997; 1998; Ben-Tor and Ben-Ami 1998. 2. The objects found by Yadin’s expedition are illustrated in the final reports (Yadin et al. 1958; 1960; 1961; Ben-Tor, ed. 1989) and are marked in this essay with an asterisk. Because the objects found in the renewed excavations are still being studied, only the four photos that have already been published are included.

Age contexts in the 1991 and 1992 excavation seasons. These four artifacts are not included, because their original location and the circumstances under which they were broken are unknown. 1. Basalt lion orthostat (reg. no. A/45206; fig. 1). This almost complete orthostat, except for the tip of the nose and front paws, was found reused as part of the foundations of a large Iron Age public building. The orthostat originally adorned the left doorjamb of an important Late Bronze Age building on the acropolis at Hazor, which may have functioned as a palace or a temple. It is identical to the lion orthostat found by Yadin’s expedition buried at the entrance to the Area H temple (see below): the two constitute a perfect right/left pair. It cannot be ascertained, however, whether the more recently discovered lion orthostat indeed represents the mate of the Area H one or whether it is one of a pair that originally stood at the entrance to the Late Bronze Age structure on the acropolis. 2. *Basalt lioness orthostat (reg. no. A/5370). Only the head and part of the mane along the shoulder were preserved (Yadin et al. 1961: pls. 13:4, 319; Yadin 1972: 125; Beck 1989: 328–29; Bonfil 1997: 74, 88). The orthostat, probably one of two that originally flanked the entrance to the Area A palace, was found in a heap of stones. It clearly belongs to the Late Bronze Age, but the circumstances of its breakage and final disposal cannot be determined. Another part of this orthostat—the adjoining middle part of the lioness—was found in the 1997 season in secondary use as a doorjamb in an Israelite dwelling located approximately 50 m west of the findspot of the head. 3. *Basalt lion orthostat (no reg. no.). This complete small orthostat was found in the niche of the Area C Late Bronze Age temple, buried under the northernmost stela (Yadin et al. 1958: pls. 29:3, 30:2; Yadin 1972: 72). It therefore clearly was not in use in the last phase of occupation, Stratum IA, but Yadin’s supposition that it was not in use even in the preceding Stratum IB (1972: 72) is unwarranted. The orthostat, which probably flanked the southern doorjamb of the Stratum IB temple, was apparently buried as a relic from the previous building by the Stratum IA settlement inhabitants. 4. *Basalt lion orthostat (reg. no. H/666). This almost complete orthostat was found buried in a pit that cut into the southern doorjamb of the Area H temple and was covered with massive stones (Yadin et al. 1961: pls. 118– 20, 328; Yadin 1972: 91; Beck 1989: 327–28). The lion originally stood at the right-hand doorjamb of the temple. Because the pit cuts into the doorjambs of both the Strata IA and IB temples, it could have been buried either “by the people of IA, who, in reconstructing the temple [of Stratum IB] had no more use for it, but buried it out of veneration. The second alternative is that this was done by

The Sad Fate of Statues and the Mutilated Statues of Hazor

5

Fig. 1. Basalt lion orthostat from Hazor (reg. no. A/45206) (Selz Foundation Hazor Excavations in Memory of Yigael Yadin).

fugitives who returned to the spot after the temple’s destruction. The possibility that it was the work of the conquerors seems less likely” (Yadin 1972: 91). As Yadin went on to note (1972: 91), the second possibility is “strangely similar” to the one suggested by Woolley for the burial of the Idrimi statue (1955: 89). Yadin’s first alternative has far-reaching implications, becausee it may indicate that during the final phase of Canaanite Hazor the temple consisted of only two units, the outer, southern one of which went out of use. This would contradict the plan of the temple reconstructed by the excavators (Ben-Tor, ed. 1989: plan 40), but the accompanying text suggests such a possibility: “It is difficult to estimate exactly the extent of the porch’s reconstruction carried out in stratum IA, but the traces of building which can be attributed to this stratum allow us to imagine the situation [of the porch] in stratum IA” (Ben-Tor, ed. 1989: 262–63). A reduction of this sort in the size of the temple would accord well with the general signs of decline that are evident in the last phase of Canaanite Hazor (Yadin 1972: 108; BenTor, ed. 1989: 170, 264, 297). It is also possible, however, that the porch did not go out of use in Stratum IA, but for some reason—perhaps because one of the two lions was damaged in the course of the destruction of Stratum IB— the Stratum IA inhabitants chose not to adorn the entrance to the porch with guardian lion orthostats.

5. *Egyptian statue (reg. no. A/6201/1). This statue, which probably originally stood in the nearby Late Bronze Age palace, was found in an Iron Age context “in the Solomonic fill” (Yadin et al. 1961: pl. 323:4–5; Yadin 1972: 126, n. 1). Although only the torso of the statue was preserved, there are clear indications that it was deliberately mutilated in antiquity: the arms, which in such statues are close to the body (in fact, joined to the body) were chiseled off. Similarly mutilated Egyptian statues found in Late Bronze Age contexts during the renewed excavations (see below), indicate that the mutilation took place sometime in the Late Bronze Age. Yadin dated this statue to the New Kingdom period, based on its style (1972: 126, n. 1). However, a reexamination of the statue, together with the three recently discovered examples, indicates that all four are of Middle Kingdom dating and thus centuries earlier than the palace in which they must have stood. The fragment of a foot found nearby is all that remains of yet another Egyptian statue (Yadin et al. 1961: pl. 323:6). 6. Egyptian statue (reg. no. A/14760; fig. 2). As in no. 5 above, only the torso of this statue was preserved, and the intentional mutilation (the chiseling away of the arms) is clearly evident. The statue was found out of its original context in an Iron Age locus just above the courtyard of the Late Bronze Age palace, where it may have originally stood.

6

Fig. 2. Torso of Egyptian statue from Hazor (reg. no. A/14760) (Selz Foundation Hazor Excavations in Memory of Yigael Yadin).

7. Egyptian statue (reg. no. A/17844). The third Egyptian statue is very similar to the above two and also bears the same indications of intentional mutilation. It was found in the so-called “throne room” in the final destruction layer of the Late Bronze Age palace. 8. Egyptian royal statue (reg. no. A/15364). Unlike the three above-mentioned examples, which probably represent either kings or dignitaries, this is certainly a royal statue, of which only the head was found; it was also found in the “throne room” in the final destruction layer of the palace. The head had been cut off and only traces of the back-pillar are preserved. 9. Egyptian sphinx (reg. no. A/14774; fig. 3a–b). A small sphinx was found reused as a stone in an Iron Age wall built directly above the forecourt of the Late Bronze Age palace, indicating that it most likely originated in the underlying palace courtyard. Like the above-mentioned

Amnon Ben-Tor Egyptian statues, the sphinx dates to the Middle Kingdom period and must have been an ancient relic that had been placed in the Late Bronze Age palace. Signs of deliberate mutilation are evident: the cartouche of Amenemhat III incised on the front of the sphinx had been slightly damaged (fig. 3b) and its head and forepaws cut off (fig. 3a). The questions how the statues reached Hazor, and when and why they were placed in the Late Bronze Age palace are beyond the scope of this study and will be addressed elsewhere. 10. Basalt statue of a standing figure (reg. no. A/15227). This statue was found immediately to the right of the entrance to the palace “throne room” (Ben-Tor 1995: 285, fig. 2), where it must have stood originally. A huge basalt bowl adjoining the front of the statue is an integral part of it. The statue was found broken into more than 100 pieces in the destruction debris caused by the fire that brought an end to Canaanite Hazor. A conical depression drilled into the middle of the span of the shoulders indicates that the now-lost head of the statue, which may have been made of a different material, was joined to the body by a protrusion in its base that fit into the depression. The left hand of the statue is missing, while most of the right hand, which originally held a circular object (probably a cup; see below), was restorable. It is not clear whether the statue was deliberately broken or whether it broke into pieces as a result of the intense heat of the fire that destroyed the palace. The fact that the head is missing and that both hands were also damaged (parts that are commonly missing in deliberately mutilated statues) seems to indicate that both intentional and unintentional factors contributed to the destruction of the statue. 11. Head of an ivory statue (reg. no. A/41219; fig. 4). The object was found covered by the destruction debris in one of the rooms of the Late Bronze Age palace. Even though the earth and ash that had accumulated in this room were carefully sifted, as throughout the palace, and a large number of beads and other small finds were recovered, no trace of the rest of the statue was found. It therefore cannot be determined with certainty whether the head was intentionally severed from the body by those who destroyed the palace, in which case the body would have been dumped somewhere in the vicinity. The circumstances of the head’s discovery seem to indicate that this is indeed what happened, but the possibility that the breakage was accidental cannot be ruled out, and if this was the case, there is no way to ascertain when it occurred. 12. Bronze statue of a seated figure (reg. no. A/44949). An almost intact bronze statue was found buried under what may have been a wooden floor in the corner of one of the rooms located to the north of the “throne room” of the Late Bronze Age palace. Judging by the cap and dress

The Sad Fate of Statues and the Mutilated Statues of Hazor

7

3a Fig. 3a–b. Egyptian sphinx from Hazor (reg. no. A/14774) (Selz Foundation Hazor Excavations in Memory of Yigael Yadin).

remnants, the statue probably portrays a king or dignitary. Before it was buried, however, the golden garments, of which traces still remain on the statue, were stripped— perhaps by a person who hid the statue prior to the palace’s destruction, thus preventing its mutilation. 13. Bronze statue of a seated figure (reg. no. A/44740). As in no. 12 above, this bronze statue was found buried, in another corner of the same room (Ben-Tor 1996a: 264, fig. 4). While no traces of the golden garments that the statue probably wore have survived, the incised lines cut into the sides of the body in order to hold garments of this sort are clearly visible. The statue’s left hand is missing, for reasons that cannot be determined: because it was buried prior to the palace’s destruction, the damage could not have been caused by those who destroyed Hazor. The surviving garments and headgear of the statue indicate that in all probability it represents a deity. 14. *Statue of a seated figure (no reg. no.). This basalt statue was discovered in the niche of the Area C temple of Stratum IA (Yadin et al. 1958: pls. 31:1, 181:13; Yadin 1972: 71–73). Yadin identified the figure as representing the moon-god Sin (1972: 73), an identification Beck considers “most unlikely” (1989: 324). The statue was placed at the extreme southern end of the niche, to the left of the viewer; its decapitated head was found lying nearby. The mutilation was clearly intentional. 15. Statue of a seated figure* (reg. no. F/68/14). According to Yadin et al. (1960: 157, pl. 197), the statue was

3b

found in Area F in a “surface layer [?], and was apparently in use in stratum IA, although it seems to have originated in stratum IB.” No reason is given for this statement. The head of the figure is missing, and although the circumstances under which this occurred cannot be established with certainty, at least partial intentional mutilation seems to be indicated by the fact that the outline of the base of a cup that the figure originally held in its right hand can still be seen.

8

Fig. 4. Head of ivory statuette from Hazor (reg. no. A/41219) (Selz Foundation Hazor Excavations in Memory of Yigael Yadin).

16. *Statue of a deity standing on a bull (reg. nos. H/760, H/526). This severely damaged statue was found in Area H in two parts: the torso near the obelisk in front of the temple entrance and the partially preserved bull in the heap of stones that covered the buried lion orthostat (no. 4) discussed above (Yadin et al. 1961: pls. 117:6–7, 324–25; Yadin 1972: 95). Yadin identified the figure as representing the temple’s deity, Hadad. It is not possible to prove that the statue was deliberately damaged, even though it seems highly likely, nor are we able to determine exactly when it was discarded. The two loci in which the different parts of the statue were found were assigned to Stratum IB (Yadin et al. 1961: pls. 324–25), but the facts that one of these loci (L. 2119) is assigned elsewhere to both Strata IB and IA (Ben-Tor, ed. 1989: plans 39–40) and that one part of the statue was found in the heap of stones covering the lion orthostat (no. 4) indicate that it may have been (broken? and) discarded in Stratum IA. 17. *Statue of a seated figure (reg. no. H/127; fig. 5). The statue was found in the Stratum IA destruction layer in the Area H temple, to the west of the entrance to the Holy of Holies (Yadin et al. 1961: pls. 123:2, 326–27; Ya-

Amnon Ben-Tor din 1972: 94; Beck 1989: 324–27). The statue, which apparently represents a lay figure, perhaps a king, was deliberately decapitated. Its head was found lying nearby, as was the head of the statue from Area C described above (no. 14). The mutilation included damaging the hands, especially the right one, in which a cup was held, only the base of which survived. The right hands of the above-mentioned statues from Areas C and F (nos. 14– 15) show similar damage, with only the bases of the cups still remaining. 18. *Statue of a seated figure (H/1027). The statue was found in a pit more than 3 m deep, cut into the cella of the Stratum IB temple in Area H (Yadin et al. 1961: pl. 330; Yadin 1972: 84; Beck 1989: 322–24). Based on stylistic grounds, the statue must already have been a relic in the Stratum IB temple, originating centuries before it was placed in the pit. The statue’s head is missing and both hands (the right one originally held a cup, of which only the base still remains) are severely damaged. The missing head and damaged hands, features that were typically the targets of intentional mutilation, also indicate a deliberate, hostile act in the case of this statue: “Intentional decapitation [and cutting off the hands] would corroborate the figure’s special status, already deduced from the context in which it was found” (Beck 1989: 327). Nevertheless, the possibility that the damage to the statue may have been the result of its very long period of use cannot entirely be ruled out.

Where Have All the Statues Gone? Having surveyed the fate of the Hazor statues, let us now view it in the broader context of the neighboring ancient Near Eastern cultures. As noted above, the three main causes for the “disappearance” of statues were exile, burial, and mutilation/demolition. All three are widely observed in the archaeological record. Exile of Statues The pillaging of statues of deities and the subsequent restitution are already attested in a text dating to the third millennium bce, in which Ensakusanna, king of Uruk, states, “the king of Kish, whom in his city I have destroyed, has pillaged the belongings of the gods; I have returned them; his statues, his silver and lapis-lazuli, his wood and all other goods, to Enlil in Nippur I have brought” (Sollberger 1971: 91, IH1b). A text from Nineveh dating to the second millennium describes the events that led to the exile of the statue of the god Marduk from Babylon, first to Hatti, then to Assyria, and finally to Elam (Borger 1971). Of the many inscriptions relating to the exile of statues in the first millennium bce, the accounts of

The Sad Fate of Statues and the Mutilated Statues of Hazor

9

Fig. 5. Decapitated basalt statue found in the final destruction layer of Late Bronze Age Hazor (reg. no. H/127).

Tiglath-pileser III describing the exile of the statues of the gods of the city of Gaza and of Sennacherib describing the deportation of statues of the gods of Ashkelon are particularly noteworthy. These events are known not only from written texts but also from artistic representations on the reliefs decorating the walls of the palaces of these two monarchs (Uehlinger 1997: 124–27, figs. 45–46). Among the exiled statues—mainly of kings—and other major works of art found in archaeological contexts far from their places of origin, the following are noteworthy: a statue of Manistusu, king of Akkad; of a prince from Eshnunna; and of a prince from Babylon, all taken to Susa by the Elamite king Shutruk-Nahunte in the 12th century bce. He usurped several of these statues by carving his own name on them (Strommenger 1964: figs. 148–49; Moortgat 1969: 49), as the Assyrian king Esarhaddon did several centuries later on the statues of the gods of the Aramean king Hazael that he deported to Assyria (Borger 1956: 53). Among other works of art exiled, most noteworthy are the stele of the Code of Hammurabi and the victory stele of Naram-Sin, both exiled to Susa from Babylon by ShutrukNahunte, who incised his own name on the latter stele and dedicated it to his own god, Inshushinak (Strommenger 1964: 405, 420, figs. 122–23, 158–59).

Burial of Statues The statues placed in deeply cut pits found at ºAin Ghazal in Jordan constitute the earliest examples of the burial of statues (Rollefson 1983; 1986; see also Garfinkel 1994: 163–64). The phenomenon of burial of statues and other cult objects has been observed throughout the ancient Near East from the Neolithic period onward and in fact continues to the present day—for example, in the form of the genizah for damaged Torah books. Whether the motive for this practice was “positive” (intended to preserve, protect, or dispose of a damaged cult object in a dignified manner, as often attested in temple favissae) or “negative” (in “an attempt to nullify or de-sanctify its role,” Peltenburg 1989: 166) cannot in most cases be determined with certainty. 3 Let us review a few of the many known cases of statue burial that may have a bearing on the issue of intent. Textual evidence of the burial of gods is to be found quite often in myths of dying fertility gods and their repeated seasonal resurrection, among which the myth describing the death and resurrection of Dumuzi is the most 3. For a discussion of these and other motives for burial and for additional bibliography, see Garfinkel 1994: 178–80.

10 prominent (Jacobsen 1976: 25–73); however, there is no text indicating that a statue of Dumuzi was ever actually buried. Similarly, the description of the death of Baal, a central theme in Ugaritic myth, never refers to a ritual in which a statue is actually buried. The closest we get is a text in which the goddess “bewails him, and buries him”— the god, not the statue—and “sets him in a divine pit in the earth” (Smith 1997: 152). That the actual burial of statues of such gods may in fact have been performed as part of cultic rituals seems to be indicated by the Punic inscription from Pyrgi referring to what appears to have been an annual rite, probably associated with Tammuz/Adonis: “in the month of krr, on the day of the burial of the god.” That the god’s statue is probably mentioned a few words later is yet another indication that a statue was indeed buried (Soggin 1975). Finally, the burial of the statues of “foreign gods” described in Gen 35:1–4 has been widely treated in scholarly publications (see, e.g., Keel 1973, with extensive bibliography). Keel suggests that even in this case, the motivation for the interment of the statues was positive, because the verb used is tmn and not qbr, the former always indicating an action taken with the intention of later retrieving that which was interred (1973: 307–8). One of the most striking caches of buried statues is the group recovered from the temple of Abu at Tell Asmar (Frankfort 1939: 2–3, 13–16, pl. 1; Delougaz and Lloyd 1942: 188–92, figs. 149–51). The majority of the statues represent mortals; the question whether statues of deities are also included in the group—as argued by some scholars (e.g., Frankfort 1963: 24; Jacobsen 1976: frontispiece; 1989) and denied by others (e.g., Spycket 1968: 11–13; Hallo 1993: 19–22)—is not pertinent to the present study. The important point is that the statues were buried next to the altar, and “the heavier statues, including the two cult figures, lay at the bottom; the remainder were packed one upon another, apparently with considerable care” (Delougaz and Lloyd 1942: 190). From the second millennium bce, the Egyptian Execration Texts, in particular the texts published by Posener (1940), clearly represent the mass burial of objects intended for use in some kind of cultic ritual, most likely in the form of “sympathetic magic.” The roughly-made clay statuettes on which these texts appear, which represented kings considered to be enemies of Egypt, were ritually broken and then buried in the Saqqara necropolis (for further discussion, see Nielsen 1955: 104–5; Keel 1973: 308–9). Three additional instances of statue burials warrant special attention: the burial of a statue of a king, of a deity, and of a group of statues of undetermined identity. 1. The nearly complete statue of Idrimi, king of Alalakh, was found buried in a pit in the Stratum I temple at Alalakh (Woolley 1949). The statue itself dates to the

Amnon Ben-Tor earlier part of the 14th century bce, that is, to the period of Stratum IV at Alalakh, but must have survived the violent end of that stratum and been put on display in the temples(?) of Strata III, II, and I, the last of which was violently destroyed sometime in the early 12th century bce. The statue had been knocked off its elaborate base and broken into several pieces (see further below on the breaking of statues), when “some survivor of the disaster took the trouble, and perhaps the risk, of hunting through the ruins for the fragments of the figure, and hiding them in a hastily dug pit, under great stones so that they might be recovered if and when the remnants of the inhabitants settled again in the town” (Woolley 1949: 7). There is of course no way to verify this speculation; the last part seems highly unlikely, since, as Woolley himself points out, the pit was filled with earth and large stones, some of them weighing more than one-and-a-half tons (1949: 2). 2. The complete bronze statue of a horned god unearthed at Enkomi was also buried later than the period in which it was first venerated: the statue was originally manufactured and worshiped in Period III of the ashlar building in which it was found (Dikaios 1962: 38) and had been placed in a pit “as if to be hidden” (Dikaios 1962: 24) one hundred years later, during Period I. It is difficult to accept the excavator’s suggestion that it was still worshiped during this period, when it was located in the pit (Dikaios 1962: 26). 3. A group of three statues—one complete, one missing the head and feet, and the third of which only the torso could be restored from the many broken pieces—was found buried in a pit in front of the entrance to Temple P2 at Ebla. The complete statue probably portrays a high official, while the other two probably represent royal figures (Matthiae 1990: 51–56). The statues were apparently broken when the city was sacked, around 1600 bce, and buried later, probably as an act of respect. From the first millennium bce, the following examples, mostly from Neo-Hittite contexts, are especially noteworthy (for a general discussion, see Ussishkin 1970). A carefully buried statue—the identity of which, be it god or king, is disputed (von Luschan 1911: 363; Frankfort 1963: 180, pl. 163)—was uncovered next to King Kilamuwa’s palace at Zinjirli. The face, hands, and scepter of the statue had been damaged when it fell after being knocked off its base (von Luschan 1911: 288–89, 362–69, figs. 194, 261–68, pl. 64). The burial of a royal statue beneath the Lion’s Gate at Arslantepe-Malatya in ancient Milid (Delaporte 1940: 35– 38, pls. 14–15, 26, 28) is quite similar. The statue was identified as that of Muttallu, king of Kummuh, who was awarded the city of Milid by Sargon II and ruled it from 712 to 708 bce, until he was dethroned by the Assyrian

The Sad Fate of Statues and the Mutilated Statues of Hazor king following an alleged conspiracy (Landsberger 1948: 77; Güterbock 1961). The statue, which originally stood in the city gate, was knocked off its base and damaged by the fall, like the above-mentioned statue at Zinjirli. Also like the latter, this statue was subsequently buried, and a carefully built stone tomb was constructed around it. At Zinjirli, lion statues that had protected the city gates were also buried. Following the destruction of the city, five lion statues of this sort were gathered from the different buildings in which they had been incorporated into the gates and were ritually buried in a specially dug pit (Koldewey 1898: 128–30; von Luschan 1902: 230–36; Landsberger 1948: 79–83). Finally, a good example of the burial of statues of deities comes from Egypt, from a temple dedicated to Montu at Armant (Mond and Myers 1940: 16, pl. 11, figs. 1:L, 2). Several statues and statue fragments were found carefully buried in the courtyard of this temple. The excavators’ suggestion that the statues were broken by a conquering enemy (Assyrian or Persian) and buried during the restoration work that was undertaken in the Ptolemaic period (Mond and Myers 1940: 16) seems reasonable. Mutilation/Demolition of Statues The main reason for the “disappearance” of statues from the archaeological record is that they were deliberately destroyed. Statues were burned, broken, smelted, stripped of their components of value, or a combination of these methods of destruction, depending on the material/s from which they were made. We are not concerned here with statues that were broken accidentally or as a result of the ravages of time or extensive use: these were, as a rule, buried in temple favissae (which may also explain the burial of some of the objects discussed in the previous section). In the present context, we shall discuss only objects that were demonstrably deliberately damaged, which in many instances is not easily established. As in the case of the exile of statues, textual evidence for the intentional destruction of statues goes back to the third millennium bce. An inscription of Urukagina, king of Lagash, describes the atrocities inflicted on the sanctuaries of Lagash by Lugalzageszi, “the man from Umma”: “The temple of the goddess Gatumdug he has burnt, its silver and lapis lazuli he has pillaged, its statues he has destroyed. . . . The temple of the goddess Amageshtinna he has pillaged, from the (statue of) the goddess Amageshtinna he has taken its silver and lapis lazuli, and he has thrown (the statue) into the pits” (Sollberger 1956: IV:4– 10, VI:11–VII:6). This must have been the fate of numerous statues in subsequent periods throughout the ancient Near East, as is evident in the archaeological record (see below). Two later texts that describe contemporary phe-

11

nomena but undoubtedly also reflect earlier realities should be mentioned in this context. The first is the biblical verse, “Fallen, fallen is Babylon, and all the images of her gods lie shattered on the ground” (Isa 21:9), and the second comes from the Mishna, discussing the issue of the desecration of “idols”: “How does one nullify it? [If] he has cut off the tip of its ear, the tip of its nose, the tip of its finger, if he battered it, even though he did not break off [any part of] it—he has nullified it. [If] he spit in its face, urinated in front of it, scraped it, threw excrement at it, lo, this does not constitute an act of nullification” (m. ºAbod. Zar. 4:5 [Neusner 1988: 668]). The idea expressed in this clearly religiously motivated discussion is that in order to desecrate a statue, one has to cause irreversible damage to it, as indeed is apparent in so many statues in the archaeological record. The capture of the Urartian city of Musasir and the pillaging of the rich assemblage from its temple dedicated to Khaldi are described in minute detail in an inscription of King Sargon II (Thureau-Dangin 1912: lines 362–405). The weight in minas and talents of the bronze artifacts that were seized, including the weight of the various statues, is repeatedly mentioned, clearly demonstrating Sargon’s interest, not in the artifacts themselves, but in their value as raw metal. The pillaging of Musasir is vividly portrayed on one of the bas reliefs in Sargon II’s palace at Khorsabad (= Dur-Sharrukin), which includes a scene showing the destruction of a statue, either of the king or of the city’s patron deity Khaldi (or his spouse Bagbartu): the statue is being hacked to pieces by two Assyrian soldiers (Keel 1972: §317; Brandes 1980: 30–31). Despite the difficulty in ascertaining whether a damaged statue indeed reflects deliberate mutilation, the following are a few examples from the archaeological record that clearly demonstrate such abuse. One of the best-known cases is the mutilated head of “Sargon” (Nylander 1980). The bronze head clearly belongs to a third-millennium monarch, although his identity is still disputed. The answers to such questions as who mutilated the head and when can only be speculation, but it is certainly clear that the damage to the left eye, the bridge and point of the nose, both ears, and the tip of the beard was intentional: the marks left by the blows of the chisel are evident. In the three cases of the burial of statues mentioned above, all the statues had been intentionally broken before they were buried. At Ebla, of the three statues that had been placed in a pit, two (probably portraying a king and a queen) were found broken, and the third, identified by the excavator as representing a dignitary, was complete (Matthiae 1990: 51–54). The fragments of the broken statues had been

Amnon Ben-Tor

12 carefully placed in the pit. Whether the burial took place immediately after the sack of the city to which the statues belonged, ca. 1600 bce, or a long time later, in the Persian period (Matthiae 1990: 51), is not entirely clear: the Persian period sherds found in the pit could have been accidentally deposited when the stones of Temple P2 were robbed at the time. There is no doubt, however, that “the breakings at the base of the necks of the two statues, as well as the base of the cloak of the female figure—date to the end of the Middle Bronze II” (Matthiae 1990: 53). The excavator concludes that “the complete state of the male sitting statue of a dignitary or priest is certainly due to the fact that it was saved because it was not royal, as the mutilation of royal images was, in the sacks of towns, a rule perhaps without exceptions” (Matthiae 1990: 53, n. 121). The statue of Idrimi in Temple I at Alalakh had been knocked off its base, which was left standing (Woolley 1955: pls. 12a, 47b). The statue had been buried in a pit: “The head had been broken off and was set beside the body, together with two smaller fragments, one of the beard, the other of a foot. Only part of one foot was missing” (Woolley 1949: 2). At Armant, several massive statues, among them one of Osiris, were buried in the courtyard of the Temple of Montu. The statues, which were probably intentionally broken following the Assyrian or Persian conquest, were buried during the restoration of the temple in the Ptolemaic period: “The statues were deliberately buried. In the two examples where heads and torsos were found together, the heads did not belong to the torsos with which they were buried” (Mond and Myers 1940: 16).

The Question of Motivation There can be no doubt that the motivation for the exile of statues of both deities and kings was negative, namely, for the purposes of desecration and humiliation. The exile of these statues, which sometimes were placed in the temples of the gods of the victors, was an act that clearly demonstrated the supremacy of the victor over the vanquished. This concept is nicely illustrated in 1 Sam 5:1–4, the account of the events that transpired after the victory of the Philistines over the Israelites at the battle of Eben-ezer: the ark of God of the Israelites, the equivalent of their divine statue, was taken by the victorious Philistines to the city of Ashdod, where it was placed in the Temple of Dagon. The motivation for the burial of statues is more complicated. As noted above, several of the buried statues had been deliberately broken prior to their interment, while others had not. The breaking of statues can certainly be attributed, in most cases, to a conquering enemy; but who buried them and when? The burial of the broken statues

can as a rule be attributed to the survivors of the disaster but not necessarily so: for example, who buried the statue of King Muttallu in Malatya after it fell to the Assyrians? As mentioned above, Muttallu was originally the king of Kummuh, who was imposed on the local population of Milid by a foreign power. He was later deposed by force, after having conspired against that same power. King Muttallu was therefore not a “popular” figure with the local population, nor with the Assyrians, yet his statue was respectfully buried. It follows that in some cases even statues of fallen foes may have been treated with respect. A similar observation was made by Koldewey with regard to the burial of the gate lions at Zinjirli: “Our gate lions were thus treated by the conqueror with special respect . . . resulting from superstitious fear” (1898: 130). The burial of undamaged statues may in most cases have been motivated by the desire to spare them from the fate of desecration and humiliation. This act may therefore usually have been performed by the local population prior to the sack of a city or even afterwards, in the case of statues that somehow escaped destruction. It thus appears that, as a rule, most burials of statues, whether complete or damaged, were positively motivated, to save them from further harm. The Egyptian Execration Texts are a clear exception: the burial of the clay figures of rulers was negatively motivated, in a “voodoo-like” act of sympathetic magic, with the intention of breaking their power. The motivation for the third phenomenon—the deliberate mutilation of statues—is of course exclusively negative. While the exile and burial of statues virtually do not occur at all in modern times, examples of breaking the statues and burning the pictures of rulers that have fallen from grace are common even today: the destruction of statues such as those of Stalin, Lenin, Mobutu Sese-Seko, and others is well documented in the media. As a rule, this is today, as it was in ancient times, exclusively a politically motivated iconoclasm. Naturally, this applies to the statues of kings in the ancient Near East, but it apparently also holds true for the destruction of statues of deities, which seems to have occurred less frequently. The statue of Hadad, for example, may have been destroyed, not because it represented Hadad, but because it was a statue of Hadad of the particular city that was conquered. A similar statue of Hadad could still be worshiped in the city of the victor. The intentional destruction of statues was also widely practiced in Egypt, where, as elsewhere, it was almost entirely politically motivated. At Deir el-Bahari alone, more than 200 statues and sphinxes of Hatshepsut “were savagely and methodically obliterated and smashed to pieces” by Thutmose III (Hayes 1959: 2.89). This act was motivated by personal animosity combined with a political in-

The Sad Fate of Statues and the Mutilated Statues of Hazor centive (Nims 1966; Robins 1993: 52). The usurpation of statues of kings and officials by erasing the names of the original owners and substituting new ones is a very common phenomenon in Egypt (Hayes 1959: 2.235; Sourouzian 1988: 229–54). The mutilation or even complete destruction of statues is also well attested: one example is the mutilation of a statue of Thutmose III by cutting off its head and both hands (Scott 1946: fig. 17 [Metropolitan Museum of Art reg. no. 23.2.34). Another statue of the same king with the head and hands similarly cut off was found at Luxor (El-Saghir 1992: 69–70, figs. 149–52). Interestingly, several of the statues found at Hazor had been mutilated in exactly the same manner (see further below). Statues of deities were also subject to mutilation in Egypt: wood or wax effigies of the god Seth, inscribed with the phrase “Seth the miserable,” were spat upon, beaten, and finally thrown into the fire (Nielsen 1955: 105–6; Brunner-Traut 1982: 338–41). The wording of the inscription indicates that Seth—who was generally identified with chaos and all negative forces—was identified in this particular ritual as representing Egypt’s Asiatic enemies. The destruction of Seth’s effigies was thus politically motivated, very similar to the motivation behind the burial of the statues bearing Execration Texts. Of particular interest for the present study is the destruction of statues of deities and the obliteration of their names from monuments throughout Egypt during the Amarna period, when, as part of the religious reforms introduced by Akhenaten, the older religion was officially banned. The names of Amun and his consort Mut were erased from their statues and monuments throughout Egypt (Allen 1996: 4), and in some cases, this obliteration went so far as the complete destruction of their images (Brunner-Traut 1982: 339). The practice of name erasure went to such extremes as obliterating the “imn ” component of the name of Amenemhat III from the very top of his pyramid at Dahshur (Arnold 1987: 14, pls. 38–39). Another example of the fate of representations of deities “banned” during the Amarna period is the cutting out of the figure of Hathor from a painted limestone relief discovered at Deir el-Bahari. In the Ramesside period, a new figure of Hathor, made of plaster, was restored on the damaged relief (Hayes 1959: 1.157, fig. 94; Arnold 1974: 11, pl. 26). 4 Such acts in the Amarna period were clearly religiously motivated and, as such, are an isolated phenomenon in the Bronze Age. They are exceptional in terms of this early 4. I wish to thank Dorothea Arnold and James Allen of the Department of Egyptian Art at the Metropolitan Museum of Art, New York, and my wife, Daphna, for their help and important comments on the subject of the treatment of statues and monuments in ancient Egypt.

13

date, since they reflect a monotheistic concept of only one God, to the exclusion of all others. As a rule, religiously motivated iconoclasm makes its appearance only with the establishment of nation-states in the first millennium bce and the introduction of monotheism, whose adherents believe that there is only one “true” God. One recently excavated assemblage may provide a good example of religiously motivated iconoclasm but at the same time may also demonstrate the difficulty in trying to assess the motivation for the destruction and/or burial of religious artifacts, including statues: at ºEn Hazevah in southern Israel, 67 clay objects and 7 stone altars were found buried in a pit dug next to a building identified by the excavators as an Edomite shrine (Cohen and Yisrael 1995: 26). Among these were 3 anthropomorphic cult stands that, like all other clay artifacts in the assemblage, had been smashed to pieces (Cohen and Yisrael 1995: 24). The excavators conclude that “the smashing of the Edomite cult vessels may have been part of Josiah’s wide ranging religious reforms” (Cohen and Yisrael 1995: 27), which may indeed have been the case. On the other hand, the pit could be interpreted as the shrine’s favissa, in which the religious artifacts were piously buried after they were damaged or, indeed, to protect them from just such impending religious reforms. Religious iconoclasm still exists today, side-by-side with politically motivated iconoclasm, as demonstrated, among other things, by the burning of churches, mosques, or synagogues and the deliberate destruction of objects of religious significance.

Conclusions The following points sum up the information gleaned from the 17 objects from Hazor discussed above. We are unaware of any statues that may have been exiled from Hazor. There is, however, abundant evidence that statues were buried and mutilated. No further information regarding the lion and lioness orthostats from Area A (nos. 1–2) is available. Five of the objects were buried. The small lion orthostat (no. 3) from Area C that most likely stood in the Stratum IB temple was buried in the Stratum IA temple in the course of restoration of this temple following its earlier destruction. The Area H lion orthostat (no. 4) was buried either by the inhabitants of the Stratum IA settlement under similar circumstances to the burial of the Area C lion orthostat or, in the event that the former was in use in the Stratum IA temple, by the survivors after the destruction of Stratum IA. It should be noted that, except for the Area A lioness (no. 2), which might have been broken as a result of large-scale building activities in the Iron Age

Amnon Ben-Tor

14 (Yadin 1972: 125), none of the lion orthostats were damaged. Regarding the statues, the two bronze statues found in the Area A palace (nos. 12–13) were probably buried prior to the final destruction of the Stratum IA city in order to save them from mutilation by the attacking enemy. The statue found in the cella of the Area H temple (no. 18) was buried in a deep pit by the Stratum IB inhabitants. This is the only clear instance of the burial of a statue at Hazor attributable to Stratum IB, which may indicate an intention of protecting it from those responsible for the destruction of Stratum IB. Eleven objects seem to have been mutilated. Two (nos. 5–6) of the three Egyptian statues and the sphinx (no. 9) that were found in Iron Age contexts bear signs of deliberate damage, but it cannot be determined when this was caused. The same applies to the seated figure from Area F (no. 15) and the statue of the deity standing on a bull from Area H (no. 16), although the latter does seem to belong to the group deliberately mutilated by those who destroyed Stratum IA. The six remaining statues—the torso of an Egyptian statue (no. 7) and the head of an Egyptian royal statue (no. 8), both found in the destruction debris in the “throne room”; the large standing statue (no. 10) also found in the “throne room”; the head of the ivory statue (no. 11) found in the destruction debris of one of the side rooms of the palace; the seated figure (no. 14) from the Area C Stratum IA temple; and the seated figure (no. 17) from the Area H Stratum IA temple—were all most likely

mutilated by those who brought about the final destruction of Late Bronze Age Hazor. Of the four Egyptian statues, one is clearly a royal statue, while the others may have represented either kings or dignitaries. Of the locally made anthropomorphic statues, four are of kings or dignitaries: one (no. 12) of the two bronze statues from Area A, the seated statue (no. 15) from Area F, and the two seated basalt statues (nos. 17– 18) from Area H. Four statues apparently represent deities: one (no. 13) of the two bronze statues and the large standing basalt statue (no. 10) from Area A, the seated basalt statue (no. 14) from the Area C temple, and the figure standing on a bull (no. 16) from the Area H temple (see the citations above for the view maintaining that the Area C and H statues are not divine). The motive for the burial of all statues seems to have been veneration, while the motive for mutilation was exclusively political. In all cases of mutilation, the heads and hands of the statues were the primary targets. This practice brings to mind the description of the fate of the statue of Dagon in his Temple at Ashdod in 1 Sam 5:4: “Dagon’s head and both his hands were broken off upon the threshold; only his trunk was left intact.” This account of the struggle between the god of Israel and the god of the Philistines was composed much later than the period of the Hazor statues and was religiously motivated. It does, however, echo practices common in much earlier times.

References Allen, J. P. 1996 The Religion of Amarna. Pp. 3–6 in Images of Beauty from Ancient Egypt: The Royal Women of Amarna, by D. Arnold. New York: Metropolitan Museum of Art. Arnold, D. 1974 Der Tempel des Königs Mentuhotep von Deir el-Bahari, vol. 2: Die Wandreliefs des Sanktuares. Mainz am Rhein: von Zabern. 1987 Der Pyramidenbezirk des Königs Amenemhet III in Dahschur. Vol. 1. Archäologische Veröffentlichungen des Deutschen Archäologischen Instituts Abteilung Kairo. Mainz am Rhein: von Zabern. Beck, P. 1989 Stone Ritual Artifacts and Statues from Areas A and H. Pp. 322– 38 in Hazor III–IV: An Account of the Third and Fourth Seasons of Excavation, 1957–1958 (Text), ed. A. Ben-Tor. Jerusalem: Israel Exploration Society. Ben-Tor, A. 1995 Notes & News: Tel Hazor 1995. Israel Exploration Journal 45: 283–87. 1996a Notes & News: Tel Hazor 1996. Israel Exploration Journal 46: 262–68. 1996b The Renewed Excavations at Tel Hazor, 1990–1995. Qadmoniot 29/111: 2–18 (Hebrew).

1997 The Yigael Yadin Memorial Excavations at Hazor (1990–1993): Aims and Preliminary Results. Pp. 107–27 in The Archaeology of Israel: Constructing the Past, Interpreting the Present, ed. N. Silberman and D. Small. Sheffield: Sheffield Academic Press. 1998 The Fall of Canaanite Hazor: The “Who” and “When” Questions. Pp. 456–67 in Mediterranean Peoples in Transition: Thirteenth to Early Tenth Centuries bce, ed. S. Gitin, A. Mazar, and E. Stern. Jerusalem: Israel Exploration Society. Ben-Tor, A., and Ben-Ami, D. 1998 Hazor and the Archaeology of the Tenth Century b.c.e. Israel Exploration Journal 48: 1–37. Ben-Tor, A., ed. 1989 Hazor III–IV: An Account of the Third and Fourth Seasons of Excavation, 1957–1958 (Text). Jerusalem: Israel Exploration Society. Bonfil, R. 1997 Area A. Pp. 15–176 in Hazor V: An Account of the Fifth Season of Excavation, 1968, ed. A. Ben-Tor and R. Bonfil. Jerusalem: Israel Exploration Society. Borger, R. 1956 Die Inschriften Asarhaddons Königs von Assyrien. Archiv für Orientforschung 9. Graz: Weidner. 1971 Gott Marduk und Gott-König Sulgi als Propheten: Zwei Prophetische Texte. Bibliotheca Orientalis 28: 3–24.

The Sad Fate of Statues and the Mutilated Statues of Hazor Brandes, M. 1980 Destruction et mutilation de statues en Mésopotamie. Akkadica 16: 28–41. Brunner-Traut, E. 1982 Namenstilgung. Cols. 338–41 in vol. 4 of Lexikon der Ägyptologie, ed. H. W. Helck and W. Westendorf. Wiesbaden: Harrassowitz. Buccellati, G. 1982 The Descent of Inanna as a Ritual Journey to Khuta? SyroMesopotamian Studies 4/3: 3–7. Cohen, R., and Yisrael, Y. 1995 On the Road to Edom: Discoveries from ºEn Haz5evah. Israel Museum Catalog 370. Jerusalem: Israel Museum. Delaporte, L. 1940 Malatya, Arslantepe I. Paris: De Boccard. Delougaz, P., and Lloyd, S. 1942 Pre-Sargonid Temples in the Diyala Region. Chicago: University of Chicago Press. Dikaios, P. 1962 The Bronze Statue of a Horned God from Enkomi. Archäologischer Anzeiger 1962: 1–40. El-Saghir, M. 1992 Das Statuenversteck im Luxortempel. Mainz am Rhein: von Zabern. Foster, B. 1993 Before the Muses: An Anthology of Akkadian Literature. Bethesda: C.D.L. Frankfort, H. 1939 Sculpture of the Third Millennium b.c. from Tell Asmar and Khafajah. Chicago: University of Chicago Press. 1963 The Art and Architecture of the Ancient Orient. Baltimore: Penguin. Freedberg, D. 1989 The Power of Images. Chicago: University of Chicago Press. Garfinkel, Y. 1994 Ritual Burial of Cultic Objects: The Earliest Evidence. Cambridge Archaeological Journal 4/2: 159–88. Güterbock, H. 1961 When Was the Late Hittite Palace at Sakçagözü Built? Bulletin of the American Schools of Oriental Research 162: 49–50. Hallo, W. 1993 Sumerian Religion. Pp. 15–35 in Kinattutu sa darâti: Raphael Kutscher Memorial Volume. Institute of Archaeology Occasional Publications 1. Tel Aviv: Tel Aviv University. Hayes, W. C. 1959 The Scepter of Egypt. Vols. 1–2. New York: Metropolitan Museum of Art. Hurowitz, V. 1995 “ºAse lecha pesel.” Beth-Mikra 40: 337–47 (Hebrew). Jacobsen, T. 1976 The Treasures of Darkness. New Haven: Yale University Press. 1989 God or Worshipper. Pp. 125–33 in Essays in Ancient Civilization Presented to Helene J. Kantor, ed. A. Leonard and B. Beyer-Williams. Studies in Oriental Civilization 47. Chicago: Oriental Institute. Keel, O. 1972 Die Welt der altorientalischen Bildsymbolik und das Alte Testament. Zurich: Benziger. 1973 Das vergraben der “fremden Götter” in Genesis XXXV 4b. Vetus Tetamentum 23: 305–36. Koldewey, R. 1898 Ausgrabungen in Sendschirli. Vol. 2. Königlische Museen zu Berlin Mitteilungen aus den Orientalischen Sammlungen. Berlin: Spemann.

15

Kramer, S. N. 1951 Inanna’s Descent to the Netherworld. Journal of Cuneiform Studies 5: 1–17. Landsberger, B. 1948 Samªal: Studien zur Entdeckung der Ruinenstätte Karatepe. Ankara: Türkische historische Gesellschaft. Lichtheim, M. 1976 Ancient Egyptian Literature. Vol. 2. Berkeley: University of California Press. Luschan, F. von 1902 Ausgrabungen in Sendschirli. Vol. 3. Königliche Museen zu Berlin Mitteilungen aus den orientalischen Sammlungen. Berlin: Reimer. 1911 Ausgrabungen in Sendschirli. Vol. 4. Königliche Museen zu Berlin Mitteilungen aus den orientalischen Sammlungen. Berlin: Reimer. Matthiae, P. 1990 Masterpieces of Old Syrian Art: Discoveries of the 1988 Ebla Excavations in a Historical Perspective. Proceedings of the British Academy 75: 25–56. Mond, R., and Myers, H. 1940 Temples of Armant. Vols. 1–2. London: Egypt Exploration Society. Moortgat, A. 1969 The Art of Ancient Mesopotamia. New York: Phaidon. Negbi, O. 1976 Canaanite Gods in Metal. Tel Aviv: Tel Aviv University. Neusner, J. 1988 The Mishnah: A New Translation. New Haven: Yale University Press. Nielsen, E. 1955 The Burial of the Foreign Gods. Studia Theologica 8: 103–22. Nims, C. F. 1966 The Date of the Dishonoring of Hatshepsut. Zeitschrift für Ägyptische Sprache und Altertumskunde 93: 97–100. Nylander, C. 1980 Earless in Nineveh: Who Mutilated “Sargon’s” Head? American Journal of Archaeology 84/3: 329–33. Oppenheim, L. 1949 The Golden Garments of the Gods. Journal of Near Eastern Studies 8: 172–92. Peltenburg, E. 1989 The Beginnings of Religion in Cyprus. Pp. 108–26 in Early Society in Cyprus, ed. E. Peltenburg. Edinburgh: National Museums of Scotland. Posener, G. 1940 Princes et pays d’asie et de Nubie: Textes hiératiques sur des figurines d’envoútement du Moyen Empire. Brussels: Fondation Égyptologique Reine Elisabeth. Reeves, N. 1990 The Complete Tutankhamun: The King, the Tomb, the Royal Treasure. London: Thames & Hudson. Robins, G. 1993 Women in Ancient Egypt. Cambridge: Harvard University Press. Roehrig, C. 1996 Ancient Art from the Shumei Family Collection. New York: Metropolitan Museum of Art. Rollefson, G. O. 1983 Ritual and Ceremony at Neolithic ºAin Ghazal (Jordan). Paléorient 9/2: 29–38. 1986 Neolithic ºAin Gazal (Jordan): Ritual and Ceremony II. Paléorient 12/1: 45–52. Scott, N. 1946 Egyptian Statuettes. New York: Metropolitan Museum of Art.

16 Smith, M. S. 1997 The Baal Cycle. Pp. 81–180 in Ugaritic Narrative Poetry, ed. S. B. Parker. Writings from the Ancient World 9. Atlanta: Society for Biblical Literature. Soggin, J. A. 1975 ‘The Burial of the Godhead’ in the Inscription from Pyrgi (Lines 8–9) and Parallel Motifs in the Old Testament. Biblica et Orientalia 29: 112–19. Sollberger, E. 1956 Corpus des inscriptions ‘royales’ présrgoniques de Lagas. Geneva: Droz. 1971 Inscriptions Royales Sumeriennes et Akkadienes. Paris: Cerf. Sourouzian, H. 1988 Standing Royal Collosi of the MK Reused by Rameses II. Mitteilungen des Deutschen Archäologischen Instituts, Abteilung Kairo 44: 229–54. Spycket, A. 1968 Les statues de cult dans les textes Mèsopotamiens des origines à la 1ere dynastie de Babylone. Paris: Gabalda. 1981 La statuaire du proche-orient ancien. Leiden: Brill. Strommenger, E. 1964 The Art of Mesopotamia. London: Thames & Hudson. Thureau-Dangin, F. 1912 Une relation de la huitième campagne de Sargon (714 av. J.-C.). Textes cunéiformes du Louvre 3. Paris: Geuthner.

Amnon Ben-Tor Uehlinger, C. 1997 Anthropomorphic Cult Statuary in Iron Age Palestine and the Search for Yahweh’s Cult Images. Pp. 97–156 in The Image and the Book: Iconic Cults, Aniconism, and the Rise of Book Religion in Israel and the Ancient Near East, ed. K. van der Toorn. Leuven: Peeters. Ussishkin, D. 1970 The Syro-Hittite Ritual Burial of Monuments. Journal of Near Eastern Studies 29: 124–28. Weiss, H., ed. 1985 From Ebla to Damascus. Washington: Smithsonian Institution. Woolley, L. 1949 Introduction: i. The Statue. Pp. 1–9 in The Statue of Idri-mi, by S. Smith. London: British Institute of Archaeology in Ankara. 1955 Alalakh: An Account of the Excavations at Tell Atchana in the Hatay, 1937–1949. London: Society of Antiquaries. Yadin, Y. 1972 Hazor: The Head of All Those Kingdoms. The Schweich Lectures 1970. London: Oxford University Press. Yadin, Y., et al. 1958 Hazor I: An Account of the First Season of Excavations, 1955. Jerusalem: Hebrew University. 1960 Hazor II: An Account of the Second Season of Excavations, 1956. Jerusalem: Hebrew University. 1961 Hazor III–IV: An Account of the Third and Fourth Seasons of Excavations, 1957–1958 (Plates). Jerusalem: Hebrew University.

An Archaeologist on Mars Giorgio Buccellati Cotsen Institute of Archaeology University of California at Los Angeles

Go up any of the ancient tells and walk about, see the skulls of people from ages ago and from yesteryear: can you tell the difference? 1

Archaeology has been with us, it seems reasonable to imagine, ever since humans left traces for other humans to find. This bond across centuries is rooted in the awareness of what a human trace is: culture as evidenced by material remains. The instant recognition of a common past, which the text from ancient Mesopotamia cited above evokes in us, does not need scholarly support. But sorting out the differences does, and this is where archaeology as a discipline begins. This is also where reflection about archaeology begins—the philosophy, if you will, of archaeology. It is one of William G. Dever’s many merits to have contributed to this issue in a programmatic way, particularly in his recent volume What Did the Biblical Writers Know and When Did They Know It? (2001), in which he takes up a wide range of issues dealing with the epistemological underpinnings of the discipline. He especially confronts the implicit assumptions that condition, more or less unwittingly, the mapping of archaeology onto history, with regard to ancient Israel. As a tribute to his efforts, I would like to pursue here a similar line of reasoning, taking up a single, core issue and drawing some conclusions that directly pertain to Dever’s central argument.

Definition and Definitions The core issue is the very understanding of what archaeology is. It is not merely a question of semantics but of principles and presuppositions that condition any dis-

1. I am taking some liberty in the translation of this passage from the Mesopotamian Dialogue of Pessimism (Lambert 1967: 149, lines 76– 78). The original of the last line cited reads, “who was the doer of evil deeds and who the one who brought help?” It is remarkable how the text explicitly assumes that one can differentiate between ancient and recent finds (“the skulls of the ancient and the recent ones”: gulgulle sa arkûti u panûti) and then asks the question about possible higher-level interpretations of the finds (the moral quality of the individuals behind the skulls: ayyû bel lemuttim-ma ayyû bel usati?).

course about further inferences—such as, precisely, the inferences addressed by Dever. If it might at first seem otherwise, it is perhaps because of a confusion between definition and definitions. The latter, and looser, term, definitions, implies no more than a common-sense approach to the task of segmenting into conceptual categories the universe of discourse. Such are, for instance, the definitions in a dictionary, which in fact are paraphrases or descriptions. Having established that a term refers to a given item (let us say, a horse), one singles out the attributes of the same item that appear most distinctive (a quadruped, with a certain type of tail, etc.), and one evokes thereby a picture to which the term can properly be said to apply. On the other hand, definition, understood in the stricter sense of the term as the act of defining, implies that we make explicit the boundaries of a concept by way of contrasting it with all pertinent and related concepts within a given structural system. One might say that in this second approach one is concerned not so much with what is within the boundaries of a concept as with what is outside of them. It is in fact this contrast with immediate neighbors that maximally and explicitly circumscribes any part of a given whole. One may remark in this connection that the application of a binary system of contrasts is often preferable in insuring a proper definition of concepts. By focusing on complete dichotomies, a binary system provides for the most explicit boundaries. Two components of a binary pair are intrinsically self-limiting, contrasting parts of a given whole—they are, as one would properly say, mutually exclusive. Such complete discreetness is the best guarantee for effective analysis, a process whereby we “dissolve” (as the etymology of the term analysis would have it) a given whole into its component parts. Analysis or “dissolution” of this sort will obviously be most successful if no overlapping of “dissolved” components is permitted—or, as one again would properly say, if the attributes are indeed mutually exclusive.

Giorgio Buccellati

18 A concept, then, is defined by having its boundaries established vis-à-vis other concepts—the sharper the definition, the more powerful the concept. A concept exists as a function of its boundaries; that is, we recognize the boundaries first and then we name the concept that those boundaries in fact delimit. The term used for a concept is no more than a label applied to a segment of a structural system: the label as such has no conceptual power, other than as an index. It emerges from all of this that a definition, understood as the process of de-fining (etymologically, to articulate the limit), is something more meaningful than a simple paraphrase: we are concerned with the very structure of a conceptual system as a whole and with the correlations and reciprocal delimitations of its component parts, rather than with individual concepts in themselves.

The Grip of the Earth In practical terms, the concepts discussed above may be understood to mean that archaeology ought to be defined in terms that make explicit what it is that no one other than an archaeologist does. Instead of “archaeology as anthropology” or as anything else, we ought to look explicitly at “archaeology as archaeology.” There is, in my view, only one answer to this question, or rather, two closely related answers. The first is rooted in the physical dimension and defines archaeology stricto sensu as the stratigraphic analysis, through excavation, of cultural remains embedded in the ground. The second builds on the inferences that can be drawn from the disentangled stratigraphy and defines archaeology lato sensu as the study of broken traditions. 1. In a strict sense, then, archaeology should be viewed as the stratigraphic analysis of cultural remains, first as they are found embedded in the ground (emplacement) and then as they can be understood to have gotten there (deposition). Since emplacement is the only aspect of the archaeological universe that we can properly document, it is the foundation of everything else that can be said about archaeology. This statement may seem too radical, but in fact no one else but archaeologists can, and must, document what is in the ground and how it relates to everything else that is there as well. The grip of the earth masks everything. It must be lifted, and this process of disentanglement itself must be documented, not just the pieces that emerge from it. Emplacement refers to how “things” are in a complex matrix that includes other “things”—whether objects (typologically identifiable separately from the context in which they are embedded) or not (not so identifiable). As archaeologists, we do not “enter” a room. We disentangle a brick today, another tomorrow (or maybe

next year), until the room is reconstituted in its typological identity and can then be “entered”—by the architect, the anthropologist, the art historian, the historian of religion (if the room happens to have a cultic function). It may well be (in fact, it ought to be) the same person who reasons along all of these lines of inquiry, or who at least has a sensibility for all the pertinent dimensions. But the fundamental commitment that must precede everything else is to emplacement and the fundamental obligation is to the record that is made of it. This no one else but the archaeologist can do. Halfway between the emplacement record and functional analysis (artistic, sociological, religious, etc.) lies the reconstruction of depositional history. Deposition is how things got to where they are when found. It is important to note that we do not observe, and therefore we cannot document, deposition. It can only be inferred from emplacement. When we call a given feature a “foundation,” we are making a depositional judgment. What we observe is the emplacement of elements such as (a) the juxtaposition of two different kinds of soil, (b) different planes that define the boundaries within which the two types of soil are contained, some aligned vertically and others horizontally, (c) the presence of bricks laid within the inner soil, and (d) the regularity with which these bricks continue above the upper horizontal boundary of the soil. All of this can be observed and documented. On this we build a depositional inference, and we speak (in slightly different order) of (b) a trench defined by a cut and the floor above it, (a) the fill within the trench, and (c) the brick foundation of (d) a brick wall. This inference is by definition interpretive, and must be kept separate from what is in fact observed and documented (the emplacement). It goes without saying that in practice we shortcircuit the process and speak of a foundation the moment we see it. But this ought to be only a mental shortcut, not a mixing of levels of analysis. The inferential argument that takes us from emplacement to deposition is also something that the archaeologist alone can adequately produce. And this is what is known as stratigraphic analysis. It rests on a full understanding of how observation ought to be carried out and on a sophisticated recognition of depositional patterns and laws. There is a constant interchange between the two, emplacement and deposition. In particular, the strategy of excavation (the unraveling of the emplacement) rests directly on presuppositions about how things that we still only partially see in the ground have gotten there (the inference about the depositional sequence). Thus, we must argue emplacement from deposition, even while keeping the two lines of reasoning quite distinct. These concerns add up to a proper theory of excavation, something to which very little atten-

An Archaeologist on Mars tion is being paid, even in an era when theory dominates so much of our scholarly attention. 2. In a broader and secondary sense, archaeology can be understood as the study of broken traditions. The cultural remains that we extricate from their own collapse exhibit no direct line of continuity with our current experience; there are no living carriers of the culture from which they originate. They bear within themselves a cipher, as it were, that we must discover, so as to “de-cipher” their meaning. And so we must recompose a cultural whole within which they may make sense once more. Just as there are “dead languages,” so there are “dead cultures”— not because they never lived, but simply because no living person has today the linguistic or cultural competence to tell us what that whole is. What is done to retrieve meaning within a dead culture only the archaeologist can do. For it is the archaeologist who is trained to correlate assemblages that are linked stratigraphically and to insert each new fragment that is brought to light within its own pertinent framework. In this secondary sense of “archaeology,” the time gap between deposition and excavation is not that critical—in other words, it is not necessary that the archaeological material be “ancient.” There can be very recent cultures, the meaning of the material remains of which escapes us, and more ancient cultures for which filaments of continuity can be detected, generally among populations that inhabit the same territory. And this is the philosophical presupposition for the validity of ethno-archaeology—the recovery of these filaments.

The Great Disconnect Relating the case of an extraordinary autistic person—a woman by the name of Temple Gardin who went on to join the faculty of Colorado State University—Sacks quotes her as saying: “Much of the time I feel like an anthropologist on Mars” (1995: 259). 2 This is a profound statement on which the author elaborates with great insight, describing the ways in which Gardin could relate to the world of feelings only via abstract thought. Thus, he writes that “she could understand ‘simple, strong, universal’ emotions but was stumped by more complex emotions and the games people play” (Sacks 1995: 259). For her, “ ‘normality’ had been revealed more and more . . . as a sort of front, or façade . . . albeit a brave and often brilliant front, behind which she remained, in some ways, as far ‘outside,’ as unconnected, as ever. ‘I can really relate to Data,’ she said . . . an android [from the TV series Star Trek] who, for 2. The same quotation is also cited in Sacks 1995: 269, 292, and is obviously the quotation borrowed for the title of Sacks’s book (and of my article).

19 all his emotionlessness, has a great curiosity, a wistfulness, about being human. He observes human behavior minutely, and sometimes impersonates it, but longs, above all, to be human” (Sacks 1995: 275). Sacks relates that to obviate the great disconnect that marked Gardin’s experiential range, she “built up a vast library of experiences. . . . They were like a library of videotapes, which she could play in her mind and inspect at any time—‘videos’ of how people behaved in different circumstances. She would play these over and over again and learn, by degrees, to correlate what she saw, so that she could then predict how people in similar circumstances might act” (1995: 259–60). Sacks explains that, “when she was younger, she was hardly able to interpret even the simplest expressions of emotion; she learned to ‘decode’ them later, without necessarily feeling them” (1995: 269), and that “she has . . . to ‘compute’ others’ intentions and states of mind, to try to make algorithmic, explicit, what for the rest of us is second nature” (1995: 270). The parallel with our case is illuminating. The cultural remains unearthed by the archaeologist are buried not only physically but also metaphorically. They are separated from our reach by the physical cover of their own collapse and of the accumulations that followed on top; but they are also separated from our immediate understanding by the great rift caused by a time gap that has severed direct continuity. They are the mute evidence of, precisely, broken traditions. For the archaeologist, the break is vertical, as it were, across time, whereas for the autistic person studied by Sacks, it is horizontal, because of the makeup of her personality. Thus, we as scholars approach the broken traditions recovered from the earth much as an autistic person would approach normal human interaction—before them, we remain “far outside,” “unconnnected,” “emotionless.” We are always “archaeologists on Mars,” because we cannot presume continuity but must instead supply the argument that takes the place of intuitive understanding. We do this through the methodical and laborious identification of patterns that we know must have carried meaning. Even when we do not comprehend the meaning, we may legitimately assume it. But we can never presume to suggest meaning if we have not first established the regularity of patterns. The alternative is pure fantasy, the domain of storytellers. As scholars, we can only start from established patterns. Akin to semiotics, cognitive archaeology lays the groundwork for this kind of effort. In linguistics, semiotics reads value into signs that are shown to have significance because they are recurrent. Archaeology in and of itself can attribute meaning without presuming to define its proper value. But cognitive archaeology accomplishes this by establishing higher levels of correlation. Interestingly, Gardin described her effort at reading meaning

Giorgio Buccellati

20 into emotions in terms of cognitive analysis: “I had to learn it cognitively. I could put two and two together [in order to know that a colleague was jealous], but I couldn’t see the jealous look on his face” (Sacks 1995: 260). But can we, too, as archaeologists, go beyond the cold description of these patterns and presume to propose meaning? Can we re-embed past experiences in our own experience without fantasizing about them?

Healing the Rift The distinction and relation between social sciences and humanities lie at the core of our answer. A humanistic approach aims at recapturing for our live experience another experience that is “dead,” not in the sense that it never lived, but rather in the sense that its carriers no longer live. The humanistic conviction is that patterns originate in experience, that they are handed down through culture, and that assimilating the patterns can lead us back to their experiential source. The scholarly dimension of humanism is in the careful grounding of such renewed experience upon definable patterns. These can be described as part of the factual record—hence documented, dissected, and analyzed. But beyond the patterns, or beneath them, we can identify with that magical moment that is the recreated experience. It is as if the humanist could lend a sense of wholeness to the fragments. Against the backdrop of the experiences described by Sacks, we can think of the social scientist as laboring in the way an autistic person does, creating libraries of data and identifying correlations that presume meaning, and of the humanist as pointing to the nexus that holds the fragments together. Once more, this reminds us of Temple Gardin: as a child, she knew that “something was going on between the other kids, something swift, subtle, constantly changing—an exchange of meanings, a negotiation, a swiftness of understanding so remarkable that she wondered if they were all telepathic. She is not aware of the existence of these social signals. She can infer them, she says, but she herself cannot perceive them, cannot participate in this magical communication directly, or conceive the many-leveled kaleidoscopic states of mind behind it. Knowing this intellectually, she does her best to compensate, bringing immense intellectual effort and computational power to bear on matters that others understand with unthinking ease. This is why she often feels excluded, an alien” (Sacks 1995: 272). Not so with us. We need not be either social scientists or humanists. We must be both. We can and must internalize once again the “exchange of meanings.” The experience of the ancients, broken though it is, can be reinserted in ours—not in a fantastic mode but through scholarly reasoning.

In this sense, archaeology can be seen as paradigmatic. It starts with the greatest distance—the mute testimony of a material culture cut off from our experience—and ends with the resurrection of a culture and, through it, of the experience behind that culture. For this remains, ultimately, the goal of every archaeological endeavor. We may feel like archaeologists on Mars, without initial points of reference except for the internal coherence of the data and the recognition of patterns that emerge from them. But unlike the autistic experience, and unlike, as a matter of fact, an archaeologist on Mars, we can expect to find clues that can establish some continuity and thus heal the rift. In its most minimal form, this is the continuity of the most basic of human experiences. But if we can go beyond this and appropriate more specific moments and more articulate expressions of this experience, all the better.

Biblical and Other Archaeologies The many qualifications that are in current use or have been introduced to modify the concept of archaeology may be grouped into three major categories. On the one hand, there are terms that reflect a global approach to the discipline, such as “new archaeology.” On the other, there are terms that apply to specific sectors of the discipline in reference to methods or techniques, such as “underwater archaeology,” “ethno-archaeology,” and even “archaeometry.” Finally, there are qualifications that refer to specific sectors of cultural data to which archaeology applies, as in “classical” or “Near Eastern” archaeology. The notion of “Biblical Archaeology” would seem to belong to the last. But if one thinks about it, it is rather anomalous to speak of the archaeology of a book. It would be like speaking of “Livian Archaeology.” These terms might be understood as answering the question what can archaeology tell us about the facts referred to in the Bible or in Livy. And to this extent, they are acceptable concepts. But two major problems quickly arise if one goes beyond this simple understanding of the term. 1. As documents, the Bible and Livy belong to a continuous manuscript tradition that began at some remove from the time of the initial composition. The stream of actual philological witnesses (essentially codices) begins later, at some distance in time from the original “edition.” The portion of the textual tradition that precedes the extant manuscript tradition is indeed presupposed by the latter but not documented. It is “broken” in a literal sense. Thus, the first problem is that we are tempted to deal with these particular textual traditions as if they instead were closed traditions, carrying with them a documentable contextual understanding. In addition, the manuscript tradition, as we have it, is a cultural carrier that elaborates, interprets, and interjects at the same time as it hands down the text—not

An Archaeologist on Mars

21

within the text itself, but as an interpretive apparatus that develops concurrently with the text. Thus it is that the Bible or the Livy of the manuscript tradition is an artifact that must also be understood in stratigraphic terms, as it were—that is, detached from the (later) chain of witnesses (interestingly, this is the thrust of Dever’s 2001 volume title, What Did the Biblical Writers Know and When Did They Know It?). 2. Another consideration is that, while the concept of “Biblical” or “Livian” Archaeology is valid and legitimate, it is of limited power and potentially harmful as well. It has limited conceptual power because it properly refers only to the correlation between stratigraphic data on the one hand and cultural data filtered through a partly undocumented manuscript tradition on the other. It is potentially harmful because it may lead to an undue superposition of goals. Imagine a dig or a series of digs undertaken with the express purpose of illustrating Livy’s work, let alone of proving him right. The overlap would be so limited that if the excavators were to be true to their stated intents, they would effectively destroy and leave undocumented most of the archaeological record. While we can hardly expect this to happen in actual practice, the mixing of levels of analysis is sufficiently dangerous to elicit special attention. Thus, the case of Biblical Archaeology is the most dramatic example of the potential contrast between a broken and a continuous tradition. The pertinent archaeological record is, in effect, a broken tradition that is claimed by a continuous tradition as part of itself. And while there are compelling reasons that this often ends up being the case, we cannot accept that one should override the other.

To conclude, we may consider a comparison that is instructive and ironic at the same time—the comparison between Biblical Archaeology and ethno-archaeology. Ethno-archaeology pays careful attention to the survival, however opaque, of ancient cultural features in the life of modern living groups. The analogy with Biblical Archaeology is that the latter pays just as much attention to the survival of equally ancient cultural features in a later written record. The irony is that ethno-archaeology should be accepted with an enthusiasm that is often only matched by the skepticism directed instead toward Biblical Archaeology. The reason is that the biblical written record has a higher degree of awareness of its past than ethnographic survivals, and this generates the suspicion of a hidden agenda. But precisely for this reason, the comparison with ethno-archaeology is instructive, because ethno-archaeology may serve as a good paradigm for understanding the merits, and limits, of Biblical Archaeology. In both cases, we seek to verify the validity of assumed continuities that bridge the gap, the brokenness. Such assumed continuities cannot be accepted at face value: the validity of a link across the hiatus of the broken tradition must first be proved. To do this, no one is properly equipped but the archaeologist, in the sense that I have described above. Thus, biblical archaeologists should not be understood as the antiquarians who relate text to artifacts on the basis of mere typological analogies. Rather, as if after the fashion of a Kantian “Critique of Archaeological Reason,” they ought to serve as those who calibrate the stratigraphic context of the material finds against the textual context of the written record.

References Dever, W. G. 2001 What Did the Biblical Writers Know and When Did They Know It? What Archaeology Can Tell Us about the Reality of Ancient Israel. Grand Rapids, MI: Eerdmans.

Lambert, W. G. 1967 Babylonian Wisdom Literature. Oxford: Clarendon. [repr. Winona Lake, IN: Eisenbrauns, 1996] Sacks, O. 1995 An Anthropologist on Mars: Seven Paradoxical Tales. New York: Vintage.

Of Pots and Paradigms: Interpreting the Intermediate Bronze Age in Israel/Palestine Shlomo Bunimovitz and Raphael Greenberg Department of Archaeology and Ancient Near Eastern Cultures Tel Aviv University

Well over half a century ago, the Classical archaeologist and idealist philosopher Robin Collingwood stated that “no historical problem should be studied without studying . . . the history of historical thought about it” (1939: 132). Indeed, in recent years, a growing number of archaeologists have come to agree with Collingwood’s dictum. Behind this current consensus stands a central postmodernist realization—which by now encompasses the entire scientific world—that all knowledge is socially constructed. Postmodernism, with its subjectivist agenda, encouraged the belief that archaeology is not independent of the social and cultural context in which it is practiced. The questions that archaeologists investigate, the evidence they are predisposed to accept as conclusive, and even what they recognize as evidence are all influenced, whether consciously or unconsciously, by their intellectual persuasions, disciplinary paradigms, class interests, ethnic loyalties, and gender prejudices (see, e.g., Knapp 1996; Johnson 1999; Hodder 2002). These ideas have had a great impact on the way the history of archaeological thought is currently conceived. In a classic essay concerning the image of the American Indian in archaeology, Trigger convincingly demonstrated how the interpretation of archaeological data is shaped by contemporary archaeological paradigms and changes with them (1980a). This lesson is repeated in recent histories of archaeological thought, whether general (e.g., Trigger 1989; Patterson 1995), or specifically treating the archaeology of Israel/Palestine (e.g., Moorey 1991; Silberman 1995; Bunimovitz 1995). Coauthoring an essay about the social and symbolic meaning of the ubiquitous drinking vessels of the Intermediate Bronze Age (IBA, ca. 2300–2000 bce; also termed Early Bronze Age IV, Intermediate Early Bronze–Middle Bronze, or Middle Bronze I), we have noticed three major conceptual shifts in the interpretation of this phenomenon (Bunimovitz and Greenberg 2004). Put within the historical perspective of archaeological thought, these shifts seem to be the product of changes in the discipline’s para-

digmatic framework, rather than of new data. Indeed, some of the details concerning the introduction of foreign drinking habits into Canaan were established already in the 1920s and are still viable today (see below). Taking the long-term study of the IBA as a case in point, we wish to illustrate what has become, in some archaeological circles, almost a truism: “facts” (or rather, pots) do not speak for themselves; it is only through theory that their mute voice can be heard. Moreover, in light of the postpositivist stance detailed above, we further show that the very facts concerning the cups and teapots of the IBA are perceptiondependent—the same data are perceived differently when investigated under different conceptual paradigms (cf. Kuhn 1970; Hodder 1991: 15–18; 165–81). We present this critical essay in honor of Bill Dever, a friend and colleague who has for many years championed the study of both the IBA/EB IV period and the methodological development of the archaeology of Israel/Palestine, and whose periodic state-of-research essays serve as a constant challenge to our professional perceptions.

Diffusion and Migration: Culture-History and the “Caliciform Culture” in Canaan The founding figures of Palestinian archaeology identified intrusive cultural elements in the material culture of the IBA with affinities to the distant north. Albright (1932: 8–14; 1935: 220; 1960: 80) and Wright (1938) were the first to comment on similarities between the period’s pottery—in particular the decorated cups common in southern Canaan—and the so-called “caliciform” ceramics of latethird-/early-second-millennium bce Syria and northern Mesopotamia. Each of them, however, suggested a different mechanism for cultural transmission, namely, diffusion versus invasion. As we shall see, both concepts haunted IBA research for the next half century, until the 1980s, and faithfully reflected the culture-historical paradigm that ruled the archaeology of Palestine at the time.

24 For Albright, the key concept explaining the arrival of Syro-Mesopotamian elements in IBA Canaan was diffusion: from a “focus” of Amorite culture established in Mesopotamia during the 22nd–21st centuries bce, cultural elements “streamed” back into Syria and Canaan (Albright 1935: 220). In successive editions of his celebrated book, The Archaeology of Palestine—a building block in the professional education of generations of archaeologists in the region—Albright emphatically repeated the idea that the descent of the “caliciform ceramic culture” from Syria was “a cultural drift associated with diffusion of the SyroMesopotamian culture,” rather than a shift of peoples (1960: 80 [emphasis added]). 1 Wright was not so sure about the causes of the change in the ceramic industry between the EB II–III and the EB IV–MB I periods. Yet he offered a more drastic explanation than Albright: “It is not impossible that the shift is the result of an invasion of Amorite barbarians” (Wright 1938: 34 [emphasis added]). Indeed, whether by Amorites or other “barbarians,” invasion soon became the favored explanation for the cultural change evident in Canaan at the end of the Early Bronze Age. Following her excavations at Jericho in the 1950s, Kenyon advocated the idea of Intermediate Early Bronze–Middle Bronze Age “nomadic invaders” and of “such a wholesale incursion of newcomers that the existing population was completely submerged” (1957: 187). The North Syrian/Mesopotamian connection of the pottery found at Jericho and at other sites indicated, according to her influential statement of the case, that the new groups were Amorite pastoralists (1957: 186–209; 1971; 1979: 119–47). Surprising as it may seem in light of his later criticism of Kenyon’s “Amorite Invasion” hypothesis (see below), Dever also reconstructed incursions of Amorite pastoralists from the Syrian periphery into Canaan in his early treatments of the IBA (1970; 1971). According to Dever (1971: 220): (1) there was a natural geographical route for the migration of the Amorites into Canaan; (2) the Amorite way of life accounts remarkably well for the physical evidence from “MB I” Canaan; and (3) the preference for steppe regions for settlement in MB I Canaan reflects the newcomers’ adaptation to such a terrain in Syria. 2 The invasion theory was taken a step further by Tufnell, who linked the “invaders of the Caliciform Culture,” as she called them, with the Beaker Folk of Europe. Both 1. On another occasion, Albright proposed both a diffusion of Syrian culture into western Palestine and the settling of the country by “nomadic tribes from the desert,” who overran it during the last quarter of the third millennium bce (1938: 16). 2. Dever later described his early perception of the IBA as “an attenuated version of the ‘Amorite hypothesis’,” with further qualifications (1980: 55).

Shlomo Bunimovitz and Raphael Greenberg groups seem to have favored beaker-shaped vessels decorated with patterns arranged in zones and often executed with a sort of comb. In Europe, too, the ware is different from the wares current immediately before and after the Beaker episode. Finally, both peoples were also active in the practice of metallurgy. Tufnell anticipated that future research would establish further links between the two groups “living at the same time and on the same economic plane” (1958: 31). A link between the IBA material culture and the Beaker Folk cultural complex of the western Mediterranean was also investigated by Lapp. Finding the evidence inconclusive, he preferred the idea of non-Semitic invaders from beyond the Caucasus in Central Asia (1966: 88–116). In the same vein, Kochavi (1967) portrayed the IBA people and their “nomadic culture” as an offshoot of one of the great waves of Indo-Europeans moving from Central Asia toward Europe at the end of the third millennium bce. The above survey of the main opinions about the IBA prevalent during the formative years of the archaeology of Israel/Palestine until the end of 1970s shows them to be part and parcel of the culture-historical paradigm. Culturehistory is the bread and butter of archaeology, a foundation course upon which all further research is based. The paradigm involves the building of chronological and developmental sequences for ancient cultures, relying on material finds and their affinities with neighboring cultures. The result of this procedure is a mosaic of cultures whose chronological and geographical distribution in a certain region can be shown on comparative charts and maps. The paradigm was personified in Europe, the Mediterranean, and the Near East by the towering figure of V. G. Childe, although its roots go back to late-19th–early-20th-century European scholars (see Trigger 1980b; 1989: 148–74; Jones 1997: 15–26; for culture-history in North American archaeology, see Lyman, O’Brien, and Dunnel 1997). The cornerstone of the culture-historical paradigm is Childe’s definition of archaeological culture as recurring sets of associated artifacts or traits held to represent a people or society (1929: v–vi). Such an idea of culture has been called normative, because its basic assumption is that artifacts are expressions of cultural norms or mental templates shared by all members of a certain group and transmitted over generations. One important implication of this view is that cultures seem to be unchanging; change is brought in from the outside, from another human group. The outside influence can come in two forms: migration of peoples or diffusion—the spread of new ideas through contact between groups. Both forms of change can be gauged archaeologically: migration by a wholesale change in the material culture assemblage and diffusion by the appearance of new types of artifacts (mainly pottery), the

Of Pots and Paradigms: Interpreting the Intermediate Bronze Age in Israel/Palestine origins of which need to be sought outside the culture investigated (cf. Renfrew and Bahn 1991: 407). The ideas about IBA culture reviewed above conform to the normative view of culture. A “focus” of (Amorite) culture from which cultural elements “streamed,” “cultural drift,” and “diffusion of (Syro-Mesopotamian) culture” (Albright, above) are all phrases typical of the “aquatic view of culture,” as Binford derisively called the interpretive framework of culture-history (1965: 205). What Binford meant was that the normative school of culturehistorical archaeology viewed the ancient world as a large pool of water. When an innovation occurred for whatever reason in a given place, it would tend to spread through the process of “influence” or diffusion in all directions, like the ripples from a stone dropped in the pool. There is little explicit explanation in such a description of why this or that pottery style diffused, or why this or that culture spread or changed (cf. Johnson 1999: 19). The same normative discourse also produced a second group of key words: “invasion/migration” of “nomadic barbarians/invaders.” Whether caused by Semitic Amorites from Syria (Wright, Kenyon, Dever) or Indo-European Beaker Folk/Transcaucasian emigrants spreading from their remote native homeland all over Europe and the Mediterranean (Tufnell, Lapp, Kochavi), culture change in Canaan at the end of the Early Bronze Age was related to new populations occupying the land. It is not our intention to criticize the above ideas but to expose them as mirroring the archaeological paradigm of their time. One cannot ignore, however, the simplicity and naïveté of the “explanations” offered for the conspicuous culture changes in Canaan at the end of the Early Bronze Age. As recent research has shown, invasion, migration, and diffusion are legitimate mechanisms of culture change, but each of these processes must be thoroughly studied and not merely assumed (see, e.g., Adams, Van Gerven, and Levy 1978; Anthony 1990; Schortman and Urban, eds. 1992; Chapman and Hamerow, eds. 1997; Burmeister 2000). In any event, the new, provocative archaeological paradigm that emerged in North America during the 1960s and 1970s rendered the existing hypotheses about the IBA ill conceived and outdated in the eyes of a young generation of Syro-Palestinian archaeologists, led by their teacher and mentor, W. G. Dever. New ideas and a new discourse now characterized the investigation of this period.

