Ur in the Twenty-First Century CE: Proceedings of the 62nd Rencontre Assyriologique Internationale at Philadelphia, July 11–15, 2016 9781646021512

Citation preview

Ur in the Twenty-First Century CE

Rencontre Assyriologique Internationale Proceedings

1. Language in the Ancient Near East, edited by L. Kogan, N. Koslova, S. Loesov, and S. Tischchenko. 2 volumes (RAI 53) 2. City Administration in the Ancient Near East, edited by L. Kogan, N. Koslova, S. Loesov, and S. Tischchenko (RAI 53) 3. Organization, Representation, and Symbols of Power in the Ancient Near East, edited by Gernot Wilhelm (RAI 54) 4. La famille dans le Proche-Orient ancien: réalités, symbolismes et images, edited by Lionel Marti (RAI 55) 5. Time and History in the Ancient Near East, edited by Lluis Feliu, J. Llop, A. Millet Albà, and Joaquin Sanmartín (RAI 56) 6. Tradition and Innovation in the Ancient Near East, edited by Alfonso Archi (RAI 57) 7. Private and State in the Ancient Near East, edited by R. DeBoer and J. G. Dercksen (RAI 58) 8. Fortune and Misfortune in the Ancient Near East, edited by Olga Drewnowska and Małgorzata Sandowicz (RAI 60) 9. Ur in the Twenty-First Century CE, edited by Grant Frame, Joshua Jeffers, and Holly Pittman (RAI 62)

Associated Volumes

1. Divination in the Ancient Near East, edited by Jeanette C. Fincke (RAI 54) 2. Divination as Science, edited by Jeanette C. Fincke (RAI 60) 3. Studying Gender in the Ancient Near East, edited by Saana Svärd and Agnès Garcia-Ventura (RAI 59 and 60) 4. Distant Impressions: The Senses in the Ancient Near East, edited by Ainsley Hawthorn and Anne-​Caroline Rendu Loisel (RAI 61) 5. Patients and Performative Identities: At the Intersection of the Mesopotamian Technical Disciplines and Their Clients, edited by J. Cale Johnson (RAI 60 and 61)

Ur in the Twenty-First Century CE Proceedings of the 62nd Rencontre Assyriologique Internationale at Philadelphia, July 11–15, 2016

Edited by Grant Frame, Joshua Jeffers, and Holly Pittman

eisenbrauns   |  University Park, Pennsylvania

Library of Congress Cataloging-in-Publication Data Names: Rencontre assyriologique internationale (62nd : 2016 : Philadelphia, Pa.), author. | Frame, Grant, editor. | Jeffers, Joshua, 1977– editor. | Pittman, Holly, editor. Title: Ur in the twenty-first century CE : proceedings of the 62nd Rencontre assyriologique internationale at Philadelphia, July 11–15, 2016 / edited by Grant Frame, Joshua Jeffers, and Holly Pittman. Description: University Park, Pennsylvania : Eisenbrauns, 2020. | Includes bibliographical references and index. Summary: “Papers presented at the 62nd Rencontre Assyriologique Internationale, held in Philadelphia on July 11–15, 2016, examining archaeological, artistic, cultural, economic, historical, and textual matters related to the ancient city of Ur in southern Iraq”—Provided by publisher. Identifiers: LCCN 2020035378 | ISBN 9781646021062 (hardback) Subjects: LCSH: Excavations (Archaeology)—Iraq—Ur (Extinct city)—Congresses. | Ur (Extinct city)—Antiquities—Congresses. | LCGFT: Conference papers and proceedings. Classification: LCC DS70.5.U7 R46 2016 | DDC 935/.5—dc23 LC record available at https://lccn.loc.gov/2020035378 Copyright © 2021 The Pennsylvania State University All rights reserved Printed in the United States of America Published by The Pennsylvania State University Press, University Park, PA 16802–1003 Eisenbrauns is an imprint of The Pennsylvania State University Press. The Pennsylvania State University Press is a member of the Association of University Presses. It is the policy of The Pennsylvania State University Press to use acid-free paper. Publications on uncoated stock satisfy the minimum requirements of American National Standard for Information Sciences—Permanence of Paper for Printed Library Material, ansi z 39.48–1992.

Contents Foreword . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ix Abbreviations . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . xi Program . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . xxi Welcome to Participants of the 62nd Rencontre Assyriologique Internationale . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1 Grant Frame

Plenary Papers Woolley’s Excavations at Ur: New Perspectives from Artifact Inventories, Field Records, and Archival Documentation . . . . . . . . . . . . . . . . . . . . . . . . 7 Richard L. Zettler Sîn-City: New Light from Old Excavations at Ur . . . . . . . . . . . . . . . . . . . . . . . . . . . 35 Jonathan Taylor Bad Moon Rising: The Changing Fortunes of Early Second-Millennium BCE Ur . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 49 Katrien De Graef

Conference Papers

(Alphabetical by Author) The Moon Watching Over the Sun and Venus: Revisiting the Attributes and Functions of Nanna/Sîn in Mesopotamia . . . . . . . . . . . . . . . . . . . . . . . 91 Isabel Gomes de Almeida & Maria de Fátima Rosa Detecting Social Tensions in the Archaeological Record: Official and Vernacular Figurine-Making Traditions at Ur in the First Millennium BCE . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 105 Anastasia Amrhein Old Babylonian Terracottas from Ur: Ancient and New Perspectives . . . . . . . . . 135 Laura Battini The City of Ur and the Neo-Babylonian Empire . . . . . . . . . . . . . . . . . . . . . . . . . . 153 Paul-Alain Beaulieu The Ziqqurats of Ur and Babylon and the Place Where the Ark Moors After the Flood (The Epic of Gilgameš XI 158) . . . . . . . . . . . . . . . . . . . . . 171 Daniel Bodi v

vi

Contents

Epigraphy of Ur: Past, Present, and Future . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 181 Dominique Charpin Signs from Silence: Ur of the First Sumerians (Late Uruk Through ED I) . . . . . 195 Petr Charvát Breakers and Enforcers of the Oath of the King . . . . . . . . . . . . . . . . . . . . . . . . . . 205 Laura Culbertson Utopic and Dystopic Images in Mesopotamian Literature: The Conflict Between Order and Chaos in Ur III . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 219 Maria de Fátima Rosa & Isabel Gomes de Almeida In Search of Ur in iqqur īpuš . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 233 Jeanette C. Fincke The Kingdom as Sheepfold: Frontier Strategy Under the Third Dynasty of Ur; A View from the Center . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 245 Steven J. Garfinkle From Uruk to Ur: Automated Matching of Virtual Tablet Fragments . . . . . . . . . 253 Erlend Gehlken, Tim Collins, Sandra Woolley, Laurence Hanes, Andrew Lewis, Luis Hernandez-Munoz, and Eugene Ch’ng The City of Ur: The Possibilities for Statistical Analysis on the Use of Space in Domestic Households . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 263 Nicole Grunert Trace Elements and Isotopes: The Origin of Gold from Ur from a Geochemical Point of View . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 273 Moritz Jansen, Andreas Hauptmann, Sabine Klein, and Richard L. Zettler Science and Technology: Using Ur-Online to Aid in Scientific Analysis . . . . . . . . 293 Kyra Kaercher and Tessa de Alarcon Towards Archival Reconstruction of Ur III Cuneiform Tablets . . . . . . . . . . . . . . 309 Yudong Liu and James Hearne Excavating Ur in Children’s Literature . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 315 Kevin M. McGeough and Elizabeth A. Galway Aspekte der Sklaverei im altbabylonischen Ur: Untersuchungen zu den a-ru-a-Texten . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 329 Christin Möllenbeck Glyptic Art from the Ur III to the Šimaški Periods: Heritage and Overtaking of the Models . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 341 Clélia Paladre The Status of Real Estate in Neo-Babylonian Ur: The Case of the Gallābu Family . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 371 Olga V. Popova Palace and “Private” Sectors in Old Babylonian Ur . . . . . . . . . . . . . . . . . . . . . . . . 379 Mitchell S. Rothman

Contents

vii

Materiality and Mark-Making: Sealing and Community in Early Dynastic Ur . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 395 Sarah J. Scott Eclipse of a King: A New Interpretation of the Bull-Headed Lyre . . . . . . . . . . . . 419 JoAnn Scurlock Local and Imported Religion at Ur Late in the Reign of Shulgi . . . . . . . . . . . . . . 429 Tonia M. Sharlach The Third Millennium Banquet Scene at Ur and Its Archaeological Correlates . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 441 Diana L. Stein New Excavations at Ur . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 475 Elizabeth C. Stone, Paul Zimansky, Katheryn Twiss, Michael Charles, and Melina Seabrook Ur and Other Cities in Some Sumerian and Akkadian Personal Names: The Pre-Sargonic and Sargonic Periods . . . . . . . . . . . . . . . . . . . . 485 Jan Tavernier Implications of Intertextuality: Erra and Išum and the Lamentation Over the Destruction of Sumer and Ur . . . . . . . . . . . . . . . . . . . . . . . . . . . . 503 Selena Wisnom Was the Karzida of Ur’s Akītu Festival at Tell Sakhariya? . . . . . . . . . . . . . . . . . 525 Paul Zimansky

Foreword The Rencontre Assyriologique Internationale met for the second time at the University of Pennsylvania in Philadelphia on July 11–15, 2016 and had the theme “Ur in the Twenty-First Century CE.” Approximately 300 people attended, coming from over twenty different countries, and over 150 papers were presented, as well as seven posters. Nine workshops took place during the conference and there were special sessions in honor of Åke Sjøberg and Miguel Civil. The sessions on the Wednesday afternoon were held at Bryn Mawr College with the collaboration of Peter Magee. In association with the Rencontre, a workshop for middle and high school teachers was held at the Penn Museum, organized by Ann Guinan and Emily Hirshorn: “Mesopotamian Madness: Teachers and Scholars Creating Curriculum Together.” Six Assyriologists gave talks at that workshop which was intended to help teachers deal with the Ancient Near Eastern world in their classrooms. Two special events also took place during the conference: a presentation of the play “Sennacherib, King of Assyria” by Selena Wisnom and a sneak preview of the film on Gertrude Bell “Letters from Baghdad.” The Penn Museum hosted a reception for attendees on the Sunday evening, July 10, following a presentation on the museum’s Ur Online Project. The Institute for the Study of the Ancient World (New York) co-sponsored the Opening Reception at Huntsman Hall on Monday, July 11; an informal pizza evening for students organized by Irene Sibbing Plantholt was held on Tuesday, July 12; Bryn Mawr College hosted a reception at Bryn Mawr on Wednesday July 13; and a BBQ dinner was held in the Stoner Courtyard of the Penn Museum on Thursday July 14. We gratefully acknowledge support for the Rencontre from the University of Pennsylvania, the Penn Museum, Bryn Mawr College, the Institute for the Study of the Ancient World (New York), and the Center for Ancient Studies of the University of Pennsylvania. The organizing committee for the Rencontre consisted of Grant Frame (Chairman), Joshua Jeffers, Holly Pittman, Lauren Ristvet, Steve Tinney, and Richard Zettler. However, the conference could not have been held without the help and support of numerous other individuals, in particular Linda Greene, Peter Magee, Tena Thomason, Kellie Zimmerman, and all the student assistants. This volume includes thirty-two of the papers given at the Rencontre, most of those presented on the main theme. The technical formatting of the articles for the volume was carried out by Darren Ashby, Tegan Bunsu Ashby, and Joshua Jeffers. Financial support for the preparation of these Proceedings came from the School of Arts and Sciences of the University of Pennsylvania and our thanks must be expressed to Jeffrey Kallberg, Associate Dean for Arts and Letters. Grant Frame, Joshua Jeffers, and Holly Pittman ix

Abbreviations AbB AbB 8 ABC ADFU AfO AfO Beih. AJ AJSL AnOr AnOr 2 AnSt APA AOAT AOAT 25 AOS ARM ARM 28 AS ASJ ATU AUCT AUCT 2 AUCT 3 AUCT 4 AUWE BAH BaM BBVO

Altbabylonische Briefe in Umschrift und Übersetzung L. Cagni, Briefe aus dem Iraq Museum (TIM II). AbB 8. Leiden: Brill, 1980 A. K. Grayson, Assyrian and Babylonian Chronicles. TCS 5. Locust Valley, NY: J. J. Augustin, 1975 Ausgrabungen der Deutschen Forschungsgemeinschaft in Uruk-Warka Archiv für Orientforschung Archiv für Orientforschung, Beiheft The Antiquaries Journal American Journal of Semitic Languages and Literatures Analecta Orientalia A. Deimel, Šumerische Tempelwirtschaft zur Zeit Urukaginas und seiner Vorgänger: Abschluss der Einzelstudien und Zusammenfassung der Hauptresultate. AnOr 2. Roma: Pontificio Istituto Biblico, 1931 Anatolian Studies Acta Praehistorica et Archaeologica Alter Orient und Altes Testament B. L. Eichler, et al. (eds.), Kramer Anniversary Volume. Cuneiform Studies in Honor of Samuel Noah Kramer. AOAT 25. Kevelaer: Butzon & Bercker, 1976 American Oriental Series Archives royales de Mari J.-R. Kupper, Lettres royales du temps de Zimri-Lim. ARM 28. Paris: Librairie orientaliste Paul Geuthner, 1941 Assyriological Studies Acta Sumerologica (Japan) Archaische Texte aus Uruk Andrews University Cuneiform Texts M. Sigrist, Neo-Sumerian Account Texts in the Horn Archaeological Museum. AUCT 2. IAPAS 5. Berrien Springs, MI: Andrews University Press, 1988 M. Sigrist, C. E. S. Gavin, D. Stein, and C. Menard, Neo-Sumerian Account Texts in the Horn Archaeological Museum. AUCT 3. IAPAS 6. Berrien Springs, MI: Andrews University Press, 1988 M. Sigrist, Old Babylonian Account Texts in the Horn Archaeology Museum. AUCT 4. IAPAS 7. Berrien Springs, MI: Andrews University Press, 1990 Ausgrabungen in Uruk-Warka. Endberichte Bibliothèque archéologique et historique, Institut Français d’Archéologie du Proche-Orient Baghdader Mitteilungen Berliner Beiträge zum Vorderer Orient

xi

xii BCSMS BCT BCT 1 BCT 2 BDTNS BE BE 1 BE 3/1 BE 6/1 BIN BIN 1 BIN 8 BIN 9 BiOr BM BPOA BPOA 1 BRM BRM 3 BRM 4 BSA BSNA CAD CDLI CDLN CM CNRS CPOA CT CT 4 CT 48 CT 50 CTMMA

Abbreviations Bulletin of the Canadian Society for Mesopotamian Studies Catalogue of Cuneiform Tablets in the Birmingham City Museum P. J. Watson, Neo-Sumerian Texts from Drehem. BCT 1. Warminster: Aris & Phillips, 1986 P. J. Watson, Neo-Sumerian Texts from Umma and Other Sites. BCT 2. Warminster: Aris & Phillips, 1993 Database of Neo-Sumerian Texts Babylonian Expedition of the University of Pennsylvania H. Hilprecht, Old Babylonian Inscriptions, Chiefly from Nippur, Part I. BE 1. Philadelphia: University of Pennsylvania, 1893 D. W. Myhrman, Sumerian Administrative Documents Dated in the Reigns of the Second Dynasty of Ur from the Temple Archives of Nippur. BE 3/1. Philadelphia: University of Pennsylvania, 1910 H. Ranke, Babylonian Legal and Business Documents from the Time of the First Dynasty of Babylon, Chiefly from Sippar. BE 6/1. Philadelphia: University of Pennsylvania, 1906 Babylonian Inscriptions in the Collection of B. J. Nies C. E. Keiser, Letters and Contracts from Erech Written in the NeoBabylonian Period. BIN 1. New Haven: Yale University Press, 1917 G. G. Hackman, Sumerian and Akkadian Administrative Texts from Predynastic Times to the End of the Akkad Dynasty. BIN 8. New Haven: Yale University Press, 1958 V. E. Crawford, Sumerian Economic Texts from the First Dynasty of Isin. BIN 9. New Haven: Yale University Press, 1954 Bibliotheca Orientalis Tablets in the collections of the British Museum Biblioteca del Próximo Oriente Antiguo T. Ozaki and M. Sigrist, Ur III Administrative Tablets from the British Museum, Part One. BPOA 1. Madrid: Consejo Superior de Investigaciones Científicas, 2006 Babylonian Records in the Library of J. Pierpont Morgan C. E. Keiser, Cuneiform Bullae of the Third Millennium B.C. BRM 3. New Haven: Yale University Press, 1914 A. T. Clay, Epics, Hymns, Omens and Other Texts. BRM 4. New Haven: Yale University Press, 1923 Bulletin on Sumerian Agriculture Biblical Scholarship in North America The Assyrian Dictionary of the Oriental Institute of the University of Chicago, 1956–2010 Cuneiform Digital Library Initiative Cuneiform Digital Library Notes Cuneiform Monographs Centre National de la Recherche Scientifique Civilisations du Proche-Orient Cuneiform Texts from Babylonian Tablets in the British Museum T. G. Pinches, Cuneiform Texts from Babylonian Tablets in the British Museum 4. CT 4. London: Trustees of the British Museum, 1898 J. J. Finkelstein, Old-Babylonian Legal Documents. London: Trustees of the British Museum, 1968 E. Sollberger, Pre-Sargonic and Sargonic Economic Texts. London: Trustees of the British Museum, 1972 Cuneiform Texts in the Metropolitan Museum of Art

Abbreviations CTMMA 1 CTNMC CUNES CUSAS CUSAS 3 CUSAS 6 CUSAS 11 CUSAS 16 CUSAS 19 CUSAS 20 CUSAS 23 CUSAS 26 CUSAS 27 CUSAS 32 CUSAS 33 DoCu EPHE DP ECTJ

ERC ETCSL FAOS FAOS 15/1 FM

xiii

I. Spar (ed.), Tablets, Cones, and Bricks of the Third and Second Millennia B.C. CTMMA 1. New York: The Metropolitan Museum of Art, 1988 Th. Jacobsen, Cuneiform Texts in the National Museum, Copenhagen, Chiefly of Economic Content. Copenhagen: C. T. Thomsens bogtrykkeri, 1939 Cornell University Near Eastern Studies Cornell University Studies in Assyriology and Sumerology D. I. Owen and R. Mayr, The Garšana Archives. CUSAS 3. Bethesda, MD: CDL Press, 2007 D. I. Owen (ed.), Garšana Studies. CUSAS 6. Bethesda, MD: CDL Press, 2011 G. Visicato and A. Westenholz, Early Dynastic and Early Sargonic Tablets from Adab. CUSAS 11. Bethesda, MD: CDL Press, 2010 S. Garfinkle, H. Sauren, and M. Van De Mieroop, Ur III Tablets from the Columbia University Library. CUSAS 16. Bethesda, MD: CDL Press, 2010 M. Maiocchi and G. Visicato, Classical Sargonic Tablets Chiefly from Adab, Part II. CUSAS 19. Bethesda, MD: CDL Press, 2012 F. Pomponio and G. Visicato, Middle Sargonic Tablets Chiefly from Adab in the Cornell University Collections. CUSAS 20. Bethesda, MD: CDL Press, 2015 V. Bartash, Miscellaneous Early Dynastic and Sargonic Texts in the Cornell University Collections. CUSAS 23. Bethesda, MD: CDL Press, 2013 A. Westenholz, A Third-Millennium Miscellany of Cuneiform Texts. CUSAS 26. Bethesda, MD: CDL Press, 2014 L. Milano and A. Westenholz, The “Suilisu Archive” and Other Sargonic Texts in Akkadian. CUSAS 27. Bethesda, MD: CDL Press, 2015 A. R. George, Mesopotamian Incantations and Related Texts in the Schøyen Collection. CUSAS 32. Bethesda, MD: CDL Press, 2016 P. Notizia and G. Visicato, Early Dynastic and Early Sargonic Administrative Texts Mainly from the Umma Region. CUSAS 33. Bethesda, MD: CDL Press, 2016 J.-M. Durand, Documents cunéiformes de la IVe section de l’École pratique des Hautes Études, Tome I: Catalogue et copies cunéiformes. HEO 18. Genève: Droz, 1982 F. M. Allotte de la Fuye, Documents présargoniques. Paris: E. Leroux, 1908 A. Westenholz, Early Cuneiform Texts in Jena: Pre-Sargonic and Sargonic Documents from Nippur and Fara in the Hilprecht-Sammlung vorderasiatischer Altertümer, Institut für Altertumswissenschaften der Friedrich-Schiller-Universität, Jena. Historisk-filosofiske skrifter 7/3. København: Munksgaard, 1975 Édition Recherche sur les Civilisations J. A. Black, G. Cunningham, J. Ebeling, E. Flückiger-Hawker, E. Robson, J. Taylor, and G. Zólyomi. The Electronic Text Corpus of Sumerian Literature (http://etcsl.orinst.ox.ac.uk) Freiburger Altorientalische Studien G. J. Selz, Die altsumerischen Wirtschaftsurkunden der Ermitage zu Leningrad. FAOS 15/1. Stuttgart: F. Steiner Wiesbaden, 1989 Florilegium Marianum

xiv FM 4 FRLANT FTP GMTR HEO HEO 12 HSS HSS 3 HSS 10 HUCA IAPAS ITT JANER JANES JAOS JCS JEOL JESHO JNES JRAS KAR Kaskal MAD MAD 1 MAD 5 MARI MC MC 4 MCS MDP MDP 14 MDP 23 MHEM MSVO

Abbreviations N. Ziegler, Le Harem de Zimrî-Lim: la population feminine des palais d’apres les archives royales de Mari. FM 4. Antony, France: Société pour l’Étude du Proche-Orient ancien, 1999 Forschungen zur Religion und Literatur des Alten und Neuen Testaments H. P. Martin et al., The Fara Tablets in the University of Pennsylvania Museum of Archaeology and Anthropology. Bethesda, MD: CDL Press, 2001 Guides to the Mesopotamian Textual Record Hautes Études Orientales D. Charpin, Archives familiales et propriété privée en Babylonie ancienne: étude des documents de “Tell Sifr.” HEO 12. Genève: Droz, 1980 Harvard Semitic Studies M. Hussey, Sumerian Tablets in the Harvard Semitic Museum. Part I: Chiefly from the Reigns of Lugalanda and Urukagina of Lagash. HSS 3. Cambridge: Harvard University, 1912 T. J. Meek, Old Akkadian, Sumerian, and Cappadocian Texts from Nuzi. HSS 10. Cambridge, Harvard University, 1935 Hebrew Union College Annual Institute of Archaeology Publications, Assyriological Series İstanbul Arkeoloji Müzeleri, Inventaire des tablettes de Tello conservées au Musée Impérial Ottoman. Paris: E. Leroux, 1910 Journal of Ancient Near Eastern Religions Journal of the Ancient Near Eastern Society Journal of the American Oriental Society Journal of Cuneiform Studies Jaarbericht van het Vooraziatisch-Egyptisch Genootschap “Ex Oriente Lux” Journal of the Economic and Social History of the Orient Journal of Near Eastern Studies Journal of the Royal Asiatic Society of Great Britain and Ireland E. Ebeling, Keilschrifttexte aus Assur religiösen Inhalts I/II. WVDOG 28 and 34. Leipzig: J. C. Hinrich, 1919 and 1923 Kaskal. Rivista di storia, ambiente e culture del Vicino Oriente antico Materials for the Assyrian Dictionary I. J. Gelb, Sargonic Texts from the Diyala Region. MAD 1. Chicago: University of Chicago Press, 1952 I. J. Gelb, Sargonic Texts in the Ashmolean Museum. MAD 5. Chicago: University of Chicago Press, 1970 Mari Annales de Recherches Interdisciplinaires Mesopotamian Civilizations P. Steinkeller and J. N. Postgate, Third Millennium Legal and Administrative Texts in the Iraq Museum, Baghdad. MC 4. Winona Lake: Eisenbranus, 1992 Manchester Cuneiform Studies Mémoires de la Délégation de Perse V. Scheil, Textes élamites-sémitiques (Cinquième série). MDP 14. Paris: Leroux, 1913 V. Scheil, Actes juridiques susiens. MDP 23. Paris: Leroux, 1932 Mesopotamian History and Environment, Series 2: Memoirs Materialien zu den frühen Schriftzeugnissen des Vorderen Orients

Abbreviations MVN MVN 3 MVN 6 MVN 7 MVN 8 MVN 10 MVN 11 MVN 13 MVN 15 MVN 18 MVN 20 NABU Nakahara NATN NEA Nik. Nisaba 8 Nisaba 15/2 Nisaba 26 NRVN 1

NSGU

NTSŠ

xv

Materiali per il Vocabolario Neo-Sumerico D. Owen, The John Frederick Lewis Collection. MVN 3. Roma: Multigrafica editrice, 1975 G. Pettinato, Testi economici di Lagaš del Museo di Istanbul, Parte I: La. 7001-7600. MVN 6. Roma: Multigrafica editrice, 1977 G. Pettinato and S.A. Picchioni, Testi economici di Lagaš del Museo di Istanbul, Parte II: La. 7601-8200. MVN 7. Roma: Multigrafica editrice, 1978 D. Calvot, Textes économiques du Selluš-Dagan du Musée du Louvre et du College de France. MVN 8. Roma: Multigrafica editrice, 1979 J. P. Grégoire, Inscriptions et archives administratives cunéiformes, Ie Partie. MVN 10. Roma: Multigrafica editrice, 1981 D. Owen, Selected Ur III Texts from the Harvard Semitic Museum. MVN 11. Roma: Multigrafica editrice, 1982 M. Sigrist, D. Owen, and G. D. Young, The John Frederick Lewis Collection, Part II. MVN 13. Roma: Multigrafica editrice, 1984 D. Owen, Neo-Sumerian Texts from American Collections. MVN 15. Roma: Multigrafica editrice, 1991 M. Molina, Tablillas administrativas neosumerias de la Abadía de Montserrat (Barcelona). Copias Cuneiformes. MVN 18. Roma: Multigrafica editrice, 1993 F. D’Agostino, Testi amministrativi della III Dinastia di Ur dal Museo Statale Ermitage. San Pietroburgo—Russia. MVN 20. Roma: Multigrafica editrice, 1997 Notes Assyriologiques Brèves et Utilitaires Y. Nakahara, The Sumerian Tablets in the Imperial University of Kyoto. Memoirs of the Research Department of the Toyo Bunko 3. Tokyo: Toyo Bunko, 1928 D. I. Owen, Neo-Sumerian Archival Texts Primarily from Nippur in the University Museum, the Oriental Institute and the Iraq Museum. Winona Lake: Eisenbrauns, 1982 Near Eastern Archaeology M. V. Nikolskij, Drevnosti Vostočnyja. St. Petersburg: Moskau, 1908–1915. J. Politi and L. Verderame, The Drehem Texts in the British Museum (DTBM). Nisaba 8. Messina: DiScAM, 2005 D. I. Owen, Cuneiform Texts Primarily from Iri-Saĝrig/Āl-Šarrākī and the History of the Ur III Period. Part 2: Catalogue and Texts. Nisaba 15/2. Messina: DiScAM, 2013 F. N. H. al-Rawi, F. Gorello, and P. Notizia, Neo-Sumerian Administrative Texts from Umma Kept in the British Museum. Part Five (NATU V). Nisaba 26. Messina: DiScAM, 2013 M. Çig and H. Kizilyay, Neusumerische Rechts- und Verwaltungsurkunden aus Nippur I = Yeni Sumer Çağina ait Nippur Hukukî ve Idarî Belgeleri I. Ankara: Türk Tarih Kurumu Basimevi, 1965 A. Falkenstein, Die neusumerischen Gerichtsurkunden. Abhandlungen (Bayerische Akademie der Wissenschaften, Philosophisch-historische Klasse), NF 39. München: Verlag der Bayerischen Akademie der Wissenschaften, in Kommission bei Beck, 1956 R. Jestin, Nouvelles tablettes sumériennes de Šuruppak au Musée d’Istanbul. Bibliothèque archéologique et historique de l’Institut français d’archéologie d’Istanbul 2. Paris: Librairie Adrien Maisonneuve, 1957

xvi OAIC OBO OIP OIP 14 OIP 104 OIP 115

OIS OLA Ontario 2 OPSNKF Or. (NS) OrAnt OSP 1

OSP 2

PBS PBS 8/2 PBS 9/1 PIHANS PPAC PPAC 5

Princeton 2 PSBA RA RAI RGTC RGTC 8

Abbreviations I. J. Gelb, Old Akkadian Inscriptions in Chicago Natural History Museum: Texts of Legal and Business Interest. Fieldiana, Anthropology 44/2. Chicago: Chicago Natural History Museum, 1955 Orbis Biblicus et Orientalis Oriental Institute Publications D. D. Luckenbill, Inscriptions from Adab. OIP 14. Chicago: University of Chicago Press, 1930 I. J. Gelb, P. Steinkeller, and R. M. Whiting, Earliest Land Tenure Systems in the Near East: Ancient Kudurrus, 2 vols. OIP 104. Chicago: Oriental Institute of the University of Chicago, 1991 M. Hilgert, Cuneiform Texts from the Ur III Period in the Oriental Institute, Vol. 1: Drehem Administrative Documents from the Reign of Šulgi. OIP 115. Chicago: Oriental Institute of the University of Chicago, 1998 Oriental Institute Seminars Orientalia Lovaniensia Analecta M. Sigrist, Neo-Sumerian Texts from the Royal Ontario Museum II: Administrative Texts Mainly from Umma. Bethesda, MD: CDL Press, 1995 Occasional Publications of the Samuel Noah Kramer Fund Orientalia (Nova Series) Oriens Antiquus: Rivista del Centro per la antichità e la storia dell’arte del Vicino Oriente A. Westenholz, Old Sumerian and Old Akkadian Texts in Philadelphia, Chiefly from Nippur, I: Literary and Lexical Texts and the Earliest Administrative Documents from Nippur. Bibliotheca Mesopotamica 1. Malibu: Undena Publications, 1975 A. Westenholz, Old Sumerian and Old Akkadian Texts in Philadelphia, Chiefly from Nippur, II: The “Akkadian Texts,” the Enlilemaba Texts, and the Onion Archive. CNI Publications 3. Copenhagen: Museum Tusculanum Press, University of Copenhagen, 1987 Publications of the Babylonian Section E. Chiera, Old Babylonian Contracts. PBS 8/2. Philadelphia: University Museum, 1922 G. A. Barton, Sumerian Business and Administrative Documents from the Earliest Times to the Dynasty of Agade. PBS 9/1. Philadelphia: University Museum, 1915 Publications de l’Institut historique et archéologique néerlandais de Stamboul Periodic Publications on Ancient Civilisations T. Ozaki and M. Sigrist, Administrative Ur III Texts in the British Museum, 2 vols. PPAC 5. Supplement to Journal of Ancient Civilizations 3. Changchun, China: Institute for the History of Ancient Civilizations, 2013 M. Sigrist, Tablets from the Princeton Theological Seminary. Ur III Period, Part 2. OPSNKF 18. Philadelphia: University of Pennsylvania Museum, 2005 Proceedings of the Society of Biblical Archaeology Revue d’Assyriologie et d’Archéologie Orientale Rencontre Assyriologique Internationale Répertoire géographique des textes cunéiformes R. Zadok, Geographical Names According to New- and Late-Babylonian Texts. RGTC 8. Wiesbaden: L. Reichert, 1985

Abbreviations RIMB RIMB 2 RIME RIME 4 RlA RT RTC SAA SAACT SANER SAOC SAT SD SET SNAT SRU Studies Salvini

TCBI 1 TCBI 2

TCL TCL 5 TCL 10 TCS TÉL TIM TIM 1 TIM 6 TLOB

xvii

Royal Inscriptions of Mesopotamia, Babylonian Periods G. Frame, Rulers of Babylonia: From the Second Dynasty of Isin to the End of Assyrian Domination (1157–612 BC). RIMB 2. Toronto University of Toronto Press, 1995 Royal Inscriptions of Mesopotamia, Early Periods D. Frayne, Old Babylonian Period (2003–1595 B.C.). RIME 4. Toronto University of Toronto Press, 1990 Reallexikon der Assyriologie und Vorderasiatischen Archäologie Recueil de travaux relatifs à la philologie et à l’archéologie égyptiennes Fr. Thureau-Dangin, Receuil de tablettes chaldéennes. Paris: E. Leroux, 1903 State Archives of Assyria State Archives of Assyria, Cuneiform Texts Studies in Ancient Near Eastern Records Studies in Ancient Oriental Civilization Sumerian Archival Texts Studia et documenta ad iura Orientis antiqui pertinentia T. B. Jones and J. W. Snyder, Sumerian Economic Texts from the Third Ur Dynasty. A Catalogue and Discussion of Documents from Various Collections. Minneapolis: University of Minnesota Press, 1961 T. Gomi and S. Sato, Selected Neo-Sumerian Administrative Texts from the British Museum. Chiba, Japan: 1990 D. O. Edzard, Sumerische Rechtsurkunden des III. Jahrtausends: Aus der Zeit vor der III. Dynastie von Ur. Munich: Verlag der Bayerischen Akademie der Wissenschaften, Beck in Kommission, 1968 P. S. Avetisyan, R. Dan, and Y. H. Grekyan (eds.), Over the Mountains and Far Away. Studies in Near Eastern History and Archaeology Presented to Mirijo Salvini on the Occasion of His 80th Birthday. Oxford: Archaeopress, 2019 F. Pomponio et al., Tavolette cuneiformi di Adab delle collezioni della Banca d’Italia I: Le tavolette cuneiformi di Adab delle collezioni della Banca d’Italia. Roma: Banca d’Italia, 2006 F. Pomponio et al., Tavolette cuneiformi di varia provenienza delle collezioni della Banca d’Italia II: Tavolette cuneiformi di varia provenienza delle collezioni della Banca d’Italia. Roma: Banca d’Italia, 2006 Textes cunéiformes. Musée du Louvre, Département des Antiquités Orientales H. de Genouillac, Textes économiques d’Oumma de l’époque d’Our. TCL 5. Paris: P. Geuthner, 1922 C.-F. Jean, Contrats de Larsa, Part I. TCL 10. Paris: P. Geuthner, 1926 Texts from Cuneiform Sources C. Virolleaud/M. Lambert, Tablettes économiques de Lagash (époque de la III e dynastie d’Ur). Cahiers de la Société asiatique 19. Paris: Société asiatique, 1968 Texts in the Iraq Museum A. al-Zeebari, Old Babylonian Letters, Part I. TIM 1. Baghdad: Directorate General of Antiquities, 1964 F. Rashid, Administrative Texts from the Ur III Dynasty. TIM 6. Baghdad: Directorate General of Antiquities, 1971 S. Richardson, Texts from the Late Old Babylonian Period. JCS Supplemental Series 2. Boston: American Schools of Oriental Research, 2010

xviii TMH NF 1–2

TRU TS TSA TSŠ UAVA UE UE 2 UE 3 UE 5

UE 6 UE 7 UE 8

UE 9 UE 10 UET UET 1 UET 2 UET 3 UET 4

Abbreviations A. Pohl, Rechts- und Verwaltungsurkunden der III. Dynastie von Ur. Texte und Materialien der Frau Professor Hilprecht Collection of Babylonian Antiquities im Eigentum der Universität Jena, NF 1–2. Leipzig: J. C. Hinrichs, 1937 L. Legrain, Le Temps des rois d’Ur. Paris: H. Champion, 1912 Tablets from Tell Sifr H. de Genouillac, Tablettes sumeriennes archaiques. Paris: P. Geuthner, 1909 R. Jestin, Tablettes sumériennes de Šuruppak conservées au Musée de Stamboul. Mémoires de l’Institut français d’archéologie de Stamboul 3. Paris: E. de Boccard, 1937 Untersuchungen zur Assyriologie und Vorderasiatischen Archäologie Ur Excavations E. Burrows, Archaic Texts. UE 2. London: Trustees of the British Museum and of the University Museum, University of Pennsylvania, Philadelphia, 1935 L. Legrain, Archaic Seal Impressions. UE 3. Oxford: Trustees of the British Museum and of the University Museum, University of Pennsylvania, Philadelphia; Oxford University Press, 1936 C. L. Woolley, The Ziggurat and its Surroundings. UE 5. Oxford: Trustees of the British Museum and of the University Museum, University of Pennsylvania, Philadelphia; Oxford University Press, 1939 C. L. Woolley, The Buildings of the Third Dynasty. UE 6. London: Trustees of the British Museum and of the University Museum, University of Pennsylvania, Philadelphia, 1974 C. L. Woolley, The Old Babylonian Period. UE 7. London: Trustees of the British Museum and of the University Museum, University of Pennsylvania, Philadelphia; British Museum Publications, 1976 C. L. Woolley, The Kassite Period and the Period of the Assyrian Kings. UE 8. Oxford: Trustees of the British Museum and of the University Museum, University of Pennsylvania, Philadelphia; Oxford University Press, 1965 C. L. Woolley, The Neo-Babylonian and Persian Periods. UE 9. London: Trustees of the British Museum and of the University Museum, University of Pennsylvania, Philadelphia, 1962 L. Legrain, Seal Cylinders. UE 10. London: Trustees of the British Museum and of the University Museum, University of Pennsylvania, Philadelphia, 1951 Ur Excavations: Texts C. J. Gadd, Royal Inscriptions from Ur. UET 1. London: Trustees of the British Museum and of the University Museum, University of Pennsylvania, Philadelphia, 1928 E. Burrows, Archaic Texts. UET 2. London: Trustees of the British Museum and of the University Museum, University of Pennsylvania, Philadelphia, 1935 L. Legrain, Business Documents of the Third Dynasty of Ur. UET 3. London: Trustees of the British Museum and of the University Museum, University of Pennsylvania, Philadelphia, 1937 H. H. Figulla, Business Documents of the New-Babylonian Period. UET 4. London: Trustees of the British Museum and of the University Museum, University of Pennsylvania, Philadelphia, 1949

Abbreviations UET 5

UET 6/1–2 UET 6/3

UET 7

UET 8 UET 9

UF UTI 6 VAS VAS 8 VAS 13 VAS 14 WOO WZKM YNER YOR YOS YOS 1 YOS 4 YOS 5 YOS 6 YOS 8 YOS 12 YOS 13

xix

H. H. Figulla and W. J. Martin, Letters and Documents of the OldBabylonian Period. UET 5. London: Trustees of the British Museum and of the University Museum, University of Pennsylvania, Philadelphia, 1953 C. J. Gadd and S. Kramer, Literary and Religious Texts, Parts I–II. UET 6/1–2. London: Trustees of the British Museum and of the University Museum, University of Pennsylvania, Philadelphia, 1963 and 1966 A. Schaffer and M.-Chr. Ludwig, Literary and Religious Texts, Part III. UET 6/3. London: Trustees of the British Museum and of the University Museum, University of Pennsylvania, Philadelphia; British Museum Press, 2006 O. R. Gurney, Middle Babylonian Legal Documents and Other Texts. UET 7. London: Trustees of the British Museum and of the University Museum, University of Pennsylvania, Philadelphia; British Museum Publications, 1974 E. Sollberger, Royal Inscriptions, Part II. UET 8. London: Trustees of the British Museum and of the University Museum, University of Pennsylvania, Philadelphia, 1965 D. Loding, Economic Texts from the Third Dynasty. UET 9. Philadelphia: Trustees of the British Museum and of the University Museum, University of Pennsylvania, Philadelphia; the Babylonian Fund, 1976 Ugarit-Forschungen F. Yildiz and T. Ozaki, Die Umma-Texte aus den Archäologischen Museen zu Istanbul, Band VI (Nr. 3501–3834). Bethesda, MD: CDL Press, 2001 Vorderasiatische Schriftdenkmäler der königlichen Museen zu Berlin A. Ungnad, Vorderasiatische Schriftdenkmäler der königlichen Museen zu Berlin 8. VAS 8. Leipzig: J. C. Hinrichs, 1909 H. H. Figulla, Altbabylonische Verträge.VAS 13. Leipzig: J. C. Hinrichs, 1914 W. Förtsch, Altbabylonische Wirtschaftstexte aus der Zeit Lugalanda’s und Urukagina’s. VAS 14. Leipzig: J. C. Hinrichs, 1916 Wiener Offene Orientalistik Wiener Zeitschrift für die Kunde des Morgenlandes Yale Near Eastern Researches Yale Oriental Series, Researches Yale Oriental Series A. T. Clay, Miscellaneous Inscriptions in the Yale Babylonian Collection. YOS 1. New Haven: Yale University Press, 1915 C. E. Keiser, Selected Temple Documents of the Ur Dynasty. YOS 4. New Haven: Yale University Press, 1919 E. M. Grice, Records from Ur and Larsa Dated in the Larsa Dynasty. YOS 5. New Haven: Yale University Press, 1919 R. P. Dougherty, Records from Erech, Time of Nabonidus (555–538 B.C.). YOS 6. New Haven: Yale University Press, 1920 D. E. Faust, Contracts from Larsa, Dated in the Reign of Rim-Sin. YOS 8. New Haven: Yale University Press, 1941 S. I. Feigin, Legal and Administrative Texts of the Reign of Samsuiluna. YOS 12. New Haven: Yale University Press, 1979 J. J. Finkelstein, Late Old Babylonian Documents and Letters. YOS 13. New Haven: Yale University Press, 1972

xx YOS 14 YOS 19 ZA ZAR ZATU

Abbreviations S. D. Simmons, Early Old Babylonian Documents. YOS 14. New Haven: Yale University Press, 1978 P.-A. Beaulieu, Legal and Administrative Texts from the Reign of Nabonidus. YOS 19. New Haven: Yale University Press, 2000 Zeitschrift für Assyriologie und Vorderasiatische Archäologie Zeitschrift für Altorientalische und Biblische Rechtsgeschichte M. W. Green and H. Nissen, Zeichenliste der archaischen Texte aus Uruk. ADFU 11. Berlin: Mann, 1987

Program * Sunday, July 10 The Dietrich Gallery of the Penn Museum 4:00–7:00 Registration The Widener Lecture Hall of the Penn Museum 5:15–6:00 Presentations of the Ur Online Project The Dietrich Gallery of the Penn Museum 6:00–7:00 Reception

Monday, July 11 The Fox Art Gallery of Cohen Hall 9:00–4:00 Registration Opening Session and Plenary Papers (Cohen G17) 10:00–10:45 Opening Session Associate Dean for Arts and Letters - J. Kallberg Museum Director - J. Siggers President of IAA - C. Michel Chair of Local Organizing Committee - G. Frame* Plenary Papers 11:00–11:30 R. Zettler, Woolley’s Excavations at Ur: New Perspectives from Artifact Inventories, Field Records and Ancillary Archival Documentation* 11:30–12:00 J. Taylor, Sîn City: New Light from Old Excavations at Ur* 12:00–12:30 K. De Graef, Bad Moon Rising: Changing Fortunes of Early 2nd Millennium BC Ur* 12:30–2:00 Lunch Break Scientific Analyses of Artifacts from Ur (I) (Cohen G17) Chair: A. McMahon 2:00–2:30 E. Salzmann, S. Klein, and A. Hauptman, Provenance Studies of Copper, Bronze and Silver Artefacts from the Royal Tombs of Ur, Mid-3rd Millennium BC *  Lectures published in this volume.

xxi

Program

xxii

2:30–3:00 M. Jansen, et al., Trace Elements and Isotopes: The Origin of Gold from Ur from a Geochemical Point of View* 3:00–3:30 S. Klein and A. Hauptmann, Ur, Mesopotamia: The Lead Metal from Pit X 3:30–4:00 A. Hauptmann, S. Klein, and R. Zettler, Sorts of Gold, Sorts of Silver from the Royal Tombs of Ur, Mesopotamia Workshop: Art/ifacts and Artworks in the ANE (I) (Cohen 402) Chair: K. Sonik 2:30–3:00 C. Thornton, Art and Artifact in the Ancient Near East 3:00–3:30 B. Pongratz-Leisten, Cultural Concepts and the Deictic Power of the Image: Seeing and Knowing in the Experience, Reception, and Production of Art in Mesopotamia 3:30–4:00 R. Wright, Puabi’s “Cloak”: What Lies Beneath Literary and Scientific Texts (I) (College 200) Chair: J. Miller 2:00–2:30 E. Koubková, Washing the Mouth of a Kettledrum 2:30–3:00 L. Talbot, Textual History Between Innovation and Fidelity: A Case Study from the Therapeutic Corpus 3:00–3:30 J. Fincke, In Search of Ur in iqqur īpuš* 3:30–4:00 I. Freedman, Babylonian Astronomers Predicted Lunar and Planetary Positions Using Iterated Maps, Forerunners of Deterministic Chaos Workshop: As Above, So Below: Religion and Geography (I) (College 314) Chair: S. Zaia 2:00–2:30 G. Konstantopoulos, Gods in the Margins: Religion and the Fictionalized Frontier in Sumerian Literary Texts 2:30–3:00 S. Richardson, Place and Portability: Divine Emblems in Old Babylonian Law 3:00–3:30 E. Knott, Mari’s Investiture Scene and the Ideology of Conquest in the Time of Samsi-Addu 3:30–4:00 H. Peighambari, Between Geography and Religion of Middle Elam: How the Middle Elamite Territory Shaped the Superiority of the Trinity of Deities 4:00–4:30 Break Scientific Analyses of Artifacts from Ur (II) (Cohen G17) Chair: P. Zimansky 4:30–5:00 A. Hauptman, S. Klein, and R. Zettler, On the Making and Provenancing of Pigments from the Early Dynastic Royal Tombs of Ur, Mesopotamia 5:00–5:30 S. Klein, S. La Niece, and A. Hauptmann, Objects from the Ur Collection of the British Museum: Sampling and Analytical Investigations 5:30–6:00 K. Kaercher and T. de Alarcon, Science and Technology: Using UrOnline to Aid in Scientific Analysis*

Program

xxiii

Workshop: Art/ifacts and Artworks in the ANE (II) (Cohen 402) Chair: C. Thornton 4:30–5:00 D. Kertai, Idols of Assyria: The Reception of the Neo-Assyrian Palace Colossi in the West 5:00–5:30 F. Rojas, Cuneiform Under Cross and Crescent 5:30–6:00 X. Wu, The Arts of Resistance: Central Asian Trade and Conflict in the Achaemenid Period Literary and Scientific Texts (II) (College 200) Chair: E. Robson 4:30–5:00 T. Abusch, The Form and History of a Babylonian Prayer to Nabû 5:00–5:30 J. Miller and A. George, A New Old Babylonian Incantation Against a Snake 5:30–6:00 D. Campbell, A Hurrian Incantation for a Toothache and Mesopotamian Folk Medicine Workshop: As Above, So Below: Religion and Geography (II) (College 314) Chair: G. Konstantopoulos 4:30–5:00 A. Shafer, Topographical Peripheries in the Assyrian Center: Abstract Ornament and the Sacred in Architecture 5:00–5:30 S. Zaia, Divine Foundations: Religion and Assyrian Capital Cities 5:30–6:00 A. Hausleiter, Foreign and Local Gods in Northwest Arabia 6:30–9:00 Opening Reception Huntsman Hall, 38th and Walnut Streets (Co-sponsored by the Institute for the Study of the Ancient World)

Tuesday, July 12 Old Babylonian Period (Cohen G17) Chair: N. Ziegler 9:00–9:30 C. Möllenbeck, Aspekte der Sklaverei im altbabylonischen Ur* 9:30–10:00 A. Goddeeris, A Feast for God and Man: The Redistribution of Sacrificed Animals in Old Babylonian Nippur 10:30–11:00 M. Rothman, State and Private Sectors in Old Babylonian Ur* Sjöberg Honorary Session (I) (Cohen 402) Chair: P. Delnero 9:00–9:30 M. Roth, Åke Sjöberg, Oneiromancy, and the River Ordeal 9:30–10:00 N. Brisch, Women as Authors of Sumerian Literary Texts 10:30–11:00 J. Klein, Šulgi K: A Unique Poem Referring to Nin-Isina, Demons and (Perhaps) Sick People* Workshop: What Can Objects Do that Words Can’t? Recent Contributions of Art History to the Study of the ANE (I) (College 200) Chair: M. Feldman 9:00–9:30 Opening Remarks by M. Feldman and S. Langin-Hooper

xxiv

Program

9:30–10:00 K. Neumann, Beyond Impressions: Perceiving and Experiencing Cylinder Seals 10:30–11:00 D. Stein, Reflections on the Banquet Scene: An Interdisciplinary Approach to Thematic Investigations* 10:30–11:00 Break Poster presenters will be available in the Fox Art Gallery to discuss their presentations Ur (Cohen G17) Chair: K. van Lerberghe 11:00–11:30 D. Charpin, Epigraphy of Ur: Past, Present, and Future 11:30–12:00 S. Ito, The Sibling Rivalry in Ur Around the Time of the Revolt of Šamaš-šumu-ukīn 12:00–12:30 A. Hätinen, Sîn-balāssu-iqbi and the Exaltation of the Moon God in Ur Sjöberg Honorary Session (II) (Cohen 402) Chair: J. Klein 11:00–11:30 J. Crisostomo, Close Enough? The Mixed Vocabularies and Sumerian Literature 11:30–12:00 R. Averbeck, The é-PA, the é-ub-imin, the Eninnu, and the Shape of Ningirsu’s Temple in Girsu 12:00–12:30 J. Cooper, Goings on in the Tablet Room: Editing Sumerian Texts Workshop: What Can Objects Do that Words Can’t? Recent Contributions of Art History to the Study of the ANE (II) (College 200) Chair: M. Feldman 11:00–11:30 M. Seymour, The Illusion of Place in Neo-Assyrian Campaign Reliefs 11:30–12:00 A. Thomason, Sensing ‘Nature’ in Neo-Assyrian Art 12:00–12:30 P. Collins, History as Ritual in Mesopotamia 12:30–2:00 Lunch 1:00–1:50 Play: Sennacherib, King of Assyria (Part 1) by S. Wisnom (Cohen G17) First Millennium (I) (Cohen G17) Chair: R. Zadok 2:00–2:30 H. Baker, The Historical Topography of Ur in the First Millennium BC 2:30–3:00 P. Beaulieu, The City of Ur and the Neo-Babylonian Empire* 3:00–3:30 O. Popova, The Status of Land Property in Neo-Babylonian Ur (the Case of the Gallābu Family)* 3:30–4:00 B. Wells, Divorce, Demotion, and Economic Status in Neo-Babylonian Marriage Contracts

Program

xxv

Sjöberg Honorary Session (III) (Cohen 402) Chair: N. Brisch 2:00–2:30 U. Gabbay, The Balaĝ Prayers to the God Enki: Text, Transmission, and Theology 2:30–3:00 M. Cohen, Of Gods and Glosses 3:00–3:30 A. Kleinerman and A. Gadotti, Beyond Nippur and Ur: Early Education in Southern Mesopotamia During the Old Babylonian Period 3:30–4:00 Closing Remarks by P. Michalowski Workshop: What Can Objects Do that Words Can’t? Recent Contributions of Art History to the Study of the ANE (III) (College 200) Chair: S. Langin-Hooper 2:00–2:30 Y. Rakic, The Ushumgal Stele: Examining Representation in the Past and its Presentation in the Present 2:30–3:00 I. Winter, Shulgi’s Stela of Ur-Namma 3:00–3:30 C. Suter, When Visual Images are Described in Texts: Who Represents Whom on the Daduša Stela? 3:30–4:00 M. Eppihimer, Invoking the Akkadians in the Stele of Shamshi-Adad I Workshop: Oracc Session (College 314) 1:00–4:00 An open session which will feature the release of a new Oracc application and will offer Oracc users and content creators the opportunity to learn more about working with Oracc. 4:00–4:30 Break First Millennium (II) (Cohen G17) Chair: J. Lauinger 4:30–5:00 A. Knapp, Royal Succession in the Ancient Near East 5:00–5:30 K. L. Younger, Arameans in the Zagros: The Cuneiform and Aramaic Evidence 5:30–6:00 J. Jeffers, Re-evaluating the Recensions of Ashurbanipal’s Prism F Inscription: Elucidating the Compositional History of the Prism F Class Gertrude Bell Sneak Preview Film Screening (Cohen 402) 4:30–6:00 Sneak preview screening of Letters from Baghdad: A Documentary Film About Gertrude Bell (Voiced by Tilda Swinton), created by S. Krayenbühl and Z. Oelbaum Workshop: What Can Objects Do that Words Can’t? Recent Contributions of Art History to the Study of the ANE (IV) (College 200) Chair: S. Langin-Hooper 4:30–5:00 H. Pittman, Art Without Writing: Binary or Continuum in the Early Bronze Age? 5:00–5:30 D. Nadali, Feeling Pictures and Words: On the Interplay of Communication in Ancient Mesopotamia 5:30–6:00 K. Sonik, The Copy as (Art) Object and the Copyist’s Art

Program

xxvi

Tablets, Writing, and Ur (College 314) Chair: W. Sallaberger 4:30–5:00 E. Gehlken, T. Collins, et al., From Uruk to Ur: Automated Matching of Virtual Tablet Fragments* 5:00–5:30 A. Bramanti and P. Notizia, New Light on the Paleography of the Early Dynastic Umma Region 5:30–6:00 P. Charvat, Signs from Silence: The Beginning of History of the Sumerian Ur* 6:30–9:00 Student-organized gathering Harvest Seasonal Grill and Wine Bar, 40th and Walnut Streets

Wednesday, July 13 Ur: Archaeology (I) (Cohen G17) Chair: A. Archi 9:00–9:30 E. Hammer and J. Ur, The City and Landscape of Ur: An Aerial and Satellite Reassessment 9:30–10:00 E. Stone, New Excavations at Ur* 10:00–10:30 N. Grunert, Space Syntax Analysis as a Spatial Analysis Tool for Breathing Life into a Dead City* Seals and Sealings (Cohen 402) Chair: R. Carter 9:00–9:30 N. al-Mutawalli Mahmood and W. Sallaberger, Larsa Period Bullae from the Shara Temple of Umma (Djokha) 9:30–10:00 C. Paladre, La glyptique d’Iran du Sud-Ouest au début du 2e millénaire avant notre ère: héritage et dépassement des schémas hérités d’Ur III* 10:00–10:30 K. Burge, The Early Development of the Mitanni Common Style Civil Honorary Session (I) (College 200) Chair: G. Rubio 9:00–9:30 P. Michalowski, At the Origins of Mesopotamian Narrative: The First Few Hundred Years 9:30–10:00 F. Karahashi, Royal Nurses and Midwives in Pre-Sargonic Lagaš Texts 10:00–10:30 J. Cooper, Enlil and Namzitara 10:30–11:00 Break Poster presenters will be available in the Fox Art Gallery to discuss their presentations Ur: Archaeology (II) (Cohen G17) Chair: M. Rothman 11:00–11:30 S. Scott, Agency Inscribed: Sealings and Community in Early Dynastic Ur*

Program

xxvii

11:30–12:00 P. Zimansky, Was the Karzida of Ur’s Akiti Festival at Tell Sakhariya?* 12:00–12:30 L. Battini, Terracottas from Old Babylonian Ur* Sealand (I) (Cohen 402) Chair: S. Richardson 11:00–11:30 A. al-Hamdani, The Shadow State: The Settlement System in UrEridu Region During the First Sealand Dynasty (1721–1340 BCE) 11:30–12:00 E. Robson, The Hinterland of Ur in the Sealand Period: Evidence from Current Excavations at Tell Khaiber 12:00–12:30 O. Boivin, The Palace as Economic Production Unit in the Sealand I Kingdom Civil Honorary Session (II) (College 200) Chair: G. Rubio 11:00–11:30 C. Woods, In the Mind’s Eye: Possible Mental Abacus Calculations in Mesopotamia 11:30–12:00 L. Feliu (read by G. Rubio), Miguel Civil: The Road to Sumer 12:00–12:30 N. Veldhuis, Computational Text Analysis: Comparing Lexical and Literary Vocabularies 12:30–2:30 Lunch 1:00–1:30 Gather at buses to transfer to Bryn Mawr Spruce Street in front of Houston Hall Literature (I) (Thomas 110) Chair: J. Sasson 2:30–3:00 G. Gabriel, Mytho-historiography from Ur III to the OB Period: Manuscript-based Textual and Ideological History of the Sumerian King List 3:00–3:30 M. Frazer, “Gilgamesh, King of Ur”: The Gilgamesh Letter in Context 3:30–4:00 M. de Fátima Rosa and I. Almeida, Utopic and Dystopic Images in Mesopotamian Literature: The Conflict Between Order and Chaos in Ur III* Mesopotamia: Now and Then (Thomas 224) Chair: P. Machinist 2:30–3:00 A. Lenzi, Assyriology at the Liberal Arts College: A Report from the Field 3:00–3:30 K. McGeough and E. Galway, Excavating Ur in Children’s Literature* 3:30–4:00 A. Schmitt, Sites of Memory in 2nd and 1st Millennium Ur Workshop: Ur and the Persian Gulf (I) (Carpenter 21) Chair: P. Magee 2:30–2:55 R. Carter, Ur and the Arabian Frontier: Cultural Interactions During the ’Ubaid Period 2:55–3:20 S. Karacic, P. Magee, and P. Drechsler, New Evidence on UbaidPeriod Trade with the Arabian Gulf

Program

xxviii

3:20–3:45 J. Pournelle, Ur in its Ecological Context: The Evolution of the Lower Euphrates from the 3rd Millennium BCE 3:45–4:10 H. Pittman, Interaction Between Ur and the Iranian Plateau via the Persian Gulf in the Early Dynastic Period: A Review of the Evidence 4:00–4:30/4:10–4:45 Break Literature (II) (Thomas 110) Chair: N. Veldhuis 4:30–5:00 E. Jiménez, The Story of the Poor, Forlorn Wren 5:00–5:30 M. Piccin, The Superlative Case in ludlul bēl nēmeqi and Babylonian Theodicy Early Texts (Thomas 224) Chair: H. Neumann 4:30–5:00 C. Lecompte, G. Benati, and E. Clevenstine, Non-administrative Documents from Archaic Ur and from the Early Dynastic I-II Period 5:00–5:30 G. Benati and C. Lecompte, The Ties that Bind: Interdisciplinary Perspectives on the Archaic Texts from Ur 5:30–6:00 D. Musgrave, A Sumerian Administrative Tablet Workshop: Ur and the Persian Gulf (II) (Carpenter 21) Chair: H. Pittman 4:45–5:10 J. Hershkovitz, The Absolute Chronology of the First Sealand Dynasty and its Influence on the Persian Gulf 5:10–5:35 E. Gorris, Traces of Neo-Elamite Presence in the Persian Gulf: A Preliminary Approach 5:35–6:00 P. Magee, The Donkey of the Sea: New Evidence for the Domestication of the Dromedary Camel and its Appearance in Southern Mesopotamia 6:30–9:00 Reception The Cloisters of the Thomas Building 9:00 Load buses for return trip to Penn

Thursday, July 14 (A special session, “Mesopotamian Madness: Teachers and Scholars Creating Curriculum Together,” will be held on Thursday and Friday at the Penn Museum. See below for details.) Workshop: Ur III (I) (Cohen G17) Chair: X. Ouyang 9:00–9:30 M. Molina, The Routes to Ur: Communications Between the Capital City and the Core Provinces of the Ur III State 9:30–10:00 T. Sharlach, Local and Imported Elements in the Religious Observances at Ur Late in the Reign of Shulgi* 10:00–10:30 R. Winters, The Royal Cowherds of Ur: A Study in ŠUKU

Program

xxix

Ur (Cohen 402) Chair: C. Suter 9:00–9:30 J. Barrabee, The Bearded Bull of Ur: Context, Comparanda, and Connotations 9:30–10:00 J. Scurlock, Eclipse of a King: A New Interpretation of the Bull Headed Lyre* 10:00–10:30 D. Bodi, The Ziqqurats of Ur and Babylon and the Place Where the Ark Moors After the Flood* Workshop: Kassites (I) (College 200) Chair: W. van Soldt 9:00–9:30 T. Clayden, Ur in the Kassite Period 9:30–10:00 H. Malko, Investigation into the Impacts of Foreign Ruling Minorities in Traditional State Societies: The Case of the Kassite State in Babylonia 10:00–10:30 I. Sibbing Plantholt, “She is Able, so I am Capable”: The Healing Goddess Gula and the Embodiment of Scholarly Knowledge, Practice and Tradition in the Kassite Period 10:30–11:00 Break Workshop: Ur III (II) (Cohen G17) Chair: D. Owen 11:00–11:30 L. Culbertson, Breakers and Enforcers of the “Oath of the King” (and the King Who Wasn’t There)* 11:30–12:00 P. Notizia, Rural Estates of the Girsu/Lagaš Province: The Case of the House of Ur-DUN 12:00–12:30 N. Borrelli, Reconstructing a Socio-economic Landscape: The Agricultural Estate of the Lagaš Governor Kurdistan (Cohen 402) Chair: L. Ristvet 11:00–11:30 S. Franke, The Temple of Haldi in Muṣaṣir and the Haldi Temples in Urartu 11:30–12:00 C. Pappi, Evidence from the Dark Age: Circulation and Production of Artefacts Between Assyria and West-Iran 12:00–12:30 S. Renette, Once Upon a Time in the East: The Early History of the Zagros Region Workshop: Kassites (II) (College 200) Chair: T. Clayden 11:00–11:30 E. Devecchi, Managing the Harvest in Kassite Babylonia: The View from the Texts in the Rosen Collection at Cornell University 11:30–12:00 S. Paulus, Redistribution Revisited: Administration of Kassite Nippur 12:00–12:30 B. Schneider, The Archaeology of the Kassite Ekur 12:30–1:30 Lunch

Program

xxx

12:55–1:50 Play: Sennacherib, King of Assyria (Part 2) by S. Wisnom (Cohen G17) Workshop: Ur III (III) (Cohen G17) Chair: P. Michalowski 2:00–2:30 L. Allred, An Archive from the Vicinity of Nippur 2:30–3:00 J. Hearne and Y. Liu, Dating Tablets of the Ur III Period* Workshop: Culture Contact and the History of Ideas: Comparing Scribal Ideologies in the Persian and Hellenistic Periods (Cohen 402) Chair: S. Sanders 2:00–2:30 J. Myerston, Assyro-Babylonian Nominal Semantics in Greece: A History of Reception, Appropriation, and Rejection 2:30–3:00 S. Sanders, Heavenly Writing and Earthly Scribes: The Aramaic Republic of Letters in Hellenistic Babylonia and Judea Workshop: Kassites (III) (College 200) Chair: S. Paulus 1:30–2:00 C. Hess, Springtime for Adad: SEM 117 and the Middle Babylonian Hymnic Tradition 2:00–2:30 K. Sternitzke, Babylon in the Old Babylonian and Kassite Periods: An Archaeological Study of the 2nd Millennium Settlement 2:30–3:00 W. van Soldt, Grammatical Irregularities in Middle Babylonian Texts and their Chronology Grammar and Onomastics (College 314) Chair: M. Roth 1:30–2:00 J. Tavernier, Ur and Other Cities in Sumerian and Akkadian Personal Names: An Update* 2:00–2:30 J. Cooley, A Scribalized Past: VR 44’s Onomastic Reflections on the History of the Land 2:30–3:00 J. Wolfe, Between Cause and Effect: The Instruments of Action 3:00–3:30 Break Workshop: Ur III (IV) (Cohen G17) Chair: P. Michalowski 3:30–4:00 S. Nett, The Administration of Regular Offerings in the Ur III Period: People, Networks, Institutions 4:00–4:30 S. Garfinkle, The Kingdom as Sheepfold: Frontier Strategy Under the Third Dynasty of Ur, A View from the Center* Persian Seals (Cohen 402) Chair: T. Kawami 3:30–4:00 X. Ouyang and A. Berlin, The Connection Between Persian Nippur and Tyrian Kedesh: The Story Behind a Seal Impression 4:00–4:30 T. Mikolajczak, Seal Inscriptions of the Accounting Seals in the Persepolis Fortification Archive

Program

xxxi

Sealand (II) (College 200) Chair: P. Beaulieu 3:30–4:00 S. Dalley (read by T. Clayden), The First Sealand Dynasty: New Evidence, and the Potential for Trade 4:00–4:30 R. Zadok, On the Economic Relations of Eanna with the Sealand, and Related Issues Sumerian Lexicography (College 314) Chair: P. Jones 3:30–4:00 A. Meskhi, Whom Ninlil Named at his Birth Šulgi (the Kartvelian Origin of Šulgi) 4:00–4:30 A. Shuke, The River and the City: On their Origins and Semantics in Sumerian 4:30–5:30 General Meeting of the IAA (Cohen G17) 6:15–6:30 Group photo The Warden Garden of the Penn Museum 6:30–9:00 BBQ dinner The Stoner Courtyard of the Penn Museum (Co-sponsored by Penn’s Center for Ancient Studies) Mesopotamian Madness: Teachers and Scholars Creating Curriculum Together Organized by A. Guinan and E. Hirshorn Thursday, July 14 and Friday, July 15 at the Penn Museum This is a workshop designed for middle and high school teachers to come together and discuss how best to teach students about the Ancient Near Eastern world in their classrooms. The event will be held in the Widener Auditorium of the Penn Museum, and a number of Assyriologists will give short presentations on various topics. July 14—Thursday Schedule 9:00–9:30 Introduction to Mesopotamia: A. Guinan 9:30–10:00 First Cities: L. Ristvet 10:00–10:30 Break 10:15–11:00 Writing Workshop: C. Michel 11:00–11:30 Hammurabi’s Code: M. Roth 11:30–12:00 Small Group, Structured Debriefs about Take-aways from Morning Sessions 12:00–1:00 Lunch 1:00–1:30 Empires: A. Podany 1:30–2:15 Jewelry Workshop: J. Bjorkman

xxxii

Program

2:15–2:30 Small Group, Structured Debriefs about Take-aways from Afternoon Sessions 2:30–3:00 Group-based Discussions on Lesson Topics and Development of Lesson Ideas ** The schedule for the Friday, July 15 sessions will be distributed to the participants of this workshop during the week of the conference.

Friday, July 15 Deities and Supernatural Beings (I) (Cohen G17) Chair: E. Stone 9:00–9:30 T. Kawami, The Sumerian Goddess Ninḫursaǧa and the Puzzle of her Temple at Tell al-Obaid 9:30–10:00 I. Almeida and M. de Fátima Rosa, The Moon Watching over Sun and Venus: Revisiting the Attributes and Functions of Nanna/Sîn in Mesopotamia* 10:00–10:30 A. Podany, A Goddess and her Community: Insights into a Late Old Babylonian Neighborhood in Terqa Fourth and Third Millennia (Cohen 402) Chair: P. Collins 9:00–9:30 J. Fuensanta and A. Mederos Martín, The Surtepe Late Chalcolithic/ Uruk Levels (Birecik, SE Turkey) and their Parallels with Other Artifacts from the Middle-Euphrates 9:30–10:00 J. Weber, More than Just Good Eats: Animals as Administrative Artifacts 10:00–10:30 G. Neumann, A Journey to Lorestān: Some New Thoughts to the Historical Topography and Development of Lorestān in the Later 3rd Millennium BCE Workshop: Intertextuality in Cuneiform Scholarship (I) (College 200) Chair: Z. Wainer 9:30–10:00 J. Steele, Citation and Use of MUL.APIN in the Neo-Assyrian and Late Babylonian Periods 10:00–10:30 S. Wisnom, Implications of Intertextuality: Erra and Išum and the Lamentation over the Destruction of Sumer and Ur* 10:30–11:00 Break Deities and Supernatural Beings (II) (Cohen G17) Chair: D. Stein 11:00–11:30 J. Bjorkman, The Case of the Missing Cult Statue 11:30–12:00 A. Amrhein, Detecting Social Tensions in the Archaeological Record: Official and Vernacular Figurine-Making Traditions at Ur in the 1st Millennium BCE

Program

xxxiii

Early Dynastic Period (Cohen 402) Chair: M. Gibson 11:00–11:30 D. Ashby, Food and Beverage Production in the ED III Bagara of Ningirsu at Tell al-Hiba, Ancient Lagash 11:30–12:00 R. Goodman, Reprising the Lagash-Umma Border Conflict in Light of New Research: The Political Ecology of Water Management in Early Dynastic Mesopotamia 12:00–12:30 A. McMahon, T. Greenfield, and H. Reade, The Oxen Project: Herd Animal Management in Early Dynastic III Ur Workshop: Intertextuality in Cuneiform Scholarship (II) (College 200) Chair: M. Rutz 11:00–11:30 Z. Wainer and M. Rutz, Quotation, Interpretation, and Intertext in the Neo-Assyrian Scholarly Reports 11:30–12:00 M. W. Monroe, Calendar Texts and the Micro-Zodiac: Intertextuality in Seleucid Astrology 12:00–12:30 E. Frahm and E. Jiménez, Cuneiform Commentaries: New Discoveries 12:30–1:00 Closing session (Cohen G17)

Welcome to Participants of the 62nd Rencontre Assyriologique Internationale Grant Frame On behalf of your fellow Assyriologists and Near Eastern archaeologists and art historians at the University of Pennsylvania, I wish to welcome you to the 62nd Rencontre Assyriologique Internationale in Philadelphia, the only World Heritage city in the United States. This is only the fifth time the Rencontre has met in North America and it is the second time it has come to the city of “Brotherly Love.” In 1988, twenty-eight years ago, it also met on July 11–15 and we hope that this Rencontre on “Ur in the 21st Century CE” will be as successful as the last one, on “Nippur, the Holy City of the Sumerians.” The University of Pennsylvania and the University of Pennsylvania Museum of Archaeology and Anthropology (or the Penn Museum as we now call it) have long and distinguished histories with regard to their involvement in the study of ancient Mesopotamia and the ancient Near East in general. These histories have been described in great detail in such works as C. Wade Meade’s Road to Babylon: Development of U.S. Assyriology (1974); Cyrus H. Gordon’s The Pennsylvania Tradition of Semitics: A Century of Near Eastern and Biblical Studies at the University of Pennsylvania (1986); Dilys Pegler Windgard’s Through Time, Across Continents: A Hundred Years of Archaeology and Anthropology at the University Museum (1993); Bruce Kuklick’s Puritans in Babylon: The Ancient Near East and American Intellectual Live, 1880–1930 (1996); and most recently Barry Eichler’s article “Cuneiform Studies at Penn: From Hilprecht to Leichty” in the volume If a Man Builds a Joyful House: Assyriological Studies in Honor of Erle Verdun Leichty (2006). Leichty was the individual most responsible for the 1988 Rencontre here in Philadelphia. One can say that the study of ancient Mesopotamia at Penn began one hundred and thirty years ago, in 1886, when John Punnett Peters and Morris Jastrow gave a series of public lectures on “Ancient Civilization of Babylon” and “Assyrian Literature and its Bearing on the Old Testament” respectively. However, it was the appointment of Hermann Vollrath Hilprecht, also in 1886, that truly began the teaching of Assyriology at the University of Pennsylvania. He became the first holder of the Clark Research Professorship in Assyriology, the first endowed chair of Assyriology in the United States and one now held by my colleague Steve Tinney. Even before Hilprecht’s appointment, there had been discussions of sending an expedition to excavate a site in Babylonia and in 1888 the “First Babylonian Expedition” set out under the direction of Peters. I won’t repeat the fascinating story of those early excavations or the ensuing Peters-Hilprecht controversy, but 1

2

Grant Frame

will note that Penn’s four seasons work at Nippur until 1900 resulted in the creation of the Penn Museum, as a place to store and display its discoveries. In 1887, Penn’s Provost William Pepper had persuaded the Trustees of the University to erect a fireproof building to house artifacts from the upcoming expedition to the ancient site of Nippur; this was necessary to ensure the funding for the expedition. One could argue that American archaeological work in Mesopotamia really began at Penn, and Penn continues to be at the forefront of work there. Over the years, numerous renowned Assyriologists and Near Eastern archae­ ologists and art historians have taught at the University of Pennsylvania or been curators in the Penn Museum, and much important work has been carried out here. I may mention the names of such individuals as: John Peters, Morris Jastrow, Hermann Hilprecht, John Henry Haynes, Albert Tobias Clay, Edward Chiera, Arthur Ungnad, Léon Legrain, George Aaron Barton, Ephraim Avigdor Speiser, Cyrus Herzl Gordon, James A. Montgomery, Samuel Noah Kramer, Francis Rue Steele, Robert H. Dyson, Irene Winter, James Muhly, Bruce Routledge, Fred Hiebert, Åke Sjøberg, Erle Verdun Leichty, and Barry Eichler. Steve Tinney, the current professor of Sumerology at Penn, and myself, the professor of Assyriology, stand in the footsteps of numerous distinguished predecessors, and it is truly humbling to do so. I am sure that Holly Pittman, Lauren Ristvet, and Richard Zettler, Penn’s faculty in Near Eastern art and archaeology, feel the same way. The five of us come from three different University departments—Near Eastern Languages and Civilizations, Anthropology, and History of Art—and from two different sections of the museum—the Babylonian Section and the Near East Section. The list of scholars that I just read did not include any of the research associates who have spent time at Penn working on various Penn-related projects. Among these, one can mention, for example, Miguel Civil, Piotr Michalowski, Darlene Loding, Stephen Lieberman, Hermann Behrens, Margaret Green, Jacob Klein, Antoine Cavigneaux, Niek Veldhuis, Wu Yuhong, Tonia Sharlach, Fumi Karahashi, Jeremiah Peterson, Matthew Rutz, and Jamie Novotny. The list of scholars who received some or all of their academic training in Assyriology and ANE art and archaeology is even more extensive and impossible to recite here. Åke Sjøberg and Erle Leichty, former curators of the Babylonian Section, fostered a tradition of hospitality to outside scholars and a policy of open access to our collection, and Steve Tinney and I have tried to continue this. The number of scholarly publications by the individuals already mentioned and by numerous others who have come to Penn to work on the cuneiform texts in the Museum’s Babylonian Section (a.k.a. the Tablet Room) or on archaeological materials in its Near East Section, or who have been associated with Penn excavations and scholarly projects, is truly amazing. I will only mention the fourteen volumes of the series “Babylonian Expedition of the University of Pennsylvania, Series A Cuneiform Texts;” four volumes of “Series D Researches and Treatises;” the fifteen volumes of “Publication of the Babylonian Section;” the seventeen volumes of the “Occasional Publications of the Babylonian Fund” and later “Occasional Publications of the Samuel Noah Kramer Fund;” and seventeen years of issues of the Journal of Cuneiform Studies that Leichty edited beginning in 1972. Archaeological work has been carried out throughout the Near East under the sponsorship, or with the support, of Penn and the Penn Museum. I can mention:

Welcome to Participants

3

Nippur (Nuffar), Ur (Tell al-Muqayyar), Tepe Gawra, Nuzi (Yorghan Tepe), Tell Umar, Shibaniba (Tell Billa), Tureng Tepe, Shuruppak (Fara), Damghan, Tepe Hissar, Khafaje, Tell Agrab, Rayy, Persepolis, Hasanlu, Tell al-Rimah, Anshan (Tall-i Malyan), Tell al-Hiba, Qabr Sheykheyn, Tell es-Sweyhat, the Bat Archaeological Project (Oman), Gordion, the Naxcivan Archaeological Project (Azerbaijan), Konar Sandal South and North, the Rowanduz Archaeological Project, Idu (Satu Qala), and Kani Shaie. In addition, current faculty, staff, and students have been involved with work at such other places as: Arslantepe, Hacinebi Tepe, Togoluk (Turkey), Tell Fadous-Kfarabida (Lebanon), Tell Leilan, and Zenjirli. Current archaeological, but non-fieldwork projects include the already men­ tioned Ur Digitization and Ur Online project (with William B. Hafford as the project manager at Penn), the Al-Hiba Publication Project (led by Holly Pittman), the Analysis of Skeletal Materials from Hasanlu and Tepe Hissar project (led by Janet Monge and Page Selinsky), and the Tepe Hissar Ceramic Chronology project (led by Ayşe Gürsan-Salzmann). Present textual research projects include the ePSD (the electronic Pennsylvania Sumerian Dictionary) led by my colleague Steve Tinney, with the aid of Philip Jones; this is of course a development from the Pennsylvania Sumerian Dictionary project founded by the late, lamented Åke Sjøberg. There is also my own RINAP (Royal Inscriptions of the Neo-Assyrian Period) project that is currently being carried on with the help of Joshua Jeffers. And, of course, Oracc, the Open Richly Annotated Cuneiform Corpus, was created by Tinney, is hosted on a Penn site, and mounts various Penn-related projects, including the online version of RINAP and Tinney’s BLMS: Bilinguals in Late Mesopotamian Scholarship. I can also mention our current work on designing three new permanent Near Eastern galleries in the Penn Museum, work that involves about ten of us curators. These galleries are to open in November 2017 and I hope that you will all come back to see them and some of the many fabulous items discovered by Penn excavations, from the Ubaid temple frieze to slipper coffins from Parthian Nippur, from the Roy­al Cemetery’s so-called Ram in the Thicket to the Ur-Nammu Stele, and from the Sumerian Flood tablet to some of the Murashu documents and incantation bowls. Perhaps I should apologize for reading the aforementioned lists of people, places and projects, but remember that I am simply following in the tradition of the scribes of ancient Mesopotamia who were noted for making lists. I am sure I have neglected to mention numerous relevant people, places, and projects, and I apologize to those forgotten, but do remember that the ancient Mesopotamian scribes also made mistakes and occasionally omitted items in one exemplar that were found in a parallel exemplar. To conclude, the University of Pennsylvania and the Penn Museum have had a one-hundred and thirty year-long history in the study of the ancient Near East and in particular ancient Mesopotamia, a history that continues today with its current cohort of scholars who are involved in numerous and varied projects, and who have plans for many more. Long may Assyriology and ancient Near Eastern art and archaeology thrive at Penn! I hope that one hundred and thirty years from now the 192nd Rencontre will be held here at Penn, and with ten times the number of participants as today, and that Penn will again stake its claim to be in the vanguard of the study of ancient Mesopotamia and the ancient Near East.

Woolley’s Excavations at Ur: New Perspectives from Artifact Inventories, Field Records, and Archival Documentation Richard L. Zettler

University of Pennsylvania Charles Leonard Woolley’s excavations at Tell al-Muqayyar (ancient Ur), near Nasiriyah, were the first excavations in the modern nation-state of Iraq and arguably the largest field project in an era of “big digs” between World War I and World War II. Sponsored by the British Museum and the University of Pennsylvania Museum of Archaeology and Anthropology (Penn Museum), Woolley worked on a notable scale, digging for twelve consecutive years, 1922–1934, most years for four or four-and-a-half months (November to February or March), and with hundreds of local workmen. While he usually hired 150–200 workmen, Woolley had 240–300 workmen on his payroll in at least four field seasons. He engaged Hamoudi ibn Ibrahim, foreman at Carchemish before World War I, and his sons to organize the workmen, but Woolley himself, his wife (Katharine), and just one or two field assistants oversaw the excavations day to day. The architecture Woolley uncovered, in particular the stepped temple tower (zig­ gurat) that Iraq’s Directorate-General of Antiquities stabilized and restored in the early 1960s (Baqir 1961: 7–8 and 1962: 7; al-Wailly 1963: 7), stands as a landmark, towering above the floodplain, today. And, the tens of thousands of artifacts he cat­ aloged provided the core collection of the newly founded Iraq Museum in Baghdad and substantial additions to the collections of the British Museum and Penn Museum. Both Woolley’s provisional permit for excavations—dated October 31, 1922 and signed by Sabih Bey Nash’at, Minister of Communications and Works—and the permit issued under Iraq’s Antiquities Law provided for a division of finds between Iraq and the excavators, though the Antiquities Law, approved in 1924, reserved any artifacts “needed for the scientific completeness of the Iraq Museum” for the government. The topic for the 62nd Rencontre Assyriologique Internationale is “Ur in the Twenty-First Century,” but before introducing the Ur Digitization Project and Ur Online (www.ur-online.org) that we hope will encourage and facilitate new re­ search on the site and ancient Mesopotamia, more generally, I want to step back in time to the early twentieth century to recap briefly Woolley’s excavations at Tell al-Muqayyar.

7

8

Richard L. Zettler

Woolley began his excavations in 1922 with two long, narrow trial trenches (Fig. 1): Trial Trench A (TTA), running south from near the eastern corner of Building B (e2-ḫur-sag) that H. R. Hall had uncovered in 1918, and Trial Trench B (TTB), oriented northeast by southwest, to the east or southeast of the ziggurat (Woolley 1934b: 5). He found broken (Early Dynastic) burials that he initially dated to the seventh century BCE in Trial Trench A (Woolley 1923: 318 and pl. XXIX), but he uncovered a large and well-preserved building (e2-nun-maḫ or ga2-nun-maḫ) with walls and pavements of baked bricks, and a hoard of gold jewelry below a late floor, in TTB. In a decision perhaps best referred to as a “no brainer,” Woolley expanded TTB, and subsequently spent his first four field seasons and the initial months of his fifth season (1922–1926) uncovering the ziggurat complex and the large public buildings—e.g., e2-gi6-par3 or giparu—within the temenos or enclosure wall dating to the time of Nebuchadnezzar II (604–562 BCE). By the middle of his fifth (1926–1927) field season, Woolley was back where he started, TTA in the southeastern corner of the temenos being the only area in close proximity to the ziggurat that remained unexcavated (Fig. 2). Woolley, by his own account, began his investigations of the area by running two long search trenches off TTA (Woolley 1928: 1–2). Both trenches began at the head or northeast end of TTA, with one (TTD) extending to the east angle of Nebuchadnezzar II’s enclosure

TTB TTA

Fig. 1. RAF Photograph, taken November 22, 1922 (Woolley 1923: pl. XXIV), showing the location of Woolley’s initial Trial Trench A (TTA) and Trial Trench B (TTB).

Woolley’s Excavations at Ur

9

TTA

Fig. 2. RAF Photograph, taken May 21, 1926, showing the extent of Woolley’s excavations in the vicinity of the ziggurat and the original Trial Trench A (TTA) (BM-Ur-GN-915).

wall and the other (TTE), perpendicular to the first, running southwest toward a gate (“south gate”) in the wall. Woolley uncovered little in TTD and stopped the excavations at a tamped earth floor ca. 1.75 m below the surface. He discovered the cemetery and the first of the late Early Dynastic “royal” burials (PG 337 and PG 580) in the other trench. Woolley was pre-occupied with the cemetery’s late Early Dynastic and Akkadian burials in the middle years of his time at Ur. He uncovered 600 or so burials in three months in 1926–1927; about 300 in 1927–1928; 454 in 1928–1929; and 350 in 1929– 1930. A Royal Air Force photograph shows the area of the Royal Cemetery below the southeastern corner of Nebuchadnezzar’s temenos near the end of Woolley’s eighth (1929–1930) field season (Fig. 3). Though he abandoned the cemetery excavations in 1930–1931, when he excavated the mausoleum of the Third Dynasty kings, on the northeastern periphery of the cemetery, and Old Babylonian and Neo-Babylonian houses to the southeast in AH (AH is an acronym for Abraham’s Housing) and NH (Nebuchadnezzar’s Housing), respectively, he returned to it in 1931–1932, as he was preparing his final report, and uncovered several large and complex burials (PG 1848, 1849, and 1850) that he thought chronologically significant near the mausoleum of the Third Dynasty kings (Woolley 1932: 357–369). He found an additional 264 burials in a southward extension of the Royal Cemetery excavations that he dubbed Pit X in his last season of excavations. These included 164 Akkadian and 100 late Early Dynastic interments, below which were even earlier burials he attributed to the so-called Jemdet Nasr Cemetery (Woolley 1955: 77–79, 127–145).

10

Richard L. Zettler

Fig. 3. RAF Photograph, taken March 12, 1930, showing the area of the Royal Cemetery (BM-Ur-GN-1616).

With the excavation of the cemetery having cleared a large area to a depth of ten to fifteen meters below the surface of the mound, Woolley took the opportunity to investigate the mound’s earlier occupation levels. He made initial soundings below the floor levels of excavated burials already in 1928–1929, and his focus on Ur’s “prehistoric” strata seemingly dominated his attention in the later years of his excavations, even though he continued to carry out extensive excavations at various locations around the site. Woolley’s deep soundings yielded a sensational discovery: a thick layer of water-laid clay that sealed strata containing Ubaid painted pottery. Woolley quickly associated the stratum with the flood known from the Sumerian and Akkadian lit­ erary compositions, as well as the Hebrew Bible, and rushed to press with the news, stealing the thunder of Stephen Langdon and the Oxford-Field Museum excavators of Kish in the northern portion of the floodplain, near Babylon, who had discovered evidence of a flood (or floods) of their own. The Kish excavators felt that Woolley had cheated them of credit for the discovery of the flood. In a letter to the Director of the Field Museum dated March 20, 1929, Stephen Langdon, Director of the Field Museum-Oxford University Joint Expedition to Kish, pointed out that Woolley had been at Kish on January 26–27 and Charles Watelin, Field Director of the Kish excavations, had shown him the alluvial deposits he had uncovered. When Woolley returned to Ur he discovered his own alluvial deposit and claimed credit for the discovery of the flood in the popular press without mention of Kish (Woolley 1929b). Penn Museum’s Director, Horace H. F. Jayne, urged Woolley to end his excavations with the eleventh (1932–1933) field season, in large part because the Great Depression had depleted Penn Museum’s financial resources, but Woolley asked for

Woolley’s Excavations at Ur

11

one more year. He brought the excavations to an end on February 25, 1934, having completed a deep sounding (Pit X) that went down ca. eighteen meters from the surface of the mound and involved removing 13,160 m3 of soil in a little more than a month (Woolley 1934a: 356–372). Woolley’s energy and stamina were, as M. E. L. Mallowan wrote in an obituary, prodigious, and his excavations a tour de force that earned him a knighthood, announced in the Birthday Honors List June 1935. But, perhaps as remarkable as the excavations is Woolley’s conscientiousness in promptly publishing reports on his work, with highlights appearing each year in The Illustrated London News and detailed preliminary reports in The Antiquaries Journal. Woolley and his collabor­ ators published final reports on the excavations in a series, Ur Excavations (UE), of ten oversized volumes. Four volumes (UE 6–9) based on his drafts appeared after his death February 20, 1960, the last volume that Terrance Mitchell, then Deputy Keeper of Western Asiatic Antiquities in the British Museum, edited and annotated, appearing in 1976. I am not sure we could ever replicate Woolley’s reports with the field records he left behind, but the British Museum and Penn Museum, with critical lead support from the Leon Levy Foundation have, nevertheless, been sifting his notes, photo­ graphs, and artifact catalogs, as well as ancillary archival documentation, for the last four years with the aim of digitally reuniting as much published and unpub­ lished information on the excavations as possible. We have also been inventorying the artifacts from Woolley’s work in our respective collections, checking and correct­ ing data, for example, on find spots, and conducting conservation assessments and opportunistic analyses of our artifacts. Woolley’s field records—the notes he took while digging and his object catalogs—are housed in the British Museum’s Department of the Middle East. They consist of ca. 20,000 blue-line (graph) cards, apparently cut to fit into his jacket pocket, with handwriting and sketches, usually in pencil, that have been tipped into seventy-eight bound volumes. Seventy-five volumes contain records of Woolley’s excavations at Ur and three records of his excavations at al-Ubaid in the first months of his 1923–1924 excavations. Of the former seventy-five volumes, twenty contain cards with his field notes and fifty-five his handwritten object catalogs. The Brit­ ish Museum also has original plans, consisting of some twenty-five rolls of large paper, mylar, and other tracing materials, many with multiple plans and even artifact drawings. No duplicate copies of these original records exist, though both the British Museum and Penn Museum have second (or third) generation typescript copies of his catalog cards. The glass plate negatives of the 2,350+ field photographs, most taken by Hamoudi’s son, Fig. 4. Hamoudi’s son, Yahya, holding Yahya (Fig. 4), are also in the British Museum, camera and tripod (Penn Museum and both the British Museum and Penn Museum Archives, Legrain Photo 39, http://www. have albums with prints of the field photographs. ur-online.org/media_item/189344/).

12

Richard L. Zettler

The British Museum and Penn Museum also have extensive ancillary archival documentation relating to the Ur excavations. The British Museum’s Department of the Middle East has bound volumes of letters, many relevant to the Ur excavations, arranged alphabetically by addressee; its Central Archives has two major collections of Woolley’s papers, one with the “Personal and Business Papers of Sir Leonard Woolley” (seven boxes containing seventy-one file folders) and the other “Documents relating to the Excavations at Ur Led by Sir Leonard Woolley” (seven boxes containing fifty-one folders). Both collections have detailed indices. The Penn Museum’s Archives has two boxes of correspondence, one containing twelve and the other fifteen folders. Additional boxes contain copies of field catalogs and correspondence and other records related to publications. 1 Léon Legrain’s papers also contain relevant documents and photographs. As for the Iraq Museum, our late colleague, Donny George, wrote in an email dated January 18, 2004 that the Iraq Museum had copies of inventories of artifacts from Woolley’s excavations, but little other original documentation. I began working with Woolley’s field records in the mid 1990s, when Penn Museum organized a traveling exhibit on the Royal Cemetery of Ur, but a Museum Loan Network grant to make a representative sample of our collections available for routine loan in late 2002 made the necessity of ready access to the originals of Woolley’s field records critical. In February 2003 I wrote John E. Curtis, Keeper of Western Asiatic Antiquities, as the Department of the Middle East was then called, asking if we could come to London in late March or April to investigate the feasibility of scanning Woolley’s field records. By sheer coincidence, our trip to London occurred at the time of the looting of the Iraq Museum in April 2003, and we were in London when John Curtis, who had gone to Baghdad hard on the heels of the looting, escorted Donny George out of Iraq. We ended up with a front row seat for Donny’s account of what had happened in Baghdad at a public event held at the British Museum on April 29. In time John Curtis and I agreed that one way we could potentially help the Iraq Museum inventory its collections in the wake of the looting was by digitizing Woolley’s field records, particularly his object catalogs, and by inventorying our own collections. The Iraq Museum needed copies of the field catalogs, and we reasoned —quite naively, I now understand—that if we could determine what we had from Woolley’s excavations, we would by default reveal what the Iraq Museum ought to have had. I was able to put together some research funds in the following years to scan all of Woolley’s field records, which we completed by the end of 2007. However, we were only able to scan and save them as (relatively) low-resolution jpegs. The portable hardware and storage capacity we needed for such a large number of scans did not really exist a decade ago, but the scans were nevertheless very legible (if Woolley’s handwriting was less so), and I was able to make ready use of them. We now have higher resolution scans of the field records available at ur-online.org. We had hoped to expand the scale of our digitization efforts, but we made little progress in doing so until the Leon Levy Foundation, then focused on partage, expressed an interest in helping us reunite the field records and artifacts from Ur in Baghdad, London and Philadelphia in 2010. 1. http://dla.library.upenn.edu/dla/ead/detail.html?id=EAD_upenn_museum_PUMu1018.

Woolley’s Excavations at Ur

13

After an initial two-day meeting here in Philadelphia in January 2011, the Leon Levy Foundation provided funding for a feasibility study and subsequently provided lead support for the multi-year effort to reunite digitally the field records and artifacts from Woolley’s excavations at Ur that we concluded in June 2016. I think we’ve created a valuable web resource, Ur Online (www.ur-online.org), for specialists and the general public alike to investigate Woolley’s excavations at Ur, as well as the longue durée of Mesopotamian history through the lens of his fieldwork. The conclusion of the initial phase of the Ur Digitization Project presents an opportune time to ask what new insights or perspectives on Woolley’s excavations our efforts have yielded. What have we learned from inventorying objects from the excavations (or at least substantial portions of the collections) in the British Museum and Penn Museum? And, what did we learn about the Woolley’s field records? What light, if any, do the notes Woolley took while digging (or his field photographs) provide about Woolley’s accounts of his discoveries as detailed in both his popular accounts and, more important, his preliminary and final reports on his excavations?

Artifact Inventories Woolley used 20,094 field registration numbers—one-to-five-digit numbers with a “U.” prefix (hereafter U-number)—to catalog the objects he uncovered in his twelve years of excavations. Most of the objects were from Tell al-Muqayyar proper, but just over two thousand were from Diqdiqqeh, low mounds immediately east and northeast of Tell al-Muqayyar, and other nearby sites (see http://www.ur-online. org/location/1803/). Workmen or their children collected the overwhelming majority of objects from these sites and brought them to Woolley, who gave them token compensation or bakshish for the finds (Fig. 5).

Fig. 5. Photograph showing Woolley recording finds workers brought in for bakshish, 1928–1929 (BM Ur GN 1589).

14

Richard L. Zettler

As the list of U-numbers (Table 1) for Woolley’s excavation seasons shows, Woolley left gaps in the sequence of field registration numbers between seasons, with a particularly large block of numbers between his third (1924–1925) and fourth (1925–1926) seasons. He expected most of the numbers to be assigned post-excavation to the considerable number of cuneiform tablets, most dating to the time of the Third Dynasty of Ur, he had found below baked brick pavements in rooms to the south of the courtyard of the Kassite Edublalmah in his third season (Woolley 1925: 392–394 and 1965: 25–31; Jacobsen 1953). Woolley’s epigraphers—Sydney Smith (1922–1923), C. J. Gadd (1923–1924), Léon Legrain (1924–1925 and 1925–1926), Eric Burrows (1926–1927, 1927–1928, 1928–1929 and 1929–1930), Chauncey Winkworth (1930–1931) and Cyrus H. Gordon (1931–1932) 2—weren’t able to catalog all of the tablets and inscribed objects while on site, and Woolley routinely exported them for study, with a division between Iraq and the excavators following their publication. Woolley skipped a total of 4,758 catalog numbers between seasons, but Table 1. Woolley’s field catalog numbers by excavation season.

Season I

Years 1922–1923

Catalog Numbers

Gaps

U.1–1054 U.1055–1100

II

1923–1924

U.1101–1788

III

1924–1925

U.2501–3374 (3373)

IV

1925–1926

U.6001–7145 (7143)

U.1789–2500 (3374) U.3375–6000 U.7146–7499 V

1926–1927

U.7500–9365 (9360) U.9366–9500

VI

1927–1928

U.9501–11231 U.11233–11399

VII

1928–1929

U.11400–13108

VIII

1929–1930

U.13500–15817

IX

1930–1931

U.16001–17448

U13109–13499 U.15818–16000 U.17449–17600 X

1931–1932

U.17601–18208

XI

1932–1933

U.18209–18723

XII

1933–1934

U.18724–20094

2.  Woolley did not engage an epigrapher in 1931–1932 because he did not anticipate that the work he was planning would yield many tablets or other inscribed material. However, he made arrangement with E. A. Speiser, Director of the Penn Museum, American Schools of Oriental Research and Dropsie College excavations at Tell Billa and Tepe Gawra, to employ Cyrus H. Gordon, the epigrapher at Tell Billa, as might be necessary. Gordon was at Ur for a short period of time in February 1932 (Gordon 2000: 35–36).

Woolley’s Excavations at Ur

15

his epigraphers subsequently assigned most of the numbers he skipped between his third and fourth, fourth and fifth, and fifth and sixth seasons to tablets, leaving no more than 1,645 or so numbers open. 3 We can assume then that Woolley used a total of just under 18,500 cards to document the finds he uncovered in his excavations. Woolley not only left catalog numbers unused between seasons, but cards missing from the sequence he compiled each season initially suggested that he may also have skipped individual numbers or blocks of numbers in the course of particular field seasons. However, we were able to track down a substantial number of the artifacts linked to missing cards in typed copies of Woolley’s catalog or various other lists of objects, implying that the missing cards were probably lost over the years. It is reasonable to assume, if by no means certain, that we will eventually find artifacts associated with the 500 or so field cards not yet accounted for in one or other of the three museums that divided the finds or with the hundreds of other institutions (and individuals) that acquired artifacts from Woolley’s excavations. 4 In general, Woolley assigned U-numbers to individual objects, sometimes adding a drawing to his field card, but he also used single numbers to record multiple artifacts related by find spot, e.g., earrings, pins, chokers, necklaces or cuffs, etc. associated with each of the seventy-four bodies in PG 1237 (Great Death Pit), including Body 61 with nineteen objects (U.12380), two or three times the number with other bodies, or, in some cases, by type of object, e.g., arrowheads, weights, and stone tools. Woolley usually recorded the number of objects he had grouped under a single U-number, but not always. For example, his card for U.17938 records “Celts. Flint. With Concave tang. (From) X Reijiba” and includes a drawing of the type. We have to date located twenty-four celts with the field number U.17983 written on them. For our Ur Online catalog of objects, we created separate records for each of the artifacts on field cards with multiple entries; so, our catalog will eventually provide a more or less accurate count of the number of objects from the excavations. Currently, tabulating the number of artifacts in our Ur Online database by season, the 18,500 or so numbers Woolley assigned in digging record a total of 27,236 artifacts from his twelve years of excavations. That number includes tablets and inscribed objects cataloged between seasons, but not the “ersatz field numbers” U.21000–21423 and U.30000–31620 that Penn Museum and the British Museum assigned to more than 2,000 uncataloged tablets, formally on loan to the two museums, prior to their publication (Loding 1976; D’Agostino, Pomponio, and Laurito 2004; Spade 2006; and Black and Spada 2008). All of these tablets have been (or will be) returned to the Iraq Museum since the two Museums waived rights to a division in the late 1970s. 5 The objects Woolley included in his field catalog represent only the proverbial “tip of the iceberg.” Woolley did not include all of the objects he found in his field 3.  Woolley’s epigraphers also inadvertently used field registration numbers, e.g., U.7000–7145, for tablets that he had already assigned to objects in the field. 4.  The British Museum, Penn Museum and, apparently, even the Iraq Museum sent artifacts from Woolley’s excavations around the world. A former Penn Museum employee visiting Australia on business, for example, recently appraised us of the collection of artifacts from Ur at the Nicholson Museum, University of Sydney. For correspondence between the British Museum and other museums in Britain and the Commonwealth to whom it was giving gifts of antiquities from Ur, see British Museum Central Archives, Documents relating to the Excavations at Ur Led by Sir Leonard Woolley (1922–1939), Box 18. 5.  Penn Museum still holds 4,000–5,000 fragmentary tablets, most Ur III in date, that have not been published and that will be returned to the Iraq Museum in the near future.