Internal Social Process: The Impact of the “New Archaeology” In a 1980 issue of the Bulletin of the American Schools of Oriental Research, a programmatic collection of essays concerning the MB I–MB IIA periods was published by

25

Dever and his colleagues and Ph.D. students, in the hope that it would “mark a shift in the stance from which Palestine (in particular) is viewed ca. 2500–1800 b.c.” (Dever 1980: 35; see also the “editorial”). Indeed, the essays specifically dealing with the IBA in Israel/Palestine–Transjordan (Dever 1980; Richard 1980) mark a watershed in the study of the period. Their major innovation was the uncompromising rejection of the prevalent invasion/diffusion perception of culture change at the end of the Early Bronze Age and its replacement with the idea of internal change (see also Richard 1987: 34–40). Concomitant empirical constituents of this view were the contention that “virtually the entire post-EB III assemblage represents an indigenous, not an intrusive culture” (Richard 1980: 8) and the adoption of an EB IV terminology implying continuity with the Early Bronze Age cultural tradition. Both arguments were based first and foremost on a formal ceramic typology of the EB III–EB IV pottery assemblages in Israel/Palestine and Transjordan, emphasizing the Early Bronze Age antecedents of the EB IV ceramic tradition. The same continuity was also claimed for other spheres of material culture, such as the lithic and metallic industries, as well as mortuary architecture (Dever 1973; 1980: 38; 1985: 123, figs. 9–11; Richard 1980: 12–22; 1987: 35). In a context of this sort, the “caliciform” elements in the pottery assemblage—teapots and cups—were viewed as local forms, and elements previously considered to attest to northern influence or migration were downgraded to an “awareness of a tradition in vogue in Syria” and to “an intermingling of influences [that] simply arose from close cultural contact between Syro-Palestinian peoples on the EB III/EB IV horizon” (Richard 1980: 18, 23). The apparently imported, black, wheel-made teapots and goblets found at Megiddo and at other northern sites (see, e.g., Guy and Engberg 1938: 148; Tadmor 1978) were tagged as “anomalous” imports or “luxury goods” (Dever 1980: 46, 50) or as testimony of continuing trade with Syria, “although on a relatively small scale” (Richard 1987: 38). Thus, they were left outside the discussion of the local cultural context. The new perception of the IBA also relied on two important theoretical components. First, a systemic view of the collapse of Early Bronze Age urban society. The meaning of this stance is that the breakdown that occurred in EB III and the ensuing deurbanization were seen as the outcome of imbalances and complex interactions between the subsystems (e.g., environment, sociopolitical structure, subsistence economy, foreign relations, etc.) that constituted the Early Bronze Age cultural system rather than as the result of a supposed invasion (Fargo 1979; Richard 1980: 25; 1987: 34; Dever 1989). A systemic approach

26 was also used to assess socioeconomic change during the IBA itself (Dever 1985). Second, modern ethnographic and anthropological models concerning nomadism were applied to the deurbanized situation in Palestine in the late third millennium bce (Dever 1980: 53–58; Kamp and Yoffee 1980; Richard 1980: 9–10; 1987: 36, 39–40). New research on nomadism in Mesopotamia and sociobiblical studies of the Israelite conquest of Canaan provided a better understanding of seminomadic life in the Near East and its symbiotic relationships with both the natural environment and urban society and cast new light on the vexed problems of the posturban Early Bronze Age era. Socioeconomic change during the EB III–EB IV transition in Canaan—increased pastoralism and village life at the expense of urbanism—was now understood as an adaptive strategy pursued in the wake of the demise of the urban centers. This view rendered the association of change in the period’s material culture with new people “one of the classic examples of a misused ‘pottery-to-peoples’ argument” (Richard 1980: 10). Apparently, “Amorite invasion” as an adequate explanation for socioeconomic and culture change during the IBA was “out,” whereas indigenous pastoral nomadism and de-specialized economy were “in.” This change in the perception of the IBA is most distinctly manifested in two maps of MB I/EB IV Canaan published by Dever within a time span of two decades. The maps are identical in all the graphic details, yet the early version instructs the viewer that the bold black arrows drawn on the map denote “initial movements of MB I peoples into Palestine” (Dever 1971: 200), while in the later publication they “indicate movements of cultural influence, not ethnic groups” (Dever 1985: fig. 1). The new trends in IBA scholarship were also mirrored in fieldwork. The Central Negev Highlands Project, a joint American-Israeli undertaking directed by Dever and R. Cohen from 1978 to 1980, was devised “specifically to field-test the newer ethnographic and environmental theories of EB IV settlement sites in the Negev” (Dever 1980: 55). 3 The research program of this multidisciplinary project brought to the fore yet another new methodological aspect: the formulation of hypotheses and an ambition to test them empirically (Cohen and Dever 1978: 29–31). 4 3. The interdisciplinary spirit of this regional project and its emphasis on “man-land” relationships echo American projects conducted in the same vein during the 1970s—for example, the Minnesota Messenia Expedition in Greece (McDonald and Rapp, eds. 1972; compare especially fig. 1–1 there with Cohen and Dever 1979: fig. 4). 4. Some of the above ideas, especially the claim of continuity from the Early Bronze Age to the IBA, were already voiced at the beginning of the 1970s (see, e.g., Dever 1973; Oren 1973a; 1973b; Prag 1974). Yet these important treatments of certain facets of the period were limited in scope and lacked the holistic perception that characterizes the later works discussed above.

Shlomo Bunimovitz and Raphael Greenberg Clearly, the innovations in IBA scholarship reflected new ideas presented within the wider context of contemporary archaeological thought. A growing dissatisfaction with the culture-historical paradigm had led to the appearance of a new paradigm in the 1960s —the New Archaeology (for its sociological and intellectual background, see Binford 1972: 1–14; Trigger 1989: 244–328; Redman 1991; Patterson 1995). First and foremost, the “new”—or processual—archaeology was explicitly antinormative. Culture was defined in terms of adaptive responses to the natural environment, rather than as a collection of arbitrary norms and ideas held in common by members of a particular social group. Furthermore, culture was viewed as a system comprising many subsystems (e.g., economic, social, ideational, etc.), joined together by a network of mutual relations. Diffusion and migration were denied a central role in culture change. Instead, self-generated change was seen as society’s response to disequilibrium. Cultures were viewed as classifiable along an evolutionary scale, from simple to complex. Emphasis was put on the explanation of long-term culture processes, because these were conceived as better suited to archaeology’s time scale than ephemeral, historical events. The “New Archaeology” aspired to assimilate anthropology for its ability to generalize about human cultures and the natural sciences for their positivist and rigorous “scientific” procedures— mainly deductive reasoning and hypothesis testing—supposedly ensuring the objectivity of explanation (see, e.g., Trigger 1989: 294–312; Johnson 1999: 12–33). While the novel trends in IBA research at the late 1970s obviously echoed the American processual archaeology in vogue at the time, the pertinent IBA literature oddly lacks explicit reference to the fundamental works of this paradigm. However, a contemporary examination of the impact of the New Archaeology on Syro-Palestinian archaeology—one of Dever’s periodic “state-of-the-art” essays—sheds light on the theoretical backdrop, or subtext, of the (then) latest IBA studies (1981: 15–16): I would argue that the major innovations of the “new archaeology” among the more avant-garde Syro-Palestinian specialists in the 1970s were all based on assumptions borrowed—often naively or even subconsciously—from American anthropology and archaeology. The changes were seen most obviously in: (1) the multi-disciplinary orientation; (2) the broader consideration of environmental factors (“ecology”); (3) the recognition of the value of ethnographic parallels (“ethnoarchaeology”); (4) the employment of general systems-theory (a “holistic” or systemic theory of culture), with its quantitative method; (5) the logic of “explicitly scientific” method, with its hypothesis testing; and (6) the adoption of the “behavioral-processualist” goal (from one school of cultural anthropology). These are precisely the fundamental tenets of the “new

Of Pots and Paradigms: Interpreting the Intermediate Bronze Age in Israel/Palestine archaeology” of the 1960s in America (Willey and Sabloff 1974: 183–97), only the overriding evolutionary framework was missing, and that had been more or less assumed in Near Eastern archaeology from the beginning.

Later works concerning the IBA were usually more explicit about their link with processual archaeology. Moreover, they enthusiastically embraced the “missing” attribute diagnosed by Dever—a cultural-evolutionary framework. Discussions about the collapse of the Early Bronze Age and the IBA societal fabric now brimmed with neoevolutionary tribe, chiefdom, and state organizations, as well as “egalitarian” and “ranked” societies, the “evolution” and “devolution” of political systems, and so on (see, e.g., Shay 1983; Palumbo 1987; Dever 1989; 1992). However—and here we return to the IBA drinking vessels that serve as a leitmotif of our essay—there was a dramatic lacuna in the reproduction of “New Archaeology” in IBA research in Palestine. One of the main characteristics of “new” processual archaeology was its strong functionalism, expressed in systems-thinking and adaptive explanations (Hodder 1982: 1– 3; Johnson 1999: 64–79). Within such a conceptual framework, one would expect that the appearance of a great quantity and variety of drinking vessels in the IBA would raise interest in their function within the new social order (“retribalization”) that emerged after the collapse of the Early Bronze Age “city-states” (see Dever 1989 for an archetypal functional and cultural-evolutionary analysis of the EB II–IV). Furthermore, because the “new” archaeologists were concerned with human behavior, they were obsessed with data quantification as a detector of patterning in material culture remains. Similar zeal would have been expected from the adherents of the “New Archaeology” in IBA studies for associating quantitative and qualitative changes in the ceramic repertoire of the post–Early Bronze Age era with new behavioral patterns. This unfortunately did not materialize, because they were satisfied with “ceramic typology in the classic Albrightian manner” (“Editorial,” Bulletin of the American Schools of Oriental Research 237 [1980] 2). Thus, an important indicator for culture change was seemingly overlooked.

Why a Cup? A Postprocessual Appreciation of the “Caliciform Culture” The processual agenda had only barely begun to be realized in the archaeology of Israel/Palestine in the early 1980s, when the postprocessual critique broke upon the scene. Championed by an influential coterie of Cambridge University–bred scholars, the critique challenged the positivist and materialist program that underwrote some of the most visible aspects of processual archaeology—for ex-

27

ample, functionalist-adaptationist approaches to culture, the search for cross-cultural generalizations of human behavior (antihistoricism), a systems approach to ancient cultures, and explanation through hypothesis testing. Instead, postprocessualists emphasized the symbolic role of material culture and its active role in the structuring and reproduction of society and naturalization of ideology, the importance of contingency and agency in the generation of social change, the importance of historical context for understanding past activities as well as present research, the impact of present ideologies on the creation of knowledge, and the value-ladenness of archaeological “data” (Hodder 1982; 1991; 1999; 2002; Shanks and Tilley 1992; Johnson 1999: 98–115). These ideas were very slow in penetrating the fabric of archaeological practice in both North America and Israel/Palestine, especially because they were couched in a rhetoric completely unfamiliar both to staunch processualists and to traditional Near Eastern archaeologists, most of whom led a sheltered existence in the cloisters of text- and philology-oriented academic departments or the basements of the great museums. Many felt threatened by the new agenda, interpreting its call for recognition of the social and political context in which knowledge is produced as an overt challenge to objectivity and to scientific methodology. This was rather unfortunate on several counts. For one, the avoidance of self-understanding hampered the ability of scholars to defend themselves against those who challenged the privileged position of academic scholarship on postmodernist or postcolonialist grounds. For another, their reaction prevented the establishment of a dialogue with other archaeological approaches. Rather than considering it a threat, historians and archaeologists of the ancient Near East should have recognized that the postprocessualist agenda was reintroducing normative archaeology or rehabilitating the treatment of a number of aspects of ancient society that had been shelved by the processualists. These include the role of the individual—or the agent—in the introduction of change, the manner in which migration and interaction across regional borders can influence culture, and the importance of history and cultural context for the interpretation of artifacts and their significance. In dealing with the IBA, a postprocessualist approach would find ample compass in treating questions such as the ideological impact of the Early Bronze Age collapse, the significance of the diverse burial customs practiced, the symbolic dimension of metalcraft, the identification of an IBA habitus, and so on. Surprisingly, such studies have failed to materialize beyond a modest attempt to interpret the cognitive map of IBA pastoralists by, once again, W. G. Dever (2003).

28 The authors of the present essay were both bred within the culture-history paradigm and witnessed the impact of processual archaeology on excavation practice and analysis in Israel/Palestine. Converging on the IBA from research interests principally in the second and third millennia, respectively, we found common ground in a shared interest in the postprocessual critique. Independently, we each found ourselves phrasing similar questions: Why a cup? What practice lies behind the largescale introduction of new drinking paraphernalia into the IBA ceramic repertoire? How did this practice reflect a significant disposition or structure in IBA society? In the concluding section of this contribution, we wish to demonstrate briefly how a postprocessual approach might contribute toward a fuller understanding of this important IBA phenomenon. As noted above, the incomplete realization of the processual agenda in the archaeology of the IBA left some central questions unanswered regarding the pottery of the period: the relative quantity of various types, the possible association of life-style and environment with changes in the makeup of the assemblage, and the function of newly introduced types. In a separate study (Bunimovitz and Greenberg 2004), we argue at length for a significant shift in the composition of the ceramic assemblages during the transition from the EB III to IBA. Large platters, the hallmark of EB III, give way to cups and other drinking vessels in the IBA, testifying to a shift of emphasis in food consumption from feasting to drinking. That having been established, the postprocessual discourse kicks in: why did IBA people choose this specific mode at this specific time, and what element of social interaction is underlined by this practice? Recent work on the social role of drinking, especially of alcoholic beverages, by Sherratt (1986), Dietler (1990), Vencl (1994), Joffe (1998), and Arnold (1999), among others, looks at drink as an important source of symbolic, social, and political capital in ancient societies. Drinking practices are governed by rules and expectations that determine the times and modes of drinking, the age, gender, and status of drinkers, and so on. Convivial consumption of alcoholic drinks plays a key role as a social lubricant, establishing and maintaining social and political relationships, as well as reciprocal obligations among drinkers (Sherratt 1986: 90–91; Dietler 1990: 361; Vencl 1994: 312; Woolf and Eldridge 1994: 327–30). Because of the articulation of drinking with social, economic, and political institutions, changes in these social phenomena will often be reflected in changes in drinking patterns. Conversely, changes in drinking patterns may themselves be implicated in social change, thus serving both as a product and as a structuring principle of the new order. Exotic drinking practices may

Shlomo Bunimovitz and Raphael Greenberg be adopted in order to differentiate groups or classes symbolically within a society or to provide a symbolic link between groups. Thus, if members of one society begin to define their status by association with another society, they may attempt wholesale emulation of the customs of that society (Dietler 1990: 372–79). Following this line of thought, it becomes evident that emulation could be the key to understanding the introduction of the cup and teapot into the IBA ceramic repertoire. These vessels appear in two basic configurations: as “exotic” Black Wheelmade teapots, cups, beakers, and bottles, and as part of the local assemblage in each subregion of Israel/Palestine. Black Wheelmade Ware (sometimes called “Megiddo Ware”) has long been associated with the Painted Simple Ware of EB IV Syria, otherwise known as “caliciform.” The local development of cups and teapots appears to be derived from this initial contact. In Syria, mass production of fine wares is an EB IV innovation and is particularly prominent at important urban sites such as Mardikh and Hama. The emphasis in this highly specialized pottery is on vessels for liquids, including teapots, or for drinking, especially cups and goblets (Mazzoni 1985; 1994; 2002; Jamieson 1993: 52). The development of the standardized fine ware repertoire in Syria has been ascribed by some scholars to the growth of a prosperous and sophisticated urban culture in EB IV (Mazzoni 1985: 1, 12–13; 1994: 245–46; 2003; Carter and Parker 1995: 112). According to this view, specialized fine ware pouring and drinking vessels expressed the importance of formal drinking to the social elite within the context of growing cultural, economic, and political complexity. Highly visible in both palatial and more mundane settings, the fine ware of central Syria and the social ceremony associated with it constituted an accessible cultural package, inviting emulation in less-sophisticated neighboring societies. The putative emulation of Syrian practices in Canaan leads us to suggest a linkage between the increased importance of drinking and the wide-ranging social and ethnic transformations affecting the southern Levant at the end of the third millennium bce. Early Bronze Age Canaanite social order and hierarchy had been based mainly on a palatial/urban elite that used a variety of symbols of power and patronage, including feasting, to manifest and reproduce the existing order. When that structure was rejected and Palestine reverted to tribalism and agropastoralism (Dever 1995: 289–95), a new, alternative system of symbols was needed to express the dominant ideology of the posturban society. To this end, Syrian drinking practices were appropriated by the inhabitants of the southern Levant. This appropriation would have been no simple matter. In the first place, grafting the commodified Syrian cup

Of Pots and Paradigms: Interpreting the Intermediate Bronze Age in Israel/Palestine onto local, small-scale ceramic traditions must have resulted in a product with connotations decidedly different from those in its original context. In the second place, this appropriation required specific agents; cups, as has often been pointed out, did not sprout legs and travel on their own. Because the most accessible venue of ceremonial drinking, foreign to the Early Bronze Age culture of Canaan, was to be found in palatial/urban centers in Syria as early as the 24th century bce, the introduction of elements of Syrian drinking paraphernalia into Canaan in the IBA may be attributed to people—seminomadic pastoralists, perhaps—straddling the interface between Canaan and the urban centers of central Syria (Mazzoni 1985: 13–15; cf. Dever 1970: 145; 1971: 210–20). Their acquaintance with the highly visible drinking behavior that characterized Syria at the beginning of the EB IV and a wish to emulate certain values of the Syrian elite could explain their initial adoption of these practices. Within Canaan, the translation of the Syrian morphemes into local ceramic idioms resulted in the development of a drinking repertoire only generally reminiscent of the Syrian original. The emulation of habits and rank symbols associated with the remote Syrian elite thus became part of a series of local responses to the disintegration of the long-established Early Bronze Age urban system and of the ensuing restructuring of social hierarchies in the period that followed.

29

Conclusions The approach outlined above reveals some of the potential contained in postprocessual or interpretive archaeology for a more nuanced understanding of cultural change in the Levant. Admittedly, our revisiting of the IBA pots and paradigms brought us back to artifacts and issues discussed in the earliest archaeological treatments of this period. Yet this is not “back to square one,” as some skeptics might claim. Although one can hardly escape the conclusion that the paradigms about the IBA are true reflections of their zeitgeist, there can also be no doubt that the field has come a long way since the early days of culture-historical archaeology. Our understanding of key concepts such as culture, diffusion, and migration is, or should be, more robust and multidimensional, as is our reading of material texts. The projects of uncovering the “genetic code” of the IBA habitus, the social and cultural ideologies as embodied in material culture, and the manner in which people constitute knowledgeable agents engaged in the production and reproduction of society all offer a boundless prospect for future research. If taken up, they will surely result in the emergence of a new interpretive framework, integrated into broader anthropological theory and offering a clearer view of the life and deeds of the people of IBA Israel/Palestine.

References Adams, W. Y.; Van Gerven, D. P.; and Levy, R. S. 1978 The Retreat from Migrationism. Annual Review of Anthropology 7: 483–532. Albright, W. F. 1932 The Excavation of Tell Beit Mirsim in Palestine 1: The Pottery of the First Three Campaigns. Annual of the American Schools of Oriental Research 12. New Haven: American Schools of Oriental Research. 1935 Palestine in the Earliest Historical Period. Journal of the Palestine Oriental Society 15: 193–234. 1938 The Excavation of Tell Beit Mirsim 2: The Bronze Age. Annual of the American Schools of Oriental Research 17. New Haven: American Schools of Oriental Research. 1960 The Archaeology of Palestine. Harmondsworth: Penguin. Anthony, D. W. 1990 Migration in Archaeology: The Baby and the Bathwater. American Anthropologist 92: 895–914. Arnold, B. 1999 ‘Drinking the Feast’: Alcohol and the Legitimation of Power in Celtic Europe. Cambridge Archaeological Journal 9: 71–93. Binford, L. R. 1965 Archaeological Systematics and the Study of Culture Change. American Antiquity 31: 203–10. 1972 An Archaeological Perspective. New York: Seminar. Burmeister, S. 2000 Archaeology and Migration. Current Anthropology 41: 539–67.

Bunimovitz, S. 1995 How Mute Stones Speak: Interpreting What We Dig Up. Biblical Archaeology Review 21/2: 58–67, 96–100. Bunimovitz, S., and Greenberg, R. 2004 Revealed in Their Cups: Syrian Drinking Customs in Intermediate Bronze Age Canaan. Bulletin of the American Schools of Oriental Research 334: 19–31. Carter, E., and Parker, A. 1995 Pots, People and the Archaeology of Death in Northern Syria and Southern Anatolia in the Latter Half of the Third Millennium bc. Pp. 96–116 in The Archaeology of Death in the Ancient Near East, ed. S. Campbell and A. Green. Oxbow Monographs 51. Oxford: Oxbow. Chapman, J., and Hamerow, H., eds. 1997 Migrations and Invasions in Archaeological Explanation. British Archaeological Reports International Series 664. Oxford: British Archaeological Reports. Childe, V. G. 1929 The Danube in Prehistory. Oxford: Oxford University Press. Cohen, R., and Dever, W. G. 1978 Preliminary Report of the Pilot Season of the “Central Negev Highlands Project.” Bulletin of the American Schools of Oriental Research 232: 29–45. 1979 Preliminary Report of the Second Season of the “Central Negev Highlands Project.” Bulletin of the American Schools of Oriental Research 236: 41–60.

30 Collingwood, R. G. 1939 An Autobiography. Oxford: Oxford University Press. Dever, W. G. 1970 The “Middle Bronze I” Period in Syria and Palestine. Pp. 132– 63 in Near Eastern Archaeology in the Twentieth Century: Essays in Honor of Nelson Glueck, ed. J. A. Sanders. Garden City, NY: Doubleday. 1971 The Peoples of Palestine in the Middle Bronze I Period. Harvard Theological Review 64: 197–226. 1973 The EB IV–MB I Horizon in Transjordan and Southern Palestine. Bulletin of the American Schools of Oriental Research 210: 37–62. 1980 New Vistas on the EB IV (“MB I”) Horizon in Syria–Palestine. Bulletin of the American Schools of Oriental Research 237: 35– 64. 1981 The Impact of the “New Archaeology” on Syro-Palestinian Archaeology. Bulletin of the American Schools of Oriental Research 242: 15–29. 1985 From the End of the Early Bronze Age to the Beginning of the Middle Bronze Age. Pp. 113–35 in Biblical Archaeology Today: Proceedings of the International Congress on Biblical Archaeology, Jerusalem, April 1984, ed. J. Amitai. Jerusalem: Israel Exploration Society. 1989 The Collapse of the Urban Early Bronze Age in Palestine: Toward a Systemic Analysis. Pp. 225–46 in L’urbanisation de la Palestine à l’âge du Bronze ancien, ed. P. de Miroschedji. British Archaeological Reports International Series 527. Oxford: British Archaeological Reports. 1992 Pastoralism and the End of the Urban Early Bronze Age in Palestine. Pp. 83–92 in Pastoralism in the Levant: Archaeological Materials in Anthropological Perspectives, ed. O. Bar-Yosef and A. Khazanov. Monographs in World Archaeology 10. Madison: Prehistory Press. 1995 Social Structure in the Early Bronze Age IV Period in Palestine. Pp. 282–96 in The Archaeology of Society in the Holy Land, ed. T. E. Levy. London: Leicester University Press. 2003 The Rural Landscape of Palestine in the Early Bronze IV Period. Pp. 43–59 in The Rural Landscape of Ancient Israel, ed. A. M. Maeir, S. Dar, and Z. Safrai. British Archaeological Reports International Series 1121. Oxford: Archaeopress. Dietler, M. 1990 Driven by Drink: The Role of Drinking in the Political Economy and the Case of Early Iron Age France. Journal of Anthropological Archaeology 9: 352–406. Fargo, V. 1979 Settlement in Southern Palestine during Early Bronze III. Unpublished Ph.D. dissertation, University of Chicago. Guy, P. L. O., and Engberg, R. M. 1938 Megiddo Tombs. Oriental Institute Publications 33. Chicago: University of Chicago Press. Hodder, I. 1982 Theoretical Archaeology: A Reactionary View. Pp. 1–16 in Symbolic and Structural Archaeology, ed. I. Hodder. Cambridge: Cambridge University Press. 1991 Reading the Past. 2nd ed. Cambridge: Cambridge University Press. 1999 The Archaeological Process: An Introduction. Oxford: Blackwell. 2002 Archaeological Theory. Pp. 77–90 in Archaeology: The Widening Debate, ed. B. Cunliff, W. Davies, and C. Renfrew. Oxford: Oxford University Press. Jamieson, A. S. 1993 Euphrates Valley and Early Bronze Age Ceramic Traditions. Abr-Nahrain 31: 36–92.

Shlomo Bunimovitz and Raphael Greenberg Joffe, A. H. 1998 Alcohol and Social Complexity in Ancient Western Asia. Current Anthropology 39: 297–322. Johnson, M. 1999 Archaeological Theory: An Introduction. Oxford: Blackwell. Jones, S. 1997 The Archaeology of Ethnicity. London: Routledge. Kamp, K., and Yoffee, N. 1980 Ethnicity in Western Asia during the Early Second Millennium b.c.: Archaeological Assessments and Ethnoarchaeological Prospectives. Bulletin of the American Schools of Oriental Research 237: 85–103. Kenyon, K. M. 1957 Digging Up Jericho. London: Benn. 1971 Syria and Palestine, c. 2160–1780: The Archaeological Sites. Pp. 567–94 in Early History of the Middle East, ed. I. E. S. Edwards, C. J. Gadd, and N. G. L. Hammond. 3rd ed. Volume 1/2 of Cambridge Ancient History. Cambridge: Cambridge University Press. 1979 The Archaeology of the Holy Land. 4th ed. London: Benn. Knapp, A. B. 1996 Archaeology without Gravity: Postmodernism and the Past. Journal of Archaeological Method and Theory 3: 127–58. Kochavi, M. 1967 The Settlement of the Negev in the Middle Bronze (Canaanite) I Age. Unpublished Ph.D. dissertation, Hebrew University of Jerusalem. Kuhn, T. 1970 The Structure of Scientific Revolutions. Chicago: University of Chicago Press. Lapp, P. 1966 The Dhahr Mirzbaneh Tombs: Three Intermediate Bronze Age Cemeteries in Jordan. Publications of the Jerusalem School 4. New Haven: American Schools of Oriental Research. Lyman, R. L.; O’Brien, M. J.; and Dunnel, R. C. 1997 The Rise and Fall of Culture History. New York: Plenum. Mazzoni, S. 1985 Elements of the Ceramic Culture of Early Syrian Ebla in Comparison with Syro-Palestinian EB IV. Bulletin of the American Schools of Oriental Research 257: 1–18. 1994 Drinking Vessels in Syria: Ebla and the Early Bronze Age. Pp. 245–76 in Drinking in Ancient Societies: History and Culture of Drinks in the Ancient Near East. Papers of a Symposium Held in Rome, May 17–19, 1990, ed. L. Milano. Padua: Sargon. 2002 The Ancient Bronze Age Pottery Tradition in Northwestern Central Syria. Pp. 69–96 in Céramique de l’âge du Bronze en Syrie I: La Syrie du Sud et la vallée de l’Oronte, ed. M. alMaqdissi, V. Matoïan, and C. Nicolle. Beirut: IFAPO. 2003 Crafts and Power in an Emergent State of 3rd Millennium Syria. Journal of Mediterranean Archaeology 16: 173–91. McDonald, W. A., and Rapp, G. R., eds. 1972 The Minnesota Messenia Expedition: Reconstructing a Bronze Age Regional Environment. Minneapolis: University of Minnesota Press. Moorey, P. R. S. 1991 A Century of Biblical Archaeology. Cambridge: Lutterworth. Oren, E. D. 1973a The Northern Cemetery of Beth Shan. Leiden: Brill. 1973b The Early Bronze IV Period in Northern Palestine in Its Cultural and Chronological Setting. Bulletin of the American Schools of Oriental Research 210: 20–37. Palumbo, G. 1987 “Egalitarian” or “Stratified” Society? Some Notes on Mortuary Practices and Social Structure at Jericho in EB IV. Bulletin of the American Schools of Oriental Research 267: 43–59.

Of Pots and Paradigms: Interpreting the Intermediate Bronze Age in Israel/Palestine Patterson, T. C. 1995 Towards a Social History of Archaeology in the United States. Fort Worth: Harcourt Brace. Prag, K. 1974 The Intermediate Early Bronze–Middle Bronze Age: An Interpretation of the Evidence from Transjordan, Syria and Lebanon. Levant 6: 69–116. Redman, C. L. 1991 Distinguished Lecture in Archaeology: In Defense of the Seventies—The Adolescence of New Archaeology. American Anthropologist 93: 295–307. Renfrew, C., and Bahn, P. 1991 Archaeology: Theories, Methods, and Practice. London: Thames & Hudson. Richard, S. 1980 Toward a Consensus of Opinion on the End of the Early Bronze Age in Palestine–Transjordan. Bulletin of the American Schools of Oriental Research 237: 5–34. 1987 The Early Bronze Age: The Rise and Collapse of Urbanism. Biblical Archaeologist 50: 22–43. Schortman, E. M., and Urban, P. A., eds. 1992 Resources, Power, and Interregional Interaction. New York: Plenum. Shanks, M., and Tilley, C. 1992 Re-Constructing Archaeology. 2nd ed. London: Routledge. Shay, T. 1983 Burial Customs at Jericho in the Intermediate Bronze Age: A Componential Analysis. Tel Aviv 10: 26–37. Sherratt, A. 1986 Cups That Cheered. Pp. 81–106 in Bell Beakers of the Western Mediterranean, ed. W. H. Waldren and R. C. Kennard. British Archaeological Reports International Series 331. Oxford: British Archaeological Reports.

31

Silberman, N. A. 1995 Power, Politics and the Past: The Social Construction of Antiquity in the Holy Land. Pp. 7–23 in The Archaeology of Society in the Holy Land, ed. T. E. Levy. London: Leicester University Press. Tadmor, M. 1978 A Cult Cave of the Middle Bronze I near Qedesh. Israel Exploration Journal 28: 1–30. Trigger, B. G. 1980a Archaeology and the Image of the American Indian. American Antiquity 45: 662–76. 1980b Gordon Childe: Revolutions in Archaeology. London: Thames & Hudson. 1989 A History of Archaeological Thought. Cambridge: Cambridge University Press. Tufnell, O. 1958 Lachish IV: The Bronze Age. London: Oxford University Press. Vencl, S. 1994 The Archaeology of Thirst. Journal of European Archaeology 2: 299–326. Willey, G. R., and Sabloff, J. A. 1974 A History of American Archaeology. San Francisco: Freeman. Woolf, A., and Eldridge, R. 1994 Sharing a Drink with Marcel Mauss: The Uses and Abuses of Alcohol in Early Medieval Europe. Journal of European Archaeology 2: 327–40. Wright, G. E. 1938 The Chronology of Palestinian Pottery in Middle Bronze I. Bulletin of the American Schools of Oriental Research 71: 27–34.

A Decorated Ivory Lid from Tel Miqne–Ekron Trude Dothan Institute of Archaeology, The Hebrew University of Jerusalem

In this essay, I discuss an elaborate, round, ivory pyxis lid found at Tel Miqne–Ekron, decorated in an unmistakably Aegean style with scenes depicting battles between various animals: a griffin, a lion, and bulls. The lid was excavated in Field IV, the elite area at Tel Miqne–Ekron and no doubt its administrative center (fig. 1). Well-planned structures of large dimensions were uncovered in this area, undoubtedly palaces or temples. The ivory lid was found in the fill between two monumental structures (fig. 2): Building 351 of Stratum VI, dating to the fourth quarter of the 12th century/beginning of the 11th century bce, and the foundations of Building 350 of Stratum V, dating to the first half of the 11th century bce (T. Dothan and M. Dothan 1992: 245–46; T. Dothan and Gitin 1993: 1055–56; 1994: 13–14). Building 351, a well-planned mud-brick structure, was partially destroyed by the construction of the overlying Building 350. Most of the area of Building 351 was occupied by a large hall (10 x 8 m), with a number of smaller rooms to its east. The mud bricks of the walls were laid as stretchers, and traces of white plaster were still evident. The beaten-earth floor was covered with layers of black ash and potsherds. The only indication of the hall’s function was a number of large open basins, thick fragments of which were unearthed on the floor. A white-plastered niche with steps leading up to it was uncovered in the southwestern corner of the hall; the niche probably had a cultic function. One of the small rooms in the eastern part of the structure contained many Philistine Bichrome vessels and some Mycenaean IIIC:1 ware, attributed to the fourth quarter of the 12th century/beginning of the 11th century bce, con-

Author’s note: The joint Tel Miqne–Ekron Excavation and Publications projects are directed by Trude Dothan of the Hebrew University of Jerusalem and Seymour Gitin of the W. F. Albright Institute of Archaeological Research, Jerusalem. I would like to thank Tamar Nachmias-Lotan for her assistance in the scientific research and in preparing this article for publication; Anna de Vincenz for collecting additional material on the subject; Yosef Garfinkel, the field archaeologist in charge of the Field IV Iron Age I excavations; and Celia Bergoffen, the supervisor of Square NW.9, where the object was found. Figures 1–2 are by J. Rosenberg, fig. 3 by Ilan Sztulman, and figs. 4–5, and 7–11 by Sarah Halbreicht.

temporaneous with Stratum VI. The ivory lid discussed in this essay was uncovered in this area, as were a zoomorphic clay vessel and a miniature pomegranate made of bone. Several additional ivory objects were found in Field IV, mostly in Building 350 of Strata V–IVA. These include an earlobe plug and a figurine head (T. Dothan 1989: 11, fig. 1.9) and inlays. Ivory items were also found in Iron I strata in Field III, including birds’ heads (presumably belonging to cosmetic boxes) and bands incised with Egyptian-style decorations, such as the “swimming girl” motif. Particularly elaborate ivories were found in the destruction layer of the Iron Age II Philistine city (Field IV), in the precinct of Temple complex 650 and its environs. The Ekron royal dedicatory inscription was unearthed in the sanctuary of this building (Gitin, T. Dothan, and Naveh 1997). 1

The Ivory Lid The lid (Object No. 3176; Locus 9060; Basket No. IV.NW.9.472) is round but not entirely symmetrical (figs. 3–5). Due to the shape of the tusk, the lid’s diameter changes slightly, ranging from 12.4 to 13.3 cm. The lid is flat, 0.3 cm thick. It presumably belonged to a pyxis (a cylindrical box with a flat round base and a separate lid), a vessel common on Cyprus, in the Aegean, and in the Levant (see below). 1. The auxiliary rooms behind the sanctuary contained hundreds of vessels—many complete and others restorable—as well as silver, gold, and bronze objects and a large number of ivory fragments. In addition, several ivory objects were found: a female figurine, an ivory knob with a cartouche of Ramesses VIII, a cylinder seal with a Horus motif, a carved figurine head (the largest ivory item found to date in Israel), an ivory wheel with a bronze pin, and an elephant tusk, which probably served as the body of a headless figurine. The tusk bears a relief of a princess or goddess on one side and a carved cartouche of Merneptah on the other. Other unique finds are a gold Egyptian uraeus (23 cm long) and a Ptahpatecus amulet. The Egyptian objects were part of the temple assemblage from the final phase of the city (Gitin, T. Dothan, and Naveh 1997: 7–8). They may have been curated, originating in the Late Bronze Age, when Egypt’s influence in Philistia was still strong, but it is more likely that they represent booty plundered by the Assyrians from Egyptian temples or mortuary temples during the first half of the 7th century bce (Gitin 2003: 60*).

Trude Dothan

34

Fig. 3. Photograph of ivory lid.

Fig. 1. Topographical plan of Tel Miqne–Ekron, showing excavation areas.

Fig. 4. Drawing of ivory lid. Fig. 2. Plan of Stratum VI Building 351, showing findspot of ivory lid.

The perimeter of the lid is incised with five concentric circles, probably executed with a compass. In the center is an eight-petalled rosette (2.6 cm in diameter), with four additional concentric circles encompassing it. Between the two groups of circles is a wide area (4–4.3 cm in diameter) decorated with an animal combat scene featuring a lion, a griffin, and two bulls.

A round hole was drilled close to the perimeter, perforating the griffin’s hind leg. The hole was used for a pivot, which attached the lid to the object and permitted it to rotate. The rotation marks are at an angle of 180°, suggesting that the lid did not open completely. According to François Poplin of the Louvre Museum, Paris, who examined the lid, it is made of elephant-tusk ivory (personal communication).

A Decorated Ivory Lid from Tel Miqne–Ekron

Fig. 5. Painted drawing of ivory lid.

Ivory-Working and Decoration Techniques Elephant tusks are the animal’s upper incisors, which continue to grow throughout its lifetime. One-third of the tusk at the proximal end, closest to the elephant’s head, is the hollow pulp cavity (fig. 6). This part was used for the body of a vessel, such as our ivory pyxis. The base and lid were produced from the external part of the tusk, which is covered with hard enamel. The enamel was scaled off by filing, grinding, or rubbing it with an abrasive, such as sandstone or the like. The carving and incision of the decoration followed the removal of this layer. Elephant tusks are composed of dentine and calcified connective tissue. As mentioned above, the tusk is in a state of perpetual growth, with the dentine and calcified connective tissue forming a “cone-in-cone” pattern. As a result, longitudinal grooves are evident in the tusk’s long section, while the cross-section has oval rings (Krzyszkowska 1990: 33–34). The decoration may involve any of several techniques or combinations thereof: carving in high relief, carving in low relief, and/or incision. The lid from Ekron was decorated with incised lines: the ivory worker “drew” the design on the surface by means of a sharp instrument, such as a chisel. The concentric circles were incised with a compass (fig. 3). After the incised decoration of the various areas was completed, the undecorated base was connected to the box by means of ivory pegs. Two methods were used to attach the lid to the vessel if it was not left loose (as on a pyxis from Athens [Kantor 1960: fig. 2]): (1) it could be attached to the handle by

35

Fig. 6. The three parts of an elephant tusk (right, Caubet and Poplin 1987: fig. 16; center and left, Krzyszkowska 1990: fig. 12).

means of a central perforation (as on a pyxis from Palaepaphos [Maier and Wartburg 1985: pl. 14:8]); or (2) it could be attached to the vessel by means of a lateral perforation (as on the pyxis from Ekron). The Decorative Motifs As mentioned above, the center of the lid is incised with a rosette inside four concentric circles, and five additional concentric circles appear on the perimeter. Room was left between the two groups of concentric circles for the animal combat scene, an endless scene in torsion, corresponding to the lid’s circular shape (figs. 3–5). The designs and the use of space are in the Aegean tradition, which is characterized by a vivid depiction of the animals’ motion and momentum (Kantor 1956: 167). However, despite the excellent execution of the design, it appears that the craftsperson who incised the decoration did not plan the division of the space accurately; hence, the griffin, sandwiched between the two bulls, is poorly executed. The lion (fig. 7). This animal is depicted in a “flying gallop,” with its forelegs stretched forward and its hind legs extending backward. Its head is turned backward, toward the bull, and its jaws are open. The muscles of its forelegs are highlighted by oval lines. While the hind legs are extended outward, parallel to one another, the forelegs form an inward revolving motion, somewhat like an “embrace.” It seems that the craftsperson used the model of a lion grasping a hunted animal without adapting it to the scene in question. The lion’s tail is trapped between the

36

Trude Dothan

Fig. 7. Depiction of lion on Ekron ivory lid.

Fig. 8. Depiction of bull on Ekron ivory lid.

Fig. 11. Drawing of woman carrying pyxis from wall fresco at Tiryns (after Rodenwaldt 1912: pl. 8).

Fig. 9. Depiction of bull on Ekron ivory lid.

Fig. 10. Depiction of griffin on Ekron ivory lid.

hind legs and is barely discernible. The lion has typical Mycenean characteristics, such as the pouncing stance, the flying posture, and the mane, which is stylized with a flame pattern (Kantor 1960: 21). However, other typical Mycenean attributes are absent, such as the plait that is often braided along the stomach and the hind part and the whorl of hair on the shoulder (Kantor 1956: 21). The bulls (figs. 8–9). The two bulls are in ramming positions. One is standing firm, its forelegs on the ground and its head bent in an attacking position, with both horns on the griffin. The bull’s head and horns are combined with the depiction of the griffin and are incised on its back, and the bull’s tail is hidden between its legs (fig. 8). The other bull is attacking the lion and is depicted with forelegs extended forward and horns (one of which is not vis-

A Decorated Ivory Lid from Tel Miqne–Ekron

37

Fig. 12. Photograph of pyxis from Athens-Aerophagus, lid and full design (Barnett 1982: pl. 29).

ible) embedded in the lion, which, in turn, is leaping over it (fig. 9). The genitalia of both bulls are pronounced, and their hind legs are depicted with oval lines highlighting their muscles. The bulls’ snouts are highlighted with meshed lines, and their shoulder blades are horn-shaped. The griffin (fig. 10). This creature is depicted in a “flying posture,” sandwiched between the two bulls and being attacked by one of them. One of its wings is spread out frontally. One foreleg is placed on the neck of a bull, while the other is bent forward, its hind part on the back of the other bull. The griffin is turning backward in an unnatural manner. Its head, held high, is depicted in profile, with an open beak—a well-known portrayal of griffins in battle. The griffin faces the lion, and similarly creates an “embracing” movement with its legs. The execution of the griffin is poor: the feathers of one wing are schematic, depicted with a broken line, and the other wing is not clearly defined. The body appears to be divided in half, and only a close examination reveals the connection between the two parts. The two halves of the body are disproportionate in size: the rear part, depicted in a “flying gallop,” is much smaller than the front. There is a perforation in the rear part of the griffin, through which the lid’s axis was inserted. Despite the faulty spatial planning, this griffin recalls the superbly executed griffin depicted on the pyxis from Aerophagus (Kantor 1960: 16).

Fig. 13. Pyxis lid from Palaepaphos, Evreti (Maier and Wartburg 1985: fig. 9).

Fig. 14. Pyxis lid from Palaepaphos, Kouklia (Maier and Wartburg 1985: fig. 10).

Trude Dothan

38

Fig. 15. Depiction of attacking griffin on ivory from Megiddo (Kantor 1956: fig. 3c).

Fig. 16. Depiction of attacking lion on ivory from Megiddo (Kantor 1956: fig. 3b).

Other Ivory Pyxides and Lids A reconstructed wall fresco from Tiryns (Rodenwaldt 1912: pl. 8) depicts a woman carrying a decorated ivory pyxis (fig. 11). A few complete pyxides have been found in tombs in Greece, such as the example from the burial chamber at Aerophagus, Athens (fig. 12), dated to the 14th century bce (Shear 1940; Barnett 1982: 37, pl. 29a–c); at Menidi, Attica, dated to the Late Mycenaean IIIB (Barnett 1982: 37, pl. 30f); and at Myrsinochorion-Routsi, near Pylos, dated to the same period (Marinatos 1960: fig. 223). Pyxis lids have been uncovered in various archaeological excavations in the eastern Mediterranean basin—in the Aegean and the Levant—including the well-known lid bearing a relief of the Minoan animal goddess found at Mina al-Baida (the port of Ugarit) (Kantor 1960: 86). The closest parallel to the lid from Ekron was unearthed during the excavations directed by Maier at Late Bronze Age Palaepaphos (Evreti and Kouklia), depicting a battle between a lion and a bull, which is dated to Late Cypriote IIIA (Maier, Beckedorf, and Gesche 1969: 405, fig. 34; Maier and Wartburg 1985: 112, fig. 9) (see here fig. 13). An assemblage of fragments of ivory pyxis lids dating to the same period was found in wells (not burials) at Kouklia, Cyprus. These are carved with scenes of animal battles, depicting a lion and bull, a creature that is part eagle and part griffin perched on a tree (fig. 14), and a bull and griffin (Maier and Wartburg 1985: figs. 9–19). They clearly belong to a group that combines Aegean elements with Eastern characteristics, and point to a flourishing ivory industry in Late Bronze Age Cyprus.

Fig. 17. Depiction of lion attacking bull on ivory from Megiddo (Kantor 1956: fig. 3d).

Parallels for the Decoration A close parallel for the decorative design on the pyxis lid from Ekron comes from the Stratum VIIA treasury at Megiddo, which went out of use in the first half of the 12th century bce (on the basis of a cartouche of Ramesses III on a carved ivory pen box) (Loud 1939: 11, pl. 62, no. 377). The Megiddo ivories bear depictions of a griffin (fig. 15) and a lion (fig. 16) in confrontational postures, as well as a lion attacking a bull (fig. 17) (Kantor 1956: fig. 3b–d; see also Loud 1939: pls. 5–7). The Megiddo griffins are depicted in typical 13th-century bce Aegean style and are considered to be the latest examples in the Aegean world. The rectangular ivory plaques from Megiddo, also typically Aegean in both style and depiction (Kantor 1956: 169–70), are not discussed in this essay. Two ivory plaques depicting animal battles from the Temple of Artemis at Delos are examples of the “coexistence” of different cultural elements: one plaque (fig. 18) depicts a bull attacking a lion, while the other (fig. 19) depicts a lion attacking a griffin (Kantor 1956: figs. 2a–2b; Kantor 1960: figs. 10a–10b). These plaques are dated to the Late Helladic IIIB. All the elements on the former, including the trefoil pattern, are Aegean, except for the whorl on the lion’s back, an Eastern element that first appeared in the western Mediterranean basin during the second millennium bce and was incorporated into Aegean art (Kantor 1960: 21). The animal battle is a typical motif in Western art, and presumably reached Cyprus via Crete. The animals’ vigorous and assertive character is based on Aegean prototypes

A Decorated Ivory Lid from Tel Miqne–Ekron

39

Fig. 18. Depiction of bull attacking lion on plaque from Delos (Kantor 1956: fig. 2a).

Fig. 19. Depiction of lion attacking griffin on plaque from Delos (Kantor 1956: fig. 2b).

(Kantor 1960). The “flying gallop,” a decorative element in Aegean art, was popular in the second millennium bce in Egypt and in western Asia (Kantor 1947: 93, 97, 106). Lions resembling that on the Ekron lid are depicted on ivories from Delos (Kantor 1960: fig. 10), Palaepaphos (Barnett 1982: pl. 33c) and Megiddo (Loud 1939: pls. 5–7). Bulls are a common motif in the Mycenaean world: examples are known on ivory pyxis lids from Kouklia (Poursat 1977: pl. 15:5–6), on a gold cup from Vaphio, Laconia (Marinatos 1960: 183), and on a faience rhyton from Kition (Karageorghis 1982: 65). Griffins frequently appear in depictions of animal battles, for example, on an ivory pyxis lid from Delos (Poursat 1977: pl. 12:1), as well as on a mirror handle from Enkomi (Poursat 1977: pl. 16:3). A solitary griffin is depicted on a Mycenaean plaque (Barnett 1982: pl. 31c). Bulls also occur by themselves, for example on a plaque from Megiddo (Poursat 1977: pl. 18:1).

Conclusions The closest parallels to the ivory lid from Ekron are the incised lids from Palaepaphos, Cyprus, which resemble our object in terms of decorative theme, combination of motifs, and design.

The ivory industry flourished in Late Bronze Age Cyprus. Ivory-carving workshops dating to ca. 1200 bce excavated at Kouklia and Kition yielded numerous ivories with Aegean and Eastern characteristics, and it seems that they exported their products throughout the Mediterranean basin. The lids bearing scenes of animal combat are reminiscent of plaques from Delos and Megiddo in terms of both incising techniques and motifs. The lid from Ekron discussed in this essay is a prime example of this variegated tradition. The excavations of Iron I Ekron have revealed a large, fortified city encompassing both the earlier Canaanite upper city of ca. 40 dunams and the lower Iron I city of ca. 160 dunams, providing evidence of a well-organized society with a strong economic structure. This is compatible with our assumptions regarding the military supremacy and highly developed technological abilities of the “Sea Peoples.” The material culture from Ekron, Ashkelon, and Ashdod indicates a high level of artistic expression evident in architectural elements, cultic installations, and the ceramic repertoire, and a material culture emanating from the Aegean world. Nevertheless, the archaeological record is not compatible with a either sharp transition or a uniform,

Trude Dothan

40 homogeneous change. Instead, it suggests a broader picture of the coexistence of diverse Canaanite, Egyptian, Israelite, and Philistine elements, which is characteristic of

the transition from the Late Bronze Age to Iron Age I in Israel.

References Barnett, R. D. 1982 Ancient Ivories in the Middle East. Qedem 14. Jerusalem: Hebrew University. Caubet, A., and Poplin, F. 1987 Les objets de matière dure animale: Étude des materiaux. Pp. 273–308 in Ras Shamra–Ougarit III: Le centre de la ville, 38 e–44 e campagnes (1978–1984), ed. M. Yon. Paris: Association pour la Diffusion de la Pensée Française. Dothan, T. 1989 The Arrival of the Sea Peoples: Cultural Diversity in Early Iron Age Canaan. Pp. 1–22 in Recent Excavations in Israel: Studies in Iron Age Archaeology, ed. S. Gitin and W. G. Dever. Annual of the American Schools of Oriental Research 49. Winona Lake, IN: Eisenbrauns. Dothan, T., and Dothan, M. 1992 People of the Sea: The Search for the Philistines. New York: Macmillan. Dothan, T., and Gitin, S. 1993 Tel Miqne/Ekron. Pp. 1051–59 in vol. 3 of The New Encyclopedia of Archaeological Excavations in the Holy Land, ed. E. Stern. Jerusalem: Israel Exploration Society. 1994 Tel Miqne/Ekron: The Rise and Fall of a Philistine City. Qadmoniot 105–6: 2–28 (Hebrew). Gitin, S. 2003 Neo-Assyrian and Egyptian Hegemony over Ekron in the Seventh Century bce: A Response to Lawrence E. Stager. EretzIsrael 27 (Hayim and Miriam Tadmor volume): 55*–61*. Gitin, S.; Dothan, T.; and Naveh, J. 1997 A Royal Dedicatory Inscription from Ekron. Israel Exploration Journal 47: 1–16.

Kantor, H. J. 1947 The Aegean and the Orient in the Second Millennium b.c. Bloomington, IN: Principia. 1956 Syro-Palestinian Ivories. Journal of Near Eastern Studies 15: 153–74. 1960 Ivory Carving in the Mycenaean Period. Archaeology 13: 14– 25. Karageorghis, V. 1982 Cyprus from the Stone Age to the Romans. London: Thames & Hudson. Krzyszkowska, O. 1990 Ivory and Related Material: An Illustrated Guide. Institute of Classical Studies Bulletin Supplement 59. London: Institute of Classical Studies. Loud, G. 1939 The Megiddo Ivories. Chicago: University of Chicago Press. Maier, F. G.; Beckedorf, H.; and Gesche, H. 1969 Die Brunnen von Evreti. Archäologischer Anzeiger 84: 398– 406. Maier, F. G., and Wartburg, M. L. 1985 Excavations at Kouklia (Palaepaphos)—Thirteenth Preliminary Report: Seasons 1983 and 1984. Report of the Department of Antiquities, Cyprus 1985: 100–25. Marinatos, S. 1960 Crete and Mycenae. New York: Abrams. Poursat, J. C. 1977 Les ivoires myceniens. Athens: École française d’Athénes. Rodenwaldt, G. 1912 Tiryns II. Athens: Eleutheroudakis & Barth. Shear, T. L. 1940 The Campaign of 1939: The Mycenaean Tomb. Hesperia 9: 274–93, figs. 27–31.

Har Resisim 126: An Ephemeral Early Bronze Age Site Barry M. Gittlen Baltimore Hebrew University

Introduction The excavation of the small Early Bronze Age settlement of Har Resisim 126 took place during the final season (1980) of the Central Negev Highlands Project. 1 Somewhat different in environmental setting, architecture, and ceramic repertoire from other Early Bronze Age settlements in southern Israel, Har Resisim 126 poses unique problems of interpretation. The excavation of seemingly similar EB II settlements in south-central Sinai and the central Negev revealed links between Har Resisim 126 and these sites. Nevertheless, Har Resisim 126 does not appear to be an industrial or commercial site like the sites in Sinai, nor does it appear to be a station on a known route from Canaan proper to these Sinai settlements. This study presented in honor of William G. Dever clarifies the character and purpose of the settlement at Har Resisim 126, places it within the chronological framework of the Early Bronze Age, and defines its relationships to its environment and to other settlements in the Negev and Sinai.

Har Resisim 126 The small Early Bronze Age village that the Beªer Resisim survey discovered in 1979, 1 km south of Beªer Resisim, was listed as Har Resisim Site 126 (map reference 1094–0193). The site sits at 525 m above sea level on the southwestern protrusion of a broad plateau overlooking 1. The Beªer Resisim regional archaeological survey was conducted by Jerome B. Schaefer of the University of Arizona. For preliminary reports on this survey and on Har Resisim 126, see Cohen and Dever 1979: 41–46; 1981: 70–71 and the map fig. 2 on p. 59; Cohen 1999: 63, fig. 45. The Central Negev Highlands Project was a joint venture of the University of Arizona and the Israel Department of Antiquities (now the Israel Antiquities Authority) under the shared direction of R. Cohen and W. G. Dever. I participated in the 1978 season and was invited by Dever to direct the excavation of Har Resisim 126 in the 1980 season. I supervised a nine-member team that included assistant supervisors Robin Cover and Robert Hodgman. My participation was made possible by a grant from the Herman and Rosa L. Cohen Fund of Baltimore Hebrew University.

the Nahal Nissana far below. The surface of the plateau has probably changed little since the village was founded. It consists of flat, shelf-like outcrops of the limestone bedrock, especially prominent at the fringes of the settlement, with an accumulated wind-blown loess soil deposit ranging in depth from 1 to at least 50 cm in the central parts of the settlement area. In the areas in which the loess accumulation is shallow, natural flint scatters darken the surface of the plateau. Despite the stark character of the environment, there seems to be ample vegetation for the grazing of sheep or goats, 2 not only on the plateau itself, but also along the myriad ridges descending to the Nahal Nissana floor— ridges that are today laced with goat paths. Clusters of desert shrubs today dot the eastern parts of the plateau and its slopes, but no water source or water-catchment installation could be identified. The absence of water resources in this area of 100 mm average annual rainfall suggests a temporary, impermanent settlement. The 1979 survey of Har Resisim 126 identified a cluster of several circular structures and recovered a few EB IV sherds and one EB II ledge handle, all unpublished. Despite the apparent similarity to the ceramics found at Beªer Resisim, the site’s different architecture, size, and topographical location led to the decision to excavate Har Resisim 126 in the summer of 1980. The project team sought to define the character of Har Resisim 126, ascertain the chronology of its settlement, and understand the relationship between Beªer Resisim and Har Resisim 126. Upon closer investigation, the Har Resisim 126 settlement was found to cover an area of 90 m north–south and 55 m east–west (fig. 1). 3 The excavations exposed four small, shallow, rectangular cairns (and another three possible cairns located 50 m east of the settlement), all in disturbed condition; five simple, single-room, circular structures; two large, multiroom buildings; one enigmatic 2. See Ron Gerson’s preliminary report in Cohen and Dever 1981: 73. 3. Figure 1 was hastily drawn in the field and does not include all the features discussed; nor does it have the accuracy of the individual building plans.

Barry M. Gittlen

42 row of stones; and a series of petroglyphs and more-recent inscriptions. As the excavation of Har Resisim 126 proceeded, it became clear that neither its architecture nor its ceramics were identical to those found at Beªer Resisim. Indeed, fundamental differences between the two sites led to the conclusion that Har Resisim 126 was not an agricultural village contemporary with EB IV Beªer Resisim. The excavated data indicate that Har Resisim 126 was instead a shepherd’s village or pastoral encampment settled prior to Beªer Resisim and abandoned before the latter became a flourishing EB IV settlement.

The Structures Building 1 (figs. 1–2) The walls of roughly circular Building 1 were built of large limestones set mostly on soil above bedrock, except for its southern projection, which was based on a shelf of bedrock. Oriented north–south, the structure’s internal north–south diameter was 4.88 m (3.75m without the southern projection), and its east–west diameter was 4.50 m. The entryway into Building 1 was impossible to determine precisely, but it seems to have been in the center of the northern side of the wall. The stones of the wall, a single course of which was found in situ, range in size from 22 x 15 x 7 to 95 x 50 x 30 cm (averaging 49 x 34 x 19 cm). The stones of the circular wall were chinked in position on the bottom and sides by small pebbles. Whether the mud deposited between the stones resulted from entirely natural processes following the construction of the walls or represents the remains of the builders’ mud plaster together with later deposition remains unresolved. There were not enough stones in the ruins either inside or outside the structure to postulate walls originally over 1 m (2–3 courses) in height. The characteristics of the Building 1 wall parallel the characteristics of all the walls of the Har Resisim single-room and multiroom buildings presented below. The question of the superstructure of this building, like that of all of the others at Har Resisim 126, remains unsolved. The ruins lacked large flat stones that could be postulated as roofing materials. 4 We found no indication of a central pillar (whether of wood or stone) to support a hypothetical wooden-beam ceiling—the structure is too large to be spanned by a single beam. Although direct evidence thereof is lacking, the structure may have been roofed with woven tent-cloth, animal skins, or branches and mud, or perhaps it remained unroofed.

4. As at Beªer Resisim (Cohen and Dever 1981: 58–61 and fig. 7 on p. 64.

The living surface of Building 1 was earth (Soil 1010), upon which its stone wall (Wall 1006) was built (no foundation trenches were observed in the excavations or in the sections), several centimeters above the bedrock in most places (on Bedrock 1012, 1013, and 1017 in the south). This bedrock had an unusual composition: a mixture of large limestone pieces, chunky-rubble lime, and infiltrated soil. 5 Most of the pottery was lying on or above the earth surface (Soil 1010) in the collapse of the wall debris. From between the stones of the Building 1 wall came a copper pin, the only metal object found in the excavation of Har Resisim 126. Built up from this earth surface on the eastern side of the interior of the building was a stone bin (Bin 1027) of undetermined function. Building 2 (fig. 1) Oval-shaped Building 2, oriented southwest–northeast, had internal dimensions of 4.63 x 3.75 m. This stone-built structure possibly had an entryway in the southwestern sector of the wall, where two large upright stones were found. The circular wall of the structure was built of stones ranging in size from 48 x 12 x 11 to 85 x 44 x 20 cm (averaging 59 x 35 x 24 cm). The entire structure was based on virgin Soil 2006, as much as 30 cm above bedrock. The soil accumulation above this hard-earth living surface was also deep, reaching 30 cm in places. The paucity of fallen stones both inside and outside the circumference of the structure’s wall provides evidence that the wall was at most one or two courses above the remaining single course of stone found in situ. There was no central pillar or any other architectural feature within the structure. The debris on and above the living surface was sparse: some fallen stone but nothing else—neither artifacts nor ecofacts. The building was entirely sterile. Nor were there any finds in the area excavated down to bedrock or in the section made outside the building. Although it was built as well as the rest of the structures and was apparently contemporary in terms of its construction and use, its function remains obscure. 6 Building 3 (figs. 1, 3) A massive stone-built complex, comprising nine units (labeled A–I) and covering an area of 23.13 x 17.50 m (northeast–southwest/northwest–southeast) constituted Building 3. The complex includes two large lobes (Units E–F) that may have been courtyards or enclosures and five 5. Initially, this bedrock seemed to be a man-made layering of limeplaster, soil, and flagstones, but eventually, geologist Ron Gerson (who worked with us during the 1980 excavations) determined that its composition was entirely natural and was like the bedrock found elsewhere in the unoccupied environs of Beªer Resisim (personal communication). 6. At the time of excavation, Dever thought that this structure might have been a sheep enclosure, used only briefly.

Har Resisim 126: An Ephemeral Early Bronze Age Site

Fig. 1. Har Resisim 126 site plan.

43

44

Fig. 2. Building 1.

dwelling units (Units A–D, G) that approximate the size of the simple, single units, such as Buildings 1, 2, and 8. Smaller Units H and I remain unexcavated. Between Units C and D sat a stone-built bin, actually part of Unit C. There are only two clear passageways: from Unit A into Unit E and from Units C–D to the outside. No other doorways were clearly identifiable, and no two small units were interconnected. Buildings 3 and 4 were most revealing of the nature of this community during EB II, as well as during later periods. Although these complexes were composed of a series of circular units attached to much larger enclosures, the construction techniques and state of preservation paralleled those of the single-room structures. If the single structures and the similar-sized units within the complexes were living quarters, could the larger enclosures be interpreted as courtyards or animal corrals? Well-built and preserved doorways opened from Building 4 Unit 4B into its courtyard (Unit 4D) and from Building 3 Unit 3A into its courtyard (Unit 3E). However, no doorway linking any other rooms or even leading to the outside could be defined with certainty except for the doorway of Building 3 Units 3C–D. Nevertheless, the nature of the wall bonding indicated that all of the stone-built components within each of the complexes were contemporary. Indeed, in terms of the walls alone, there is evidence of no more than one phase of occupation within each structure.

spread is 12 points short

Barry M. Gittlen Unit 3A. The stones of the circular wall enclosing an interior area 2.75 m in diameter were based partly on bedrock and partly on virgin soil; they vary in size from 32 x 19 x 17 to 88 x 42 x 39 cm (averaging 51 x 31 x 23 cm). The stones of the wall shared with Unit 3B (Wall 3063) are smaller than those of the walls of the eastern and western sectors of the room. The floor was difficult to define, and we could not determine whether it was at or below the level of the threshold opening onto Unit 3E. Very little pottery was recovered from this feature. There was no central pillar or other evidence of roofing. Unit 3B. Built on bedrock that sloped sharply to the south, the roughly circular stone wall of Unit 3B may have had an entryway in its northern part, where the wall was too disturbed to be certain. The stones of the wall range from 25 x 13 x 12 to 93 x 36 x 27 cm (averaging 53 x 32 x 19 cm). There was no central pillar or other evidence of roofing in this room, the interior dimensions of which were 4.75 m north–south by 2.50 m east–west. EB II pottery fragments were found just above bedrock on a thin layer of hard-packed earth (Surface 3070) beneath the tumbled wall debris. This earthen layer represents the occupation surface of Unit 3B. Unit 3C. The circular stone wall of Unit 3C was built on soil just above bedrock. A hard-earth layer (Soil 3101) in this room (with interior dimensions of 1.75 x 5 m north– south x east–west) marked the living surface or “floor” at precisely the foundation level of the unit’s stone wall. Scattered potsherds were recovered from this earth “floor.” In places, the bedrock also formed part of this living surface. The stones of the wall range in size from 28 x 10 x 9 to 90 x 70 x 27 cm (averaging 50 x 37 x 20 cm). There was no central pillar or any other evidence of roofing. This oblong room was entered via a doorway at its northern end, an entryway that also leads, via a small alcove, to Unit 3D. Immediately south of this doorway and of the Unit 3C–3D partition wall is a large, stone-built bin (Bin 3107), the northern half of which sits on bedrock and the southern half on virgin soil. Unit 3D. Unit 3D shares an entryway with Unit 3C, on the northern side of both units. The main entrance, however, leads first into a small alcove and then through a threshold and doorjambs into Unit 3D. The circular stone wall of this unit was based on virgin soil above bedrock and enclosed an area of 3.75 x 2.50 m (north–south x east– west). The stones of the wall range from 16 x 12 x 10 to 76 x 60 x 33 cm (averaging 46 x 32 x 18 cm). Despite the numerous potsherds recovered from its debris, the earth floor of this unit was difficult to determine. There was no central pillar or other evidence of roofing, nor could we detect an entryway from Units 3C–3D into Unit 3E. There was in-

Har Resisim 126: An Ephemeral Early Bronze Age Site sufficient stone tumble in Units 3A–3D to project walls more than 2–3 courses in height. Units 3E and 3F. Unit 3E (11.50 x 5.75 m north– south x east–west) and Unit 3F (5.25 x 12.00 m north– south x east–west), comparable in size only to Building 4 Unit 4D, were far larger than the other Building 3 units and the rest of the Har Resisim 126 structures. As mentioned above, these units may have served as courtyards and/or animal enclosures. Their roughly circular perimeter walls were built of stones much larger in size than those of the rest of the Building 3 walls. Only 1–2 courses remain of these walls, and there was not enough stone debris to project more than another 1–2 original courses. No entrance or exit from these enclosures into the surrounding areas of the settlement could be identified. A probe to bedrock in Unit 3E, outside the doorway into Unit 3A, proved inconclusive. Only one artifact was recovered—a potsherd found on topsoil. The rest of the probe was sterile. In fact, an intensive surface survey of the entire area of Units 3E and 3F, including the area outside the walls, did not recover a single potsherd or any other artifact. Units 3H and 3I. Two small, circular, stone-built rooms within Unit 3F were identified only at the end of the project, when the tumbled stones of Unit 3F were removed to clarify the original wall lines. Unit 3H (1.38 x 2.25 m north–south x east–west) and Unit 3I (2.00 x 2.13 m north–south x east–west) were not excavated. Their small size indicates that they may have been storage areas or bins. Unit 3G. Unit 3G was at first thought to be a rectangular cairn appended to the western wall of Unit 3B. As the excavation proceeded, Unit 3G (2.13 x 2.63 m north–south x east–west) took on the appearance of an original room that may have been reused in a later phase as a cairn or for other purposes. The stones of its wall were founded on bedrock, and the wall itself was more rectangular than circular in shape. The stones of the wall vary from 23 x 12 x 8 to 75 x 60 x 27 cm (averaging 51 x 33 x 20 cm). The unit was sterile except for two traces of habitation found in the bedrock that constituted the living surface of this unit. The first is a cupmark (Cupmark 3305) with a line on two opposite sides cut into the bedrock in the southwestern corner of the room. This cupmark may originally have functioned as a grinding installation. In the soil inside and above the bottom of the cupmark was an EB II (?) potsherd, together with a shell bead. The second indication of habitation was found in a natural crevasse (Crevasse 3302) located on the northern side of the bedrock surface. The crevasse contained a few EB IV sherds, clear evidence of a later phase of use of Unit 3G.

45

Fig. 3. Building 3.

Building 4 (fig. 1) The units of Building 4, although it is much smaller than Building 3, were arrayed in similar fashion. At least two and possibly four dwelling units were arranged around a larger courtyard or enclosure, Unit 4D. The only clearly distinguishable doorway found was in Unit 4B, connecting it directly to Unit 4D. Unit 4A. Built on soil above bedrock, Unit 4A (5.75 x 2.75 m north–south x east–west) lacked a clear doorway, a central pillar, or any evidence of roofing. The structure’s circular stone wall was preserved 1–2 courses high, and the remaining stone debris could allow up to two additional original courses. Hard-packed earth (Surface 4002) constituted the living surface of Unit 4A, as well as the foundation level of its stone wall. Numerous EB II potsherds and the site’s single flint fan scraper were found on this surface. Above and sealing this deposit was a layer of well-laid cobbles/flagstones (Locus 4001), on which there was a stone-built bin (Bin 4010, 50 x 75 cm interior area) along the southwestern wall of the unit. Unfortunately, the debris on and above the bin and cobbles was devoid of

46 artifacts and ecofacts. Clearly, however, we have two distinct phases of use in Unit 4A: the earlier phase datable by its EB II pottery, and the later phase of Bin 4010 undatable. Unit 4B. The roughly circular stone wall that defines Unit 4B (2.25 x 2.25 m north–south x east–west) had a doorway opening into Unit 4D. The wall was built on bedrock on the west and on soil elsewhere. The stones of the wall separating Units 4A and 4B are on average smaller than those of the other Unit 4A and 4B walls, ranging in size from 33 x 20 x 18 to 70 x 45 x 31 cm (averaging 55 x 40 x 20 cm). The Unit 4B living surface was difficult to identify. This surface consisted partly of bedrock and partly of the soil naturally deposited around the bedrock. Unit 4C. Unit 4C was not completely preserved and was not excavated. This feature was defined by a circular stone wall similar to the walls of Units 4A and 4B. Unit 4D. Unit 4D was a large enclosure or courtyard, similar to Building 3 Units 3E and 3F. Its stone wall was founded on virgin soil and, in places, on the southwest and north, bedrock. The stones of this wall were larger than stones found elsewhere at the site, ranging in size from 36 x 18 x 16 to 184 x 57 x 20 cm (averaging 77 x 37 x 21 cm). In the southeastern corner of Unit 4D, a scatter of minute potsherds on the topsoil encouraged excavation of the unit. However, no sherds were found below the topsoil surface. Moreover, all of the surface sherds were EB IV in date, while sherds from Units 4A and 4B belonged to EB II. No EB II pottery was recovered from our probe in the southeastern corner of Unit 4D. Unit 4E. Unit 4E was an irregularly shaped, stonewalled room attached to the eastern end of Unit 4D. Excavation of a large section of this room down to bedrock yielded neither artifacts nor ecofacts. The walls of Unit 4E were founded well above bedrock in most places, and somewhat above the foundation level of these walls, as in Unit 4A, was a cobbled surface (which did not cover all of the excavated area of Unit 4E). Unfortunately, no artifacts were found above, on, or below this cobbled surface. Building 5 (fig. 1) Building 5 was a cairn to the east of Building 1. 7 Upon excavation, this cairn proved to be barren, containing neither artifacts nor ecofacts: only a few centimeters of sterile soil covered the bedrock surface upon which it had been constructed.

7. The cairn was excavated by Elizabeth Bloch-Smith on June 16, 1980.

Barry M. Gittlen Building 6 (fig. 1) The oblong stone-walled Building 6 (2.75 x 4.88 m north–south x east–west), perhaps a large cairn rather than a dwelling, was situated southwest of Building 3. The stones of the Building 6 walls range from 26 x 12 x 9 to 110 x 70 x 38 cm (averaging 60 x 39 x 20 cm). The excavation of half of this structure down to bedrock yielded few data: only three potsherds. No living surfaces, doorways, or roofing materials were identifiable. A large stone circle (2 x 1.5 m north–south x east–west) was built against the western wall segment of the structure, but its function could not be determined. Although the original function of Building 6 remains unknown, the presence of the interior stone circle suggests a dwelling, rather than a cairn. Building 7 (fig. 1) Building 7 was a relatively straight row of stones, 9 m long, running southeast–northwest to the north of Building 4. Only one course of stones was preserved, and there was no tumble that would suggest additional courses. The size of the stones range from 39 x 34 x 15 to 106 x 50 x 25 cm (averaging 67 x 45 x 19 cm). The stones were founded on virgin soil, slightly below the topsoil surface. A probe on both sides of the line of stones yielded no artifacts or ecofacts, and its purpose or function remains obscure. However, this line of standing stones is comparable with the stones cataloged in Avner 2002: 115–75, especially p. 118, where Avner states that one-third of the maßßebot sites he cataloged were located adjacent to habitation sites and that 72% of these were eastward-facing. Building 8 (fig. 1) Building 8 is a circular, stone-walled structure with an interior area 3 m in diameter, founded on virgin soil. Although the beaten-earth occupation surface (Surface 8001) was found at the foundation level of the stones, little artifactual material was recovered from the floor or from the soil above it. Moreover, nothing was found below this beaten-earth surface. Indeed, almost all of the recovered potsherds came from the topsoil surface. The relative sterility of this structure and the fact that its foundation is entirely on soil are similar to the case of Building 2. No doorway could be defined with certainty, although there is a place in the wall on the eastern side where stones are missing. To the south of this possible “doorway” was a well-constructed stone bin (Bin 8004, 50 x 50 cm interior area) that was also devoid of finds. The bin was sealed below an apparently recent, semicircular, stone-built fire/windbreak. The secondary reuse of this structure as a fire/windbreak could date to relatively modern times. The small exterior bin on the eastern side beneath the fire/windbreak

Har Resisim 126: An Ephemeral Early Bronze Age Site seems to be similar to construction features that I observed in EB II structures on the alluvial fans within Nahal Resisim, approximately 4–6 km to the east. Building 9 (fig. 1) Building 9 was rectangular, had been competely robbed, and was possibly a cairn, 1 m south of Building 4. Like the Building 5 cairn, it had an interior area of 1 x 1 m. It was not excavated. Building 10 Building 10 was also rectangular, had been robbed, and was possibly a cairn south of Building 2 in the southwestern corner of the site, similar to cairn Buildings 5 and 9. This structure was not excavated. Building 11 (fig. 1) Building 11 was a small, circular dwelling, or perhaps a cairn, 2.60 x 2 m in interior area, to the west of Building 6. It was not excavated. Building 12 Building 12 was a fragmentary, entirely robbed cairn to the west of Building 3 Unit 3G. It was not excavated. Additional Features at Har Resisim 126 The plateau area of Har Resisim 126 yielded additional remains in the vicinity of the above-mentioned structures. Several unexcavated cairns (Buildings 13–15) were discovered 50–100 m to the northeast of the settlement. A very large cairn was found several hundred meters to the east, surrounded by several smaller ones and a number of petroglyphs. Four petroglyph fields were found in and near the settlement itself. 8 Petroglyph Fields 1 and 2 lie within the settlement: the former, just west of Building 2, includes both animal drawings and inscriptions in Arabic; the latter, 1 m south of Building 1, includes animal and human representations. Petroglyph Field 3 lies a few hundred meters to the southeast along the edge of the plateau and includes human and animal representations. Petroglyph Field 4 lies toward the northeastern edge of the slope between Buildings 1 and 13–15. It includes figures and an old “Arabic” (South Arabic or Thamudic?) inscription.

Conclusions The inhabitants of Har Resisim 126 gathered large stones from the surface limestone outcrops to build their circular structures. Some of the structures were founded partly on bedrock, but most were built entirely on the loess accumulation above the bedrock. These circular structures had few or no common features other than their circular 8. All of these petroglyphs have been photographed.

47 stone construction. The walls were rarely preserved to more than 1–2 courses high, and the quantities of abutting stone tumble indicate that they were originally built no more than 4 courses high. There is no evidence of roofing material or roof support. Doorways were difficult and in some cases impossible to identify. Floors were not easy to determine because they consisted of hard-packed loess just above the bedrock (in only one case was the bedrock actually used: in Building 3 Unit 3G). In a few cases, when the bedrock sloped down sharply, the floor of a room consisted partly of bedrock and partly of soil. The average interior space of the 14 single-room buildings and units within Buildings 3 and 4 (excluding cairns and courtyards) was 8–16 sq. m, suggesting a population (calculated at 3–5 people in each) of 40–70 individuals. 9 The structures were erected quickly and simply: there is no evidence of foundation trenches, mortar, or wall plaster. Indeed, they are reminiscent of the Bedouin structures noted by Bar-Yosef in northern Sinai: circular walls of large stones erected inside tents (1981: 157). Bar-Yosef (1981) describes those Bedouin complexes as settlements of hunters or herdsmen and thus they may provide useful analogues to the Har Resisim 126 settlement. The character of the two complex Buildings, 3 and 4, was similar to the character of the simple, single-room structures. Composed of a series of circular or semicircular lobes attached to larger structural units, their construction and state of preservation paralleled similar structures illustrated in Beit-Arieh 2003: 43, 368, 441. Several of the rooms surrounding the larger enclosures in Buildings 3 and 4 opened onto these enclosures via clear, well-preserved doorways. However, no doorway between rooms could be identified. Nevertheless, the nature of the bonding of the walls clearly determined that all the components of Buildings 3 and 4 were contemporary. The architectural evidence of the walls indicates a single construction phase for both the complex and the simple buildings. However, as described below, the evidence of the surfaces suggests two phases of occupation in a few of the rooms. Buildings 3 and 4 provided most evidence of the nature of this community during the EB II and later periods. The finds on the floors/surfaces and in the tumbled debris above the floors/surfaces of the structures are uniform, albeit meager. Roughly 50 potsherds constitute the entire ceramic corpus from Har Resisim 126. The pottery consists of sherds, predominantly body fragments, mostly small, and almost entirely of holemouth cooking pots. The few rim and burnished fragments recovered from the excavation of 9. This population estimate is based on Broshi and Gophna 1984; Haiman 1992: 101; Finkelstein 2001: 60; Beit-Arieh 2003: 108–9; as well as Shiloh 1980; Stager 1985; and Lehmann 2004.

Barry M. Gittlen

48

Fig. 4. Har Resisim 126 diagnostic pottery: holemouth jars.

these buildings clearly antedate the EB IV occupation of Beªer Resisim. Even the ware of the body sherds is significantly different from the assemblage from Beªer Resisim. A preliminary search through the literature disclosed only a few parallels for the rim shapes, and these came from Arad. However, an examination of several assemblages from the Negev and Sinai revealed some startling parallels. In 1981, with the assistance of Rudolph Cohen, I examined material from Arad and from a number of EB II sites discovered during the Israel Survey of the Central Negev Highlands. Although their rim and base profiles bore some similarities to their counterparts at Har Resisim 126, the Har Resisim 126 clays were quite different from the heavily included clays from those sites. The EB II materials I examined from Eliezer Oren’s surveys of northern Sinai also proved to be completely dissimilar. Also in 1981, with the assistance of Beit-Arieh, I examined the materials from his excavations of the EB II copper-mining complex at Nabi Salah (1974; 2003). We were both astonished to discover that, in terms of both the profiles and the finely levigated ware, our ceramic assemblages were quite similar. Indeed, despite the greater diversity of profiles and wares in the Nabi Salah assemblage, it appears to be the only EB II ceramic group with which the Har Resisim 126 assemblage finds close parallels. Although we are not certain, it appears that the pottery belongs to the EB II tradition best represented by the assemblage from Nabi Salah in southern Sinai and also seen in the Arad and the Central Negev Highlands forms but not their ware (fig. 4). The small finds—including two beads, a sandstone palette, a flint fan scraper, and a copper pin—probably also belong to EB II but do not help to pinpoint further the date of occupation at Har Resisim 126. While the architectural evidence points to a single building phase datable to EB II, other evidence strongly suggests more than one phase of occupation at Har Re-

sisim 126. In Building 4 Unit 4A, hard-packed earth Floor 4002, which was covered with EB II sherds and yielded a flint fan scraper, was itself covered and sealed by the welllaid flagstone Floor 4001 on which the stone-built Bin 4010 was constructed. Unfortunately, not a single find came from this upper phase of settlement. Similarly, above the sterile soil in Building 4 Unit 4E was a well-laid flagstone floor, but this floor and the soil above it also proved to be barren. However, just to the east, on the other side of the wall that separated Units 4E and 4D, a scatter of miniscule EB IV potsherds, including two with combed decoration, was found on the earth surface of Unit 4D. A probe beneath the surface of the soil down to bedrock revealed no other pottery. Clearly, there was some activity at Har Resisim 126 in the EB IV period. The flagstone reflooring of Units 4A and 4E may have been contemporary with the EB IV floor of Unit 4D. Additional evidence for a shortlived EB IV phase at Har Resisim 126 came from a deposit in a depression/ditch at the northern edge of Building 3 Unit 3G. Apparently, the cupmarked bedrock on which the Unit 3G walls were built represents the occupational surface of the room. Within Cupmark 3305 was a small shell bead and an EB II sherd, but Crevasse 3302 in the bedrock in the northwestern part of this room contained a few clearly EB IV sherds. Har Resisim 126 was a transitory, briefly occupied EB II village or encampment, perhaps related to sheep or goat herding (although there is no evidence that animals were kept in the large enclosures). Because the site lacks any depth of deposit (indeed, there is hardly any deposit), its occupation could not have been very long. All of the structures appear to be contemporary; where they yield any pottery at all, it is EB II pottery. The spartan locale and ephemeral nature of the occupation may suggest that Har Resisim was a winter pastoral encampment. Yet the rather difficult ascent undoubtedly reduced severely whatever

Har Resisim 126: An Ephemeral Early Bronze Age Site benefit was derived from the seclusion of its locale. Moreover, there is little reason to conclude that the large enclosures, let alone the single-room structures, were animal corrals. Finally, the distinct lack of osteological data and the absence of evidence of hearths or fires argue against positing a long stay at this site. The EB II settlement at Har Resisim 126 may have been built in this dramatic location in order to overlook the route along the Nahal Nissana, 500 m below it. From Har Resisim 126, one could see for miles up and down the nahal. Might the site have been a lookout (mizpeh), perhaps for bandits? Assuming that the EB II copper trade was significant (see, e.g., Beit-Arieh 1984), it must have attracted some banditry, and one need look no further than the Late Bronze Age Amarna Letters for parallels (see, e.g., Gittlen 1985; Moran 1992: 16–17).

49 The meager EB IV occupation was even more ephemeral in nature than that of the EB II. The few cairns on the fringes of the site may be dated to EB IV or later. Building 5 was the only cairn intact enough to warrant excavation, and it proved to be completely empty. Moreover, the upwind placement of the cairns may indicate construction and use after the abandonment of EB II Har Resisim 126. Perhaps the cairns were erected during the EB IV period by villagers from nearby Beªer Resisim, whose few broken vessels lodged among the ruins of Har Resisim 126, which they were mining for their cairn-building stone. After EB IV, Har Resisim 126 lay abandoned, only occasionally visited by passers-by, whose presence is recorded in the numerous undatable animal, human, and abstract figures that they incised into the bedrock ledges surrounding the ruins and overlooking the Nahal Nissana.

References Avner, U. 2002 Studies in the Material and Spiritual Culture of the Negev and Sinai Populations during the 6th–3rd Millennia b.c. Unpublished Ph.D. dissertation, Hebrew University of Jerusalem. Bar-Yosef, O. 1981 From Hunter to Herder in Southern Sinai. Pp. 156–57 in Temples and High Places in Biblical Times, ed. A. Biran. Jerusalem: Hebrew Union College. Beit Arieh, I. 1974 An Early Bronze Age II Site at Nabi Salah in Southern Sinai. Tel Aviv 1: 144–56. 1984 New Evidence on the Relations between Canaan and Egypt during the Proto-Dynastic Period. Israel Exploration Journal 34: 20–23. 2003 Archaeology of Sinai: The Ophir Expedition. Tel Aviv: Tel Aviv University Press. Broshi, M., and Gophna, R. 1984 The Settlements and Populations of Palestine during the Early Bronze Age II–III. Bulletin of the American Schools of Oriental Research 253: 41–53. Cohen, R. 1999 Ancient Settlement of the Central Negev. Jerusalem: Israel Antiquities Authority. Cohen, R., and Dever, W. G. 1979 Preliminary Report of the Second Season of the “Central Negev Highlands Project.” Bulletin of the American Schools of Oriental Research 236: 41–60. 1981 Preliminary Report of the Third and Final Season of the “Central Negev Highlands Project.” Bulletin of the American Schools of Oriental Research 243: 57–77.

Finkelstein, I. 2001 Living on the Fringe: The Archaeology and History of the Negev, Sinai and Neighbouring Regions in the Bronze and Iron Ages. Sheffield: Sheffield Academic Press. Gittlen, B. M. 1985 The Murder of the Merchants near Akko. Pp. 63–72 in Biblical and Related Studies Presented to Samuel Iwry, ed. A. Kort and S. Morschauser. Winona Lake, IN: Eisenbrauns. Haiman, M. 1992 Sedentarism and Pastoralism in the Negev Highlands in the Early Bronze Age: Results of the Western Negev Highlands Emergency Survey. Pp. 93–104 in Pastoralism in the Levant: Archaeological Materials in Anthropological Perspectives, ed. O. Bar-Yosef and A. Khazanov. Madison: Prehistory Press. Lehmann, G. 2004 Reconstructing the Social Landscape of Early Israel: Rural Marriage Alliances in the Central Hill Country. Tel Aviv 31: 141–93. Moran, W. 1992 The Amarna Letters. Baltimore: Johns Hopkins University Press. Shiloh, Y. 1980 The Population of Iron Age Palestine in the Light of a Sample Analysis of Urban Plans, Areas, and Population Density. Bulletin of the American Schools of Oriental Research 239: 25–35. Stager, L. E. 1985 The Archaeology of the Family in Ancient Israel. Bulletin of the American Schools of Oriental Research 260: 1–35.

Gezer Rectified: The Dating of the South Gate Complex James W. Hardin and Joe D. Seger Cobb Institute of Archaeology Mississippi State University

In an article published in 2002, “Gezer Revisited and Revised,” Israel Finkelstein censures the Hebrew Union College (HUC) Gezer excavation team for presumed deficiencies in its excavation work and scholarship. He also presumes to render a “revised” analysis of Gezer’s stratigraphic history. Dever has already responded on behalf of the Gezer team in “Visiting the Real Gezer” (Dever 2003). Dever’s article exposes the often anachronistic use and misappropriations of data that typify Finkelstein’s approach, as well as correcting his most egregious mischaracterizations. The goal of this essay is to complement and supplement Dever’s response with a more focused review of Finkelstein’s treatment of the South Gate complex of the Middle Bronze Age Inner Wall system at the site. In the opening section of his article, “The HUC Team at Gezer, Period by Period” (2002: 263–68), Finkelstein levels the following charges regarding the treatment of the end of the Middle Bronze Age (MB IIC, his MB III) (Finkelstein 2002: 264): [T]he elaborate, fortified city of the Middle Bronze III has not been put in context: The excavators did not discuss the role of Gezer in the city-state system of the Middle Bronze Age and did not attempt to reconstruct the size and demographic characteristics of the Gezer territory at that time. . . . The nature of the main element in the Middle Bronze city—the fortifications (which had already been discovered by Macalister)—has long remained a matter of dispute. Tower 5017 is one of the most elaborate monuments built in Canaan in the Bronze Age. Yet, its precise date and function have not been clarified.

The broad strokes of these introductory comments are seriously misleading. While we do not maintain that everything that needs to be said about the context and role of Gezer in the late Middle Bronze Age has been written, 1 it is significant that, in advancing these initial remarks, Finkel1. The authors are currently preparing the final report on these materials. They wish to acknowledge the assistance of Mississippi State University Philosophy and Religion major Michael Stewart, who provided an editorial review for the present essay, and of technical illustrator Dylan Kargas, who prepared the accompanying illustration.

stein omits reference to several key studies on these topics that have been published. He does not directly refer to Seger’s 1975 comparative study of Gezer and Shechem in the Hyksos period (he only cites it later, on a different topic [Finkelstein 2002: 277–78]), and he fails to acknowledge Dever’s study of this period in Gezer IV (1986: 20–36) and his 1987 survey article in the Biblical Archaeologist. The last two are ignored throughout. Regarding the dating and function of the Inner Wall fortifications, a general consensus has long existed about the relative dating of the system within the Hyksos period (see already Parr 1968; Gezer I: 42; Seger 1975; 1993; Kempinski 1992a: 189). Regarding Tower 5017, more specifically, Finkelstein cites only the 1972 review by Kempinski of the first Gezer publication (Gezer I), again ignoring Dever’s pointed rebuttal to the review, exposing Kempinski’s faulty reasoning (Dever 1973b). Curiously, later in the article, Finkelstein himself rejects Kempinski’s approach (2002: 277 [see further below]).

History Revised? Finkelstein presumes to claim the field with these introductory jousts and then proceeds to provide what he characterizes as a “significantly different” and revised history of Gezer during the Middle Bronze Age. Curiously, most of his “revisions” are no more than rehearsals of what the excavation team has already suggested. With regard to the South Gate complex—which includes Tower 5017, its series of connecting defensive walls, its exterior glacis, and the flanking towers of the South Gate itself (fig. 1)— Finkelstein raises three main issues. (1) He takes exception to the excavation staff’s dating of some of the fortification elements of the South Gate complex and of the associated Middle Bronze Age strata. (2) He accuses the Gezer team of misunderstanding the function of some of the components of this system (2002: 277–78) and argues that this misunderstanding leads to erroneous interpretations and overly complicated reconstructions of the history and use of the South Gate and its associated fortifications.

52

James W. Hardin and Joe D. Seger

Fig. 1. Plan of Gezer Field IV.

(3) He maintains that this misunderstanding was brought about by “a series of methodological and disciplinary deficiencies” (2002: 268). It is this misunderstanding, he argues, that ultimately leads to the “debacle” of the excavation staff’s attempts to understand the gate complex and to place Middle Bronze Age Gezer in its geopolitical context (2002: 268, 277–79). Each of these issues is addressed below. Dating the Components of the South Gate Complex As stated above, there has long been a general consensus regarding the relative dating of the components of the South Gate complex within the Hyksos period. Despite this, Finkelstein takes issue with the excavation team’s dating of these elements and their associated Middle Bronze Age strata. His argument stems from his lack of confidence in the team’s ability to understand correctly the construction history of the Inner Wall system. He begins by questioning the approach used by the team to date the Inner Wall, suggesting that their method

of dating “according to the level from which the foundation trench was cut . . . is not only difficult, it is in fact impossible” (Finkelstein 2002: 277). This suggestion seems to derive from his belief that, as a result of the various filling activities and the cutting of the Inner Wall’s foundation trench (Trench 4168), it is “difficult to associate settlement strata with the face of the inner wall” (Finkelstein 2002: 277). These statements are both surprising and confusing. The means used by the excavation team to date the Inner Wall were initially based on the analysis of material collected from Foundation Trench 4168 in Field I, which was obviously cut to prepare for the construction of Tower 5017. This is the same means of dating that was subsequently used to investigate the Inner Wall in Fields IV, V, and VIII (Seger 1975: 39; 1993: 561–67). It is difficult to understand with what Finkelstein finds fault in the team’s approach to this problem, because he also states that the use of this dating method is far better “than Kempinski’s search (1972; 1976) for traces of a superstructure on top of Tower 5017” (Finkelstein 2002: 277).

Gezer Rectified: The Dating of the South Gate Complex The dating methods used by the Gezer team are clear cut. The team first uses the latest pottery from foundation trench deposits. This pottery is then compared with later materials associated with the floors abutting the Inner Wall and with pottery from earlier floors cut by related construction activity (Seger 1975: 39; 1993: 565). A foundation trench, perforce, is cut from and through a preexistent layer or level, which represents a post quem threshold for the wall’s construction. A surface that directly covers and seals the foundation trench and abuts the wall must represent an associated settlement stratum, with materials providing a terminus post quem. The details that illustrate this sequence are clear from Field I, where Tower 5017 was first investigated (see Gezer I: plan II). Floors cut by Foundation Trench 4168 include the floors of Phase 9 (Stratum XIX), which date to the late MB IIB (MB II). The earliest living surfaces associated with the wall system are those of Phase 8 (Stratum XVIII), representing the MB IIB/C (MB II/III) transition. The floors of this phase run up to and abut the internal face of the Inner Wall, sealing Foundation Trench 4168. They thus provide a terminus post quem for the construction of Tower 5017. The latest materials in Foundation Trench 4168 and the Phase 8 pottery therefore represent the horizon in which the construction of the Inner Wall and its first phase of use took place. In Field I at Gezer, Surface 3044 of Phase 9A and Surface 4055 of Phase 9B represent the latest pre-Tower 5017 occupation levels (Gezer I: plan II). These and earlier levels were cut by Foundation Trench 4168 in preparation for the building of Tower 5017. Foundation Trench 4168 is associated with Phase 8B and represents the period of the tower’s construction and trench backfilling. Surface 4032 represents the first stage of related occupation. In addition, Wall 4024, built directly on Surface 4032, along with Surfaces 4041 and 4026 that abut it offer evidence of the subsequent Phase 8A occupation (Gezer II: 33, plan II). Taken together with the results from the work conducted in Field IV, these remains provide definitive evidence supporting the team’s understanding of the Inner Wall system’s construction and dating. Again, it is unclear with what exactly Finkelstein finds fault or why he believes the methods used by the team to provide a terminus post quem for the Inner Wall’s construction and use are “impossible.” Having muddled his way to this point, Finkelstein proceeds to analyze briefly the ceramic data for dating Tower 5017 and the Inner Wall system. He asserts that “the Gezer plates must always be suspected of contamination” and that this “may be the reason for Dever’s errors in dating Middle Bronze pottery” (Finkelstein 2002: 269). The possibility of contamination, according to Finkelstein, is due to the team’s excavation methodology and its reliance on deep narrow probes. Finkelstein believes that methods of

53 this sort only yield sherd data, rather than large corpora of restorable ceramics. Although this methodological critique is addressed further below, it must be noted here that the presence of sherd data is inherent to the nature of the excavated material and not to the excavation method. Finkelstein points out how readily sherd data can be mixed; citing examples from his excavations at Megiddo, he notes how difficult it can be to separate mixed materials that are temporally proximate. But no matter how difficult it may be, separation of this sort is possible. Finkelstein himself demonstrates this in his own analysis of a number of the published sherds from Gezer (2002: 277). He refers to numerous objects and ceramic sherds that he believes—and we agree—are significant for determining the date of the Inner Wall system’s construction and use. It is notable that at this point he uses the Trench 4168 data in the way that he earlier dubbed “impossible.” He similarly refers to sherd materials from Glacis 8012, which he erroneously associates with the Tower’s founding phase (see below). However, the point remains that, despite his arguments to the contrary, Finkelstein dates the construction of the South Gate complex, as the excavation team did before him, by means of the recovered materials, thereby countering his own position by demonstrating the utility of the excavation method. Finkelstein’s analysis of the ceramic data from the South Gate complex and Inner Wall Middle Bronze Age defenses thus warrant his conclusion that (2002: 277): All in all, it seems that the Inner Wall was constructed in the late Middle Bronze II or early Middle Bronze III. This date fits what we know about cyclopean fortifications in other parts of the country: in the highlands sites of Shechem, Shiloh and Jerusalem . . . and in Megiddo (Stratum X) in the lowlands.

There is absolutely nothing new in this dating of the genesis of the Inner Wall system; it merely echoes the earlier conclusions of the Gezer staff (Seger 1975: 39; 1993: 563; Dever 1992: 999; 1993: 501; 1997: 397). In this analysis, rehearsal rather than revision again characterizes his presentation. The Function and History of the South Gate Components In advancing his second main issue, however, Finkelstein does put forward a revised view, arguing against the sequence of development and dating of the various elements of the South Gate complex proposed by the Gezer team. His view is based on what he considers to be a misunderstanding of the function of the architectural components. This misunderstanding, he argues, results in erroneous interpretations and overly complicated reconstructions of the South Gate system’s history and use.

James W. Hardin and Joe D. Seger

54 Specifically, he argues against the excavation staff’s conclusion that two separate phases were involved in the construction and use of the South Gate complex, opting instead for a “one-phase-fits-all” model. He begins with a succinct outline of the team’s early arguments for the twophase theory, which warrants repetition in toto (Finkelstein 2002: 278): The two phases theory was based on the following arguments (Seger 1975: 40–41): • Wall 3097A [the lower extender wall2] is built of larger, cyclopean stones, similar to those used in Tower 5017 and different from most of the stones in the upper extender wall [Wall 13004]. • Wall 3097A and Tower 5017 are bonded. • Wall 3097A and Tower 5017 are founded in the same level. • Wall 3097A is not bonded to the foundations of the Southern Gate, “but appeared to pass independently under it,” while Wall 13004 does bond to the gate tower. • The latest sherds in Glacis 8012 seem to post-date the latest sherds in Foundation Trench 4168.

Finkelstein’s arguments are largely dismissive of these data. His one-phase model presupposes that all the components of the system (the tower, glacis, gateway, and two extender wall elements) belong to the same fortification complex. In one sense, this is clearly true. But the issue is whether all the components always functioned within the system at the same time. Finkelstein contends that they did and consequently concludes that they were constructed together at the same time. This conclusion is apparently based on the interpretation of the primary functions of fortification elements that he derives from his observations of Middle Bronze Age fortification systems at other sites, most notably Shiloh and Shechem (1992). He accordingly believes that only when all components of the South Gate complex at Gezer are seen as functioning together at the same time can they be understood properly. We contend that the various elements of the fortification system at Gezer did not always function together. In the first stage, the system included only Tower 5017, extender Wall 3097, and a putative early gateway. The second stage, in which the complex was strengthened and consolidated, included Tower 5017 and the elevated extender Wall 13004, together with complementary new gate towers and the exterior glacis. Finkelstein submits three points to refute this interpretation: (1) He asserts that differences in construction techniques within the Inner Wall fortification system are not chronologically significant. (2) He suggests that, because 2. The wall originally designated Locus 3097A by the Gezer team in the field is now simply called Wall 3097.

the Inner Wall was built on a slope, it could not have survived for long without a glacis counterbalancing the massive fills contained by the wall. (3) He proposes that Wall 3097 always served as a foundation for and was contemporary with Wall 13004. Differences in Construction Techniques. In advancing point 1, Finkelstein states that “differences in construction of the same system do not necessarily [italics ours] indicate different dates; they rather point to structural considerations” (2002: 278). Were the emphasis on the conditional nature of the statement, it could readily be accepted. However, Finkelstein’s extension of this caveat to apply to bonded structures (2002: 278) is clearly wrong. Bonded structures are by definition built together, at the same time. Although a system at any given time may be constructed with bonded and unbonded elements, bonded features are a sure indication of contemporaneity. As evidence in support of his idea, Finkelstein refers to other fortification systems in which the combined use of different construction techniques has been observed. He cites specifically the case he made in 1994 for a revised dating of the Outer Wall system at Gezer (contra Dever 1990a; see also Finkelstein 1981; 1990). But even while making his case for a different dating of the initial Outer Wall construction, he also affirms that there are differences in construction techniques within this wall system that are chronologically significant, pointing to the long history of its repair and reuse. However, irrespective of the confusion he introduces with these observations, the point to be stressed is that no conclusive interpretation of stratigraphic relationships and architectural history can be made based solely on observations of differences in construction style and technique. All the related stratigraphic data must be taken into account. With respect to the South Gate complex, Finkelstein provides no evidence indicating that he has done this. Wall Construction and Glacis Function. Finkelstein’s point 2, much like his point 1, also looks to considerations other than the related stratigraphy. What he proposes is a “functional” explanation for understanding the wall system’s history. He believes that “the Inner Wall was built on the slope, with a huge accumulation of debris—fills and living surfaces—leaning on its inner face” (2002: 278). In addition, he believes that debris was laid as a large fill to create a “type of podium extending the site in the direction of the slopes of the hill” (2002: 277). Thus, according to him, one of the main functions of the wall system was to contain the massive fills laid in preparation for this podium. He argues that for the wall to function in this manner an exterior glacis was necessary to provide adequate support. He thus assigns the glacis in Fields I and IV at Gezer to the first (and for him the only) stage of fortifica-

Gezer Rectified: The Dating of the South Gate Complex tion construction. The glacis was ostensibly added to serve the needed function of supporting Tower 5017 and the eastward extending Walls 3097 and 13004 in counterbalancing the huge fills laid behind their inner faces. Finkelstein’s main argument for this reconstruction arises from his analysis of the comparative data from Shiloh and Shechem (1992). 3 At these sites, elements in the Middle Bronze Age wall systems were understood by their excavators as functioning to create extended living areas within the fortifications (Seger 1975; Lederman 1985). This was accomplished by building broad walls farther out from the settlement and then laying substantial leveling fills against their interior faces. These fills created new raised areas onto which settlement could be expanded. Finkelstein believes that the pressures exerted on city wall elements at Shiloh and Shechem by these fills were so massive that they necessitated building glacis structures for external support. The glacis at Shiloh. Finkelstein’s interpretation of the primary function of Glacis 723 in Field D at Shiloh as a counterbalance to the fills laid internally to the settlement is both interesting and convincing (1985; 1992; see also Lederman 1985). With a great width (25 m) and gentle incline (25 degrees, closely matching the slope of the bedrock), it ascended to a height (i.e., depth of deposit) of 6.3 m (Finkelstein 1985: 161; Lederman 1985: 141– 43). It therefore would probably not have presented much of an obstacle to an attacking foe. This would most assuredly be true if, moreover, Finkelstein’s conclusion that the fortification wall against which the glacis was laid had no additional brick or other superstructure (1992: 203) is correct, because this wall was preserved to but a few stone courses (less than a meter) above the glacis summit. The glacis at Shechem. The excavators of Shechem identified a glacis on the eastern side of the tell (Seger 1974: 118; 1975: 38; Campbell: 2002: 113–14). In 1989, Ussishkin proposed (followed by Finkelstein 1992) that a similar glacis structure existed on the northwestern periphery of the site in front of cyclopean Wall A. As recently argued by Campbell, this is not convincing for several reasons (2002: 112–13). First, the stones on the face of Wall A on the northwest were secondarily trimmed to create a smooth facade. Second, it is unlikely that Sellin and Welter, then Wright, and then Dever all missed a deliberately laid, anterior fill that may have served as a glacis as3. This approach follows the revised understanding of the Middle Bronze Age defense systems at Jericho and Shechem proposed by Ussishkin (1989), who argued that they should be understood as functioning similarly to the Middle Bronze Age fortification system at Shiloh as based on Finkelstein’s preliminary reports (Finkelstein 1985; see also Lederman 1985). Finkelstein now seems to believe that every system must fit the single-phase scenario.

55 sociated with the construction of Wall A. Third, the sizable windows cut through Wall A on the northwest to allow for drainage or “weeping” could not have functioned if the outer face of the wall had been covered. Finally, it is also worth noting that, since its original excavation in the 1930s, the 40-meter-long section of Wall A on the northwest has stood with its 7-meter-high outer face completely exposed. Still standing for over 70 years, it continues to hold back the massive fills of redeposited soil and chalk laid behind it over 3,600 years ago. The glacis at Gezer. We believe that the main purpose of the glacis at Gezer was to stabilize and fortify the outer slopes of the settlement, while at the same time inhibiting an attacking enemy’s access to the outer perimeter of the Inner Wall line—deterring efforts to sap its lower/basal courses. This follows earlier understandings of the development of MB IIB and C (MB II–III) fortification systems (Yadin 1955: 25–31; 1963: 67; Parr 1968; Kenyon 1979: 161; Pennels 1983; Mazar 1990: 198–205; Stager 1991; Kempinski 1992b: 129; Ilan 1993). While a glacis may indeed provide stabilizing support for wall foundations in some instances, perhaps even at Gezer, this appears to have been of secondary concern for the builders. Where the slope of the glacis is gradual and its horizontal extent is great, as at Shiloh, glacis construction may also have served as a means of providing counterbalancing support. The glacis at Gezer, however, creates an almost 45-degree slope, far steeper than the bedrock on which it rests. Its total height (i.e., depth of deposit) is between 6 and 7 m, about the same as the glacis at Shiloh, but it is only half as wide. It is also apparent that the wall sections against which it rests continue upward to rise to substantial heights above it. We believe that the differences observed in the glacis systems at Shiloh, Shechem, and Gezer represent adjustments in defensive strategies related to the topographical features at the individual sites in addition to their primary function of being filling and support structures for their related fortification walls. Although filling activities clearly took place at Gezer (Gezer I: 41–42; Gezer II: 33), they were not of the magnitude or for the purpose suggested by Finkelstein. In fact, the stratigraphic data from Gezer Field I related to Tower 5017 present a different picture. Foundation Trench 4168 is indeed broad and deep, cutting through earlier levels all the way down to bedrock (Gezer II: plan II). However, except for the girth of the wall itself, it is not obvious that the site was much expanded at this point. As the tower was constructed, the trench was backfilled until it reached the level from which it was cut. The trench was finally sealed by the living floors and other occupation remains of Phase 8 (Stratum XIX). In Field IV, within the line of connecting walls, the situation is similar. But it is obvious from the

56 published section drawings that the backfilling and leveling against the inner face of Tower 5017 and against Inner Wall elements elsewhere at Gezer are not at all like the massive filling activities observed in Field D at Shiloh or along Wall A near the Northwest Gate at Shechem. Whatever pressure these backfills may have exerted on the Gezer Inner Wall system, the necessity of support from the outside in the form of a glacis, as Finkelstein argues, is clearly contestable. Finkelstein evidently cannot imagine that the 3+-meterwide Wall 3097, much less Wall 13004 mounted above it, could have withstood the pressure of internal fills without the added support of the glacis. However, his logic is faulty. At other sites, there are many structural walls of even less girth than 3 m that retain substantial fills without the addition of a glacis. At Gezer itself, long stretches of the 3-meter-wide Outer Wall have no such external glacis support. In Fields I, II, IV, and XI, sections extending across the Outer Wall show that it is constructed in a cut made down and through the fills forming the Inner Wall’s glacis. This positions it farther down-slope, where it retains even larger fills than the fills inside the Inner Wall. Yet there is no glacis system to support it. Moreover, there are features at Gezer other than the glacis that add to the structural integrity of the Inner Wall line. In Fields I and IV, the 3+-meter Walls 3097 and 13004 are strengthened and consolidated by the massive flanking structures of Tower 5017 and the towers of the South Gate. Finkelstein’s argument (2002: 277) that a massive structure such as Tower 5017, with 15.6-meterwide foundations built of well-fitted cyclopean stones, would need this sort of glacis for “counterbalancing the pressures of the fills” is most unreasonable. Furthermore, with regard to his “one-phase-fits-all” assessment, applying his logic actually works against his argument. It is probable that some perceived weakness in Wall 3097, whatever the causes (possibly including the settling and pressure from the fills within), was in fact the impetus for the second-phase reconstruction effort represented by the additions of Wall 13004 and the glacis. 4 As a final point in support of his contention that Gezer’s glacis was constructed primarily as a counterbalance, Finkelstein concludes with the statement: “The fact 4. Engineering/structural problems discovered in a gate built in the southeastern corner of Tel Dan lend tacit support to this conclusion (Biran 1984: 7–9). The gate was part of a slightly earlier fortification system (MB IIA), constructed with a variation of the “same principle of construction”—that is, a stone core with sloping earthen ramparts (Biran 1990: 103). Either a structural fault in the gate itself or slippage of the rampart materials made it necessary first to buttress the structure heavily and then to abandon it completely and to cover it with the rampart. All of this took place within a short period of time (Biran 1984: 8–9, 18).

James W. Hardin and Joe D. Seger that Glacis 8012 creates a plateau in front of the fortification (Seger 1975: 41) proves that its main function was supportive, not defensive, since a plateau-like glacis would only facilitate the approach of an enemy to the wall” (Finkelstein 2002: 278). This assertion is facile and easily challenged. First, only in the offset sections between the towers of the Inner Wall system, such as the sections along Wall 13004, does the glacis at Gezer form a modest plateau. Otherwise, the slope fronting the fortifications is maintained at a daunting 45-degree angle. As noted above, this is designed to inhibit an attacking enemy’s access to the outer perimeter of the wall and to deter efforts at sapping its lower courses. Second, along the plateau between Tower 5017 and the South Gate in front of Wall 13004, traces of modest parapet-type walls were identified (Wall 14003; Seger 1972b: 240; 1975: 41) (see here fig. 1). 5 As Yadin’s 1955 and 1963 studies both clearly demonstrate, systems including low revetment walls in front of major fortification structures along the top of a glacis were routine elements in the defense systems of the late Middle Bronze Age (1955: 26; 1963: 67, with illustrations pp. 160–61). Militarily, these walls formed a kind of “bunker” system for first-line defense on the glacis summit. Third, the narrow plateau sections actually provided “traps” for enfilading fire by defenders from the flanking towers. Thus, contrary to what Finkelstein suggests, rather than facilitating the approach of an attacking army, these plateaus in fact formed an integral part of Middle Bronze Age strategies for city defense. Certainly, as Finkelstein states, many of the Middle Bronze Age fortification systems, at Gezer and elsewhere, were “buried in the ground, with fills leveling the areas on their inner sides” (2002: 277), and again as at Gezer, a glacis was added to the outer face of some. But we believe that to attribute the Middle Bronze Age glacis primarily to activities associated with counteracting the outward-exerting forces of fills in the tell matrix is overly simplistic. It represents a serious misreading of the architectural history of the Bronze Age glacis and embankment fortification systems and overlooks important primary features involved in their development and use. The Contemporaneity of Walls 3097 and 13004. Finkelstein’s point C, that Wall 3097 was foundational to Wall 13004, is related to his first point regarding observed differences in architectural features. Both are raised to suggest that the lower connector Wall 3097 (which was founded at the same general level as Tower 5017) and the upper connector Wall 13004 (which was built above Wall

5. A similar parapet structure was excavated along the glacis summit in Field II in 1973 (Seger 1973: 249).

Gezer Rectified: The Dating of the South Gate Complex 3097) could have been (which for Finkelstein in fact means were) constructed together at the same time. To sustain this argument, however, Finkelstein has to ignore the overall stratigraphic picture. As we asserted above, the relationship of a number of features associated with the walls cannot be explained by his one-phase interpretation. These include most notably the several courses of mud-brick superstructure preserved atop Wall 3097, separating it from Wall 13004, and the sequence of occupation phases represented by surfaces and walls abutting these Inner Wall elements. In materials already published, these occupation surfaces include Field IV Area 3 Surfaces 3069 (Phase 8B) and 3066 (Phase 8A), both of which abut Wall 3097, and Surface 3025 (Phase 7), associated with Wall 13004 (Seger 1975: 36; 1993: 562, fig. 3). In order to accept Finkelstein’s hypothesis, one has to ignore the remains of the mud-brick superstructure on top of the Wall 3097 stone foundations. More specifically, one has to assume that a large stone foundation was laid, then several courses of mud brick were laid on this foundation, and then a slightly narrower stone wall (also foundational) was constructed on top of these mud bricks, which in turn supported another massive mud-brick superstructure above them. As calculated from the debris of its collapse, the mud-brick superstructure that topped Wall 13004 must have risen yet another 5 or more meters. With regard to the occupational sequence, it is then necessary to presume that the room associated with Phase 8B Surface 3069 (the earliest occupation surface, which runs up to the top foundation stones of Wall 3097) had a southern wall with a brick base of more than 1 m high, above which rose the additional 2.5 m of the Wall 13004 stone structure. Above this, furthermore, would have been the 5-plus m of the Wall 13004 brick superstructure. In sum, the back wall of this structure would have risen well over 8 m above the Phase 8B occupation surface. Ironically, this would have left an exceedingly high wall with no countervailing internal fill to be balanced by the pressures of the external glacis buildup. Clearly, according to Finkelstein’s own argument, this would have been structurally untenable. To sum up, Finkelstein puts forward no real argument here. He simply offers a flat statement reiterating yet again his belief that Wall 3097 is only a lower foundation element for Wall 13004 and that “differences in construction have no chronological meaning.” Although he acknowledges that he looked at the published sections from Field IV Areas 3 and 16, which illustrate the internal and external stratigraphy against these walls (Seger 1975: fig 3; 1993: figs. 3–4), he does not comment on the crucial relevance of the remnants of the brick superstructure of Wall 3097 or on the phasing of the internal occupation levels. Thus, he adds nothing that would advance his case in any way.

57 Methods and Results The third major issue raised by Finkelstein involves the problems and disciplinary deficiencies that he believes were inherent to the field methods employed at Gezer. In his second section (“Reasons for the Gezer Debacle” [2002: 268–74]), Finkelstein outlines three “drawbacks” resulting from the use of the “baulk/debris-layer” method of excavation at Gezer (2002: 269): First, the excavation of limited sections hindered the possibility of deciphering the stratigraphic connection, date and function of large monuments such as Tower 5017 of the Middle Bronze and the gate and system of fortifications of the Iron II. Second, the concentration on narrow sections prevented the Gezer team from conducting modern research in the spirit of the processual archaeology, which requires the excavation of large enough fields. . . . Third, the “baulk/debris-layer” method prevented the excavations from uncovering significant assemblages of vessels from clearly defined layers.

In each case, Finkelstein is simply mistaken. First, even a casual look at the results of the HUC Gezer project’s 10plus seasons, involving work in 11 excavation fields at the site, belies these assertions. Regarding the South Gate complex, they are very specifically contradicted by the combined Field I and Field IV materials. In the case of Tower 5017 and the walls connecting it to the South Gate, an area of over 1,400 sq. m was exposed. The foundations of the tower, the glacis, and the connecting walls were tested in depth at three points externally and at four points internally. Second, with regard to “the conduct of modern research in the spirit of processual archaeology,” Finkelstein fails to acknowledge that the “New Archaeology” was only in its beginning stages in the early 1970s, when the excavations at Gezer were already coming to an end. Moreover, at the time, the work at Gezer—which in its later seasons included a wide range of material-culture sampling, such as the study of floral, faunal, and geological evidence—was in the vanguard of promoting these new avenues of research in Syria–Palestine. To be sure, the team grappled with addressing the methodological and theoretical gap between the new goals of the “New Archaeology” and with the problem of how best to deal with an ever-increasing database produced by this new approach (Dever 1981). But this was symptomatic of almost all excavations that adopted the methods of the “New Archaeology” in the “heady glow of the rhetorical fireworks of the 1960s” (Sabloff 1981: 2). Problems of this sort were in fact just beginning to be recognized in the early 1980s (Sabloff 1981), and this recognition led to the development of a number of methodological and theoretical movements concerned with explanation. These included, among others, the rigorous

James W. Hardin and Joe D. Seger

58 study of natural and cultural formation processes and their effects on the archaeological record (Schiffer 1976; 1983; 1987), ethnoarchaeological studies to improve understanding of past material culture (e.g., Binford 1978; Kramer, ed. 1979; Watson 1979), behavioral approaches that provided strategies for addressing both nomothetic and ideographic questions (Reid, Schiffer, and Rathje 1975; Schiffer 1976), neoevolutionary models of social change (Dunnell 1980), and, ultimately, the various “Post-Processual” movements (e.g., Leone 1982; 1986; Hodder 1986; Shanks and Tilley 1987). But all of these came after the HUC-sponsored Gezer excavation work had already been completed. Regarding broad exposures, the “New Archaeology” in fact moved away from this. It focused instead on retrieval of meaningful sample groups that could be analyzed in discrete detailed studies. Sampling strategies thus guided excavations in economizing the time, labor, and expense of field research (Trigger 1989: 310; see, e.g., Schiffer and House 1975). In any case, Finkelstein’s final charge that the method used at Gezer “prevented the excavations from uncovering significant assemblages of vessels from clearly defined layers” is quite specifically refuted by the work in Field IV. There, the wide exposure across the intramural room complex between Tower 5017 and the South Gate produced a most significant assemblage of pottery, artifacts, and other material culture elements documenting its final phase of use. Although not all of the Field IV data has yet been presented in final form, the preliminary reports seem sufficient to have made Finkelstein aware of its general character (Seger 1972a; 1972b; 1973; 1974; 1975; 1976; 1983; 1993). Overall, in advancing these criticisms of the Gezer excavation methods, Finkelstein merely offers a bland reiteration of the Tel Aviv School’s long-asserted architectural approach. Criticism from this quarter has frequently been countered over past decades, perhaps most effectively by Dever already in 1973 (1973a). Even more than three decades later, Finkelstein breaks no new ground, seeming not to grasp the importance of detailed stratigraphic observations provided by the balk/debris excavation method.

History Rectified In summary, we can do no more than simply restate our long-standing conclusions. The development and use of the South Gate complex occurred over an extended period, for at least one and one-half centuries. In the mid-17th century bc, a broad foundation trench was cut along the southwestern flanks of the site, removing preexisting tell strata almost to bedrock. Within this clearing, the massive foun-

dations of Tower 5017 were laid. Extending from its eastern end, the substantial but much narrower Wall 3097 was contemporaneously constructed. A mud-brick superstructure was added to the wall’s massive stone foundations. 6 At the same time, the internal foundation trench was backfilled and leveled in preparation for a first stage of use (Phase 8, City Stratum XIX). Then, sometime near the turn of the century, as part of the Hyksos Empire’s developing response to challenges from Seventeenth Dynasty Egyptians and/or other changing sociopolitical dynamics in Palestine, a second stage of fortification development was undertaken (Seger 1975: 44; Bunimovitz 1992; Redford 1992: 93–97). This included partial leveling of the Wall 3097 mud-brick superstructure, the construction of the 2.5meter-high stone foundations of Wall 13004, and the addition of its 5+-meter mud-brick superstructure. Associated with this rebuilding was the development and elevation of the South Gate tower structures and the addition of the exterior glacis to enhance and reinforce the slopes fronting the wall and gate system. Inside the wall system, this second stage involved the importation of additional preparation fills and the establishment of new Phase 7 (City Stratum XVIII) architecture. As the excavations in Areas 3 and 4 show, this was followed a generation or so later by additional localized rebuilding. Then, in the final phase of occupation, the central area was further redeveloped to serve as a broad open courtyard. This courtyard complex was in use for a somewhat longer time, until the city fell and was completely destroyed in the late 16th century bc. Finkelstein addresses the matter of dating the final demise of the South Gate complex in his penultimate paragraph on Middle Bronze Age Gezer (2002: 279). Again, he offers little more than rehearsal, citing the discussions on the subject by the Gezer staff through the mid-1970s. These considerations, which were based on the limited Field I remains and preliminary readings of the Field IV ceramic materials, allowed for a range of possible dates within the LB IA (1550–1475 bc). Although Finkelstein agrees with Seger’s proposed date of just before 1500 bc (1975: 42, 45), Dever’s suggestion that the fall of the city correlates with the campaigns of Thutmosis III several decades later (1990b) still remains viable. The study of the ceramics that mark the MB IIC/LB IA transitional period still remains fluid. Certainly, the appearance of Tell el-ºAjjul type bichrome and chocolate-on-white wares provide a 6. It is almost certain that Tower 5017 was also completed with a high mud-brick superstructure. While, as noted previously (Dever 1973b; Seger 1975), all evidence of this had been removed by quarrying activities in the Hellenistic and Roman periods, it must still be supposed that a gigantic podium of this sort carried high brick walls at least on its outer perimeters.

Gezer Rectified: The Dating of the South Gate Complex useful LB I benchmark. However, absolute dates for the first appearances of these fabrics are not sure, and in any case, their relative absence in the deposits at Gezer does not represent absolute chronological proof. Generally, a much more thorough comparative analysis of all the relevant MB IIC and LB I data related to this critical topic is still needed. In the case of Gezer, a better index for this discussion will be provided as the more complete corpus of the pottery sealed beneath the destruction debris in Field IV is published. The coauthors are currently at work on this project. Finkelstein closes his Middle Bronze Age section with a more positive contribution. His last paragraph (2002: 279–80) addresses issues relating to Gezer’s role within the dramatically expanded settlement system of the later Middle Bronze Age. He is almost certainly correct that in the mid-to-late-17th century, the city grew rapidly as the region’s main center. It also is reasonable to argue that a

59 growing population base provided the sizable labor force that was required to carry out the massive Inner Wall construction work. As argued elsewhere (Seger 1975: 43–45), the impetus for these dramatic changes correlates fully with the emergent polity of the Fifteenth Dynasty Hyksos in the Egyptian Delta and the accompanying growth of their political sway both in central Palestine and more generally throughout the southern Levant. Although it can be acknowledged that, allowing for hindsight, Finkelstein identifies some issues regarding the late Middle Bronze Age at Gezer that remain to be more fully addressed (1992), his brand of moralizing scholarship overall is neither convincing nor constructive. It is our hope that this review of his treatment of the Gezer South Gate complex will serve to rectify his faulty assessment of its stratigraphy and history and help to redirect future discussions to a more positive course.

References Binford, L. R. 1978 Nunamiut Ethnoarchaeology. New York: Academic Press. Biran, A. 1984 The Triple-Arched Gate of Laish at Tel Dan. Israel Exploration Journal 34: 1–19. 1990 The Middle Bronze Age Ramparts of Tel Dan. Eretz-Israel 21 (Ruth Amiran volume): 56–65, *103 (Hebrew, with English summary). Bunimovitz, S. 1992 The Middle Bronze Age Fortifications in Palestine as a Social Phenomenon. Tel Aviv 19: 221–34. Campbell, E. F. 2002 Shechem III/I. Boston: American Schools of Oriental Research. Dever, W. G. 1973a Two Approaches to Archaeological Method—The Architectural and the Stratigraphic. Eretz-Israel 11 (Dunayevsky volume): 1–8. 1973b Tower 5017 at Gezer: A Rejoinder. Israel Exploration Journal 23: 23–26. 1981 The Impact of the “New Archaeology” on Syro-Palestinian Archaeology. Bulletin of the American Schools of Oriental Research 242: 15–30. 1987 Palestine in the Middle Bronze Age: The Zenith of the Urban Canaanite Era. Biblical Archaeologist 50: 149–76. 1990a Of Myths and Methods. Bulletin of the American Schools of Oriental Research 277/278: 121–30. 1990b ‘Hyksos’, Egyptian Destructions, and the End of the Palestinian Middle Bronze Age. Levant 22: 75–81. 1992 Gezer. Pp. 998–1003 in vol. 2 of The Anchor Bible Dictionary, ed. D. N. Freedman. New York: Doubleday. 1993 Gezer. Pp. 496–506 in vol. 2 of The New Encyclopedia of Archaeological Excavations in the Holy Land, ed. E. Stern. Jerusalem: Israel Exploration Society. 1997 Gezer. Pp. 396–400 in vol. 2 of The Oxford Encyclopedia of Archaeology in the Near East, ed. E. M. Meyers. New York: Oxford University Press.

2003 Visiting the Real Gezer: A Reply to Israel Finkelstein. Tel Aviv 30/2: 259–82. Dunnell, R. C. 1980 Evolutionary Theory and Archaeology. Pp. 35–99 in vol. 3 of Advances in Archaeological Method and Theory, ed. M. B. Schiffer. New York: Academic Press. Finkelstein, I. 1981 The Date of Gezer’s Outer Wall. Tel Aviv 8: 136–45. 1985 Summary and Conclusions: History of Shiloh from Middle Bronze Age II to Iron II. Excavations at Shiloh 1981–1984: Preliminary Report, ed. I. Finkelstein, S. Bunimovitz, and Z. Lederman. Tel Aviv 12: 159–77. 1990 On Archaeological Methods and Historical Considerations: Iron Age II Gezer and Samaria. Bulletin of the American Schools of Oriental Research 277/278: 109–30. 1992 Middle Bronze Age ‘Fortifications’: A Reflection of Social Organization and Political Formations. Tel Aviv 19: 201–20. 1994 Penelope’s Shroud Unravelled: Iron II Date of Gezer’s Outer Wall Established. Tel Aviv: 21: 276–82. 2002 Gezer Revisited and Revised. Tel Aviv 29: 262–96. Gezer I = Dever, W. G.; Lance, H. D.; and Wright, G. E. 1970 Gezer I: Preliminary Report of the 1964–66 Seasons. Jerusalem: Hebrew Union College. Gezer II = Dever, W. G., et al. 1974 Gezer II: Report of the 1967–70 Seasons in Fields I and II. Jerusalem: Hebrew Union College. Gezer IV = Dever, W. G. 1986 Gezer IV: The 1969–71 Seasons in Field VI, the “Acropolis.” Jerusalem: Hebrew Union College. Hodder, I. 1986 Reading the Past. Cambridge: Cambridge University Press. Ilan, D. 1993 The Dawn of Internationalism: The Middle Bronze Age. Pp. 297–319 in The Archaeology of Society in the Holy Land, ed. T. E. Levy. New York: Facts on File.

60 Kempinski, A. 1972 Review of Gezer I, by W. G. Dever, H. D. Lance, and G. E. Wright. Israel Exploration Journal 22: 183–86. 1976 Review of Gezer II, by W. G. Dever et al. Israel Exploration Journal 26: 210–14. 1992a The Middle Bronze Age. Pp. 159–210 in The Archaeology of Ancient Israel, ed. A. Ben-Tor. Trans. R. Greenberg. New Haven: Yale University Press. 1992b Middle and Late Bronze Age Fortifications. Pp. 127–42 in The Architecture of Ancient Israel: From the Prehistoric to the Persian Periods, ed. A. Kempinski and R. Reich. Jerusalem: Israel Exploration Society. Kenyon, K. 1979 Archaeology in the Holy Land. New York: Norton. Kramer, C., ed. 1979 Ethnoarchaeology: Implications of Ethnography for Archaeology. New York: Columbia University Press. Lederman, Z. 1985 The Middle Bronze Age IIC Defense System. Excavations at Shiloh 1981–1984: Preliminary Report, ed. I. Finkelstein, S. Bunimovitz, and Z. Lederman. Tel Aviv 12: 140–46. Leone, M. P. 1982 Some Opinions about Recovering Mind. American Antiquity 47: 742–60. 1986 Symbolic, Structural, and Critical Archaeology. Pp. 415–38 in American Archaeology Past and Future, ed. D. J. Meltzer, D. D. Fowler, and J. A. Sabloff. Washington, DC: Smithsonian Institution. Mazar, A. 1990 Archaeology of the Land of the Bible, 10,000–586 b.c.e. New York: Doubleday. Parr, P. 1968 The Origins of the Rampart Fortifications of Middle Bronze Age Palestine and Syria. Zeitschrift des Deutschen Palästina-Vereins 84: 18–45. Pennels, E. 1983 Middle Bronze Age Earthworks: A Contemporary Engineering Evaluation. Biblical Archaeologist 45: 57–61. Redford, D. B. 1992 Egypt, Canaan, and Israel in Ancient Times. Princeton: Princeton University Press. Reid, J. J.; Schiffer, M. B.; and Rathje, W. L. 1975 Behavioral Archaeology: Four Strategies. American Anthropologist 77: 864–69. Sabloff, J. A. 1981 When the Rhetoric Fades: A Brief Appraisal of Intellectual Trends in American Archaeology during the Past Two Decades. Bulletin of the American Schools of Oriental Research 242: 1–6. Schiffer, M. B. 1976 Behavioral Archaeology. New York: Academic Press. 1983 Toward the Identification of Formation Processes. American Antiquity 48: 675–706.

James W. Hardin and Joe D. Seger 1987 Formation Processes of the Archaeological Record. Albuquerque: University of New Mexico Press. Schiffer, M. B., and House, J. H. 1975 The Cache River Archaeological Project: An Experiment in Contract Archaeology. Arkansas Archaeological Survey Research Series 8. Fayetteville: Arkansas Archaeological Survey. Seger, J. D. 1972a Notes and News: Tel Gezer (Spring 1972). Israel Exploration Journal 22: 160–61. 1972b Notes and News: Tel Gezer (Summer 1972). Israel Exploration Journal 22: 240–42. 1973 Notes and News: Tel Gezer. Israel Exploration Journal 23: 247– 51. 1974 The Middle Bronze IIC Date of the East Gate at Shechem. Levant 6: 117–30. 1975 The MB II Fortifications at Shechem and Gezer: A Hyksos Retrospective. Eretz-Israel 12 (Glueck volume): 34–45. 1976 Reflections on the Gold Hoard from Gezer. Bulletin of the American Schools of Oriental Research 221: 133–40. 1983 The Gezer Jar Signs: New Evidence of the Earliest Alphabet. Pp. 477–95 in The Word of the Lord Shall Go Forth: Essays in Honor of David Noel Freedman in Celebration of His Sixtieth Birthday, ed. C. L. Meyers and M. O’Connor. Winona Lake, IN: Eisenbrauns. 1993 Gezer: A Twice Told Tell, 1902–1909 and 1964–1974. Pp. 559– 74 in Biblical Archaeology Today, 1990: Proceedings of the Second International Congress on Biblical Archaeology, ed. A. Biran and J. Aviram. Jerusalem: Israel Exploration Society. Shanks, M., and Tilley, C. 1987 Re-Constructing Archaeology: Theory and Practice. Cambridge: Cambridge University Press. Stager, L. E. 1991 When Canaanites and Philistines Ruled Ashkelon. Biblical Archaeology Review 17/2: 25–43. Trigger, B. G. 1989 A History of Archaeological Thought. Cambridge: Cambridge University Press. Ussishkin, D. 1989 Notes on the Fortifications of the Middle Bronze II Period at Jericho and Shechem. Bulletin of the American Schools of Oriental Research 276: 29–53. Watson, P. J. 1979 Archaeological Ethnography in Western Iran. Viking Fund Publications in Anthropology 57. Tucson: University of Arizona Press. Yadin, Y. 1955 Hyksos Fortifications and the Battering Ram. Bulletin of the American Schools of Oriental Research 137: 23–31 1963 The Art of Warfare in Bible Lands. 2 vols. New York: McGrawHill.

An Early Iron Age I House with a Cultic Corner at Tall al-ºUmayri, Jordan Larry G. Herr Canadian University College Lacombe, Alberta

After the Early Bronze Age IV, the early Iron Age is probably the most important period of study in the scholarship of Bill Dever. Partly because of the interesting early Iron I finds at Tall al-ºUmayri, Bill has visited our site on numerous occasions and has sent several of his students to excavate with us. He has lectured to our group several times, and we have shared many a memorable hour in friendship sitting under the trees at our campsite or traveling from one site to another visiting excavations. It is an honor to publish a preliminary report on this outstandingly preserved house with its interesting religious overtones in a volume celebrating Bill’s distinguished career and scholarship.

The House The house is situated in Field B near the northwestern corner of the site, where our cut through the fortification system reaches the perimeter wall (labeled “Rampart” on fig. 1). Designated “Building A” (fig. 1), the house is immediately to the south of a partially reconstructed fourroom house, “Building B” (Herr and Clark 2001), with which it shares a common wall. Although the eastern limit of the house is not yet clear, the building was probably approximately the same length as Building B and, together with the latter, may have used the huge contemporary “refuse pit” to the northeast. This pit, filled with almost 25,000 meat-bearing bones and a heavy concentration of cooking pot fragments discolored by burning, extended at least another 5 m north of its location on fig. 1, as the excavations in 2002 showed. Author’s note: This paper is based on the excavations of the Madaba Plains Project (MPP)–ºUmayri directed by Douglas R. Clark and me, with Lawrence T. Geraty as senior consultant. MPP is sponsored by La Sierra University in consortium with Canadian University College and Walla Walla College. Douglas R. Clark and John I. Lawlor were the field supervisors responsible for excavating the house under discussion in this paper. The pottery plates were prepared under my supervision at various times by my assistants, Elena Reimche, Akasha Spenst, Heather Till, and Michaela Holkova. Warren C. Trenchard took the digital photographs.

Composed of five isolatable spaces (or “rooms”), Building A has external dimensions of approximately 17 m in length (using the eastern wall of Building B as the approximate eastern terminus) by 8.5 m at its widest point (Rooms A2 and A4 combined) (fig. 2). The house was much better preserved in the west than the east (fig. 3). Indeed, 1.5 m of roofing mud and brick walls fell into Rooms A2 and A3, suggesting that, with the surviving walls preserved at 1.5 m high when excavated, there must have been a second story in the western part of the house. The wall between Rooms A2 and A3 has lost two to four courses during the years since excavation, especially near its northern limit. As in the case of Building B, the inhabitants most likely entered the house from the east into a courtyard (Room A5) with a possible animal pen or other structure in the southeastern corner, supported by posts set on at least three surviving bases (fig. 3). Unlike Building B, however, there were no paving stones close to the post bases (compare with fig. 1). It is as yet uncertain where the southern extent of this probable courtyard was located. The wall at lower left in fig. 3 appears to be later, and the wall at lower right may be the eastern wall of Room A5, but it seems to stem from earlier phases. The western wall of the room is only weakly attested (the wall in fig. 3 is a later wall, but fragmentary remains below it may be a thin curtain wall [fig. 2]). A thick accumulation of pottery was found on the surface in the northern part of the room, to the north of the post bases. The relatively thick beaten-earth surface (or surfaces—there were at least two phases) was hard, with a high clay content. The doorway into Room A1 may have been in the southwestern corner of Room A5, although the separation of Rooms A1 and A5 may have been similar to the space between the courtyard and house of Building B, which we suggest was an open wall, perhaps hung with a curtain or rug (Herr and Clark 2001: especially the illustrations on p. 40). Room A1, with a beaten-earth surface, was clearly domestic in nature, containing two bins and other domestic

62

Larry G. Herr

Fig. 2. Building A (with part of Building B to the north). The perimeter wall is on the left and two late Iron I walls form a corner just inside the perimeter wall, to the southwest of Room A4.

Fig. 1. Iron I remains at Tall al-ºUmayri in Fields A (southern half) and B (northern half). Building B and half of Building A were in Field B. Both houses and the refuse pit to the northeast date to the very early Iron I period. South of Building A are late Iron I features, such as the “Pillared Room” and the small walls immediately to its northwest. The perimeter wall, however, was in use in early Iron I, along with Buildings A and B.

finds (compare figs. 3 and 4). The first was a pie-shaped bin paved with pebbles in the northwestern corner, while the second was a square structure, divided into two parts, along the south wall. Southeast of the first bin were the fragmentary remains of a hearth, near which a grinding stone was found. A concentration of smashed pottery lay on the surface between the hearth and the square bin (fig. 5). Room A2 was separated from Room A1 by three post bases that sat on top of the eastern edge of the flagstone pavement of Room A2 (figs. 2, 4). The most remarkable feature of this room was the standing stone leaning against the western wall (figs. 3–4). It was a tall (ca. 1 m high), almost perfectly rectangular natural stone of hard limestone coated with a veneer of calcrete. Many stones at ºUmayri

Fig. 3. Photograph of Building A looking west, with Room A5 in the foreground separated from Room A1 by a later wall (the original separation between Rooms A5 and A1 lies beneath this wall). Room A1 has a beaten-earth surface, while Room A2, behind, with the standing stone, is paved with flagstones. Room A4 is on the upper left and Room A3 at top right. The northern wall of the building is visible at top right.

An Early Iron Age I House with a Cultic Corner at Tall al-ºUmayri, Jordan

63

Fig. 4. The western four rooms of Building A, looking north. To the right is Room A1; at the top center is Room A2 with its standing stone; at the bottom center is Room A4; and at the upper left is Room A3.

have this veneer of calcrete, but none has it on all sides; nor are any others as perfectly formed. Clearly, the stone was chosen for its natural form and apparently placed in a position of veneration. Standing stones of this sort are normally associated with divinities and are often placed near gates— for example, the standing stones in Iron II contexts at Dan (Biran 1994: 244; Zevit 2001: 191–96, with references) and Bethsaida (Arav, Freund, and Schroder 2000: 49–50; Zevit 2001: 149–53, with references), among others. In front of the standing stone was another, similar stone, but more oval in shape. Also completely covered with a veneer of calcrete, it lay flat in front of and at 90 degrees to the standing stone, as if it were intended as a presentation altar. Unfortunately, no other cultic remains were found on the flagstone floor. A pile of carbonized barley was found immediately to the south of the standing stone. The floor of the southern part of Room A2, apparently where it formed a hallway between Rooms A1 and A3, was made of beaten earth (fig. 4). A pair of stones somewhat higher than the flagstones to the north formed the edge of the stone paving. A small alcove, Room A4, protruded from the southern wall of the building (figs. 4, 6). The beaten-earth surface of the “hallway” south of Room A2 continued into the northern portion of Room A4 but was interrupted by a line of long stones laid in a somewhat chaotic stretcher fashion with other stones. These stones are, however, laid too haphazardly to constitute a wall, and it is still unclear whether they were laid on the surface or whether the surface runs up to them. Behind this line of stones was a group of seven flat-lying stones similar to but smaller and more amorphously shaped than the standing stone in Room A2. An-

other stone was found leaning against the southern wall of the alcove. Are these a series of standing stones that were no longer in use, or were they stored in the alcove awaiting use? Are they simply paving stones in the alcove? If so, they are unlike any other stones at ºUmayri from this, or any other, period. The western room of the building, Room A3, used the perimeter wall for its western boundary. It was entered from the “hallway” between Rooms A2 and A4 at its southeastern corner. Immediately opposite the entrance was a plastered platform with a series of three small, narrow steps on the northern side (fig. 4; this is clearer in Clark 1994: 143, photo 11). We have long interpreted it as a platform for a ladder ascending to the second floor, preferring a relatively mundane interpretation to other, more exotic possibilities. But because Building B had no such platform for ladders, this interpretation no longer seems as likely. Instead, a less mundane interpretation may make better sense: it may have been a small altar associated with the cultic features in Rooms A2 and A4 to the east. However, there was no sign of burning on the hard plaster that topped the stones. Moreover, the room must have been covered and, being an inner room, would not have allowed for smoke to exit easily. The northern part of the room was used for storage immediately prior to its destruction: two well-preserved collared pithoi stood against the northern wall, with four or five others smashed in front of them (Clark 1994: 144, photo 13). The overall plan of the building seems to be unparalleled. It certainly does not belong to the common group of houses usually divided into three or four rooms (like the neighboring house, Building B). Instead, it appears to

64

Larry G. Herr

Fig. 5. Concentration of pottery on Surface 7K71:51 in Room A1 after the northern balk of Square 7K71 was excavated in the 2000 season. Smaller concentrations of pottery had been found on either side of the balk in earlier seasons.

Fig. 6. Room A4 with a line of stretcher stones in front of a group of seven flat-lying, long stones, with another propped against the back wall.

consist of four rooms in a line, with a small alcove projecting to the side. This plan may be better understood when we are able to excavate outside the structure to the south in future seasons. Like Building B to the north, Building A met with a violent destruction that burned several of the stones in the walls to lime, and almost all the stones were charred gray and cracked by the heat of the fire. The carbonized barley was so hard that we needed a hammer to break the grains. All the rooms, but especially those in the west, were filled with several layers of destruction. First came the ceiling

and roof debris, about 0.4–0.5 m deep, which included a concentration of pottery and artifacts that fell from the second story and roof. On top of this debris, and containing almost no pottery or artifacts, were the fallen mud bricks of the second story. This destruction debris was deepest in Rooms A2 and A3, where it was about 1.3–1.5 m deep and included clear examples of stacked bricks fallen at various angles from the upper-story walls. The destruction layer sloped down toward the east (and somewhat to the south) in Rooms A1 and A4 until, in Room A5, the walls were only 0.4–0.5 m high, suggesting that, although there were

An Early Iron Age I House with a Cultic Corner at Tall al-ºUmayri, Jordan

Fig. 7. Group 1: Pithos from Room A3 (Clark 1997: fig. 4.14:1).

probably two stories in Rooms A3 and A2, only one existed in the eastern parts of the house and, perhaps, in Room A4.

The Pottery Concentrations of pottery were found on the floors of four rooms of the house. The first includes the collared pithoi and other vessels found in the northern half of Room A3. The pithoi were embedded in the surface of the room, while the rest of the pottery, including many more rims from other pithoi, came from the roof debris (figs. 7– 15). A second, much smaller concentration of sherds from collared pithoi was found just east of the door between Rooms A2 and A3. Because they were found in the roof debris, they and the other vessels found with them must have fallen from the second story, probably above Room

65

Fig. 8. Group 1: Pithos from Room A3 (Clark 1997: fig. 4.16:1).

A2. The third concentration was a thick scatter across the floor of Room A1 (fig. 5), and the fourth was at the northern end of Room A5. Pottery from the second story was found only in Rooms A3 and A2. There was probably no second story above Rooms A1 and A5. We excavated all these ceramic concentrations as the bottom parts of the destruction layer, interpreting them as vessels that were in use in the structure when it was destroyed. Thus, we are not in these instances able to separate second-story groups from those of the first story. All four groups contained pottery similar to the pottery found on the floors and in the destruction debris of Building B. The cooking pots were similar to the pots found in the refuse pit to the northeast. All the pottery concentrations were surrounded by destruction debris that included clay, silt, and roofing beams, as well as hundreds of smaller carbon bits and ash lenses from the construction material of the ceilings and/or roofs.

66

Larry G. Herr

Fig. 9. Group 1: Pithos from Room A3 (Clark 1997: fig. 4.17:1).

Fig. 10. Group 1: Pithos from Room A3 (Clark 1997: fig. 4.18:1).

Group 1: Pottery from Room A3, First and Second Stories (figs. 7–15) As mentioned above, the largest group was found in Room A3. The four complete pithoi (figs. 7–10) came from the surface, while the rest of the pottery (figs. 11–15) was found in the collapse of the roof debris and thus probably originated on the second story. The four complete pithoi along with the rims of vessels from the ceiling collapse (figs. 11–12) represent a wide range of rim and collar forms. The classic early Iron I pithos shape is seen in fig. 7, with a long neck; flaring rim; single, distinctive collar at the base of the neck; and a relatively pointed base. The other three complete pithoi are more bag-shaped, with handles lower on the body; a triangular-shaped rim; wavy, extended collar; and relatively blunt base. The wavy collars are quite unique. The rims and collars in figs. 11–12 illustrate a similar diversity of form. Virtually all rims are relatively long, flaring, and thickened on the exterior. The rims go from slightly thickened simple forms (figs. 11:1–

2; 12:2, 7) to a rounded thickening (figs. 11:3–4; 12:5), triangular thickening (fig. 11:5–10), and rectangular thickening (fig. 12:1, 3–4, 6). Several of these forms sport a ridge below the rim (fig. 12:1–2, 6–7). Collars are just as diverse. The common form with a simple ridge low on the neck is present (figs. 11:1–2, 9; 12:1–2, 7), sometimes with a thickened collar (fig. 12:2, 7), but wavy varieties (figs. 11:5–8, 10; 12:5–6) and double collars (fig. 12:3–4) also occur. Despite this huge variety of forms, there are no examples with the short neck and upright rim that became dominant later in Iron I. The variety of forms also suggests a multiplicity of potters, most likely itinerant, who manufactured vessels at the site. Each complete pithos weighs approximately 40–50 kg when empty. They would be much too awkward as trade items or as vessels to transport traded goods. Most of the fragmented (or partially fragmented) pottery in Room A3 came from the ceiling collapse (figs. 13– 15). Like the pithoi, the jar rims are flaring (fig. 13:1, 3),

An Early Iron Age I House with a Cultic Corner at Tall al-ºUmayri, Jordan

67

Fig. 11. Group 1: Pithoi from the second story of Room A3 (Clark 1997: fig. 4.19).

Fig. 12. Group 1: Pithoi from the second story of Room A3 (Clark 1997: fig. 4.20).

except for one that is everted (fig. 13:4). Jugs also tend to have flaring rims (figs. 13:6–8; 14:1–6, 8) and handles that rise slightly above the rim (fig. 13:6–8). Most jar and jug bases tend to be rounded or small and flat (figs. 13:2, 5; 14:9–10). Krater bases were most often made with a wide, round bottom and an added ring (fig. 14:11–12). They are normally very thick with the ring added in a crude fashion. Bowls include a carinated form (fig. 14:14) and two examples with a squared, in-curving rim (fig. 14:15–16). One large and two small shallow bowls with strongly everted rims were probably chalices (fig. 14:17–19), perhaps used in cultic activities associated with Room A2 or the platform in Room A3, but stored on the upper story. One bowl base is simply a smaller form of the common krater base (fig. 14:20). The cooking pots are direct descendents of Late Bronze Age carinated forms with out-turned, triangular rims. But the outer edges of our vessels are beginning to lengthen into the Iron I flanged form (fig. 14:21–25). No rims show any hint of moving toward the inward-leaning stances so common later in Iron I.

The single example of a lamp (fig. 15:1) is typical of lamps from the central Jordan plateau, with a carinated upper body and a slightly flaring rim opposite the pinched nozzle. The flask (fig. 15:2), with its high neck and pieshaped decoration, is also typical of the central Jordan plateau. The basket-handled jug with a wide, sieved spout (fig. 15:3) is an unusual element in the corpus. The large numbers of pithoi from both stories indicate storage on both levels of the broadroom along the perimeter wall, similar to the function of the corresponding rooms in Building B (Herr and Clark 2001). However, the presence of more unique forms, such as the chalices, the flask, and the basket-handled jug, may be associated with ceremonial or other specialized activities in the house. Group 2: Pottery from Room A2, Second Story (figs. 16–18) None of the pottery in this group came from the flagstone pavement of Room A2 but from the ceiling and roof debris immediately above the floor, and consequently it originated on the second story. Although there are new

68

Larry G. Herr

Fig. 15. Group 1: Pottery from the second story of Room A3 (Clark 1997: fig. 4.28).

Fig. 13. Group 1: Pottery from the second story of Room A3 (from Clark 1997: fig. 4.26).

Fig. 14. Group 1: Pottery from the second story of Room A3 (Clark 1997: fig. 4.27).

forms in this group, it is identical in ware and character to the preceding group. The pithos (fig. 16:1) has a gently grooved triangular rim (much like fig. 11:9), a long, flaring neck, and probably a single-ridge collar at the base of the neck. One of the jars (fig. 16:4) has a similar rim form to a jar in Group 1 (fig. 13:1). The other jar (fig. 16:3) has a flaring triangular rim, typical of jars from our building. When the amphora in fig. 16:5 was discovered, we believed that it had only one handle, and it was initially published as such (Clark 2000: fig. 4.30:16). However, during the 2000 season, the left side of the vessel was found together with a second handle. Given the presence of a second amphora (fig. 16:2), I suspect that the vessel in fig. 17:1 (only half of which was found) is similarly two-handled. The squat form of this vessel type is not frequent. Other jar rims (fig. 17:2–3) have everted or flaring rims. The jug rims are similar (fig. 17:4–6) but are mostly simple or thickened in a square or slightly triangular shape. As in the case of the Group 1 jugs, the single handle rises slightly above the rim. The single juglet (fig. 17:7) is a typical rim form of a dipper juglet. Also unique at the site is the oil-separation krater (fig. 17:8) with a spout near the top and a drain hole near the bottom. Even the rim form, with its grooved rectangular thickening, is not the usual krater rim of this period. The presence of an oil-separating krater may indicate agricultural activities but could just as well suggest a cultic interpretation, because oil from the first pressing was an important part of many cultic activities (Zevit 2001: 139, n. 28). The two bowls (fig. 17:9–10) are similar to those in Group 1. Because its rim is slightly out-turned, fig. 17:2

An Early Iron Age I House with a Cultic Corner at Tall al-ºUmayri, Jordan

69

Fig. 16. Group 2: Pottery from the second story of Room A2 (from Clark 2000: figs. 4.29–31).

could represent another chalice, but the rim is not as strongly everted as other examples (fig. 14:17). The squared rim of fig. 17:3 is from a shallower bowl than the bowls in Group 1 (fig. 14:15–16). The small cup (fig. 17:11) could also come from a narrow jug rim. The cooking pots (figs. 17:12–15; 18:1–2) are very similar to those of Group 1 (fig. 14:21–25). The bases in fig. 18:3–4 come from a pithos and krater, respectively. Although there was a cultic corner in Room A2, the pottery appears to be mostly domestic in function; a few items may have been used for specialized activities. Group 3: Pottery from Room A1, First Story (figs. 19–20) Most of this pottery was found on the surface level of Room A1 and reflects the pottery in use on the main floor of this room. There was probably no second story in this location. Only four pithoi were found, but they exhibit the same variety of forms as the examples in Group 1. There are three varieties of triangular rims (fig. 19:1–3) and a wavy form with a slight interior thickening (fig. 19:4). The fig. 19:3 ex-

ample looks similar to a Middle Bronze Age pithos rim, but it is of the same ware as the other pithoi from our period. The rims of all the jars (fig. 19:5–13) are flared and thickened in rounded, rectangular, or triangular sections. Most of the jugs (fig. 19:14–17) are similar, except for fig. 19:17, which may be a small chalice. The rim of the juglet (fig. 20:1) flares more than the example in Group 2 (fig. 17:7). The kraters (fig. 20:2–5) with a rounded upright rim are unlike other kraters at the site, the rims of which tend to have exterior and interior thickening in a variety of triangular shapes. Carinated bowls occur (fig. 20:7–9), as well as more simple rims (fig. 20:10–11, 15). One example is thickly everted (fig. 20:12). Two examples of pyxides were found (fig. 20:13–14), both similarly shaped, with painted metopes on the shoulders. Only two cooking pots (fig. 20:16–17) were discovered, both very similar to those found in Groups 1 and 2 (figs. 14:21–25; 17:12–15; 18:1). Neither of the two lamps (fig. 20:18–19) is of the carinated form, and both display a wider flange than the lamp in Group 1 (fig. 15:1). The large number of jars seem to correspond functionally with the two bins in Room A1. The small number of

70

Larry G. Herr

Fig. 17. Group 2: Pottery from the second story of Room A2 (from Clark 2000: figs. 31–32).

Fig. 18. Group 2: Pottery from the second story of Room A2 (from Clark 2000: figs. 31–32).

pithoi may suggest that they had fallen eastward from the second story over Room A2 onto the floor of Room A1. The presence of several kraters in this but not the other rooms may suggest specialized activity patterns, perhaps

food preparation in association with the hearth and grinding stones. However, the relative absence of cooking pots in the room that contained a hearth is surprising. It may suggest that, although cooking pots were used with fires,

An Early Iron Age I House with a Cultic Corner at Tall al-ºUmayri, Jordan

71

Fig. 19. Group 3: Pottery from the floor of Room A1 (from Clark 2000: figs. 4.29–31; Lawlor forthcoming a; forthcoming b).

they were used elsewhere for eating or were stored in other locations. Group 4: Pottery from Room A5, First Story (fig. 21) The pottery from Room A5 was found on the floor, to the north of the possible post bases. As would be expected in a location far from the broadroom at the western extent of the house, no pithoi were found in Room A5, but the jars and/or amphorae in Group 4 (fig. 21:1–7) are very similar to those in Group 3 (fig. 19:5–13), with flaring, slightly thickened rims. The jugs (fig. 21:8–10), however, are significantly different. Indeed, the fig. 21:9–10 examples flare so strongly that, had these vessels not been found on the floor, we would be tempted to date them to the Middle Bronze Age. Note also the painted rim on fig. 21:9, which closely resembles “chocolate-on-white” ware, but there is no trace of a white slip. The two kraters (fig. 21:11–12) in Group 4 are of the more common type that we find in early Iron I layers. The rims are thickened on the exterior in triangular shapes, and one example (fig. 21:12) also has a small thickening on the

interior. Only two bowls (fig. 21:13–14) are present: fig. 21:13 is similar to a bowl in Group 3 (fig. 20:15) and fig. 21:14 sports a heavy hammer-headed rim. One cooking pot (fig. 21:15) is similar to our other forms, but the other (fig. 21:16) is totally unique. Indeed, were it not for its obvious cooking-pot ware and presence on the floor of our building, it would seem to us like a small Byzantine pithos. Odd for this period is also its holemouth shape. The high base (fig. 21:19) and the painted body sherd (fig. 21:20) that seems to depict trees or a lattice pattern are more at home in the Late Bronze Age and may be remnants from that period. Parallels Although the best parallels come from ºUmayri itself, especially Field B (Herr 1997: 237–44; 2000: 187–97; 2002: 138–47), they are also found elsewhere in late13th–early-12th-century deposits (see the many parallels presented in the articles cited above). The strongest evidence for dating purposes is the cooking pots, with their everted stance and flanged rim. The relatively lengthy

Larry G. Herr

72

Fig. 20. Group 3: Pottery from the floor of Room A1 (from Clark 2000: figs. 4.31–32; Lawlor forthcoming a; forthcoming b).

flange on the rim is later than most Late Bronze Age assemblages, but the everted stance is earlier than most 11thcentury groups. All the collared pithoi sport a high neck, flaring rim, and low, strong collar. These traits are all typical of the late-13th–12th-century assemblages but are not found in later deposits (Herr 2001).

Conclusions The four groups of pottery from Building A are from the same time period and strongly indicate a domestic function of the building, with hints of other specialized activities, perhaps ceremonial in nature. The architectural

features of the building also suggest this dual function. Whether the ceremonial activity was performed solely for the inhabitants of the house or for others in the community, however, cannot be determined. The date of this corpus of pottery suggests a time span within Iron I (some of the cooking pots clearly postdate the Late Bronze Age), but very near its beginning. Consequently, we have dated this assemblage to the very end of the 13th and the first half of the 12th centuries bc. The pottery from Building B and the refuse pit elsewhere on the site is virtually identical to the pottery from the house studied here (Clark 1997; 2000).

An Early Iron Age I House with a Cultic Corner at Tall al-ºUmayri, Jordan

73

Fig. 21. Group 4: Pottery from the floor of Room A5 (from Lawlor forthcoming a; forthcoming b).

References Arav, R.; Freund, R. A.; and Shroder, J. F. 2000 Bethsaida Rediscovered. Biblical Archaeology Review 26/1: 44–56. Biran, A. 1994 Biblical Dan. Jerusalem: Israel Exploration Society. Clark, D. R. 1994 The Iron I Western Defense System at Tell el-ºUmeiri, Jordan. Biblical Archaeologist 57: 138–48. 1997 Field B: The Western Defensive System. Pp. 53–98 in Madaba Plains Project: The 1989 Season at Tell el-ºUmeiri and Vicinity and Subsequent Studies, ed. L. G. Herr et al. Madaba Plains Project Series 3. Berrien Springs, MI: Andrews University Press. 2000 Field B: The Western Defensive System. Pp. 59–94 in Madaba Plains Project: The 1992 Season at Tall al-ºUmayri and Vicinity and Subsequent Studies, ed. L. G. Herr et al. Madaba Plains Project Series 4. Berrien Springs, MI: Andrews University Press. Herr, L. G. 1997 The Pottery. Pp. 228–49 in Madaba Plains Project: The 1989 Season at Tell el-ºUmeiri and Vicinity and Subsequent Studies, ed. L. G. Herr et al. Madaba Plains Project Series 3. Berrien Springs, MI: Andrews University Press. 2000 The Pottery. Pp. 185–203 in Madaba Plains Project: The 1992 Season at Tall al-ºUmayri and Vicinity and Subsequent Studies, ed. L. G. Herr et al. Madaba Plains Project Series 4. Berrien Springs, MI: Andrews University Press.

2001 The History of the Collared Pithos at Tell el-ºUmeiri, Jordan. Pp. 237–50 in Studies in the Archaeology of Israel and Neighboring Lands in Memory of Douglas L. Esse, ed. S. R. Wolff. Studies in Ancient Oriental Civilization 59. Chicago: Oriental Institute. 2002 The Pottery. Pp. 135–55 in Madaba Plains Project: The 1994 Season at Tall al-ºUmayri and Vicinity and Subsequent Studies, ed. L. G. Herr et al. Madaba Plains Project Series 5. Berrien Springs, MI: Andrews University Press. Herr, L. G., and Clark, D. R. 2001 Excavating the Tribe of Reuben. Biblical Archaeology Review 27/2: 36–47. Lawlor, J. I. Forthcoming a Field A The Western Side. In Madaba Plains Project: The 1996–1998 Season at Tall al-ºUmayri and Vicinity and Subsequent Studies, ed. L. G. Herr et al. Madaba Plains Project Series 6. Berrien Springs, MI: Andrews University Press. Forthcoming b Field A: The Western Side. In Madaba Plains Project: The 2000 Season at Tall al-ºUmayri and Vicinity and Subsequent Studies, ed. L. G. Herr et al. Madaba Plains Project Series 7. Berrien Springs, MI: Andrews University Press. Zevit, Z. 2001 The Religions of Ancient Israel. New York: Continuum.

Aphrodite/Astarte on Horseback Vassos Karageorghis Foundation Anastasios G. Leventis Nicosia, Cyprus

The iconography of the “Great Goddess” of Cyprus enjoys a long tradition in the island’s coroplastic art. I should state from the outset, however, that although I refer to this female figure as a “goddess,” I am fully aware that the idea of a god or goddess may not be the same in later periods as it was in the prehistoric periods. In her earliest appearance (in the fourth millennium bc, the Chalcolithic period), Aphrodite is represented as a nude female figure in various positions, standing or seated, occasionally pregnant or giving birth, squeezing her breasts from which milk flows into a basin placed on her lap, and touching her nipples in a gesture of fertility and regeneration. The same iconography, more or less, persists during the Early and Middle Bronze Ages (third–early-second millennium bc), with the additional novelty of a nude female figure holding an infant (kourotrophos). One must also mention some rare examples of a male and female couple, shown embracing one another (hieros gamos or ‘sacred marriage’). This appears in the decoration, both relief and in the round, of large Red Polished ware vases. During the Late Bronze Age, the predominant Aphrodite/Astarte types are nude female figures with accentuated genitals in Base-ring ware. Variations include standing or seated figures, some holding their breasts with both hands and others holding an infant. In many cases, the Near Eastern influence is apparent in the iconography (especially what we consider “bird-faced” figures), but influences from Aegean art are distinguishable in other figurines (V. Karageorghis 1993: 21–23). In the Aegean, we have a rare type that is relevant to the specific case we shall discuss in this paper—a female figure seated sideways on a saddled horse or other equid. One such figurine found at Attica is now in the Stathatos Collection of the National Museum in Athens. There is no reason to question its authenticity (for a discussion and references, see Crouwel 1981: 52, pl. 46). Found in a tomb, it probably dates to the 14th–13th centuries bc; the Author’s note: I am grateful to Lawrence Stager, Nota Kourou, Georgos Rethemiotakis, and Joost Crouwell, who read this brief essay, for their very useful suggestions and additional bibliographic references.

female figure is represented seated sideways on the back of a saddled equid, with her arms raised in the fashion of the well-known Minoan “goddess with up-raised arms.” The figurine was first published by Levi (1951), who labeled it “Dea Micenea a Cavallo,” the name by which it is known in the archaeological literature. Two other figurines, from a looted Late Minoan III tomb at Archanes on Crete, represent a female figure with one arm stretched out, touching the top of an equid’s mane (Sakellarakis and Sapouna-Sakellaraki 1997: 522–23). The figures are seated sideways on the back of the saddled equid. The horse wears a collar, which was probably fastened to the saddle to keep it in place. Several other similar representations are known from the Aegean (Sakellarakis and Sapouna-Sakellaraki 1997: 524). It is not clear whether there is a continuation of this type in later periods in the Aegean, but we know of a similar figurine from the Geometric period found in a child’s burial at Gavalomouri, Crete (Andreadaki-Vlasaki 1987: 315–16, pl. 6.1, with further bibliography; see also below). On Cyprus, there are as yet no examples to speak of with regard to Late Bronze Age female figures riding astride a saddled horse or other equid. Two similar examples in “Base-ring” ware, dating to ca. the 14th–13th centuries bc, include a complete male rider seated sideways on a saddle with arched bows fore and aft (fig. 1). The horse wears a collar shown across its chest and is harnessed (V. Karageorghis 1980). The second saddled horse displays a similar arched saddle, but the rider is not preserved (V. Karageorghis 1980); it is currently in the collection of the John and Mabel Ringling Museum of Art in Sarasota, Florida. Questions arise regarding how this particular type reached Cyprus. When I first published these figurines, I did not exclude the possibility that they may have been inspired by Aegean prototypes, as is evident in their striking similarity to the two Aegean examples cited above. Further evidence of an Aegean “artistic provenance” are several seals from this region bearing a female figure wearing a Minoan flounced skirt, seated sideways on a mythical beast known as the Minoan dragon (Crouwel 1981: 51). I agree with Crouwel, however, that the

76

Vassos Karageorghis

Fig. 1. “Base-ring” ware horse-and-rider figurine (inv. no. 1979/XII-8/2, Cyprus Museum, Nicosia).

sideways seat is more frequent and appears earlier in the Orient (Crouwel 1981: 51). I should also mention a terracotta figurine, probably from the Proto-Geometric period, found in a cave at Inatos, Crete (Marinatos 1996: 136, fig. 2). I suggest that this particular type had a parallel development in both regions and that the iconography was adopted by women and dignitaries who wore long robes, the elderly and the infirm, thus explaining the figures’ stance with both legs on one side of the horse. A terra-cotta figurine representing a female figure riding sideways on a horse was recently found in a sanctuary of Artemis at Ano Mazaraki in Achaia (Touchais, Hubert, and Phillipa-Touchais 2001: 867, fig. 92). It may date to ca. 700 bc. Whether this is a distant descendant of the Mycenaean type is difficult to determine. A second figurine of bronze was found at Olympia (Crouwel 1981: 51, pl. 115). Even today, this fashion of riding is widely adopted, not only among women riding a mule or a donkey in Greece and Cyprus, but even by the Queen of England in the “Trooping the Colour” procession (cf. Crouwel 1981: 51; Crouwel and Tatton-Brown 1988: 84–85). A nude female figure is known from the Late Bronze Age in Egypt and Canaan. She is depicted as a war-goddess, sitting on a galloping horse. But she is also a goddess of love, fertility, and healing, often related to Astarte (for a discussion and references, see Clamer 1980: 159–60). In Egypt and the Near East, there are several representations of the goddess Astarte riding sideways, nude or draped, both in terra-cotta and other materials (e.g., Leclant 1960: 23, pl. 1A, 41–42, figs. 18–19; Littauer and

Fig. 2. Bichrome III–IV hollow figurine of an enthroned “Astarte” (in a private collection in Nicosia).

Crouwel 1979: 96–97; Badre 1980: 99, pl. 47:200–201). Occasionally armed like the Hittite bisexual divinity Pirva of the second millennium bc (Otten 1953; Leclant 1960: 24–25, 60–61), Astarte continues to personify a warlike divinity in the first millennium bc (Lipinski 1995: 131–32, 135–36). In Syria and Cyprus, a nude female figure is depicted on metallic parts of horse ornaments (Orchard 1967: figs. 135–42; V. Karageorghis 1974: 83–84), thus continuing her role as a war-goddess. An unusual representation of the nude goddess standing on a horse appears on a Late Bronze Age terra-cotta mold found on the surface of Tel Qarnayim, near Beth-Shean (Ben-Arieh 1983, with references). Like some of the Egyptian steles that highlight the triad of Qudshu flanked by Min and Reshef, this mold features a triad of deities, with a large nude goddess standing on a horse in the center of the representation; two much smaller warrior-gods stand at her side, just below each of her elbows, brandishing weapons and emblems. On an Iron Age II scarab found near Acco, we see a clothed goddess riding a horse side-saddle (Keel

Aphrodite/Astarte on Horseback

77

Fig. 3. Cypro-Archaic I terra-cotta figurine of an enthroned “Astarte” (inv. no. 12204, National Archaeological Museum of Athens).