16

Richard L. Zettler

catalog, and both the British Museum and Penn Museum—presumably also the Iraq Museum—have substantial numbers of artifacts in their collections without field registration numbers. We knew when we initiated our inventory that Penn Museum had numerous objects from Ur on shelves in our storeroom that had either lost their U-number over time or never had U-numbers. For example, many of the artifacts from Woolley’s twelfth (1933–1934) season had “U.” written on them, but no field numbers. Penn Museum’s EMu collections database contains 8,930 records for Ur though these include records for electrotype copies of gold objects from the Royal Cemetery (29-22-2 to 29-22-11 and 31-18-1 to 31-18-11) and other reproductions, exhibit props, etc. Of 8,930 records, 8,756 records link to objects, including cuneiform tablets and inscribed artifacts, from Woolley’s excavations. 6 Some of these represent objects, for example, Puabi’s “diadem,” that have been divided into multiple components. We “deconstructed” Puabi’s “diadem” (U.10948), originally registered as B16684 (March 1, 1929), on the supposition that what Woolley (or, more likely, his wife Katharine) “created” as a single elaborate headband, after the fashion of the flapper era of the 1920s, was, in fact, an ensemble of thematically-linked pieces of jewelry (Pittman and Miller 2014). We concomitantly subdivided B16684 into B16684.1–71, giving its components—appliques, amulets and beads—unique numbers for internal tracking purposes. Other numbers mask multiple artifacts, the most extreme example being 87-28-26, a museum registration number that covers 7,070 Uruk baked clay mosaic cones, stored in 643 plastic bags. We conducted a preliminary audit of our cylinder seals and more thorough inventories of our ceramic vessels, baked clay or terracotta objects and metal (copper alloy, silver, lead, and iron) tools and weapons, as well as vessels, in 2013–2014 and 2014–2015, though not objects made of stone, shell, frit or glass, etc. We determined that little more than half (4,745 or ca. 54%) of Penn Museum’s 8,756 collections records link to objects with U-numbers, while a little less than half (4,011 or ca. 46%) to objects that do not have field numbers. 7 When we began inventorying our collections, we guessed that Woolley would more likely have assigned U-numbers to objects in groups with large numbers of unique or distinctive exemplars than other sorts of objects that he could readily organize by types. We expected, for example, that more cylinder seals would have U-numbers than ceramic vessels. 8 6.  Searching Ur Online yields 7,692 artifacts in Penn Museum’s collections. The difference can be explained, at least in part, by the fact that where Penn Museum’s EMu collections database is organized by museum registration numbers, Ur Online’s database is organized by field number. Ur Online does not always divide artifacts with multiple components the way Emu database does, and it also does not tabulate, for example, groups of objects like the more than 300 artifacts, still in Penn Museum’s collections database, that were deaccessioned over the years. These can, nevertheless, be found by searching for the field numbers in question or by searching under the museums that currently have artifacts that were deaccessioned, e.g., the Metropolitan Museum of Art or Nelson-Atkins Museum, Kansas City. 7.  We were not able to recover U-numbers for more than a handful or objects, for example 31-43-59 (U.16794). Our Ur Digitization Projector Conservator (Tessa de Alarcon) examined a few objects using infrared (IR) reflectance and ultraviolet (UV) photography. The photographs revealed writing in pencil that enabled her to confirm the field numbers of B17246 (U.9239), a limestone footed stand from PG 348, long thought to be a plaster cast of the original, and B17539 (U.10954), a copper alloy jar from PG 800. 8.  Woolley nowhere detailed his criteria for including (or not including) objects in his field catalog. However, Legrain (1951: 56, n. 1) noted that “badly worn or decayed seals whose find-spot afforded no dating evidence” were not included in the field catalog, implying that condition and provenience were factors. Apropos of pottery vessels, Woolley (1962: 105) wrote, “... in the case of clay vessels, all of which

Woolley’s Excavations at Ur

17

Our work on Penn Museum’s collections more or less confirmed our assumptions. Of the cylinder seals in our collections, 272 of 376 (72%) have U-numbers, while 104 (28%) do not. The British Museum did not inventory its cylinder seals as part of the Ur Digitization Project, but its collection is well published; so, it is possible to calculate roughly the number of cylinder seals from Woolley’s excavations that do and do not have U-numbers. Of the cylinder seals from Ur in the series Catalogue of Western Asiatic Seals in the British Museum: Cylinder Seals I-V, 211 of 355 (60%) have U-numbers, while 144 (40%) have either lost or, more likely, never had field numbers. Penn Museum and the British Museum together have about half of the terracotta figurines and plaques of various periods from Woolley’s excavations. Penn Museum has 532 figurines and plaques, of which 348 (65%) have field numbers and 185 (35%) have either lost their U-numbers or were never included in the field catalog. The British Museum’s larger collection is more evenly split, in part because of ninety or so plaques from the last season of excavations with no field numbers. Of the 757 figurines and plaques in Bloomsbury, 384 (51%) have U-numbers. Compiling statistics on ceramic vessels from Woolley’s excavations in Penn Museum is complicated by the seemingly bewildering array of terms used to describe their forms in the collections database over the years. Be that as it may, of 1,123 whole or nearly complete ceramic vessels in the collection I was able to identify, less than a quarter (259 or 23%) have U-numbers. Except for being from Woolley’s excavations, the more than ten thousand objects from Ur (or nearby sites) in Baghdad, London and Philadelphia without U-numbers, even if intrinsically interesting, cannot be linked to particular sites or specific find spots and are, for all intents and purposes, unprovenienced. Inventorying our collections yielded more than merely dry statistics. We also made some noteworthy discoveries about the objects themselves that we often detailed in monthly blog posts over the course of the project. For example, as Woolley noted now and then on his catalog cards, a substantial number of the copper or copper-alloy objects had pseudomorphs that preserved traces of organic materials (or objects made of organic materials) through their replacement by corrosion products of the metal. The most common and noticeable pseudomorphs on Penn Museum objects are of wood, reeds, leather and textiles (Kaercher 2015a-b). The most interesting of the pseudomorphs revealed elaborately decorated leather and textile scabbards of daggers that resemble the scabbard of the gold dagger U.9361 from PG 580 (de Alarcon 2016). In addition to pseudomorphs, the x-radiographs our Ur Project conservator took of both the gold and copper or copper-alloy objects revealed extensive details of their manufacture (Zettler et al. 2017). The British Museum’s Ur Digitization Project’s conservator also found pseudo­ morphs of woven fibers on one side of a large copper-alloy cauldron, perhaps the cauldron (U.12487) found at the foot of the wooden coffin in PG 1422, a late Akkadian burial on the northeastern edge of the Royal Cemetery. Woolley noted that the floor of PG 1422’s pit had been covered by reed matting with an unusual patterned are fully recorded in the field, the individual specimen loses importance as soon as the class or type has been established, and the types can be adequately dealt with in the section on Pottery, so that a clay vessel is entered in the catalogue only for some special reason.”

18

Richard L. Zettler

weave. Unfortunately, the paraffin wax Woolley had used to consolidate the cauldron obscured the pattern of the matting; so, we cannot confirm that the cauldron is from PG 1422. However, x-radiography revealed details of how the cauldron was manufactured, as well as the rivets fixing the upright handles to the neck and the round base to the bowl (Gardiner 2016 and 2017). As it happened, our inventory of the metal tools, weapons, and vessels in 2015–2016 coincided with the last year of an intensive collaboration between Penn Museum and Andreas Hauptmann and Sabine Klein (Deutsches Bergbau Museum), supported by a grant from the German Research Foundation (Deutsche Forschungsgemeinschaft), to analyze the metals (gold, silver, and copper or copper-alloy arti­ facts) from the Royal Cemetery in our collections (Hauptmann and Klein 2016; Hauptmann et al. 2016). 9 The grant funded two Ph.D. students: Moritz Jansen (Penn Museum’s Center for the Analysis of Archaeological Materials), who worked on the gold objects (Jansen et al. 2016 and Jansen 2019), and Eveline Salzmann (2019), who analyzed the copper and copper-alloy artifacts. Though not really linked to our inventories and conservation assessments, other Ur Digitization Project activities led inadvertently to the rediscovery of human skeletal remains that the Penn Museum had gotten in the division of finds from Woolley’s 1929–1930 excavations with the British Museum that had long been listed in our collections database as “not accounted for.” In digitizing archival records in 2014, we came across Léon Legrain’s letter to Horace Jane, dated October 19, 1930, reporting on the division of finds from the 1929–1930 excavations, 10 as well as Legrain’s list, dated March 21, 1931, of the Penn Museum’s share of the finds that had arrived in Philadelphia five months after the division that had taken place in London. 11 In addition to an inscribed foundation statuette (U.15065 or 31-17-8) and inscribed stone tablet (U.15064 or 31-17-7) of Rim-Sin from the temple of Enki (Woolley and Mallowan 1976: 64–67), a soil sample from the flood stratum, and a silver bowl (3117-14) from an Achaemenid Persian bathtub-shaped clay coffin (Woolley 1962: 131 and pl. 23), the list included a tray of human bones and two human skeletons, with no additional information. Searching our collections records turned up 31-17-172, the complete skeleton of a young attendant from PG 1648 (Woolley 1934: 133–134; Molleson and Hodgson 2003: 100–105); 31-17-403, a human skeleton “in a crouched position;” and 31-17-404, a human skeleton “in a stretched position.” Working with Woolley’s reports and field photographs, we tentatively identified 31-17-403 as Jemdet Nasr (Cemetery) Grave (JNG) 361, 12 and we conclusively identified 31-17-404 as a late Ubaid burial (PFG/Z) that cut into the flood layer in Pit F. 13 We eventually located 31-17-403 and 31-17-404 in their original wooden shipping crates in our Physical Anthropology storerooms. They had lost all association with the museum registration numbers Legrain assigned them in 1931. 9.  Hauptmann and Klein were also able to do non-destructive testing of some of the Ur metals in the British Museum with the help of Susan La Niece, British Museum, Conservations and Scientific Research (Klein, La Niece, and Hauptmann, 2016). 10.  Penn Museum Archives, Horace F. Jayne Director’s Office Records, Box 10. 11.  Penn Museum Archives, Ur Box 4. 12.  Woolley 1955: p. 125 and pl. 8d. See field photographs BM-Ur-GN-1518 and 1594. 13.  Woolley 1955, pp. 93–94 and pl. 5c. Field photograph BM-Ur-GN-1484 shows PFG/Z as excavated; BM-Ur-GN-1485–1488 show Woolley consolidating, lifting and transporting the burial out of Pit F. BM-Ur-GN-1609 shows PFG/Z on exhibit at the British Museum in the eighth temporary exhibition of Woolley’s latest discoveries in Summer 1930.

Woolley’s Excavations at Ur

19

Fieldnotes and Photographs As noted above, Woolley’s field records are bound in seventy-five volumes that are housed in the British Museum’s Department of the Middle East, but now accessible at Ur Online. Fifty-five volumes contain his catalog cards. Twenty volumes contain cards with the notes Woolley took while digging, and sixteen of twenty volumes—more than 3,100 cards—record his excavations of the Royal Cemetery. The cards are organized by PG-numbers and document the overwhelming majority, if not all, of the burials Woolley included in the three tabular analyses of graves in his final report (Woolley 1934b: 411–509). Woolley needed a substantial number of cards for the larger royal burials, with ninety-six cards for PG 789, eighty-three for PG 800, and 109 for PG 1237. By contrast, only four volumes—just over a thousand cards—include the notes Woolley took on other excavation areas. They include a large number of cards (and probably full documentation) for some operations such as TTB and Enunmah, KP (Giparu) and KPS, as well as the Harbor Temple, but the notes on other areas are incomplete or missing entirely. Notes on the Old Babylonian private houses in AH, on the southeastern side of the mound, for example, appear to be incomplete. Woolley’s records include 230 cards with notes on burials found below the houses (Luby 1990). These cards have isolated, if significant, details on the stratigraphy and architecture of the houses, e.g., a section through No. 4 Straight Street, Room 4. 14 However, the cards do not have the detailed data Woolley would have needed to write the description of the houses in his final publication (Woolley and Mallowan 1976: 118–166). Cards with such notes must have existed. The back of a card with notes on the burials has the heading AH House Notes, as if those notes would have followed in the sequence of cards. 15 Three cards with notes on the houses actually exist: one with information on a Late Pavement NW of House I 16 and two with a description of the find spot of U.16959, an oversize baked clay plaque depicting a goddess with a flow vase (Woolley and Mallowan 1976: pl. 64:1). The latter two cards were bound with the object’s catalog card. 17 The cards have the header AH. But, cards with details of the stratigraphy and architecture of the houses in AH do not exist among Woolley’s records today. In addition to AH, Woolley uncovered Old Babylonian private houses at the southern corner of the temenos (EM) and over the mausolea of the kings of the Third Dynasty of Ur (BC). His notes for the fifteen or so houses in EM are, like those for AH, incomplete, with twenty cards recording burials under the floors of the houses, but no cards with details of their stratigraphy and architecture. By contrast, Woolley’s notes on the scrappy Old Babylonian houses in BC, cut by Nebuchadnezzar’s temenos wall, would appear to be (more or less) complete. 18 Woolley dug the 14.  British Museum, Department of the Middle East, Ur Fieldnotes 194-03, p. 027 (see http://www. ur-online.org/file-detail/21129/). 15.  British Museum, Department of the Middle East, Ur Field Notes 194-03, p. 133 (see http://www. ur-online.org/file-detail/21455/). 16.  British Museum, Department of the Middle East, Ur Field Notes 194-04, p. 107 (see http://www. ur-online.org/media_item/23996/). 17.  British Museum, Department of the Middle East, Ur Fielkd Notes, 194-67, pp. 59–60 (see http:// www.ur-online.org/media_item/254106/ and http://www.ur-online.org/media_item/254107/). 18.  British Museum, Department of the Middle East, Ur Field Notes, 194-02, p. 110_v and 194-02, pp. 272–322.

20

Richard L. Zettler

houses as a single unit he called House 30, but he divided them into five houses (House 30/A–E) for publication. His account of the houses (Woolley and Mallowan 1976: 166–168) is perfunctory, but his description of House 30/B incorporates all the detail on the cards that also include a sketch plan of the building. Woolley’s notes on House 30/B provide a basis for estimating how many cards might originally have existed for AH and EM. Woolley uncovered more than fifty houses in AH and fifteen in EM. If Woolley used at least six cards per house as he did for the stratigraphy and architecture of House 30/B (he used three cards for notes on the burials associated with House 30/B), Woolley must have had at least four hundred cards with notes on the stratigraphy and architecture of the houses in AH and EM, and probably many more since the houses in AH and EM are larger, better preserved, and yielded many more objects than House 30/B. These cards either got separated from his field records over the years or, perhaps, discarded when Woolley completed UE 7 (The Old Babylonian Period). He drafted the manuscript before World War II, but he revised it between 1953 and 1959. Woolley’s field cards were bound into books sometime after 1975 or 1976, when Max Mallowan returned some note cards integrated into the bound volumes that had been with his own papers. Woolley published three-quarters of the 2,350 Ur field photographs in either his preliminary and/or his final reports. A few of the published field photographs have proven critical to conservation and restoration projects we have undertaken on iconic objects from excavations here in Penn Museum, for example, the photograph showing the upper part of the ram-caught-in-a-thicket (U.12357B or 30-12-702) from PG 1237 in situ (Woolley, 1934b: pl. 77b, BM-Ur-GN-1216). The photograph shows the front hooves of the animal squarely on the branches of the tree, as did the photograph of the British Museum’s ram U.12357A (BM 122200). Yet Woolley’s restoration of U.12357B, the first of the two statuettes he worked on, left them dangling in the air as if the goat were walking on its hind legs. Woolley defended his restoration (1934b: 265), but never referred to the field photograph. He had to have known the restoration was inaccurate, and he was doubtless relieved when it fell to the Penn Museum in the division of finds with the British Museum. Our overriding concern in the restoration work we undertook in the late 1990s was to stabilize the ram for travel, but our major goal was to position the animal’s front legs on the branches of the tree as in the field photograph (Rakic 1998). Woolley’s photograph also led us to adjust the wrapping of the sheet(s) of gold around the tree, where the branches join the trunk, to more closely match the original than Woolley’s restoration. Woolley’s unpublished field photographs do not add markedly to our knowledge of Ur’s stratigraphy and architecture or the objects Woolley uncovered. Photographs provide some additional views of architectural features like the Warad-Sin bastion or objects like lyres from PG 1237 in situ and the bull’s head of the gold lyre. “Under­ utilized” photographs of the silver rein ring with the electrum statuette of an equid U.10439 (BM 121348), the gold spouted bowl U.10451 (B17692), the gold tumbler U.10453 (B17691), and the gold cup with upturned pouring spout U.10454 (BM 121346), all from PG 800’s death pit, as well as the gold oval bowl U.10930 (B16707) and fluted oval bowl U.10850 (B17693) from PG 800’s tomb chamber, show the condition of the objects when excavated (Fig. 6a–e). These photographs appeared in The Illustrated London News (January 21, 1928 and March 3, 1928) while Woolley

Woolley’s Excavations at Ur

Fig. 6a. BM-Ur-GN1042a is U.10454 (BM 121346).

21

Fig. 6d. BM-Ur-GN-1091 is U.10850 (B17693), left, and Fig. 6b. BMU.10931 (BM 121347), right. Ur-GN-1043 is U.10453 (B17691).

Fig. 6c. BM-Ur-GN-1044 is U.10451 (B17692).

Fig. 6e. BM-Ur-GN-1095 is U.10930 (B16707), left, and U.10851 (BM 121344), right.

Fig. 6. Gold vessels from PG 800 death pit and tomb chamber as excavated.

was still in the field. But, Woolley, with his emphasis on restoration for the annual exhibition at the British Museum to attract public attention and financial support, never used them subsequently. Woolley’s preliminary report (1928) included photographs of the objects after the goldsmith James. R. Ogden (presumably) had restored them (Millerman 2008). His final report on the Royal Cemetery featured only Mary Louise Baker’s water colors of the rein ring and gold vessels (Woolley 1934: pls. 162, 164, and 166). Woolley’s field photographs include 139 group shots of cylinder seals (or really their modern impressions) from the excavations; they also include 109 photographs of groups of terracotta figurines and plaques. Only seventy or so photographs show digging in progress, the workers, or the excavators in the field, and only a handful of these photographs have been published to date.

Revisiting the Published Record Woolley’s field records may represent an incomplete data set, but they are, nevertheless, an incalculable resource, as Edward Luby’s Ph.D. dissertation on Old Babylonian mortuary remains at Ur (see above) convincingly demonstrated more than three decades ago. Woolley’s notes provide us with the “unfiltered” observations he put on paper while digging, notes that enable us to amend and augment his pub­ lished accounts and that now and again raise questions about his understandings of his discoveries. A few concrete examples based on our work with his records over the last few years will illustrate their enduring utility.

22

Richard L. Zettler

PG 789 and PG 800 Woolley’s unpublished field notes and reports, in particular his biweekly reports to the Directors of the British Museum and Penn Museum, now available at Ur Online, are key to disentangling the complex stratigraphic relationship between Woolley’s two proto-typical late Early Dynastic royal tombs, PG 789 and PG 800. His records conclusively substantiate the suggestion made two decades ago (and greeted with some skepticism since) that the two seemingly identical and contemporary royal tombs represent not two tombs belonging to an unnamed king and his queen, Puabi, as Woolley posited, but a sequence of three royal burials (Zimmerman 1998a–b). I summarized the unpublished information in Woolley’s records in my plenary address at the 62nd Rencontre Assyriologique Internationale, but Paul C. Zimmerman and I cited it in full in an article that will appear in From Sherds to Landscapes: Studies on the Ancient Near East in Honor of McGuire Gibson to be published by the Oriental Institute in the near future.

Enunmah Woolley’s enunmah, or, more probably, ga2-nun-maḫ or e2 ga2-nun-maḫ, “(House), Exalted Storehouse,” lay to the southeast of the ziggurat complex. 19 Woolley excavated the building in 1922–1923 and 1926–1927. In the late third and second millennia BCE, Ganunmah, whose northwestern side was destroyed by a large horizontal drain dating to the Neo-Babylonian period, was a square building, measuring 57 m on a side, with a central complex of five rooms, ringed by a corridor, off which opened a series of long, narrow rooms. It probably functioned as a warehouse (Van De Mieroop 1992: 125). Nabonidus rebuilt the building, radically altering its plan and seemingly its function. He called it e2-nun-maḫ, “House of the Lordly Prince.” Woolley published descriptions, with plans, of the earlier Ganunmah) and later Enunmah in his preliminary report (Woolley 1923: 319–330, figs. 2, 3, and 5). He included a detailed report on the earlier building in UE 6, The Buildings of the Third Dynasty (Woolley 1974: 45–54, but see also Woolley 1965: 5–8), with an account of the later building in UE 9, The Neo-Babylonian and Persian Periods (Woolley 1962: 23–33). In preparing his final report on the earlier Ganunmah, completed in 1935, Woolley renumbered the rooms on the plan of the building, and his description of the earlier Ganunmah reflects his renumbering. But Woolley never produced a plan of the earlier Ganunmah with the renumbered rooms and apparently did not (systematically) change the find spots of objects in the catalog he planned to include in the volume. Penn Museum took responsibility for publishing The Buildings of the Third Dynasty of Ur. With no updated plan at hand (and no concordance of room numbers assigned in the field and room numbers intended for the final report), the Publications Committee (Robert H. Dyson, Jr., Samuel Noah Kramer, and James B. Pritchard) reproduced the plan Woolley had published fifty years earlier in his preliminary 19.  Inscriptions of various Isin-Larsa rulers, e.g., Nur-Adad (Frayne 1990: E4.2.8.2) and his son Sin-iddinam (Frayne 1990: E4.2.9.10), and Kudur-Mabug/Warad-Sin (Frayne 1990: E4.2.13.9 and E4.2.13.10), refer to the building as ga2-nun-maḫ. Inscriptions of Kurigalzu I (Gadd 1928: 49, nos. 162–163) and Marduk-nadin-ahhe (Frame 1995: B2.6.1) record their rebuilding of e2 ga2-nun-maḫ, with e2 likely a determinative. (Personal communication, Irene Sibbing Plantholt, Jan. 2013.)

Woolley’s Excavations at Ur

23

report. The Committee recognized that Woolley’s text description did not match the plan, as well as discrepancies between Woolley’s text and the original field catalog. In cases where Woolley specifically noted an artifact as coming from a particular room in his text, the Publications Committee gave primacy to Woolley’s location, but nevertheless included the find spot in his original catalog in parentheses. In other cases where Woolley did not reference an object in his catalog as from a particular room in his text description, the Committee cited the find spot assigned in the field. 20 Öhnan Tunça (1986: 286–287) recognized the discrepancies between the plan of the earlier Ganunmah and Woolley’s text description and catalog. He attempted to create a concordance of room numbers on the preliminary plan with room numbers in Woolley’s text description. His concordance was largely, though not completely, correct. The Penn Museum’s Publication Committee had a second generation typed copy of Woolley’s field catalog cards, but apparently did not have ready access to (or never consulted) Woolley’s original field records in the British Museum. Nor, apparently, did Tunça! Had they consulted the original cards, they would have found that someone—the handwriting is not Woolley’s—had changed the room numbers and find spots on the cards (or at least the majority of them) to correspond to Woolley’s renumbering. Woolley’s original field cards are a critical resource not just for his notes, but also for the annotations added to them post-excavation, and his cards need to be consulted in any reassessment of his excavations. 21 By way of an addendum, in preparing his final report on the later Enunmah, Woolley renumbered the rooms on the plans of the building just as he had re­ numbered the rooms on the plan of the earlier Ganunmah. His description reflects the renumbering, but in the case of the late rebuildings he produced new plans for publication. He also corrected the find spots of objects he included in the Catalogue. For example, U.450, given on the field catalog card as from “T. T. B. 7, between shell pavement and Nebuchadnezzar mud floor” is described in the published Catalogue as “Neo-Babylonian enunmah room 10,” though no one corrected the original catalog card (Woolley 1962: 106).

The Great Death Pit (PG 1237) Woolley not only renumbered the rooms of the Ganunmah and later Enunmah, but he also renumbered the bodies in PG 1237 for his final report. He was, in all 20.  The Publications Committee explained in a note preceding the Catalog (Woolley 1974: 83), “A special problem arises with the designation of room numbers in E-nun-maḫ. In several instances (U.195, 246, 256, 296, 406, 599, 7903) there is a conflict or discrepancy between Woolley’s text and the catalogue. In each instance, the Woolley’s text location is given in the catalogue, followed by the location given in the original field catalogue in parentheses. In a number of other instances, the text gives the locations using room numbers as assigned by Woolley while the catalogue gives the original coded field location (e.g. T. T. B 17). While clearly some reasoned correlation between the two exists, it is not entirely certain what the complete correspondence for all numbers is, or why some changes appear to have been made. For that reason, again, we have listed the text location first, followed by the field location (in parentheses) as given in the original field catalogue. Where Woolley’s text fails to mention cataloged items specifically under his various room descriptions, those items appear in the catalogue as originally given with only their field location designations. The resulting report thus leaves some unanswered questions of provenience but at the present time there seems no way to resolve the problem further.” 21.  Aaron W. Schmitt (Johannes Gutenberg-Universität Mainz) is currently undertaking a detailed study of the Ganunmah, accessing Woolley field records at Ur Online.

24

Richard L. Zettler

likelihood, trying to rationalize the sequence of numbers and, possibly, make his excavations of the seventy-four bodies in the death pit look more systematic than they were in reality. 22 But, Woolley himself annotated his field note cards, adding the new numbers, and he produced a plan of PG 1237, with the bodies numbered, for publication. If Woolley’s renumbering was of little consequence for most readers, it had a bigger impact on me as a curator responsible for the Penn Museum’s collections. The 130 or so artifacts from PG 1237 allotted Penn Museum in the division of finds from 1928–1929, carried out in London July 22–27, 1929, were accessioned (with tripartite 30-12 numbers) before Woolley renumbered the bodies, and the museum’s registration cards (from which our earliest computerized databases were created) still had the original body numbers as find spots in the late 1990s. We had to go to London to consult Woolley’s original field records to update our collections database in the late 1990s. We worked extensively with Woolley’s field notes on PG 1237 for the staging of the Royal Cemetery of Ur in Penn Museum’s new Middle East Galleries, and quickly came to appreciate the corrections Woolley’s records enabled us to make to his published account of the Great Death Pit, as well as the little details they added to it. A few examples will perhaps suffice to underscore the critical importance of familiarity with his field notes. 1. The two statuettes of a goat rampant in a tree that Woolley famously dubbed “ram caught in a thicket,” found in the west corner of the PG 1237, and the three lyres, found on top of one another at the southeast end of the death pit, are the bestknown objects from PG 1237 and, arguably, the Royal Cemetery, more generally. Woolley published detailed descriptions of the lyres and at least five of the eight photographs of the lyres in situ. But, Woolley’s field notes also include scale (1/10) drawings of the gold lyre (U.12353) that went to the Iraq Museum in the division of finds; the silver lyre (U.12354), assigned to the British Museum; and the boat lyre (U.12355), on exhibition here at Penn Museum (de Schauensee 2002: pl. 7b). Woolley specifically noted that he made “measured drawings” of the gold lyre, with its elaborately decorated sound box and uprights, before lifting it from the ground (Woolley 1934b: 123), and the note card’s creases and wax stains suggest he kept it close by in the field and/or the lab. Woolley’s notes also contain a scale (1/10) drawing of copper or copper alloy statuette (88 cm high) of a stag rampant in a tree (U.12356) found nearby. After a month of work in the lab Woolley discovered that the stag was actually a pair of stags collapsed on one another. He surmised that the two stags had decorated a wooden lyre (Woolley 1934b: 123). 2. Woolley identified two of the bodies in PG 1237 as juveniles or sub-adults in his field notes. He noted Body 8, in the east corner of the pit, was “apparently a young person as the bones seem small.” He identified the age of Body 72, against 22.  A plan of PG 1237 with the numbers assigned to the bodies in digging ought, at least in theory, to show the order in which Woolley actually excavated the death pit. The plan Woolley published in his preliminary report (Woolley 1929a: pl. XXX1, 2) did not have the bodies numbered, though he did refer to Body 22 (renumbered Body 61) in his text description (Woolley 1929a: 318). A few of his field note cards include sketches of the bodies in the west central part of the pit with both the field and publication numbers, while a sketch plan, found loose among Woolley’s records in the British Museum, shows the whole of the pit with skulls in lieu of bodies and their original field numbers. That plan indicates Woolley started the excavation of PG 1237 in the south corner of the death pit and proceeded to the northeast, where he uncovered the three lyres. He then moved to the west corner of the death pit.

Woolley’s Excavations at Ur

25

the southwest edge of the pit, near the south corner, based on the teeth. He wrote, “Judging from the teeth, which were much discolored—roots being blue—body was that of a quite young person.” Woolley mistakenly identified not Body 8, but Body 9 nearby as having “small bones as of a very young person” in his list of objects found with each body in his final report (Woolley 1934b: 117). He inexplicably didn’t identify Body 72 as a young person in the same list of objects (Woolley 1934b: 120). Mixing up Bodies 8 and 9 in his final report may be of some real consequence inasmuch as the age disparity might explain the two bodies’ different ensembles of jewelry. Body 8 had an unusual ensemble that included a pair of “rather small” gold lunate earrings; two gold wire hair coils, one of which was inside an earring; two silver pins; two gold finger rings made of spiral wire coils, both on one finger; and two necklaces including one of gold, silver and lapis lazuli double conoids with a gold pendant amulet (?) that Woolley described as imitating a “grape-cluster” inlaid with two small pieces of lapis lazuli, the only such oddity with any of the bodies in the death pit. 23 Body 9, on the other hand, had jewelry typical of the other women in PG 1237, including a gold hair ribbon, wreath of pendant poplar leaves, a silver comb with inlaid flowers, gold earrings, a beaded cuff, necklaces and pins, though she wasn’t wearing a choker—Woolley’s “dog-collar”—of triangular gold and lapis beads. She also had two cockle shells with green pigments. 3. Woolley suggested that Body 7, in the east corner of PG 1237, might have been a musician in his field notes, but not in his publications. He wrote, “This may be the harpist, but the bones were not actually mixed up with the harp, though the head was almost touching it.” 24 Woolley’s hesitation in identifying Body 7 as a musician alludes to the fact that he had found the finger bones of one of the women near the musical instruments in PG 800’s death pit “actually in the place of the strings” (Woolley 1934b: 74). If Body 7 was, in fact, a musician, the scene carved on her lapis lazuli cylinder seal (U.12374, 30-12-2) might be taken as depicting her. The seal was divided into two registers. The upper register depicted banqueters, while the lower register showed the banquet’s musical accompaniment, including a female playing a bull-headed lyre, while two children (or, arguably, dwarfs) danced beneath (or in front of) it. The woman playing the lyre may be Body 7; the children might be the two juveniles in PG 1237 (Body 8, next to Body 7, and Body 72). 4. Even a casual glance at Woolley’s field records for the bodies in PG 1237 will reveal, not surprisingly, that the notes Woolley took while excavating the seventy-four bodies in the death pit, including his meticulous description of the jewelry they wore and the objects associated with them represent the primary, most detailed record of what he found. 25 Woolley’s field catalog cards, drawn up later in the house, provide an abridged record of the objects, if occasionally adding details perhaps uncovered in cleaning them, and his published description (Woolley 1934b: 23.  Rather than imitating a grape-cluster, the amulet may be imitating a spiny cockle shell. Similar cockle shell pendants with inlaid eyes come from Khafaje, Houses 2: Kh 9-85 and Kh 9-86 (38-10-98). 24.  Body 7 is visible in the field photograph BM-Ur-GN-1211 (Woolley 1929a: pl. XXXII, 2). Woolley drew Body 7 on its left side, facing northwest, but the photograph would suggest that Body 7’s skull was face up. 25.  A letter from Woolley to H. R. Hall, dated December 28, 1926, perhaps conveys the tedium of the detailed work Woolley had to do in digging PG 1237. Woolley wrote, “We are doing marvelously well: I am sick to death of getting out gold headdresses, but the other things are wonderful: if only we find the tomb to which all that we are finding now belongs ...” (British Museum, Department of Egyptian and Assyrian Antiquities letter books).

26

Richard L. Zettler

116–120) a consolidated version of the field catalog cards. Woolley’s notes on Body 61, in the west corner of PG 1237, will perhaps serve to make the case. 26 Body 61 (Woolley 1934b: pl. 144, BM-Ur-GN-1332) was the most elaborately dressed of the sixty-eight females in the death pit. Woolley did not include the objects with the attendants in PG 1237 (U.12359–12429) in the Catalogue in his final report, but he referred readers instead to the list of objects with the bodies he embedded in his detailed description of PG 1237. Woolley’s published list of objects with Body 61 (Woolley 1934b: 120) reads: Body 61. Gold hair ribbon, gold leaf pendant wreath, wreath of gold triple willowleaves, wreath of gold rings set with lapis disks (cf. Pl. 132 [sic]), silver comb with inlaid flowers, gold earrings, two gold hair rings, three gold flower rosettes on silver stems, “dog collar” of gold and lapis, necklace of gold and lapis tubular beads and carnelian rings, necklace of large gold and lapis fluted balls and carnelian rings, necklace of gold and carnelian and gold double conoids and carnelian rings, necklace of gold balls, lapis double conoids, and carnelian rings, six gold finger rings, bracelet of gold and lapis diamonds and carnelian rings, bead cuff, gold pin of Type 1 with lapis ball head, silver pin of Type 7 b with lapis head, lapis cylinder seal, Pl. 195, silver tumbler. U.12380 (L. BM 122300), (P. 30-12.551).

Woolley’s field notes provide a much more specific and detailed description of the position and components of the objects. His description of Body 61’s frontlet and hair wreaths (1–4 on his field cards) reads: (1) gold ribbon headdress (2) on front of head over everything gold medallion with large lapis centre from four stringed beads. 2 carnelian one bugle [tubular] lapis between [DRAWING] (3) next wreath long shaped gold leaves with veining marked. 3 together. from 4 stringed beads (4) gold beech [poplar] leaves—from 2 stringed beads the usual small carnelian w(ith) 2 bugle [tubular] lapis and one carnelian again.

Woolley’s description of Body 61’s four necklaces is specific about the order of the beads and presumably records what he saw in the ground as he was cleaning them. His descriptions of the first and second of the four necklaces in the published list of objects (14 and 5, respectively, on his field cards), 27 for example, reads: 26.  Hamoudi’s son Yahya took two photographs of Body 61 (dug as Body 22) in the ground: BMUr-GN-1222 and 1223. They are, to all intents and purposes, the same photograph, though one (BM-UrGN-1223) has a card reading “PG 1237 Body 22” in the frame. Woolley 1929a: pl. XXXII, 1 and Woolley 1934: pl. 74b reproduce BM-Ur-GN-1222. 27.  Woolley changed the order in which he listed the objects with the bodies in PG 1237 from his field note cards to his object cards and published catalog. But, it is at least reasonable to assume that the order in which Woolley listed the objects with the bodies in PG 1237 corresponded to the sequence in which he excavated them and/or took them out of the ground. He probably removed those with the lowest numbers in the sequence on the cards before those with higher numbers. That is clear as regards the frontlet and wreaths that Body 61 wore. And, it is probably also the case for her necklaces: three necklaces (5, 6, and 7) probably lay over the necklace of gold, lapis lazuli and carnelian beads, nine rows

Woolley’s Excavations at Ur

27

14) necklace of gold lapis & carn bugle [tubular] beads nine rows wide : the order in each row : carn ring, blue bugle, 2 carn rings, gold bugle, 2 carn rings, carn bugle (?), 2 carn rings, carn bugle, 2 carn rings, gold bugle. 5) necklace of fluted ball beads lapis & gold & carnelian rings : gradated : order of larger beads, carn & 3 gold alternating, carn. lapis (perhaps six, or more?) carn gold etc. : of the smaller beads, gold carn. lapis carn. gold carn lapis.

Woolley’s field catalog cards condense the descriptions and refer to the field note cards for the order in which the beads were strung: for 14, for example, “Necklace of gold and lapis tubular beads & carnelian rings (for order see Field Notes);” for 5, “Necklace of large fluted ball beads gold & lapis, & carnelian rings (for order see Field Notes).” As regards Body 61’s six finger rings (No. 10 on his field note cards), Woolley noted that Body 61 had “gold finger rings, 6 in sets of 3,” presumably implying that she had three rings on each hand, and the field photographs BM-Ur-GN-1222–1223 (see above) confirm the assumption. The field catalog card is specific in describing her rings as “all of cable pattern between plain rims : different sizes.” 28 Woolley’s summary list of objects (1934b: 120) merely lists the presence of a silver tumbler with Body 61, but his notes provide an interesting additional detail, “silver cup held to the mouth : this was hopelessly decayed.” Holding the silver tumbler to her mouth, Body 61 recalls a female banqueter with an attendant in the upper register of Body 7’s cylinder seal (U.12374). The female in the center of the domed chamber in PG 1054 apparently also held a cup—a fluted gold tumbler—to her mouth. In his final report Woolley (1934b: 106) writes that the woman held a gold tumbler in her hands, while his note card describes the gold tumbler as next to the head. Along more general lines, Woolley assumed that PG 1237 was the death pit of a typical royal burial with a tomb chamber and death pit, perhaps on the model of PG 800 whose tomb chamber and death pit he presumed to have been at differ­ ent absolute elevations. He suggested that PG 1237’s tomb chamber was likely de­stroyed in antiquity, but it might have been to the southeast of the death pit, ca. 4 m from its east corner, with its floor at an elevation ca. 1.5 m above the floor of the death pit (Woolley 1934b: 113–114). It had been destroyed in antiquity, and all that remained were “a number of loose blocks of limestone rubble and a short length of a ruined wall of rubble masonry.” Scattered nearby were the presumed remains of its contents, “a great many beads of gold, carnelian, and lapis lazuli of unusually fine quality and color.” As Paul Zimmerman and I argued elsewhere (see above), the tomb chamber and death pit Woolley linked as PG 800 were probably not a single burial, but components of two different royal burials. PG 800 cannot, therefore, provide a model for PG 1237. We have suggested elsewhere that PG 1237, laid out as a tableau mort of a banquet that we tried to recreate for Penn Museum’s new Middle East Galleries, may, in fact, be an intact royal burial of a type that or strands wide (14), with the choker (8) tight around the neck. This corresponds to the order in which Woolley (or, more likely, Katharine) exhibited them on the mannequin head 1934b: pl. 144. 28.  Penn Museum has three of Body 61’s gold rings (30-12-553 and 30-12-555A-B). The Iraq Museum probably has the other three rings IM 8314, 8315, and 8316 (Government of Iraq, Directorate-General of Antiquities, 1942: pl. XX).

28

Richard L. Zettler

Woolley did not originally recognize, in which the principle (royal) figure and his/ her attendants were buried together in an open pit. Body 61, a female holding, as noted above, a silver cup to her mouth, in the west corner of the burial pit, near the two “rams caught in a thicket,” may be the principle royal burial. Body 61 (U.12380) is more elaborately dressed than any of the other females in PG 1237, and her jew­ elry closely resembles that worn by Puabi (Baadsgaard 2008: 186–187, 217, 224, and 226–227). She wore a three-tiered headdress (see above) that included a frontlet and two wreaths, one with willow leaves and one with poplar leaves. The wreath with willow leaves is the only example of the type other than the one Puabi wore and it had alternating eight-petalled rosettes and clusters of willow leaves. Moreover, Body 61, like Puabi, had a higher frequency of carnelian beads and fewer lapis lazuli beads compared to other royal attendants. She also had six gold rings, three on each hand, all of the cable pattern between plain rims 29 and a lapis lazuli cylinder seal (two registers with the upper featuring a banquet scene and the lower an animal contest). The specific identity of Body 61 remains to be determined.

Woolley’s Excavations and Ceramics Woolley claims to have “fully recorded” pottery vessels in the field, but, in fact, he paid scant attention to ceramics, at least in comparison to the emphasis archae­ ologists put on collecting and recording pottery today. Except for painted Ubaid pottery, Woolley does not appear to have collected sherds. He developed a typology—or, judging by type numbers on his field cards that differ from his published types, a series of typologies—based largely on form, and recorded whole vessels by these abstract types (Fig. 7a–b). As he noted, individual vessels, with their seemingly minor variations, lost any real value once a class or type had been established (see above). Woolley’s chapters on pottery in his final reports are perfunctory and seem­ ingly intended to do little more than introduce his period-specific type series. His comments on Kassite pottery are particularly derogatory (Woolley 1965: 98), but, for archaeologists who have excavated Kassite occupation levels, ring true. As he wrote:

Fig. 7a. Woolley and Mallowan sorting whole Fig. 7b. Packing ceramic vessels for shipment. ceramic vessels from the seventh (1928–1929) field season. Fig. 7. BM-Ur-GN-1327 and 1328. 29.  Body 61’s three rings in Penn Museum are roughly the same size as Puabi’s rings, but weigh substantially more (Baadsgaard 2008: 342, Table 6.26). 30-12-553 has a diameter of 2.1 cm and weighs 6.3 g. 30-12-555A has a diameter of 2.1 cm and weighs 4.8 g. 30-12-555B has a diameter of 1.8 cm weighs 4.0 g.

Woolley’s Excavations at Ur

29

Taken as a whole, Kassite pottery is uninteresting. It is altogether utilitarian and possesses the technical merits of utilitarian ware, being well turned, very evenly fired, and with a surface adequate to its purpose, rough, water-smoothed or with a plain engobbage (true slip does not seem to be used); but the shapes are uninspired, and there is no attempt at decoration of any sort.

Woolley noted pottery types from particular contexts in his text description, presumably where he found whole vessels on floors (see, for example, EM No. 1 Gay Street [Woolley and Mallowan 1976: 95] or AH No. 1 Broad street [Woolley and Mallowan 1976: 36]) and he recorded the pottery from burials in the various tabular analysis of graves he included in his final reports. For the late Early Dynastic and Sargonic burials from the Royal Cemetery he cited just type numbers; for the Old Babylonian burials under the floors of private houses in AH, EM and BC he cited type numbers, with the number of exemplars of each type in parentheses. 30 Woolley’s seeming indifference to pottery is perhaps belied by the numerous sketches of whole pottery vessels, with measurements and (brief) notes on wares, e.g., drab clay, light drab clay or red clay, and surface treatments, e.g., hematite wash or burnished hematite wash (a surface treatment typical of certain types in late Akkadian burials), (rim) painted black (common in the later Old Babylonian period), or, incised decoration, on his field note cards recording burials. For example, the late Akkadian burial PG 1422 (see above) included eighteen or nineteen pottery vessels, and Woolley recorded Royal Cemetery or RC types 44, 60, 76, 83, and 200 in his tabular analysis of Second Dynasty Graves (Woolley 1934: 482–483). Woolley sketched seven of those eighteen or nineteen vessels on his note cards. All had Roman numeral type numbers written over or next to them; one had LI (51) crossed out and “new 83” written next to it, with 83 being his published type number for the jar in question. None had U-numbers that Woolley routinely wrote next to objects he put in his field catalog. The Old Babylonian LG 157 (BCG/10), a child’s larnax burial under the floor of House 30/B had seven pottery vessels against the outside of the larnax: (Larsa) types 44, 58a (two), 69a (two), 90, and 125. 31 Woolley drew five of the seven on the single note card for the burial. All had Arabic numeral 30. Woolley included only a small number of the whole ceramic vessels in his field catalog. He presumably consigned representative samples to Baghdad, London and Philadelphia, but may not have kept all of the whole vessels he collected. Arthur M. Mintier (Waynesburg College), who represented Penn Museum on the excavations at Seleucia (Tell Umar) in 1929–1930, visited Woolley’s excavations in early December 1929. In a post-script to a letter to Horace Jayne, dated December 8, 1929, Mintier wrote, “On my visit to Ur, I noticed a great deal of pottery intact and perfect shape that was to be broken up because they had duplicates at your museum or in London or Baghdad—I wonder if some of those could be available for the college where I teach—If we had ten or twelve, just as examples of pottery to use in my courses I think it would greatly enhance the effectiveness of my courses particularly—Mr. Wooley [sic] said he would be glad to send them to me if you gave him permission.” (Penn Museum Archives, Horace H. F. Jayne Director’s Office Records, Box 10). Even if Woolley did not actually break up unwanted ceramics, he certainly considered duplicates “disposable,” as his offer to give them to Mintier makes clear. Legrain’s letter to Horace Jayne, dated July 27–29, 1929, reporting on the division of finds, suggests he shared Woolley’s attitude. “I insisted on the wish of our museum to get a fair share of all the undivided uninteresting surplus of pots and small objects for distribution to minor museums in America.” (Penn Museum Archives, Horace H. F. Hayne Director’s Office Records, Box 10, Folder 5). 31. Woolley recorded burials by excavation areas, but later split them into burials of different periods. So, for example, he used the prefix BCG to record the burials he found on the Mausoleum site; he included BCG/10 with the Larsa Graves (LG-prefix) in Woolley 1976: 209.

30

Richard L. Zettler

type numbers on or next to them that do not correspond to his published Larsa type numbers and some an RC (Royal Cemetery) number. As with PG 1422, none had a U-number. I assume Woolley himself drew the pottery on the cards. Since the sketches are on the note cards he kept in his jacket pocket, he probably made them on site as he was digging. The sketches are particularized, as if he were trying to render a whole or largely complete vessel sitting in front of him, and it may be that Woolley used his sketches to construct his type series. In terms of their utility today, Woolley’s sketch­es, measurements and ware descriptions could be extracted from his note cards to create a visual record of the pottery that had been found in particular burials as we often do for publication today. The illustrations would certainly have a bigger impact than a list of type numbers. It might also be useful to collect and compare all of Woolley’s drawings of particular pottery types to determine how much variability existed in his types and how his types compare to types, for example, that Armstrong and Gasche (2014) distinguished in their authoritative study of second millennium BCE pottery. Armstrong and Gasche tried to match their types to Woolley’s types, but had only his published plates of period-specific pottery types to work with. It is perhaps worth noting, parenthetically, that in addition to sketches of whole pottery vessels, Woolley’s note cards (and field catalog cards) also include numerous sketches of metal vessels. PG 1422, for example, included seventeen metal (silver and copper or copper-alloy) vessels, including a hoard of ten copper or copper-alloy vessels, found outside the wooden coffin, on its southwest side, corroded together (Woolley 1934b: pl. 184a). Woolley’s field note cards contain sketches of four of the metal vessels, while his catalog cards add seemingly more meticulous drawings of five others from the burial. The sketches all have Roman numeral type numbers written on them and a couple have published type numbers in parentheses; two of the four on the field note cards have U-numbers and one has “not kept” written next to it. Like Woolley’s sketches of whole pottery vessels, his drawings of metal vessels illustrate individual objects not idealized types. They are particularly valuable today because most of the silver and almost all of the copper or copper-alloy vessels were mineralized and in poor condition when Woolley found them. Their shipment from Basra to London (and, subsequently, from London to Philadelphia) and their storage in less than optimal conditions, certainly in Penn Museum, has resulted in considerable losses over the years.

Conclusion Woolley’s preliminary and final reports on his excavations at Tell al-Muqayyar, as well as his popular narratives, e.g., Ur of the Chaldees (1929) and Excavations at Ur (1954), stand as an authoritative and exemplary account of his twelve years of arduous work at the site. Nevertheless, Woolley’s unpublished field records, if an incomplete dataset, are an incalculable resource that can still be mined today to correct and supplement his published reports and occasionally challenge his interpretation of his findings. Thanks to the critical underwriting of the Leon Levy Foundation, as well as the generous support of the Kowalski Family Foundation and the Hagop Kevorkian Foundation, Woolley’s field notes and catalog cards, as well as his field photographs, are accessible at Ur Online.

Woolley’s Excavations at Ur

31

The tens of thousands of objects Woolley excavated, particularly those he included in his field catalog, still have much to tell us, as we pose fresh research questions and exploit advanced technologies to analyze them. Most of the objects from Woolley’s excavations at the British Museum and Penn Museum are now searchable at Ur Online, though both museums have more work to do to complete inventories and photographic documentation of the collections. We hope in the future the Iraq Museum will contribute the objects from Woolley’s excavations in its collections to Ur Online, and we encourage museums and other public institutions around the world that may have acquired objects from Ur, and especially objects with U-numbers, to include them in our database. Woolley’s reports, ledgers and letters, both personal and professional, in the British Museum’s Central Archives and Penn Museum’s Archives represent a rich and as yet underexploited source of information on the dynamics of Woolley’s excavations and publication efforts. The British Museum’s Central Archives, as noted, has detailed indices to its holdings, but the papers (and photographs) have not yet been digitized. Penn Museum’s archival records, some of which duplicate documents, e.g., Woolley’s reports and accounts, in the British Museum have been scanned and added to Ur Online, but they are not yet fully accessible. The papers and recollections of individuals who collaborated with Woolley (or just visited the site when he was working there) may also contribute to a fuller picture of his excavations, as letters and photographs of Leon Legrain and the papers of James R. Ogden (Millerman 2015) have made so evident. Making these ancillary archival documents accessible at Ur Online is a desideratum, but one that will entail considerable time and effort over the long term. Sir Leonard Woolley left behind a rich record of an astonishing twelve years of large-scale excavations in the “cradle of civilization,” and we hope our efforts over the last few years and into the future will preserve and make as much of that record as possible available to future generations of archaeologists and the general public alike.

Bibliography Alarcon, T. de. 2016 Of Daggers and Scabbards: Evidence from Organic Pseudomorphs and X-Radiography. Penn Museum Blog, April 16, 2016. See https://www.penn.museum/blog/ collection/conservation/of-daggers-and-scabbards-evidence-from-organic-pseudomorphs-and-x-radiography/ (accessed June 27, 2018). Armstrong, J. A. and Gasche, H. 2014 Mesopotamian Pottery: A Guide to the Babylonian Tradition in the 2nd Millennium B. C. Ghent and Chicago: The Oriental Institute. Baadsgaard, A. 2008 Trends, Traditions and Transformations: Fashions in Dress in Early Dynastic Mesopotamia. Ph.D. dissertation, University of Pennsylvania. Baqir, T. 1961 Forward. Sumer 17: 1–12. 1962 Forward. Sumer 18: 5–14. Black, J. and Spada, G. 2008 Texts from Ur: Kept in the Iraq Museum and the British Museum. Nisaba 19. Messina: Dipartimento di Scienze dell’Antichitá.

32

Richard L. Zettler

D’Agostino, F., Pomponio, F., and Laurito, R. 2004 Neo-Sumerian Texts from Ur in the British Museum. Nisaba 5. Messina: Dipartimento di Scienze dell’Antichitá. Frame, G. 1995 Rulers of Babylonia: From the Second Dynasty of Isin to the End of Assyrian Domination (1157–612 BC). RIMB 2. Toronto: University of Toronto Press. Frayne, D. 1990 Old Babylonian Period (2003–1595 BC). RIME 4. Toronto: University of Toronto Press. Gadd, C. J. 1928 Royal Inscriptions. UET 1. London: Trustees of the British Museum and of the University Museum, University of Pennsylvania, Philadelphia. Gardiner, H. 2016 Corroded Ruin or Hidden Treasure? The British Museum Blog, March 3. See https://blog.britishmuseum.org/corroded-ruin-or-hidden-treasure-a-third-millennium-bc-copper-alloy-cauldron-from-ur/ (accessed June 27, 2018). 2017 Conservation and Observation: More on a Copper-Alloy Cauldron from Ur. The British Museum Blog, February 20. See https://blog.britishmuseum.org/conservation-and-observation-more-on-a-third-millennium-bc-copper-alloy-cauldronfrom-ur/ (accessed June 27, 2018). Gordon, C. H. 2000 A Scholar’s Odyssey. Atlanta, GA: Society of Biblical Literature. Government of Iraq, Directorate-General of Antiquities 1942 A Guide to the ‘Iraq Museum Collections. Baghdad: Government Press. Hauptmann, A. and Klein, S. 2016 Golden Artifacts from the Royal Tombs of Ur. Metalla 22: 84–88. Hauptmann, A., Klein, S., Zettler, R., Baumer, U., and Dietemann, P. 2016 On the Making and Provenancing of Pigments from Early Dynastic Royal Tombs of Ur, Mesopotamia. Metalla 22: 41–74. Jacobsen, T. 1953 Review of Léon Legrain, Ur Excavation Texts III: Business Documents of the Third Dynasty of Ur. American Journal of Archaeology 57: 125–128. Jansen, M. 2019 Geochemie und Archäometallurgie des Goldes der Bronzezeit in Vorderasien. Ph.D. dissertation. Ruhr-Universität Bochum. See https://hss-opus.ub.ruhr-unibochum.de/opus4/frontdoor/index/index/docId/6686. Jansen, M., Aulback, S., Hauptmann, A., Höfer, H. E., Klein, S., Krüger, M., and Zettler, R. L. 2016 Platinum group placer minerals in ancient gold artifacts—Geochemistry and osmium isotopes of inclusions in Early Bronze Age gold from Ur/Mesopotamia.” Journal of Archaeological Science 68: 12–23. Kaercher, K. 2015a Pseudomorphs on Objects from Ur. Penn Museum Blog, February 2. See https:// www.penn.museum/blog/collection/conservation/ur-project-january-2015/ (accessed June 27, 2018). 2015b Daggers and Fighting Knives—Ur Project: July 2015. Penn Museum Blog, August 4. See https://www.penn.museum/blog/museum/daggers-and-fighting-knives-urproject-july-2015/ (accessed June 27, 2018). Klein, S., La Niece, S., and Hauptmann, A. 2016 Objects from the Ur Collection of the British Museum: Sampling and Analytical Investigations. Metalla 22: 89–97.

Woolley’s Excavations at Ur Legrain, L. 1951 Seal Cylinders. UE 10. Oxford: Trustees of the British Museum and of the University Museum, University of Pennsylvania, Philadelphia. Loding, D. 1976 Economic Texts from the Third Dynasty. UET 9. Philadelphia: Trustees of the British Museum and of the University Museum, University of Pennsylvania, Philadelphia; the Babylonian Fund. Luby, E. M. 1990 Social Variation in Ancient Mesopotamia: An Architectural and Mortuary Analysis of Ur in the Early Second Millennium B. C. Ph.D. dissertation. State University of New York at Stony Brook. Millerman, A. 2008 Interpreting the Royal Cemetery of Ur Metalwork: A Contemporary Perspective from the Archives of James R. Ogden. Iraq 70: 1–12. 2015 The Spinning of Ur. Ph.D. dissertation. University of Manchester. Molleson, T. and Hodgson, D. 2003 The Human Remains from Woolley’s Excavations at Ur. Iraq 65: 91–129. Pittman, H. and Naomi, F. M. 2014 Puabi’s Diadem(s): The Deconstruction of a Mesopotamian Icon. Pp. 106–131 in From Ancient to Modern: Archaeology and Aesthetics, eds. J. Y. Chi and P. Azara. Princeton: Princeton University Press. Rakic, Y. 1998 Rescue and Restoration: A History of the Philadelphia “Ram Caught in a Thicket.” Expedition 40: 51–59. Salzmann, E. 2019 Silver, Copper and Bronze in Early Dynastic Ur, Mesopotamia. Bochum: Verlag Marie Leidorf, GmbH. Schauensee, M. de 2002 Two Lyres from Ur. Philadelphia: University of Pennsylvania Museum of Archaeology and Anthropology. Spade, G. 2006 Testi economici da Ur di period paleo-babilonese. Nisaba 12. Messina: Dipartimento di Scienze dell’Antichitá. Tunça, Ö. 1986 Remarques sur l’architecture des temenos d’Uruk et d’Ur a la periode de la IIIe Dynastie d’Ur. BaM 17: 255–292. Van De Mieroop, M. 1992 Old Babylonian Ur: Portrait of an Ancient Mesopotamian City. Journal of the Ancient Near East Society 21: 119–130. Wailly, F. al-. 1963 Forward. Sumer 19: 1–7. Woolley, C. L. 1923 Excavations at Ur of the Chaldees. AJ 3: 311–333. 1925 The Excavations at Ur, 1924–1925. AJ 5: 347–402. 1928 Excavations at Ur, 1927–8. AJ 8: 415–448. 1929a Excavations at Ur, 1928–9. AJ 9 (1929): 305–343. 1929b The Flood. Times. 16 Mar. 1929: 13. 1932 Excavations at Ur, 1931–2. AJ 12 (1932): 355–392. 1934a Excavations at Ur, 1933–4. AJ 14 (1934): 355–378. 1934b The Royal Cemetery. UE 2. London: Trustees of the British Museum and of the University Museum, University of Pennsylvania, Philadelphia.

33

34

Richard L. Zettler

1955 The Early Periods. UE 4. London: Trustees of the British Museum and of the University Museum, University of Pennsylvania, Philadelphia. 1962 The Neo-Babylonian and Persian Periods. UE 9. London: Trustees of the British Museum and of the University Museum, University of Pennsylvania, Philadelphia. 1965 The Kassite Period and the Period of the Assyrian Kings. UE 8. Oxford: Trustees of the British Museum and of the University Museum, University of Pennsyl­ vania, Philadelphia; Oxford University Press. 1974 The Buildings of the Third Dynasty. UE 6. Philadelphia: Trustees of the British Museum and of the University Museum, University of Pennsylvania, Philadelphia. Woolley, Sir L. and Mallowan, Sir M. 1976 The Old Babylonian Period. UE 7. London: Trustees of the British Museum and of the University Museum, University of Pennsylvania, Philadelphia; British Museum Publications. Zettler, R. L., Alarcon, T. de, Hafford, W. B., Jansen, M., and Miller, N. F. 2017 The Ur Digitization Project: Examination of the Metals from an Akkadian Tomb at Ur. Pp. 161–170 in Engaging Conservation: Collaboration across Disciplines, eds. N. Owczarek, M. Gleason, and L. A. Grant. London: Archetype Publication. Zimmerman, P. C. 1998a A Critical Re-examination of the Early Dynastic “Royal Tomb” Architecture from Ur. M. A. Research Paper, Department of Anthropology, University of Pennsylvania. 1998b Two Tombs or Three? P. 39 in Treasures from the Royal Tombs of Ur, eds. R. L. Zettler and L. Horne. Philadelphia: University of Pennsylvania Museum of Archaeology and Anthropology.

Sîn-City: New Light from Old Excavations at Ur Jonathan Taylor British Museum

Introduction The city of Ur first commanded serious Assyriological attention in the 1850s. A series of explorations by British consul Taylor showed the promise held by the site and would reveal the city’s ancient name. Ur was soon enthusiastically connected with Ur of the Chaldees, known from the Bible as home to the patriarch Abraham. Yet it was not until the 1920s that extensive excavations were undertaken. Leonard Woolley headed a joint British Museum-University Museum (now Penn Museum) expedition that would capture the imagination of scholars and public alike. Almost a century later, a renewed British Museum-Penn Museum collaboration has brought Woolley’s excavations into the twenty-first century. The Joint Expedition’s finds, together with its rich archive of contemporary records and photographs, have been digitally reunited in an open access web resource. Researchers have unparalleled access to the finds and associated records from a major site spanning classical Mesopotamian history. New avenues of research have opened, and a wealth of new material is available.

Background to the Ur of the Chaldees Project In 2011 a collaborative project was initiated between the British Museum and the Penn Museum. Thanks to generous lead funding from the Leon Levy Foundation, 1 Ur of the Chaldees: A Virtual Vision of Woolley’s Excavations would digitize the finds from Leonard Woolley’s 1922–1934 expedition to Ur, together with the associated archives in the British Museum. 2 The primary aim of the project was to facilitate and stimulate new research on one of the most important archaeological excavations in Iraq. The initiative for this project came from Richard Zettler, who had long identified the potential offered by linking object photographs with the archival records relating to them; he laid the foundations that made our endeavor possible. Although the finds from Ur have been extensively published, work on them has been complicated by the difficulty in correlating their excavation numbers with their museum registration numbers: where was any given object, and what identity 1.  Further supported by the Kowalski Family Foundation, and the Hagop Kevorkian Fund. 2.  The results are available at http://www.ur-online.org.

35

36

Jonathan Taylor

did it have in that museum? An additional level of significance was added to this issue by the looting of the Iraq Museum in 2003. Many objects were stolen, storerooms were left in disorder, and registration documentation was damaged. By clarifying which objects had been allocated as the Expedition’s share of the division, it would be possible to determine which objects should be in the Iraq Museum collection; in turn, this would allow staff to identify any losses. At the British Museum Sarah Collins (then Curator of Early Mesopotamia) had been compiling notes on the finds from Ur in response to enquiries from colleagues. In 2007 she began to formalize this material into a comprehensive concordance. This incorporated a set of file-cards that had been created by John W. Parsley, a former General Post Office employee who put his organizational skills to use dur­ ing the 1950s and 1960s as Executive Officer in what was then the Department of Western Asiatic Antiquities at the British Museum. 3 His role included handling photographic requests, so to him fell the task of managing the Joint Expedition’s albums of excavation photographs. While Woolley had drawn extensively from these to illustrate his reports, many were never published. 4 They had proven especially useful for the photographs they contained of objects not in the British Museum collection. Parsley’s index listed the excavation numbers (U.-), the museum to which the object had been assigned in the division, its registration number there (where known), publication references, and any other useful information. In 2011 a preliminary phase funded by the Leon Levy Foundation allowed Birger Helgestad (Project Curator at the British Museum) to complete the digitization of these records, and to convert them into a sophisticated database. Building on the success of this work, Ur of the Chaldees (2013–2016) would give the initiative greater ambition. Comprehensive digitization of the complete corpus of finds in London and Philadelphia would make the excavations accessible in a way unparalleled for any other site in Iraq.

The First Hundred Years of Explorations and Excavations at Ur 5 The site of Ur became known to western scholarship through Pietro della Valle’s travels in 1625. 6 The collection of objects he brought back, some of the earliest Meso­ potamian objects to reach Europe, included an inscribed brick and some cylinder seals. James Baillie Fraser visited in 1835, 7 and the steamer Euphrates passed by during its survey of that river in 1835–1837. 8 William Kennett Loftus visited Ur in 1850,  9 in connection with the Turco-Persian Frontier Commission (1849–1852), and again in 1853, whilst excavating in Mesopotamia on behalf of the Assyrian Excavation Fund. Major-General Henry Creswicke Rawlinson, East India Company Resident in Baghdad, persuaded the British Museum to direct John George Taylor, British Vice-consul at Basra, to include Ur among the targets of his exploration of 3.  I am grateful to Terence Mitchell for providing information relating to Parsley and his index. 4.  The full set is made available for the first time now through our project. 5. Convenient summaries of the site and excavations there, with bibliography, are available in Zettler 1998 and Zettler and Hafford 2015. 6.  A translation of his findings is given in Zettler 1998: 10. 7.  See Fraser 1840: 90–94; referring to measurements taken by Ross (1842: 129–130). 8.  See Ainsworth 1838: 179–80; cf. Chesney 1850: I 93. 9.  Described in Loftus 1857: 127–35.

Sîn-City: New Light from Old Excavations at Ur

37

southern Mesopotamia. Taylor visited Ur in January and February–March 1853, launching excavations in December 1853–February 1854 and October–December 1854. Sufficient funds remained for him to dig again in January–February 1858. 10 Crucially, Taylor discovered foundation cylinders of Nabonidus in the corners of the ziggurat mound. These were subsequently read by Rawlinson, revealing the site’s ancient name, Urim, thus establishing its connection with the Ur Kasdim of the Bible (Genesis 11), whence its persistent moniker, “of the Chaldees.” 11 Taylor published reports on his work with commendable speed. His first report was written in March 1854, just one month after his season ended. Having been read on his behalf at a public meeting of the Royal Asiatic Society in July 1854, it appeared in print in the Journal of the Royal Asiatic Society the following year. The second report, authored in January 1855 was read in May that year and likewise published in the 1855 volume of the journal. Those reports include notes on where key objects were found, and even measurements and annotated drawings. While not adequate by modern standards (or indeed those of Woolley’s time), his efforts do demonstrate a concern for what we are now able to understand as context, and an attempt to record data that can be scrutinized by peers. His reports provide valuable information not just about where objects were found, but also how they were found. For example, he tells us that those important foundation cylinders of Nabonidus found by him at the corners of the ziggurat had been stood on their end (rather than laid on their sides). This kind of information is lacking from archaeological reports until the Austrian excavations at Borsippa over 100 years later. Together they indicate that cylinders were conceived as “vertical” objects. 12 The “horizontal” orientation engrained in our minds is a product of privileging the text over its vehicle. Reading orientation forces the cylinder onto its side, and this is how it has typically been displayed in museums 13 and presented in publications. Packing lists sent by Taylor to the British Museum show careful labeling and include details of provenance not just by site, but also more specifically, together with contextualizing information, such as “The coffin is packed with clay tablets; inscribed and pieces of inscribed Priapi [i.e., cones] from Tomb Mound. These articles were not found in the coffin,” (cf. Taylor 1855a: 274) or “Containing two pieces of wooden beams dug out of the solid brick composing principal building at Mugeyer. Just above the beam was found one cylinder. The pieces form one beam (cf. Taylor 1855a: 264). N.B. all the jars and vases marked with the same letter were found in one coffin.” Documentation of the time also yields details of instructions issued to ships’ captains on how to safeguard objects from damage due to rolling of the vessel or water in the hold, as well as security once the objects reached their first port of call in Mumbai. 14 Despite the potential shown by Taylor’s research, Ur would be undisturbed (by archaeological explorers) for decades to come. Woolley (1929: 14–15) relates the 10.  The finds from which were registered as the 1859-10-14 collection. 11.  Skepticism about the identification was voiced already by Oppert (1863: 259 and n. 3). Loftus (1857: 131) noted that Rawlinson had frequently identified Uruk as the biblical Ur of the Chaldees (as in Rawlinson 1850: 481). 12.  Note also the copper cylinders of Nur-Adad found in situ at Ur, positioned upright (Woolley 1939: p. 38, pl. 18a) and the archaising foundation cones of Sin-balassu-iqbi, found in situ, on end (Woolley 1939: pp. 63–64, pl. 27b). 13.  An exception is the display in Yale, shown in photos published in Foster 2013. 14. For a discussion of Assyrian objects arriving in Mumbai around this time, see Guha 2015: 67–98.

38

Jonathan Taylor

existence of a Philadelphia expedition to Ur, the fruits of which are said to remain unpublished. This seems to be a misunderstanding of two brief visits by a Philadelphia team: 1) a two-hour visit in February 1885 by the Wolfe Expedition, which examined tells between Baghdad and Ur; 15 and 2) a visit by John Punnett Peters late in May 1890 in the course of a tour of the south after the second expedition to Nippur. 16 Peters reports that on this tour he intended to dig sondages and supplement previous work by other diggers; Ur was among his principal targets. He notes (1897: 296) that locals had been using the site (but not the ziggurat 17) as a brick mine; thousands had been piled up at the time of his arrival. Peters (1898: 297) does report taking a door socket he had found on the surface. 18 He said of Ur (1898: 300), “I have seen no mound which seemed easier and safer to excavate, or promised richer results than Mughair.” Security was a persistent concern in this region at the fringes of Ottoman authority, as reported by Taylor, 19 Loftus (1857: 134–35)—with great mirth at the jitteriness of his Turkish escort—and Hilprecht (1903: 172, 177). Similar concerns were listed by Woolley (1929: 14) in explanation of the previous lack of work at Ur, alongside the timeless “want of pence which vexes public institutions” (1929: 15). A proposed American expedition to Ur was aborted when in 1901 the sultan refused to issue Edgar Banks with a firman (Hilprecht 1903: 177 n. 1). At the end of the First World War, the British Museum was keen to return to what had been normal business before the War. This included fieldwork in Mesopotamia. The Museum persuaded the army Intelligence Service to second former curator Reginald Campbell Thompson to archaeological duties. He made soundings for one week in 1918, before moving on to nearby Eridu. Leonard W. King had been scheduled to replace Thompson in 1919, but this was rendered impossible by King’s ill health; he died in August that year. 20 Into the breach stepped Henry R. Hall, likewise released from the Intelligence Service. Hall excavated at Ur, Eridu, and Ubaid between February and May 1919, publishing his results in traditional format in a series of preliminary reports as well as a final publication in 1930, again providing a wealth of detail about the circumstances in which the excavations took place. This expedition produced a more accurate map of Ur. The sketch plan by C. O. Waterhouse was based on 1922 aerial photography (then a new technology), notes by the expedition architect F. G. Newton, and an original sketch by 2nd Lieutenant O. D. O’Sullivan RE. The changing times were also reflected in the workforce. While Campbell Thompson had hired men from the local Muntafiq tribe, Hall supplemented these men, boys, and girls 21 with a group of seventy Turkish POWs, headed by two senior NCOs, under a British sergeant-major. Hall also benefitted 15.  Peters 1897: 345–46; Budge (1920: I 241) would make a brief visit in 1888. 16.  Peters 1897: 296–98. 17.  Fraser (1840: 91) had likewise reported that the ziggurat had not been mined for bricks. 18.  He apparently left behind several stone tablets he found defaced to varying degrees. 19.  Letter of 14th January 1854 (Sollberger 1972: 137, n. 21) and Rawlinson’s letter of 4th March 1853 (Sollberger 1972: 131). Taylor reports, in relation to his planned work at nearby Nuwaweis during this expedition, that his workers were armed with muskets and swords and would be able to withstand siege in the trenches until help arrived. 20.  Ironically, following a series of inoculations designed to protect him against typhoid and other serious diseases that posed a risk to people in the region (Budge 1925: 178). 21.  The system Hall describes differs from that employed by Rassam (for which see Rassam 1897: 199). Men were paid 1 rupee/day for their efforts, children half that. Each digger had the assistance of three children to clear the spoil he produced.

Sîn-City: New Light from Old Excavations at Ur

39

from the services of ‘Amran ibn-Hamud and three other headmen from the German excavations at Babylon that had been halted by the British army’s advance in 1917. The excavations of Hall and Campbell Thompson took place in the immediate aftermath of the main fighting in the Mesopotamia campaign; they can be seen as the last excavations of the old era. 22 Archaeological work was then put on hold by the India Office until a treaty with the region’s former colonial power, Turkey, could be concluded (the Treaty of Sèvres, August 1920) and new antiquities laws implemented. New nations in the Middle East were being formed from the ruins of the Ottoman empire. The old provinces of Baghdad and Basra, and after some discussion also Mosul, became Iraq. Britain was given a League of Nations Mandate, the stated intention of which was to help Iraq manage the transition to full statehood. Responsibility was subsequently transferred to the Colonial Office, where T. E. Lawrence, Woolley’s old colleague from the Carchemish excavations before the War, was in a pivotal position. A committee (headed by British Museum Director, Frederick Kenyon) would quality control applications for archaeological permits. The University Museum had been keen to undertake fieldwork in Mesopotamia, and saw the British Museum as a partner which could help them achieve this now. The initial idea was that Philadelphia would provide the majority of the funds, while London would provide access and expertise. Thus the Joint Expedition was formed, Leonard Woolley selected to lead its excavations, and Ur chosen as the focus of work. 23 Meanwhile in Iraq, Gertrude Bell was made Honorary Director of Antiquities. The antiquities laws she drafted were not passed by the Iraqi cabinet at the first attempt in November 1922, so a provisional permit had to be issued in advance of Woolley’s first season; the team at this point already being in the field, waiting to start. Half the finds would be assigned to the Expedition—which would in turn be split equally between the British Museum and Penn Museum—but the other half would remain in Baghdad. The new antiquities laws, eventually passed in 1924, stipulated instead that all objects deemed necessary to ensure that the future Iraq Museum collection was scientifically adequate would be retained; other objects could be divided with the Expedition such that they received a representative share. Each season the Director of Antiquities would be sent an inventory of finds prior to their visit to the site to inspect the finds, which were laid out on tables. 24 Woolley issued

22. An account of the transitional period that followed can be found in Bernhardsson 2005 and Goode 2007, especially Chapter 9. 23.  New light on the origins and workings of the Joint Expedition is being shed by Agnes Henriksen’s research using the archives from London and Philadelphia. The Central Archives of the British Museum contain two sets of papers relating to the Ur excavations: 1) reports and correspondence from Woolley (the “CE 32 series”); personal and business papers of Woolley (the “WY 1” series). The CE 32 archive consists of seven boxes of material: reports to the Directors (including the kinds of practical details that did not reach the published reports; also drafts of popular reports for the press), correspondence with the Antiquities Department in Baghdad, the Colonial Office, the Foreign Office, the Department of Egyptian and Assyrian Antiquities at the British Museum, other museums, scholars, the press and the expedition funders. The WY 1 archive consists of seven boxes containing Woolley’s copies of official correspondence (similar to the above), as well as personal material such as correspondence with other archaeologists working in Iraq and family members too. The CE 32 and WY 1 archives fell outside the remit of the project and remain undigitized. 24.  See Bell 1927: II plate opposite p. 724 for a photo of this.

40

Jonathan Taylor

field numbers to such finds as survived in reasonable state, from U 1–20094; 25 not all numbers in the sequence were assigned to objects, however. Other finds that could not be preserved or were not wanted were recorded and left on site. A few minor objects without number or provenance were kept. Bell’s opinions on the division process during the first few seasons of work at Ur are recorded in her private correspondence. She notes that the first division lasted an entire day, although the results were deemed satisfactory: “… Mr. Woolley was an angel. We had to claim the best things for ourselves but we did our best to make it up to him and I don’t think he was very much dissatisfied. We, for our part, were well pleased.” 26 The second season was quicker, but Bell was determined that “In my capacity as Director of Antiquities I’m an Iraqi official and bound by the terms on which we gave the permit for excavation. J.M. [Wilson 27] backed me but it broke Mr. Woolley’s heart, though he expected the decision. I’ve written to Sir F. Kenyon explaining.” 28 For his part, Woolley reports a mixed reaction: “The division of objects, however painful in process, was not I think so unfavourable to ourselves as I had feared…. Certainly we retain a very fine collection of objects.” 29 By the third season, “The division was rather difficult but I think J.M. and I were very fair and reasonable—I hope Mr. Woolley thinks the same.” 30 By the fourth season Bell’s approach had matured: “I had to take the best thing they have got … I’m getting much more knowing with practice. I now can place cylindrical and other seals at more or less their comparative date and value, so that I don’t choose wildly according to prettiness.” 31 The Iraqi share forms the primary founding collection of the new national museum fought for and nurtured by Bell. Although initially only a small room in a government building, Bell proudly organized an exhibition with labels in English and Arabic; in this she was assisted by Abdul Qadir Pachanji, a former employee of the antiquities museum in Istanbul. A private viewing was arranged for the king and VIPs, and Woolley came to talk about the first season’s finds. The success of this endeavor was demonstrated by the formal establishment of the museum in 1926, in purpose-built accommodation. Bell had known Woolley from his days at Carchemish, describing him in a letter written to her father in November 1922 as “a tiresome little man but a first class digger and an archaeologist after my own heart—i.e. he entirely backs me up in the way I’m conducting the department.” 32 Indeed Woolley helped Bell design the museum’s display spaces. 33 The Expedition’s share was transported to the British Museum, where another division took place. The principal division was an equal one between the British Museum and University Museum, 34 but Woolley took care to assign small, representative collections to other museums, and some individual objects of very common types to the private sponsors who had funded the excavation. This explains the wide 25. Numbers outside this range were set aside for subsequent registrations by the University Museum (20,000s) or British Museum (30,000s). 26.  Bell 1927: II 665–66; March 1923. 27.  Director of Public Works in Iraq. 28.  Bell 1927: II 686–87; March 1924. 29.  Report 8th March 1924. Woolley’s responses are detailed in Bernhardsson 2005: 142–43. 30.  Bell 1927: II 724; March 1925. 31.  Bell 1927: II 754–55; March 1926. 32.  Bell 1927: II 654. 33.  See, e.g., Bell 1927: II 755. 34.  Occasionally, an object would remain unassigned until an equivalent for it would be found.

Sîn-City: New Light from Old Excavations at Ur

41

spread of small numbers of objects from Ur around the world. There were exceptions to the division. Given the lack of conservation expertise in Baghdad at that time, metalwork was conserved by the British Museum, which was granted more than a half share in return. The metalwork preserves organic remains as impressions or pseudomorphs, where the organic remains are replaced by minerals. In these ways valuable evidence of reed matting, textiles, and woods is preserved. These are being studied by Caroline Cartwright at the British Museum, in collaboration with Nicole Reifarth of the University of Tübingen. Preliminary work has found sheep’s wool (efforts being made to recover the weave structure of the textiles), and identified the wood used for ax shafts as ash. Cuneiform expertise was also lacking in Baghdad at that time, but the tablets did not require the same level of conservation, so while all the tablets were exported for study, the ultimate division remained unaffected. Once studied, the Iraq Museum tablets would be returned (see further below). At each museum, an exhibition of the latest finds would be held after the end of the season. In the British Museum’s case, they were staged in the “Assyrian basement,” where they attracted significant audiences. Short guidebooks were printed with updates on the progress of the excavations. Sets of postcards were soon produced, catering to the still strong demand for related souvenirs. Images from Ur circulated more widely through such objects as cigarette cards, like the 1937 Churchman “Treasure Trove” set, for example. 35 Archaeology had been taking shape as a more scientific discipline, and Woolley was keen to stress the superiority of his methods over those of previous excavators at Ur. He employed a team of professionals: archaeologists (it was here that Max Mallowan would be trained), architects (F. G. Newton and the little known Algernon Whitburn, who also clearly helped keep team morale high), illustrators (including Katharine Keeling 36 and Mary Louise Baker 37) and assyriologists (initially Sidney Smith). While the more travelogue-style elements of earlier reports no longer feature in Woolley’s reports, 38 he did publish a lot of information about his techniques and the rationale for them in his Digging up the Past (1930), itself derived from a series of radio programs broadcast by BBC. 39 Woolley maintained a large workforce (now drawn again from the Muntafiq tribe), justifying this by a perceived need for large gangs in order to obtain sufficient results and on the costs of establishing a dig abroad, especially given the shortness of the work season (November–February), and the difficulty of raising funds for excavation (Woolley 1930: 41–44). But he notes that this workforce would be only as large as was suitable for the task: 300 men 35.  No. 35 features a “Ram in the Thicket” over a colorized image of the excavations; no. 36 features Pu-abi’s headdress. 36.  Although best known as Woolley’s wife (apparently having married due to pressure from University Museum director, Byron Gordon, who objected to unmarried men and women living in the same house), Katharine was an accomplished illustrator. After initially serving as a volunteer, her abilities to draw objects accurately saw her formally become part of the team. 37.  Baker, staff artist at the University Museum, was responsible for producing the many wonderful watercolours illustrating the finds from Ur, and reconstructions of life in ancient Ur, such as the famous Death Pit scene. 38.  The final publications appeared in the Ur Excavations and Ur Excavations: Texts series, which are also available in digital form through the project. Thanks are due to the British Institute for the Study of Iraq for making available the digital version of Gurney 1983. 39.  Such details can be found in the reports to the Directors. Further accounts are available from Mallowan 1977: 28–67 and Christie 1977.

42

Jonathan Taylor

under 5 supervisors when a large volume of earth needed to be dug and few finds expected; maybe 180 under the same 5 supervisors when less earth was to be moved and more objects expected. 40 The workers’ salary was supplemented by baksheesh, intended both to encourage careful digging and to prevent the theft of objects from site (Woolley 1930: 44–47; pl. 4). It was also paid for objects found outside the excavations themselves. At Ur, this meant the discovery of many terracottas in the Diqdiqqeh area, which lay between the workers’ homes and the site. They were found by the children who brought the workers food from home during the day. The workers also received healthcare. Excavation photos show the workers being vaccinated against cholera, which had long been a serious public health problem in southern Iraq (and still can be today). Woolley controlled his workforce through three Arab overseers who had worked with him during his time at Carchemish. They would be called whenever a pick man found anything significant such as the face of a wall or an object. Foremost among them was Hammoudi ibn Sheikh Ibrahim, with whom Woolley evidently had a strong relationship. Hammoudi brought his sons with him; Yahia became the site photographer. Security was an ongoing concern, and not without good cause. At the beginning of the mission, the team’s traditional tent camp was raided; in response a brick dig house was built. Apart from this early episode, the Expedition was kept safe thanks to an agreement with the local sheikh, who would also ensure the site was protected during the team’s absence over the summer, to prevent looting (apparently with success). Woolley produced detailed records of his work, including 19,988 pages of notes handwritten on millimeter-paper file-cards, subsequently bound into 76 volumes. 41 They take the form of field notes and field catalogues, describing loci and objects, which give numbers, descriptions, measurements, find spots, and sketches. These records are curated at the British Museum, where they were all digitized for this project. All 15,458 pages of catalogue and 4,530 pages of notes are now available. Yahia Hammoudi took 2,340 photos of the excavations. These very high quality images reveal a wealth of detail about life on the dig, both intentionally and incidentally. Some key finds can be seen at the point of excavation. Other photos show the exceptional preservation of the mudbrick walls, giving an impression of the ancient city far more vivid than can be gained by visiting the site today. Many photos are staged in some way, as was typical for the time. People often serve as scale; they can be arranged for effect, and there are examples of multiple attempts to shoot the same scene, visible through costume change. The glass negatives are curated at the British Museum, where all were digitized for this project, along with their original captions and data relating to the objects in the images. Woolley would send regular reports to the Directors in London and Philadelphia. He also published preliminary reports in the Antiquaries Journal, Museum Journal and Museum Bulletin, as well as newspapers in the UK (especially the Illustrated London News and Times) and the USA. News of his discoveries spread around the world, rivaling the dramatic finds at Tutankhamun’s tomb in Egypt. 40.  Woolley’s workforce was again arranged along lines familiar from Rassam’s time. Each gang was led by a pick man, who needed to have experience and not be “so old as to have grown stupid nor too young to exercise authority.” He was assisted by a spade man (who should spot any finds not noticed by the pick man), with three or four basket men clearing the soil. 41.  Typescript registers were compiled from these notes shortly after the excavations, with copies presented to Baghdad, London, and Philadelphia; annotations are known from the latter two copies.

Sîn-City: New Light from Old Excavations at Ur

43

Text Corpus The corpus of cuneiform texts from Ur currently stands at around 13,000 inscriptions. 42 Around 3,500 derive from the earlier campaigns of Taylor, Campbell Thompson, and Hall, but most were excavated at Ur by Woolley’s team. The majority of course are clay tablets, but there are also inscribed bricks, foundation deposits (clay cones, cylinders, and stone and metal tablets), door sockets, cylinder seals, and clay vessels, plus a variety of miscellaneous other object types. They span almost the entire length of cuneiform history, from around 2900 BCE to 317 BCE. The Ur corpus includes examples from a broad range of genres. Inevitably, the majority are administrative texts. The earliest group is that of the archaic texts found in the Seal Impressions Strata, recording land cultivation, cereals, and cereal-­ based products, as well as livestock. The Ur III period is well documented, with texts relating to agriculture, animal husbandry, and craft production, as well as letters and legal texts. Old Babylonian texts record the functioning of the Ganunmah (“Great Storage House”), where precious goods, metals, and foodstuffs for offerings and rations were kept. There are also private letters such as those detailing the activities of the famous “rogue trader” Ea-Nasir. Among the more unusual and interesting groups is that of the Kassite legal texts (for which see Gurney 1983) recording not just regular transactions but also criminal proceedings. The first millennium documents represent private business archives. The royal inscriptions range from simple labels such as the name of Meskalamdug on a ceremonial dagger to longer building inscriptions. Copies of such inscriptions are among the many school texts found. There are also several school texts among the archaic group, many hundreds of important lexical and literary texts and fragments from the Old Babylonian period (published in the UET 6 volumes), and also excerpt tablets from the Neo-Babylonian period. Among the hundreds of Old Babylonian school tablets from Ur are six that can securely be identified as having been written by a scribe named Damiq-ilišu. Apart from having his name at the end, these tablets are characterized by several textual (Delnero 2012: Chapter 5) and non-textual features (Ludwig 2009: 9–10). They thus offer a valuable opportunity to study the handwriting and practices of an individual cuneiform scribe. 43 Additionally, they may allow us to identify further tablets written, but not signed, by the same scribe. Analysis of the clustering of features might even suggest the order in which the tablets had been written, yielding evidence about the “curriculum” followed by this scribe. One of the most significant fea­ tures of his handwriting is the confirmation it provides that a scribe would master more than one version of the script. This manifests not just between tablets, but also within tablets. It is not unusual for more than one form of a sign to be found with­in only a few lines. Generally speaking, there is a tendency for “archaizing” and “cursive” forms of signs to cohere in sets. So, for example, UET 6 33, 34, and 163 contain the cursive form of AN and the cursive form of IM, while UET 6 38 contains the archaizing forms of AN and IM. Even this is not consistent, however. UET 6 131 contains the archaizing form of AN and the cursive form of IM. Likewise, while UET 42.  A digital corpus of texts from Ur is available at the Oracc site Ur Text Corpus (http://oracc.org/ urtc/). An estimated 9,000 further fragments were exported to Philadelphia. 43.  This research is being conducted by Marie-Christine Ludwig and the author, and will be published in detail elsewhere.

44

Jonathan Taylor

6 38 duly contains the archaizing form of HA and UET 6 163 contains the cursive form (matching their forms of AN and IM), UET 6 131 again contains the cursive form of HA. Tablets can contain a mixture of archaizing and cursive forms of the same sign. Sometimes these variations can be found among iterations of a sign that are adjacent, e.g. UET 6 131: 55–57, which in three exactly aligned lines writes AN in the archaizing then the cursive then the archaizing form. The variation is at least sometimes conscious and deliberate, and should be seen as stylistic variation; the scribe displays his erudition. It is plausible to hypothesize other factors that may influence the variation. One might wonder if a scribe might tend to begin with archaiz­ ing forms, before shifting to cursive forms as he “tired.” Yet UET 6 163 suggests the opposite. The cursive forms of HA and GAR are found on the obverse (twice each), while the archaizing forms are found on the reverse (twice and once, respectively). Although only relatively few instances are under study, they do seem to cluster meaningfully. Yet this pattern is not fully consistent across tablets. In UET 6 131, written predominantly in archaizing script, the cursive forms of AN, BA, and IM, for example, are found strewn throughout the text. Other patterns emerge, such as the habit of writing the colophon in cursive script, unlike the practice known later in the second millennium, where a scribe may display his learning by switching to archaiz­ ing script for the colophon. Additional depth to the scribal hand is evidenced in the use of multiple variations in sign form. That is, beyond general characterization as archaizing or contemporary, instances of a sign may further vary from each other in the construction of elements or number of wedges. Sometimes these are explicable as simplifications of the ancestral form. Aligning Damiq-ilišu’s handwriting against that of another scribe from the Old Babylonian school texts, Ningishzi[da-…], the differences are readily apparent. The Damiq-ilišu group also offers an opportunity to test wedge order variation within a single scribe’s hand. The results show a high degree of consistency. Where variation does occur, it correlates with variation in sign forms, and in each case the sequences are predictable based on the principles detected in Taylor (2015). At the time of the excavations, sufficient expertise in cuneiform was not avail­ able in Baghdad. For this reason, all cuneiform tablets were sent to London (some being sent in turn to Philadelphia) for study. Most of the tablets sent to London have since been returned, but several thousand tablets and fragments still remain. These include the tablets published in the three UET 6 volumes, which await the Iraq Museum’s judgment that the situation is appropriate and convenient for a return to take place. Beyond them, there are three large groups of tablets. The first is a large collection of mostly fragmentary Ur III administrative texts. Often they are only small fragments; numerous joins can be made among them. The project has facilitated research by a team originally based at the University of Birmingham that seeks to automate the joining process. 3D scans of the fragments are produced by combining multiple digital images; these can be produced with either a digital camera or a phone camera. Algorithms then compare the profiles of the fragments to find possible matches. The initial results of this work are promising. 44 The second group is a similar collection of Old Babylonian documents. 45 The third is a better

44.  See Ch’ng et al. 2014; a further publication is forthcoming. 45.  These will be published by the Archibab project.

Sîn-City: New Light from Old Excavations at Ur

45

preserved collection of Neo-Babylonian documents. 46 All such study tablets in the British Museum have been photographed for the project. Although they have been left till last by generations of scholars, these tablets contain a wealth of interesting new information. One, for example, is an uninscribed school tablet containing re­ peat­ed, deep stylus impressions, apparently the product of a young student stabbing the clay out of boredom or frustration. These impressions provide valuable evidence for the shape of the stylus with which they were made (on which see most recently Bramanti 2015). It is demonstrable that there are relationships between the shape of a text vehicle and factors such as date and genre. This is clearly the case among the inscribed cones found at Ur. The smaller-headed cones typical of the Ur III period are of course clearly distinguishable from the larger-headed equivalents from the Isin-Larsa period, where the inscription from the shaft is repeated on the head. Even within these groups, however, differences are visible. The cones produced for Warad-Sin and Rim-Sin are very similar, which may itself be a deliberate decision to project continuity, but those of Rim-Sin typically have a more rounded/ less beveled head than those of his brother, for example. Woolley (1939: 24, 43) provides rare testimony as to how they were placed in the walls, illustrated with the excellent photos produced by Yahia Hammoudi (Woolley 1939: pls. 15a, 29a, 31b). Thirty-six examples of Ur-Namma’s cones were found in a grid pattern, spaced seventy cm both horizontally and vertically, with the heads proud of the wall. Eight Warad-Sin cones were found. Two were embedded in the wall, at the same height, with their heads proud of the plaster. A further six were found on the top of the brickwork, some loose but some still in situ, at the same height as the first two examples. Frustratingly, what we are not told is how they were placed in position. That is, what the orientation of the cones was. For the Ur-Namma examples, it is now impossible to deduce this from the existing evidence, since the heads are uninscribed and otherwise featureless. For the Warad-Sin examples, however, the photos offer an opportunity. High resolution scans of the glass negatives reveal that these cones were oriented with care. The text on each is arranged in what must be either the older (horizontal) or younger (vertical) reading orientation; it is difficult to examine with confidence at a greater level of detail from the photos, but the orientation seems rather to be in the older fashion. The objects themselves also offer evidence. It is noticeable that in each case the text on the shaft is written in one of only two positions: when held in the hands below eye level, the beginning of the text is visible centrally in the field of view when the text on the shaft is oriented either in the older or in the younger fashion. One might have expected consistency in this regard, but the variation could be understood as evidence for display in vertical orientation, and reading in horizontal orientation. The archaizing cones produced for Sin-balassu-iqbi—reviving a practice dead for a thousand years already—mimic Ur III predecessors. They were presumably the oldest examples which his builders were able to find. They are nevertheless clearly distinguishable from them. The shaft is shorter and wider. The head is flat, with a decorative bevel finish. This same finish is applied to the pointed end of the shaft, making them quite unlike any other cone type. Hammoudi’s photos show the different arrangement of these cones. They were deposited under the floor, standing on end. This pattern reflects not 46.  These are due for publication by Paul-Alain Beaulieu.

46

Jonathan Taylor

third or second millennium practice, but contemporary first millennium practice, as observed for cylinders by Taylor in the 1850s (see above).

Conclusion The Ur excavations provoked scientific marvel and press sensation. The fruits of those excavations are among the essentials for every student of Mesopotamia to know about, and are among the best known objects outside of the expert community. The site of Ur remains one of the most important and most visited cultural heritage sites in Iraq, and is a celebrated element of Iraqi identity. Although almost 100 years have passed since the Joint Expedition, new knowledge continues to flow from the excavations. The renewed collaboration between the Expedition partners has inspired fresh research, and with the digitization of the collections and the archives, has provided researchers with unparalleled access to the material, opening the way for scholars from around the world to undertake further new programs of research. Colleagues in Baghdad have expressed interest in joining the collaboration, which would provide significant further stimulus, and lead to a fuller and more contextualized picture of the excavations and their results.

Bibliography Ainsworth, W. F. 1838 Researches in Assyria, Babylonia, and Chaldæa. London: Parker. Bell, G. 1927 The Letters of Gertrude Bell. London: Ernest Benn [selected and edited by Lady Bell, D.B.E.]. Bernhardsson, M. T. 2005 Reclaiming a Plundered Past: Archaeology and Nation Building in Modern Iraq. Austin, TX: University of Texas Press. Bramanti, A. 2015 The Cuneiform Stylus. Some Addenda. CDLN 2015:12. Budge, E. A. W. 1920 By Nile and Tigris: A Narrative of Journeys in Egypt and Mesopotamia on Behalf of the British Museum Between the Years 1886 and 1913, etc. London: Murray. 1925 The Rise and Progress of Assyriology. London: Hopkinson. Chesney, F. R. 1850 The Expedition for the Survey of the Rivers Euphrates and Tigris. London: Longman, Brown, Green and Longmans. Ch’ng, E., S. I. Woolley, L. Hernandez-Munoz, T. Collins, A. Lewis, E. Gehlken 2014 The development of a collaborative virtual environment for 3D reconstruction of cuneiform tablets. Pp. 35–42 in Virtual Systems & Multimedia (VSMM). Christie, A. 1977 Agatha Christie: An Autobiography. London: Collins. Foster, B. R. 2013 Albert T. Clay and His Babylonian Collection. Pp. 121–135 in Beyond Hatti: A Tribute to Gary Beckman, eds. B. J. Collins and P. Michalowski. Bristol, CT: Lockwood. Fraser, J. B. 1840 Travels in Koordistan and Mesopotamia. London: Bentley.

Sîn-City: New Light from Old Excavations at Ur 1842 Mesopotamia and Assyria from the Earliest Ages to the Present Time. Edinburgh: Oliver & Boyd. Goode, J. F. 2007 Negotiating for the Past: Archaeology, Nationalism, and Diplomacy in the Middle East, 1919–1941. Austin, Texas: University of Texas Press. Guha, S. 2015 Artefacts of History. Archaeology, Historiography and Indian Pasts. New Delhi: Sage. Gurney, O. 1983 The Middle Babylonian Legal and Economic Texts from Ur. London: British School of Archaeology in Iraq. Hall, H. R. 1930 A Season’s Work at Ur, Al-ʿUbaid, Abu Shahrain - Eridu - and elsewhere. Being an unofficial account of the British Museum Archaeological Mission to Babylonia, 1919. London: Methuen. Hilprecht, H. V. 1903 Explorations in Bible Lands during the 19th Century. Philadelphia: A. J. Holman and Company. Loftus, W. K. 1857 Travels and Researches in Chaldea and Susiana. London: Nisbet. Ludwig, M.-C. 2009 Literarische Texte aus Ur: Kollationen und Kommentare zu UET 6/1–2. UAVA 9. Berlin: de Gruyter. Mallowan, M. E. L. 1977 Mallowan’s Memoirs. London: Collins. Oppert, J. 1863 Expédition scientifique en Mésopotamie exécutée par ordre du gouvernement de 1851 à 1854, par Fulgence Fresnel, Felix Thomas et Jules Oppert. Paris: Imprimerie Impériale. Peters, J. P. 1898 Nippur: or, Exploration and Adventures on the Euphrates: The Narrative of the University of Pennsylvania Expedition to Babylonia, 1888–1890. New York: Putman. Rassam, H. 1897 Asshur and the Land of Nimrod: Being an Account of the Discoveries Made in the Ancient Ruins of Nineveh, Asshur ... and Van. New York: Eaton & Mains. Rawlinson, H. C. 1850 On the Inscriptions of Assyria and Babylonia. JRAS 12: 401–483. Sollberger, E. 1972 Mr. Taylor in Chaldaea. AnSt 22: 129–139. Taylor, J. G. 1855a Notes on the Ruins of Muqeyer. JRAS 15: 260–76. 1855b Notes on Abu Shahrein and Tell el Lahm. JRAS 15: 404–415. Taylor, J. J. 2015 Wedge Order in Cuneiform: A Preliminary Survey. Pp. 5–34 in Current Research in Cuneiform Palaeography. Proceedings of a Workshop held at the 60th Rencontre Assyriologique Internationale, Warsaw 2014, eds. E. Devecchi, G. G. W. Müller, and J. Mynářová. Gladbeck: PeWe. Woolley, C. L. 1929 Ur of the Chaldees: A Record of Seven Years of Excavation. London: Penguin. 1930 Digging Up the Past. London: Penguin.

47

48

Jonathan Taylor

Zettler, R. 1998 Ur of the Chaldees. Pp. 9–19 in Treasures from the Royal Tombs of Ur, eds. R. L. Zettler and L. Horne. Philadelphia: University of Pennsylvania Museum of Archaeology and Anthropology. Zettler, R. and Hafford, W. B. 2015 Ur. B. Archäologisch, Reallexikon der Assyriologie und Vorderasiatischen Archäologie 14/5–6: 367–385.

Bad Moon Rising: The Changing Fortunes of Early Second-Millennium BCE Ur Katrien De Graef Ghent University

I see a bad moon rising, I see trouble on the way I see earthquakes and lightning, I see bad times today Don’t go around tonight, it’s bound to take your life There’s a bad moon on the rise (J. Fogerty)

Introduction Over the last thirty years, excellent, profound, and comprehensive studies by Charpin (1986), Diakonoff (1985, 1990, and 1995), and Van De Mieroop (1992a and 1992b) have been devoted to the history and society of early second millennium BCE Ur, and it is hard, not to say impossible, to add something to their exhaustive and well-considered analyses, hypotheses, and conclusions. Whereas Charpin’s book is an in-depth study on the clergy of Ur, Diakonoff’s and Van De Mieroop’s works deal with the society of Ur in a more general sense. While going through these studies on early second millennium BCE Ur, which primarily focus on the city’s socio-economic situation, abundantly and clearly exemplified by in-depth analyses of private and other archives, where possible linked to their archaeological context, a strange paradox struck me. For it seems that the city of Ur was relatively prosperous, both economically and culturally, as is shown by the economic texts, the production of a literary corpus, and inscriptions referring to building activities, whereas this period is generally characterised by instability and chaos caused by wars between competing dynasties, involving invasions and raids, and resulting repeatedly in devastations, famine, and exile. How is it that economy, religion, and culture could flourish and an apparent social stability could be attained in such politically instable times? How is it that life apparently went on in such a turbulent period? Therefore, the question arises of how we can approach this contradiction, how we can understand and interpret the seemingly evident but no doubt continuously evolving disequilibrium, and how

49

50

Katrien De Graef

such an approach could contribute to our understanding of early second millennium BCE Ur.

Unfortunate Times Indeed, the early second millennium BCE did not work out very favourably for the city of Ur. It both started and ended with a violent destruction of the city. In between, the city was subject to the power struggle between the competing dynasties of Isin and Larsa, under the control of which it went to and fro, to end up being incorporated in the Babylonian empire of Ḫammurabi, which soon started to disintegrate under his successor Samsu-iluna. 1 At the dawn of the second millennium, the Ur III dynasty collapsed due to a combination of internal crises, causing massive inflation, and the invasion of the Elamites and Šimaškians from the East, who were able to attack the Ur III state at its heart and destroy its capital. 2 The Amorites, who entered the Ur III empire at the end of the third millennium BCE—in spite of the wall the Ur III kings built to stop them—seem not to have confronted Ibbi-Sîn’s troops directly, but contributed to the disorganization of the empire by cutting the lines of communication between the cities. 3 After this major blow to the city, Ur came under the control of the dynasty of Isin. This dynasty was founded about fifteen years earlier by a previous Ur III official, Išbi-Erra (2019–1987 BCE), who took advantage of the rapidly declining state of the empire to create his own power base. Eight years after the fall of Ur, Išbi-Erra retook the city, expelled the Elamites, 4 and started a policy of reconstruction. Išbi-Erra as well as his successors directed

1.  Recent general overviews of the history of Babylonia in the early second millennium BCE, the socalled Isin-Larsa (or early Old Babylonian) and Old-Babylonian periods, can be found in, among others, Liverani 2014 and Van De Mieroop 2016. An in-depth analysis of the Old Babylonian period, covering political history, literature, religion, economy, and society, is to be found in Charpin, Edzard, and Stol 2004. An overview of the early seceond millennium BCE political history of the city of Ur in particular can be found in Van De Mieroop 1992a: 45–71. It is by no means my intention here to add to the history of early second millennium BCE Ur as we know it. This short overview, which draws largely on Charpin 2004 and Van De Mieroop 1992a, serves merely to demonstrate the general political instability and chaos that characterizes this period as well as the significance the city of Ur, which had been downgraded from the capital of a large bureaucratic empire to one city-state among many, still had for the consecutive dynasties it was ruled by. For convenience’s sake, all dates referred to in this article follow the Middle Chronology. However, even then the absolute chronology of the early second millennium BCE is still problematic and most studies differ slightly in dating the reigns of the Isin and Larsa rulers. Since the exact absolute dating of the rulers and events is of less importance for my purpose in this article, this is not a large problem, and I will not go into this. I will use the absolute dates given in Van De Mieroop 2016: 249–350, as this is the most recent list published. 2.  See, among others, Wilcke 1970; Gomi 1984; Lafont 1995; and Sallaberger 1999: 174–178. 3.  See Charpin 2004: 57–60. 4.  This is commemorated in the year names of the twenty-sixth and twenty-seventh years of reign of Išbi-Erra: mu (ús.sa) elam šà úriki.ma.ka durun.a gištukul kal.ga.ni im.ta.e11 Year (after the year) he (i.e. Išbi-Erra) brought out of Ur, with his strong weapon, the Elamite who was dwelling in its midst (see Sigrist 1988: 19–20), as well as in Išbi-Erra “Hymn B,” see van Dijk 1978; Frayne 1982; and Vanstiphout 1989–90.