Fig. 4. Cypro-Archaic I terra-cotta figurine of an enthroned “Astarte” from Ayia Phyla Tomb 7, no. 81 (Limassol District Museum).

and Uehlinger 1998: ill. 163a). This is probably an EgyptoPhoenician representation of Astarte as a war-goddess. The warlike character of Aphrodite is echoed in Homer’s Iliad, in which Aphrodite shows her warlike character in siding with the Trojans, the enemy of the Greeks, in opposition to Athena, who sides with the Achaeans. This is relevant to current theories regarding the Oriental provenance and, more specifically, the Cypriote origin of Aphrodite (J. Karageorghis and V. Karageorghis 2002). This mode of riding may be connected with religious ritual, as suggested by a rhyton of Proto-White Painted ware (11th century bc) representing a nude male figure riding sideways on a horse beneath a canopy decorated with coiled serpents. Because in almost all the cases in which the rider is seated sideways there are no reins in their hands, the horse may have been led by a second person, perhaps in a procession. This rhyton was found in all probability in a tomb at Palaepaphos. Catling suggests that, if this is a grave offering, it may represent the deceased himself, setting out on his journey to the next world, accompanied by a packhorse carrying sustenance for this journey; symbolically, the snakes are a reminder of immortality (1974: 111). Riding side-saddle continued to appear in iconography in Greece, Cyprus, and Phoenicia throughout the first mil-

lennium bc, during the Cypro-Geometric and CyproArchaic periods (Hermary 1984: 292, nos. 8–9; Yon and Caubet 1985: 17, no. 53; V. Karageorghis 1993: 89–90). In some of these representations, the rider plays a lyre (see, e.g., V. Karageorghis 1993: pl. 39:1), which may suggest that these figures are associated with a special feast or ceremony. I have published a terra-cotta figurine of a female figure riding sideways on a wheeled horse; rather than seated on a saddle, the figure sits on a richly ornamented throne, replete with a footstool. She wears a crown of flowers on her head and holds a small quadruped in her lap (fig. 2). This terra-cotta, now in a private collection in Nicosia, may date to the Cypro-Geometric period and was probably found in Palaepaphos (V. Karageorghis 1997). The goddess (or her priestess [see below]) wears necklaces and displays a red spot on each cheek; she wears a “scarf” above her left shoulder, which may form part of her hieratic dress (cf. J. Karageorghis 1999: 198). I have previously stressed the association of Aphrodite/Astarte with horses; here, however, she is not connected with war but with a particular ceremony: she is the goddess who has already been endowed with her sensual attributes and has become the principal goddess of fertility (Lipinski 1995: 132). In addition to the novelty of Aphrodite/Astarte riding

Vassos Karageorghis

78 on horseback, she is shown placed on a magnificent throne, recalling the well-known “throne of Astarte” from the Phoenician, the Punic, and later, the Greek world (Lipinski 1995: 198–201). Does the seated figure described above represent Aphrodite/Astarte or her priestess? Perhaps the latter, shown as a female dressed as the goddess herself, bearing an offering in her hands as she is taken ceremonially to the temple. A later terra-cotta figurine from Khan Sheikoun, dating to the Persian period (5th–4th century bc), represents a nude female figure seated on horseback under a canopy, with a footstool at her feet, now in the Musée du Louvre collection. Du Mesnil du Buisson (1932) suggests that this figurine represents a procession during which the statue of Aphrodite/Astarte is carried on horseback (for references see V. Karageorghis 1997: 199), a custom that is deeply rooted in the Near East from the second millennium bc onwards (V. Karageorghis 1997: 200) and spread to the Greek world (Romano 1988: 132), perhaps via Cyprus. The recent “rediscovery” of a Cypriote terra-cotta figurine of an enthroned Aphrodite/Astarte on horseback, now in the National Archaeological Museum of Athens, offers a useful addition to the group discussed above. It stands approximately 14 cm high (fig. 3). The figure is shown seated sideways on a throne on a horse’s back. The horse has a high-arched neck and a slightly raised, short tail; its head is missing and its lower legs restored. Aphrodite/Astarte is represented with a flat body and short flat legs shown separately; probably the coroplast intended to represent her nude. Both arms stretch out (the right arm is missing) and touch the sides of the throne, which are

curved and made of two joined rolls of clay. The resemblance to the Late Bronze Age figurine in the Cyprus Museum referred to above is striking. The backrest is relatively low. The goddess wears a flat, high tiara decorated with two horizontal rows of pellets, probably representing flowers. She has a flat face with a pinched nose, and her eyes are rendered in red paint. Traces of red paint remain on her legs and on bands across her body. Around her neck is a partly damaged necklace rendered with added clay and painted red. Red and black linear decoration highlights the horse’s body and red the throne. A newly discovered terra-cotta figurine comes from a tomb at Ayia Phyla (in the Limassol district) (fig. 4). This figure also sits sideways and wears a long robe. Unfortunately, there is no indication of breasts, and the head is missing, making it impossible to determine the sex of the figure; the long robe, however, may suggest a female (Christou 1992: 805–6, fig. 50). The above remarks certainly do not completely exhaust the iconographic phenomenon of Aphrodite/Astarte on horseback. One must broaden the study of the goddess’s cult to extend beyond Cyprus and deeper into the Near East. By analyzing both texts and material culture, one can fully understand this particular iconographic peculiarity in Cyprus and throughout the Mediterranean world. This task is beyond the scope of this essay and certainly beyond the capabilities of the present writer, who wishes hereby to express cordial greetings to Bill Dever, a great scholar and a good friend, with whom he has shared happy occasions in Tucson and Cyprus.

References Andreadaki-Vlasaki, M. 1987 An Early Greek Child Burial at Gavalomouri, Kissamos. Studi Micenei ed Egeo-Anatolici 26: 307–35. Badre, L. 1980 Les figurines anthropomorphes en terre cuite à l’Age du Bronze en Syrie. Bibliothèque Archéologique et Historique 103. Paris: Geuthner. Ben-Arieh, S. 1983 A Mould for a Goddess Plaque. Israel Exploration Journal 33: 72–77, pl. 8a. Catling, H. W. 1974 The Bomford Horse-and-Rider. Report of the Department of Antiquities, Cyprus 1974: 95–111. Christou, D. 1992 Chronique des fouilles et découvertes archéologiques à Chypre en 1991. Bulletin de Correspondance Hellénique 116: 793–831. Clamer, C. 1980 A Gold Plaque from Tel Lachish. Tel Aviv 7: 152–62. Crouwel, J. H. 1981 Chariots and Other Means of Land Transport in Bronze Age Greece. Allard Pierson Series 3. Amsterdam: Allard Pierson.

Crouwel, J. H., and Tatton-Brown, V. 1988 Ridden Horses in Iron Age Cyprus. Report of the Department of Antiquities, Cyprus 1988/2: 77–87. Du Mesnil du Buisson, R. 1932 Une Campagne de fouilles à Khan Sheikhoun. Syria 13: 171–88. Hermary, A. 1984 Un nouveau relief “gréco-perse” en Cilicie. Revue Archéologique 1984: 289–300. Karageorghis, J. 1999 The Coroplastic Art of Ancient Cyprus V. The Cypro-Archaic Period. Small Female Figurines. B. Figurines moulées. Nicosia: Leventis Foundation. Karageorghis, J., and Karageorghis, V. 2002 The Great Goddess of Cyprus or the Genesis of Aphrodite in Cyprus. Pp. 263–82 in Sex and Gender in the Near East, ed. S. Parpola and R. M. Whiting. Vol. 1 of Proceedings of the XLVII e Rencontre Assyriologique Internationale, Helsinki. Helsinki: Neo-Assyrian Text Corpus. Karageorghis, V. 1974 Excavations in the Necropolis of Salamis. Vol. 3. Nicosia: Cyprus Department of Antiquities.

Aphrodite/Astarte on Horseback 1980 Kypriaka V. Report of the Department of Antiquities, Cyprus 1980: 128–32. 1993 The Coroplastic Art of Ancient Cyprus II. Late Cypriote II– Cypro-Geometric III. Nicosia: Leventis Foundation. 1997 An Enthroned Astarte on Horseback? Report of the Department of Antiquities, Cyprus 1997: 195–202. Keel, O., and Uehlinger, C. 1998 Gods, Goddesses, and Images of God in Ancient Israel. Minneapolis: Fortress. Leclant, J. 1960 Astarté à cheval d’après les représentations égyptiennes. Syria 37: 1–67. Levi, D. 1951 La dea micenea a cavallo. Pp. 108–25 in Studies Presented to D. M. Robinson on His Seventieth Birthday, ed. G. E. Mylonas. Saint Louis, MO: Washington University Press. Lipinski, E. 1995 Dieux et déesses de l’univers phénicien et punique. Studia Phoenicia 14. Leuven: Peeters. Littauer, M. A., and Crouwel, J. H. 1979 Wheeled Vehicles and Ridden Animals in the Near East. Leiden: Brill. Marinatos, N. 1996 Cult by the Seashore: What Happened at Amnissos? Pp. 135–39 in The Role of Religion in the Early Greek Polis, ed. R. Hägg. Stockholm: Swedish Institute at Athens.

79 Orchard, J. J. 1967 Equestrian Bridle-Harness Ornaments. Vol. 1/2 of Ivories from Nimrud (1949–1963). Aberdeen: British School of Archaeology in Iraq. Otten, H. 1953 Pirva, der Gott auf dem Pferde. Jahrbuch für Kleinasiatische Forschung 2: 62–73. Romano, I. B. 1988 Early Greek Cult Images and Cult Practices. Pp. 127–34 in Early Greek Cult Practice, ed. R. Hägg et al. Stockholm: Swedish Institute at Athens. Sakellarakis, Y., and Sapouna-Sakellaraki, E. 1997 ArcaneÍÚ Miav neva matiav sth Minwikhv Krhvth. Vols. 1–2. Athens: Ammos Editions/Eleni Nakou Foundation. Touchais, G.; Hubert, S.; and Phillipa-Touchais, A. 2001 Chronique des fouilles et découvertes archéologiques en Grèce en 2000. Bulletin de Correspondance Hellénique 125: 779– 1063. Yon, M., and Caubet, A. 1985 Kition-Bamboula, vol. 3: Le sondage L-N13: Bronze récent et géometrique 1. Paris: Éditions Recherche sur les Civilisations.

Planning a Paleolithic Picnic Albert Leonard, Jr. Department of Classics University of Arizona, Tucson

It could be said that every day was a picnic for the prehistoric people of the Nile Valley—not a picnic in the modern sense of the word, perhaps, but definitely a picnic in the sense of a group of people striving to assemble nutritious, life-sustaining, and portable meals designed to be eaten al fresco. The ability of these early groups to plan their communal movements in a manner that would coincide with seasonally available food sources marked a major watershed in human history.

Introduction During the Paleolithic period, before our early ancestors had adopted a settled existence in villages and towns, they passed through a time in which they were required to wander in a never-ending search for food. At first, the routes of these “hunter/gatherers” would have been haphazard, and the success of the enterprise would have been mainly in the hands of chance. Over time, however, the cumulative experience and knowledge acquired by each group would have led them to repeat a certain cyclical pattern based on years of observation. This accumulated knowledge, no doubt passed down through generations, would have allowed a group to plan their movements so as to coincide with known food sources (ripening fruit, spawning fish, etc.) according to predictable patterns of availability. This was a watershed in human history, because it is to this point in time—when our ancestors had achieved or developed the ability to plan and predict the presence of a certain food source in a predetermined place at a specific time of year—that we can trace our own patterns of food acquisition. In a way, it was not very different from our modern knowledge that a quart of milk, a loaf of bread, or a dozen eggs will be available at the corner convenience store on our way home from work. An excellent glimpse of these early hunter/gatherers in the Nile Valley comes from a series of small, relatively contemporary Upper Paleolithic (ca. 16,000 bc) sites located close to the point at which the Wadi Kubbaniya joins the Nile, about 10 miles north of Aswan. In this area, in the late 1970s and early 1980s, Fred Wendorf and his colleagues excavated and/or collected from the surface arti-

facts and ecofacts that produced evidence of some of the oldest food procurement and preparation strategies in northeastern Africa (Close, ed. 1986). Due to the meticulous nature of their work, this was one of the rare instances in which a substantially large number of fish and animal bones as well as a significant amount of vegetal material and its residue were recovered from such an early site. These finds represent the archaeological and culinary “footprints” of the human population(s) that seasonally visited the area in order to exploit its food resources and, when finished, moved on to places unknown. Because there was no evidence for human settlement at Wadi Kubbaniya—even of the most rudimentary nature—these groups are considered to have been true transients, living the classic hunter/gatherer life-style. According to the excavators, however, by this point in time, these early wanderers had already developed a very complex and seasonally diverse diet. Even more impressive is the nutritional balance that these early people were able to achieve, especially in the face of such limited dietary options.

The Wadi Kubbaniya Diet The diet of the Upper Paleolithic peoples who frequented the sites along the Wadi Kubbaniya consisted of a surprisingly sophisticated nutritional balance between the protein and fat provided by fish from the waters of the flooding Nile and the carbohydrates and dietary fiber supplied by wetland tubers and the fruit of local stands of doum palm. The story these important nutritional building blocks interacted to feed the Wadi Kubbaniya population is a credit to the great ingenuity of our early forebears. Fish The Nile began its annual inundation in early June, and by the end of August, in places such as the Wadi Kubbaniya, it would have reached a depth of between 15 and 20 feet above its low-water mark. Surging across the flood plain, these rising waters would have traveled a considerable distance up the dried riverbeds (wadis) that led from the valley into the desert. Along with these rushing waters, massive schools of adult catfish (Clarias gariepinus) and, less frequently, tilapia (Tilapia nilotica and related species)

82 would have come in search of food and/or a place to spawn, while eels and other predators would have followed in their wake, eagerly anticipating a meal of the newly hatched small-fry. This concentration of fish life in the shallow waters that flooded the Wadi Kubbaniya would have certainly caught the attention of the Paleolithic peoples of the area. They were drawn to its banks each summer, where, by hand or with the most primitive fishing “tackle” (stones, clubs, or basket nets), they would have been able to secure a sizable harvest, not only at the beginning of the flood, but for its entire duration and even afterward, when many fish would have been trapped in isolated pools as the flood waters receded into the main channel of the Nile. Although it is impossible to estimate the size of those annual harvests when viewed from so vast a chronological distance, it must have been substantial, for the fish-bone count at just one of the Wadi Kubbaniya sites exceeded 130,000 individual pieces (Gauthier and Van Neer 1986: 121, Site E-81-4). How this figure would translate into the quantity of individual fish caught or the number of single servings obtained is also impossible to calculate, but it seems to be more than sufficient to satisfy the immediate needs of the group. This strongly suggests that some process of food preservation and storage was already being practiced by the people at the site. Given the very high summer temperatures and extremely low humidity, sun-drying (dehydrating) the fish— either whole or in “portion-controlled” fillets—would have been the simplest method of preservation available at Wadi Kubbaniya in the Upper Paleolithic period. Millennia later, in describing the customs of the peoples of the Nile Valley in his day, the Greek historian Herodotus noted that “some of these people live . . . upon nothing but fish, which they gut as soon as they catch them, and eat after drying them in the sun” (The Histories 2.92; see de Sélincourt 1971: 135). Another option, smoking the fish over an open fire, might also have been practiced, in light of the large amount of carbonized tamarisk (Tamarix spp.) wood that was recovered from these sites (Tomczynska 1986). Combining the two techniques would have accomplished the task more quickly, and the increased speed would have been quite beneficial in retarding or eliminating the natural development of bacteria that would have led to rapid spoilage. Thus, the group could have quickly and efficiently assembled a store of provisions that would last them for five or six months. While the group was “putting up” catfish for the future, they would also have been able to enjoy the freshly caught fish either raw, as a kind of fresh-water carpaccio, or baked in the ashes of a campfire. Perhaps, in the latter case, the fish were wrapped in fronds from neighboring stands of the doum palm (Hyphaene thebaica). In addition to the catfish, spread is 1 line long

Albert Leonard, Jr. the archaeological assemblages of the period also contain remains of tilapia, eel (Anguilla anguilla or A. vulgaris) and even a few edible freshwater mollusks (Unio abyssinicus or Caelatura aegyptiaca). Treats of this sort, however, were much less numerous than the ubiquitous catfish and most likely represent fortuitous encounters with an Upper Paleolithic version of the “Catch of the Day.” When the females of these aquatic species were caught before they had the opportunity to spawn, they would have presented the group with an added attraction: the eggs— caviar—that were about to be released. Of necessity, this boon would have had to be consumed almost immediately, before it spoiled, but any “leftover” roe would surely have been processed in the same way as the fish, resulting in an early ancestor of botarga, a delicacy that is still produced and savored in many parts of the Mediterranean world today. Fowl It is not known whether every member of a specific group would have moved away from the Wadi Kubbaniya sites after the climax of the inundation toward the end of August. Some, or possibly all, may have stayed on in order to take advantage of the fish—especially the tilapia—that could be taken from the residual (and slowly deoxygenizing) pools that would have remained behind as the Nile gradually retreated back into its main channel. Thereafter, fresh food sources would become much scarcer. It would not have been until October, with the arrival of flocks of migratory birds escaping the falling temperatures in Europe, that another easily obtainable protein source would have become available in the area. Exhausted from the long flight across the Mediterranean, many of these winter visitors would literally have dropped at the feet of the potential diner. The presence of fowl remains at the Wadi Kubbaniya sites again demonstrates that, by the Upper Paleolithic period, a pool of knowledge had developed that enabled a group to plan its movements to be in the right place at the right time in order to take advantage of these culinary gifts from the heavens. Although very fragmentary, the bones of a large number of individual, nonresident (i.e., migratory) bird species were recovered. In most cases, the number of individuals of a given species is quite small, often just one or two specimens of each, but there are some exceptions. Over 50% of the archaeological sample consisted of bones of the ralid family (mostly Fulica atra, the “common coot”), but at least 40% of the remaining sample were bones of ducks and geese (several different species of the Anatid family) (Gauthier and Van Neer 1986: 154), and it is this latter group that would be more attractive on a modern menu. Ducks and geese could easily have been caught using stones, throwing sticks, or rudimentary nets and subse-

Planning a Paleolithic Picnic quently processed in the same ways as suggested above for the fish. However, in light of the small number of potential “servings” found at these sites, a roasted or baked meal of freshly caught fowl is more likely than drying or smoking. Meat Although the people were surrounded by a bounty of fish, the appetites of the Upper Paleolithic meat-eaters at Wadi Kubbaniya would not have gone completely unsated; however, red meat must have appeared on the menu far less frequently than the (by then) ubiquitous fish fillet. Bones of small mammals such as hare and fox most likely represent chance encounters with meals of opportunity, but three game animals are noted somewhat regularly in the archaeological assemblage and do seem to have been hunted intentionally. These are hartebeest (Alcelaphus buselaphus), constituting 66% of the assemblage; auroch or wild cattle (Bos primigenius), accounting for ca. 18% of the sample; and gazelle (Gazella dorcas), representing only some 16% of the ancient remains (Gauthier and Van Neer 1986: 155). Both hartebeest and cattle would have shared (or perhaps, more correctly, competed for) sustenance in the same type of landscape. Both feed on grasses and other bulk roughage that grew along the fringes of the vegetation, where the hartebeest, especially, should have been a familiar sight to those who stopped to fish along the banks of the Wadi Kubbaniya. Gazelles, on the other hand, are more of a desert dweller, able to go for extensive periods of time without water. Unless hunting parties were sent out into the desert to procure them, gazelles would have been encountered by the human population only when they approached the fertile zone for food and rehydration. That the frequency of this contact would have increased with the expansion of the inundating waters across the flood plain is a good possibility, and thus, like the fish, the gazelle may represent another known seasonal food source that was intentionally included in the Wadi Kubbaniya diet. Regardless of how this game was acquired, the taking of one of these relatively large animals (hartebeests range in weight between 150 and 200 kg, and the dorcas gazelle averages around 100 kg) would have provided an enormous amount of meat protein at one time, and the members of the group would have needed to consume it quickly before it spoiled. It seems that such an event would have called for a celebration: perhaps a day of joyous song, festive dance, and a feast of roasted meat. As the embers of campfires dimmed and the dancing slowed, any leftovers could have been gathered and then processed into “jerky” to be stored with the fish for future consumption. Vegetables Of the more than 20 wild food-plants recovered from the Wadi Kubbaniya sites, it is members of the sedge fam-

83 ily (the Cyperaceae), which still grow along the banks of the Nile today, that appear to have played the most significant role in the diet of those who were drawn to the area for the annual harvest of fish. Encompassing more than 100 species, Cyperus is characterized by a subterranean system of rhizomes and tubers in which a rich supply of carbohydrate is stored. This is especially true of Cyperus rotundus, popularly known in English as “nut grass” or “purple nut sage,” the tubers of which could be harvested in September and October, as the waters of the Nile flood began to retreat into their original channel. Cyperus rotundus is presently treated as a food source in several parts of eastern Asia, where it is known as “wild yam” or “Chinese yam,” and its taste is described as spicy, bitter, and pungent. However, its fibrous tubers must be processed before they can be transformed into anything even resembling a culinary treat. This is because they contain a variety of toxins, such as alkaloids and phenols, that seem to increase in concentration as the plant matures. Thus, older rhizomes/tubers require more processing (grinding and leaching) than younger plants (Hillman, Madeyska, and Hather 1986: 182–85). Modern ethnographic parallels suggest that, after the tubers were collected, they were cleaned in water and either spread on the earth under the summer sun or parched in a campfire in order to burn away the fibrous root system and leave the carbohydrate-rich tubers ready for consumption. Some indigenous groups in Australia simply roast them in an open fire and then roll them on the ground to remove the skin, producing a foodstuff, the taste of which has been compared to that of a baked potato (Hillman, Madeyska, and Hather 1986: 185). The fact that several of the ancient pieces of C. rotundus found at Wadi Kubbaniya show evidence of charring coupled with the presence of several ancient grinding stones (querns) on the surface of the site suggest that a similar manner of food processing and/or consumption was practiced there during the Paleolithic period. It has been determined that when dried, these tubers will retain their food value for several months. In fact, the density with which the subterranean portions of the plant proliferate suggests that a small group or family could accumulate a large store of food in a very short period of time. Both a charred rhizome and the nutlets of another member of the carbohydrate-rich sedge family, the club-rush (Scirpus maritimus or S. tuberosus), were found at the site and must also have formed part of the Wadi Kubbaniya diet. Similar to, but much less toxic than C. rotundus, the club-rush was most likely processed in the same manner, but whether this was done separately in order to preserve its culinary individuality or whether it was treated together with other rhizomes will probably never be known. Two other close relatives of C. rotundus are known to have thrived in the Wadi Kubbaniya area at a slightly later

Albert Leonard, Jr.

84 date (Hillman, Madeyska, and Hather 1986: 178–79, 208, fig. 7.25) and may also have been exploited by our Paleolithic menu-planners. They are the much less toxic Cyperus esculentes or “yellow nut-sedge” (also known as “tiger nuts” or “chufa”), used in modern kitchens in the Hispanic tradition; and Cyperus papyrus, best known for having provided us with the first form of paper, but which, according to Herodotus, was also eaten by the ancient Egyptians (The Histories 2.92; see de Sélincourt 1971: 135): (They) pull up the annual crop of papyrus-reed which grows in the marshes, cut the stalks in two, and eat the lower part, about eighteen inches in length, first baking it in a closed pan, heated red-hot, if they want to enjoy it to a perfection.

In addition, the stems of the sedges (Cyperus spp.) as well as the roots of the doum palm (H. thebaica) can be used for making baskets and fish traps, thus providing the means to catch and/or store other components of the diet. The doum palm is distinct from the date palm (Phoenix dactylifera) and produces brown, apple-sized fruit that is another rich source of carbohydrates. It is in its role as a major source of potassium, however, that this sweet but fibrous fruit made its major contribution to the Wadi Kubbaniya diet. The charred remains of a single umbrell (flower-bud receptacle) of a lotus (water lily, Nymphaea sp.) were found at Wadi Kubbaniya (Hillman, Madeyska, and Hather 1986: 223–25). This suggests that, as in many other primitive and modern societies, it too was used as a food source. Again, Herodotus adds some detail (The Histories 2.92; see de Sélincourt 1971: 135): They gather the water lilies . . . which grow in great abundance when the river is full and floods the neighboring flats, and dry them in the sun; then from the center of each blossom they pick out something which resembles a poppy-head, grind it, and make it into loaves which they bake. The root of this plant is also edible; it is round, about as big as an apple, and tastes sweet.

In addition to the plant material noted above, single or small samples of fern (Pteridophyte), chamomile (Athemis nobilis), and even wild asparagus (Asparagus spp.) were found at Wadi Kubbaniya, but whether they had been used for culinary purposes could not be determined (Hillman, Madeyska, and Hather 1986: 169–72, table 7.1).

Conclusions The diet of the wanderers who seasonally visited the Wadi Kubbaniya area contains much to be admired. The protein- and fat-rich fish was balanced by vegetal tubers that offered a rich supply of carbohydrates as well as a large amount of essential amino acids, such as lysine, a feature that it shares with the more familiar potato of New World origin. These tubers also contain small but significant amounts of magnesium, calcium, and iron but are devoid of potassium, an important element that was brought into the Wadi Kubbaniya diet by means of the fruit of the doum palm. In a way, the culinary regimen practiced by the people who frequented the Wadi Kubbaniya was a diet of necessity, for the Paleolithic landscape of the Nile Valley offered a fixed and severely limited market in which to shop. But it was also a diet of intelligence, built on a corpus of knowledge that had been developed and passed down from generation to generation, a tradition about not only what to eat but also where and when to find it. We can imagine wizened elders under starry skies mystifying wide-eyed youngsters with tales of how the movement of certain stars would dictate the length of time that they would spend at a given campsite and how these same stars would reposition themselves to signal the time for them to pack up and move on. This could only be accomplished through the structure and security of a plan. If successfully followed, however, this plan would provide them not only with food for sustenance but, perhaps more importantly, with food for thought. And the rest, as they say, is history.

References Close, A. E., ed. 1986 The Prehistory of Wadi Kubbaniya 2: Stratigraphy, Paleoeconomy, and Environment. Assembled by F. Wendorf and R. Schild. Dallas: Southern Methodist University Press. Gauthier, A., and Van Neer, W. 1986 Animal Remains from the Late Paleolithic Sequence at Wadi Kubbaniya. Pp. 119–61 (chapter 6) in The Prehistory of Wadi Kubbaniya 2: Stratigraphy, Paleoeconomy, and Environment. Assembled by F. Wendorf and R. Schild, ed. A. E. Close. Dallas: Southern Methodist University Press. Hillman, G.; Madeyska, E.; and Hather, J. 1986 Wild Plant Foods and Diet at Late Paleolithic Wadi Kubbaniya: The Evidence from Charred Remains. Pp. 162–242 (chapter 7)

in The Prehistory of Wadi Kubbaniya 2: Stratigraphy, Paleoeconomy, and Environment. Assembled by F. Wendorf and R. Schild, ed. A. E. Close. Dallas: Southern Methodist University Press. Sélincourt, A. de 1971 Herodotus: The Histories, trans. A. de Sélincourt. Baltimore: Penguin. Tomczynska, Z. 1986 Identification of Charcoal Fragments from Late Paleolithic Sites in Wadi Kubbaniya. Pp. 252–59 (chapter 9) in The Prehistory of Wadi Kubbaniya 2: Stratigraphy, Paleoeconomy, and Environment. Assembled by F. Wendorf and R. Schild, ed. A. E. Close. Dallas: Southern Methodist University Press.

The Archaic Palace at Ebla: A Royal Building between Early Bronze Age IVB and Middle Bronze Age I Paolo Matthiae Università degli Studi di Roma “La Sapienza” Rome

Of the goals achieved during the most recent years of the systematic exploration of Tell Mardikh, two stand out more prominently. On the one hand, we wished methodically to bring to light the important secular and religious public buildings of the northwest Lower Town of MB I (ca. 2000–1800 bc) and MB II (ca. 1800–1600 bc)—that is, of the Archaic and Classical Old Syrian periods (Matthiae 1997a). 1 On the other hand, we wished to open several new excavation areas on top of the earthwork rampart in order to identify the buildings that complemented, in an unprecedented way, the fortification system of the MB I–II town (Matthiae 1997b). The main purpose of these excavations was to clarify as exhaustively as possible the urban pattern of the Old Syrian town within the encompassing Lower Town and the massive outer fortifications (Matthiae 1998; 2000a) immediately prior to the heavy destruction that finally put an end to the great Classical Old Syrian town ca. 1600 bc, according to the conventional Middle Chronology (Matthiae 1989a: 55–56). 2

1. These buildings are: to the north, the Northern Palace, an important building with a royal function, with the entrance to the west and an east–west axis; to the south of a much-used street with the same orientation, the noteworthy religious complex of Ishtar’s Sacred Area, probably encompassing more than 1 ha, including the great Temple P2 to the northeast, which is oriented, like most of the Old Syrian temples at Ebla, with the entrance to the south; and to the southwest, the imposing cultic terrace called Monument P3. The two cultic buildings of Ishtar’s public sanctuary in the Lower Town, Temple P2 and Monument P3, opened onto the large Cisterns Square that extended along the foot of the northwestern part of the Citadel fortifications (Matthiae 1991). 2. The main result with regard to the urban pattern, dramatically confirmed by the identification in 2002 of the Southern Palace in the Lower Town, was the discovery of a virtually uninterrupted belt of important public secular and religious buildings (Matthiae 1986) that stretched from north to southwest at the foot of the inner fortifications of the Citadel: Temple N (dedicated to Shamash), the Northern Palace, Ishtar’s Sacred Area (with Temple P2 and Monument P3), the Western Palace, Temple B (dedicated to Rashap), Sanctuary B2 (devoted to the cult of the deified royal ancestors), and the Southern Palace found in 2002 (Matthiae in press a).

An unexpected historical aspect of the destruction of Ebla by the end of MB II can actually be deduced from the analysis of the great bilingual Hittite-Hurrian poem called “Epos der Freilassung” (Neu 1996a). From this analysis, one may infer that, ca. 1600 bc, Ebla was destroyed by Pizikarra, a thus-far unknown king of Nineveh mentioned in the proem, who is said to have decisively conquered the Old Syrian town thanks to the god Teshub of Kumme, a town certainly located on the Upper Tigris (Matthiae in press b). The poem was quite likely composed in an eastern Hurrian milieu, in the Hurrian language, possibly shortly after the conquest, in order to celebrate the great enterprise of a king (and a god) of the eastern Hurrians (Neu 1996b: 479–83), probably in close alliance with Mursili I. A few decades later, quite likely at the time of Tudhaliya I, who certainly wished to repeat Mursili I’s great deeds in upper Syria (Klengel 1992: 80–83, 89), the poem was translated into Hittite at Hattusa, and the duallanguage version was kept in the capital town of the empire precisely because it celebrated in a mythical way an event whose protagonists were the allies Pizikarra of Nineveh and Mursili I of Hattusa (Matthiae in press b). 3 As a consequence of the intensified excavations in the Lower Town, two spectacular monumental complexes of Classical Old Syrian Ebla were brought to light: one secular, the Northern Palace in Sector P North of Area P; and one religious, Ishtar’s Sacred Area in Sector P Central and South. Ishtar’s Sacred Area includes Temple P2, with a classical Long-Room plan (Matthiae 1990c), and the very 3. The interpretation of the “Epos der Freilassung” as a poetic composition celebrating the conquest of Classical Old Syrian Ebla by the end of MB II as an accomplishment of Pizikarra of Nineveh under the protection of the god Teshub of Kumme was proposed and illustrated in detail by the author in a paper read at the Accademia Nazionale dei Lincei in Rome, Nov. 8, 2001 (Matthiae in press b). In the same paper, I also reported some astonishing analogies between the “Epos der Freilassung” and Homer’s Iliad, which have been mentioned and accepted by Godart (2001: 1577–79).

Paolo Matthiae

86

Fig. 1. Ebla, Archaic Palace Phase I, central sector with throne room L. 5715, looking south.

peculiar and unique cultic terrace called Monument P3 (Matthiae 1993). The Northern Palace is an extended building, more than 3,400 sq. m in area, with a royal function, quite likely erected at the beginning of MB II, during the first half of the 18th century bc (Matthiae 1989a: 171–75). The Northern Palace was not the only palatial building with a monumental structure and royal function that was found at Old Syrian Ebla at the time of the final destruction, around 1600 bc. Two other palaces were discovered some time ago, and the excavation of another began in 2002. The first and certainly most important palace, exposed in only a limited area, is the Royal Palace in Area E on the Citadel, of which only the northern peripheral wing is known: it was probably quite articulated and large and included the royal residence quarters as well as wings devoted to the palace administration and probably also a reception suite, storerooms, and workshops (Matthiae 1989a: 160–62). The second building is the extensive Western Palace in Area Q in the Lower Town West, systematically explored and recently restored (Matthiae 2002b: 558–65), which was nearly 7,200 sq. m in area (Matthiae 1989a: 162–71): it was probably the crown prince’s residence, as indicated among other things by the discovery of a number of jars with impressions of the cylinder-seal of a son of King Indilimgur/Indilimma, possibly the last king of the Old Syrian town, and by its proximity to

the hypogea of the Royal Cemetery and to Sanctuary B2, devoted to the cult of deified royal ancestors (Matthiae 1991: 328–30). 4 The third building is the Southern Palace in Area FF in the Lower Town South, discovered in 2002, which is definitely larger than 1,000 sq. m and probably reached 2,000–2,500 sq. m in area. The function of this building is still difficult to define, but the construction technique, the presence of jars inscribed with the word é.gal ‘palace’, and our discovery of a palatial administrative cuneiform tablet in it lead to the conclusion that the building accommodated important elements of the central administration and was probably the residence of a very high official in the town government (Matthiae in press a). 5 4. It seems certain that each palatial building at Old Syrian Ebla was associated with a temple or a sacred area. Royal Palace E on the Acropolis was definitely closely related to the great Temple D, Ishtar’s true dynastic sanctuary in the Citadel; the Northern Palace, specifically conceived for ceremonial functions, was surely connected to Ishtar’s Sacred Area in the Lower Town, which was the great goddess’s public sanctuary, like Temple D in the Citadel that served in some way as the king’s private chapel. The relationship between the Western Palace and Rashap’s cultic area, including Temple B and Sanctuary B2 for the cult of deified royal ancestors, can be understood through the crown prince’s specific role in the successful accomplishment of the deceased king’s funeral, in which he became a rapiªu, and his function as an effective protector of the community of the living. 5. The Southern Palace identified in 2002 by means of a north–south trench, 9 m wide and 30 m long, was constructed using very fine masonry,

The Archaic Palace at Ebla: A Royal Building between EB IVB and MB I

87

Fig. 2. Ebla, Archaic Palace Phases I–II, northeastern sector, looking south.

The expansion of the excavations in Sector P North, to the north of the Northern Palace, led to an important and unexpected discovery of special significance for understanding the previously obscure history and topography of the town in the period between the destruction of the mature Early Syrian town of the Archives period, ca. 2300 bc (Matthiae 1989b), and the reconstruction of the Classical Old Syrian town, which reached its final form only around 1800 bc. This is the Archaic Palace (fig. 1), initially designated Building P5, excavated between 1993 and 1996 (Matthiae 1993: 638–40; 1995: 659–73; 1998: 564–68). The Archaic Palace is an extensive building that was founded during EB IVB at the time of the late Early Syrian town, between 2300 and 2000 bc, and was used during three successive phases, during most of MB I. Only its northern peripheral wing could be excavated, because to the south, it stretches below the MB II Northern Palace, where the excavation is obviously impossible, 6 for as with almost all of the walls covered with limestone and basalt orthostats, large rooms, and doors with jambs and thresholds made of monolithic slabs; it was probably equally well developed along the east–west axis. It has not yet been possible, however, to identify the location of the entrance to this monumental building (Matthiae in press a). 6. It is impossible to ascertain how far the Archaic Palace stretches below the Northern Palace, but it is sure that this hidden part was not smaller and may perhaps have been even slightly larger than the excavated

much as two-thirds of its total surface (Matthiae 1995: 662–67, figs. 6–17). The EB IVB Archaic Palace is the earliest palatial building in the area of the Lower Town. It is quite likely that this monumental edifice, only partially explored because of the overlapping of the Northern Palace, was the Royal Palace (or one of the palaces with a royal function) of the town at the end of EB IVB, the late Early Syrian period. It certainly functioned during the 21st century bc, several decades after the destruction of Royal Palace G of the Archives period, the mature Early Syrian period, corresponding to EB IVA (Matthiae 1989a: 52–55; Schwartz and Weiss 1992: 237–40). 7 It is quite likely that the Archaic Palace of the first phase (Phase I) was not completed according to its original plan, which probably included a northern façade with three large piers—two narrower ones at the eastern and

sector and that it corresponded at most to the surface area of the later Northern Palace. On the basis of these considerations, the total surface area of the Archaic Palace can be calculated as at least 1,700–2,000 sq. m, but the possibility cannot be ruled out that it even reached 4,500 sq. m. 7. Royal Palace G of the EB IVA was most likely destroyed by Sargon of Akkad, perhaps right at the beginning of his reign, according to a long Old Babylonian tradition, as maintained in Matthiae 1989a: 125– 32, 241–50. Against Astour’s proposal (1992) that the destruction dates to approximately 20 years before Sargon and was the result of natural causes, see Matthiae 1995: 655; 2000b: 578.

Fig. 3. Ebla, Archaic Palace, schematic plan of Phases I and II.

88

Paolo Matthiae

The Archaic Palace at Ebla: A Royal Building between EB IVB and MB I

Fig. 4. Ebla, Archaic Palace, schematic plan of Phases I, II, and III.

89

90

Paolo Matthiae

western edges, and one twice as large in the middle (fig. 3). After the heavy destruction at the end of EB IVB, probably ca. 2000 bc (Matthiae 1989a: 125–32; Schwartz and Weiss 1992: 240), the Archaic Palace had three more phases (Phases II, III, and IV), with refurbishing and restorations (fig. 2), all of which were carried out during MB I, possibly in the 20th and first half of the 19th centuries bc (fig. 4). Probably the Phase II Archaic Palace, although in a different historical and socioeconomic setting, retained its function as a royal residence and official building with an audience hall, while Phases III and IV show clear marks of decline. 8 The Phase IV Archaic Palace (fig. 5) was probably abandoned by the end of MB I, perhaps shortly before 1800 bc, as a consequence of an earthquake that caused subsiding and collapses in the underground limestone layer, rich in natural caves, on which the structure had been built (Matthiae 1995: 671–73, fig. 17). The entire north wing of the Archaic Palace was correctly considered to be too dangerous for the erection of a new palace. In fact, the new Intermediate Palace, of which only the tops of some walls were exposed below the floors of the later Northern Palace, was constructed only over the central and southern wings of the Archaic Palace (Matthiae 1995: 675–76, fig. 19). 9 The whole northern peripheral wing was abandoned, which allowed for the recovery of precisely this part of the earliest building. It is quite likely that the Intermediate Palace, the plan of which is unknown except for the eastern, northern, and western perimeter walls and two sections of interior walls, was built at the very beginning of MB II. The eastern and western perimeter walls converged to the north, precisely like the Archaic Palace, because their foundations, unlike those of the northern wall, definitely shifted to the south and were erected exactly over the foundations of the Archaic Palace. For reasons we do not know, shortly afterward, during an early phase of MB II, perhaps by the middle of the 18th century bc, the Intermediate Palace was

sealed, without being destroyed, and the Northern Palace was built. This palace is largely preserved. It was in use until the end of MB II but was pillaged and destroyed by fire during the great destruction of ca. 1600 bc (Matthiae 1989a: 55–57). 10 The north wing is what is thus far known of the Phase I Archaic Palace, of which the central and southern sectors are concealed by the superimposition of the Intermediate and Northern Palaces. It includes two large halls to the west (L. 5893 and L. 5892), approximately 10 m long, oriented east–west, as well as a smaller room (L. 5711), behind one of the buttresses of the northern façade; and, to the south, a magnificent, large hall (L. 5715) with a north–south axis that has a bench and a central dais against its northern wall. One can safely assume that, in the original scheme that was never fully implemented, two other east–west-oriented halls were planned, as well as a third room behind the third central buttress of the north façade, in a perfectly symmetrical pattern. This hypothesis is based on two important elements. First, the placement of hall L. 5715 is precisely in the middle between the eastern and western perimeter walls. Second, the eastern limit of the small room, L. 5711, is exactly at the “midriff” of the building. One must, therefore, infer that in the original plan this was the central north–south axis of the entire building and that hall L. 5715 was in the center of the edifice. This important central sector of the Phase I Archaic Palace was probably designed to have a mirror-like correspondence between the three rooms L. 5893, L. 5892, and L. 5711 on the west and three rooms that were never built, which we might designate L. 5893a, L. 5892a, and L. 5711a, on the east. 11 It is clear that this quite-likely original plan was never realized, although the orientation and placement of the eastern perimeter wall had already been sketched out; the wall was built in one of the later phases, all of which were poorer in architectural terms. Moreover, the eastern perimeter wall of the Northern Palace was subsequently built with the same alignment.

8. Because there is usually quite a clear gap between the end of EB IVB and the beginning of MB I with regard to the ceramic horizons of the two periods, the continuity of use of a royal palatial building, albeit unfinished, is of the greatest interest and must be considered an important element in the evaluation of the transition between the late Early Syrian and the beginning of the Archaic Old Syrian period, which was evidently much more complex and less clear-cut than is usually believed, at Ebla at least, but probably in the whole of northern Syria. 9. The northern limit of the Intermediate Palace was certainly the same as the limit of the later Northern Palace, and it was located nearly 15 m to the south of the northern limit of the Archaic Palace. The northern limit of the Intermediate Palace was exactly parallel to the Archaic Palace and was based on the foundations of its eastern and western perimeter walls, creating obtuse angles identical to the junctures of these walls in the original plan of the Archaic Palace, which, as mentioned above, was not fully realized.

10. A special feature of the Intermediate Palace—observed even in the very limited spaces in which its structures, in a very good state of preservation, were identified—is the fact that the inner walls of this earlier building corresponded to the rooms of the later Northern Palace and not to its walls (Matthiae 1995: fig. 19). This is in contrast to the perimeter walls of the Intermediate and Northern Palaces, which, in fact, are placed over each other. 11. This quite bold reconstruction can be proposed on the following basis: we ascertained that if we used a north–south line passing through the middle of throne room L. 5715 as an axis and reversed the entire west wing of the Archaic Palace to the east, then the reconstructed eastern perimeter wall of the complex ran exactly along the line of and had the same orientation as the actual eastern perimeter wall of the building that was built later on, probably during Phase II. It is perfectly symmetrical with the western perimeter wall in position and orientation, but it is clearly different in terms of its size and construction technique.

Fig. 5. Ebla, Archaic Palace, schematic plan of Phases I, II, IV, and V.

The Archaic Palace at Ebla: A Royal Building between EB IVB and MB I 91

92 The great central hall, L. 5715 was 9.50–9.80 m wide and more than 9.50 m long on the north–south axis, because its southern limit is below the Northern Palace. It has two important special features: first, the bench against the northern wall, nearly 0.85–1.05 m thick, has an almost 0.40-meter-wide dais in the middle; second, the gypsum plaster floor ends at a regular distance of 0.30–0.40 m from the foot of the eastern and western walls and 1.10– 1.20 m from the bench against the northern wall. From these features one may infer that hall L. 5715 was an audience hall: the dais was probably for the throne, and the walls may have been covered with wooden planks placed against the vertical mud bricks. It is quite likely that this was the throne room, probably rectangular in shape, and the royal dais was probably located against one short wall, precisely as in the later audience hall of the Northern Palace, also located in the center of the building (Matthiae 1990b: 406–9, fig. 12). 12 The general pattern of the Archaic Palace in historical perspective is quite exceptional because of its trapezoidal plan and the three unusual buttresses on the rear façade. However, the most relevant characteristic is certainly the thickness of its walls, which corresponds to the relevant width of its rooms: the main walls are between 1.50 and 2.50 m wide, while the thinner partition walls are never less than 1.20 m thick. The date of the foundation of the Archaic Palace is securely established by the original floors found in western rooms L. 5892 and L. 5893, which were 0.35–0.40 m below the level of the last phase of the building. Quite a few homogeneous sherds belonging exclusively to the EB IVB were found on these floors (fig. 6). Quite common in this assemblage are fragments of Painted Ware goblets with the characteristic horizontal bands of black or red paint (fig. 6:1–5), typical of Hama J5–1 (Fugmann 1958: 62–85, figs. 74–75, 85, 93, 98), Amuq J (Braidwood and Braidwood 1960: 442–46, figs. 340, 342–44), and Late Selenkahiye (Schwartz 2001: 252, fig. 5A.27), which are not present in the horizon of the Royal Palace G of Mardikh IIB1, namely, EB IVA Ebla (Mazzoni 1985) and are unique to the horizon of Mardikh IIB2, that is, the EB IVB. The 12. Some hints lead us to believe it likely that the original throne room L. 5715 of the Archaic Palace was approximately 20 m long on the north–south axis, exactly like the throne room of the later Northern Palace, which, like the throne room of the Western Palace and all the royal buildings known thus far from MB II Syria, had an entrance of the bentaxis approach type—that is, not corresponding to the main axis. Although the correspondences are not completely satisfactory, it cannot be ruled out that throne room L. 5715 of the Archaic Palace was, in a form not yet canonical, the model for the characteristic type of Old Syrian royal reception suite (Matthiae 1990a), which I recently suggested as a parallel for an important religious and palatial building at Neo-Sumerian Ur (Matthiae 2002a).

Paolo Matthiae painted goblets of Mardikh IIB2 also have unique bases that are bell-shaped or have moldings (Schwartz and Weiss 1992: 240). The same floors of the two rooms of the Archaic Palace also yielded a fair number of the two most distinctive shapes of open bowls of Mardikh IIB2: bowls with a vertical rim marked by two horizontal grooves (fig. 6:6) and bowls with an out-turned molded rim (fig. 6:7, 9; see also Fugmann 1958: figs. 98, 103; Matthiae 1989a: fig. 25; Schwartz 2001: pls. 5a.2–3, 5B.40b). The above evidence leads us to the conclusion that the Archaic Palace was founded sometime after the destruction of the town of Royal Palace G and after the Archives period of EB IVA, most likely during the 21st century bc. Another indication of this is the fact that its northwestern wing, which was never completed, was immediately reused at the beginning of MB I. Thus, the evidence regarding the first phase of the Archaic Palace has far-reaching historical implications; in fact, from around the mid-21st century bc, the Neo-Sumerian administrative texts, written between the end of Shulgi’s and the beginning of ShuSin’s reign, repeatedly attest to the presence of messengers and merchants from Ebla at the court of the most important sovereigns of the Third Dynasty of Ur (Owen 1992). 13 One can reasonably presume that the Archaic Palace was the royal palace that dispatched these persons to the NeoSumerian capital town, if our understanding of the correspondence between the EB IVB archaeological horizon in northern Syria and, in general, the post-Akkad and NeoSumerian periods in Mesopotamia holds true. The Archaic Palace might have been built in the northern zone of the Lower Town after the possible partial abandonment of the great Early Syrian town, destroyed by Sargon of Akkad, perhaps because subsequently, the settlement was limited to the northern quarters (Dolce 1999). 14 The topographical choice might have been determined by the proximity of Ishtar’s Sacred Area. The latter consideration seems to be 13. These important epigraphic data concerning the relations between Ebla and Ur in the Neo-Sumerian period should by no means be interpreted as proof of Ur’s political domination over important western cities such as Byblos and Ebla, as initially assumed by Sollberger (1959– 60), but rather as simply evidence of important, frequent, and perhaps regular economic, diplomatic, and political relations between the two towns, as correctly proposed by Klengel (1992: 36–38). It is possible that these relations had a stronger impact than previously supposed also in the cultural sphere, in particular with regard to architecture (Matthiae 2002a). 14. A secure assessment of the extent of the EB IVB settlement in the Lower Town is very difficult, because it is quite clear that the ruins of the late Early Syrian town, apparently destroyed around 2000 bc, were removed and spread around in order to accumulate soil—always, in fact, thickened with ash—along the line of the earthenwork ramparts, built in part with materials collected inside the settlement and in part with materials that had piled up during the digging of a kind of external ditch at the foot of the rampart itself (Matthiae 2001).

The Archaic Palace at Ebla: A Royal Building between EB IVB and MB I

Fig. 6. Ebla, Archaic Palace Phase I, EV IVB pottery.

93

Paolo Matthiae

94

Fig. 7. Ebla, Archaic Palace Phase II, northeastern sector, looking northeast.

important, because, on the one hand, it implies a continuity between the late Early Syrian period (corresponding to EB IVB) and the high Early Syrian tradition of EB IVA; 15 and on the other, it offers an explanation for the continued presence of palatial buildings with a royal function in this area of the Lower Town during the Old Syrian period. After the heavy destruction that marked the end of EB IVB, between the final years of the 21st century and the first years of the 20th century bc, at the beginning of MB I, the Archaic Palace was not disregarded but, rather, entered its Phase II: the east wing was completed with the construction of four new, smaller rooms, a large external cobbled floor to the north, and most importantly, a kind of open-air trapezoidal room, L. 6867 (fig. 7). This special room, 12.5 m wide, which opened onto the court with the cobbled floor, had a dais against the eastern wall. It was probably covered by a tent that opened directly to the outside, to the west. 15. Similarly, in fact, in the Classical Early Syrian period, Royal Palace G of the Archives period on the summit and slopes of the Acropolis was very close to or even included the predecessor of Temple D, perhaps dedicated at that time to the goddess Ishkhara (Archi 1992; Prechel 1996; Pomponio and Xella 1997: 215–17), whose divine figure certainly underwent a process of syncretism with Ishtar at the beginning of the MB I, to whom Temple D was in fact dedicated from the beginning of the Archaic Old Syrian period (Matthiae 2003).

The very distinctive arrangement of hall L. 6867 (fig. 8) reveals an unusual social setting: the political authority apparently wished to appear in public, not in a traditional audience hall inside a palatial building, but in an open, temporary hall located on the periphery of the ancient royal palace of the previous period. This social situation seems to imply an intense, critical phase of urban demographic and economic development that can be related to the reconstruction of the town at the beginning of the MB I, when, in particular, the building of huge earthwork ramparts (Matthiae 2001) was a priority for prestige and defense needs (Bunimovitz 1992; Finkelstein 1992; Kempinski 1992; Herzog 1997: 116–18). 16 16. The archaeological evidence from Ebla in favor of the erection of the massive ramparts of the outer fortifications in the first years of the MB I corresponds exactly to what is documented for the beginning of the dynasties of Isin, Larsa, and Babylon: Ishbi-Erra, the founder of the Isin dynasty, mentions that he built the wall of Isin in his Year 12 (Sigrist 1988: 14); Gungunum, the first independent king of Larsa, built the fortifications of the town in his Year 21 (Sigrist 1990: 9–10); Sumuabum, the founder of the First Dynasty of Babylon, mentions the erection of the wall of Babylon in his Year 1 (Ungnad 1938: 175). The huge amount of work needed for the construction of the ramparts, which at Ebla required extensive cleaning and rearrangement of the entire Lower Town, might well have obliged the sovereign who founded the dynasty at the beginning of the MB I to have a kind of temporary open-air throne room from which he could view progress of the work at all times.

The Archaic Palace at Ebla: A Royal Building between EB IVB and MB I

95

Fig. 8. Ebla, Archaic Palace Phase II, throne hall L. 6867, looking southeast.

At that time, when the new seat of the dynasty certainly had not yet been built—that is, the later Royal Palace E, which was probably constructed on top of the Citadel shortly thereafter (Matthiae 1989a: 160–62, fig. 36)—the Archaic Palace in all likelihood has to be used as a mark of clear continuity with the EB IVB kingship (Dolce 2001; Pinnock in press b). 17 Moreover, the need to break with the past was probably a factor, because the previous throne room was replaced by a new, open, external reception hall. The pottery of Archaic Palace Phases II (figs. 9–10) and III includes shapes unique to MB I (Nigro 2000). Much of the Simple Ware that appears in Phases II and IV consisted of closed-shape bowls with a strong, well-defined carina17. The analysis of the few known names of kings of Ebla of the early Old Syrian period—Igrish-Kheb(at) and his son Ibbit-Lim, mentioned in the inscription on the well-known statue found in 1968 that allowed for the identification of Tell Mardikh as Ebla (Matthiae and Pettinato 1972; Gelb 1984); and Ib-Damu, known from the impression of a royal cylinder seal reused at Kültepe, found in the Kanesh II level (Teissier 1994: 177, 233, n. 529a; Pinnock 2000)—clearly shows that the dynastic gods of this period were the same as the gods of the previous dynasty of the Archives period, when Damu and Lim very frequently were elements in royal personal names (Pomponio and Xella 1997: 387– 88, 454). The analysis also shows that the lexical repertoire and verbal forms of those ancient times were still in use; for instance, IgrishKheb(at) clearly recalls King Igrish-Khalab’s name, which appeared in the first phase of the Archives period.

tion (fig. 10:5); open-shape bowls with a short, out-turned rim (figs. 9; 10:3–4, 6–9), representing several Archaic types of carinated bowls that are very common throughout inner Syria from the region of Aleppo to Hama; and small, closed-shape bowls with an unusual thin, almost vertical rim with threefold molding on the outer face (fig. 11:7). Among the medium-sized vessels, including bowls with a rounded, slightly protruding rim and ribbed body (fig. 11:1–2), are typical biconical juglets with a thickened, everted, flat, lip-like rim, triangular in section, and frequently with combing on the shoulder (fig. 12:11–15); and jars, perhaps almost globular, with a high neck and thickened rim, rounded on the edge. On the other hand, eggshaped jugs with a definitive out-turned rim with the edge modeled on the exterior (fig. 11:4–5); and small, open, socalled funnel-shaped bowls with an in-turned rim are quite rare and only beginning to appear (Pinnock in press a). In a later phase of MB I, when the construction of a new royal residence had quite likely begun, probably on top of the Acropolis, which at precisely this time became a fortified citadel (Matthiae 2001), the Archaic Palace entered Phase III. 18 Modest refurbishing of some walls and 18. The evidence from Archaic Palace Phases II–III seems to support the hypothesis that the urban renaissance of Ebla during the first decades

text continues on p. 100

Paolo Matthiae

96

Fig. 9. Ebla, Archaic Palace Phase II, MB IA pottery.

The Archaic Palace at Ebla: A Royal Building between EB IVB and MB I

Fig. 10. Ebla, Archaic Palace Phase II, MB IA pottery.

97

Paolo Matthiae

98

Fig. 11. Ebla, Archaic Palace Phase IV, MB IB pottery.

The Archaic Palace at Ebla: A Royal Building between EB IVB and MB I

Fig. 12. Ebla, Archaic Palace Phase IV, MB IB pottery.

99

100

Paolo Matthiae

Fig. 14. Idlib, Archaelogical Museum, Hyksos nefer-ka-ra scarab, postpalatial phase.

Fig. 13. Ebla, Archaic Palace Phase IV, oven T. 6203 in the northeastern sector, looking east.

floors in this northeast wing reveal decay of the building, which is logical, considering the probability that it retained only a secondary royal function. The final development of the Archaic Palace is documented by the subsequent Phase IV, which probably dates to a late MB I phase, when the new northern delimiting wall of the building, M. 5884, was built and some smaller rooms surrounded the cobbled court, L. 6034, in front of the ancient external reception hall. The hall was reduced in

of the MB I, corresponding to the early Old Syrian period, developed in a similar way to the urban development of the dynasties of Isin, Larsa, and Babylon I, documented by year names, in the contemporary, Old Babylonian period. In fact, in these great centers of Babylonia, the erection of the town fortifications was accomplished by the founders of the dynasties, while the construction of palaces, rarely mentioned in the year names, was undertaken only by their successors—not even the direct successor—evidently because, in the beginning, the first sovereigns of the Old Babylonian Amorite dynasties continued to occupy their older royal buildings.

size, had more regular layout, and was surrounded by very plain structures. During this last phase, the function of the eastern wing changed completely: room L. 6061 contained a small oven used for firing clay figurines, some of which were found in situ, that has been dated to the MB I (fig. 13). The ceramic horizon of Phase IV of the Archaic Palace (figs. 11–12) is characterized, with regard to Simple Ware, by a few archaic closed-carinated bowls (fig. 12:3, 5); a large number of typical small open-carinated bowls, with several variants that are typical of a large part of the MB I in its middle and final phases (fig. 12:1–2, 4); as well as by the presence of medium-sized bowls with almost oblique walls and a distinct rib near the rim, which is quite thick and down-turned (Nigro 2002: 298–305, figs. 3, 7). The Kitchen Ware includes many quite large, polished, open bowls, gray-greenish in color, with a large, oblique, rounded rim, thickened on the exterior and interior, and displaying a strong rib below the rim exterior (fig. 12:6–9): this is a very typical MB I shape that disappears in the MB II, the horizons of which are usually rich in polished gray-black or dark brown Kitchen Ware. Toward the end of MB I, the Archaic Palace was completely abandoned: in Phase V, a postpalatial phase, at least two small rooms were built against the northeastern wing over the cobbled floor. In this last phase, bronze tools were produced in a workshop above the poorly preserved remains of the structures of the ancient building, abutting the western wall of the great central throne room. 19

19. This limited industrial installation must not have been used very long, perhaps only a few years. In fact, certainly by the end of the 18th century bc, this area, like the entire zone to the north of the northern limit of the Intermediate Palace and then of the Northern Palace, became a large waste pit. Its contents piled up over the poorly preserved remains of the walls of the last phase of the Archaic Palace, with its almost completely missing mud-brick structures, over the meager walls of the small, postpalatial rooms, and over the small workshop, soon completely covering the top of the stone foundations of the Archaic Palace.

The Archaic Palace at Ebla: A Royal Building between EB IVB and MB I

101

Only shortly thereafter, when construction began on the Intermediate Palace, the entire north wing of the Archaic Palace fell into disuse except as a waste pit, which was gradually filled up with layers of waste rich in MB II pottery as well as some precious objects that were discarded because they had deteriorated. Among these we may mention an Egyptian scarab of the Hyksos period of the type called nefer-ka-ra (Scandone Matthiae 2004), based on the three hieroglyphic signs with an apothropaic function (fig. 14); and, more importantly, a wonderful complete ivory figurine of an offering-bearer carrying a gazelle (fig. 15; see also Matthiae 1995: 677–78, fig. 21). This noteworthy piece is one of the oldest specimens of ivory carving from inner Syria, probably dating to the second half of the 18th century bc, and can definitely be attributed to an Old Syrian workshop. The history of the building phases in the area of the Archaic Palace comes to an end with the erection of the MB II Northern Palace, probably in the second half of the 18th century bc. It is certain that this building had a royal function. It is quite well preserved, with the exception of the peripheral southwestern and southeastern wings. It is likely that the Northern Palace (like the preceding, Archaic Palace) was related to Ishtar’s great Sacred Area, including Temple P2 and the cultic terrace of Monument P3, located immediately to the south—although we do not yet know precisely how. The Northern Palace must have had a special function and may be defined in general terms as a luxurious royal pavilion for audiences and official ceremonies; there is no hint that it was used as a royal residence or as a seat of the royal administration (Matthiae 1990b: 403–9).

Fig. 15. Idlib, Archaeological Museum, ivory offering bearer, postpalatial phase.

References Archi, A. 1992 Substrate: Some Remarks on the Formation of the West Hurrian Pantheon. Pp. 7–14 in Hittite and Other Anatolian and Near Eastern Studies in Honour of Sedat Alp, ed. H. Otten et al. Ankara: Türk Tarih Kurumu. Astour, M. C. 1992 The Date of the Destruction of Palace G at Ebla. Pp. 23–40 in New Horizons in the Study of Ancient Syria, ed. M. W. Chavalas and J. L. Hayes. Bibliotheca Mesopotamica 21. Malibu: Undena. Braidwood, R. J., and Braidwood, L. S. 1960 Excavations in the Plain of Antioch I. The Earlier Assemblages: Phases A–J. Oriental Institute Publications 61. Chicago: University of Chicago Press.

Bunimovitz, S. 1992 The Middle Bronze Age Fortifications in Palestine as a Social Phenomenon. Tel Aviv 19: 221–34. Dolce, R. 1999 The “Second Ebla”: A View on the EB IVB City. ISIMU 2: 293– 304. 2001 Ebla et ses souverains: Du renouvellement de la ville protosyrienne tardive à l’épanouissement du règne amorrhéen. Pp. 211– 38 in Anatolia antica. Studi in memoria di F. Imparati, ed. S. de Martino and F. Pecchioli Daddi. Florence: Eothen. Finkelstein, I. 1992 Middle Bronze Age Fortifications: A Reflection of Social Organization and Political Formations. Tel Aviv 19: 201–21.

102 Fugmann, E. 1958 Hama. Fouilles et recherches de la Fondation Carlsberg 1931– 1938, vol. 2/1: L’architecture des périodes pré-hellénistiques. Nationalmuseets Skrifter 4. Copenhagen: Nationalmuseet. Gelb, I. J. 1984 The Inscription of Jibbit-Lim, King of Ebla. Studia Orientalia 55: 213–29. Godart, L. 2001 Littérature mycénienne et épopée homérique. Comptes Rendus de l’Académie des Inscriptions et Belles-Lettres: 1561–79. Herzog, Z. 1997 Archaeology of the City: Urban Planning in Ancient Israel and Its Social Implications. Institute of Archaeology Monographs 13. Tel Aviv: Tel Aviv University Press. Kempinski, A. 1992 Middle and Late Bronze Age Fortifications. Pp. 127–42 in The Architecture of Ancient Israel from the Prehistoric to the Persian Periods, ed. A. Kempinski and R. Reich. Jerusalem: Israel Exploration Society. Klengel, H. 1992 Syria 3000 to 300 b.c.: A Handbook of Political History. Berlin: Akademie Verlag. Matthiae, P. 1986 Sull’identità degli dei titolari dei templi paleosiriani di Ebla. Contributi e Materiali di Archeologia Orientale 1: 335–62. 1989a Ebla, un impero ritrovato. Dai primi scavi alle ultime scoperte. 2nd ed. Turin: Einaudi. 1989b The Destruction of Ebla Royal Palace: Interconnections between Syria, Mesopotamia and Egypt in the Late EB IVA. Pp. 163–69 in High, Middle or Low? Acts of an International Colloquium on Absolute Chronology Held at the University of Gothenburg, 20th–22nd August 1987, ed. P. Åström. Gothenburg: Åströms. 1990a The Reception Suite of the Old Syrian Palaces. Pp. 209–28 in De la Babylonie à la Syrie en passant par Mari: Mélanges offerts à J.-R. Kupper, ed. Ö. Tunca. Liège: Université de Liège. 1990b Nouvelles fouilles à Ebla en 1987–1989. Comptes Rendus de l’Académie des Inscriptions et Belles-Lettres: 384–431. 1990c A New Monumental Temple of Middle Bronze I at Ebla and the Unity of the Architectural Tradition of Syria–Palestine. Annales Archéologiques Arabes Syriennes 40: 111–21. 1991 Architettura e urbanistica di Ebla paleosiriana. La Parola del Passato 46: 304–71. 1993 L’aire sacrée d’Ishtar à Ebla: Résultats des fouilles de 1990– 1992. Comptes Rendus de l’Académie des Inscriptions et BellesLettres: 613–62. 1995 Fouilles à Ebla en 1993–1994: Les palais de la Ville Basse nord. Comptes Rendus de l’Académie des Inscriptions et BellesLettres: 651–81. 1997a Tell Mardikh, 1977–1996: Vingt ans de fouilles et de découvertes. La renaissance d’Ebla à l’époque des dynasties amorrhéennes. Akkadica 101: 1–29. 1997b Ebla and Syria in the Middle Bronze Age. Pp. 379–414 in The Hyksos: New Historical and Archaeological Perspectives, ed. E. D. Oren. University Museum Monographs 96. Philadelphia: University Museum. 1998 Les fortifications d’Ebla paléo-syrienne: Fouilles à Tell Mardikh (1995–1997). Comptes Rendus de l’Académie des Inscriptions et Belles-Lettres: 557–88. 2000a Monuments séculaires et religieux dans la Ville Basse et fortifications d’Ebla au BM I–II: Fouilles à Tell Mardikh, 1991–1997. Pp. 1029–41 in vol. 2 of Proceedings of the First International Congress on the Archaeology of the Ancient Near East, Rome May 18–23, 1998, ed. P. Matthiae et al. Rome: Università di Roma “La Sapienza.”

Paolo Matthiae 2000b Nouvelles fouilles à Ebla (1998–1999): Forts et palais de l’enceinte urbaine. Comptes Rendus de l’Académie des Inscriptions et Belles-Lettres: 567–610. 2001 A Preliminary Note on the MB I–II Fortification System at Ebla. Damaszener Mitteilungen (Festschrift für Ali Abu Assaf) 13: 29–51. 2002a About the Formation of Old Syrian Architectural Tradition. Pp. 191–209 in Of Pots and Plans: Papers on the Archaeology and History of Mesopotamia and Syria Presented to D. Oates, ed. L. al-Gailani Werr et al. London: Nabu. 2002b Fouilles et restaurations à Ebla en 2000–2001: Le Palais Occidental, la Résidence Occidentale et l’urbanisme de la ville paléosyrienne. Comptes Rendus de l’Académie des Inscriptions et Belles-Lettres: 531–74. 2003 Ishtar of Ebla and Hadad of Aleppo: Notes on Terminology, Politics and Religion of Old Syrian Ebla. Pp. 381–402 in Semitic and Assyriological Studies Presented to Pelio Fronzaroli by Pupils and Colleagues, ed. P. Marrassini. Wiesbaden: Harrassowitz. In press a Nuove scoperte archeologiche a Ebla: Il Palazzo Meridionale e l’urbanistica della città paleosiriana. In vol. 9 of Rendiconti dell’Accademia Nazionale dei Lincei, Classe di Scienze Morali Storiche e Filologiche. In press b Assedi divini e ostaggi mortali da Ebla a Troia. Sulla tradizione epica prima dell’Iliade. In vol. 9 of Rendiconti dell’Accademia Nazionale dei Lincei: Classe di Scienze Morali Storiche e Filologiche. Matthiae, P., and Pettinato, G. 1972 Il torso di Ibbit-Lim, re di Ebla. Rome: Università di Roma. Mazzoni, S. 1985 Elements of the Ceramic Culture of Early Syrian Ebla in Comparison with Syro/Palestinian EB IV. Bulletin of the American Schools of Oriental Research 257: 1–18. Neu, E. 1996a La bilingue hourro-hittite de Hattousha: Contenu et sens. Amurru 1: 189–95. 1996b Das hurritische Epos der Freilassung I: Untersuchungen zu einem hurritisch-hethitischen Textensemble aus Hattusha. Studien zu den Boghazköy-Texten 32. Wiesbaden: Harrassowitz. Nigro, L. 2000 Coordinating the MB I Pottery Horizon of Syria and Palestine. Pp. 1187–1212 in vol. 2 of Proceedings of the First International Congress on the Archaeology of the Ancient Near East, Rome, May 18–23, 1998, ed. P. Matthiae et al. Rome: Università di Roma “La Sapienza.” 2002 The MB Pottery Horizon of Tell Mardikh/Ancient Ebla in a Chronological Perspective. Pp. 297–328 in The Middle Bronze Age in the Levant: Proceedings of an International Conference on MB IIA Ceramic Material, Vienna, January 24–26, 2001, ed. M. Bietak. Vienna: Österreichische Akademie der Wissenschaften. Owen, D. 1992 Syrians in Sumerian Sources from the Ur III Period. Pp. 108–76 in New Horizons in the Study of Ancient Syria, ed. M. W. Chavalas and J. L. Hayes. Bibliotheca Mesopotamica 25. Malibu: Undena. Pinnock, F. 2000 Some Thoughts about the Transmission of Iconographies between North Syria and Cappadocia, End of the Third–Beginning of the Second Millennium b.c. Pp. 1397–1415 in vol. 2 of Proceedings of the First International Congress on the Archaeology of the Ancient Near East, Rome, May 18–23, 1998, ed. P. Matthiae et al. Rome: Università di Roma “La Sapienza.” In press a Middle Bronze Ceramic Horizon at Ebla: Typology and Chronology. Proceedings of the International Symposium, “From Relative Chronology to Absolute Chronology: The

The Archaic Palace at Ebla: A Royal Building between EB IVB and MB I Second Millennium bc in Syria–Palestine,” Rome, Nov. 29– Dec. 1, 2001, ed. P. Matthiae et al. Rome: Accademia Nazionale dei Lincei. In press b Change and Continuity in Art Viewed from Ebla. 2000 v. Chr.—Politische, wirtschaftliche und kulturelle Entwicklung im Zeichen einer Jahrtausende, ed. J.-W. Meyer and W. Sommerfeld. Frankfurt. Pomponio, F., and Xella, P. 1997 Les dieux d’Ebla: Étude analytique des divinités éblaïtes à l’époque des archives royales du IIIe millénaire. Alter Orient und Altes Testament 245. Münster: Ugarit-Verlag. Prechel, D. 1996 Die Göttin Ishara: Ein Beitrag zur altorientalischen Religionsgeschichte. Abhandlungen zur Literatur Alt-Syrien–Palästinas und Mesopotamiens 11. Münster: Ugarit-Verlag. Scandone Matthiae, G. 2004 Les scarabées d’Ebla. Pp. 195–202 in Scarabs of the Second Millennium bc from Egypt, Nubia, Crete and the Levant: Chronological and Historical Implications. Papers of a Symposium, Vienna, 10th–13th of January 2002, ed. M. Bietak and E. Czerny. Vienna: Akademie der Wissenschaften. Schwartz, G. M. 2001 The Pottery from Selenkahiye 1972, 1974 and 1975. Pp. 223– 339 in Selenkahiye: Final Report on the University of Chicago and University of Amsterdam Excavations in the Tabqa Reser-

103

voir, Northern Syria, 1967–1975, ed. M. N. van Loon. Uitgaven van het Nederlands Historisch-Archaeologisch Instituut te Istanbul 91. Leiden: Nederlands Historisch-Archaeologisch Instituut te Instanbul. Schwartz, G. M., and Weiss, H. 1992 Syria, ca. 10,000–2000 b.c. Pp. 221–43 in vol. 1 of Chronologies in Old World Archaeology, ed. R. W. Ehrich. 3rd ed. Chicago: University of Chicago Press. Sigrist, M. 1988 Isin Year Names. Institute of Archaeology Assyriological Series 2. Berrien Springs, MI: Andrews University Press. 1990 Larsa Year Names. Institute of Archaeology Assyriological Series 3. Berrien Springs, MI: Andrews University Press. Sollberger, E. 1959–60 Byblos sous les rois d’Ur. Archiv für Orientforschung 19: 120–22. Teissier, B. 1994 Sealing and Seals on Texts from Kültepe karum Level 2. Uitgaven van het Nederlands Historisch-Archaeologisch Instituut te Istanbul 70. Leiden: Nederlands Historisch-Archaeologisch Instituut te Instanbul. Ungnad, A. 1938 Datenlisten. Pp. 131–94 in vol. 2 of Reallexikon der Assyriologie, ed. E. Ebeling and B. Meissner. Berlin: de Gruyter.

Tel Beth-Shean and the Fate of Mounds in the Intermediate Bronze Age Amihai Mazar Institute of Archaeology, The Hebrew University of Jerusalem

William G. Dever dedicated much of his academic work to the period following the EB III, which he initially designated MB I and subsequently EB IV (1970; 1971; 1973; 1980; 1985; 1987; 1995; 2003). In this paper in his honor, I present the finds from this period at Tel BethShean discovered in the Hebrew University excavations. I will then examine the situation at other sites that were urban centers during EB III and compare these in light of our current knowledge of settlement patterns and the nature of society during the late third millennium bce in the Mediterranean regions of the southern Levant.

Terminology Albright (1960: 80–81) and Wright (1961: 87) referred to this period as the “Middle Bronze I” (MB I), and this term was in common use until the early 1980s and is still used in major reference works (e.g., Stern, ed. 1993: 1529). Following her excavations at Jericho, Kenyon designated the period “Intermediate Early Bronze–Middle Bronze” (1960: 180), and this term was accepted by her students, for example, Prag (1974). Lapp modified the term to “Intermediate Bronze Age” (IBA) (1966: vii), and this designation has been accepted by several Israeli scholars since the 1980s and in 1992 was recommended as the standard term for the period by the editors of Tel Aviv 19: 131. Other Israeli scholars continue to use the traditional term, “Middle Bronze I” (e.g., Gal and Covello-Paran 1996: 38). In the early 1970s, Oren proposed the term EB IV for this period (1973a: 58–60; 1973b), a designation that gained popularity among scholars, mainly in the U.S.A. (for example, Dever 1973 [and thereafter]; 1 Richard 1980; Palumbo 1990; Meyers, ed. 1997: 411). At present, all three terms are used simultaneously, resulting in confusion concerning both this and the following periods. In this paper, I use IBA (Intermediate Bronze Age), be1. It is noteworthy that in 1970, although he still used the MB I terminology, Dever suggested that the term EB IV be used, following Wright’s earlier proposal. Dever wrote: “We should call MB I instead ‘EB IV/MB I’ and expand it to make a composite period embracing former ‘EB IV’ and ‘MB I’ ” (1970: 150).

cause I find it most appropriate for the nature of this period, mainly from the socioeconomic and settlement pattern points of view. 2

Tel Beth-Shean during the IBA Stratigraphy The EB and MBA levels excavated by the University Museum of the University of Pennsylvania expedition at Tel Beth-Shean between 1921 and 1933 were only partially published. In FitzGerald’s publication of selected pottery from Levels XVIII–XI, between 9 and 11 items belonging to the IBA can be identified from among the 30 items attributed to Level XI (FitzGerald 1935: pls. 8–10), the others being EB III and MB II forms. 3 Unpublished plates prepared by F. James and kept in the University Museum show a similar picture: Level XI is mixed and contains EB III, IBA, and MB II forms. 4 The reason for this mixture was clarified by our renewed excavations at Tel BethShean, where we detected IBA material in parts of our Area R. 5 This excavation area included the northeastern and western parts of the LB I/IIA Level IX sanctuary excavated by A. Rowe (for summaries, see Mazar 1997: 64–66; 2003: especially 327, map fig. 3). The topography of this area dictated the nature of the stratigraphy: the summit of 2. In the general book I wrote in 1988–89 (Archaeology of the Land of the Bible, 10,000–586 b.c.e.), I used the combined term EB IV/MB I (Mazar 1990: 151). 3. The IBA forms are: FitzGerald 1935: pl. 9:10, 15 ( jar necks), 13– 14 (“envelope handles” of jars), 18–19 (troughs—perhaps EB III?); pl. 10:8–10, 15 (amphoriskoi). Plate 9:4, a hand-made cooking pot, could belong to either the IBA or MB II, and some of the holemouth jars could also date to the IBA. Some MB II pottery was also attributed to Level XI (pl. 8:11, 14, 17). 4. My thanks go to P. McGovern and the Department of Near Eastern Archaeology at the University Museum for allowing me to study these plates. 5. The excavations at Tel Beth-Shean were directed by the author on behalf of the Institute of Archaeology of the Hebrew University of Jerusalem between 1989 and 1996. The excavations were supported by the Beth-Shean Tourism Development Administration. R. Mullins helped in processing the stratigraphic data, and A. Ziv-Esudri helped in processing the finds from the EBA and IBA periods.

106

Amihai Mazar

Fig. 1. Tel Beth-Shean, view of Area R Squares P–Q/9–10, looking south: center, floor and pits of Stratum R-6 (IBA period); above: Hall 1230 of Level IX (LB II, 14th century bce).

the natural hill was in the southeastern corner of the mound, and the occupation levels descended to the northwest. In the northwestern part of Area R, we detected three MB II strata (R-5–R-3), all dated to the 17th–16th centuries bce (Maeir 1997: 65–101; Mazar 1997: 66–67; 2003: 326–29). A limited area of 25 sq. m excavated below these layers revealed five successive EB III phases, designated Strata R-12–R-7 (Mazar, Ziv-Esudri, and Cohen-Weinberger 2000). IBA remains were found in only certain parts of this area and were attributed to Stratum R-6 (figs. 1–2). It should be emphasized that these IBA remains were found in only a small part of Area R; in most of the area, MB II occupation levels were located immediately above the EB III levels, with no IBA remains separating them. In one place only (in the northeastern corner of Square P/10), two superimposed surfaces were detected, 0.45 m apart, both belonging to the IBA (Loci 10337 and 10322). The lower floor, set on a hard-packed earthen foundation, was above the last EB III occupation phase (R-7a). The floor was covered with a 0.20-meter-thick layer of accumulation or collapse (Locus 10328), including fallen mud

bricks and ash. To the west was a poorly preserved circular installation (10329) with a thin clay wall (outer diameter of 1.60 m, preserved to a height of 0.35 m). Inside and around this installation were mud-brick debris and collapse. The higher floor (Locus 10322) was white plastered, and the debris above it contained some IBA pottery, including an open bowl (fig. 3:2) and a teapot (fig. 5:1). In Squares Q/10 and part of Q/9 (figs. 1–2), we uncovered a very thin IBA occupation layer immediately below the pavement of a large MB II piazza (Stratum R-5–R-4) and above EB III occupation debris. The remains included a beaten earth floor (58115) covered by a thin layer of ash. Two whole IBA globular-shaped jars (figs. 3:3–4, 6:3) were found sunken in specially fitted pits in the floor, their rims flush with the floor surface. On the same floor were a teapot (figs. 5:2, 6:1), two jugs (figs. 5:3–4, 6:6), and two smashed globular jars (fig. 3:5–6). Some stones related to the floor did not have any apparent architectural function other than perhaps to support a tent or another kind of temporary structure. A few stones in the western part of the square seemed as though they could have been part of a

Tel Beth-Shean and the Fate of Mounds in the Intermediate Bronze Age

107

Fig. 2. Tel Beth-Shean, plan of Stratum R-6 in Area R Squares P–Q/9–10.

stone pavement. Square Q/9 included a floor surface (Locus 58098) at 88.57 m, where an MB II lime floor sealed a concentration of IBA finds in a thin layer immediately above the top of the EB III accumulation. The IBA finds included an oil lamp (figs. 5:7, 6:4) and several copper objects (fig. 7:1–7). In Squares J–K/6–7 in the western part of Area R, Locus 18310 yielded three flat basalt stones that could have been part of a pavement, covered by an ash layer containing fragments of a storage jar with envelope handles. In Locus 10506 to the east, several additional sherds from the IBA were found at 88.08–87.60 m. This locus, however, was mixed and also yielded some Early Bronze Age and MB II sherds. The IBA remains in Area R are thus scattered and scant and probably represent an ephemeral settlement. The nature of this settlement on the mound stands in contrast to

the elaborate burial caves containing rich finds in the Northern Cemetery of Beth-Shean (Oren 1973a: 19–60). The Pottery Bowls. The two vessels in fig. 3:1–2 are large open bowls or vats, with a diameter of 34 and 40 cm, respectively, and a height of ca. 10 cm. Number 2 has two elongated thumb-indented ledge handles located close to the rim and is decorated with diagonal stripes in a light red color, a decorative style typical of the Jordan Valley (dubbed “trickle painting” in Falconer 1987: 253; Palumbo 1990: 81). A red strip decorates the grooved part of the rim. While open vessels of this sort are known in the IBA pottery repertoire, they are rare. The closest parallels come from Tell Abu en-Niaj, some 10 km southeast of Beth-Shean (Palumbo 1990: fig. 37:1, with an identical grooved rim, but unpainted; see Palumbo 1990: 901 for a

Amihai Mazar

108

Fig. 3. Beth-Shean Stratum R-6 Pottery: Bowls and Globular Jars.