Bad Moon Rising

51

their attention to the recovery of the ancient capital, 5 and especially to its trade with the Gulf. His successor, Šū-ilīšu (1986–1977 BCE), claimed to have returned the statue of Nanna, which the Elamites took as booty to their homeland, 6 and to have resettled the scattered people of Ur, 7 implying that there was an exile of at least a part of Ur’s population. The kings of Isin were able to recover only the heart of the ancient Ur III empire and large parts of periphery became independent. 8 This newly established dynasty no doubt went through turbulent times, trying to hold control over its core territory, maintaining relations with the periphery and especially the East where a new, vast, and powerful Šimaškian empire had firmly established itself, 9 and facing a new social reality in which the autochthonous people of Sumer and the newcomers, in casu the Amorites, had to find a way to cohabitate. The city of Ur, although it no longer played any political role, remained important for the Isin kings, who, although Amorites, saw themselves as the legitimate successors of the Ur III kings and most of them took the title “king of Ur” or other epithets/titles mentioning the city of Ur. 10 Ur was not only important economically, due to its trade with the Gulf, but also in the realms of religion and culture, as is shown for example by the fact that Išme-Dagan (1955–1937 BCE) installed as

5.  Various Isin kings claim to have restored and/or built structures at Ur, such as Šū-ilīšu, who restored the Edublamaḫ and built a standard for Nanna (see RIME 4.1.2.1–2) and Iddin-Dagan, who made a throne dais for the Edublamaḫ of Nanna (year names l and m: mu (ús.sa) di-din-dda-gan lugal.e giš gu.za bára dub.lá.maḫ dšeš.ki mu.na.dím, see Sigrist 1988: 25). 6. See RIME 4.1.2.1, esp. ll. 8–11: u4 dnanna an-ša-anki-ta uri5ki-šè mu-un-túm-ma-a “when he brought (back the statue of) the god Nanna from Anšan to Ur.” 7.  See RIME 4.1.2.2, esp. ll. ii 1–7: [u4 …] ⸢mu⸣ […] ⸢uri5⸣[ki …] x […] zà.⸢an⸣.[ša.anki.na.šè] ság. ⸢du11⸣.[ga] ki.tuš.ba gi.⸢na⸣.[a] mu.na.dím “he fashioned for him [when] he establish[ed in] U[r the people] scattered as far as A[nšan], in their abode.” 8.  Išbi-Erra’s territory extended from Marad, Apiak, Borsippa, and Kazallu in the north, to Uruk, Ur, Eridu, and Larsa in the south, but it seems that Iddin-Dagan lost control over Nippur and Uruk (Charpin 2004: 62). 9.  One way to maintain relations with the East was through diplomatic marriages. According to Iddin-Dagan’s year names a and b, he gave his daughter Matumniatum to the king of Anšan in marriage: mu (ús.sa) di-din-dda-gan ma-tum-ni-a-tum dumu.munus.a.ni lugal an.ša.anki ba.an.tuk.a (see Sigrist 1988: 24). This, however, might also be an imitation of the famous Ur III king Šulgi. 10. Šū-ilīšu was the first Isin king to call himself “king of Ur,” see RIME 4.1.2.1, 2 and 4. His successor Iddin-Dagan called himself “king of Ur” (see RIME 4.1.3.2) as well as “king of Isin” (see RIME 4.1.3.1) alongside the more general “king of the land of Sumer and Akkad.” His successor Išme-Dagan called himself “constant (attendant) of Ur” (sag.ús.uri5ki.ma, see RIME 4.1.4.1–2, 11–12 and 15) alongside various other epithets and titles mentioning Nippur, Eridu, and Uruk, “king of Isin” and “king of the land of Sumer and Akkad.” Išme-Dagan is called “king of Ur” only once in the inscription on a stone bowl dedicated by one of his servants (see RIME 4.1.4.2001). His successor Lipit-Ištar called himself “true farmer of Ur” (engar.zi.uri5ki.ma, see RIME 4.1.5.1–6), alongside various other epithets and titles mentioning Nippur, Eridu, and Uruk, “king of Isin” and “king of the land of Sumer and Akkad.” He is called “king of Ur” in the inscription on two seals of his servants (see RIME 4.1.5.2001 and 2003). Even after the Isin kings lost the city of Ur to the Larsa dynasty, some of them continued using epithets/ titles mentioning Ur: Ur-Ninurta called himself “herdsman of Ur” (na.gada.uri5ki.ma, see RIME 4.1.6.1), Bur-Sîn, who retook Ur but lost it again very quickly, called himself “mighty farmer of Ur” (engar.kala. ga.úriki.ma, see RIME 4.1.7.1), and Enlil-bani called himself “farmer (who grows) tall grain for Ur” (engar še.maḫ.uri5ki.ma, see RIME 4.1.10.1, 4–5 and 9).

52

Katrien De Graef

entum-priestess of Nanna at Ur his daughter En-ana-tuma, who is said to have contributed greatly to the embellishment of the city. 11 However, the Isin dynasty was granted only a short life, and the city of Ur soon fell into the hands of the competing Larsa dynasty. The main cause for the confrontation between Isin and Larsa was control over water rights, as is shown by the letters between Lipit-Ištar of Isin (1936–1926 BCE) and his general Nanna-kiaga. 12 According to Sigrist (1987: 28), Lipit-Ištar was obliged to re-dig the Ninki canal and restore the city of Ur after a first campaign by Gungunum of Larsa. 13 In his seventh year of reign, 14 Gungunum (1932–1906 BCE) managed to conquer Ur, after which he called himself “king of Ur.” 15 It seems that this new leadership did not cause too much disruption in the city as officials installed by the Isin kings were maintained in their posts, just as Išme-Dagan’s daughter stayed on as entum priestess of Nanna and continued her policy of reconstruction (see n. 7). Gungunum even installed Lipit-Ištar’s daughter Enninsunzi as entum-priestess of Ningublaga at Ur. 16 Like his predecessors from the Isin dynasty, Gungunum pursued the embellishment of Ur through public works. 17 However, the loss of Ur was of great importance for the Isin kings, not only symbolically, but especially economically, as it lost access to the Gulf and the thri­ ving maritime trade. Moreover, the conflict between Isin and Larsa continued, as 11. See RIME 4.1.4.3–4 and 13. The appointment of his daughter as entum-priestess of Nanna at Ur is commemorated in no less than four consecutive year names of Išme-Dagan, viz. a, b, c, and d: mu (ús.sa / mu.4.kam) diš-me-dda-gan lugal.e en.dšeš.ki úriki.ma ba.ḫun.gá (mu ús.sa.bi) “Year (after the year / third year / fourth year) in which Išme-Dagan installed the en-priestess of Nanna in Ur” (see Sigrist 1988: 26). After Ur was conquered by Gungunum of Larsa, An-ana-tuma remained in office and continued her policy of (re)construction, see RIME 4.2.5.1–2. 12. See ETCSL c./t.3.2.03–4 for transliterations, translations, and references. See also Rowton 1967: 273. 13.  This is commemorated in year names b and d of Lipit-Ištar: mu dli-pí-it-iš8-tár lugal.e inim den. líl dšeš.ki.ta úriki ki.bé bín.in.gi4.a “Year in which on the order of Enlil and Nanna (the city of) Ur was restored,” and mu i7 dnin.ki ba.ba.al “Year the canal of Ninki was dug” (see Sigrist 1988: 28), as well as in RIME 4.1.5.5, ll. 17–21: du11.du11.ga.den.líl.dnanna.ta uri5ki ki.bé ḫé.bi.gi4 ḫi-ri-tum-bi ḫu.mu.ba.al “by the decree of the gods Enlil and Nanna, I restored Ur, I dug its moat.” 14.  According to Charpin (2004: 71 especially n. 223). Note, however, Frayne (1990: 114) according to whom Gungunum took Ur in his tenth year of reign. 15.  See RIME 4.2.5.1–2. His successors Abi-sare and Sumu-el also called themselves “king of Ur” in their inscriptions (see RIME 4.1.6.1 and 2001 and 4.2.7.1–2 and 2001–3). Sumu-el’s successor Nūr-Adad called himself “provider of Ur” (ú.a.úriki.ma, see RIME 4.2.8.1, 3–4 and 6), “farmer of Ur” (engar.uri5ki. ma, see RIME 4.2.8.2) and “true farmer of Ur” (engar.zi.uri5ki.ma, see RIME 4.2.8.5) always alongside the title of “king of Larsa.” The same goes for his successors Sîn-iddinam and Sîn-iqīšam, who stick with “provider of Ur,” alongside “king of Larsa” and sometimes “king of the land of Sumer and Akkad” (see RIME 4.2.9.1–7 and 9–15, and 4.2.11.1–3). Ṣilli-Adad called himself “provider of Nippur” and “governor of Ur, Larsa, Lagaš, and the land of Kutalla” (ú.a.nibruki énsi.úriki larsaki lagaški ù ma.da-ku-ta-al-laki-a, see RIME 4.2.12.1–2). 16.  See the year name of Gungunum’s thirteenth year of reign: mu en.dnin.sún.zi en dnin.gublaga ba.ḫun.gá “Year in which Enninsunzi was installed as en-priestess of Nin-gublaga,” and year name g of Lipit-Ištar: mu dli-pí-it-iš8-tár lugal.e en.nin.sún.zi en dnin.gublaga úriki.ma máš.e ì.pàd “Year king Lipit-Ištar chose by means of the omens Enninsunzi for en-priestess of Nin-gublaga in Ur” (see Sigrist 1990a: 8 and 1988: 28). 17.  As is shown in the year names of Gungunum’s twentieth and twenty-fifth years of reign: mu ká.gal maḫ úriki.ma ba.dù “Year the magnificent gate of Ur was built,” and mu é.šútum (é.gi.na.ab.tum) kù dšeš.ki šà úriki.ma ba.dù “Year the holy Ešutum (Eginabtum) of Nanna in Ur was built” (see Sigrist 1990a: 9–10).

Bad Moon Rising

53

Gungunum was able to take hold of some key cities of Isin’s territory, such as Uruk and Nippur, but Ur-Ninurta (1925–1898 BCE) managed to retake some of the places lost by his predecessor Lipit-Ištar. The struggle for power between Isin and Larsa persisted during the reigns of Gungunum’s successors Abi-sare (1905–1895 BCE) and Sumu-el (1894–1866 BCE). Bur-Sîn of Isin (1897–1876 BCE) succeeded in reconquering Ur temporarily, 18 but the city soon came back in the hands of Sumu-el, who installed his daughter there as entum-priestess of Nanna. 19 The following period is characterised by ups and downs. The struggle for water seems to have turned out advantageously for the kings of Larsa who, after twenty years of major works, succeeded in diverting a branch of the Euphrates that pre­ viously watered the region of Isin in their favour, as a result of which new areas could be opened up and Larsa’s economy grew significantly. 20 However, things changed rapidly for the worse when an unidentified enemy attacked the country, took various cities, and cut off the water supply for Larsa, causing a famine in the city as a result of which the population broke out in revolt. It is as yet not certain when exactly this catastrophe took place. According to the royal inscriptions of NūrAdad (1865–1850 BCE) and his son and successor Sîn-iddinam (1849–1843 BCE), Utu selected Nūr-Adad as leader of the people, who cleared the Euphrates from the barrages built by the enemy, restored the water supply, reconquered the countryside, and returned the dispersed population of Larsa, 21 implying the disaster took place before Nūr-Adad came to power, viz. at the end of Sumu-el’s reign. However, as Sîn-iddinam also claimed to have put down a rebellion, 22 and the restoration works 18.  Probably during Sumu-el’s seventeenth to twenty-first years of reign as the year names of these years are not attested in Ur, see Van De Mieroop 1992a: 57 and Charpin 2004: 77. 19. According to Charpin (2004: 77), Sumu-el retook Ur at the latest in his twenty-first year of reign, after which he installed his daughter Enšakiag-Nanna as entum-priestess of Nanna in Ur, which is commemorated in his twenty-second (and following) year name(s). However, as the lists with his year names contradict each other, the exact order is not clear (see Charpin 2004: 76 n. 247 with references). According to Sigrist (1990a: 20–21) the installation of Sume-el’s daughter as entum-priestess of Nanna in Ur is commemorated in his twenty-third to twenty-ninth year names: mu (ús.sa / .a.bi / .4.bi / .5.bi/ kam / .6.bi/kam / .7.kam) en dšeš.ki en.šà.ki.ág.dšeš.ki ba.ḫun.gá “Year (after the year / 3rd / 4th / 5th / 6th / 7th year) in which Enšakiag-Nanna the en-priestess of Nanna was installed.” 20.  See the so-called “archive of Lu-igisa,” consisting of more than 100 administrative texts and letters from Larsa, to be dated between Abi-sare 7 and Sumu-el 17, edited by Walters (1970). See also Stol 1971 and 1981, in which he republished the letters of the archive. In some of these texts, the building of a weir at the junction of the Isin canal and the Euphrates involving the use of 1.3 million bricks in Sumu-el 14 is mentioned (see Dight 2002: 119). See also Rowton 1967 and Frayne 1989. 21.  See the letter-prayer of Sîn-iddinam to Utu, addressed to the statue of his father Nūr-Adad, published by van Dijk (1965), in which Sîn-iddinam describes the conquest of Larsa by a foreign enemy by obstructing the water supply and the consequent revolt in the city, the election of his father Nūr-Adad, and the latter’s reconquest and restoration of the city. It is yet not clear who the foreign enemy was. Van Dijk (1965: 12–13) argues that the enemy must have attacked from the north, as Zabalam, Lagaš, and Ennegi, cities watered by the Tigris, were not affected. This leads him to suggest the possibility that it was Babylon. However, as Sîn-iddinam starts his letter-prayer mentioning the fact that his father seized [GN?] and usurped the power of the Sukkal (ll. 50–53; […].⸢íb⸣.dab5.dab5 nam.sukkal.⸢bé⸣ sá bí.in.[du11]) he suggests that the enemy might also have been Elamite. 22.  See RIME 4.2.9.11 in which Sîn-iddinam claims to have defeated all of his enemies with weapons and provided perpetual water and abundance for his city and land by digging the Tigris and RIME 4.2.9.13 in which he claims to have bowed down and smitten with weapons the land that rebelled against him. Note that both inscriptions also mention public works in Ur, the construction of the Enamnuna and the wall of Ur respectively, which might allude to the fact that the city of Ur had also been affected by the foreign enemy who attacked Larsa, as already suggested by van Dijk (1965: 13). See in this respect

54

Katrien De Graef

started by his father continued under his reign, it is possible that the problems started during Nūr-Adad’s reign and were not yet completely resolved when his son succeeded him. 23 Whatever the case may have been, the Larsa dynasty had clearly been badly hit, and it seems that Nūr-Adad temporarily lost control over the city of Ur, as only six of his sixteen year names are attested on tablets from Ur. Sîn-iddinam continued his father’s restoration works, among which were those of the Ganunmaḫ and Ur’s city walls. 24 At the end of his reign, though, he expressed his worries about the incessant battles and despair of the people in another letter-prayer to Utu, 25 and although we have to take the usual exaggeration into account here, the situation does not seem to have been too favourable at the end of his reign. 26 His successors

also MS 5000, published by Volk (2011), in which Sîn-iddinam claims to have built the walls of Ur on virgin soil, elevated its battlements like a mountain, and extended its residential estates in a single year in order that its citizens would be able to sleep well (see especially i 39–54, ii 65–72, and iv 26–28, 31–34, and 44–48). 23.  According to Van De Mieroop (1992a: 59) the disaster did not take place at the beginning of Nūr-Adad’s reign, as Sîn-iddinam’s letter-prayer published by van Dijk (1965) seems to imply, but at the end of his reign and his son and successor Sîn-iddinam was the one who settled things. 24.  Both Nūr-Adad and Sîn-iddinam were highly involved in the organization of great public works, among which various constructions and a statue in Ur, such as the Ganunmaḫ (see RIME 4.2.8.2 and 4.2.9.10), an oven and cauldron for Nanna (see RIME 4.2.8.3), the Agun.kù for Ningal (see RIME 4.2.8.4), the Enamnuna (see RIME 4.2.9.11), a statue for Nanna (see RIME 4.2.9.12), and the wall of Ur (see RIME 4.2.9.13 and Volk 2011). 25.  See Hallo 1982 and 2010: 335–340 with references for this and other letter-prayers. 26.  One might even ask oneself whether the disasters mentioned in the two letter-prayers of Sîniddinam refer to two different incidents or one and the same. Might it be possible that a foreign enemy attacked Larsa and cut off its water supply by damming the river, causing years of fighting, famine, epidemics, and uprisings among the people? It seems that Nūr-Adad and his son Sîn-iddinam co-ruled for a period: in Sîn-iddinam’s royal inscription commemorating the restoration of the Ganunmaḫ in Ur (RIME 4.2.9.10), he calls himself king of Larsa but dedicates his reconstruction to his and his father’s life, indicating that Nūr-Adad is still alive at the time. He adds moreover that no royal ancestor of his ever restored the Ganunmah before, which is odd, to say the least, as his very father claims to have restored the Ganunmaḫ in Ur (see RIME 4.2.8.2). Might it be that both inscriptions refer to one and the same restoration of the Ganunmaḫ by Nūr-Adad and his son Sîn-iddinam? As such, they would date from the time that both were in power. In that case, the foreign enemy would have attacked Larsa during the period that Nūr-Adad and Sîn-iddinam co-ruled or just before, which would explain why Sîn-iddinam attributes the rescuing of Larsa to his father—who was still in power or at least still alive—in the first letter-prayer, but claims to have defeated the enemy and put down a rebellion himself in his royal inscriptions. In the other letter-prayer (see supra), Sîn-iddinam mentions that the plague was already raging for seven (or five, according to another manuscript of the text) years, which has been associated with the duration of his reign, viz. seven years, and is hence believed to date from the end of his reign. This, however, does not necessarily hold true: if the attack causing the seven-year period of total turmoil happened during the period that Nūr-Adad and Sîn-iddinam co-ruled or just before, it might have ended early on or in the middle of Sîn-iddinam’s solo reign—unless Sîn-iddinam’s seven years of reign were all years of co-rule with his father. In RIME 4.2.9.11, Sîn-iddinam claims to have defeated his enemies after which he provided perpetual water and abundance for his city and land by digging the Tigris. This last achievement can be linked to his second year name (mu i7 idigna ba.ba.al “Year the Tigris was dug,” see Sigrist 1990a: 24), implying that at least a part of the difficulties might have been overcome at the beginning of his solo reign. As for the unidentified foreign enemy, it is remarkable that in both letter-prayers Sîn-iddinam refers to Elam: the first refers to his father usurping the power of the Sukkal and the other puts Larsa’s tragic fate in contrast to the joyful fate of Elam and the godless Subarians and Šimaškians. Especially so since less than ten years after the end of Sîn-iddinam’s reign, an Elamite (or at least “Elamitizing”) dynasty under Kudur-Mabuk took power in Larsa. It goes without saying that this hypothesis needs further and thorough investigation.

Bad Moon Rising

55

only reigned briefly; 27 the third of whom was removed from the throne by a new dynasty after only a couple of months of reign. The founder of this new dynasty, Kudur-Mabuk, who might have been of Elamite origin, 28 did not become king himself, although he seems to have played a very active role in the government, 29 but put his son Warad-Sîn (1834–1823 BCE) on the throne in Larsa. The change of dynasty did not cause any interruption, as is shown by both the literary and documentary texts. Following his predecessors, Kudur-Mabuk started an extensive building programme at Ur that was continued by his sons and successors, 30 and installed his daughter as entum of Nanna at Ur. 31 Warad-Sîn claimed to have enlarged the city of Ur and built a new city wall. 32 After a short reign of only twelve years, his brother, Rīm-Sîn (1822–1763 BCE), succeeded him. During the first half of his sixty-year-long reign, he managed to reconquer 27.  Sîn-eribam (1842–1841 BCE), Sîn-iqīšam (1840–1836 BCE), and Ṣilli-Adad (1835 BCE). 28. The origin of the Kudur-Mabuk dynasty poses problems: his name and patronymic, SimtiŠilḫak, are both Elamite, but the titles he bears, viz. “father of the Amorite land” (ad.da.kur.mar.tu, see RIME 4.2.12.3, 5–7, 9–10, 13, and 30) and “father of Emutbala” (ad.da-e-mu-ut-ba-la, see RIME 4.2.12.15, 17–20, 23–24, 26 and 29; RIME 4.2.13a.2; and RIME 4.2.14.2–3) refer clearly to the Amorites. According to Steinkeller (2004: 30–33), Kudur-Mabuk was of genuine Elamite extraction. Residing in Maškanšāpir, in charge of the Emutbala tribes, he was in contact with Larsa already during Sîn-iddinam’s reign, positioning himself to take Larsa’s throne. 29.  See the letter of Kudur-Mabuk to Ur-Nanna in which he orders to gild the statue of the entumpriestess of Nanna and send goldsmiths to Ur to do so (see Charpin 1986a: 43–44). Moreover, almost all royal inscriptions of Warad-Sîn and those of Rīm-Sîn prior to his eighth year of reign mention their father Kudur-Mabuk. In some of them, he is the agent responsible for restorations and reconstructions of buildings (see for example RIME 4.2.13.3, 5–7, 9–10, 13, and 15). Last but not least, it was his daughter who was installed as entum-priestess in Ur (see infra), a position usually preserved for the daughter of the king. 30.  Kudur-Marduk claims to have restored the Eeškite shrine for Nanna (RIME 4.2.13.5–7) and the Ganunmaḫ (RIME 4.2.13.9–10), and to have built a throne and statue for Nanna (RIME 4.2.13.13). Warad-Sîn claims to have restored the Eilurugukalama for Ningal (RIME 4.2.13.1), the Egabura for Ningubalag (RIME 4.2.13.4 and 26), the Egalmaḫ for Ningal (RIME 4.2.13.2), the Etemeniguru for Nanna (RIME 4.2.13.16), the Ekuga for Nergal (RIME 4.2.13.23), the Ekituššatenbi for Zababa (RIME 4.2.13.24), the Nanna-ḫul canal in the harbor district (RIME 4.2.13.25), and the Etilmun for Inana (RIME 4.2.13.27). 31. See RIME 4.2.13.15. This is also commemorated in Warad-Sîn’s seventh year name: mu en.an.e.du7 en dšeš.ki úriki ba.ḫun.gá “Year in which Enanedu was installed as en-priestess of Nanna in Ur” (according to Sigrist 1990a: 33–34 this is Warad-Sîn’s eighth year name contra Sigrist’s website [cdli.ucla.edu/tools/year names] where it is his seventh year name; see also Charpin 2004: 68 n. 202). The seal of Enanedu is impressed on UET 5 272, a donation of an orchard in the harbour district of Ur by Enanedu, to be dated in the eleventh year of Rīm-Sîn I: [en.an.e.du7] ⸢en⸣ d[nanna] ⸢uri5⸣ki.[ma] dumu ku-du-ur-ma-bu-[uk] ⸢šeš⸣ ir11-dEN.[ZU] lugal larsaki.⸢ma⸣ “Enanedu, entum-priestess of Nanna of Ur, son of Kudur-Mabuk, brother of Warad-Sîn, king of Larsa” (see Charpin 1986a: 60–61 and RIME 4.2.13.32). For the use of “son” (dumu) and “brother” (šeš) for a woman, see Lion 2009: 166–169. As already suggested by Steinkeller (2004: 30–31), it might be possible that Enanedu is the same person as the daughter of Kudur-Mabuk called Manzi-wartaš in a text from Uruk. This would imply that she originally bore an Elamite name, like her father and her grandfather, and adopted the Sumerian name of Enanedu when she became entum-priestess, which was not uncommon among priests (see Charpin 1986a: 396–402, Janssen 1992: 47–48, and Tanret 2010: 201). One might ask oneself whether KudurMabuk’s sons, Warad-Sîn and Rīm-Sîn, originally also bore Elamite names, but adopted Akkadian names when they became king. 32. See RIME 4.2.13.18–24. The construction of this wall, which was called “Nanna makes the foundation of the land firm” (Nanna-suḫuš-mada-gengen), is commemorated in the year name of WaradSîn’s tenth year of reign: mu ᵈìr-ᵈEN.ZU lugal.e bàd gal úriki.ma mu.dù.a “Year king Warad-Sîn built the great city wall of Ur” (see cdli.ucla.edu/tools/year names).

56

Katrien De Graef

Uruk and Isin. As for Ur, the archive of the merchant Ea-nāṣir mentions enemy activity that made trade difficult, 33 and Sîn-muballiṭ of Babylon claimed to have defeated the army of Ur in his fourteenth year name. 34 It only got worse in the second half of Rīm-Sîn’s reign, when several kingdoms vied with each other for dom­ ination: Elam, Ešnunna, Mari, Babylon, and Larsa raided each other’s territories and entered into strategic alliances in order to gain supremacy. Problems with the water supply, whether related to the hostilities or not, caused a shortage of barley, resulting in a famine in the south between Rīm-Sîn’s thirty-ninth and forty-second years of reign. 35 Eventually, Ḫammurabi of Babylon (1792–1750 BCE) was able to conquer the kingdom of Larsa, after which Ur became part of Ḫammurabi’s Babylonian Em­ pire. 36 This, however, did not mend matters for the city of Ur, which had fallen into decline some twenty-five years earlier due to Babylonian incursions and the fact that the Larsa kingdom had been forced to shift its centre of gravity to the north. Ḫammurabi set up a large project to restore the water supply for the devastated southern cities and claimed to have resettled the scattered population, 37 but did not 33.  Two letters, UET 5 20 and 81, refer to hostile activities and enemy territory (see Oppenheim 1954: 9 n. 11; Leemans 1960: 39–40; Van De Mieroop 1992a: 64–65; and Garfinkle 2010: 198). 34.  mu ugnim úriki gištukul ba.sìg “The year: he defeated the army of Ur” (see Horsnell 1999 [vol. II]: 98). 35.  Various letters (AbB 8 12 and 14; TIM 1 22, 26, and 22; Rowton 1967: 269–270; and A.7535–52 in Christian 1969) show that merchants of Larsa had to buy large amounts of barley in the kingdom of Ešnunna. This caused difficulties, as the transportation of the barley to Larsa by boats implied that the level of the water in the canals needed to rise, which Ešnunna opposed strongly. However, the political decision to obstruct the canals in order to prevent the transportation of the barley to Larsa also caused a shortage of water for watering the fields. Rīm-Sîn sent messengers to the king of Ešnunna, but even after five months, the canals had not been opened yet (See Charpin 1983: 104–106 and 2004: 126). 36.  It is still difficult to pinpoint the exact date of the fall of Larsa (see Charpin 2004: 319–323). Recently, Dercksen (2014) suggested that Larsa fell shortly after the twenty-first day of the tenth month of Hammurabi’s thirtieth year of reign. This fits in well with the year name of Ḫammurabi’s thirty-first year of reign, commemorating his defeat of Rīm-Sîn in his thirtieth year of reign (mu ḫa-am-mu-ra-pí lugal.e giskim.ti an den.líl.bi.ta igi éren.na.šè ì.gin.na.àm usu maḫ dingir gal.gal.e.ne mu.un.na.an.sum. uš.àm (ì.si.in.na) ma.da e-mu-ut-ba-lumki ù lugal.bi ri-im-dEN.ZU šu.ni sá bí.in.du11.ga … x.bi.šè zi.ni x ba.è … x ki.en.gi ki.uri du11.ga.né bí.in.tuš.e “The year: Hammurapi, the king, with the help of An and Enlil, went before the army , by the supreme power which the greatest gods had given to him, conquered (Isin), the land of Emutbal and its king, Rīm-Sîn; he brought forth his life to its …; caused … Sumer and Akkad to dwell at his command,” see Horsnell 1999 [vol. 2]: 141). Apart from Larsa and Isin, it is very likely that Ḫammurabi also captured Ur and Nippur at the end of his thirtieth year of reign. A text from Ur (UET 5 419) is dated “the year Ḫammurabi became king” (mu ḫa-am-mu--ra-pí lugal), which refers to Ḫammurabi’s thirty-first year of reign; see Horsnell 1999 (vol. 1): 41–44. 37.  See the year name of Ḫammurabi’s thirty-third year of reign: mu ḫa-am-mu-ra-pí lugal.e i7 ḫaam-mu-ra-pí-nu-ḫu-uš-ni-ši-šà.ge.túm.àm.(an).den.líl.lá mu.un.ba.al.lá a da.rí ḫé.gál.ka nibruki eriduki úriki larsaki.ma unuki.ga ì.si.in.naki mu.un.ga.ra.àm ki.en.gi ki.uri bir.bir.re.a ki.bi.šè bí.in.gi4.a ugnim ma.ríki ù ma.al.giki mè.ta bí.íb.šub.bé ma.ríki ù á.dam.bi ù uru.didliki (ma.da kur) su.bir4ki (é-kál-la-tumki (kìlib) bu.ru.un.daki ù ma.da za-al-ma-qumki gú i7 idigna en.na i7 buranun gú ki.šè mi.ni.gar) du11-ga.né ku.li.bi bí.in.tuš … “The year: Hammurapi, the king, dug the canal ‘Hammurapi means abundance for the people, the beloved of (An) and Enlil’; provided perennial water of abundance for Nippur, Eridu, Ur, Larsa, Uruk, and Isin; restored Sumer and Akkad, which had been scattered; overthrew in battle the army of Mari and Malgium; subjugated Mari and its villages and the many cities (of the mountain land) of Subartu, (Ekallatum, (all of) Burunda, and the land of Zalmaqum on the bank of the Tigris up to the Euphrates); and caused to dwell at his command in friendship …,” see Horsnell 1999 (vol. 2): 146–147. His digging of the “Hammurabi means abundance for the people” canal is also commemorated

Bad Moon Rising

57

seem to have a special interest in the city of Ur. He did mention the city and its god in the prologue of his Code, but merely as one of the cities dedicated to an astral deity in the enumeration of twenty-five cities grouped on the basis of theology and geography. 38 Under the reign of Ḫammurabi’s son and successor Samsu-iluna, the Baby­ lonian empire, if it can be called an empire at all, 39 soon started to disintegrate due to a combination of internal economical and institutional crises and the threat of invading Kassites. In the middle of Samsu-iluna’s eighth year of reign, a revolt broke out in the South, 40 and Rīm-Sîn II, who declared himself king in Larsa, soon started to conquer other southern areas. He soon managed to take control of Ur, where he ruled for at least twelve months, 41 although it seems that there were plenty of competitors, as is shown by the attestation of a king named Iluni in a seal inscription and year name found in Ur. 42 This might have been the ruler of Ešnunna, who also rebelled against Samsu-iluna, and possibly took control of Ur for a short period of time during Samsu-iluna’s seventh year of reign. After having dealt with the Kassites, Samu-iluna started the reconquest of the south in his ninth year of reign and managed to recapture Ur in his tenth year, which was done with great force as he claimed to have destroyed Ur’s walls. 43 The recapture of Ur involved its general destruction, and although some resettlement must have taken place, as is shown by a few texts dating from the eleventh and twelfth year of reign of Samsu-iluna, 44 Ur headed for a long period of general decline. Samsu-iluna was confronted with more revolts in the south; these revolts eventually led to the loss in one of Ḫammurabi’s royal inscriptions, together with him gathering the scattered peoples of Sumer and Akkad and providing them pastures and watering places; see RIME 4.3.6.7, especially ll. 17–37. 38.  See Roth 1997: 77 and Charpin 2004: 334. 39.  See Charpin 2004: 334–335. 40.  See Stol 1976: 44–58 and Charpin 2004: 337–339 with references. According to Charpin (2004: 337), the difficulties were probably due to the increasing threat of floods in northern Babylonia, which Samsu-iluna tried to prevent by digging a canal to drain the swollen water of the Euphrates into the depression of the present-day Lake Habbaniyah; but in doing so, he caused a shortage in the water supply in the South, resulting in poor harvests, debts, and eventually revolts. The digging of the canal is commemorated in his third year name: mu sa-am-su-i-lu-na lugal.e ki.in.gub.(a) nì.u4.ul.(lí).a.ta ba.šub. ba i7 sa-am-su-i-lu-na-(lugal)-na-qá-ab-nu-uḫ-ši mu.un.ba.al.a šà ma.da gán dagal im.ta.an.è.a bar kalam è.a kur.kur.ra “The year: Samsu-iluna, the king, dug the canal ‘Samsu-iluna(, the king,) is the source of abundance for the people,’ a watercourse which from ancient times had fallen into disrepair, and made it flow forth in the broad fields in the heart of the country, overflowing the land and going forth into the foreign countries,” see Horsnell 1999 (vol. 2): 177–178. See also Cole and Gasche 1998: 11–13. 41.  See Stol 1976: 57 and Van De Mieroop 1992a: 68 n. 144. Rīm-Sîn II calls himself “king of Ur” in his second year name: mu dri-im-dEN.ZU lugal úriki.ma é.mud.kur.ra.ke4 ki.eden.šè bí.in.gar.ra.a “Year Rīm-Sîn, king of Ur laid the foundations of the Emudkurra at ki.edin,” see Sigrist 1990a: 61 and Charpin 2004: 338 n. 1762. According to Frayne (1992: 33 n. 246), ki.edin is to be identified with the city of Edina in the Sealand. 42.  See the seal inscription on a small receipt dated Si 8/ix/28: dEN.ZU-i-bi-⸢šu⸣ dumu dEN.ZU-iqí-ša-[am] ìr di-lu-[ni] (see Ormsby 1972: 95 no. 11 and 99 seal 3; and Charpin 1986a: 127) and the year name on the sale contract HE 167: mu i-lu-ni lugal (see Charpin 1986a: 174–175). 43.  This is commemorated in Samsu-iluna’s eleventh year name: mu sa-am-su-i-lu-na lugal.e du11. ga an den.líl.bi.ta bàd (gal) úriki (larsaki) (ù) unuki.ga mu.un.gul.la ugnim ki.uri a.rá … gištukul.ta in.sìg. ga “The year: Samsu-iluna, the king, at the command of An and Enlil, destroyed the (great) walls of Ur, (Larsa) (and) Uruk and defeated the army of Akkad time,” see Horsnell 1999 (vol. 2): 195. 44.  Three texts from Ur date from Samsu-iluna’s eleventh year of reign (UET 5 268 and 485; and YOS 12 356) and only one text from his twelfth year of reign (UET 5 868), see Charpin 1986a: 155–156,

58

Katrien De Graef

of the south, which suffered a great economic decline due to a collapse of the entire system. Urban centres such as Ur could no longer be supported and settlement was reduced to dispersed villages, causing a massive emigration to the north and an, at least partial, abandonment of the south for a long period of time. 45

Fortune Against All Odds? Before being destroyed and consequently abandoned for several centuries, Ur lived through more than two and a half centuries of political instability, came under the control of various dynasties, and fell victim to several violent raids and attacks, water supply problems, famine, and revolts. Nevertheless, in spite of these turbulent and seemingly unfortunate times, textual evidence from the early second millennium BCE shows that the city of Ur lived through periods of economic, religious, and cultural prosperity and remained important and prestigious as a commercial and sacred urban centre and port. Indeed, literary production was booming in the early second millennium. Following their predecessors of the Ur III empire, rulers of especially the Isin but also the Larsa dynasty, commissioned the composition of many literary works: hymns, praise poems, literary letters, and letter-prayers alike. 46 Lipit-Ištar of Isin even ordered the composition of a so-called law collection, 47 no doubt in imitation of his illustrious predecessor Ur-Namma. A particular literary genre is that of the so-called city laments, one for Ur and one for Sumer and Ur, in which the fall of Ur and its empire is presented in terms of divine abandonment. 48 These laments, together with Išbi-Erra’s hymns and royal letters of Ibbi-Sîn and Išbi-Erra, 49 clearly fit in with the attempts to legitimize the new dynasty. As keeper of the tradition and legacy of the Ur empire, Išbi-Erra need­ed to justify the new royal lineage as well as the new seat of power, namely Isin. As such, the fall and destruction of Ur was presented as an inevitable fact, for it was decided by the gods, who allowed the enemy to destroy the empire. The fact that the kings of both the Isin and Larsa dynasties attached great importance to the ancient capital is shown in their royal inscriptions and year names, which commemorated the restoration, building, and embellishment activities they

167–169, and 209–210. After Si 12, written documentation stops at Ur, Uruk, and Larsa for several centuries. 45.  According to Gasche (1989: 109–143), archaeological prospecting shows a complete absence of late Old Babylonian material, implying that Ur (as well as Uruk and Larsa) had been abandoned for a long period. The reasons for this abandonment, and consequent massive emigration to the north, are as yet unclear. It might have been caused by a foreign (Elamite) attack, by kings of the Sealand dynasty, who are thought to have conquered Nippur late in Samsu-iluna’s reign, forcing the populations of the south to abandon their cities, by a considerable shift of the course of the Euphrates causing problems with the water supply; see Charpin 2004: 342 with references. 46.  See ETCSL: City Laments (2.2.2–2.2.6), Royal Praise Poetry, Isin Dynasty (2.5.1.2–2.5.8.1) and Larsa Dynasty (2.6.2.1–2.6.9.a), Literary Letters and Letter-prayers, Isin, Larsa, and other dynasties (3.2.01–3.2.05) with extensive bibliography. 47.  See Roth 1997: 23–35. 48.  See Michalowski 1989. 49.  See Römer 1965; Hallo 1966; van Dijk 1978; Vanstiphout 1989–90; Michalowski 1980, 1993, and 2011; and Tinney 1996.

Bad Moon Rising

59

had carried out in Ur (see supra). This is corroborated by the fact that daughters from both Isin and Larsa rulers were ordained entum-priestesses at Ur. 50 Moreover, apart from Nippur, the city of Ur is the largest source for manuscripts of early second millennium literature, with at least one, but probably more houses serving as scribal schools, which points to the importance and prestige of Ur as a cultural, intellectual, and educational centre. 51 However, the most important information on Ur’s prosperity comes from the economic and legal tablets from private and other archives, as is shown by the works of Charpin (1986), Diakonoff (1985, 1990, and 1995) and Van De Mieroop (1992a and 1992b). Indeed, soon after the fall of the Ur III empire, economic activities started up again. Maritime trade with Dilmun was pursued by Išbi-Erra even before the Elamites were expelled from the city, implying that their control was only minimal. A possible explanation lies in the fact that the sea harbour was no doubt located outside of the city of Ur. 52 Early in the reign of his successor Šū-ilīšu, the first texts reappear at Ur, showing a continuation of administrative practices and a sound economy, and implying that the former capital, or at least parts of it, recuperated rather quickly. 53 Throughout the next two centuries, until the middle of Rīm-Sîn’s reign, Ur’s economy was clearly thriving in all of its sectors—temple, palace, and private citizenry, which were largely interdependent. 54 An important part of Ur’s economy was controlled by the temples, whose involvement comprised various sectors, from agriculture to metallurgy. 55 However, the role of the temple administration changed over time, due to the fragmentation of temple offices, from a system originally completely supervised by temple personnel to one in which private entrepreneurs played a major role. This is apparent from the change in function that the Ganunmaḫ or “Great Storage House,” which played an important role in the administration of the temple, underwent, as is shown by Van De Mieroop (1989; 1992a: 77– 81; 1992b: 125; and 2015: 361–362). From the reign of Šū-ilīšu of Isin until that of Sîn-iddinam of Larsa, the main func­ 50. See supra and Van De Mieroop 1992a: 116–117. 51.  See Charpin 1986a: 420–486; Tinney 1998; and Brusasco 1999–2000. 52.  Two (unfortunately unprovenanced, according to CDLI possibly from Isin) administrative texts, BIN 9 403–4, dated to the twentieth day of the sixth month of the nineteenth year of reign of Išbi-Erra, show that the maritime trade with Dilmun was pursued by Išbi-Erra even before he was able to expel the Elamites from Ur and reconquer the city, implying that Ur’s kārum (sea harbour and commercial district) was still functioning at the time. The kārum of Ur has not been located archaeologically, but is likely to have been situated at the so-called Diqdiqqah site, located to the north-northeast of the inner city of Ur (see Van De Mieroop 1992a: 194 and 1992b: 121). Its extramural location may explain the fact that soon after the fall of the Ur III empire the maritime trade with Dilmun resumed. 53.  The first text from the Isin dynasty found in Ur dates from the fifth year of reign of Šū-ilīšu (Loding 1976: 237 no. 1, ŠI 05-x-/), implying there was an interruption of some twenty years in the written documentation, the last texts from the Ur III dynasty dating from the twenty-third year of reign of Ibbi-Sîn (UET 3 711, IS 23-xii-22). However, if Ur’s sea harbour was indeed still functioning eighteen years before (see supra), there might have been hardly any interruption of trade activities after the fall of Ur. According to Stone and Zimansky’s latest findings during their recent excavations at Ur, the Ur III–Isin/Larsa transition suggests the fall of Ur did not involve a violent demise as described in the Lamentation over the Destruction of Ur, notwithstanding the fact that there was at least some discontinuity in architecture and perhaps in subsistence patterns (see Stone and Zimansky 2016). 54.  As is apparent from both the public and private archives excavated by Woolley, see Charpin 1986a; Van De Mieroop 1989, 1992a, 1992b, and 2015. 55.  See Van De Mieroop 1992a: 77–105.

60

Katrien De Graef

tion of the Ganunmaḫ was twofold, viz. (1) the collection and storage of durable materials, payments in silver by various people for rents and taxes, and valuable objects, received as ex-votos, and (2) the collection and storage of perishable food products, such as dairy, oil, dates, and spices, which were issued for offerings and rations. However, from the start of Warad-Sîn’s reign on, the role of the Ganunmaḫ was restricted to the collection and storage of durable materials and valuable objects, whereas the collection and storage of perishable food products were, at least in part, handled by private entrepreneurs. According to Van De Mieroop (1992a: 79–81), this change in function is to be attributed to Sîn-iddinam’s restoration of the Ganunmaḫ. A simultaneous change is perceptible in the role played by the palace in the administration of Ur’s economic resources. When the temple started to privatize many of its functions, the palace drew administrative duties, originally executed by the temple, away and relocated them in their own Larsa-based offices. 56 Ur had a unique position as seaport, and controlled all access routes for products entering Mesopotamia from overseas. From old, Ur’s trading connections extended as far as the coasts of the Persian Gulf and the Indian peninsula. From the early second millennium onward, Dilmun, whose exact location remains uncertain but is to be situated in the area from Falaika to Bahrain, functioned as a kind of entrepôt where products from Oman, Arabia, and India, such as copper, semi-precious stones, shells, ivory, and spices, were stored to be imported into Mesopotamia in exchange for silver and agricultural and craft products. 57 Both Mesopotamian and Dilmunite traders were involved. Private merchants of Ur, often called ālik Dilmun, Dilmun traders, collected investments from numerous individuals to finance their expeditions. Overall these investments were small, no doubt to reduce the risks, but in case of larger sums a certain return would be guaranteed. Although this trade was undertaken by private citizens, the temple maintained some control by levying taxes on all imported products, a role which was taken over by the palace over time. 58 The imported goods, as well as agricultural and crafts products from Ur, were then traded further inland by boat or caravan. Ur had extensive contacts inland, up to the far north, but the overland trade was subject to the political situation and sometimes warfare and instability prevented merchants from reaching certain areas. 59 Locally, the urban population traded movables as well as immovables. Bulky agricultural products, such as barley and dates, were traded in the harbour area, no doubt for reasons of storage, whereas retail sale took place in shops and booths located in the city. 60 Houses in the city, orchards outside of the city, as well as temple prebends were traded and/or let. Remarkably, there are very few field sales 61 and no inheritance divisions mentioning fields, implying private ownership 56.  See Van De Mieroop 1992a: 111–115. 57.  See, among others, Oppenheim 1954; Leemans 1960 and 1968; Heimpel 1987; Crawford 1998 and 2005; Potts 1994; and Potts 1990, 2009, and 2016. 58.  See Oppenheim 1954 and Van De Mieroop 1992a: 194–197 with references. 59.  See Leemans 1960 and 1968; and Van De Mieroop 1992a: 198–199 and 202–203. 60.  See Van De Mieroop 1992a: 199–203. 61.  Van De Mieroop (1992a: 169) mentions three field sales: UET 5 167, YOS 12 277 (see Charpin 1986a: 173–174), and TS 4 (see Charpin 1980: 36–37 and 203). According to Van De Mieroop (1992a: 169), all three were of a special nature, viz. uncultivated (ki.gál), subject to flooding (ú.sal), and very small. Note, however, that UET 5 167 is probably not from Ur but from Larsa (see Van De Mieroop 1992a: 169 n.

Bad Moon Rising

61

of fields was greatly restricted. Possibly, the arable land was primarily owned by the central institutions, or by landowners who did not reside in the city, as the business contracts between the urban population and people living outside of the city walls seem to imply. 62

How to Approach This Paradox? How can Ur’s economic and cultural prosperity and the political instability the city underwent be reconciled? How can we approach this paradox? Usually, the periodic collapse and rise of dynasties, and consequently cities, is considered in political terms or attributed to external factors such as invasions or climate problems. As such, our main focus lies on pinpointing the causes of crises, disruptions, and implosions, 63 assuming that the normal state of affairs is stability and prosperity. In other words, we tend to focus on trying to explain where it went wrong. However, is this the right question to ask? Taking the course of events in early second millennium BCE Ur into account, and the course of history in general for that matter, would we not better ask ourselves why it did not go wrong? Would it not be worthwhile trying to understand why the Elamite invasion followed by centuries of political instability due to the competing dynasties, involving invasions and raids, did not cause Ur’s economy to collapse? A promising way to approach this paradox is la théorie de la régulation or régulation theory. 64 Régulation theory is a systems theory developed in the early 1970s in response to the economic crisis following the long period of seemingly stable post-war economic growth on the one hand and the inability of neoclassical economic theory to understand and respond to that crisis on the other hand. Régulation theory, which is influenced by structural Marxism and the Annales school, among others, studies “the transformation of social relations as it creates new forms, both economic and non-economic, that are organized in structures.” 65 These new forms then reproduce a determinant structure, the mode of production. 66 Lipietz (1987: 18–19) formulated the key methodological principles of régulation theory as follows: 67 1), and that both the area and price of TS 4 are broken (see transliteration in Charpin 1980: 203), which makes it impossible to know how large the field was. 62.  The controversy on private individual or collective ownership of arable land vs. the ownership of arable land by the State and/or temples in early Old Babylonian Ur in particular and southern Babylonia in general, goes far back in time and continues until today. As noted by Renger (1995–96: 288) this controversy is based on the “uneven distribution of the sources and heuristic problems with regard to their interpretation.” Various aspects of this controversy are treated by, among many others, Liverani 1984; Diakonoff 1982 and 1985; Renger 1995–96; van Driel 1998; and Garfinkle 2012: Chs. 2, 3, and 8. 63.  See among others Adams 1988; and Yoffee 1988 and 2010. 64.  This and the following paragraphs of this paper largely draw upon Boer 2015, especially his first chapter. A general overview of the history and applications of régulation theory is to be found in Boyer and Saillard 2002. The French term régulation designates “the social, institutional, and ideological factors that determine the stabilities and transformations of a system as a whole” (Boer 2015: 32). Being not analogous to the English term ‘regulation’, and hard to translate, I follow Boer in using the French term régulation. See also Jessop 1997: 288. 65.  See Aglietta 1979: 16. 66.  See Boer 2015: Ch. 1, esp. pp. 31–41; Boyer 2002; Boyer and Saillard 2002; Jessop 1997; and Hein, Dodig, and Budyldina 2014: 2–9. 67.  See also Boer 2015: Ch. 1, esp. pp. 34–35; and Boyer and Saillard 2002.

62

Katrien De Graef 1.  Society, and within society economic activities, is a network of social relations: individuals connecting with each other and engaging in exchanges 2.  Each of these social relations is in se contradictory, as a result of which living within this network of social relations is difficult. Therefore, crisis is the normal, natural state and non-crisis is a rather chance-event. 3.  There are, however, long periods of time when all goes well, when living within this network of social relations, however difficult, works out fine. This is called a régime d’accumulation or accumulation regime. 4.  The question is thus how an accumulation regime was achieved. In order for an accumulation regime to be achieved, individual expectations and behaviour must be in line with the needs of this accumulation regime. This is a process with two aspects, the first one functioning as habit, or “habitus, as Bourdieu would say, in the minds of individuals with a particular culture and willingness to play by the rules of the game;” the second one “through a set of institutional forms or codification of the fundamental social relations that underpin economics and may vary widely.” 68 A set of such behavioural patterns and institutional forms is called a mode de régulation or mode of régulation.

In other words, the starting point of régulation theory is that all social relations making up society are in se contradictory. Hence, crisis is the normal state of affairs and non-crisis the abnormal or exceptional state of affairs. The key question is therefore how stability is obtained, or as Boer (2015: 37) puts it “how can a system reproduce itself if it is unstable, if the social relations at its core are inherently contradictory and crisis-ridden?” Régulation theory considers the economy to be inter­connected with society in its political, social, and cultural systems. The specific shape of these systems at one time constitutes the mode of regulation of that time. Within such a mode, an accumulation regime is defined and can function until regime and mode conflict, leading to a crisis. Régulation theory distinguishes four types of crises, described by Boer (2015: 36–37) as follows: 1.  Exogenous crises are due to an external event, such as invasions or climate change. Although they can be very perturbing, they cannot endanger the mode of régulation. They are not threatening in themselves but become real perils during structural crises. 2.  Endogenous crises are internal, cyclical crises that are inevitable and necessary for they “perform a useful function in managing the extremes of the economic system.” 3.  A structural crisis is a crisis in the accumulation regime and mode of régulation, and as such a special form of endogenous crisis in which the “internal socially determined tensions that enabled the system to rise to dominance, now threaten to tear it apart.” Contrary to the normal endogenous crises, a system in a structural crisis cannot recover due to its internal dynamics, as a result of which a thorough reconfiguration or shift in the mode of régulation is needed. Such a crisis is often a forerunner for the fourth type of crisis, namely, the ultimate crisis, a crisis of modes of production when the whole economic structure can no longer be sustained and the system collapses completely. 4.  Ultimate crisis. 68.  See Lipietz 1987: 19.

Bad Moon Rising

63

How to Apply Régulation Theory to Early Second Millennium Ur? So how can régulation theory be applied to the early second millennium BCE history of Ur? How can the fundamental changes the city of Ur underwent during this period be discussed in terms of the two central concepts of accumulation regime and mode of régulation? And can it help us to approach and understand the con­ tradiction between the politically unstable climate and the apparent economic and social stability? It goes without saying that the answer to these questions requires an extensive and thorough research into not only the Ur situation, but also that of the rest of the territories controlled by the Isin, Larsa, and Babylonian dynasties, in order to get a complete picture of early second millennium BCE society. It goes without saying that this largely exceeds the present contribution, which restricts itself to the presentation of some preliminary outlines on how the régulation approach can possibly contribute to our understanding of the Ur situation. Boer (2015: 111) describes ancient Near Eastern economy as basically dominated by two basic institutional forms that were constantly in tension with each other: 1.  “subsistence survival:” the crucial allocative institutional form focused on agriculture and socially determined by kinship household 2.  “state estate”: the extractive system of temple and palace estates

This certainly works for the early second millennium BCE southern-Mesopotamian economy, where the majority of the population was engaged in subsistence agriculture. 69 Although these rural agricultural villages and communities no doubt comprised the bulk of all socio-economic life, they are almost completely absent in the written documentation. The reason is obvious: the sources originate from urban centres and are thus mainly concerned with the urban elites. Very much present in the written record is the system of estates, which ap­ pear­ed together with the first urban centres where individuals became disengaged from direct production and needed others to produce food, clothing, etc. At first there were temple estates, but over time the palace took over the existing patterns of estates, along with control of the temples, and gradually members of the wealthy urban elite got involved. 70 The estates employed permanently and periodically indentured labourers in exchange for rations, as a result of which they had direct control over both labour and the means of production. Other estate employees were assigned sustenance fields in exchange for services rendered. 71 According to Boer (2015: 118–121), the tension between two institutional forms is in part due to the estates having a constant shortage of labour. The need for estates to gain ever-new labourers not only put an increasing strain on the agricul-

69.  See, among others, Diakonoff 1975, 1982, 1985, and 1988; Adams 1982; and Liverani 1984. 70.  See, among others, Diakonoff 1982; Renger 1979 and 1994; Silver 1995, esp. Ch. 1; and Boer 2015: 110–145. For early Old Babylonian Ur in particular, see Charpin 1986a: 233–302; Van De Mieroop 1992a: 77–106; and Diakonoff 1995. 71.  See, among others, Gelb 1965; Diakonoff 1982 and 1985; Renger 1979, 1994, and 1995–6; Klengel 1987; and, most recently, on the city of Larsa, Richardson 2015.

64

Katrien De Graef

ture of village communities, but also led to the ruling classes seeking other sources for their own needs, namely the institutional form of “tribute exchange.” 72 This may have played a part in the crisis which would lead to the downfall of the Ur III state, as it largely involved a system of “tribute exchange,” using taxation, tribute, and exchange as mechanisms for appropriating goods. We might even speak of a form of exploitation colonialism when it comes to parts of the periphery. 73 However, both the estate and tribute systems relied on the subtle combination of persuasion and force provided by the state. This clearly went wrong as governors of subject provinces ceased contributing their shares for the support of the palace and the temples and showed their independence by omitting the use of year names proclaimed by the king. 74 The constant shortage of labourers in the estate system, and the influx of newcomers, the Amorites, semi-nomads who would establish self-governing communities and only occasionally entered military service if it served their interest, led to a substantial loss of control over both labour and the means of production by the ruling classes, resulting in a massive economic crisis. 75 Exogenous crises such as foreign invasions and poor harvests only added to this structural crisis, resulting in a crisis in the accumulation regime and mode of régulation. However, this did not lead to a total collapse of the system, as rather soon after the fall of Ur, a certain degree of recovery had been achieved, as is shown by the continuation of administrative and economic practices apparent in the written documentation. But, whereas we would expect to see a thorough reconfiguration or shift in the mode of régulation after such a structural crisis, this is not really the case. For, a similar system, made up of “subsistence survival” and “state estate,” dominated the following two centuries. As already noted, the estate system underwent some changes in this period, leading to an increasing role of private entrepreneurs and the privatization by the temple of a part of its functions. This did not change the system in se, for these private entrepreneurs were largely interconnected with the temple and the palace, and acted within the boundaries of this very estate system. It is, however, a symp­ tom of another development. For, the urban elites, whose wealth was generated through the estate system, started to invest their wealth in order to gain profit. As such, a private commercial circuit, parallel to the estate system, came into being. Its origins go back to the third millennium, but it developed further and became more impor­tant in the course of the second millennium, although it always remained small-scale obviously. This parallel circuit of private economy was largely interconnected with and dependent on the estate system, of which it was actually a kind of outgrowth, as it was for the most part operated by the same agents: members of the urban elite, who often combined an official temple or palace function with 72.  An extractive form using tribute, taxation, and exchange as mechanisms for appropriate goods that one did not produce, see Boer 2015: 139–163 and 236. 73.  See, among others, Wilcke 1970; Gomi 1984; Steinkeller 1987; Yoffee 1988; Sallaberger 1999; and Sharlach 2004. 74.  Išbi-Erra, who was asked by the last king of the Ur III empire, Ibbi-Sîn, to defend Isin against the Amorites, declared himself king in the eighth year of reign of Ibbi-Sîn and, as such, founded the dynasty of Isin. The first year name of Išbi-Erra known to us is that of his fourth year of reign (mu gir13-tabki ba-ḫul “The Year Girtab was destroyed,” see Sigrist 1988: 13). By then, the extent of the Ur III empire had shrunk to the city’s environs and Ibbi-Sîn tolerated a parallel dynasty under his former general Išbi-Erra. See Charpin 2004: 57–60. 75.  See Gomi 1984.

Bad Moon Rising

65

private entrepreneurship. 76 The same goes for the private merchants involved in overseas trade, in which the city of Ur being the only seaport played a major role. In order to obtain luxury goods, the estates engaged in international trade. This was contracted out to private merchants who obviously got a piece of the pie. The estate system and private commercial circuit benefitted from each other, for they had mutual interests. At the same time, the disruption of one system affected the other. When the estate system was put under pressure, which, as we saw earlier, was mainly caused by the constant shortage of labour, the private commercial circuit dried up. The same applies vice versa. Investments and speculation, after all, involve risks. Especially those parts of the urban population that were not fabulously rich, but simply well-to-do, the lower strata of the urban elites, could easily get into temporarily financial trouble, which is apparent from the many loans that are preserved. 77 For the creditors, mostly temple, private entrepreneurs, and merchants, lending was a lucrative business due to the very high interest rates. Although not all debtors were people in economic distress—borrowing could have been necessary and even beneficial in case of an investment with a guaranteed return—the exorbitant interest rates cannot but have led to the insolvency of many a debtor, resulting in them losing their properties and eventually even being forced to do manual labour. As the mode of régulation in the early second millennium BCE was not fundamentally different from that before the fall of Ur, we cannot but assume that it was bound to go wrong again. In other words, that the built-in limitations of the institutional form of estates kept economic crisis an ever-present reality. And that seems to be exactly what happened, as is shown by the references to the edicts the kings of Isin, Larsa, and Babylon issued. 78 These edicts functioned as bypasses to ensure that the crisis-ridden system continued to survive, and were as such barometers of economic instability. No less than five kings of Isin—Išme-Dagan, Lipit-Ištar, Ur-Ninurta, Irra-imitti, and Enlil-bani—issued decrees in which they freed inhabitants and temples from duties, services, and dues, cancelled tithes, lowered the days of corvée labour to be rendered annually, and/or established justice in another way. 79 As most of these edicts mention the inhabitants and/or temples of Nippur, and to a lesser degree Isin, one might question the degree to which these edicts reflected social justice realm-wide, as Renger (2002: 150–151) already noted. One might even wonder whether they were in part political means in order to (re-)gain the allegiance of Nippur, which was after all the religious capital.

76. For the situation in early Old Babylonian Ur in particular, see Van De Mieroop 1992a, esp. 169–210. 77. Loan documents are by far the most commonly preserved records from the Old Babylonian period, see Skaist 1994: 11–21 and Van De Mieroop 2002b. As for Old Babylonian Ur, there are some ninety loan contracts preserved, the majority of which are dated to the reign of Rīm-Sîn I (see Charpin 1986a: 472–481; Van De Mieroop 1992a: 253–299; and Skaist 1994: 85–87, 96, 196–170, 235–236, and 288–289). Charpin (2015) mentions the discovery of one silver loan dated to the reign of Samsu-iluna during the recent excavations (autumn 2015) at Ur. It is generally believed that loan documents were broken after repayment, implying that the preserved ones are those that were not paid back. 78.  See Kraus 1984 and Renger 2002 for a general overview and analysis of the Old Babylonian royal edicts. 79.  See Kraus 1984: 16–30 and Renger 2002: 145–151.

66

Katrien De Graef

The only king of Isin who explicitly mentions Ur is Lipit-Ištar. This, however, does not come as a surprise, as Ur is lost to the rulers of the Isin dynasty after Lipit-Ištar. In the prologue of his law collection, Lipit-Ištar states that he set free the inhabitants of Nippur, Ur, and Isin, both male and female, who were put under the yoke of doing bond service, and put them into their previous social status. He made family members support each other and ordered service obligations to be shared by the father and his sons. He also claimed to have lowered the number of working days in the estates to ten per month for the labourers and seventy per year for those who have to render corvée, probably those who were assigned sustenance fields. 80 Six royal inscriptions and two royal hymns refer to the fact that Lipit-Ištar established justice in the land of Sumer and Akkad. 81 The same phrase is used in his year name a. 82 Renger (2002: 147) argues that the establishing of justice by Lipit-Ištar mentioned in his royal inscriptions must refer to a royal decree issued later in his reign, as they refer to building activities that must have taken some time to complete. Year names e and f, in which he claims to have released the arrears of the land of Sumer and Akkad, 83 might refer to such a later decree. Contrary to later royal edicts, none of the edicts by the Isin kings are referred to in the contemporary administrative, economic, and legal texts. It is, in other words, unclear how these measurements affected daily life. 84 80. ii 1–40: “At that time, the inhabitants of Nippur, both male and female, the inhabitants of Ur, both male and female, the inhabitants of Isin, both male and female, as well as (or: that is) the inhabitants (of the entire lands) of Sumer and Akkad, both male and female, people put under the yoke of doing bond service, I set them free, put them into their previous status (of free persons). According to the personnel roster (dub.sag) (I decreed) that a father would stand in for his son, that a son would stand in for his father, that a father would do service together with his sons, and that a son would do service together with his father. I imposed service (equally) on the household of a living father and on the undivided household of brothers. I, Lipit-Ištar, son of Enlil, obliged those in the household of a living father and those living in the undivided households of brothers to do service seventy days (per year). I made those living in a household of institutional subjects (guruš) serve ten days per month. …” (translation from Renger 2002: 145–146). A slightly different translation can be found in Roth 1997: 25–26 (note in particular the passage ii 10–15 [lú gú].⸢bi⸣.a [šudul(?)] nam.arad [ḫu].⸢mu⸣.ni.ib.ak ⸢ama⸣. ar.gi4.bi [ḫu].mu.gar ki.bi.šè ḫé.bí.dab5, translated by Roth as “I liberated, …, who were subjugated [by the yoke(?)], and I restored order.” vs. Renger’s “people put under the yoke of doing bond service, I set them free, put them into their previous status (of free persons).”). See also Kraus 1984: 19–27 and Renger 2002: 145–148. 81. u4 nì.si.sá ki.en.gi.ki.uri.a i.ni.in.gar.ra.a “When I established justice in the land of Sumer and Akkad” (see RIME E4.1.5.2–6 and E4.1.5.8) and nì.si.sá ki.en.gi.ki.uri.a mu.ni.gar “I/You established justice in the land of Sumer and Akkad” (Lipit-Ištar A 92, see Römer 1965: 36, and Lipit-Ištar B: 38, see Vanstiphout 1978: 38–39). 82. mu dli-pí-it-iš8-tár lugal.e nì.si.sá ki.en.gi ki.uri.a mu.ni.in.gar ‘Year in which king Lipit-Ištar established justice in the land of Sumer and Akkad,” see Sigrist 1988: 28. 83.  mu (ús.sa) dli-pí-it-iš8-tár lugal.e láxu ki.en.gi ki.uri i.in.gál.la gá.ra “Year (after the year) king Lipit-Ištar released the arrears of the land of Sumer and Akkad,” see Sigrist 1988: 28. 84. This might be due to the fact that a part of Lipit-Ištar’s measures as expressed in the prologue of his law collection concern obligations within the estate system. However, the liberation of the inhabitants of Ur who were put under the yoke of debt bondage and the restoration to their previous social status—at least according to Renger’s (2002: 146) interpretation of this passage—might refer to insolvent debtors who had to cede their property to their creditors and/or were even enslaved by their creditors in order to pay off their debts, which we would expect to be referred to in the economic and legal texts, as is the case for later edicts. This is not the case, which might imply that this passage does not refer to insolvent debtors, but it might also be due to the gaps in our sources.

Bad Moon Rising

67

The only king of the Larsa dynasty who is known to have proclaimed remission of debts is Rīm-Sîn I, who issued no less than three royal edicts. 85 Contrary to the Isin dynasty, the royal edicts issued by Rīm-Sîn of Larsa are never mentioned in his year names, royal inscriptions, or literary texts, but are only known from the economic and legal texts from private archives, in which they are referred to as ṣimdat šarrim “royal decree.”  86 The texts from Ur only refer to two of his edicts, but it seems logical that they were all valid throughout the kingdom of Larsa, including at Ur. Before his three “official edicts”—i.e. those explicitly referred to as ṣimdat šarrim “royal decree”—Rīm-Sîn already seems to have taken similar measures, which are referred to in a rather enigmatic way in TCL 10 40, a sales contract from Larsa. 87 This contract is dated to the last month of Rīm-Sîn’s fifteenth year of reign and concerns the sale of a date orchard by Ku-Ninšubur to Gaz-Sîn “after the forehead of the land became clean and the sealed documents had been discarded.” 88 Although it remains as yet unclear what is exactly meant with “the forehead of the land having become clean,” Kraus (1984: 31–33) believed it to have been an earlier edict by Rīm-Sîn regulating debts incurred by private individuals. Whether a distinction was made between sustenance loans, which were declared null and void, and business loans, which were not, as was the case in the edict of the later king Ammi-ṣaduqa, 89 is not known. Taking into account Landsberger’s (1939: 230) statement that adding such a phrase expressly marked the contract as not falling under this special measure, Kraus (1984: 33) suggested that the cause for drawing up this document might have been the remission of debts incurred among private individuals: by ceding his orchard to his creditor, Ku-Ninšubur paid off the debt he owed Gaz-Sîn. In other words, notwithstanding the fact that measures cancelling probably both government and private debts had been taken by the king, Ku-Ninšubur settled his debts by selling (giving) his orchard to Gaz-Sîn. If this interpretation holds true, one might ask whether this was an exceptional situation for reasons unknown to us, or a clever way to by-pass the cancellation of debts by 85.  UET 5 253 (RS 35/xii/—) refers to Rīm-Sîn’s second royal decree: ina ṣimdat šarrim warkītim (ll. 10–11), and VAS 13 81 (RS 41/i/—) refers to Rīm-Sîn’s third royal decree: warki ṣimdat šarrim 3.kam. ma (l. 9), see also Kraus 1984: 33–50. 86.  There is no consensus on the exact meaning of ṣimdat šarrim and to what it refers when used in legal and economic texts and letters, viz. general legal custom or a particular royal regulation such as the mīšarum acts, which were issued in order to restore equity in times of crisis and restricted in time, or other royal decrees, which were probably restricted to a particular legal problem but not time sensitive. An excellent overview of various interpretations by various scholars is to be found in Graetz 2015: 254–260 and especially n. 74. I tend to follow Kraus (1979: 62) who interprets ṣimdat šarrim as “the present royal regulation” when used in the phrase ana/kīma/warki ṣimdat šarrim in legal and economic texts. The fact that there are specific references to the second and third ṣimdat šarrim of Rīm-Sîn seems to confirm this. See Kraus 1984: 31–50 and Renger 2002: 151–152. 87.  According to Jean, the tablets published in TCL 10 originate from Larsa, which is confirmed for the series AO 6346–6442–among which TCL 10 40 being AO 6394–by the merchant who sold them to the Louvre Museum (see TCL 10 p. 1). Kraus (1984: 31) also considers this tablet to originate from Larsa. However, in his “Renger Corpus Berlin” available at archibab.fr, Renger believes this contract to be from Ur instead of Larsa (“wohl nicht Larsa sondern Ur,” see http://pix.archibab.fr/4Dcgi/13937D4577.pdf). 88.  TCL 10 40 (RS 15/xii/–): 19–20: ištu pūt mātim īliluma u kunukkātum ittabkā. For translations and interpretations of this passage, see Kraus 1984: 32–33; CAD E sub elēlu 1b and ellu 1e, CAD P sub pūtu 1b2′ and CAD T sub tabāku 10a. 89.  See Kraus 1958, 1965, and 1984: 184–288; Olivier 1997; and Gaertner 2008.

68

Katrien De Graef

royal decree–which was resolved in the later decrees by adding the cancellation of all expropriation of property due to insolvency. However, the phraseology of this contract differs from that of later documents referring explicitly to a ṣimdat šarrim “royal decree,” which complicates the anal­ ysis of its meaning and does not rule out the possibility that it concerns a one-off settlement between individuals. The phrase “after the forehead of the land became clean and the sealed documents had been discarded” might simply refer to an earlier agreement between Ku-Ninšubur and Gaz-Sîn, as a result of which the date orchard was sold. Rīm-Sîn’s three official edicts, issued in his twenty-fifth, thirty-fourth, and before his forty-first year, all concern the alienation of urban real estate, orchards, and temple prebends. Kraus (1984: 42) suggested that this measure was taken because insolvent debtors had to cede their property to their creditors and thus lost the basis of their livelihood. As this became a widespread situation causing social imbalance, intervention was needed to re-establish the equilibrium by reinstating the debtors in their former properties. References to at least two of these edicts are found in the Ur texts, mostly claims, lawsuits, and verdicts in which, according to the royal edict, property had to return to the original owner (or his heirs), as he had been forced to sell his property to pay off his debts. TS22/22a (RS 28/viii/—) 90 states the return of one sar of a built house, adjacent to that of Ku-Ninšubur and that of Narām-ilīšu, from the children of Sâsiya to Sîn-imguranni, son of Ipquša, according to the royal edict (ana ṣimdat šarrim). Charpin’s study of the Sâsiya dossier revealed that Sâsiya owned the house since the sixth year of reign of Rīm-Sîn. 91 In other words, twenty-two years after Sîn-imguranni’s father Ipquša was forced to cede his house to Sâsiya, to whom he must have been indebted, Sâsiya’s children were forced by the royal decree to return the house to Ipquša’s son. This, however, does not seem to have taken place without difficulties, as Rīm-Sîn’s edict was issued in his twenty-fifth year of reign 92 but the house was returned only three years later to its rightful owner. It seems therefore plausible that Sâsiya’s children were sued by Sîn-imguranni and a trial preceded the actual return. Actually, this would not be the end, as, seven years later, on the occasion of Rīm-Sîn’s third royal edict, Sâsiya’s children would attempt to regain the one sar house from Sîn-imguranni (see infra). The fact that Sîn-imguranni is only able to reclaim his property after Rīm-Sîn’s first official edict (in RS 25), and not after his possible earlier edict (in RS 15) seems to indicate that there was in fact no earlier edict, or at least that it did not concern the alienation of urban real estate. UET 5 263 is an undated fragmentary tablet stating that, after having investigated the matter, the king returned a temple prebend to Kiag-madana from whom it had been taken by Appâ, son of Bēlî according to the royal edict. 93 Because the beginning of the text is lost, we know that Kiag-madana originally bought the 90.  See Charpin 1980: 29–30 and 214–215. 91.  See Charpin 1980: 28–34. 92.  According to Kraus (1984: 38) this edict was issued at the latest in the seventh month and before the nineteenth day of the eighth month of Rīm-Sîn’s twenty-fifth year of reign. 93.  UET 5 263 1′–12′: [mar.za a-ap-pa-a(?)] Iki.ág-⸢ma⸣.da.⸢na⸣ id-di-in-ma i-ša-am i-na ṣí-im-da-at šar-ri-⸢im⸣ Ia-ap-pa-a dumu be-lí-i mar.za i-ki-im-šu!-ú šar-ru-um wa-ar-ka-ta-ša ip!-ru-is-([x?])-ma ša mar.za [a-ap-pa-a(?)] mar.za ⸢gá.nun.maḫ(?)⸣ um.e 10.kam mar.za tuk.lú.lú šar-ru-um a-na ki.ág-ma.

Bad Moon Rising

69

prebend, but it is not clear from whom: this might have been Appâ, to whom it was returned after the royal edict, but it also might have been another person, whose legal heir was Appâ. Ellis (1972: 79) believes this to be a case in which the king reversed an earlier decision made on the basis a ṣimdat šarrim, which causes some difficulties in interpreting ṣimdat šarrim as a royal decision (or royal edict) in her opinion. In my opinion, the king did not overrule his own edict, as a result of which the interpretation of a ṣimdat šarrim is not affected. From this and other texts, it seems clear that these returns of property and prebends were not applied automatically: people had to go to court and sue their former creditors. It is obvious that the claimant had to produce proof that hardship was the cause of the sale and that this proof had to be evaluated by judges. In the present case, if Appâ was the original seller, he would have retrieved his prebend by trial, invoking the royal decree. Kiag-madana would then have appealed to a higher authority, the king, who—on the basis of more or better evidence?—would have annulled the first judgement, no doubt by stating that his edict was not applicable here, and returned the prebend to him. This means that the king would not have overruled his own edict. The tablet can be dated during Rīm-Sîn’s reign, as one of the witnesses, Namti-­ nibani, as well as Kiag-madana, also appear in UET 5 194 (RS 19/xii/—), stating the sale of a temple prebend of Amurru bought by Kiag-madana from Qīštum and Sîn-imguranni, sons of Ur-Ningirsu. 94 The edict referred to in UET 5 263 is likely to have been Rīm-Sîn’s first official one, which would allow us to date UET 5 263 post RS 25. TS 24 (RS 34/vi/—) 95 is the equal division by the children of Sâsiya and Sîn-­ imguranni  96 of Ipquša’s orchard “as much as they (i.e. the judges) reverted according to the royal edict.” 97 In other words, at some point in time, Ipquša, Sîn-imguranni’s father, was forced to cede an orchard to Sâsiya, to whom he must have been indebt­ed. After Rīm-Sin’s second official edict, Sîn-imguranni reclaimed his property, but the judges decided that the orchard, more precisely as much of it as was concerned by the edict, should be divided between both parties, viz. Sîn-imguranni and the children of Sâsiya (see infra). Again, this is clearly the outcome of a trial in which both parties are deemed to have equal rights on the contested property. UET 5 253 (RS 35/xii/—) relates how Šēp-Sîn and Ili-imguranni approached the judges of Ur and Larsa concerning the sale of two plots of land, a one bur orchard with date palms and a one bur empty lot, by Sîn-iqīšam, son of Dān-Lulal, to Iškur-gugal, son of Ilšu-bani, according to the second royal edict (ina ṣimdat šarrim warkītim), 98 referring as such clearly to Rīm-Sîn’s second official edict. A part of the tablet is broken, and the relationship between the plaintiffs on the one hand, and the buyer and seller on the other, is therefore not clear. 99 da.na ú-te-er (transcription based on the “Renger Corpus Berlin” available at archibab.fr and the photos of the tablet available at https://cdli.ucla.edu/dl/photo/P415148.jpg). See also Kraus 1984: 49–50. 94.  See Kraus 1984: 50. In his “Renger Corpus Berlin” available at archibab.fr, Renger dates UET 5 263 between RS 15 and RS 30, based on the attestation of Namti-nibani. 95.  See Charpin 1980: 30 and 215–216. 96.  TS 24 5–6 and TS 24a 6–7: mitḫāriš izzuzzū “they divided equally,” see Charpin 1980: 215–216. 97.  TS 24a 2–3: mala ana ṣimdat šarrim ušēṣūnim, see Charpin 1980: 30 and 215–216. 98.  CAD Ṣ sub ṣimdatu 1a. 99.  It would only make sense for Šēp-Sîn and Ili-imguranni to contest the ownership of the two plots of land Iškur-gugal bought from Sîn-iqīsam if they were Sîn-iqīšam’s heirs (and thus his sons?)

70

Katrien De Graef

It seems, however, that these edicts incited people to claim property even if they had no right to it. Indeed, some of the texts from Ur do not explicitly refer to a ṣimdat šarrim issued by Rīm-Sîn but can be linked to one, such as UET 5 252 (RS 35/v/—), in which the judge confirmed before witnesses the right of Ea-gāmil and his brother Apil-kittim to detain and possess ½ sar of a built house. 100 In other words, after Rīm-Sîn’s second official edict, Ea-gāmil and Apil-kittim, brothers of Ku-Ningal (see infra), were sued concerning the possession of ½ sar of a built house, but they were determined to be in the right. The same goes for UET 5 124 (RS 35/xi/—), in which Sîn-ereš, son of Annanesag, was able to obtain compensation for a house that his father had to sell earlier to Ku-Ningal. Ku-Ningal’s children did not want to return the original plot, as it was now part of a larger house, but gave Sîn-ereš a larger plot elsewhere. 101 Another example is TS 25 (RS 35/xii/30), in which the children of Sâsiya attempt­ed to regain real estate and various temple prebends from Sîn-imguranni, son of Ipquša, and Sîn-uselli, son of Šumi-abum. 102 The text concerns one sar of a built house, adjacent to that of Ku-Ninšubur and that of Narām-ilīšu, which Sâsiya’s children had to return to Sîn-imguranni, son of Ipquša, seven years and four months earlier, as a consequence of Rīm-Sîn’s first official edict, as stated in TS 22/22a (RS 28/viii/—), and three temple prebends, of respectively five, sixteen, and eight days per year in the temples of Nanna, Ninlil, and Gula. Sâsiya’s children pressed charges against Sîn-imguranni, son of Ipquša, (about the one-sar house) and Sîn-uselli, son of Šumi-abum, (about the three temple prebends) but the royal judges of Larsa and Ur disallowed their claims. Remarkably, the case states that the judges of Larsa and Ur removed them from the house, prebends, and orchard, 103 whereas there is no mention of an orchard in the claim. As Charpin (1980: 32) already remarked, this must have been the orchard mentioned in TS 24 (RS 34/vi/—), stating the equal division by the children of Sâsiya and Sîn-imguranni of Ipquša’s orchard “as much as they (i.e. the judges) reverted according to the royal edict” (see supra). It is clear that the dispute between Ipquša’s son, Sîn-imguranni, and Sâsiya’s children went back a while. Due to his insolvency, Ipquša was forced to sell an and Sîn-iqīšam had been forced earlier to sell his plots to Iškur-gugal due to insolvency. However, we do not know when Iškur-gugal bought the plots of land from Sîn-iqīsam, nor do we know whether Šēp-Sîn and Ili-imguranni were related to Sîn-iqīsam. Iškur-gugal, son of Ilšu-bani, is attested in three other texts from Ur. In UET 5 144 (RS 08/vii/–), he bought ½ sar nine gín of a built house from Ur-ḫegal, son of Sîn-nūr-māti; the house he bought is adjacent to another house of his, implying he was expanding his property. In UET 5 145 (RS 09/viii/–), he bought one sar of a built house from Sîn-pater, son of Sîn-iqīsam. About a year later, he also expanded this part of his property by buying another 1 1/3 sar of a built house from Sîn-pater, son of Sîn-iqīsam, as recorded in UET 5 146 (RS 10/v/–). In other words, about twenty-five years before Šēp-Sîn and Ili-imguranni went to court to contest the ownership of the plots of land Iškur-gugal bought, the same Iškur-gugal was expanding his property. However, given the fact that Rīm-Sîn issued a first official edict concerning the alienation of property in his twenty-fifth year of reign, and Šēp-Sîn and Ili-imguranni, not Sîn-iqīsam, contested the ownership of the plots only after Rīm-Sîn’s second official edict in his thirty-fourth year of reign, it seems probable that Iškur-gugal bought the later disputed plots of land from Sîn-iqīsam between RS 25/viii/20 and RS 34. See also Ellis 1972: 75 n. 11–12 and 80; and Kraus 1984: 44–45. 100.  See Charpin 1986a: 72–74. 101.  See Charpin 1986a: 72–74. 102.  See Charpin 1980: 30–34 and 216–217. 103. TS 25a 15–18: di.ku5.e.ne lugal ša larsaki ù úriki.ma ina é mar.za ù kiri6 issuḫūšunūṭi, see Charpin 1980: 217.

Bad Moon Rising

71

orchard and one sar of house to Sâšiya. In RS 28, after Rīm-Sîn’s first official edict, Sîn-imguranni recovered the one-sar house. In RS 34, after Rīm-sîn’s second official edict, the judges decided that the orchard should be divided equally between Sîn-imguranni and the children of Sâsiya. However, one and a half year later, the children of Sâsiya tried to regain the house and the orchard, based on the same second official edict of Rīm-sîn, but this attempt was unsuccessful. This indicates how fundamental the measures must have been and how families and individuals were affected by the royal edict. Another example is PBS 8/2 264 (RS 35/xii/20), 104 a lawsuit in which Lu-dingira claimed two temple prebends, both of twenty days per year in the temples of Bawa and Nergal, from the children of Waqar-abušu. The latter came into possession of the prebends when he was adopted by Ur-Ningirsu. Lu-dingira claimed that this adoption had been invalidated, as a result of which Waqar-abušu, and after him his heirs, were no longer entitled to Ur-Ningirsu’s prebends. However, Lu-dingira’s claim was declined by the servants of the king and the judges, who appointed the heirs of Waqar-abušu as the rightful owners, because, contrary to Lu-dingira’s assertion, the adoption had never been invalidated. The relation between the plaintiff, Lu-dingira, and the original owner of the prebends, Ur-Ningirsu, is not clear. However, Lu-dingira was clearly confident that he was entitled to Ur-Ningirsu’s prebends if the latter had not adopted, or had unadopted, Waqar-abušu, implying he must have been biologically related to Ur-Ningirsu. Lu-dingira’s patronymic is not mentioned in the text, and the legend on his seal, although very fragmentary, seems to suggest his father’s name started with dingir. 105 Lu-dingira might have been Ur-Ningirsu’s nephew or grandson. After his conquest of Larsa, Ḫammurabi issued an edict in which debts and sales were annulled. 106 According to Charpin (2000a: 188), Hammurabi issued this edict in the middle of his thirty-second year of reign, on the occasion of his accession to the throne in the south—he did not simply conquer the kingdom of Larsa but succeeded Rīm-Sîn as king. This edict applied throughout the kingdom, including the city of Ur. There are, however, no preserved documents from Ur referring to this edict. His successor Samsu-iluna did the same on the occasion of his accession to the throne 107 and in his eighth year of reign, four years before Ur would be abandoned. The latter is one of the few actual royal decrees that have been preserved, albeit in a fragmentary state. 108 There are, again, no preserved documents from Ur referring to Samsu-iluna’s edicts, which is not surprising at all, given the political crises and revolts that took place in southern Mesopotamia at the time. Apart from the economic and legal documents explicitly referring to a royal edict or those that can be linked to one based on their content and date, another genre 104.  See Charpin 1986a: 169–172. 105.  See Charpin 1986a: 170–171 and 173 n. 2. 106.  See Kraus 1984: 58–62; Charpin 1991 and 2000a: 187–189. 107.  See Charpin 1988 and Charpin 2000a: 189–190. Samsu-iluna’s first edict is commemorated in his second year name: mu sa-am-su-i-lu-na lugal.e še.ga dingir gal.gal.e.ne amar.ar.gi ki.en.gi ki.uri.(a) i.ni.in.gar.ra šà ma.da du10 mu.ni.in.du10.du10 nì.si.sá pa.è bí.in.ak “The year Samsu-iluna, the king, the favourite of the greatest gods, established the freedom of Sumer and Akkad, made the heart of the good land very glad and made justice manifest,” see Horsnell 1999: 177. 108.  See Kraus 1965, 1984: 154–159; and Hallo 1995.

72

Katrien De Graef

of texts is linked to and can give information on royal edicts, viz. loan contracts. Indeed, as already mentioned above, loan contracts are by far the most commonly preserved records from the Old Babylonian period. Charpin (2000a and 2001) was able to show that there was a peak in the frequency of preserved loan documents in the years preceding a royal edict, showing that loans that were declared null and void by the royal edict, and as a result were not paid back, were kept by the creditors in their archives. If we take a look at the loan documents from Ur, we see a peak of sixteen loan documents that were retrieved from Dumuzi-gamil’s house. 109 Half of them are dated to Rīm-Sîn’s thirty-third year, four to his thirty-second year, three to his thirty-fourth year and one to his thirty-first year. Is it possible that these loans had been cancelled by Rīm-Sîn’s second royal edict, and were therefore kept by Dumuzi-­ gamil, as Van De Mieroop (1992a: 118) already suggested? According to Kraus (1984: 45), the latest possible date for Rīm-Sîn’s second edict is the eleventh day of the first month of his thirty-fourth year. Dumuzi-gamil’s archive, however, contains three loans dated respectively in the fifth, tenth, and eleventh months of RS 34. If Kraus’s hypothesis is correct, and Rīm-Sîn’s second royal edict was indeed issued at the very beginning of his thirty-fourth year, these three loans could not have been cancelled by the edict. Still, they were kept by Dumuzi-gamil, implying that they had not been paid back. How can we explain this? Charpin (2000a: 197) came across the same phenomenon in two dossiers of pre­served loans prior to Samsu-iluna’s second edict, that of Awilīya, rāʾibānum, and that of Marduk-muballiṭ of Lagaba: both dossiers contain one loan document that can be dated after Samsu-iluna issued his second edict in the third month of his eighth year of reign. 110 The same phenomenon can be observed in the archive found in the house of Ur-Utu during the Belgian excavations at Tell ed-Dēr (Sippar-Amnānum), 111 where we see an enormous peak of preserved private loans prior to Ammi-ṣaduqa’s first edict on the occasion of his accession to the throne, after which the number of preserved, and thus unpaid, loans goes right back to “normal.” 112 Charpin (2000a: 197, n. 44) suggested the possibility that the creditors had new, fictitious loan documents drawn up after the edict with the consent of the debtors in order to recover those cancelled by the edict. This is of course possible—although contrary to the very purpose of the edict—but maybe the explanation is simpler: whereas the edict did indeed solve acute economic problems—all outstanding debts were cancelled, giving people some breathing room for a while—it obviously could not solve the structural problem, and soon enough a new crisis occurred. In the case of Dumuzi-gamil, it seems that it took less than half a year before people were not able to repay their debts again, which explains why Rīm-Sîn had to issue a third official royal edict rather soon, viz. before his forty-first year of reign. 109.  UET 5 365 (RS 31/viii/—), 311 (RS 32/vii/—), 312 (RS 32/ix/—), 347 (RS 32/xi/30), 348 (RS 32/xii/—), 349 (RS 33/ii/30), 350 (RS 33/iii/—), 351 (RS 33/iii/—), 313 (RS 33/vii/13), 352 (RS 33/vii/30), 314 (RS 33/x/21), 315 (RS 33/x/21), 353 (RS 33/xii/—), 363 (RS 34/v/—), 354 (RS 34/x/—), and 317 (RS 34/xi). All these tablets were found in AH, 3 Niche Lane, room 3, see Van De Mieroop 1992a: 275–277. For the well-known entrepreneur from Ur called Dumuzi-gamil, see Van De Mieroop 1992a: 132–136. 110.  See Charpin 2000a: 191–192 and 194–197. 111.  See Gasche 1989. 112.  A study on the loan documents found in the archive of Ur-Utu by M. Tanret and myself is forthcoming.

Bad Moon Rising

73

It is, however, remarkable that there are no later loans preserved from Dumuzi-­ gamil, whom we know to have been active from RS 27 to RS 36. 113 Is it possible that Rīm-Sîn’s second edict was issued later, at the end of his thirty-fourth or the start of his thirty-fifth year of reign? In that case, the three loans from RS 34 would have been cancelled by this edict. Kraus’ hypothesis to date Rīm-Sîn’s second edict at the very beginning of his thirty-fourth year is based on YOS 8 141 (RS 34/i/11), a text from Larsa recording the donation of property by Ṣilli-Aḫua to his wife Burtum, who lost her inheritance rights as a result of a family dispute. The property is said to have been reverted for the second time by Ṣilli-Aḫua on the orders of the king after the death of his father. 114 As Kraus (1984: 44) already pointed out, it is not at all certain that this refers to Rīm-Sîn’s second edict, for it is not known when this event took place, and how long before this the text was written. Moreover, if Rīm-Sîn’s edict was issued at the beginning of his thirty-fourth year, Ṣilli-Aḫua’s claim would have been settled within ten days, which seems rather improbable. It is, therefore, in my opinion, unlikely that this text refers to Rīm-Sîn’s second edict: it does not refer to an edict (ṣimdatum) but to the orders of the king (awat šarrim), hinting rather at a lengthy process concerning the inheritance of Ṣilli-Aḫua’s father in which the king ulti­mately passed judgement, especially as it also states that his brothers have no claims against him. 115 Given the fact that the above-mentioned court cases in order to reclaim property all took place during Rīm-Sîn’s thirty-fifth year of reign, it seems indeed possible that Rīm-Sîn’s second edict was issued at the end of his thirty-fourth or the start of his thirty-fifth year of reign. The only document implying an earlier date of issue is TS 24, recording the equal division of Ipquša’s orchard between the children of Sâsiya and Sîn-imguranni and referring explicitly to a royal edict, which is dated in the sixth month of RS 34 (cf. supra). However, it is not inconceivable that the edict referred to in TS 24 is actually Rīm-Sîn’s first royal edict, and that it took the children of Sâsiya and Sîn-­ imguranni almost nine years of litigating to come to a mutual agreement concerning the orchard of Ipquša. As such, the children of Sâsiya tried to regain the house and orchard, which were returned to Sîn-imguranni and equally divided between both parties after Rīm-Sîn’s first royal edict, on the basis of Rīm-Sîn’s second edict, which would make much more sense indeed. Another problem to be tackled is the fact that all but four loans issued by Dumuzi-­gamil were short-term loans. Apart from four šu.lá loans—which were in all probability commercial loans and therefore not qualified for cancellation on the basis of a royal edict 116—which were due in two months or after the end of the expedition, 117 all but one of the loans issued by Dumuzi-gamil were due within 113.  See Van De Mieroop 1992a: 275–278. 114.  YOS 8 141 22–24: ša Ṣilli-Aḫua warki abīšu šinīšu ina awat šarrim ušēṣûma, see Kraus 1984: 43–44. 115.  YOS 8 141 25–26: aḫḫūšu mimma elīšu lā išû. 116.  For šu.lá loans being commercial, see Skaist 1994: 41–51. In the edict of the late Old Baby­ lonian king Ammi-ṣaduqa, a distinction is made between private and commercial loans. Only the private loans, those given for consumption or as soft loans, which produce interest for the creditor, are cancelled by the decree. See Kraus 1958 and 1984: 168–288. 117.  UET 5 365 (RS 31/viii/—) due in two months (iti.2.kam), and UET 5 313 (RS 33/vii/13), 314, and 315 (both RS 33/x/21), due after the end of the expedition (kaskal.(ta).silim.ma.(bi)), UET 5 314

74

Katrien De Graef

one month. 118 This means, that whenever the edict was issued, it would only have affected one or two actually outstanding loans. In other words, the edict must have concerned not only the outstanding loans, but also those that were expired but not yet paid back. Charpin (2000a: 203) already noted that the peak of preserved loan documents before a royal edict did not only contain loans concluded in the months preceding the edict but also those that went back several years and were long ex­ pired when the edict was issued. This accumulation of unpaid loan documents, starting years before and reaching its peak in the year preceding the royal decree, can only be explained if we assume that the royal decree was issued because of this accumulation. When too many loans were not repaid, and people undoubtedly stacked debt on debt, resulting in the sale of their property, the loss of their livelihood, or even debt bondage, the authorities could not but intervene to get the economy up and running again. As such, not only the debtors, but also the creditors benefitted from the royal decree: the cancellation of the outstanding debts prevented the economy from complete stagnation or even collapse. The reason why these invalid loan documents were kept by the creditors is as yet unclear. We would expect them to be broken, as is shown by CT 48 15, a legal text from Sippar, dated to the thirteenth year of reign of Ḫammurabi, stating that the loan document concerning 1/3 mina of silver that Nuṭṭuptum, daughter of Warad-ilīšu, loaned, was lost. and could therefore not be broken according to the royal edict. In order for the debtors not to risk being asked repayment later on, a clod of earth was broken instead. 119 Charpin (2000a: 187, 197, and 204) quotes two other texts showing that loan documents cancelled by a royal edict were supposed to be broken. The first one is a tablet from Larsa (Anbar and Stol 1991 no. 4, Ḫa 32/i/—) stating that the document concerning the loan of four mina of silver that Apil-Kittim lent from Ibbi-Sîn could not have been broken as it had not been stored, but is consigned to be broken, (as) it dates from RS 59. 120 According to Charpin (2000a: 197) this tablet shows that Ḫammurabi’s second edict was, although only issued in the middle of his thirty-second year, already applicable from the first month. This seems rather unlikely, especially as AUCT 4 80 records the partial repayment of a loan in the fourth month of Ḫa 32. 121 Why would a debtor repay a part of his debt to his creditor when the royal edict was already applicable? Moreover, there is no reason to assume that the breaking of the loan document mentioned in the above quoted Larsa tablet is related to a adding within thirty days (iti .30.kam). 118.  UET 5 312 (RS 32/ix/—) does not mention when the loan has to be paid back. UET 5 311 (RS 32/vii/—), 347 (RS 32/xi/30), 348 (RS 32/xii/—), 349 (RS 33/ii/30), 350 (RS 33/iii/—), 351 (RS 33/iii/—), 352 (RS 33/vii/30), 353 (RS 33/xii/—), 363 (RS 34/v/—), 354 (RS 34/x/—), and 317 (RS 34/xi) state that they are to be paid back on the thirtieth day of the month (iti u4.30.šè or iti u4.30.kam). 119. CT 48 15 5b–11: ina ṣimdat šarrim ṭuppam ana ḫepîm ul iddin ḫaliq iqbīma kīma ṭuppi kirbānam ḫepê “she did not give the tablet to be broken according to the royal decree, ‘it is lost’ she said ‘break a clod instead of the tablet’,” see Charpin 2000a: 204 and 2000b: 89–90. The edict referred to in this text was Ḫammurabi’s second edict, issued in his twelfth year of reign. According to Kraus (1984: 56) the latest possible date for Ḫammurabi’s second edict was the ninth month of his twelfth year of reign. 120.  Anbar and Stol 1991 no. 4 6–12: ṭuppu ul qurrumma ul iḫḫepi ṭuppu ana ḫepê nadi ša mu.30. kam ì-si-inki ba.an.díb. 121.  AUCT 4 80 1–7: 1 gín kù.babbar šu.ti.a Lipit-Amurru ki Sîn-uselli šà.ba 5 2/3 gín kù.babbar ša elīšu išû.

Bad Moon Rising

75

royal edict. What happened and why was it drawn up? In RS 59, Ibbi-Sîn lent four mina of silver to Apil-Kittim. When Apil-Kittim paid his loan back, Ibbi-Sîn was not able to provide the loan document in order to (give it to the debtor to) destroy it. At that point, a tablet similar to CT 48 15 must have been drawn up, stating that although Apil-Kittim paid back his loan, the document was lost and could not be broken, in order to safeguard Apil-Kittim against later claims of repayment should the original loan document turn up. About two years later, in the first month of Ḫa 32, the original loan document did turn up, but as the loan was already repaid, and in line with the earlier document, the tablet was now consigned to be broken. 122 The second one is BE 6/1 103 (Aṣ 01/x/03), a tablet from Sippar stating a sequence of claims and statements concerning sixty gur of barley that Warad-Sîn, son of Eṭerum, was loaned by Gimillum, the servant of Marduk-mušallim, son of UtulIštar, who was replaced 123 by Iluni, son of Sizzatum. Although the king issued an edict, cancelling Warad-Sîn’s debt, the creditor and his representative insisted on getting their barley back. When Warad-Sîn argued that he on his turn lent out the barley, which had also been cancelled by the king’s edict, Gimillum and Iluni took barley that was heaped up in Warad-Sîn’s house. Warad-Sîn went to the judges in Babylon, was determined to be in the right, and brought the tablets of the judges of Babylon to those of Sippar, after which Iluni accused Gimillum of having stolen the barley from Warad-Sîn’s house. After further examination of their case, they came to a mutual agreement. Finally, it is stated that should the sealed document concerning the sixty gur of barley that Warad-Sîn, son of Eṭerum, was loaned, appear in the basket of Marduk-mušallim, son of Utul-Ištar, it is invalid. 124 Note that this tablet does not mention witnesses, implying it was a draft or a copy of the original. It seems thus, as Charpin (2000a: 204–205) already noted, that the legal status of loan documents was similar to that of title deeds. When a seller of a property could not produce all the necessary title deeds that had to be handed over to the new owner according to the principle of the “chains of transmission,” 125 a document had to be drawn up listing the missing tablet(s) in order to replace it/them. The same goes for loan documents: if the creditor was not able to produce a loan document that needed to be broken, the debtor was given a document stating the original loan document had not been broken, but was to be considered invalid and was to be broken whenever it turned up. This would also explain the presence of UET 5 361 and 126 in Dumuzi-gamil’s archive. 126 UET 5 126 (RS 32/iii/—) records that Dumuzi-gamil paid back his half of the one mina of silver he borrowed together with Šumi-abīya from Šumi-abum five years earlier, including interest for five years, to Nūr-ilīšu and Sîn-ašarēd. UET 5 361 (RS 27/ii/03) records the loan of one mina of silver by Dumuzi-gamil and Šumi-­ abīya from Šumi-abum and the respective amount both have to pay back, viz. ½ 122.  Both Anbar and Stol (1991: 17) as Charpin (2000a: 187 n. 12) interpret ana ḫepê nadi as a future tense (“sera donnée pour être brisée” and “est destinée à être brisée”). However, nadi being a stative, it seems more appropriate to translate “it is now being consigned.” 123.  See Tanret and Janssen 1992 for the interpretation of ana qabê as “replaced by.” 124.  BE 6/1 103: 40–42: kanīk 60 še.gur ša Warad-Sîn dumu Eṭerum ina gipisan Marduk-muššalim dumu Utul-Ištar illiam ḫepi. Charpin (2000a: 204) translates ḫepi ‘il sera brisé’. Being a stative, it seems more appropriate to translate ‘it is broken’ and understand this in a figurative sense ‘it is invalid’. 125.  See Charpin 1986b. 126.  See Leemans 1955: 117–118 and Van De Mieroop 1992a: 132–133.