Tel Beth-Shean and the Fate of Mounds in the Intermediate Bronze Age

109

Figure 3 Pottery Descriptions No. 1

Vessel Bowl

Reg. No. 183110

Locus 18310

2

Bowl

103160

10322

3

Jar

580522/1

58115

4

Jar

580498/1

58115

5 6

Jar Jar

580506/16 580506/5

58115 58115

Description Very pale brown clay, gray core, many inclusions, concentric wetsmoothing lines Very pale brown clay, pink core, many inclusions, diagonal light-red lines on exterior and red line of rim, trimming Pale brown clay and core, many inclusions, concentric and irregular smoothing lines Pale brown clay and core, some inclusions, concentric smoothing lines, very light brown wash Brown and dark gray clay, dark brown core, some inclusions Pale brown clay and core, some inclusions, irregular smoothing lines

list of parallels, mainly from sites in the Jordan Valley). At Megiddo, similar flat bowls appear in a mixed context related to Stratum XIV (Loud 1948: pl. 15:16) and in Tomb 1101B Lower (Guy 1938: pl. 6:19). Examples of a grooved rim on bowls like our fig. 3:2 appear in domestic contexts at ºAfula (Gal and Covello-Paran 1996: 41, fig. 10:2) and at Óorbat Qishron in Lower Galilee, which also yielded a variety of other bowls (Smithline 2002: 26*– 29*, figs. 10:9–11, 11:4). Variations of these bowls at Qishron were made of cooking ware and are defined as “cooking bowls” (Smithline 2002: 30*, 33*, fig. 13:3–10). Some of these have a ledge handle below the rim, as in fig. 3:2. Smithline notes the scarcity of bowls of this sort in publications, because most published IBA pottery comes from tombs, while the daily domestic pottery is less well represented (2002: 26*). Large bowls like ours are known from a domestic context at Jebel Qaºaqir (Gitin 1975: 52*, fig. 2:14–16), but they differ in the details. At Iktanu in central Transjordan, a bowl similar to our fig. 3:1 appears in an earlier phase of the IBA (Prag 1974: 82, fig. 3:22). The rim of our fig. 3:2 recalls bowl rims from ºArºair in Transjordan (Prag 1974: 88, fig. 6:17), as well as from Jericho (Kenyon 1965: 47, fig. 21:1–2). Jars. Figure 3:3–4 and 6 are handleless jars with a rounded bottom. Figure 3:4 (photo fig. 6:3) is sack-shaped, while fig. 3:3 is more globular. Both have a flaring simple rim. A jar from the Northern Cemetery has a similar rim, but its body is more elongated (Oren 1973a: 34, Type M2, 178, fig. 22:1). Oren provides Syrian parallels for this type. Similar jars are common in northern assemblages (Dever’s Families North [N] and North-Central [NC] [1971: 198–205]), for example, from Megiddo Tomb 1101B Lower (Guy 1938: pl. 7:3) and a tomb at GevaºCarmel (Amiran 1974: 2, fig. 1:1–2). Figure 3:5 is the body of globular jar with two folded (“envelope”) ledge handles, typical of the period, also

found in the Northern Cemetery at Beth-Shean (Oren 1973a: 37, Types R1–R2, 174, fig. 20:1, 178; fig. 22:6), at Dahr Mirzbaneh (Lapp 1966: fig. 18:2), and at many other sites. Figure 4:1 is a small jar with an elongated body, flat base, flaring rim, and two “pushed” ledge handles (Oren 1973a: 37, Type R2; cf. 174, fig. 20:1 for similar handles on a different vessel form). The shape is common in the IBA (see, e.g., Oren 1973a: 170, fig. 18:1–3, among others), but in our case, there is a sharp angle between the shoulder and the neck, unlike most of the jars from the Northern Cemetery. While this sharp angle is a rare feature in the north, it is common in the central hill country and southern sites (for exceptions from ºAfula, see Gal and Covello-Paran 1996: 43, fig. 11:7, 10–11; and from Óorbat Qishron, Smithline 2002: 38*, fig. 16:2; at both sites, these jars are considered to be rare and to represent southern forms). Close parallels can be found at Jericho (Kenyon 1960: 228, fig. 86:1). Figure 4:4 is the neck of a jar with an incised decoration and a diagonally cut rim. A similar decoration, but without the borders, appears on jars from the Northern Cemetery (Oren 1973a: 176, figs. 21:2, 6; 23:3). Similar jar rims appear in domestic contexts at northern sites such as ºAfula (Gal and Covello-Paran 1996: 43, fig. 11:12, 15) and Óorbat Qishron (Smithline 2002: 35*–38*, fig. 16:6, and see discussion there). Cooking vessels. Figure 4:2–3 are cooking vessels with a short, out-turned rim. These vessels are rather large, with a rim diameter of 18 cm and a possible height of ca. 35 cm. Figure 4:2 has five horizontal grooves incised on the upper part of the body. These cooking vessels are typical of IBA northern groups; examples are known from various sites, such as Kedesh, Hanita, Maºayan Barukh, ºEnan, ºAfula, Óorbat Qishron, and the Golan Heights (see, respectively, Tadmor 1978: 15, 17, fig. 5; Eisenberg

Amihai Mazar

110

Fig. 4. Beth-Shean Stratum R-6 Pottery: Storage and Holemouth Jars and Cooking Vessels.

No. 1 2

Vessel Storage jar Cooking jar

Reg. No. 183110/99 103161

Locus 18310 10322

3

Cooking jar

580426/4

58102

4

Storage jar

580426/3

58102

5

Holemouth jar

183178/5

18310

Description Pale brown clay, concentric and irregular smoothing lines Discolored clay, some inclusions, burned, five incised concentric circles on body Pale brown and discolored clay, brown core, some inclusions, burned, concentric smoothing lines Pale brown clay and core, some inclusions, concentric smoothing lines, light reddish-brown band, incised diagonal lines between borders Reddish-brown clay and core, many inclusions, smoothing lines, very pale brown slip, red band on rim, horizontal red strips below rim

Tel Beth-Shean and the Fate of Mounds in the Intermediate Bronze Age

111

Fig. 5. Beth-Shean Stratum R-6 Pottery: Various Closed Vessels and a Lamp. .

No. 1 2 3 4 5 6 7

Vessel Teapot Teapot Jug Jug Amphoriskos Amphoriskos Lamp

Reg. No. 103160/99 580389/1 580441/15 580416/1 580415/1 580417/1 580380/1

Locus 10322 58115 58115 58115 58102 58107 58113

Description Pale brown clay, many inclusions, encrusted, two incised grooves Pale brown clay, some inclusions, smoothed surface Not available Very pale brown clay, some inclusions, encrusted, smoothing lines Discolored clay, some inclusions, burned black patches Pale brown clay, encrusted Discolored clay, many inclusions, traces of red slip, burned black mouth

1985: 62, fig. 3:5; Gal and Covello-Paran 1996: 39–41, fig. 10:12–17; Smithline 2002: 30*; Epstein 1985: 49, fig. 3:11–12). Beth-Shean is the southernmost distribution limit of this form (Oren 1973a: 34, 182, figs. 22:2–3; 24:2, 5–6). There are variations in the form and decoration: some of the vessels from Qishron have a thumb-print decoration on the neck and the vessels from Kedesh are smaller. Our rims are lower and the angle is sharper than in most of the known examples. The above-mentioned incised decoration on our fig. 4:2 is an unusual feature on this type. A jar from Megiddo (Guy 1938: pl. 21:1) is reminiscent of our fig. 4:2 in terms of rim shape and has a rope decoration that is not very different from our fig. 4:5. Oren (1973a: 35) and Tadmor (1978: 17) present parallels for this type from Syria (Hama K and J5, Tell Bila, and Til Barsip), while Gal and Covello-Paran (1996: 40) empha-

size their indigenous origin in local Early Bronze Age cooking pots. These vessels can be considered as successors of Early Bronze Age holemouth jars and forerunners of MB II holemouth cooking pots with a rilled rim. Holemouth jar. Figure 4:5 is a rim sherd of a small holemouth jar. Jars of this sort are common in IBA domestic contexts (Gitin 1975: 49*, fig. 1:15; Palumbo 1990: fig. 32:1; Smithline 2002: 30*–31*, 34*, fig. 14, with references to other sites). Teapots. Figure 5:1–2 are two small teapots, one without handles and the other with two small handles between the lower neck and shoulder (fig. 6:1). This form is common in all IBA assemblages and can be related to Syrian teapots (Oren 1973a: 30; for a few of the many parallels, see Guy 1938: pl. 12:2–4; Oren 1973a: 170, fig. 18:11–12; Tadmor 1978: 18, fig. 7).

Amihai Mazar

112

3

1

6 2 4

5

Fig. 6. Photo of pottery vessels from Beth-Shean Stratum R-6: 1. teapot (fig. 5:2); 2. amphoriskos (fig. 5:5); 3. jar (fig. 3:4); 4. lamp (fig. 5:7); 5. amphoriskos (fig. 5:6); 6. jug (fig. 5:4). Photo: G. Laron.

Jugs. Figure 5:3–4 are two jugs with a handle attached to the rim or at mid-neck, a flat base, and a somewhat squat body. Various jugs of this type appear in northern contexts during the IBA (see Oren 1973a: 29 for a discussion and parallels; for examples from the Northern Cemetery, see Oren 1973a: 170, fig. 18:7, 172; fig. 19:8; from Kedesh, Tadmor 1978: 18, fig. 7:70–482, 70–489; from the Golan, Epstein 1985: 47, fig. 2:3). A few examples come from the south (Dever 1980: 43, fig. 4:12) and from Transjordan (Dever 1980: 41, fig. 3:12). The more elaborate jugs from Yavneªel, with a strip handle and red slip, may represent an earlier phase of the IBA (Liebowitz and Porat 1992: 5, fig. 1:2–7). Amphoriskoi. Figure 5:5 (photo fig. 6:2) is a small amphoriskos (height of 10 cm) with a globular body, simple rim, and a rounded, slightly pointed base, a rare feature in the IBA (cf. Bahat 1976: 30, fig. 3:4); most IBA amphoriskoi have a flat base (Oren 1973a: 170, fig. 18:10). Figure 5:6 (photo fig. 6:5) is a small amphoriskos with a flat base and two debased ledge handles. Amphoriskoi of this general shape are common in the Northern Cemetery at Beth-Shean, but they usually have two lug handles on the

lower neck (e.g., Oren 1973a: 176, 178, figs. 21:16, 22:11); none have ledge handles like our examples. Lamp. Figure 5:7 (photo fig. 6:4) is a small lamp (8 cm in diameter) with a single spout. This is a rare form in the IBA, when most lamps had four spouts (like all the examples from the Northern Cemetery at Beth-Shean [Oren 1973a: 33]). Several single-spouted lamps come from IBA tombs at Megiddo (Guy 1938: pls. 10:27, 12:10–11), but these have a rounded base and are reminiscent of MB II lamps. A group of lamps with a high foot found in the cave at Kedesh also have only one spout (Tadmor 1978: 23, fig. 9). This assemblage belongs to the regional group discussed by Oren (1973a; 1973b) and defined by Dever as Family NC (1980: 40). It is typical of the Jordan Valley on both sides of the Jordan River (e.g., in the Beth-Shean Northern Cemetery and at Tell Abu en-Niaj). Although most of the examples in this group come from tombs, our assemblage is from a domestic context. This provides a rare opportunity to compare tomb deposits with domestic assemblages during this period. It is remarkable that small vessels such as teapots and amphoriskoi, typical of grave

Tel Beth-Shean and the Fate of Mounds in the Intermediate Bronze Age

113

Fig. 7. Beth-Shean Stratum R-6 copper and stone objects: 1. curled pin (left, as found; right, reconstructed); 2–3. chisels; 4–5. bracelets; 6. perforated stone; 7. grinding stone.

offerings, appear in a domestic context alongside larger jars, large open bowls, and cooking ware, the last two of which are almost entirely absent in the tomb assemblages. Metal and Stone Objects Several copper objects were found together in Locus 58098. Pin. Figure 7:1 is a pin with curled top, 14 cm long, found bent. Pins of this sort are common in IBA northern Israel, Transjordan, and Syria (Palumbo 1990: 111). A single example was found in the Northern Cemetery at Beth-Shean (Oren 1973a: 40, 172, fig. 19:11). Additional examples from northern Israel are known from the Megiddo tombs, the Kedesh cave, the tomb near ºEnan, the dolmens in the Golan, and the tombs near Galºed (respectively, Guy 1938: pl. 114:5; Tadmor 1978: 25, fig. 10; Eisenberg 1985: 70, fig. 10:62–65; Epstein 1985: 47, fig. 2:8; Meir 1974: 32, fig. 7:9–10; for Syrian examples, see Eisenberg 1985: 71, n. 31). Only a few pins of this type

reached sites south of the Jezreel Valley on the Sharon Plain (Gophna and Ayalon 2003: 62, fig. 12:3) and at Dahr Mirzbaneh (Lapp 1966: 54, fig. 2:8), as well as northern Transjordan (Prag 1974: 94, fig. 9:2–9, from Tell el-Óußn). Chisels. Figure 7:2–3 are two small chisels (height of ca. 7 cm) with a flattened upper part and bearing marks of working. No parallels for these objects were found among the IBA metal objects published to date. Bracelets. Figure 7:4–5 are two simple bracelets, opened on one side, 4.8 cm in diameter. Simple bracelets of this sort are known from several contexts in the Upper Jordan Valley, such as the Kedesh cave, the ºEnan tomb, and the dolmens in the Golan (respectively, Tadmor 1978: 25, fig. 10; Eisenberg 1985: 70, fig. 10:66–72; Epstein 1985: 47, fig. 2:1), as well as in northern Syria (Eisenberg 1985: 72, n. 35; for a discussion and references, see also Palumbo 1990: 112). Perforated stone. Figure 7:6 is a perforated, ringshaped basalt stone, 11 cm in diameter and 6 cm high.

Amihai Mazar

114 Similar stone objects were found at several IBA sites and were interpreted as digging-stick weights (Palumbo 1990: 117, fig. 59:1; Smithline 2002: 41*–42*, fig. 18:4; see also Tadmor 1978: 27, fig. 11:17–18). Grinding stone. Figure 7:7 is a “loaf-shaped” upper grinding stone made of basalt. Summary Following the destruction of the EB III city at BethShean, there was a brief period of occupation during the IBA. This was a patchy, scattered settlement, with almost no architectural features. It may represent seasonal occupation and was of short duration. The life-style this settlement reflects and the pottery and metal objects from this period at Beth-Shean are distinctly different from those of the EB III period. Was there a gap in time between the end of the EB III city and the foundation of the IBA settlement? How long could such a gap have lasted? Who were the inhabitants of the IBA settlement? The contrast between the IBA pottery assemblage and the EB III assemblage found only a few cm lower is very sharp. Although some of the IBA vessels recall EB III forms, the two assemblages differ in many respects; for example, in the appearance of new shapes and in clay composition, firing, color, and surface treatment. Thus, there must have been a break between these two periods, and this break appears to have been both temporal and cultural. The short-term IBA settlement was followed by an occupation gap at Beth-Shean that lasted until an advanced stage of MB II—that is, ca. 1700 bce.

The Fate of Other EB III Cities in the IBA The results of the Tel Beth-Shean excavations described above call for a comparative survey of the fate of other EB III cities in the IBA, in light of the general settlement pattern of the period. This issue is barely addressed in the current literature on the IBA, except most recently by Dever (2003: 46–49). 6 This comparison shows, in my view, a contrast between the presence of many newly founded sedentary villages in the IBA and the abandonment or very meager IBA settlement at sites that had been flourishing cities in the EB III cities. At the following EB III urban sites, only sparse IBA remains were found, as at Beth-Shean: Tel Dan. A few scattered IBA sherds were found on top of the ruins of the EB III city in two of the excavation areas (Biran 1994: 44–45). 6. Dever 2003, which appeared after the completion of the first draft of this paper, addresses similar questions to those raised above, although his answers differ somewhat.

Hazor. Stratum XVIII represents the IBA, but no architectural remains could be attributed to this “stratum.” Scattered IBA sherds (including of black Syrian teapots) were found in Areas A and L (Ben-Tor et al. 1997: 194). Beth Yerah. A few IBA sherds are reported to have been found in the southeastern tip of the mound (Raphael Greenberg, personal communication). Megiddo. Fifteen–sixteen published pottery vessels attributed to Strata XV–XIIIB belong to the IBA (Loud 1948: pls. 8:2–4, 9:12–13, 20 [Stratum XV]; pls. 13:6–7, 15:14, 16–17(?), 18, 22 [Stratum XIV]; pl. 16:16–18 [Stratum XIIIb]). These appear on the pottery plates together with EB III or MB IIA (MB I) pottery and thus probably represent a very poor settlement that could not be well defined stratigraphically. Dunayevsky and Kempinski (1973: 172–73) have suggested that EB III Temple 4040 was filled and a small cult chamber created inside it during the IBA, but this dating is based on a single fenestrated axe found in the fill, and the object may provide only a terminus post quem for the fill (cf. Aharoni 1993: 1008). Kempinski (1989: 39–40) wrote, on the one hand, that Megiddo was inhabited by nomads for a short period of time, but on the other hand, he attempted to reconstruct a plan with some substantial buildings (1989: 40, fig. 17). This reconstruction is not based on any actual data, because none of the structures on this plan can securely be dated to the IBA, and they could well belong to the subsequent MB IIA (MB I) period. Thus, it appears that the IBA settlement at Megiddo was not much different from that at Beth-Shean: a sparse, nonurban occupation on top of the ruins of the EB III city. Yoqneºam. A few EB III and IBA sherds were found out of context in MB II fills and were thought to represent settlement strata from this period and were assigned to “Stratum XXV” and “Stratum XXIV,” respectively (BenTor 1993: 811). However, the final report (Ben-Tor, BenAmi, and Livneh 2005) does not mention any finds from the IBA, and Stratum XXIV is assigned to the following MB IIA period. Jericho. Kenyon attributed three stratigraphic phases of poor mud-brick buildings to this period (1981: 105–8, 166–67, 213–15). Prag summarized these finds as follows: “The EB–MB occupation seems, therefore, to have covered the whole of the tell area in an uneven and perhaps ephemeral pattern of wide-spread occupation, with thinwalled and poorly built mud brick houses set into terraces in the slopes . . . there seems no certain evidence for reuse of Early Bronze houses” (Prag 1986: 65). Thus, Jericho appears to be an exception. At many other EB III sites, no IBA activity was found whatsoever. Excavated examples include Rosh Haniqra, Tel Yakush, Tel Qashish, Tel Dothan, ºAi, Gezer, Yarmuth,

Tel Beth-Shean and the Fate of Mounds in the Intermediate Bronze Age Tel Halif, Tell el-Hesi, and Tel ºIra. In several cases (for example, at Tel Rehov, Gezer, Lachish, and Bab edhDhraº), there is evidence of IBA settlement close to but outside the location of the EB III fortified city. 7 This picture of poor, patchy, and ephemeral occupation (probably in the second half of the period, after a gap in time) at earlier EB III urban sites or their complete abandonment, therefore, reflects a general phenomenon throughout the country. In Transjordan, it appears that continuity from the EB III to the IBA was more pronounced and has been observed at several sites: Khirbet Zeraqun. An IBA village is reported to have existed on top of the ruins of the EB III fortified city (Palumbo 1990: 58). Feqeiqes (in the Karak region). An IBA level was found on top of the large EB II–III site (Palumbo 1990: 100). Ader (in Moab). There may have been continuity from EB III to IBA, although this has not been supported by a thorough stratigraphic excavation (Palumbo 1990: 98–99). Bab edh-Dhraº. A sparse IBA settlement was found on top of the ruins of the EB III town in only one area (Field XVI, perhaps a cultic site), while in other areas there was clearly no settlement after the end of EB III. On the other hand, evidence of settlement was found in three different locations outside the fortified EB III town (Schaub 1993: 135–36; Rast and Schaub 2003: 398–448). 8 Both EB II–III pottery and IBA pottery were found at two additional unexcavated sites in Transjordan: Tell ºAlla and Zeazºiyyeh in the ºAjlun (Palumbo 1990: 91–92).

Sedentism versus Pastoralism in the IBA Unlike the earlier image of the IBA as lacking permanent settlements, it has become clear that many agricultural villages existed during this period throughout the 7. At Tel Rehov, evidence of an IBA settlement was found on the alluvial plain west of the mound; the mound itself is known to have been fortified during EB III. Near Gezer, the site of ºEn Yered is an IBA settlement (Shavit 2000: 209). At Lachish, only a few sherds of this period were found on the mound. Ussishkin (2004: 54) follows Tufnell in negating any actual settlement on the mound during the IBA, while Gophna and Blockman (in Ussishkin 2004: 892–96) anticipate a sparse settlement on the mound and in its vicinity. It is clear, however, that the main IBA settlement was in Area 1500, northwest of the mound. 8. The following statement by Rast and Schaub regarding the EB IV stratum at Bab edh-Dhraº (Stratum I) is of interest in this context: “two situations present themselves to anyone considering Stratum I. . . . One was the choice by the Stratum I settlers to occupy areas outside the confines of the old town, the other to use a section within the area of the former town under very limited conditions” (2003: 398).

115

Mediterranean region, as well as in the semiarid zones of both Cisjordan and Transjordan (as anticipated by Prag [1974: 102]; the main study is Palumbo 1990: especially pp. 41–62, 132, 163–87; see also Kamaisky 1995; Dever 2003). Many of these settlements were discovered during surface surveys or salvage operations in places where there was no earlier EBA occupation, and in most cases, they were abandoned after the end of the IBA. The settlers preferred new locations, unrelated to the previous EB III settlements on mounds. They chose various types of ecological niches: alluvial plains, mountainous zones, and semiarid regions. Thus, for example, Zori (1962) identified 22 IBA sites in the Beth-Shean Valley, and in the semiarid zone of the northeastern Samaria hills, southwest of the Beth-Shean Valley, Zertal discovered 39 sites dating to the IBA, following a decline in settlement in EB III (1996: 74–75, 807). Dagan (2000) identified 23 small settlements in the Shephelah, scattered in a pattern very different from that of the EB III settlements in this area. Only after they were excavated or examined in salvage operations (in the Ramat Beth-Shemesh area) could he determine that some of these were agricultural settlements situated near fertile land and water sources (Dagan 2000: 251–52). Several additional sites are known from salvage operations and surveys on the Coastal Plain and in the foothills of Samaria (for a list, see Shavit 2000: 207–9). Many IBA sites known from surface surveys yielded pottery only from this period, indicating that they were single-period sites, but the nature of these settlements cannot be defined without excavation. Therefore, the few sites excavated over the past decade are of particular interest. Following is a short list. Shaºar Hagolan. A large, well-planned village was found (although not systematically excavated) on top of the Pottery Neolithic period site (Eisenberg 1993). Óorbat Qishron. A single-period IBA settlement on the alluvial plain near the multiperiod site of Qishron (Smithline 2002). Two additional IBA sites were discovered close to these sites, one of them near a spring (1 km away; Smithline 2002: 44*). Smithline suggests that the choice of the wetter alluvial plain for the site indicates dryer climatic conditions during this period (2002: 44*). 9 Near Migdal HaºEmeq. An IBA settlement was excavated by K. Covello-Paran under the auspices of the Israel Antiquities Authority.

9. The following quotation from the Óorbat Qishron excavation report is noteworthy: “The idea of the IBA as an essentially pastoral society is not supported by the finds of this excavation. The quality of construction of the walls and floors, the structural changes undertaken on certain structures and the material finds—ceramic, flint and basalt—indicate the existence of a permanently based, agriculturally oriented settlement which experienced a relatively long existence” (Smithline 2002: 45*).

Amihai Mazar

116 ºAfula. A meager settlement was found on top of the EB I occupation after a gap in EB II–III; this settlement, in turn, was followed by an occupation gap in MB IIA (MB I) (Gal and Covello-Paran 1996: 38–44). Emek Rephaim. An extensive settlement along the wide wadi to the west of Jerusalem was excavated, including massive stone buildings. The faunal evidence of pigraising is understood as indicating a permanent sedentary settlement (Eisenberg 1989–90; Horwitz 1989). Tel Ashir. A small site on the Sharon Plain that yielded evidence of an open-air cultic place with standing stones (Gophna and Ayalon 2003). Some of the IBA sites were fortified, such as Khirbet Iskander (Richard 1993), 10 Khirbet el-Mite in Wadi Malikh, and er-Rahwe near the Zebabdeh Valley in northern Samaria (Zertal 1996: 270–71, 126–27, respectively), 11 and possibly a few additional sites (Palumbo 1990: 132). These may have been regional centers of power and control. It thus appears that a large part of the IBA population was composed of sedentary farmers who lived permanently in newly established villages based on a mixed subsistence economy (agriculture and animal-raising, including pigs). Agricultural villages of this sort seem to be abundant during this period in all the Mediterranean zones of the country. However, as in any other period in the history of the Levant, pastoralists were an important component of the dimorphic society characteristic of the region throughout history. Thus, many of the cemeteries of this period, including the burial caves used for secondary burials of individuals, may have belonged to such a nonsedentary population (Dever 1987; 1995), although this cannot be ascertained with certainty. Some of these cemeteries are found near mounds that were sparsely inhabited (Beth-Shean, Megiddo, Jericho, and Lachish). The proportion of the population represented by the pastoralists or seminomads may have increased in this period due to the collapse of the previous urban structure, but it is doubtful that they constituted the majority. It also appears that in this new settlement pattern the ruined EB III cities were deliberately abandoned because they were considered inappropriate as settlements for the sedentary population, which preferred new locations for its villages. The poor, patchy, and perhaps ephemeral settlements that did exist at the sites of abandoned EB III cities, such as the settlement found at Tel Beth-Shean, may be attributed to the pastoral seminomadic component of 10. The results of recent excavations at the site appear to indicate that the fortifications attributed to the IBA at Khirbet Iskander were in fact reused EB II walls (Dever 2003: 49, n. 11). 11. Both of these sites are located on hilltops, are ca. 2 acres in area, and are surrounded by a stone wall. At the first, a stone glacis and gate were identified. The ceramic finds from both date only to the IBA.

the population or to small sedentary groups that returned to these sites for a short time and on a small scale after an occupation gap, while most of the sedentary population deliberately avoided these sites.

Conclusions How can this complete change in settlement pattern and what appears to be the avoidance of previous sites of EB III cities by the IBA sedentary population be explained? The answer must be rooted in the processes and events that led to the collapse of the EB III urban system. This vast subject is beyond the scope of this paper, but a few points should be made. In the major excavated EB III cities, such as Beth Yera˙, ºAi, and Yarmuth, no indication of decline prior to the destruction/abandonment was found, and it appears that these thriving EB III cities were suddenly abandoned or destroyed when they were at their peak. Thus, the end of the EB III urban system should not be defined as a “process” but, rather, as an “event.” Various reasons for this catastrophe have been suggested: outside invasions (“Amorite,” Egyptian), internal strife, inner system collapse, or drought due to climatic changes (see the essays by Dever, Esse, Kochavi, and Rosen in Miroschedji, ed. 1989; Rosen 1995; Greenberg 2002: 97–105). I would add the possibility of plagues as a factor. Combined factors (such as a drought that could have led to severe competition over food resources that in turn could have led to wars between cities) could have caused their destruction, the disappearance of the centralized authorities, and anarchy. Plagues could easily have erupted under crisis conditions. Such an assumed chain of events could have caused the destruction and abandonment of cities, massive death, and population decline. A dramatic end of this sort to the centralized, authoritative, EB III urban system may explain the desertion or near desertion of many of these cities during the IBA. The memories of the previous regimes and their catastrophic end were probably strong enough to keep the sedentary population of the IBA away from most of these city sites, which perhaps were considered “cursed.” 12 Thus, a conscious selection of new settlement sites and abandonment of the old city sites may be explained on cognitive grounds as being related to human beliefs, fears, and feelings rather than to environmental/ecological factors. This cognitive explanation cannot be tested or proved, but it may provide a possible avenue for understanding both the end of the EB III urban system and the new settlement pattern and life-style during the IBA. 12. Compare with the biblical narrative regarding Jericho in Josh 6:26.

Tel Beth-Shean and the Fate of Mounds in the Intermediate Bronze Age

117

References Aharoni, Y. 1993 Megiddo (Bronze Age). Pp. 1003–12 in vol. 3 of The New Encyclopedia of Archaeological Excavations in the Holy Land, ed. E. Stern. Jerusalem: Israel Exploration Society. Albright, W. F. 1960 The Archaeology of Palestine. Rev. ed. Harmondsworth: Penguin. Amiran, R. 1974 A Tomb Group from Gevaº-Carmel. ºAtiqot 7: 1–12. Bahat, D. 1976 A Middle Bronze Age I Cemetery at Menahamiya. ºAtiqot 11: 27–33. Ben-Tor, A. 1993 Jokneam. Pp. 805–11 in vol. 3 of The New Encyclopedia of Archaeological Excavations in the Holy Land, ed. E. Stern. Jerusalem: Israel Exploration Society. Ben-Tor, A., et al. 1997 Hazor V: An Account of the Fifth Season of Excavation, 1968. Jerusalem: Israel exploration Society. Ben-Tor, A.; Ben-Ami, D.; and Livneh, E. 2005 Yoqneºam III: The Middle Bronze and Late Bronze Ages. Qedem Reports 7. Jerusalem: The Hebrew University of Jerusalem. Biran, A. 1994 Biblical Dan. Jerusalem: Israel Exploration Society. Dagan, Y. 2000 The Settlement in the Judean Shephela in the Second and First Millennium b.c. Unpublished Ph.D. dissertation, Tel Aviv University (Hebrew). Dever, W. G. 1970 The Middle Bronze I Period in Syria and Palestine. Pp. 132–63 in Near Eastern Archaeology in the Twentieth Century: Essays in Honor of Nelson Glueck, ed. J. A. Sanders. New York: Doubleday. 1971 The Peoples of Palestine in the Middle Bronze I Period. Harvard Theological Review 64: 197–226. 1973 The EB IV–MB I Horizon in Transjordan and Southern Palestine. Bulletin of the American Schools of Oriental Research 210: 37–63. 1980 New Vistas on the EB IV (MB I) Horizon. Bulletin of the American Schools of Oriental Research 237: 55–64. 1985 From the End of the Early Bronze Age to the Beginning of the Middle Bronze. Pp. 113–35 in Biblical Archaeology Today, ed. J. Amitai. Jerusalem: Israel Exploration Society. 1987 Funerary Practices in EB IV (MB I) Palestine: A Study in Cultural Discontinuity. Pp. 9–19 in Love and Death in the Ancient Near East: Essays in Honor of Marvin H. Pope, ed. J. H. Marks and R. M. Good. Guilford, CT: Four Quarters. 1995 Social Structure in the Early Bronze IV Period in Palestine. Pp. 282–96 in The Archaeology of Society in the Holy Land, ed. T. E. Levy. London: Leicester University Press. 2003 The Rural Landscape of Palestine in the Early Bronze IV Period. Pp. 43–60 in The Rural Landcape of Ancient Israel, ed. A. M. Maeir, S. Dar, and Z. Safrai. British Archaeological Reports International Series 1121. Oxford: Archaeopress. Dunayevsky, I., and Kempinski, A. 1973 The Megiddo Temples. Zeitschrift des deutschen PalästinaVereins 89: 161–87. Eisenberg, E. 1985 A Burial Cave of the Early Bronze IV (MB I) Near ºEnan. ºAtiqot 17: 59–74.

1989–90 Nahal Refaªim. Excavations and Surveys in Israel 9: 150–56. 1993 Shaºar ha-Golan: The Middle Bronze I Age Site. Pp. 1342–43 in vol. 4 of The New Encyclopedia of Archaeological Excavations in the Holy Land, ed. E. Stern. Jerusalem: Israel Exploration Society. Epstein, C. 1985 Dolmens Excavated in the Golan. ºAtiqot 17: 20–58. Falconer, S. E. 1987 Village Pottery Production and Exchange: A Jordan Valley Perspective. Pp. 251–59 in vol. 3 of Studies in the History and Archaeology of Jordan, ed. A. Hadidi. Amman: Department of Antiquities of Jordan. FitzGerald, G. M. 1935 The Earliest Pottery of Beth-shan. The Museum Journal 24: 5–22. Gal, Z., and Covello-Paran, K. 1996 Excavations at ºAfula , 1989. ºAtiqot 30: 25–68. Gitin, S. 1975 Middle Bronze I “Domestic” Pottery at Jebel Qaºaqir. EretzIsrael 12 (Nelson Glueck volume): 46*–62*. Gophna, R., and Ayalon, E. 2003 Tel Ashir: An Open Cult Site of the Intermediate Bronze Age on the Bank of the Poleg Stream. Eretz-Israel 27 (Miriam and Hayim Tadmor volume): 55–65. Greenberg, R. 2002 Early Urbanizations in the Levant: A Regional Narrative. London: Leicester University Press. Guy, P. L. O. 1938 Megiddo Tombs. Chicago: Oriental Institute. Horwitz, L. K. 1989 Sedentism in the Early Bronze IV: A Faunal Perspective. Bulletin of the American Schools of Oriental Research 275: 15–25. Kamaisky, Y. 1995 The Settlement of the Land of Israel during the Middle Bronze I Period. Unpublished M.A. thesis, Hebrew University of Jerusalem. Kempinski, A. 1989 Megiddo: A City State and Royal Centre in North Israel. Munich: Beck. Kenyon, K. M. 1960 Excavations at Jericho I. London: British School of Archaeology in Jerusalem. 1965 Excavations at Jericho II. London: British School of Archaeology in Jerusalem. 1981 Excavations at Jericho III: The Architecture and Stratigraphy of the Tell. London: British School of Archaeology in Jerusalem. Lapp, P. W. 1966 The Dahr Mirzbaneh Tombs: Three Intermediate Bronze Age Cemeteries in Jordan. New Haven: American Schools of Oriental Research. Liebowitz, H., and Porat, P. 1992 Early Bronze IV Tombs from Yavneªel. ºAtiqot 21: 1–9. Loud, G. 1948 Megiddo II: Seasons of 1955–1939. Chicago: Oriental Institute. Maeir, A. 1997 The Material Culture of the Central Jordan Valley during the Middle Bronze II Period: Pottery and Settlement Pattern. Unpublished Ph.D. dissertation, Hebrew University of Jerusalem. Mazar, A. 1990 Archaeology of the Land of the Bible, 10,000–586 b.c.e. New York: Doubleday. 1997 Four Thousand Years of History at Tel Beth-Shean: An Account of the Renewed Excavations. Biblical Archaeologist 60: 62–76.

118 2003 Beth-Shean in the Second Millennium b.c.e.: From Canaanite Town to Egyptian Stronghold. Pp. 323–40 in The Synchronisation of Civilization in the Eastern Mediterranean in the Second Millennium b.c. II: Proceedings of the SCIEM 2000–EuroConference, Haindorff, 2nd–7th May 2001, ed. M. Bietak. Vienna: Austrian Academy. Mazar, A.; Ziv-Esudri, A.; and Cohen-Weinberger, A. 2000 The Early Bronze II–III at Tel Beth-Shean: Preliminary Observations. Pp. 55–278 in Ceramics and Change in the Early Bronze Age of the Southern Levant, ed. G. Philip and D. Baird. Levantine Archaeology 2. Sheffield: Sheffield Academic Press. Meir, M. M. 1974 Ancient Tombs near Kibbutz Galºed. Galºed: Kibbutz Galºed (Hebrew [pamphlet]). Meyers, E. M., ed. 1997 The Oxford Encyclopedia of Archaeology in the Near East. New York: Oxford University Press. Miroschedji, P. de, ed. 1989 L’urbanisation de la Palestine à l’âge du Bronze ancien: Bilan et perspectives des recherches actuelles. 2 vols. British Archaeological Reports International Series 527. Oxford: British Archaeological Reports. Oren, E. D. 1973a The Northern Cemetery of Beth Shan. Leiden: Brill. 1973b The Early Bronze IV Period in Northern Palestine and Its Cultural and Chronological Setting. Bulletin of the American Schools of Oriental Research 210: 20–36. Palumbo, G. 1990 The Early Bronze Age IV in the Southern Levant: Settlement Patterns, Economy and Material Culture of a ‘Dark Age’. Contributi e Materiali di Archeologia Orientale 3. Rome: Università degli Studi di Roma “La Sapienza.” Prag, K. 1974 The Intermediate Early Bronze–Middle Bronze Age: An Interpretation of the Evidence from Transjordan, Syria and Lebanon. Levant 6: 69–116. 1986 The Intermediate Early Bronze–Middle Bronze Age Sequence at Jericho and Tell Iktanu Reviewed. Bulletin of the American Schools of Oriental Research 264: 61–72. Rast, W. E., and Schaub, R. T. 2003 Bâb edh-Dhrâº: Excavations at the Town Site (1975–1981). 2 vols. Winona Lake, IN: Eisenbrauns.

Amihai Mazar Richard, S. 1980 Towards a Consensus of Opinion on the End of the Early Bronze Age in Palestine–Transjordan. Bulletin of the American Schools of Oriental Research 237: 5–34. 1993 Iskander, Khirbet. Pp. 649–52 in vol. 2 of The New Encyclopedia of Archaeological Excavations in the Holy Land, ed. E. Stern. Jerusalem: Israel Exploration Society. Rosen, A. M. 1995 The Social Response to Environmental Change in Early Bronze Age Canaan. Journal of Anthropological Archaeology 14: 26– 44. Schaub, R. T. 1993 Bab edh-Dhraº. Pp. 130–36 in vol. 1 of The New Encyclopedia of Archaeological Excavations in the Holy Land, ed. E. Stern. Jerusalem: Israel Exploration Society. Shavit, A. 2000 Settlement Patterns in the Ayalon Valley in the Bronze and Iron Ages. Tel Aviv 27: 189–230. Smithline, H. 2002 An Intermediate Bronze Age Site at Óorbat Qishron. Pp. 21*– 46* in Eretz Zafon: Studies in Galilean Archaeology, ed. Z. Gal. Jerusalem: Israel Antiquities Authority. Stern, E., ed. 1993 The New Encyclopedia of Archaeological Excavations in the Holy Land. Jerusalem: Israel Exploration Society. Tadmor, M. 1978 A Cult Cave of the Middle Bronze Age I near Qedesh. Israel Exploration Journal 28:1–30. Ussishkin, D. 2004 The Renewed Archaeological Excavations at Lachish (1973– 1994). Tel Aviv: Tel Aviv University Press. Wright, G. E. 1961 The Archaeology of Palestine. Pp. 73–112 in The Bible and the Ancient Near East: Essays in Honor of William Foxwell Albright, ed. G. E. Wright. Garden City, NY: Doubleday. Zertal, A. 1996 The Manasseh Hill Country Survey. Tel Aviv: Ministry of Defense (Hebrew). Zori, N. 1962 An Archeological Survey of the Beth-Shean Valley. Pp. 135–98 in The Beth-Shean Valley: The 17th Archaeological Convention of the Israel Exploration Society. Jerusalem: Israel Exploration Society.

Early Bronze Age IV Transitions: An Archaeometallurgical Study Suzanne Richard Institute of Archaeological Research, Gannon University Erie, Pennsylvania

It is an honor to offer this essay on an Early Bronze Age IV artifact as a tribute to William G. Dever, the scholar who, arguably, has done more to advance our knowledge of the EB IV period than any other. It is quite apropos for me as well, because I am reassessing the topic of metal weapons that I discussed long ago in my dissertation on the Early Bronze Age (1978), supervised by Dever. Moreover, what could be more appropriate than to discuss a topic on which Dever has made significant contributions to the field (as with everything else in EB IV), namely, metal artifacts and specialized metallurgy (see, e.g., Dever 1972; 1973; 1975; Dever and Tadmor 1976; Cohen and Dever 1979; Merkel and Dever 1989). On a final, more personal note, I would like officially to voice my appreciation for a mentor/teacher/scholar who has always supported and promoted the work of his students, while graciously accepting criticism by the same students! He has been a leader in the field for over 40 years, persuasively arguing the merit of his views, yet always one of the first to modify long-held positions if warranted by new data and theory—the mark of a true scholar. One need only point to his developing views conceptualizing the EB IV period to support that contention (Dever 1970; 1973; 1980; 1989; 1995; 2003). Indeed, he will probably agree with the proposed revisions to his views reported in this essay on a metal artifact from Khirbet Iskander. The artifact is significant because it provides new information Author’s note: I gratefully acknowledge the generosity of the Materials Science and Engineering Department, University of Arizona, for carrying out the metallurgical analysis as a tribute to William G. Dever. Special thanks go to Brent Hiskey, who graciously spent time discussing the results of the analysis with me, and to James Muhly for reading an earlier draft of this paper and for the helpful discussions that ensued. Likewise, special thanks also go to my colleague, Jesse C. Long, Jr., for his insights and suggestions on an earlier draft of this essay. Of course, none of these scholars bears any responsibility for the opinions expressed in this final product. I wish to thank the Department of Antiquities of Jordan, in particular Ghazi Bisheh, then Director General, who kindly permitted the loan of the artifact for study. Finally, I would like to acknowledge the fine work of our staff artist, Sarah White, who drew the original illustration of the Iskander spearhead.

on the transition to tin/bronze-alloying processes in the southern Levant at the end of the third millennium bce.

Introduction A bronze socketed spearhead was found in the 1997 excavations at Khirbet Iskander, a site in Transjordan located on the banks of the Wadi el-Wala at a crossing point on the ancient “King’s Highway,” some 56 km south of Amman (fig. 1). The artifact dates to the EB IV, ca. 2300–2000 bce. The site is best known for this cultural time span and well-preserved remains from the period; however, the continuing excavations have revealed urban EB II–III occupation, as well as an EB I stratum (Richard and Long 1995; 1998; 2000; Richard, Long, and Libby 2001; Richard 2002). Elemental analysis by means of Energy Dispersive Spectroscopy (EDS) demonstrates that this spearhead is of a standard tin-bronze alloy. Given its clear EB IV date, the Khirbet Iskander spearhead is a remarkable find that can inform us about archaeometallurgical processes in the southern Levant at the end of the third millennium bce. The archaeology of metallurgy has become an increasingly critical field of study for the Near East in general (see Pigott’s 1996 overview), but it is the recent discovery in our region—particularly in the area of the Faynan, south of the Dead Sea, in the Wadi ºArabah—of “the largest evidence of copper production during this period [Early Bronze Age] in the ancient Near East” (Levy et al. 2002: 427) that stands to impact significantly our understanding and interpretation of the culture and society of the Early Bronze Age peoples, in particular the mechanisms operative in the rise and collapse of urbanism. This impact continues to the end of the period— the EB IV; 1 it has now been confirmed that the mines were still in operation then (Adams 2000; Levy et al. 2002).

1. Elsewhere, the term Intermediate Early Bronze–Middle Bronze is used. In older terminology, the period was often referred to as Middle Bronze I or Early Bronze IV/Middle Bronze I. For the beginning of the second millennium bce, the term Middle Bronze I (MB I) is used here (MB IIA in the older terminology).

120

Suzanne Richard 5; Gerstenblith 1980: 65–66; Richard 1980: 8; Mazar 1990: 188), seems to reflect more material aspects of continuity than formerly believed: for example, in ceramics, tomb types, architecture, and metals (Amiran 1960: 205; Prag 1974: 101–3; 1984: 68; 1985: 87; Tubb 1982: 1; 1983; Falconer and Magness-Gardiner 1984: 57–58; Philip 1989: 199, 207). This current view is a trend that I support in this essay, which will highlight transitions in metallurgy. Prior to considering the social implications of the latter, we shall discuss the archaeological context of the spearhead, its description, and the metallurgical analysis, as well as review the still-critical topics of chronology and typology with regard to bronze metallurgy and weapon type.

The Archaeological Context

Fig. 1. Map of Jordan, showing the location of Khirbet Iskander.

The archaeometallurgical analysis of the Iskander spearhead provides us with valuable data that further the discussion concerning the role of metalworking and its socioeconomic implications in reconstructions of the EB IV. This was a posturban period once thought to have been an insignificant interlude, now more correctly viewed as an important period of transition from EB III, reflecting the way in which an urban society adapts or readapts to lesshighly centralized, more regionalized political and economic institutions (Richard 1987; 2003). Similarly, the transition from EB IV to the MB, once thought to be a complete break (Kenyon 1966: 36; Dever 1970: 144; 1976:

The metal object was found in Phase B, the penultimate EB IV phase in Area B in the northwestern corner of the mound (fig. 2). In Square B5, close to the western perimeter of the site, the blade lay on a plaster surface (L. 5007) associated with a wall (L. 5005) that jutted out from the southwestern corner of the square. At the northern end of the square, a nearly complete EB IV holemouth storejar was discovered lying on this surface. Surface 5007 ran under and was apparently the first-use surface with Wall 5005. The stratigraphic profile in Area B consistently demonstrates that the latest EB IV settlement (Phase A) seals the destruction debris of Phase B. That debris consisted of a 30–50-centimeter-thick matrix of plaster and roofing material, as well as rock, and preserved a wealth of whole and restorable vessels associated with a plaster floor traced in most of the structures excavated thus far (see Richard and Boraas 1988; Richard 1990). Moreover, the Phase B settlement seals the destruction level of Phase C, mostly collapsed mud brick and detritus. The restorable vessels found in Square B5, mostly domestic types, complement the previously discussed corpus of Phase B vessels from an apparent public storeroom, where large numbers of pithoi and unusually fine vessels were found (Richard 2000).

The Socketed Spearhead Figures 3:1 and 4–5 illustrate the socketed spearhead from several different perspectives. The blade is extremely small and broad, 38 mm in length and 17 mm in width. There is no midrib; rather, the profile of the blade shows a thickness of 4 mm at the center, with the blade thinning out gradually toward the edges, the section profile being more flattened than triangular. The overall appearance is of a broad, leaf-shaped spearhead of greatest width at the center, thus giving the blade a “shouldered” or biconical appearance. The blade edges appear more curved than

Early Bronze Age IV Transitions: An Archaeometallurgical Study

121

Fig. 2. Plan of the Area B, Phase B “public complex” at upper right; enlarged area shows location of the socketed spearhead in Square B5.

straight or sloping, and the blade itself narrows toward a somewhat angled, but extremely pointed tip. The remains of a socket (ca. 7 mm in length) have an outer diameter of 5 mm and an inner diameter ca. 2–3 mm. The socket, clearly folded in with sloping shoulders near the blade, is not fully closed longitudinally in the section that is preserved. The weapon is a very lightweight, delicate-looking metal artifact. What is unusual about this spearhead, besides its small size, is its socket, which, in profile, gives an asymmetrical

appearance to the weapon. This asymmetry provides a clue to the manufacturing technique used by the smith: hammering and rolling rather than casting. There are two types of technique associated with the manufacturing of socketed spearheads in the late third–early second millennia bce (Philip 1989: 88). The more typical form is the socketed spearhead cast in a bivalve mold, such as the MB I example from ºAin es-Samiyeh (Dever 1975: fig. 1:2); the less-well-attested type, exemplified by the Iskander specimen, is beaten from sheet metal, with the socket

122

Suzanne Richard

Fig. 3. Socketed spearheads. 1. Khirbet Iskander, Area B, Square B5 (A587); 2. Megiddo T. 1101B Lower, redrawn after Guy 1938: pl. 86:3; 3. ºAin es-Samiyeh, redrawn after Dever 1975: fig. 1:3; 4. Hama redrawn after Fugmann 1958: pl. 10 (6A358); 5. tanged (poker-butted) spearhead or javelin from Lachish, redrawn after Tufnell 1958: 75, pl. 22.1. Scale: 1:2, except for no. 1 at 1:1. Artifacts redrawn by Tina Curtis.

Early Bronze Age IV Transitions: An Archaeometallurgical Study walls rolled rather than cast. Interestingly, an example of the latter, “rolled” type was also found in the same ºAin esSamiyeh context noted above (Dever 1975: fig. 13) (fig. 3:3). In light of our spearhead, and given the fact that the corpus of EB IV and MB I metals and pottery from ºAin es-Samiyeh published by Dever (1975; see also 1972) was purchased from an antiquities dealer, it is conceivable that the “rolled” socketed spearhead from ºAin es-Samiyeh is EB IV rather than MB I in date (see further below). According to Philip’s extremely valuable compilation and analysis of metal weapons from the Early and Middle Bronze Ages, the Iskander blade would belong to his Type 10 (1989: 96–99), classified separately from other socketed spearheads on the basis of manufacturing technique, that is, hammered and rolled, not cast. When compared with his cluster analysis of all socketed spearheads (1989: fig. 61), the Iskander piece may be one of the smallest spearhead points known (if not the smallest). The socket size (5 mm outer diameter) is likewise one of the smallest sockets thus far published, one of the reasons why I originally identified the artifact as an arrowhead and, indeed, according to R. Miller, a hafting diameter below 8 mm is feasible for arrowheads (as mentioned in Philip 1989: 94). However, for several reasons, it is unlikely that this weapon is an arrowhead. When differentiating arrowheads from small spearheads or javelins, size and weight should be disregarded (Miller, McEwen, and Bergman 1986: 189), but more compelling is the fact that arrowheads are tanged projectiles mounted in a particular way, the tang stuck into a reed shaft or affixed and bound in some way to an arrowshaft. Although it is not impossible, it would be highly unusual to mount an arrowhead with a socket (see also Philip 1989: 94), but note a reference to a possible Aegean socketed arrowhead (Dickinson 1976: 98, fig. 5.47b:4). Thus, we consider the Iskander piece to be a small spearhead point or, more likely, a javelin head—that is, a thrown rather than a thrust weapon. Whether this “weapon” was actually used is questionable; its size strongly suggests that it is a miniature. Although most researchers dealing with the topic agree that the socketed spearhead superseded the tanged spearhead tradition at the beginning of the MB, the presence of earlier examples has generally been noted (Oren 1971: 115; Dever 1975: 23; Tubb 1985: 189; Philip 1989: 98). Indeed, on the basis of two EB IV (Philip’s Type 10) examples—one from Megiddo T. 1101B Lower (Guy 1938: pl. 86:3) (fig. 3:2) and the other an unpublished blade from Tell Brak—Philip correctly dates the beginning of the socketed-spearhead tradition to the late third millennium bce (1989: 97–98). On the evidence of the early EB IV context at Iskander and the generally acknowledged early EB IV date of Megiddo T. 1101B Lower, we may plausi-

123

Fig. 4. EB IV socketed spearhead from Khirbet Iskander, reverse view (photo by E. Skinner).

Fig. 5. EB IV socketed spearhead from Khirbet Iskander, showing folded-down shoulders of socket (photo by M. Rohaly).

bly suggest a date for the new weapon type as sometime shortly after 2300 bce, following the end of the EB III period. Thus, along with the introduction of tin-bronze alloying processes in the southern Levant in EB IV, we must now conclude that, although the MB I period represents the floruit of socketed spearheads, clearly the critical transition occurred at the end of the third millennium bce, that is, in the EB IV period. Thus, there are three Type 10 weapons known from the southern Levant in late-third- and/or early-secondmillennium bce contexts: the ºAin es-Samiyeh, Megiddo, and Iskander spearheads. As for the examples beyond our region, according to Philip, except for one from Hama

Suzanne Richard

124 Tomb G X (Fugmann 1958: pl. 10) (fig. 3:4), the remaining seven “hammered and rolled” socketed spearheads come from Depot Beta at Byblos (Dunand 1954: 187–88, pl. 58; see Philip 1989: 372–73 for descriptive details on these weapons, all of which he dates to MB I). We must, however, add to this list a number of miniature socketed spearheads (both cast and rolled, apparently) from Ugarit–Ras Shamra. These probable votives—along with miniature fenestrated axeheads and daggers—were found on the acropolis (Schaeffer 1949: fig. 25) and span the Ugarit Moyen 1 (late EB IV) and 2 periods. They are probably hoards or deliberate deposits (Philip 1988: 196). Type 10 spearheads are remarkably similar, all showing a preference for a broad, flat (or, rarely, very low Vshaped), clearly beaten blade, folded-down shoulders, and a usually incompletely closed socket. Yet it is clear from fig. 3 that each shows the peculiarities of probably local production. The Megiddo T. 1101B Lower spearhead (fig. 3:2) has a tapered blade ending in shoulders near the socket; somewhat differently, the Iskander, ºAin esSamiyeh, and Hama examples (fig. 3:1, 3–4, respectively) show the lower blade merging into a socket of approximately the same width. The “biconical” shape of the Iskander blade further distinguishes it from the others. The miniature votives from Ras Shamra are remarkably close parallels (Schaeffer 1949: fig. 25:f, 1–6). As for parallels from further afield, note the examples from Anatolia (Özgüç 1950: 200, 220, Abb. 369, 373) and Chagar Bazar (Mallowan 1937: 118, fig. 13:15), all of MB I date, and, as mentioned above, from Tell Brak, the last being especially significant, since it is of late-thirdmillennium bce date. Type 10 weapons do reappear briefly in the local LBA and later—for example, the Lachish folded spearhead from Tomb 555 (Tufnell 1958: pl. 23:12, photo pl. 54:39); see also Tell el-Farºah (S) (Petrie 1930: pls. 48:1, 55:293) and Tell Beit Mirsim (Albright 1941– 43: pl. 62:10).

Metallurgical Analysis Elemental analysis of the Iskander spearhead was conducted using the SEM-EDS system. 2 Scanned for analysis was a 100 microgram by 100 microgram area in the center of the fragment, which came from the socket, using 25 kw to penetrate the surface, in effect, “drilling” into the mantel of the sample in order to detect its elemental composition. A subsequent scan confirmed the readings of the first scan. The EDS system is considered a reliable method for 2. The analysis was carried out by the Department of Materials Science and Engineering, University of Arizona, using a Scanning Electron Microscope and Energy Dispersive Spectroscopy (SEM-EDS).

detecting different alloying elements in metal, although it is semiquantitative; increasingly, the preferred (quantitative) method is atomic absorption spectroscopy. Nonetheless, recent examination of a hoard of artifacts from Pella using both quantitative analysis of metallographic samples and semiquantitative analysis of drilled samples showed remarkable correspondence (Philip, Clogg, and Dungworth 2003: 73–81). There were four trace elements detected in the Iskander spearhead: copper (Cu), tin (SN), lead (PB), and silicon (Si) (table 1). The most significant reading is the high tin content (11.5%)—12% tin is near the upper limit of a desirable alloy, above which the metal is brittle and more difficult to work (Stech, Muhly, and Maddin 1985: 77). The tin-copper alloy of the Khirbet Iskander spearhead (fig. 4) makes it a standard bronze more typical of Middle Bronze weapons (Philip 1989: table 1). The presence of lead in the Iskander piece at a level of 2.27% is likewise of considerable interest. The earliest evidence for leaded bronzes comes from Mesopotamia, at Susa and Ur, dating to the second half of the third millennium (Malfoy and Menu 1987: 362), but the odd example of copper-arsenic-lead alloys from the fourth millennium has come to light (Moorey 1985: 122). Recent research has shown that, as early as MB I, a peculiar copper-tinlead alloying tendency also appeared in the southern Levant (Philip 1991; 1995a: 527–29; Rosenfeld, Ilani, and Dvoracheck 1997; Shalev 2000: 279; Philip, Clogg, and Dungworth 2003: 88). A comparative study of MB I metalwork from Jericho and Tell el-Dabºa shows an interesting correlation of tin-bronzes associated with lead alloys ranging from 1 to 4%, whereas arsenic-copper alloys contained lead levels of 0.5% or less (Philip 1995a: 527). Any lead in concentration above 2% does not add appreciable advantages to the final product (Craddock and GiumliaMare 1988: 319). It is therefore not inconceivable that the Iskander spearhead could be a true tin-copper-lead alloy. There is even the possibility that a tin-copper-lead axehead from Tell es-Shuna, with 6.50% tin and 1.70% lead, could be the earliest such alloy (Philip, Clogg, and Dungworth 2003: 88), although its provenience and date (EB III) are questionable. The late-third-millennium Iskander spearhead could be one of the earliest leaded bronzes in the southern Levant (fig. 5). As such, it would appreciably close the chronological gap with Mesopotamia. In this regard, it is interesting that as early as 1945, Schaeffer suggested that not only was there evidence at Ras Shamra for late-third-millennium bce (that is, pre-MB) tin-copper alloys, but also for tin-lead-copper alloys. This was questioned by Oren, however, given the assumption at the time that bronze metallurgy began with the MB (1971: 130–31).

Early Bronze Age IV Transitions: An Archaeometallurgical Study

125

Table 1. Results of EDS analysis of the Khirbet Iskander socketed spearhead, showing a high tin additive (11.5%), as well as lead (2.27%) in this alloy Element Si-K PB-M Sn-L Cu-K

k-ratio (calc.) 0.0024 0.0143 0.0951 0.8525

ZAF

Atom %

3.424 1.591 1.210 1.002

1.95 0.74 6.54 90.77

It should be noted that low levels of lead can be associated with particular ore bodies (Maddin and StechWheeler 1976: 171) or can result from the copper-smelting process (Segal, Halicz, and Cohen 1999: 180). One of the daggers at ºEnan, for example, contained 8.66% tin and 2.36% lead, as well as arsenic and iron, while three other objects contained from 1.35 to 1.70% lead. The conclusion drawn was that all the artifacts came from the same ores (probably in Anatolia) containing arsenic, lead, and iron, although it is clear that tin was added to the copper during the manufacturing process to produce several of the weapons (Stech, Muhly, and Maddin 1985: 77–79). The question of whether the quantity of lead found in EB IV metals represents deliberate alloying practices or reflects ore sources must, of course, remain open. Nevertheless, as evidence of innovation in technology in the EB IV mounts, it is intriguing to view assumed MB I practices once again as seemingly originating at the end of the third millennium bce.

Bronze Metallurgy in the EB IV Period The quantity of tin needed for it to be considered an alloying agent varies, as reported in the literature, from 5– 10% (Mazar 1990: 185) to 2–3% (Philip 1991: 101) to 1% (Maddin, Muhly, and Stech 2003: 516) to 0.5% (Moorey 1999: 251–52). For example, a dagger from Bab edhDhraº contained 0.30% tin, a level considered higher than that probably occurring in the ore and lower than the amount needed to be considered an alloy, that is, 1% (Maddin, Muhly, and Stech 2003: 516). On the basis of that analysis, it appears that previously analyzed EB IV items, for example, the Megiddo T. 1101B Lower Type 10 spearhead (Guy 1938: 160–61) and many others as well, may actually be tin-bronze alloys. Regardless, as table 1 clearly shows, the Iskander spearhead must be added to a growing corpus of EB IV tin-bronze alloyed artifacts. Previously, this corpus contained one weapon type, the dagger, found virtually only in tombs. Of the 12 known

Element Wt. % 0.81 2.27 11.50 85.42 100.00%

Wt. % Err. (1–Sigma) +/- 0.09 +/- 0.29 +/- 0.23 +/- 0.79

tin-bronze daggers, one example has been found at each of the following sites: Jericho T. G. 83a (Moorey and Schweizer 1972: 193), Bab edh-Dhraº, the only example from a settlement (Maddin, Muhly, and Stech 2003: 514), Menahemiya Tomb 1 (Bahat 1976: 32), Maªabarot (Dar 1977: 44), and ºAin es-Samiyeh (Branigan, McKerrell, and Tylecote 1976: 17); 2 come from Motza (Bahat 1975); and 5 of the 6 tested daggers from ºEnan were found to be tinbronzes (Stech, Muhly, and Maddin 1985). One can expand the corpus of tin-bronze artifacts from the period if we consider nonweapon types, for example, several items from Motza (Bahat 1975) as well as from Jericho (Khalil 1983). As others have already noted, it is therefore misleading to equate the onset of tin-bronze metallurgy with MB I, as has often been assumed in the past (Oren 1971: 130; Dever 1975: 30; Gerstenblith 1980: 66; Mazar 1990: 188), and it is surely a risky proposition now to assume the date of a metal piece based on tin content, especially when deciding between assigning an EB IV or MB I date to an object from a mixed tomb or questionable deposit (Oren 1971: 127–31; Dever 1975: 30). Although the evidence presented above for pre-Middle Bronze tin-copper-lead alloying practices is still somewhat slim, with increased metallurgical analyses, it would not be surprising to find more examples in the EB IV period in the future. Archaeometallurgical studies today increasingly stress the importance of lead isotope analysis as the best means to ascertain the provenience of the ore source used for a particular artifact. This analysis is critical in reconstructing ancient trade networks, cross-cultural influences, procurement processes, modes and place of production, and so on (Pernicka 1999). Lead isotope analysis is lacking for the Iskander spearhead, unfortunately, and we can therefore only speculate about its ore source. Based on the low level of impurities and lack of detectable arsenic in the blade, it is highly unlikely that the copper derives from the high-arsenic ores of Anatolia or elsewhere (de Jesus 1980: 233, 259). More plausibly, the provenience of the metal ores is the area of the Faynan or Timna (Hauptmann and

126 Weisgerber 1987; 1992; Rothenberg 1990; 1999), both in the Wadi ºArabah rift valley. Both areas, however, share a very similar geological profile; that is, the DolomiteLimestone-Shale (DLS) layer at Faynan is similar in composition to the Timna Late Cambrian ores. Thus, even with the aid of Lead Isotope Analysis, it is not possible to be definitive about provenience (Hauptmann, Begemann, and Schmitt-Strecker 1999; Segal, Halicz, and Cohen 1999; Levy et al. 2002). Several factors suggest that the Faynan is the more likely source. In addition to its obvious proximity to the south-central plateau site of Khirbet Iskander, the fact is that its mines were clearly operating most fully between the EB II and EB IV periods (Adams 2000; Levy et al. 2002). This evidence is generally lacking for the Timna copper mines, with the exception of one EB IV site found in the vicinity of Timna (Rothenberg and Shaw 1990). Moreover, there is growing evidence documenting connections in EB IV between south-central Transjordanian settlement sites, the copper-producing areas of the Faynan, probable production, manufacturing, and trading outposts in the Negev, and trade with Egypt (Goren 1996; Haiman 1996; Segal, Halicz, and Cohen 1999; Adams 2000; Levy et al. 2002). As for the tin, it must derive ultimately from wellknown northern sources, either in Anatolia, for example, the Kestel/Göltepe mines (Earl and Özbal 1996; Yener 2000), or in Afghanistan (see generally Muhly 1973; 1985; Stech and Pigott 1986). Whether the procurement of such alloying materials was through direct trade, which seems unlikely, or through the intermediary of Syrian traders cannot be determined (see Pinnock 1985 on Ebla trade generally; Philip 1989: 202). Although previously assumed to be imports, the tin-bronze weapons in EB IV more than likely attest to a local industry. Support for this inference derives from the mounting evidence for raw materials, ingots, molds, and workshops in the region, for example, in the Faynan (Adams 1997; 2000; Levy et al. 2002) and the Negev (Cohen and Dever 1979; Haiman 1996). The sum of this evidence combined makes it highly unlikely that tin-bronze alloying and manufacturing of EB IV artifacts occurred extraregionally. Rather more plausible is the view that posits the existence of trade routes between our region and the sources of tin to the north. Buttressing this view is the established documentation for the continued widespread importation of arsenic copper ores and/or arsenic as an alloying agent, indicative of trade with northern sources in the EB IV period (Stech, Muhly, and Maddin 1985: 81; Philip 1991: 100–101; Hauptmann, Begemann, and Schmitt-Strecker 1999: 9; Maddin, Muhly, and Stech 2003: 521; Philip, Clogg, and Dungworth 2003: 88–89). The fact is that, de-

Suzanne Richard spite the onset of tin-bronze metallurgy in the EB IV and the evidence of high-arsenic alloys, the traditional unalloyed copper practices of the Early Bronze Age persist overwhelmingly. The inescapable conclusions to draw from all of the above are that, in the EB IV, metalworking and production were far more complex and specialized than hitherto believed. Clearly, multiple trade networks were in operation, and—despite the absence of centralized sociopolitical authority—some organizational hierarchy must have been driving the metals industry, which appears to be the primary basis for the economy.

Typology/Chronology of Spearheads On the basis of the early (EB IV) date for the Type 10 socketed spearhead, it is reasonable to suggest that it was the prototype of the MB I cast or partly-cast socketed spearhead. Likewise, in the absence of better information about the Tell Brak example (see above), one could surmise with some degree of confidence that the two, or possibly three, early EB IV Type 10 examples from the southern Levant suggest the place of origin of this innovative weapon type. As unorthodox as that may sound, and although the general consensus among scholars is that socketed (bronze) spearheads define the beginning of the Middle Bronze urban era (Oren 1971: 130–31; Dever 1975: 23; Gerstenblith 1980: 66, 73, n. 14; Philip 1989: 86–87), why not consider the possibility that (a) there is innovation in the EB IV period, and (b) the Type 10 socketed spearhead is one of these innovations? After all, like the late Chalcolithic period, the EB IV culture is astonishing for the wealth of metal artifacts known from the archaeological record, far more than from the EB I–III periods combined. Unfortunately, certain presuppositions about the EB IV and the MB I periods still persist that the industry of the southern Levant was derivative and probably lagging behind Syria chronologically. The evidence of the early Type 10 spearheads from Iskander and Megiddo leads us to question the view that similarities between metal types of the Syrian MB I and the local EB IV suggest a secondmillennium date for the latter (Philip 1989: 100–101, 201). The bulk of the radiometric dates for the period strongly support a late-third-millennium date for EB IV (Adams 2000; Levy et al. 2002; Weinstein 2003). Although the suggestion that the place of origin of the Type 10 socketed spearhead is the southern Levant seems reasonable (they are the earliest examples, after all), the proposal that the cast socketed spearhead of MB I is the typological descendant of the EB IV Type 10 spearhead, although an attractive proposition, seems tenuous. It is, in fact, highly probable that both types of socketed spear-

Early Bronze Age IV Transitions: An Archaeometallurgical Study heads overlap at the end of the third millennium. An examination of the late-third-millennium contexts of cast socketed spearheads—albeit usually termed questionable—bears this out. A socketed spearhead was found in Dolmen 10 in the Golan in association with clear EB IV pottery and no later materials (Epstein 1985: 43, fig. 2:9–15). Furthermore, in Dolmen 13 (Epstein 1985: fig. 3), which also yielded several MB I pottery vessels, another socketed spearhead came to light in association with an EB IV broad-bladed spearhead (arrowhead?) and typical awl. Taken at face value, except for the few MB I ceramic pieces, all the pottery and the socketed spearhead and other metals were found “pushed aside to make room for later burials” (Epstein 1985: 33, 44). There are several other questionable contexts that may indeed be of pre-Middle Bronze date. In my dissertation, I had noted the possibility of dating several socketed spearheads from Ugarit–Ras Shamra to the late third millennium (Richard 1978: 236–38). At that site, four socketed spearheads (pointes de lance ou javelot à douelle) dated earlier than Ugarit Moyen I were discovered in two separate deposits. Two of the weapons lay at the bottom of the poche au Bronze (Courtois 1962b: 342–45, fig. 10; Schaeffer et al. 1962: 232, fig. 28:E, J [fig. 28:I is a third example, albeit a fragment]) and the others in another silo (Courtois 1962a: 457–59, figs. 48–49). The excavators dated these deposits to the Bronze Ancien 3 period. The weapons have been variously dated to EB III (Hennessy 1967: 81), the late third millennium (Tubb 1985: 189), and MB I (Oren 1971: 115; Philip 1989: 97–98). Furthermore, there is a late-third-millennium Ur III example from Tomb 1850 (Nissen 1966: 191) and another Mesopotamian example now reported from the late third millennium bce (Philip 1995b: 135). Another questionable context is Megiddo Stratum XVIII (Loud 1948: pl. 173:1), in which a socketed spearhead was found that was dated by Hennessy to EB III on the basis of the above-mentioned Ras Shamra spearheads (1967: 81). I had originally dismissed Hennessy’s suggestion and dated this weapon instead to the MB I period, agreeing with Oren that it probably belonged to a grave (4052) that had been dug into an earlier horizon (1971: 127). It is worthwhile reconsidering this assessment (presupposition?) in light of new information on EB IV. Moreover, in another tomb at Megiddo, T. 84C, a socketed spearhead was found alongside a classic fenestrated-“eye” EB IV axehead (Guy 1938: pl. 163:8–9). The latter has been dated to the MB I, with the explanation that there is probably an overlap between “eye” and “duckbill” (MB I) axes (Oren 1971: 137, n. 141; Philip 1989: 52, 167). The fact is that presuppositions about the MB I origin of the

127

socketed spearhead, as well as about the onset of tinbronze alloys, may have influenced the dating of the Megiddo and other socketed spearheads. The purpose of this discussion has been to raise awareness of an important issue for further scholarly debate. Regardless of the outcome of this debate, the late-thirdmillennium bce appearance of the Type 10 socketed spearhead in the southern Levant seems unequivocal. It follows, therefore, that the Type 10 socketed spearheads in Depot Beta at Byblos, if dated correctly to the MB I period, are later than the Khirbet Iskander and Megiddo T. 1101B Lower socketed spearheads. However, a more attractive alternative may be to date the Byblos Type 10 socketed spearheads (and the Ras Shamra deposit examples cited above) to the EB IV period, on the basis of the (earlier) Megiddo and Iskander examples. The deposits at Byblos (and Ras Shamra) clearly span the late third/early second millennia, that is, the EB IV and MB I periods (Kenyon 1966: 208; Tufnell and Ward 1966: 227; Kenyon, Posener, and Bottéro 1971: 587–92; Dever 1975: 23; Tubb 1985: 49). Redating the Byblos spearheads, although only a suggestion, would allow for the more or less contemporaneous appearance of the (relatively unique) Type 10 socketed spearhead throughout Syria, Mesopotamia, and the southern Levant. It would further bolster the idea that the floruit of the cast socketed spearhead was the early second millennium bce, that is, the MB I period. This correlation strengthens the view that strong links existed between the repertoires of coastal Syria and the southern Levant in the late third millennium and early second millennium bce (Kenyon, Posener, and Bottéro 1971: 592–93). Finally, as I concluded in 1978, the impetus to develop a socketed spearhead remains enigmatic. Probably the best explanation is to see its appearance as an innovation of the late third millennium bce, by all accounts a period of widespread trade and the diffusion of tin-bronze technology (Stech and Pigott 1986; Philip 1989; Moorey 1999). Metalsmiths in all parts of the Near East were experimenting with the “new” technology, including some of the dispersed communities in the southern Levant in the EB IV, at a time when metals and the metal trade apparently undergirded the regional economy. The mounting evidence of the use of tin-bronze as well as arsenic alloys in the EB IV attests to this experimentation. Additional evidence of experimentation is illustrated by the typological variants within the corpus of metal types. Examples include the tanged (poker-butted) spearhead/javelin (fig. 3:5); the fenestrated-“eye” axehead, the earliest appearance of which seems to be in the local EB IV period (Tubb 1982: 10; Philip 1989: 52); bar- or crescent-shaped ingots (Segal, Halicz, and Cohen 1999: 179; Adams 2000: 395); a variety

Suzanne Richard

128 of dagger types (Dever 1972: 106–7; Philip 1989: 102– 25); and, of course, the Type 10 socketed spearhead. If the Type 10 spearhead is a southern Levantine innovation, one could suggest local inspiration from the “poker-butted” javelin (fig. 3:5) and the broad, flat, leafshaped spearhead or projectile, a point I made about the Megiddo T. 1101B Lower example (Richard 1978: 238). The Iskander blade compares well with the contemporary typical javelin point (fig. 3:5). With regard to the socket, the contemporary fenestrated-“eye” (socketed) axehead attests to the experimentation with a hollow rather than a solid shaft. This proposed derivation resonates with Philip’s suggestion that the “poker-butted” spearhead (his Type 6) reflects a “hammered” technology similar to the Type 10 socketed spearhead (1989: 100). However, I disagree that the “southern” Type 6 reflects a “lag” in technology behind the “northern” Type 10 tradition, thus adding further support to a dichotomy between the two areas (Philip 1989: 86–76, 166–67, 201). A discussion of northern and southern regional traditions in the EB IV goes beyond the parameters of this essay; however, I would simply note that the typical “band-combed” ware found at, for example, ºAin es-Samiyeh (a central site) and Lachish (a southern site) is comparable with the Phase B assemblage found in a destruction layer at Khirbet Iskander, which likewise shares many parallels with “northern” assemblages as well (Richard 2000). The discovery of the Iskander “northern” spearhead as well as the tinbronze dagger at Bab edh-Dhraº, seem to bridge the gap between the north and the south.

Summary and Conclusions By way of a summary of the above discussion, the stratified Khirbet Iskander Type 10 “rolled and hammered” socketed spearhead comes from an early EB IV context and is an excellent parallel for and affirmation of the other early EB IV example found in Megiddo T. 1101B Lower. These two and perhaps the ºAin es-Samiyeh spearhead as well may be the earliest socketed spearheads attested in the Near East or, given the scattered references to late-thirdmillennium examples in Mesopotamia and the possible EB IV examples in Syria, may at least attest to a contemporaneous development of the innovative type in various regions. The Khirbet Iskander example brings to a total of 13 the corpus of EB IV tin-bronze alloyed weapons found in the southern Levant. It is the first such non-dagger weapon to appear in the period and, notably, may attest to an early experimentation with tin-copper-lead alloys. Of the 13 weapons, all are from tombs except for 2: the Khirbet Iskander spearhead and a dagger from Bab edh-Dhraº.

spread is 12 points short

On the basis of the undeniable early EB IV appearance of the two Type 10 socketed spearheads, I have speculated on the innovativeness of the local industry and raised cautionary flags concerning certain presuppositions connected with the rise of metallurgy in MB I, at the beginning of the second millennium bce. These presuppositions may have led to the misdating of certain artifacts and contexts. At the very least, the Iskander and Megiddo socketed spearheads provide more evidence to support the view that a close connection existed with coastal Syria (that is, the Byblos and Ras Shamra deposits), and, further, that no appreciable lag behind Mesopotamia is apparent. Against the backdrop of the unusually large corpus of metal artifacts from the EB IV period as well as the mounting evidence for copper mining, production, distribution, and trade (from the Faynan and the Negev), I have concluded that the Iskander spearhead is a product of local manufacture rather than being an import. This view gains support from the parallel local production of arsenic-copper alloys in the EB IV. It is becoming clearer that trade routes with the north were the conduit for raw materials, such as arseniccoppers and arsenic as an alloying agent, as well as tin. That mining and manufacture can be the impetus for local innovation, not to mention a boon to the economy, has been noted elsewhere (see, in particular, the influence of the Kestel mines on the local vicinity [Yener 2000: 12– 15]). Finally, on the basis of a number of cast socketed spearheads in early (albeit questionable) contexts, I speculated that the two types developed contemporaneously in the late third millennium bce, possibly suggesting a period of early experimentation. Not yet considered but germane to our understanding of the people and the town of Khirbet Iskander and the EB IV period are the social implications that we might infer from the presence of the Iskander metal artifact. First, that the socketed spearhead—early, rare, and of tin-bronze—is an item of prestige seems incontestable. Recent studies suggest that the specific use of tin-bronze for daggers no doubt reflects a special value attached to the object (Stech, Muhly, and Maddin 1985; Philip 1991). It appears likely that it is the silvery surface of tin-bronze and tin-leadbronze (mimicking prestigious silver and gold objects) that imparts prestige to these weapons (Philip 1991; Rehren, Hess, and Philip 1997). In fact, it has been argued that the deliberate selection of an arsenic-copper alloy in the EB IV period, virtually used only in the production of daggers, likewise illustrates the purposeful attempt to achieve a silvery appearance in prestige items (Philip 1991: 96, Philip, Clogg, and Dungworth 2003: 88–89). Information from the Ebla texts concerning the prestige of bronze daggers supports this view (Archi 1985; Waetzoldt

Early Bronze Age IV Transitions: An Archaeometallurgical Study 1990; see also Stech and Pigott 1986 for a general discussion of tin-bronze as a prestige alloy). The entire subject of daggers as prestige items (that is, of high, possibly warrior, status, as in MB I) needs further study, given the fact that in EB IV, they appear in pure copper, arsenic-copper alloy, and tin-bronze alloy, often in the same context—for example, at the Bab edh-Dhraº settlement, where an unalloyed (pure) copper dagger was found alongside a tin-bronze dagger (Maddin, Muhly, and Stech 2003: 519). Although our weapon is neither a dagger nor a burial gift and is not MB I in date, the arguments adduced for daggers are congenial and suggestive. It is possible that, were it not for the destruction of Phase B, the ultimate destination of the bronze spearhead might have been a tomb. All speculation aside, the discovery of the weapon in the Phase B settlement in the immediate vicinity of what we have termed a “public” structure adds further support to the view espoused for some time that elites were living at the site of Khirbet Iskander (Richard and Boraas 1988; Richard 1990). The bronze socketed spearhead could very well epitomize an outward or conspicuous display of wealth on the part of an elite individual living in the area. One possibility to consider is that the valued possession attests to the practice of gifting between elites at Iskander or with trading partners, given the site’s strategic location guarding a crossing of an important trade route. The above scenario has much to commend it. Items of prestige were valuable because they were rare, whether circulation was kept down by gifting, by the use of hoards, or by burial deposit. Given its context (a destruction layer), the Iskander spearhead probably was not a “deliberate deposit,” that is, a hoard, whether utilitarian or nonutilitarian (Knapp, Muhly, and Muhly 1988: 35). Hoards do occur in the late Chalcolithic period (Moorey 1988), the Early and Middle Bronze Ages (Philip 1988), and later (Kletter 2003), although by the Iron Age, most are composed of precious metals, that is, silver or gold, rather than copper or bronze. As for burial deposit, I have speculated above that the bronze artifact could have been intended for a tomb. This scenario resonates well with the evidence from Megiddo T. 1101B

129

Lower and ºAin es-Samiyeh, the latter cemetery already well known for a very prestigious item, a decorated silver cup (Yadin 1971; Yeivin 1971). Yet the findspot in Square B5 militates against this view as well. It is also unlikely that the Iskander spearhead is indicative of metalworking at the site, for, although an apparent miniature crucible was found in the neighboring Phase B “public building,” evidence thereof remains elusive. Having considered (and rejected) all of the above scenarios, I suggest that the best explanation relates to the “public building” itself. If we view the spearhead in the context of the neighboring public building, where unusual stores of foodstuffs, high-quality vessels, offering deposits, possible favissae, and miniature vessels (including the miniature crucible), as well as a ceramic bull’s head were found not far from the Square B5 floor, we could surmise that the very costly and prestigious item was intended as a cult offering. Its very size and delicacy suggest that it should be considered a miniature, a nonutilitarian weapon probably meant to be an offering. In this regard, the miniature votives found in a late-third/early-second-millennium context at Ras Shamra are especially instructive (Schaeffer 1949: fig. 25). At this stage of our research, prior to definitively analyzing the entire Phase B settlement in context, we are wont to say that, whether treasured gift, actual weapon, or votive, the spearhead is a clear indication of a prestige item bestowing status in the community on the owner. Presuming it to be an elite, cultic item, one can probably connect the spearhead with the range of unusual, probably cultic material found in the Phase B “public building.” This one, small artifact brings Khirbet Iskander and the Transjordanian plateau into the larger orbit of both the northern sites (with the parallel at Megiddo and connections with Hama, Byblos, and Ras Shamra) and the southern areas (probably the trade and distribution of metals in connection with the Faynan and Negev) but especially the greater Near East in terms of the concomitant appearance of parallels over a vast area. And last but not least, it lends more credence to a smooth transition to the Middle Bronze age.

References Adams, R. B. 1997 On Early Copper Metallurgy in the Levant: A Response to Claims of Neolithic Metallurgy. Pp. 651–56 in The Prehistory of Jordan II: Perspectives from 1997, ed. H. G. K Gebel, Z. Kafafi, and G. O. Rollefson. Studies in Early Near Eastern Production, Subsistence, and Environment 4. Berlin: Ex Oriente.

2000 The Early Bronze Age III–IV Transition in Southern Jordan: Evidence from Khirbet Hamra Ifdan. Pp. 379–401 in Ceramics and Change in the Early Bronze Age in the Southern Levant, ed. R. B. Adams and Y. Goren. Sheffield: Sheffield Academic Press.