76

Katrien De Graef

mina 7 1/6 sheqel and ½ mina 9 sheqels. The fact that Šumi-abīya’s interest rate is higher implies a longer term of repayment and explains the existence of UET 5 126: the loan document—originally kept by Šumi-abum, but handed over to Nūr-ilīšu and Sîn-ašarēd who had taken over the loan—could not be broken, as Šumi-abīya’s part was not yet repaid, but UET 5 126 provided Dumuzi-gamil with the necessary proof his part was repaid. Note that Dumuzi-gamil paid back his own debt in RS 32, whereas four of his debtors were not able to do so, corroborating the phenomenon of the accumulation of unpaid debts, starting years before but reaching its peak only in the year preceding the royal decree. This shows that in special circumstances, when loans were lost or otherwise unavailable for breaking, documents were drawn up ensuring the rights of the debtors, whereas in “normal” circumstances we can sometimes observe that rather large groups of cancelled loans would simply have been kept in the creditor’s archive. As is shown by YOS 5 58, a cylindrical tag (possibly from Ur according to cdli), to be dated to the eleventh year of Warad-Sîn, tablets that had lost their value were collected in baskets in order to be broken. 127 In the case of YOS 5 58, these were accounting tablets, such as receipts (mu.DU) and arrears (lál), whose data had probably been incorporated into summary lists, and therefore had lost their value. Charpin (2000a: 205) argues that the same may have happened for loans cancelled by a royal edict: creditors would have stored their loan documents in baskets in order to be destroyed, which eventually did not happen. The debtors were safeguarded against claims of repayment by the pre-edict date of the loans, the creditors must have kept them for archival and bookkeeping reasons. We do not know whether this was a general rule or whether loans were eventually broken or discarded. But, as Charpin (2000a: 204) noted, the main difficulty in providing a clear explanation of the phenomenon is that we are dealing mainly with tablets from illicit excavations. Without archaeological context, it is hard to know whether they were actually kept in the archive of the creditors, were buried under the floor of a room, 128 or were disposed of in another way. However, the few Old Babylonian private archives that were found during licit excavations and for which archaeological context is available seem to show exactly the same peak of loan documents prior to a royal decree. This goes for the tablets found during the Iraqi excavations at Abu Ḥabbah (Sippar-Yaḫrūrum) 129 in the late 70s and early 80s in room 2 of house 11 (level III of U 106), among which a group of (at least) twenty loan documents, belonging to the archive of Ḫumṭi-Adad and Tarīb-ilīšu, children of Ili-kīma-abīya, the majority of which are to be dated in the sixth and seventh years of reign of Samsu-iluna, prior to his second edict (in Si 8). 130 It is, however, not entirely clear where exactly these tablets were found—in a 127.  YOS 5 58 1–3: gipisan im.sar.ra siḫiātum zi.re.dè “basket with invalid tablets, to be destroyed,” see Veenhof 1986: 18 and Charpin 2000a: 205. 128.  Loan documents have been found under the floor in Ur-Utu’s house at Sippar-Amnānum (see Tanret 2004) and in Igibuni’s house at Susa (see De Graef 2008). 129.  See al-Jadir and Abdullah 1983. The Old Babylonian tablets found in private houses during the Iraqi excavations at Abu Ḥabbah in the late 70s and early 80s are published in al-Rawi and Dalley 2000. 130.  Charpin (2001) enumerates nineteen loan documents, four in which Tarīb-ilīšu (aka Tarībīya) is creditor: nos. 16 (Si 7/iii/26), 17 (Si 7/iii/26), 14 (Si 7/ix/23), and 20 (Si 7/xi/—), and fifteen in which Ḫumti-Adad (aka Ḫumṭīya) is creditor: nos. 58 (Si 1?/[…]/[…]), 34 (Si 4/xi/[…]), 10 (Si 6/iii/01), 7 (Si 6/ iii/25), 11 (Si 6/vii/22), 23 (Si 6/viii/[…]), 42 (Si 6/x/09), 31 (Si 7/[…]/[…]), 9 (Si 7/iii/01), 35 (Si 7/iii/14),

Bad Moon Rising

77

(broken) jar? 131—and if they were found together, as the catalogue in al-Rawi and Dalley (2000: 27–33) does not mention a findspot for every tablet, 132 which makes it difficult to reconstruct their archival context. As mentioned earlier, this also goes for the archive found in the house of Ur-Utu, where we see a peak of preserved private loans prior to Ammi-ṣaduqa’s first edict on the occasion of his accession to the throne. In other words, creditors indeed tended to keep the loan documents that were cancelled by a royal decree, as they no doubt wanted to keep their accounts, and book the loss they suffered due to the cancellation of all outstanding debts. The total amount of more than 330 unpaid and/or cancelled loan documents that Ur-Utu kept in his archive seems to indicate that, although the intention might have been to destroy or break them once they were booked as loss, this never actually happened. The reason is obvious: archives were only sorted out on the occasion of a move or renovation of a house. Moreover, although the creditors kept the loan documents, the debtors were safeguarded against claims of repayment by the pre-edict date of the documents, which made them de facto invalid. One might therefore wonder whether ḫepûm is always to be understood in its literal sense “to break.” It is clear that in some contexts ḫepûm is to be understood in a figurative sense “to be invalid:” Ammi-ṣaquda’s edict (§3) states that when an Akkadian or Amorite has lent barley or silver for interest (…) and had a tablet drawn up, his tablet is void (ḫepi), and he cannot have barley or silver collected according to the wording of this tablet. 133 The same goes for the above-mentioned BE 6/1 103 and CT 4 42a, an adoption contract from Sippar, mentioning at the end, after the list of witnesses: “after Sumu-la-ila voided the tablets” (ištu Sumu-la-ila kunukkāti iḫpû), where it is difficult to suppose that the king broke all the tablets. In other words, the following preliminary conclusions can be drawn. Loans that were paid back disappeared from the creditor’s archive, as there was no need to keep them for accounting purposes. It seems logical that debtors received the loan documents upon repayment, after which they could be broken or discarded. Loans that were expired but not paid back, even though they were cancelled by a royal edict, were kept by the creditor for accounting purposes. Loan documents that were lost and could not be given to the debtor upon repayment in order to be broken, or could not be kept by the creditor after cancellation by a royal edict, were replaced 30 (Si 7/iii/29), 8 (Si 7/iii/30), 21 (Si 7 iv/11), 36 (Si 7/xi/12), and 32 (Si 7/xii-bis/02). To this last group should be added no. 22 (Si 7/iii/12). The same goes for nos. 24 (Si […]/viii/23) and 37 (no date), which are probably to be dated between Si 1 and 7, and possibly for no. 47, the name of the creditor of which is unfortunately lost, but the date of which (Si 7/ix/03) matches perfectly. 131.  See al-Rawi and Dalley 2000: 5, “Some or all these tablets with seal impressions were found in a jar.” In their catalogue, only one tablet (no. 38) is said to have been found in “level III house 11 room 2 in a broken jar.” 132.  According to the catalogue in al-Rawi and Dalley 2000, only nos. 7, 10, 14, 16, 17, 20, 21, 22, 23, 24, 30, 31, 35, and 37 are found in room 2 of house 11. For nos. 8, 11, 32, 34, and 42, no findspot is given. Moreover, no. 58 is said to have been found in room 26 of area 3. 133. See Kraus 1984: 170–171: [ša š]e’am u kaspam ana lú akkadî ù lú amurrî […] ⸢máš⸣ ulu ana melqētim […] iddinuma ṭuppam ušēzibu aššum šarrum mīšaram ana mātim iškunu ṭuppašu ḫepi “Wer Gerste oder Silber einem Akkader oder Amurräer [als Darlehen, auf] Zins oder zur ‘Entgegennahme’ […] ausgeliehen hat und (sich darüber) eine Urkunde hat ausstellen lassen—weil der König Gerechtigkeit für das Land wiederhergestellt hat, ist seine Urkunde hinfällig; Gerste oder Silber kann ern ach dem Wortlaute eben der Urkunde nicht eintreiben (lassen).” Kraus translates ḫepi by “hinfällig,” invalid, not broken.

78

Katrien De Graef

by another document stating that should the original document reappear, it was invalid. Going back to the main subject of this paper, the city of Ur, we witness an accumulation of internal crises during the first centuries of the second millennium BCE, leading up eventually to a structural crisis in the second half of the eighteenth century, which caused massive emigration to the north and an, at least partial, abandonment of the south for a long period of time. If Lipit-Ištar’s edict mentioned in the prologue of his law collection is to be understood as the liberation of those living in debt-bondage, Ur lived through its first economic crisis less than fifty years after its recovery. Hereafter, there seems to have been a long century of relative economic stability. Economic troubles start again under the reign of Rīm-Sîn I, who had to intervene at least three (but possibly four) times in order to restore the economic balance. Here we see a clear cycle of internal economic crises that succeeded each other at a high speed. Various documents from private archives found at Ur concern claims in order to retrieve property that was lost due, in all probability, to enduring debts. It seems, moreover, that thirteen outstanding private loans, given by Dumuzi-gamil, were cancelled by Rīmsîn’s second edict. Things did not change for the better once Ur became part of the Babylonian kingdom of Ḫammurabi; on the contrary, the cycle of internal crises continued, but now at an even higher speed: Ḫammurabi issued an edict on the occasion of the fall of Larsa at the end of his thirtieth year of reign, but his son and successor, Samsu-iluna, deemed it necessary to issue an edict thirteen years later, when he took over the throne, and again eight years after that. Four years later, exogenous crises such as continuing raids, water supply problems, revolts, conquests, and reconquests resulted in a crisis in the accumulation regime and mode of régulation. It is important to stress that, although the royal edicts were no doubt issued to restore the means of self-support of the agrarian population, who ended up in debt-bondage, as a result of which the estates had a constant shortage of labour, 134 the edicts also affected the urban elites, who, although largely interconnected with the temple and the palace, and acting within the boundaries of this very estate system, were involved in a private commercial circuit, parallel to the estate system. This is apparent from the documents concerning the property of Ku-Ningal, Ipquša, and Sâsiya discussed above. Not only the individuals reclaiming property that had to be sold due to debts belong to the urban elites, the same goes for the creditors and debtors: quite often they seem to belong to the same high class of urban elites. Lending and borrowing seems, at least partly, to have been a means of investing. Indeed, as mentioned earlier, Dumuzi-gamil, well-known entrepreneur and creditor of Ur, borrowed ½ mina of silver from Šumi-abum. Renger (2002: 139–141 and 153–154) argues that royal edicts were so frequent in the Old Babylonian period because of the privatization by the palace of a part of its functions and the increasing role of private entrepreneurs. The frequency of the need for debt remission reflects the precarious economic situation created by the franchise system through which the palace outsourced various economic activities (the so-called “Palastgeschäfte”). The business operations entrusted to private entrepreneurs did not produce enough to pay the stipulated annual dues, as a result 134.  See Hudson 2002: 29–30.

Bad Moon Rising

79

of which arrears accrued that had to be remitted by the palace, which was dependent on the service of these entrepreneurs. The documents from private archives mainly inform us on how this precarious economic situation affected the urbanites, some of whom stacked debt on debt and had to sell property, and show once more how the parallel circuit of private economy was interconnected with and dependent on the estate system. However, whereas the cancellation of outstanding debts prevented the economy from complete stagnation and collapse, it was only a short-term solution: soon enough, debts began to accumulate again, resulting in a new need for debt remission to keep the economy going. By issuing royal edicts, the Old Babylonian rulers temporarily counteracted the symptoms, but were never able to solve the causes of the failing economic system. After the structural crisis in the second half of the eighteenth century, which caused massive emigration to the north and an, at least partal, abandonment of the south for a long period of time, internal crises kept accumulating in the north, eventually leading to an ultimate crisis in the middle of the second millennium BCE when the system collapsed completely and the Old Babylonian period came to an end.

Conclusions The régulation approach can certainly be used in the study of Mesopotamian economy. In our case, it opened up new ways in looking at the early second millennium BCE, especially as it reveals the internal dysfunctions inherent in the system, which caused it to ultimately implode. As such, our paradox is maybe less a paradox than we initially thought. First, our view is biased by our sources, which are overwhelmingly concerned with the estate system, whereas the subsistence agriculture in the rural villages and communities is almost completely absent, although they must have comprised the bulk of all socio-economic life. Moreover, our sources provide a warped picture of the economy, seen from the perspective of the ruling classes, written by and for the elites. Second, whereas the times were indeed turbulent, this was not the main prob­ lem. The main problem seems to be internal, namely the inability to adjust the economic system when necessary. As stated above, our textual sources paint a picture of a thriving community: commerce, sales, leases, and rents depict a prosperous economy, but only as long as it lasts, between crises. Each moment of crisis is shown by peaks in lending and is short-lived, leaving only subtle traces in our documentation, thanks to the royal intervention, but these crises return and are only bypassed, not cured. Internal cyclical crises are supposed to perform a useful function in managing and adapting the economic system, and restoring the economic balance. However, the royal edicts, cancelling debts and sales, do not restore the economic balance–at least not fundamentally. They soften the blow, but do not cure the disease. They treat the symptoms on a temporary basis, but do not solve the underlying problem. Once the immediate crisis has been averted, the same system starts over again. As a result, the crises repeat themselves over and over again, eventually leading to a structural crisis. Their inability to reconfigure the mode of régulation due to a lack of insight

80

Katrien De Graef

into their own economic system will eventually lead to a total collapse of the system in the middle of the second millennium.

Bibliography Adams, M. R. 1982 Property Rights and Functional Tenure in Mesopotamian Rural Communities. Pp. 1–14 in Societies and Languages of the Ancient Near East. Studies in Honour of I.M. Diakonoff, ed. M. Dandamaev. Warminster: Aris and Phillips. 1988 Contexts of Civilizational Collapse: A Mesopotamian View. Pp. 20–43 in The Collapse of Ancient States and Civilizations, eds. N. Yoffee and G. L. Cowgill. Tuscon: University of Arizona Press. Aglietta, M. 1979 A Theory of Capitalist Regulation: The US Experience. London-New York: Verso. Anbar, M. and Stol, M. 1991 Textes de l’époque babylonienne ancienne III. RA 85: 13–48. Boer, R. 2015 The Sacred Economy of Ancient Israel. Louisville: Westminster John Knox Press. Boyer, R. 2002 The Origins of Régulation Theory. Pp. 13–20 in Régulation Theory: The State of the Art, eds. R. Boyer and Y. Saillard. London and New York: Routledge. Boyer, R. and Saillard, Y., eds. 2002 Régulation Theory: The State of the Art. Updated edition of the original French Théorie de la régulation: l’état des savoirs from 1995; translated by C. Shread. London and New York: Routledge. Boyer, R. and Saillard, Y. 2002 A Summary of Régulation Theory. Pp. 36–44 in Régulation Theory: The State of the Art, eds. R. Boyer and Y. Saillard. London and New York: Routledge. Brusasco, P. 1999/2000 Family Archives and the Social Use of Space in Old Babylonian Houses at Ur. Mesopotamia 34–35: 3–173. Cagni, L. 1980 Briefe aus dem Iraq Museum (TIM II). AbB 8. Leiden: Brill. Charpin, D. 1980 Archives familiales et propiété privée en Babylonie ancienne: étude des documents de «Tell Sifr». HEO 12. Genève: Droz. 1983 Review of Briefe aus dem Iraq Museum (TIM II). (Altbabylonische Briefe, Heft 8) by L. Cagni. AfO 29/30: 103–108. 1986a Le clergé d’Ur au siècle d’Hammurabi (XIXe-XVIIIe siècles av. J.-C.). HEO 22. Geneva: Librairie Droz. 1986b Transmission des titres de propriété et constitution des archives privies en Babylonie ancienne. Pp. 121–140 in Cuneiform Archives and Libraries: Papers Read at the 30e Rencontre Assyriologique Internationale, Leiden, 4–8 July 1983, ed. K. R. Veenhof. PIHANS 57. Istanbul: Nederlands Historisch-Archeologisch Instituut. 1988 Le mîšarum à l’avènement de Samsu-iluna. NABU 1988/4: 52–53 no. 76. 1991 La mîšarum d’Hammu-rabi après la conquête de Larsa: un nouvel indice. NABU 1991/4: 71 no. 102. 2000a Les prêteurs et le palais: les édits de mîšarum des rois de Babylone et leurs traces dans les archives privées. Pp. 185–211 in Interdependency of Institutions and Private Entrepreneurs: Proceedings of the Second MOS Symposium (Leiden 1998), eds. A. C. V. M. Bongenaar and G. van Driel. PIHANS 87. Istanbul: Nederlands historisch-archeologisch instituut.

Bad Moon Rising 2000b Lettres et procès paleo-babyloniens. Pp. 69–111 and 249–250 in Rendre la justice en Mésopotamie: archives judiciaires du Proche-orient ancien (IIIe-Ier millénaires avant J.-C.), ed. F. Joannès. Saint-Denis: Presses universitaires de Vincennes. 2001 Les prêteurs et le palais (suite). NABU 2001/3: 51–52 no. 51. 2004 Histoire politique du Proche-Orient Amorrite (2002–1595). Pp. 25–480 in Die altbabylonische Zeit, eds. D. Charpin, D. O. Edzard, and M. Stol. OBO 160/4. Fribourg: Academic Press/Göttingen: Vandenhoeck & Ruprecht. 2015 Découvertes épigraphiques à Ur (octobre-décembre 2015). NABU 2015/4: 178 no. 109. Charpin, D., Edzard, D.O., and Stol, M. 2004 Die altbabylonische Zeit. OBO 160/4. Fribourg: Academic Press/Göttingen: Vandenhoeck & Ruprecht. Chiera, E. 1922 Old Babylonian Contracts. PBS 8/2. Philadelphia: University of Pennsylvania. Christian, J. B. 1969 Some Unpublished Old Babylonian Letters at the University of Chicago. Master’s thesis, University of Chicago. Cole, S. W. and Gasche, H. 1998 Second- and First-Millennium BC Rivers in Northern Babylonia. Pp. 1–64 in Changing Watercourses in Babylonia. Towards a Reconstruction of the Ancient Environment in Lower Mesopotamia, eds. H. Gasche and M. Tanret. MHEM 5. Ghent/Chicago: Ghent University/Oriental Institute. Crawford, H. 1998 Dilmun and Its Gulf Neighbours. Cambridge: Cambridge University Press. 2005 Mesopotamia and the Gulf: The History of a Relationship. Iraq 67: 41–46. Crawford, V. E. 1954 Sumerian Economic Texts from the First Dynasty of Isin. BIN 9. New Haven: Yale University Press. De Graef, K. 2008 Rest in Pieces. The Archive of Igibuni. Pp. 225–234 in The Growth of an Early State in Mesopotamia: Studies in Ur III Administration, eds. S. J. Garfinkle and J. C. Johnson. BPOA 5. Madrid: CSIC. Dercksen, J. G. 2014 The Date of Hammurabi’s Conquest of Larsa. NABU 2014/4: 128 no. 79. Diakonoff, I. M. 1975 The Rural Community in the Ancient Near East. JESHO 18: 121–133. 1982 The Structure of Near Eastern Society before the Middle of the 2nd Millennium BC. Oikumene 3: 7–100 1985 Extended Families in Old Babylonian Ur. ZA 75: 47–65. 1988 Three Ways of Development in the Ancient Oriental Society. Pp. 1–8 in Stato, economia, lavoro nel Vicino Oriente antico. Atti del convegno dell’Instituto Gramsci Toscano, Firenze, 29–31 October 1984, ed. A. Zanardo. Milano: Francoangeli. 1990 люди города Урa, Москва: Наука. [Lyudi goroda Ura, Moscow: Nauka]. 1995 Old Babylonian Ur. JESHO 38: 91–94. Dight, R. J. W. 2002 The Construction and Use of Canal Regulators in Ancient Sumer. AuOr 20: 115–122. Dijk, J. J. van. 1965a Une insurrection générale au pays de Larša avant l’avènement de Nūr-Adad. JCS 19: 1–25. 1978 Išbi’Erra, Kindattu, l’homme d’Elam et la chute de la ville d’Ur. JCS 30: 189–207.

81

82

Katrien De Graef

Driel, G. van. 1998 Land in Ancient Mesopotamia: ‘That What Remains Undocumented Does Not Exist’. Pp. 19–49 in Landless and Hungry? Access to Land in Early and Traditional Societies, eds. R. Haring and R. de Maaijer. Istanbul: Nederlands Historisch Archeologsich Instituut. Ellis, M. deJong 1972 Ṣimdatu in the Old Babylonian Sources. JCS 24: 74–82. Faust, D. E. 1941 Contracts from Larsa Dated in the Reign of Rīm-Sîn. YOS 8. New Haven: Yale University Press. Feigin, S. I. 1979 Legal and Administrative Texts of the Reign of Samsu-iluna. YOS 12. New Haven: Yale University Press. Figulla, H. H. and Martin, W. J. 1953 Letters and Documents of the Old Babylonian Period. UET 5. London: Trustees of the British Museum and of the University Museum, University of Pennsylvania, Philadelphia. Finkelstein, J. J. 1968 Cuneiform Texts from the Babylonian Tablets in the British Museum. Old Baby­ lonian Legal Documents. CT 48. London: British Museum. Frayne, D. R. 1982 New Light on the Reign of Išbi-Erra. Pp. 25–32 in Vorträge gehalten auf der 28. Rencontre assyriologiue internationale in Wien, 6.-10. Juli 1981, eds. H. Hirsch and H. Hunger. AfO Beih. 19. Wien: Ferdinand Berger & Söhne. 1989 A Struggle for Water: A Case Study from the Historical Records of the Cities Isin and Larsa (1900–1800 BC). BCSMS 17: 17–28. 1990 Old Babylonian Period (2003–1595 BC). RIME 4. Toronto: University of Toronto Press. Gaertner, L. 2008 The Motives for the Mēšarum Edict of King Ammiṣaduqa of the Old Babylonian Period: Ethics, Ego or Economics? Ph.D. dissertation. Stellenbosch University. Garfinkle, S. J. 2010 Merchants and State Formation in Early Mesopotamia. Pp. 185–202 in Opening the Tablet Box. Near Eastern Studies in Honor of Benjamin R. Foster, eds. S. C. Melville and A. L. Slotsky. Leiden-Boston: Brill. 2012 Entrepreneurs and Enterprise in Early Mesopotamia. A Study of Three Archives from the Third Dynasty of Ur. CUSAS 22. Bethesda: CDL Press. Gasche, H. 1989 La Babylonie au 17e siècle avant notre ère: approche archéologique, problems et perspectives. MHEM 1. Ghent: Ghent University. Gelb, I. J. 1965 The Ancient Mesopotamian Ration System. JNES 24: 230–243. Gomi, T. 1984 On the Critical Economic Situation at Ur Early in the Reign of Ibbisin. JCS 36: 211–242. Graetz, S. 2015 To Whom Can a Wronged Person Turn for Help in the Old Babylonian Period? Pp. 237–260 in Marbeh Ḥokmah Studies in the Bible and the Ancient Near East in Loving Memory of Victor Avigdor Hurowitz, eds. S. Yona, E. L. Greenstein, M. I. Gruber, P. Machinist, and S. M. Paul. Winona Lake, IN: Eisenbrauns.

Bad Moon Rising Grice, M. E. 1919 Records from Ur and Larsa Dated in the Larsa Dynasty. YOS 5. New Haven: Yale University Press. Hallo, W. W 1966 New Hymns to the Kings of Isin. BiOr 23: 239–247. 1982 The Royal Correspondence of Larsa: II. The Appeal to Utu. Pp. 95–109 in Zikir Šumim. Assyriological Studies Presented to F. R. Kraus on the Occasion of his Seventieth Birthday, eds. G. van Driel, Th. J. H. Krispijn, M. Stol, and K. R. Veenhof. Leiden: Brill. 1995 Slave Release in the Biblical World in Light of a New Text. Pp. 79–94 in Solving Riddles and Untying Knots: Biblical, Epigraphic, and Semitic Studies in Honor of Jonas C. Greenfield, eds. Z. Zevit, S. Gitin, and M. Sokoloff. Winona Lake, IN: Eisenbrauns. 2010 The World’s Oldest Literature. Studies in Sumerian Belles-Lettres. Leiden-Boston: Brill. Heimpel, W. 1987 Das Untere Meer. ZA 77: 22–91. Hein, E., Dodig, N., and Budyldina, N. 2014 Financial, Economic and Social Systems: French Regulation School, Social Structures of Accumulation and Post-Keynesian Approaches Compared, Working Paper, Institute for International Political Economy Berlin, No. 34/2014. See http://hdl.handle.net/10419/92910. Horsnell, M. J. A 1999 The Year-Names of the First Dynasty of Babylon (2 vols.). Hamilton, Ontario: McMaster University Press. Hudson, M. 2002 Reconstructing the Origins of Interest-Bearing Debt and the Logic of Clean Slates. Pp. 7–58 in Debt and Economic Renewal in the Ancient Near East, eds. M. Hudson and M. Van De Mieroop. International Scholars Conference on Ancient Near Eastern Economies 3. Bethesda: CDL Press. Jadir, W. al- and Abdullah, Z. R. 1983 University of Baghdad Excavation of Sippar. Sumer 39: 97–122. Janssen, C. 1992 Inanna-mansum et ses fils: relation d’une succession turbulente dans les archives d’Ur-Utu. RA 86: 19–52. Jean, C.-F. 1926 Contrats de Larsa. TCL 10. Paris: Geuthner. Jessop, B. 1997 Survey Article: The Regulation Approach. Journal of Political Philosophy 5: 287–326. Klengel, H. 1987 Non-Slave Labour in the Old Babylonian Period: The Basic Outlines. Pp. 159–166 in Labor in the Ancient Near East, ed. M. A. Powell. AOS 68. New Haven: American Oriental Society. Kraus, F. R. 1958 Ein Edikt des Königs Ammi-ṣaduqa von Babylon. SD 5. Leiden: Brill. 1965 Ein Edikt des Königs Samsu-iluna von Babylon. Pp. 225–231 in Studies in Honor of Benno Landsberger on his Seventy-fifth Birthday, April 21, 1965, eds. H. G. Güterbock and T. Jacobsen. AS 16. Chicago: Oriental Institute. 1979 Akkadische Wörter und Ausdrücke, XII. RA 73: 51–62. 1984 Königliche Verfügungen in altbabylonischer Zeit. SD 11. Leiden: Brill.

83

84

Katrien De Graef

Lafont, B. 1995 La chute des rois d’Ur et la fin des archives dans les grands centres administratifs de leur empire. RA 89: 3–13. Landsberger, B. 1939 Die babylonischen Termini für Gesetz und Recht. Pp. 219–234 in Symbolae ad iura Orientis antiqui pertinentes Paulo Koschaker dedicatae, eds. J. Friedrich, P. Koschaker, and Th. Folkers. SD 2. Leiden: Brill. Leemans, W. F. 1955 Review of Letters and Documents of the Old Babylonian Period (UET 5) by Figulla, H. H. and Martin, W. J. BiOr 12: 112–122. 1960 Foreign Trade in the Old Babylonian Period. SD 6. Leiden: Brill. 1968 Old Babylonian Letters and Economic History. A Review Article with a Digression on Foreign Trade. JESHO 11: 171–226. Legrain, L. 1937 Business Documents of the Third Dynasty of Ur. UET 3. London: Trustees of the British Museum and of the University Museum, University of Pennsylvania, Philadelphia. Lion, B. 2009 Sexe et genre (2). Des prêtresses fils de rois. Pp. 165–182 in Femmes, cultures et sociétés dans les civilisations méditerranéennes et proche-orientales de l’Antiquité, eds. F. Briquel-Chatonnet, S. Farès, B. Lion, and C. Michel. Topoi Suppl. 10. Lyon: Persée. Lipietz, A. 1987 Rebel Sons: The Regulation School (an interview with A. Lipietz conducted by J. Jenson). French Politics and Society 5: 17–26. Liverani, M. 1984 Land Tenure and Inheritance in the Ancient Near East: The Interaction between ‘Palace’ and ‘Family’ Sectors. Pp. 33–44 in Land Tenure and Social Transformation in the Middle East, ed. T. Khalidi. Beirut: American University of Beirut. 2014 The Ancient Near East. History, Society and Economy. London-New York: Routledge. Michalowski, P. 1980 Königsbriefe. RlA 6/1–2: 51–59. 1989 The Lamentation over the Destruction of Sumer and Ur. MC 1. Winona Lake, IN: Eisenbrauns. 1993 Letters from Early Mesopotamia. Writings from the Ancient World 3. Atlanta: Scholars Press. 2011 The Correspondence of the Kings of Ur: An Epistolary History of an Ancient Mesopotamian Kingdom. MC 15. Winona Lake, IN: Eisenbrauns. Olivier, H. 1997 Restitution as Economic Redress: The Fine Print of the Old Babylonian mēšarumedict of Ammiṣaduqa. ZAR 3: 12–25. Oppenheim, A. L. 1954 The Seafaring Merchants of Ur. JAOS 74: 6–17. Ormsby, D. L. 1972 An Old Babylonian Business Archive of Historical Interest. JCS 24: 89–99. Pinches, T. G. 1898 Cuneiform Texts from Babylonian Tablets in the British Museum. CT 4. London: British Museum. Potts, D. T. 1990 The Arabian Gulf in Antiquity. 1: From Prehistory to the Fall of the Achaemenid Empire. Oxford: Clarendon Press.

Bad Moon Rising 2009 The Archaeology and Early History of the Persian Gulf. Pp. 27–56 in The Persian Gulf in History, ed. L. G. Potter. New York: Palgrave Macmillan. 2016 Cultural, Economic and Political Relations Between Mesopotamia, the Gulf Region and India Before Alexander. Pp. 109–118 in Megasthenes und seine Zeit, eds. J. Wiesehöfer, H. Brinkhaus, and R. Bichler. Classica et Orientalia 13. Wiesbaden: Harrassowitz Verlag. Potts, T. 1994 Mesopotamia and the East: An Archaeological and Historical Study of Foreign relations ca. 3400–2000 BC. Oxford: Oxford University Committee for Archaeology. Ranke, H. 1906 Babylonian Legal and Business Documents from the Time of the First Dynasty of Babylon Chiefly from Sippar. BE 6/1. Philadelphia: University of Pennsylvania. Rawi, F. N. H. al- and Dalley, S. 2000 Old Babylonian Texts from Private Houses at Abu Ḥabbah Ancient Sippar Baghdad University Excavations. EDUBBA 7. London: NABU. Renger, J. 1979 Interaction of Temple, Palace and ‘Private Enterprise’ in the Old Babylonian Economy. Pp. 249–256 in State and Temple Economy in the Ancient Near East. Proceedings of the International Conference Organized by the Katholieke Universiteit Leuven from the 10th to the 14th of April 1978, ed. E. Lipinski. OLA 5 and 6. Leuven: Peeters. 1994 On Economic Structures in Ancient Mesopotamia. Or. NS 63: 157–208. 1995/6 Institutional, Communal, and Individual Ownership or Possession of Arable Land in Ancient Mesopotamia from the End of the Fourth to the End of the First Millennium B.C. Chicago-Kent Law Review 71: 269–319. 2002 Royal Edicts of the Old Babylonian Period. Structural Background. Pp. 139–162 in Debt and Economic Renewal in the Ancient Near East, eds. M. Hudson and M. Van De Mieroop. International Scholars Conference on Ancient Near Eastern Economies 3. Bethesda: CDL Press. Richardson, S. 2015 Building Larsa: Labor-Value, Scale and Scope-of-Economy in Ancient Mesopotamia. Pp. 237–328 in Labor in the Ancient World, eds. P. Steinkeller and M. Hudson. The International Scholars Conference on Ancient Near Eastern Economies 5. Dresden: Islet Verlag. Römer, W. H. Ph. 1965 Sumerische Königshymnen der Isin-Zeit. Leiden: Brill. Rowton, M. B. 1967 Watercourses and Water Rights in the Official Correspondence from Larsa and Isin. JCS 21: 267–274. Roth, M. T. 1997 Law Collections from Mesopotamia and Asia Minor. Second edition. Atlanta: Scholars Press. Sallaberger, W. 1999 Ur-III-Zeit. Pp. 121–390 in Akkade-Zeit und Ur III-Zeit, eds. W. Sallaberger and A. Westenholz. OBO 160/3. Fribourg: Academic Press/Göttingen: Vandenhoeck & Ruprecht. Sharlach, T. M. 2004 Provincial Taxation and the Ur III State. CM 26. Leiden: Brill. Sigrist, M. 1987 Lipit-Ištar. RlA 7/1–2: 28–30. 1988 Isin Year Names. IAPAS 2. Berrien Springs: Andrews University Press. 1990a Larsa Year Names. IAPAS 3. Berrien Springs: Andrews University Press.

85

86

Katrien De Graef

1990b Old Babylonian Account Texts in the Horn Archaeological Museum. AUCT 4. IAPAS 7. Berrien Springs: Andrews University Press. Skaist, A. 1994 The Old Babylonian Loan Contract: Its History and Geography. Ramat Gan: BarIlan University Press. Steinkeller, P. 1987 The Administration and Economic Organization of the Ur III State: The Core and the Periphery. Pp. 19–41 in The Organization of Power: Aspects of Bureaucracy in the Ancient Near East, eds. McG. Gibson and R. D. Biggs. SAOC 46. Chicago: The Oriental Institute. 2004 A History of Mashkan-Shapir and Its Role in the Kingdom of Larsa. Pp. 26–42 in The Anatomy of a Mesopotamian City: Survey and Soundings at MashkanShapir, eds. E. C. Stone and P. E. Zimansky. Winona Lake, IN: Eisenbrauns. Stol, M. 1971 Review of Walters, S. D. 1970. BiOr 28: 365–369. 1976 Studies in Old Babylonian History. PIHANS 40. Istanbul: Nederlands historischarcheologisch instituut. 1981 Letters from Yale. AbB 9. Leiden: Brill. Stone, E. C. and Zimansky, P. 2016 Archaeology Returns to Ur: A New Dialog with Old Houses. NEA 97: 246–259. Tanret, M. 2004 Verba volant, scripta non manent Tablettes nomads dans les archives des gala. maḫ à Sippar-Amnānum. Pp. 249–270 in Nomades et Sédentaires dans le ProcheOrient ancien. Compte rendu de la XLVIe Rencontre Assyriologique Internationale (Paris, 10–13 juillet 2000), ed. C. Nicolle. Amurru 3. Paris: ERC. 2010 The Seal of the Sanga. On the Old Babylonian Sangas of Šamaš of SipparJaḫrūrum and Sippar-Amnānum. CM 40. Leiden-Boston: Brill. Tanret, M. and Janssen, C. 1992 ana qabê—qui remplace qui? NABU 1992/3: 63–66 no. 85. Tinney, S. 1996 The Nippur Lament: Royal Rhetoric and Divine Legitimation in the Reign of IšmeDagan of Isin (1953–1935 B.C.). OPSNKF 16. Philadelphia: Samuel Noah Kramer Fund. 1998 Texts, Tablets, and Teaching: Scribal Education in Nippur and Ur. Expedition 40: 40–50. Van De Mieroop, M. 1989 Gifts and Tithes to the Temples in Ur. Pp. 397–401 in DUMU-E2-DUB-BA-A Studies in Honor of Åke W. Sjöberg, eds. H. Behrens, D. Loding, and M. T. Roth. OPSNKF 11. Philadelphia: Samuel Noah Kramer Fund. 1992a Society and Enterprise in Old Babylonian Ur. BBVO 12. Berlin: Reimer. 1992b Old Babylonian Ur: Portrait of an Ancient Mesopotamian City. JANES 21: 119–130. 2002a A History of Near Eastern Debt? Pp. 59–94 in Debt and Economic Renewal in the Ancient Near East, eds. M. Hudson and M. Van De Mieroop. International Scholars Conference on Ancient Near Eastern Economies 3. Bethesda: CDL Press. 2002b Credit as a Facilitator of Exchange in Old Babylonian Mesopotamia. Pp. 163–173 in Debt and Economic Renewal in the Ancient Near East, eds. M. Hudson and M. Van De Mieroop. International Scholars Conference on Ancient Near Eastern Economies 3. Bethesda: CDL Press. 2015 Ur. A. II. Philologisch. Im frühen 2. Jahrtausend. RlA 14/5–6: 361–364. 2016 A History of the Ancient Near East. ca. 3000–323 BC. Third edition. Chichester: Wiley Blackwell.

Bad Moon Rising Vanstiphout, H. L. J. 1989/90 The Man from Elam. A Reconsideration of Išbi-Erra “Hymn B.” JEOL 31: 53–62. 1978 Lipit-Eštar’s Praise in the Edubba. JCS 30: 33–61. Veenhof, K. R. 1986 Cuneiform Archives. An Introduction. Pp. 1–36 in Cuneiform Archives and Libraries: Papers Read at the 30e Rencontre Assyriologique Internationale, Leiden, 4–8 July 1983, ed. K. R. Veenhof. PIHANS 57. Istanbul: Nederlands HistorischArcheologisch Instituut. Volk, K. 2011 Eine neue Inschrift des Königs Sîn-iddinam von Larsa. Pp. 59–88 in Cuneiform Royal Inscriptions and Related Texts in the Schøyen Collection, ed. A. R. George. CUSAS 17. Bethesda: CDL Press. Walters, S. D. 1970 Water for Larsa. An Old Babylonian Archive Dealing with Irrigation. YNER 4. New Haven: Yale University Press. Wilcke, C. 1970 Drei Phasen des Niedergangs des Reiches von Ur III. ZA 60: 54–69. Yoffee, N. 1988 The Collapse of Ancient Mesopotamian States and Civilization. Pp. 44–68 in The Collapse of Ancient States and Civilizations, eds. N. Yoffee and G. L. Cowgill Tuscon: University of Arizona Press. 2010 Collapse in Ancient Mesopotamia. What Happened, What Didn’t. Pp. 176–203 in Questioning Collapse: Human Resilience, Ecological Vulnerability, and the Aftermath of Empire, eds. P. A. McAnany and N. Yoffee. Cambridge: Cambridge University Press. Zeebari, A. al-. 1964 Old Babylonian Letters. TIM 1. Baghdad: The Directorate General of Antiquities.

87

The Moon Watching Over the Sun and Venus: Revisiting the Attributes and Functions of Nanna/Sîn in Mesopotamia Isabel Gomes de Almeida & Maria de Fátima Rosa CHAM & DH, FCSH, Universidade NOVA de Lisboa 1

Moonlight drowns out all but the brightest stars J. R. R. Tolkien

Introduction This paper intends to explore the role of the patron god of Ur, Nanna/Sîn, 2 in the Mesopotamian religious system, by focusing on the relation between his numinous features and those of his divine progeny. Inanna/Ištar and Utu/Šamaš, two of the most influent deities in the “land between the rivers,” shared an astral identity with their divine father, which, in turn, led to multiple associations between them. These were visible not only in their attributes and functions, but also in iconographic depictions. 3 Since family ties were a crucial devise for identity and for legitimation in ancient Mesopotamia, we believe there is a profound symbolic meaning on the construction of this triad, where the fatherly role of the Moon allowed for the rise to prominence of its offspring. Thus, it is our goal to examine the nature of their divine kinship, particularly on what concerns the third millennium BCE literary accounts, concentrating our attention in two main symbolic vectors: the luminosity and the movement of this divine cluster.

1.  This paper had the support of CHAM (FCSH/NOVA-UAc), through the strategic project sponsored by FCT (UID/HIS/04666/2013). 2.  Given the profound and ancient syncretism between the Sumerian and the Semitic divine figures, which makes it impossible to isolate the role of each background on the construction of Mesopotamian deities, we prefer to use the deities’ combined names. 3.  Glyptic representations insist on displaying their three emblems closely together. Though it is not our purpose to analyse Mesopotamian visual arts in this paper, we suggest that this iconographic association helped to stress, among other aspects, the preponderance of family identity.

91

92

Isabel Gomes de Almeida & Maria de Fátima Rosa

Preliminary Remarks Addressing any aspect of the history of the Mesopotamian religious system is no easy task. Oppenheim’s argumentation on the nature of the sources available and on the conceptual barriers which constrain the understanding of this complex, and rather distant religion, still echoes in one’s mind (1976: 171–183). As pointed out by many, though extremely negative and somewhat dated, Oppenheim’s warn­ ings should be kept in mind, given the necessary caution one must have when examining such a matter. Veldhuis (2004: 15–16), for instance, stressed how the personal/individual religious manifestations constitute a more tangible expression of one’s religion, when compared with the elaborations of the religious elite of the time. As we know, the written data concerning Mesopotamian personal religion 4 is rather fragmentary, which prevents a balanced examination between the official and the private/individual spheres. On the other hand, Seymor (2011: 787) underlines the problems regarding the complexity of the written sources in a wide time and space spectrum of analysis, given their differences regarding dating, place of origin, plus the variances on the contents. Though these discrepancies are rich if one is to develop a comparative approach about a confined period or region, they constitute a deep barrier to achieve a general picture of the Mesopotamian religious thought. On another level, there are two central and entangled idiosyncrasies of the Mesopotamian religious system that any researcher should be alerted to: its rather profound theocentric nature, and its cumulative character. These features should always be taken into account when applying the History of Religions theoretical and methodological framework to the Mesopotamian case. Hence, some preliminary remarks on how we shaped our approach to this subject are needed. The data pertaining to the inhabitants of the “land between the rivers” clearly show they were the very definition of homo religiosus. 5 In fact, since the beginning of this civilization and throughout time, one can say that the Mesopotamians exper­ ienced the presence of a numinous element, whose fascinating but simultaneously daunting nature required a reaction. This “experience of a mysterium tremendum et fascinosum, a confrontation with a ‘Wholly Other’ outside of normal experience and indescribable in its terms” (Jacobsen 1976: 3) was, thus, the trigger for the creation of a mythical discourse expressed both in ritual practices and in oral/written narratives. Hence, Mesopotamians fashioned a system of knowledge, which helped them to appease their anxieties regarding the numen’s presence but also regarding the apprehensions of their own existence. Through a “controlled and calculated imagination” (Bottéro 1998: 55) answers to their multiple doubts were provided, by the edification of a “philosophy in images” (Bottéro 2004: 55). This system of knowledge was constructed on the premise that the natural world was impregnated by the presence of divine figures, whose existence was indisputable, and who played central roles on the phenomenological cause-effect processes. Naturally, the understanding of divine will and actions became central to the Mesopotamian homo religiosus, urging the elaborations on the nature, relations, behaviours, attributes, and any other aspect which concerned deities. 4.  We are applying this expression in a broader sense, without any disregard for the other categories pertaining to a more private religious behavior, such as “individual,” or “familiar.” 5.  On the theoretical framework followed here, see Dumézil 1958; Otto 1966; Bouillard 1974; Eliade 1987, 1989; and Ries 1992.

The Moon Watching Over the Sun and Venus

93

Considering the transcendental character of the divine figures, which made them unintelligible, the Mesopotamians were forced to recur to their own reality, to formulate their mythical discourses. The concrete aspects were, subsequently, deeply transformed, through exercises which allowed the metamorphosis of tangible into symbolic. However, the reality and the experiences of the individuals and/or the community were, somehow, reflected in these metaphorical significances. There­ fore, any mythical discourse must be understood as an account with its own logic and reason, “presenting a form of truth” (Hatab 1990: 10) deeply connected with the historical context it was produced in. Naturally, as scholars, we should question the processes undertaken to transpose the concrete aspects into the religious discourse, through a scrupulous hermeneutical process. Yet, our point is we should never address a religious system (and its discourses) as products independent of their historical processes. By doing so, and by understanding myths as elaborations of and for a given context, the danger of simplifications, another aspect Oppenheim alerted to, will surely decrease. Thus it is imperative to integrate the Mesopotamian mythic production in its historical context, which is the goal of any study within the History of Religions field. But to do so, we should recall the cumulative nature of the religious system developed in the “land between the rivers.” The fact is we are dealing with a millen­arian world, where several cultural backgrounds and religious experiences circu­lated, originating systematic syncretic processes. 6 Moreover, the theocentric (and even centrifugal 7) nature of the Mesopotamian thought, makes it impossible to discard any historic action from the religious sphere. As stressed by Lenzi (2007: 126) the economic, the political, the cultural, and the social events had strong implications on Mesopotamian religion, which led to complex and continuous accumulations. Old and new visions were, thus, embraced, by recurring to the devise of accommodation shaping a multi-layered construct, where “the sense of tradition brings legitimacy” (Odisho 2004: 3). Therefore, as in so many other contexts, change and tradition were encompassed by the Mesopotamians, so that a sense of identity and order was maintained. 8 Given the above, we agree with Seymor’s perspective (2011: 787) that it is wiser to develop partials studies, in order to better identify and interpret the several layers of this religion. Moreover, it is our understanding that when it comes to the field of religions, or any cultural expression for that matter, an historical analysis should privilege a longue durée methodological approach. As a result, we believe the identification of changes within continuity and of gradations within layers can become perceptible. 6.  Besides the crucial contributes of Sumerian and Semites, who were considered by Bottéro and Kramer (1989: 3) as a kind of founding fathers of this civilization, other protagonists were important in the edification of this religious building, such as the Gutians, who had a strong presence in the south, during the second half of the third millennium BCE, or the Hittites, who were central players in the Near and Middle East, during the second millennium BCE. The agency of these and other actors, combined with the Semitic and Sumerian ones, contributed for the extreme dynamism of the Mesopotamian religious system. 7.  Bottéro 1998: 30. 8.  Naturally, the postulates of Erik Hobsbawm on the “invention of tradition” can and should be apply to this line of thought. On this matter, see the different contributions in Hobsbawm, and Ranger 1983.

94

Isabel Gomes de Almeida & Maria de Fátima Rosa

Hence, the reflections presented in this paper, regarding the relation between the symbolic meanings of the Mesopotamian divine Moon, Sun, and Venus, were guided by the following assumptions: a) mythic thought always reports to reality; b) mythological discourses are created in deep relation to its historical context; c) Mesopotamian religious system, due to its historical processes, was cumulative; d) Mesopotamian deities were the product of multiple and systematic processes of syncretism; e) changes and continuities were embraced by Mesopotamians in the edification of traditions; and finally, f) the long duration analysis is the method­ ological approach which better suits our present goals.

Moon, Sun, and Venus: A Celestial Divine Triad Throughout time, when looking at the sky, homo religiosus quickly perceived the influence and importance of three main celestial bodies in their daily life: the Moon, the Sun, and the planet/star Venus. These entities, easily observed by the naked eye, acquired a profound role within most religious systems, 9 the Mesopotamian one being no exception. By the banks of the Tigris and the Euphrates, these celestial bodies were understood as a triad, forming a household, where Nanna/Sîn, the Moon god, was the father of the twins Utu/Šamaš, the Sun god, and Inanna/ Ištar, the “morning/evening star.” Though the three deities were worshiped since the dawn of Mesopotamian history, and they all occupied a relevant position within the pantheon, diachronically, when one focus on the mythological narratives, the Moon god seems to be overshadowed by his progeny. Inanna/Ištar was probably one of the main protagonists of Mesopotamian literature, assuring remarkable roles, no matter what, no matter when. As for Utu/Šamaš, though his cult only became well established at the beginning of the second millennium BCE, he was always granted a decisive role, whether as the epitome of justice or as the radiant light who brought prosperity. Despite the seemingly moderated role of Nanna/Sîn, when compared to those of this progeny, the affiliation between them endured. It is our belief that the numinous character of Nanna/Sîn, as envisioned by the Mesopotamian homo religiosus, made this possible. To explore this hypothesis, we should start out by asking ourselves one crucial question: why did the Mesopotamians feel the need to bind these deities by family ties? A closer look into the third millennium BCE mythical narratives can help to pinpoint the motives for this kinship, given that it seems it was sometime during this long period that those ties were consolidated. On one hand, the hymns attributed to Enheduanna 10 identify Inanna/Ištar as offspring of Nanna/Sîn, 11 discarding other divine fatherly figures of the powerful patron goddess of Uruk (such as her Urukian companion An/Anu, or even her friendly rival Enki/Ea 12). On the other hand, the 9. For instance, in the Maya religious system, the Sun and the Moon were regarded as divine lovers, whereas Venus and the Sun were divine brothers. It is important to note that these deities’ gender identity changed from one religious context to another. For an introduction to the Maya religious manifestations, see Thompson 1970. 10.  On the discussion regarding Enheduanna’s authorship, see Civil 1980. 11.  “Impetuous wild cow, great daughter of Suen” (ETCSL 4.07.2), “the great daughter of Suen, exalted among the Great Princes” (ETCSL 4.07.2). 12.  On the local or precedent traditions, which ascribed An/Anu or Enki/Ea as fathers of the goddess, see Westenholz 1998 and Almeida 2015.

The Moon Watching Over the Sun and Venus

95

Third Dynasty of Ur literary accounts establish a close connection between the Moon and the Sun gods. 13 Given the political context in which the above-mentioned narratives were produced, it is possible to identify political motivations in the way the family ties of the triad were established. As it is well known, in Mesopotamia, kingship was a sacred mission, whereas deities delegated the right to rule the land on the royal persona. Despite being consid­ered human, 14 a subjective affiliation between the Mesopotamian monarch and the patron god or goddess of the dynasty was fashioned. In this sense, each ruling power felt the need to crystalize the symbolic genealogical bounds with the divine figures, to legitimate their city’s rise to prominence. That was the case of the Akkadian Dynasty and of the Third Dynasty of Ur. During the Akkadian unification, it seems that Ištar-Annunītum 15 was envisioned as the patron goddess of the Akkadian rulers. Given the importance of Sumerian deities, at that time, it was important to straighten the ties between the Akkadian ruling family and the gods and goddesses of Sumer. Hence, it seems the Akkadian rulers took advantage of the profound syncretism between the Semitic Ištar and the Sumerian Inanna, already under construction, emphasizing the confusion between the two feminine divine figures. Simultaneously, Enheduanna, the daughter of Sargon and aunt of Narām-Sîn, was placed at the service of Nanna/Sîn, at his temple in Ur (Winter 1987: 200). There, she took the role of high priestess and of dam, 16 that is, the symbolic wife of the Moon god. 17 Through this metaphorical marital liaison, Enheduanna helped to further legitimate the Akkadian Dynasty, given that she was in close contact to no other than the divine father of Inanna/ Ištar. This intricate process of legitimation, whereas divine and human genealogies were entangled, had to have an impact in the consolidation of their father-daughter relationship. Thus, we believe the emphasis on their kinship displayed by the compositions attributed to the daughter of Sargon expressed, amongst other aspects, these political motivations. In what concerns the Third Dynasty of Ur, we identify a paralleled process. In this case, it was Nanna/Sîn, the patron god of the new capital of the land, who under­went a rearrangement. His tie with the traditional leader of the Mesopotamian pantheon of the time, Enlil, was highlighted, given that this affiliation would help to promote him into a higher position within the divine universe. Consequently, the Ur III king’s claims over the land of Sumer and Akkad would gain other vigour. At the time, Utu/Šamaš was also being promoted amongst the divine hierarchy. Though his role as judge of the dead was only fully developed during the Old Babylonian period (Katz 2003: 28–29), the Sun god was assigned with the function of being “the judge who searches out verdicts for the gods” (ETCSL 1.1.3) during the Ur III 13.  For instance, in Enki and the World Order, a composition which concerns the Third Dynasty of Ur, Utu/Šamaš is identified as son of Ningal, who was the traditional spouse of the Moon god. See the composition in ETCSL 1.1.3. 14.  On the human or divine character of the Mesopotamian monarch, see the different contributions in Brisch 2008. 15.  One of Narām-Sîn’s epithets was “warrior of Aštar-annunītum,” which points to this goddess’s role as a patron deity of the Akkadian dynasty. On Ištar-Annunītum, see Selz 2000. 16.  “Spouse” (ePSD). On the role of Enheduanna as symbolic spouse of Nanna/Sîn, see Westenholz 1989. 17.  For the multiple interpretations on the Mesopotamian sacred marriage ritual, see the different contributions in Nissinen, and Uro 2010.

96

Isabel Gomes de Almeida & Maria de Fátima Rosa

period. On another level, some Akkadian accounts depict Utu/Šamaš as the son of Enlil (or even An/Anu) (Black and Green 1992: 182–184). As we know, the Third Dynasty of Ur was a time where Sumerian traditions were revitalized, probably to draw a line between the new power and the Akkadian one. So, it is fair to speculate that the reaffirmation of Nanna/Sîn as father of the Sun god helped to reinforce this Sumerian tradition, and at the same time, it allowed to control the rising power of Utu/Šamaš, given that he was no longer son of higher deities, but offspring of the patron god of the land. However, one must recall that the political motivations always had to consider the previous traditions, 18 or else it would be difficult for these changes to become widely accepted. As stated above, the cumulative nature of this religious system worked upon the premise that new and old notions should be accommodated in the logic of continuity. As Jean Bottéro argued, “il n’est pas facile de changer de dieux” (1998: 73). Hence, we should look beyond the political context, in order to understand why this celestial triad was envisioned as a family. Because mythical discourse finds its roots within reality, Mesopotamians seem to have transposed their own social organization, which was grounded in family clusters, to the divine universe. One can easily imagine the pantheon as a macro-­ family, where every deity belonged to a household, which in turn was connected with the rest of the divine clan. Just like in the human world, each member of a divine cluster reinforced its position by sharing and continuing a common identity. To some extent, each deity was thus a reflection of his own ancestors. 19 So, it is fair to say that for the Mesopotamians the sharing of functions and/or attributes was the main vector that influenced the construction process of this or any other divine family, in the first place. In what concerns the divine triad under analysis, it is our understanding that the ties expressed by their luminosity, and by their movement in the skies, was the motive that led to their affiliation. In this perspective, Nanna/ Sîn was not overshadowed by his progeny; instead we argue he was perceived as a respectable patriarch, who passed on his own numinous features into the next generation, and, like any other progenitor, continuously watched over them.

There Shall Be Light! Let us start with the significance of light which we find in Mesopotamian mythological literature. As expressed in Atra-Hasis, complete darkness was a char­ acteristic pertaining to the diluvium chaos, a cosmic event that brought turmoil and that destroyed life. 20 This connection of darkness with death is emphasized in the descriptions of the Netherworld, as “…the house where those who entered are deprived of life…they see no light, they dwelt in darkness” (Dalley 2000: 155). In opposition, the divine heavenly domains and the fertile terrestrial world were 18.  Klein (2001: 289, 292) states that Enlil as a father of Nanna/Sîn is an older tradition, given the reference “impetuous calf of Enlil” which appears on Eannatum’s Stele of the Vultures. 19.  “Ever since the gods acquired their anthropomorphic character, the theologians have engaged in grouping them into generations and families, whereby the rank and function of a given god was determined by his genetic relationship to another god in the pantheon” (Klein 2001: 279). 20.  “No one could see anyone else, they could not be recognized in the catastrophe. The Flood roared like a bull, like a wild ass screaming the winds [howled], the darkness was total, there was no sun” (Dalley 2000: 31).

The Moon Watching Over the Sun and Venus

97

conceived as overflown by brightness. Since luminosity and existence were obviously linked to a sense of order, it is fair to say that in the symbolic Mesopotamian framework, these terms were equated. Hence, deities intrinsically connected with light, such as Nanna/Sîn, were naturally associated with life and order. One of the obvious functions of the Moon god was to shed light into the night. 21 Diachronically, night-time was always considered a moment of anxiety, since the lack of light opened the possibility of one’s intimate fears to become real. In what concerns the Mesopotamians, whether recalling the diluvium darkness or the obscu­ rity of the Netherworld, night-time was associated with barrenness, and therefore considered a dreadful moment. Nanna/Sîn’s brightness secured the possibility of definition within night, thus appeasing the worries of the Mesopotamians. Consequently, the Moon god received other functions related with his role within night skies, whose symbolic meanings lead us to think he was envisioned has having a wider spectrum of action. The Nanna F hymn, for instance, depicts Nanna/Sîn as overseeing the stars, which were considered his herd. 22 This task conveys profound significances, given that as a cowherd the Moon god was considered the ruler of celestial bodies. 23 Simultaneously, this deity was understood as impregnating the stars. 24 In fact, as Mark Glenn Hall (1985: 593–594) already stated, the renewal that came along with the moon’s cycle, allowed the fertility of the cattle. Moreover, Mesopotamian early understanding on the impact the full moon had on the tide of the rivers (Ornan 2001: 3), plus the relation established between the moon and the women’s cycle, reinforced the god’s fertile nature. Consequently, his disappearance during the new moon was understood as a period where Nanna/Sîn was conceived as sleeping in the Netherworld, a place where the god could not shine, and accordingly, could not act as a fertility patron (ETCSL 5.5.2). The god’s return from the realm of the dead was thus conceived as a time full of opportunities for life to flourish. On another level, the well-known relation between the Moon god and the bull, underlines his celestial power: the crescent, which was the traditional emblem of the “Divine 30,” was equated with the bull’s horns, a symbolic motive identifiable, since the Neolithic period, 25 as connected with governance and, of course, fertility. More than just shining through the night, Nanna/Sîn’s light bestowed him sexual and ruling dimensions. Thus, the multiple significances of luminosity allow us to argue that the starry night was envisioned as a promise of abundance, controlled and protected by Nanna/Sîn’s numinous features. Given the above, we argue that the Mesopotamian homo religiosus was naturally impelled to envision the divine Moon as the father of the divine Sun and Venus, who were both conceived as connected with governance and prosperity. 21.  “The lord has burnished the heavens, he has embellished the night” (ETCSL 4.13.06). 22.  “The cows are driven together in herds for him. His various types of cow number 39600. His young cows and calves … number 108000. His young bulls number 126000. The sparkling-eyed cows number 50400. The white cows number 126000. The cows for the evening meal are in four groups of five each. Such are the various types of cow of father Nanna” (ibid.). 23.  It must be recalled that the metaphor of the shepherd was applied to the Mesopotamian ruler. See, for instance, Westenholz 2004. 24.  “He is ever able to increase the butter of abundance in the holy animal pens of (…) and goats” (ETCSL 4.13.06). 25.  On the Near and Middle Eastern Neolithic repertoire, see the different contributions in Bolger and Maguire 2010, especially Stuart Campbell’s reflections on its symbolic significances (2010: 147–155).

98

Isabel Gomes de Almeida & Maria de Fátima Rosa

Their celestial father, who ruled the skies, could have been envisioned as transferring his governance feature into his progeny, which in turn developed their own role as divine sovereigns. Utu/Šamaš frequently appears handing over “the rod and the ring” emblem to the Mesopotamian king, which were the symbols for the equity of kingship; and Inanna/Ištar, the “Queen of Heaven and Earth,” played a central role on the renewal of the king’s right to rule, namely in what concerns the hierogamy symbolic ritual. Combined, the divine twins stood for the Mesopotamian sovereignty principles, which they bestowed upon the human “shepherd of the land” improving his father ruling role, as it was expected by the new generations. In what concerns fertility, in his fatherly role, Nanna/Sîn appears giving instructions to the rising sun for him to bring the necessary warmth, which allowed crops to grow. 26 Together, father and son had a complementary role regarding cosmic fertility, a trait transposed into visual arts. As exposed by Collon (1992: 28–29) and by Braun-Holzinger (1993) some glyptic exemplars of the Early Dynastic and Akkadian periods, presents an anthropomorphic Nanna/Sîn with rays of light rising from his shoulders, which parallels the traditional iconographic depiction of Utu/Šamaš. As for Inanna/Ištar, we can identify an equivalent symbolic meaning regarding the father and daughter disappearance from the celestial domains. When Venus was not observed in the skies, Mesopotamians envisioned that this celestial body was in the Netherworld, just like Nanna/Sîn, during the new moon. Both disappearances, though at different levels and with different impacts, had consequences on the fertility of the land. 27 We can, thus, establish that luminosity, with all its underlying symbolic significances, was a vector shared by the celestial triad. This identification by the Mesopotamian homo religiosus helped to shape the close connection between Moon, Sun, and Venus, whereas their radiance was conceived as a “family trade.”

The Movement Across the Universe The central role ascribed to deities, combined with the belief that the divine figures painted nature with signs that bore their will, encouraged Mesopotamians to thoroughly observe the natural world, promoting the development of divination techniques very early in time. 28 Naturally, the movement of the celestial bodies was object of a profound examination, which helped to further develop symbolic meanings regarding the deities that crossed the skies. So, it comes with no surprise that every member of the triad under analysis displays a voyager nature. Just like 26.  “Father Nanna gives the direction for the rising Utu” (ETCSL 1.8.2.1). 27. The symbolic significances conveyed in the Sumerian and Semitic narratives which describe Inanna/Ištar’s descent to the Netherworld deeply surpass the concrete event of Venus’ disappearance from the sky. Thus, we are not, by any means, reducing those rich accounts to just an explanation of a natural phenomenon. Moreover, we are aware of the significant differences between Nanna/Sîn’s presence in the realm of the dead—conceived as asleep—and of Inanna/Ištar’s one—held captive as a consequence of her defiance of the cosmic rules. Nevertheless, given the polysemy nature of the mythological discourse, we believe this parallel can be established, stressing its non-exclusive character. 28.  According to Seth Richardson (2010), the first known Mesopotamian references regarding the divination techniques appear in the professions’ list discovered in Tell Fara, ancient Šuruppak, and are dated to ca. 2600 BCE. In the introduction of this book, however, Amar Annus points to the strong possibility that divination was older than the written records which attest it, being known and transmitted by a previous oral tradition.

The Moon Watching Over the Sun and Venus

99

luminosity, there are several significances pertaining to this feature, the liminal one being extremely important. If we consider the geographic disposition of the “land between the rivers,” we identify the Zagros mountain range as enclosing Mesopotamia from eastern neighbours. The distant lands within the Iranian plateau, from which, now and then, threats emerged, ascribed a liminal character to the mountains: it was the place where the other dwelt. Nanna/Sîn had to have an obvious presence in this geographic accident, given the path the moon follows during its cycle, which was certainly observed by the Mesopotamians. In this natural observation can reside the earlier motive for the association between Nanna/Sîn’s disappearance, during new moon, and his voyage to the Netherworld. During the third millennium BCE, there was still an association between the mountains (Sumerian: kur) and the realm of the dead, probably a legacy of older times, as Katz (2003: 45) pointed out. Whether geographical or cosmic, the symbolic notions associated with mountain stood for alterity, which in turn bestowed to Nanna/Sîn a liminal character. The Moon god crossed borders between the civilized and the wild, between life and death, between order and chaos. On another angle, the voyage through the skies allowed Nanna/Sîn to control every aspect of the world beneath. For this reason, he could easily foresee the future and grant prosperity to the land. His link with divination, once and again, was more than natural. This control of the world beneath, through movement, enclosed yet another meaning: the possibility to divide time in several segments. Again, Nanna/ Sîn was prone to act upon it: the rise and the set of the moon differentiated day and night; its several phases (new moon, waxing, full moon, and waning) differentiated units within its natural cycle; and its full cycle allowed the yearly calendar to be defined. Therefore, Nanna/Sîn’s movement was a feature that ascribed him the role of time regulator. Again, it was only natural that the deity bound to these features was conceived as begetting children who not only shared but also enhanced them. Given that the sun’s movement divided the days, once and again, father and son displayed a complementary role as time regulators. The divine Sun’s daily crossing path through the skies also secured the possibility for him to know everything that happened beneath, just as his father did during the night. Moreover, in his voyage through the cosmos, Utu/Šamaš arose from the mountains, which made him a regular presence in the liminal kur, manifesting, therefore, a liminal character. Hence, Utu/Šamaš had the necessary requirements to excel as a judge, both in life and in death, a combined function the Sun god received, as we have seen, from the Old Babylonian period on. Inanna/Ištar, on her side, had a profound liminal character which was intrinsic to her astral form. As we have mentioned, the goddess travelled to the Netherworld, and in her warrior function was described as charging against the mountains (ETCSL 1.3.2), features that stressed her connection with the liminal kur. 29 On another level, the presence of the “morning and evening star” at the dawn of day, and at the beginning of the night, could be envisioned as Venus accompanying her divine brother and father in their functions as time regulators. Inanna/ 29.  Amongst the data pertaining to Uruk IV, a rather mysterious epithet for the goddess appears: “Inanna-kur.” As discussed elsewhere (Almeida 2015), the several symbolic significances of kur and of this goddess’s itinerant character, point to the definition of her liminal character very early in time.

100

Isabel Gomes de Almeida & Maria de Fátima Rosa

Ištar’s itinerant character through the skies also secured the possibility to control the world beneath, and therefore to predict the “yet to come” events, which was manifested in the development of her prophetic nature. Combined, the features pertaining to the symbolic meanings of movement across the skies granted Nanna/Sîn and his progeny a relevant role within the Mesopotamian religious framework, and can help us to understand how the family bounds between the triad were shaped by the Mesopotamians.

Conclusions When we first started to think about this paper, one question was always present: in the long duration, was Nanna/Sîn overshadowed by his children? We concluded that he was not. As stated at the beginning, the Mesopotamian social order envisioned family ties as a device to maintain and legitimate the sense of identity, throughout generations. It was expected that progeny continued their parents’ paths, developing and perfecting what was already achieved. Given that myth finds its roots on the mundane experiences, Nanna/Sîn’s relation with his offspring was guided by the very same logic Mesopotamians had regarding their own relations between old and new generations. Moreover, an identical logic is also displayed in the theogonic process, where each new divine generation was more defined and powerful than the previous one. Therefore, Nanna/Sîn seems to have been understood as an honourable patriarch, who by transferring his features onto his progeny, allowed for them to surpass him. Utu/Šamaš and Inanna/Ištar extraordinary protagonism within mythical accounts, regarding the domains of abundance, justice, royal power, and so on, was only possible because of their close relation with their father. The construction of these genealogical bounds, we argue, finds its roots on the identification by the Mesopotamian homo religiosus of luminosity and movement as shared traits of this divine triad. Recognizing the difficulties related to the search for older layers underneath the (re)construction processes of divine figures, as well as the deficiencies on fully understanding past actors’ religious feelings and motivations, we hope the reflections presented above can contribute to further discussions about the patron deity of Ur and, in a broader sense, to the Mesopotamian religious system, from a History of Religions perspective.

Bibliography Almeida, I. 2015 A construção da figura de Inanna/Ištar na Mesopotâmia: IV–II milénios a.C. Ph.D. dissertation. New University of Lisbon. Black, J. and Green, A. 1992 Gods, Demons and Symbols of Ancient Mesopotamia. An Illustrated Dictionary. London: The British Museum Press. Bolger, D. and Maguire, L.C. 2010 The Development of Pre-State Communities in the Ancient Near East: Studies in Honor of Edgar Peltenburg. Oxford & Oakville: Oxbow Books. Bottéro, J. 1998 Religiosité et raison en Mésopotamie. Pp. 15–91 in L’Orient ancien et nous, eds. J. Bottéro, C. Herrenschmidt, and J.-P. Vernant. Paris: Hachette Littératures.

The Moon Watching Over the Sun and Venus

101

2004 Au commencement étaient les dieux. Paris: Hachette Littératures. Bottéro, J. and Kramer, S. N. 1989 Lorsque les dieux faisaient l’homme. Paris: Éditions Gallimard. Bouillard, H. 1974 La catégorie du sacré dans la science des religions. Pp. 33–56 in Sacré, ed. E. Castelli. Paris: Aubier-Montaigne. Braun-Holzinger, E. A. 1993 Die Ikonographie des Mondgottes in der Glyptik des III. Jahrtausends v. Chr. ZA 89: 119–135. Brisch, N. 2008 Religion and Power: Divine Kingship in Ancient World and Beyond. OIS 4. Chicago: The Oriental Institute of the University of Chicago. Campbell, S. 2010 Understanding Symbols: Putting Meaning into the Painted Pottery of Prehistoric Northern Mesopotamia. Pp. 147–155 in The Development of Pre-State Communities in the Ancient Near East: Studies in Honor of Edgar Peltenburg, eds. D. Bolger and L. C. Maguire. Oxford & Oakville: Oxbow Books. Civil, M. 1980 Les limites de l’information textuelle. Pp. 225–232 in L’archéologie de l’Iraq au Début de l’Époche Néolitique à 333 avant Notre Ère, ed. M.-T. Barrelet. Paris: CNRS Éditions. Collon, D. 1992 The Near Eastern Moon God. Pp. 19–37 in Natural Phenomena: Their Meaning, Depiction, and Description in the Ancient Near East, ed. D. J. W. Meijer. Amsterdam, New York: Royal Netherlands Academy of Arts and Sciences. Dalley, S. 2000 Myths from Mesopotamia: Creation, the Flood, Gilgamesh and Others. Oxford: Oxford University Press. Dumézil, G. 1958 L’idéologie tripartite des Indo-européens. Bruxelles: Latomus. Eliade, M. 1987 Le Sacré et le Profane. Paris: Gallimard. 1989 Traité d’histoire des religions. Paris: Payot & Rivages. ePSD 2006 The Electronic Pennsylvania Sumerian Dictionary. Pennsylvania: University of Pennsylvania. See http://psd.museum.upenn.edu/epsd1/nepsd-frame.html [Accessed: 2017/06/25]. ETCSL 2006 The Electronic Text of Sumerian Literature. Oxford: Faculty of Oriental Studies, University of Oxford. See http://etcsl.orinst.ox.ac.uk/ [Accessed: 2017/06/25]. Hall, M. G. 1985 A Study of the Sumerian Moon-god, Nanna/Suen. Ph.D. dissertation. University of Pennsylvania. Hatab, L. 1990 Myth and Philosophy. A Contest of Truths. La Salle, IL: Open Court Publishing Company. Hobsbawm, E. and Ranger, T. 1983 The Invention of Tradition. Cambridge: Cambridge University Press. Jacobsen, T. 1976 The Treasures of Darkness: A History of Mesopotamian Religion. New Haven/ London: Yale University Press.

102

Isabel Gomes de Almeida & Maria de Fátima Rosa

Katz, D. 2003 The Image of the Netherworld in the Sumerian Sources. Bethesda, MD: CDL Press. Klein, J. 2001 The Genealogy of Nanna-Suen and Its Historical Background. Pp. 279–301 in Proceedings of the XLVe Rencontre Assyriologique Internationale, Part 1: Historiography in the Cuneiform World, eds. T. Abusch et al. Bethesda, MD: CDL Press. Lenzi, A. 2007 Dead Religion and Contemporary Perspectives: Commending Mesopotamian Data to the Religious Studies Classroom. Method & Theory in the Study of Religion 19: 121–133. Nissinen, M. and Uro, R. 2008 Sacred Marriages: The Divine-Human Sexual Metaphor from Sumer to Early Christianity. Winona Lake: Eisenbrauns. Odisho, A. Y. 2004 The Akitu Festival in Mesopotamia: The Expression of Royal Ideology through Religion, Ritual and Architecture. Ph.D. dissertation. University of California, Berkeley. Oppenheim, A. L. 1976 Ancient Mesopotamia. Portrait of a Dead Civilization. Chicago: University of Chicago Press. Ornan, T. 2001 The Bull and its Two Masters: Moon and Storm Deities in Relation to the Bull in Ancient Near Eastern Art. Israel Exploration Journal 51/1: 1–26. Otto, R. 1966 The Idea of the Holy. New York: Oxford University Press. Richardson, S. 2010 On Seeing and Believing: Liver Divination and the Era of the Warring States. Pp. 225–266 in Divination and Interpretation of Signs in the Ancient World, ed. A. Annus. OIS 6. Chicago: Oriental Institute of the University of Chicago. Ries, J. 1992 Traité d’Anthropologie du Sacré I. Paris: Desclée. Selz, G. J. 2000 Five Divine Ladies: Thoughts on Inanna(k), Ištar, In(n)nin, Annunitītum and Anat, and the Origin of the Title “Queen of Heaven.” NIN: Journal of Gender Studies in Antiquity 1: 29–59. Seymor, M. 2011 Mesopotamia. Pp. 775–794 in The Oxford Handbook of the Archaeology of Ritual and Religion, ed. T. Insoll. Oxford: Oxford University Press. Stol, M. 1992 The Moon as Seen by the Babylonians. Pp. 245–277 in Natural Phenomena: Their Meaning, Depiction, and Description in the Ancient Near East, ed. D. J. W. Meijer. Amsterdam/New York: Royal Netherlands Academy of Arts and Sciences. Talon, P. 2005 Enūma Eliš. SAACT 4. Helsinki: The Neo-Assyrian Text Corpus Project. Thompson, J. E. S. 1970 Maya History and Religion. Oklahoma: University of Oklahoma Press. Veldhuis, N. 2004 Religion, Literature, and Scholarship: The Sumerian Composition “Nanše and the Birds.” CM 22. Leiden/Boston: Brill-Styx.

The Moon Watching Over the Sun and Venus

103

Westenholz, J. G. 1989 Enheduanna, En-Priestess, Hen of Nanna, Spouse of Nanna». Pp. 539–556 in DUMU-E2-DUB-BA-A: Studies in Honor of Ake W. Sjoberg, eds. A. W. Sjöberg et al. Philadelphia: Samuel Noah Kramer Fund. 1998 Goddesses of the Ancient Near East (3000–1000 BC). Pp. 63–82 in Ancient Goddesses: The Myths and the Evidence, eds. L. Goodison and C. Morris. London: British Museum Press. 2004 The Good Shepherd. Pp. 281–310 in Proceedings of the Fourth Annual Symposium of the Assyrian and Babylonian Intellectual Heritage Project, eds. A. Panaino and A. Piras. Milan: Università di Bologna. Winter, I. J. 1987 Women in Public: The Disk of Enheduanna, the Beginning of the Office of enpriestess and the Weight of Visual Evidence. Pp. 189–201 in La femme dans le Proche Orient Antique. RAI, ed. J.-M. Durand. Paris: Éditions Recherche sur les Civilizations.

Detecting Social Tensions in the Archaeological Record: Official and Vernacular Figurine-Making Traditions at Ur in the First Millennium BCE Anastasia Amrhein

University of Pennsylvania

Introduction The expansion of the Neo-Assyrian empire across the Middle East in the first millennium BCE has been studied largely through documents of the dominant cultural order, typically carved in stone and intended for posterity: including royal inscriptions; 1 infrastructural interventions into the landscape; 2 and monumental art. 3 Analyzing these canonical images of power, scholars have been able to reconstruct the material and ideological practices of domination deployed by Assyrian kings: that imperial expansion (supported by the gods) was the imposition of order onto an otherwise chaotic and uncivilized world; that the Assyrian army never suffered defeat; and that the supreme rule of the Assyrian king was inevitable and without rival. Although it has been recognized that the world-view presented by these documents is a very partial one 4—that Assyrian rule was neither assured nor absolute—it is still difficult to assess what Hans Belting has termed “mental colonization” 5—or the extent of penetration of imperial imagery and ideology into the minds and daily activities of the diverse peoples ostensibly conquered and Assyrianized. One way to approach this question is through the close study of clay figurines, which are found at nearly all archaeological sites. Neo-Assyrian and Neo-Babylonian 1. Royal inscriptions, for example, have been used to estimate the extent of the Neo-Assyrian empire by identifying areas where kings campaigned (intimidating polities into becoming vassals and tributaries), as well as the moment in time when specific territories entered the imperial fold through the installation of governors. More broadly, the rich textual corpus preserved from the Neo-Assyrian period has been instrumental in reconstructing various practical aspects of imperial governance. See, for example, Liverani 1988 and Postgate 1992. 2.  Such as waterworks and roads. See, for example, Wilkinson et al. 2005. 3.  Especially the architecture and relief decoration of palaces, as well as rock reliefs. See, for example, Porter 2001, and Winter 1983 and 1997. 4.  See, for example, the characterization of Assyrian imperial expansion as a communication network rather than a spreading oil stain: Liverani 1988 and Postgate 1992. 5.  Belting 2005: 302–304.

105

106

Anastasia Amrhein

foundation figurines and plaques unearthed in official structures such as palaces, temples, and priestly houses have been studied extensively. These objects, depicting apotropaic, divine beings, are, however, representative of practices originated by male scholar-priests, and as such, constitute only a small segment of the fig­urinemaking traditions that existed in Mesopotamia. Figurines that are the material remains of popular practices not documented in the written record—which predominantly depict women—have, however, received considerably less critical attention. As Richard Lesure has observed in the context of Meso-American archaeology, clay figurines are “points of reference in the negotiation and reproduction of actual social relationships”—which can be only revealed, however, through the consideration of entire assemblages. 6 The clay figurine assemblage from Ur presents a unique opportunity to investigate the power dynamics between different social and cultural groups and genders in the context of Assyrian domination.

Overview of the Archaeological Contexts Official clay figurines—associated with the domain of the palace and/or temple— dated to the period of Assyrian domination were excavated in the so-called area KP at Ur by Leonard Woolley during the 1924–25 and 1925–26 seasons of the joint expedition of the British Museum and the University of Pennsylvania (Fig. 1). 7 The figurines were likely deposited into the foundation of a building (no longer extant) during the time of Sîn-balāssu-iqbi, an Assyrian governor of Ur under Ashurbanipal, who carried out extensive restorations of the sacred precinct c. 650 BCE. 8

6.  Lesure 1997: 228. 7.  Woolley 1925b; Woolley 1926a; and Woolley 1926b: 691, 695ff. A detailed account of the figurine types and their findspots is presented in Woolley 1926b, along with a translation of the related incantation text, K.A.R. no. 298 by Sidney Smith. 8.  Archaeological evidence of Sîn-balāssu-iqbi’s restorations includes the Temenos wall; the Great Court of Nanna/Sîn and the ziggurat terrace wall; the Ningal Temple located southeast of the ziggurat; and area KP (the site of the Gipar and Ningal temple in pre-Kassite times); the É-dub-lá-mah temple and surrounding structures that were later incorporated into the Neo-Babylonian Gipar (located to the east of the earlier Gipar); the Nin-ezen temple on the city wall; as well as structures in the area northwest of the Ziggurat, including likely a temple or shrine of Nanna/Sîn. Some structures contained stamped bricks or cones of Sîn-balāssu-iqbi, while others are associated with the governor on the basis of brick size and type. Sîn-balāssu-iqbi’s (re)constructions seem to represent the only building activity between the Kassite king Kurigalzu I (c. 1400 BCE) and the Neo-Babylonian kings Nebuchadnezzar and Nabonidus (Woolley 1925a: 4ff; Woolley 1925b: 349, 358–359, 362, 368–375, 377, 381, 383–385, and 390; Woolley 1926a: 378–379; Woolley 1926b: 691; Woolley 1932: 384; Woolley 1939a: 53ff, 59ff, 89ff; Woolley 1962: 2–6, 11–15, 35; and Woolley 1965: 4, 15, 35ff, 93–94). In room 5 of the Neo-Babylonian Gipar/É-dub-lá-mah a cylindrical clay “pedestal” was found (U.2757) that was inscribed with copies of several inscriptions of Būr-Sîn, which had been found at Ur by Sîn-balāssu-iqbi in the course of his reconstructions of the É-kiš-nu-gál (the temenos enclosure). Woolley had posited that rooms ES 3, 4, and 5 functioned as a scribal school and “museum exhibit” at the time of En-nigaldi-Nanna, daughter of Nabonidus and entu-priestess of Nanna/Sîn (in room 5, for example, a cut-down diorite statue fragment with an inscription of Šulgi was also found). Additionally, the Assyrian governor’s restorations of the É-temen-ní-gùr (the ziggurat terrace) and the gate É-sag-dili of the É-dub-lá-mah, are detailed in an inscription on a stone door socket found inside the western side door of the outer chamber (room 1) of the Neo-Babylonian É-dub-lá-mah sanctuary (Woolley 1925b: 383–385, 396–397; pls. XL, 1; XLI, 1). Furthermore, according to brick inscriptions, the governor rebuilt the É-lugal-galga-si-sá, the ziggurat of Nanna/Sîn (Woolley 1962: 14ff, 111, pl. 29; Woolley 1965: 4).

Detecting Social Tensions in the Archaeological Record

107

Prior to the Kassite period (during the time of the Ur III and Isin-Larsa kings) area KP had been the Gipar (the residence of the entu-priestess), which also contained the Temple of Ningal (the consort of Nanna/Sîn). The function of this structure in Neo-Assyrian times is uncertain, however, as very little of the structure has been preserved. According to Woolley, the building’s location inside the Temenos points to an important government building. The better-preserved Kassite structure underneath was Fig. 1. Map of the site of Ur with main archaeological areas. perhaps a priestly residence—characterized by Woolley as “purely domestic [in] character”—although it also incorporated a sanctuary at the northwest end of the building (the Ningal temple had, however, been moved across the street to the northwest, closer to the ziggurat, by Kurigalzu I). 9 Penelope Weadock argues that despite the changes in layout, area KP retained its function as the Gipar during the Kassite period, and even into the Neo-Assyrian period, as female burials, presumably of entu-­priestesses, were found here. 10 While Nabonidus claims to have rebuilt the original Gipar in his texts, hardly any Neo-Babylonian architectural remains were found in area KP. Rather, it appears that the residence of Nabonidus’ daughter, the entu-priestess En-nigaldiNanna, was a structure located to the northeast of area KP that incorporated the É-dub-lá-mah shrine. Another possibility for the residence of Nabonidus’ daughter is the so-called Palace of En-nigaldi-Nanna, or the North Harbor Palace, evidently a residential structure (the largest excavated at Ur) that was located outside of the Temenos to the northeast, and has been compared to the “Great House” at Merkes, Babylon. This new construction was built by Nabonidus and was connected to a temple. Woolley had discovered in this building the inscribed bricks of Nabonidus mention­ ing the rebuilding of the Gipar. Such bricks had also been found, however, in the Neo-­Babylonian Gipar building that incorporated the É-dub-lá-mah. According to 9.  The Neo-Assyrian structure in area KP is cut by the great Temenos Wall built by Nebuchadnezzar c. 600 BCE. During the time of Kurigalzu, the gate entrance of the Ningal Temple was directly opposite that of the KP building (Woolley 1925b: 349–350, 371–375; Woolley 1926a, 366ff, 378–379; Woolley 1926b: 691ff; Woolley 1962: 11ff; Woolley 1965: 33ff; Zettler and Hafford 2015: 372; Weadock: 1975: 101, 111–113). Regarding the identification of the Kassite constructions with Kurigalzu I, see Brinkman 1969. 10.  Weadock also cites textual evidence of the continuation of the office of the entu-priestess in the Kassite and following periods (Weadock 1975: 111–112; cf. Woolley 1962: 11).

108

Anastasia Amrhein

Woolley, the latter was the “titular residence” of the priestess, containing a scribal school and possibly a museum, but was “scarcely adequate to the household of a princess.” Weadock observes, however, that Nabonidus’ inscriptions state that he built the residence for his daughter next to the (original) Gipar—thus not likely on the North Harbor, which she suggests was perhaps the residence of a governor or the king when he traveled to Ur. Richard Zettler and William Hafford also point out that the so-called palace may in fact simply be a large Neo-Babylonian domestic structure. 11 Following the fall of the Neo-Assyrian empire, clay figurines continued to be utilized in official contexts at Ur. Foundation deposits of clay figurines dated to the Neo-Babylonian period were found in the so-called palace of En-nigaldi-Nanna, or North Harbor palace, described above. 12 Figurines of a more “vernacular” nature (judging from their abundance through­ out the city) were found largely on or near the surface of the site. Since the levels associated with the Kassite and later periods were poorly preserved, in most cases, no firm dating or archaeological context is available for these figurines. Furthermore, since such figurines were not, as a rule, ritually deposited into the foundations of buildings, but rather, remained in use, their findspots are in most cases indicative of a secondary (rubble) context. Nonetheless, some of the primary contexts evidently include houses (in most cases roughly dated Kassite to Neo-Babylonian): located outside the northwest wall of the Temenos” / the Nebuchadnezzer Corner Fort (area NNCF); in area AH in the southeast of the site; in area NH (directly southwest of AH); and in area EM (Extra-Mural area outside the southwest Temenos Wall). Moreover, some figurines “were definitely associated with late temple sites,” according to Woolley—unfortunately, the excavation reports do not list all of the figurines with context, but merely “types.” 13 I have been able to examine a large number of the figurines housed in the Penn Museum, however, as part of my dissertation research, which has been greatly facilitated by the archival information pertaining to archaeological context available in the Ur-online database. The vast majority of the vernacular figurines, however, were found at Diqdiqqeh, a series of low-lying mounds located a mile and a half outside the city proper, to the north and northeast. Because this area had been cultivated, the archaeologists were not able to distinguish any stratification. The excavations yielded a large number of figurines ranging in date from the Ur III to the Persian period, based on comparanda. 14

11. Woolley 1931:374–381; Woolley 1962: 2, 18–19, 41ff; Woolley 1965: 34; Weadock 1975: 113; Zettler and Hafford 2015: 374, 381–382. 12.  Woolley 1962: 42, n. 1; Woolley 1965: 93–94. In addition, outside the entrance of room 6 in the Great Courtyard, two empty foundation deposit boxes were found that were made of inscribed Nebuchadnezzar bricks; boxes of similar, but un-inscribed bricks, likewise empty, were found at the entrance to room 5 (Woolley 1925a: 8). According to Woolley 1926b: 690, some earth was found in these boxes—likely the remains of unbaked clay figurines. 13.  Woolley 1962: 101–102. For area NNCF see Woolley 1932: 384ff, 390–391. For areas AH and NH, see Woolley 1931 (the latter is dated by tablet archives to the time of Nebuchadnezzar and Nabonidus). Also, see Woolley 1965: 77ff for areas AH and EM. 14. Woolley 1962: 101–102, pls. 26–-28; Woolley 1965: 91–93; Woolley 1925a: 18–20, pls. VII, 2; VIII, 1–2; Woolley and Mallowan 1976; Zettler and Hafford 2015: 368.

Detecting Social Tensions in the Archaeological Record

109

The Material Evidence from Ur in Broader Context During the 1924– 25 season, two brick foundation boxes were found in the area later labeled as KP (Fig. 2), each containing seven clay figures of fish-skin-clad apkallū sages, painted with black and red pigment (Type no. 1 in Woolley 1926b). Although the figurines quickly crumbled upon removal, according to Woolley they were comparable to Assyrian examples of this type. 15 These fig­ urines are not included Fig. 2. Neo-Assyrian architectural remains in Area KP (detail and locaon Woolley’s (1926b: tion within temenos wall). fig. 28) plan of area KP in the Neo-Assyrian period, with the foundation boxes labeled by figurine type (I have augmented this plan in Fig. 3 with images of the figurine types to make the arrangement visually accessible). Nonetheless, I believe that the fish apkallū were found in the area of KP that lay directly above the Kassite rooms 31 and 33, based on Woolley’s description of burnt remains and a thick layer of ash in this area—indeed, two boxes are shown on Woolley’s 1962 map of the entire KP area that fall outside of the Neo-Assyrian building wall remains, and are located slightly removed from the main group, to the north (Fig. 2). The main group of figurines were not recovered until the subsequent 1925–26 season, when more of the Neo-Assyrian building (and underlying Kassite building) in area KP was revealed—they can be placed above the Kassite rooms 59–62. During 1924–25 season, however, only the rooms along the northwestern and southwestern outer walls of the Kassite building (as far as room 61)—and thus the Assyrian remains on top—were excavated. 16 15. Woolley’s (1926b) figurine type no. 1 (Woolley 1925b: 374–375; Woolley 1965: 94; Woolley 1926b: 691, 693, pl. IX, 1) shows comparative images of two figurines currently in the British Museum: 91836 and 91837 (a group of six figurines are preserved, likely from a total of seven that were deposited together), which were almost certainly excavated at the North Palace at Nineveh and date to the time of Ashurbanipal; see Rittig 1977: 85. See Wiggermann 1992 regarding the identification of the various mythological beings represented by the figurines. 16.  The map of the Kassite building in KP (Woolley 1965: pl. 52) can be overlaid with the Assyrian remains (pl. 53 in the same publication). For the 1924–25 season, see the map fig. 4 in Woolley 1925b and compare with Woolley 1965: pl. 52. Of the post-Kassite period, Woolley notes in the preliminary report: “traces of wall…were not sufficient to afford a ground-plan” (hence the Assyrian foundation boxes are shown on the 1965 map without any accompanying walls). Woolley also comments on the “enormous quantity of ashes that buried” the Kurigalzu-period building “almost to the modern surface,” and posits that the remains above the burnt layer, belonged to a Neo-Babylonian or Persian building, placing the

110

Anastasia Amrhein

Fig. 3. Foundation figurine types found in Area KP at Ur.

During the 1925–26 season, a greater variety of apotropaic figurine types were discovered in the foundation of the Assyrian building in area KP. They were placed in boxes composed of three plano-convex bricks, with the fourth side open to the center of the room, and the top consisting of the pavement brick of the floor above. conflagration some time later than the Kassite but before the Neo-Babylonian/Persian period. The foundation boxes with the non-extant fish apkallū were found “near the surface and connected with the latest remains … underlying what had been the floor of a late building.” Compare these preliminary report observations with the final report (Woolley 1965: 37, n.2): “Over the whole area from the west corner of the [Kassite] building [i.e., room 10] to rooms 31 and 36 we encountered, immediately below the surface sand, a very heavy bed of ashes.” In the final 1965 report (published after Woolley’s death in 1960) the latest remains are likewise associated with the Neo-Babylonian period, although the editors of the volume note the incomplete and contradictory nature of Woolley’s manuscripts and notes. However, in the 1926b publication of the figurines, Woolley connects these figurine finds with those excavated in area KP during the 1925–26 season—and thus all of the figurines and the architectural remains to the Assyrian period. In fact in the 1926a preliminary report for that season, Woolley states that post-Assyrian remains in area KP constituted only the temenos wall and “one or two terracotta drains.” Furthermore, in the 1925b report, Woolley describes the excavation of several graves in the top level of KP, one of which was a female burial in a “clay sarcophagus with one flat and one oval end” with a bronze and glazed frit vessels, bronze bracelet, gold necklace, and sardonyx beads. Additional female burials were found the following season and associated with the Neo-Assyrian period as they cut across a wall built by Kurigalzu I; they featured comparable grave goods (agate and gold beads, gold earrings, bronze and glazed vessels). Thus, the fire must have occurred sometime after Kassite occupation, but before the Neo-Assyrian rebuilding of area KP. Weadock suggests that the women buried here were likely entu-priestesses. See Woolley 1925b: 349–350, 371–375; Woolley 1926a: 378–379; Woolley 1926b: 691ff; Woolley 1962: 11ff; Woolley 1965: 33ff, 94; Clayden 1995: 61–63; Zettler and Hafford 2015: 372; Weadock: 1975: 101, 111–113.

Detecting Social Tensions in the Archaeological Record

111

Each box contained a single figurine. The boxes were arranged in a specific order around the perimeter of the walls and/or at door jambs: either alternating or in pairs, “two of a sort coming next to each other, or two of one sort alternate with two of another.” 17 This figurine assemblage included stand­ing bearded figures (Fig. 4; Woolley’s Type no. 3), either holding vases (originally with “wavy black lines painted on the white ground of the garment [to] represent water”) or a staff—in either case, their hands are stacked right on top of left. 18 Dessa Rittig categorizes these Ur figurines as deities, separate from other vase and staff bearers, some of which she interprets as the lahmu, or the hero with six hair locks, and others simply as human men. It is likely, however, that all of these figurines, although variously stylized, are intended to represent the same being—a divine apotropaic being, judging from their deposition in the foundations of official buildings. Stylistically similar examples have been excavated in the priestly houses at Assur dated to the eighth and seventh centuries. 19 Examples from both sites have roughly the same proportions and appear to be wearing horned headdresses; the bodies are modeled in the round. The assemblage furthermore included several types of lion-demons (ugallū) (Fig. 5; Woolley’s nos. 7–8) brandishing a weapon in the right hand, with the left hand held at the chest. These figurines were originally painted entirely black. 20 Although very poorly preserved, they are generally comparable to examples from official contexts at Khorsabad, Fort Shalmaneser, and Assur, found in palaces and priestly houses, and some likewise dated to the seventh century. 21

Fig. 4. Standing bearded figures from Ur.

Fig. 5. Lion-demon figurines from Ur.

17.  Woolley posited that the bricks of the boxes “must have been obtained for the purpose by excavation on early sites” or possibly made intentionally in an archaizing style (Woolley 1926a: 378–379; Woolley 1926b: 691ff; Woolley 1965: 93–94. Harriet Crawford has suggested that these types of bricks may have held “special powers themselves” (Crawford 2015: 119). 18. The boxes follow a different numbering scheme in the 1926b and 1965 publications. I have deferred to the earlier publication as it was published by Woolley during his lifetime (Woolley 1965: 94, no. 2, U.6767 A–G, pls. 33–34; Woolley 1926b: 694, pl. IX, no. 3). A selection of these are preserved in the Penn Museum: B16270a-f. This figurine type is incorrectly identified as “Papsukkal” by Woolley, following Koldewey. See Wiggermann 1975 and below regarding the so-called Papsukkal or Ninšubur figure. 19. See Rittig 1977: 49–65. Assur comparanda: for example, nos. VA 8191, 8194, 9197 (in the Vorderasiatisches Museum, Berlin), published in Klengel-Brandt 1968 and in Rittig 1977: 59–62, 64–65. See also other, stylistically different, examples of the ninth through seventh centuries from palatial contexts at Nineveh and Nimrud in Rittig 1977: 56, 62–64. Regarding the priestly houses at Assur, see Preusser 1994 and Nakamura 2008. 20.  Woolley 1965: 94, no. 3, U.6768 A-D, pl. 33; Woolley 1926b: 695, pl. XI, 7–8. In the Penn Museum: B16272A-B. 21.  N8287 / Nap. 3427 (in the Louvre) from the Palace of Sargon at Khorsabad. ND 8190 from Fort Shalmaneser, seventh century context. VA Assur 3603 (in the Vorderasiatisches Museum, Berlin) from a priestly house at Assur. Rittig 1977: 102–110. See also Klengel-Brandt 1968.

112

Anastasia Amrhein

Fig. 7. Serpent figurines from Ur.

Fig. 6. Mušhuššu dragon figurine from Ur.

Mušhuššu dragons (Fig. 6; Woolley’s no. 9) 22 from area KP at Ur are very similar to an example excavated at Fort Shalmaneser, Nimrud, dated to the eighth–seventh century. 23 Figurines of serpents (Fig. 7; Woolley’s no. 10)  24 from Ur are comparable to those found in a priestly house at Assur, dated to the eighth–seventh century, 25 as well as those found in the bottoms of Neo-Babylonian boat models at Babylon. 26 The bull-man (kusarikku) (Fig. 8; Woolley’s no. 5) fig­ urine from Ur is described by Woolley as a “nude male figure wearing a high (apparently horned) headdress with long hair falling over the shoulders and a curled bearded”; with bull’s legs and a long tail. The figure’s right hand is held at the chest, while the left one is lowered. 27 Similar figurines have been found in the eighth–seventh century priestly houses at Assur and Sargon’s palace at Khorsabad. 28 Furthermore, some of the Ur figures from area KP possibly represent the scorpion-man (girtablullû) (Woolley’s nos. 4 and 6). They are described by Woolley as nude bearded male figures: U.6773–4 (Woolley’s no. 4) is human except for a tail, wearing a horned headdress; his hands are folded at the chest, possibly holding a vessel. Another example of this type, B16273 in the Penn Museum, clearly has bird feet (which are a join, B16270e), however, and a short, curled,

Fig. 8. Bull-man figurine from Ur.

22.  Woolley 1965: 94, no. 4, U.6769–70, pl. 34 (described as “dogs or possibly dragons with dogs’ bodies and snakes’ heads.”); Woolley 1926b: 695, pl.XI, no. 9. In the Penn Museum: B16274; also a broken head: 31-16-967. 23.  ND 8194 published in Rittig 1977: 114–115. See also Klengel-Brandt 1968. 24.  Woolley 1965: 94, no. 5, U.6771, pl. 34; Woolley 1926b: 695, pl. XI, no. 10. In the Penn Museum: B16275a,b, and a broken head: 31-43-327. 25.  For example, VA 8197, 8198, 8192 (in the Vorderasiatisches Museum, Berlin) published in Rittig 1977: 122. See also Klengel-Brandt 1968. 26.  See Reuther 1926: fig. 35. 27.  Woolley 1965: 94, no. 6, U.6772, pl. 33; Woolley 1926b: 694, pl. X, 5. 28.  For example, VA Assur 3601 (Vorderasiatisches Museum) from Assur and BNF A.391 and A. 391.2 (in the Bibliothèque Nationale, Paris) from Khorsabad. A further example is known from the Burnt Palace at Nimrud but is unpublished. See Rittig 1977: 98–99; see also Klengel-Brandt 1968. For the Khorsabad figurines, see Botta and Flandin 1849: pl. 153a, c; Botta and Flandin 1850: 169; and Barrelet 1968.

Detecting Social Tensions in the Archaeological Record

113

pig-like tail (Fig. 9). Woolley’s no. 6 is described as human above the waist and a bird below the waist, wearing a horned headdress; his right arm extended out to the side, while the left is broken below the shoulder. The latter featured details rendered in black pigment. 29 Comparable examples are not known to me from the Assyrian corpus. Finally, the Ur assemblage includes bird-headed apkallū sages (Fig. 10; Woolley’s no. 2), who hold a bucket in the left hand and raise the right hand to the chest. 30 While bird-headed apkallū are well known from Assyrian sites, from palaces and priestly houses of the ninth through seventh centuries, the examples from Ur are unique in that they are modeled in the round, whereas Assyrian examples are all mold-made—either as flat figurines or plaques. 31

Fig. 9. Scorpion-man figurine from Ur.

Fig. 10. Figurines of bird-headed apkallū sages from Ur.

Both sets of the foundation deposit figurines from area KP, those found during the 1924–25 and 1925–26 seasons, were coated with “white lime” and many showed the remains of black and/or red paint. Additionally the bones of small animals or birds as well as grain and a broken piece of pottery were contained inside the boxes. 32 In addition, clay dog figurines, both sitting and standing, were reportedly found in area KP. These are not, however, included in Woolley’s mapping of the foundation boxes by type, because they were placed directly into the earth under the floor. 33 Unfortunately, none of the Neo-Assyrian dog figurines are published nor are their current whereabouts known. However, clay dog figurines were also 29.  Woolley 1965: 94, nos. 7–8, U.6773–5 pl. 33; Woolley 1926b: 694–695, pl. X, 4, 6. In the Penn Museum: B16273. 30.  Woolley 1965: 94, no. 3, U.6768D, pl. 34; Woolley 1926b: 691, 693–694, pl. IX, no. 2. In the Penn Museum: B16271A-C. 31.  For example, 54.117.26 (in the Metropolitan Museum of Art), a flat figurine from the Burnt Palace at Nimrud (reign of Adad-nirari III) and plaques from the priestly houses at Assur (such as VA Assur 3606 in the Vorderasiatisches Museum, Berlin), Rittig 1977: 70–73. See also Klengel-Brandt 1968. 32.  Woolley 1925b: 375; Woolley 1926a: 378–379; Woolley 1926b: 692ff; Woolley 1965: 93–94. 33.  Woolley 1965: 94, 106; Crawford 2015: 119. In Woolley 1965: pl. 34, U.6769–70 are described as standing dogs but in reality are mušhuššu dragons. This must have been a mistake in this posthumous publication. In Woolley 1926b, these figures are correctly identified as mušhuššu dragons. However, dogs were evidently present as well, based on Woolley’s description, because the mušhuššū are in fact placed on the plan of KP as foundation boxes labeled with the number 9 (fig. 28 in Woolley 1926b). The boxes follow a different numbering scheme in the 1926b and 1965 publications. As stated above, I have deferred to the earlier publication.

114

Anastasia Amrhein

found in the Neo-Babylonian North Harbor palace—two of the five examples from foundation boxes under room 4 are currently in the Penn Museum (Fig. 11). 34 Dogs were also among the figurines that were found in the eight brick boxes placed beneath the pavement of a small postern gate on the north side of the North Harbor palace. 35 Although the clay dog figurines of the Neo-Assyrian Fig. 11. Dog figurines from Ur. and Neo-Babylonian periods cannot be fruitfully compared, dogs in copper/bronze were also found in both Neo-Assyrian and Neo-Babylonian contexts at Ur and were reportedly closely comparable to each other. 36 The use of dog figurines may, thus, have been introduced by the Assyrians. The Neo-Babylonian dog figurines that were found under the door jambs of room 4 of the North Harbor palace were originally each painted with a different color, including red, blue, and green (four dogs were placed on one side of the entrance, and five on the other). 37 This arrangement is very similar to the set of five mastiffs found in the North Palace of Ashurbanipal (beneath door D, Room S), likewise painted with black, red, white and blue. 38 The Neo-Assyrian dogs are standing however; thus, the examples from the so-called library at Kish (dated to the reign of Nebuchadnezzar) are more closely comparable to those from Ur. 39 The boxes under the gate of the North Harbor palace also contained “painted mud Papsukkal figures,” and an additional three foundation boxes were found at the entrance to room 4, and under other doorways. The latter had been disturbed, how­ ever. Woolley also does not provide U. numbers nor images for these figurines, simply stating that they are discussed in Volume VIII of the final reports. 40 Since Woolley referred to foundation figurines of all types as “prophylactic Papsukkal fig­ures,” it is uncertain exactly what types of figurines were found in the Neo-­Babylonian context 34.  The clay seated dogs published in Woolley 1965: 94, 106, pl. 32 (U. 16159 A-E) all come from the North Harbor Palace and are dated to the Neo-Babylonian period, although they are discussed in the context of Neo-Assyrian figurines. See also Ur-online database. In the Penn Museum: 31-43-488 and 31-43-489. 35.  Woolley 1962: 42, n. 1. 36.  In the Ningal temple, a copper/bronze dog figurine was found “in-situ” (but without a foundation box) beneath the floor of room 7. Woolley suggests that this deposit is to be dated to the Neo-Assyrian restoration of the temple as it was found under the lower of two pavements (Woolley 1925b: 369; Woolley 1939a: 64; Woolley 1965: 94, 103 (U.3372), cf. pl. 28 (U.2853). Comparable bronze dog figurines (some with gold leaf), accompanied by a bearded vase-bearer in the same material, were found in the NeoBabylonian Gipar/É-dub-lá-mah building under the Nabonidus floor, also without boxes (in the doorway between ES 1 and ES 3, and in room ES 4, where a number of school exercise tablets were found as well as artifacts of disparate time periods, leading Woolley to interpret this area as a museum and scribal school)—U.2853, U.2867 (87-28-17), U.2963, U.3107, and U.2854 (Woolley 1925b: 383, pl. XL, 2; Woolley 1962: 16–17, 111, pl. 25). The vase bearer is comparable to Neo-Assyrian period terracotta examples—see Woolley’s type no. 3 above—suggesting that this Neo-Assyrian figurine type lived on in the post-Assyrian period, at least in this medium. 37.  Woolley 1965: 94, nos. 9–10, pl. 32; U.16159–60. In the Penn Museum: 31-43-488 and 31-43-489. 38.  British Museum nos. 30001–30005. Rittig 1977: 116. 39.  AN 1924.302–4 (in the Ashmolean Museum). Rittig 1977: 118. 40.  Woolley 1962: 42, n. 1.