130 Albright, W. F. 1941–43 The Excavations at Tell Beit Mirsim III: The Iron Age. Annual of the American Schools of Oriental Research 21–22. New Haven: American Schools of Oriental Research. Amiran, R. 1960 The Pottery of the Middle Bronze I in Palestine. Israel Exploration Journal 10: 204–25. Archi, A. 1985 Circulation d’objects en métal précieux de poids standardisé à Ébla. Pp. 25–34 in Miscellanea Babylonica: Mélanges offerts à Maurice Birot, ed. J.-M. Durand and J. R. Kupper. Paris: Éditions Recherche sur les Civilisations. Bahat, D. 1975 A Middle Bronze I Tomb-Cave at Motza. Eretz-Israel 12 (Glueck volume): 18–23, 117* (Hebrew, with English summary). 1976 A Middle Bronze Age I Cemetery at Menahemiya. ºAtiqot (English series) 11: 27–33. Branigan, K.; McKerrell, H.; and Tylecote, F. R. 1976 An Examination of Some Palestinian Bronzes. Journal of the Historical Mining Society 10/1: 15–23. Cohen, R., and Dever, W. G. 1979 Preliminary Report of the Second Season of the “Central Negev Highlands Project.” Bulletin of the American Schools of Oriental Research 236: 41–60. Courtois, J. C. 1962a Contribution à l’étude des civilizations du Bronze Ancien à Ras Shamra–Ugarit. Pp. 415–75 in Ugaritica IV, ed. C. F. A. Schaeffer. Paris: Guethner. 1962b Contribution à l’étude des niveaux II et III à Ras Shamra. Pp. 329–414 in Ugaritica IV, ed. C. F. A. Schaeffer. Paris: Geuthner. Craddock, P. T., and Giumlia-Mare, A. R. 1988 Problems and Possibilities for Provenancing Bronzes by Chemical Composition, with Special Reference to Western Asia and the Mediterranean in the Early Iron Age. Pp. 317–28 in BronzeWorking Centres of Western Asia c. 1000–539 bc, ed. J. E. Curtis. London: Kegan Paul. Dar, S. 1977 Ancient Settlements of ºEmeq Hefer: The Sites of Tell Nurit, Nahal Alexander and the Rockcut Tombs of Maªabarot. Kibbutz Maªabarot. Dever, W. G. 1970 The Middle Bronze I Period in Syria and Palestine. Pp. 132–63 in Near Eastern Archaeology in the Twentieth Century: Essays in Honor of Nelson Glueck, ed. J. A. Sanders. Garden City: Doubleday. 1972 Middle Bronze Age I Cemeteries at Mirzbaneh and ºAinSamiya. Israel Exploration Journal 22: 95–112. 1973 The EB IV–MB I Horizon in Transjordan and Southern Palestine. Bulletin of the American Schools of Oriental Research 210: 37–63. 1975 MB IIA Cemeteries at ºAin es-Samiyeh and Sinjil. Bulletin of the American Schools of Oriental Research 217: 23–36. 1976 The Beginning of the Middle Bronze Age. Pp. 3–38 in Magnalia Dei: The Mighty Acts of God. Essays on the Bible and Archaeology in Memory of G. Ernest Wright, ed. F. M. Cross, W. E. Lemke, and P. D. Miller. Garden City, NY: Doubleday. 1980 New Vistas on the EB IV (“MB I”) Horizon in Syria–Palestine. Bulletin of the American Schools of Oriental Research 237: 35– 64. 1989 The Collapse of the Urban Early Bronze Age in Palestine: Toward a Systemic Analysis. Pp. 225–46 in L’urbanisation de la Palestine à l’âge du Bronze ancien, ed. P. de Miroschedji. British Archaeological Reports International Series 527. Oxford: British Archaeological Reports.

Suzanne Richard 1995 Social Structure in the Early Bronze IV Period in Palestine. Pp. 281–96 in The Archaeology of Society in the Holy Land, ed. T. E. Levy. New York: Facts on File. 2003 The Rural Landscape of Palestine in the Early Bronze IV Period. Pp. 43–59 in The Rural Landscape of Ancient Israel, ed. A. M. Maeir, S. Dar, and Z. Safrai. British Archaeological Reports International Series 1121. Oxford: Archaeopress. Dever, W. G., and Tadmor, M. 1976 A Copper Hoard of the Middle Bronze I. Israel Exploration Journal 26: 163–69. Dickinson, O. 1976 The Aegean Bronze Age. Cambridge: Cambridge University Press. Dunand, M. 1954 Fouilles de Byblos 1933–1938. Vol. 2. Paris: Geuthner. Earl, B., and Özbal, H. 1996 Early Bronze Age Tin Processing at Kestel/Göltepe, Anatolia. Archaeometry 38/2: 289–303. Epstein, C. 1985 Dolmens Excavated in the Golan. ºAtiqot (English Series) 17: 20–58. Falconer, S., and Magness-Gardiner, B. 1984 Preliminary Report of the First Season of the Tell el-Hayyat Project. Bulletin of the American Schools of Oriental Research 255: 49–74. Fugmann, E. 1958 Hama: Fouilles et récherches, 1931–1938 II i: l’Architecture des périodes pré-hellenistiques. Copenhagen: Nationalmuseet. Gerstenblith, P. 1980 A Reassessment of the Beginning of the Middle Bronze Age in Syria–Palestine. Bulletin of the American Schools of Oriental Research 237: 65–84. Goren, Y. 1996 The Southern Levant in the Early Bronze Age IV: The Petrographic Perspective. Bulletin of the American Schools of Oriental Research 202: 1–32. Guy, P. L. O. 1938 Megiddo Tombs. Oriental Institute Publications 33. Chicago: University of Chicago Press. Haiman, M. 1996 Early Bronze Age IV Settlement Pattern of the Negev and Sinai Deserts: View from Small Marginal Temporary Sites. Bulletin of the American Schools of Oriental Research 310: 33–72. Hauptmann, A., and Weisgerber, G. 1987 Archaeometallurgical and Mining-Archaeological Investigations in the Area of Feinan, Wadi ºArabah (Jordan). Annual of the Department of Antiquities of Jordan 31: 419–37. 1992 Periods of Ore Exploitation and Metal Production in the Area of Feinan, Wadi ºArabah, Jordan. Pp. 61–66 in Studies in the History and Archaeology of Jordan IV, ed. G. Bisheh, M. Zaghloul, and I. Kehrberg. Amman: Department of Antiquities of Jordan. Hauptmann, A.; Begemann, F.; and Schmitt-Strecker, S. S. 1999 Copper Objects from Arad: Their Composition and Provenance. Bulletin of the American Schools of Oriental Research 314: 1–18. Hennessy, J. B. 1967 The Foreign Relations of Palestine during the Early Bronze Age. London: Colt Archaeological Institute. Jesus, P. S. de 1980 The Development of Prehistoric Mining and Metallurgy in Anatolia. British Archaeological Reports International Series 74. Oxford: British Archaeological Reports. Kenyon, K. 1966 Amorites and Canaanites. The Schweich Lectures of the British Academy. London: Oxford University Press.

Early Bronze Age IV Transitions: An Archaeometallurgical Study Kenyon, K.; Posener, G.; and Bottéro, J. 1971 Syria and Palestine c. 2160–1780 b.c. Pp. 532–94 in Early History of the Middle East. 3rd rev. ed. Vol. 1/2 of Cambridge Ancient History. Cambridge: Cambridge University Press. Khalil, L. 1983 Copper Metallurgy from Jericho. Pp. 777–80 in Excavations at Jericho 5: The Pottery Phases of the Tell and Other Finds, ed. K. Kenyon and T. A. Holland. London: British School of Archaeology in Jerusalem. Kletter, R. 2003 Iron Age Hoards of Precious Metals in Palestine: An “Underground Economy”? Levant 35: 139–52. Knapp, B.; Muhly, J. D.; and Muhly, P. M. 1988 To Hoard Is Human: Late Bronze Age Metal Deposits in Cyprus and the Aegean. Report of the Department of Antiquities, Cyprus 1988: 233–62. Levy, T. E., et al. 2002 Early Bronze Age Metallurgy: A Newly Discovered Copper Manufactory in Southern Jordan. Antiquity 76: 425–37. Loud, G. 1948 Megiddo II: Seasons of 1935–1939. Oriental Institute Publications 62. Chicago: University of Chicago Press. Maddin, R., and Stech-Wheeler, T. 1976 Metallurgical Study of Seven Bar Ingots. Israel Exploration Journal 26: 170–73. Maddin, R.; Muhly, J. D.; and Stech, T. 2003 Metallurgical Studies on Copper Artifacts from Bab edh-Dhraº. Pp. 513–21 in Bab edh-Dhraº: Excavations at the Town Site (1975–1981), by W. E. Rast and R. T. Schaub. Winona Lake, IN: Eisenbrauns. Malfoy, J.-M., and Menu, M. 1987 La métallurgie du cuivre à Suse aux IVe et IIIe millénaires: Analyses en laboratoire. Pp. 355–73 in Métallurgie Susienne I et II, ed. F. Tallon. Notes et Documents de musée de France 15. Paris: Éditions de la Reunion des musées nationaux. Mallowan, M. E. L. 1937 The Excavations at Tall Chagar Bazar and an Archaeological Survey of the Habur Region, Second Campaign. Iraq 4: 91–154. Mazar, A. 1990 Archaeology of the Land of the Bible (ca. 10000–586 b.c.e.). New York: Doubleday. Merkel, J. F., and Dever, W. G. 1989 The Metalworking Technology at the End of the Early Bronze Age in the Southern Levant. Institute for Archaeo-Metallurgical Studies Newsletter 14: 1–4. Miller, R.; McEwen, E.; and Bergman, C. 1986 Experimental Approaches to Ancient Near Eastern Archery. World Archaeology 18/2: 178–95. Moorey, P. R. S. 1985 Materials and Manufacture in Ancient Mesopotamia: The Evidence of Archaeology and Art. Metals and Metalwork, Glazed Material and Glass. British Archaeological Reports International Series 237. Oxford: British Archaeological Reports. 1988 The Chalcolithic Hoard from Nahal Mishmar, Israel, in Context. World Archaeology 20: 171–89. 1999 Ancient Mesopotamian Materials and Industries: The Archaeological Evidence. Winona Lake, IN: Eisenbrauns. Moorey, P. R. S., and Schweizer, F. 1972 Copper and Copper Alloys in Ancient Iraq, Syria and Palestine: Some New Analyses. Archaeometry 14: 177–98. Muhly, J. D. 1973 Copper and Tin. Transactions of the Connecticut Academy of Arts and Sciences 43. Hamden, CT: Archon. 1985 Sources of Tin and the Beginnings of Bronze Metallurgy. American Journal of Archaeology 89: 275–91.

131

Nissen, H. J. 1966 Zur Datierung des Königsfreidhofes von Ur: Unter besonderer Berücksichtigung der Stratigraphie der Privatgräber. Bonn: Habelt. Oren, E. D. 1971 A Middle Bronze Age I Warrior Tomb at Beth-Shan. Zeitschrift des Deutschen Palästina-Vereins 87: 109–39. Özgüç, N. 1950 Kültepe Kazisi Raporu 1948. Ankara: T.T.K. Pernicka, E. 1999 Trace Element Fingerprinting of Ancient Copper: A Guide to Technology or Provenance? Pp. 163–71 in Metals in Antiquity, ed. S. M. M. Young et al. British Archaeological Reports International Series 792. Oxford: British Archaeological Reports. Petrie, W. M. F. 1930 Beth Pelet I. London: British School of Archaeology in Egypt. Philip, G. 1988 Hoards of the Early and Middle Bronze Ages in the Levant. World Archaeology 20: 190–207. 1989 Metal Weapons of the Early and Middle Bronze Ages in Syria– Palestine. British Archaeological Reports International Series 526. Oxford: British Archaeological Reports. 1991 Tin, Arsenic, Lead: Alloying Practices in Syria–Palestine around 2000 b.c. Levant 23: 93–104. 1995a The Same but Different: A Comparison of Middle Bronze Age Metalwork from Jericho and Tell el-Dabºa. Pp. 523–30 in Studies in the History and Archaeology of Jordan V, ed. E. Amr, E. Zayadine, and M. Zaghloul. Amman: Department of Antiquities of Jordan. 1995b New Light on North Mesopotamia in the Earlier Second Millennium bc: Metalwork from the Hamrin. Iraq 57: 119–44. Philip, G.; Clogg, P.; and Dungworth, D. 2003 Copper Metallurgy in the Jordan Valley from the Third to the First Millennia bc: Chemical, Metallographic and Lead Isotope Analyses of Artefacts from Pella. Levant 35: 71–100. Pigott, V. C. 1996 Near Eastern Archaeometallurgy: Modern Research and Future Directions. Pp. 139–76 in The Study of the Ancient Near East in the Twenty-First Century: The William Foxwell Albright Centennial Conference, ed. J. S. Cooper and G. M. Schwartz. Winona Lake, IN: Eisenbrauns. Pinnock, F. 1985 About the Trade of Early Syrian Ebla. Pp. 85–92 in À propos d’un cinqantenaire: Mari, bilan et perspectives. Mari: Annales des recherches interdisciplinaires 4. Actes du colloque internationale du Centre National de la Recherche Scientifique 620. Paris: Éditions Recherches sur les civilisations. Prag, K. 1974 The Intermediate Early Bronze–Middle Bronze Age: An Interpretation of the Evidence from Transjordan, Syria and Lebanon. Levant 6: 69–116. 1984 Review Article: Continuity and Migration in the South Levant in the Late Third Millennium: A Review of T. L. Thompson’s and Some Other Views. Palestine Exploration Quarterly 116: 58–68. 1985 Ancient and Modern Pastoral Migration in the Levant. Levant 17: 81–88. Rehren, T.; Hess, K.; and Philip, G. 1997 Fourth Millennium bc Copper Metallurgy in Northern Jordan: The Evidence from Tell esh-Shuna. Pp. 625–40 in The Prehistory of Jordan II: Perspectives from 1997, ed. H. G. K. Gebel, Z. Kafafi, and G. O. Rollefson. Studies in Early Near Eastern Production, Subsistence, and Environment 4. Berlin: Ex Oriente.

132 Richard, S. 1978 The End of the Early Bronze Age in Palestine–Transjordan: A Study of the Post–EB III Cultural Complex. Unpublished Ph.D. dissertation, Johns Hopkins University. 1980 Towards a Consensus of Opinion on the End of the Early Bronze Age in Palestine–Transjordan. Bulletin of the American Schools of Oriental Research 237: 5–34. 1987 The Early Bronze Age in Palestine: The Rise and Collapse of Urbanism. Biblical Archaeologist 50: 22–43. 1990 The 1987 Expedition to Khirbet Iskander and Its Vicinity: Fourth Preliminary Report. Bulletin of the American Schools of Oriental Research Supplement 26: 35–56. 2000 Chronology versus Regionalism in the Early Bronze IV: An Assemblage of Whole and Restored Vessels from the Public Building at Khirbet Iskander. Pp. 399–417 in The Archaeology of Jordan and Beyond: Essays in Honor of James A. Sauer, ed. L. E. Stager, J. A. Greene, and M. D. Coogan. Studies in the Archaeology and History of the Levant 1. Winona Lake, IN: Eisenbrauns. 2002 2000 Season of Excavations at Khirbet Iskander, Jordan. Pp. 105–14 in Proceedings, ed. B. Fiore. Eastern Great Lakes and Midwest Biblical Societies 21. Buffalo: Canisius College. 2003 The Early Bronze Age in the Southern Levant. Pp. 286–302 in Near Eastern Archaeology: A Reader, ed. S. Richard. Winona Lake, IN: Eisenbrauns. Richard, S., and Boraas, R. S. 1988 The Early Bronze IV Fortified Site of Khirbet Iskander, Jordan: Third Preliminary Report, 1984 Season. Pp. 107–30 in Preliminary Reports of ASOR-Sponsored Excavations 1982–85. Bulletin of the American Schools of Oriental Research Supplement 25. Baltimore: Johns Hopkins University Press. Richard, S., and Long, J. C. 1995 Archaeological Expedition to Khirbet Iskander and Its Vicinity, 1994. Annual of the Department of Antiqities of Jordan 39: 81– 92. 1998 Khirbet Iskander. Archaeology in Jordan. American Journal of Archaeology 102: 587–88. 2000 Khirbet Iskander. ACOR Newsletter 12/1: 7–8. Richard, S.; Long, J.; and Libby, B. 2001 Khirbet Iskander. Archaeology in Jordan. American Journal of Archaeology 105: 440–41. Rosenfeld, A.; Ilani, S.; and Dvoracheck, M. 1997 Bronze Alloys from Canaan during the Middle Bronze Age. Journal of Archaeological Science 24: 857–64. Rothenberg, B. 1990 The Ancient Metallurgy of Copper. London: Institute of Archaeo-Metallurgical Studies. 1999 Archaeo-Metallurgical Researches in the Southern Arabah. Part I: Late Pottery Neolithic to Early Bronze IV. Palestine Exploration Quarterly 131: 68–89. Rothenberg, B., and Shaw, C. T. 1990 The Discovery of a Copper Mine and a Smelter from the End of the Early Bronze Age in the Timna Valley. Institute for ArchaeoMetallurgical Studies Newsletter 15–16: 1–8. Schaeffer, C. F. A. 1945 La contribution de la Syrie ancienne à l’invention du bronze. Journal of Egyptian Archaeology 31: 92–95.

Suzanne Richard 1949 Ugaritica II: Nouvelles études relatives aux découvertes de Ras Shamra. Mission de Ras Shamra 5. Paris: Geuthner. Schaeffer, C. F. A., et al. 1962 Les fondements préhistoriques d’Ugarit. Pp. 151–251 in Ugaritica IV, ed. C. F. A. Schaeffer. Bibliotheque Archéologique et Historique 74. Paris: Geuthner. Segal, I.; Halicz, L.; and Cohen, R. 1999 A Study of Ingots and Metallurgical Remains from ºEin Ziq and Beªer Resisim, Central Negev, Israel. Pp. 179–86 in Metals in Antiquity, ed. S. M. M. Young et al. British Archaeological Reports International Series 792. Oxford: British Archaeological Reports. Shalev, S. 2000 Metalurgical Analysis. Pp. 278–87 in Aphek-Antipatris I: Excavation of Areas A and B, the 1972–1976 Seasons, ed. M. Kochavi. Tel Aviv: Tel Aviv University Press. Stech, T.; Muhly, J. D.; and Maddin, R. 1985 Metallurgical Studies on Artifacts from the Tomb Near ºEnan. ºAtiqot (English Series) 17: 75–82. Stech, T., and Pigott, V. C. 1986 The Metals Trade in Southwest Asia in the Third Millennium b.c. Iraq 48: 39–64. Tubb, J. N. 1982 A Crescentic Axehead from Amarna (Syria) and an Examination of Similar Axeheads from the Near East. Iraq 44: 1–12. 1983 The MB IIA Period in Palestine: Its Relationship with Syria and Its Origin. Levant 15: 49–62. 1985 Some Observations on Spearheads in Palestine in the Middle and Late Bronze Ages. Pp. 189–96 in Palestine in the Bronze and Iron Ages, ed. J. N. Tubb. London: Institute of Archaeology. Tufnell, O. 1958 Lachish IV (Tell ed-Duweir): The Bronze Age. London: Oxford University Press. Tufnell, O., and Ward, W. A. 1966 Relations between Byblos, Egypt and Mesopotamia at the End of the Third Millennium bc. Syria 43: 165–241. Waetzoldt, H. 1990 Zur Bewaffnung des heeres von Ebla. Oriens Antiquus 29: 1–38. Ward, W. A. 1971 Egypt and the East Mediterranean World, 2200–1900 bc. Beirut: American University of Beirut. Weinstein, J. W. 2003 A New Set of Radiocarbon Dates from the Town Site. Pp. 638– 48 in Bâb edh-Dhrâº: Excavations at the Town Site (1975– 1981), ed. W. E. Rast and R. T. Schaub. 2 vols. Winona Lake, IN: Eisenbrauns. Yadin, Y. 1971 A Note on the Scenes Depicted on the ºAin-Samiya Cup. Israel Exploration Journal 21: 82–83. Yeivin, Z. 1971 A Silver Cup from Tomb 204a at ºAin-Samiya. Israel Exploration Journal 21: 78–81. Yener, K. A. 2000 The Domestication of Metals: The Rise of Complex Metal Industries in Anatolia. Leiden: Brill.

The Chipped Stone Assemblage from Beªer Resisim in the Negev Highlands: A Preliminary Study Steven A. Rosen, Sorin Hermon, Jacob Vardi, and Yael Abadi Archaeological Division Ben-Gurion University of the Negev

Introduction Almost 70 years after Albright’s first systematic attempt to characterize the nonurban period between the urban Early Bronze Age and the urban Middle Bronze Age, the nature of terminal third-millennium societies is still debated. This is reflected, for example, in the disputes over what to call the period, variously Middle Bronze Age I (e.g., Albright 1938; 1962; Cohen 1999: 83–84), Intermediate Bronze Age (Kochavi 1969), Early Bronze–Middle Bronze Age (Kenyon 1980: 119–20), and, of course, Early Bronze Age IV (Dever 1973; Richard 1980). Although on one level all agree on certain basic characteristics of the period and its associated cultures—the nonurban aspect, the basic ceramic repertoire, and elements of continuity with the preceding Early Bronze Age—research on other issues, such as developmental origins, subsistence systems, patterns of movement, and socioeconomic organization and interactions, may still bear much fruit. Within the general study of the societies of the terminal third millennium, the investigation of sites in the southern Levantine deserts has played an especially important role. The seeming anomaly of settlement florescence in a desert region during what appears to be a climatically deteriorating period (A. M. Rosen 1989; 1995) in itself begs explanation. In addition, the apparent contrasts with the preceding Early Bronze Age system in settlement patterns, architecture, material culture, and so on (e.g., Haiman 1996; Goren 1996; Cohen 1999: 283–98) pose important questions for anyone concerned with archaeological culture change and/or desert adaptations. Authors’ note: The authors are grateful to William G. Dever and Rudolf Cohen for the opportunity to study the collection from Beªer Resisim. The Israel Antiquities Authority was helpful in providing access to the materials. Support for the work was provided by a grant to William Dever for the publication of the site, and the research was supported by the Israel Science Foundation (Grant 84–9/02).

Investigations at the site of Beªer Resisim (Cohen and Dever 1978; 1979; 1981) have played a major role in our understanding of this period in the Negev (fig. 1). Following Glueck’s surveys (e.g., 1959), which established the existence of the terminal third-millennium horizon in the area, and Kochavi’s excavations at Yeroham (1967), the data from Beªer Resisim have both added significantly to our knowledge of the basic materials of the period and provided new insights and perspectives. Methodologically, in particular, the work at Beªer Resisim stands out, a direct result of which was the collection of all chipped-stone materials, including industrial waste. 1 Thus, the lithic assemblage from Beªer Resisim provides our first good sample of a large domestic chipped-stone assemblage from the end of the third millennium bce in the Negev Highlands. Consequently, the analytical potential of this assemblage is great. At the simplest level, these materials provide a baseline for characterizing the lithic industries in the Negev during this period. Beyond this, the assemblage can be compared with the assemblages of preceding and succeeding periods and adjacent regions, providing new data on cultural and chronological affinities and attributions. Analysis of waste assemblages and raw materials can indicate import and/or export of stone tools and/or onsite manufacture. The basic structure of lithic production can (and should) be incorporated into any model of the economy of these desert societies. Functional analysis of the tool assemblages provides important insights into the activities that actually took place on the site, activities perhaps not reflected in other elements of the material culture. In short, chipped-stone tools played a major role in the basic lifeway of third-millennium desert societies, and the

1. Although the sieving and collection of the smallest artifacts was not exhaustive, the absence of a significant microlithic component suggests that this would not have resulted in an important sampling bias. The microdrills and other bladelets were nevertheless recovered.

Steven A. Rosen, Sorin Hermon, Jacob Vardi, and Yael Abadi

134

Fig. 2. Cores from Beªer Resisim.

Fig. 1. Map showing location of sites mentioned in the text.

materials from Beªer Resisim offer a new opportunity for examining this role.

Description of the Assemblage The chipped-stone assemblage from Beªer Resisim consists of 11,972 pieces. Almost half of them were classified as debris (chunks and chips) and somewhat more than half as debitage (unmodified flake products). The 906 tools represent 7.6% of the lithic assemblage. Basic typological counts are presented in table 1. Raw Materials The raw materials exploited for the manufacture of chipped-stone tools at Beªer Resisim are dominated by relatively high-quality Eocene flint, found in the vicinity of the site in the form of large nodules. These materials are brown in color, varying from light to dark, occasionally with a dark patina. In northern Israel, this type of raw ma-

terial was exploited for the production of Canaanean blades, a technology absent from Beªer Resisim and the Negev in general. Other materials include gray flint, often with a white patina, banded flints on wadi cobbles, and some hard limestone. The textures of all the materials are medium-to-fine grained. All of the materials are found in the vicinity of the site in flint exposures in the local limestone strata or as cobbles in local wadis. In general, this picture of local materials accords with the technological conclusions for on-site production. Indeed, the presence of many large chunks of raw material along with cortical waste suggests import of raw flint nodules and blocks and on-site reduction, rather than off-site core preparation or import of finished tools. Technology Lithic production at Beªer Resisim is dominated by flake products, constituting 73% of the debitage and core assemblage. Flakes vary considerably in size and are not standardized, falling into the general category of ad hoc production (e.g., S. A. Rosen 1997a: 110–12). The relatively high proportion of primary elements to flakes (0.20)

The Chipped Stone Assemblage from Beªer Resisim in the Negev Highlands

135

Table 1. Beªer Resisim Lithic Frequencies Waste flakes primary blades bladelets core trimming Total debitage Cores chunks chips Total debris

4,511 900 471 48 88 6,018 131 1,678 3,239 4,917 11,066

Total waste Tools

Scrapers

Tabular scrapers Retouched blades

Sickles Borers

Truncations

Retouched flakes Notches/denticulates Varia Total tools Total lithics

reflects a reduction sequence extending only a few flakes beyond initial cortex removal, again a sign of basically ad hoc production. Similarly, in their lack of technical and

simple cortical side thumb on blade steep denticulated beaked massive fragments Total

17 4 17 8 8 24 23 4 5 10

simple backed massive varia Total

43 33 8 8

simple small massive drill microdrill Total straight pointed convex other Total

40 21 6 1 40

120 8

13.2% 0.9%

92 18

10.2% 2.0%

108

11.9%

4 22 7 14 47 5.2% 221 24.4% 267 29.5% 25 2.8% 906 100.1% 11,972

metric standardization, blades and bladelets seem more the by-products of flake manufacture than a discrete production process.

136

Steven A. Rosen, Sorin Hermon, Jacob Vardi, and Yael Abadi

Fig. 3. Scrapers from Beªer Resisim: 1. denticulated scraper; 2. beaked steep scraper; 3. simple scraper; 4.–5. tabular scrapers.

Cores are largely amorphous (fig. 2:1–2) and are the result of the import of raw materials to the site, informal preparation of a striking platform, and ad hoc flake production. Beyond the simple platform preparation, there is no evidence of core modification intended to guide or predetermine flake shape, although after the initial flake removal, some of the cores were further modified for the production of blades and bladelets. Thus, while no formal blade or bladelet core was recovered, some cores show mixed flake-blade-bladelet removals (fig. 2:2). This expedient approach to production is also reflected in the low proportion of core-trimming elements (CTE:Core = 0.67), indicating little core preparation or rejuvenation. Given the

large size of the original nodules and the high quality of the flint, the knappers apparently preferred to create new striking platforms and prepare new cores rather than rejuvenate old platforms, which occurred only on a few occasions probably related to the preparation of the cores for blade and bladelet removals. The Tool Assemblage The tool assemblage at Beªer Resisim is dominated by manufacture on flakes, constituting some 84% of the retouched pieces. With the exception of isolated scrapers on blades and a few borers, blade production was used exclusively for sickle segments and other retouched blades,

The Chipped Stone Assemblage from Beªer Resisim in the Negev Highlands providing long, sharp cutting edges. Bladelet tools are represented by the microlithic drills and their manufacture reflected in the presence of bladelet debitage. Finally, a few chunks were modified into crude bifacial elements (not to be confused with Chalcolithic and Neolithic axes). Expedient tools, especially the notches/denticulates and retouched flakes, dominate the assemblage. In fact, considering the crude nature of even the more formal types (the scrapers, retouched blades, and borers), even though they exhibit more standardized attributes, with but a few exceptions (e.g., microdrills) none is actually standardized morphologically. Thus, virtually the entire assemblage can be characterized as ad hoc. Scrapers (120; 13.2%). The great variability (read lack of standardization) in the scraper category is reflected in the nine subtypes discerned, defined by variations in the morphology of retouch and shape of tool (fig. 3:1–3). Denticulated (fig. 3:1) and steep (fig. 3:2) scrapers dominate. The dividing line between these and notches/denticulates is the extent of retouch, the denticulated and steep scrapers having all or most of their circumferences modified, while notches/denticulates have in most cases only one edge modified. It seems likely, in fact, that all of these belong to a single group, reflecting a spectrum of use, repair, and reuse, with the scrapers reflecting later stages in the use-cycle of the notches/denticulates. Of the other subtypes, the massive scrapers are of interest. They were made on large flakes, usually double the size of the regular scrapers. Tabular scrapers (8; 0.9%). Tabular scrapers (fig. 3:4–5) are distinct from other scrapers both technologically and functionally (S. A. Rosen 1983a; 1997a: 71–80). In the microwear study of this type from the Early Bronze Age site of Bab edh-Dhraº, McConaughy concluded that they functioned as knives, not scrapers (1979: 304), and morphologically this seems to make sense. They differ specifically from the cortical scrapers in both raw materials and technology: the cortical scrapers are smaller, derived from smaller cobbles, do not show especially flat dorsal surfaces, and were produced on site. The tabular scrapers show flat cortical surfaces and in fact derive from large nodules of fine-grained Eocene flint (as opposed to tabular flint), and seem to have been manufactured elsewhere and brought to the site. Of the six more or less complete pieces, four are fan-shaped (e.g., fig. 3:5), one is round, and one is elongated with a point (fig. 3:4). Their butt is flat, and no thinning of the bulb of percussion occurs; the retouch is usually parallel, extending over the distal end and some of the distal parts of the edges. Given the apparent absence of tabular scrapers at other EB IV sites in the Negev and Sinai (e.g., Kozloff 1972–73; Gilead 1973; S. A. Rosen 1983a; 1997a: 75), the relative

137

scarcity of the type at Beªer Resisim, and the proximity of Early Bronze Age sites at which these tools are typical, the tabular scrapers in this assemblage may represent some kind of contemporary collecting or perhaps evidence of an ephemeral EB I–II presence on the ridge prior to the primary EB IV settlement. It seems unlikely that they were a product of EB IV knapping. Retouched blades (92; 10.2%). Retouched blades (fig. 4:5–8) were manufactured for the most part on highquality Eocene flint. Almost half are simple blades with partial fine retouch along one of the edges (fig. 4:8). Onethird of the retouched blades are backed, with unipolar, parallel retouch along one of the edges, in most cases resulting in a curved back (fig. 4:5, 7), similar to the archedbacked blades prominent in the EB I–II assemblages from Biqat Uvda (e.g., S. A. Rosen 1997a: fig. 3.21:5–9). On the other hand, these examples are mostly smaller than those in the Biqat Uvda assemblages, being only around 3 cm in length (see especially fig. 4:5). In a few cases, the back forms a straight edge. These items tend to be smaller and are made on a dark-gray pebble flint. Only a single bladelet tool was recovered, an intrusive Ramon point, attributable to the Ramonian culture of the Epipaleolithic period. Eight blades are large and retain some cortex on their dorsal face and along one of the edges, recalling naturally backed knives (fig. 4:6). The possibility that they were knapped off large cores in the first stages of core reduction as part of the preparation of the raw-material chunk to be further used as a core cannot be excluded. In the varia category, six of the eight pieces are irregularly retouched, and the other two are broad, bitruncated blades, reminiscent of Egyptian types. Sickle segments (18; 2.0%). One group of retouched blades consists of items with gloss along their working edge (fig. 4:1–4). Fourteen items with sickle gloss (see especially fig. 4:4) were collected, representing 2% of all tools. A striking feature of the sickle segments is their high diversity in terms of morphology, size, and raw materials. Some have a straight and some a curved back, while others have a ventral retouch that forms the hafted edge. Some have a finely denticulated working edge, some show a ventral retouch, and others have no retouch at all along the working edge. Their only common attribute is the gloss, always along one of the edges. Although using a single attribute—sickle gloss—to define sickle segments in contradistinction to other retouched blades is perhaps methodologically problematic, experiments have shown that gloss takes little time to develop in the course of cereal harvesting, generally from only a few hours’ work (UngerHamilton 1984). Thus, although it is possible that other retouched blades were sickles-to-be, the ratio of 5:1 of

138

Steven A. Rosen, Sorin Hermon, Jacob Vardi, and Yael Abadi

Fig. 4. Sickle segments and blade tools from Beªer Resisim: 1.–4. sickle segments with gloss; 5. small arched-back blade; 6. backed knife; 7. arched-back blade; 8. simple retouched blade.

unused sickle blades : used sickle blades argues against a single group. The association between gloss and reaping (see, e.g., Witthoft 1967; Anderson 1980; see also S. A. Rosen 1997a: 55–58) allows for a neat functional definition as well. One item of particular interest retained bitumen along its hafted edge (fig. 4:1). It is a long, narrow, truncated blade, made on brown Eocene flint with a white patina, a unipolar back, and a working edge shaped by ventral scaled retouch covering almost half of the ventral face at its maximum extent. The gloss is straight, extending along

the entire working edge and covering half the dorsal face as well as parts of the ventral face, hinting that after first use the item was turned and reused. The extent of the gloss and the remains of asphalt suggest that half the item was covered by the haft. The presence of bitumen suggests connections with the Dead Sea area, which is not especially surprising. Borers (108; 11.9%). Borers are dominated by two groups—simple awls and microdrills. The simple awls were made on medium-sized flakes with a single point obtained by the retouch of two notches at one of the ends of

The Chipped Stone Assemblage from Beªer Resisim in the Negev Highlands

139

Fig. 5. Borers and retouched flakes from Beªer Resisim: 1.–2. small awls; 3. simple borer; 4. denticulate, 5.–6. retouched flakes.

the item (fig. 5:3). At the extreme ends of this group are 3 massive and 18 small awls (e.g., fig. 5:1–2). Six items are made on blades, one end narrowed by abrupt retouch along part of their edges, forming a point. One of these is classified as a broken drill.

A concentration of some 40 microlithic drills (see fig. 6) was recovered from a single room (B22), in association with a large number of seashells and ostrich eggshell fragments (Cohen 1999: 210), constituting the only occurrence of microdrills at the site. The drills, clearly intended for

140

Steven A. Rosen, Sorin Hermon, Jacob Vardi, and Yael Abadi

Fig. 6. Microdrills from Beªer Resisim.

hafting and use in composite bow drills (e.g., Burian and Friedman 1985; cf. S. A. Rosen 1997b) were manufactured on small blades/large bladelets. They show bilateral abrupt retouch, converging at the end to form the drill bit or point. They differ significantly from the Early Bronze Age microdrills recovered at the Camel Site (S. A. Rosen 1997b) and Rekhes Nafha (Saidel 2002), both technologically (in their generally larger size) and typologically (in the general scarcity of double- and single-shouldered pieces). The drills are usually triangular and pointed, unlike the longer parallel drill points of the Early Bronze Age microdrills. It is worth noting that they contrast with Chalcolithic microdrills as well, because of their larger size and generally cruder aspect. Whether these contrasts represent differences in function (used on seashell, stone, or ostrich eggshell), lithic raw material, lithic tradition (i.e., chronology), or some combination thereof needs to be tested against other assemblages. Regardless, the clear presence of flint microdrills and bead production in the Beªer Resisim assemblage represents a new aspect of the EB IV economy. Truncations (47; 5.2%). Truncations are flakes with regular and continuous retouch along one end or short edge. There were items with typical truncations, such as

spread is 12 points short

straight, convex, or concave, alongside a few items that were defined as having pointed truncations, since the aim seems not to have been to form a tip, as in the case of awls, but rather a tool with a pointed end. This was obtained by an oblique truncation at the distal end and fine retouch along one or both of the edges. In some cases, the truncation is straight or slightly convex, and the edge has parallel abrupt retouching, thus forming a tool similar to a backed knife, either with a straight or curved back. Retouched flakes and notches/denticulates (488; 53.9%). This composite group represents more than half of all the tools, with each type appearing in almost equal quantity. Retouched flakes are simple flakes in various sizes and shapes, with partial, irregular retouching along one edge (fig. 5:5–6). In a few cases, the retouch is ventral. Notches are simple flakes of a nondetermined shape and size, with one or more retouched notches. Denticulates constitute a particular case, with more than two notches aligned along the same edge (fig. 5:4). In fact, the distinction between denticulated scrapers and denticulates is blurred, apparently representing variations within the continuum of a single tool type, the denticulates at one end of the range and the denticulated scrapers at the other.

The Chipped Stone Assemblage from Beªer Resisim in the Negev Highlands

141

Fig. 7. Tool-class comparison of the Beªer Resisim lithic assemblage with nearby EB I–II assemblages. For the sake of comparability, in the Beªer Resisim assemblage, truncations and varia were included in the retouched flake category; and in the EB I–II assemblages, miscellaneous trimmed pieces were included in the retouched flake category.

Varia (25; 2.8%). This group compriss two classes— bifacially shaped chunks and apparent burins. The bifacials are cortical pieces with large flake removals extending over both faces (not to be confused with the bifacial axes and similar tools of the Neolithic and Chalcolithic periods). Only five apparent burins were found, all on naturally truncated flakes. Their low numbers and association with natural breaks suggest that they were probably unintentional.

Discussion and Summary The large size of the lithic assemblage from Beªer Resisim and the efforts made in its collection provide a reliable sample on which to base conclusions concerning various aspects of EB IV society in the Negev, such as domestic economy, technological traditions, and cultural affinities. With the exception of the limited collection and report on the materials from Har Yero˙am (Gilead 1973), the assemblage from Beªer Resisim offers the first detailed look at the lithic industry of this period in the region. The economic structure of Beªer Resisim society as reflected in the lithic assemblage is a domestic or household economy. The reconstruction of the reduction sequences of the different tool types, the chaînes opératoires, especially as reflected in the configuration of the debitage assemblage, indicates the virtual absence of off-site manufac-

ture. Local raw materials were brought short distances to the site, where they were reduced to blanks—flakes and occasionally blades and bladelets. These were then quickly and simply modified into a range of ad hoc tools, used, and discarded. Even without spatial analysis of the materials, there is no evidence of expert or specialized manufacture of any of the tool types (with the possible exception of the tabular scrapers, discussed below) or of the use of chipped-stone tools for nondomestic tasks, and even the production of beads should best be considered a domestic cottage industry—that is, lithic manufacture and use occurred at the household level. This is not only of inherent interest and importance but also contrasts significantly with other production activities known from the period, such as ceramics and metallurgy, which clearly transcended the domestic. The ad hoc or expedient flake technology associated with virtually the entire assemblage is consistent with this household level of production. The wide range of variability in types such as scrapers, borers, and retouched flakes reflects the lack of standardization rather than the complexities of specialized functions. Even the seemingly more-standardized blade tools, microdrills, and sickles seem to indicate technological continuity with the flake industry, as opposed to a discrete and standardized technology or industry. Similarly, the simple, direct percussion,

142

Steven A. Rosen, Sorin Hermon, Jacob Vardi, and Yael Abadi

hard hammer knapping and retouch manufacturing techniques reflect little complexity in the structure of production and little technical sophistication on the part of the people who actually knapped the flint. The expedient nature of use and the related low value of the tools are also reflected in the generally low intensity of retouch. Even considering the possibility of a range of scraper types representing a repair/reuse cycle (cf. Dibble 1987), this sequence is best seen as the opportunistic reuse of available raw material rather than the curation of valuable resources. As mentioned above, the technological structure of the lithic system contrasts with the complexity of and value associated with, for example, copper production, use, and recycling—both at Beªer Resisim and in the EB IV in general. Functionally, the range of tasks associated with the Beªer Resisim assemblage can also be characterized as domestic. With the exceptions of the sickle segments and microdrills, the various tool types—scrapers, retouched blades and flakes, awls and borers, notches and denticulates—can all be ascribed to light-duty activities associated with basic household consumption and maintenance. These include tasks such as food preparation, leatherworking, light woodworking and whittling, and light cutting of various soft materials (e.g., McConaughy 1979: 334–44). With the possible exception of bead production, there is no evidence of manufacture for export, either of stone tools or of materials produced by stone tools. The only exception is the presence of bitumen derived from the Dead Sea area on the back of one of the sickle segments. The sickle segments constitute an interesting addition to this picture. The presence of gloss on flint blades is unquestionably indicative of grass-cutting, usually associated with agriculture (e.g., Witthoft 1967; Anderson 1980; S. A. Rosen 1997a: 55–58). On the other hand, for a site as large as Beªer Resisim, 18 sickle segments, which constituted only 2% of the tool assemblage, in fact suggest that reaping played little role in the Beªer Resisim subsistence economy. Specifically, the 18 sickle segments represent at most 18 sickles—18 harvesters—and most likely only 3 or 4, given an average of 3–6 blades per composite sickle. Compared with roughly contemporary, comparable sites farther north, for example, Shaºar Hagolan (S. A. Rosen 1997a: 129), this is both proportionally and absolutely quite a small number of sickles. Based on the sickle segments in the assemblage, if agriculture was practiced at Beªer Resisim, it was basically an opportunistic activity, and the possibility that the sickles in fact reflect some aspect of plant gathering, perhaps of wild cereals, cane, or reeds, and are not connected at all with agriculture per se cannot be ruled out.

spread is 12 points short

The lithic assemblage from Beªer Resisim also provides material for comparative study. In general, the assemblage seems quite similar to the assemblage from the roughly contemporary site of Har Yero˙am (Gilead 1973), the only other EB IV desert site published. However, only 19 tools are described in this report, and the waste assemblage, numbering fewer than 200 artifacts, is only mentioned in passing. Gilead (1973) also notes the presence of some 88 pounding stones (cf. Cohen 1999: 266). This large quantity is paralleled in the high number of similar objects in the Beªer Resisim assemblage (which were not analyzed for this study). The presence of ingots, other copper objects, and circumstantial evidence for metallurgical production at Har Yero˙am and at Beªer Resisim may explain the pounding stones, which perhaps were used in the production of copper. The continued use of Canaanean sickles during EB IV in central and northern Canaan constitutes a significant contrast between north and south. Functionally, sickle segments form significantly higher proportions of the EB IV assemblages in the Mediterranean zone (S. A. Rosen 1997a: 127–30), reflecting the basic difference between those agricultural economies and the more pastoral-based economies of the desert. Economically, Canaanean technology reflects a specialized mode of manufacture (S. A. Rosen 1983b; 1997a: 107–9), again in contrast to the domestic mode dominant in the lithic assemblages of the south. The fact that Canaanean blades did not penetrate the societies of the desert, in spite of their status as trade items, suggests that at least on this level the desert exchange economies were not fully integrated with the exchange economies of the Mediterranean zone. Comparison of the assemblage from Beªer Resisim with assemblages from the preceding EB I–II in the Central Negev reveals important contrasts that serve all the more to illustrate the ad hoc domestic nature of the Beªer Resisim materials. Figure 7 summarizes the basic toolclass proportions in the assemblages from Beªer Resisim, Har Horsha (Haiman 1991; S. A. Rosen 1991), and two Nahal Mitnan sites (combined to increase sample size; Haiman 1993; S. A. Rosen 1993), the last three EB I–II sites in the general vicinity of Beªer Resisim. Although the basically expedient nature of all these assemblages is evident and suggests basic underlying technological and functional continuities, two major contrasts are apparent. First, as indicated above, the extremely low proportion of tabular scrapers at Beªer Resisim contrasts significantly with the proportion in the Early Bronze Age sites. In EB I– II, tabular scrapers represent an off-site production system distinct from the ad hoc, on-site production system reflected by other tool types in the Negev Highlands in this

The Chipped Stone Assemblage from Beªer Resisim in the Negev Highlands period (S. A. Rosen 1983a; 1997a: 111; see also Quintero, Wilke, and Rollefson 2002). Although it is likely that the tabular scrapers at Beªer Resisim are intrusive, even if they were associated with EB IV occupation it is clear that their low numbers would represent a major change in the structure of production as compared with the earlier system. In addition to the contrast represented by the tabular scrapers, the Beªer Resisim assemblage exhibits higher proportions of both notches/denticulates and, especially, retouched flakes. Specifically, if the different tool classes are ranked in some order of degree of expediency, these elements can be considered more expedient than the more formal types—the blade tools and sickles, scrapers, and borers. Thus, the Beªer Resisim assemblage shows a trend toward increasing expediency of production. The Mitnan and Horsha EB I–II assemblages are small and relatively limited, but the sites are geographically close to Beªer Resisism. If the comparison is expanded to include other EB I–II sites in the desert regions, it is appar-

143

ent that an additional component of the EB I–II lithic system drops out of the EB IV lithic repertoire. The absence of microlithic lunates and other small and transverse arrowheads contrasts to their presence at various EB I–II sites (S. A. Rosen 1983c; 1997a: 39–44). It is tempting to suggest that this disappearance somehow reflects a change in hunting systems, either a decline in the importance of hunting or perhaps a technological change that no longer required stone points. In any event, the disappearance of this component results in a basic “simplification” of the total lithic system. More detailed work on the Beªer Resisim assemblage, including spatial analysis of the lithic artifacts and a detailed technological study, will add more to our picture of the basic lifeways at the site. Studies of the lithic industries from contemporary sites will hopefully provide information on the variability inherent in the system. This report is only preliminary, and much remains to be learned from the lithic industries of EB IV.

References Albright, W. F. 1938 The Excavation of Tell Beit Mirsim II. Annual of the American Schools of Oriental Research 17. New Haven: American Schools of Oriental Research. 1962 The Chronology of Middle Bronze I (Early Bronze–Middle Bronze). Bulletin of the American Schools of Oriental Research 168: 36–42. Anderson, P. C. 1980 A Testimony of Prehistoric Tasks: Diagnostic Residues on Stone Tool Working Edges. World Archaeology 2: 181–94. Burian, F., and Friedman, E. 1985 Flint Borers: Methods of Production and Use. Mitekufat Haªeven 18: 63–66. Cohen, R. 1999 Ancient Settlement of the Central Negev. Israel Antiquities Authority Reports 6. Jerusalem: Israel Antiquities Authority (Hebrew). Cohen, R., and Dever, W. G. 1978 Preliminary Report of the Pilot Season of the “Central Negev Highlands Project.” Bulletin of the American Schools of Oriental Research 232: 29–45. 1979 Preliminary Report of the Second Season of the “Central Negev Highlands Project.” Bulletin of the American Schools of Oriental Research 236: 41–60. 1981 Preliminary Report of the Third and Final Season of the “Central Negev Highlands Project.” Bulletin of the American Schools of Oriental Research 243: 57–77. Dever, W. G. 1973 The EBIV–MBI Horizon in Transjordan and Southern Palestine. Bulletin of the American Schools of Oriental Research 210: 37–63. Dibble, H. L. 1987 The Interpretation of Middle Paleolithic Scraper Morphology. American Antiquity 52: 109–17.

Gilead, D. 1973 Flint Industry of the Bronze Age from Har Yeru˙am and Tell Nagila. Pp. 133–43 in Excavations and Studies, ed. Y. Aharoni. Tel Aviv: Tel Aviv University (Hebrew with English summary). Glueck, N. 1959 Rivers in the Desert. Philadelphia: Jewish Publication Society. Goren, Y. 1996 The Southern Levant in the Early Bronze Age IV: The Petrographic Perspective. Bulletin of the American Schools of Oriental Research 303: 33–72. Haiman, M. 1991 A Site of the Early Bronze Age at the Foot of Har Horsha. ºAtiqot 20: 1–20 (Hebrew with English abstract). 1993 An Early Bronze Age Cairn Field at Nahal Mitnan. ºAtiqot 22: 49–61 (Hebrew). 1996 Early Bronze Age IV Settlement Pattern of the Negev and Sinai Deserts: View from Small Marginal Temporary Sites. Bulletin of the American Schools of Oriental Research 303: 1–32. Kenyon, K. 1980 Archaeology in the Holy Land. 4th ed. London: Benn Norton. Kochavi, M. 1967 The Wave of Settlement in the Middle Bronze (Canaanite) Age I in the Negev. Unpublished Ph.D. dissertation, Hebrew University of Jerusalem (Hebrew). 1969 The Middle Bronze I Period (The Intermediate Bronze Age). Qadmoniot 6: 38–44 (Hebrew). Kozloff, B. 1972–73 A Brief Note on the Lithic Industries of Sinai. Museum Haªaretz Yearbook 15/16: 35–49. McConaughy, M. 1979 Formal and Functional Analysis of Chipped Stone Tools from Bab edh-Dhra, Jordan. Unpublished Ph.D. dissertation. Pittsburgh: University of Pittsburgh.

144

Steven A. Rosen, Sorin Hermon, Jacob Vardi, and Yael Abadi

Quintero, L. A.; Wilke, P. J.; and Rollefson, G. O. 2002 From Flint Mine to Fanscraper: The Late Prehistoric Jafr Industrial Complex. Bulletin of the American Schools of Oriental Research 327: 17–48. Richard, S. 1980 Toward a Consensus of Opinion on the End of the Early Bronze Age in Palestine–Transjordan. Bulletin of the American Schools of Oriental Research 237: 5–34. Rosen, A. M. 1989 Environmental Change at the End of Early Bronze Age Palestine. Pp. 247–56 in L’Urbanisation de la Palestine à l’Age du Bronz Ancien, ed. P. de Miroschedji. British Archaeological Reports International Series 527. Oxford: British Archaeological Reports. 1995 The Social Response to Environmental Change in Early Bronze Age Canaan. Journal of Anthropological Archaeology 14: 26– 44. Rosen, S. A. 1983a The Tabular Scraper Trade: A Model for Material Culture Dispersion. Bulletin of the American Schools of Oriental Research 249: 79–86.

1983b The Canaanean Blade and the Early Bronze Age. Israel Exploration Journal 33: 15–29. 1983c The Microlithic Lunate: An Old-New Tool Type from the Negev, Israel. Paleorient 9: 81–83. 1991 The Lithic Assemblage from the EB Site at Har Horsha. ºAtiqot 20: 169–76. 1993 Lithic Assemblages from Nahal Mitnan. ºAtiqot 22: 62–69. 1997a Lithics after the Stone Age. Walnut Creek, CA: Altamira. 1997b Beyond Milk and Meat: Lithic Evidence for Economic Specialization in the Early Bronze Age Pastoral Periphery in the Levant. Lithic Technology 22: 105–15. Saidel, B. A. 2002 The Excavations at Rekhes Nafha 396 in the Negev Highlands, Israel. Bulletin of the American Schools of Oriental Research 325: 37–63. Unger-Hamilton, R. 1984 The Formation of Use-Wear Polish on Flint: Beyond the “Deposit versus Abrasion” Controversy. Journal of Archaeological Science 11: 91–98. Witthoft, J. 1967 Glazed Polish on Flint Tools. American Antiquity 32: 383–89.

Podium Structures with Lateral Access: Authority Ploys in Royal Architecture in the Iron Age Levant Ilan Sharon and Anabel Zarzecki-Peleg Institute of Archaeology The Hebrew University of Jerusalem

Introduction Bit Hilani: The Typical Israelite Administrative Structure? It has been almost 40 years since Ussishkin first suggested that Palace 1723 at Megiddo was a Syrian bit hilani–style palace (1966; 1973). This afforded a modicum of archaeological support for a conjecture presented some 70 years ago that the somewhat obscure description of Solomon’s palace in 1 Kgs 7:1–12 was a description of a bit hilani (Watzinger 1933: 95–97; Galling 1937: 411). A local and purportedly contemporary parallel for the hypothesized Solomonic edifice thus appeared to have been established. The bit hilani hypothesis was expanded to include other structures—for example, Palace 6000 at Megiddo and the structure at Bethsaida—and conflated by repetition in almost all textbooks (e.g., Dever 1990: 107–8; A. Mazar 1990: 382–84; Barkay 1992: 311; Reich 1992: 204–6; Herzog 1997: 214), thus promulgating the notion that the typical palatial structure of the southern Levant was bit hilani in style. This notion, however, was questioned by some scholars (e.g., Wright 1985: 276–77; Kempinski 1989: 162) and rejected by others (e.g., Fritz 1983a; 1983b: 22–37; 1995: 153), but a consistent typological alternative was never put forward (although Nigro [1995: 193–291, 327–71] does propose alternatives, he does not suggest a consistent type). In the following essay, we attempt to demonstrate that bit hilanis are not characteristic of the southern Levant and suggest an alternative building type, what we call a Lateral-Access Podium (LAP) structure. We then discuss the significance of this building type for the ideology of power in the Iron Age Levant. Bit Hilani: Definition The term bit hilani is taken from an Assyrian text that states that Sargon II built his palace at Dur Sharukin as a Authors’ note: The authors thank Elizabeth Bloch-Smith, Ruhama Bonfil, and Talia Goldman for their helpful comments on earlier drafts of this article, and Ruhama Bonfil, Talia Goldman, and Svetlana Maskevich for the graphics.

“bit hilani in the Syrian style.” The term is not known in Akkadian. Singer maintains that hilani is taken from Luwian and means ‘opening’ or ‘portico’ (1975), referring to the colonnaded porch characteristic of these buildings in the Neo-Hittite kingdoms in northern Syria. Whatever the Assyrians may have meant by the term, the Iron Age palaces at Zinjirli, Tell Taºyinat, Tell Halaf, and Sakje Guezi are typically broad-house buildings, 30– 60 m x 25–35 m in size, with the main frontal portal of the building perpendicular to the long axis (figs. 1–2). They exhibit three basic characteristics: (1) they are approached from a courtyard directly or via a broad shallow stairway and a broad portico with one to three columns dominating the façade; (2) the portico leads to a broad throne room with the throne situated at the narrow end of the room, usually to the left of the entrance from the portico; and (3) subsidiary rooms are accessed from the throne room. Symmetry is not of importance in the plan of these buildings: the portico and entrance may be either at the center of the broad façade or off-center, with the access symmetry often being marred by the existence of a stairwell on one side of the portico and the long secondary axis of the building being purposefully asymmetrical, with the entrance portico on one side and either nothing or various service rooms on the other. The examples from Syria that the southern Levantine examples supposedly emulate are all dated to the late 9th– 6th centuries bce. 1 No palatial buildings in Israel or in the southern Levant as a whole can be demonstrated to have these defining characteristics, and even the ones that might have them are atypical (Reich 1992: 206). The reason for this uncertainty is that the crucial evidence of the superstructure of these buildings did not survive.

Lateral-Access Podium (LAP) Structures: Definition and Setting Some previous attempts to define palatial architecture— including the bit hilani—have been based on matching 1. For possible Late Bronze Age antecedents of this plan, see Frankfort 1952; Margueron 1979; Zarzecki-Peleg 2005a; in press.

146

Ilan Sharon and Anabel Zarzecki-Peleg

Fig. 1. Examples of bit hilani buildings in northern Syria (reprinted by permission from Margueron 1979) and LAP structures at sites in Israel (scale 1:1250).

Authority Ploys in Royal Architecture in the Iron Age Levant

147

Fig. 2. Isometric reconstructions of (a) hypothetical LAP structure and (b) bit hilani.

architectural remains to textual descriptions, while others —including ours—focus purely on archaeological typology. The latter method runs into inherent difficulties. The structures under discussion constitute a polythetic set; that is, it is perfectly feasible that no single building displays all of the defining characteristics of the type, and no single attribute is present in all of the buildings. This is further complicated by the poor state of preservation of most of the structures. Even the position of the doorways and access through the building are unknown. Hence, the usual typological classifications of buildings—for example, broad house vis-à-vis long house—become indeterminate. 2 We begin by defining a virtual archetypal “palace”/administrative center for the southern Levant and then discuss the structures that we propose belong in this category Definitions Our virtual building is a high, tower-like structure (fig. 2:a), typically 15–20 m x 15–20 m in size. It can be— and has been—characterized as a “palace,” “fort,” or “citadel” (se, e.g., Lamon and Shipton 1939: 17–24; Yadin 2. In order to deal with this issue, we use the following definitions: the long main axis of a structure is the line that crosses it parallel to its longest walls. When this is impossible to determine (e.g., when the building is square or the plan is incomplete), we refer to the internal long axis, the line that bisects the main space of the building lengthwise. The secondary or broad axis is perpendicular to the long one, bisecting the building in the middle of its long sides. Axial (mirror) symmetry refers to a plan bisected by a line so that the two halves are symmetrical. Where perfect symmetry does not exist, an axis of maximal symmetry can be defined. In cases in which the line perpendicular to the axis of maximal symmetry also displays a degree of symmetry, we refer to secondary axial symmetry. Finally, the portal axis of the building is the line perpendicular to the center of its central doorway.

et al. 1960: 43–53; Ussishkin 2004: 768). It is conspicuous first and foremost for its monumentality. It is at least one order of magnitude larger than ordinary houses in almost every respect—overall size, thickness of walls, size of the stones used for construction, and depth of foundation. Especially significant is the ratio of the thick, load-bearing walls to the narrow spaces between them. These elements point to a considerable height of the original structure. Another distinguishing characteristic of the building is its ashlar construction, whether entirely or partially (see Shiloh 1979a; Sharon 1987; Stern 1988; Reich 1992: 211–12; Elayi 1996). In Iron Age Israel and Judah, ashlar architecture is generally limited to public or “royal” buildings and their enclosure walls and gatehouses, and to fortifications. Our virtual structure sits atop a high podium (either filled with earth or hollow, made up of cellars) against an enclosure wall or the city wall itself. Access to the podium is gained by climbing a long, narrow ramp attached to the side of the podium. At the top of the ramp, one would have to turn and enter the back corner room or one of the side rooms of the building, go through this room to the central space, and again turn again to face the internal long axis. Close to the entrance is an internal stairwell providing access to the upper story/ies and/or roof. The stairwell may have protruded above the roof to form a tower. The plan of the building is either almost square or rectangular. The long axis of the building and the internal long axis of its main central space correspond. It also exhibits strong axial symmetry, especially in relation to its long main axis, and the secondary axis often displays nearsymmetry. The structure is thus focused around its central space. As this space is usually surrounded by rooms, the

148

Ilan Sharon and Anabel Zarzecki-Peleg

Fig. 3. Administrative enclosures: (a) Megiddo Stratum IVA (reprinted by permission from Herzog 1992: fig. 16) and (b) the Samaria acropolis (reprinted by permission from Herzog 1992: fig. 14) (scale 1:3000).

only access for light and air is from above, so it is either a courtyard (e.g., Fritz 1995: 153) or a high, clerestorylighted hall (the excavators of Megiddo provide illustrations for both possibilities in their reconstruction of Palace 1723 [Lamon and Shipton 1939: figs. 29, 43]). In either case, this is the only space in which activities involving more than a few people could have been carried out within the building. Most of the subsidiary rooms around the central space are small and typically square in shape, and their

only door leads into the central space. More elaborate examples may have several such halls/courtyards that are interconnected and additional subsidiary rooms outside this nucleus. This undermines the symmetry exhibited by the simple structure. Buildings of this sort may have been in use, with renovations, for as long as several centuries. Due to their monumentality and the fact that they were raised high above the rest of the town, the podia tended to be reused without

Authority Ploys in Royal Architecture in the Iron Age Levant

Fig. 4. Administrative enclosures: (a) Lachish Stratum III (reprinted by permission from Herzog 1992: fig. 18) and (b) Jezreel (reprinted by permission from Ussishkin and Woodhead 1997: fig. 4) (scale 1:3000).

149

150 raising the floor levels. The extant floor level represents only the last of a long series of occupations. Expansion or alterations may have been implemented either by successive extensions of the podia and/or rebuilding within the confines of the previous podium. Alternatively, the previous superstructure may have been razed. Moreover, the massive podium would have remained intact even after the building was destroyed, and other uses might have been found for it in subsequent strata. In the case of Lachish, for instance, the only preserved floor level above the podium dates to the Persian period (Tufnell 1953: 131–35), and the last phase of the bamah at Dan is dated to the Roman period (Biran 1994: 218). In searching for a name that would capture the essence of these buildings, we did not want a loaded term that would presume too much about their function—while these structures are related to ruling and administration, we have no evidence that they are palaces in the usual sense of the word (i.e., royal residences). Nor did we want to confuse literary interpretation with an archaeological terminus technicum by adopting a term from the ancient literature (such as bit hilani or Beit Millo) for an etically-defined type. On the other hand, we wanted the name to reflect the difference in strategies for the control of access between these and the so-called bit hilani–type structures. The regulation of accessibility and space in our structure is achieved by raising it high above the surrounding courtyard and denying direct access to it—entrance requires going around the side of the structure and up a ramp. We therefore chose the name Lateral-Access Podium (LAP) structure. The main differences between the bit hilani–buildings in Syria and the LAP structures in the southern Levant as defined here are (fig. 2): • The bit hilanis in northern Syria are much larger than the structures known in the southern Levant (with the exception of the Lachish complex) (fig. 1). The difference in size may imply a different function. • The bit hilani is a direct-access building (i.e., its main portal is on its broad main axis), where access to the LAP building is from the side, with the doorway typically near the rear corner. • The bit hilani is a broad-house and the LAP structure is either square or elongated. • The LAP structure is almost symmetrical in terms of both its long main and broad secondary axes. If the bit hilani displays any symmetry at all, it is only in relation to its broad main axis. • In LAP structures, the main functional floor of the building is raised high above ground level either on a high podium or above a ground floor consisting of cellars. The bit hilani is usually raised by only two–three steps.

Ilan Sharon and Anabel Zarzecki-Peleg • The steps bridging the elevation difference in the bit hilani are broad and frontal, whereas the ramp in the LAP building is long and lateral. • At the top of the steps in the bit hilani is a broad, open portico with columns; the ramp in the LAP structure probably led to a single doorway. • While some of the Syrian bit hilani buildings have elaborate stonework and/or sculpture on the portico, ashlar construction is not one of their characteristics. Most of the LAP buildings are ashlar, but architectural ornamentation other than the austere proto-Aeolic capitals and perhaps recessed windows and crenellations is absent. Setting Administrative buildings of the LAP type usually occur in very specific contexts, clearly associated with other structures of distinct types (figs. 3–4). The building is typically the main structure in a vast enclosure located at the highest point of the tell. The enclosure may take up a significant portion of the inhabitable space atop the mound, and in some cases all of it (for example, at Jezreel and perhaps Samaria). It is separated from the rest of the town by a wall and/or a ring of subsidiary buildings (see further below) and is accessed via a gatehouse. There are usually subsidiary buildings adjacent to the main podium structure and/or around the enclosure, with a large open courtyard between them, immediately in front of the main building. Several distinct types of subsidiary buildings may be noted: • Tripartite buildings are a well-documented and much-discussed architectural type in Iron Age Israel, particularly with regard to their function. It has been suggested that they were used as stables (Yadin 1976; Holladay 1986), storehouses (Pritchard 1970; Yadin 1972: 168), barracks (Fritz 1995: 142), and marketplaces (Herr 1988). We have little to add on this subject at present, except to reaffirm the association of at least some of these buildings with the “royal”/administrative enclosures under discussion. • It is clear that large-scale storage capacity constituted one of the functions of most—if not all—of these enclosures, for which the cellars within the podium (where these exist) and groups of long-hall complexes were used. In contrast to the tripartite colonnaded buildings, the long halls are not interconnected except at their entryways. These long dark halls that lead nowhere could hardly have been used for any other purpose than storage. Herzog calls them “treasuries” (1992: 228). • Buildings of the type sometimes referred to as scribal chambers are also associated with these administrative complexes. The ascription of scribal activities to this type of building is based on the renowned Samaria Ostraca that were found in one of these units (Herzog 1992: 229–30).

Authority Ploys in Royal Architecture in the Iron Age Levant

151

• In one case, at Dan, there is evidence of cultic activity in the courtyard in front of the podium building (see further below).

Case Studies The evidence is presented below beginning with sites at which there are buildings that exhibit most of the core attributes of the LAP group. The less-certain examples are those in which the central building has fewer of the defining properties, structures that were not fully excavated, LAP structures in a nonurban setting, and complexes for which attribution to this group is debatable. Hierarchy among the sites/structures, their spatial distribution/attribution to polities, and their temporal development are dealt with in the ensuing discussion. Because most of the buildings have been amply described elsewhere, a detailed discussion here would be redundant (for syntheses and further bibliography, see Reich 1992; Herzog 1997). Megiddo Palace 1723, Palace 6000, and Building 338 Palace 1723 and Building 338 at Megiddo definitely fall within the LAP structure category; Palace 6000 was not completely excavated, and its attribution to this group is tentative. Palaces 1723 and 6000 are attributed to Stratum IVB–VA (an advanced phase of the Iron IIA) and Building 338 to Stratum IVA (Iron IIB). The literature on Megiddo Strata IVB–VA and IVA is, of course, vast and lacks consensus (for a detailed discussion of these buildings and their stratigraphy, see Zarzecki-Peleg 2005a, with references; in press). Palace 1723 (figs. 1, 5) is a massive, almost square podium, with its back on the southern edge of the tell on what had been its highest side. The Chicago expedition found only the foundations, except a few places in which a single course of ashlar superstructure remained (Lamon and Shipton 1939: 21–23, fig. 12). In order to identify Palace 1723 as a bit hilani type, Ussishkin (1966) reconstructed a frontal entryway through Room H, a doublecolumned portico, to a broad throne room, Room K. The relationship between these two rooms and the rest of the building, however, does not fit the bit hilani plan (for other critiques of the reconstruction of Palace 1723 as a bit hilani, see Fritz 1983a: 58; 1983b: 22–24; Kempinski 1989: 162–63; Nigro 1995: 223–29; Arav and Bernett 2000: 72– 74). We suggest that Palace 1723 may be the earliest example of a LAP structure. Protruding from the square plan of the building is a “porch” or platform (1728), a solid mass of large stones, which is the base of an entranceramp that is a defining characteristic of the LAP group. Thus, the building would have had an entryway on the side, Room 1723 would be the wide central space of the LAP structure, and Room M would include the square

Fig. 5. Megiddo Palace 1723 (reprinted by permission from Reich 1992: fig. 2) (scale 1:750).

stairwell. However, the defining axial symmetry around the central space is not as strong in Palace 1723 as in other structures. 3 Palace 6000 (figs. 1, 6), like Palace 1723, is an ashlarbuilt monumental structure, which is backed against the northern edge of the tell. It was partly excavated by Yadin (1966; 1967; 1970; 1972: 150–58; see also Zarzecki-Peleg 2005a: 27–44), and the exposure was enlarged in the course of the recent Tel Aviv University excavations (Finkelstein 2000: 120–21). Nevertheless, our knowledge of this building remains limited. To the west of the main structure are three additional square rooms with ashlar walls in header construction that Yadin interpreted as part of a casemate city wall (1967: 120; 1972: 156–58). While this interpretation is no longer generally accepted, it does indicate that Palace 6000 was part of a larger complex (see Zarzecki-Peleg 2005a: 78–80 for possible reconstructions). Yadin characterized Palace 6000 as a bit hilani type, citing affinities with Palace III at Zinjirli (1970: 45). Zarzecki-Peleg (2005a: 59–75; in press) suggests that more appropriate parallels are selected units in Building I at Tell Taºyinat (Margueron 1979: 168, bit hilani Type C; see also Haines 1971: 44–66, pls. 102–3). While several architectural elements of this building are indeed compatible with a bit hilani interpretation, reconstructing it as a LAP type is equally feasible. If it is a LAP structure, it would be more elongated than square.

3. Detailed stratigraphic analysis by Zarzecki-Peleg (2005a: 146–52) suggests that Palace 1723 is earlier than the surrounding enclosure (consisting of Courtyard 1693, Wall 1610, Gatehouse 1567, and the “scribalchambers” complex Building 1842). This accords with previous evaluations (e.g., Kenyon 1964: 150; 1971: 64–66; Currid 1991: 35) maintaining that Courtyard 1693 must be stratigraphically later than Palace 1723, because its thick floors run up to another building constructed over Palace 1723, of which only a few walls survived. Although a monumental enclosure around Palace 1723 would nicely fit the model we are suggesting, there is no evidence for (or against) the presence of a complex of this sort beneath Courtyard 1693.

Ilan Sharon and Anabel Zarzecki-Peleg

152

N 0

20 m

Fig. 6. Megiddo Palace 6000 (scale 1:750).

Megiddo Building 338 (figs. 1, 3:a, 7). The transition from Stratum IVB (= VA–IVB) to IV (= IVA) at Megiddo is represented by extensive renovations and public works. In both the southern and northern parts of the mound, similar complexes were built, each consisting of a large, enclosed administrative structure and large “stables.” The southern stables were associated with the complex that succeeded Palace 1723 (see n. 3). Not enough of the central building remained to draw a coherent plan. The northern stables were built above Palace 6000. Associated with these stables, with its back at the eastern edge of the tell, was Building 338, with its own—possibly casemated—enclosure (Plaza 313) and gatehouse (355). 4 Its clear association with the stables indicates an administrative use. Building 338 exhibits several distinctive features that indicate that it is of the LAP type, first and foremost the fact that the structure is a podium: the floor levels inside are considerably higher than the plaza outside. Other indicators are its ashlar construction and associated protoAeolic capitals, as well as the solid-stone entrance ramp attached laterally to the main structure and the square stairwell next to the entrance. 4. Building 338 and its complex were excavated piecemeal by several different excavators (Schumacher 1908: 110–21; Fisher 1929: 68– 74; Lamon and Shipton 1939: 47–59); for the incorporation of the data into one plan, see Ussishkin 1989: 149–62, fig. 4. Ussishkin has argued that Building 338 should be reassigned to Stratum IVB–VA (1989: 161– 62; see also McClellan 1975: 90–91), unlike most other investigators, who ascribe it to Stratum IVA (including the authors; see the detailed discussion in Zarzecki-Peleg 2005a: 153–62). Furthermore, we see no compelling evidence for a cultic use of this building, as Ussishkin maintains.

Fig. 7. Megiddo Building 338 (reprinted by permission from Herzog 1992: fig. 16) (scale 1:750).

Hazor: The Citadel (figs. 1, 8) The citadel complex at Hazor, comprising a central building and several “scribal chamber” units, is located on the western edge of the tell (Area B) at the highest point of the acropolis (Yadin et al. 1989: 88–112, with previous bibliography). The complex continued in use, with alterations, from Stratum VIII through Stratum V, when the city was destroyed. Stratum VIII at Hazor (the Iron IIA–B transition) appears to include a fully developed administrative center, with storeroom complexes and silos (Areas A and M) (Ben-Tor 1993: 253, fig. 1; 1999: 34, 37; Bonfil 1997: 114–19, plans 2.13–2.14). The citadel building, the largest single structure at Iron Age Hazor, is almost square (25 x 21.5 m). Because the preserved part of the building has no access into it at ground level, the plan is of the cellars, showing that the building was raised on a hollow podium. It stands to reason that the central division of the cellar into two long, narrow halls indicates that the walls provided structural support for the floor of a wide central space in the main story above them. Abutting the front of the main citadel building is an annex (fig. 8) comprising frontal tower 3168 (Yadin et al. 1989: plan 22), structures 3188 and 3156, and ramp 3661—a solid stone platform lined with ashlars. 5 In con5. The excavators interpreted at least some of the annex structures as late additions but had difficulties in assigning them to a specific stratum

Authority Ploys in Royal Architecture in the Iron Age Levant

153

Fig. 8. Hazor citadel (reprinted by permission from Yadin et al. 1989: plans 20–21) (scale 1:750).

Fig. 9. Dan Stratum IVA pre-LAP structure installations (reprinted by permission from Biran 1994: fig. 143) (scale 1:750).

trast to the excavators, we consider all of these to be structural phases of the same primary construction. A staircase abuts the northwestern corner of the building, presumably leading to a doorway on the second story. We believe that the original main access to the structure was via the southeastern corner and that the northwestern access ramp was secondary. Whether it was constructed only when the original ramp went out of use or whether the two were in fact used together is impossible to determine on the basis of the available evidence. In Stratum VIII, the excavators found sections of floors that they reconstructed together as a wide, open plaza in front of the citadel. How large it may have been and whether it had an enclosure wall separating it from the rest of the town is unknown. This plaza is built over by Stratum V (Yadin et al. 1960: 47–69, plan 205; Yadin et al. 1989: 105, plan 24). Two proto-Aeolic capitals were found in secondary use in this area in Stratum VII (Yadin et al. 1989: 93; for suggestions regarding the original placement of the capitals, see Yadin 1959: 79; 1972: 171; Reich 1992: 213). The defining characteristics of the Hazor citadel are that the main level is above the cellars, raised high above the surrounding courtyard, and is accessed by a ramp or

ramps attached to the side of the main structure. The ground plan, consisting of a long central space with small square rooms on three sides of it and a stairwell, is a classic LAP structure plan. The technique of annex structures abutting the podium is also part of LAP architecture, as is the ashlar construction in the corners and access ramp.

(Yadin et al. 1989: 95–96, 99–100, plans 20–21). However, parts of this annex, including Ramp 3661, went out of use sometime before the citadel did: a four-room house attributed to Stratum VA was partially built over it, and there are even earlier walls (attributed by the excavators to Stratum VB) that straddle both Ramp 3661 and Building 3188 (Yadin et al. 1989: 104, plans 23–24). Although the excavators show the northwestern stairway on all of their plans beginning with Stratum VIII (Yadin et al. 1989: plan 20), the earliest excavated floor is the floor in Stratum VI (Yadin et al. 1989: plan 22).

Dan: The Bamah (figs. 1, 9–12) The central area at Dan, Area T, includes a monumental cultic/administrative complex consisting of a central building with a small square structure in front of it and a series of casemate rooms forming an enclosure around both. The excavators interpreted these structures together as a bamah (“high place”) which was built at Dan, according to 1 Kgs 12:25, by Jeroboam I (Biran 1994: 159– 233). 6 Based on the complex’s affinities with the group of structures under discussion, Barkay has argued that the cultic place was limited to the enclosure, and the building itself was a palace/administrative center (1992: 312), an interpretation with which we agree. The only extant floor relating to the main structure dates to the Roman period. Immediately beneath it is a confusing mass of foundation walls, dating all the way back to Iron IIA. Although preliminary attempts have been made to disentangle the stratigraphy of these remains (Biran 1982; 1994: 161–228; 1996: 32–49; 2002: 28–30), we suggest a different interpretation in table 1 (figs. 9–12).

6. Biran initially reconstructed the main structure as an open-air solid podium (e.g., 1971: 10) but subsequently amended his view to interpret it as a temple, with open-air cultic activities taking place in the enclosure in front of it (1994: figs. 149, 163; see also Shiloh 1979b: 153; Stager and Wolff 1981: 99).

154

Ilan Sharon and Anabel Zarzecki-Peleg

Fig. 10 (left). Dan, first monumental buildings, post-Stratum IVA/ pre-Stratum III (reprinted by permission from Biran 1994: fig. 143) (scale 1:750). Fig. 11 (middle). Dan, LAP structure, Strata III–I (reprinted by permission from Biran 1994: figs. 149, 163) (scale 1:750). Fig. 12 (right). Dan, prostyle temple phase, Hellenistic (?)/ Early Roman period (reprinted by permission from Biran 1994: figs. 163, 176) (scale 1:750).

There is a conceptual change in the layout of Area T between Stratum IVA, crowded with thin fieldstone walls and installations (fig. 9), and the large Stratum III monumental ashlar structures with a wide, open courtyard. In our opinion, Bamah A represents an intermediate phase both stratigraphically and conceptually. We believe that the ashlar construction of Bamah A and the earliest “altar” (fig. 10) are later than stratum IVA, and that shortly thereafter, the area was restructured and rebuilt on an even grander scale (fig. 11). An intermediate stage of this sort, during which an embryonic plan is reorganized and developed into a full-fledged administrative complex, may also exist elsewhere (e.g., at Megiddo and perhaps Lachish [see below]). Thus, the entire complex—with its central structure, “altar” platform, open courtyard, and enclosure wall with subsidiary rooms in its casemates—reaches its final form in Stratum III. Thereafter, it remains basically unchanged, except for structural modifications in the enclosure wall and its subsidiary rooms, until the late Hellenistic (or perhaps Herodian/Early Roman?) period. At that time, the entire precinct was renovated in the spirit of the new era, and a prostyle temple with a monumental frontal staircase was built on the ancient podium, with a massive high altar enclosed by its own temenos in front of it (fig. 12). This is similar to the development of the royal enclosure at Samaria and perhaps of the Jerusalem palace and temple.

The excavators interpret the solid stone platform abutting the western side of Bamah B as a side staircase or ramp leading up to it, replaced in Stratum II by the monumental frontal staircase. According to our interpretation, the frontal staircase was not added until the Hellenistic period, and access to the podium structure was gained by means of the lateral ramp throughout the Iron Age. 7 Notwithstanding the finds of a cultic nature, we suggest that the enclosure as a whole displays many of the defining attributes of an administrative complex: a casemated enclosure wall abutting the main structure with an open courtyard in front of it situated on the crest of the mound and probably integrated into the town’s defensive system. Evidence that the central building may have been a LAP structure is represented by its ashlar podium and long, narrow side-ramp. Samaria: The Acropolis (figs. 1, 3:b, 13) Iron Age remains were excavated at Samaria–Sebaste by the Harvard University expedition (Reisner, Fisher, and Lyon 1924) and by the joint expedition (J. W. Crowfoot, 7. Two fragmentary walls appear to abut the front of the podium in Stratum III. These are not the banister walls of the frontal staircase but, rather, are sealed by it. Nor are they evidence for an earlier frontal staircase, because sections of the “yellow floor” of Stratum III appear between them (i.e., under the putative staircase). Rather, these walls are part of a subsidiary structure abutting the front of the podium, perhaps in the same manner that the frontal tower abuts the Hazor structure.

Authority Ploys in Royal Architecture in the Iron Age Levant

155

Table 1. Dan: Excavators’ Stratigraphic Scheme versus Our Proposed Alternative Date

Stratum

Published interpretation

Our suggestion

Iron IIA

IVA

Bamah A—solid ashlar construction (beneath and offset slightly to the south of the later bamah structures); earliest of the “altar” constructions—a smaller, square, solid podium/pavement; casemated enclosure wall; various rooms and installations, including one that contained two “snake” pithoi and a “water libation installation” (Biran 1994: 165–83).a

Various irregular rooms with thin walls made of fieldstones and installations (fig. 9). b

Post-IVA / pre-III c

First monumental ashlar construction—Bamah A and the earliest “altar” (fig. 10).d

Iron IIB

III

Rebuilding of main structure (Bamah B); central “altar” platform enlarged and extended north and east; continued use of Stratum IV casemate enclosure, now paved with “yellow floor” (Biran 1994: 184–89).

Modified plan of complex (fig. 11): construction of Bamah B; extension of “altar” platform; construction of casemate enclosure and open courtyard with “yellow floor” that connects these features. e

Late Iron IIB

II (destroyed by the Assyrians)

Monumental ashlar staircase added in front of Bamah B; “altar” platform rebuilt as high podium with two narrow staircases abutting it; f “altar” complex surrounded by ashlar temenos wall. g

No change in main components; rooms adjacent to the enclosure wall undergo structural changes (?).

Iron IIC

I

Changes in structures surrounding enclosure (Biran 1994: 210–14; 1996: fig. 1.37).

Changes in structures surrounding enclosure.

Persian period

Site abandoned but cultic activity continues—some artifacts but no architecture.

No architectural changes.

Hellenistic period

Renovation of cultic center; smaller building constructed above Iron Age bamah; frontal stairway and its banister walls extended (Biran 1994: 214–28; 1996: fig. 1.38).

Construction of prostyle temple on top of Iron Age podium, including: broad-room cella; monumental frontal staircase in its entirety; high altar with two staircases; ashlar temenos wall around altar (fig. 12). h

Roman period

Extant floor above podium.

Extant floor above podium.

a. Stager and Wolff have suggested that the “water libation” installation is an olive-oil press (1981). b. According to the plans (Biran 1994: fig. 143; 1996: fig. 1.34), there is no direct relationship between these installations and Bamah A or the “altar” platform. The latter is not only oriented differently from the walls of the installations, but some of the walls are built too close to it to allow for comfortable access. We therefore suggest that they are earlier than Bamah A and the first “altar.” The excavators date Bamah A on the basis of the pottery found mainly in the subsidiary rooms and installations (Biran 1982: figs. 24–27; 1994: figs. 129, 131). In our proposed stratigraphic sequence, this pottery would represent a terminus post quem for the construction of Bamah A. c. We introduce this nomenclature in order to make the point that the first bamah is later than the other Stratum IVA features and earlier than Stratum III. d. We use “altar” in quotation marks because the excavators note that there were plaster marks on top of the ashlar platform that indicate column bases—we therefore suggest that this feature was an ashlar pavement with some sort of canopy above it, rather than a solid, high altar. e. Although the casemate enclosure already appears in the published plans of Stratum IV, because the enclosure wall cleanly abuts the northwestern corner of Bamah B (Biran 1994: fig. 149; 1996: fig. 1.35), it apparently is a Stratum III construction. f. The “horn” found reused in a later wall is supposed to have stood at one corner of this altar (Biran 1994: 203). g. The dating evidence comes only from the rooms adjacent to the casemated enclosure wall (Biran 1994: 192–204; 1996: fig. 1.36). h. The excavators maintain that, when they removed the frontal staircase, they found Hellenistic pottery beneath only the lowermost three stairs (Biran 1996: 41). The published plans and photographs, however, do not show any noticeable differences in the construction of the frontal staircase or banister walls, and in any event, it is the upper parts of the banisters that clearly straddle Bamah B and are bonded to the Hellenistic structure above it. Furthermore, the monumental staircase is not aligned with the Bamah B podium, and there was no structural reason for this if it was built in Stratum II. Its orientation accords better with the orientation of the Hellenistic structure.