Detecting Social Tensions in the Archaeological Record

115

besides the dogs, as these are the only types mentioned in Volume VIII as originating from the North Harbor palace. One figurine that does appear in the publication, albeit without explicit context, is recorded as originating from the northeast gate of the North Harbor palace in the Ur-online database: this is a figure of a protective bearded deity in a smiting position, with the right hand raised, originally hold­ ing a weapon, and the left at his side (Fig. 12). He is outfitted with a belt, harness, and sword of copper, and is possibly wear­ ing a horned headdress. 41 Miniature copper/bronze weapons (U.17060, U.17432) were also found in foundation boxes in the northwest gate of the North Harbor palace. 42 These likely belonged to similar clay protective deities who guarded the entrance to the structure. Comparable fig­urines Fig. 12. Figurine of a are known from Sargon’s palace at Khorsabad and the protective bearded deity priestly houses at Assur, as well as other Neo-Babylonian- from Ur. period sites, including the so-called library at Kish and the palace of Nabopolassar at Babylon. 43 Overall, the dense, particular arrangement of the Ur fig­urines is comparable to the finds in Neo-Assyrian palaces as well as the priestly houses at Assur. 44 The practice of depositing apotropaic clay demons and sages beneath the floors of official buildings other than temples is an Assyrian practice that appears to have been adopt­ed in Babylonia for a short while during the period of Assyrian domination, and even then with certain idiosyncrasies, as described above. The bearded vase or staff bearer figurines (Woolley’s type no. 3) are a good example of the melding of Assyrian and Babylonian traditions at Ur during the time of Assyrian domination. While the Ur examples find the closest stylistic comparanda with figurines from Assur, they are at the same time closely related in theme to the various Neo-Babylonian fig­urines of vase-bearers, often found in temples (see below). These figurines, however, were never buried as foundation deposits in Babylonia, but rather appear to have functioned as votives. 45 Unlike the Assyrian examples, and those from Ur, these votive vase-bearer figurines are also oftentimes beardless. Unfortunately, the Babylon­ian examples are not as well dated as the Assyrian ones, but it is unlikely that they appeared only after the time of Assyrian domination, as findspots at Babylon range in date from the ninth through sixth centuries. 46 The Assyrian examples (in a style different than that of the Ur figurines) can also be dated as far back as the

41.  Woolley 1965: 94, no. 11, U.17075, pl. 34. Currently in the Penn Museum: 31-43-490. 42.  Woolley 1962: 124, 126. 43.  For example, N8281 from Khorsabad (in the Louvre); AN 1924.701 from Kish (in the Ashmolean Museum); VA Bab 44 400 from Babylon and VA 5473 from Assur (both in the Vorderasiatisches Museum), Rittig 1977: 44–48. 44.  See, for example, Mallowan 1954, 1966; Preusser 1954; Nakamura 2008. 45.  For example, figurines from a Neo-Babylonian priestly house near the Eanna temple at Uruk: VAM 11740, W13740, VAM11618, in the Vorderasiatisches Museum, Berlin (Van Buren 1931: 38; Barrelet 1968: 121). See also other examples from Uruk dated to the first half of the first millennium, such as VAM 11680 (Ziegler 1962: 164–165). Some examples from Babylon were found in the Ninurta temple and Ishtar temple; see Klengel-Brandt and Cholidis 2006: pls. 22–23 46.  Klengel-Brandt and Cholidis 2006; Pruß 2013: 608.

116

Anastasia Amrhein

ninth century. 47 It appears, thus, that an image well known to the Babylonians was utilized in a different context at the time of Assyrian domination. According to Woolley, all of the foundation figurines that he found at Ur were unbaked (“mud”), “sometimes hand modeled, sometimes cast in a mold.” 48 Having closely studied the examples in the Penn Museum, it is clear that all of the figurines are indeed unbaked, which is also the case for the majority of foundation figurines and plaques from other sites, both Assyrian and Babylonian. This is fitting for the figures’ operation against demons beneath the earth. According to Carolyn Nakam­ ura, “as a mixture of water and earth,” clay “locates a likely border … between the earth and the Apsu”—a very liminal area where demons can gain access to the human world. 49 Most of the Ur figurines, however, appear to have been hand-modeled (or at least heavily augmented by hand), as they are fully three-dimensional. Some do have molded or stamped faces, which is most evident in the case of the vase or staff bearers. While some three-dimensional Neo-Assyrian foundation figurines do exist (such as the vase or staff bearers, lion demons, and bull-men cited above), proportionally, the majority of the known Assyrian examples are mold-made—typically rectangular, flat plaques. 50 Figurines from the reign of Adad-nirari III (810–783 BCE) are unique in that they are essentially molded relief plaques, even though the background has been cut away and some features have been minimally articulated on the back of the “figurine” through the use of pigment. 51 No comparable moldmade plaques are known from Ur, nor greater Babylonia with the exception of one unique example of a plaque from Babylon that is incised and/or molded by hand. This example is associated with Esarhaddon’s rebuilding of the Temple of Marduk. 52 The limited number of hand-modeled examples in the round seem to be restricted to the eighth and seventh centuries in Assyria, and may have been inspired by the Babylonian iteration of these figures, as seen at Ur. It is not surprising that the cultural exchange of ideas and images should travel in both directions—even in the context of imperial domination—and that some of the features of the foundation figures were in turn adopted in Assyria. During the Neo-Babylonian period, the Assyrian figurine tradition seems to have been pared down, and rulers yet again concentrated on protecting the temples, 47.  For example, 54.117.24 (in the Metropolitan Museum of Art), from the Burnt Palace at Nimrud (reign of Adad-nirari III), which is mold-made, flatter, and more angular and schematic in its representation of the figure’s body and hairstyle; published in Rittig 1977: 64 (3.2.23-24). See other ninth through seventh century examples from Nineveh and Nimrud in Rittig 1977: 56, 62–64. 48.  Woolley 1925b: 375; Woolley 1926a: 378–379; Woolley 1926b: 692ff; Woolley 1965: 93–94. 49.  Nakamura 2008: 134. 50.  See Klengel-Brandt 1968; Rittig 1977; Barrelet 1968. 51. For example, 54.117.25 and 54.117.26 (in the Metropolitan Museum of Art) from the Burnt Palace (810–783 BCE); published in Rittig 1977: 64, 71. See also Mallowan 1954. 52.  VA Bab 2354 (in the Vorderasiatischs Museum, Berlin), found under the door of the Ea sanctuary. See Koldewey 1911: 44, fig. 63; Rittig 1977: 99 (11.2.1); Klengel-Brandt and Cholidis 2006: no. 696. In addition, a very fragmentary plaque (U.367, B15295) of unbaked clay, comparable in technique (likewise incised), was found by Woolley in room 10 of the Neo-Babylonian É-nun-mah at Ur, under the shell mosaic floor, which was dated to the Persian period. This example shows the profile of a bearded male figure wearing a horned headdress, and perhaps holding a staff. The exact context of this piece is unclear from Woolley’s description: it may have been a ritual deposit of the Persian period, or may have been found in the fill underneath the Persian-period pavement, which consisted of broken mud bricks. (Woolley 1962: 23ff, 106, pl. 31).

Detecting Social Tensions in the Archaeological Record

117

in keeping with southern Mesopotamian tradition—however, the use of dog fig­ urines, both in clay and metal, as well as protective deities brandishing weapons, may have been Assyrian traditions that continued into the Neo-Babylonian period in palatial contexts, as suggested by the archaeological evidence discussed above. 53 The hand-modeled god with staff—previously known as Papsukkal, but more recently identified by Wiggermann as the messenger god Ninšubur—is frequently found in Neo-Babylonian temples, located exclusively beneath the cult statue, facilitating communication between the divine and human realms. 54 This practice is not known from Assyrian sites in the Neo-Assyrian period. However, one example of this appar­ently Babylonian tradition is dated to Aššur-rēša-iši I or II (1132–1115 or 971–967 BCE, from the Ištar temple at Aššur), briefly adopting a tradition that had perhaps originated in Babylonia during the Kassite period—specifically at Ur, where a fragment of a figurine bearing an inscription typical for Ninšubur was found beneath the cult statue in Kurigalzu’s Ningal temple. 55 53.  See Van Buren 1931; Ellis 1968; and Rashid 1971. 54.  Ellis 1967; Wiggermann 1987. 55.  Ellis lists a figurine from the Ningal Temple at Ur as the earliest example of the Papsukkal, subsequently identified as Ninšubur (Ellis 1967). The unbaked, inscribed figurine of Ninšubur was found with a gold wire (likely the figure’s staff) and fragments of embossed gold leaf (perhaps a diadem). The figurine was part of a deposit found in a baked brick cult statue base located at the back the sanctuary (room 8), in a niche opposite the entrance, dated to the Kassite period by Woolley. Tim Clayden has questioned the dating of this deposit to the Kassite period, and proposes a date in the eighth or seventh century BCE, dated to the time of Assyrian restoration of the temple, citing doubts by Ellis as well as Rittig 1977: 20–21, 41, and the Neo-Babylonian comparanda. The brick base had two phases of construction, as noted by Woolley. Clayden observes that two different types of bricks were used: in Phase One: 0.33 × 0.33 × 0.09 m and added in Phase Two: 0.395 × 0.07–0.08 m. According to the excavation cards, Clayden notes that the deposit was found in the Second Phase. He states that while the first phase is clearly associated with Kurigalzu I, “the date of the Phase Two alter is less easily determined.” Following Woolley’s observation that, by the seventh century, the Kassite “walls had fallen, or had been intentionally razed, down to two or three courses” (Woolley 1925b: 371–372), Clayden reasons that the structure “was so ruined that it had to be rebuilt rather than repaired.” Nonetheless he observes that if any repairs had been made, “they would have been concentrated in the cultic heart of the building—the altar in room 8,” thus dating the altar “any time between the fourteenth and seventh centuries” (Clayden 1995: 63–64). Indeed, Kassite restorations post-Kurigalzu have also been observed in area KP at Ur (Woolley 1965: 35), and are thus also possible for the Ningal temple. Clayden’s argument is, thus, ultimately inconclusive. According to Woolley, in the Ningal temple, “many of the bricks both in the walls and in the pavements bore the stamp of Kurigalzu.” Furthermore, the paving of room 8 was composed of 0.41 m. sq. bricks, was consistent with other Kassite bricks from other rooms of the temple, some of which were stamped with a text of Kurigalzu: 0.41–0.45 m. sq., typically baked and covered with bitumen (cf. Sînbalāssu-iqbi’s pavement bricks measuring 0.255–0.26 m. sq. × 0.07 m). Woolley furthermore notes that “the lower building lay about a meter and a half beneath the foundations of Sîn-balāssu-iqbi’s and was quite distinct from it.” In addition, according to Woolley, several aspects of room 8 were retained from the Kassite period, such as the northeast wall of room 8, and northwest wall of the whole temple (Woolley 1925b: 371–372; Woolley 1939a: 53ff, 57, 60ff; Woolley 1965: 32ff, 103, U.3304–8). The dimensions of the Phase One bricks are comparable to other Kassite period bricks from the site, especially from the reign of Kurigalzu: 0.33–0.44 m. sq. × 0.085 m. (Woolley 1965: 52); 0.33–0.34 × 0.08–0.09 m; 0.31–0.33 m. sq. × 0.07–0.09 m. (Woolley 1939a: 81, 92). However, the closest comparanda for the Phase Two bricks are the bricks of Ur-Namma (not Sîn-balāssu-iqbi): 0.39 m. sq. × 0.07 m. (Woolley 1939a: 81); 0.395 m. sq. × 0.07–0.075 m. (Woolley 1939a: 80). The bricks of Sîn-balāssu-iqbi, on the other hand, typically measure: 0.30 × 0.27–0.28 × 0.09 m.; 0.33–0.44 m. sq. × 0.085 m (Woolley 1965: 56); 0.30 m. × 0.15 m. (Woolley 1962: 20); 0.33–0.34 m. sq. × 0.11 m (Woolley 1939a: 81). Inscribed bricks of the governor measure 0.255–0.60 m. sq. × 0.070–0.075 m., or 0.120–0.125 m. × 0.255–0.260 m. × 0.070–0.075 m. for half-bricks (for example: B16483, U.3161, U.3296 in the Ur-online database).

118

Anastasia Amrhein

Judging from the discovery of foundation boxes filled with loose earth and tiny metal weapons, Neo-Assyrian temples did feature clay figurines, although not Ninšubur, the god with the staff. 56 Such boxes are never found directly beneath cult statues and thus must have depicted protective deities brandishing weapons, similar to an example from a priestly house at Assur (cited above). The primary function of clay figurines in Assyria was to protect the king—and the scholar-priests who worked towards this same purpose—as indicated by the large number and variety of figurines found in palaces and priestly houses. At Khorsabad and Nimrud, for example, boxes of figurines were found deposited around the perimeter of the king’s throne. 57 Indeed, a protective god brandishing a weapon was found in a foundation deposit at the palace at Khorsabad (see above). Such figures were also a common feature of Neo-Babylonian buildings, as suggested by the Ur example from the North Harbor palace—but interestingly, none of the Babylonian examples originate from temples (see above). This figurine type may have been another Assyrian import. As stated in the introduction, the use of clay figurines was however, not exclusively a royal prerogative. In fact, the appearance of clay foundation deposits beginning in the Kassite period can be seen as the appropriation of a vernacular, magical material, since earlier Mesopotamian foundation deposits were typically made of metal. Popular clay figurine types have been consistently geographically defined since the late third / early second millennium BCE: hand-modeled, bell-­ shaped figurines predominated in northern Mesopotamia and Syria, while moldmade examples were prevalent in southern Mesopotamia. 58 These distinctions persist into the first millennium. The vernacular figurines from Ur include a wide range of mold-made female figures (Fig. 13)—this is the most abundant type of figurine found at Diqdiqqeh, for example. The women are all depicted frontally and may be dressed but are typically nude and standing, although occasional examples appear seated in a cross-legged position. They frequently hold infants (much more often than in the second mil­ It must also be considered that bricks from earlier periods were frequently reused in the sacred precinct, including bricks of Ur-Namma in the period of Kurigalzu (see, for example, Woolley 1965: 7, 64, 67; Woolley 1939a: 126; for mixed bricks in the Ningal temple, Woolley 1939a: 55), and also that foundation deposits were likely to be reburied when discovered by subsequent rebuilders. In any case, whether Kassite period or later, the Ninšubur deposit can be understood as a southern Mesopotamian cultural phenomenon as it is unknown in Assyria except for the single Middle Assyrian example. It is of interest to note that another Kassite foundation deposit box, “made of Ur-Namma bricks” was found in room 2 of the Ningal temple, against the northeast jamb of the door. Inside were only a lapis lazuli bead and some gold leaf. This box was placed “in a hole dug down into the Third Dynasty mud-brick platform, [and] touched and rose above the footings of the Kurigalzu wall” (Woolley 1939a: 55–56). A further foundation deposit box, albeit empty, was found in a narrow room (20) located at the back of the central gate in the southwest wall of the Great Courtyard. This wall, decorated with half-columns, can be securely dated to Kurigalzu I via stamped bricks. The function of this room is unclear, but according to Woolley, the whitewashed floor and recessed back wall may suggest a shrine, where, “a statue of the god might stand to supervise the…business transacted” in the Great Courtyard. The box of fired red bricks was found below the foundations of the back wall (Woolley 1925a: 4–9; Woolley 1925b: 364–366; Woolley 1939a: 48ff, 88ff). If indeed the space was a shrine, the location of the box would suggest a Ninšubur figurine, directly beneath a cult statue. 56.  See Van Buren 1931; Mallowan 1966: 266, 292; Oates and Oates 2001: 112–113, n. 19, 116–119, 130–132; Loud 1936: 101, 107–108, 111–112, 119, 125; Loud 1938: 62–63. 57.  Oates and Oates 2001: 124–126, n.46, 253; Oates and Reid 1956; Loud 1936: 62. 58.  See Klengel-Brandt 1978; Barrelet 1968; Pruß 2013; Woolley and Mallowan 1976; McCown et. al. 1967.

Detecting Social Tensions in the Archaeological Record

119

Fig. 13. Female figurine types found at Ur compared to those found at other southern Mesopotamian sites.

lennium) or clasp their hands below the breasts; less frequently, they support their breasts with their hands or hold their hands by their sides. The figurines are typically highly naturalistic, with round faces, a row of curls along the forehead indicated by scallops, and tight rows of curls on top of the head receding back from the hairline (often described as a melon-like ribbing); their hair falls on the shoulders in neat, bead-like rows of curls. Sex is marked through the presence of breasts, but the pubic triangle is rarely indicated in contrast to the third and second millennium figurines. Although likewise mold-made, the Neo-Babylonian examples are in higher relief, more three-dimensional than the Old Babylonian plaques depicting frontal nude females; the background has also been cut away around the first millennium figures, creating a more three-dimensional figurine rather than a plaque. The figures have slightly fuller bodies than the lean, elongated Old Babylonian examples. Creases of on the abdomen are frequently rendered, perhaps indicating that the woman has recently given birth. The quality of the mold-made impressions greatly varies, however, effacing many of these details on some examples. Some stylistic variations are also evident: for example, some figures wear hairstyles that are more Egyptian in appearance; some have more schematized bodies and squat proportions. Highly comparable examples—some perhaps even fashioned from the same molds—have been found at other Babylonian sites including Kish, Tello, Babylon, Uruk, Larsa, and Nippur. Such figurines were evidently produced consistently through the Neo-Assyrian and Neo-Babylonian periods (ninth–sixth centuries BCE) without any detectable evolution in style or iconography. 59 Thus, figurines produced under Assyrian domination cannot be distinguished from those of the Neo-Baby­­lonian 59.  Klengel-Brandt and Cholidis 2006: 80–130, pls. 16–21; Ziegler 1962: 165–166, pl. 15–18; For examples in the Louvre, see Barrelet 1968: nos. 407, 452 (from Tello); nos. 589, 592–95 (from Sinkara); as well as nos. 805, 820–826 (purchased); McCown et. al. 1967; Pruß 2013; Koldewey 1911.

120

Anastasia Amrhein

period. Some of the female figurines (especially at Ur and Babylon) were found in temples. Most of the female figurines, however (both at Ur and other Babylonian sites), are simply found scattered throughout the urban areas—in areas that in some cases have been identified as domestic—as well as in secondary contexts. Interestingly, the only known representations of women that served as foundation deposits come from Ur. One figurine—“a woman suckling a child”—was Fig. 14. Female figurine from found in a brick box against one of the door jambs Ur. underneath the pavement of the Nabonidus Gate of the Temenos Wall (the other box contained a diorite socket-stone with an inscription of Nabonidus referencing the Gipar). 60 Another figurine, which I have examined in the Penn Museum (Fig. 14), is stated to have come from a brick box below the Neo-Babylonian pavement of area NNCF, outside the northwest wall of the Temenos at Ur; it was accompanied by another such figurine, which is not preserved. According to Woolley, this figurine was made using a double mold, but in actuality, upon closer inspection, only the face is mold-made. The rest of the figurine is very finely hand-modeled, fully in the round, and some features are finished by hand: the hair seamlessly continues on the sides and back of the figurine, completed by hand incising to match the mold-made representation on the front; the ears also appear to have been finished by hand. According to Woolley, the figurine was paint­ed white with black hair and eyes. 61 Another, mold-made female figurine of “a naked woman suckling an infant” was found in room ES 4 of the Neo-Babylonian Gipar/É-dub-lá-mah building. Although this figurine was evidently not found as a foundation deposit, it is another example of a female figurine that appears in an official context. Woolley had interpreted rooms ES 3, 4, and 5 as a scribal school and museum of sorts, as they contained a variety of artifacts from disparate time periods. In room ES 4, for example, a foundation cone of the Larsa king Kudur-Mabuk as well as Ur III tablets were found. 62 A further glimpse into the status and meaning of the female image in Babylonia, specifically Ur, is offered by a gold pin topped with a standing female figure (U.456). 63 60.  Woolley 1962: 11. No U. number or image is provided. 61.  Woolley 1962: 101, pl. 26, 7. 62. Woolley 1962: 17. No U. number or image provided for the figurine in text, although it may be U.2802 (no. 8, p. 101, pl. 26), which is listed as originating from the southeast end of the Gipar of Nabonidus. The foundation deposits of copper/bronze dogs and a bearded male figure were also found in this area beneath the floor of Nabonidus; the clay cylinder inscribed with copies of Būr-Sîn texts, which had been found at Ur by Sîn-balāssu-iqbi, was found in another room in this area (see notes above). Another hand-modeled figurine fragment (U.311, B14991)—either a female, exceptionally, wearing a horned headdress, or possibly a vase bearer of uncertain gender—was found in a Neo-Babylonian level of the É-nun-mah (Trial Trench B, locus/room 27); Woolley 1962: 101. Furthermore, very similar examples of mold-made plaques (but from different molds, according to Woolley) depicting a female seated cross-legged holding an infant were found in two potentially official contexts: by the North Harbor (U.16445, 31-43-407) and “in the rubble against the northwest face of the Ziggurat” (U.17819); Woolley 1962: 102, nos. 18–19, pl. 28. 63.  Woolley 1962: 29–30, 106, pl. 21. Woolley argued that this object should be dated earlier due to its “fine Sumerian modeling.” The object was found as part of a hoard of jewelry and other precious objects dating as early as the Ur III period, between the floors of the Persian period and of Nebuchadnezzar in room 5 of the É-nun-mah, which had been a treasury or storehouse until the time of Nebuchadnezzar when it became a temple (Zettler and Hafford 2015: 372–373; Crawford 2015: 125).

Detecting Social Tensions in the Archaeological Record

121

Although the woman is dressed in a long gown, her facial features, hairstyle and hand gesture (folded beneath the breasts) find parallel in the terracotta corpus. While the material—gold—suggests an elite object, this iconography clearly spanned media and social contexts in Babylonia. Finally, another figurine type that occurs less commonly than the female image at Ur, but might likewise be categorized as broadly vernacular, is the frontal image of a standing vase-bearer (Fig. 15). These are described by Woolley as male figures, although many are beardless and some have clear indications of breasts, and may thus be either female or gender-ambiguous individuals. All of these figures are dressed in long gowns, but some appear nude above the waist. Some wear horned headdresses and are thus clearly divine. 64 These examples, akin to the female figurines, find close comparanda with other Babylonian sites, including Babylon, Borsippa, Uruk, Larsa, Kish, and Nippur and are found in the same types of contexts. 65

Fig. 15. Vase-bearer figurine types found at Ur compared to those found at other southern Mesopotamian sites.

I have briefly referred to this figurine type above, in the context of vase and staff bearer foundation figurines found in area KP at Ur. Unlike these foundation figurines, which are mostly hand-modeled (except for the face) in unbaked clay, figurines of the type discussed here are all mold-made and kiln-baked and are in this way similar to the Babylonian female figurines (the vase and staff bearers found as foundation deposits at Assyrian sites are of unbaked clay, as they are at Ur, although some can be mold-made). At Uruk, a number of such vase-bearers were found buried underneath the floor of a priestly house near the Eanna: because the figurines were broken and not found in foundation deposit boxes, it appears that 64.  Woolley 1962: pls. 26–28; Woolley 1925a: 19. 65.  Klengel-Brandt and Cholidis 2006: 80–130, pls. 22–23; Ziegler 1962: pls. 14–16, 19–20; Barrelet 1968; McCown et al. 1967; Pruß 2013; Van Buren 1931: 38.

122

Anastasia Amrhein

they were de-consecrated votives (see above). Such vase-bearers are not known from Assyrian sites. Although I have categorized the mold-made, kiln-baked Babylonian female and vase-bearer figurines as “vernacular”—i.e., found in domestic and broadly urban, public areas—some examples have also been found in temple contexts (see above). This suggests a close connection between official and popular religion in Babylonia—a phenomenon that cannot be observed in Assyria, where vernacular figurines (again, predominantly representing women—see below) were found throughout the city, but notably not in temple areas. 66 In Babylonia, both the female and vase bearer figurines could evidently be used in domestic ritual as well as set up in temples as votives. It appears that in Babylonia, only Ninšubur, the smiting god figures, and dogs were restricted to official contexts (as foundation deposits). During the period of Assyrian domination—as evidenced by Ur—this restricted, official group was briefly expanded to include a wider variety of Assyrian-type images.

Interpretation The difference in manufacturing technique between vernacular and official fig­ urines is more strongly visible in Assyria than in Babylonia and can be con­nected with the rise of the Neo-Assyrian kings. Neo-Assyrian society might also be inter­preted as more overtly patriarchal than in Babylonia (where, for example, the daughter of the ruler traditionally occupied the position of priestess of the moon god at Ur and held high status; see also the female foundation figurines cited above). As I suggest below, the figurine assemblages show that in Assyria, an effort was made to sharply segregate the official ritual practices—of men—from the vernacular ritual practices of women. In Babylonia, however, such a distinction is less readily observable. What I find especially significant is that the vernacular figurines—from all time periods, but especially in the first millennium—predominantly represent mortal, human women (although female figurines have traditionally been interpreted as goddesses, there is no evidence to support this: they do not wear horned headdresses, nor are they provided with any recognizable divine attributes 67). This is a marked contrast to the near-absence of women in official representations at this time. In reliefs for example, Assyrian women are never depicted, except for the queen and her attendants, and foreign women who have been taken captive are portrayed as compliant and devoid of agency. 68 This is not surprising because there is little evidence for the presence of female artists, scholars, or scribes in the Assyrian court. 69 66.  See Klengel-Brandt 1978. 67.  Woolley, for example, at various times describes the figures as goddesses, votaresses, priestesses, and “servants of love” (Woolley 1962: 101–106; Woolley 1965: 91). See also Albright 1939 and Pritchard 1943; cf. however more recent studies such as Tadmor 1982; Clayton 2014; Conkey and Tringham 1995. 68.  See, for example, representations on the Balawat Gates of Shalmaneser III (British Museum no. 124655); scenes on Sennacherib’s reliefs depicting the siege of Lachish, from Nineveh (British Museum no. 124907, 124908); and the so-called Banquet Scene of Ashurbanipal (British Museum no. 124920). 69. See limited examples cited in Stol 1995: 137 and Stol 2016. Legal and economic texts also indicate that positions of power outside the home available to women had greatly diminished between the third and first millennia. Only lower-class women held jobs (out of necessity); women could be sold by their husbands and fathers to pay off debts; and only widows appear to have managed their own legal and financial affairs, although this was not seen as a socially desirable situation, according to Saporetti 1979; Asher-Greve 2002: 17–18.

Detecting Social Tensions in the Archaeological Record

123

Women are also under-represented and stigmatized in Assyrian male-authored texts. Sue Rollin, for example, notes that the magic practices of women and foreign­ ers were characterized as witchcraft because these individuals were not involved with public institutions in the same way as Assyrian men and had to obtain power and authority through less direct means. The texts demonstrate that while so-called witchcraft was seen as “unjustified action,” the anti-witchcraft rituals that were performed by male āšipu priests, and utilized the very same procedures, were considered legitimate. 70 Although the use of figurines is mentioned in a number of the officiallysanc­tioned, male-authored texts, including anti-witchcraft, love incantations, heal­ ing and religious rituals, 71 the vernacular figurines under discussion cannot be con­nect­ed with any of them because the texts typically prescribe ritual destruction of the figurines. The preserved examples, however, are carefully baked and in many cases repaired with bitumen, indicating long-term use. Since literacy was largely restricted to male scholars in Assyro-Babylonian society, these anonymous objects are representative of a tradition that was transmitted orally and through practice, likely by women, and were probably used for vernacular magical rituals that have not been directly recorded in the text corpus. Already in the 1960s, utilizing ethnographic evidence from Syrian villages, Marie-Thérèse Barrelet had posited that figurines were made by female craft specialists. In Mesopotamia, the material of clay and the enterprise of claywork­ ing were also culturally gendered feminine. Goddesses appear in creation stories characterized as “potters,” molding or modeling the first humans out of clay, and the Akkadian verb—karâṣu, to model (lit., “to pinch off”)—is used to describe cre­ation and fetal formation, as well as pottery production. 72 The conceptual parallels between the manufacture of clay figurines and human conception, gestation and birth are further emphasized by the characterization of the womb as a kiln, and the placenta as clay—an initially amorphous material that is formed into an infant. 73 As part of my object-based research, I have measured the dermal ridge densities of artisan fingerprints preserved on about 200 figurines—both of official and vernac­ ular types—and my findings support female authorship for the first millennium examples (Fig. 16). This protocol has also been used for pottery and is based on data collected in the field of criminology, which has determined that the fingerprint dermal ridge density of females is consistently higher than that of males. 74 That clay work was a professional occupation is confirmed by the consistency of the material properties of the clay figurines I have studied, as well as the use of molds, kilns, and the wheel in various examples. 75 Ethnographic studies demonstrate that in some cases, men and women as well as children may be involved in pottery production, although a single individual ultimately receives credit for the final product. 76 This is interesting to consider in light of the official, apotropaic figurine making tradition, which in the ritual texts 70.  Rollin 1983. 71.  Abusch 2015; Biggs 1967; Langin-Hooper 2014: 458; Bidmead 2002. 72.  Barrelet 1968; Graff 2014; Couto-Ferreira 2013: 108; Arnaud 1996: 123–142. 73.  Couto-Ferreira 2013: 105–106; Kilmer 1987: 211–213; Graff 2014. 74.  Amrhein dissertation (forthcoming); Sanders 2015: 223–238. 75.  Barrelet 1968; Klengel-Brandt 1978. 76.  Wright 1991; Graff 2014: 379.

124

Anastasia Amrhein

Fig. 16. Examples of preserved fingerprints on the back of a clay figurine from Babylon, with detail. A ridge density of 7 is measured using a 2.5 × 2.5 mm square in the software ImageJ and doubled; recorded as 14* per 5 × 5 mm square.

is portrayed as the domain of male priests who recorded the practice. Akin to the vernacular female figurines, the apotropaic demon figurines bear predominantly female fingerprints. My hypothesis is, thus, that there were two types of femaleoper­ated clay workshops at the time of the Neo-Assyrian empire: one catering to the general public and the other to male priests tied to the palace and temple, and that the priests attempted to pass off the latter practice as entirely their own. The male priests were, however, ultimately responsible for enlivening the apotropaic images through ritual action (in the official contexts where such figurines are found). Erica Reiner has also suggested that different people were responsible for writing down the incantations and rituals and making the various objects that appear in such texts: not only figurines, but also seals and amulets. Reiner notes that when the instructions call for making any kind of image, the scholar-scribes are generally silent on the matter because image-making was not their domain. In rare instances, very rough instructional sketches may be provided. 77 Moreover, the reader of the ritual texts is instructed to not only make the clay figurines, but also weapons and accessories of bronze. 78 Although it is possible for a non-professional to mold or model clay, metalworking would have certainly required the expertise of a metalsmith. It is also worth noting that while the āšipu is instructed to take clay from the purified 77.  Reiner 1987: 30. For example, one ritual calls for a drawing of gods to be made on the bed of an individual who “wishes to see a mantic dream”—the text states “make this drawing that I have drawn here,” but leaves room for interpretation. 78.  Wiggermann 1992: 11.

Detecting Social Tensions in the Archaeological Record

125

clay pit, he does not make the figurines on site, but simply returns to the city with the clay, where perhaps the material was worked by women artisans. 79 The “official” figurine making tradition was introduced with the rise of the Neo-Assyrian empire, and the same apotropaic beings also appear on palace reliefs at this time. Each new king, with his team of scholar-priests, seems to have originated new apotropaic iconographies and experimented with modes of representation (as was the case with palace reliefs as well). 80 For example, the figurines of Adad-nerari III are made with shallow molds and feature bird apkallū and vase or staff bearers (see above). Although the frontal, molded image is carefully trimmed at the edges and the back painted to create the appearance of a figurine, the result is ultimately closer to a plaque. By contrast, during the reign of Sargon II, the figurines are unusually large and in much higher relief (perhaps intended to match the scale of the palace at Khorsabad and the style of the reliefs). 81 They depict staff-bearers whose bodies are in profile but heads en face, as well as smiting gods and lion demons. The figurines of Ashurbanipal, on the other hand, feature some unique figurine types not known from the reigns of other kings, including birds, dogs, and a two-faced figure (with a human face on one side, and a lion face on the other—perhaps a priest wearing a lion-skin). 82 It is also interesting to note that especially in Assyria, and to a lesser extent in Babylonia, the official figurine-making techniques differ from the popular ones. I suggest that this is the result of social tension between female clay artisans who were seeking to legitimate their craft / ritual tradition, and male priests who were attempting to distance themselves from the vernacular tradition. 83 Findings at Assyrian capitals and at Ur indicate that the official concepts and images did not penetrate the local cultural traditions of the population. The local vernacular traditions in the Assyrian heartland as well as further afield represent strong, geographically defined currents of image-making that persisted in spite of, and parallel to, the appearance of Assyrian imperial art. According to sociologist Anthony Giddens’ practice theory, societies are consti­ tuted and maintained through the repeated actions of individuals. Although bound by societal structures, individuals have agency within those limitations and can transform them slowly over time. 84 By making and manipulating the figurines, past individuals—specifically women—were enacting and expressing their identities on several levels, participating in or resisting group affiliations. In contrast to the very conscious development and use of official fig­urines (which evolved rapidly with each king’s reign), the shared conventions pertaining to vernacular figurines were probably never articulated, yet mutually understood—what Giddens calls “practical consciousness” (and were very slow to change). Regional variations of the female image are so consistent across time that they suggest two broad circles of shared cultural identity: one that originated in Babylonia, 79.  Woolley 1926; Wiggermann 1992: 6ff. 80.  Regarding court scholars, see Ataç 2010. 81.  See Barrelet 1968; Botta and Flandin 1849, 1850. 82.  No. 57.27. 12 in the Metropolitan Museum of Art. See Rittig 1977: 85, 124. See Ataç 2010: 42 regarding the ritual of “mummery.” 83.  Cf. Rollin 1983: 42ff, regarding anthropological and textual evidence that “witchcraft accusations are indices of social tension” within a given society. 84.  Giddens 1984.

126

Anastasia Amrhein

and the other in the Jazirah (in continuation from the late third and second millennia— see above). Further, smaller circles of group identity are signified by varied executions of the heads—the primary loci of identity—as well as additional features such as jewelry and dress (Fig. 17). Although figurines from the Jazirah and Levant are all bell-shaped, fig­ urines from Assur, Carchemish, and Fig. 17. Lachish, for example, clearly evidence disLeft: Female figurine from Assur. Middle: Female figurine from Kefrik/Kourik. tinct spheres of identity. The bell-shaped Right: Female figurine from Lachish. figurines from Assur, which span the first half of the first millennium, are recognizable by their short, linear hair. 85 Examples of the eighth and seventh centuries from the area of Carchemish wear elaborate headdresses or hairstyles represented by ap­pliqued strips and discs of clay. 86 Beginning in the late eighth century, in Syria and the Levant (especially in Judea), so-called pillar figurines appear. They represent women supporting their breasts and feature a molded or stamped face, short curls, and a hand-modeled body. They are further distinguished by the fact that the heads are usually separate pieces with a cone-like neck that is inserted into the body. 87 As shown above, at Ur, the female images for the most part demonstrates a shared identity with other Babylonian cities, although several examples appear to be unique. Examples from all of these areas occasionally hold infants. Through some isolated examples, it is also possible to identify the presence of specific groups that have been forcibly re-settled by the Assyrian kings—a topic that has been studied primarily through texts. A few female figurine heads characteristic of Megiddo have been found at Assur,  88 and some mold made female plaques characteristic of Babylonia have been found at Ashkelon 89 and at Assyrian sites (including Assur, Billa, and Nimrud). 90 At Ur, a single bell-shaped figurine has been found. 91 85.  Klengel-Brandt 1978. Only a few comparable examples of hand-made, bell-shaped figurines with appliquées have been recorded from the excavations at Nimrud (Fort Shalmaneser) and at Khor­sabad. See Barrelet 1968: 123–124, no. 716 (from Khorsabad). See also no. 31-51-195 in the Penn Museum from Billa. 86.  Iron Age II figurines from Syrian sites are dated to the eighth–seventh centuries, and are believed to have originated in the area of Carchemish (Pruß 2013: 608; Moorey 2005: 227ff; Woolley 1939b). 87.  Darby 2014; Pruß 2013: 608; Kletter 1996; Tufnell 1953: pl. 28:10. 88.  VA 8074 and VA 4965 from Assur (in the Vorderasiatisches Museum, Berlin); Klengel-Brandt 1978. 89. No. 52341 and no. 46974 in Press 2012. Somewhat similar to the Babylonian figurines are mold-made images of nude women that occur frequently in Iron Age II sites in Syria and the Levant—the so-called “Astarte plaques.” In contrast to the few Babylonian examples cited, these figurines are more comparable to figurines from Cyprus, Egypt, and the Greek world, than Babylonia. These represent a continuation from the Late Bronze Age. In the Levant, mold-made reliefs largely disappear over the course of the seventh century, only continuing in Philistia (Pruß 2013: 608; Pruß 2010: 145–151; Albright 1939). 90. VA 4918 from Assur (in the Vorderasiatisches Museum, Berlin); Klengel-Brandt 1978. Nos. 31-51-196 and 31-51-251 from Billa, in the Penn Museum. BM 1994,1105.58 from Nimrud in the British Museum. 91.  No. 31-16-885 in the Penn Museum, found in the Nebuchadnezzar temple outside the northeast side of the Temenos, according to the Ur-online database.

Detecting Social Tensions in the Archaeological Record

127

Conclusions Is it possible, however, that the vernacular figurine tradition is representative of active resistance? James Scott observes that “the hidden transcript has no reality as pure thought; it exists only to the extent it is practiced, articulated, enacted and disseminated.” 92 Figurines are certainly the material remains of such shared practices, although in many cases, it is difficult to separate resistance from folk tradition. Indeed, Scott notes that both rely on oral transmission, which offers “a kind of seclusion, control, and even anonymity.” 93 In the context of Neo-Assyrian hegemony—where forced resettlement of conquered populations was a regular practice and women are virtually non-existent in the official record—the expression of gender and geography-based identity may have been in itself a subversive act. It is evident that the Assyrian goal of dissolving regional identities and resistance by mixing populations did not succeed, however: that re-settled individuals were evidently absorbed into the culture of their new home, suggests to me that geography based identities were not diluted by this tactic, but in fact remained strong (see above). That the makers of figurines presented a threat to Assyrian hegemony is supported by the physical isolation of clay-workers in Mesopotamian cities. The site of Diqdiqqeh, for example, where numerous figurines and molds were found, lies just outside the city of Ur and Woolley suggested that it was a “working-class suburb” where craftspeople lived and worked. 94 Similarly, first millennium Nineveh had separate neighborhoods of goldsmiths, bleachers, and potters. 95 Following Max Weber, Scott emphasizes that dominant groups are especially concerned about the activities of the so-called “pariah-intelligentsia”—marginalized, often itinerant individuals such as prophets, carnival performers, diviners, and healers, as well as artists and craftspeople. Their “critical distance from dominant values,” he notes, arises not only from their skills but also from their existence “outside of the social hierarchy.” 96 These are also the types of individuals who were accused of witchcraft in the Neo-Assyrian period (see above). According to Scott, hidden transcripts often take the form of the carniva­ lesque—“the acting out in fantasy … of the anger and reciprocal aggression denied by the presence of domination.” 97 Carolyn Nakamura has argued that it is specifically the miniaturization of the figurines that allows them to be animated in such a way that gives the human user power not otherwise available in the lived world. 98 Play, however, is not “empty posturing,” but in many cases, as Scott notes, has historically led to revolts. 99 It is of course impossible to say whether the production and use of figurines lead to uprising and rebellions in Assyria, since the historical

92.  Private homes beings primary examples of offstage sites (Scott 1990: 118–121). 93.  He goes on (Scott 1990: 161): “There is no orthodoxy or center to folk culture since there is no primary text to serve as the measure of heresy… multiplicity of its authors provides its protective cover.” 94.  Woolley and Mallowan 1976: 12. 95.  Van De Mieroop 1997: 183, 78; May and Steiner 2014: 132–133. 96.  Scott 1990: 124. 97.  Scott 1990: 37–38. 98.  Nakamura 2005: 34. 99.  Scott 1990: 52, 184, 178.

128

Anastasia Amrhein

record does its best to conceal and smooth over such ruptures. 100 Nonetheless, there is some textual evidence of outsiders, especially women, disrupting the dominant order through non-professional prophecy, dream oracles, and necromancy—often predicting a coup. Such prophecies were unwelcome but were still taken seriously since they were divine communications. 101 Assyrian imperial anxiety over the practices of women—both Assyrian and foreign—is thus evident in the marginalization of women and their practices in official textual and visual sources; it is perhaps most evident, however, in the appropriation of the use of clay for magical purposes by the dominant order, re-framing the practice as legitimate, while stigmatizing its folk origins. Nonetheless, the women who were the makers and users of the figurines were responsible for the constitution and reproduction of regional social identities, which remained stable in the face of Assyrian hegemony.

Image Credits Fig. 1: After © Ur-online. Fig. 2: After Woolley 1962: pl. 60 and Woolley 1965: pl. 43. Fig. 3: After Woolley 1926b: fig. 28; pl. IX, figs. 2, 3; pl. X, fig. 4a; pl. XI, figs. 7–10. Fig. 4: Image courtesy of University of Pennsylvania Museum of Archaeology and Anthropology Near East Curator-in-Charge. © Trustees of the Two Museums (British Museum and University of Pennsylvania Museum of Archaeology and Anthropology). Fig. 5: Image courtesy of University of Pennsylvania Museum of Archaeology and Anthropology Near East Curator-in-Charge. © Trustees of the Two Museums (British Museum and University of Pennsylvania Museum of Archaeology and Anthropology). Fig. 6: Image courtesy of University of Pennsylvania Museum of Archaeology and Anthropology Near East Curator-in-Charge. © Trustees of the Two Museums (British Museum and University of Pennsylvania Museum of Archaeology and Anthropology). Fig. 7: Image courtesy of University of Pennsylvania Museum of Archaeology and Anthropology Near East Curator-in-Charge. © Trustees of the Two Museums (British Museum and University of Pennsylvania Museum of Archaeology and Anthropology). Fig. 8: Image courtesy of University of Pennsylvania Museum of Archaeology and Anthropology Near East Curator-in-Charge. © Trustees of the Two Museums (British Museum and University of Pennsylvania Museum of Archaeology and Anthropology). Fig. 9: Author’s own photo, by permission of the University of Pennsylvania Museum of Archaeology and Anthropology, B16273 (U.6774). Fig. 10: Courtesy of the University of Pennsylvania Museum of Archaeology and Anthropology, B16271A–C (U.6768A–C).

100.  King Tiglath-pileser III, for example, is interpreted as a usurper because he does not include his lineage in his royal inscriptions (Radner 2012). 101.  Pongratz-Leisten 2006: 27; cf. Scott’s observation that resistance in the form of spirit possession, ecstatic cults, and the hysteria of Victorian women, were “never directly challenging the domination at which it is aimed” because they were “experienced as involuntary” (Scott 1990: 141–142).

Detecting Social Tensions in the Archaeological Record

129

Fig. 11: Courtesy of the University of Pennsylvania Museum of Archaeology and Anthropology, 31-43-488 (U.16159A) and 31-43-489 (U.16159E). Fig. 12: Author’s Own Photo, by Permission of the University of Pennsylvania Museum of Archaeology and Anthropology, 31-43-490 (U.17075). Fig. 13: Author’s Own Photo, Reproduced with Kind Permission of Ashmolean Museum, University of Oxford: AN1927.3277. Courtesy of the University of Pennsylvania Museum of Archaeology and Anthropology: 3116-768, 31-16-769, 31-16-837, 31-43-394, 31-43-406, 32-40-20, 32-40-29, 35-1-106, 35-1-107, B15450, and B17210. © Musée du Louvre/Département des Antiquités orientales: AO 12724, AO 16963, and AO 16966. © Staatliche Museen zu Berlin – Vorderasiatisches Museum, Photo: Olaf M. Teßmer: VA 11589, VA 11591, VA 11598, VA 11608, VA Bab 25, VA Bab 70, VA Bab 333, VA Bab 781, VA Bab 842, and VA Bab 3221. Fig. 14: Author’s Own Photo, by Permission of the University of Pennsylvania Museum of Archaeology and Anthropology, 32-40-40 (U.18132). Fig. 15: Author’s Own Photos, by Permission of the University of Pennsylvania Museum of Archaeology and Anthropology: 31-16-802, 32-40-16, 32-40-19, 33-35-39, and 51-6-178. Author’s Own Photos, Reproduced with kind permission of Ashmolean Museum, University of Oxford: AN1924.699 and AN1925.152. Courtesy of the University of Pennsylvania Museum of Archaeology and Anthropology: 3116-923, 33-35-11, 51-6-152, B2998, and B15468. © Ashmolean Museum, University of Oxford: AN1931.459. © Musée du Louvre/Département des Antiquités orientales: AO 8658. © Staatliche Museen zu Berlin – Vorderasiatisches Museum, Photo: Olaf M. Teßmer: VA 11638, VA 11680, VA 11740, VA 11744, VA Bab 441, VA Bab 752, and VA Bab 3266. Fig. 16: Author’s Own Photo, Reproduced with Kind Permission of the Ashmolean Museum, University of Oxford, AN1927.4335.b. Fig. 17: Left: © Staatliche Museen zu Berlin—Vorderasiatisches Museum, VA 4943, Photo: Olaf M. Teßmer. Middle: Author’s Own Photo, Reproduced with Kind Permission of the Ashmolean Museum, University of Oxford, AN1914.795. Right: The Metropolitan Museum of Art, New York, Gift of Harris D. and H. Dunscombe Colt, 1934, 34.126.53 © Public Domain.

Bibliography Abusch, T. 2015 The Witchcraft Series Maqlû. Writings from the Ancient World 37. Atlanta: SBL Press. Albright, W. F. 1939 Astarte Plaques and Figurines from Tell Beit Mirsim. Pp. 109–117 in Mèlanges Syriens Offerts à Monsieur Renè Dussaud 1. Paris.

130

Anastasia Amrhein

Arnaud, D. 1996 Le foetus et les dieux au Proche-Orient sémitique ancient. Revue de l’histoire des religions 213: 123–142. Asher-Greve, J. 2002 Decisive Sex, Essential Gender. Pp. 11–27 in Sex and Gender in the Ancient Near East: Proceedings of the 47th Rencontre Assyriologique Internationale, eds. S. Parpola and R. M. Whiting. Helsinki: Neo-Assyrian Text Corpus Project. Ataç, M.-A. 2010 The Mythology of Kingship. Cambridge: Cambridge University Press. Barrelet, M.-T. 1968 Figurines et reliefs en terre cuite de la Mésopotamie antique. Paris: Librairie Orientaliste Paul Geuthner. Belting, H. 2005 Image, Medium, Body: A New Approach to Iconology. Critical Inquiry 31: 302–319. Bidmead, J. 2002 The Akītu Festival: Religious Continuity and Royal Legitimation in Mesopotamia. Piscataway: Gorgias Press. Biggs, R. D. 1967 Šà.zi.ga: Ancient Mesopotamian Potency Incantations. TCS 2. Locust Valley: J. J. Augustin. Botta, P. E. and Flandin, E. 1849 Monument de Ninive II. Paris: Imprimerie Nationale. 1850 Monument de Ninive V. Paris: Imprimerie Nationale. Brinkman, J. A. 1969 Review: Ur: “The Kassite Period and the Period of the Assyrian Kings.” Or. NS 38: 310–348. Clayden, T. 1995 The Date of the Foundation Deposit in the Temple of Ningal at Ur. Iraq 57: 61–70. Clayton, V. 2014 Visible Bodies: Resistant Selves: The Iron Age Figurines from Tell Ahmar. Hawthorn: K and H Publishing. Conkey, M. W. and Tringham, R. E. 1995 Archaeology of the Goddess: Exploring the Contours of Feminist Archaeology. Pp. 199–247 in Feminisms in the Academy: Rethinking the Disciplines, eds. A. Steward and D. Stanton. Ann Arbor: University of Michigan Press. Couto-Ferreira, M. E 2013 The River, the Oven, the Garden: Female Body and Fertility in a Late Babylonian Ritual Text. Rivista degli studi orientali 86: 97–116. Crawford, H. 2015 Ur: The City of the Moon God. London: Bloomsbury Academic. Darby, E. 2014 Interpreting Judean Pillar Figurines. Tübingen: Mohr Siebeck. Ellis, R. 1967 “Papsukkal” Figures Beneath the Daises of Mesopotamian Temples. RA 61: 51–61. 1968 Foundation Deposits in Ancient Mesopotamia. New Haven: Yale University Press. Giddens, A. 1984 The Constitution of Society: Outline of a Theory of Structuration. Berkeley: University of California.

Detecting Social Tensions in the Archaeological Record

131

Graff, S. B. 2014 Sexuality, Reproduction and Gender in Terracotta Plaques from the Late ThirdEarly Second Millenia BCE. Pp. 371–390 in Critical Approaches to Ancient Near Eastern Art, eds. B. Brown and M. Feldman. Boston: de Gruyter. Kilmer, A. 1976 Appendix C: The Brick of Birth. JNES 46: 211–213. Klengel-Brandt, E. 1968 Apotropäische Tonfiguren aus Assur. Forschungen und Berichte 10: 19–37. 1978 Die Terrakotten aus Assur im Vorderasiatischen Museum Berlin. Berlin: VEB Deutscher Verlag der Wissenschaften. Klengel-Brandt, E. and Cholidis, N. 2006 Die Terrakotten von Babylon im Vorderasiatischen Museum in Berlin. Saarwellingen: Saarländische Druckerei. Kletter, R. 1996 The Judean Pillar-Figurines and the Archaeology of Asherah. Oxford: Tempus Reparatum. Koldewey, R. 1911 Die Tempel von Babylon und Borsippa. WVDOG 15. Leipzig: J. C. Hinrichs. Langin-Hooper, S. 2014 Terracotta Figurines and Social Identities in Hellenistic Babylonia. Pp. 451–481 in Critical Approaches to Ancient Near Eastern Art, eds. B. Brown and M. Feldman. Boston: de Gruyter. Lesure, R. 1997 Figurines and Social Identities in Early Sedentary Societies of Coastal Chiapas, Mexico: 1550–800 B.C. Pp. 227–247 in Women in Prehistory, eds. C. Claassen and R. A. Joyce. Philadelphia: University of Pennsylvania Press. Liverani, M. 1988 The Growth of the Assyrian Empire in the Habur/Middle Euphrates Area: A New Paradigm. State Archives of Assyria Bulletin 2: 81–98. Loud, G. 1936 Khorsabad, Part 1: Excavations in the Palace and at a City Gate. Chicago: Oriental Institute Publications. Loud, G. and Altman, C. B. 1938 Khorsabad, Part 2: The Citadel and the Town. Chicago: Oriental Institute Publications. Mallowan, M. E. L. 1954 The Excavations at Nimrud (Kalhu), 1953. Iraq 16: 59–114. 1966 Nimrud and its Remains. London: Collins May, N. N. and Steiner, U. 2014 The Fabric of Cities: Aspects of Urbanism, Urban Topography and Society in Mesopotamia, Greece and Rome. Leiden: Brill. McCown, D. E., Haines, R. C., and Hansen, D. 1967 Nippur I: Temple of Enlil, Scribal Quarter, and Soundings. Chicago: Oriental Institute Publications. Moorey, P. R. S 2005 Ancient Near Eastern Terracottas. Oxford: Ashmolean Museum. Nakamura, C. 2005 Mastering Matters: Magical Sense and Apotropaic Figurine Worlds of Neo-As­ syria. Pp. 18–45 in Archaeologies of Materiality, ed. L. Meskell. Malden: Blackwell. 2008 The Matter of Magic: Material Figures of Memory and Protection in Neo-Assyrian Apotropaic Figurine Rituals. Ph.D. dissertation: Columbia University.

132

Anastasia Amrhein

Oates, D. and Reid, J. H. 1956 The Burnt Palace and the Nabu Temple: Nimrud Excavations, 1955. Iraq 18: 22–39. Oates, J. and Oates, D. 2001 Nimrud: An Assyrian Imperial City Revealed. London: British School of Archaeology in Iraq. Pongratz-Leisten, B. 2006 Cassandra’s Colleagues: Prophetesses in the Neo-Assyrian Empire. The Canadian Society for Mesopotamian Studies Journal 1: 27. Porter, B. 2001 The Importance of Place: Esarhaddon’s Stelae at Til Barsip and Sam’al. Pp. 373–390 in Proceedings of the XLVe Rencontre Assyriologique Internationale, Part 1: Historiography in the Cuneiform World, eds. T. Abusch et al. Bethesda, MD: CDL Press. Postgate, N. 1992 The Land of Assur and the Yoke of Assur. World Archaeology 23: 247–263. Press, M. D. 2012 Ashkelon 4: The Iron Age Figurines of Ashkelon and Philistia. Winona Lake: Eisenbrans. Preusser, C. 1954 Die Wohnhäuser in Assur. WVDOG 64. Berlin: Deutsche Orient-Gesellschaft. Pritchard, J. B. 1943 Palestinian Figurines in Relation to Certain Goddesses known through the Literature. New Haven: American Oriental Society. Pruß, A. 2010 Die Amuq-Terrakotten: Untersuchungen zu den Terrakotta-Figuren des 2. und 1. Jahrtausends v. Chr. aus den Grabungen des Oriental Institute Chicago in der Amuq-Ebene. Turnhout: Brepols. 2013 Terrakotten. RlA 13/7–8: 603–610. Radner, K. 2012 Tiglath-pileser III, King of Assyria (744–727 BC). Assyrian Empire Builders. University College London. See http://www.ucl.ac.uk/sargon/essentials/kings/ tiglatpileseriii/ Rashid, S. A. 1971 Gründungsbeigaben. RlA 3/9: 655–662. Reiner, E. 1987 Magic Figurines, Amulets, and Talismans. Pp. 27–36 in Monsters and Demons in the Ancient and Medieval Worlds: Papers Presented in Honor of Edith Porada, eds. A. E. Farkus, et al. Mainz: Verlag Philipp von Zabern. Reuther, O. 1926 Die Innenstadt von Babylon (Merkes). Ausgrabungen der Deutschen OrientGesellschaft in Babylon 3. Leipzig: J.C. Hinrichs. Rittig, D. 1977 Assyrisch-babylonische Kleinplastik magischer Bedeutung vom 13.–6. Jh. c. Chr. München: Verlag Uni-Druck München. Rollin, S. 1983 Women and Witchcraft in Ancient Assyria. Pp. 35–45 in Images of Women in Antiquity, eds. A. Cameron and A. Kuhrt. Detroit: Wayne State University Press. Sanders, A. 2015 Fingerprints, Sex, State, and the Organization of the Tell Leilan Ceramic Industry. Journal of Archaeological Science 57: 223–238.

Detecting Social Tensions in the Archaeological Record

133

Saporetti, C. 1979 The Status of Women in the Middle Assyrian Period. Monographs on the Ancient Near East 2. Malibu: Undena. Scott, J. C. 1990 Domination and the Arts of Resistance: Hidden Transcripts. New Haven: Yale University Press. Stol, M. 1995 Women in Mesopotamia. JESHO 38/2: 123–144. 2016 Women in the Ancient Near East. Translated by Helen and Mervyn Richardson. Boston: de Gruyter. Tadmor, M. 1982 Female Cult Figurines in Late Canaan and Early Israel: Archaeological Evidence. Pp. 139–173 in Studies in the Period of David and Solomon, ed. T. Ishida. Winona Lake: Eisenbrauns. Tufnell, O. 1953 The Wellcome Archaeological Research Expedition to the Near East: Lachish (Tell ed Duweir) 3: The Iron Age. London: For the Trustees of the late Sir Henry Wel­ lcome by O.U.P. Van Buren, E. D. 1931 Clay Figurines and Offerings. Berlin: H. Schoetz & Co. Van De Mieroop, M. 1997 The Ancient Mesopotamian City. Oxford: Clarendon Press. Weadock, P. N. 1975 The Giparu at Ur. Iraq 37: 101–128. Wiggermann, F. A. M. 1992 Mesopotamian Protective Spirits: The Ritual Texts. CM 1. Groningen: Styx & PP Publications. 1987 The Staff of Ninšubura. Studies in Babylonian Demonology, II. JEOL 29: 3–34. Wilkinson, T. J., Ur, J., Wilkinson, E. B., and Altaweel, M. 2005 Landscape and Settlement in the Neo-Assyrian Empire. Bulletin of the American Schools of Oriental Research 340: 23–56. Winter, I. 1983 The Program of the Throneroom of Assurnasirpal II. Pp. 15–32 in Essays on Near Eastern Art and Archaeology in Honor of Charles Kyrle Wilkinson, eds. P. O. Harper and H. Pittman. New York: Metropolitan Museum of Art. 1997 Art in Empire: The Royal Image the Visual Dimension of Assyrian Ideology. Pp. 359–381 in Assyria 1995, eds. S. Parpola, et al. Helsinki: The Neo-Assyrian Text Corpus. Woolley, C. L. 1925a The Excavations at Ur, 1923–1924. AJ 5: 1–20. 1925b The Excavations at Ur, 1924–1925. AJ 5: 347–402. 1926a The Excavations at Ur, 1925–1926. AJ 6: 365–401. 1926b Babylonian Prophylactic Figures. JRAS 4: 689–713. 1931 The Excavations at Ur, 1930–1. AJ 11: 343–381. 1932 The Excavations at Ur, 1931–2. AJ 12: 355–392. 1939a The Ziggurat and its Surroundings. UE 5. Oxford: Trustees of the British Museum and of the University Museum, University of Pennsylvania, Philadelphia; Oxford University Press. 1939b The Iron-Age Graves of Carchemish. Annals of Archaeology and Anthropology 26: 11–37.

134

Anastasia Amrhein

1962 The Neo-Babylonian and Persian Periods. UE 9. London: Trustees of the British Museum and of the University Museum, University of Pennsylvania, Philadelphia. 1965 The Kassite Period and the Period of the Assyrian Kings. UE 8. Oxford: Trustees of the British Museum and of the University Museum, University of Pennsyl­ vania, Philadelphia; Oxford University Press. Woolley, C. L. and Mallowan, M. E. L. 1976 The Old Babylonian Period. UE 7. London: Trustees of the British Museum and of the University Museum, University of Pennsylvania, Philadelphia; British Museum Publications. Wright, R. P. 1991 Women’s Labor and Pottery Production in Prehistory. Pp. 194–223 in Engendering Archaeology: Women and Prehistory, eds. J. M. Gero and M. W. Conkey. Oxford: Blackwell. Zettler, R. L. and Hafford, W. B. 2015 Ur. B. Archäologisch. RlA 14/5–6: 367–385. Ziegler, C. 1962 Die Terrakotten von Warka. ADFU 6. Berlin: Gebr. Mann.

Old Babylonian Terracottas from Ur: Ancient and New Perspectives Laura Battini

CNRS, UMR 7192 PROCLAC- Collège de France Woolley’s excavations at Ur brought a large amount of data largely unexplored today (see the contribution of J. Taylor in this volume). As in all excavations, even in contemporary ones that appear largely “scientific” to us, reports contain certainly some mistakes, some oversights, but in any case Woolley was intellectually honest and in fact very fine for his time (Fig. 1). 1 Each time I go back to UE, thinking that this time at last I could prove Woolley’s weaknesses because he is not like a con­ temporary “scientific archaeologist,” each time I have to apologize and recognize his greatness. This is the case also in this study of Old Babylonian terracottas from Ur. Woolley did not accept the repartition of terracottas proposed by Leon Legrain, who also was wise for his time, but he gave more importance than Woolley to the stylis­ tic data. 2 And quite annoyed to contradict Legrain, Woolley inserted some terracottas as “Kassite,” while thinking that they were too similar for some of them to Larsa and for others to Neo-Babylonian terracottas. A quick look at UE 8 (Fig. 2) is sufficient to prove the rightness of Woolley. Of the thirty terracottas reproduced in UE 8, Fig. 1. Woolley and his foreman, Sheikh most are Larsa, some Neo-Babylonian and Hamoudi Ibn Ibrahim (from https://www.penn. museum/sites/expedition/sir-leonard-woolley/). only few perhaps Kassite. For example, 1.  At the same time Woolley excavated at Ur, Starr was working in Nuzi (1937 and 1939). But the final reports of both archaeologists are very different: if Woolley tried to record the maximum amount of data, the publication of the Nuzi excavations suffers from too many errors, contradictions, or unreported subjects (see among others: Lion 1999, Shortland et al. 2008, and Battini 2015a). 2.  Woolley 1965: 91. In this refusal of the stylistic criterion at nearly the same time see also Bar­ relet (1969: 96).

135

136

Laura Battini

Fig. 2. Terracottas from Ur (UE 8: pl. 31 nos. 24 and 25, and pl. 33).

UE 8: pl. 29 no. 6 offers better comparisons with UE 7: pl. 68 no. 42; and also pl. 66 no. 25; and pl. 67 no. 32 (Fig. 3). And UE 8: pl. 30 no. 13 is similar to UE 7: pl. 66 no. 24. 3

I. Number of Clay Figurines Discovered in Ur So, how many terracottas did Woolley and his team discover in Ur? It is difficult to answer this question without a certain degree of approximation. According to the Fig. 3. Terracottas from Ur, UE 8: pl. 29 no. 6 Ur online site (http://www.ur-online.org), and UE 7: pl. 68 no. 42. there are 1,976 figurines from all epochs. Even if not all are Old Babylonian, it seems possible—in the light of excavations and of results published in the final reports 4—to think that this is the case for most of them. Of these 1,976, 1,517 are anthropomorphic and 459 are animal; that is, anthropomorphic are three times more than animal. The first 304 figurines are unknown; they have never been published by Woolley, they never even had a number of field excavation, nor of museum. And the great majority are Old Babylo­ nian. So, it is perhaps easier to ask another question: how many are the published terracottas? But also this question needs a cautious answer. On one hand most of the terracottas published in UE 8 must be added to UE 7, without being sure that some terracottas could have been forgotten. On the other hand, sometimes the Old Babylonian terracottas published with a photo in UE 7 refer to other copies for which Woolley only cited the number of excavations without giving a photo and which are not described, but of which we know the existence. For example, terracotta UE 7: 3. Other similarities: UE 8 (=Woolley 1965): pl. 30 nos. 15 and 8 have comparisons with UE 7 (=Woolley 1976): pl. 68 no. 37. In the same manner, UE 8: pl. 31 nos. 24 and 25 correspond to UE 7: pl. 78 no. 125. And UE 8: pl. 30 nos. 12 and 10 are quite similar to UE 7: pl. 65 no. 13, as well as UE 8: pl. 29 no. 5 to UE 7: pl. 67 no. 34. 4.  In the UE series, the largest number of published figurines are Old Babylonian. Also the majority of areas dug is Ur III/Old Babylonian. There are three volumes concerning this period (UE 7, UE 5, and UE 6: respectively Woolley 1976, 1939, and 1974).

Old Babylonian Terracottas from Ur

Fig. 4. Figurine UE 7: pl. 65 no. 8 and its copies.

137

pl. 65 no. 8 (Fig. 4) has three other copies as well as UE 7: pl. 65 no. 10, which has one other copy. So the comparison between published photos of clay figurines and numbers of field excavations’ objects bring some 30 more Old Babylonian terracottas. Adding those considered “Kassite—but surely more ancient (eleven at least)—and adding some clay objects published in UE 7 but not inserted in the terracottas section, 5 the total number of Old Babylonian terracottas published amounts to about at least 312 (Table 1). 6 Out of these 312 terracottas, 282 are published with a photo or drawing, that is 90.4%. This is another example of the precision and wideness of Woolley: he had the merit of collecting the widest information about objects he discovered. He will rest as an example for all archaeologists, especially in his determination to publish all the results of excavations. Woolley spent only twelve seasons at Ur and then he spent the rest of his life publishing what he discovered. 7

Table 1. Number of possible Old Babylonian published terracottas from Ur. • • • • • •

261 Terracottas in the section (UE 7: pls. 64–92) + Pl. 263 + Pl. 96 U.11455, U.6476 + Pl. 97 i, j, o + Terracottas cited but without published photo or drawing + UE 8: pls. 29–31 = 312 → 90.4%

II. Ancient Perspectives on Figurines from Ur or Iconography I have been trying to understand in which way Woolley presented the terra­ cottas. 8 Not in a chronological order, nor in an order of excavations but in different ways, as Table 2 demonstrates. He began with terracottas found in 1930–1 (that is numbers 1–2), then those discovered in 1931–2 (that is number 3), and then those excavated in 1923–4 (that is number 4) and so on. Chronologically, even if there are still problems on dating clay figurines, the first two are not the most ancient. So, how did Woolley present terracottas? 5.  This is the case of ten objects published with photo or drawing in Woolley 1976: U.16993 (pl. 63), U.18199 (pl. 93.l), U.1034 (pl. 93.o), U.2788 (pl. 93.p), pl. 93.n (without number of excavation), U.11455 (pl. 96.g), U.6476 (pl. 97.h), U.6812 (pl. 97.i), U.17661 (pl. 97.j) and U.6195 (pl. 97.o). 6.  I say “at least” because in five cases Woolley (1976: 176–183) did not precisely know how many similar terracottas he found, so I calculated at least one, but there could be more. 7.  Apart the excavations in Tell Atchana (1937–1939, 1946–1949). 8.  To present objects coming from excavations, I advise people certainly to use some criteria to offer a better description of discoveries. So, I came to think that the manner in which Woolley described the figurines was not casual, by hazard, but follows some unwritten ideas. Following his ideas could let me know better his comprehension of terracottas.

138

Laura Battini

Table 2. Comparison of terracotta numbers in UE 7, excavation numbers, year of excavation, season numbers, and museum numbers. Terracotta no.

Excavation no.

Year of excavation

Season of excavation

Museum no.

No. 1

U.16959

1930–1

9th

P. 31-43-577

No. 2

U.16426

1930–1

9th

?

No. 3

U.18076

1931–2

10

B

No. 4

U.1231

1923–4

2nd

P. CBS 15672

No. 5

U.6664

1925–6

4

P. CBS 16265

No. 6

U.39

1922–3

1

B. JM31

No. 7

U.16263

1930–1

9th

P. 31-43-368

No. 8

U.18087

1931–2

10

?

Against the SW face of the temenos wall

U.608

1922–3

1st

?

Diqdiqqah

U.2974

1924–5

3rd

B.IM.986

Diqdiqqah

U.18070

1931–2

10th

P. 32-40-24

Diqdiqqah

U.18073

1931–2

10

L.BM 123237

No. 9

U.18702

1933–4

12th

?

No. 10

U.18072

1931–2

10

?

No. 11

U.16944

1930–1

9th

?

th

th st

th

th

th

The first two were chosen by order of importance and by their difference with all the others: they are the only two which exceed 10 cm; one is 75 cm high, the other 63 cm high (Fig. 5). So their use is not the same as the others clay objects. Woolley suggests that they could be used at the entrance of shrines. And it is quite possible. 9 For the others, the choice seems to have been mostly made on the basis of categories (as you can see in Table 3). Order by categories represent 99.8% of the 261 figurines assem­ bled by Woolley in the “terracottas” section. In the order by category, Woolley followed not the principle of importance (gods are not the first), but the number of clay figurines dis­ covered (Chart 1). First human figurines then divine, then couples, then grotesque body, and finally animals and objects. In reality, in the order by categories Woolley considered not only the number of terracottas found, but also their “importance;” that is, not the importance attributed to them in the Ancient Near East, Fig. 5. Terracotta no. 1 (left) and no. 2 but that of the time of Woolley. 10 This is clear (right) from Ur (UE 7: pl. 64). 9. There are numerous temples which have clay sculptures at their entrance since the Middle Bronze Age (Tell Harmal, Khafajah Mound D, Isin, Nasirya, Usiyeh, Tello, Mashkan Shapir, Haradum, Assur, Bija island), through the Late Bronze Age (Nuzi, Tell al Rimah) and until the Iron Age (Clai’a): cf. Braun-Holzinger 1999 and Battini 2009c. 10.  The importance is demonstrated by the series HAR.ra/ḫubullu (Landsberger 1934 and 1962), by the “wisdom” literature (Lambert 1960: passim), by their presence in the literature (Heimpel 1968)

Old Babylonian Terracottas from Ur

Terracotta no.

Type of Order

Percent

Nos. 1–2

IMPORTANCE

0.8%

Nos. 3–261

CATEGORIES Nos. 3–90 = Humans Nos. 91–158 = Gods Nos. 159–183 = Couples Nos. 184–208 = Grotesque Nos. 209–261 = Animals and different objects

99.2%

Table 3. Woolley’s presentation of terracottas.

139

Chart 1. Number of figurines by cate­gory in Woolley’s section “Terracottas.”

if one has a quick look at the animals category: this one is composed of a number of figurines which is more than the double of the “grotesque” and “couples” categories. But animals are presented after the humans, even after the less numerous cate­ gories of humans (“grotesque” and “couple”). Clearly, in Woolley’s opinion animals have to be separated from humans (this is not the case in the ancient Near East), they are more similar to objects than to living people (today after the diffusion of “animal studies” one would scream at this consideration). 11 But the percentage itself of animals—not so far away from that of gods (cf. Chart 1)—tends to prove that for Mesopotamians they are also important in everyday life. 12 In my analysis of how Woolley presented clay figurines and other clay objects, I believe I have discovered his way of thinking, a reflection quite important on terracottas, unstated but really deep. In fact, Woolley was not limited to drawing broad thematic categories. He also tried to distinguish for each category some trends and in this manner (that is implicit) he arranged terracottas to be published. Here is an example: in the category “humans” Woolley distinguished women from men (and women are treated before men because of their larger number). For the human masculine figures he distinguished (Table 4, Fig. 6) worshipper (nos. 63–66), worshipper in attitude of adoration (nos. 67–68), advancing men (nos. 69–70 and nos. 73–80), worshipper again (nos. 81–86), and men accompanied by monkeys (nos. 87–90). Inside the sub-category “advancing men” Woolley distinguished implicitly “simple” men (nos. 69–70) and what I consider the king. 13 and by their multiple representations in the artistic and craft objects (Breniquet 2002, which underlines how the iconological data are rich because animals are represented everywhere, even if these data are also limited because of the difficulties of an exact identification of animals). Apart from this general study on the importance of animals in the iconography of Mesopotamia (Breniquet 2002), other interest­ ing writings more specific on the subject are numerous: e.g. Van Buren (1939), Amiet (1956), Heimpel (1968), Paraire (2000), Watanabe (2002), and Straw (2005). But the importance of animal was quite soon understood: Boussac (1915), for example, underlined the figure of the lion. Digard (1990) is one of first zoologists of a new trend, who like to demonstrate the particular and special relationship between men and animals. 11.  Digard 1990; Bodson 1997: passim, especially vii–x, 43–51, 119–154. 12. In the ancient Near East animals were a constant source of inspiration because of their importance for human subsistence but also for the enjoyment of life (Limet 1997; Breniquet 2002; Bat­ tini 2009a). And the number itself of animal representations in every kind of artistic/craft production demonstrate the importance of animals for Mesopotamians (Breniquet 2002; see n. 10). 13.  One can identify without doubt this figure as royal because of his hat and other details (beards, objects accompanying him, attitudes, etc.); cf. Wrede 2003: 311–312. On the contrary, Barrelet (1969) and Opificius (1961) did not identify this character as royal.

140

Laura Battini

Terracotta no.

SERIES

Nos. 63–66

Worshipper

Nos. 69–70

Man advancing

Nos. 73–80

King advancing

Nos. 81–86

King worshipper

No. 88

King

Nos. 87, 89–90

Man with monkeys

Table 4. Reconstruction of the Woolley raisonnement in distinguishing human male figurines.

Fig. 6. Terracotta no. 63 (left) (UE 7: pl. 71), nos. 89–90 (center) (UE 7: pl. 73 ), and no. 73 (right) (UE 7: pl. 72) from Ur.

In the same manner, if there are two groups of worshippers it is due to Woolley’s implicit distinction between common men (nos. 63–66) and the figure I suggest to be the king (nos. 81–86). In the category of human female figurines I found the same “arborescence’s” 14 distinctions. Unlike sev­ eral scholars, according to whom clay feminine figures are quite the same (nude, holding her breast, in a frontal position), Woolley is very smart. He distinguished two main types, and two other less frequent ones: the two main types are one “covering (better sustaining) her breast with her hands” (for example UE 7: pl. 64 no. 6; Fig. 7) and one—more common at Ur—of a nude female holding her hands under her breasts (UE 7: pls. 66–69 nos. 20–45). In the first case the figure sup­ ports the breasts, in the second, on the contrary, she can not, because her hands are superimposed Fig. 7. Female terracotta sustaining (for example nos. 20, 25, 26; and probably 21, 29, her breast (left) and holding her hands 32, 34–44) or one hand touch the other (nos. 22, under her breasts (right) from Ur (UE 23, 45). Other nude types have hands along the 7: pl. 66 no. 25 and UE 8: pl. 30 no. 9). body or one on the belly and one raised. There are also other female figurines not naked but dressed: harpist player, mother feeding her child, woman crossing her hands on the belly, woman holding her breast, etc. (Fig. 8). Woolley arranged most of these dressed figurines after the female nude, exactly respecting the order of quantitative importance. An order of conceptual importance (not quantitative at all) is recognized in Woolley’s presentation of divine figures. In this case, despite the fact that clay goddesses are more numerous than gods, the latter are presented first. And this Fig. 8. Terracotta no. 56 (UE 7: pl. 70) and no. 50 (UE 7: pl. 69) from Ur. raises one question: if in the official religion—as 14.  For “arboriscence” I intended a series of distinctions, each one depending on the preceding and more and more detailed.

Old Babylonian Terracottas from Ur

141

reported by texts—gods are more powerful and active than goddesses and also more important, why in clay production are goddesses represented twice as much as gods? It is too simple to explain this difference by arguing a difference between official religion and a popular or personal one. Perhaps this was in part true, but after some research I have reached the opinion that in Mesopotamia official religion is not so far away from a personal one, between the two there is only a difference on stresses but both respond quite similarly to the same human necessities. 15 So, for the clay figurines I keep thinking that there is only a difference of stress: the most pressing need for Mesopotamians who lived badly, quickly and shortly, was reproduction. I wonder if this stress on females—that are human or divine—are not almost in part due to this necessity of ensuring continuity of their species. And this is not in contrast with the male predominance in official religion. It is only a matter of stress, not an opposition of popular versus official religion.

III. New Perspectives: Iconology and Other Paths Let us stop at this point for what concerns “typology.” Just one more element: the themes of Old Babylonian Ur terracottas resemble those from other Old Babylonian towns (Fig. 9). For example, the terracottas from Uruk or Tello 16—one of the richest and more recently published collections of clay figurines—present the same themes: human worshippers, man advancing with a weapon, king advancing, king wor­ shipper, female naked, female dressed, human couples, divine couples, gods with ox ears, goddess on goose, goddess with rosettes, “Humbaba”, animals, etc. 17 But what is more important than to underline this cultural unity of South Mesopota­ mia is to note the change in the different percentage of these themes from one city to another and to check for differences, because these are the better tools for arriving at a deep comprehension of mental sphere and specific needs of people. For example, out of thirty-two clay figurines of gods, eighteen (that is, Fig. 9. Terracottas from Tello (from left to right and up to down: 56.3%) are of that strange Barrelet 1968: nos. 305, 221, 530, 392, 363, and 65). 15.  Battini 2015b (and related bibliography). 16.  Wrede 2003: 253–352, nos. 892–1265; Barrelet 1969: pls. I–LI, nos. 1–541. . 17.  Another comparison is with the figurines of Tello (published in Barrelet 1969: 142–301, pls. I–LI), which give the same result: female figurines, naked or dressed, male figurines, gods, goddesses, “Humbaba,” animals, etc. According to Barrelet (1969: 233–236, 242–248, pls. XXIX–XXXI) the figurine with rosettes or with vases are not surely divine; she prefers the definition “undetermined nature.” But in comparison with Woolley’s discoveries and also based on the analysis of Opificius (1961: 76, 207–208), I tend to think that they are divine.

142

Laura Battini

personage with ox ears (Fig. 10), which raises a difficult interpretation. 18 The importance of this god at Ur—no matter his identifica­ tion—cannot be denied. And also in this case there is a small shift between official religion, dominated by Sin, and clay representations. Is Sin the ox-eared god? In texts there is no special connection between oxen and Sin. If we want to answer the question of the iden­ tification of this god we have to consider the reasons for his appearance, why in this period specifically more than in another one, and the reasons for his disappearance. Among other strange figurines, attested only or nearly only at Ur, is a terracotta representing a female harpist (Fig. 11). It is the only one known at the moment and this changes also our way to think about ancient musicians. More enigmatic and really strange is the representation of a female grotesque figure, which is the female pendant (Fig. 12) of what we call Fig. 10. Gods with ox ears from Ur (UE 7: pl. probably wrongly 19 74 nos. 91–100). “Humbaba.” This fig­urine presents the same position of legs as in the so-cal­ led “Humbaba”—slightly open but with knees wide open—, the same preeminent belly, the same position of arms—one raised, one down—, the same nudity and frontality, etc. Apart from the two examples of Ur (one broken, the other well-conserved), I found a similar figurine in Van Buren and Spycket. 20 This figurine is really in opposition with the more common female representation, it is a denial not only of some kind of beauty but especially of dignity. If we do not understand her male counterpart, it will be seriously Fig. 11. Female harpist from awkward to attempt to interpret now this female figurine. Ur (UE 7: pl. 84 no. 177). 18.  Wrede 2003: 318–323; Tomabechi 1984: 11; Barrelet 1969: 181–184, 382–383; Opificius 1961: 90–100, 214–215. 19.  In the Gilgamesh epic (most recently, George 2003), the guardian of the Forest of cedars, is a terrible monster. This is really not the case of the Humbaba we know from figurines, always represented, in a full corpse, as ridiculous and grotesque (cf. Wrede 2003: 338–342, nos. 1228–1232). The problem of the irreconcilability between the literary image and the clay figures supposedly of Humbaba was stressed by Opificius (1961: 222) and Barrelet (1969: 196–198), without arriving at a solution. An interesting one was offered by Howard-Carter (1983: 72): a possible link between the so called Humbaba clay figurines and Ishtar on the basis of the similarities between Humbaba and Bes and of Ishtar and Hator. Wrede (2003: 340) seems also to appreciate this interpretation. A completely new analysis is necessary for a bet­ ter comprehension. For a recent study of this personage see Ornan 2010. For the necessity of analysing terracottas see Hamilton et al. 1996. 20.  Van Buren 1930: 263–264; Spycket 1992b: no.745.

Old Babylonian Terracottas from Ur

143

In this paper, there is not room to discuss about the meaning of all the terracottas, so I’m only going to give some of the high­ lights of the research. In fact, typological studies serve only if they are not an end in themselves, if they are inserted in a bigger study. Certainly, as A. Schnapp recently recalled, “sans typologie il n’y a d’autre voie que celle de l’esthétique”. 21 But according to Cesare Ripa (1593), the first who used the term, “iconologia” is “lo studio, strutturato sistematicamente, del significato delle im­ magini.” Of course other scholars have the Fig. 12. Female grotesque figures from Ur (UE same conception of iconology: from Andrea 7: pl 87 nos. 205 and 204). Fulvio or Vincenzo Cartari to Aby Warburg and of course Erwin Panovsky, but also all scholars having access to Warburg’s library (E. Cassirer, R. Wittkower, E. Ernst Gombrich and F. Saxl) passing among others by Honoré Lacombe de Prézel, Ennio Quirino Visconti, Joachim Sandrart 22 to reach today Miller. If iconology can be today defined as “l’interpretazione integrale di un’opera d’arte che muove dal suo contenuto, mediante ricerche che mirano a ritrovarne le ragioni contestuali, anche nell’ambito della storia delle idee religiose, filosofiche, politiche, sociali,” I agree with Warburg about the importance of the two: iconography and iconology, the first one is essential to have then access to the iconologist analysis. 23 Only in this way can artistic/craft pro­ ductions—like here terracottas— possibly relate some of the needs, the problems, the breathings, the mental sphere of ancient people. Let us have an example with the iconographic analysis and then with iconological ones. Among the terracottas of Ur ten concern the so called “goose goddess” (Fig. 13). 24 Fig. 13. Goose goddess from Ur (UE 7: pl. 80 nos. 147 Through iconographic analysis it and 148. 21.  Schnapp 1998: 395. 22.  For a history of iconology see, among others, Recht 2013. 23. Some articles and books of this author on the subject are not yet widely read, but they are essential in the multiple approach attempted by Warburg (Lescourret 2013). See at least La Naissance de Vénus et Le Printemps de Sandro Botticelli (2007) and especially L’atlas mnémosyne, pub­ lished only eighty-three years after the death of the author (2012), but rich in interpretations and suggestions (https://warburg.sas.ac.uk/collections/warburg-institute-archive/bilderatlas-mnemosyne/ mnemosyne-atlas-october-1929). 24.  There are thirteen other examples from other southern sites, but Ur is the richest, roughly 50%. The ten terracottas found at Ur are now conserved in multiple places: two in the British Museum (BM 127484, BM 1933-10-13, 215), three or four in Philadelphia (CBS 16268, CBS 16267, 31-43-421, perhaps also Opificius no. 244 = U.978), one in Baghdad (Iraq Museum, IM 1617), and three are without sure location (U.16471, U.1162, Opificius, no. 243).

144

Laura Battini

would be possible to say that there are four types of representations, 25 and that the figure is out at the end of the Old Babylonian period. But according to iconological analysis, 26 it is also important to understand the differences of significance of these four types of representations and especially to think about the identification of the goddess, about the symbolic value of geese, and about the reasons for the disappear­ ance of the figure, thanks also to the study of the textual data. 27 One more example is a wonderful and astonishing image of a goddess with a bull head discovered in Isin and published by Agnès Spycket. 28 Another similar terracotta is today in the Yale Babylonian Collection, published by Buchanan, without sure provenance. While the iconographic analysis will consider only the description of the figure and the search for other similar examples, the iconological one will try to understand the significance and the context of this divine image. 29

IV. Final Remarks Before ending, I would like to stress the importance of the symbolic and ideo­ logical comprehension of terracottas, which would be the essential basis for a better understanding of the religious/philosophical world of ancient people. But other important elements, most often neglected by scholars, have to be considered in the analysis of clay figurines: context—Ur context, especially for Diqdiqqah, is often lacking—, colours—we have information from texts on clay figurines only for the first millennium—, light—light can highlight the terracottas and enhances their emotional impact—, and use—3D clay models serve differently from figurines and from clay plaques. 30 So, still a lot of work and—let us hope so—new data.

25. Battini 2006: The four representations are: 1) goddess represented sitting over a goose and having the feet over another one, the head frontal, the body in profile, the left hand raised in greetings, and the right hand holding a vase; 2) goddess represented sitting in full frontality over two geese and accompanied by astral symbols; 3) goddess represented standing, completely in profile—except the shoul­ ders—, holding a vase with flowing waters and accompanied by geese; 4) goddess represented sitting in full frontality over a goose, holding a vase with flowing waters. 26.  Battini 2006. 27.  This kind of analysis is followed for example by Breniquet (1996, 2007–8, 2011, 2012, and 2016), Cornelius (2004, 2009, and 2013), Dolce (1986, 1991, 1997, 2004, 2012, and 2016), Mazzoni (1977, 1986, 1992, 2002, 2005, 2008, and 2012), Muller (2002, 2010, 2012, 2013, 2016a, and 2016b), Ornan (2002, 2004–5, 2005, 2009, 2010, 2012, 2013, 2014, and 2016), Otto (1992, 2006, 2010, 2013, and 2016), Pittman (1994, 1996a, 1996, 2002, and 2014), Rova (2006, 2010, 2012, and 2016), Suter (2000, 2007, 2011, and 2014; Ead. et al. 2006) and Winter (1986, 1995a, 1995b, 1997, 1998, and 2012), and some others; specifi­ cally on the geese by Battini 2006. For another interpretation of this goddess see: Maxwell-Hyslop 1992. Her approach tends to use too much directly textual data to identify gods represented in art—on the line of Frankfort (1934)—is largely outdated today (cf. Winter 1983, 1986, 1997, 1999; Green 1997; Lambert 1997; Assante 2002; Matthiae 1992; and Battini 2009b), even if this approach was modern and really new in the forties. In comparison with Frankfort, a more cautious position was just adopted by Amiet (1961, 1976 and 1980). 28.  Spycket 1992a, 1994; and Buchanan 1962. For an analysis of this subject see Battini 2009b. 29.  Battini 2009b. 30.  Context: Hodder 1982a, 1982b, 1987; Hodder et al. 1995; Shanks 1991; Shanks and Tilley 1987. Colours: Nunn 2010 and 2015, Pastoureau 1989. Light: Matthews and Curtis 2012: 152–514, and especially Ornan 2012, Winter 2012 and Winter 1994. Use: Battini 2017.

Old Babylonian Terracottas from Ur

145

Bibliography Amiet, P. 1956 Le symbolisme cosmique du répertoire animalier en Mésopotamie. RA 50: 113–126. 1961 La glyptique mésopotamienne archaïque. Paris: CNRS Éditions. 1976 L’Art d’Agadé au Musée du Louvre. Paris: Editions des Musees Nationaux. 1980 The Mythological Repertory in Cylinder Seals of the Agade Period (c. 2335–2155 BC). Pp. 35–59 in Ancient Art in Seals, ed. E. Porada. Princeton: Princeton Uni­ versity Press. Assante, J. 2002 Style and Replication in ‘Old Babylonian’ Terracotta Plaques: Strategies for En­ trapping the Power of Images. Pp. 1–29 in Ex Mesopotamia et Syria Lux. Festschrift für Manfried Dietrich zu seinem 65. Geburtstag, eds. O. Loretz, et al. AOAT 281. Münster: Ugarit. Barrelet, M.-Th. 1969 Figurines et reliefs en terre cuite de la Mésopotamie antique I. Potiers, termes de métier, procédés de fabrication et production. BAH 78. Paris: P. Geuthner. Battini, L. 2006 La déesse aux oies: une représentation de la fertilité? RA 99: 57–70. 2009a La conception des animaux domestiques et des animaux de compagnie dans la Mésopotamie d’époque historique. Res Antiquae 6: 169–195. 2009b La terre cuite IB 1967 ou peut-on lire les images coroplastiques à travers les textes? Iraq 71: 125–138. 2009c Une inscription sur brique d’Enannatum I: un nouveau regard sur les lions gardi­ ens des portes en Mésopotamie. Anthropolozoologica 44/1: 189–209. 2015a Les maisons à Nuzi: nouvelle approche archéologique. Cahiers des thèmes transversaux d’ArScAn 12: 325–345. 2015b How Better Understanding of Ritual Practises Can Help the Comprehension of Religious Feelings? Pp. 176–183 in Defining the Sacred. Approaches to the Archaeology of Religion in the Near East, ed. N. Laneri. Oxford and Philadelphia: Oxbow Books. 2017 L’épiphanie divine en Mésopotamie a travers les terres cuites. Akkadica 138: 69–106. Bodson, L. 1997 L’animal de compagnie: ses rôle et leurs motivations au regard de l’histoire. Col­ loque d’histoire des connaissances zoologiques 8. Liège: University of Liège. Boussac, P. H. 1915 Iconographie zoologique des monuments assyro-chaldéens. Le lion. RA 12: 173–188. Braun Holzinger, E. A. 1999 Apotropaic Figures at Mesopotamian Temples in the Third and Second Millen­ nia. Pp. 149–172 in Mesopotamian Magic. Textual, Historical and Interpretative Perspectives, eds. T. Abusch and K. van der Toorn. Ancient Magic and Divination 1. Groningen: Styx Publications. Breniquet, C. 1996 Du fil à retordre: réflexions sur ‘les idoles aux yeux’ et les fileuses de l’époque d’Uruk. Pp. 31–53 in Collectanea orientalia: histoire, arts de l’espace et industrie de la terre : études offertes en hommage à Agnès Spycket, eds. B. Hrouda and H. Gasche. CPOA 3. Neuchâtel; Paris: Recherches et publications. 2002 Animals in Mesopotamian Art. Pp. 145–168 in A History of the Animal World in the Ancient Near East, ed. B. J. Collins. Leiden/Boston/Köln: Brill.

146

Laura Battini

2006 Ce lin, qui me le peignera? Enquête sur la fonction des peignes en os du Néo­ lithique précéramique levantin. Syria 83: 5–13. 2007/8 Buvait-on de la bière au chalumeau en Mésopotamie à l’époque protodynastique? Cahier des thèmes transversaux ArScAn 9: 347–353. 2011 Une plaquette ‘au harpiste’ d’Eshnunna. Pp. 283–296 in Profils d’objets. Approches d’anthropologues et d’archéologues, ed. F. Wateau. Colloques de la Maison René-Ginouvès 7. Paris: de Boccard. 2012 Objets et images. Pp. 92–98 in L’archéologie en mouvement : hommes, objets, espaces et temporalités, eds. S. de Beaune and H.-P Francfort. Paris: CNRS-Editions. 2016 Ur Nanshé, un inconnu illustre. Pp. 41–56 in Mille et une empreintes. Un Alsacien en Orient. Mélanges en l’honneur du 65e anniversaire de D. Beyer, eds. J. Patrier, P. Quenet, and P. Butterlin. Subartu 36. Turnhout: Brepols. Buchanan, B. 1962 Ancient Near Eastern Art in the Yale Babylonian Collection. Archaeology 15: 267–275. Cornelius, I. 2004 The Many Faces of the Goddess: The Iconography of the SyroPalestinian Goddesses Anat, Astarte, Qedeshet, and Asherah c.1500–1000. OBO 204. Göttingen: Vandenhoeck & Ruprecht. 2009 Aspects of the Iconography of the Warrior Goddess Ishtar and Ancient Near East­ ern Prophecies. Pp. 15–40 in Images and Prophecy in the Ancient Eastern Mediterranean, eds. M. Nissinen and C. E. Carter. FRLANT 233. Göttingen: Vanden­ hoeck & Ruprecht. 2013 ‘Trading Religion’ and ‘Visible Religion’ in the Ancient Near East. Pp. 139–165 in Religion and Trade. Religious Formation, Transformation and Cross-Cultural Exchange between East and West, eds. P. Wick and V. Rabens. Leiden: Brill. Digard, J.-P. 1990 L’homme et les animaux domestiques. Anthropologie d’une passion. Paris: Fayard. Dolce, R. 1986 Per una riconsiderazione delle opere figurative da Gebelet el-Beyda. Contributi e Materiali di Archeologia Orientale 1: 307–331. 1991 La produzione artistica e il Palazzo di Ebla nella cultura urbana della Siria del III millennio. La Parola del Passato 46: 237–269. 1997 Dualità e realtà virtuale nel Palazzo di Assurnasirpal II a Nimrud. Contributi e Materiali di Archeologia Orientale 7: 142–162. 2004 The ‘Head of the Enemy’ in the Sculptures from the Palaces of Nineveh. Iraq 66: 121–132. 2012 La comunicazione visuale come linguaggio del potere: alcuni esempi in Mesopota­ mia e in Siria. Pp. 19–23 in Atti della giornata della ricerca 2008, eds. M. Micheli and F. Rangoni. San Casciano (FI): libroCo.Italia. 2016 Sulle donne nelle società arcaiche di Mesopotamia e di Siria. Note a margine. Hen 38/2: 208–225. Frankfort, H. 1934 Gods and Myths on Sargonid Seals. Iraq 1: 2–29. George, A. R. 2003 The Babylonian Gilgamesh Epic. Oxford: Oxford University Press. Green, A. 1997 Myths in Mesopotamian Art. Pp. 135–158 in Sumerian Gods and Their Representations, eds. I. L. Finkel and M. J. Geller. CM 7. Groningen: Styx Publications. Hamilton, N. et al. 1996 Can we Interpret Figurines? Cambridge Archaeological Journal 6/2: 281–307.

Old Babylonian Terracottas from Ur

147

Heimpel, W. 1968 Tierbilder in der sumerischen Literatur. Studia Pohl 2. Roma: Pontificium Insti­ tutum Biblicum. Hodder, I. 1982a The Present Past. London: Batsford. 1982b Symbols in Action. Cambridge: Cambridge University Press. 1987 The Archaeology of Contextual Meanings. Cambridge: Cambridge University Press. Hodder, I. et al. 1995 Interpreting Archaeology: Finding Meaning in the Past. London: Routledge. Howard-Carter, T. 1983 An Interpretation of the Sculptural Decoration of the Second Millennium Temple at Tell al-Rimah. Iraq 45/1: 64–72. Lambert, W. G. 1960 Babylonian Wisdom Literature. Oxford: Oxford University Press. 1997 Sumerian Gods: Combining the Evidence of Texts and Art. Pp. 1–9 in Sumerian Gods and Their Representations, eds. I. L. Finkel and M. J. Geller. CM 7. Gronin­ gen: Styx Publications. Landsberger, B. 1934 Die Fauna des alten Mesopotamiens nach des 14. Tafeln der Serie HAR. RA=hubullu. Leipzig: S. Hirzel. 1962 The Fauna of Ancient Mesopotamia. HAR-ra=hubullu Tablets XIV and XVIII. MSL 8/2. Rome: Pontificium Institutum Biblicum. Lescourret, M.-A. 2013 Aby Warburg ou la tentation du regard. Paris: Hazan. Limet, H. 1997 Animaux compagnons ou: de compagnie? La situation dans le Proche-Orient an­ cien. Pp. 53–73 in L’animal de compagnie: ses rôles et leurs motivations au regard de l’histoire, ed. L. Bodson. Colloque d’histoire des connaissances zoologiques 8. Liège: University of Liège. Lion, B. 1999 Les archives privées d’Arrapha et de Nuzi. Pp. 35–62 in Studies on the Civilization and the Culture of Nuzi and the Hurrians, vol. 10, eds. D. I. Owen and G. Wilhelm. Bethesda, MD: CDL Press. Mallowan, M. E. L. and Wiseman, D. J. 1960 Ur in Retrospect. In Memory of Sir C. Leonard Woolley. Iraq 22. Matthews, R. and Curtis, J. 2012 Proceedings of the 7th International Congress on the Archaeology of the Ancient Near East, London, 12–16 April 2010. Volume 2: Colour & Light in Architecture, Art & Material Culture. Wiesbaden: Harrossowitz. Matthiae, P. 1992 Figurative Themes and Literary Texts. Pp. 219–241 in Literature and Literary Language at Ebla. Quaderni di semitistica 18. Florence: Department of Linguis­ tics, University of Florence. Maxwell-Hyslop, K. R. 1992 The Goddess Nanshe. An Attempt to Identify Her Representation. Iraq 54: 79–82. Mayer-Opificius, R. 1996 Feldzeichen. Pp. 213–226 in Collectanea orientalia: histoire, arts de l’espace et industrie de la terre : études offertes en hommage à Agnès Spycket, eds. B. Hrouda and H. Gasche. CPOA 3. Neuchâtel; Paris: Recherches et publications. Mazzoni, S. 1977 Studi sugli avori di Ziwiye. Studi semitici 49. Roma: Università di Roma Editrice.

148

Laura Battini

1986 La dea alata con veste paleosiriana. Egitto e Vicino Oriente 9: 83–96. 1992 Significato e ruolo del paesaggio nei rilievi di Sennacherib. Contributi e Materiali di Archeologia Orientale 4: 151–166. 2002 Percorsi figurativi tra Oriente e Occidente: immagini di donne dell’arte siro’ittita. Pp. 343–362 in Da Pyrgi a Mozia. Studi sull’archeologia del Mediterraneo in memoria di Antonia Ciasca, eds. M. G. Amadasi Guzzo, M. Liverania, and P. Mat­ thiae. Quaderni Vicino Oriente 3/2. Rome: Università degli Studi di Roma La Sapienza Editore. 2005 Pyxides and Hand-Lion Bowls: A Case of Minor Arts. Pp. 43–66 in Crafts and Images in Contact, eds. C. Suter and C. Uehlinger. OBO 210. Gottingen: Vanden­ hoeck & Ruprecht. 2008 Arts, Crafts and the State: A Dialectic Process. Pp. 37–46 in Proceedings of the 5th International Congress on the Archaeology of the Ancient Near East, eds. J. M. Cordoba, M. Molist, et al. Madrid: Ediciones Universidad Autónoma de Madrid. 2012 l re e la comunicazione del potere nell’arte siro-ittita (XI-X sec. a.C.). Pp. 11–31 in Arte-Potere. Forme artistiche, istituzioni, paradigmi interpretativi. Atti del convegno di studio tenuto a Pisa Scuola Normale Superiore, 25–27 Novembre 2010, eds. M. Castiglione and A. Poggio. Milan: LED. Muller, B. 2002 Les maquettes architecturales du Proche-Orient ancien. BAH 160. Beyrouth: Paul Geuthner. 2010 Vêtements et parures dans l’Antiquité proche-orientale. Pp. 83–90 in Décors des corps en couleurs, eds. G. Boetsch, D. Chevé, and H. Claudot-Hawad. Paris: Ed. CNRS. 2012 Mutations vers la ville, mutations dans l’art: la peinture murale. Pp. 99–116 in Du village néolithique à la ville syro-mésopotamienne, ed. J.-L. Montero Fenollós. Ferrol: Université de la Corogne. 2013 Espace paysager et espace construit dans l’iconographie syro-mésopotamienne. Topoi Suppl. 12: 371–380. 2016a Elements of War iconography at Mari. Pp. 13–28 in Making Pictures of War. Realia et Imaginaria in the Iconology of the Ancient Near East, ed. L. Battini. Oxford: Archaeopress. 2016b Du Proche-Orient ancien à l’Occident médiéval: «concordances iconographiques». Pp. 367–378 in Mille et une empreintes. Un Alsacien en Orient. Mélanges en l’honneur de Dominique Beyer à l’occasion de son 65e anniversaire, eds. J. Patrier, P. Quenet, and P. Butterlin. Subartu 36. Turnhout: Brepols. Nunn, A. 2010 Farben und Farbigkeit auf mesopotamischen Statuettes. Pp. 427–448 in Kulturlandschaft Syrien. Zentrum und Peripherie. Festschrift für Jan-Waalke Meyer, eds. J. Becker and E. Rehm. AOAT 371. Münster: Ugarit Verlag. Nunn, A. et al. 2015 Polychromy on Mesopotamian Stone Statues. Preliminary Report. Studia Mesopotamica 2: 187–206. Opificius, R. 1961 Das altbabylonische Terrakottarelief. UAVA 2. Berlin: de Gruyter. Ornan, T. 2002 The Queen in Public: Royal Women in Neo-Assyrian Art. Pp. 461–477 in Sex and Gender. Proceedings of the 47th Rencontre Assyriologique Internationale, Helsinki, July 2–7, 2011, eds. S. Parpola and R. M. Whiting. Helsinki: Neo-Assyrian Text Corpus Project. 2004/5 Expelling Demons at Nineveh: On the Visibility of Benevolent Demons in the Palace of Nineveh. Iraq 66: 83–92.

Old Babylonian Terracottas from Ur

149

2005 The Triumph of the Symbol. OBO 213. Fribourg: Academic Press. 2009 The Likeness of a Man. Reflections on the Anthropocentric Perception of the Di­ vine in Mesopotamian Art. Pp. 93–151 in What is a God?, ed. B. Porter. Winona Lake, IN: Eisenbrauns. 2010 Humbaba, the Bull of Heaven and the Contribution to Gilgamesh Epic. Pp. 229– 260 in Gilgamesch: Ikonographie eines Helden, ed. H. U. Steymans. OBO 45. Fri­ bourg/Göttingen: Fribourg Academic Press/ Vandenhoeck & Ruprecht. 2012 The Role of Gold in Royal Representation: The Case of a Bronze Statue from Hazor. Pp. 445–458 in Proceedings of the 7th International Congress on the Archaeology of the Ancient Near East, London, 12–16 April 2010, eds. R. Matthews and J. Curtis. Wiesbaden: Harrassowitz. 2013 A Silent Message: Godlike Kings in Mesopotamian Art. Pp. 569–595 in Critical Approaches to Ancient Near Eastern Art, eds. B. Brown and M. H. Feldman. Ber­ lin: Walter de Gruyter. 2014 A Mitannian Cylinder Seal with a Worshipper and Divine Images. Pp. 1020–1022 in The Smithsonian Institution Excavations at Tell Jemmeh, Israel, 1970–1990, eds. D. Ben-Shlomo and G. W. Van Beek. Washington: Smithsonian Institution Scholarly Press. 2016 The Beloved Ne’ehevet and Other Does: Reflections on the Motif of the Grazing or Browsing Wild Horned Animal. Pp. 279–302 in Alphabets, Texts and Artefacts in the Ancient Near East, eds. I. Finkelstein, C. Robin, and Th. Römer. Paris: Van Dieren. Otto, A 1992 Zur offiziellen Ikonographie auf Siegeln aus der Regierung des Königs ShamshiAdad I. APA 24: 159–171. 2006 Das Oberhaupt des westsemitischen Pantheons ohne Abbild? Überlegungen zur Darstellung des Gottes Dagan. ZA 96: 242–268. 2010 The Iconography and Use of the Ur SIS 8-4 Seals Reconsidered. Pp. 19–29 in Shepherds of the Black-headed People: The Royal Office vis-à-vis Godhead in Ancient Mesopotamia, eds. K. Saskova and P. Charvát. Plzen: University of West Bohemia in Plzen. 2013 Königssiegel als Programm. Überlegungen zum Selbstverständnis altorien­ talischer Herrscher und zur Deutung der Tierkampfszene. ZA 103: 54–68. 2016 Much More Than Just a Decorative Element: The Guilloche as Symbol of Fertil­ ity. Pp. 379–393 in Mille et une empreintes. Un Alsacien en Orient. Mélanges en l’honneur de Dominique Beyer à l’occasion de son 65e anniversaire, eds. J. Patrier, P. Quenet, and P. Butterlin. Subartu 36. Turnhout: Brepols. Parayre, D. 2000 Les animaux et les hommes dans le monde syro-mésopotamien aux époques historiques. Colloque Internationale du 4 et 5 décembre 1998 à Lille. Topoi Suppl. 2. Lyon: Maison de l’Orient. Pastoureau, M. 1989 Couleurs, Images, Symboles. Paris: Le léopard d’or. Pittman, H. 1994 The Glazed Steatite Glyptic Style: The Structure and Function of an Image System in the Administration of Protoliterate Mesopotamia. BBVO 16. Berlin: D. Reimer. 1996a The White Obelisk and the Problem of Historical Narrative in the Art of Assyria. Art Bulletin 78: 334–355. 1996b Constructing Context; the Gebel el Arak Knife. Pp. 1–20 in The Study of the Ancient Near East in the Twenty-First Century, eds. G. Schwartz and J. Cooper. Winona Lake, IN: Eisenbrauns.

150

Laura Battini

2002 Bears at Brak. Pp. 287–296 in Of Pots and Plans. Papers on the Archaeology and History of Mesopotamia and Syria Presented to David Oates in Honour of his 75th Birthday, eds. J. Werr, L. al-Gailani, et al. London: Nabu Publications. 2014 Hybrid Imagery and Cultural Identity in the Bronze Age of Exchange: Halil River Basin and Sumer meet in Margiana. Pp. 625–636 in My Life is like the Summer Rose Papers in honor of Maruizio Tose for his 70th Birthday, eds. C. C. Lamberg Karlovsky, B. Genito, and B. Cerrassti. BAR International Series 2690. Oxford: Archaeopress. Recht, R. 2013 L’iconologie avant Warburg. Images Re-vues [En ligne], Hors-série 4|2013, mis en ligne le 30 janvier 2013, consulté le 10 juillet 2016. See http://imagesrevues. revues.org/2898. Ripa, C. 1593 Iconologia overo Descrittione dell’Imagini universali cavate dall’antichità et da altri luoghi Da Cesare Ripa Perugino. Opera non meno utile, che necessaria à Poeti, Pittori, & Scultori, per rappresentare le viru, vitij, affetti, & passioni humane. Milano. Rova, E. 2006 The Eagle and the Snake: Remarks on the Iconography of Some Archaic Seals. Kaskal 3: 1–29. 2010 A Re-discovered Naked Man of the Uruk Period. Pp. 269–284 in Festschrift für Gernot Wilhelm anläßlich seines 65 Geburtstages am 28. Januar 2010, ed. J. Fincke. Dresden: ISLET. 2012 ‘Themes’ of Seals Images and their Variants: Two New Examples from Tell Bey­ dar. Pp. 745–761 in Leggo! Studies Presented to Frederick Mario Fales on the Occasion of His 65th Birthday, eds. G. B. Lanfranchi, D. Morandi Bonacossi, et al. Leipziger Altorientalische Studien 2. Wiesbaden: Otto Harrassowitz Verlag. 2016 Ishtar in Shida Kartli?: About a Recently Discovered Fragment of Stone Plaque. Pp. 511–531 in Libiamo ne’ lieti calici: Ancient Near Eastern studies presented to Lucio Milano on the occasion of his 65th birthday by Pupils, Colleagues and Friends, eds. P. Corò, et al. AOAT 436. Münster: Ugarit-Verlag. Schnapp, A. 1998 La conquête du passé: aux origins de l’archéologie. Paris: Librairie générale française. Shanks, M. 1991 Experiencing the Past: On the Character of Archaeology. London/New York: Routledge. Shanks, M. and Tilley, C. 1987 Social Theory and Archaeology. Cambridge: Cambridge University Press. Shortland, A. et al. 2008 Reassessing Bronze Age Manufacturing Technologies at Nuzi. Mater. Res. Soc. Symp. Proc. N.1047. See http://journals.cambridge.org/action/displayAbstract?fr omPage=online&aid=8025222. Spycket, A. 1992a Les figurines en terre cuite, 1986–89. Pp. 56–73 in Isin-Ishan Bahriyat IV, ed. B. Hrouda. München: Verlag der Bayerischen Akademie der Wissenschaften. 1992b Les Figurines de Suse. Les Figurines humaines IVe- IIe mill. av. J.-C., Mémoires de la Délégation archéologique en Iran 52. Paris: Gabalda. 1994. Un naos à divinité bovine. Pp. 269–272 in Beiträge zur Altorientalischen Archäologie und Altertumskunde, eds. P. Calmeyer, et al. Wiesbaden: Harrassowitz Verlag.

Old Babylonian Terracottas from Ur

151

Starr, R. F. S. 1937 Nuzi. Report on the Excavations at Yorgan Tepe near Kirkuk, Iraq. Conducted by Harvard University in Conjunction with the American Schools of Oriental Research and the University Museum of Philadelphia. 1927–1931. Volume II: Plates and Plans. Cambridge, MA: Harvard University Press. 1939 Nuzi. Report on the Excavations at Yorgan Tepe near Kirkuk, Iraq. Conducted by Harvard University in Conjunction with the American Schools of Oriental Research and the University Museum of Philadelphia. 1927–1931, Volume I: Text. Cambridge, MA: Harvard University Press. Straw, B. A. 2005 What Is Stronger Than a Lion? Leonine Image and Metaphor in the Hebrew Bible and the Ancient Near East. OBO 212. Fribourg: Academic Press Fribourg. Suter, C. E. 2000 Gudea’s Temple Building: The Representation of an Early Mesopotamian Ruler in Text and Image. CM 17. Groningen: Styx. 2007 Between Human and Divine: High Priestesses in Images from the Akkad to the Isin-Larsa Period. Pp. 315–359 in Ancient Near Eastern Art in Context: Studies in Honor of Irene J. Winter by her Students, eds. M. Feldman and J. Cheng. Culture and History of the Ancient Near East 26. Boston: Brill. 2011 Images, Tradition and Meaning: The Samaria and other Levantine Ivories of the Iron Age. Pp. 219–241 in A Common Cultural Heritage: Studies on Mesopotamia and the Biblical World in Honor of Barry L. Eichler, eds. G. Frame et al. Bethes­ da, MD: CDL Press. 2014. Human, Divine or Both? The Uruk Vase and the Problem of Ambiguity in Early Mesopotamian Visual Arts. Pp. 545–568 in Critical Approaches to Ancient Near Eastern Art, eds. M. Feldman and B. Brown. Berlin: Walter de Gruyter. Suter, C. E. and Uehlinger, C. 2006 Crafts and Images in Contact studies on Eastern Mediterranean Art of the First Millennium BCE, OBO 175. Fribourg/Göttingen: Vanderhoech/Ruprecht. Tomabechi, Y. 1984 Ancient Near Eastern Terracottas, Stone Artifacts, and Seals in the Babylonian Collection of the Lowie Museum of Anthropology. Bibliotheca Mesopotamica 15. Malibu: Undena Publications. Ur Online: http://www.ur-online.org/ Van Buren, E. 1930. Clay Figurines of Babylonia and Assyria. YOR 16. New Haven: Yale University Press. 1939 The Fauna of Ancient Mesopotamia as Represented in Art. AnOr 18. Rome: Ponti­ ficium Institutum Biblicum. 1945 Symbols of the Gods in Mesopotamian Art. AnOr 23. Rome: Pontificium Institu­ tum Biblicum. Watanabe, Ch. E. 2002 Animal Symbolism in Mesopotamia. A Contextual Approach. WOO 1. Wien: Insti­ tut für Orientalistik der Universität Wien. Winter, I. J. 1983 The Program of the Throneroom of Assurnasirpal II at Nimrud. Pp. 15–21 in Essays on Near Eastern Art and Archaeology in Honor of Charles Kyrle Wilkinson, eds. P. O. Harper and H. Pittman. New York: Metropolitan Museum of Art.

152

Laura Battini

1986 The King and the Cup: Iconography of the Royal Presentation Scene on Ur III Seals. Pp. 253–268 in Insight through Images: Studies in Honor of Edith Por­ ada, ed. M. Kelly-Buccellati. Bibliotheca Mesopotamica 21. Malibu: Undena Publications. 1994 Radience as an Aesthitic Value in the Art of Mesopotamia. Pp. 123–132 in Art: The Integral Vision. A Volume of Essay in Felicitation of Kapila Vatsyayan, eds. B. N. Saraswati, S. C. Malik, and M. Khanna. New Delhi: D. K. Printworld. 1995a Art and Visual Culture. Art Journal 54/3: 42–43. 1995b Aesthetics and Mesopotamian Art. Pp. 2569–2580 in Civilizations of the Ancient Near East, ed. J. Sasson. New York: Scribners. 1997 Art in Empire: The Royal Image and the Visual Dimensions of Assyrian Ideology. Pp. 359–382 in Assyria 1995, eds. S. Parpola and R. M. Whiting. Helsinki: The Neo-Assyrian Text Corpus Project. 1998 The Affective Properties of Styles: An Inquiry into Analytical Process and the Inscription of Meaning in Art History. Pp. 55–77 in Picturing Science, Producing Art, eds. C. A. Jones and P. Gallison. New York/London: Routledge. 1999 Reading Ritual in the Archaeological Record. Deposition Pattern and Function of the Two Artifact Types from the Royal Cemetery of Ur. Pp. 229–256 in Fluchtpunkt Uruk: Archäologische Einheit aus methodischer Vielfalt: Schriften für Hans Jörg Nissen, eds. H. Kühne, R. Bernbeck, and K. Bartl. Rahden/Westf.: M. Leidorf. 2012 Gold! Divine Light and Lustre in Ancient Mesopotamia. Pp. 153–172 in Proceedings of the 7th International Congress on the Archaeology of the Ancient Near East, London, 12–16 April 2010, eds. R. Matthews and J. Curtis. Wiesbaden: Harrassowitz. Woolley, L. 1939 The Ziggurat and Its Surroundings. UE 5. Oxford: Trustees of the British Mu­ seum and of the University Museum, University of Pennsylvania, Philadelphia; Oxford University Press. 1965 The Kassite Period and the Period of the Assyrian Kings. UE 8. Oxford: Trustees of the British Museum and of the University Museum, University of Pennsyl­ vania, Philadelphia; Oxford University Press. 1974 The Buildings of the Third Dynasty. UE 6. London: Trustees of the British Mu­ seum and of the University Museum, University of Pennsylvania, Philadelphia. 1976 The Old Babylonian Period. UE 7. London: Trustees of the British Museum and of the University Museum, University of Pennsylvania, Philadelphia; British Museum Publications. Wrede, N. 2003 Uruk. Terrakotten I. Von der ‘Ubaid- bis zur altbabylonischen Zeit. AUWE 25. Mainz am Rhein: Philip von Zabern.

The City of Ur and the Neo-Babylonian Empire Paul-Alain Beaulieu University of Toronto

Cuneiform sources from the time of the Neo-Babylonian empire are fairly abun­dant. Archival texts, found mostly in temples and private residences, form the largest corpus. Also, many royal inscriptions, commemorating chiefly the rebuilding of temples, have been recovered. 1 However, the city of Ur has yielded only limited amounts of evidence dated to that period. The main group of texts found in situ at Ur is the archive of Sîn-uballiṭ, son of Sîn-zēru-līšir. The other relevant private archive discovered at Ur is the Gallābu archive, which extends from the twenty-ninth year of Nebuchadnezzar II until the seventh year of Philip III Arrhidaeus, that is, from 576 to 317, but only two of its transactions are dated to the time of the Neo-Babylonian empire. 2 In addition, we find a limited number of references to the city in texts from Uruk, but not a single one in those from central and northern Babylonia. 3 This seems to suggest that Ur had become a purely regional center with no hinterland and limited influence, although one must refrain from hasty generalizations given the small amount of evidence we have. In addition, a few royal inscriptions of Nebuchadnezzar II and Nabonidus mention building works at Ur. The purpose of my contribution is to review what these texts tell us about Ur during the time of the Neo-Babylonian empire.

The Archive of Sîn-uballiṭ The archive of Sîn-uballiṭ was discovered during the excavations led by Sir Leonard Woolley at Ur. 4 Woolley excavated seven houses of Neo-Babylonian and Achaemenid date, all in the same area to the southeast of the temple precinct (area NH). The Gallābu archive was found in House 6 and the Sîn-uballiṭ archive in House 7. The two houses adjoin one another. House 7 is a large residence that 1.  Survey of archival sources by Jursa 2005, and of building inscriptions by Da Riva 2008. 2.  The two texts are published in UET 4 (= Figulla 1949) as nos. 7 (sale of a house) and 36 (lease of a house, dialogue document), dated respectively to the twenty-ninth year of Nebuchadnezzar II and to the reign of Nabonidus (the year number is damaged in the date formula but the house is leased until the month Simānu in the twelfth year of Nabonidus). 3.  References to Ur in Neo- and Late Babylonian texts are listed in Zadok 1985: 321–323. The vast majority of references to Ur are found either in texts from Ur published in UET 4, or in the Assyrian state correspondence of the seventh century, the latter predating the rise of the Neo-Babylonian empire. 4.  The archaeological context of the tablets and the description of the house can be found in Woolley 1962: 48. Pedersén 1998: 203 has numbered the archive as “Ur Archive no. 3” in his survey.

153

154

Paul-Alain Beaulieu

was found in poor condition and had undergone many alterations. The tablets were found in a jar in Room 15, a small closet area. According to Woolley’s report the jar was discovered lying under the foundations of the south-east wall, which belonged to the second phase of the building. The tablets in the jar were given the excavation number U.17238. Ten additional texts (U.17239), all letters, were found in the same room, at a higher level corresponding to the third and fourth courses of bricks in the wall from the second phase. In total the archive includes sixty-two texts published in UET 4 plus ten unpublished envelopes and fragments. 5 The archive of Sîn-uballiṭ has all the characteristics of a “dead archive,” a collection of documents that were stored when no longer relevant to the conduct of business. 6 This is further supported by the fact that the archive contains no transactions of long term relevance such as real estate, marriage, and inheritance. Typical of many dead archives, the chronological spread of the texts is short. All dated transactions fall between 624 and 616, from Nabopolassar year 2 to year 9. The archive is unusual in several respects. First, its contents are somewhat heterogeneous, and in some cases no connection to the presumed owner of the texts can be found; this could indicate that the archive deals in fact with the transactions of a group of individuals rather than one single person. Second, the main participant in the transactions, Sîn-uballiṭ, appears always as debtor rather than creditor. 7 And third, all transactions are dated in Babylon or other northern locations such as Dilbat, Borsippa and Šapiya. Yet the archive was found at Ur, and the majority of the participants in the transactions have personal names typical of that city, that is, theophoric names where the god Sîn, the patron deity of Ur, predominates. In sum, the texts record the undertakings of a group of citizens of Ur who had established residence in Babylon temporarily. After the cessation of their activities the main participant in the transactions, Sîn-uballiṭ, took the documents back to his house in Ur. 8 All this suggests that the circumstances surrounding Sîn-uballiṭ’s transactions were unusual and might be a reflection of the troubled political situation in Babylonia at that time. Indeed, the archive is coeval with the struggle for control of Babylonia between Nabopolassar and the Assyrian king Sîn-šarru-iškun. It is important to stress this fact since it has often been ignored in discussions of this archive. Therefore one must begin with the chronology of the period. 9 5.  A description of the Sîn-uballiṭ archive can be found in Jursa 2005: 135–137. The archive has been little studied with the exception of Cseke 2010 and 2014 and the remarks by Oelsner 2001. I am currently preparing an edition and reconstruction of this archive. I wish to thank Jonathan Taylor who sent me photographs of the Sîn-uballiṭ tablets preserved in the British Museum, and Grant Frame who facilitated my collations of those preserved in the Babylonian Section of the University Museum, Philadelphia. 6.  The characterization of the Sîn-uballiṭ archive as “dead” must be nuanced. The fact that the texts were stored in a jar and kept in Sîn-uballiṭ’s house at Ur shows that they still had some importance for their owner and could be retrieved if needed. Therefore the texts were kept for future reference, which is indeed the main characteristic of an archive, even if the transactions they recorded presumably no longer had any relevance for the economic life of the family. 7.  One promissory note on silver with interest (UET 4 69) mentions one Sîn-uballiṭ, son/descendant of Balāssu, as creditor, but this is certainly another Sîn-uballiṭ. 8.  There is every reason to believe that the house where the tablets were discovered belonged to Sînuballiṭ since some of the letters U.17239 are addressed from him to his wife Iṣṣūrtu and other members of his family. 9.  The chronology and sequence of events from the death of Ashurbanipal to the collapse of Assyria and the rise of Babylon to hegemonic power have been discussed in several publications, notably Oates

The City of Ur and the Neo-Babylonian Empire

155

Our main sources on the rise of the Neo-Babylonian state are two chronicles prob­ably written in Borsippa in the sixth century. 10 The first Chronicle is ABC 2. 11 It covers the accession year and the first three years of Nabopolassar, breaking off in the middle of the third year. This corresponds to the period from 626 to 623. The second Chronicle is ABC 3. 12 It details the assault on Assyria and the fall of Nineveh and Harran, covering years 10 to 17 of Nabopolassar, between 616 and 609. 13 The first Chronicle tells us that between 626 and 623 Assyrian and Babylonian forces were fighting to assert their control over Babylonia. When the second Chronicle resumes narration of events seven years later, in 616, we read that the Babylonians are on the offensive in the Suhum, beginning their push towards the heartland of Assyria that will fall to them and their Median allies a few years later. Unfortunately, the extant Chronicles do not cover the years 623 to 617. For that period the sequence of events must be reconstructed from archival sources. These sources indicate that Nabopolassar’s rule was effectively recognized throughout these years in northern Babylonia. The only exception is Sippar, where Sîn-šarru-iškun was recognized in a few documents dated from his accession year to his third year. In southern and central Babylonia, on the other hand, the situation was more confused. Quite early in the third year of his reign, Nabopolassar lost control of Uruk, which he regained only in the second half of his sixth year. Between 623 and 620 documents from Uruk acknowledge the rule of Sîn-šarru-iškun or are dated according to a system of year numbers accompanied by such notations as “during the closure of the city gate” or “during the hostilities in the land.” 14 It seems that pro- and anti-Assyrian factions competed for power at Uruk, but that the city was largely outside Nabopolassar’s control. The situation was similar at Nippur, where documents dated by Sîn-šarru-iškun range from his second to his sixth year, that is, from 626 to 621. 15 We do not have texts from Nippur dated after the sixth year of Sîn-šarru-iškun, and documentation from that site resumes only later, at the end of the reign of Nabopolassar, with a hiatus of about fifteen years. Thus we cannot establish when exactly the Babylonians recaptured the city. However, it seems probable that by 620 the Assyrians had been expelled from Babylonia completely, although they may have held out in Nippur and perhaps some other cities in the south until 617. The archive of Sîn-uballiṭ falls very neatly within this chronology. The earliest transaction is UET 4 56, dated at Babylon in Nabopolassar year 2, month 2, day 30. The latest one is UET 4 76, dated at Babylon in Nabopolassar year 9, month 11, day 5. This means, in Julian dates, from May 11, 624 until January 19, 616. Therefore the archive begins with the onset of the Assyrian counter-offensive which led to the recapture of Uruk, and it ends at the time of the Babylonian assault on Assyria, 1991, Na’aman 1991, Oelsner 1999, Fuchs 2014, Levavi 2017, and also Jursa 2007 for the origins of the Neo-Babylonian dynasty. 10.  On the context and time when these chronicles were composed see the recent study by Waerzeggers 2012. 11.  This Chronicle is edited by Grayson 2000: 87–90 (Chronicle 2) and by Glassner 2004: 214–218 (Chronicle 21). 12.  This Chronicle is edited by Grayson 2000: 90–96 (Chronicle 3) and by Glassner 2004: 218–224 (Chronicle 22). 13.  These crucial events have been discussed in many publications, notably by Joannès 2008. 14.  The evidence is detailed in Beaulieu 1997. More texts belonging to this troubled period have been published since, notably by Da Riva 2002 and 2003–4, and by Zawadzki 2013. 15.  These documents have been studied by Oppenheim 1955.

156

Paul-Alain Beaulieu

which according to the Chronicle ABC 3 started at the latest in the second month (Ayāru) of the tenth year of Nabopolassar, that is, in April-May 616 (although it may conceivably have begun earlier because of the gap in the Chronicles between 623 and 616). Thus, if Sîn-uballiṭ’s return to Ur occurred soon after the latest transaction in his archive, it also coincided almost exactly with the onset of the Babylonian offensive in the heartland of Assyria. In her chapter for the Cambridge Ancient History, Joan Oates had already remarked that the Sîn-uballiṭ documents were “presumably moved to Ur only after the cessation of hostilities in Babylonia, that is, sometime after 616” (Oates 1991: 177). We must now investigate whether the transactions of Sîn-uballiṭ explain his presence in Babylon and reflect the political situation of the time. Jursa (2005: 136) has described the main categories of transactions found in the archive: promissory notes on silver mostly with Sîn-uballiṭ as debtor, often with pledges of real estate or slaves; lists of soldiers; lists of aromatics and spices; itemized lists of silver expenditures; and letters. Few texts contain explicit information on the activities of Sîn-uballiṭ and his associates and the reason for their stay in Babylon. The most promising ones are those which suggest a connection of the archive with the military: UET 4 117: account of silver expenditures (1 KÙ.BABBAR te-lit-ti ŠID-nu “Silver reckoned as expenditures”); the list mentions the purchase of a bow, an išḫu, a bow-case, and a spear (7 1 ½ GÍN a-na GIŠ.BAN 8 3-ta 4-tú a-na kušiš-ḫi u kuššal-ṭu 9 4-tú a-na gišil-taḫ). 16 UET 4 118: account of silver expenditures (1 KÙ.BABBAR te-li-ti šá Id30-TIN-iṭ ana UGU Ie-ri-š[i] 2 ú-še-lu-ú MU.NI! “Itemized silver expenditures which Sîn-uballiṭ made on behalf of Erišu”); the list includes a širiam garment (4 3 GÍN 1 TÚG šir-’-aam šá qal-lat “3 shekels for a širiam for a slave girl”). UET 4 158: list of soldiers (6 [PAP 5] ERÍN.MEŠ ina ŠU.MIN Id30-TIN-iṭ “[Total: five] soldiers in the care of Sîn-uballiṭ”); Sîn-uballiṭ appears as one of the soldiers (4 Id30-TIN-iṭ A-šú šá Id30-NUMUN-GIŠ “Sîn-uballiṭ, son of Sîn-zēru-līšir”) and is also probably the one in charge of the contingent. 17 UET 4 159: list of soldiers (11 PAP 10 ERÍN.MEŠ ina ŠU.MIN Id30-NIR.GÁL-DINGIR.MEŠ “Total: ten soldiers in the care of Sîn-etel-ilī”); one of the soldiers is named Sîn-uballiṭ (6 Id30-TIN-iṭ A Id30-ABGAL-DINGIR “Sîn-uballiṭ, son/descendant of Sîn-apkal-ilī”), but this is another Sîn-uballiṭ. 18 16.  The signs copied by Figulla as 4 SÌLA on lines 4, 8, 15, and 18 of UET 4 117 must of course be read 4-tú, and this is confirmed by the photograph of the tablet, which shows a clear UD, rather than SÌLA, in all cases. 17.  Sîn-uballiṭ was the son of one Sîn-zēru-līšir, therefore his identity on line 4 of this text is almost beyond doubt. The Sîn-uballiṭ on line 6, in the summation, could also be the same individual, but the possibility of homonymy is fairly high given the frequency of the name pattern DN-uballiṭ. 18.  Since the scribes who wrote these transactions often used the sign A (= aplu) indiscriminately both for father and ancestor, one could argue that in this case our Sîn-uballiṭ appears with his ancestral name, which is otherwise unknown, but this is quite uncertain. The letter UET 4 186 is addressed by Sîn-uballiṭ to his “brother” Sîn-apkal-ilī, a further indication of the high incidence of homonymy in these texts.

The City of Ur and the Neo-Babylonian Empire

157

UET 4 160: list of soldiers (11 PAP 10 ERÍN.MEŠ ina ŠU.MIN Id30-MU-MU 12 A Id30bar-ḫi-DINGIR.MEŠ “Total: ten soldiers in the care of Sîn-šumu-iddin, descendant of Sîn-barḫi-ilī”). UET 4 182: letter, probably from Sîn-uballiṭ to his wife Iṣṣūrtu; the letter ends with a request for a širiam garment (18 1* TÚG* šir-a-am ul-tu ŠÀ-bi šu-li-i 19 ⸢di*⸣-ke-e u šup-riš “deduct and remove a širiam garment from it and send it”). 19 UET 4 188: letter (1 IM Id30-ABGAL-⸢DINGIR⸣.MEŠ 2 a-na Id30-NIR.GÁL-DINGIR EN-šú “Letter from Sîn-apkal-ilī to Sîn-etel-ilī, his lord”); one passage mentions the rab kissati, an official in charge of fodder (9 lúGAL kis-sat!-ti), and the palace (12 É. GAL). Erišu is mentioned in the letter (line 21). UET 4 192: letter from Sîn-uballiṭ to his mother; the letter mentions the confisca­ tion by the rab-ša-rēši of a house belonging to Sîn-uballiṭ. UET 4 198: seven men stand as guarantees for another man to the army commander (8 [pu-ut Iza]-kir A INÍG.DU A Iku-na-a 9 ⸢ina⸣ ŠU.MIN IŠEŠ-li-šìr lúGAL ki-ṣir 10 našu-ú ki-i iḫ-te-liq 11 ⸢1⸣ GUN KÙ.BABBAR i-tur-ru-nu “they stand as guarantee for [Zā]kir, son of Kudurru, descendant of Kūnāya, to Aḫu-līšir, the army commander. If he escapes they will pay one talent of silver”). Erišu is one of the guarantees. Jursa’s brief discussion of the Sîn-uballiṭ archive mentions the possibility of a background in military conscription and taxation, a suggestion initially made by Waerzeggers (Jursa 2005, 136). Indeed, a few texts refer to soldiers and military equipment. UET 4 117 and 118 are lists of amounts of silver expendited partly for military gear. 20 UET 4 158, 159, and 160 are lists of soldiers; text 158 in fact mentions Sîn-uballiṭ as a soldier and probably as the one responsible for the whole group. 21 Texts 159 and 160 each mention groups of ten soldiers in the care of associates of Sîn-uballiṭ; most of these individuals have personal names honoring the god Sîn, and therefore must originate from Ur. Another text, UET 4 198, records that several parties assume personal guarantee towards the rab-kiṣri for an individual named Kudurru, son of Kūnāya, to the amount of one talent of silver. Rab-kiṣri was a military title, often translated “military commander” or “commander of an army 19.  Figulla’s copy has a fairly clear ⸢40 TÚG⸣ šir-a-am on line 18, a reading which has been taken up in discussions of the tablet (e.g. Jursa 2005: 137 n. 1060; Quillien 2016: 704 n. 2216) and would imply the fabrication of textiles by the family of Sîn-uballiṭ on a significant scale, but collation indicates 1* TÚG* šir-a-am instead. 20.  The reference to a širiam garment in UET 4 118 presents difficulties. This garment, probably a tunic, is widely attested as part of the soldier’s equipment (Quillien 2016: 654–657), but the word can also refer to a generic piece of clothing (Quillien 2016: 558–562), which could be the case here if we interpret ša qallat(i) to mean “for a slave girl” as a reference to its user. Sîn-uballiṭ was involved in the fabrication and trade of textiles, including for soldiers (Quillien 2016: 704). 21.  The term translated as “soldier” (ERÍN = ṣābu) is inherently ambiguous since it can also mean “worker,” but given the context of the archive and of the particular documents where the term appears, its interpretation as “soldier” seems fairly secure. The mention of an ERÍN LUGAL in UET 4 183: 2, noted by Jursa (2005: 136 n. 1058), seems dubious (but not entirely excluded) in light of my collation of the tablet, which appears to have ṣab*-tak* instead (the second sign is clearly ŠUM, as copied by Figulla, not LUGAL).

158

Paul-Alain Beaulieu

unit.” 22 This data suggests that Sîn-uballiṭ and his associates were involved in the levying and provisioning of troops. This in turn might account for the high level of indebtedness reflected in the legal transactions concluded by Sîn-uballiṭ, who prob­ ably needed to raise a substantial amount of cash to fulfill his military obligations. What these obligations were, however, remains unclear. Some texts indicate that Sîn-uballiṭ owned real estate in Ur as well as female slaves. Therefore he must have belonged to the urban elite. In fact, Sîn-uballiṭ often raised silver by pledging his land and slaves to lenders. 23 Thus, one possibility is that Sîn-uballiṭ and his associates were simply fulfilling their military tax duties for the crown. At the same time the efforts involved appear to exceed normal taxation and may imply a deeper involvement in the war effort. Two letters, UET 4 188 and 192, hint at connections with the court and the state administration, as they mention the palace (É.GAL) and the rab-ša-rēši (LÚ.GAL.SAG) “chief of the eunuchs,” one of the high dignitaries of the palace. A certain Erišu, son of Sîn-apkal-ilī, appears in several transactions of the archive. 24 He seems to have been a close associate of Sîn-uballiṭ. 25 Why would citizens of Ur have an interest in supporting Nabopolassar? Some clues indicate that Nabopolassar started his bid for power in southern Babylonia. According to Chronicle ABC 2 his initial base was Uruk, months before he formally assumed the title of king in Babylon. Later tradition associates him with the Sealand province, which included probably Ur and perhaps Uruk as well. His con­ nection with the Sealand may explain why the governor of that province, (the šanû) Ea-dayān, is listed first among the provincial governors of Babylonia in the Court Document (Hofkalender) of Nebuchadnezzar II, dated to the year 597 (Beaulieu 2002: 114–115). These connections with the Sealand might account for the peculiar features of the archive of Sîn-uballiṭ. If Ur had fallen back into Assyrian hands, like Uruk and Nippur, in the second or third year of Nabopolassar, then Sîn-uballiṭ and his associates were probably refugees in Babylon responsible for outfitting troops made up mostly of their fellow citizens from Ur. If Ur was still in the hands of Nabopolassar throughout this period, then we can speculate that Sîn-uballiṭ and his associates were supporters of Nabopolassar who moved to Babylon to participate in the resistance to Assyria. The latter scenario seems more likely since the archive contains strong hints that connections between Sîn-uballiṭ and his family at Ur had not been cut off - Sîn-uballiṭ wrote to his wife and family during that period - but of course the situation may have been shifting all the time. 26 At any rate, communica22.  The translation proposed by CAD K: 442, s.v. rab kiṣri, is “commander of an army unit.” The title appears often in Neo-Assyrian sources and is translated “cohort commander” in Baker 2017: 99–109. Here too the Neo-Babylonian state appears to have borrowed an institution of the former Assyrian empire. 23.  UET 4 72 (pledge of field in Ur); UET 4 87 (pledge of orchard); UET 4 88 (pledge of field); UET 4 73–79 (pledges of female slave); UET 4 89 (pledge of female person of unspecified status); UET 4 197 (promise to redeem female slave held as pledge). The typology of these transactions is discussed by Oelsner 2001. 24.  UET 4 69: 2; 71: 2; 73: 10; 88: 3; 195: 10; 197: 2, 9; and 198: 6; and very probably in other texts where one Erišu is mentioned without his father’s name (UET 4 118: 1, 7, 12, 15; 176: 1; 184: 1; 188: 21). 25.  This Erišu borrows silver in UET 4 69, 71, and 88; he promises to redeem the pledged female slave of Sîn-uballiṭ in UET 4 197 (with interest penalty in case of failure); he appears as one of the guarantees to the rab kiṣri in UET 4 198; he is a witness in UET 4 73 (where Sîn-uballiṭ borrows silver) and 195 (quitclaim involving Sîn-uballiṭ). He also appears without filiation in UET 4 118, 176, 188, and is the recipient of the letter UET 4 184, where Sîn-uballiṭ addresses him as “his brother.” 26.  I am assuming that the letters U.17239 date from the same period as the transactions U.17238.

The City of Ur and the Neo-Babylonian Empire

159

tions between Babylon and Ur must have been problematic during the years when the Assyrians controlled both Nippur and Uruk. Whichever scenario we prefer, the archive of Sîn-uballiṭ constitutes, together with the “siege documents” from Uruk and Nippur, one of the most important archival witnesses to the struggle for control of Babylonia between the years 626 and 620/616.

Ur Under Nebuchadnezzar II We know almost nothing about the status of Ur in the empire built by Nabopolassar and Nebuchadnezzar II. When Babylonia became a province of the Assyrian empire in the seventh century, a dynasty of autonomous governors ruled at Ur and assumed quasi-regal status, claiming the title of “viceroy (šakkanakku) of Ur” and in one instance of “viceroy (šakkanakku) of Ur, Eridu, and the Gurasimmu.” One of them, Sîn-balāssu-iqbi, was very active in the restoration of the religious monuments of Ur and commissioned several building inscriptions to commemorate them; they are written in his name and mention Ashurbanipal on a couple of occasions, although they never acknowledge his brother Šamaš-šumu-ukīn as ruler of Babylonia. 27 Other officials are attested in documents from that period, such as “šaknu of Ur,” “šākin ṭēmi of Ur,” and “šangû of Ur.” 28 None of these titles except the last one are attested after the rise of the Neo-Babylonian empire. The city was probably governed in the sixth century by a šangû “high priest.” The title is attested in BIN 1 30 (obv. 9 lúÉ.BAR ŠEŠ.UNUGki), a letter from the Eanna archive addressed by people from Ur (one is called Sîn-mukkē-elip) to the šatammu (unnamed) of that temple. The letter concerns offerings of fish and mentions the fishermen of the Lady-of-Uruk, those of the god Sîn and also the offerings of the god Sîn. The time span of the Eanna archive is largely co-eval with the Neo-Babylonian empire and the beginning of Achaemenid rule (until ca. 520). The Hofkalender of Nebuchadnezzar lists a number of šangûs as governors of southern Babylonian cities, including Udannu, Larsa, and Kissik. Ur is not mentioned in the preserved entries but the first official listed in column VII is one Ii-rib-bi-dEN.ZU lúÉ.[BAR o o o] “Erīb-Sîn, 27.  Documents drafted at Ur, however, are dated according to the regnal years of Šamaš-šumu-ukīn; see BM 113927: 32 ina GUB-zu šá Id30-TIN-su-E GÌR.NÍTA ŠEŠ.UNUGki “in the official presence of Sînbalāssu-iqbi, viceroy of Ur” and 46 ŠEŠ.UNUGki ITI APIN UD 18-KAM 47 MU 10-KAM dGIŠ.NU11-MU-GI. NA LUGAL TIN.TIRki “Ur, month of Arahsamnu, 18th day, 10th year of Šamaš-šumu-ukīn, king of Babylon” (transliteration courtesy of Grant Frame). The building inscriptions of Sîn-balāssu-iqbi are edited by Frame 1995: 230–247; “viceroy of Ur” (šakkanakku Uri) is the habitual title of Sîn-balāssu-iqbi, the longer title with the addition of Eridu and the Gurasimmu appearing only in one inscription (no. 2015). The building works of Sîn-balāssu-iqbi are described together with those of the Kassite kings, mainly Kurigalzu I, in the final excavation report published by Woolley 1965; by Brinkman 1969: 336–342; and in the synthetic survey of the archaeology of Ur by Zettler and Hafford 2016. The history of Ur in the seventh century and the hereditary line of autonomous governors is studied by Brinkman 1965. 28.  The references are collected in Zadok 1985: 321–323. As noted by Brinkman 1965: 246 n. 3, it is probable that the titles šaknu and šakkanakku were considered more or less the same at Ur, since Sînbalāssu-iqbi retrospectively calls his father Ningal-iddin a šakkanakku in one of his inscriptions written in Sumerian (Frame 1995: 231 no. 2001: 7 I[d]nin-gal-SUM.MA 8 GÌR.NÍTA úri.KI-ma), while Ningal-iddin is called a šaknu in sources from his own lifetime (e.g. UET 4 9: 32 ina GUB-zu šá Idnin-gal-MU lúGARKUR ŠEŠ.UNUGki “in the official presence of Ningal-iddin, governor of Ur”; UET 4 27: 16 ŠEŠ.UNUGki ITI ŠE 17 UD 11-KAM MU 8-KAM Idnin-gal-MU 18 lúGAR-KUR ŠEŠ.UNUGki “Ur, month of Addaru, 11th day, 8th year of Ningal-iddin, governor of Ur”). The sequence GAR-KUR must probably be read šá-kìn rather than šākin māti.

160

Paul-Alain Beaulieu

the šan[gû of o o o]” (Da Riva 2013: 213). Thus we could plausibly restore Ur as the name of the missing city. It seems likely that Ur was included in the province of the Sealand (māt tâmti, or simply tâmti), in which case its administration and governor, the šangû, would have been partly or entirely subordinated to the šakin māt tâmti and other officials of that province (the šanû māt tâmti and the qīpu māt tâmti), but direct evidence is lacking. However, we have contemporary evidence from the Eanna archive suggesting that the administration of Uruk was answerable to the Sealand, at least for some matters. 29 In his excavations reports Woolley arributed a number of building works to Neb­ uchadnezzar: 30 1. the great wall of the temenos; 2. the rebuilding and re-modelling of Enunmaḫ; 3. the Nin-Ezen temple; 31 4. work done on Edublalmaḫ; 5. the reconstruction of the great Nannar court; and 6. work on the city wall. Stamped bricks in the name of Nebuchadnezzar have been found in these locations, attesting at least to the repair of existing structures. Nebuchadnezzar claims in his inscriptions that he restored Ekišnugal, the main sanctuary of the god Sîn. Unfortunately no extensive account of this major construction work has been discovered, and from Ur itself only inscribed bricks have been found that briefly mention it. 32 Nevertheless, analysis of the archaeological remains confirms the extent of Nebuchadnezzar’s building activities. Woolley identified two perimeter walls enclosing the Ekišnugal. He attributed the smaller one to the Ur III period and the second, larger one, to Neb­ uchadnezzar II, with six gates leading to the temple precinct. He also speculated that the reconstruction of the city along a more regular plan was due to Nebuchadnezzar, noting the similarities in urban planning with Babylon. With the exception of the city walls and the Nin-Ezen temple, the building activities of Nebuchadnezzar concentrated on the precinct of Ekišnugal, including the temenos wall and its gates, the courtyard, and the shrines Enunmaḫ and Edublalmaḫ. He considerably altered the ground plan of Enunmaḫ, leaving the shrine as it had been restored during the Kassite and Isin II period barely recognizable. This could signify a change in the function of the building or in temple rituals. 33

Ur Under Nabonidus With Nabonidus, a devotee of the moon god, Ur became even more the focus of royal attention. In the second year of his reign he revived the institution of 29.  The evidence is collected by Beaulieu 2002: 112–114; on the extent of the Sealand province see also Kleber 2008: 311–331, who discusses all the evidence in detail. 30.  These are listed by Woolley 1962: 1, and discussed in detail throughout the book. Zettler and Hafford 2016 provide summary descriptions of these building works. 31.  The so-called Nin-Ezen temple lay in area NT, in the southwestern area of the city, alongside the city wall. The Neo-Babylonian repair of the temple is described by Woolley 1962: 34–35. Earlier it was known as Eniginna and was rebuilt in the Isin-Larsa period by Sîn-iqīšam and Rīm-Sîn I (George 1993: no. 877). The temple was dedicated to the god Ningizzida and his consort the healing goddess Nin-Ezen (Nin-Isinna). The history of the shrine is summarized on the Ur Online Project: http://www. ur-online.org/location/23/ (accessed September 26, 2017). 32.  The brick inscription of Nebuchadnezzar from Ur is listed by Da Riva 2008: 117: B 10 (with references to earlier treatments). The restoration of Ekišnugal is also briefly mentioned in the Hofkalender, for which see Da Riva 2013: 210, 1. 24′–27′, as well as other inscriptions of Nebuchadnezzar. 33.  Woolley 1962: 23–33. The archaeological history of the building is summarized by Hafford and Zettler 2016: 372–373; and in the Ur Online Project: http://www.ur-online.org/location/4123/ (accessed September 26, 2017).

The City of Ur and the Neo-Babylonian Empire

161

ēntu-priestess (EREŠ.DINGIR) 34 of the god Sîn of Ur and elevated his daughter En-nigaldi-Nanna to that rank, usually translated as “high priestess.” The event is recorded in three sources: the En-nigaldi-Nanna cylinder, the Eigikalamma cylinder, and the so-called Royal Chronicle, a fragmentary historical literary text which is heavily dependent on contemporary royal inscriptions. 35 The En-nigaldi-Nanna cylinder relates in some detail the circumstances of her elevation to that office, not­ ably the lunar eclipse which indicated that the moon-god required a high priestess. We also hear about the excavations the king undertook in the giparu (Egipar), the residence of the former high priestesses at Ur. There he allegedly discovered a stele of Nebuchadnezzar (I), son of Ninurta-nādin-šumi, who reigned more than 500 years before him. An image of a high priestess with her insignias was apparently depicted on the stela. We do not know the actual name of the daughter of Nabonidus since the cylinder tells us that the king changed it to En-nigaldi-Nanna, which means in Sumerian “the high priestess requested by the god Nanna,” a learned name modelled on those of ēntu-priestesses of the third and second millennia and inspired by the omen from the series Enuma Anu Enlil which provided the basis for her elevation to that status: “If in the Month Ulūlu an eclipse (of the moon) occurs during the morning watch, (this means that) Sîn requests a high priestess.” 36 The En-nigaldi-Nanna episode raises some questions as to the role played by royal daughters in that period, especially in southern Babylonia, and also about the actual importance of Ur in the political and economic structure of the Babylonian empire. There is a priori no reason to doubt that Nabonidus is telling the truth when he claims that the institution of high priestess of Ur had long fallen into oblivion. It was apparently still alive under the Second Dynasty of Isin. Nabonidus does claim that he discovered a stele of Nebuchadnezzar (I) with a depiction of the high priestess, perhaps a daughter of that king. Evidence contemporary with the late Kassite and Isin II periods corroborates this. Adad-apla-iddina adopts the epithet of “father-in-law of the god Nanna” in two of his inscriptions from Ur, a claim which seems to suggest that he betrothed his daughter to the god and can be explained only if his daughter became ēntu, and literally, spouse of Nanna. 37 In addition, four Middle Babylonian legal and administrative texts from the Ur excavations published in UET 7 mention the EREŠ.DINGIR and the É EREŠ.DINGIR; two are undated or with dates lost, but the other two are dated to the fifth year of Kadašman-Turgu (UET 7 79; ca. 1277 BCE) and to the reign of Meli-Šipak, perhaps his fourteenth year (UET 7 70; ca. 1172 BCE). 38 Thus we must assume that the 34. The logogram is usually transliterated NIN.DINGIR, but EREŠ.DINGIR is more accurate historically and is adopted by Sallaberger and Huber Vuillet 2003–2005: 627–628. For the reading see the references collected by Steinkeller 1999: 120 n. 53. 35.  These sources are all edited by Schaudig 2001: 366–367, ii 8–14 (Eigikalamma cylinder); 373– 377 (En-nigaldi-Nanna cylinder); and 591–592, ii–iii (Royal Chronicle). The En-nigaldi-Nanna cylinder is translated and discussed in details by Reiner 1985: 1–16. The installation of En-nigaldi-Nanna is also discussed by Cousin 2016: 393–395. 36.  Schaudig 2001: 374 n. 428, with references to previous discussions. 37.  In his edition of the two inscriptions, Frame 1995: 60–61 (nos. 10 and 11) translates emu and its Sumerian equivalent mu10-ús-sá as “son-in-law.” However, the term is ambiguous and can also mean “father-in-law” (CAD E: q.v.). The transformation throughout the third millennium of the ereš-dingir, a term which originally meant something like “divine votary,” into the priestly wife of the god (by analogy with the male office of high priest, the en), is charted in detail by Steinkeller 1999. 38.  The texts are edited with extensive commentaries by Gurney 1983.

162

Paul-Alain Beaulieu

institution of high priestess of Ur disappeared during the era of political confusion and economic decline which accompanied and followed the end of Isin II. There is almost no archival documentation from Babylonia until the second half of the eighth century. It seems of interest, however, that an unpublished document dated to the thirteenth year of Nabû-šuma-iškun (ca. 748 BCE) mentions the “landed estates of the king’s daughter which are in Ur” (qaq-qarmeš šá DUMU.MÍ LUGAL šá ina URU ú-ri-⸢i⸣). 39 Although the document does not mention the institution of high priestess, it indicates that royal daughters still retained some form of economic and perhaps institutional connection with the city. Similar connections for Uruk are well documented later during the reign of Nebuchadnezzar II. His two daughters, Ba’u-asītu and especially Kaššāya, the latter very probably to be identified as the wife of the future king Neriglissar, owned estates in the vicinity of Uruk and because of this were involved in the affairs of the Eanna temple, although the nature and extent of their connection remains unknown (Beaulieu 1998). Thus far no evidence has surfaced that they played a cultic role in the temple. One text dated to the end of the first month in the first year of Nabonidus mentions the involvement of a royal daughter in the affairs of the Eanna temple. The text in question, YOS 6 10, is dated at Larsa, where Nabonidus appears to have granted audiences during his travels to southern Babylonia. Nabonidus mentions in a stele found at Babylon that he visited Uruk, Larsa, and Keš at the beginning of his reign, and the text no doubt was written on that occasion. YOS 6 10 records a number of decisions made or approved by the king at Larsa concerning the administration of the Eanna temple. The last royal order in YOS 6 10 reads as follows: 22 ŠUK.ḪÁ DUMU.MÍ LUGAL a-na qu-up-pi šá LUGAL ú-su-uk 23 UD.UNUGki ITI BÁRA UD 28-KAM MU 1-KAM dNÀ-I LUGAL TIN.TIRki “deposit the food allotments of the king’s daughter in the royal cash-box; Larsa, Month of Nisannu, 28th day, first year of Nabonidus, king of Babylon.” Unfortunately the document does not provide any information on the context of this order. The king’s daughter mentioned in the text can only be a daughter of Nabonidus; a reference to the daughters of Nebuchadnezzar II previously active at Uruk seems excluded for semantic and political reasons. 40 However, it seems probable that the unnamed daughter of Nabonidus mentioned here stepped in the role formerly played by the daughters of Nebuchadnezzar in Eanna; the estates of the previous royal family in Uruk may have been transfered to her. Since the text is dated only one and a half years before the elevation of En-nigaldi-Nanna to high priestess, it is possible that the reference is to her, which would imply that Nabonidus temporarily installed her at Uruk, but another royal daughter could be involved as well. However, that YOS 6 10 refers to the future En-nigaldi-Nanna seems plausible in light of AUAM 73.2942, also drafted at Larsa and dated to the seventh year of Nabonidus. The text mentions the house(hold) of the ēntu-priestess, and this is in all probability a reference to En-nigaldi-Nanna since the institution of ēntu-priestess had been revived specifically for her and that no 39. Brinkman 2014–2016: 366b, without reference to the museum or publication number of the text. However, this is certainly BM 108527, listed by Brinkman and Kennedy 1983: 63, AE.3 II-25, year 13, BM 108527 (1913-7-12). The text is also mentioned by Zadok 2000: 119–120, who reads URU ú-rik[i]. 40.  Popova 2015: 407 proposes a slightly different scenario; according to her, Nabonidus installed his daughter at Ur because this was the only vacant post of high priestess in the south, the ones at Uruk still being occupied by the daughters of Nebuchadnezzar, whom he could not displace because he positioned himself as his successor.

The City of Ur and the Neo-Babylonian Empire

163

other evidence for the existence of that function can be found in the Neo-Babylonian documentation. The text is preserved in the Horn Museum at Andrews University in Berrien Springs, Michigan, and was previously unpublished. It belongs to the archive of the Šamaš-bāri family, associated by marriage with the Larsa businessman Itti-Šamaš-balāṭu and his son Arad-Šamaš. 41 The text reads as follows: AUAM 73.2942 obverse 1 15-KAM u4-mumeš lú⸢ú⸣-[ra-šú] 2 šá É míen-ti IdUTU-DU-A 3 A-šú šá Ie-tel-lu ina ŠU.MIN IdUTU-NUMUN-BE 4 A-šú šá Ie-tel-lu ma-ḫe-er 5 lú ⸢mu⸣-kin-nu IdUTU-ŠEŠ.⸢MEŠ-SU⸣ 6 ⸢A-šú šá IdNÀ-ú⸣-ter-ri reverse 7 I mar-duk A-šú šá IdNÀ-ú-šib-ši 8 u lúUMBISAG INÍG.DU-nu A-šú šá 9 A-šú šá (ditto) IdNÀ-NUMUN-MU UD.UNUGki 10 ITI ŠU UD 24-KAM MU 7-KAM 11 d NÀ-NÍ.TUK LUGAL TIN.TIRki 12 ul-tu UD 24-KAM šá ITI ŠU upper edge 13 Id UTU-NUMUN-GÁL-ši pi-ir-ri 14 šá ⸢lú⸣[ú]-ra-šú šá É míen-ti 15 ul-te-er-ra-⸢a’⸣ “Šamaš-mukīn-apli, son of Etellu, has received from Šamaš-zēru-ušabši, son of Etellu, (the responsibility to lead for) fifteen days the cor[vée workers] of the household of the ēntu-priestess. 42 Witnesses: Šamaš-aḫḫē-erība, son of Nabû-uterri; Marduk, son of Nabû-ušebši; and the scribe is Kudurrānu, son of Nabû-zēru-iddin. Larsa, month of Dûzu, 24th day, 7th year of Nabonidus, king of Babylon. From the 24th day of the month of Dûzu, Šamaš-zēru-ušabši has begun (leading) the team of corvée workers of the household of the ēntu-priestess.” Line 2: the spelling en-ti, without female determinative, occurs in the En-nigaldi-­ Nanna cylinder: i 26 áš-šú iš-tu UD.MEŠ ru-qu-tì pa-ra-aṣ en-ti ma-šu-ú-ma “because since distant days the ritual of the high priestess had been forgotten” (Schaudig 2001: 375). Other instances of the same spelling, with or without determinative, are listed in CAD E: 172–173, s.v. ēntu. Line 15: this is probably the verb šurrû A “to begin” (CAD Š/3, pp. 358–360), attest­ ed in Neo- and Late Babylonian.

41.  The combined archive totals more than 200 documents (Jursa 2005: 108–109). A full edition and study is currently being prepared by Odette Boivin. 42.  The “household of the ēntu-priestess” (É EREŠ.DINGIR) is mentioned in UET 7 70: rev. 6′ and 8′, dated to the reign of the late Kassite king Meli-Šipak.

164

Paul-Alain Beaulieu

The term translated as “corvée worker” is urāšu (urāsu in Assyrian), which also means “corvée obligation,” one of the several Neo-Babylonian words referring to taxes in cash or in the form of labor service. There are two ways to understand the expression urāšu ša bīt ēnti. One possibility is to translate it “the corvée workers (or corvée obligation) attached to the household of the ēntu-priestess.” This interpretation is strongly supported by parallel expressions such as urāšu ša Šamaš “corvée workers of the god Šamaš,” urāšu ša Larsa “corvée workers of Larsa,” and ḫadru ša urāšu ša bīt mašenni “ḫadru of the corvée workers of the household of the mašennu official,” which all certainly refer to the institution to which the corvée duty is attached. The other possibility is that the expression refers to punctual corvée labor for building a residence for the ēntu-priestess: “the corvée workers assigned to (building) the residence of the ēntu-priestess.” Parallel examples are urāšu ša gišri “corvée workers for the bridge” and urāšu ša dullu epiri “corvée workers for the earthwork,” which refer to the specific task to be accomplished by the conscripted workers. 43 The first interpretation seems more likely for AUAM 73.2942, implying that urāšu taxes were levied to maintain the household of the ēntu-priestess, in all probability the daughter of Nabonidus installed at Ur. 44 The wording of the document is not very explicit about the background of the transaction. The urāšu tax was paid by performing various types of corvée labor, but it could also be acquitted with a silver payment in lieu of service. Tax farming was common in that period. A number of transactions deal with collection of the urāšu tax. 45 They indicate that its collection could be leased to entrepreneurs or entrusted to subordinates, and this may provide the background of the transaction recorded in our text, where the responsibility (or the right) to lead the corvée workers is transfered to another party for a limited amount of time without the cause for the transaction being stated. 46 The interesting fact, of course, is that the text is not from Ur but from Larsa. Therefore the combined evidence from YOS 6 10 and AUAM 73.2942 suggests that the household of En-nigaldi-Nanna in Ur was supported with resources from Uruk and Larsa, or that she established her temporary living quarters in that area before moving to Ur. Woolley (1962: xi) noted in his final excavation report that “Nebuchadnezzar practically rebuilt the city. His successor Nabonidus completed all that Nebuchadnezzar had left unfinished and in some cases at least re-organised Nebuchadnezzar’s work on more orthodox lines, and in addition set up entirely new buildings of his own.” He attributed the following repair and construction work to Nabonidus: 1. the rebuilding of the ziggurat Elugalmalgasisa; 2. the restoration of the temple of Ningal rebuilt in the previous century on the south-east side of the ziggurat by Sîn-balāssu-iqbi; 3. the restoration of Edublalmaḫ and Egipar (giparu), the latter 43.  The examples which precede are gleaned from CAD U: 209, s.v. urāsu. I left the term ḫadru untranslated; basically it refers to a taxation unit in the widespread system of leases of land for service to the state. 44.  If we adopt the second interpretation, we could tentatively identify the “residence of the ēntupriestess” as the dwellings built by Nabonidus for his daughter at Ur in the Ekišnugal precinct (see below). 45. The urāšu institution is discussed by van Driel 2002: 264–268, with many textual references, and more briefly by Jursa 2011: 442 and 2011–2013: 173; collection of the urāšu tax is discussed in various places by Jursa 2010: 647–668. 46.  Since both parties in the transaction are sons of one Etellu, it is possible that they were in fact brothers.

The City of Ur and the Neo-Babylonian Empire

165

being the residence of his daughter as high priestess; 4. the construction of a new palace, also for his daughter, in the northeast area of the walled city; and 5. the restoration of Enunmaḫ which had been initiated earlier by Nebuchadnezzar. These projects are documented not only by archaeological evidence, with stamped bricks of Nabonidus found in situ, but also by building inscriptions which name the buildings and give some general account of their construction. The En-nigaldi-Nanna cylinder provides some information on the excavation and rebuilding of Egipar; it is also commemorated in two other short inscriptions, one on bricks and one on a doorpost. 47 The En-nigaldi-Nanna cylinder claims that Nabonidus built the residence for his daughter alongside the restored giparu: ii 5 É.GE6.PÀR ki-ma la-bi-ri-im-ma e-eš-ši-iš e-pú-uš 6 BÁRA.MEŠ-šu ù GIŠ.ḪUR. MEŠ-šu ki-ma la-bi-ri-im-ma e-eš-ši-iš ab-ni 7 a-na i-te-e É.GE6.PÀR É EN-NÍG.AL. DI-dŠEŠ.KI DUMU.MÍ-ia 8 EREŠ.DINGIR.RA d30 e-eš-ši-iš e-pú-uš “I built anew the giparu as in olden days; I restored its sacred location and ground plan as in former times; alongside the giparu I built anew a house for my daughter En-nigaldi-Nanna, the high priestess” (Schaudig 2001: 374). This is corroborated by the excavated remains. In her study of the archaeological history of the giparu, Weadock (1975: 112–114) has noted that the remains of the Neo-Babylonian rebuilding of the old giparu by Nabonidus are scant, but that the residence he built for his daughter can be identified with a complex of apparently domestic character north-east of (and therefore alongside) the old giparu and forming the southeast extension of the Edublalmaḫ shrine. Inscribed bricks of Nabonidus were found there recording the building of the giparu, making the identification of this structure as the residence he built for his daughter quite secure. Another brick inscription of Nabonidus commemorates the restoration of Enunmaḫ, which is specifically named as the bīt ḫilṣi of the goddess Ningal (or Nikkal), the consort of the moon god Sîn, in the Ekišnugal. 48 In the last years of his reign Nabonidus also completed the restoration of Elugalmalgasisa, the ziggurat of Ur. This is commemorated by two inscriptions, one on bricks and one on cylinders. 49 The cylinder makes no reference to En-nigaldi-Nanna but ends with a prayer for Bēl-šarru-uṣur (Belshazzar), his eldest and perhaps only son. The inscription also refers to the finding of foundation deposits of Ur-Namma, the original builder of the ziggurat, and of his successor Šulgi. The restoration of Elugalmalgasisa was a defining moment in the theological program introduced by Nabonidus after his return to Babylon in the thirteenth year of his reign. The cylinder from Ur praises the moon-god Sîn with epithets which proclaim his status as “king” and “lord” of all the gods, and even exalt him as “god of gods.” 50 Nabonidus also built a large residence in the northeastern part of the site. The residence generally adheres to 47.  Edition and discussion by Schaudig 2001: 341–342 (1.8. Ziegel), and 344 (1.12. Türangelstein). 48. Edition and discussion by Schaudig 2001: 339–340 (1.6. Ziegel). As already mentioned Neb­ uchadnezzar had previously restored the Enunmaḫ and altered its plan, but no building inscription specifically mentioning this work has been found. 49. Edition and discussion by Schaudig 2001: 340–342 (1.7. Ziegel), and 350–353 (Elugalmalgasisa Zylinder); translation of the cylinder by Beaulieu 2000: 313–314. Several copies of the cylinder were discovered at Ur in the remains of the ziggurat and are preserved in the British Museum and the Iraq Museum. An additional exemplar has been published by George 2011: 185–186. 50.  Schaudig 2001: 480–481 ascribes to Nabonidus a cylinder fragment found at Ur and now in the British Museum (BM 120526). The fragment has been ascribed to Cyrus in the past, but given the formulaic nature of the few lines that are preserved, no decisive atttibution can be made.

166

Paul-Alain Beaulieu

the ground plan of large private dwellings of the period and thus resembles the “Great House” in the Merkes at Babylon or a much larger version of the houses excavated at Ur in area NH. Its ground plan also bears a close resemblance to the South Palace in Babylon. Unfortunately no inscriptional evidence informs us on the function of this building. Originally Woolley speculated that Nabonidus had built it as living quarters for his daughter, the giparu being in his view too small for that purpose. Hence he dubbed it the “palace of Bel-shalti-Nannar” after the accepted reading of the name En-nigaldi-Nanna at that time. 51 It seems more likely that this was the palace of the local governor. 52

Conclusions In the seventh century, when the Assyrian empire reached its apex under the Sargonids, the city of Ur played some significant political role at the regional level and even experienced a revival of its architectural patrimony during the tenure of Sîn-balāssu-iqbi as governor. After Šamaš-šumu-ukīn rebelled against his brother Ashurbanipal the city appears to have sided with the Assyrians and acted as regional watchdog for their interests, even when the countryside and other cities of the Sealand defected to the insurrectionists. With the rise of Nabopolassar one generation later, the situation at Ur seems less clear, mainly because we do not have any source equivalent to the Assyrian state correspondence. By then the hereditary line of local governors seems to have disappeared. We do not know the situation at Ur during the war between Sîn-šar-iškun and Nabopolassar for control of Babylonia, but we can surmise that if Nippur and Uruk stayed under Assyrian control for a few years, Ur may have known a similar fate. This may account for the peculiar circumstances of the Sîn-uballiṭ archive, which can be viewed in this context as a testimony to the efforts of citizens of Ur, exiled (forcibly or voluntarily) to the capital, in the levying of troops to support Nabopolassar. After the end of the civil war in Babylonia they took the documents attesting to their war efforts back to Ur and stored them, perhaps as proof of their loyalty to the new dynasty during the years of conflict. After the collapse of Assyria and the consolidation of Babylon’s power, Neb­ uchadnezzar launched a vast building program which included Ur, especially the renovation of Ekišnugal. This was continued by Nabonidus, whose involvement with Ur was motivated primarily by personal inclinations. The two rulers restored the city to some of its erstwhile magnificence, completing the work initiated by Sînbalāssu-iqbi in the previous century. While the Neo-Babylonian houses uncovered by Woolley in area NH reflect the continued existence of the city as a living community, and in spite of the efforts deployed by Nebuchadnezzar and Nabonidus, Ur may not have been a city of major economic and political importance. It can be argued indeed 51.  The form Bel-shalti-Nannar is an erroneous Akkadian reading of the name En-nigaldi-Nanna(r), which in fact should be translated into Akkadian as Entu-erišti-Sîn (or -Nanna) “the high priestess requested by the god Sîn (Nanna in Sumerian).” 52.  Weadock 1975: 113–114 notes that bricks of Nabonidus mentioning the building of the giparu were found there, but that it was used most likely as the house of the governor or a stopping place for the king. This palace is in fact quite large. Baker 2014: Table 2.3 compares the size of Neo-Babylonian private and official dwellings: the large house no. III in the Merkes at Babylon has three courtyards and a surface of approximately 1,475 m2; the “palace of Bel-shalti-Nannar” (identified as the “local governor’s palace”) has ca. 5,743 m2; and the South Palace in Babylon ca. 43,840 m2. If the building indeed housed the local governor, it may have been the šangû of Ur.

The City of Ur and the Neo-Babylonian Empire

167

that it had become a backwater of little more than religious significance. Hopefully future excavations at Ur will uncover more material that will shed light on its role in the political, economic and cultural fabric of the Neo-Babylonian empire.

Bibliography Baker, H. D. 2014 House Size and Household Structure: Quantitative Data in the Study of Babylonian Urban Living Conditions. Pp. 7–23 in Documentary Sources in Ancient Near Eastern and Greco-Roman Economic History, eds. H. D. Baker and M. Jursa. Oxford and Philadelphia: Oxbow Books. 2017 Neo-Assyrian Specialists. Crafts, Offices, and other Professional Designations. The Prosopography of the Neo-Assyrian Empire 4/1. Helsinki. The Neo-Assyrian Text Corpus Project. Beaulieu, P.-A. 1997 The Fourth Year of Hostilities in the Land. BaM 28: 367–394. 1998 Ba’u-asītu and Kaššaya, Daughters of Nebuchadnezzar II. Or. NS 64: 173–201. 2000 Neo-Babylonian Royal Inscriptions. Pp. 306–314 in The Context of Scripture, vol. II. Monumental Inscriptions from the Biblical World, eds. W. W. Hallo and K. L. Younger. Leiden - Boston – Köln: Brill. 2002 Ea-dayān, Governor of the Sealand, and Other Dignitaries of the Neo-Babylonian Empire. JCS 54: 99–123. Brinkman, J. A. 1965 Ur: 721–605 B.C. Or. NS 34: 241–258. 1969 Ur: The Kassite Period and the Period of the Assyrian Kings. Or. NS 38: 310–348. 2014–6 Ur. A. III. Philologisch. Mitte 2.-1. Jahrtausend. Pp. 364–367 in Reallexikon der Assyriologie und vorderasiatischen Archäologie 14. Berlin: Walter de Gruyter. Brinkman, J. A. and Kennedy, D. A. 1983 Documentary Evidence for the Economic Base of Early Neo-Babylonian Society: A Survey of Dated Babylonian Economic Texts, 721–626 B.C. JCS 35: 1–90. Cousin, L. 2016 Babylone, ville du roi au premier millénaire av. J.-C. Thèse de doctorat, Université de Paris 1 Panthéon-Sorbonne. Cseke, H. 2010 Did Sîn-uballiṭ Return His Loans or Not? Methods of Archiving and Pledging Slave Women by Reason of a Neo-Babylonian Archive. RA 104: 163–168. 2014 The Economic Determination of the Changing Interests. A Survey Based on the Loan Documents of the Neo-Babylonian Sîn-uballiṭ Archive. Pp. 557–578 in Studies in Economic and Social History of the Ancient Near East in Memory of Péter Vargyas, ed. Z. Csabai. Budapest: L’Harmattan; Pecs: Department of Ancient History, The University of Pecs. Da Riva, R. 2002 BM 114354: un nuevo texto edēl bābi redactado en Uruk durante la guerra entre Nabopolassar y Sin-shar-ishkun. Aula Orientalis 20: 249–251. 2003/4 Pfründe in Eanna in der Zeit der Unruhe. AfO 50: 245–254. 2008 The Neo-Babylonian Royal Inscriptions. GMTR 4. Münster: Ugarit-Verlag. 2013 Nebuchadnezzar II’s Prism (EŞ 7834): A New Edition. ZA 103: 196–229. Driel, G. van. 2002 Elusive Silver. In Search of a Role for a Market in an Agrarian Environment. Aspects of Mesopotamia’s Society. PIHANS 95. Leiden.

168

Paul-Alain Beaulieu

Figulla, H. H. 1949 Business Documents of the New-Babylonian Period. UET 4. London: Trustees of the British Museum and of the University Museum, University of Pennsylvania, Philadelphia. Frame, G. 1995 Rulers of Babylonia, From the Second Dynasty of Isin to the End of Assyrian Domination (1157–612 BC). The Royal Inscriptions of Mesopotamia, Babylonian Periods 2. Toronto: University of Toronto Press. Fuchs, A. 2014 Die unglaubliche Geburt des neubabylonischen Reiches oder: Die Vernichtung einer Weltmacht durch den Sohn eines Niemand. Pp. 25–71 in Babylonien und seine Nachbarn in neu- und spätbabylonischer Zeit. Wissenschaftliches Kolloquium aus Anlass des 75. Geburtstags von Joachim Oelsner, Jena, 2. und 3. März 2007, AOAT 369, eds. M. Krebernik and H. Neumann. Münster: Ugarit-Verlag. George, A. R. 1993 House Most High. The Temples of Ancient Mesopotamia. CM 5. Winona Lake, IN: Eisenbrauns. 2011 Cuneiform Royal Inscriptions and Related Texts in the Schøyen Collection. CUSAS 17. Bethesda, Maryland: CDL Press. Glassner, J.-J. 2004 Mesopotamian Chronicles. Writings from the Ancient World 19. Atlanta: Society of Biblical Literature. Grayson, A. K. 2000 Assyrian and Babylonian Chronicles. TCS 5. Winona Lake, IN: Eisenbrauns. Gurney, O. R. 1983 The Middle Babylonian Legal and Economic Texts from Ur. British School of Archaeology in Iraq. Joannès, F. 2008 La stratégie des rois néo-babyloniens contre l’Assyrie. Pp. 207–218 in Les armées du Proche-Orient ancien (IIIe–Ier mill. av. J.-C.), eds. Ph. Abrahami and L. Battini. BAR International Series 1855. Oxford. Jursa, M. 2005 Neo-Babylonian Legal and Administrative Documents. GMTR 1. Münster: Ugarit-Verlag. 2007 Die Söhne Kudurrus und die Herkunft der neubabylonischen Dynastie. RA 101: 125–136. 2010 Aspects of the Economic History of Babylonia in the First Millennium BC. AOAT 377. Münster: Ugarit-Verlag. 2011 Taxation and Service Obligations in Babylonia from Nebuchadnezzar to Darius and the Evidence for Darius’ Tax Reform. Pp. 431–448 in Herodot und das Persische Weltreich - Herodotus and the Persian Empire, eds. R. Rollinger, B. Truschnegg, and R. Bichler. Classica et Orientalia 3. Wiesbaden: Harrassowitz. 2011–3 Steuer. D. Spätbabylonisch. Pp. 168–175 in Reallexikon der Assyriologie und vorderasiatischen Archäologie 13. Berlin: Walter de Gruyter. Kleber, K. 2008 Tempel und Palast. Die Beziehungen zwischen dem König und dem Eanna-Tempel im spätbabylonischen Uruk. AOAT 258. Münster: Ugarit-Verlag. Levavi, Y. 2017 Loyalty in the Early Years of the Neo-Babylonian Empire. Pp. 177–190 in Fortune and Misfortune in the Ancient Near East. Proceedings of the 60th Rencontre Assyriologique Internationale, Warsaw, 21–25 July 2014, eds. O. Drewnowska and M. Sandowicz. Winona Lake, IN: Eisenbrauns.

The City of Ur and the Neo-Babylonian Empire

169

Na’aman, N. 1991 Chronology and History in the Late Assyrian Empire (631–619 B.C.). ZA 81: 243–267. Oates, J. 1991 The Fall of Assyria (635–609 B.C.). Pp. 162–193 in The Cambridge Ancient History 3/2, 2nd Edition, eds. J. Boardman, et al. London: Cambridge University Press. Oelsner, J. 1999 Vom Assurbanipal zu Nabopolassar - das Zeugnis der Urkunden. Pp. 643–666 in Munuscula Mesopotamica. Festschrift für Johannes Renger, eds. B. Böck, et al. AOAT 267. Münster: Ugarit-Verlag. 2001 The Neo-Babylonian Period. Pp. 289–305 in Security for Debt in Ancient Near Eastern Law, eds. R. Westbrook and R. Jasnow. Culture and History of the Ancient Near East 9. Brill Academic Publishers. Oppenheim, A. L. 1955 ‘Siege Documents’ from Nippur. Iraq 17: 69–89. Pedersén, O. 1998 Archives and Libraries in the Ancient Near East 1500–500 BC. Bethesda, MD: CDL Press. Popova, O. V. 2015 The Royal Family and its Territorial Implantation during the Neo-Babylonian Period. Kaskal 12: 401–410. Quillien, L. 2016 Les textiles en Mésopotamie (750–500 av. J.-C.): techniques de production, circuits d’échanges et significations sociales. Thèse de doctorat. Université Paris 1 Panthéon-Sorbonne. Reiner, E. 1985 Your Thwarts in Pieces, Your Mooring Ropes Cut. Poetry from Babylonia and Assyria. Michigan Studies in the Humanities 5. University of Michigan. Sallaberger, W., and Hubert Vuillet, F. 2003–5 Priester A. I. Mesopotamien. Pp. 617–640 in Reallexikon der Assyriologie und vorderasiatischen Archäologie 10. Berlin: Walter de Gruyter. Schaudig, H.-P. 2001 Die Inschriften Nabonids von Babylon und Kyros’ des Grossen. AOAT 256. Münster: Ugarit-Verlag. Steinkeller, P. 1999 On Rulers, Priests and Sacred Marriage: Tracing the Evolution of Early Sumer­ ian Kingship. Pp. 103–137 in Priests and Officials in the Ancient Near East, ed. K. Watanabe. Heidelberg: C. Winter. Ur Online Project http://www.ur-online.org/ Waerzeggers, C. 2012 The Babylonian Chronicles: Classification and Provenance. JNES 71: 285–298. Weadock, P. N. 1975 The Giparu at Ur. Iraq 37: 101–128. Woolley, L. 1962 The Neo-Babylonian and Persian Periods. UE 9. London: Trustees of the British Museum and of the University Museum, University of Pennsylvania, Philadelphia. 1965 The Kassite Period and the Period of the Assyrian Kings. UE 8. Oxford: Trustees of the British Museum and of the University Museum, University of Pennsyl­ vania, Philadelphia; Oxford University Press.

170

Paul-Alain Beaulieu

Zadok, R. 1985 Geographical Names According to New- and Late Babylonian Texts. RGTC 8. Wiesbaden: Dr. Ludwig Reichert Verlag. 2000 Review of Steven W. Cole, Nippur in Late Assyrian Times, c. 755–612 BC. State Archives of Assyria Studies 4. Helsinki, University of Helsinki Press, 1996. Or. NS 69: 115–121. Zawadzki, S. 2013 “Era of Unrest” and Uruk Prebendaries in the Light of a New Document. ZA 103: 230–236. Zettler, R. L., and Hafford, W. B. 2016 Ur. B. Archäologisch. Pp. 367–385 in Reallexikon der Assyriologie und vorderasiatischen Archäologie 14. Berlin: Walter de Gruyter.

The Ziqqurats of Ur and Babylon and the Place Where the Ark Moors After the Flood (The Epic of Gilgameš XI 158) Daniel Bodi

University of Paris 4 – La Sorbonne

1. The Ziqqurats of Ur and Babylon The ziqqurats are among the most characteristic architectural achievements of ancient Mesopotamia that continue to generate an unabated production of scholarly studies and Museum expositions. The studies on the ziqqurats usually bear on the ritual function, the architectural features, the building materials and techniques, the drainage systems, the number of bricks used, the evaluation of the required work force, the amount of bricks the workers transported daily, and the pay they received in cereals, mostly issues related to the construction of the ziqqurats (Parrot 1949; Busink 1969; Schmid 1995; George 2005–6: 75–95). The present article deals mainly with the philosophical-theological meaning given to the ziqqurat in Babylon and its role as a mooring pole for the Ark according to one version of the Flood story found in Gilg. XI 158. One has to start with the ziqqurat from Ur, being one of the oldest, dating from the time of Ur-Namma and the Third Dynasty of Ur, since it has the advantage of being still fairly well preserved, in contrast to the one in Babylon where all that is left is just its imprint on the ground. The Ur ziqqurat preserved the first terrace and part of the second one. Comparably to what happened with the Coliseum in Rome, the Babylonian ziqqurat suffered even more from human abuse than from the passing of time since the bricks were apparently taken away and reused elsewhere. Martial, a Latin poet in the 1st century CE (40–104 CE), in his Liber Spectaculorum 1, heading his Epigrams, offered to the Emperor Titus for the inauguration of the Coliseum, compared Babylon’s ziqqurat to Rome’s Coliseum saying that the Roman building would become the most famous, by far surpassing the Babylonian one. Both constructions were pillaged and the Babylonian ziqqurat was utterly dismantled. The ziqqurats provided valuable high quality baked bricks since the external baked-brick mantle that covered the mud-brick core of the Ur ziqqurat, was between 1.15 m to 2.50 m thick for the first level; the baked-brick layer was 1.50 m thick at Eridu; and 1.45 m at Nippur (Sauvage 1998: 50). Besides the ziqqurat of Ur, the best partially preserved ziqqurat is found at Tchoga Zambil, forty km SE of Susa in Iran. 171

172

Daniel Bodi

The ziqqurats represented massive multi-layered towers constructed as superposed terraces, each upper one smaller in size and ending with a shrine erected on the top terrace. Numerous hypotheses of reconstruction have been offered in respect to the number and size of the terraces. Among ancient witnesses there is Herodotus who visited Babylon in 450 BCE and who mentions a religious ceremony that took place in the shrine on the top (Histories I 181–182). Herodotus probably counted the temple on the top of the ziqqurat as the eighth level on a seven-storey building. He reports that according to priests he spoke to, a woman slept there who had no commerce with men. The ziqqurats of Ur, Uruk, Nippur, and Eridu can be dated to the Third Dynasty of Ur owing to the inscribed bricks found in situ (Sauvage 1998: 45). The Ur ziqqurat, somewhat better preserved, has a base of 62.50 m x 43.90 m. Around 2,100 BCE, King Ur-Namma built a ziqqurat in honor of the moon god Sîn. It was called E-temen-ní-gùr, meaning “House whose foundation inspires awe.” On the eastern side, there was a triple staircase with two staircases along the façade and a third one positioned perpendicularly. Saddam Hussein’s botched “reconstruction” in 1980 significantly altered its original aspect since it presents the Ur ziqqurat with two stairways only. The ziqqurat was placed inside a large court on a platform whose dimensions were 140 m by 135 m and encased by a wall that served as an enclosure or a precinct with numerous chambers. The ziqqurat, the terrace, and the surrounding courts represented an imposing ensemble within the city. In Mesopotamia as in most civilizations, the notion of the divine is associated with height and elevation. From an anthropological point of view, this notion is related to an omnipresent tripartite manner the humans perceive and structure space and its content between what is up and elevated and what is on earth or on the horizontal level, as well as what is under the earth, the netherworld. The divine sphere is placed on an elevated, upper ground, implying the notion of higher, superior and more powerful beings (André-Salvini 2008: 16–23). As soon as hu­mans began building sanctuaries for divinities in Mesopotamia, at the Obeid period, around 5,000 BCE, the temples were placed on an elevated terrace. In this manner, it was distinguished visually, graphically, structurally, and symbolically from human habitations. The ziqqurats are usually perceived as natural developments of these artificially elevated terraces which as the centuries went by were built higher and higher. The analysis of the term ziqqurat can help us understand what these constructions represented. It comes from the verb zaqārum meaning, “to build high, to elevate,” while the term ziqqurat stands for the “(elevated) temple tower.”

2. The Cosmic and Religious Symbolism of the Ziqqurat in Babylon The Mesopotamian hero Atraḫasīs, the Babylonian Noah who survived the Flood and acquired immortality, is closely related to the ziqqurat as the spot where he landed and presented the first offerings to the gods. According to Gilg. XI 158 ina muḫḫi ziq-qur-ra-at šadî “on the top of the ziqqurat of the mountain” where the Ark moors after the Flood, represents the first dry land that emerges after the destruction of life and the cataclysm brought about by the Deluge. The new beginning and the place where life resumes after the Flood starts at the shrine located at the top of the temple tower. The ziqqurat represents, essentially, a cosmic symbol,

The Ziqqurats of Ur and Babylon

173

a central link, an axis mundi, a mooring pole. On account of its imposing height, reaching into the skies, it requires solid and deep foundations at the base of the building structure. In this manner, the ziqqurat connects different levels of the universe: the earth on the horizontal level, the skies above, and the deep foundations below linking the underworld. This aspect is particularly important in the city of Babylon perceived as the center of the world and where the name of the ziqqurat was E-temen-an-ki “House Foundation of Heaven and Earth.” This name together with the name of Marduk’s temple on the ground, next to the ziqqurat, and called É-sag-íl “House whose head is high,” or as stated in one of Nabopolassar’s (635–605 BCE) inscriptions who “made its top vie with the heavens” (re-e-ší-ša ša-ma-mi BE 1 84, George 2005–6: 83), has contributed to the creation of the legend of Babylon, the holy city par excellence, which has its prototype in the heavens. It explains the condemnation of excessive pride or hubris associated with the biblical story of the Tower of Babel “whose head reaches the heavens” (werōʾšô baššāmayîm) in Gen. 11:4, and whose completion God opposes. In the major part of the first millennium BCE, Babylon was the political, religious, and cultural center of Mesopotamia. The main religious ceremonies took place at the more accessible E-sagil (also transcribed Esangil). Given the exalted position of Marduk and his temple at Babylon as the supreme divinity and sanctuary of Babylonia in the first millennium BCE, it is no surprise that there survived a relatively large number of documentary sources which shed light on Marduk’s temple (George 1995: 173). The ziqqurat in Babylon had colossal dimensions for its time. It was 91 m wide and had a 60 m staircase. It was a seven-story building, over 70 m in height. The temple-tower was protected and encircled with a huge wall with various buildings that formed a square of 460 m by 420 m. The ziqqurat was higher than any other building in the city of Babylon and could be seen from afar. The ziqqurat in Babylon was originally built around the time of Hammurabi (1792–1750 BCE). The restor­ ation began under Nabopolassar, and was finished after forty-three years of work under Nebuchadnezzar II (604–562 BCE). In 538 BCE Cyrus dismantled the three stair ramps to prevent from using the tower as a fortress. In his study of the religious representation of Babylon as the replica (Abbild), image, and navel of the world in the Creation epic Enūma eliš, S. Maul points out that for the one who visited the city of Babylon the impression of being at the axis mundi came from seeing the seven-story ziqqurat, E-temen-an-ki. Moreover, according to Enūma eliš, Esagil’s roots are set in the apsû (En. el. VI 63: ziqqurrat apsî elīte “the upper ziqqurat of the Apsû”). In the Babylonian Creation Epic, the temple complex, the ziqqurat, and E-sagil are given a central cosmic position and paramount importance. It’s the place where the universe holds together as around a mast (Maul 1997: 115; Glassner 2002: 32 é.dim.an.ki “house, mast of heaven and earth”). According to En. el. V 117–130, Marduk creates Babylon as the terrestrial counterpart to Ešarra, abode of the gods in heaven. The gods are to repose there during their earthly sojourns when they come up from the Apsû or as a stopping place when they come down from heaven for the divine assembly. From these lines, it can be deduced that most of the gods from the Mesopotamian pantheon had their cellas or chapels in the E-temenaki and E-sagil (Foster 1996 I: 381–82). In the Erra Epic, Marduk, the patron god of Babylon, plays a decisive role in restoring life in the devastated city of Babylon and in the rest of Mesopotamia. In the concluding, fifth tablet of Erra V 23–27, 40–41, in an “oracle of restoration” after

174

Daniel Bodi

the god of plague Erra wrought havoc and devastation through war, life is restored in Babylon and in Mesopotamia through Marduk’s beneficial intervention (Bodi 1991: 290–292). In Gilg. XI, the Ark mooring at the ziggurat on the top of which is Marduk’s shrine serves the same purpose. After the devastation brought about by the Flood, Marduk’s beneficial life-ensuring presence restores life in Mesopotamia. The paramount theological importance of Babylon and its ziqqurat has been exploited in one version of the Flood account found in the Gilgameš Epic. This link was established through various literary means examined below.

3. The Ark Mooring at the Top of the Ziqqurat in Babylon After the Flood The argument that Utanapištim’s Ark corresponds in its structure to the ziqqurat in Babylon and that it landed at its top after the Flood is based on a series of textual indications enumerated below.

3.1. The Louvre E-sagil Tablet (AO 6555) J.-J. Glassner (2002: 32–34) has suggested that there exists a link between the symbolic numbers of the seven-story ziqqurat E-temen-an-ki, its architectural divisions (area base, length, breadth, and height) as found on the Louvre E-sagil Tablet, and the detailed specifications of the seven-story Ark and its inner compartments as described in the Gilgameš Flood Tablet XI. Both have the same seven-fold division in the vertical plane and a nine-fold division in the horizontal plane. As described in the E-sagil Tablet, Marduk’s ziqqurat is a structure with the same nine-fold division in the horizontal plane, implicit in the notation of the dimensions of the ziqqurat’s first stage as 3x3 ṣuppān, and explicit in the nine chambers listed for its seventh stage. The Late Babylonian E-sagil tablet from the Eanna temple complex in Uruk was copied in 229 BCE from an original from Borsippa. The colophon gives the name of the scribe Anu-bēlšunu, son of Nidintu-Ani, descendant of Sîn-leqe-unninnī. He claims being a descendent of the same famous scribe who compiled the canonical version of the Gilgameš Epic (George 1992: 118), probably another case of a tailored genealogy (Bodi 2015: 311–320). As pointed out by A. George, the E-sagil Tablet is not an accurate, eyewitness’ description of the ziqqurat, nor is it an architect’s blueprint for an eventual construction. The text is more abstract and academic than an architect’s plan (George 2005–6: 77). Furthermore, it’s a misnomer since the major part of the E-sagil Tablet deals with dimensions of the adjacent ziqqurat. It is a compilation of mathematical exercises, probably drawing its figures from an ideal, not from reality. Two systems of measurements are used in the text: the Kassite or early NB system (Powell 1982: 106–123; 1989: 468, the E-sagil tablet would be a “Seleucid copy of older material going back to late Kassite or early NB periods”), and the later Standard NB system, each of which uses different cubit standards. The cubit-standard was understood and used as a pedagogical tool in academic circles long after it fell out of use in practical surveying. Even if the base dimensions it gives have been verified on the ground, they remain theoretical and symbolic (enclosing a square of exactly one ikû in area) and do not compel us to accept as unquestionable the other figures recorded (George 1992: 156; 2003 I: 513). J.-J. Glassner suggested that the text may have come to have other significance. He has found in it a more

The Ziqqurats of Ur and Babylon

175

speculative purpose of a philosophical-theological nature, observing a cosmological allusion in the matching size in SB Gilg. XI, both having bases of one ikû, and both subdivided on the vertical plane into a seven-story construction. As he notes, the ikû is also the key dimension in the first section of the E-sagil Tablet, being the notional area of the Grand Court (kisalmaḫ). A. George has found an additional important astronomical feature associated with the use of ikû. The ikû was also held to be visible in the night sky, where it was the great square of Pegasus, the celestial counterpart of both E-sagil and Babylon (George 1992: 244). The measurements given in the E-sagil tablet are quoted in a Late Babylonian commentary (George 1992: text no. 62) which shows that its subject matter was known to learned scholars and used in the exposition of difficult contexts outside mathematics, implying theological speculations about the deeper significance of the dimensions of the ziqqurat (George 2005–6: 77). The fact that both the ziqqurat as a central mast where Marduk, the king of the gods resided between heaven and earth holding the world together, and the Ark, a microcosm of all the living creatures that survived the Flood, have in common a dimensional scheme is very likely to be explained in terms of cosmic symbolism. The significance of number nine in the cosmic pattern might be related to the fact that in a blueprint of nine squares it is particularly easy to identify the middle square, as the central mast or axis mundi. The eight paragraphs of the E-sagil tablet describe the seven levels of the ziqqurat, the sixth one was omitted by the scribe, giving the height of each level. In conclusion, the ziqqurat is described as a cube having 90 m on each side, each surface being 1 ikû, or 90 x 90= 8,100 m2, divided in seven levels where the first and the last levels are subdivided into nine compartments. In the description of the Ark’s surface in Gilg. XI 58 “one ikû (1.iku) was her surface kippat(gúr)-ra, 10 nindan each her sides stood high” (George 2003 I: 707). One should note, however, that kippatu can also mean “circumference of a circle” (CAD K: 397, courtesy of E. Frahm). It would bring the circular shape of the Ark closer to Finkel’s coracle in the “Ark Tablet.”

3.2. The Ark Lands on a Sacred Mountain Ni-muš Second, there is a further link mentioned in the Gilgameš Epic XI 158 where at the end of the Deluge, once the Ark was immobilized and held fast for six days by the mountain Ni-muš (Gilg. XI 142–146), on the seventh day after releasing a dove, a swallow and a raven, Uta-napištim “offered incense at the ziqqurat of the mountain:” áš-kun sur-qin-nu ina muḫḫi(ugu) ziq-qur-rat šadî(kur)i. While earlier scholars read the name of the mountain where the Ark landed after the Deluge as Ni-ṣir, the newer translations of J. Bottéro (1992: 193 n. 2) and of A. George (2003 I: 516), following Lambert’s (1986: 185) reference to a theophorous name i-din-ni-muš, now read it as Ni-muš, which would indicate a divinized or a sacred mountain.

3.3. The Number Seven The Gilgameš Epic displays a great usage of numbers to the point that D. O. Edzard (1991: 57–67 [57]) wrote an entire article with an inventory of about thirty references in the Epic of Gilgameš where numbers are mentioned. Edzard describes the ancient Babylonians as a “number friendly people:” “ein so zähl- und rechenfreudiges Volk wie die Babylonier mit dem Abzählen, Aufzählen, Auszählen”…(und Erzählen).

176

Daniel Bodi

The Herodotus account in his Histories I: 178–182.2 implies a seven-story ziggurat. The E-sagil tablet from Seleucid times also enumerates seven levels of the ziqqurat in Babylon. Finally, a NB stele from the Schøyen Collection depicts King Nebuchadnezzar II standing before a seven-story ziqqurat drawn in outline and labeled [é].tem[en].an. [z]i-qú-ra-at bābili(ká.dingir.ra)ki “E-temen-anki, ziqqurat of Babylon.” The tower is as high as it is wide (George 2011). In respect to our topic, the systematic repetition of the figure seven establishes a link between the seven-story ziqqurat, the seven-story Ark (Gilg. XI 62 [a]ptara[ss]u ana 7-šu “I divided her into seven parts” (George 2003 I: 512), the seventh day on which the birds are released and the seven incense flasks offered on the top of the ziqqurat. In the Gilgameš Epic XI 158 at the end of the Deluge, once the Ark was immobilized and held fast for six days by the mountain Ni-muš (Gilg. XI 142–146), on the seventh day after releasing a dove, a swallow, and a raven, Uta-napištim “strewed incense at the ziqqurat of the mound, seven flasks and seven I set in position”: áš-kun sur-qin-nu ina muḫḫi(ugu) ziq-qur-rat šadî(kur)i 7 u 7 dug adagurru (a.da.gur5) uk-tin (XI 158–159) (Bodi 2014: 17–43, on the various types of the numerical sequence in ancient Near Eastern literature).

3.4. The Term Mountain šadû as a Metaphor for the City of Babylon In the syntagm ziq-qur-rat šadî(kur)i (Gilg. XI 158), which is an unusual coin­ age, the term šadû seems to be a metaphor for the city of Babylon as indicated in several texts: “Babylon ša-du-ú ša ṣurri Babylon the mountain of ṣurru-colored glazed bricks” (RT 19 51 HS 1893:1 [MB Lit]). It probably refers to its ziqqurat covered with glazed bricks. Another text establishes a parallelism between šadû “mound, mountain” and “Babylon:” ana ša-di-im (var. ša-da) napišti nišī KÁ.DINGIR.RAki aškun “I made Babylon a stronghold? / mound for the sustenance (?) of people” (Sumer 3 8 ii 10 and 15 ii 13, var. from VAB 4 92 ii 15 [Nabuchadnezzar II]) (CAD Š/1: 57). Most translators of Gilg. XI 158 disregard ziq-qur-rat as a redundant metaphor for the mountain peak and translate it with “I strewed incense on the peak of the mountain” (George 2003 I: 713 with the literal meaning placed in a note). At least one translator attempted to render the line literally: “I offered incense in front of the mountain-ziggurat” (Kovacs 1989: 102). In light of the above discussion, one could suggest the following translation: “I strewed incense at the ziqqurat of Babylon,” where šadû stands as the metaphor for the city of Babylon. The term ziqqurat should probably not be rendered in a metaphoric way as “peak” but left in its literal sense. Two cylinder seals from the Baghdad Museum in Iraq, one from Tell Asmar (IM 18971) from the Early Dynastic (ED II) times, and a second (IM 31406) equally dating from ED II, could explain the link between offerings and the ziqqurat. The cylinder seals represent the ziqqurat as a triangle by two intersecting lines. Three horizontal lines indicate that the ziqqurat is a four-story construction. On the top of it is a horizontal line and above it is another triangular object that Pierre Amiet interprets as an offering. The temple tower would serve as an altar for offerings to the divinity (Amiet 1951: 80–88; Amiet and Basmachi 1952: 78–81).

The Ziqqurats of Ur and Babylon

177

4. Not One Version of the Flood but Multiple Ones The above discussion confirms the view that there is not one version of the Flood in Akkadian literature but several ones with multiple variations. There are several Old Babylonian versions of the Flood story, the first version is found in the Atraḫasīs Epic, tablet III, dating from the seventeenth century BCE, while the second version is represented by the Sumerian Flood story (Lambert and Millard 1969: 98–99). A third version is the Gilgameš Flood story found in Gilg. XI, well preserved in the Ninevite SB version. It was inserted in the Gilgameš Epic by the scribe Sîn-lēqē-unninni by the end of the Kassite times around the twelfth century BCE. There is a fourth version of the Flood found in the ninth century BCE Erra Epic where it is referred to as a major cosmic upheaval and dislocation, with a far more apocalyptic vision than the one found in the OB Atraḫasīs Epic (Bottéro and Kramer 1989: 711), when Marduk left his seat in Babylon and the “government of heavens and earth was dissolved” šipiṭ šamê u erṣetim ša uptaṭiru (Erra I 136; Bottéro 1977/1978: 118 “the bond of heaven and earth was untied”). There is now a fifth version in the so-called “Ark Tablet” published by I. Finkel (2014). According to Finkel, the Ark was a round coracle like modern Iraqi “guffa.” Therefore, there would not be one version of the Ark but at least three different ones, each one giving the Ark a different shape and dimension. First, in Gilg. XI, the Ark is a seven-story vessel in the form of a cube; second, in Finkel’s “Ark Tablet,” the Ark is a round boat comparable to a coracle, a bowl-shaped boats still in use in Iraq; third, in the Genesis Flood story (Gen. chs. 6–9) the Ark is presented as an elongated boat, 300 cubits long (=150 m), 50 cubits (=25 m) in breadth, and 30 cubits (=15 m) in height (Gen. 6:15) (Tsumura 1994: 27–57). Furthermore, the OB Atraḫāsis Ark is smaller than the SB Gilgameš XI one. I. Finkel attempted to harmonize the two accounts of the Ark and failing to do so finally states the following: “My suspicion, however, is that the Gilgameš Ark as it came to be launched from Nineveh is primarily the consequence of textual misunderstanding compounded by the insertion of narrative without care as to overall meaning coupled with interventing (sic) editing” (Finkel 2014: 327–332). However, instead of trying to harmonize the square cubic Ark from Gilgameš with the round one from the “Ark Tablet,” it is better to simply refer to two different traditions about the Ark (however, see above on the meaning of kippatu in Gilg. XI 58). Moreover, there is not one set of birds that are released from the Ark but three different birds depending on the accounts: in Gilg. XI 148, Ūta-napištim releases a dove summata(TU)mušen, in l. 151 a swallow sinnūnta(SIM)mušen, and in l. 154 a raven (ariba); in Genesis 8:7–8, Noah releases a raven (ʿōrēb) and a dove (yônâ) in that order. In the Akkadian Flood fragment from Ugarit (RS 94.2953), discovered in 1994 and published in 2007, originally attributed to the construction of Ba’al’s temple but correctly translated by A. Cavigneaux (2007: 319–320), mentions a dove and a kumâmušen “water-bird.” The Ugaritic fragment is highly pertinent as it brings a clarification about the technique of releasing several different birds to scout out for the land and explains why the dove flew back to the Ark, out of exhaustion, presumably not having found a place to rest. 10. apta ēpuš ana muḫḫī-ya 11. dannata kappī summatamušen

I made an opening above me, strong of wings, a dove–

178

Daniel Bodi 12. umaššir illik itūr-am-ma I released. She went forth and returned to me, 13. ušāniḫi kappī-ša she exhausted her wings 14. ašnī-ma anāku kumâmušen umaššir-ma Once again I released a water-bird (seagull, cormorant, or pelican?)

Moreover, the various versions report different mountains where the Ark landed after the Flood (George 2003 I: 516 for the analysis of different locations). Since there is also a thirteenth-century BCE copy of the Flood story found at Ugarit (RS 22.421) (Lambert and Millard 1969: 131–133), as well as a fragment of the Gilgameš Epic found at Megiddo in the land of Canaan (Horowitz, Takayoshi, and Sanders 2006), it implies that by the end of the second millennium BCE, the scribes in the Northwest Semitic geographical region (Ugarit and Megiddo) knew and copied the Gilgameš Epic as part of their standard scribal training. This raises the issue of the relationship between the written and oral transmission of some of the key motives from the Gilgameš Flood story to other works of literature as, for example, the Hebrew one. In the course of the transmission of the Flood story each region adapted the details to its local preferences and characteristics.

Conclusion The article argued for the existence of a link between the symbolic numbers of the seven-story ziqqurat, its architectural divisions (area base, length, breadth, and height) as found on the Louvre E-sagil Tablet and the detailed specifications of the seven-story Ark with its inner compartments as described in Gilgameš Epic XI. A series of indications found in the Gilgameš Flood Story and the E-sagil Tablet were analyzed pointing to the existence of philosophical-theological speculation. In the Gilgameš Epic XI 158 at the end of the Deluge, once the Ark was immobilized and held fast for six days by the sacred mountain Ni-muš (Gilg. XI 142–146), on the seventh day after releasing a dove, a swallow, and a raven, Uta-napištim “offered incense at the ziqqurat of the mountain,” placing seven flasks in position. The systematic repetition of the figure seven establishes a link between the seven-story ziqqurat and the seven-story Ark, the seventh day on which the birds are released and the seven incense flasks. These various elements make it more appropriate to translate the unusual collocation in Gilg. XI 158 literally and state that Uta-napištim offered incense on the “ziqqurat of the mountain,” i.e., on the sacred mountain, instead of giving the term ziqqurat a transferred meaning “mountain peak” (e.g., CAD Z: 131). Moreover, the term šadû appears in several texts as a metaphorical reference to Babylon. The myth and legend of the Flood as reflected in the Gilgameš Flood Tab­ let, do not deal with a precise geographical reference nor with a specific historical event. Rather, on this point the text establishes an important link between the Ark and the ziqqurat, and this conjunction carries a religious meaning. In the Gilgameš version of the Flood story the Ark stands for a symbolic representation of the world as a microcosm of all the surviving creatures, while the ziqqurat E-temen-an-ki in Babylon stands for the central pole of the world uniting the heavens (an) and the earth (ki) as reflected in a series of Sumerian names for temples and ziqqurats. The Ark at the end of the Flood lands at the summit of the ziqqurat at Babylon assuming the role of the mooring pole. In this manner it brings Marduk, the patron god of

The Ziqqurats of Ur and Babylon

179

Babylon, supreme god of the Mesopotamian pantheon, the creator, protector, and guarantor of life in contact with all life that survived the Flood. This conjunction between the Ark and the ziqqurat is essential for life to thrive again in Mesopotamia and on earth.

Bibliography Amiet, P. 1951 La ziggurat d’après les cylinders de l’époque archaïque. RA 45: 80–88. Amiet, P. and Basmachi, F. 1952 Deux représentations nouvelles de la ziggurat. Sumer 8: 78–81. André-Salvini, B. 2008 Introduction. L’héritage de Babylone. Pp. 16–23 in Babylone, ed. B. André-Salvini. Paris: Musée du Louvre. Bodi, D. 1991 The Book of Ezekiel and the Poem of Erra. OBO 104. Fribourg, Switzerland: Universitätsverlag; Göttingen: Vandenhoeck & Ruprecht. 2013 The Numerical Sequence x/x+1 in Aramaic Aḥiqar Proverbs and in Ancient Near Eastern Literature. ALIENTO 4: 17–43. 2015 Traditional Claims of an Illustrious Ancestor in Craftsmanship and in Wisdom. Pp. 311–320 in Tradition and Innovation in the Ancient Near East (57th RAI, Rome, 4–8 July 2011), ed. A. Archi. Winona Lake, IN: Eisenbrauns. Bottéro, J. 1977/8 L’Épopée d’Erra. Annuaire de l’École Pratique des Hautes Études: 107–164. 1992 L’Épopée de Gilgameš. La grand homme qui ne voulait pas mourir. Paris: Gallimard. Bottéro, J. and Kramer S. N. 1989 Lorsque les dieux faisaient l’homme. Mythologie mésopotamienne. Paris: Gallimard. Busink, M. T. A. 1970 L’origine et évolution de la ziggurat babylonienne. JEOL 21: 91–141. Cavigneaux, A. 2007 Les oiseaux de l’arche. Aula Orientalis 25: 319–320. Finkel, I. 2014 The Ark Before Noah. Decoding the Story of the Flood. London: Hodder and Stoughton. Foster, F. B. 1996 Before the Muses. An Anthology of Akkadian Literature. Bethesda, MD: CDL Press. George, A. R. 1992 Babylonian Topographical Texts. OLA 40. Leuven: Peeters. 1995 The Bricks of E-Sagil. Iraq 57: 173–197 2003 The Babylonian Gilgamesh Epic. Introduction, Critical Edition and Cuneiform Texts. Oxford: Oxford University Press. 2005/6 The Tower of Babel: Archaeology, History and Cuneiform Texts. AfO 51: 75–95. 2011 A Stele of Nebuchadnezzar II. Pp. 153–169 in Cuneiform Royal Inscriptions and Related Texts in the Schøyen Collection, ed. A. R. George. CUSAS 17. Bethesda, MD: CDL Press. Glassner, J.-J. 2002 L’Etemenanki, armature du cosmos. NABU 2002/2: 32–34 no. 32.