Kenyon, and Sukenik 1942; Crowfoot, Crowfoot, and Kenyon 1957). 8 Retaining walls were built around the en8. The stratigraphy and pottery sequence were established by Kenyon (1942; 1957) and are generally accepted, although controversy has arisen over her method of attributing pottery assemblages to construction phases (see Stager 1990; Tappy 1992: passim; 2001: passim).

tire hill of Samaria to enable the construction of a huge platform that covered the acropolis, a project reminiscent of Jezreel and perhaps Bethsaida in terms of scope and objectives (see below). This was done in two construction stages: first, a single ashlar retaining ring was built (Kenyon’s Building Period I) and later, the podium was extended by adding an outer ring of ashlar retaining walls

156

Ilan Sharon and Anabel Zarzecki-Peleg

Fig. 13. Samaria, central structure on the acropolis (reprinted by permission from Herzog 1992: fig. 14) (scale 1:750).

abutting the inner wall on the south and built as an independent casemate wall on the east and north (Building Period II) (J. W. Crowfoot, Kenyon, and Sukenik 1942: 5– 20). Pottery in the construction fills of the two phases of the podium dates to the Iron IIA, and, for the second, possibly to the Iron IIA–B transition (Crowfoot, Crowfoot, and Kenyon 1957: 94–134). Most of the acropolis consists of vast courtyards surrounded by several structures. On the west, on the extension provided by the construction of Podium II, is a series of at least three square “scribal chambers.” The Samaria Ostraca that list amounts of produce and place-names— probably tax-collection records—were found in this structure, attached to which are two very long halls of the type discussed above. In the northeastern sector are the disturbed remains of a series of elongated rooms, from the debris of which the Samaria ivories were recovered (Crowfoot and Crowfoot 1938; J. W. Crowfoot, Kenyon, and Sukenik 1942). The main buildings of the complex appear to be in the southwestern corner of the original podium (Podium I), and its core consists of a square building with rooms (either small and square or slightly larger and rectangular) arranged around a large, elongated central space. The core has a clear east–west symmetry that corresponds to its long axis. The other rooms and courtyards attached on the north, south, and east somewhat obscure this symmetry. The architecture of the entire complex, with its square-shaped main ashlar building arranged around a central space, vast open courtyards, and annexes, is typical of a LAP structure. Jezreel (fig. 4:b) Archaeological investigation of the Iron Age at Jezreel concentrated on the impressive fortification system that includes casemate walls, corner towers or bastions, a gatehouse, and a fosse with revetment walls separating the site from the background ridge. Ashlar construction was used liberally in the fortifications (Ussishkin and Woodhead

Fig. 14 (right). Lachish Podium A (reprinted by permission from Ussishkin 1978: fig. 7) (scale 1:750).

1992; 1994; 1997; Williamson 1996). Zimhoni (1997: 38– 39) dated the short occupation period of the complex to an advanced phase of Iron IIA (the transitional Iron IIA–B, in our opinion). 9 The structures at the top of the podium were poorly preserved, and no central building was uncovered. It is significant, however, that most of the excavation areas did not yield evidence for any architecture between the Medieval occupation and bedrock. Thus, with all due caution, we think that there was no town at Jezreel; instead, it comprised an isolated, largely empty, monumental podium/enclosure with only a few large structures. Parallels for this phenomenon may be found at Bethsaida (see below) and perhaps the Samaria acropolis. Lachish: Administrative Complex (figs. 1, 4:a, 14–16) The excavations at Lachish exposed a huge enclosure at the top of the tell with a series of massive podia at its highest point (first excavated by James L. Starkey and published by Tufnell [1953], subsequently probed by Aharoni [1975], and reexcavated by Ussishkin [1978; 1983; 2004]). These span Strata V–III. The initial core of the construction is the square Podium A (fig. 14). The area of the first podium was more than doubled with the addition of the elongated Podium B abutting its southern side (fig. 15). The final addition is a single wall abutting the eastern side of both Podia A and B (fig. 16), so that the resulting 9. The excavators view Jezreel as a central military base for the royal army during the Omride Dynasty (Ussishkin and Woodhead 1994: 47). Williamson emphasizes the propaganda value of the architecture, “intended . . . to overawe, if not intimidate, the local population” (1996: 49), a view shared by Finkelstein (2000: 132). Naªaman sees it as a royal residency and/or a royal administrative center (1997: 123), an interpretation much closer to our own.

Authority Ploys in Royal Architecture in the Iron Age Levant

157

Fig. 15. Lachish Podium B (reprinted by permission from Ussishkin 1978: fig. 7) (scale 1:750).

Fig. 16. Lachish Podium C (reprinted by permission from Ussishkin 1978: fig. 7) (scale 1:750).

Podium C is slightly wider than the first two podia. The only preserved floor level was in the last post–Iron Age palatial structure on top of the podia. 10 In the center of Podium A is a smaller square divided into two rows of three square cubicles flanking a central elongated space. Between this inner square and the eastern outer podium wall, a stairwell can be reconstructed. This is a classic LAP plan. Podium B consists of a square core the

same size as Podium A with an extension to the south about half as wide. It is crisscrossed by walls that divide it asymmetrically into long, narrow spaces. The extension of Podium B by the width of a single abutting wall (forming Podium C) does not make sense unless the buildings on top of Podia A and/or B were substantially rebuilt. It is possible that some of the foundation walls within Podium B are late additions that date to the Podium C renovations or even to the Persian period construction works. The only access from the enclosure courtyard to the top of the podia is via a solid stone ramp attached alongside the northeast corner of Podium C. Several phases were identified in the construction of this “porch,” and Ussishkin observed that at least the earliest phase predates Podium C (1978: 35). The parallels with Hazor, Megiddo Palace 1723 and Building 338, and Dan are obvious. A series of tripartite colonnaded (?) buildings (no actual columns survive) form the southern edge of the enclosure, with access to the enclosure through a six-chambered gatehouse set in between these structures. Long hallways—probably a long-hall storage complex—were built to the north of the podium (fig. 4).

10. The stratigraphic sequence of the three podia is: Podium A, late in Stratum V; Podium B, Stratum IV; and Podium C, Stratum III, although the possibility that all three podia were constructed shortly after one another within Stratum IV cannot be ruled out. The principal data used as the bases for dating these remains are: (a) that the enclosure wall abutting Podium B dovetailed with the western city wall of Strata IV–III; (b) that the fills under Podium A contained abundant Iron IIA potsherds (Tufnell 1953: 80; Zimhoni 1997: 170–74 ); and (c) that the last floor of the enclosure courtyard, which clearly runs up to Podium C from the east, must be dated to the Stratum III destruction associated with Sennacherib’s 701 bce campaign (Tufnell 1953: 86; Ussishkin 1977; 1978: 38). The Assyrians presumably destroyed the structure on top of the podia, after which the enclosure went out of use, although the podium itself was reused for the construction of the palatial residency in the Persian period (Ussishkin 1978: 41–42; erroneously attributed to the Assyrians by Amiran and Dunayevsky [1958] and Aharoni [1975: 33–40]).

158

Ilan Sharon and Anabel Zarzecki-Peleg

C

Fig. 17. Rosh Zayit: left: earlier fortress of Stratum IIb; right: later fortress of Stratum IIa (reprinted by permission from Gal and Alexandre 2000: plans 3–5) (scale 1:750).

Fig. 18. Jerusalem Ophel, early phase (reprinted by permission from Mazar and Mazar 1989: plan 7) (scale 1:750).

The Lachish complex is the largest podium structure thus far excavated. The corner access-ramp and the symmetrical core of Podium A—with rooms surrounding a wide central space, an attached stairwell, and a wide enclosure with subsidiary structures—are typical of the LAP structure architecture. One attribute, however, is missing at Lachish: no evidence was found for ashlar construction.

to the fortress was gained via a wooden ladder (Gal and Alexandre 2000: 17) does not seem reasonable. We suggest that a ramp ran along the southern wall and on top of the western extension. This ramp, the plan of the building, the ground floor storage cellars, 13 and the use of ashlar masonry in corners and doorways indicated that the Rosh Zayit fortress is of the LAP building type.

Rosh Zayit (figs. 1, 17) Rosh Zayit is a small site in the hills of the western Lower Galilee overlooking the Acco Bay. A fortress 11 (Stratum II) was built at the top of the site in the late Iron IIA, superseding a rural village (Stratum III), also of Iron IIA dating (Gal and Alexandre 2000: 12–22). The fortress was originally constructed as an almost square structure built partly of ashlars, with two small frontal towers on the west. At least three of the outside walls were buttressed by sloping revetment walls. 12 The basic plan of the building includes several small square rooms arranged to the north, east, and south of a larger, rectangular central space. All the rooms open into the central space. The excavators’ suggestion that access

Jerusalem: The Ophel (figs. 1, 18–19) This monumental complex was first excavated by Benjamin Mazar and reexcavated by Eilat Mazar (Mazar and Mazar 1989). Adjacent areas dug previously and excavated by Warren (1884: pl. 40) and Kenyon (1967: fig. 5) are probably parts of the same complex. 14 The only extant floor is the last one, dating to Iron IIC, which ended with the Babylonian destruction of Jerusalem (Mazar and Mazar 1989: 37). E. Mazar interpreted the remains to belong to three separate architectural units: Building D in the north, which she calls the “royal building”; Building C in the southwest, which she identifies as a gatehouse; and Building B—Warren’s “Great Tower”—an outer gate structure related to Building C (Mazar and Mazar 1989: plan 26). Another structure—Building A (Warren’s “Small Tower”)—abutted Building B. Building C is not a gatehouse, and Mazar’s

11. The excavators link the site to the biblical narrative of Solomon’s gift of “the Land of Cabul” to Hiram king of Tyre, suggesting that Rosh Zayit was “Cabul” itself. They offer several different interpretations of the site: first, that the fortress was a “fortified public storehouse,” “a well defended storage and administrative center” that served as the royal regional headquarters for the inner Akko plain and the hill zone. Alternatively, it could have been a Phoenician principality administering and collecting taxes in the “Land of Kabul” or a border fortress serving as a transfer point for agricultural goods en route to Phoenicia (Gal and Alexandre 2000: 8, 198–200). 12. The excavators postulate two occupation phases (Gal and Alexandre 2000: plans 3–5). In the later phase (Stratum IIa), the western side of the fortress was strengthened by another wall running parallel to and abutting both the original face of the wall and the sloping revetment and canceling out the northern tower. The division into two subphases is not necessary if we interpret the buttressing of podia with additional abutting walls as a construction technique (for a detailed critique of the stratigraphy of Rosh Zayit, see Zarzecki-Peleg 2005a: 283–91).

13. That the ground floor consisted of storage cellars is borne out by the excavators’ observation that small finds and cooking ware were found mainly in the collapse of the second story, while storage jars were predominant on the floors of the first (Gal and Alexandre 2000: 150). They estimate the total capacity of the storerooms and silos in the building at 25,000–40,000 liters (Gal and Alexandre 2000: 22, 198), which exceeds by far the needs of the inhabitants of the village of Rosh Zayit. 14. B. Mazar attempted to identify this structure as a Beit Millo, as described in 2 Kgs 12:20. The term Millo is used specifically in connection with Solomon’s building works in Jerusalem, and its Hebrew root may connote a filling or earth-moving operation (Mazar and Mazar 1989: ix). The suggestion that structures of the type under discussion are Batei Millo is attractive, although the dating of the Ophel structure to the end of the Iron Age precludes it from being the particular Beit Millo mentioned in the book of Kings.

Authority Ploys in Royal Architecture in the Iron Age Levant

159

Fig. 19 (left). Jerusalem Ophel, late phase (reprinted by permission from Mazar and Mazar 1989: plan 7) (scale 1:750). Fig. 20 (middle). Dor “Monumental Building” (scale 1:500). Fig. 21 (right). Bethsaida bit hilani (reprinted by permission from Arav and Bernett 2000: fig. 2) (scale 1:500).

hypothesis that Building B comprised the supporting walls of a ramp leading into a city gate is unlikely (see Sharon 1996: 309–25). 15 We suggest that Building C and the southwestern half of Building D belong to a single building, representing the eastern and northern edges of a monumental structure with an entryway in its southwestern corner, a structure that was extended laterally by adding new wings—the northeastern half of Building D and Buildings A and B. Extending a podium by progressively attaching abutting podia to the original core is typical of LAP structure architecture, as are the corner access and the use of ashlars (as in Warren’s “Great Tower”). Dor: The “Monumental Building” (figs. 1, 20) The corner of a massive Iron Age structure, partly eroded into the southern bay of Dor (Area D2), was first identified by Raban (1987) and excavated by Stern (2000: 353–56). The remains consist of what was probably a rectangular structure, built of boulders, with ashlars in the northwestern corner. The excavators distinguished two construction phases, with several superimposed floor levels (Area D2 Phases 10 [Iron IB]–8a [Iron IIA] or even 7 [Iron IIB]). Two walls (also made of boulders) that abut this structure may be part of an enclosure of the type under discussion. In the space circumscribed by these walls and the “Monumental Building,” a mud-brick building comprising long storage cellars was built (Phases 10–9). Sub15. The evidence for a symmetrical, four-room gate is insubstantial. On the contrary, the foundation walls indicate that the structure was not axially symmetrical. Also, there are no parallels for a corner gatehouse of the type posited by the excavators anywhere else in the Iron Age Levant. On the other hand, the corner access is the rule for the LAP plan.

sequently, most of it was replaced by an open courtyard (Phase 8c–a). 16 Features that may link this structure to the LAP group are its massive construction, including the use of ashlars, and its enclosure and subsidiary storage facilities. Bethsaida: The Bit Hilani (figs. 1, 21) The excavations at Bethsaida exposed the remains of an imposing complex dating to the Iron IIA and IIB (Strata 6 and 5)—the capital of the kingdom of Geshur, according to the excavators (Arav and Bernett 2000, with earlier bibliography). The first building operations involved a series of defensive and terrace walls around the hill. To the north of the monumental gatehouse and backed against the city wall is the only other major structure that the excavators attribute to the Iron Age in the area thus far excavated. They identify it as a bit hilani. Its basic plan consists of eight rooms arranged around a central courtyard or hall. Most of these are square, but in the southeastern corner there is a double-width room that the excavators reconstruct as a typical bit hilani vestibule. 17

16. If this complex at Dor indeed belongs to the LAP structure group, it is the only example from Iron I. This fact, taken together with its attribution to Phoenician builders (Stern 2000: 353–57; Sharon and Gilboa 1997), may point to an origin for this type of construction in the southern Levantine cultural sphere. 17. The excavators attribute extraordinary longevity to the structure as a whole, from Iron IIA to the Hellenistic period, with various structural changes along the way (Arav and Bernett 2000: 52, 81). Late artifacts were found all the way down to bedrock in the building and around it. The only exception is one room, in which an assemblage of Iron Age pottery was found that the excavators used to date the first phase of the building.

160 The plan, however, shows strong symmetry, especially if one interprets the vestibule in the southeastern corner as actually two square rooms with the partition between them missing. 18 The building at Bethsaida is apparently not a podium structure—it is more elongated than any of the LAP buildings mentioned here it is not as massive in construction, and it has no ashlar masonry. Thus, it certainly is not a typical example of either the bit hilani or the LAP group. However, the general aspect of the site as a whole—a small, heavily fortified citadel with apparently only a few large buildings in it—is reminiscent of the Jezreel enclosure and perhaps the Samaria acropolis. Tell el-Kheleifeh (figs. 1, 22) This small site on the estuary of Wadi Arabah that runs into the Gulf of Aqabah was excavated by Glueck (1938; 1939; 1940; comprehensively published by Pratico [1993]). The main structure (Strata I–IV, according to the excavator 19) is an almost square, thick-walled, mud-brick building, consisting of three small square rooms on the north with three elongated halls to their south. Both side halls were subdivided (in a late phase of the structure, according to Glueck). The most significant features, in our opinion, are the mud-brick glacis/revetments abutting the exterior walls of this building on three sides. These, together with the fact that no doorways were found, mean that the ground floor consisted of cellars. The access to the building is unclear, but it is noteworthy that “Room 19” may have been a stairwell. 18. In the eastern part of the building, the bedrock is quite close to the surface, and only one course of the walls was preserved. It is therefore impossible to determine whether sections where no walls were found are indeed wide doorways or whether the walls are in fact missing. 19. Glueck attributed the construction of this building to Stratum IA (which he suggested was Solomonic Ezion-Geber), the revetment to Stratum IB, and the square casemate wall to Stratum IC. In Stratum II, a much wider square of solid offset-inset fortification with a large, fourchambered gate was built over both the casemate wall and the eastern wall of the building. Nevertheless, both continued to function, according to Glueck, now abutting the new fortification. Only in Strata III and IV were both enclosures gradually filled with residential structures. Pratico (1993: 25) points out that the addition of revetments to the building was a construction phase, and there is no way to ascertain the stratigraphical order of that construction vis-à-vis the casemate wall. Furthermore, there is no reason to suppose that the outer offset-inset enclosure was empty for a long time, and it is impossible to determine the progression or phasing of construction within the new fortifications (Pratico 1993: 27, 34). Therefore, according to Pratico, only two phases can be distinguished: a square casemated fortress with the square structure in its center (Glueck’s Stratum I) and a fortified settlement (Glueck’s Strata II–IVA). We would add that it is not clear whether Glueck and Pinkerfeld (his architect) ever actually saw the eastern offset-inset wall superimposed on the eastern wall and revetment of the square structure or only assumed that they were. If the latter, the prima facie explanation should be that the casemate wall is structurally later than the offset-inset wall, which it appears to abut. In that case, the square building would belong only to Stratum II and later.

Ilan Sharon and Anabel Zarzecki-Peleg Glueck accepted the biblical descriptions of EzionGeber/Elath at face value: as a way-station during the exodus, as the headquarters of King Solomon’s mines, and as a port for Judean expeditions to Tarshish that was taken over by the Edomites only at the end of the 8th century. He first suggested that the square, thick-walled building was used for copper smelting but later (Glueck 1965: 73–76) accepted Rothenberg’s (1962: 44–56) evaluation that there is no evidence for copper smelting of any magnitude at Tell el-Kheleifeh. Glueck then proposed the theory that the building was a granary. Pratico calls the structure a monumental four-room building, citing the Hazor citadel as a parallel. The comparison of Tell el-Kheleifeh with Rosh Zayit (Gal and Alexandre 2000: 198) is also compelling and, although we agree with these comparisons, we do not believe that any of these buildings is a “four-room house.” The similarities to the LAP structures as defined here lie in the almostsquare plan, the massive walls, the series of small square chambers, the fact that the main space is a central courtyard/hallway, and, most importantly, the fact that the ground floor consists of cellars to which the only access was from above, while the entrance to the building must have been via the second story. Given that the square building was apparently in use—and may have been constructed—in Iron Age IIB–C, the possibility exists that this is an Edomite version of a LAP structure. 20

Authority Ploys: The Architecture and Politics of Distance and Restricted Access The establishment and maintenance of authority has much to do with the division of social space, the maintaining of proper distances within these divisions, and the limiting and controlling of access between them. Architecture, most broadly defined as the methods by which people construct durable divisions of their physical space, is therefore a primary way in which authority is symbolized in a wide variety of cultures. This is exemplified in epithets such as “the Pharaoh,” “the Pentagon,” or “No. 10 Downing Street,” naming the administration after the 20. Pratico notes that, while most of the architectural parallels are with sites west of the Jordan rift (a more-thoroughly investigated region), the ceramic and epigraphic evidence points to the Transjordanian highlands as the cultural sphere of the site. The non-local wheelmade pottery is by-and-large of eastern, rather than Judean, derivation and dated to Iron IIB–C, rather than to Iron IIA, much less Iron I, as dated by Glueck). The local Negebite pottery is prevalent in the Negev and the southern Transjordanian highlands until the end of the Iron Age (Cohen 1979: 77). Nevertheless, Pratico does not attribute the Tell elKheleifeh fortress to any specific ethnic group or political entity (Pratico 1993: 21, 72).

Authority Ploys in Royal Architecture in the Iron Age Levant

161

Fig. 22. Tell el-Kheleifeh enclosure (reprinted by permission from Pratico 1993: pl. 4) (scale 1:500).

structure that houses it, or indeed Sublime Porte, a term in which the door of a palace is used as a metaphor for the empire itself. The notion that architecture, whatever its primary purpose, is also meant to “make a statement” goes without saying, and maxim that “palatial” architecture is intended to “legitimize authority” has also been repeated so often that we no longer stop to consider how exactly a structure does this (for Iron Age administrative complexes in the Levant, see, inter alia Williamson 1996; Finkelstein 2000: 130–33). The implicit assumptions seem to be that the sheer size of a building can cow the populace into submission, that ornate decoration can persuade an observer of the opulence and munificence of the administration, and/or that a building of such magnitude constitutes proof of the ability of the ruler to command and organize the necessary logistics. These explanations do not take into consideration the reasons that one type of building is chosen as a symbol of authority over another. Although we do not deny that rulers are prone to use a show of strength to instill subordination, we maintain that more subtle ploys are also evident in the design of administrative structures. Different plans of administrative complexes may convey different strategies for empowering and legitimizing authority. The bit hilani and the LAP structure embody two different authority ploys—and both are designed to enable an authority to hold the metaphorical and physical high ground at the point of contact between administrator and administered.

The following analogy is based on Van Loon’s introduction to art history (1938: 202–3). Figure 23 shows a typical Baronial residence in 13th–15th-century Europe and fig. 24 the equivalent in the 17th–19th centuries. In the first case, the castle is perched high above the countryside, dominating it by the inaccessibility of its sheer walls. The second case seems just the opposite: the focus of the building is its wide approach, its embracing, open facade, and its ornate, prominent doorway. It could easily be assumed that the seemingly open and inviting appearance of the latter structure represents a less-authoritative social order, and this misconception may be the gist of the authority ploy it embodies. Defying its inviting appearance is its physical separation and distance from other buildings. It dominates its surroundings and dwarfs anyone approaching it, reserving the high ground for the proprietor, who waits at the door, which frames him in the center of his small “designer universe” like a benevolent lord. Indeed, most of the people who went in and out of houses of this sort would not use the ornate front doorway; tenants, servants, and tradesmen would use the back entrances. Thus, the welcoming but inaccessible doorway is an architectural metaphor for an ideology of protecting a higher social class against incursions by a lower one and at the same time of attempting to penetrate social circles of higher status. Leone (1984) has demonstrated the subtle “subploys” used by the architects to magnify these effects, creating optical illusions to make the building appear farther away

Ilan Sharon and Anabel Zarzecki-Peleg

162

Fig. 23. Typical Medieval castle representation (after Van Loon 1938: 202).

Fig. 24. Typical Renaissance palace representation (after Van Loon 1938: 203).

and higher up than it really is. He argues that these structures, and the formal gardens around them, are models of the world view of their owners/builders and bolster the social structure they embody. There are, of course, functional differences between these types of buildings as well, dictated by historical circumstances, but the general analogy applies to our discussion.

not a functional necessity. Officialdom had more space and, perhaps equally significantly, was seen to have more space than the rest of the population, which lived in rather cramped conditions in an Iron Age town. Significantly, size as an authoritarian feature does not seem to have been inherited from Bronze Age Canaanite architecture. Although the palace complex at Hazor, which represented by far the largest polity in the southern Levant in the Late Bronze Age, is the grandest structure known in the region in that period, it does not occupy a large proportion of the city as a whole. By comparison, the Iron Age city at Hazor is a fraction of the size of its predecessor, but the space accorded to officialdom is proportionally considerably greater. The allocation of space can exemplify the interplay between the overt and the symbolic use of power and privilege to promote authority. We contend that authority invested in appearances. Late Bronze Age “residencies” are not usually located in a large open space or cordoned off from the populace (Oren 1992: 117–20, figs. 17–23). In the Late Bronze Age palace at Megiddo, with a built-up

Authority Cast in Stone: Ploys Used in Levantine Iron Age Palatial Architecture Perhaps the most obvious architectural power play is the appropriation of space. An obvious hierarchy is reflected in the size of the administrative enclosures. Capital cities/royal residencies such as Samaria, Jezreel, and presumably Jerusalem had palatial enclosures. As much as one-third of the walled city was enclosed for official use even in secondary administrative centers such as Megiddo and Lachish. The appropriation of such large spaces was

Authority Ploys in Royal Architecture in the Iron Age Levant area probably significantly larger than the area of most Iron Age LAP structures, the administration “shared” its space with the administered. The palace was built contiguously with the city gate and was omnipresent. The favored ploy for authority seems to have been to sit on the main artery of the town, to be on top, as it were, of its comings and goings. In the Iron Age, however, authority elected to appear to the administered as spacious, independent, detached, distant, and unassailable. Structurally, most of the LAP buildings seem to be distinguished by the use of ashlar architecture. Architectural ornamentation is apparently limited to the use of protoAeolic capitals, either on pilasters or on free-standing columns (although none has yet been found in situ). Albeit rarely, column-bases, windowsill ornamentation, and crenellations were also employed (for the correlation between these decorative elements and Phoenician/Israelite ashlar tradition, see Shiloh 1979a; see also Stern 1977; 1998). In Iron Age Israel, this type of building technique and associated ornamentation seem to be restricted to public architecture. On the whole, however, the available evidence indicates that displays of opulence in the structures under discussion were subdued. The Syrian bit hilani style was more prone to monumental displays of visual propaganda. The most obvious ploy used in LAP building construction is the domination of the surroundings by means of elevating the main structure, visibly detaching it even further by the use of a long narrow ramp and indirect access. We assume that the indirect access, in contrast to the frontal colonnaded porch entrance of the Syrian bit hilani, was an important part of the symbolism these buildings were meant to convey. The building types represent two different ploys for limiting the access of the administered and maintaining the authority of the administrator. One is the gradual “inductive” approach, by which the structure appears to be welcoming, but the individual is then humbled by the grandeur of the portals and the stature of the symbolic gate-keepers. In the other, authority is visibly circumscribed, presenting a blank wall to the administered and, at the same time, providing a way around the side by means of which the properly circumspect may access it. Although there are straightforward functional reasons for the circumscription of authority, these do not explain the differences between the two ploys. No doubt the physical, as well as the metaphysical, protection of authority was taken into consideration by the architects of the LAP structures, which have often been described as forts or citadels, and may well have served as forts or citadels in times of war or civil turmoil. But the preference for the circumscription ploy should not necessarily be taken to mean that authority in the Iron Age Levant was constantly threatened and therefore needed to seclude and protect it-

163

self (cf., e.g., Williamson 1996; Finkelstein 2000). Indeed, we find no reason to suppose that the kings of Israel, Judah, or Edom—or their provincial governors—had any less legitimacy or more need of physical protection than their Late Bronze Age predecessors or their contemporaries in northern Syria. Function Domestic activities in the LAP structures perforce involved a small number of inhabitants, although evidence is lacking from capital cities in which kings, and presumably their courts, were in permanent residence (Jerusalem, Samaria, and perhaps Jezreel). It seems that most of these administrative complexes had some connection to the collection, storage, and possibly redistribution of bulk goods. In many cases, there is evidence for some sort of supralocal redistribution, either as taxation or allocation or both. The complexes located in central places, for example, Lachish and Megiddo, can be interpreted as regional logistical centers; while others, such as Rosh Zayit and Tell elKheleifeh, are small border sites that may have served as entrepôts for foreign trade or as customs depots. At Megiddo, the complex is connected to the stabling of horses (and presumably the quartering of troops as well). There is evidence of scribal/archival activities in one case (Samaria), and in at least one complex (Dan), there are indications of cultic activities. Can we then generalize at all on a single function of the LAP building or administrative complex as a whole? This difficulty is also inherent in the interpretation of the function of other public structures in the Iron Age Levant (such as tripartite colonnaded buildings). Perhaps public architecture was simply not unifunctional. We suggest that the ploy is one of the intrinsic functions, if not the essence, of the LAP structure and the enclosure around it. Its purpose is to embody authority. This, in our opinion, is the only reasonable explanation for two other peculiarities of building complexes of this sort. The first is thus far evident only at Megiddo—namely, the presence of more than one complex at a site both in Stratum IVB–VA (Palace 6000 in the north and Palace 1723 in the south) and in Stratum IVA (the northern stable complex connected to Building 338 and the southern complex linked to the unidentified structure above Palace 1723). The second is the ubiquitous phenomenon of extending the podium. Often, as in the case of Lachish Podium C or Samaria Podium II, the extension is a rather minor addition to the total area of the podium that may have necessitated extensive rebuilding of the structure/s on top of it. If the LAP structure signified kingship (or other types of authority [see below]), it would be equally true that it was kingly to construct LAP buildings whether or not there was a functional need for them. That large-scale construction is a

164 kingly quality is as true of ancient Israel (e.g., 1 Kings 7, 15:22–23, 16:24, 22:39; 2 Kgs 14:22) as of any other nation in the ancient Near East, regardless of whether these biblical citations represent actual royal constructions, empty boasts, or late glosses. Authority with a Capital A Most scholars view these structures as evidence of a central royal administration and we concur. Several aspects seem to indicate a supra-local and probably even supra-regional hierarchy: (a) not every town had such enclosures; (b) in the towns that did, the size of the enclosures and the complexity of the structures within them differ greatly; (c) these enclosures apparently had different functional foci in different towns; and (d) in some cases, they are not actually located in a town. Assuming that the hierarchical and functional differences do reflect a state-level administration, there is still considerable leeway in deciding who occupied these enclosures and “held court” in the LAP buildings. There are numerous biblical and epigraphic references to royal staff positions, that is, to functionaries who presumably were resident in the capital and were responsible to the king for a specific aspect of administration. But we know little about how administration worked outside the capital. Theoretically, it is possible that the “royal enclosures” were built by the king for his own use and that they functioned as royal estates. On the other hand, one might posit that the structures were built and/or used by a local aristocracy that administered a town or region on behalf of the central royal administration. Jezreel is an especially pertinent example. Its role may be encapsulated in one biblical verse: “Naboth the Jezreelite owned a vineyard in Jezreel, adjoining the palace of King Ahab of Samaria” (1 Kgs 21:1). From this text, one may surmise that the king, whose primary capital was Samaria, had a secondary palace in another town. The presence of vineyards next to the palace implies that Jezreel was an administrative center in a rural setting. This may suggest the first type of administration (royal estate). However, other than references to the Omrides and Jezreel, there is no biblical reference to a king’s residing away from his primary capital for any length of time. The archaeological evidence seems to indicate that the smaller administrative complexes lacked the facilities required to house extensive staff and retinues. The biblical narrative about the fate of Naboth also relates that, in order to confiscate his vineyard, Jezebel needed to convince “the elders and the nobles [?] who lived in the same town with Naboth” to accuse him of heresy and treason and to execute him (1 Kgs 21:8–14). This implies a dual hierarchy of administration: in the town in

Ilan Sharon and Anabel Zarzecki-Peleg which the king had a secondary palace, there was also a local elite that had at least nominal judicial independence. Where the local elite and population lived if Jezreel consisted only of a single palatial enclosure is a moot point. It is possible that the reference to the “city” included the countryside or that the writer either was not very familiar with the nonurban topography of Jezreel or did not want to bother with the details. While there are numerous biblical references to elders and nobles (see the extensive discussion in Schloen 2001), there seems to be no indication that this “nobility” held supra-local authority such as the authority implied by the administrative complexes. We, too, therefore, opt for the interpretation that these “palatial” structures were the strongholds of “governors” appointed by the king to watch over royal interests (primarily, according to the biblical text, tax-collection and provisioning, but possibly other centralized activities as well). It should be noted, however, that references to an appointed provincial administration are rare, apart from references that relate to the Solomonic provinces, which may well represent a later gloss. The possibility that this particular configuration of LAP structures and associated enclosures is indicative of state formation (as opposed to the Bronze Age city-state system) is also supported by a diachronic examination of the process of its appearance and development (for details, see Zarzecki-Peleg 2005b). In many cases, the structures and enclosures first appear as part of a general restructuring of the town plan that signifies a completely new urban concept, changing the focus from largely residential to public/administrative. The establishment of the administrative enclosure is often accompanied by the refortification of the city as a whole, the construction of monumental citygates, the creation of water systems, and the pushing of the general populace to the edge of (and sometimes completely off) the tell. These construction projects bespeak a supra-local authority in terms of the logistics involved, their function, their blatant disregard of the previous plan, land ownership, and the residential needs of the locals, and—most of all—the ideological message conveyed. The development of the enclosure construction process, where it is possible to trace it, is also significant. It typically starts toward the end of Iron IIA, after the initial nucleation of the town. In several cases (at Megiddo, Lachish, and perhaps Dan), it seems that there was an initial attempt to build a single LAP structure within the existing urban framework, which was shortly followed (during the Iron IIA/B transition) by the reorganization of the city as a whole and the rebuilding of the LAP structure itself. The phenomenon of royal enclosures and LAP-type “palaces” in major towns and liminal entrepôts was well established by the beginning of Iron IIB.

Authority Ploys in Royal Architecture in the Iron Age Levant

165

References Aharoni, Y. 1975 Lachish V: Investigations at Lachish—The Sanctuary and the Residency. Publications of the Institute of Archaeology 4. Tel Aviv: Tel Aviv University Press. Amiran, R., and Dunayevsky, I. 1958 The Assyrian Open-Court Building and Its Palestinian Derivatives. Bulletin of the American Schools of Oriental Research 149: 25–32. Arav, R., and Bernett, M. 2000 The Bit Hilani at Bethsaida: Its Place in Aramaean/Neo-Hittite and Israelite Palace Architecture in the Iron Age II. Israel Exploration Journal 50: 47–81. Barkay, G. 1992 The Iron Age II–III. Pp. 302–73 in The Archaeology of Ancient Israel, ed. A. Ben-Tor. Trans. R. Greenberg. New Haven: Yale University Press. Ben-Tor, A. 1993 Tel Hazor (Notes and News). Israel Exploration Journal 43: 253–56. 1999 Excavating Hazor, Part One: Solomon’s City Rises from the Ashes. Biblical Archaeology Review 25: 26–37, 60. Biran, A. 1971 Laish-Dan: Secrets of a Canaanite City and an Israelite City. Qadmoniot 4/13: 2–11 (Hebrew). 1982 The Temenos at Dan. Eretz-Israel 16 (Harry M. Orlinksy volume): 15–43, 252*–53* (Hebrew, with English abstract). 1994 Biblical Dan. Jerusalem: Israel Exploration Society. 1996 A Chronicle of the Excavations, 1966–1992. Pp. 7–62 in Dan I, by A. Biran, D. Ilan, and R. Greenberg. Jerusalem: Hebrew Union College. 2002 A Chronicle of the Excavations. Pp. 3–32 in Dan II, by A. Biran and R. Ben-Dov. Jerusalem: Hebrew Union College. Bonfil, R. 1997 Middle Bronze Age to Persian Period, Area A. Pp. 25–176 in Hazor V: An Account of the Fifth Season of Excavation, 1968, by A. Ben-Tor et al. Jerusalem: Israel Exploration Society. Cohen, R. 1979 The Iron Age Fortresses in the Central Negev. Bulletin of the American Schools of Oriental Research 236: 61–79. Crowfoot, J. W., and Crowfoot, G. M. 1938 Early Ivories from Samaria. Vol. 2 of Samaria–Sebaste: Reports of the Work of the Joint Expedition in 1931–1933 and the British Expedition in 1935. London: Palestine Exploration Fund. Crowfoot, J. W.; Crowfoot, G. M.; and Kenyon. K. M. 1957 The Objects from Samaria. Vol. 3 of Samaria-Sebaste: Reports of the Work of the Joint Expedition in 1931–1933 and the British Expedition in 1935. London: Palestine Exploration Fund. Crowfoot, J. W.; Kenyon, K. M.; and Sukenik, E. L. 1942 The Buildings at Samaria. Vol. 1 of Samaria-Sebaste: Reports of the Work of the Joint Expedition in 1931–1933 and the British Expedition in 1935. London: Palestine Exploration Fund. Currid, J. D. 1991 The Re-Stratification of Megiddo during the United Monarchy. Zeitschrift des Deutschen Palästina-Vereins 107: 28–38. Dever, W. G. 1990 Recent Archaeological Discoveries and Biblical Research. Seattle: University of Washington Press. Elayi, J. 1996 Nouveaux Eléménts sur le Mur à Piliers Phénicien Transeuphratène 11: 77–94.

Finkelstein, I. 2000 Omride Architecture. Zeitschrift des Deutschen PalästinaVereins 116: 114–38. Fisher, C. 1929 The Excavation of Armageddon. Oriental Institute Communications 4. Chicago: University of Chicago Press. Frankfort, H. 1952 The Origin of the “Bit Hilani.” Iraq 14: 120–31. Fritz, V. 1983a Die syrische Bauform des Hilani und die Frageseiner Verbreitung. Damaszener Mitteilungen 1: 43–58. 1983b Paläste während der Bronze- und Eisenzeit in Palästina. Zeitschrift des Deutschen Palästina-Vereins 99: 1–42. 1995 The City in Ancient Israel. Biblical Seminar 29. Sheffield: Sheffield Academic Press. Gal, Z., and Alexandre, Y. 2000 Óorbat Rosh Zayit: An Iron Age Fort and Village. Israel Antiquities Authority Reports 8. Jerusalem: Israel Antiquities Authority. Galling, K. 1937 Biblisches Reallexicon. Handbuch zum Alten Testament 1. Tübingen: Mohr. Glueck, N. 1938 The First Campaign at Tell El-Kheleifeh (Ezion-Geber). Bulletin of the American Schools of Oriental Research 71: 2–18. 1939 The Second Campaign at Tell El-Kheleifeh (Ezion-Geber: Elath). Bulletin of the American Schools of Oriental Research 75: 8–22. 1940 The Third Season of Excavation at Tell El-Kheleifeh. Bulletin of the American Schools of Oriental Research 79: 2–18. 1965 Ezion-Geber. Biblical Archaeologist 28: 70–87. Haines, R. C. 1971 Excavations in the Plain of Antioch, vol. 2: The Structural Remains of the Later Phases. Oriental Institute Publications 95. Chicago: University of Chicago Press. Herr, L. G. 1988 Tripartite Pillared Buildings and the Market Place in Iron Age Palestine. Bulletin of the American Schools of Oriental Research 272: 47–67. Herzog, Z. 1992 Settlement and Fortification Planning in the Iron Age. Pp. 231– 74 in The Architecture of Ancient Israel, ed. A. Kempinski and R. Reich. Jerusalem: Israel Exploration Society. 1997 Archaeology of the City: Urban Planning in Ancient Israel and Its Social Implications. Tel Aviv: Tel Aviv University Press. Holladay, J. S. 1986 The Stables of Ancient Israel. Pp 103–95 in The Archaeology of Jordan and Other Studies Presented to Siegfried H. Horn, ed. L. T. Geraty and L. G. Herr. Berrien Springs, MI: Andrews University Press. Kempinski, A. 1989 Megiddo: A City-State and Royal Centre in North Israel. Allgemeinen und Vergleichenden Archäologie 40. Munich: Beck. Kenyon, K. M. 1942 The Summit Buildings and Constructions. Pp. 91–139 in The Buildings at Samaria, by J. W. Crowfoot, K. M. Kenyon, and E. L. Sukenik. Vol. 1 of Samaria-Sebaste: Reports of the Work of the Joint Expedition in 1931–1933 and the British Expedition in 1935. London: Palestine Exploration Fund. 1957 Israelite Pottery: Stratified Groups. The Evidence of Samaria Pottery and Its Bearing on Finds at Others Sites. Pp. 94–134,

166 198–209 in The Objects from Samaria, by J. W. Crowfoot, G. M. Crowfoot, and K. M. Kenyon. Vol. 3 of Samaria-Sebaste: Reports of the Work of the Joint Expedition in 1931–1933 and the British Expedition in 1935. London: Palestine Exploration Fund. 1964 Megiddo, Hazor, Samaria and Chronology. Bulletin of the Institute of Archaeology 4: 143–56. 1967 Jerusalem: Excavating 3000 Years of History. London: Thames & Hudson. 1971 Royal Cities of the Old Testament. New York: Schocken. Lamon, R., and Shipton, G. 1939 Megiddo I: Seasons of 1925–34, Strata I–V. Oriental Institute Publications 42. Chicago: University of Chicago Press. Leone, M. P. 1984 Interpreting Ideology in Historical Archaeology: Using the Rules of Perspective in the William Paca Garden in Annapolis, Maryland. Pp. 25–35 in Ideology, Power and Prehistory, ed. D. Miller and C. Tilley. New Directions in Archaeology. Cambridge: Cambridge University Press. Macalister, R. A. S. 1912 The Excavation of Gezer. London: Palestine Exploration Fund. Margueron, J.-C. 1979 Un “hilani” à Emar. Pp. 153–76 in Archaeological Reports from the Tabqa Dam Project: Euphrates Valley, Syria, ed. D. N. Freedman. Annual of the American Schools of Oriental Research 44. Cambridge, MA: American Schools of Oriental Research. Mazar A. 1990 Archaeology of the Land of the Bible 10,000–586 b.c.e. The Anchor Bible Reference Library. New York: Doubleday. Mazar, E., and Mazar, B. 1989 Excavations in the South of the Temple Mount: The Ophel of Biblical Jerusalem. Qedem 29. Jerusalem: Hebrew University. McClellan, T. L. 1975 Quantitative Studies in the Iron Age Pottery of Palestine. Unpublished Ph.D. dissertation, University of Pennsylvania. Naªaman, N. 1997 Historical and Literary Notes on the Excavation of Tel Jezreel. Tel Aviv 24: 122–28. Nigro, L. 1995 Ricerche sull’ architettura palaziale della Palestina nelle età del Bronzo e del Ferro. Contributi e materiali di archaeologia orientale 5 (1994). Rome: Università degli Studi di Roma “La Sapienza.” Oren, E. D. 1992 Palaces and Patrician Houses in the Middle and Late Bronze Ages. Pp. 105–20 in The Architecture of Ancient Israel, ed. A. Kempinski and R. Reich. Jerusalem: Israel Exploration Society. Pratico, G. D. 1993 Nelson Glueck’s 1938–1940 Excavations at Tell el-Kheleifeh. Atlanta: Scholars Press. Pritchard, J. B. 1970 The Megiddo Stables: A Re-Assessment. Pp. 268–76 in Near Eastern Archaeology in the 20th Century: Essays in Honor of Nelson Glueck, ed. J. A. Sanders. Garden City, NY: Doubleday. Raban, A. 1987 The Harbor of the Sea Peoples at Dor. Biblical Archaeologist 50: 118–26. Reich, R. 1992 Palaces and Residencies in the Iron Age. Pp. 202–22 in The Architecture of Ancient Israel, ed. A. Kempinski and R. Reich. Jerusalem: Israel Exploration Society. Reisner, G. A.; Fisher, C. S.; and Lyon, D. G. 1924 Harvard Excavations at Samaria, 1908–1910. Vol. 1. Cambridge: Harvard University Press.

Ilan Sharon and Anabel Zarzecki-Peleg Rothenberg, B. 1962 Ancient Copper Industries in the Western Arabah. Palestine Exploration Quarterly 94: 5–71. Schumacher, G. 1908 Tell El-Mutesellim I: Text und Tafeln. Leipzig: Haupt. Schloen, J. D. 2001 The House of the Father as Fact and Symbol: Patrimonialism in Ugarit and the Ancient Near East. Studies in the Archaeology and History of the Levant 1. Winona Lake, IN: Eisenbrauns. Sharon, I. 1987 Phoenician and Greek Ashlar Construction Techniques at Tel Dor, Israel. Bulletin of the American Schools of Oriental Research 267: 21–41. 1996 Architectural Stratigraphy. Unpublished Ph.D. dissertation, Hebrew University of Jerusalem. Sharon, I., and Gilboa, A. 1997 Dor in the Iron I Period: A Port and Trading Emporium under Cultural and Economic Changes. Pp. 12–33 in New Studies on the Coastal Plain, ed. E. Regev. Ramat Gan: Bar-Ilan University Press (Hebrew). Shiloh, Y. 1979a The Proto-Aeolic Capital and Israelite Ashlar Masonary. Qedem 11. Jerusalem: Hebrew University. 1979b Iron Age Sanctuaries and Cult Elements in Palestine. Pp. 147– 57 in Symposia Celebrating the Seventy-Fifth Anniversary of the American Schools of Oriental Research, ed. F. M. Cross. Cambridge, MA: American Schools of Oriental Research. Singer, I. 1975 Hittite Hilammar and Hieroglyphic Luwian Hiliana. Zeitschrift für Archäologie 65: 69–103. Stager, L. E. 1990 Shemer’s Estate. Bulletin of the American Schools of Oriental Research 277–278: 93–107. Stager, L. E., and Wolff, S. R. 1981 Production and Commerce in Temple Courtyards: An Olive Press in the Sacred Precinct at Tel Dan. Bulletin of the American Schools of Oriental Research 243: 95–102. Stern, E. 1977 The Excavations at Tell Mevorach and the Late Phoenician Elements in the Architecture of Palestine. Bulletin of the American Schools of Oriental Research 225: 17–27. 1998 New Phoenician Elements in the Architecture of Tel Dor, Israel. Pp. 373–88 in Óesed ve-Emet: Studies in Honor of Ernest S. Frerichs, ed. J. Magness and S. Gitin. Brown Judaic Studies 320. Atlanta, GA: Scholars Press. 2000 Dor—Ruler of the Seas: Nineteen Years of Excavation at the Israelite-Phoenician Harbor Town on the Carmel Coast. Jerusalem: Israel Exploration Society. Tappy, R. 1992 The Archaeology of Israelite Samaria, vol. 1: Early Iron Age through the Ninth Century b.c.e. Harvard Semitic Studies 44. Atlanta, GA: Scholars Press. 2001 The Archaeology of Israelite Samaria, vol. 2: The Eighth Century b.c.e. Harvard Semitic Studies 50. Winona Lake, IN: Eisenbrauns. Tufnell, O. 1953 Lachish III: The Iron Age. London: Oxford University Press. Ussishkin, D. 1966 King Solomon’s Palace and Building 1723 in Megiddo. Israel Exploration Journal 16: 174–86. 1973 King Solomon’s Palaces. Biblical Archaeologist 36: 78–105. 1977 The Destruction of Lachish by Sennacherib and the Dating of the Royal Judean Storage Jars. Tel Aviv 4: 28–60. 1978 Excavations at Tel Lachish, 1973–1977, Preliminary Report. Tel Aviv 5: 1–97.

Authority Ploys in Royal Architecture in the Iron Age Levant 1983

Excavations at Tel Lachish, 1978–1983: Second Preliminary Report. Tel Aviv 10: 97–175. 1989 Schumacher’s Shrine in Building 338 at Megiddo. Israel Exploration Journal 39: 149–72. 2004 The Renewed Archaeological Excavations at Lachish (1973– 1994). Institute of Archaeology Monograph Series 22. Tel Aviv: Tel Aviv University Press. Ussishkin, D., and Woodhead, J. 1992 Excavations at Tel Jezreel, 1990–1991: Preliminary Report. Tel Aviv 19: 3–56. 1994 Excavations at Tel Jezreel, 1992–1993: Second Preliminary Report. Levant 26: 1–48. 1997 Excavations at Tel Jezreel, 1994–1996: Third Preliminary Report. Tel Aviv 24: 6–72. Van Loon, H. W. 1938 The Arts of Mankind. London: Harrap. Warren, C. 1884 The Survey of Western Palestine: Jerusalem. London: Palestine Exploration Fund. Watzinger, C. 1933 Denkmäler Palästinas: Eine Einfürung in die Archäologie des Heiligen Landes. Vol. 1. Leipzig: Hinrichs. Williamson, H. G. M. 1996 Tel Jezreel and the Dynasty of Omri. Palestine Exploration Quarterly 128: 41–51. Wright, G. R. H. 1985 Ancient Building in South Syria and Palestine. Leiden: Brill. Yadin, Y. 1959 Excavations at Hazor, 1958: Preliminary Communiqué. Israel Exploration Journal 9: 74–88. 1966 Megiddo (Notes and News). Israel Exploration Journal 16: 278–80. 1967 Megiddo (Notes and News). Israel Exploration Journal 17: 199–21.

167

1970 The Megiddo of the Kings of Israel. Qadmoniot 10: 38–56 (Hebrew). 1972 Hazor: The Head of All Those Kingdoms. The Schweich Lectures of the British Academy 1970. London: Oxford University Press. 1976 The Megiddo Stables. Pp. 249–52 in Magnalia Dei—The Mighty Acts of God: Essays on the Bible and Archaeology in Memory of G. Ernest Wright, ed. F. M. Cross, W. E. Lemke, and P. D. Miller. Garden City, NY: Doubleday. Yadin, Y., et al. 1960 Hazor II: An Account of the Second Season of Excavations, 1956. Jerusalem: Magnes. 1989 Hazor III–IV: An Account of the Third and Fourth Seasons of Excavation, 1957–1958 (Text). Jerusalem: Israel Exploration Society. Zarzecki-Peleg, A. 2005a Tel Megiddo during the Iron Age I and IIA–IIB: The Excavations of the Yadin Expedition at Megiddo and Their Contribution for Comprehending the History of This Site and Other Contemporary Sites in Northern Israel. Unpublished Ph.D. dissertation, Hebrew University of Jerusalem (Hebrew). 2005b Trajectories of Iron Age Settlement in North Israel and Their Implications for Chronology. Pp. 367–78 in The Bible and Radiocarbon Dating: Archaeology, Text, and Science, ed. T. E. Levy and T. Higham. London: Equinox. In press The Bit-Hilani in the Land of Israel: A Renewed Discussion. In Proceedings of the Third International Conference on the Archaeology of the Ancient Near East: Paris, May 2002, ed. J.-C. Margueron, P. de Miroschedji, and J.-P. Thalmann. Winona Lake, IN: Eisenbrauns. Zimhoni, O. 1997 Studies in the Iron Age Pottery of Israel: Typological, Archaeological and Chronological Aspects. Tel Aviv: Tel Aviv University Press.

Chariot Fittings from Philistine Ashkelon Lawrence E. Stager Department of Near Eastern Languages and Civilizations Harvard University

Recent discoveries by the Leon Levy Expedition to Ashkelon indicate that the Philistine newcomers to the coast of Canaan in the first half of the 12th century bce were well equipped for chariot warfare against the Egyptians, Canaanites, and Israelites. In the Medinet Habu reliefs of Ramesses III, the Sea Peoples, like the Egyptians, field six-spoked chariots drawn by two horses (Dothan 1982: 7). A driver and two spearmen constitute the crew of a Sea Peoples’ chariot, and their spears are about as long as the warriors are tall (Fig. 1). Unlike the Egyptians, however, the Sea Peoples do not have archers on board. The light, two-wheeled, horse-drawn chariot with spoked wheels was introduced into the Levant early in the second millennium bce (Littauer and Crouwel 1979: 71–72; Moorey 1986; Drews 1988: 96; Caubet and Yon 2001: 71), and from there the Canaanites (“Hyksos”) brought it to Egypt (for the early textual reference to chariotry at Avaris, the capital of the Hyksos, see Wilson 1969a: 554). Like the chariot, the composite bow was introduced into Egypt by the Canaanites. It greatly increased the arrow’s range of flight, velocity, and power on impact. The foundation of the composite bow was a central strip of wood, to which animal sinew was attached to the outer side and pieces of horn were glued to the inner side (McLeod 1970: 31). In the Ugaritic Tale of ºAqhat, the composite bow is described as made of “ash” wood, with sinews from wild bulls, horns from rams, tendons from bull hocks, and reeds. It took years to cure a composite bow before it was ready for use in combat (McLeod 1958: 400; Drews 1993: 110). Author’s note: I am pleased to dedicate this essay to William G. Dever, who has been at the forefront of Syro-Palestinian archaeology for the past 40 years. It was my good fortune to learn the fundamentals of field archaeology under his expert tutelage in the “trenches” at Tell Gezer. I am also delighted to see that in recent years Bill has returned to the battlefield of Biblical Archaeology, where, armed with his formidable knowledge of archaeology and history, and wielding a sword-like pen, he has challenged many of the inanities that today pass for biblical scholarship. I am grateful for the assistance provided by Sophia Chaknis and Adam Aja in preparing this manuscript. I also thank Peter Dorman of the Oriental Institute, University of Chicago, for his help with the Egyptian references to teher-warriors.

Armed with the composite bow, the light horse-drawn chariot became a formidable “mobile firing platform” for archers (Littauer and Crouwel 1979: 92; Moorey 1986: 208; Drews 1993: 110; Keegan 1993: 57; Burke 2004: 69) and brought about a “revolution in ancient warfare”: “The charioteer drove the horses while the chariot warrior shot arrow after arrow against the relatively stationary enemy formations, the chariots keeping just outside the range of the opposing infantry’s bowmen” (Drews 1993: 105). As mentioned above, Egyptian depictions of Sea Peoples’, as well as Hittite, chariots show a crew of three, with a driver and two other warriors armed with spears in their right hands, not with bows and arrows, and their chariots had neither quivers for arrows nor bowmen on board. Thus, thrusting spears seem to have been the weapon of choice for the Hittite and Sea Peoples charioteers (Littauer and Crouwel 1979: 91; Dothan 1982: 7); however, we know from other textual and archaeological sources that both the Hittites and the Philistines had skilled archers. It is possible, therefore, that the Egyptian artists chose not to acknowledge these bowmen in their renditions of the pharaoh’s enemies, an artistic convention intended to illustrate how inept the enemy was. Alternatively, the Egyptian portrayal of the chariot crews of the Hittites and the Sea Peoples may reflect the reality that, rather than being a “mobile firing platform,” the chariot served as a mobile thrusting or throwing platform from which warriors could spear the enemy and their war-horses. In addition, the lightweight chariot, designed for speed and maneuverability, might have served as a “taxi” to transport warriors in and out of battle, as described in Homer’s Iliad (cf. Drews 1993: 113–29). Textual and iconographic references to Philistine chariotry are quite meager. In Ramesses III’s description of his war against the Sea Peoples—a confederation of Philistines, Sikkils, Shekelesh, Denyen, and Weshesh—he mentions teher-warriors as part of the land forces he defeated: “They cut off their (own) land and were coming, their soul finished. They were teher-warriors on land; another (group) was on the sea” (Wilson 1969b: 263). Teherwarriors were not limited to Sea Peoples’ armies. They were associated with Hittite chariotry in the Battle of

170

Lawrence E. Stager

Fig. 1. Sea Peoples’ chariot with charioteers wearing feathered headdresses in the Medinet Habu reliefs of Ramesses III (illustration by C. Alexander and L. E. Stager [linchpin and yoke-saddle boss in black added], adapted from H. H. Nelson et al. 1930: pl. 32, “Ramses in Battle with the Land Forces of the Sea Peoples”).

Kadesh. Wilson refers to them as “chariot-warriors” (1969c: 258, n. 23). Helck views them as disorganized bands of raiders (1979: 135–37), but this does not fit the picture of Philistine and other Sea Peoples’ forces depicted in the Medinet Habu wall reliefs. O’Connor sees in these battle scenes a “high degree of effective military organization” (2000: 95). In short, teher seems to connote an elite status of warrior among the Hittites, the Egyptians, and the Sea Peoples, at times associated with horses and chariotry. It seems likely, therefore, that the spearmen in the light, two-wheeled, horse-drawn chariots of the Sea Peoples depicted in the Medinet Habu reliefs are teherwarriors (Fig. 1). The first and only direct biblical reference to Philistine chariotry appears in the early monarchic story of the Philistines mustering their forces and camping at Michmas in the hill country as they readied for battle against King Saul and the Israelites (1 Sam 13:5): “The Philistines gathered to attack Israel: 3,000 chariots (rekeb) and 6,000 chariot horses (parasîm) and troops as numerous as the sands of the seashore.” The MT reads 30,000 chariots, an impossibly high number; with McCarter (1980: 226), I follow the Lucianic recension of the Septuagint, which has 3,000. The number of chariots and horses could be considerably

lower if ªlp in this context does not mean literally one thousand but a much smaller military unit, contingent, or squadron (see, e.g., Mendenhall 1958; McCarter 1980: 107). What is important in this understanding of the passage is that the proportion of horses to chariots is exactly right for the period: two horses per chariot. In the Bible the, terms ªelep < ªalp and mispa˙â appear frequently in parallelism, but they need not be numerically equivalent. I suggest that the social term mispa˙â, referring to large families or “clans” idealized at a “thousand” members, provided the pool from which the military unit known as ªalp was derived and the young men of fighting age were drawn. This interpretation would reduce the ªalp to a much smaller number. The other reference, or rather allusion, to Philistine chariotry appears in the story of Saul’s last stand on Mount Gilboa, as the Philistines’ chariots (rekeb) and horses (parasîm) were closing in on him. Saul ordered his Amalekite aide to finish him off (2 Sam 1:6–9). Paras has been variously rendered “horseman,” “cavalry,” or the like. In this passage and the above-cited 1 Sam 13:5, paras must be translated in accordance with the context, which implies “chariot horse.” In such contexts, paras is a synonym for sûs, and when used in conjunction with rkb/

Chariot Fittings from Philistine Ashkelon

a

171

b

c

d

Fig. 2. Bronze anthropomorphic linchpin from Ashkelon: (a) side view; (b) front view; (c) side view left; (d) rear view (illustrations by T. Bartolo, photos by Z. Radovan; Leon Levy Expedition to Ashkelon).

mrkbh, it should be translated “chariot horse” (1 Sam 8:11, 2 Sam 15:1, 1 Kgs 1:5; see Niehr 2003). There is no evidence to suggest that Israel had cavalry, despite the clay figurines of the horse-and-rider type that appear in the late 8th–7th century bce. The meaning of this coroplastic art is in any event disputed, with interpretations running the gamut from riders representing a solar deity to the “horses of the sun” mentioned in 2 Kgs 23:11, to astral deities composing the militant “Host of Heaven” (see Keel and Uehlinger 1998: 341–49). Also noteworthy is the taunt of the Rabshakeh to Hezekiah’s official representatives in 2 Kgs 18:23: “Come now, make this wager with my master, the king of Assyria: I’ll give you two thousand horses if you can produce riders to mount them.” Clearly, the Assyrians did not think that Judah in the late 8th century had a military force that could be dignified with the term cavalry. That cavalry was known at the time from other countries, such as Assyria, is clear from the unambiguous use of paras in Ezek 23:6, 12. It is noteworthy that Ahab, king of Israel, contributed 2,000 chariots and 10,000 foot soldiers but no cavalry to the anti-Assyrian coalition at the Battle of Qarqar in 853 bce, while his allies from Damascus and Hamath contributed cavalry in addition to chariots and foot soldiers (Oppenheim 1969: 278–79). Direct evidence for Philistine chariotry in Iron Age I comes from the recent archaeological discoveries of chariot fittings at Ashkelon and Ekron, two leading cities of the

Philistine Pentapolis: an anthropomorphic linchpin and a yoke-saddle boss or knob were found at Ashkelon, and another anthropomorphic linchpin was found in Ekron.

Linchpins The bronze linchpin from Ashkelon (Material Culture No. 47,971) was found in Grid 50, Square 59, Layer 483 in Room 503, Philistine Bichrome Phase 9 of the 11th century bce. It is 17.4 cm in height (Fig. 2a–d). The linchpin and the yoke-saddle knob were both found in Room 503, ca. 4 m apart. Their proximity suggests that they served as fittings for the same chariot. There was nothing in the room or the building to suggest that it was anything other than domestic quarters. Spool weights for weaving were common. The painted pottery belonged to the Philistine Bichrome tradition. Phase 9 in Grid 50 is roughly contemporary with Phase 18 in Grid 38, which yielded a pictorial krater with a depiction including an abbreviated chariot (see below), as well as with Stratum V at Ekron, in which the other linchpin was found. It is not by chance that the only two linchpins excavated in Israel come from the same cultural horizon in Philistia. In the Ashkelon example, the lower half of the rounded shaft tapers to a point. The shaft widens to its largest diameter at midpoint, above which is an elaborately decorated upper torso and human-like head. It seems likely that the

172

Lawrence E. Stager

Fig. 3. Ivory yoke-saddle boss from Ashkelon (photo by I. Sztulman; Leon Levy Expedition to Ashkelon).

undecorated lower half of the pin went through and below the hole at the end of the axle as far as the thickened midsection. A cord or leather thong would have been used to lash the pin in place. The loop at the back of the head was probably also used to secure the linchpin. The long neck or torso of the pin is decorated with either some sort of “necklace” or a “corslet” of scale armor (for the development of the mail corslet and the war chariot, see Yadin 1963: 84), above which is the head of the figure. The face has a prominent nose and bulging eyes. The hair falls in a circular pattern at the back, below five tightly wound horizontal bands. A mushroom-shaped coiffure or headdress spreads above and beyond the bands (Fig. 2a–d). The long neck or torso with a corslet (if that is what is actually represented in this metalwork) and the prominent facial features are, in my opinion, reminiscent of the terra-cotta specimens of the so-called “Ashdoda” figurines, which very likely represent a leading goddess of the Philistines (Dothan 1982: 234–37). A dozen or more fragments of this type of clay figurine have been found at Ashkelon. The upper portion of these, that is, the part that rises above the throne of the deity, is characterized by a very long neck, bulging eyes (applied clay pellets), and a long aquiline nose (also applied separately). Most of the terra-cottas, unlike our metal example, have a flat headdress or polos. An exception among the clay figurine fragments found at Ashkelon is a longnecked version of the goddess with pellet eyes and a pellet nose and the head crowned with an elaborate headdress or hairstyle. I therefore believe it likely that the figure on the upper half of the Ashkelon linchpin represents the leading goddess of the Philistines in Iron Age I. She provided protection for the chariot, its team of horses, and its chario-

Fig. 4. Ivory yoke-saddle boss from Ashkelon (illustration by C. Alexander; Leon Levy Expedition to Ashkelon).

teers, much like the Astarte and Hathor representations emblazoned on horse blinkers and frontlets. The royal chariot of Ramesses II at the Battle of Kadesh depicted in the wall reliefs on the second pylon of the Ramesseum has a linchpin topped with the image of the cow-goddess Hathor wearing a headdress with the sun disk between two horns (Ellis 1966: 42–43, n. 11, fig. 1). Other decorative linchpins from New Kingdom Egypt depict prisoners, animals, and deities (Yadin 1963: 213; Ellis 1966: 43–44, figs. 2–4; Littauer and Crouwel 1985: 75; 1987). At Ekron, a Janus-headed bronze linchpin was found together with other cultic paraphernalia in Stratum V Temple Building 350, dating to the 11th century bce (Dothan 2002: 11–14; 2003: 208–9). The Ekron linchpin is 10 cm in height. A flat cap covers the antithetical heads. The neck has a horizontal perforation, probably for the insertion of a cord or thong. Although the faces of the two heads are somewhat eroded, their flat heads and facial profiles resemble sphinxes on bronze-wheeled cult stands from Cyprus (Dothan 2002: 11). The sphinx was a frightening mythological creature used to guard and protect royal and sacred portals. On linchpins, with one sphinx facing ahead and the other behind, they served much the same purpose in protecting the chariot.

Chariot Fittings from Philistine Ashkelon

173

Table 1. Yoke-Saddle Bosses from the Levant Site

Period

Source

Alalakh

MB II–LB II

Woolley 1955: 296, pl. 82:27–29

Ugarit

LB II

Caubet 1991; Caubet and Yon 2001: 77, fig. 1

Beth-Shean

MB II–LB II

James and McGovern 1993: fig. 114:1–6

Gezer



Macalister 1912: 252, fig. 401 (“furniture knob”)

Tell el-ºAjjul

Late Bronze Age

Petrie 1933: pl. 27:82–83; 1934: pl. 41:120

Yoke-Saddle Boss The elephant-ivory yoke-saddle knob (Material Culture No. 53,595) was found in Grid 50, Square 58, Feature 490 in Room 503, Philistine Bichrome Phase 9 of the 11th century bce. It is 6.2 cm in height. Its diameter is 5.7 cm at its widest and 3 cm at its narrowest, and the diameter of the perforation is 1.2 cm (Figs. 3–4). The yoke-saddle is shaped like an inverted slingshot, which helps adapt the yoke to the horse’s neck ahead of the withers (Littauer and Crouwel 1979: fig. 42). The handle of the “slingshot” was lashed to each arm of the yoke; the forked part was fitted like a small saddle with a blanket over the horse’s shoulders (Littauer and Crouwel 1985: 81–82). The inverted handle of the yoke-saddle terminated in a boss or knob made of stone (Littauer and Crouwel 1985: 4, fig. 2, pls. 33 [close-up], 72 [upper]; James and McGovern 1993: 186; see also James 1974; 1978) (Fig. 1). The boss was attached to the yoke-saddle by a nail 14– 15 cm long, which passed through the hole in the knob and into the yoke (Littauer and Crouwel 1985: 82, fig. 19, pl. 74 [upper left]; James and McGovern 1993: 186). The Ashkelon yoke-saddle boss was found wedged against the fieldstones of a flimsy interior wall in Room 503. At first we believed it was the pommel of a dagger, but the perforation, 1.2 cm in diameter, that continued from the top to the bottom of the reel-shaped artifact disabused us of this interpretation. Artifacts of this type have been variously labeled “doorknobs,” bobbins, reels, furniture parts, and the like (James 1974; James and McGovern 1993: 186; Caubet and Yon 2001: 69–70). The Ashkelon exemplar has a convex top, spool-shaped profile, and a flat base with three horizontal grooves around it (Figs. 3–4). It compares nicely with other exemplars from the Levant, for example, from Late Bronze Age Ugarit (Caubet and Yon 2001: 77, fig. 1 [designated pommeaux], the one on the far right being the closest parallel) and from LB II Beth-Shean Levels VII and VIII (Type A-1) (James and McGovern 1993: fig. 114:1, 6 [designated “bosses”]). The most secure contexts for understanding the bosses as chariot fittings are the well-preserved remains of Tutankhamun’s chariots, which include almost exact parallels for the Ash-

kelon boss (Littauer and Crouwel 1985: 81, pls. 33–35 [designated “yoke-saddle finials”]). Most of the examples from the Levant and Egypt were carved from “alabaster,” either Egyptian calcite or Canaanite gypsum. The Ashkelon boss is a very rare specimen, carved from an elephant tusk (Paula Wapnish, personal communication) (for the infrequent use of ivory, see the survey of Near Eastern bosses [pommeaux] and finials [embouts] in Caubet and Yon 2001: 70; see also Caubet 1991). In addition to the bosses found in the Levant (Table 1), some sites have yielded Late Bronze Age yoke terminals, representing further archaeological evidence for chariotry in Canaan, for example, Ashdod, Gezer, Hazor, Lachish, and Megiddo (James and McGovern 1993: 186). James was the first to recognize, on the basis of her analysis of the finds from Beth-Shean, that the bosses and terminals found in the Levant probably were chariot fittings (1974). But a decade later, the two leading authorities on the chariot, Littauer and Crouwel, remained unconvinced by the Levantine parallels and declared: “On the basis of their resemblance to the known examples, these have been tentatively identified as yoke-saddle finials, but pending finds that associate them clearly with other chariot material this must remain uncertain” (1985: 81). We now have the combinatory contextual evidence at Ashkelon: an ivory yoke-saddle boss and an elaborate bronze anthropomorphic linchpin, both found in the same room. In light of these finds, I suggest that the Ashkelon chariot must have belonged to one of the elite corps of charioteers, perhaps a teher-warrior.

Pictorial Krater From another part of the Iron I city comes a supplementary piece of iconographic evidence for Philistine chariotry at Ashkelon. In Grid 38, at the center of the ancient seaport, a sequence of Iron Age settlements were excavated that began in the second quarter of the 12th century bce and ended with the destruction of the seaport by the armies of Nebuchadrezzar in 604 bce. Following the initial Philistine settlement, characterized by locally-made Mycenean IIIC or Philistine Monochrome pottery (Phase 20), the

Lawrence E. Stager

174

Fig. 5. Side B of Bichrome ware pictorial krater, depicting abbreviated chariot with one six-spoked wheel and charioteer wearing feathered headdress (illustration by P. Mountjoy, photo by Z. Radovan; Leon Levy Expedition to Ashkelon).

builders laid out a new architectural scheme, which continued with various modifications until ca. 1000 bce. Phase 19 consisted of two domestic complexes designated the north and south villas, separated by an open courtyard. To the Mycenean IIIC pottery was added, in approximately equal amounts, the well-known Philistine Bichrome ware. The two villas were rebuilt in Phase 18, when the local Monochrome ware was no longer being produced, and only the Bichrome pottery remained of the painted tradition. Phase 18 is roughly contemporary with Phase 9 in Grid 50.

Where a tree had once stood in the Phase 18 courtyard, we excavated a large pit filled with refuse. Among several sherds retrieved from the pit were fragments of a rare pictorial Bichrome krater, dating to the late 12th or 11th century bce (Stager and Mountjoy in press). On Side A of the krater, a warrior/sailor wearing a feathered headdress typical of some of the Sea Peoples, in particular the Philistines, confronts a dolphin. On Side B of the vessel, another human figure, also adorned with a feathered headdress, sits astride a vehicle with one six-spoked wheel, probably representing an abbreviated chariot (Fig. 5). Egyptian and

Chariot Fittings from Philistine Ashkelon

175

Canaanite chariots had four-spoked wheels in the early part of the 18th Dynasty and LB I (Yadin 1963: 86–88), but by LB II, the six-spoked wheel prevailed. On the other hand, Mycenean chariots were usually depicted with fourspoked wheels throughout the Late Bronze Age. After surveying the evidence of chariot fittings from the Levant, all dating to the Late Bronze Age or slightly earlier, James and McGovern wondered why, “inexplicably, no bosses or terminals are recorded from early Iron Palestinian sites” (1993: 186). Now, with the discovery of char-

iot fittings in the Iron I Philistine cities of Ashkelon and Ekron and with the contemporary iconographic evidence of the pictorial krater from Ashkelon, we need no longer wonder. Drews states that “Greek chariot warfare virtually disappeared” after the Mycenean IIIC period: “the day of chariot warfare was over, and the day of the infantryman had arrived” (1993: 118). Clearly, the Philistines of Mycenean background had not yet received this message in the 11th century bce.

References Burke, A. 2004 The Architecture of Defense: Fortified Settlements of the Levant during the Middle Bronze Age. Unpublished Ph.D. dissertation, University of Chicago. Caubet, A. 1991 Objets et instruments d’albâtre. Pp. 265–72 in Ras Shamra Ougarit VI: Arts et industries de la pierre, ed. M. Yon. Paris: Éditions Recherche sur les Civilisations. Caubet, A., and Yon, M. 2001 Pommeaux de chars, du Levant à la Mésopotamie et à l’Élam. Pp. 69–78 in Études Mésopotamiennes: Recueil de Textes Offert à Jean-Louis Huot. Paris: Éditions Rechercher sur les Civilisations. Dothan, T. 1982 The Philistines and Their Material Culture. New Haven: Yale University Press. 2002 Bronze and Iron Objects with Cultic Connotations from Philistine Temple Building 350 at Ekron. Israel Exploration Journal 52: 1–27. 2003 The Aegean and the Orient: Cultic Interactions. Pp. 189–213 in Symbiosis, Symbolism, and the Power of the Past: Canaan, Ancient Israel, and Their Neighbors from the Late Bronze Age through Roman Palaestina. Proceedings of the Centennial Symposium, W. F. Albright Institute of Archaeological Research and American Schools of Oriental Research, Jerusalem, May 2000, ed. W. G. Dever and S. Gitin. Winona Lake, IN: Eisenbrauns. Drews, R. 1988 The Coming of the Greeks: Indo-European Conquests in the Aegean and the Near East. Princeton: Princeton University Press. 1993 The End of the Bronze Age: Changes in Warfare and the Catastrophe ca. 1200 b.c. Princeton: Princeton University Press. Ellis, R. 1966 A Note on Some Near Eastern Linch Pins. Berytus 16: 42–48. Helck, W. 1979 Die Beziehungen Ägyptens und Vorderasiens zur Ägäis: Bis ins 7. Jh. v. Chr. Darmstadt: Wissenschaftliche Buchgesellschaft. James, F. W. 1974 Stone Knobs and Chariot Tracks. Expedition 16: 31–39. 1978 Chariot Fittings from Late Bronze Age Beth-Shan. Pp. 102–15 in Archaeology in the Levant: Essays for Kathleen Kenyon, ed. R. Moorey and P. Parr. Warminster: Aris & Phillips. James, F. W., and McGovern, P. E. 1993 The Late Bronze Egyptian Garrison at Beth-Shan: A Study of Levels VII and VIII. 2 vols. Philadelphia: University Museum.

Keegan, J. 1993 A History of Warfare. New York: Knopf. Keel, O., and Uehlinger, C. 1998 Gods, Goddesses, and Images of God in Ancient Israel. Minneapolis: Fortress. Littauer, M. A., and Crouwel, J. H. 1979 Wheeled Vehicles and Ridden Animals in the Ancient Near East. Leiden: Brill. 1985 Chariots and Related Equipment from the Tomb of Tut’ankhamun. Oxford: Griffith Institute. 1987 Unrecognized Linchpins from the Tomb of Tut’ankhamun and Amenophis II: A Reply. Göttinger Miszellen 100: 57–61. Macalister, R. A. S. 1912 The Excavation of Gezer, 1902–1905 and 1907–1909 II. London: Palestine Exploration Fund. McCarter, P. K. 1980 I Samuel: A New Translation. Anchor Bible 8. Garden City, NY: Doubleday. McLeod, W. 1958 An Unpublished Egyptian Composite Bow in the Brooklyn Museum. American Journal of Archaeology 62: 397–401. 1970 Composite Bows from the Tomb of Tut’ankhamun. Oxford: Oxford University Press. Mendenhall, G. E. 1958 The Census Lists of Numbers 1 and 26. Journal of Biblical Literature 77: 52–56. Moorey, P. R. S. 1986 The Emergence of the Light, Horse-Drawn Chariot in the Near East c. 2000–1500 bc. World Archaeology 18: 196–215. Nelson, H. H., et al. 1930 Medinet Habu I: Earlier Historical Records of Ramses III. Oriental Institute Publications 8. Chicago: University of Chicago Press. Niehr, H. 2003 Paras. Pp. 124–28 in vol. 12 of the Theological Dictionary of the Old Testament, ed. G. J. Botterweck and H. Ringgren. Grand Rapids, MI: Eerdmans. O’Connor, D. 2000 The Sea Peoples and the Egyptian Sources. Pp. 85–102 in The Sea Peoples and Their World: A Reassessment, ed. E. Oren. Philadelphia: University Museum. Oppenheim, A. L. 1969 Shalmaneser III (858–824): The Fight against the Aramean Coalition. Pp. 276–81 in Ancient Near Eastern Texts Relating to the Old Testament, ed. J. B. Pritchard. Princeton: Princeton University Press.

176 Petrie, W. M. F. 1933 Ancient Gaza III. Egyptian Research Account and British School of Archaeology in Egypt 55. London: British School of Archaeology in Egypt. 1934 Ancient Gaza IV. Egyptian Research Account and British School of Archaeology in Egypt 56. London: British School of Archaeology in Egypt. Stager, L. E., and Mountjoy, P. A. In press A Pictorial Krater from Philistine Ashkelon. In “Up to the Gates of Ekron”: Essays in Honor of Seymour Gitin, ed. S. White Crawford et al. Jerusalem: Israel Exploration Society. Wilson, J. A. 1969a The War against the Hyksos. Pp. 232–33, 554–55 in Ancient Near Eastern Texts Relating to the Old Testament, ed. J. B. Pritchard. Princeton: Princeton University Press.

Lawrence E. Stager 1969b The War against the Peoples of the Sea. Pp. 262–63 in Ancient Near Eastern Texts Relating to the Old Testament, ed. J. B. Pritchard. Princeton: Princeton University Press. 1969c The Asiatic Campaigning of Ramses II. Pp. 255–58 in Ancient Near Eastern Texts Relating to the Old Testament, ed. J. B. Pritchard. Princeton: Princeton University Press. Woolley, C. L. 1955 Alalakh: An Account of the Excavations at Tell Atchana in the Hatay, 1937–1949. London: Society of Antiquaries. Yadin, Y. 1963 The Art of Warfare in Biblical Lands in Light of Archaeological Study. New York: McGraw-Hill.

Goddesses and Cults at Tel Dor Ephraim Stern Institute of Archaeology, The Hebrew University of Jerusalem

In the final season of excavation at Tel Dor, 1 a number of interesting objects were found that represent the figures of two important goddesses, both of whom are not indigenous to the country. These finds shed light on the cult practiced by the inhabitants of the coastal cities of Palestine in the Persian and Hellenistic periods and on the diverse composition of the populations of these cities. The two objects discussed below were uncovered in a large public building in Area D1 in the southwestern part of the tell (part of the city’s acropolis at the time) (fig. 1). The building contained several levels, beginning in the Persian and continuing into the Hellenistic period.

Lead Weight (fig. 2a–b) The item shown in fig. 2 is a square lead weight (5 x 5 cm) with a loop at the top for suspending it from the rod of a balance. The weight itself and its significance are not discussed here but only the deities depicted on it. On the face of the weight, on a lattice-pattern background, is a schematic representation of the Phoenician–Punic goddess Tanit (fig. 2a). She is depicted with a circular head, triangular body, and outstretched arms. On the reverse side (fig. 2b) is a heavy club, one of the characteristic signs of Heracles. Next to it are the letters AGO, an abbreviation of the Greek agoranomos, meaning the official in charge of the marketplace and responsible for verifying the accuracy of the weights. The reverse side also includes the date of the weight’s manufacture—Year 12 of the era of Tyre (corresponding to 115 bce). A discussion of these subjects (see, e.g., Elayi and Elayi 1997; Kushnir-Stein 1997; 2002) is also beyond the scope of this paper. Heracles, the Greek hero, was in this period identified by the Phoenicians with their god Baal-Melqart (‘King of the City’), the principal god of Tyre. Thus, despite the presence of the heavy Greek club, it is the Phoenician deity, the consort of Tanit, who is represented on the weight. For each of the Phoenician cities, the name of a second deity was attached to the name of Baal (the princiAuthor’s Note: This paper is dedicated to Bill Dever, long-time colleague and close friend. 1. The excavations at Tel Dor were conducted under my direction for 20 years, from 1980 to 2000.

pal deity of the Phoenicians from the beginning of the Iron Age) indicating his status as the chief god of that particular city (e.g., Baal-Melqart was the chief god of Tyre, Baal-Eshmun the god of Sidon, Baal-Hamon the god of Carthage, etc.). Our main interest here is the depiction of the figure of Tanit on the face of the weight. A survey of the study of this goddess is an instructive example of the role that chance plays in the discovery of archaeological finds. Before the 1970s, dozens (if not hundreds) of signs and inscriptions with Tanit’s name had been found at all of the major archaeological sites of the Phoenician colonies in the western Mediterranean, primarily at Carthage itself, but also in other cities in North Africa, Sicily, Sardinia, Malta, and Spain (Moscati 1988). The standard text of these dedicatory inscriptions reads: “The great lady, Tanit, Panei Baal (‘Face of Baal’),” because she was known in the west as the devoted consort of the Carthaginian BaalHamon (Yadin 1970). In the east, on the other hand, in Phoenicia itself, and even on Cyprus, where many Phoenician outposts were established, no artifacts attesting to the worship of Tanit were discovered. Consequently, for many years, scholars were virtually unanimous in maintaining that the goddess Tanit had no affinities with the eastern Canaanite–Phoenician goddess Astarte and was unique to the inhabitants of the Phoenician–Punic colonies in the western Mediterranean, especially to Carthage, the capital. Early on, some scholars even sought a Lybian North African origin for her name, although in this period, many proper names known from Phoenicia included Tanit as a component (such as “Name, the servant of Tanit”). These names were simply interpreted as belonging to Carthaginians who had emigrated to or settled in the east as traders (Harden 1963: 87–90). Doubts about this interpretation first began to arise with the discovery of a dedicatory inscription in Carthage that reads: “To our lady Astarte and Tanit in Lebanon” (Slouschz 1942: 156, no. 127; Stieglitz 1990) and subsequently, in the early 1970s, archaeological excavations conducted along the Phoenician and Palestinian coast uncovered artifacts bearing the name and sign of Tanit with surprising frequency. Later, her name was found on stone funerary steles in the Tophet (a burial site with

Ephraim Stern

178

Fig. 1. Plan of Tel Dor, showing excavated areas.

urns containing the ashes of the deceased and stone steles above them) at Tyre (Sader 1990) and perhaps also in the Tophet at Achziv (Stern 2001: 89, fig. 1:47). Her symbol also survived on several dozen of the hundreds of pottery figurines of various sizes recovered from the shipwrecks near the port of Tyre and off the coast of Shavei Zion (Linder, who excavated the shipwreck at Shavei Zion, believed that it was a Carthaginian boat that had capsized along the Phoenician coast [1973]). The sign of Tanit appears on seals, seal impressions, bullae, and lead weights that are turning up in excavations at all the Phoenician coastal towns, from north to south, such as Byblos, Sarepta (biblical Zarephath, the town of the prophet Elijah) (Pritchard 1975: fig. 59:2), the Phoenician temple at Umm el-ºAmed (Rosh ha-Niqra), Acco (Dothan 1974), Dor (Stern 2000: 370), Ashdod Yam (Stern 1982: 216, fig. 363), and Ashkelon (Stager 1991: 31). The reverse side of the weight from the last site has not been published but, according to the description provided, it apparently bears the sign of Tanit. Dothan interpreted it as a “Punic inscription” and as evidence of trade between Ashdod and Carthage (1974). The sign of Tanit also appears on stone and metal pendants from the Persian period found at Ashkelon (Stager 1991: 31). Recently, it has become evident that Tanit was also worshiped in Galilee, which was also settled by the Phoe-

Fig. 2a–b. Tanit lead weight from Dor.