180

Daniel Bodi

Horowitz, W., Takayoshi, O., and Sanders, S. 2006 Cuneiform in Canaan, Cuneiform Sources in the Land of Israel in Ancient Times. Jerusalem: Israel Exploration Society. Kovacs, M. G. 1989 The Epic of Gilgamesh. Stanford: Stanford University Press. Lambert, W. G. 1986 Note breve, Niṣir or Nimuš. RA 80: 185–186. Lambert, W. G. and Millard, A. R. 1969 Atra-ḫasīs. The Babylonian Story of the Flood, with the Sumerian Story by Miguel Civil. Oxford: Clarendon. Martial 2006 Liber Spectaculorum, ed. K. M. Coleman. Oxford: Oxford University Press. Maul, M. M. 1997 Die altorientalische Hauptstadt – Abbibild und Nabel der Welt. Pp. 109–124 in Die orientalische Stadt. Kontinuität, Wandel, Bruch, ed. G. Wilhelm. Saarbrücken: Saarbrücker Druckerei und Verlag. Parrot, A. 1949 Ziggurats et la tour de Babel. Paris: Albin Michel. Powell, M. A. 1982 Metrological notes on the Esagila tablet and related matters. ZA 72: 106–123 1989 Masse und Gewichte. RlA 7/5–6: 457–517. Sauvage, M. 1998 La construction des ziggurats sous la Troisième Dynastie d’Ur. Iraq 60: 45–63. Schmid, H. 1995 Der Tempelturm Etemenanki in Babylon. Baghdader Forschungen 17. Mainz am Rhein: P. von Zabern.

Epigraphy of Ur: Past, Present, and Future Dominique Charpin Collège de France

Introduction 1 May 2015: One more e-mail from Elizabeth Stone…but this time, a definitive one: I was invited as epigrapher for the next season, in the autumn! “You had indi­ cated that you would be happy to come to Ur”: yes, I was! I visited the site for the first time in autumn 1974; I was at that time a very young student, excavating at nearby Larsa under the direction of Jean-Louis Huot. When I came back to Larsa as an epigrapher thirteen years later, in 1987 and 1989, circumstances did not allow me to visit Ur again, to which I had in the meantime devoted two books (Charpin 1980a and 1986). So I was very excited to take part in this new page of the excavation of a fascinating site… As shown by its title, my paper will focus on the epigraphy of Ur. I will first recall the past, then describe the present efforts to make the Old Babylonian texts of Ur more accessible, and finally give a brief report on the epigraphical results of the last season, at the end of 2015.

The Past… The First Excavations by Taylor After a very short exploration by Loftus, “J. G. Taylor, British Vice-Consul at Basra, conducted the first excavations at Tell al-Muqayyar from 1853 to 1854.” I quote from the website Ur Online: 2 He focused on the ziggurat and recovered inscribed clay cylinders and bricks stamped with inscriptions that led Rawlinson to identify Tell al-Muqayyar as the fabled city known as ‘Ur of the Chaldees,’ the traditional home of the biblical patriarch, Abraham. 1.  This contribution has been written in the frame of the ANR funded “EcritUr” project (see http:// digitorient.com/?p=3341). It includes sections added after the oral presentation during the Rencontre. I wish to thank Brad Hafford, who helped me find the information I needed in the treasures stored in the Ur Online website. 2. http://www.ur-online.org/about/5/. The initials of the publication in Taylor 1854 (J. E.) were wrong and the correct ones are J. G. (see Sollberger 1972: 129 n. 2).

181

182

Dominique Charpin

There is here no mention of the private archives that were discovered by Taylor in what he called the “Tomb Mound,” a part of the site later designated as EH by Woolley. 3 What Taylor thought to be a big cemetery was in fact the vaulted burials built under Isin-Larsa houses which later disappeared because of erosion. He described his findings in these terms: 4 At the back of the the vault described above I found, close to the foundation, a small unbaked inscribed clay tablet, and I also dug up, at two feet only below the surface, about thirty small and large fragments of the same. From being so close to the surface they were of course considerably damaged.

A report of Rawlinson to the Trustees of the British Museum testifies that these tablets were sent to London: 5 The collection of bricks, tablets and cylinders which he has packed for transport to England promises to be of great interest.

It happened that the discoveries of Taylor at Ur, at their arrival in the British Museum, were mixed with the tablets found by Loftus at the same time at Tell Sifr, ancient Kutalla (Loftus 1857: 270). Ultimately, these hundred or so tablets were published by Strassmaier as coming from Warka! I want to stress that this book (Strassmaier 1882) was the first one where Old Babylonian archives were ever published. Fifty years later, the same tablets were copied again by Jean (1931), who corrected the attribution to Warka into Tell Sifr, without realising that one-third of the tablets had no connection with the rest. In August 1978, I was working on my PhD: the idea was to write a family mon­ ograph based on the archives of Ṣilli-Ištar and his brother discovered at Tell Sifr. But thirty-four tablets could not be linked to the seventy-two forming the bulk of the archives. Kraus had already noted that some contracts followed the same drafting as those of Ur (Kraus 1955), and the CAD made some prosopographical connections between texts supposedly from Tell Sifr and others published in UET 5. 6 And suddenly, I understood that the explanation was very simple: I had to deal with texts written at Ur! After some more research, I found evidence that these tablets were those discovered by Taylor (which were at that time completely forgotten). My book, published in 1980, has been very positively received. I only regret its subtitle (Étude des documents de «Tell Sifr»), where I put the name of Tell Sifr in quotes: it would have been better to write Étude de documents d’Ur et de Kutalla (Tell Sifr). 7 Since 1980, our knowledge of Old Babylonian archives has increased: some connections between these thirty-four tablets can be better explained, especially by the phenomenon of transfer of property titles which is now much better understood. 8 3.  For the explanation of the sigla EM and EH, see Woolley and Mallowan 1976: XV. 4.  Taylor 1854: 274. 5. Letter of Sir H. C. Rawlinson, 19th February, 1855, to the Trustees of the British Museum; quoted in Sollberger 1972: 139 (italics mine). 6.  In the article “kakikku,” CAD K: 44b (no author is credited in the CAD, but one may suppose that this section is due to J. Renger, one of the contributors of CAD K). 7.  The list of the thirty-four texts from Ur published as found at Tell Sifr is: 1–2, 4–9, 11, 13–16, 20–26, 87–94, 96–98, and 105–107. I also regret now that I did not make separate indexes for the Ur texts (Charpin 1980a: 299–347). 8.  For instance, despite what I wrote (Charpin 1980a: 34–36), there is no archive to be attributed to “Abi-ṭabum fils d’Abiya.” The archive is to be ascribed to Igmil-Sin son of Buzazum. Igmil-Sin bought

Epigraphy of Ur: Past, Present, and Future

183

Campbell Thompson and Hall (1918–1919) Just after World War I, excavations were resumed at Ur by the British Museum. In 1918, Campbell Thompson made some “test-excavations” for about a week, but his attention was mainly devoted to nearby Eridu. One year later, Hall excavated also at Ur; he benefitted from the work of seventy Turkish prisoners of war. Hall excavated some houses in the neighborhood later called EM by Woolley and described his epigraphic findings in this way: 9 On February 19, more tablets were found lying by the stump of a burnt brick wall, including one very fine one (No. 113915), now in the British Museum, which contains a serie of prognostications derived from the observation of the flight of birds, including the eagle, the hawk, and the mountain-bird. It dates from about 2000 B.C. Other tablets from the same place contained medical prescriptions, business and legal documents of the same period.

The omen tablet is the only one to have been published (Weisberg 1969). More details on this discovery have been given by Irving Finkel (2000: 145 n. 21): One can point here to a batch of real treasure from the British Museum now being prepared for publication by the present writer, namely the Isin-Larsa period medical tablets mentioned by H. R. Hall in A Season’s Work at Ur (London, 1930) 173. The latter contain many helpful explicit writings, and are purely therapeutic: there is no hint of any magic in the identified pieces.

But here, we are already in the future, because these texts are still unpublished… 10

Woolley’s Excavations Back to the past. During his twelve seasons of excavation, between 1922 and 1934, Leonard Woolley was not only successful in exploring “the ziggurat and its surroundings” or excavating the famous royal graves. He also discovered a great number of texts, and in most of his seasons, his very small staff—never more than five people—included an epigrapher. Let us stress that this was unusual at that time: André Parrot excavated at Mari since 1933 during four seasons before he asked Georges Dossin to go with him for the fifth and sixth seasons, in 1937 and 1938, after the bulk of the royal archives had already been excavated…. Let me recall the six epigraphers who worked for Woolley’s expedition. Field Epigraphers 11 During the first season (1922–23), “Mr. Sidney Smith of the British Museum dealt with the inscriptions” (Woolley 1923: 311). In his obituary of Sidney Smith, Wiseman wrote: “The experience introduced him to Iraq and its archaeological possibilities, though the restrictive nature of some of the expedition’s life and practices were not altogether to his liking” (Wiseman 1981: 466). This did not prevent him a ruined house from his uncle Abi-ṭabum (HEO 12 7), which the latter had received as inheritance share one month before (HEO 12 8: 10–13); at the time of the sale, Igmil-Sin received the property title of Abi-ṭabum. 9.  Hall 1930: 172; see Charpin 1986: 129–130. 10. I. Finkel gave a very interesting paper on them in Paris in 2006, but unfortunately did not publish it in the proceedings of the colloquium (Attia, Buisson and Geller 2009). 11.  See the list established by Mallowan (1960: 17).

184

Dominique Charpin

from becoming soon afterward Director of Antiquities and Director of the Iraq Museum in Baghdad. For the second season (1923–24), Woolley wrote: “This time I had with me Mr. Cyril J. Gadd, of the British Museum, for work upon the inscriptions” (Woolley 1924: 329). Gadd was to publish as soon as 1928 (with Legrain) the first volume of the series Ur Excavations Texts, 12 as well as one year later his book History and Monuments of Ur. He acted as editor for the series Ur Excavations Texts till his death in 1969, publishing himself, with the collaboration of S. N. Kramer, two of the three volumes devoted to Literary and Religious Texts (UET 6/1 [1963] and 6/2 [1966]). For the third season (1924–25), Woolley was more enthusiastic in his report: “For the epigraphical side of the work I had associated with me Dr. L[éon] Legrain, of the University Museum, to whose help I owe much more than I can express: even in this preliminary report it will be clear how greatly our discoveries gained in interest and value from his study of the inscriptions” (Woolley 1925a: 347). This explains perhaps why Legrain came back for the fourth season (1925–26): “For the second year in succession, Dr. L[éon] Legrain, of the University Museum, was with the Expedition in charge of the inscriptional material” (Woolley 1925b: 365). Legrain published the royal inscriptions discovered during these two seasons in 1928 in UET 1. He later published the two huge volumes of UET 3 (1937 and 1947) devoted to 1800 Ur III tablets discovered between 1924 and 1931. He also contributed to the archaeological series, with a book devoted to Archaic Seal Impressions (UE 3, 1936) and another one to Seal Cylinders (UE 10, 1951). When Woolley dug the “EM” quarter during the fifth season (1926–27), a new cuneiformist was in the field: “This year the inscribed material was dealt by the Rev. E[ric] Burrows, S.J.” (Woolley 1927: 385). Burrows had studied Assyriology in Oxford with Langdon. He had already some field experience as epigrapher, since he took part to the Kish expedition in 1924–25 13. Burrows came back to Ur for the sixth (Woolley 1928: 415), seventh (Woolley 1929: 305), and eighth (Woolley 1930: 315) seasons. It was during the fifth and above all the seventh seasons that the archaic tablets had been found, which Burrows published in 1935 in UET 2. He died untimely in 1938 in a car accident. Woolley wrote in the Times: 14 It is impossible to describe his worth as a scholar without adding a word of regret for one of the most charming personalities that ever triumphed over the difficulties of camp life, for the gaiety and whimsical humour that overlay his deeply-felt convictions.

It must be emphasized that Burrows was the first of Woolley’s epigraphers— and the only one—who signed contributions published as appendices to Woolley’s annual reports in The Antiquaries Journal. 15 12.  A volume which also included copies by S. Smith and E. Burrows, as explained by F. G. Kenyon in his preface. 13.  See the obituary by R. Campbell Thompson in the Journal of the Royal Asiatic Society 70 (1938): 644. 14. According to the quotation I found in http://archive.thetablet.co.uk/article/2nd-july-1938/26/ obituaries. 15. “The Inscriptions By Rev. E. Burrows, S.J.,” in Woolley 1927: 404–406; “The Inscriptions With Observations on the Chronology By E. Burrows,” in Woolley 1929: 339–343; “Tablets and Sealimpressions By E. Burrows,” in Woolley 1930: 341–343.

Epigraphy of Ur: Past, Present, and Future

185

The ninth season (1930–31) was one of the longest led by Woolley; 16 the epigrapher was Chauncey Winckworth (Woolley 1931: 343), who was lecturer in Assyri­ ology at Cambridge. Indeed, Winckworth is not very famous, 17 and as far as I know he has only left copies of three inscribed bricks, published by Sollberger in UET 8. 18 Apparently, Woolley was not very satisfied with his work: 19 I regret to say that early in the month Mr. Winckworth was ill and as the local treatment failed of effect had to be sent to the hospital in Basra, where he was kept for a fortnight suffering from dysentery; he is now back, and fit for work again, but his absence, at a time when inscribed material was coming in fairly freely has meant that on the epigraphical side work has fallen into arrears; Mr. Railton’s knowledge of cuneiform proved most useful in the interim, but much copying has been held over. In this case again I have had to meet unexpected expenditure.

It is certainly unfortunate that the “AH quarter” was excavated with such an unexperienced epigrapher: thanks to Ur Online, the field files relating to the tablets then discovered are available and the quality of many of them for that season is rather poor—and too often, we find the mention “No data.” For the tenth season (1931–32), “no epigraphist was engaged, for the work contemplated was not expected to produce much in the way of inscriptions; but an arrangement was made whereby Dr. Cyrus B. Gordon, epigraphist on the Tell Billah Expedition of the Museum of the University of Pennsylvania, could be called upon to give his services when required; actually a single visit enabled him to do all that was essential” (Woolley 1932: 355). Cyrus Gordon was then 23 years old… and he was rather proud later to recall that he had been Woolley’s epigrapher at Ur. 20 Apparently, no epigrapher took part to the last two seasons (1932–33 and 1933–34), which were intended to close the digging. The Publication of the Texts As is well known, the publication of the texts discovered by Woolley at Ur was made in a typological and chronological manner, paying little attention to their provenience. Three categories can be distinguished: A.  Royal Inscriptions •  C. J. Gadd & L. Legrain, with contributions by Sidney Smith and E. R. Burrows, Royal Inscriptions, UET 1, London, 1928. •  E. Sollberger, Royal Inscriptions Part II, UET 8, London, 1965. 16.  “The Joint Expedition of the British Museum and of the Museum of the University of Pennsylvania started its ninth season at Ur on 1st November 1930, and continued in the field until 20th March 1931. The season was a long one and a larger number of workmen than usual was employed, the average for the first three and a half months being 280 and for the remainder of the time 200; the amount of actual excavation done was in consequence greater than in any previous year” (Woolley 1931: 343). 17.  I remark that no biographical data has been given by Ur Online for Winckworth among the people taking part to the ninth season (http://www.ur-online.org/term/779/). 18. As a curiosity, I quote the brief obituary which has been published anonymously in Revue d’Assyriologie in 1954: “Chaucey Winckworth était depuis plus de vingt-cinq ans maître de conférences en Assyriologie à l’Université de Cambridge. Il avait participé comme épigraphiste à la 9e saison de fouilles à Ur en 1930–31. De santé fragile, il s’était consacré à l’enseignement, et chacun sait qu’il était toujours prêt à mettre son savoir au service des étudiants.” (RA 48 [1954]: 206). 19. http://www.ur-online.org/media_item/205792/. 20.  Gordon 2000: 35–36.

186

Dominique Charpin B.  Archives: From the Early Dynastic to the Persian Period •  E. Burrows, Archaic Texts, UET 2, London, 1935. See now Pomponio & Alberti 1986; Lecompte 2013 and 2015; Benati and Lecompte 2016. •  L. Legrain, Business Documents of the Third Dynasty of Ur. Plates, UET 3, London, 1937; Indexes, Vocabulary, Catalogue, Lists, London, 1947. •  D. Loding, Economic Texts from the Third Dynasty, UET 9, Philadelphia, 1976. Supplemented by D’Agostino, Pomponio and Laurito 2004; Black and Spada 2008 (nos. 43–45, 60, 91–95, 166, 202, 215, 278–281, 284, 292, 297). •  H. H. Figulla & W. J. Martin, Letters and Documents of the Old-Baby­lonian Period, UET 5, London, 1953. Supplemented by Figulla 1953; Ormsby 1972 (texts reedited in Charpin 1986: 125–129); Loding 1976; Loding 1989. See now the two (partly complementary) books Spada 2007 and Black and Spada 2008. •  O. R. Gurney, Middle Babylonian Legal Documents and Other Texts, UET 7, London, 1974. The MB texts have been edited and commented in Gurney 1983. The book also contains numerous OB lexical and other school tablets. •  H. Figulla, Business Documents of the New Babylonian Period, UET 4, London, 1949. C.  Literary and Religious Texts •  C. J. Gadd & S. N. Kramer, Literary and Religious Texts. First Part, UET 6/1, London, 1963. •  C. J. Gadd & S. N. Kramer, Literary and Religious Texts. Second Part, UET 6/2, London, 1966. See now also Ludwig 2009. •  A. Shaffer, Literary and Religious Texts. Third Part, UET 6/3, London, 2006.

This splitting of the documentation was obviously unavoidable: no Assyriologist had anymore the possibility to edit texts from every period and every genre. But the drawback of such a publication policy is obvious: too little attention was paid to context. This was especially a pity for the Old Babylonian period. Contextual Approaches The final report on the Old Babylonian period had been prepared by Woolley before he died in 1960, with a supplement by Mallowan; it had been edited by Terence Mitchell in 1969. But for lack of money, the publication was delayed till 1976 (Woolley and Mallowan 1976): finally, thanks to the huge “catalogue of objects” (pp. 214–254), it was possible to reconstruct the findings according to their findspot. I began to investigate the matter as soon as I finished my PhD, in 1979 (see Charpin 1980b). I delivered a paper at the Rencontre in Paris in 1980, another one in Chicago in February 1983, where I received much support; just after, I collated tablets in Phil­adelphia—and I cannot fail to mention the unforgettable welcome I then received from Åke Sjöberg and his wife, and from the team of the PSD, especially Hermann Behrens. I also collated tablets at the British Museum and the late Jeremy Black collated for me tablets kept in the Iraqi Museum. Finally, my book entitled Le Clergé d’Ur appeared in 1986. It was the work of a cuneiformist with some experience of archaeology; its main defect is the fact it has been written in French…, which could explain why it is nowadays sometimes no more quoted—for instance by Harriet Crawford in her recent book on Ur. 21 At the same time, two young American archaeologists did the same kind of work about 21.  Crawford 2015; see the review in RA 110 (2016): 190–191.

Epigraphy of Ur: Past, Present, and Future

187

Nippur: Elizabeth Stone, whose book on Nippur Neighborhoods appeared one year later (Stone 1987), and Richard Zettler, who published articles on the Inanna temple since 1984 and finally his book in 1992 (Zettler 1992). In the three cases, the idea was to study the texts taking into account their archaeological contexts: this idea may look very basic today, but at that time it was something new in Assyriology, a lot of philologists being unaware of the importance of that matter. And the results were very rewarding. In the case of Ur, I was especially lucky with house No. 7 Quiet Street (EM site), the home of a family of priests, especially E-šuluhbi-uru and E-namti-sud; they proudly bore the title of abrig2 den-ki-eriduki-ga, stressing that their tradition as abriqqum-purification priests was linked to the god Enki-ofEridu. 22 And I can still remember how Sjöberg laughed, when I explained him—with some anxiety—that a text he published as a bilingual temple hymn (UET 6/1 117) was in fact…a school exercise were Sumerian names of priests were translated into Akkadian (Charpin 1986: 397–402). Legal documents found in the same house (such as UET 5 191, mentioning some of these names) 23 could give the key for the right interpretation of a supposedly literary text. A few years later, M. Van De Mieroop published a general synthesis about the city of Ur, with a special attention paid to economic history (Van De Mieroop 1992a), as well as a number of articles (Van De Mieroop 1989, 1992b, 1993). Unfortunately, the project he initiated at Columbia University, in collaboration with S. Richardson and S. Garfinkle, aiming to the publication of a book entitled Private Archives from Ur (Richardson 2005: 47 n. 22) has been given up. 24

Present and Future: Ur Old Babylonian Texts in ARCHIBAB Let us go now to the present. I would like to stress how the ARCHIBAB website is currently making efforts regarding the accessibility of Old Babylonian texts found at Ur. The main problem with the tablets discovered by Woolley is that most of them were published only as hand copies, with very short catalogue notices and indices. For the Ur III period, the problem is now solved thanks to the very useful website “BDTNS” of Manuel Molina. But a lot of work is still necessary for the Old Babylonian period.

Legal and Administrative Texts: The “Berlin Files” Thanks to Johannes Renger and Eva Cancik-Kirschbaum, an agreement has been made to have the thousands of transliterations of Old Babylonian legal and administrative texts made in Berlin available for the ARCHIBAB project. As far as Ur is concerned, the files were made by J. Renger and two former students and collaborators: •  Howard Farber, who prepared a dissertation at Chicago in the middle of the seventies, which was never finished; •  and Kilian Butz, who worked in Berlin with J. Renger and published some important articles (Butz 1973, 1978–9, 1979, 1980, and 1981). 22.  Cf. Charpin 1986: 392–395. 23.  See the edition and commentary in Charpin 1986: 85–88. 24.  Pers. comm. M. Van De Mieroop, New York, April 2013.

188

Dominique Charpin

It is possible through the ARCHIBAB website to have access to these provisory transliterations as pdf-documents. Little by little, they will be checked, provided with metadata, and lemmatised. I included already all the texts which were edited in my Clergé d’Ur (nearly one hundred texts).

Letters of UET 5: the “Leiden Team” Woolley was aware of the importance of the letters he discovered in the houses of the AH quarter. In his report of March 1, 1931, he wrote on page 3: 25 But the most important objects found in the houses are the tablets. We have now a collection of about seventy private letters in Sumerian (sic) dating to the latter half of the twentieth century B.C., found in different houses and giving information about their owners, a collection important enough to warrant a special volume in our publications of texts.

Finally, the letters—no less than eighty-three—have been included as the first section of UET 5; unfortunately, the hand copies of W. J. Martin were rather unreli­able, and this may explain why these letters did not attract very much attention—the correspondence of Ea-naṣir, the famous merchant, being the exception (Leemans 1960: 39–47). Thanks to the efforts of Marten Stol and Rients de Boer, all the letters from UET 5 are now fully edited (with an English translation) in the ARCHIBAB database and accessible on the web since April 2016 (Boer 2016).

Texts Put Online (Deriving from Woolley’s Excavations or Not) The rest of the OB tablets from Ur will be included in the near future, those deriving from the official diggings as well as tablets found in other ways. It seems that between 1854 and 1918, the local people excavated different buildings on the tell, finding archives for an amount of more than 200 tablets as far as the Old Babylonian period is concerned. In her foreword to YOS 5, E. Grice (1919: 13–14) wrote: In this large collection of texts of this period there is a group of one hundred and thirty-six which were found at Mugheir, or Muqayyar, on the site of the ancient city of Ur. The first one hundred and ten texts in the present volume have been selected from this group, and form the first considerable number of texts to be published from this site. […] The texts from Ur are nearly all official records from the principal temple of the city, Ê-nun-maḫ, the temple of the moon-god Nannar or Sin and his consort Ningal. […] Texts 1–110 are from Ur, according to both the testimony of those from whom they were bought, and the contents of the texts themselves.

Allusion is made here to the texts deriving from the storehouse of the temple (Ganun-mah). But other texts of the Yale Babylonian Collection found at Ur, published either in YOS 5, YOS 8, or YOS 12, 26 belong to private archives; they seem to have been found in what Woolley later called the “EH” and “EM” quarters (Charpin 1986: 158–192). 27 25. http://www.ur-online.org/media_item/205792/. 26.  And even YOS 14 311 (Charpin 1979: 200b): this text is close to UET 5 849 and 604 (cf. Butz 1973: 30). 27.  See Charpin 1986: 158 n. 2. Let us recall that the distinction between the two neighborhoods is due to the presence of the Neo-Babylonian “temenos” wall, which did not exist in the OB period (Charpin

Epigraphy of Ur: Past, Present, and Future

189

Numerous other tablets of this kind found their way in a great number of smaller collections. 28 The most interesting case has been published by W. Farber (2007): he found in Bremen a tablet (A 13120) originating from Ur, whose envelope, broken in two pieces, is preserved in Paris in the collection of the Ecole pratique (HE 316+317). 29 Identifying and editing all those tablets is very important, because they complement very often the information given by the texts discovered by Woolley. Let me come back to the example of the abriqqum-purification priests of No. 7 Quiet Street. Soon after the publication of my book on Le Clergé d’Ur, when collating tablets from YOS 12 at Yale, I had the surprise to discover on two administrative texts dated in the reign of Samsu-iluna the impressions of the seal of E-igidubi-isilim, also qualified as abrig2 den-ki-eriduki-ga. 30 This purification priest, colleague of E-šuluhbi-uru and E-namti-sud, figured thirty years earlier among the witnesses of UET 5 191 (RimSin 54). And we know thanks to a tablet kept in the Louvre (TCL 10 52 [Rim-Sin 21], see Charpin 1986: 178), that his father Nanna-addani was also an abriqqum: this demonstrates how such religious traditions were transmitted within families, generation after generation. This example shows that integrating the tablets found during illicit diggings is an important task to be done.

Present and Future: The New Excavations Back to the field! The new season of excavation of 2015 took place in the sector called AH by Woolley. 31 I will here give some information about the twenty-eight tablets that have been discovered. 32 When I arrived in the field, twelve were already dug up, but every day more were discovered, till the last day of the excavation…. So I was a busy and happy epigrapher!

Old Babylonian Period For the Old Babylonian period, four tablets, unfortunately not well preserved, were discovered: one tablet dated in Rim-Sin 24; one loan dated from Samsu-iluna; two lists of proper names.

Ur III Period The preservation of the Ur III tablets was much better, for a simple reason: the tablets were already baked in antiquity. This situation was by no mean accidental: the tablets had been deliberately baked. This observation (reiterated during the 2017 campaign for the three Ur III tablets discovered in Area 4) leads me to a hypothetical explanation for the preservation of so many Ur III tablets: during that period, tablets would have been often baked. If this remark has—as far as I 1986: 142). 28.  A list has been given in Charpin 1980a: 55–60. It has been increased by the texts published in 1979 in YOS 12 (see Charpin 1981); other additions have been made in Charpin 1986 (see the index p. 508–509). The list given by M. Van De Mieroop (1992: 8 n. 16) includes those additions without credit. 29.  First quotation by Scheil 1918; publication by Durand 1982, pl. 68. 30.  Charpin 1987. YOS 12 349 is dated 10/vi/Samsu-iluna 11, while the date of YOS 12 543 has been partly destroyed. E-igidubi-isilim is not mentionned in any of these tablets; his seal may have then been used by Muhegal (who could possibly be his son). 31.  For a first brief report, see Stone and Zimansky 2016. 32.  See already Charpin 2015.

190

Dominique Charpin

know—not been made till now, it could be for two reasons. First, for a long time, tablets have been baked in museums, and it has not been made explicit in the case of baked tablets if this was due to a modern treatment or not. 33 Second, a great number of Ur III tablets derive from illicit diggings, so that observations on their appearance when they were discovered have not been possible. If I am right, the Ur III administration did not recycle its tablets as later administrations did, because a lot of them were baked before being stored, and this could be the explanation for the great number of preserved tablets. 34 Four Ur III tablets were discovered in 2015: •  two tablets of gold accountings: one, dated Amar-Sin 9, pertains to more than one mina of red gold; 35 another one, dated Amar-Sin 8, is very similar to UET 3 316; 36 •  one copper ring weighing 21 1/2 minas (Šulgi 22); •  and a small fragment dated in Amar-Sin 9.

All the inscribed bricks discovered were already known. It is clear that till now there is nothing to make a revolution in Assyriology…. The more interesting find­ ings were made on the lower level.

Old Akkadian Period In the AH quarter, in “Niche Lane,” at approximately four meters deep, a building was reached where no more than eighteen tablets have till now been discov­ ered. None of them are dated, but their paleography leaves no doubt: they belong to the Old Akkadian period. This period till now was poorly represented at Ur: in the supplement to UE 2, only nine tablets are assigned to the Sargonic period (Alberti and Pomponio 1986). Between the fourth and twelfth of December, eighteen accounting tablets were discovered: silver accounting; lists of goods with indication of their price; account­ ing of wool, oil, grain, textiles; documents pertaining to agriculture (field surfaces with proper names); documents pertaining to livestock (accountings of cows); and finally some documents pertaining to fishing (accountings of fish, supplemented by account­ings of nets). 33.  In the case of Ur, let us recall that Woolley baked some tablets in the field before sending them for study to London. This is for example what he did for the OB tablets discovered in the “school” (= AH Broad street no. 1) according to page three of his report of March 1. 1931: “All had to be baked before cleaning, and this necessarily slow process has left little time for their study” (see above n. 25). But I found another indication weakening the above proposal, concerning the Ur III tablets discarded in the Dublamah: “We found a very large collection of unbaked clay tablets (pl. XLII, 2 and XLIII, 1) dating from the last years of the Third Dynasty of Ur; they lay five and six deep, closely packed together, just as they had fallen from the shelves where they had been stored; many, especially those of the upper layers nearest to the pavement bricks, were in very bad condition, reduced by infiltering moisture to the consistency of soft mud and impregnated with salts, and most had been broken or chipped in their fall; but many hundreds were saved and after being baked in a furnace and properly cleaned form a most interesting series of records” (Woolley 1925b: 392 [italics mine]). 34.  It has been proposed that the Ur III Inanna temple at Nippur contained a large tank-like structure used to recycle tablets; but this idea has been criticised by Taylor and Cartwright (2011: 301–303). 35.  A photo of the reverse has been published in Stone and Zimansky 2016: 254 fig. 15E (BDTNS 196360). Correct the date in Charpin 2015. 36.  A photo of the obverse has been published in Stone and Zimansky 2016: 254 fig. 15B (BDTNS 196359).

Epigraphy of Ur: Past, Present, and Future

191

This gives the feeling that we are dealing with the administration of some domain. Till now, there is no indication about the nature of this domain nor about the identity of its owner; but the excavation of the building is far from complete. The main interest of this discovery lies in the fact that the tablets were discovered in a well stratified context, with ceramics, and even two seal impressions.

Conclusion I would like to conclude with a remark about the importance of the collaboration on the field. The presence of epigraphers during excavations on such historical sites as Tell al-Muqayyer has been felt as a necessity since the beginning, and I am happy that E. Stone wished to continue that tradition. For the epigraphist, the clearance and cleaning of tablets are crucial operations to which it is best he can take part. Having seen the progress of the excavation, he can be more sensible about the contexts in which the texts were discovered. For the archaeologist, knowing immediately the date of each strata and the social background of the buildings occupants is gratifying. And the ensuing collaboration between archaeologists and epigraphers is even closer, when begun during the excavations. P.S. The 2017 season has been even more fruitful than the previous one: see Charpin 2017.

Bibliography Alberti, A. and Pomponio, F. 1986 Pre-Sargonic and Sargonic Texts from Ur Edited in UET 2, Supplement. Studia Pohl SM 13. Rome: Biblical Institute Press. Attia, A., Buisson, G., and Geller, M. J. 2009 Advances in Mesopotamian Medicine from Hammurabi to Hippocrates. Proceedings of the International Conference “Oeil malade et mauvais oeil”, Collège de France, Paris, 23rd June 2006. CM 37. Leiden: Brill. Benati, G. and Lecompte C. 2016 New Light on the Archaic Texts from Ur: The “Ancient Room” Tablet Hoard. Pp. 13–30 in Proceedings of the 9th International Congress on the Archaeology of the Ancient Near East, June 9–13, 2014, University of Basel. Volume 3: Reports, eds. R. Stucky, O. Kaelin, and H.-P. Mathys. Wiesbaden: Harrasowitz Verlag. Black, J. A. and Spada, G. 2008 Texts from Ur Kept in the Iraq Museum and in the British Museum. Nisaba 19. Messina: Di. Sc. A.M. Boer, R. de 2016 Old Babylonian Letters from UET 5 in the Archibab Database. NABU 2016/1: 11 no. 5. Butz, K. 1973 Konzentrationen wirtschaftlicher Macht im Königreich Larsa: Der NannaNingal-Tempelkomplex in Ur. WZKM 65/66: 1–58. 1978 Fischabgabe und Feldabgabe in Fischen und Vögeln in den Nanna-Tempel in Ur in altbabylonischer Zeit? Ein Versuch. AfO 26: 30–44.

192

Dominique Charpin

1979 Ur in altbabylonischer Zeit als Wirtschaftsfaktor. Pp. 257–409 in State and Temple Economy in the Ancient Near East I. Proceedings of the International Conference organized by the Katholieke Universiteit Leuven from the 10th to the 14th of April 1978, ed. E. Lipinski. OLA 5. Leuven: Departement Oriëntalistiek. 1980 Zwei Urkunden aus dem aB Ur, Frauenrecht betreffend. OrAnt 19: 29–35. 1981 Eine altbabylonisch Erbteilungsurkunde aus Ur, angeblich aus Larsa. OrAnt 20: 195–201. Charpin, D. 1980a Archives familiales et propriété privée en Babylonie ancienne: étude des documents de «Tell Sifr». HEO 12. Geneva-Paris: Librairie Droz. 1980b Note sur les archives des quartiers d’habitation d’Ur. Akkadica 16: 26–27. 1981 La Babylonie de Samsu-iluna à la lumière de nouveaux documents. BiOr 38: 517–547. 1986 Le Clergé d’Ur au siècle d’Hammurabi (XIXe-XVIIIe siècles av. J.-C.). HEO 22. Geneva-Paris: Librairie Droz. 1987 Notices prosopographiques, 1: une nouvelle famille d’abrig d’Enki-d’Eridu. NABU 1987/2: 19 no. 35. 2015 Découvertes épigraphiques à Ur (octobre-décembre 2015). NABU 2015/4: 178 no. 109. 2017 Nouvelles découvertes épigraphiques à Ur (2015 et 2017). Comptes rendus de l’Académie des Inscriptions et Belles-Lettres: 1063–1081. Crawford, H. 2015 Ur. The City of the Moon God. London and New York: Bloomsbury Academic. D’Agostino, F., Pomponio, F., and Laurito, R. 2004 Neo-Sumerian Texts from Ur in the British Museum. Nisaba 5. Messina: DiCAM. Durand, J.-M. 1982 Documents cunéiformes de la IVe Section de l’Ecole pratique des Hautes Etudes. Tome I. Catalogue et copies cunéiformes. HEO 18. Paris: Librairie Droz. Farber, W. 2007 Imgur-Sîn und seine beiden Söhne. Eine (nicht ganz) neue altbabylonische Erbteilungsurkunde aus Ur, gefunden wahrscheinlich in Larsa. Pp. 65–79 in Studies Presented to Robert D. Biggs. June 4, 2004. From the Workshop of the Chicago Assyrian Dictionary, Volume 2, eds. M. Roth, W. Farber, M. W. Stolper, and P. von Bechtolsheim. AS 27. Chicago: Oriental Institute of the University of Chicago. Faust, D. E. 1941 Contracts from Larsa Dated in the Reign of Rîm-Sin. YOS 8. New Haven: Yale University Press. Feigin, S. I. 1979 Legal and Administrative Texts of the Reign of Samsu-iluna. YOS 12. New Haven and London: Yale University Press. Figulla, H. H. 1953 Accounts Concerning Allocation of Provisions for Offerings in the Ningal-Temple at Ur. Iraq 15: 88–122; 171–192. Finkel, I. L. 2000 On Late Babylonian Medical Training. Pp. 137–223 in Wisdom, Gods and Literature. Studies in Assyriology in Honour of W. G. Lambert, eds. A. R. George and I. L. Finkel. Winona Lake, IN: Eisenbrauns. Gordon, C. H. 2000 A Scholar’s Odyssey. BSNA 20. Atlanta: Society of Biblical Literature. Grice, E. M. 1919 Records from Ur and Larsa Dated in the Larsa Dynasty. YOS 5. New Haven: Yale University Press.

Epigraphy of Ur: Past, Present, and Future

193

Gurney, O. R. 1983 The Middle Babylonian Legal and Economic Texts from Ur. Oxford: Alden Press. Hall, H. R. 1930 A Season’s Work at Ur, Al-‘Ubaid, Abu Sharain (Eridu), and Elsewhere Being an Unofficial Account of the British Museum Archaeological Mission to Babylonia, 1919. London: Methuen Co., Ltd. Jean, Ch.-F. 1931 Tell Sifr. Textes cunéiformes conservés au British Museum. Paris: Librairie Paul Geuthner. Kraus, F. R. 1955 Neue rechsturkunden der altbabylonischer Zeit. WO 2: 120–136. Lecompte, C. 2013 Archaic Tablets and Fragments from Ur (ATFU) from L. Woolley’s Excavations at the Royal Cemetery. Nisaba 25. Messina: DiCAM. 2015 Remarques sur les textes archaïques d’Ur. Akkadica 136: 141–156. Leemans, W. F. 1960 Foreign Trade in the Old Babylonian Period. SD 6. Leiden: Brill. Loding, D. 1976 Old Babylonian Texts from Ur I. JCS 28: 233–242. 1989 Two Old Babylonian Texts from Ur. Pp. 367–370 in DUMU-E2-DUB-BA-A, Studies in Honor of Å. W. Sjöberg, eds. H. Behrens, D. Loding, and M. T. Roth. OPSNKF 11. Philadelphia: Samuel Noah Kramer Fund. Loftus, W. K. 1857 Travels and Researches in Chaldaea and Susiana. London: Robert Carter Brothers. Ludwig, M.-C. 2009 Literarische Texte aus Ur. Kollationen und Kommentare zu UET 6/1–2. UAVA 9. Boston and Berlin: Walther de Gruyter. Mallowan, M. E. L. 1960 Memories of Ur. Iraq 22: 1–19. Ormsby, D. 1972 An Old Babylonian Business Archive of Historical Interest. JCS 24: 89–99. Richardson, S. 2005 Axes against Ešnunna. Or. NS 74: 42–50. Scheil V. 1918 Notules XLV. L’expression NU-ḤA.SA.ṢI. RA 15: 80–81. Sollberger, E. 1972 Mr. Taylor in Chaldaea. AnSt 22: 129–139. Spada, G. 2007 Testi economici da Ur di periodo paleo-babilonese. Nisaba 12. Messina: DiCAM. Stone, E. 1987 Nippur Neighborhoods. SAOC 44. Chicago: Oriental Institute of the University of Chicago. Stone, E. and Zimansky, P. 2016 Archaeology Returns to Ur. A New Dialog with Old Houses. NEA 79/4: 246–259. Strassmaier, J. N. 1882 Die altbabylonischen Verträge aus Warka. Abhandlungen und Vorträge des fünften Internationalen Orientalisten-Congresses. Berlin: A. Asher Co. Taylor, J. and Cartwright, C. 2011 The Making and Re-making of Clay Tablets, Scienze dell’Antichità. Storia Archeologia Antropologia 17: 297–324.

194

Dominique Charpin

Taylor, J. E. 1854 Notes on the Ruins of Muqeyer. JRAS 15: 260–276. Van De Mieroop, M. 1989 Gifts and Tithes to the Temples in Ur. Pp. 397–401 in DUMU-E2-DUB-BA-A, Studies in Honor of Å. W. Sjöberg, eds. H. Behrens, D. Loding, and M. T. Roth. OPSNKF 11. Philadelphia: Samuel Noah Kramer Fund. 1992a Society and Enterprise in Old Babylonian Ur. BBVO 12. Berlin: D. Reimer. 1992b Reed in the Old Babylonian Texts from Ur. BSA 6: 147–153. 1993 Sheep and Goat Herding According to the Old Babylonian Texts from Ur. BSA 7: 161–182. Weisberg, D. B. 1969 An Old Babylonian Forerunner to šumma ālu. HUCA 40: 87–104. Wiseman, D. J. 1981 Proceedings of the British Academy, 1981: 466. Woolley, C. L. 1923 Excavations at Ur of the Chaldees. AJ 3: 312–333 and pl. XXIV. 1924 Excavations at Tell el Obeid. AJ 4: 329–346. 1925a The Excavations at Ur, 1923–1924. AJ 5: 1–20. 1925b The Excavations at Ur, 1924–1925. AJ 5: 347–402. 1926 The Excavations at Ur, 1925–1926. AJ 6: 365–401. 1927 The Excavations at Ur, 1926–1927. AJ 7: 385–423. 1928 Excavations at Ur, 1927–1928. AJ 8: 415–448. 1929 Excavations at Ur 1928–1929. AJ 9: 305–343. 1930 Excavations at Ur, 1929–1930. AJ 10: 315–343 and pl. XXVIII. 1931 Excavations at Ur, 1930–1931. AJ 11: 343–381. 1932 Excavations at Ur, 1931–1932. AJ 12: 355–392 and pl. LVIII. Woolley, C. L. and Mallowan, M. 1976 The Old Babylonian Period (edited by T. C. Mitchell). UE 7. London: Trustees of the British Museum and of the University Museum, University of Pennsylvania, Philadelphia; British Museum Publications. Zettler, R. L. 1992 The Ur III Temple of Inanna at Nippur. BBVO 11. Berlin: D. Reimer.

Signs from Silence: Ur of the First Sumerians (Late Uruk Through ED I) Petr Charvát

University of West Bohemia at Plzeň and Charles University Prague The formation of “primary states” during the fourth millennium BCE (or Late Uruk period) represents a major epoch of deep social transformation within southern Mesopotamia. The onset of the third millennium BCE, then, offers evidence for the consolidation and stabilization of the pristine statehood born in the previous time period. The factors which exercised decisive influence on these fundamental processes do, however, remain largely unknown. The site of Tell al-Muqayyar, ancient Ur, in southern Iraq, is best known for its late Early Dynastic “Royal Cemetery” (ca. 2500–2350 BCE) but it had already been a political centre at the onset of the third millennium (Early Dynastic I period, ca. 2900–2700 BCE). Due to the rich archaeological evidence at our disposal, Ur is an ideal test case for analysis of the the freshly formed Mesopotamian statehood for this specific time period. Here I am addressing the internal economic and political organization, as well as spiritual orientation and achievement, of archaic Ur. The interplay between economic and socio-cultural actions are analysed here on the basis of three main lines of evidence: archaeology, written sources, and iconographic data. Excavating within the central sacred precinct of the city, the Leonard Woolley expedition (1922–1934) explored also the archaeological strata below the famous “Royal Cemetery,” but above the so-called “Jemdet Nasr cemetery” (see below). These strata have yielded alternate layers of both domestic and administrative refuse including numerous find groups of seal impressions on clay, referred to as “Seal-impression strata” (SIS). 1 The seal-impression bearing strata constitute a most useful source category for the investigation of economic, administrative, social, and spiritual structures streamlining the life of the major successor state of the Uruk-age culture. The anal­ ysis of the images borne by the sealed surfaces will give fairly instructive data for understanding the symbolic messages of the relevant seals. Inscribed seal impressions excavated from reasonably well-dated contexts will offer a historian the best possible material for studies of the socially engineered goods-exchange practices available. On the other hand, impressions which the sealed objects left behind on

1.  On which see Woolley in Legrain 1936: 1–2, and Woolley 1955: passim; for recent revisions of the situation see Sürenhagen 1999; Dittmann 2006; Benati 2015.

195

196

Petr Charvát

the reverses of the clay sealings will reveal the type of seal carrier, and thus provide first-rate evidence of social and administrative practices of the relevant period. These seal impressions link up with remains of a series of houses, of which Woolley has suggested that they may even have been store-rooms of a temporary nature (Woolley 1955: 71). Stratigraphic evidence does, however, imply that during the first quarter of the third millennium, the relevant part of the Ur mound, i.e. the SE slope of the temenos-area enclosure, assumed the form of an urban area with courtyard houses, alley­ ways and a thick boundary wall (cf. Woolley 1956: pl. 73). The adjacent landscape southeast from here, below the SE edge of the temenos-area enclosure—occupied by steeply sloping rubbish heaps—remained in use as a waste-disposal area for a long period. Garbage deposits covered here a low-lying area used as burial ground since the late fourth millennium (the so-called “Jemdet Nasr Cemetery”). The debris of SIS 8–4 formed a slope following a northwest/southeast direction, suggesting that the dumping originated from the area lying immediately to the back of Woolley’s Pit Z. Repeated phases of garbage deposition gradually levelled the gap between the terrace and the lower ground, and formed a rubbish heap in the area comprised between Woolley’s Pits Z-Y, W and X (Benati 2015: 2–4). Chronologically speaking, Giacomo Benati places the relevant stratigraphic sequence (SIS 8–4) into the earlier segment of the period comprised between, by and large, 2,900 and 2,700 BCE (Benati 2015: 3, Table 1). In the descriptions and evaluations of the reverse traces, and especially imprints of cords and ties of all kinds, I am following the procedure proposed by Roger Matthews (Martin-Matthews 1993, esp. p. 37; Matthews 1993: 44–46). Let us now pass in review the particular development phases of the earliest city of Ur within the state-formation phase of ancient Mesopotamia.

Phase I: Late Uruk Age (Before 3,100 BCE) 2 In this time period, Ur probably represented a local or regional central place with evidence for material production. In Woolley´s Pit F—located to the back of the Royal Cemetery (henceforth RC) area—a pottery production area with kilns, in use during the late fourth millennium, came to light (Benati 2015: 2). This “Great Sherd Dump” might be characterized as a production area with a sequence of layers of vessel fragments, pottery-production wasters, kiln facilities and tools of various kind, several metres thick and beginning in (at the end of?) the Ubaid culture. The dump went out of use sometime in the Jemdet Nasr period, that is, around 3000 BCE. (Sürenhagen 1999, esp. pp. 188–208; Dittmann 2006: passim, esp. Table 1 on p. 36). A cylinder seal found in Pit F, level 7.80 cm (Dittmann 2006: 29 = “ausgehende Späturuk-Zeit”; see Sürenhagen 1999: 200, 286 sub no. 79, Taf. 57: 7) may, in fact, depict women weavers, hinting thus at a possibility of the presence of textile production in coeval Ur. In addition to this, a quantity of terra-cotta cones, used in Late Uruk as architectural decoration, came to light during Woolley’s excavations of his Archaic II–III layers (University of Pennsylvania Museum of Archaeology and Anthropol­ 2.  See Benati 2015: 3, Table 1.

Signs from Silence: Ur of the First Sumerians

197

ogy, Photographic Archives, photo No. 2018). A considerable quantity of these are now deposited in the University of Pennsylvania Museum of Archaeology and Anthropology (UM 87-28-26). Collections of the same museum also contain a pottery group designated “Found with Ur wall cones” (UM 92-4-16) featuring items datable between the Ubaid period through to Parthian/Sasanian times.

Phase II: Jemdet Nasr Age (ca. 3,100–2,900 BCE) 3 Here the crucial evidence is supplied by one of Woolley’s soundings, namely “Pit G, 7 to 7. m (alt. 7.10–8 m) below brick pavement dated c. 3100 BC in the prehistoric settlement.” 4 The summary inventories of the cultural layer (?) characterizing this “prehistoric settlement” do include an artefact category with a rather short circulation period. These are the “clay copies of shell bugle beads.” Roger Matthews (2002: 31, fig. 54: 8–31; pl. 43) refers to them as to “baked clay beads with distinctive spiral grooving along the length,” and notes that they appear to be restricted to the Jemdet Nasr period. An even more interesting piece of information can be culled from the proto-­ cuneiform sign which two of the sealings found here bear (UM 31-17-352B and UM 31-17-352R). The ŠAM2 sign is usually translated as “to buy,” “to pay,” “to sell.” 5 This evidence for trade, offered by the earliest attestations of the use of writing at archaic Ur, finds an echo in archaeological materials retrieved from the site. The same find-spot as that of the sealings has yielded finds of obsidian (Quenet 2008: 161). The graves of the “Jemdet Nasr cemetery” at Ur brought forth a quantity of lapis lazuli products, as well as those of chalcedony and chalcedony-related min­ erals, carnelian and chlorite (Sürenhagen 1999, esp. pg. 237, Tab. 52; Quenet 2008: 97, 135, 132–137, 143). At that time period, Ur must have served—on the testimony of cylinder-seal style—as a dependency of Uruk. It probably played the role of a port of trade opened chiefly on the Persian Gulf, where selling and buying represented a component of administrative goings-on registered in writing.

Phase III: Earlier Part of ED I (ca. 2,900–2,700 BCE) 6 At this age, Ur grew into a self-sustaining economic system in which a set of receptors conveyed goods in sealed containers into the centre where they were unsealed. What went as compensation back to the suppliers cannot be judged from the archaeological record. A number of the inscribed sealings name dignitaries and institutions, or denote deliveries of comestibles (Charvát 2010: 40–43). Out of the total of twenty-nine available seal impressions, twelve stem from movable objects and nine from immovable structures. In three cases the same seal closed movable and immovable items, and the sealing carrier cannot be identified in five finds. 3.  See Benati 2015: 3, Table 1. 4.  See Legrain 1936: 42–43, apud nos. 480, 481 and 483; for the situation of Pit G in the southern corner of the Ur temenos see Woolley 1955: pl. 1, and Sürenhagen 1999: 178; see also Charvát 2014. 5.  See http://psd.museum.upenn.edu/epsd/nepsd-frame.html s. v. sa (accessed April 25, 2012). 6.  See Benati 2015: 3, Table 1.

198

Petr Charvát

As to the character of the pristine cylinder seals of early Ur, introduced by this date for the first time, it has been observed that: • first, the SIS seals and sealings constitute a stylistic unity, in spite of the stratigraphic distinction between the SIS 8–6 and SIS 5–4 groups, and • second, that this group of artistic creations possesses all the stylistic traits of the incipient, or, at any rate, early Early Dynastic I period (Scott 2005, esp. pp. 52, 63; Otto 2010, esp. p. 26). If the SIS material displays the stylistic unity commented on above, then the conclusion that it represents an outcome of one single artistic centre, or a glyptic workshop, is possible. As the activities of this glyptic workshop apparently followed out of the designing, and building up, of the entire administrative apparatus of— most probably—the central office of post-3000 Ur, we shall probably not miss the point too widely if we assume that the stylistic unity of the SIS sealings ensued out of a “contrat social,” of a commission by one single organ of the central administration of Nannar´s city from one single glyptic workshop. This gives us the chance to see in the SIS-seal imagery the manifestation of ideational strategy of one single governance centre. At this time, Ur acquired the full paraphernalia of the culture of writing, including a scriptorium and an archive. Witness the sealing UE 3 1 with a series of official titles and basket impressions on reverse, probably a pisandubba (Charvát 2010: 40). The local writing school fully respected the usages of the Late Uruk lit­ eracy, with an exception of inscriptions carved in cylinder seals, an innovation as against the previous time.

Phase IV: Later Part of ED I (ca. 2,900–2,700 BCE) 7 Ur has by then become a major centre of production. Both field management and livestock-keeping showed an effort at a more intense grasp of the resources at hand, rationalization of the production procedures such as shortening of the circuits of goods circulation, for instance. The registrars of archaic Ur seem to have noted down arrears in payment. Technological progress may be discerned. For instance, for the first time in Sumerian history, the scribes of the texts of archaic Ur distin­ guished between copper (URUDU) and bronze (zabar = UD+KA+BAR) (UET 2 373; Moorey 1985: 127–128; on URUDU now Marchesi 2011: 191, sub no. 5). Large quantities of commodities were gathered in storage spaces for redistribution. A part of the goods redistributed possibly went to the Inanna shrine(s) at Uruk. The managers of the city of Nannar followed now a fixed administrative procedure, during which the sealings from opened storage spaces went to a control(?) procedure by the central office, and then flew to the rubbish heaps every season (Charvát 2010: 43 fn. 57; Charvát 2013: 635–636). A number of archaic Ur sealings bore, superimposed over the first cylinder-seal impressions, additional prints of stamp seals(?); I refer to this procedure as “counter-marking.” It is to be noted that we know archaic Ur seals which, though repeatedly impressed, never received counter-marks. Counter-marked versus non-counter-marked sealings outline two circulation areas of the supplies—the counter-marked “outer sector,” pertaining possibly to the “City League” matters and 7.  See Benati 2015: 3, Table 1.

Signs from Silence: Ur of the First Sumerians

199

cultic affairs, and the non-counter-marked “inner sector,” which might have linked up with one of the local Ur offices, possibly that of the LUGAL, or, alternatively, of the Nannar temple or the city of Ur. The “outer sector” shows a multitude of counter-marked sealings with seals repeated infrequently; does this represent a shifting network of political alliances? The fact that most of the counter-marking seals bear the rosette symbol lends support to Piotr Steinkeller’s assumption that a connection with the cult of Inanna at Uruk might exist (Steinkeller 2002b), especially in connection with the new text from archaic Ur referring to personnel sent to Uruk from there (Benati 2015: 11–12). However, the Ur archaic text UET 2 51, a ration list, bears on its obverse a depiction of a rosette and, in addition to other marks representing possibly the grand total, the toponym URI5. This may indicate that the rosette stood for the city of Ur, and counter-marking seals bearing this emblem denoted that share in the “City League” (or other) consignments which fell to the abode of Nannar. The question needs further study. As against this, the non-countermarked “inner sector” displays few seals, but repeated many times (up to thirty-two), pointing to more rigorous exploitation of the sources concerned. The bearer(s) of the “City League” seal visited the member communities, collected contributions in kind and confirmed their receipt by sealed documents (Steinkeller 2002a: 256). Such contributions consisted of usually standardized quantities of figs, apples, wine (grapes, raisins?), and GA2 + GEŠTUG = /adaku’a/, which seems to have been a fish product (Steinkeller 2002a: 252). The “City League” might have borne the name of EREŠ2, denoting possibly the ki-en-gi confederation (UE 3 431; Visicato 1989: ereš2ki = ŠE.NAGA, identical with the Fara domain) and its affairs could be managed by an official with the NAMEŠDA title (see Charvát 2012). It also wielded an elementary administrative apparatus (chancery with an official seal or official seals). A varied social and political structure is revealed by the “City League” seals. The “City League” possibly collected taxes: the sealing Matthews 1993: no. 14 refers to IL2 (Charvát 2010: 53). A handwritten gloss on sealing UE 3 114 refers possibly to bows and daggers or swords, hinting at a possibility of the existence of an arsenal, and so perhaps indicating the possibility that the “City League” could wage war (Charvát 2010: 48). We know next to nothing about any “foreign policy” of the “City League,” but one of its sealings turned up at the rather remote Konar Sandal, southern Iran (Matthews 2013). The “City League” appears to have been involved in the UR2 religious ceremony (UE 3 402). 8 The “City League” seals of this time adhered firmly to the writing usages of Late Uruk age, and the vivacity of the tradition is borne out by the penetration of singular signs of writing into the pictorial compositions of an-epigraphic seals. In these cases, the signs referred either to the items depicted, or possibly to the purpose of delivery of the goods consignment carrying the relevant seals. Sometimes they may even describe the procedure shown on the seals 9 or objects depicted therein (flour or cords in an offering scene UE 3 323). 10 In this they differ from the 8.  See Charvát 2010: 61; on the UR2 ceremony see Charvát 2010: 63–64. 9.  See BAD = http://psd.museum.upenn.edu/epsd/nepsd-frame.html (accessed September 27, 2014) s. v. bad = be open, thresh grain in UE 3 315 and 325. 10.  See http://psd.museum.upenn.edu/epsd/nepsd-frame.html (accessed September 28, 2014) s. v. eš2 = flour, signs under the spread eagle, right of the double-handled vessel.

200

Petr Charvát

subsequent Fara/Šuruppak seals, in which the cuneiform signs spell out personal names. As against this, the central office of Ur itself, possibly that of the LUGAL, already used fully syllabic writing. This is not to say that its scribes would avoid using traditional, Late-Uruk terminology. A case in point is the še mú = “(Feld) mit Gerste bestanden,” turning up in text UET 2 160: 1, and UET 2 359: 1, with a parallel in an earlier text VAT 17785+119 (Bauer 2001–2002: 173). Giacomo Benati (2015: 8) observes that the metrology of the archaic Ur texts follows Late Uruk textual usage. The output of the archaic Ur scriptorium included thirty school tablets, and forty others bearing only seal impressions. There were also 290 records of administrative character including those tracing transfers of cereals (11%), transactions in land (23%), rations of bread, flour and beer to temple(?) employees (27%), personnel matters (6%), sheep-rearing texts (6%), registrations of movement of such materials as wood, metal, leather, and reeds (3%), and 23% of texts of undetermined character (Visicato 2000: 14–18). Some kind of literary tradition must have been cultivated at archaic Ur, as is shown by the “Götterlisten” of Fara, incorporating elements of the Ur pantheon into a set of deities probably of traditional, Late Uruk-Jemdet Nasr character (Steible-Yildiz 2008: 185 s. v. dzu-en). Four or five examples of the LU2 A list from the archaic Ur, apparently modernized as against the Uruk predecessor, bear witness to the scientific output of the local intellectual workshop. In this case the Ur scribes probably established a tradition, as the later versions of this list point rather to Ur than to an Uruk archetype (Lecompte 2013: 148–149). However, materials of both bureaux ended up in the same archaeological context (SIS 5 and 4). Who was the LUGAL of early Ur? The evidence of personal names in the texts of archaic Ur makes the LUGAL first and foremost a provider of plenty, a guarantee of activation of forces of fertility and fecundity of the nature and its living components including people (Andersson 2012: 134–137). The LUGAL of Ur seems to have been primarily a temple official, chosen possibly by a divine omen (Andersson 2012: 246), and, much as the moon god, was ritually re-born every month (Andersson 2012: 172). The EN of Ur served the community in the city temple as a faithful consort and servant of Nannar, the moon god (see Weadock 1975; Lion 2009; Westenholz 2013). As to the Ur temple, a hoard of possibly disused cultic inventory went into the earth in the courtyard of the temple precinct close to its southwestern corner, between the ziggurat and the N complex (Tunca 1984: 266). This included stamp seals in the traditional style, animal and human statuettes and beads, as well as an object resembling an “eye idol,” omitted by Tunca but mentioned (though not depicted) by Leonard Woolley (1955: 43, 187, U.17836, pl. 15). This indicates a break with earlier religious symbolism, but when did it occur is most difficult to say. The personal name contained in text UET 2 281 ii 5, read by Gianni Marchesi (2004: 195–197) as pabilgax-mes-utu-pà-da = “Gilgameš is the one whom Utu has selected,” shows that the bilateral and “telescoped” kinship system known from later Mesopotamian texts, with pabilgax as father’s father or father’s brother (Götzelt 1995, esp. pp. 179–180), is likely to have structured (at least some of) the broad strata of Ur society as early as this time. Works of art demonstrating the creativity of the Ur masters are represented by a statuette of a wild boar which might have once decorated a standard or a piece of

Signs from Silence: Ur of the First Sumerians

201

furniture (Woolley 1955: 31, 67, pl. 37: U.14459; Sürenhagen 1999: 200, Taf. 57: 5). The hooves of a bull statue, mounted perhaps originally on a pedestal, turned up probably in one of the SIS levels (4–5?; Sürenhagen 1999: 186 no. 8 = Woolley 1955: 38, pl. 29, U.14462). This statue represented a superior-quality product, implying external procurement of rare raw materials, skilled craftsmen, and complex casting activities in this phase at Ur (Benati 2015: 20).

Acknowledgments The material is published herewith with kind consent of the Near Eastern Section of the University Museum of Archaeology and Anthropology, University of Pennsylvania, Philadelphia, PA, USA. I could not complete the research presented here without support of several academic sponsoring bodies. The academic year 2003–2004 I spent in Philadelphia thanks to a research grant from the Prague office of the John William Fulbright Foundation (grant No. 2003-28-02, Fulbright No. ME659). In 2005, I could pursue my goals further with the aid of the American Philosophical Society (grant No. Franklin 2005), as well as of the Grant Agency of the Academy of Sciences of the Czech Republic at Prague (grant No. A8021401). No less helpful was another grant project conferred on me jointly by the Grant Agency of the Czech Republic and the Deutsche Forschungsgemeinschaft (DFG) under the No. (GA CR) 404/08/J013. I am deeply obliged to Holly Pittman, Curator of the Near Eastern Section of the said University Museum of Archaeology and Anthropology, as well as to Richard Zettler and Shannon White of the Near Eastern Section of the said University Museum of Archaeology and Anthropology, for their friendly assistance. I have carried out the research for this article also thanks to a generous annual grant by the Internationales Kolleg MORPHOMATA of the University of Cologne, Germany; I wish to declare my debt of gratitude to MORPHOMATA, especially in the persons of its both Directors, Dietrich Boschung and Günther Blamberger.

Bibliography Andersson, J. 2012 Kingship in the Early Mesopotamian Onomasticon 2800–2000 BCE. Acta Universitatis Uppsaliensis, Studia Semitica Uppsaliensia 28. Uppsala: Uppsala Universitet. Bauer, J. 2001/2 Review of ATU 5. AfO 48–49: 168–173. Benati, G. 2015 Re-modeling Political Economy in Early 3rd Millennium BC Mesopotamia: Patterns of Socio-Economic Organization in Archaic Ur (Tell al-Muqayyar, Iraq). Cuneiform Digital Library Journal 2015:2: 1–37. Burrows, E. 1935 Archaic Texts. UE 2. London and Philadelphia: Trustees of the British Museum and of the University Museum, University of Pennsylvania, Philadelphia. Charvát, P. 2010 Inscriptions on Sealings from Archaic Ur. Pp. 39–74 in Shepherds of the Blackheaded People: The Royal Office Vis-à-vis Godhead in Ancient Mesopotamia, eds. K. Šašková, L. Pecha, and P. Charvát. Plzeň: Západočeská univerzita.

202

Petr Charvát

2012 From King to God: The NAMEŠDA Title in Archaic Ur. Pp. 265–274 in Organization, Representation, and Symbols of Power in the Ancient Near East, Proceedings of the 54th Rencontre Assyriologique Internationale at Würzburg, 20–25 July 2008, ed. G. Wilhelm. RAI 54. Winona Lake, IN: Eisenbrauns. 2013 Three Kings of the Orient in Archaic Ur. Pp. 635–644 in Time and History in the Ancient Near East, Proceedings of the 56th Rencontre Assyriologique Internationale at Barcelona, 26–30 July 2010, eds. L. Feliu, J. Llop, A. Millet Albà, and J. Sanmartín. Winona Lake, IN: Eisenbrauns. 2014 The Deep Well of the Past: Earliest Seal Impressions from the City of Ur. Pp. 603–612 in Proceedings of the 8th International Congress on the Archaeology of the Ancient Near East 30 April– 4 May 2012, University of Warsaw Volume 1, Plenary Sessions, Township and Villages, High and Low: The Minor Arts for the Elite and for the Populace, eds. P. Bieliński, M. Gawlikowski, R. Koliński, D. Ławecka, A. Sołtysiak, and Z. Wygnańska. Wiesbaden: Otto Harrassowitz GmbH & Co. Dittman, R. 2006 Bemerkungen zu Ur, Pit F. BaM 37: 23–44. Götzelt, T. 1995 Zur sumerischen und akkadischen Verwandschaftsterminologie. Pp. 177–182 in Beiträge zur Kulturgeschichte Vorderasiens: Festschrift für Rainer Michael Boehmer, eds. U. Finkbeiner, R. Dittmann, and H. Hauptmann. Mainz: Verlag Philipp von Zabern. Lecompte, C. 2013 Temps, mémoire et évolution des cultures aux époques archaiques: écriture du passé et listes lexicales. Pp. 143–158 in Time and History in the Ancient Near East: Proceedings of the 56th Rencontre Assyriologique Internationale at Barcelona, 26–30 July 2010, eds. L. Feliu, J. Llop, A. Millet Albà, and J. Sanmartín. Winona Lake, IN: Eisenbrauns. Legrain, L. 1936 Archaic Seal Impressions. UE 3. Oxford: Trustees of the British Museum and of the University Museum, University of Pennsylvania, Philadelphia; Oxford University Press. Lion, B. 2009 Sexe et Genre (2): Des prêtresses fils de roi. Pp. 165–182 in Topoi Orient Occident: Supplément 10, Femmes, cultures et sociétés dans les civilisations méditerranéennes et proche-orientales de l’Antiquité, eds. F. Briquel-Chatonnet, S. Farès, B. Lion, and C. Michel. Lyon: Maison de l´Orient Méditerranéen–Jean Pouilloux (Institut Fernand-Courby). Marchesi, G. 2004 Who Was Buried in the Royal Tombs of Ur? The Epigraphic and Textual Data. Or. NS 73/2: 153–197. 2011 Goods from the Queen of Dilmun. Pp. 189–200 in Akkade is King: A Collection of Papers by Friends and Colleagues Presented to Aage Westenholz on the Occasion of His 70th Birthday 15th May 2009, eds. J. G. Dercksen, J. Eidem, K. van der Toorn, and K. R. Veenhof. PIHANS 118. Leiden: Nederlands Instituut van het Nabije Oosten. Martin H. P. and Matthews, R. J. 1993 Seals and Sealings. Pp. 23–81 in Abu Salabikh Excavations Volume 4. The 6G Ash-Tip and its Contents: Cultic and Administrative Discard from the Temple?, ed. A. Green. London: British School of Archaeology in Iraq.

Signs from Silence: Ur of the First Sumerians

203

Matthews, R. J. 1993 Cities, Seals and Writing: Archaic Seal Impressions from Jemdet Nasr and Ur. MSVO 2. Berlin: Gebr. Mann Verlag. 2002 Secrets of the Dark Mound: Jemdet Nasr 1926–1928. Iraq Archaeological Reports No. 6. Warminster: British School of Archaeology in Iraq. 2013 “Archaische Siegel” aus Mesopotamien und Iran. Pp. 146–147 in Uruk: 5000 Jahre Megacity, eds. N. Crüsemann, M. van Ess, M. Hilgert, and B. Salje. Petersberg: Michael Imhoff Verlag. Moorey, P. R. S. 1985 Materials and Manufacture in Ancient Mesopotamia: The Evidence of Archaeology and Art. BAR International Series 237. Oxford: BAR Publications. Otto, A. 2010 The Iconography and Use of the Ur SIS 8–4 Seals Reconsidered. Pp. 19–29 in Shepherds of the Black-headed People: The Royal Office Vis-à-vis Godhead in Ancient Mesopotamia, eds. K. Šašková, L. Pecha, and P. Charvát. Plzeň: Západočeská univerzita. Quenet, P. 2008 Les Échanges du nord de la Mésopotamie avec ses voisins proche-orientaux au IIIe millénaire (ca. 3100–2300 av. J. C.). Subartu 22. Turnhout: Brepols. Scott, S. J. 2005 Figure, Symbol and Sign: Semiotics and Function of Early Dynastic I Cylinder Seal Imagery from Ur. Ph.D. dissertation. University of Pennsylvania. Steible, H. and Yildiz, F. 2008 Die Listen mit Abgaben an Götter aus Šuruppak. Mitteilungen der Deutschen Orient-Gesellschaft zu Berlin 140: 151–204. Steinkeller, P. 2002a Archaic City Seals and the Question of Early Babylonian Unity. Pp. 249–257 in Riches Hidden in Secret Places, Ancient Near Eastern Studies in Memory of Thorkild Jacobsen, ed. T. Abusch. Winona Lake, IN: Eisenbrauns. 2002b More on the Archaic City Seals. NABU 2002/2: 29 no. 30. 2013 An Archaic “Prisoner Plaque” from Kiš. RA 107: 131–157. Sürenhagen, D. 1999 Untersuchungen zur relativen Chronologie Babyloniens und angrenzender Gebiete von der ausgehenden ´Ubaidzeit bis zum Beginn der Frühdynastisch II-Zeit 1, Studien zur Chronostratigraphie der südbabylonischen Stadtruinen von Uruk und Ur. Heidelberg: Heidelberger Orientverlag. Tunca, Ö. 1984 L’Architecture religieuse protodynastique en Mésopotamie. Akkadica Supplementum 2. Leeuven: Peeters. Visicato, G. 1989 Fara ed Ebla: nuove prospettive storico-politiche nel periodo protodinastico. OrAnt 28: 169–176. 2000 The Power and the Writing: The Early Scribes of Mesopotamia. Bethesda, MD: CDL Press. Weadock, P. N. 1975 The Giparu at Ur. Iraq 37: 101–128. Westenholz, J. G. 2013 In the Service of the Gods: The Ministering Clergy. Pp. 246–274 in The Sumerian World, ed. H. Crawford. London and New York: Routledge, Taylor and Francis Group.

204

Petr Charvát

Woolley, C. L. 1955 The Early Periods: A Report on the Sites and Objects Prior in Date to the Third Dynasty of Ur Discovered in the Course of the Excavations. UE 4. London: Trustees of the British Museum and of the University Museum, University of Pennsylvania, Philadelphia.

Breakers and Enforcers of the Oath of the King Laura Culbertson

American Public University

Introduction On the occasion of a celebration of Ur, this short paper considers aspects of the projection of royal authority into judicial activities outside the royal sphere during the Ur III period (2112–2004 BCE). 1 Of course the dearth of documentation from the capital Ur itself poses an immediate challenge to this objective, as the king’s own daily judicial activities are largely “invisible” to us. Nevertheless, royal power and authority can be detected from at least two ways. First, we get an impression of royal interest in provincial judicial affairs from evidence of numerous state officials dispatched around the provinces to attend judicial proceedings. But more prevalently, we can note the idea of the king’s authority in the regular and ubiquitous invocation of the king’s name in oaths. With these two possible ways of examining royal authority in judicial contexts, this study aims to add to previous inquiries on the political contexts of the mu lugal oath—the oath of the “the name of the king”—and to revisit what this can tell us about the extent and limitation of royal authority in provincial contexts. Across ancient cultures, taking an oath subjected the swearer to higher temporal or divine powers and involved self-sanctioning with curses. To take an oath calling on the name of a king, then, is to invoke his jurisdiction and authority as the enforcer of the oath. In one of the first compilations of such oaths, Mercer wrote that invoking the name of the king was done with the “purpose of making the oath more solemn and binding, and, learning from experience that oaths were sometimes broken, to guarantee its preservation” (1913: 38). Thus, the invocation of the king had symbolic and pragmatic aspects. But it has also been noted that the oath of the king in Sargonic and Ur III times may have reflected real political forces and central­ization, such as the “suppression of local governments” (Steinkeller 1989: 79), and by implication the transposition of royal authority onto local settings. And

1.  I wish to thank Steven Garfinkle, Xiaoli Ouyang, and Manuel Molina for organizing the workshop at the Philadelphia RAI in 2016, and Garfinkle for his guidance on the topic. Abbreviations throughout follow the indispensable Database of Neo-Sumerian Texts (BDTNS), bdtns.filol.csic.es.

205

206

Laura Culbertson

it has been argued that invoking the king went beyond a symbolic gesture by establishing that one would appear before the king upon breaching the oath. 2 The ideology of the king as a judicial authority is undeniable, but on a practical level, the question is here asked: who specifically enforced oaths that invoked the king’s name when they were breached? Thanks to studies on judicial officials and the databases of Sumerian texts, we can apply a contextualized approach to the oaths, noting specifically who enforced them when they were not fulfilled. Enforcing the oath of the king in court involved many political elites of various statuses and affiliations. The context-seeking approach may also allow us to identify aspects of how enforcement was practiced and whether the enforcement of oaths amounted to a unification of legal practices and norms. The latter topics are interesting to consider in light of the long trend of questioning the notion of ancient states as absolutist, regulatory regimes and the efforts to understand the political dynamics of the Ur III state in particular. 3 The first four sections attempt to build a context before getting to examples of unfulfilled promises sworn by the name of the king.

Promissory Oath Terminology and Purpose, Briefly The philology and grammar of oaths in Ur III sources, their positioning in judicial procedures, and the ritual aspects of oath-taking have already been well treated in many studies, so only a brief synopsis is needed here. 4 It is conventional to differentiate the two main idioms used in Ur III judicial proceedings, although there is debate over the basis of difference. First, the promissory oath was sworn in the name of the king or sometimes by the “life of the king” (zi lugal). Taking the oath is represented with the formula mu lugal-bi in-pad3, but in various cases the name of the king (mu lugal) is recorded in a statement without the full formula. Second, the nam-erim2 idiom, when employed in the course of judicial proceedings, is considered a testamentary oath and self-curse (also called the assertative, declaratory, or evidentiary oath). 5 Both idioms related to the act of enshrining a statement as 2.  Kitz (2013: 47) highlighted TMH NF 1–2 69 to show how the oath represented intent to submit to royal jurisdiction. 3.  Recently Richardson (2014: 75) wrote: “Ancient states were not legally and spatially integrated; they were networks of local and professional actors with closely circumscribed goals and interests.” Studies on Ur III politics, administration, and state-building have been interested in the relationships between networks and individual actors, and how individuals operated within the state. Garfinkle (2008: 154) suggested that state-building required “cooperation with local elites whose longstanding authority within their communities was co-opted by the crown.” And Michalowski (2008: 186) wrote: “The structure of power and control [in the Ur III state] was less monolithic than is often assumed, and the various overlapping networks, in order to function properly, were often fluid, with uneven concentrations and distribution of authority.” These and the Ur III studies cited throughout prompted the question of how the authority to protect and adjudicate disputed oaths of the king would have been distributed in such a setting. 4. Studies on the Sumerian oaths include (by no means exhaustive): Edzard 1975; Falkenstein 1956–57: 63–66; B. Lafont 1996; Mercer 1913; Sallaberger 2008 and 2015b; Steinkeller 1989: 71–79; and Yoda 1993. Other useful studies of oaths in legal and political contexts include the contributions in S. Lafont 1996; Wells 2008; Wells, Magdalene, and Wunsch’s 2010 and 2012 studies; the essays in Barta, Lang, and Rollinger 2015; and Kitz’s (2013) work on curses across the Near East, which includes a study on Sumerian oaths (38ff). 5.  As Kitz (2013: 39) pointed out, the differentiation of oaths between promissory and evidentiary function is a “convenient, modern construct.”

Breakers and Enforcers of the Oath of the King

207

true, whether it concerned a past, present, or future matter as in the case of promises. Both could be undertaken voluntarily, but higher authorities such as judicial officials or managers could prompt or coerce an individual to swear. And both are characterized by a reliance on a higher jurisdiction, as in most third millennium oaths (Feder 2010: 127). As mentioned, the above-cited scholars differ on how to classify the oaths, with some suggesting the difference lies in timing (past versus future) and others pinpointing the matter to a difference in jurisdiction. The former idiom appealed to the temporal authority of the king (who of course was divinized in this period) and the latter provoked the divine. 6 Kitz (2013: 41) surveyed the arguments and suggested classifying the oaths “according to context and not necessarily phraseology.”

The Many Contexts of Oath-Secured Promises It is well known that “oath-based political relations” 7 in general were central to state- and empire-building throughout the ancient Near East and beyond. They fastened people into enforceable hierarchies and secured domestic and international political networks, economic transactions, and social statuses. In the Ur III context in particular, oaths had political and judicial dimensions (B. Lafont 1996). Promissory oaths invoking the king connected or enchained individuals into webs of obligations and allegiances and served as the means of holding individuals account­ able. Taking the oath was a public act involving witnesses, rituals, and, in some contexts, feasting. The oath appears in an array of activities. Promissory oaths were used in management practices in the realms of political economy and state infrastructure. Several recently discussed texts provide examples to add to the many ones discussed in the works cited in note 4. On one occasion in Amar-Sin’s year 7, generals were gathered to swear loyalty oaths to the king at a feast (Steinkeller 2008: 185). Although its record employs the nam-erim2 idiom, a loyalty oath has a promissory nature. Six documents dating to Amar-Sin or later indicate that different high-ranking officials swore promissory oaths, by the “life of the king” to the sukkal mah (grand vizier) regarding the irrigation and draining of fields. 8 The oaths obligated these officials to operate with accountability to the king’s vizier, under penalty of death (gaz). The interesting document BCT 2 156 from Umma tells us that officials gathered in the house of Kiag, a judge whom Molina (2014: 127) links to the royal sphere. The purpose was for a merchant to swear by the king before witnesses regarding a missing seal. In a handful of letters, state officials recorded oaths of the king with names of witnesses; such letters established the “personal liability” of individuals 6.  It is preferable to some scholars to describe these as types of idioms rather than as types of oaths, as Steinkeller (1989: 75–76) argued that differentiation of the idioms into categories of promissory versus declarative does not hold strictly for all Sumerian sources. Falkenstein (1956–57: 64), who described a strict division of oaths, also suggested that two cases contained oaths that blurred the classification scheme (NSGU 103 and 190). But Sallaberger (2008: 159–160) argued that the idioms do nevertheless indicate distinctly different situations (160) and situated the oaths in judicial procedures using Umma court records. See B. Lafont 1996 on this discussion. 7.  I borrowed this phrasing from Stefan Esders (2016: 357), who wrote about the use of oaths in the building of new political orders after the fall of Rome. The works in Sommerstein and Bayliss (2013) on oaths and state in ancient Greece also relate to the matter of oaths in political contexts. 8.  After Steinkeller (2015: 98). See also Lafont (2010: 167–168) on these events.

208

Laura Culbertson

within the administration and thus they relate to legal matters. 9 One example, MVN 3 377, relates to an archive of unclear provenience of Guzana, whom Steinkeller identified as a “state official in charge of a large depot or storehouse” (1982: 642). The letter-sender Zazana includes a promissory oath before five witnesses to return stock and grain. 10 In- and outside institutional affairs, promissory oaths invoking the king could be used to establish new economic and social relationships between people of a variety of statuses. The oath occurred during the formalizing of agreements as well. 11 Mercer (1913) described promissory oaths as “ratifying” transactions (e.g., pg. 40), and the term is helpful for understanding how the oaths cemented agreements as enforceable and liable for litigation. Some sales thus included promises in the name of the king not to contest or raise a dispute, 12 and some had seals. 13 Rentals and loans also required promissory oaths of the king. 14 Creditors required the oath from debtors, even if they were conducting business outside official state institutions. Such oaths possibly indicated mistrust and that “creditors counted on the possibility of settling disputed matters in a court of law” (Widell 2008: 219–220). Regarding the social relationships and statuses, oaths of the king secured marriage agreements. 15 Free persons swore them to promise to overturn the slave status of their slaves. 16 People referred to the king in promissory statements reported in direct speech (Kitz 2013: 45). The slave Puka, for example, invoked the name of the king in a promise not to run away before seven witnesses (BE 3/1 1). As mentioned, taking the oath was a public, witnessed event. Falkenstein (1956: 64) suggested that one did not need to find a specific setting or authority to invoke the name of the king, which depicts this oath as more of an oral act than a ritual act. However, we are not reaching to assume that ritual acts accompanied the invocation even if we don’t have ample information on this, especially based on the argument that the verb pad3 refers not just to the utterance but to the act of completing the mu lugal (Steinkeller 1989: 75). In any case, a seal or royal official was not needed for the oath to be conducted, witnessed, and recorded. It is not to be suggested that the oath was therefore casual or weak. The expectation of fidelity and honor is well evident from the recording of the oaths along with names of witnesses who could attest to the details of a promise later. A theme of the above examples is that uttering the name of the king and completing the associated acts opened the speaker up to future accountability and 9. Sallaberger 2015a: 28. The four examples among the 600+ letters assembled by Sallaberger (2015a: 27 n. 63) employ the mu lugal idiom: BPOA 1 368, MVN 3 377, Nakahara 42: pl. 11, and YOS 4 275. 10. Another example of promissory oaths in the context of management and economy includes Ontario 2 156, discussed by Steinkeller (2013: 354). 11.  See Garfinkle 2004: 20 on the assumption that public acts occurred for contract-making, with the example of loans. 12.  Examples are enumerated in Steinkeller 1989: 71. 13. Like ZA 25: 209 no. 2. 14.  An example is in RA 98: 8 no. 10, from Umma, concerning a rented boat. Garfinkle (2004: 16 n. 41) lists examples of oaths used in contracting interest-free loans. 15.  A few examples regarding marriage declarations: NSGU 1, 2, 3, 4 and BM 106498 are court records and ZA 53: 76–78 nos. 17a and 17b contain “private” witnessed oaths regarding marriage. In NSGU 14–16, 18, and 19 the oath and marriage were not fulfilled and led to court proceedings. 16.  NSGU 75 contains an example of the oath of the king in the context of manumitting a slave.

Breakers and Enforcers of the Oath of the King

209

litigation, and when drafting contracts, the oath established the expectation that the parties could end up disputing in the public sphere, as Kitz (2013: 48) noted, requiring the intervention of higher officials, which brings us closer to the matter of how the oath was protected when broken.

The Oath in Court Records As Sallaberger explained, invoking a higher authority means the swearer submits to the supreme judicial authority of that power, such as the authority of the king (2008: 159). Court records (di-til-la), which tell us about judicial procedures concerning disputes, list the kinds of authorities who enforced the oaths and thus serve as our sources for the mechanics of oath enforcing. The oath of the king could make appearances at several stages in the life of a dispute. But there are two primary reasons the promissory oath of the king appears in court records: 1.  The oath was taken in past transactions, and then broken, as it were, when the promised act was unfulfilled, which prompted the dissatisfied party to seek the intervention of officials (discussed below). If someone denied having sworn by the king’s name, a witness could be called to swear the evidentiary nam-erim2 oath to prove them false (lul). 17 2.  The oath was prompted or coerced during the course of judicial proceedings to secure a resolution, such as a promise of one party not to raise claims again in the future or a promise to show up with supporting evidence. 18

Unfortunately, we have a disproportionate amount of documentation of taking promissory oaths versus documents telling us anything about the practiced consequences of breaking them (let alone on matters of the kinds of social stigmas or religious problems that could ensue). In neither situation do the documents spell out any self-reflexive conditional curses, even though curses are intrinsic to the oaths (Kitz 2013: 38). Our understanding of consequences is restricted to the matter of how judicial proceedings treated the situations. We don’t find indications of anything as extreme as death as punishments for compromising a promise, but failing to uphold the commitment enshrined by the oath was nevertheless serious and consequential. In fact, the breach could amount to a NIR-da/šer-da: “criminal act” or “transgression.” 19 One document even phrases the act of breaking the promise as “transgression of the king,” 20 signaling the role of royal jurisdiction. So, we return to the idea that the king, having been invoked and transgressed, would presumably be the actor charged with enforcing the oath through his representatives. But as noted above, there are gaps in our picture of how royal power was exercised on a daily basis in judicial contexts. Thus a short survey of what we can say about the projection of royal authority in judicial activities is needed to frame the ensuing discussion.

17.  Examples are in Table 1 below. MVN 11 185 also concerns past promissory oaths. 18.  Examples include NSGU 156, 157, and 164. In the context of judicial proceedings, it can also be used to promise to supply evidence or witnesses, as in NSGU 109 from Lagaš and BM 106551 from Umma. 19.  In the above-mentioned BE 3/1 1. See also AUCT 2 138 and JCS 12: 106, which includes an oath of the king not to go before the vizier with a claim. 20.  NATN 366: 2–5 mu lugal in-na-pa3 / tukum-bi nu-ag2 / NIR-da lugal-a-mu / in-na-du11.

210

Laura Culbertson

Royal Judicial Power “On the Ground” We are fortunate to have several recent studies to draw from on the sociopolitical context of oaths and judicial officials, even if the difficulties of mapping out the role of the kings or Ur in the daily administration of legal matters remain. Our first problem is of course the very limited amount of documentation from Ur. In one of the rare records from Ur, UET 3 45 (NSGU 114), we have reference to a judicial case taken over by the king (di dab5-ba lugal). Thus, the king retained power to seize cases but we lack a corpus to illuminate this activity. A second problem concerns the limitations of ideological texts for telling us about the practiced execution of authority. The law collection (the so-called Laws of Ur-Namma) establishes that kings were patrons of justice and order. Clues like the personal name Lugal-di-ku5, found in a small number of economic documents, echo the idea of the king as a judicial authority. But on the level of the actual practice of adjudication, we are without many court records that could tell us about the position of the king and his entourage in judicial affairs. Here, we can consider two other ways to examine royal participation and power projection in judicial happenings outside of the royal sphere: 1.  Invisible judicial activities. With this term we can refer to court activities that certainly occurred, but for which we have no corresponding court record to tell us about the dispute and its resolution. Instead we learn of the “invisible” events from other types of documentation or from references to past events in provincially archived court records. For example, a court record from Girsu indicates that Shulgi appointed judges and that there was a “place of the king’s judges” (ki di-ku5 lugal) where judicial events occurred. We learn of this place in two references to past events there, reported in a court record for a later proceeding held in Girsu, and aside from these references, the activities carried out at this setting are unknown. 21 There are also only two references to named royal judges at Nippur dating to this reign, and the full scope of their activities is missing to us. 22 An intriguing series of “invisible judicial activities” comes from the IriSaĝrig archive reconstructed by Owen (2013). In messenger texts, Owen identified orders for the provisioning of thirty-four meetings between royal officials (mostly messengers) with local judiciary officials (maškim lu2 di-daka) at this site. A number of the visits concentrated during the second regnal year of Ibbi-Sin, hinting that a major “judicial event” was in preparation (2013: 93–95). 23 One of the visitors in Amar-Sin 7 was Nanna-maba dumu lugal (Nisaba 15/2 60); the presence of a prince implies royal interest in the purpose of these meetings (ibid.). 24

21.  NSGU 113 ii 9. Interestingly, the references to the king’s judges indicate that Ur-Lamma the local governor has altered their decision (ii 15). 22.  Both as di-ku5 lugal: Ur-Dumuzida in BE 3/1 14 and Lu-dingira in NRVN 1 249. The name UrDumuzi di-ku5 Nibruki appears in a Girsu account early in the reign of Amar-Sin in PPAC 5 268. NSGU 117 is a court record closing with di-til-la di-ku5 lugal imin-ba (rev. 4′–5′), but the judges aren’t named. 23.  Another example of such a meeting is found in the tablet with Studies Salvini 375–376 2: 13, mentioning Ad-mu-um lu2-kin-gi4-a lugal, also dating to Ibbi-Sin 2. 24.  In Amar-Sin 7, Nabi-Enlil dumu lugal is listed with Ur-mu di-ku5 (BDTNS no. 040023). It may be reaching to assume the provisioning involved an “invisible” judicial event.

Breakers and Enforcers of the Oath of the King

211

2.  Extensions of the crown. Next and related, we can see the projection of the royal sphere into provincial judicial activities from the numerous examples of royal officials who attended proceedings and appeared in lists at the end of court records. Of course we should not assume that everyone with a royal title or affiliation shared a collective mind with unvaried interests, but as they were actors from the same spheres, we can infer that they were conduits of royal authority who represented the interest and images of the kings to provincial communities.

Many officials were dispatched specifically with the purpose of attending proceedings (like the aforementioned messengers), whereas others may have been included by invitation or because other affairs put them “in town” during times of proceedings. Molina’s (2014a: 144–146) compilation of court officials at Umma indicated that men with royal titles and affiliations attended proceedings alongside provincial, military, and local officials; panels of judicial officials represented a blend of institutions in many cases. His study also showed that various kinds of officials and judges moved around and some attended cases in multiple provinces. At Umma, royal princes make a few appearances in proceedings alongside local officials, 25 but royal messengers (lu2-kin-gi4-a lugal) are attested more frequently among those dispatched to judicial events at Umma, especially in the time of AmarSin. The messenger Ur-Nungal appears among the judicial officials in a court record from Umma, as well as in preparation for one of the aforementioned “invisible” events at Iri-Saĝrig, for which he met with the local maškim lu2 di-da-ka. 26 At Girsu, we also find mixed compositions of judicial officials, consisting of people link­ed to the royal household and mostly people from a range of provincial occupations; when the king’s vizier Arad-Nanna assumed the position of governor at Girsu of the Lagash province, he officiated judicial proceedings with various mixes of local officials, judges, and others, although he was not the sole judiciary authority in the city. 27 Clearly, the kings and those they dispatched must have been powerful actors in Ur III judicial affairs even if we cannot recover a complete picture of their authority and various interests. They officiated alongside officials from other spheres and were entrusted as custodians of judicial cases for the local governors (see note 26). Future studies can explore the degree of centralized coordination of officials’ migrations and visits. The court records from the province Umma made accessible and studied by Molina (listed in 2014a: 141–143) illustrate that there were differences in court procedures between Umma and Lagash, differences that go beyond superficial matters 25.  Lu-duga dumu lugal appears three times in lists among other officials: SNAT 374 (AS 6), BM 106540 (AS 6), and MVN 18 515 (AS 7). 26. In BM 106551 (AS 6) and Nisaba 15/2 143 (AS 9) respectively, assuming this is the same person. Another example of a royal messenger, dispatched to Umma, is Ur-dNanše lu2-kin-gi4-a lugal, active during the reign of Amar-Sin. In BPOA 1 852 he is the maškim for a case of the provincial gov­ernor. Undated NSGU 69 lists him as official 5 out of 6 after local authorities. Erra-nuIB, royal messenger, is maškim in NSGU 60 its duplicate BM 110379. Other messengers serve as maškim for the governor (e.g., BM 105339 and from Lagaš NSGU 106). There is also one attestation of a royal musician, Lu-Šara acting as maškim for the Umma governor (NSGU 110). 27.  Royal messengers in court records from the Lagaš province are attested in NSGU 33, 34, 106 (as maškim), and 178. In one of the few court records from Garšana, we find AN-sukkal as one of two officials (BPOA 1 602). MVN 11 185 also includes a royal messenger.

212

Laura Culbertson

of how tablets were drafted. For example, using the same documents, Sallaberger (2008: 174) showed that the timing of different types of oaths vis-à-vis other court procedures also varied between these two provinces. Not only did procedures vary, but court compositions also varied from province to province. Groups of court officials at Umma rarely included a person titled as judge, whereas the Lagash court proceedings almost never occurred without one. Thus, we do not see evidence for a complete homogenization of judicial norms among the provinces, even if there are common cultural practices. Molina’s (2014b) dissection of TCL 5 6047 and related cases from Umma shows the interplay of royal dependents, provincial officials, local officials, and military officials over the course of a series of legal investigations, which suggests that the political dynamics between sectors were under regular negotiation. In short, the royal sphere cast its presence throughout the provinces as we would expect, interacting and colliding with local political networks and even reformulating them to install royal officials. The king’s image as a judicial authority was widespread, even if he remains partially invisible in the textual record judicial proceedings. But at the same time, regional variations and the participation of mil­ itary, provincial, and local officials prevent us from adopting a picture of uniform regulation. Some variations are certainly a consequence of having incomplete and inconsistent documentation, but there is also enough to suggest that interactions among royal and other actors was negotiated as opposed to involving a forced conformity and standardization.

Adjudicated Oaths At last we can turn to the initial question—who enforced the oath of the king? If we cannot find the king himself in this role, who served as the judicial authority when someone raised a claim about a broken mu lugal oath and wanted fulfillment? There are hundreds of references to mu lugal oaths in court records, but this part will draw on twenty examples that refer to oaths of the king that have been left unfulfilled. That is, our sources are court records that indicate a promissory oath was taken in the name of the king, its corresponding promise broken, and the transgressor called before judicial officials to account for their non-fulfillment of the promise. Three examples come from Umma and seventeen from Lagaš. These deal with city “citizens” in these provinces, by which I mean people of non-royal or non-official status. The asterisk by the tablet citation indicates cases in which the past oaths/promises were investigated and litigated in the “current” proceedings recorded in the court record. Table 1. Instances of investigated or litigated mu lugal promises Texts, Date, and Provenience

Purpose of Oath

Witnesses to the Oath

Judicial Officials

*NSGU 20, Šulgi 48

Marriage dissolution

Dugidu and UNga, engar

Provincial governor (Ur-Lamma)

*NSGU 109, date broken (Šulgi era based on governor)

To produce a tablet about a sale in seven days

Urdamu and UNga

Provincial governor (Ur-Lamma)

Breakers and Enforcers of the Oath of the King

213

Texts, Date, and Provenience

Purpose of Oath

Witnesses to the Oath

Judicial Officials

BM 106551 (an AmarSin year) Umma

Unfulfilled promise to bring witnesses in three days (regarding contested slave)

Nidamu, Arad-Nanna gu-za-la2, Ur-Nungal, lu2-kin2-gi4-a lugal

Same three officials who witnessed the oath, Nidamu, a throne-bearer, one royal messenger

*NSGU 140 AmarSin 4

A transaction of silver

At least one witness (broken)

1 maškim (Lu-Girsu)

*BM 106498, Amar-Sin 5, Umma

Marriage arrangement

Agi and Uda

None listed

*BM 110171, Amar-Sin 7, Umma

Ownership of palm grove

Unclear/unstated

Governor and eight local officials 28

* NSGU 197, Šu-Sin 1

Promise of slave delivery

Duga-Baba and KaLamma, lu2-zadim

Four judges 29

* NSGU 17 Šu-Sin 2

Broken marriage arrangement (one mina settlement)

Lugal-igihuš dumu Four judges Ur-Baba nu-banda3 and Lu-Šara dumu Niurum nar

NSGU 146, Šu-Sin 2

Witnessed recording of an oath

Ur-Baba dumu [broken], Mati-ili, and Duga (of them Ur-Baba swore)

Four judges

* NSGU 18 Šu-Sin 3

Marriage arrangement

Lugina dumu Urrani and Kuku dam Urgar ma2-lah4

Four judges

* NSGU 131 Šu-Sin 3

Promise to deliver a chair in three days

Kalla the lu2-giš-gi lugal and Babamu maškim

Two judges and three attending officials 30

* NSGU 205: 27–42, Šu-Sin 4

Slave status

A tablet with the oath promising manumission, submitted by a female slave

Arad-Nanna, vizier and four attending officials 31

* NSGU 25, Ibbi-Sin 1

Marriage prohibition?

(broken)

Four judges

* NSGU 16 (dates broken for this and the following entries)

Marriage arrangement

Maga, Nineinzu, Ur-Ninazu, and Lu-Gatumdu

Broken

* NSGU 19

Marriage arrangement

Broken

Broken

* NSGU 26

Concerning marriage

Ur-Alla, Lugalhegal, Nig-Baba, Ninkugani

Broken

28.  Of mixed affiliations, both provincial and possibly royal, plus a merchant (dam-gar) and a city elder (abba iri). 29. The judge panels in this one and the next three cases are identical: Lu-Šara, Lu-ibgal, Ludingira, and Ur-Ištaran. 30.  One of whom is Ur-Nanše dumu Lugaligimaše, who appears as an attending official in three separate cases: AOAT 25 442 7, NSGU 3, and the case cited here NSGU 131. Another is Ra dumu Luigisasa, who also appears in the same function in TÉL 110 and NSGU 153, 155, and 171. 31. One of whom is Lu-dingira dumu Lugal-barabe, whom Molina (2014a: 135 nn. 59 and 110) suggested is the same one to appear as an official in Umma judicial proceedings.

214

Laura Culbertson

Texts, Date, and Provenience

Purpose of Oath

Witnesses to the Oath

Judicial Officials

* NSGU 28

Slave inheritance

Namudadi and [broken]

Broken

* NSGU 29

Slave inheritance

Namti’inzu guda2 and Ur-Enki dumu Atu

Broken

* NSGU 63

Slave sale

Lugalkagina

Broken; approximately eight broken lines leave space for eight officials’ names

* NSGU 206

Marriage

Dugazida, Urgigira, and Ninkalla

Broken

The sample is quite small, but we can consider some patterns and non-patterns in light of the broader picture I attempted to offer in the previous sections. First, noting the takers and breakers of the oaths, we find a variety of “citizens” from the provincial cities, including both men and women. The witnesses of the oath of the king were, as expected, not necessarily of special status, but people with a social or professional relationship to the swearer (save for the case in BM 106551 in which the officials who prompted the oath also adjudicate it). As we should expect given the above synopsis, the adjudicating officials hold a variety of important affiliations, ranging from the local, to provincial, to royal admin­ istrations, just as we see in the court records in general. Thus, we might infer that the authority to protect sworn promises on behalf of the king’s name was diffused through many interlinking administrations. In the case of BM 106551 from Umma, the three officials listed are representative of the blend of affiliations one finds in judicial officials from this province. 32 The panel of judges found in the Šu-Sin-era cases at Lagaš includes the judge Lu-Šara, who also had the title sukkal according to his Ibbi-Sin-era ARAD-zu seal (RTC 431). Molina (2014a: 126) suggested the panels of judges were installed in this province by the king as a deliberate “judicial reform” and noted that they linked to both the royal and provincial administrations. Zólyomi (2017: 6) has also shown that at least one member of the regular judge panels at Girsu had provincial origins and activities (Lu-Ningirsu, who is not listed in the panels represented here). Thus, it seems the backgrounds of the judges was complex and we cannot link them exclusively with one sphere of the state, even if it was the king who established their position and authority. Next we can consider the matter of what precisely these officials were enforcing. The unfulfilled oaths recorded in these court records—aside from those involving promises sworn about supplying evidence—represent agreements established by the disputing parties before their witnesses. These conditions and only these conditions are what the judicial authorities investigated and addressed in proceedings. When the oath was unfulfilled, the individual who suffered the breach initiated the proceeding before court officials; thus, the first level of enforcement of the oath of the king is self-initiated by the man or woman afflicted by the broken promise. The

32.  Molina (2014a: 146) classified Arad-Nanna (there as Urdu-Nanna) as “provincial?”

Breakers and Enforcers of the Oath of the King

215

judicial officials usually restricted their role to getting to the “truth” of statements through various strategies rather than imposing punishments. In general in the Ur III court records, subjecting a disputed matter to higher authorities did not necessarily create an opening for the transposition of outside rules and regulations. NSGU 15 (not included above), about a dissolved marriage agreement, indicates that an oath of the king was taken to secure a marriage agreement, but essentially nullified because the parents of the groom were un­ aware of the oath. Taking the oath and completing whatever formalities it entailed did not overrule social expectations of parental knowledge. Many of the marriage dissolutions show a parallel between custom and the Laws of Ur-Namma: in the payment of 1 mina of silver to the abandoned woman. 33 But in the case of NSGU 20, overseen by the governor of Lagaš, the couple dissolving their marriage agreement had established their own agreement: Geme-Enlila swore by the king’s name that she would accept ten shekels. When the matter went before the governor, he simply investigated whether she was paid the ten shekels and confirmed that she had based on the oath of two witnesses (of course we are lacking the details of the dissolution and its causes). These cases differ from the managerial use of the oath of the king mentioned above in that they are not about the top-down imposition of regulations and obligations, but about protection of the conditions already established. In sum, adjudicating these oaths was not necessarily an avenue for transposing external rules, or to impose new rules, but simply a means of investigating and enforcing the conditions that individuals established for themselves.