Goddesses and Cults at Tel Dor

179

nicians after the destruction of the First Temple. This continued at least until the 2nd century, the date of our weight, when Galilee was conquered by the Hasmoneans. At Tel Kedesh, a large public building excavated a short while ago yielded over 1,800 bullae (small clay lumps used for sealing papyrus documents), apparently from the archives of the governor of the region. These bullae included many with Phoenician inscriptions and signs of Tanit (Herbert and Berlin 2000: 122–23, fig. 3:d). The above examples represent only a few of the artifacts that have been discovered: many more dozen weights and other artifacts bearing the sign of Tanit uncovered on the Phoenician–Palestinian coast are at present in museums or private collections, and no information is available regarding their exact provenance or the circumstances of their discovery. How long the cult of Tanit continued in Palestine is not certain, but a 2nd-century ce coin found at Ashkelon mentions a temple of the goddess Paneablos, that is, Panei Baal, which as we have seen is Tanit’s title (Meshorer 1984: 27). Final proof of the identification of the “eastern” Astarte with the “western” Tanit is provided by a Phoenician inscription from Temple 1 at Sarepta, situated midway between Tyre and Sidon. This temple, dated to the 7th century bce, yielded an inscription dedicated to “Tanit Ashtoret”—that is, Tanit who is also Astarte (Pritchard 1982).

Pottery Figurine (fig. 3) This object is a complete pottery figurine of quite large dimensions (15 cm high) that represents the Phrygian goddess Cybele. Upon discovery, it was immediately identified by Andrew Stewart of the University of California at Berkeley, the Associate Director of the Dor excavations, a leading Classical art expert. His identification was based mainly on the fact that the figurine depicts a goddess seated on a throne supported at the sides by a pair of lions, animals that invariably accompany representations of this goddess. She holds a large drum in one hand, and in the open palm of her other hand is a thus far unidentified object. Her hair is arranged in typical Greek style and she wears a hat of the type common among women in Anatolia and northern Syria, a legacy from the Hittite and NeoHittite world. The figure of this goddess accompanied by a pair of lions is known from Anatolia beginning in prehistoric times. On stone and pottery figurines found there at the important Neolithic sites of the eighth and seventh millennia bce, such as Çatal Hüyük and Hacilar (Ankara 1997: 27, 32), she is depicted with lions (and occasionally tigers). From the outset, she was regarded as the mother

Fig. 3. Cybele clay figurine from Dor.

goddess (fertility goddess); her place in the Anatolian pantheon continued through the Bronze and Iron Ages. In the second millennium bce, she was called Kubaba by the Hittites. In the first millennium bce, she became the chief goddess of the central Anatolian kingdoms, especially of Phrygia, and had a young consort, Attis, who had his own independent cult. From Phrygia, her cult spread westward to the kingdom of Lydia and the Mediterranean coast, which were settled by the Greeks, until she gradually became accepted. In the Roman period, the worship of her cult extended to Rome, acquiring the status of an established religion in 204 ce (Barnett 1967). In Phrygia, Cybele was also regarded as a mountain goddess. Her sanctuaries were hewn in rock on mountaintops; their facades occasionally portrayed her with lions but, on the whole, their decoration was restricted to

Ephraim Stern

180 geometric motifs. These are among the most impressive remains of the great Phrygian kingdom. The generally held opinion is that her worship was connected with rites of purification and ecstasy, in which music played a central role (Barnett 1967: 21–24); on many of the Cybele figurines found at Phrygia, various musicians replace the lions at the sides of her throne (Ankara 1997: 183, pl. 303). The Cybele figurines from Dor similarly depict the goddess with a large drum. In the period under discussion—late Persian–early Hellenistic—the worship of Cybele was still relatively rare outside the borders of Anatolia. The Dor excavations yielded, in addition to the complete figurine with a drum, an identical drum on another figurine. A third drum was

found during Garstang’s excavations at the site in the 1920s (1924: 42, pl. 3:8), apparently in the trench dug near the findspots of our figurines. These finds suggest that this goddess was also popular at Dor. If we assume that in this relatively early period Cybele had affiliations mainly with the people of central Anatolia (Phrygia and Lydia), the Cybele figurines from Dor can provide evidence that the city’s Greek residents, generally known as “East Greeks,” also included Lydians and Phrygians who had settled in the city prior to the Macedonian conquest (Waldbaum 1994). This assumption has important implications for the study of the diverse composition of the population of the Palestinian coast in the Persian and Hellenistic periods.

References Ankara 1997 A Guide to the Museum of Anatolian Civilizations. Ankara. Barnett, R. D. 1967 Phrygia and the Peoples of Anatolia in the Iron Age. Fasc. 56 of The Cambridge Ancient History (rev. ed.). Cambridge: Cambridge University Press. Dothan, M. 1974 The Sign of Tanit from Tel ºAkko. Israel Exploration Journal 24: 44–49. Elayi, J., and Elayi, A. G. 1997 Recherches sur les poids phéniciens. Supplement à Transeuphratène 5a. Paris: Gabalda. Garstang, J. 1924 Tanturah (Dora) II: Archaelogical Results. Bulletin of the British School of Jerusalem 6: 65–75. Harden, D. 1963 The Phoenicians. London: Thames & Hudson. Herbert, S., and Berlin, A. 2000 Tel Qedesh. Israel Exploration Journal 50: 118–23. Kushnir-Stein, A. 1997 On the Chronology of Some Inscribed Lead Weights from Palestine. Zeitschrift des deutschen Palästina-Vereins 113: 88–91. 2002 New Hellenistic Lead Weights from Palestine and Phoenicia. Israel Exploration Journal 52: 225–30. Linder, E. 1973 A Cargo of Phoenician Punic Figurines. Archaeology 26: 182– 87. Meshorer, Y. 1984 The City Coins of Eretz-Israel and the Decapolis in the Roman Period. Jerusalem: Israel Museum (Hebrew). Moscati, S. 1988 Stelae. Pp. 304–27 in The Phoenicians, ed. S. Moscati et al. Milan: Bompiani.

Pritchard, J. B. 1975 Sarepta: A Preliminary Report on the Iron Age: Excavations of the University Museum of the University of Pennsylvania, 1970– 1972. Philadelphia: University of Pennsylvania Museum. 1982 The Tanit Inscription from Sarepta. Pp. 83–94 in Phoenizien im Westen, ed. H. G. Niemeyer. Mainz am Rhein: von Zabern. Sader, H. 1990 Phoenician Stelae from Tyre. Berytus 39: 105–26. Slouschz, N. 1942 Thesaurus of Phoenician Inscriptions. Tel Aviv: Dvir (Hebrew). Stager, L. E. 1991 Ashkelon Discovered: From Canaanites and Philistines to Romans and Moslems. Washington, DC: Biblical Archaeology Society. Stern, E. 1982 Material Culture of the Land of the Bible in the Persian Period 538–332, b.c. Warminster: Aris & Phillips. 2000 Dor, the Ruler of the Seas. Jerusalem: Israel Exploration Society. 2001 The Archaeology of the Land of the Bible, vol. 2: The Assyrian, Babylonian, and Persian Periods, 732–332 b.c.e. Anchor Bible Reference Library. New York: Doubleday. Stieglitz, R. 1990 Die Gottin Tanit im Orient. Antike Welt 21: 106–9. Waldbaum, J. C. 1994 Early Greek Contacts with the Southern Levant, 1000–600 bc: The Eastern Perspective. Bulletin of the American Schools of Oriental Research 293: 53–65. Yadin, Y. 1970 Symbols of Deities at Zinjirli, Carthage and Hazor. Pp. 199–231 in Near Eastern Archaeology in the Twentieth Century: Essays in Honor of Nelson Glueck, ed. J. A. Sanders. Garden City, NY: Doubleday.

Cypriot Anthropomorphic Figurines of a Certain Type Stuart Swiny Institute of Cypriot Studies State University of New York at Albany

I knew Dever the scholar well from his alreadynumerous publications, especially his work on chronology, which had a bearing on my own research on the Early and Middle Bronze Age of Cyprus, a 900-year time span beginning ca. 2500 bce, but Dever the man only came into my life as the result of a lecture trip to snow-covered Jerusalem in the winter of 1981. At the time, he was nursing a badly bruised hand, the result of an altercation (over noise, I believe) with a neighbor next door to the flat he occupied when he was a Guggenheim Fellow in Jerusalem. I found his spirit, sense of humor, appreciation for certain things Scottish, and generosity most appealing, and a friendship was born. The following year, in his capacity as the American Schools of Oriental Research (ASOR) Chairman of the Committee on Archaeological Policy, he invited met to join the tour of ASOR-affiliated and other excavations in Israel and Jordan, and what a tour it was! I well remember one midafternoon (some eight hours after leaving the W. F. Albright Institute of Archaeological Research in Jerusalem at first light), stuck in traffic in the middle of Ramallah or Nazareth and long after we should have had lunch (Sy Gitin, the Albright Director, was a stern taskmaster), Bill became testy. I inquired with feigned concern about the reason for his discomfiture, and his response was pure Deveresque: “I like to eat, and mostly every day!” When the editors invited me to contribute to William Dever’s festschrift, initially I was at a loss for an appropriate subject, having just finished the final publication of the Early Bronze Age excavations at Sotira Kaminoudhia, and not yet having any suitable subject for an essay, from the renewed work since 2001. However, upon reflection, I decided that there was a single find perhaps worthy of an offering to Bill, especially in light of one of the other contributions to this festschrift. 1 One of the most remarkable, intriguing, and intellectually challenging artifacts of the ceramic Neolithic period in

Cyprus was discovered in the spring of 1981 by Arestis Aristidou while plowing his fields a mile northwest of Sotira village, where he then served as the agrofilikas or local “rural constable” (fig. 1). That summer, I had begun excavating Kaminoudhia on the northern outskirts of the village and remember clearly the afternoon that Mr. Aristidou showed me two carved pieces of limestone, with the somewhat coy query: “What is this thing?” The patinacovered “thing” was indeed unlike any artifact I had ever seen before, and I was initially at a loss to provide him with an answer. At first glance, when both pieces were held together, it resembled a highly stylized human figure with the hint of a slanting face atop an elongated neck, which had broken off the body when it was dropped by his grandson a short while earlier. 2 But when viewed from another angle, it was unmistakably phallic (fig. 1:c) and as such, immediately invited comparison with the still-more-schematic limestone anthropomorphic figurine excavated by Porphyrios Dikaios at Sotira Teppes, overlooking the very spot where we stood that hot August afternoon (fig. 5). 3 Part of Dikaios’s description bears repeating here: “Stone figure. It is fiddle-shaped and shows a minimum indication of anatomy. The central part of the body shows straight sides which converge slightly upwards and join the head which is somewhat thicker and wider, the top being rounded. The lower part of the body is similarly thicker and a vertical groove on both front and back indicated the division of the legs” (1961: 202, pls. 91, 102, 109). Was Dikaios being prudish in omitting any allusion to the phallic appearance of the object? One wonders whether he would have drawn the connection between the Arkolies and Teppes figures that immediately seemed obvious to me; furthermore, both objects are of almost identical height. Convinced at least of its interest as a potential antiquity, I handed the find over to the Department of Antiquities, and the then Director, Vassos Karageorghis, agreed that he

Author’s note: Many thanks go to Jessie Pellerin, the excavation draftsperson in 2001, for the drawing of all the figures, with the exception of fig. 1, and to Alessandra Swiny for scanning them to scale. 1. See P. J. King, “Gezer and Circumcision,” in this volume, pp. 333–340.

2. The break is clearly visible in the photographs showing the object head-on in Swiny and Swiny 1983: pl. 6:4 and the color view of the object from the side in S. Swiny 1989: 183. 3. This is a new rendition of the figurine specifically prepared for this publication.

Stuart Swiny

182

Fig. 1. Limestone anthropomorphic figurine from Sotira Arkolies.

had never seen anything like it either. We both felt, on the basis of its findspot and similarity to the Teppes piece, that it appeared to be a genuine ceramic Neolithic figurine of unique type and that it should be published in a forthcoming issue of the Report of the Department of Antiquities, Cyprus. Karageorghis reported on the find in the following year’s Annual Report of the Director of the Department of

Antiquities (1982a: 46, fig. 56) and Bulletin de Correspondance Hellénique (1982b: 690, fig. 10), and in 1983, it was published by Swiny and Swiny in the Report of the Department of Antiquities, Cyprus. Unfortunately, the photographs showing the figurine from the front and side (Swiny and Swiny 1983: pl. 6, top register) have it sitting on a prop, albeit airbrushed out of the picture, which

Cypriot Anthropomorphic Figurines of a Certain Type

183

Fig. 2. Object S796, limestone figurine from Sotira Kaminoudhia.

changes the angle of the “neck.” It should really be shown as in the drawings or in S. Swiny 1989: 183. The purpose of this essay, however, is not to review the Arkolies find and other comparative material that has been published since its discovery but, as mentioned above, to discuss a find from the 2001 season at Kaminoudhia that appears to belong to the same class of object. Object S796 (fig. 2) was discovered against Wall AE in the fill of an alleyway (Unit 33) in Area A. It lay next to an upside-down saddle quern, evidently discarded in this spot

as no longer being of any use. The object was carved of hard white limestone and exhibited a rough, slightly pockmarked surface. As preserved, it is 12.4 cm long and 6.3 cm wide and somewhat resembles a potato in size and shape. It has rounded angles, with an inverted V-shaped concave base and a ridge just below the “neck” or at the juncture of the body and the now missing shaft, which presumably ended in a phallic-shaped head (e.g., figs. 1:b, 3). When placed on a flat surface it stands easily and remains steady. Using the same proportions as the intact Arkolies

Stuart Swiny

184

Fig. 3 (left). Limestone figurine in private collection (drawing based on Flourentzos 2003a: fig. 10). Fig. 4 (center). Picrolite Chalcolithic figurine from Erimi Pamboula (drawing based on Bolger 1988: fig. 34:342). Fig. 5 (right). Ceramic Neolithic limestone figurine from Sotira Teppes (after Dikaios 1961: pls. 91:106, 102:106).

figurine, I have drawn a neck and head to scale for Object S796, as shown in fig. 6. Although fragmentary, the preserved part of Object S796 is immediately recognizable as the “body” or base of a figurine belonging to the same category as the figurine from Arkolies. It is certainly larger, less carefully finished, and more abstract, but the angular contours, concave base, and positioning of the now-missing neck are similar. The detailing of the Arkolies piece, especially the carefully etched groove that separates the legs and feet of the frontal view that defines the scrotum when observed from the other side, is lacking, and we shall never know what form the head took, unless the rest of the object is discovered during future excavations at Kaminoudhia. 4 The overall similarity, however, firmly links the two objects, which were discovered only 800 m apart on the outskirts of the village of Sotira. Since nothing remotely similar has been recorded from an Early or Middle Cypriot context, either domestic or funerary, and in view of its similarity to the Arkolies figurine, the dating of which has not been questioned since its full publication more than 20 years ago, it is logical to presume that this is a ceramic Neolithic artifact that had been found either at Teppes or in the general vicinity by an inhabitant of the Kaminoudhia settlement and brought home as an interesting curio. If this was the case, one must also presume that it was then intact and thus more recognizably phallic, a suggestion supported by the freshness of the break, indicating that it had not been handled or left lying 4. Two additional seasons of excavation have now been completed without the discovery of the missing piece; no further fieldwork at Kaminoudhia is planned.

around for a long period prior to being discarded in the alleyway as an unwanted, fragmentary, and thus uninteresting object. The shape of Object S796 is comparable to a recently published object from a private collection, also attributed to the ceramic Neolithic (Flourentzos 2003a: figs. 10a–b). It is reported to come from a locality 10 km northeast of Polis. The line drawing prepared for this essay (fig. 3) and based on the published photograph shows a somewhat more realistic phallic representation—complete with the urethral opening—but one that seems to have lost some of the androgynous characteristics noted in connection with the Arkolies figurine (Swiny and Swiny 1983: 58). Flourentzos compares the piece in the private collection to the find from Arkolies and dates it to the Ceramic Neolithic (2003a: 16). An important parallel may be drawn with a Chalcolithic figurine from Erimi Pamboula carved out of picrolite (fig. 4), first published by Bolger (1988: 114, fig. 34:342, pl. 18) 53 years after it was excavated by Dikaios. Being of picrolite, which was collected as pebbles or small cobbles from the Kouris riverbed next to Erimi Pamboula, it was more diminutive, but the outline of the “body” is very similar to Object S796, and the incised lines defining the “legs” and “buttocks” are reminiscent of the Arkolies piece. Bolger refers to the Swiny and Swiny 1983 publication of the Arkolies find but incorrectly states that it was dated to the Chalcolithic rather than the Ceramic Neolithic period (Bolger 1988: 114). The object from Pamboula (5.5 cm high, 4.0 cm wide, and 2.6 cm thick) is abraded and lacking in a recognizable head or face, but the angle and proportions of the phallic neck are very similar to the piece from Arkolies.

Cypriot Anthropomorphic Figurines of a Certain Type Obviously, this class of figurine—which, when viewed from the side, resembles a sitting human form with its legs drawn up close to the buttocks and a long neck jutting out at an angle, and when observed from the front or back, an erect phallus and scrotum—had enduring symbolism in fifth- and fourth-millennia bce Cyprus. The deeper meaning of the blatant dualism escapes us, but human sexuality and presumably fertility surely have something to do with the figurines’ shape. It was suggested that the Arkolies figurine, when viewed from below, might also show female genitalia (Swiny and Swiny 1983: 58). If this is correct, then the figurine is truly androgynous, a suggestion endorsed by no less an authority than Gimbutas (1989: 232, fig. 359). And, finally, I also believe that the piece could be seen as representing a caprovid (Swiny and Swiny 1983: 58), in which case these representations could embody a remarkable number of messages. The phallic overtones of Cypriot Aceramic Neolithic figurines are well known (Morris 1985: 118, figs. 106–10), with some of the best examples coming from Khirokitia (Dikaios 1953: pl. 95). Recent discoveries of Neolithic phalli or phallic-shaped anthropomorphic objects are recorded from Parekklisha Schillourokambos (Guilaine 2003: 332, fig. 1:d; Stordeur 2003: 364, fig. 6) and Paralimni Nissia (Flourentzos 2003b: 80, fig. 5.2). Later Chalcolithic pieces are equally explicit (Peltenburg 1979: pl. 11:1), so the genre has a long history. Prehistoric phalli, in some instances very abstract and in others easily recognizable, are not unknown on Cyprus and seem to be more common in the Ceramic Neolithic and Chalcolithic periods. The Ceramic Neolithic settlement of Ayios Epiktitos Vrysi yielded several grooved stones up to 60 cm long, somewhat resembling phalli (Peltenburg 1989: fig. 15.2; Flourentzos 2003a: fig. 4). A more representational piece, but of cruder manufacture, came from the Chalcolithic cemetery of Souskiou Vathyrkakas. It is described as having been “broken at the base of the shaft” (Niklasson 1991: 223, fig. 91), so the original shape of the object is unknown, and there is no way of de-

185

Fig. 6. Reconstructed shape of Object S796 before breakage.

termining whether it belongs to the same class of highly stylized, dual-natured figurine. At this stage, little more can be said about these anthropomorphic figurines with phallic overtones, because the contexts and finds associated with these objects from Sotira Teppes and Ayios Epiktitos Vrysi do not allow us to take the argument much further. But as more excavations are undertaken, we can look forward to a better understanding of the meaning of these pieces, if more of them are found in a context suggestive of ritual or a context that imparted a special status to the individuals who had access to the structure. Peltenburg already suggested this connection when discussing the findspot and associated features of the Sotira Teppes piece and the Ayios Epiktitos Vrysi baetyls (Peltenburg 1989: 111–12).

References Bolger, D. 1988 Erimi-Pamboula: A Chalcolithic Settlement in Cyprus. British Archaeological Reports International Series 443. Oxford: British Archaeological Reports. Christou, D. 1989 The Chalcolithic Cemetery 1 at Souskiou-Vathyrkakas. Pp. 82– 94 in Early Society in Cyprus, ed. E. Peltenburg. Edinburgh: Edinburgh University Press. Dikaios, P. 1953 Khirokitia. Oxford: Oxford University Press.

1961 Sotira. University Museum Monograph. Philadelphia: University of Pennsylvania Museum. Flourentzos, P. 2003a H ERWTIKH ZWH STHN ARCAIA TECNH THZ KUPROU. Nicosia. 2003b Paralimni-Nissia: A Unique Neolithic Settlement in Cyprus. Pp. 73–83 in Le Néolithique de Chypre: Actes du Colloque International Organisé par le Département des Antiquités de Chypre et l’École Française d’Athènes, Nicosie, 17–19 Mai, 2001, ed. J. Guilaine and A. Le Brun. Bulletin de Correspondance Hellénique Supplement 43. Athens: École Française d’Athènes.

186 Gimbutas, M. A. 1989 The Language of the Goddess: Unearthing the Hidden Symbols of Western Civilization. London: Thames & Hudson. Guilaine, J. 2003 Objects “symboliques” et parures de Parekklisha-Shillourokambos. Pp. 329–40 in Le Néolithique de Chypre: Actes du Colloque International Organisé par le Département des Antiquités de Chypre et l’École Française d’Athènes, Nicosie, 17–19 Mai, 2001, ed. J. Guilaine and A. Le Brun. Bulletin de Correspondance Hellénique Supplement 43. Athens: École Française d’Athènes. Karageorghis, V. 1982a Annual Report of the Director of the Department of Antiquities. Nicosia: Department of Antiquities. 1982b Chronique des Fouilles et Découvertes Archéologiques à Chypre en 1981. Bulletin de Correspondance Hellénique 106: 685–751. Morris, D. 1985 The Art of Ancient Cyprus. Oxford: Phaidon. Niklasson, K. 1991 Early Prehistoric Burials in Cyprus. Studies in Mediterranean Archaeology 96. Jonsered: Åströms.

Stuart Swiny Peltenburg, E. 1979 The Prehistory of West Cyprus: Ktima Lowlands Investigations 1976–1978. Report of the Department of Antiquities, Cyprus 1979: 69–99. 1989 The Beginnings of Religion in Cyprus. Pp. 108–26 in Early Society in Cyprus, ed. E. Peltenburg. Edinburgh: Edinburgh University Press. Stordeur, D. 2003 De la vallée de l’Euphrate à Chypre? À la recherche d’indices de relations au Néolithique. Pp. 353–71 in Le Néolithique de Chypre: Actes du Colloque International Organisé par le Département des Antiquités de Chypre et l’École Française d’Athènes, Nicosie, 17–19 Mai, 2001, ed. J. Guilaine and A. Le Brun. Bulletin de Correspondance Hellénique Supplement 43. Athens: École Française d’Athènes. Swiny, H. W., and Swiny, S. 1983 An Anthropomorphic Figurine from the Sotira Area. Report of the Department of Antiquities, Cyprus 1983: 56–59. Swiny, S. 1989 Prehistoric Cyprus: A Current Perspective. Biblical Archaeologist 52: 178–89.

Women and the Worship of Yahweh in Ancient Israel Susan Ackerman Department of Religion, Dartmouth College Hanover, New Hampshire

According to both the books of Kings and Chronicles, a seminal event in the history of Israelite religion was the reformation movement promulgated by Judah’s King Josiah during the last third of the 7th century bce. 1 Although modern scholars have cautioned that Josiah’s reform efforts may have been neither as sweeping nor as lasting as our texts suggest (see Lohfink 1987 and the references in Albertz 1992: 345, n. 5), they have generally tended to agree with the biblical accounts’ overall assessment. Cogan and Tadmor, for example, characterize features of the Josianic reform as marking “a new departure in the history of Judah” (1988: 296). My purpose in this essay is to examine what effects this “new departure” may have had on the participation of women in the cult of the Israelite national god, Yahweh. Because my deeply beloved colleague Bill Dever has published an important article on Josiah’s reforms as described in 2 Kings 23 (1994a), as well as several articles and book chapters discussing the worship of the goddess Asherah that is condemned in the 2 Kings passage and its Chronicles counterpart (e.g., 1982; 1983; 1984; 1990: 140–49; 1994b: 112–13, 121–22; 1995: 40–42; 1997; 1999; 2001: 173–98; 2002: 26–27), it is a pleasure and indeed an honor to dedicate this essay to him as part of this Festschrift’s celebration of his long and distinguished scholarly career.

Women and the Production of Cultic Textiles Women are explicitly mentioned only once in the accounts of Josiah’s reforms found in 2 Kgs 23:4–25 and 2 Chr 34:3–7, namely, in 2 Kgs 23:7, where we are told that, as part of his efforts to purge forms of worship that he 1. The dates associated with Josiah’s reformation program are reported differently in Kings and in Chronicles. According to 2 Kings, Josiah’s reformation program began with the finding of the so-called “Book of the Law” in Year 18 of his reign (i.e., 622 bce) and culminated with the great Passover celebration held later that same year (2 Kgs 22:3, 23:22–23). According to 2 Chr 34:3, Josiah began his reformation efforts already in ca. 628 bce, six years prior to the finding of the “Book of the Law” in 622 bce and the great Passover celebration of that year (2 Chr 34:8, 35:19). Scholars often assume that the Chronicles chronology is the more reliable, although parts of it may be tendentiously shaped (see Cogan 1980; Cogan and Tadmor 1988: 298).

found apostate from the Jerusalem temple, Josiah destroyed the houses that were within the temple compound “where the women wove garments for Asherah” 2— houses, that is, in which women wove clothing to drape over a cult statue dedicated to the goddess Asherah that stood somewhere within the temple complex. I do take MT lªsrh to be the proper name of the ancient Canaanite mother goddess Asherah in this verse; the alternative is to read—as the Masoretic pointing (which includes the definite article) suggests—a reference to the cult object known as the ªåserâ. But because I maintain, along with most commentators, that Asherah’s adherents understood this cult object, which was in the shape of a stylized wooden pole or tree (Deut 16:21; 1 Kgs 14:15, 23; 16:33; 2 Kgs 17:10, 16; 21:3; 2 Chr 33:3), to represent the goddess in the cult (although cf. Smith 1990: 80–94; see also the useful survey of scholarly positions in Day 1986: 398–404), and because I further maintain that most ancient Israelites made little differentiation between a deity’s cultic representation and the actual deity (Olyan 1988: 31–32; Dever 1999: 11*; Ackerman 2003: 457–58), I argue that, even if one were to follow the Masoretic pointing and read 2 Kgs 23:7 as referring to the ªåserâ cult object, the text refers ultimately to a statue representing Asherah herself. I count myself among those scholars who claim, moreover, that Asherah’s cult image was being adorned—and therefore the goddess Asherah was being worshiped—in the Jerusalem temple during Josiah’s reign because many in ancient Israel (including, presumably, the temple’s priests) understood this goddess to be the consort of the Israelite national god Yahweh who was therefore to receive devotions along with Yahweh in Yahweh’s national sanctuary. Like many colleagues who have reached this conclusion, I have found the inscriptional evidence pairing Yahweh with Asherah from the sites of Kuntillet ºAjrûd and Khirbet el-Qôm to be particularly significant (Dever 2. I read in this verse battîm, cognate with Arabic batt ‘woven garment’, for the MT battîm, the plural of bayit ‘house’, a reading originally proposed by Sanda (1912: 344; pointed out in Day 1986: 407) and by Driver (1936: 107; pointed out in Gray 1964: 664, note b; and in Cogan and Tadmor 1988: 286). This revocalization is supported by the Lucianic recension of the LXX, which reads stolas ‘garment, robe’.

190 1969–70; Meshel 1976; 1978a; 1978b; 1979; Meshel and Meyers 1976; Lemaire 1977; 1984a; 1984b). 3 Equally significant, I have found, are the many reevaluations of the 40 or so biblical texts mentioning the ªåserâ cult object or the goddess Asherah that have been inspired by the Kuntillet ºAjrûd and Khirbet el-Qôm discoveries. Olyan has argued, for example, that while earlier interpretations of texts that mention Asherah worship in Jerusalem (e.g., 1 Kgs 15:13; 2 Kgs 18:4; 23:4, 6–7, 14) saw in these passages evidence of aberrant and heterodox religious practices imported into Judah from elsewhere, the Kuntillet ºAjrûd and Khirbet el-Qôm evidence suggests instead that for some ancient Israelites, at least at some points during their history, Asherah worship was a perfectly appropriate part of the Yahwistic cult, including the cult observed in Jerusalem and in the Jerusalem temple (Olyan 1988: 9). Olyan similarly proposes that a reevaluation of the biblical record suggests that for many, Asherah worship was a perfectly appropriate part of the Yahwistic cult practiced in Northern Kingdom shrines. He argues, for example, that the pairing of Asherah with “Yahweh of Samaria” in one of the Kuntillet ºAjrûd inscriptions, together with the biblical texts that attest that an ªåserâ cult object stood in Samaria, demonstrate that Asherah was worshiped alongside Yahweh in a Yahwistic temple in the Northern capital city. As in the South, moreover, this seems to have been considered a perfectly appropriate part of Yahwistic tradition, acceptable, indeed, to even the most zealous of worshipers: for example, the reformer King Jehu, who allowed the ªåserâ cult object to remain standing in Samaria, while otherwise purging all non-Yahwistic worship from the city (Olyan 1988: 6–7, 32–35). In Yahweh’s shrine at Bethel as well, there seems to have been an ªåserâ (at least there was in the late 7th century bce, according to 2 Kgs 23:15), indicating that both Bethel’s Yahwistic priests and the Yahwists who worshiped at that sanctuary found Asherah’s presence alongside Yahweh acceptable. Moreover, the Northern prophets Amos and Hosea, who otherwise derided many

3. One must admit, to be sure, that the phrase ‘his [Yahweh’s] ªåserâ’ found in these corpora is grammatically difficult. For some, the solution has been to understand ªåserâ in the inscriptions as a common noun, rather than the divine appellation Asherah, so as to avoid the grammatical improbability of a pronominal suffix’s being affixed to a proper name. Commentators have thus proposed translating ªåserâ as ‘shrine’ (Lipinski 1972); as ‘the female aspect of Yahweh’, reified and given a separate identity (Miller 1986: 246; McCarter 1987: 149); or as the ªåserâ cult object. Because ‘cult object’ is the most commonly attested meaning for ªåserâ in the Hebrew Bible, this translation would be my preference among these three alternatives, but I would again argue that, because the ancient Israelites understood a deity’s cult symbol and the deity to be essentially the same, we can and should understand the Kuntillet ºAjrûd and Khirbet el-Qôm inscriptions as ultimately referring to Yahweh’s goddess consort.

Susan Ackerman aspects of the Bethel cult (e.g., Hos 8:5–6, 10:5–6, 13:2; Amos 3:13–15, 4:4–5, 5:4–5, 7:10–13), made no complaint about Bethel’s ªåserâ, which at least “indirectly suggests approval and legitimacy” (Olyan 1988: 7). Unfortunately, none of these biblical texts other than 2 Kgs 23:7 makes any mention of the practice of clothing the ªåserâ cult statue. Still, there are grounds for assuming that this was a typical aspect of Asherah worship. The clothing of cult statues is mentioned elsewhere in the Bible in Jer 10:9 (Gressmann 1924: 325–26, pointed out by Cogan and Tadmor 1988: 286), and Jeremiah gives no indication that he sees this as anything but normal in cults that use divine images. Much the same is implied in Ezek 16:18 (Dick 1999: 19, note j). Indeed, the making of garments for divine images is a well-known practice in other ancient Near Eastern and eastern Mediterranean communities, attested in sources from Mesopotamia (Oppenheim 1949: 172 [pointed out in Cogan and Tadmor 1988: 286]; 1977: 184; Matsushima 1993: 211–18), Greece (Barber 1992: 106, citing Mansfield 1985: 443), and Egypt (Lorton 1999: 133–45). At least one set of these materials, moreover, indicates that, as in 2 Kgs 23:7, it was women who undertook the responsibility for producing a cult statue’s garments: these are the many records from Classical Greece that document how select Athenian women wove and offered to the goddess Athena the new robe or peplos that was draped over her cult statue as the central event of the annual Panathenaia celebration (Barber 1992: 112–17; Lefkowitz 1996: 79; Katz 2000: 515). Katz further notes that as part of the Panathenaia, “[t]he girls and women of Athens . . . undertook the ceremonial cleansing of the goddess’s cult statue” (2000: 515), and we might thus speculate that the women who served Asherah in the Jerusalem temple similarly gave ritual baths to Asherah’s cult image. Certainly, the tradition of bathing cult statues, like the tradition of clothing them, is well known from elsewhere in the ancient Near East. In addition, our ancient Near Eastern sources, as well as sources from Greece, make clear that the role of those who bathed and clothed the cult image, like the role of the priests who took responsibility for feeding the gods, was of major significance within a deity’s religious community (Matsushima 1993: 216–18). Indeed, in Athens, “[t]he arrhephoroi [the young girls chosen to help set the warp of Athena’s peplos] were considered sufficiently important to be housed, during the term of their service, near the temple of Athena Polias on the Acropolis” (Lefkowitz 1996: 79). The parallel with 2 Kgs 23:7 is striking. What, then, is the effect of Josiah’s decision in 2 Kgs 23:7 that the houses within the temple grounds in which women wove garments for Asherah’s cult statue must be destroyed? At a minimum, obviously, Josiah is displacing

Women and the Worship of Yahweh in Ancient Israel

191

these women and their work from what I believe we must assume was an important and honored location within the temple compound. More significantly, however, Josiah’s actions have the effect of prohibiting women from performing a cultic function within the Jerusalem temple that, our comparative evidence suggests, was an important and honored task within Israelite religion. Even more significantly, Josiah’s reforms work to remove women from the religiously important and honored positions they may have held as Asherah’s cultic weavers elsewhere in ancient Israel, and, most significantly of all, his efforts ultimately have the effect of removing Israelite women from their positions as Asherah’s weavers both in Jerusalem and elsewhere for good. For not only is Josiah described in 2 Kings 23 as displacing Asherah’s women weavers from their houses within the temple complex; he is described as acting aggressively to remove any trace of Asherah worship from the temple, from Jerusalem, and throughout Israelite territory. Thus, Josiah is said to have removed the cult image of Asherah from the temple and to have burned it at the Wadi Kidron (2 Kgs 23:6), to have removed from the temple all the vessels that would have been used to make offerings to Asherah (2 Kgs 23:4), and to have torn down the sacred images of Asherah that stood east of Jerusalem, immediately south of the Mount of Olives (2 Kgs 23:14). So zealously, in fact, does Josiah seek to eliminate Asherah worship throughout Israel that he is described as crossing over the border of Judah to go to the old Northern Kingdom shrine of Bethel to destroy the ªåserâ cult object that stood there (2 Kgs 23:15). Moreover, while I believe that we have some evidence that Asherah worship survived Josiah’s reformation efforts and so reemerged after his death (e.g., Jer 17:2, Ezek 8:3–5, Isa 65:3 [on the last two, see Ackerman 1992: 55–66, 185–94]), it is sporadic and sparse, and it completely disappears by the early postexilic period. Josiah, that is, may not have been as immediately successful in eradicating Asherah worship among the Israelites as he would have wished, but his goal was certainly realized within the span of two or three generations. By the last quarter of the 6th century bce, no opportunities remained for Israelite women to serve Asherah as weavers or in any other capacity. To be sure, no opportunities remained for men to serve Asherah either, because the priests, for example, presumably would have made the offerings to Asherah in the Jerusalem temple that are alluded to in 2 Kgs 23:4. I would argue that such men, however, would still have been able to serve as priests in the absence of an Asherah cult, because of the Israelites’ understanding that Asherah was Yahweh’s consort and was therefore worshiped in conjunction with the male god. The priests, that is, who served Asherah would of necessity also have been servants of her

divine husband, Yahweh, and so, with the elimination of the goddess’s cult, would simply have begun to offer their religious service to Yahweh alone. Note in this regard that, while Josiah is said in 2 Kgs 23:5 to have deposed the priests of the god Baal, as well as those who made offerings to the sun, the moon, the constellations, and all the host of heaven, there is no mention of his deposing the priests of Asherah when her cult was removed from the Jerusalem temple. Rather, I would suggest, the priests who served Yahweh’s female consort, as opposed to those who served Baal and other gods who were Yahweh’s rivals, were readily able to transfer their service from Asherah’s cult to the cult of Yahweh. But because of the aniconic traditions of Yahwism, which are insisted upon in the biblical text and seem generally to be confirmed by the absence of identifiable Yahweh images in the archaeological record (Hendel 1988: 366–68; 1997; Mettinger 1995; Dever 1996: 94; Sasson 2002; cf. Schmidt 1995; van der Toorn 2002: 47–51), the work done by the women weavers who served Asherah could not as easily be relocated within Yahwistic tradition. In the Yahwism without Asherah that Josiah promoted, no cult statue remained to be clothed. The Yahwistic cult, however, did use woven goods and additional kinds of textiles for many other purposes: for priestly vestments, for the curtains and other draperies that hung in and around Yahweh’s sanctuary, and to cover the table of the bread of presence. We have some evidence, moreover, that women may have participated in the production of these cultic fabrics. In Exod 35:25–26, every ‘skillful woman’ (kol-ªissâ ˙akmat-leb) among the people encamped at the base of Mount Sinai is said to have brought the colored yarns, fine linen, and spun goats’ hair she had produced as offerings to be used in the making of “the tent of meeting, and for all its service, and for the sacred vestments” (Exod 35:21; see Meyers 2000b). In Exod 36:1, moreover, we are told that kol-ªîs ˙akam-leb “to whom Yahweh has given skill and understanding to know how to do all the work in the making of the sanctuary shall work in accordance with all that Yahweh has commanded.” Although the Hebrew here translates literally as ‘every skillful man . . . shall work’, we can just as easily translate ‘every skillful individual’, which would suggest that this verse alludes back to the “skillful women” mentioned in Exod 35:25–26, who are seen as capable of producing the various textile products for the tabernacle and its service that Yahweh has commissioned. Similarly, in Exod 36:8, we are told that “all those with skill among the workers made the tabernacle with ten curtains,” which certainly can be interpreted to mean that the women described in Exod 35:25–26 as skilled in textile production were seen as helping to make the tabernacle’s elaborate collection of draperies.

192 The fact remains, however, that in the Exodus accounts of the construction of the tabernacle and all its furnishings, the only explicit mention of women’s participation in the creation of cultic textiles is the description in Exod 35:25– 26 of them bringing what we might consider “raw materials” as offerings for other fabricators to use. When the actual making of the tabernacle’s textile inventory is described, it is men who are more readily cited as doing this work. As we have seen above, for example, Exod 36:1, if translated literally, refers only to skilled men in designating those who will do the work of constructing Yahweh’s sanctuary. Subsequently, once the construction begins, it is Bezalel, the chief of the two main craftsmen in charge of the construction, who is said to make the curtains of goats’ hair that form the tent that was draped over the tabernacle’s wooden frame (36:14) and who likewise is said (in 36:35) to make an elaborately decorated curtain that seems meant to be hung at the entrance of the inner sanctum of the tabernacle in which the ark was housed. Similarly, in 36:37, Bezalel makes an elaborate cloth screen that is to stand at the entrance of the tabernacle’s outer chamber. Also, according to Exod 39:2, 8, and 22, he makes the priestly ephod, breastplate, and robe of the ephod using colored yarns and fine linen. He is said to have made as well, along with the sanctuary’s other master craftsman, Oholiab, the linen tunics, turbans, headdresses, undergarments, and sashes worn by Aaron and his priestly descendants (39:27–29). Both Bezalel (in 35:35) and Oholiab (in 35:35 and again in 38:23) are moreover lauded as skilled in doing embroidery work in colored yarns and in fine linen and (in 35:35) in doing the work of weaving. All of these Exodus materials, as is well known, come from the 6th/5th-century bce Priestly source, and it may well be that, in stressing Bezalel’s and Oholiab’s skill in textile work, they signal (in seeming contrast to the preexilic evidence found in 2 Kgs 23:7) a postexilic bias toward having men assume the primary and perhaps ultimately the exclusive responsibility for cultic weaving and other types of cultic textile production. In the postexilic books of Chronicles, too, cultic textile production is presented as a responsibility assumed by men, with the Phoenician artisan Hurum-abi, who had been sent from Tyre to aid in Solomon’s temple-building project, described as able to “work . . . in purple, blue, and crimson fabrics and fine linen” (2 Chr 2:13 [2:14 in most of the English versions of the Bible]; see also 2 Chr 2:6 [2:7 in English versions]). The actual creation of cultic fabrics in the 2 Chronicles temple-building account, however, is ascribed to Solomon himself (3:14). There are no counterparts to these verses in the earlier book of Kings’ temple-building account, which does not mention any fabrics at all. As commentators have pointed out, these details, among others, were added in the

Susan Ackerman Chronicles text in order to suggest parallels between Solomon’s temple and the Exodus description of the tabernacle sanctuary (e.g., Japhet 1993: 545). But what we have to ask for our purposes is whether the emphasis in Chronicles—like Exodus’s P materials—on the role of men in cultic textile production indicates that Chronicles shares P’s seeming bias that it is men who should properly and perhaps exclusively assume responsibility for the fabrication of cultic cloth. Indeed, does Chronicles, by ignoring the Exodus account’s preliminary mention of women’s role in generating the raw materials used for cultic fabric, suggest an even stronger bias that cultic textile production should be a responsibility assumed exclusively by men than Exodus’s P materials? In the same vein, we must also ask why Chronicles, in its revisions of the 2 Kings account of Josiah’s reform, omits any mention of Josiah’s deposing of the women who wove cultic garments for Asherah, even though it otherwise liberally cites Josiah’s attempts to eradicate Asherah worship in Israel (2 Chr 34:3–4, 7). Is this omission a coincidence? Is it explained by Chronicles’ general tendency to see Josiah’s reforms as concerned only with a purge of the cults considered non-Yahwistic that were found outside of the temple, with the temple having already been purified, according to Chronicles’ chronology, during Manasseh’s reign (2 Chr 33:15–16; see Japhet 1993: 1019)? Or might we again be seeing in the Chronicles’ omission a postexilic bias (in seeming contrast to the preexilic evidence of 2 Kgs 23:7) that men should assume the primary and perhaps ultimately the exclusive responsibility for cultic weaving and other types of cultic textile production? It is, unfortunately, impossible to say for sure, and arguments from silence are of course notoriously weak. Still, if the last of the three explanations advanced above has any merit, then we have to ask again about the effects that certain aspects of Josiah’s reform efforts had on women’s participation in the cult of Yahweh. If one of the effects was to displace women from their service as weavers in the cult of Asherah, and if there was a roughly concurrent tendency to assign the primary and perhaps exclusive responsibility for textile production to men in the associated cult of Yahweh, then what is ultimately lost to women through the agency of Josiah’s reform is an important and honored role as fabricators of cultic cloth within ancient Israelite religion.

Women and Music Athenian women, in addition to weaving the peplos for the annual Panathenaia festival and undertaking the ceremonial cleansing of the goddess’s cult statue, participated in dances in Athena’s honor in the course of the Panathenaic celebration (Katz 2000: 515; Lefkowitz 1996:

Women and the Worship of Yahweh in Ancient Israel

193

79). Various other scholars and I have similarly documented women’s roles in festal dancing and, more generally, music-making within the cult of Yahweh. I have described, for example, the special role women seemed to play as dancers and more generally music-makers during the fall harvest festival of Succoth (Ackerman 1998: 253– 87). Poethig, followed by Meyers, has likewise described the special role of women in performing the victory hymns that celebrate Yahweh’s triumphs in holy war (Exod 15:20–21; Judg 5:1–30, 11:34; 1 Sam 18:6–7; see Poethig 1985; Meyers 1991; 1999; 2000a). Meyers has also noted that in Ps 68:25 women playing hand drums are among the musicians participating in a procession into Yahweh’s sanctuary in order to celebrate God as a divine warrior who has vanquished Israel’s enemies (2000a: 190), and she further argues that, because hand drums are an instrument characteristically played by women, their mention in Ps 81:2 in the context of an exhortation to make music at the festivals of the new moon, the full moon, and other festal days also demonstrates women’s role as music-makers within the Yahwistic cult (1991; 2000a: 190). The references to hand drums along with other musical instruments again in Pss 149:3 and 150:3–5, both hymns urging that praise be sung to God, provides yet more evidence suggesting that women participated in Yahwistic musical performances; Psalm 150, moreover, like Ps 65:25, explicitly places its musicians, including, by implication, its women musicians, within Yahweh’s temple compound (Meyers 2000a: 190). We can finally note that, while not as explicitly cultic, women’s musical performance is well attested as a part of certain life-cycle rituals, especially weddings, funerals, and other occasions of lamentation (Judg 11:37– 40, 2 Sam 1:24, 2 Chr 35:25, Ps 78:63, Jer 9:16–20 [9:17– 21 in English versions], Ezek 32:16; on lamentation, see Bird 1991: 106). It is striking from the point of view of our inquiry, however, that almost all of the texts attesting to women’s cultic musicianship cited above come from the preexilic or early exilic period (although the Psalms, it must be granted, are notoriously difficult to date). When we turn to look for evidence of women’s role as cultic musicians in the late exilic or postexilic period, it is much more difficult to find. We do still have one text that attests to women’s participation in rituals of lamentation, although even this text, 2 Chr 35:25, seems to stress women’s role as mourners less than its preexilic counterparts by describing how both “the singing men and the singing women” chanted the laments sung upon the death of King Josiah (cf. Jer 9:16 [9:17 in English versions], in which only “mourning women” are summoned to come to sing a dirge over the destruction of Jerusalem). Meyers speculates that the addition of men to the retinue of mourners in Chronicles is due to the fact that

“the mourning for Josiah, one of the few kings considered a righteous ruler, called for an extraordinary outpouring of mourning, which would have involved professional singers of both genders” (2000c: 284). My own sense, however, is that what may have happened to women’s roles as cultic weavers in the aftermath of the Josianic reform also happened with regard to women’s roles as cultic and quasi-cultic musicians: namely, a responsibility assumed by women in the prereformation period became primarily and eventually perhaps even exclusively available to men. More specifically, at least according to the books of Chronicles, the responsibility for music-making in the Yahwistic cult came to lie solely with the male members of certain clans of the tribe of Levi (e.g., 1 Chr 15:16–24; 16:4, 41–42). The kindred books of Ezra and Nehemiah also seem to describe the temple’s singers as men from various Levite clans, although these texts are not unambiguous (cf., e.g., Ezra 2:41 and 2:70). Ezekiel apparently also believes that the singers in the temple should be male Levites (Ezek 40:44–46), although he, in contrast to Chronicles and the admittedly ambiguous books of Ezra and Nehemiah, envisions the temple’s levitical musicians as tracing their lineage back both to the Solomonic-era priest Zadok and to other ancestors within the levitical line. Chronicles, conversely, along with multiple texts in Ezra and Nehemiah, suggests that the temple’s musicians are drawn solely from the Levites’ non-Zadokite lineages. Particularly noteworthy according to the Chronicles-EzraNehemiah tradition is the line of Asaph (see, e.g., 1 Chr 16:7, 2 Chr 35:15, Ezra 2:41, Neh 11:22), and, consequently, a significant corpus of hymns within the Psalter (Psalms 50, 70–83) is attributed to him. Determining the preexilic history of the Levites has proved a daunting task for scholars. Nevertheless, a consensus seems to have emerged that the “priests of the towns of Judah” mentioned in 2 Kgs 23:8 are members of a non-Zadokite line or lines of Levites who had been serving in shrines in the Judean countryside, while priests who traced their ancestry back to Zadok (and ultimately to Aaron) had, since the time of Solomon, been serving in the temple in Jerusalem (Cross 1973: 195–215, 237–38). Josiah, however, whose reforms, all scholars agree, were associated in some way with the discovery in the temple of a version of the book of Deuteronomy (called in both Kings and Chronicles the “Book of the Law”), 4 follows 4. The exact relationship between the finding of the “Book of the Law” and Josiah’s program of reform is described differently in Kings and Chronicles. In Kings, some fairly modest attempts at religious rejuvenation (i.e., a series of repairs to the temple) seem to have been underway before the reform had begun, but the full-scale reform—that is, the community’s making a covenant that committed it to obeying the terms of the newly discovered law book, purging the cult to bring it into accord

194 the dictates of Deuteronomy (e.g., Deut 12:2–28) in promoting the centralization of Yahwistic worship in a place God has chosen “for the dwelling of his name,” that is, for Josiah, the temple in Jerusalem. Josiah, therefore, at least according to 2 Kgs 23:8 (material not paralleled in Chronicles), “brought all of the priests out of the towns of Judah” and to the temple in Jerusalem. Deuteronomy further mandates that a Levite who leaves “wherever he has been residing in Israel and comes to the place that Yahweh will choose . . . may minister in the name of Yahweh his God, like all his fellow Levites who stand there before Yahweh” (Deut 18:6–7), which is to say, in the language of Deuteronomy, that any Levite may serve Yahweh in God’s central sanctuary in the same way as all his fellow Levites serve, whether these Levites be of the line of the priest Zadok or not. According to 2 Kgs 23:9, however, the nonZadokite Levites brought to Jerusalem by Josiah “did not come up to the altar of Yahweh”; they were, that is, not consecrated to offer the animal sacrifices to God that were the centerpiece of Yahwistic ritual. Instead, according to 2 Kgs 23:9, they “ate unleavened bread among their kindred,” which may mean that they were allowed to make grain offerings along with other non-Zadokite Levites as part of the temple ritual or that they, along with these nonZadokite Levites and other temple personnel, consumed unleavened cakes during the great Passover celebration held in Jerusalem as a part of Josiah’s reformation efforts (Cogan and Tadmor 1988: 287). Yet however we interpret the specifics of 2 Kgs 23:9, the fact remains that according to this text, a substantial number of Levites were brought to the Jerusalem temple as a result of Josiah’s reformation efforts, but there they were deemed unable to perform the central ritual of Yahwistic animal sacrifice for which they had previously been responsible in countryside shrines. They consequently found themselves in want of something to do, as also the temple’s more established priesthood found itself in want of something to do with them. Although our sources are limited, I believe we must presume that the origins of the traditions that see the Levites as a class of second-order temple functionaries lie in this period. Certainly, within 50 years of the Josianic reforms, in 573 bce (Ezek 40:1), Ezekiel enviwith Josiah’s/Deuteronomy’s vision of what Yahwism should be, and the celebration of a great Passover—occurs only after the “Book of the Law” is found and is inspired by it. In Chronicles, the beginning of the reformation efforts date back even before the time of Josiah, to the final years of the reign of his grandfather Manasseh (2 Chr 33:15–16), and many of Josiah’s own efforts to eliminate from Israelite religion all types of worship that he sees as non-Yahwistic, as well as his efforts to repair and restore the temple complex predate the finding of the book as well (2 Chr 34:3–13). According to Chronicles, what the book inspires is only a ceremony of covenant affirmation and an enormous celebration of the Passover in Jerusalem (see further n. 1 above).

Susan Ackerman sions the non-Zadokite Levites as responsible for only secondary temple rituals. They may not approach Yahweh’s altar or come near anything that has been consecrated to Yahweh: this privilege is reserved for Zadokite descendants alone (Ezek 44:13, 15–16). Instead, the Levites serve as gatekeepers, they slaughter sacrificial animals, and they stand before the people to serve them (Ezek 44:11). Moreover, they serve as singers, although in Ezekiel, as I have already noted, both non-Zadokite Levites and Zadokites assume this function (Ezek 40:44–46). Still, from Ezekiel’s preliminary attempts at status differentiation, it is but a short jump to the hierarchy that is established in Chronicles and also somewhat in certain P materials in the Pentateuch (e.g., Numbers 3–4) and in Ezra and Nehemiah, whereby Zadokite priests take responsibility for the central rituals of the Yahwistic cult, while the non-Zadokite Levites serve as the temple’s gatekeepers, musicians, and temple servants. Yet if Josiah’s reforms establish a trajectory that by the 5th and 4th centuries culminates in the conviction that it is the men of certain levitical lineages who are responsible for musical performance in the Yahwistic cult, then what of the earlier traditions that assigned important roles as music-makers within Yahwistic ritual to women? Our answer must be that women’s ability to participate in the cult as musicians is greatly diminished by the postexilic period. In some cases, as suggested above, this may mean that the rituals of music-making associated with mourning that were previously the responsibility primarily or perhaps even exclusively of women were reconceptualized so as to include both men and women (2 Chr 35:25). In other cases, I suggest, older traditions that described women as the actual makers of music at certain cultic celebrations came to be rendered in far more metaphorical language. For example, in Jer 30:23–24, amid a series of oracles (Jer 30:1– 31:40) that almost all commentators date (at least in the form in which they have come down to us) to the early exilic period (although published some time ago, the comments in Bright 1965: 284–87 may still be regarded as representative), Jeremiah speaks of how Yahweh, although now requiring that God’s people endure suffering, will eventually come like a storm against the Israelites’ enemies and destroy them. Following this triumph of Yahweh, the divine warrior, comes the same sort of victory celebration that we have already noted is described in older sources, a celebration in which women, accompanying themselves on hand drums, dance and sing hymns praising Yahweh’s saving acts (Exod 15:20–21; Judg 5:1–30, 11:34; 1 Sam 18:6–7). Still, Jeremiah, in drawing on these earlier traditions of women’s music-making after a victory in holy war, changes them not insignificantly by suggesting that the celebratory music is no longer made by actual

Women and the Worship of Yahweh in Ancient Israel

195

women, but by the nation Israel, personified as a woman (31:4). Thus, while the notion of women taking responsibility for the musical performance that follows a holy war is retained, the idea that it is real women who would actually go forth to sing and dance is eroded. To put the matter another way, we might ask of Jeremiah what is left, according to his vision, for the actual women of ancient Israel who would make the ritual music that celebrates Yahweh’s triumphs in battle? In the same vein, we might ask of Ezekiel, the Priestly tradition, and the ChroniclesEzra-Nehemiah corpus what is left, according to their vision, for the women of ancient Israel who would render service as musicians as a part of Yahwistic worship? Our evidence may well suggest that the opportunities available to these women in the aftermath of Josiah’s reforms were relatively few.

these three traditions, I have intimated above, should be understood as categorically different from the latter two, because women who worshiped Asherah did so as part of their devotion to the Israelite national god, Yahweh. The women who worshiped the Queen of Heaven, conversely, explicitly describe themselves as abandoning the worship of Yahweh in favor of this goddess, whom they perceived as more powerful (Jer 44:15–19), and I believe we must presume that the women who devoted themselves to Tammuz worship also saw themselves as abandoning Yahweh in favor of a rival deity (Ackerman 1992: 91). Quite commonly, this movement away from Yahwistic worship as described in the late-7th/early-6th-century bce texts of Jeremiah and the early-6th-century bce text of Ezekiel is understood as a response to the growing Babylonian threat faced by Judah from ca. 605 bce onward: as Judah became more and more endangered by Babylonian military might, the people sought more and more to protect themselves by offering their devotions to gods either wholly (in the case of Tammuz) or partially (in the case of the Queen of Heaven) of Babylonian provenance. I do not necessarily mean to disagree with this explanation, but I nevertheless find it quite curious that it is women especially who are portrayed in the late 7th and early 6th centuries bce as abandoning the worship of Yahweh in favor of other deities (although cf. Ezek 8:14). I also find it quite curious that, despite the two centuries of prophetic hyperbole condemning women as apostate cited above, it is only in the late 7th and early 6th centuries bce that we have solid evidence of women worshiping other gods rather than Yahweh. In light of the analysis I have offered in this essay, I am moved to ask whether the devotions offered to other gods by these late-7th- and early-6thcentury women, in addition (perhaps) to being inspired by the threat of the Babylonian invasion, were a response to certain aspects of Josiah’s reformation efforts. If the Josianic reform ultimately did have the effect of distancing or even removing women from some important and honored functions they had earlier performed within the cult of Yahweh—the manufacture of cultic textiles for Yahweh’s goddess consort and/or the making of cultic music as part of Yahwistic ritual celebrations—then the women of the post-Josianic era may well have felt themselves compelled to abandon Yahwistic worship to find more meaningful venues for religious expression elsewhere.

Conclusions In the biblical text, especially in the prophetic books, women are often derided as responsible for behaviors that the prophets deem apostate in the religion of Israel (e.g., Hosea 1–3, 4:13; Amos 4:1–3; Isa 3:16–24, 4:1, 32:9–14). A variant motif personifies Jerusalem, also called Zion, as a female apostate or “harlot” who represents the failure of the entire Yahwistic community to live up to the religious standards required by the prophets (Isa 1:21–23, 3:25–26, 57:3, 6–13; Jer 4:30, 13:20–27, 22:20–23; Lam 1:8–9; Ezekiel 16; 23:1–4, 11–49). In particular, these texts tend to condemn women, or the woman Jerusalem, for worshiping gods other than the Israelite national god, Yahweh. Ironically, however, when we look beyond this corpus of prophetic diatribe, which dates from the 8th to 6th centuries bce, for actual descriptions of women worshiping gods other than Yahweh, we find only three sets of materials: (1) descriptions of women who served the goddess Asherah (e.g., 1 Kgs 15:13, 2 Kgs 23:7); (2) descriptions of women who participated in the cult of a goddess known only as the Queen of Heaven (Jer 7:16–20; 44:15–19, 25), probably a syncretistic deity incorporating elements of both East Semitic Ishtar and West Semitic Astarte (Ackerman 1992: 20–35); and (3) descriptions of women who sat at the northern entrance of the Jerusalem temple’s inner courtyard to mourn the god Tammuz’s death that is described in Mesopotamian myth (Ezek 8:14). The first of

References Ackerman, S. 1992 Under Every Green Tree: Popular Religion in Sixth-Century Judah. Harvard Semitic Monographs 46. Atlanta: Scholars Press. Reprinted Winona Lake, IN: Eisenbrauns, 2001.

1998 Warrior, Dancer, Seductress, Queen: Women in Judges and Biblical Israel. Anchor Bible Reference Library. New York: Doubleday.

196 2003 At Home with the Goddess. Pp. 455–68 in Symbiosis, Symbolism, and the Power of the Past: Canaan, Ancient Israel, and Their Neighbors from the Late Bronze Age through Roman Palaestina, ed. W. G. Dever and S. Gitin. Winona Lake, IN: Eisenbrauns. Albertz, R. 1992 From the Beginnings to the End of the Monarchy. Vol. 1 of A History of Israelite Religion in the Old Testament Period. Louisville, KY: Westminster John Knox. Barber, E. J. W. 1992 The Peplos of Athena. Pp. 103–17 in Goddess and Polis: The Panathenaic Festival in Ancient Athens, ed. J. Neils. Hanover, NH: Hood Museum of Art, Dartmouth College. Bird, P. A. 1991 Israelite Religion and the Faith of Israel’s Daughters: Reflections on Gender and Religious Definition. Pp. 97–108 in The Bible and the Politics of Exegesis: Essays in Honor of Norman K. Gottwald on His Sixty-Fifth Birthday, ed. D. Jobling, P. L. Day, and G. T. Sheppard. Cleveland, OH: Pilgrim. Bright, J. 1965 Jeremiah: A New Translation with Introduction and Commentary. Anchor Bible 21. Garden City, NY: Doubleday. Cogan, M. 1980 Tendentious Chronology in the Book of Chronicles. Zion 45: 165–72. Cogan, M., and Tadmor, H. 1988 II Kings. Anchor Bible 11. Garden City, NY: Doubleday. Cross, F. M. 1973 Canaanite Myth and Hebrew Epic: Essays in the History of the Religion of Israel. Cambridge: Harvard University Press. Day, J. 1986 Asherah in the Hebrew Bible and in Northwest Semitic Literature. Journal of Biblical Literature 105: 385–408. Dever, W. G. 1969–70 Iron Age Epigraphic Material from the Area of Khirbet elKôm. Hebrew Union College Annual 40/41: 158–89. 1982 Recent Archaeological Confirmation of the Cult of Asherah in Ancient Israel. Hebrew Studies 23: 37–43. 1983 Material Remains and the Cult in Ancient Israel: An Essay in Archeological Systematics. Pp. 571–87 in The Word of the Lord Shall Go Forth: Essays in Honor of David Noel Freedman in Celebration of His Sixtieth Birthday, ed. C. L. Meyers and M. O’Connor. Winona Lake, IN: Eisenbrauns. 1984 Asherah, Consort of Yahweh? New Evidence from Kuntillet ºAjrûd. Bulletin of the American Schools of Oriental Research 255: 21–37. 1990 Recent Archaeological Discoveries and Biblical Research. Seattle: University of Washington Press. 1994a The Silence of the Text: An Archaeological Commentary on 2 Kings 23. Pp. 143–68 in Scripture and Other Artifacts: Essays on the Bible and Archaeology in Honor of Philip J. King, ed. M. D. Coogan, J. C. Exum, and L. E. Stager. Louisville, KY: Westminster John Knox. 1994b Ancient Israelite Religion: How to Reconcile the Differing Textual and Artifactual Portraits? Pp. 105–25 in Ein Gott allein? JHWH-Verehrung und biblischer Monotheismus im Kontext der israelitischen und altorientalische Religionsgeschichte, ed. W. Dietrich and M. A. Klopfenstein. Göttingen: Vandenhoeck & Ruprecht. 1995 “Will the Real Israel Please Stand Up?” Part II: Archaeology and the Religions of Ancient Israel. Bulletin of the American Schools of Oriental Research 298: 37–58. 1996 Review of T. N. D. Mettinger, No Graven Image: Israelite Aniconism in Its Ancient Near Eastern Context, Stockholm: Almqvist & Wiksell, 1995. Bulletin of the American Schools of Oriental Research 302: 93–94.

Susan Ackerman 1997 Folk Religion in Early Israel: Did Yahweh Have a Consort? Pp. 27–56 in Aspects of Monotheism: How God Is One, ed. H. Shanks. Washington, DC: Biblical Archaeology Society. 1999 Archaeology and the Ancient Israelite Cult: How the Kh. elQôm and Kuntillet ºAjrûd “Asherah” Texts Have Changed the Picture. Pp. 9*–15* in Eretz-Israel 26 (Frank Moore Cross volume), ed. B. A. Levine, J. Naveh, and E. Stern. Jerusalem: Israel Exploration Society. 2001 What Did the Biblical Writers Know and When Did They Know It? What Archaeology Can Tell Us about the Reality of Ancient Israel. Grand Rapids, MI: Eerdmans. 2002 Theology, Philology, and Archaeology: In the Pursuit of Ancient Israelite Religion. Pp. 11–33 in Sacred Time, Sacred Place: Archaeology and the Religion of Israel, ed. B. M. Gittlen. Winona Lake, IN: Eisenbrauns. Dick, M. B. 1999 Prophetic Parodies of Making the Cult Image. Pp. 1–53 in Born in Heaven, Made on Earth: The Making of the Cult Image in the Ancient Near East, ed. M. B. Dick. Winona Lake, IN: Eisenbrauns. Driver, G. R. 1936 Supposed Arabisms in the Old Testament. Journal of Biblical Literature 55: 101–20. Gray, J. 1964 I & II Kings. Old Testament Library. London: SCM. Gressmann, H. 1924 Josia und das Deuteronomium. Zeitschrift für die Alttestamentliche Wissenschaft 44: 313–37. Hendel, R. S. 1988 The Social Origins of the Aniconic Tradition in Early Israel. Catholic Biblical Quarterly 50: 365–82. 1997 Aniconism and Anthropomorphism in Ancient Israel. Pp. 205– 28 in The Image and the Book: Iconic Cults, Aniconism, and the Rise of Book Religion in Israel and the Ancient Near East, ed. K. van der Toorn. Leuven: Peeters. Japhet, S. 1993 I and II Chronicles: A Commentary. Old Testament Library. Louisville, KY: Westminster John Knox. Katz, M. A. 2000 Sappho and Her Sisters: Women in Ancient Greece. Signs 25: 505–31. Lefkowitz, M. 1996 Women in the Panathenaic and Other Festivals. Pp. 78–91 in Worshipping Athena: Panathenaia and Parthenon, ed. J. Neils. Madison, WI: University of Wisconsin Press. Lemaire, A. 1977 Les inscriptions de Khirbet el-Qôm et l’Ashérah de Yhwh. Revue Biblique 84: 597–608. 1984a Date et origine des inscriptiones hébraïques et phéniciennes de Kuntillet ºAjrud. Studi epigraphici e linguistici 1: 131–43. 1984b Who or What Was Yahweh’s Asherah? Biblical Archaeology Review 10/6: 42–51. Lipinski, E. 1972 The Goddess Atirat in Ancient Arabia, in Babylon, and in Ugarit. Orientalia lovaniensia analecta 3: 101–19. Lohfink, N. 1987 The Cult Reform of Josiah of Judah: 2 Kings 22–23 as a Source for the History of Israelite Religion. Pp. 459–75 in Ancient Israelite Religion: Essays in Honor of Frank Moore Cross, ed. P. D. Miller, P. D. Hanson, and S. D. McBride. Philadelphia: Fortress. Lorton, D. 1999 The Theology of Cult Statues in Ancient Egypt. Pp. 123–210 in Born in Heaven, Made on Earth: The Making of the Cult Image in the Ancient Near East, ed. M. B. Dick. Winona Lake, IN: Eisenbrauns.

Women and the Worship of Yahweh in Ancient Israel

197

Mansfield, J. M. 1985 The Robe of Athena and the Panathenaic Peplos. Unpublished Ph.D. dissertation, University of California at Berkeley. Matsushima, E. 1993 Divine Statues in Ancient Mesopotamia: Their Fashioning and Clothing and Their Interaction with Society. Pp. 209–19 in Official Cult and Popular Religion in the Ancient Near East: Papers of the First Colloquium on the Ancient Near East—The City and Its Life—Held at the Middle Eastern Cultural Center in Japan (Mitaka, Tokyo), March 20–22, 1992, ed. E. Matsushima. Heidelberg: Carl Winter. McCarter, P. K. 1987 Aspects of the Religion of the Israelite Monarchy: Biblical and Epigraphic Data. Pp. 137–55 in Ancient Israelite Religion: Essays in Honor of Frank Moore Cross, ed. P. D. Miller, P. D. Hanson, and S. D. McBride. Philadelphia: Fortress. Meshel, Z. 1976 Kuntillet ºAjrûd: An Israelite Site from the Monarchical Period on the Sinai Border. Qadmoniot 9: 118–24 (Hebrew). 1978a Kuntillet ºAjrûd: An Israelite Religious Center in Northern Sinai. Expedition 20: 50–54. 1978b Kuntillet ºAjrûd: A Religious Center from the Time of the Judean Monarchy. Israel Museum Catalogue 175. Jerusalem: Israel Museum. 1979 Did Yahweh Have a Consort? The New Religious Inscriptions from Sinai. Biblical Archaeology Review 5/2: 24–35. Meshel, Z., and Meyers, C. 1976 The Name of God in the Wilderness of Zin. Biblical Archaeologist 39: 6–10. Mettinger, T. N. D. 1995 No Graven Image? Israelite Aniconism in Its Ancient Near Eastern Context. Stockholm: Almqvist & Wiksell. Meyers, C. 1991 Of Drums and Damsels: Women’s Performance in Ancient Israel. Biblical Archaeologist 54:16–27. 1999 Mother to Muse: An Archaeomusicological Study of Women’s Performance in Israel. Pp. 50–78 in Recycling Biblical Figures: Papers Read at a Noster Colloquium in Amsterdam, 12–13 May, 1997, ed. A. Brenner and J. W. van Henten. Studies in Theology and Religion 1. Leiden: Deo. 2000a Women with Hand-Drums, Dancing: Exod 15:20; 1 Sam 18:6– 7; 2 Sam 1:20; Ps 68:25; Jer 31:4, 13. Pp. 189–91 in Women in Scripture: A Dictionary of Named and Unnamed Women in the Hebrew Bible, the Apocryphal/Deuterocanonical Books, and the New Testament, ed. C. Meyers. Boston: Houghton Mifflin.

2000b Skilled Women (and Men): Exod 35:25–26; 36:6. Pp. 201–2 in Women in Scripture: A Dictionary of Named and Unnamed Women in the Hebrew Bible, the Apocryphal/Deuterocanonical Books, and the New Testament, ed. C. Meyers. Boston: Houghton Mifflin. 2000c Singing Women (and Men) Who Lament: 2 Chr 35:25. P. 284 in Women in Scripture: A Dictionary of Named and Unnamed Women in the Hebrew Bible, the Apocryphal/Deuterocanonical Books, and the New Testament, ed. C. Meyers. Boston: Houghton Mifflin. Miller, P. D. 1986 The Absence of the Goddess in Israelite Religion. Hebrew Annual Review 10: 239–48. Olyan, S. M. 1988 Asherah and the Cult of Yahweh in Israel. Society of Biblical Literature Monograph Series 34. Atlanta: Scholars Press. Oppenheim, A. L. 1949 The Golden Garments of the Gods. Journal of Near Eastern Studies 8: 172–93. 1977 Ancient Mesopotamia: Portrait of a Dead Civilization. Rev. ed. Chicago: University of Chicago Press. Poethig, E. B. 1985 The Victory Song Tradition of the Women of Israel. Unpublished Ph.D. dissertation, Union Theological Seminary. Sanda, A. 1912 Die Bucher der Könige. Vol. 2. Münster: Aschendorff. Sasson, J. 2002 On the Use of Images in Israel and the Ancient Near East: A Response to Karel van der Toorn. Pp. 63–70 in Sacred Time, Sacred Place: Archaeology and the Religion of Israel, ed. B. M. Gittlen. Winona Lake, IN: Eisenbrauns. Schmidt, B. B. 1995 The Aniconic Tradition: On Reading Images and Viewing Texts. Pp. 75–105 in The Triumph of Elohim: From Yahwisms to Judaisms, ed. D. V. Edelman. Grand Rapids, MI: Eerdmans. Smith, M. S. 1990 The Early History of God: Yahweh and the Other Deities in Israel. San Francisco: Harper & Row. Toorn, K. van der 2002 Israelite Figurines: A View from the Texts. Pp. 45–62 in Sacred Time, Sacred Place: Archaeology and the Religion of Israel, ed. B. M. Gittlen. Winona Lake, IN: Eisenbrauns.

Purity Strategies and Political Interests in the Policy of Nehemiah Rainer Albertz Protestant Theological Faculty, University of Münster

In his recent book Rites and Rank: Hierarchy in Biblical Representations of Cult, Olyan (2000) convincingly shows the crucial importance of binary oppositions such as “holy/common,” “unclean/clean,” “self/other,” and “whole/ blemished,” not only for biblical thought, but also for determining the reality in and the structure of ancient Israelite society. Binary classifications determined access to sanctuaries, feasts, and rites, created a certain hierarchy in the society, and established an individual’s affiliation with the community. He has demonstrated that the borderlines between these oppositions were not rigid but shifted in the course of history. Normally, a set of rites and strategies was available to reestablish a broken borderline or to negotiate between the strict oppositions. However, Olyan notes that the “self/other” opposition was very much emphasized in Ezra–Nehemiah during the 5th century bc: “ ‘Alien,’ understood in texts antedating the fifth century to apply to persons of foreign birth or persons lacking an Israelite father, becomes in Ezra–Nehemiah a term applied to persons with any foreign blood” (Olyan 2000: 89). He has observed that this intensified rigidity has to do with the transfer of two other binary oppositions—“holy/common” and “clean/unclean”—from their original cultic or quasicultic context into the area of social relations, especially regarding mixed marriages (2000: 60). In his view, the first step in this direction was taken in the Holiness Code, which stresses Israel’s separation from all other peoples and calls for avoiding polluting actions and things (Lev 18:24–30, 20:22–26; see Olyan 2000: 90). Conceiving of Israel as “the holy seed” (Ezra 9:2) that was in constant danger of being polluted by its neighbors in their uncleanness (Ezra 9:10–12), Ezra–Nehemiah went one step further: “If aliens are defiling, their ability to pollute justifies Author’s note: This study is part of a research project, “Sonderforschungsbereich 493: Funktionen von Religion in antiken Gesellschaften des Vorderen Orients,” in which Ralph Rothenbusch of the University of Freiburg and I investigate the formation of religious groups in early Judaism. Parts of this paper were read at the SBL International Meeting held in Berlin, 19–22 July, 2002. The biblical citations follow the Hebrew versification.

their absolute exclusion from the cult and community, both of which are constructed as holy” (Olyan 2000: 89). In contrast to the Holiness Code, holiness became “in Ezra–Nehemiah a matter of lineage” (Olyan 2000: 89). I believe that Olyan is correct in his analysis of the patterns of thought in what he calls “an innovative program of exclusion” in Ezra–Nehemiah. But he omitted the question of why such a rigid “program,” which went beyond the earlier legal regulations (Deut 7:3–4, 23:4–9), developed at all. When did it emerge? By whom was it promoted? Can we distinguish various stages of its development? These are the questions I wish to address in the present essay in order to offer more historical support for Olyan’s analysis.

The Nehemiah Memoir as an Older Source In seeking an earlier stage of the tradition, we find a solid base in the so-called Nehemiah Memoir. Because the entire book of Nehemiah bears the superscription “The words of Nehemiah, son of Hacaliah,” it is highly probable that the editor of Ezra–Nehemiah used an older source, written by Nehemiah himself or by an intimate friend from his immediate circle. The existence of this older source is widely accepted (Mowinckel 1923: 278–22; Rudolph 1949: xxiv; Kellermann 1967: 4–56; Williamson 1985: xxiv–xxviii; Gunneweg 1987: 176–80; Blenkinsopp 1988: 46–47; Karrer 2001: 128–213). Becker’s recent attempt (1998) to prove it to be an invention of the Chronicler is not convincing, because the style and concept of the older source clearly differ from the Chronicler’s editorial approach. Moreover, there is considerable agreement regarding its textual reconstruction: most scholars ascribe the I-passages in Neh 1:1–7:5a; 12:31–32, 37–40; and 13:4– 31 to the Nehemiah Memoir. There is only some disagreement over whether certain verses, such as 7:1b and 13:6, 7a, 22, and 30a and other shorter phrases should be seen as later additions of the editor. A more significant problem is that the reconstructed text does not constitute a complete,

200 well-structured unit. In Neh 1:1–7:3, we find a rather consistent report about Nehemiah’s success in reconstructing the wall of Jerusalem against the militant opposition of his foreign enemies. This report, however, is interrupted twice, once by the long list of builders included in Neh 3:1–32, which looks like an official document, and the second time by a digression presenting two reports on Nehemiah’s social reforms and generosity that have virtually nothing to do with the construction of the wall (cf. 5:16). In any event, the major problems appear thereafter: the report on the repopulation of Jerusalem and the registration of the people that begins in Neh 7:4–5a is not carried through but is replaced by a list of repatriates (7:6–72)— also given in Ezra 2—and a small note in Neh 11:1–2, probably written by the editor. The passage about the dedication of the wall (12:27–43), which may have concluded the building report originally, is now far removed and heavily reworked by the editor. Only 12:31–32 and 37–40 may actually have belonged to the Nehemiah Memoir. The report on Nehemiah’s cultic and ritual reforms in Neh 13:4–31, dated to his second stay in Judah (v. 6), is totally set apart. Here, the editor of Ezra–Nehemiah wrote an introduction mentioning similar cultic and ritual traditions practiced by all the people (Neh 12:44–13:3), using the older text as a kind of retrospective glance at the bad state of affairs confronting Nehemiah that was later remedied by the solemn commitment of all the people (Nehemiah 10; note the correspondence with Nehemiah 13). Thus, the original place of Neh 13:4–31 in the Nehemiah Memoir can no longer be determined. It may have been a supplement, like Nehemiah 5, and we must reckon with the possibility that the editor, who used this passage for his own purposes, made some minor additions and alignments with legal texts. In my view, the inclusion of Ammonite and Moabite women in Nehemiah’s attack on mixed marriages in 13:23, which is otherwise concerned only with Ashdodite women, should be seen to be an alignment of this sort, in this case with the legal clause in Deut 23:4–7 cited by the editor in Neh 13:1–3. Therefore, the misplaced generalization in 13:24b (“or according to the language of the other people”) is likewise an addition or even a later gloss (Mowinckel 1964: 41; Kellermann 1967: 53; Williamson 1985: 393; Gunneweg 1987: 172; Blenkinsopp 1988: 362). The summary remark in 13:30a, “So I purified them from everything foreign”—somewhat awkward after the votive clause (v. 29) and before the notes about additional cultic measures (v. 31a)—should be viewed as a redactional link to 13:3 (“they separated from Israel all those of mixed descent”), inserted by the editor in order to integrate the older source into his composition (Rudolph 1949: 210; Gunneweg 1987: 175; Blenkinsopp 1988: 366). In any case, the command that the Levites

Rainer Albertz should guard the gates during the Sabbath, suspect according to some scholars (13:22a), in my view belonged to the Nehemiah Memoir. 1 Williamson may be correct in suggesting that the Nehemiah Memoir consists of different units, an older building report from the very beginning of his activity in Judah and one or more supplements from his later office (Nehemiah 5; 13:4–31*), of which only the latter carry some characteristics of dedicatory inscriptions (Williamson 1985: xxiv–xxviii). 2 In any case, even if the Nehemiah Memoir cannot be completely reconstructed, and its emergence was somewhat more complicated than we now recognize, we have in it a nearly contemporary source that was written not later than approximately 25 years after the events it reports. This does not mean that it reflects the historical events in an objective manner—quite the contrary. Its descriptions are extremely biased, because Nehemiah wrote the report with the clear intention of justifying his controversial policy in the eyes of his Jewish countrymen and in the eyes of his God. Regarding its extremely subjective perspective, the Nehemiah Memoir may be problematic as a source for an unbiased historical reconstruction of the conflict, but it is a good source for investigating Nehemiah’s political interests and the ideological concepts underlying his policy.

The Political Interests of Nehemiah We do not know what the political status of Judah was before the arrival of Nehemiah. Alt (1953) developed the hypothesis that Judah was included within the province of Samaria during the Neo-Babylonian and early Persian periods and had not won its status as an independent province before the governorship of Nehemiah (see also McEvenue 1981; Stern 1984: esp. 82–83; Grabbe 1992: 1.79–83). This hypothesis was questioned by M. Smith (1987) and Williamson (1988), with good reason: not only was Zerubbabel called a governor (hj:P