Conclusions Even though we have not recovered a complete picture of how the Ur III kings projected power into provincial judicial contexts by any means, we can draw an impression from the foregoing. On the one hand, we’ve seen that promises secured by oaths of the king had a role in managing various state affairs, but in other contexts, regulation of the oath came from ground-up initiative. Concerning judicial activity, we noted the indications of variation in judicial traditions between provinces. In this context, the oath of the name of the king reflects a real projection of power from the royal sphere and its use indicates trust in the protection of promises. But in practice, protection of sworn promises was not consolidated to a single office or representative. From the pieces we can splice together above, the oath was enforced by a spectrum of actors and authorities. One does not need to assume that the oath of the king, when taken by people outside the state administration, was about regulation and the homogenization of legal norms across the state. Individuals and their associates opted to take advantage of the oath in the protection of their affairs. The custom helped with dispute resolution because of the shared consent to the power of the king, and, more pragmatically, because of the participation of many officials in judicial events.

33.  For example, NSGU 205 has a settlement at one mina, which parallels the Laws of Ur-Namma §9, which specifies the same amount (Civil 2012: 247).

216

Laura Culbertson

Bibliography Barta, H., Lang, M., and Rollinger, R. 2015 Prozessrecht und Eid: Recht und Rechtsfindung in antiken Kulturen. Philippika - Altertumswissenschaftliche Abhandlungen / Contributions to the Study of Ancient World Cultures 86. Wiesbaden: Harrassowitz Verlag. Civil, M. 2012 The Law Collection of Ur-Namma. Pp. 221–286 in Cuneiform Royal Inscriptions and Related Texts in the Schøyen Collection, ed. A. R. George. CUSAS 17. Bethesda, MD: CDL Press. Edzard, D. O. 1975 Zum sumerischen Eid. Pp. 63–98 in Sumerological Studies in Honor of Thorkild Jacobsen on His Seventieth Birthday, June 7 1974, ed. S. J. Lieberman. AS 20. Chicago: University of Chicago Press. Esders, S. 2016 ‘Faithful believers’: Oaths of Allegiance in Post-Roman Societies as Evidence for Eastern and Western ‘Visions of Community.’ Pp. 357–374 in Visions of Community in the Post-Roman World: The West, Byzantium and the Islamic World, 300–1100, eds. C. Ganter, R. Payne, and W. Pohl. New York: Routledge. Feder, Y. 2010 The Mechanics of Retribution in Hittite, Mesopotamian, and Ancient Israelite Sources. JANER 10/2: 119–157. Garfinkle, S. 2004 Shepherds, Merchants, and Credit: Some Observations on Lending Practices in Ur III Mesopotamia. JESHO 47/1: 1–30. 2008 The Third Dynasty of Ur and the Limits of State Power in Early Mesopotamia. Pp. 153–168 in From the 21st Century B.C. to the 21st Century A.D.: Proceedings of the International Conference on Neo-Sumerian Studies Held in Madrid 22–24 July 2010, eds. S. J. Garfinkle and M. Molina. Winona Lake, IN: Eisenbrauns. Kitz, A. M. 2013 Cursed Are You! The Phenomenology of Cursing in Cuneiform and Hebrew Texts. Winona Lake, IN: Eisenbrauns. Lafont, B. 1996 Serments politiques et serments judiciaires à l’époque sumérienne. Pp. 31–47 in Jurer et Maudire: pratiques politiques et usages juridiques du serment dans le Proche-Orient ancien, ed. S. Lafont. Mediterranees 10–11. Paris: L’Harmattan. 2010 Sur quelques Dossiers des Archives de Girsu. Pp. 167–180 in “Why Should Someone Who Knows Something Conceal It?” Cuneiform Studies in Honor of David I. Owen on His 70th Birthday, eds. A. Kleinerman and J. M. Sasson. Bethesda, MD: CDL Press. Lafont, S. 1996 Jurer et Maudire: pratiques politiques et usages juridiques du serment dans le Proche-Orient ancient. Mediterranees 10–11. Paris: L’Harmattan. Mercer, S. A. B. 1913 The Oath in Cuneiform Inscriptions. JAOS 33: 33–50. Michalowski, P. 2008 Networks of Authority and Power in Ur III Times. Pp. 169–206 in From the 21st Century B.C. to the 21st Century A.D.: Proceedings of the International Conference on Neo-Sumerian Studies Held in Madrid 22–24 July 2010, eds. S. J. Garfinkle and M. Molina. Winona Lake, IN: Eisenbrauns.

Breakers and Enforcers of the Oath of the King

217

Molina, M. 2010 Court Records from Umma. Pp. 201–217 in “Why Should Someone Who Knows Something Conceal It?” Cuneiform Studies in Honor of David I. Owen on His 70th Birthday, eds. A. Kleinerman and J. M. Sasson. Bethesda, MD: CDL Press. 2014a Court Officials at Umma in Ur III Times. ZA 103/2: 125–148. 2014b From Court Records to Sammelurkunden: A New Tablet from Umma and TCL 5, 6047. Pp. 399–421 in Studies in Sumerian Language and Literature. Festscrift für Joachim Krecher, eds. N. Koslova, E. Vizirova, and G. Zólyomi. Babel Und Bibel 8. Winona Lake, IN: Eisenbrauns. Owen, D. I. 2013 Cuneiform Texts Primarily from Iri-Saĝrig / Āl-Šarrākī and the History of the Ur III Period. Nisaba 15. Bethesda, MD: CDL Press. Richardson, S. 2014 Mesopotamian Political History: The Perversities. Journal of Ancient Near Eastern History 1/1: 61–93. Sallaberger, W. 2008 Der Eid im Gerichtsverfahren in Neusumerischen Umma. Pp. 159–176 in On the Third Dynasty of Ur: Studies in Honor of Marcel Sigrist, ed. by P. Michalowski. JCS Supplemental Series 1. Boston: American Schools of Oriental Research. 2015a Special Cases and Legal Matters: Diction and Function of Letters in the State of the Third Dynasty of Ur (2110–2003 BC). Pp. 15–30 in Official Epistolography and the Language(S) Of Power: Proceedings of The First International Conference of the Research Network Imperium & Officium, eds. S. Procházka, L. Reinfandt, and S. Tost. Papyrologica Vindobonensia 8. Vienna: Osterreichische Akademie der Wissenschaften. 2015b Sumerische und altbabylonische Eidesformeln in lexikalischer und kulturhistor­ ischer Perspektiv. Pp. 179–192 in Prozessrecht und Eid: Recht und Rechtsfindung in antiken Kulturen, eds. H. Barta, M. Lang, and R. Rollinger. Wiesbaden: Harrassowitz Verlag. Sommerstein, A. H. and A. J. Bayliss 2013 Oath and State in Ancient Greece. Berlin/Boston: de Gruyter. Steinkeller, P. 1982 On Editing Ur III Economic Texts. JAOS 102/4: 639–644. 1989 Sale Documents of the Ur-III Period. FAOS 17. Wiesbaden: F. Steiner Verlag. 2008 Joys of Cooking in Ur III Babyonia. Pp. 185–192 in On the Third Dynasty of Ur: Studies in Honor of Marcel Sigrist, ed. by P. Michalowski. JCS Supplemental Series 1. Boston: American Schools of Oriental Research. 2013 Corvée Labor in Ur III Times. Pp. 347–424 in From the 21st Century B.C. to the 21st Century A.D.: Proceedings of the International Conference on Neo-Sumerian Studies Held in Madrid 22–24 July 2010, eds. S. J. Garfinkle and M. Molina. Winona Lake, IN: Eisenbrauns. 2015 A Group of Girsu/Lagash Texts Dealing with Irrigation. NABU 2015/3: 98–100 no. 61. Wells, B. 2008 The Cultic vs. the Forensic: Judahite and Mesopotamian Judicial Procedure in the First Millennium BCE. JAOS 128: 205–232. Wells, B., Magdalene, R. F., and Wunsch, C. 2010 The Assertory Oath in Neo-Babylonian and Persian Administrative Texts. Revue Internationale des Droits de L’antiquité 57: 13–29. 2012 The Grammar of the Neo-Babylonian Assertory Oath. JNES 71/2: 275–284.

218

Laura Culbertson

Widell, M. 2008 The Ur III Metal Loans from Ur. Pp. 207–224 in The Growth of an Early State: Studies in Ur III Administration: Proceedings of the First and Second Ur III Workshops at the 49th and 51St Rencontre Assyriologique Internationale, London July 10, 2003 and Chicago July 19, 2005, eds. S. Garfinkle and J. C. Johnson. Madrid: Consejo Superior de Investigaciones Cientificas. Yoda, I. 1993 Oaths in Sumerian Archival Texts: A Case Study in Ur III Nippur. PhD dissertation, Yale University. Zólyomi, G. 2017 The Secret Life of Lu-Ningirsu, the Judge. Cuneiform Digital Library Bulletin 2017: 2.

Utopic and Dystopic Images in Mesopotamian Literature: The Conflict Between Order and Chaos in Ur III Maria de Fátima Rosa & Isabel Gomes de Almeida Universidade NOVA de Lisboa

Introduction During the last decades, Utopian Studies cemented its position as an important field of research within the Humanities. Though its scholars commonly agree that utopic literary and iconographic imagery was developed long before Thomas More’s formulation, 1 traditionally their enquiries do not reach literary production that predates Classical Antiquity. Hence, some questions have arisen in our minds: was the “good place” that was envisioned by More, and developed by his successors, in any way present within the cultural Mesopotamian framework? Can these past actors be understood as having notions of utopia and dystopia, given that they were guided by the ideas of order and chaos? We believe that Mesopotamian literature presents itself as a rich ground for interdisciplinary approaches, given the abundant reflections displayed about daily life, governance, cosmic order, and so many other topics which constitute the basis of the Humanities. Therefore, we have decided to make a reverse journey, from a modern concept to an ancient and distant past, hoping to shed light on different aspects of the Mesopotamian literary production, and thus contribute to further interdisciplinary discussions on the matter. The literary accounts regarding the Third Dynasty of Ur, a period characterized by the (re)building of a new political unification, but, simultaneously, struck by disruptions, seems to be an ideal starting point. Therefore, we will focus our analysis on the compositions Enki and the World Order, The Lament for Sumer and Ur, The Death of Ur-Namma, and on Šulgi’s praise poetry.

Authors’ note: This paper had the support of CHAM (FCSH/NOVA-UAc), through the strategic project sponsored by FCT (UID/HIS/04666/2013). 1. Vieira speaks of the anamnestic nature of utopia (2010: 5).

219

220

Maria de Fátima Rosa & Isabel Gomes de Almeida

Utopianism and Mesopotamian Literature Five hundred years ago, Thomas More published his magnus opus, Utopia, where the hopes of the Renaissance historical actors took shape in a non-existent yet wonderful place. 2 Within it, society strived by obeying strict rules which secured the possibility for everyone to have a rightful existence. Consequently, More’s book can be considered as an epitome of order. And this was a vital notion in Mesopotamia, profusely displayed in this civilization’s literature. 3 Through it, the ancient scribes aimed to present proper guidelines for society to achieve a perfect, orderly existence, which in this profoundly religious world meant, first of all, to be in harmony with the divine will. 4 Sasson, who worked on this same subject matter regarding Mari’s epistolary corpus, pointed out how the focus of those accounts was on elaborating paths that led to order or away from chaos (2006: 40). So, one can say that Mesopotamian literature contained rules for human society to follow so that the cosmos, as a whole, obtained and maintained its rightful order. 5 Gilgameš’s existential journey, for instance, depicts the difficulties and the arduous path undertaken by the hero in his search for immortality. The narrative exhibits a thoughtful message about the purpose of one’s life. When he met the ale-wife, she declared to him: O Gilgamesh, where are you wandering? You cannot find the life that you seek! When the gods created mankind, For mankind they established death, Life they kept for themselves. You, Gilgamesh, let your belly be full, Keep enjoying yourself, day and night, Every day make merry, Dance and play day and night! Let your clothes be clean! Let your head be washed, may you be bathed in water! Gaze on the little one who holds your hand, Let a wife enjoy your repeated embrace! Such is the destiny [of mortal men,] (…) 6

2.  For a recent translation of More’s Utopia, see Miller 2014. During 2016, the five-hundred year anniversary of the publication of Utopia was celebrated around the world, in numerous conferences and colloquiums. We must refer to the 17th Conference of the Utopian Studies Society/Europe, entitled “500 Years of Utopias: Commemorating the 500th Anniversary of Thomas More’s Utopia,” held at NOVA FCSH, in Lisbon, Portugal. 3. This vital notion was also depicted in Mesopotamian visual arts, such as the Assyrian royal reliefs. However, to work on this subject matter upon Mesopotamian iconography we would have to produce another paper. 4.  It must be recalled that Mesopotamian thought was deeply theocentric. Deities were understood as ruling every single aspect of the cosmos (Bottéro 2004: 55). 5.  Utopian scholars propose that utopic and dystopic narratives are completely focused on human actions, lacking divine intervention. Naturally, this aspect cannot be applied to the Mesopotamian mental framework, given its theocentric nature. 6.  Translation by George 2003: 279.

Utopic and Dystopic Images in Mesopotamian Literature

221

Siduri’s message summarizes the wise and instructive tone of the composition, 7 by revealing what deities intended for humanity to seek: to create a prosperous household, where intimacy with the family’s members and the joyful consumption of the fruits of labour could be appreciated. 8 To lead an orderly life was, in this perspective, to create and maintain a “good place” within the household, and, in a broader sense, within society. Hence, we believe a close connection between the Mesopotamian notions of order and chaos and the modern concepts of utopia and dystopia can be established. But, first of all, what is utopia? The term constitutes a neologism created by More that can be interpreted as referring to two different concepts: on the one hand, More combined the Greek word topos (“place”) with the negative particle ouk or u, pointing out to “a non-place,” that is a place that could only be imagined and was nowhere to be found; on the other hand, he also added to topos the term eu, which means “good” or “well,” thus forming “the good place,” eutopia, where everything was ideal. Fátima Vieira (2010: 5) considers that this combination makes the term utopia simultaneously an affirmation and a denial, which leads to the identification of a profound tension within itself: the word evoked a non-existent, yet a rather good place. On another level, it should be noted that the famous island of Utopia was considered to be located on the border of reality. This liminal location underlines the tension of the term, and at same time provokes another one amongst the audience, who was forced to dwell between real and imaginary places. In fact, following the conventions of travel literature of his time, Thomas More depicted its main char­ acter, Raphael Hythlodaeus, 9 as embarking on a journey which began in a real place, to a distant and imaginary one. The plausibility tone of the narrative was achieved by the establishment of a pact between author and audience, where the latter agreed to accept the confusion amid reality and imaginary. 10 So, let us focus on the Mesopotamian literary accounts in search of a liminal, nonexistent good place. The distant land of Dilmun naturally comes to mind. Many decades ago, Kramer (1945 and 1956: 141–147) considered the depictions of this region in Enki and Ninhursag 11 as an expression of the Sumerian paradise. 12 On the contrary, Lambert and Tournay (1949: 123) considered that this composition evoked not an edenic but a primeval land, an idea further developed by Dickson (2005 and 2007). The repetitive adverb “yet,” which we find throughout the poem, may indeed point out to an earlier stage, where things were still in formation. Nevertheless, aren’t the earlier stages of the world, in many literary contexts, often regarded as pleasant and ideal places for life? 13 In fact, Mesopotamian literary compositions reg­ 7.  Toorn 2007: 22. 8.  For a different interpretation of this passage, see Abusch 1993: 4–5. 9.  Another tension is identifiable in the main character’s name: Raphael stands for “healer from God” and Hythlodaeus means “speaker of nonsense.” Hence, a pun between his blessed and mischievous character is achieved by the name symbolic meaning (Sargent 2010). 10.  Vieira 2010: 8. 11.  ETCSL 1.1.1 12.  It should be noted that Kramer, as other scholars of his time, sought to identify archetypes of the Old Test­ ament tradition in Mesopotamian literary accounts. The expression “paradise” that Kramer applied to Dilmun must be understood in this historiographic context. 13.  Regarding nineteenth and twentieth centuries American literature, Leonard Lutwack identified the primeval past as one of the places used by authors to represent golden or dark ages, in order to launch

222

Maria de Fátima Rosa & Isabel Gomes de Almeida

ularly begin with the evocation of the primeval time, where the “yet to be” opened the possibility for bliss. Besides the famous enūma eliš, the Babylonian poem that discusses the many levels of tension between order and chaos, and which begins by evoking the earlier stages of the cosmos, other paralleled examples can be presented. Take the letter A.1968, from the epistolary archive of Mari, for instance. To achieve the main goal, the legitimation of Zimrî-Lîm’s reign, the letter goes back to the primordial battle between Têmtum and Addu, and the way the latter restrains the first with his weapons. 14 The aftermath of the battle secured the essence of the cosmos, and thus life was ready to begin in its most pure and effective way. As discussed elsewhere, it is our understanding that in Mesopotamian mental framework, the very nature of the universe was repeatedly searched for in its beginning (Rosa 2015). In this perspective, Dilmun as a place of origin (and whether or not an expression of a paradise) was certainly an idealization of a “pristine” land, where the materialization of the “good place” was imminent. There, “the lion did not slay,” “no eye-diseases said ‘I am the eye disease’,” “no old woman said ‘I am an old woman’.” 15 Consequently, there was no death, no diseases, no old age, making it the ideal place to live in. It was not by chance that Ziusudra, the Sumerian deluge hero, ascended to this honourable land, as it its described in The Flood Story. 16 The correlation between the “good” features of Dilmun and its liminal location, where real and imag­ inary were confused by the very presence of the flood survivor, a liminal character caught between two cosmic orders, lead us to think of this land as a sort of utopian place. Let us now go back to the Gilgameš Epic. Just like Raphael Hythlodaeus, the hero undertakes a journey that begins in a real dwelling, the city of Uruk, to a distant, far-out of reach one, where Utnapištim was placed to live his immortality. The Gilgameš Epic alternates between real and imaginary, between the hope for immortality and the orderly human existence, between the right and the wrong concretization of kingship. Within Utopianism, the distrust on human character led to the construction of an antithetical form of More’s utopia. By adding the Greek term dus (“bad”) to topos, the good place turned into a bad one—dystopia. However, utopian scholars postulate that more than just the contrary, dystopia is contained within utopia, and vice-versa. 17 In this line of thought, Laurence Davis proposes that dystopia can be considered as the alter-ego of utopia “always pulling its dreamy companion back to earth” (2013: 23). A parallel between the tension and even complementary character of utopia and dystopia can thus be made with the notions of Mesopotamian order and chaos, where one only exists within the other. As stated above, enūma eliš is a fine example of this entangled vision: for instance, the opening lines, which allude to the passage from the chaotic primeval waters to an orderly first stage of the cosmos, as well as the depiction of the battle between the chaotic forces led by Tiāmat and the orderly army commanded by Marduk.

their critics about the real, contemporary world (1984: 56). 14.  Durand 1993. 15.  ETCSL 1.1.1. 16.  ETCSL 1.7.4. 17.  In the opinion of Gregory Claeys “rather than being the negation of utopia, dystopia may paradoxically be its essence” (2013: 15).

Utopic and Dystopic Images in Mesopotamian Literature

223

Moreover, in a dystopian context the powers in charge are normally envisioned as oppressive and, therefore, object of criticism. Lyman Tower Sargent (2013: 12) characterizes some dystopias as “jeremiads,” borrowing the designation attributed to the early Puritan sermons, in the form of a lament, produced in New England, during the seventeenth century. For this author, most dystopian narratives display, even if in a subtle way, an accusative tone towards the human fault that led to the undesirable world. As we know, laments for the destruction of a city were a category of the Mesopotamian literature. In those compositions, deities are depicted as having abandoned the city and thus the disruption of order, which was the motive for their abandonment to begin with, was intensified. That disruption had to be provoked by humans, who failed to satisfy divine will. The concept of human failure was thus perceived a posteriori, 18 when deities acted upon it. 19 In this logic, the laments for the fall of Mesopotamian cities can be understood as an expression of a jeremiad dystopic narrative. Other aspect utopian scholars have been working on concerns the development of the temporal dimension of the narratives. Most of the Renaissance utopian compositions displayed a sort of “frozen present,” where the boundaries of time were abolished. Though these earliest narratives criticized the present, they did not offer a possibility of change in the future. However, this atemporal dimension began to change at the end of the eighteen century as a consequence of hope being promoted by the Enlightenment movement. Utopian compositions gradually began to project themselves into the near future, conferring to the narratives other level of dynamism. Vieira (2010: 23) stresses the importance of this development, given that the “not yet” conceptualization promotes “the universe as an open system, where nothing is static and where everything is in a constant process of formation.” Utopia was thus developed into a euchronia—a pun on the Greek word chronos. Mesopotamian literary accounts express a temporal dimension interesting to recall. In the Mesopotamian mental framework, present and future belong to the same conceptual range, which in grammatical terms is consubstantiated in the durative. As pointed out above, the “yet to be” opens up the possibility of change, of hope in (re)establishing order. Moreover, the mythical character of this civilization’s literature conveys continuity on its elaborations, given that myth, amongst other aspects, is also an explanation about situations that every individual have or will deal with, in one way or another, on the course of their lives. 20 Hence, the Mesopotamian literary accounts dealing with order/utopia and chaos/dystopia can also been seen as a type of euchronia. This general overview, where we tried to apply the main features of the Utopi­ anism theoretical framework on the Mesopotamian literary production opens up multiple possibilities to further analysis on this subject matter. For now, let us focus on the literary accounts which concern the Third Dynasty of Ur.

18.  Bottéro 1992: 188. 19. In Atra-Hasis it was the noise of humankind that disturbed deities. The human’s fault was unconscious, given that noise is part of the normal daily life’s activities. See the composition in Dalley 2000. 20.  Meslin 1973: 235.

224

Maria de Fátima Rosa & Isabel Gomes de Almeida

Utopia and Dystopia in the Third Dynasty of Ur The conflict between order and chaos was crucial for Mesopotamian thought. As mentioned above, it was the balance and the complementarity relation between human action and divine will that dictated the orderly or chaotic state of the land, in a relative perspective, and of the cosmos, in a general one. In this perspective, the Third Dynasty of Ur, which was a relatively short period, was an epoch of many contrasts: on the one hand, it was a time of cultural and political renewal, where the Sumerian archetypes were revitalized and the political unification of the “land between the rivers” re-established; on the other, it was also a time of economic and political disintegration, especially during the reigns of Šu-Sîn and Ibbî-Sîn, where amongst others threats, the one regarding nomadic populations re-emerged. 21 It should thus be expected to find in the literary corpus related to this period images that might transport us to the Utopianism notions described above. We propose to explore two entangled perspectives: a cosmic one, which conveys the universe as a whole, and a royal one, focused on the king’s persona.

Enki and the World Order According to Bottéro and Kramer (1989: 181), and though the main copies of this composition were found in Nippur and dated to the beginnings of the second millennium BCE, its main subject is the Third Dynasty of Ur. In the narrative, we find Enki, the god of wisdom, carrying out an ordainment mission on behalf of Enlil, the chief deity in charge. Having been delegated with the power to decree “good destinies,” 22 Enki acted upon Enlil’s will, by means of his controlling power over the pure waters. 23 The god begins his task by building his holy shrine, which we must remember was considered the earthly abode of Mesopotamian deities. The household, as referred above, was nuclear for the Mesopotamian conception of an orderly existence, which makes Enki’s first action in the composition rather significant: for the cosmos to become a good place it was imperative to begin with the ordainment of one’s house. 24 Afterwards, Enki embarks on a journey to inspect other earthly domains. Sumer, Ur, Meluḫḫa, Magan, Dilmun, Marḫaši, Elam and even the region of those “who have no city, to those who have no houses, to the Martu nomads” 25 are visited by the god. Again, it is significant that Enki travels to areas which were often enemies of the Mesopotamian powers, such as Elam, and to regions not only threatening but also considered uncivilized, such as Martu’s land. 26 The goal, it seems, was to integrate the menacing alterity into an ordained world. And by accomplishing that, the land would become good. On his return to Nippur, Enki proceeds to ascribe attributes and functions to different deities. By deciding what each god and goddess should be 21.  For the historical context of this period, see Roux 1992. 22. ETCSL 1.1.3. 23. “When I approach heaven, a rain of abundance rains from heaven. When I approach earth, there is a high carp-flood” (ibid). 24.  “In a pure place, and name it with a good name” (ibid). 25. Ibid. 26. Many Mesopotamian literary accounts, such as the The Marriage of Martu, depict nomadic people as uncivilized: “He lives in the mountains and ignores the places of gods (…), and eats raw flesh. He has no house during his life, and when he dies he will not be carried to a burial-place” (ETCSL 1.7.1).

Utopic and Dystopic Images in Mesopotamian Literature

225

responsible for, the division of cosmic domains and the consequent establishing of cosmic rules was complete. Everything was in its right place. It becomes obvious how the narrative intended to underline the legitimate right to rule of the Third Dynasty of Ur. It should be recalled that the founder of this dynasty, Ur-Nammu, delineated an important political, economic, cultural and religious program, which allowed the restructuration of its kingdom, centralized in the port city of Ur. Amongst other actions, we should stress the building of ziggurats in four major cities: Ur, Eridu, Nippur, and Uruk, and the development of commercial routes with distant trading partners, which was vital for the sustainment of the country. Vantisphout (1997: 122) identifies a relation between these commercial trades and the inspection character of the journey undertaken by Enki on the above mentioned composition. In fact, from Meluḫḫa, Magan, Dilmun, and the land of Martu provisions arrived to Ur, such as fish and meat, and from Marḫaši, and Elam also arrived precious metals and stones. 27 Thus, the composition can be seen not only as a discourse of legitimation but also as the description of a eutopia—a good place, where peace was established an order maintained. At the end of the poem we find a curious episode: the goddess Inanna inquires Enki, complaining she hasn’t been assigned with any specific function. Enki’s reply is interesting: he dismisses Inanna’s claims by evoking she already had a warrior function: “Inana, you heap up human heads like piles of dust, you sow heads like seed. Inana, you destroy what should not be destroyed; you create what should not be created.” 28 Moreover, the great god states: “But now, the heart has overflowed, the Land is restored.” 29 As Hermann Vanstiphout argued (1997: 131), although war had to be present, there was no need for such a task, at least in an active form, in the newly ordained world by Enki. Again, everything was in its right place. Accepting the utopian aspect of the composition, this episode should be regarded as an evocation of the “yet to be” feature of utopian narratives. In this case, if the orderly state of the land was threated, Inanna, whose warrior features were for now constrained, could act upon it, defending the idealized place, and ensuring Enki’s vision of harmony. An alternative, or better a possibility for hope was, thus, secured, in case the ever latent menace of chaos should emerge. The dystopic warrior features of the goddess were integrated in the utopic construction of Enki, underlying how one could not survive without the other; how one was contained within the other. These ideas are present in the famous garden scene of Ashurbanipal. 30 In the relief of Nineveh, king and queen are depicted enjoying a feast in the gardens of the palace after the victory against the kingdom of Elam. However, suspending from one of the trees, in this idyllic and harmonious environment, is none other than the head of the defeated king Teumman. Hence, in an atmosphere of amity and calmness a dystopic element is present, reminding the Assyrians of the daily dangers which threatened the peace and quiet of the good place they lived in. Order and chaos were thus two faces of the natural course of the cosmos.

27.  Bottéro and Kramer 1989: 184–185. 28.  ETCSL 1.1.3. 29. Ibid. 30.  The Banquet Scene (BM 124920).

226

Maria de Fátima Rosa & Isabel Gomes de Almeida

The Lament for Sumer and Ur The composition in form of a lament for the fall of Sumer and Ur, probably produced during Isin’s rise to power in Mesopotamia, 31 displays an antithetical image. It refers to the decay and consequent ruin of the Third Dynasty of Ur, by depicting the divine abandonment of the land. As a consequence, the once good place was exposed to all kinds of tribulations. The withdrawn of the divine favour upon Ur and its ruler was consubstantiated by the diluvium, and the world lost its shape. The opening line of the composition “To overturn the appointed times, to obliterate the divine plans, the storms gather to strike like a flood” 32 is clearly indicative of the dystopic tone of all the narrative. In fact, in the middle of the catastrophe, it is stated: “The flood dashing a hoe on the ground was levelling everything. Like a great storm it roared over the earth—who could escape it?” 33 Curiously, the dystopic character of the lament is a reverse depiction of the very same features which were envisioned as conducting to a harmonious and good society/place, and which were evoked in Enki and the World Order. The divine protection, materialized in Enki’s action, was withdrawn (and thus, the religious system failed); the once active symbols of monarchy held by Enki, 34 were powerless, given the immobilization of the king (and thus, monarchy as the link between human and divine collapsed); the nuclear cell of society, the shining household of the god, fell into ruin, given that people could not recognize each other in the middle of the disaster 35 (and thus, society and its rules were left to abandonment). In short, the universe returned to darkness and dust. Notwithstanding, we identify a profound dynamism at the end of the poem, where the chaotic state is overturned by the supreme gods An and Enlil. The possibility of reconstruction, of the “yet to be,” is again present. As Michalowski argued (1989: 7), if this lament was composed during the reign of Išbi-Erra of Isin, it must be seen as an “attempt to legitimize the new dynasty.” This power must have wanted to present itself as carrying out the mission of re-establishing divine favour, and so it was important to draw a line between the new rulers and the Ur predecessors. Thus, the dystopian manner in which these monarchs are displayed has the clear intention of stressing the efforts undertaken by Isin to make Sumer inhabitable and pleasant again. The strong assertions, in form of a plea, we find in the conclusion of the poem show a deep contrast with the dystopic vision of the fall of Sumer and Ur: That the orchards should bear syrup and grapes, that the high plain should bear the mašgurum tree, that there should be long life in the palace, that the sea should bring forth every abundance: may An not change it. The land densely populated from south to uplands: may An not change it. May An and Enlil not change it, may An not change it… 31.  Michaloswki 1989: 7. 32.  ETCSL 2.2.3. 33. Ibid. 34.  “The lord put on the diadem as a sign of lordship, he put on the good crown as a sign of kingship, touching the ground on his left side” (ETCSL 1.1.3). 35.  “The father turned away from his wife saying ‘This is not my wife!’ The mother turned away from her child saying ‘This is not my child!’ ” (ETCSL 2.2.3).

Utopic and Dystopic Images in Mesopotamian Literature

227

That cities should be rebuilt, that people should be numerous, that in the whole universe the people should be cared for; O Nanna, your kingship is sweet, return to your place. May a good abundant reign be long-lasting in Urim. Let its people lie down in safe pastures, let them reproduce... 36

As we have mentioned, regardless of its dystopic nature, the poem ends with a clear message of hope, with Sumer regaining divine protection. The Lament for Sumer and Ur underlines the continuity of order. Once again dystopia needs utopia, one being the essence of the other. Both the compositions manifest, therefore, vivid and entangled images of order and chaos regarding the same political period. The government of the Third Dynasty of Ur was, on the one hand, envisioned as fulfilling divine will, and on the other, disregarding it. The responsibility relied, naturally, on the figure of the king. The rhetorical discourse (whether utopic or dystopic) expressed on these narratives was obviously dependent on the power which commissioned its redaction. And the past, as often regarded by the Mesopotamian eye, reproduces a context to follow or one to avoid in the present and near future.

The Death of Ur-Namma 37 Let us now focus on the utopic and dystopic images conveyed in the Mesopotamian royal persona. The Death of Ur-Namma, composed in the form of a lament, is the ideal narrative to start our approach to this second vector. The narrative describes the journey of the deceased king to the Netherworld. Since Ur-Nammu died unexpectedly in battle, 38 a sign that could only mean he failed in obeying divine will, the message of this poem is rather negative. Its tone underlines the desolation and despair that struck the land after the violent death of the king. Again, we identify parallel images to those observed above, namely regarding the abandonment of deities, which led to the collapse of the monarchy and of the social order. 39 Fertility ceases, people lose their happiness, and ruin is imminent. However, the journey undertaken by the sovereign to the realm of the dead constitutes the last natural stage in the cycle of life of those to whom immortality had not been granted. The dystopic images expressed in this lament are but a necessary interregnum, given that mourning the deceased king was essential to the renewal of the dynasty. Just like in the Gilgameš Epic, where the hero’s grief upon Enkidu’s death drives him to embark in an existential journey that will allow his renovation as a monarch, and thus the restoration of Uruk as a good place, the death of Ur Nammu allows the possibility for his son, Šulgi, to restore order in Ur. From this perspective, the dystopic state of the land upon Ur Nammu’s failure as a ruler conveys a “yet to be” feature, whereas Šulgi could reverse the matter of affairs. 36. Ibid. 37.  For an explanation on this narrative’s structure, see Black, Cunningham, Robson, and Zólyomi 2004: 56. 38.  “(He) passed away in his prime” (ETCSL 2.4.1.1). 39.  For instance: “Enkimdu, the lord of levees and ditches, took away the levees and ditches from Urim;” “the Land became demolished,” “(Ur-Namma) could no longer bring pleasure to his wife with his embrace, that he could not bring up his sons on his knees” (ibid).

228

Maria de Fátima Rosa & Isabel Gomes de Almeida

Šulgi’s Praise Poetry It is only natural that Šulgi’s praise poetry appears as overflowing with images of order. His reign was an intense time of reformation, at all levels, in the wake of his father’s death. The first part of his long ruling of forty-eight years was dedicated to the political, military, and administrative reorganization of the kingdom. Amongst his measures, the king developed a complex program of construction of temples, some of which were certainly founded by his father. 40 The second part of his reign was focused on several military campaigns against powers such as the Lullubi, the Hurrites, and the princes of Luristan. It was during this period that Šulgi’s name began to be preceded by the divine ideogram d (dingir). His process of deification must be understood, as Michalowski stated, as an expression of his complex strategy to “reinvented his state” (2008: 38). The anthropological utopia of Šulgi is achieved, in his praise poetry, by means of his association with the semi-divine hero Gilgameš and his family. 41 The new status of the king of Ur granted him a special place amongst the gods, a place near the holy Ninsun. So, contrary to his father, Šulgi could undertake a different journey: after his death, he was not to enter in the domain of the Netherworld, reserved to human kings, but instead was meant to join deities in the restricted celestial sphere. In life and in death, Šulgi reached his desirable “good place.” Father and son were depicted, at first glance, as static images of chaos and order. However, if we take a closer look, we can perceive some latent nuances. As we stated above, the death in battle of Ur Nammu (a dystopic event) allowed the possibility of change, of reversion of the chaotic state of the land. On the contrary, the references to the divine nature of Šulgi had a clear disadvantage—his deification disrupted the liminal character inherent to the king’s persona. This feature was, as Michalowski stated (2008: 41), imperative for the monarch’s role as mediator. By transferring himself from the realm of humans to the divine domain, the sacred nature of his own office could be put into question. As a consequence, Šulgi’s divine essence conveyed a profound disruption. So, even though his praise poetry depicts a perfected image of his persona, in fact, his ambition could lead to severe dystopic consequences. 42 For the “good place” to be achieved it was imperative that the king, as any other human, should follow the natural, orderly existential cycle.

Final Remarks Our main goal with this paper was to present an interdisciplinary approach be­ tween Utopian and Mesopotamian studies. So, it should be asked: are these literary examples eligible to be regarded as utopian/dystopian narratives? Do they convey the general features defined by Utopian studies? We believe so. First, let us start by stressing the real historical backgrounds upon which these compositions were constructed. All of them refer to a concrete place, the land of Sumer and Ur, yet projecting an idealistic (or not) image of it. The necessary link with the audience 40.  Roux 1992: 195. 41.  “Šulgi, roaring like a rising flood against the rebel lands, embraces Gilgameš, his brother and friend, his comrade,” “Shepherd Šulgi, when your seed was placed in the holy womb, your mother Ninsumun gave birth to you; your personal god, holy Lugalbanda, fashioned you” (ETCSL 2.4.2.04). 42.  For some reason, the king’s deification attempted by Šulgi was abandoned by later monarchs.

Utopic and Dystopic Images in Mesopotamian Literature

229

is thus possible to be established. Moreover, the journey’s motive, which appears in Enki and the World Order and in The Death of Ur-Namma, stresses the liminal nature of the narratives between the real (human) and the imaginary (divine) worlds. We should also note that both the Lament for Sumer and Ur and The Death of Ur-Namma shows an implicit accusation towards human failure. Because society/king failed, divine favour was withdrawn, leading to the decay of Ur and to the violent death of the king. These two poems can, thus, fall into the jeremiad category postulated by Sargent. On another level, all the compositions reveal a dynamic tension between utopia and dystopia. The constant search for improvement and the possibility of alternative (into a good place or into a bad one) is always present, underlying the still in formation character of the universe. The utopic and dystopic images displayed are not static, on the contrary, they transmit realities where the “yet to be” factor is latent. Finally, it has to be mentioned two major differences between the Ur III literary accounts and the traditional utopianism framework: 1) the temporal dimension; 2) the human-centered action. First, we should state that instead of being projected into the future, these compositions are always depicted as contemporary or as retrospective euchronias. As we know, Mesopotamian thought valued what was “in front” (the pānānum), that is, the past. Hence, present and future were always shaped with regards to what had already happened. On the other hand, since the Mesopotamian mental framework revolved around divine will and power, a world without divine intervention was inconceivable. Thus, the deities were always present and active. But we do not believe these characteristics should constitute a barrier for Mesopotamian literature to be worked upon a utopianism perspective. They should be understood as they are: features of a different and distant mentality, whereas the modern theorizations, of any kind, should accommodate themselves to be applied into a different structure of thought, thus enrichening several fields of research. Hence, it is our understanding that Utopian Studies can and should be applied to the Mesopotamian literary accounts. We hope this preliminary contribution can lead to new analytical paths, allowing us to embark in more journeys to a “good place.”

Bibliography Abusch, T. 1993 Gilgamesh’s Request and Siduri’s Denial. Part I. The Meaning of the Dialogue and its Implications for the History of the Epic. Pp. 1–14 in The Tablet and the Scroll: Near Eastern Studies in Honor of W.W. Hallo, eds. M. E. Cohen, et al. Bethesda, MD: CDL. Bottéro, J. 1992 Mesopotamia: Writing, Reasoning and the Gods. Chicago: The University of Chicago Press. 2004 Au commencement étaient les dieux. Paris: Hachette Littératures. Bottéro, J. and Kramer, S. N. 1989 Lorsque les dieux faisant l’homme. Paris: Éditions Gallimard 2004 The Literature of Ancient Sumer. Oxford: Oxford University Press.

230

Maria de Fátima Rosa & Isabel Gomes de Almeida

Claeys, G. 2013 Three Variants on the Concept of Utopia. Pp. 14–18 in Dystopia(n) Matters: On the Page, on Screen, on Stage, ed. F. Vieira. Cambridge: Cambridge Scholars Publishing. Dalley, S. 2000 Myths from Mesopotamia. Creation, the Flood, Gilgamesh and Others. Oxford: Oxford University Press. Davis, L. 2013 Dystopia, Utopia and Sancho Panza. Pp. 23–27 in Dystopia(n) Matters: On the Page, on Screen, on Stage, ed. F. Vieira. Cambridge: Cambridge Scholars Publishing. Dickson, K. 2007 Enki and Ninhursag: The Trickster in Paradise. JNES 66/1: 1–32. Durand, J.-M. 1993 Le Combat entre le Dieu de l’Orage et la Mer. MARI 7: 41–61. ETCSL 2006 The Electronic Text Corpus of Sumerian Literature. Oxford: Faculty of Oriental Studies, University of Oxford. See http://etcsl.orinst.ox.ac.uk/ [Accessed: 2017/06/25]. George, A. 2003 The Babylonian Gilgamesh Epic: Introduction, Critical Edition and Cuneiform Texts. Oxford: Oxford University Press. Kramer, S. N. 1945 Enki and Ninḫursag: A Sumerian “Paradise” Myth. Bulletin of the American Schools of Oriental Research. Supplementary Studies 1. New Haven: American Schools of Oriental Research. 1956 History Begins at Sumer. Philadelphia: University of Pennsylvania Press. 1964 The Indus Civilization and Dilmun, the Sumerian Land. Expedition Magazine 6/3: 44–52. Lambert, M. and Tournay, J. 1949 Enki et Ninḫursag. RA 43: 105–136. Lutwack, L. 1984 The Role of Place in Literature. Syracuse NY: Syracuse University Press. Meslin, M. 1973 Pour une science des religions. Paris: Éditions du Seuil. Michalowski, P. 1989 The Lamentation over the Destruction of Sumer and Ur. MC 1. Winona Lake, IN: Eisenbrauns. 2008 The Mortal Kings of Ur: A Short Century of Divine Rule in Mesopotamia. Pp. 33–45 in Religion and Power: Divine Kingship in the Ancient World and Beyond, ed. N. Brisch. OIS 4. Chicago: The Oriental Institute of the University of Chicago. More, T. 2014 Utopia. Translated by C. H. Miller. Yale: Yale University Press. Rosa, M. F. 2015 A percepção da ordem e a consciência do tempo em Mari no período paleo-babilónico. Ph.D. dissertation. New University of Lisbon. Roux, G. 1992 La Mésopotamie. Paris: Éditions du Seuil. Sargent, L. T. 2010 Utopianism. A Very Short Introduction. Oxford: Oxford University Press. 2013 Do Dystopias Matter? Pp. 10–13 in Dystopia(n) Matters: On the Page, on Screen, on Stage, ed. F. Vieira. Cambridge: Cambridge Scholars Publishing.

Utopic and Dystopic Images in Mesopotamian Literature

231

Sasson, J. M. 2006 Utopian and Dystopian Images in Mari Prophetic Texts. Pp. 27–40 in Utopia and Dystopia in Prophetic Literature, ed. E. Ben Zvi. Helsinki: Finnish Exegetical Society. Talon, P. 2005 Enūma Eliš. SAACT 4. Helsinki: The Neo-Assyrian Text Corpus Project. Toorn, K. van der. 2007 Why Wisdom Became a Secret: On Wisdom as a Written Genre. Pp. 21–32 in Wisdom Literature in Mesopotamia and Israel, ed. R. J. Clifford. Atlanta: Society of Biblical Literature. Vanstiphout, H. L. J. 1997 Why Did Enki Organize the World? Pp. 117–134 in Sumerian Gods and Their Representations, eds. I. Finkel and M. Geller. CM 7. Groningen: Styx. Vieira, F. 2010 The Concept of Utopia. Pp. 3–27 in Cambridge Companion to Utopian Literature, ed. G. Claeys. Cambridge: Cambridge University Press. 2013 Introduction. Pp. 1–7 in Dystopia(n) Matters: On the Page, on Screen, on Stage, ed. F. Vieira. Cambridge: Cambridge Scholars Publishing.

In Search of Ur in iqqur īpuš Jeanette C. Fincke

CNRS (IVRY-SUR-SEINE)

Introduction The written documentation of omens begins with the widespread usage of the Akkadian language for literary texts on cuneiform tablets. Thus, although we know that the Sumerians used divination, they never wrote down the ominous signs together with their prediction in their language, and there are no omen texts from the third millennium BCE. Tablets with omens are attested from the Old Babylonian period in the first half of the second millennium BCE until the Seleucid (late fourth– second century) and even the Arsacid (second–first century) periods towards the end of the first millennium BCE. They can be found in the libraries of private scholars, but also in those of temples and—where most of our sources come from—the royal libraries of the Neo-Assyrian kings.

The Series iqqur īpuš (Table 1) The series iqqur īpuš (Labat 1965) is one of the most peculiar omen series of the cuneiform culture. Not only is its title not taken from the first words of its composition as would be normal in Mesopotamian literature, it also combines events and ominous signs from two spheres, the earth and the sky, in one series. The other omen series focus on one of these spheres only. And since the possible ominous signs that could be observed on earth are so multifarious, there are different omen series dealing with the individual bodies where ominous signs can occur (see, e.g., Maul 2003: 45b–88a). The omen series iqqur īpuš interprets these ominous signs strictly in relation to the month in which they occur, while the other omens series only occasionally do so. An even more peculiar feature of this omen series is the representation in three different formats (see Table 1). The so-called série mensuelle organizes the omens according to the months in which the ominous signs occur, resulting in twelve indiAuthor’s note: This article has been written as part of the ERC project Florential directed by Dr. habil. Robert Hawley at the Centre Nationale des Researches Scientifique in Ivry-sur-Seine. While completing the article in Leiden, I was able to draw on the excellent library of the Nederlands Instituut voor het Nabije Oosten as a Visiting Research Fellow, for which I would like to thank the director of the NINO, Prof. Dr. Jesper Eidem.

233

234

Jeanette C. Fincke

vidual tablets, one for each month of the year; a tablet referring to the intercalary month has not yet been identified. The série générale uses the ominous sign as the organisation principle, resulting in paragraphs of twelve to thirteen lines of omens—depending on whether or not the intercalary month has been considered. The third format of iqqur īpuš is the tabular form. This presentation form places the ominous sign in the abscissa and the months in the ordinate. Thus the scribes were able to simply tick the boxes formed by the lines and single out the positive months by writing “positive” (še) for them. Contrary to the other two representation forms, these tablets seem to use only terrestrial signs and omit the celestial omens. 1 Many of these fragments combine this tabular form with a short version of the Babylonian Almanac or rather the List of Lucky Days on the reverse of the tablets by listing the positive or lucky days of each month. 2 Some even combine such a compilation with a commentary. 3 However, there are also tablets in tabular form that clearly have ippuš īpuš omens written on both sides next to the List of Lucky Days. 4 Of course, there are also excerpt tablets, commentaries, clear forerunners and related compositions to the series iqqur īpuš that use the ominous sign as the organising principle. Either the apodoses or a positive interpretation (še) predict the consequences of the omen. Table 1. The omen series iqqur īpuš. Compilation of omens in two parts: [A] terrestrial and [B] celestial events.

Presentation Format

Organization Principle

Prediction

Série mensuelle [A and B]

The Month

Apodoses

Série générale [A and B]

The Ominous Sign

Apodoses

Tabular Form [A] + List of Lucky Days

Ominous Sign + Month

Positive (še)

Excerpts, Commentaries, etc. [A, B, or A and B]

The Ominous Signs

Apodoses or Positive (še)

The iqqur īpuš Tablets from Ur Both the série mensuelle and the série générale were attested until the Late Babylonian period. Two fragmentary examples from the série mensuelle have been found in Ur. One of the fragments refers to an unknown month (A), while the other one deals with month simānu (B). The preserved omens on both fragments com1. One fragment found in Tell Tayinat mentions the last night watch on its reverse (T-1930 rev. 5) in a phraseology that gives reason to expect that the other two night watches have also been mentioned. Such a reference is usually part of the description of celestial phenomena, but can also be used in terrestrial observations. Since neither the beginning of this line nor any of the other preceding or following protases are preserved the context must remain unclear. For this text see Lauinger 2016: 243–44. 2.  For the Babylonian Almanac and the List of Lucky Days see Livingstone 2013: 5–101. 3.  The fragmentary early Neo-Assyrian example from Kalḫu written by the famous scribe Nabûzuqup-kēnu son of Marduk-šuma-libši towards the end of the eighth century BCE (MS 2226+ K. 98; CUSAS 32, no. 76) is a good example for such a compilation, since it says explicitly it has extracted text from the series šumma ālu ina mēlê šakin, iqqur īpuš and ab.še.ge.da (“it will be favorable”), see George 2016: 169a–172b, pls. CXLVIII–CXLIX. 4.  See the Neo-Assyrian examples found at Tell Tayinat (Lauinger 2016: 229–248); text no. 1 is the exemplar that is best preserved (see pp. 235–241).

In Search of Ur in iqqur īpuš

235

plement each other in their sequence. In fact, taken together we see that a large middle part with approximately twenty-six lines is missing. These two fragments are the only ones of the series iqqur īpuš fragments found in Ur, and although it seems hardly probable they are most likely not from the same tablet. 5 The way the left edge and the beginning of the lines are formed by impressing the first signs of the omens differ.

Text A: UET 6/3 921 The first fragment has been published as UET 6/3 921 (text A) (Schaffer-Ludwig 2016). This fragment written in Late Babylonian ductus is from the upper left part of the tablet with the upper edge missing. On the obverse the beginning of eleven protases are preserved, and on the reverse the beginnings of thirteen. None of the apodoses are preserved and therefore the month these omens refer to cannot be detected. The surface is rubbed off in large parts in a way that indicates the tablet was not fired in antiquity. The fragment measures 30+ × 51+ × 23(+) mm. Transliteration (collated) (The left column gives the numbers of the paragraphs within the series iqqur īpuš based on the first edition of this series by Labat 1965.) Obverse § 8 A 1′ [diš é i]ṣ-ṣ[í … § 9 A 2′ ⸢diš⸣ [é i]s-[suḫ § 10 A 3′ ⸢diš é tam-la⸣ [si.a … § 11 A 4′ ⸢diš é bi⸣ -i[r … § 12 A 5′ diš é bi man-m[a kúr-ir … § 13 A 6′ diš é bi man-ma [kúr-ir-ma lá ... § 15 A 7′ ⸢diš⸣ ana é bi gur-ú[r … § 16 A 8′ diš ana é ⸢bi⸣ ku4-u[b … § 17 A 9′ diš ana é bi gibil ku4-u[b .. § 18 A 10′ [diš an]a é ⸢bi gibil⸣-i[š ... § 22 A 11′ [diš id]im b[i kúr-ir … (remainder is missing) Reverse § 92 A 1′ [diš diškur g]ù-⸢šú⸣ š[ub-ma ki zé i-ḫi-il § 93 A 2′ [diš diškur g]ù-šú š[ub-ma ki làl i-ḫi-il/-la § 94 A 3′ ⸢diš diškur gù⸣-š[ú šub-ma ki lu esir lu esir.ud.da lu ì.kur i-ḫi-il § 95 A 4′ diš an-e š[ur-nun … § 97 A 5′ diš ⸢ud⸣.7.kám an-⸢u⸣ [gin-ik … § 96 A 6′ ⸢diš⸣ u[d.29].kám : ud.30.[kám an-ú šur-nun … § 98 A 7′ ⸢diš⸣ [im].dugud ⸢iq⸣-t[ur … § 99 A 8′ diš ⸢im⸣.u18.lu g[in-ik … § 100 A 9′ diš ri-i-bu i-r[u-ub … § 101 A 10′ diš ki i-ru-ub [… § 102 A 11′ [diš i]m.⸢gú⸣ kur i[s-ḫup … 5.  I would like to thank Christopher Walker for sharing his opinion on a possible non-physical join of both fragments. He agreed with the given estimation.

236

Jeanette C. Fincke § 103 A 12′ [diš il]lu ⸢gin⸣-k[am … § 104 A 13′ [diš ill]u gi[n-kam a-šú gim úš ṣar-pu … (remainder is missing)

Translation

Obverse § 8 A 1′ [If he pu]lls dow[n a house, …] § 9 A 2′ If [he di]gs [out a house, …] § 10 A 3′ If [he fills] the terrace (for) a house, […] § 11 A 4′ If he ges his house, […] § 12 A 5′ If [he moves] into anothe[r] house of his, […] § 13 A 6′ If [he moves] into another house of his [and watches over it, …] § 15 A 7′ If he return[s] to his house, […] § 16 A 8′ If he (re-)enter[s] his house, […] 6 § 17 A 9′ If he enter[s] his renewed house, […] § 18 A 10′ [If] he renew[s] (something) [f]or his house, […] § 22 A 11′ [If he changes] hi[s id]im, […] (remainder is missing) Reverse § 92 A 1′ [If Adad] lets his [thu]nder (lit. cry) res[ound and the earth exudes bile, …] § 93 A 2′ [If Adad] lets his [thu]nder (lit. cry) res[ound and the earth exudes honey, …] § 94 A 3′ If Adad [lets] hi[s] thunder (lit. cry) [resound and the earth exudes iṭṭu-bitumen or kupru-bitumen or naphta, …] § 95 A 4′ If the sky ra[ins, …] § 97 A 5′ If at the 7th day the sky [moves, …] § 96 A 6′ If at the [29th] da[y] or the 30th day [the sky rains, …] § 98 A 7′ If [f]og “smok[es (the sky),” …] § 99 A 8′ If the south wind bl[ows, …] § 100 A 9′ If an earthquake quak[es, …] § 101 A 10′ If the earth quakes, […] § 102 A 11′ [If] silt sp[reads over] the earth, […] § 103 A 12′ [If high] water come[s, …] § 104 A 13′ [If high wa]ter com[es and its water is like blood, …] (remainder is missing)

Comments The preserved parts of this fragment run basically parallel to other manuscripts of the série mensuelles from Assur and Babylonia. Therefore, no conclusions can be drawn from this fragment concerning the tradition of the series.

6.  This protasis has been explained by the Babylonian commentary BRM 4 24 (MLC 2627) II 16′-17′ as follows: 16′ diš ana é-šú ku4 šá saḫar.šub.ba si.a-ma 17′ dadag-ma ana é-šú ir-ru-ub, “If he (re-)enters his house (means) that he was full of saḫar.šub.ba (a skin disease most probably characterized by patches of silvery-white scaly skin), he (then) was cleaned from it and (now) (re-)enters his house.”

In Search of Ur in iqqur īpuš

237

Text B: U 30714 (Figs. 1 and 2) The second fragment of a série mensuelles tablet of iqqur īpuš from Ur (U 30714, text B, see Figs. 1–2) is the lower left corner of a single column tablet measuring 40+ × 43+ × 23(+) mm. It is written in a beautiful hand in Late Babylonian ductus. The surface of the fragment is extraordinarily well preserved; it must have been fired at some point. On both the obverse and the reverse the beginnings of fourteen lines are preserved, giving a sequence of twenty-seven omens. The month this copy of the série mensuelle is dealing with can only be traced on reverse line 1, since the protases of all other lines are lost or unspecific. This fragment refers to month simānu, the third month of the Babylonian year that is corresponding to May / June. Transliteration (The left column gives the numbers of the paragraphs within the series iqqur īpuš based on the first edition of this series by Labat 1965. This I have expanded to include paragraphs from new sources.) Obverse § 66 B 1′ ⸢diš izi⸣ [ana é.gal šub-ut … § 65 B 2′ diš izi a[na é na šub … § 59 B 3′ diš 29.ká[m ká nu è … § 52 B 4′ diš ki iz[i(over erasure) šub-di … § 44 B 5′ diš ud-diš ⸢dù⸣-[uš … § 47 B 6′ diš giš.kiri6 i[na šà uru iz-qup … § 45 B 7′ diš giš.kiri6 [iz-qup … § 46 B 8′ diš giš.ki[ri6 giš.gišimmar iz-qup … § 48/49 B 9′ diš giš.ki[ri6 gibil-iš .... : diš giš.kiri6 ud-diš … § 57’’ B 10′ diš sag.ḫu[l.ḫa.za igi … § 58 B 11′ diš muš i[gi … § 42 B 12′ diš qí-šub-⸢ba-a bad⸣-[ma ana a.šà ú-tir … § ? B 13′ diš nun? ud-⸢diš NE?⸣-x [… § 54’ B 14′ ⸢diš⸣ zé [… Reverse § 54’’ B 15′ diš šu.si záḫ n[am.gilim.ma ina kur gar kur tur] § 75 B 16′ diš 30 ina igi.lá-šú a-d[ir … § 77 B 17′ ⸢diš⸣ an.ge6 guš.ḫur nigin-m[i … § 77 B 18′ diš an.ge6 giš.ḫur nigin-mi gu[run qu-ma-nu dib-bat] § 78 B 19′ diš 30 giš.ḫur nigin-mi an.ge6 [en.nun ud.zal.le gar šèg.meš u illu.meš] B 20′ l[ál.meš] § 80 B 21′ diš 30 lu tùr lu ⸢amaš⸣ n[ígin … § 67 B 22′ diš 30 ud.14.kám lu ud.15.kám k[i 20 nu igi … § 68 B 23′ diš 30 ud.30.kám igi ṭu[ḫ-da mar.tu.ki aḫ-la-mu-ú gu7] § 69 B 24′ diš 30 an.ge6 gar-un illu [ina kur gál-ši bi-bil a.meš kur ub-bal] § 70 B 25′ diš an.ge6 bi sa5 gú.[gal si.sá] § 71 B 26′ diš an.ge6 en.nun ⸢an⸣.ú[san gar-un … § 72 B 27′ diš an.ge6 en.⸢nun⸣ [murub4.ba gar-un … § 73 B 28′ ⸢diš an.ge6 en⸣.[nun ud.zal.la gar-un … (remainder is missing)

238

Jeanette C. Fincke

Fig. 1. U 30714, obverse.

Translation

Fig. 2. U 30714, reverse.

Obverse § 66 B 1′ If fire [falls on the palace, …] § 65 B 2′ If fire [falls] o[n the house of a man, …] § 59 B 3′ If [he does not goes out of the door] on the 29th , […] § 52 B 4′ If [he sets up the brazi[er, …] § 44 B 5′ If he renews and “make[s,” …] § 47 B 6′ If [he plants] an orchard i[n the middle of a city, …] § 45 B 7′ If [he plants] an orchard, […] § 46 B 8′ If [he plants] an orch[ard with date palms, …] § 48/49 B 9′ If [he renews] an orch[ard, … : If he renews an orchard, …] § 57’’ B 10′ If [he sees] an sag.ḫ[ul.ḫa.za-demon, …] § 58 B 11′ If he s[ees] a snake, […] § 42 B 12′ If he opens waste land [and turns it into agricultural area, …] § ? B 13′ If the ruler renews, ..[…] § 54’ B 14′ If the gall bladder [is lost, …] Reverse § 54’’ B 15′ If the “finger” is lost, de[struction will be placed into the country; the country will become small (i.e. insignificant).] § 75 B 16′ If the moon is blur[red] at its appearance, […] § 77 B 17′ If a (lunar) eclipse is surrounded by an uṣurtu-halo, […] § 77 B 18′ If a (lunar) eclipse is surrounded by an uṣurtu-halo, [the qummānu-disease will affect] the fru[it]. § 78 B 19′ If the moon is surrounded by an uṣurtu-halo and a (lunar) eclipse [occurs during the morning watch, rain and high water]

In Search of Ur in iqqur īpuš

239

B 20′ will be red[uced.] § 80 B 21′ If the moon is sur[rounded] by a “cattle pen”-halo or by a “sheepfold”-halo, […] § 67 B 22′ If the moon [is not seen] toget[her with the sun] on the 14th day, […] § 68 B 23′ If the moon is visible on the 30th day, [the Aḫlameans will consume] the abund[ance of Amurru.] § 69 B 24′ If the moon eclipses (lit. sets an eclipse), [there will be] high water [in the country; the devastating flood of the water will carry away the soil.] § 70 B 25′ If this (lunar) eclipse is red, the (chick)peas [will thrive.] § 71 B 26′ If a (lunar) eclipse [occurs] during the eveni[ng watch, …] § 72 B 27′ If a (lunar) eclipse [occurs] during the [middle] watch, […] § 73 B 28′ If a (lunar) eclipse [occurs] during the [morning] wa[tch, …] (remainder is missing)

Comments Paragraph numbers of § 77 (ll. 17′-18′) and § 80 (l. 21′) are written in italics becaus the protases vary slightly from those of the other manuscripts. § 77 usually refers to the moon encircled with an uṣurtu-halo, and not a lunar eclipse as is the case here. § 80 usually begins with an uṣurtu-halo, followed by what is written here: a tarbāṣu or animal-stall halo (tùr) and a supūru or sheepfold-halo (amaš). Contrary to fragment A (see above), this fragment has a sequence of omens that differs in the various manuscripts from Nineveh, Assur, Babylon and Sippar. An analysis of the sequence can be informative for the stream of tradition. Both parts, the terrestrial and the celestial, will be examined separately.

The Group of Terrestrial Omens from Text B (Table 2) The omens referring to events on earth can be found in lines 1′–15′ of text B. They follow a sequence found towards the end of the terrestrial omens in the série mensuelles. Table 2 shows an overview on this sequence of omens preserved in various manuscripts dating to the first millennium BCE. The four columns on the left of this chart refer to texts written in Neo-Assyrian ductus found in Nineveh and Assur. The fourth column refers to the omens attested in the related series inbu bēl arḫi, “Fruit, Lord of the Month,” a series that was compiled for the Neo-Assyrian king by borrowing from the terrestrial omens of iqqur īpuš and hemerologies (see Livingstone 2013: 199–248). The six columns to the right refer to sources from Babylonia written in Neo- or Late Babylonian ductus, originating from an unknown site in Babylonia, from Sippar and from Babylon. The extreme right column refers to fragment B from Ur. In Table 2, those paragraphs that are attested in the different copies in different positions within the sequences are underlined grey. In general, the fragment from Ur follows the sequence attested in Nineveh—second column from the left—as well as in Sippar and Babylon. Still, there are some differences. The two paragraphs

240

Jeanette C. Fincke

Table 2. Sequence of omens towards the end of the terrestrial part.

iqqur īpuš – série mensuelle (1st millennium BCE) NA sources Nineveh month I

month II + II

§ 41 § 42

§ 41 § 42

§ 43 § 52

§ [...] § 43 § 52

§ 38 § 32 § 30

NB and LB sources

Assur

i.b.a.

Babyl. (?)

month III + I

month VIII / III

month I

§ 46 § 49 §A

§ 42 § 46 § 49 §A

§ 52

§ 52

§ 55

§ 38 § 32 § 30

§ 33 § 66 § 65 § 59

§ 32

§ 55

§ 54′ § 54″ § 44 § 47 § 45 § 46 § 48 § 49

§ 60

§ 44 § 47 § 45 § 46 § 48 § 49 § 57’’ § 58 § 60

§ 55

§ 54′ § 54″

§ 57’’ § 60 §B

Sippar

Bab.

Ur

month I month month month I month IV V/X III § 41

§ 41 § 42

§ 41

§ 43 § 52 § [...] § 57 § 56 § 55 § 38 § 32

§ 43 § 52

§ 38 § 32 § 30

§ 43 § 52 § 58 § 57'' § 56 § 55 § 38 § 32 § 30

§ 33 § 66 § 65 § 59

§ 57 § 56 § 55 § 38 § 32

§ 33 § 66 § 65 § 59

§ 33 § 66 § 65 § 59

§ 56 § 55 § 33 § 66 § 65 § 59

§ 54′ § 54″

§ 54′ § 54″

§ 54′ § 54″

§ 44 § 47 § 45 § 46

§ 44

§ 44 § 47 § 45 § 46 § 48 § 49

§ 60

§ 54′ § 54″ § 42 § 44 § 47 § 45 § 46 § 48 § 49 § 57’’ § 58 § 60

§ 66 § 65 § 59 § 52

§ 44 § 47 § 45 § 46 § 48 § 49 § 57’’ § 58 § 42 §? § 54′ § 54″

In Search of Ur in iqqur īpuš

241

referring to extispicy, § 54′ (lost gall bladder) and § 54′′ (lost “finger”), that follow paragraph § 59 (not going out through a gate on day 29) in the other sources, is moved to the end of the sequence of terrestrial omens in the fragment B from Ur. Furthermore, § 52 (setting up a brazier) that appears much earlier in the other sources, is moved down to follow § 59 (not going out through a gate on day 29) in the fragment from Ur. Finally, § 42 (opening of waste land and turning it into agricultural area) that comes much earlier in the copies from Nineveh, Assur, and Sippar, but after § 54′ (lost gall bladder) and § 54’’ (lost “finger”) in the version from Babylon, has been moved towards the end in the fragment from Ur. Summing up, within the terrestrial omens, the version from Ur has undergone a redaction resulting in minor changes of the sequences, but in general it is quite close to the one from Sippar.

The Group of Celestial Omens from Text B (Table 3) Lines 16′–28′ of fragment B from Ur has omens referring to celestial phenomena. Table 3 has an overview of the relevant omens found in other manuscripts from the first millennium BCE from various places in Assyria and Babylonia. As is the case with Table 2, the first four columns on the left refer to sources written in Neo-Assyrian ductus from Assyrian places. Instead of using the series inbu bēl arḫim (which does not include celestial omens) an example from Ḫuzirīna is given in the fourth column. The right columns all refer to sources written in Neo- or Late Babylonian ductus, with fragments B from Ur in the very last column. Again, those paragraphs that occur in different positions within the sequences of the different sources are marked in grey. In general, the chart of this part of the series has more grey paragraphs than the one referring to the terrestrial omens, but even so the variation is tolerable. The source from Babylon referring to month 8—the second column from the right—differs considerably from all the other sources; it represents a kind of short version of the série mensuelle. The fragment from Ur begins its sequence of celestial omens with § 75 (a moon blurred as it rises), a paragraph that has been inserted later in the other sources; this paragraph can occur in various other positions in the different manuscripts. The same is true for two other paragraphs, § 77 (an eclipse surrounded with an uṣurtu-halo) and § 68 (the moon visible on day 30). The sequence §§ 75 – 76 – 77 – 78 7 or §§ 75 – 76 – 78 can appear in various positions. This indicates that the scribe of the tablet from Ur could have made a mistake by omitting § 76 (blurred moon) and writing a variant of § 77 twice instead. Another paragraph that can occur in different positions within the sequence is § 67 (the moon and the sun not been seen together on day 14). In this case, however, the fragment from Ur follows the majority of the other sources by positioning § 67 directly after § 80 (the moon surrounded by a “cattle-pen” or by a “sheepfold”-halo), and not before as one copy from Sippar does.

7.  § 75 (a moon blurred as it rises) – § 76 (blurred moon) – § 77 (the moon / an eclipse surrounded with an uṣurtu-halo) – § 78 (the moon surrounded with an uṣurtu-halo in combination with a lunar eclipse in the morning).

242

Jeanette C. Fincke

Table 3. Sequence of paragraphs of the first celestial omens.

iqqur īpuš – série mensuelle (1st millennium BCE) NA sources Nineveh

Assur

month I month month III IX

NB and LB sources Ḫuzirīna Babyl. (?) month VI

§ 77 § [...] § 80 § 67 § 68

§ 67 § 68 § [...] § [...]

§ 69

§ 69

§ 71 § 72 § 73 § 74

§ 70 § 71 § 72 § 73 § 74

§ 75

§ 79

§ 79

Sippar

Babylon

Ur

month I month month month month month month IV V XI IV VIII III § 75 § 77 § 77 § 77a § 77 § 77a § 77 § 67 § 78 § 80 § 80 § 80 § 80 § 80 § 77 § 77 § 67 § 67 § 67 § 67 § 68 § 68 § 68 § 68 § 68

§ 75 § 69

§ 69 § 81′ § 70 § 71 § 72 § 73 § 74

§ 69

§ 69

§ 70 § 71 § 72 § 73 § 74

§ 70 § 71 § 72 § 73 § 74

§ 76

§ 75 § 76

§ 75 § 76

§ 78 § 79

§ 78 § 79

§ 78 § 79

§ 70 § 71 § 72 § 73 § 74 § [...] § [...]

§ 79 § 75 § 76 § 78

§ 69

§ 69 § 70 § 71 § 72 § 73

§ 77

Summing up, we note that the sequences of omens in the série mensuelles from iqqur īpuš is not exactly fixed except for the terrestrial ones always preceding the celestial ones. Within these two parts a generally consistent sequence is followed, with individual paragraphs that are movable. Clusters of sequences were moved en bloc without separating individual paragraphs.

Unilingual Akkadian Literary Tablets from Ur (Table 4) Since both fragments from the series iqqur īpuš are written in Akkadian just as all other omen texts, the unilingual Akkadian literary tablets and fragments un­

In Search of Ur in iqqur īpuš

243

earthed at Ur need be reviewed to understand the significance of these copies. Table 4 gives an overview based on the publication of the tablets in the series Ur Excavation Texts. 8 Not many unilingual Akkadian literary texts have been unearth­ed at Ur. Dating to the Old Babylonian period there are only thirty examples, and from the Late Babylonian period only twenty-two. Among these is only one omen text from the Old Babylonian period, and three from the Late Babylonian period. The Old Babylonian omen fragment is a compendium of liver omens focussing on the ṣibtum, the processus pappilaris of the liver (U 443 (7752)+U 7759; see Jeyes 1989: 3, 137–143, pls. 10–11 no. 10). The third Late Babylonian omen text is an almost complete tablet in landscape format with lunar eclipse omens written in the tenth year of Šamaš-šum-ukīn (UET 6/2 413). Dr. Jonathan Taylor, curator of the British Museum’s Middle East Department who is conducting a project on Ur and its finds together with a team in Philadelphia, has told me that among the unpublished pieces housed at the British Museum there are many more fragments from lexical texts, but no omens text. So we should not expect to find any more fragments of iqqur īpuš tablets from Ur at the British Museum. Table 4. Unilingual Akkadian literary texts and fragments from Ur (based on the published material).

OB

LB

Religious Texts (Hymns, Myths, Prayers, Etc.)

9

2

Epics

1

1

Medical Texts

5

1

Incantations

1

6

Musical Instructions, Dialogues | Monologue

2

1

Law Code

1

1

Lexical Texts

1+?

1+?

List of ziqpu Stars

—

1

Omens

1

3

Undetermined Content

9

5

Total

30+?

22+?

Findspots of the Late Babylonian Akkadian Literary Texts from Ur The number of Late Babylonian unilingual Akkadian texts is small compared with the number of Late Babylonian Sumerian and bilingual texts, which is three to four times larger. This raises the question of where these tablets came from. According to publications, the find spot of only a very small number of these tablets is identified, but they came from different areas in Ur. Since temple libraries usually have literary texts, the excavation reports should have given more information: both 8.  UET 6/1 (Gadd – Kramer 1963), UET 6/2 (Gadd – Kramer 1966) and UET 6/3 (Schaffer – Ludwig 2006). See the very detailed overview on the relevant material published not in the series UET, but in other places provided by Marie-Christine Ludwig in UET 6/3: 25–26.

244

Jeanette C. Fincke

the Gipar temple and the Enunmaḫ temple have Neo- and Late Babylonian levels. More Neo- and Late Babylonian levels have been traced in the south and in the north-east of the temple area. This means that there are at least three possible locations for libraries, two temples and the house of Bēl-šalti-Nannar north-east of the temples. Small collections of literary texts could also have been stored in the private houses towards the south. The results of the project on Ur conducted by the British Museum and the University Museum in Philadelphia, will hopefully give us more information about the scholarly libraries at Ur. But for the moment we consider that the very small number of twenty-two unilingual Akkadian literary texts found in these areas shows that there must have been many more libraries of Neo- and Late Babylonian scholars at Ur than those we know today. There is still much work to be done on Ur.

Bibliography Gadd, C. J. and Kramer, S. 1963 Literary and Religious Texts First Part. UET 6/1. London: Trustees of the British Museum and of the University Museum, University of Pennsylvania, Philadelphia. 1966 Literary and Religious Texts Second Part. UET 6/2. London: Trustees of the British Museum and of the University Museum, University of Pennsylvania, Philadelphia. George, A. R. 2016 Mesopotamian Incantations and Related Texts in the Schøyen Collection. CUSAS 32. Bethesda, MD: CDL. Jeyes, U. 1989 Old Babylonian Extispicy: Omen Texts in the British Museum. PIHANS 64. Leiden: NINO. Labat, R. 1965 Un calendrier babylonien des travaux des signes et des mois (séries iqqur īpuš). Paris: Libraire Honoré Champion. Lauinger, J. 2016 Iqqur īpuš at Tell Tayinat. JCS 68: 229–248. Livingstone, A. 2013 Hemerologies of Assyrian and Babylonian Scholars. CUSAS 25. Bethesda, MD: CDL. Maul, S. M. 2003 Omina und Orakel. A. Mesopotamien. RlA 10/1–2: 45b–88a. Schaffer, A. and Ludwig, M.-Chr. 2006 Literary and Religious Texts Part Three. UET 6/3. London: Trustees of the British Museum and of the University Museum, University of Pennsylvania, Philadelphia; British Museum Press.

The Kingdom as Sheepfold: Frontier Strategy Under the Third Dynasty of Ur; A View from the Center Steven J. Garfinkle

Western Washington University My goal in this contribution is to examine what we can deduce from the vast cor­ pus of texts from the last century of the Third Millennium BCE about the strategic goals of the Ur III kings and their state. My basic premise is that the kingdom of Ur was a failed experiment in state building that relied at home on the charisma and patronage of its kings, and abroad on the ability of those kings to extract tribute and booty from the people who lived along its frontiers and from outlying communities beyond the boundaries of southern Mesopotamia. Frontiers are important to our historical reconstructions of ancient Mesopota­ mia. They are the places where we imagine both threats and opportunities lay for the growing states of southern Mesopotamia in the Third and early Second Millen­ nia BCE. Piotr Michalowski (2011: 127) noted, The notion of the frontier is critical to our understanding of salient aspects of the Mesopotamian geographical imagination. A frontier is not a line in the sand but a liminal zone, an area of multifaceted interaction that can unite as much as it separates, and is often the place where new cultural and political orders can establish themselves and grow…

The view presented here suggests that the relationship with the frontier was critical to the internal development of early Mesopotamian conceptions of community and political order. Frontiers, or more properly for our consideration today, the space beyond the irrigated core of southern Mesopotamia, were also significant to their understand­ ing of the world around them. In literature, the highland frontiers were the locus for the activity of mythical creatures, both divine and monstrous. The archetypal text about identity in ancient Mesopotamia, the Epic of Gilgamesh, contrasted the civilization of the urban space and its immediate hinterland with the dangerous places outside of that zone. We see this clearly in tablet ten when Gilgamesh was so transformed by his passage through steppe and mountain that Shiduri saw him and barred the gate to her tavern. 1 And yet, it is important to be precise about where Author’s note: A version of this paper was presented at an earlier symposium, and I am grateful to Piotr Michalowski for suggesting improvements. I dedicate this contribution to him for all that he has added to our knowledge of early Mesopotamia. All errors are my own. 1.  Tablet X 13–16, see George 1999: 76.

245

246

Steven J. Garfinkle

the boundaries were placed in the epic between civilization and chaos. In the first tablet, Gilgamesh was reminded of his responsibility to protect the hunters who lived beyond the watered fields; and in his entry into Mesopotamian society, Enkidu took on the task of protecting the shepherds, whose mobile pastoralism would seem to lie outside of our easy definitions of Mesopotamian space. This takes on added sig­ nificance with the repeated declaration that Uruk, where Enkidu can take his place among men, is not only the “Town Square” but is also more often, “The Sheepfold”! 2 This idea of the sheepfold is one to which I will repeatedly return, and of course, it correlates well with one of the most frequent epithets for kingship in the late Third and early Second Millennia: the king as shepherd of the people. 3 This imagery was a rich part of the literary record associated with the Ur III kings. The Shulgi hymns make clear that the king was not only shepherd of the people, but someone familiar with shepherding. 4 He protected the sheepfold and was the killer of lions too numerous to count. And he was connected not just to shepherding, but to Gilgamesh: “I am Shulgi, good shepherd of Sumer. Like my brother and friend Gilgamesh, I can recognize the virtuous and I can recognize the wicked.” 5 The idea of the king as shepherd, which was so prominent in the surviving royal inscriptions and hymns, also tied the deified kings of Ur to their legendary predecessors and to the gods. Nanna, the moon god of the city of Ur, was himself closely associated with livestock. The hymn Nanna A ends with the following line: “Your moonlight is holy and bright, and because like Utu you are a shepherd of the Land, Nanna, it shines forth for the king like the daylight.” 6 In Nanna F, also known as “The herds of Nanna,” he is shown to possess enormous cattle pens, and the text numbers his penned and his wild cows in the hundreds of thousands. Near the end of the text, Nanna received the following praise: “He is ever able to increase the butter of abundance in the holy animal pens…” 7 The kings of Ur had a divine mandate to increase the size of their herds, and to protect the pens and pasturelands in which this wealth was maintained. My goal here is to explore how the frontier and frontier strategy were connected to the act of being a “good shepherd of Sumer” and how this in turn influenced the development of community under the Third Dynasty of Ur. One of the most pressing problems that the Ur III dynasts faced in consolidating their kingdom was spatial, and here I am referring not only to physical space but also to social space. 2. See, for example, Tablet I for the introduction of Uruk-the-Sheepfold, and see Tablet II for Uruk-the-Town-Square. 3.  The kings of Ur, from Ur-Namma onwards, styled themselves in royal inscriptions as the “Mighty man, king of Ur, king of Sumer and Akkad;” increasingly, from Shulgi onwards, they added “king of the four quarters.” And by the time of Shu-Suen, the king was also “shepherd of the land.” To survey the Ur III royal inscriptions, see Frayne 1997. In the epilogue to the laws of Ur-Namma, the punishments for erasing the inscription include “may no one build a sheepfold,” see Civil 2011: 252. By the beginning of the Second Millennium BCE, we find the king as shepherd included among the royal epithets in the prologues to the so-called Law Codes, see, for example, Roth 1997: 25 and 77. 4.  The king as shepherd also appeared in the royal correspondence. For example, one of Aradmu’s letters to Shulgi names him as the “rightful shepherd of the people,” see Michalowski 2011: 341–342. 5.  Shulgi C, see http://etcsl.orinst.ox.ac.uk/cgi-bin/etcsl.cgi?text=t.2.4.2.03#. For the praise poetry of the Ur III period, see the Electronic Text Corpus of Sumerian Literature: http://etcsl.orinst.ox.ac.uk. 6.  For Nanna A, see http://etcsl.orinst.ox.ac.uk/cgi-bin/etcsl.cgi?text=t.4.13.01#. 7.  For Nanna F, see http://etcsl.orinst.ox.ac.uk/cgi-bin/etcsl.cgi?text=t.4.13.06#.

The Kingdom as Sheepfold

247

In the old cities of the heartland, there was not a lot of available space, practically or conceptually. The kings of Ur needed to create room for new constituencies bound to the larger idea of the kingdom, and they needed to find ways to reward those who carried the banner of the dynasty. The military arena was the primary area for this royal activity and advancement. The traditional elites in the cities of the heartland, despite their political subordination, remained entrenched players in the local economies, but campaigns beyond the frontier offered opportunity for the new royal elite. 8 The Ur III kingdom as we know it was a consequence of the social and economic effects of a state of near constant warfare. 9 The campaigns of the kings, which became virtually annual midway through Shulgi’s reign, allowed for the creation of new social networks and new sources of wealth. The latter was especially important both to the kings and the royal elite. Marc Van De Mieroop recently observed, In pre-industrial societies agriculture was the basis of the economy and most of it was subsistence agriculture in autarkic village communities and manors. There was very little surplus for trade in foodstuffs and fibers and whenever such trade existed, it was under the auspices of central institutions. 10

This is a crucial reminder that the fields of barley, which made urban life possible for the great cities of Mesopotamia, did not immediately translate into portable wealth. The tremendous herds of sheep and goats provided a great measure of the wealth that made commerce possible and profitable in early Mesopotamia. One of the most notable features of the various Ur III corpora is the prominence of livestock and their movement between households and institutions. The repeated military raids across the frontier, documented most clearly in the year names and administrative texts, brought in a great amount of livestock, as well as slaves. The former added to the enormous institutional herds of the kingdom while the latter provided labor in the workshops that processed the wool and hair into textiles. The texts from Puzrish-Dagan frequently document this new wealth as booty, tribute, and offerings from groups of men associated with the royal and military sector. The latter texts are especially significant to my reconstruction of the social networks that attended larger state formation in early Mesopotamia. 11 The core constituents of the new state were brought together on these tablets in a system of patronage. The lists of notables offering animals to the crown at Puzrish-Dagan, 8.  Garfinkle 2015: 157–160. 9.  Part of the larger argument I am making about the Ur III state is that it was a martial society engaged in nearly constant warfare, but that the goal of these military forays was not outright conquest and administration of frontier zones. The most decisive evidence that we have that the Ur III kingdom was not imperial in nature comes from the most frequent and mundane historical evidence at hand: the year names. I have discussed this elsewhere, so I will not belabor the point, but the year names demonstrate two things most directly: first, the frequency of warfare during the height of the dynasty’s success; and second, the indecisive nature of those wars. Let me explain the latter statement more fully: the Ur III kings went to war virtually every year of the kingdom, or at least they went on campaigns annually, and these campaigns were frequently directed against the same places year after year. Shulgi’s forty-fourth regnal year was named for the ninth defeat of Simurrum. On the raids, see Michalowski 2011 and Garfinkle 2014. 10.  Van De Mieroop 2015: 80. I am not convinced that central institutions must always have under­ taken this trade, but the larger point is critical to our understanding of early Mesopotamia. 11.  See Garfinkle 2013: 160–162; and 2015.

248

Steven J. Garfinkle

which might bring together governors, merchants, and Amorite dignitaries like Naplanum, memorialized their participation in this reciprocal arrangement. The lists were acts of commemoration. In light of the deification of the kings of Ur, this language takes on added significance. Indeed, they were delivering actual sheep to their royal shepherd. This new wealth was founded on the activities of two sig­ nificant groups that were central to the expanded notion of state and community: soldiers and shepherds. The soldiers carried out the raids that led to the acquisition of all of this booty and tribute. The shepherds were responsible for safeguarding all of this newfound bounty. These activities were critical not only to the socio-economic success of the state, but also to its ideological conception. And here, I will return to the idea of the state as sheepfold. The building of the line of fortifications that composed the bad3 mada was about the expansion of the sheepfold. 12 No longer could the state in southern Mesopotamia simply reflect the city as autarkic unit, now the kings of Ur oversaw a larger community that included not only the urban space, but also the mada that lay beyond it. The Ur III kings expanded the size of their state community in many ways. Some of this was conceptual, but clearly some of it was physical as well. The physical expansion of the kingdom that I envision here was not the imperial con­ quest that is often suggested. 13 Instead, I see an imperative to protect communities of mobile pastoralists and the different environmental zones to which they needed access: lowlands, steppe, and highlands. When we view Shulgi’s engagements with locations like Simurrum as acts of conquest, then the repeated return to the same place imagines a restless periphery eager to throw off the encroaching influence of an expansive state. When we see those same forays as raids they take on significance as the boasts of a king who could repeatedly ravage the same countryside and bring home booty to be spread among the court’s clients. The kings of Ur and their elites had a great deal of their wealth tied up on the hoof, so to speak. Anne Porter noted, It is inconceivable that even a significant portion of the numbers of animals im­ plied here would be herded as locally as the perimeters of the irrigated lands that constitute the locale of these city-states, for several reasons. Contested land us­ age and issues of carrying capacity are but two. 14

This highlights the dual problems that the kings of Ur sought to solve with their raids in the highlands. Within the old established heartland of Mesopotamia, there was less room for them to maneuver. They could not upset the traditional local and regional hierarchies and patterns of land use without undermining the local econo­ mies, and the principles of status on which their own power was based. So they needed new room for social networks bound directly to the crown. 12.  See Michalowski 2011: 122 ff. 13.  See note 10 above. Michalowski (2011: 96) reminded us that we are often at the mercy of ancient propaganda, most notably, in the victories celebrated in the year names; and this is compounded by our own assumptions expressed in the language we use in translation. The description of the Ur III state as empire implies military effort aimed at conquest, as when Porter (2012: 251) cautiously described the Third Dynasty of Ur as “history’s first great but short lived empire…” The idea of rapid territorial expansion under Shulgi has its clearest expression in Steinkeller 1991. 14.  Porter 2012: 243.

The Kingdom as Sheepfold

249

Moreover, they needed additional sources of wealth to support their royal en­ deavors and the new royal elites. The emphasis on military campaigning and booty solved both of these problems, but it not only relied on the success (ultimately the charisma) of the kings, it also created the problem of what to do to safeguard the animals. The areas to the east in which Shulgi campaigned “were, and still are, prime pastoralist territory.” 15 Indeed, good pasturage can still be found in much of what we regard as the frontier zone. Following Michalowski, I consider it no coincidence that the bad3 mada was built in the years immediately preceding the construction of Puzrish-Dagan. 16 The line of fortifications that Shulgi built was designed to protect the expanded sheepfold that was the Ur III kingdom and to provide opportunities to collect more booty. The expansion of the fortifications under Shu-Suen shows how important this idea was, but also how fragile. In this view, Iri-Sagrig, home to a dizzying array of soldiers, messengers, and merchants, became the prime city at the inner edge of this operational zone for organizing and staging the campaigns into the east, and for managing the wealth that these raids produced. Shulgi’s bad3 mada and Shu-Suen’s bad3 Martu Muriq-Tidnim were not ramparts in the traditional sense, but they served to create a divider between land that was under the protection of the king, and the wild lands on the other side. In Shulgi’s day, this was a successful venture, providing a staging ground for numerous raids into the lands beyond the frontier and providing protection for the growing flocks and their pasturelands. Shu-Suen’s efforts, by contrast, were an ambitious failure. After all, it is clear that building walls and securing the sheepfold had tremendous symbolic value to the kings of Ur, and the successes of Shulgi and Amar-Suen cre­ ated herds and allies, especially in the Diyala region, that needed to be secured. But the rest of the textual evidence from Shu-Suen’s reign suggests that he did not secure this area, nor was he able as consistently as his predecessors to raid the lands beyond the mada and bring back both the success and the booty that the state and its elites required. Michalowski and others have shown that for the late Third Millennium BCE there was no Amorite phenomenon that we can regard in a monolithic manner. Amorites were insiders and outsiders when viewed from the Ur III court. They included mobile pastoralists as well as individuals who were clearly more settled. Significantly, many of them must have lived and worked in the mada, on the protec­ ted side of the fortifications. Were the Tidnum then a kind of stand in for the sort of larger Amorite collective identity that has been a prominent part of the way historians have imagined this era and posited so strong a distinction between settled and nomad? Or, is it possible that they were an unallied group of mobile pastoralists who needed to be kept out of the expanded sheepfold of the Ur III kingdom and kept away from the Martu who already served within it, for example, men like Naplanum? In this view, one group of allied pastoralists was being offered shelter behind the fortifications, while 15.  Porter 2012: 296. Porter (2012: 243–244) also highlighted that this could not be done entirely within the land between the rivers, “we must understand the majority of the animals, that were used for wool… as having been variably dispersed across all available ranges, including upstream of the Tigris and Euphrates, in the Transtigridian lowlands and the Zagros mountains… wherever, in short, there was available pasturage not in use by, or that could be successfully wrested from, others.” 16.  See Michalowski 2011: 133.

250

Steven J. Garfinkle

another was being kept at bay. And it makes sense that all of this was taking place within the kur Martu (part of which was within the frontier zone) because of the herds and pasturelands there, the control of which had become embedded in notions of space and identity in the Ur III kingdom. Ultimately, my approach here asks us to envision what the projection of state power meant to kings like Shulgi, Amar-Suen, and Shu-Suen. Was it an imperial expansion aimed at direct control of outlying peoples and polities, or was it the crea­ tion of new space, both ideologically and on the ground, in which an expanded royal elite could operate and through which they could protect their divinely sanctioned positions and their property? The Ur III defense zone is, in my view, transformed into an Ur III operational zone, in which room was created for the royal military elite to exercise their new roles and to extract the resources that enriched the kingdom and allowed for pat­ ronage to be richly bestowed and rewarded. 17 These activities did not require the conquest and administration of established polities, but they did require regular successful campaigns, raids beyond the frontier that filled the pastures of the sheep­ fold with animals, that staffed the weaving establishments, and that brought both wealth and renown to the kings of Ur. The shift in strategy in Shu-Suen’s reign was an attempt to safeguard this zone and its pasturelands in an environment of diminished resources combined with more established resistance in the east. The focus of royal military activity in the operational zone of the frontier opened up new space, but also brought in new resources, herds and people, that had to be carefully shepherded by the kings. The eventual fall of the kingdom resulted from failures inside the state as much as it did from outside intervention. Certainly, the frontier policies of the kings of Ur created resistance, but this came more from established polities across the Zagros than from the Amorites. Moreover, the inabil­ ity of Shu-Suen to carry on the required annual campaigning beyond the frontier, his inability to maintain the greater sheepfold that his predecessors had created, spelled the end of this experimental state.

Bibliography Civil, M. 2011 The Law Collection of Ur-Namma. Pp. 221–286 in Cuneiform Royal Inscriptions and Related Texts in the Schoyen Collection, ed. A. R. George. CUSAS 17. Bethesda, MD: CDL Press. Frayne, D. R. 1997 Ur III Period (2012–2004 BC). RIME 3/2. Toronto: University of Toronto Press. Garfinkle, S. J. 2013 The Third Dynasty of Ur and the Limits of State Power. Pp. 153–167 in From the 21st Century B.C. to the 21st Century A.D. Proceedings of the International Conference on Neo-Sumerian Studies Held in Madrid, 22–24 July 2010, eds. S. J. Garfinkle and M. Molina. Winona Lake, IN: Eisenbrauns.

17.  This notion of the frontier as an operational zone is fairly consistent with earlier discussions of the mada by Michalowski 1978 and Maeda 1992, and, in particular, with the former’s conclusions (2011: 122 ff.) about this as a staging area for the military.

The Kingdom as Sheepfold

251

2014 The Economy of Warfare in Southern Iraq at the End of the Third Millennium BC. Pp. 353–362 in Krieg und Frieden im Alten Vorderasien, 52nd Rencontre Assyriologiue International, Münster, 17.–21. Juli 2006, eds. H. Neumann, et al. AOAT 401. Münster: Ugarit-Verlag. 2015 Ur III Administrative Texts, Building Blocks of State Community. Pp. 143–165 in Texts and Contexts: The Circulation and Transmission of Cuneiform Texts in Social Space, eds. P. Delnero and J. Lauinger. SANER 9. Berlin: de Gruyter. George, A. R. 1999 The Epic of Gilgamesh. London: Penguin Books. Maeda, T. 1992 The Defense Zone During the Rule of the Ur III Dynasty. ASJ 14: 135–172. Michalowski, P. 1978 Foreign Tribute to Sumer in the Ur III Period. ZA 68: 34–49. 2011 The Correspondence of the Kings of Ur: An Epistolary History of an Ancient Mesopotamian Kingdom. MC 15. Winona Lake, IN: Eisenbrauns. Porter, A. 2012 Mobile Pastoralism and the Formation of Near Eastern Civilizations, Weaving Together Society. Cambridge: Cambridge University Press. Roth, M. T. 1997 Law Collections from Mesopotamia and Asia Minor. Second Edition. SBL Writ­ ings from the Ancient World Series. Atlanta: Scholars Press. Steinkeller, P. 1991 The Administrative and Economic Organization of the Ur III State: The Core and the Periphery. Pp. 15–33 in The Organization of Power, Aspects of Bureaucracy in the Ancient Near East, eds. McG. Gibson and R. D. Biggs. SAOC 46. Chicago: The Oriental Institute. Van De Mieroop, M. 2015 Production and Commerce in the Old Babylonian Period. Rivista Di Storia Economica 31/1: 79–96.

From Uruk to Ur: Automated Matching of Virtual Tablet Fragments Erlend Gehlken, Tim Collins, Sandra Woolley, Laurence Hanes, Andrew Lewis, Luis Hernandez-Munoz, and Eugene Ch’ng Goethe-Universität, Manchester Metropolitan University, Keele University, University of Nottingham-Ningbo

Introduction Uruk is the town in which—according to our knowledge—writing was created more than 5,000 years ago. That was the beginning of a new era, the written record­ ing of information marked the start of a tempestuous development in almost all fields of human intellectual activities, from administration to science. Hundreds of thousands of clay tablet fragments have been unearthed, further pieces will be excavated in times to come. They hold an archive of an ancient civilization which is in many respects strongly connected to our modern culture. In order to make this archive fully accessible the fragments have to be joined. Manual joining is very time consuming because the number of join possibilities (P) is not proportional to the number of given pieces (N) but to the square of them (P = ½ N (N – 1)). To imagine that the fragments housed in the museums all over the world could be matched manually over the next decades is simply wishful thinking. We saw this reconstruct­ ion task as a challenge; one which we could attempt using virtual models of the fragments. We started our project with fragments from Uruk, not because writing was invented in Uruk, but because we had permission from the German Archaeological Institute to use Uruk fragments which are housed at Heidelberg University.

Virtual 3D Computer Models The first stage in virtual reconstruction is the creation of virtual fragments (Woolley et al. 2001). Computers represent three-dimensional (3D) objects by a set of points (“vertices”) distributed around the surface of the object (Fig. 1(i)). A “wireframe mesh” is formed by linking groups of points into triangles, turning the points into a virtual solid object. The wireframe mesh shown in Fig. 1(ii) is a low-resolution example of 1,000 vertices. The solid mesh in Fig. 1(iii) is a highresolution example of the same fragment comprising one million vertices. Accuracies of approximately one thousandth of an inch (25 microns) are typical with modern 3D scanners. 253

254

Erlend Gehlken, Tim Collins, Sandra Woolley, et al.

Fig. 1. Computer representation of 3D objects.

The vision of our project, “The Virtual Cuneiform Reconstruction Project” (http://virtualcuneiform.org), is an internationally accessible online environment for virtual tablet reconstruction (Ch’ng et al. 2014) where reconstructions may be done by individual users, collaborative groups, or by automatic means (Lewis et al. 2015; Woolley et al. 2017). The information used would be: •  3D models of scanned fragments (comprising vertex coordinates, mesh indices, and texturing) so that virtual fragments can be rendered graphically, •  Catalogue information about the fragment (period, excavation site, language, contents, etc.), and •  Physical descriptors about the nature of the fragment that could be useful for human or computer reconstructors (actual physical fragment size, estimated tablet size, preserved part of the tablet, etc.). The more virtual joins that can be excluded before starting the process of automated joining, the faster the joining can be achieved.

Automated Joining Computer-aided reconstruction of fragmented artefacts has been an active area of research in recent years (Papaioannou et al. 2017). However, much of the published work has been specific to the reconstructions of fragments of simpler geometry, such as frescos (Vidal and Belenguer 2014) and pottery (Rasheed and Nordin 2015), rather complex free-form 3D fragments which can produce large numbers of incorrect join recommendations (Huang et al. 2006; Belenguer and Vidal 2012). The 3D joining problem can be expressed as follows. If one fragment of a potential pair is fixed in space, the other will require moving to the correct position and rotating to the correct orientation in order to make a join. As discussed by Blanco (2010), the most efficient way of describing this positioning and orientation is in terms of six degrees of freedom (x, y, z translations and φ, θ, ψ rotations), see Fig. 2.

From Uruk to Ur: Automated Matching of Virtual Tablet Fragments

255

Fig. 2. Translations and rotations that comprise the six degrees of freedom.

In order to find the correct values of x, y, z, φ, θ, and ψ, a six-dimensional “search space” of possible solutions must be searched. This huge search space seemed to be nearly inexhaustible, one has the feeling of being faced with an insurmountable task. But, If it may be done, I will cross the ocean, if it may not be done, I will roam the wild!

(George 2003: 683, X 76–77; 687, X 153–154)

We were presumptuous enough to regard this firm intention of Gilgamesh, who set out from Uruk to search for the secret of eternal life, as permanent incentive. With six degrees of freedom, fragments can be described at absolutely any location and any orientation. We are only interested in potential joins, so we can eliminate one of the degrees of freedom by only considering positions and orientations where the two fragments touch one another (but do not overlap). The finite size of fragments means that practical bounds can be placed on the five remaining dimensions, but this is still a very big search space. A combination of a lower dimensional search in space followed by an iterative search in orientation was our solution. Our approach to matching virtual fragments begins by automatically enclosing each fragment in a “minimum bounding box” (O’Rourke 1985), i.e. the smallest rectangular-sided box (or cuboid) that the fragment could fit in (Fig. 3). The flatter faces of the tablet will usually align, approximately, with one of the sides of the box—these sides can be identified and rejected as possible joining edges. For each of the remaining box sides we •  search through a grid of x (front-back) and y (left-right) translations, •  z (up-down) translation found to make virtual fragments just touch, and •  find the optimal rotation iteratively.

256

Erlend Gehlken, Tim Collins, Sandra Woolley, et al.

Finding the optimal rotation uses an optimization algorithm (Nelder and Mead 1965) that iteratively adjusts the orientation angles aiming to progressively reduce a “cost function.” Cost func­ tions (also known as “loss functions”) are commonly used in optimization algorithms to calculate the penalty of making a particular decision (Wald 1949). All possible orientations have associated costs that reduce as the optimal solution is approached, i.e. the lower the cost, the better the match. The cost func­ tion we used is based on the distances between opposite points on each surface Fig. 3. A pair of virtual fragments enclosed inside to be joined as illustrated in Fig. 4. If the their minimum bounding boxes. two fragments match and are perfectly oriented, each of the distances in the figure will be equal. Calculating these distances is a computationally intensive operation; however, it is one of the operations that is essential in 3D graphics processing pipelines. We can use the optimised hardware in standard computer Graphical Processing Units (GPUs) to perform this calculation efficiently. Fig. 5 shows examples of “depthmaps” calculated by GPU computation. The shading represents the depth to the surface (one can think of it as the height, or elevation, of the surface— much as one would find on a geographical map). In order to test for a join, we add the two candidate depth-maps together and then subtract the smallest sum. If they join perfectly, as is the case in the fragments in Fig.5, the result is zero (black) at all points on the surface. The goodness of a join is assessed by the cost function, the sum of all mea­ Fig. 4. Depth measurements calculated between sured values of f(z), which is derived virtual fragments. from the depth measurements, z. The simplest method would be to simply use a linear or squaring function to relate the distance between the surfaces to the cost function. This works well for clean and complete breaks but not well where there is incomplete matching due to multiple breaks (Fig. 6). We want to focus on minimising the distance between nearby parts of the surfaces. Sections that are far apart are probably not matched and reducing the distance between these is not helpful. Our piecewise linear cost function works well (Fig. 7). Beyond a certain depth threshold, ϵ, surfaces are judged unlikely to match and there is a much lesser impact in increasing or decreasing the distance between pairs of points compared with nearby point-pairs below that threshold. The technical details of the algorithm are published by Collins et al. (2014).

From Uruk to Ur: Automated Matching of Virtual Tablet Fragments

257

Fig. 5. Depth maps for surfaces of joining fragments and (right) their summation when correctly joined.

Fig 6. (i) Two matching fragments but with lower right corner of fragment0 missing; (ii) the optimal match based on minimising the average of the depths; (iii) the optimal match using our piecewise linear cost function.

Fig. 7. The piecewise linear cost function for fragment join evaluation.

258

Erlend Gehlken, Tim Collins, Sandra Woolley, et al.

Joining Uruk Fragments Our first automated joining results were from virtual fragments from Uruk (Gehlken et al. 2014). The two halves were already known to match, but had not yet been glued together so we were able to obtain scans of the individual fragments (Fig. 8, left and middle). We inputted the two separate scans into the algorithm and it automatically produced the resulting optimal match, shown in Fig. 8 on the right, without any human assistance. The rendering shown here has no photographic surface texture information, but the quality of the match is obvious nonetheless.

Fig. 8. Program screenshot for the first automated join.

Joining Ur Fragments After achieving automated joining we were still not completely satisfied. The method worked well but the required processing time was too long, especially for use on large collections where many potential matches would need to be checked. Fortunately, the British Museum opened up to us the undreamed opportunity to scan more fragments in the frame of the Ur project, the new forward-looking collaboration between London and Philadelphia (Helgestad and Taylor 2016). That encouraged us to increase our efforts.

Photogrammetric Acquisition Gaining access to the collection at the British Museum also allowed us to test our new 3D acquisition system. It is based on photogrammetry—the reconstruction of 3D models from sets of conventional photographs. This is a popular technique for surveying large sites or capturing statues or monuments, but not so widely used for small objects. The system makes use of a small, very low-cost turntable and a digital camera. The system does not require calibrating or accurate camera positioning and

From Uruk to Ur: Automated Matching of Virtual Tablet Fragments

259

Fig. 9. The Smartphone Photogrammetric Acquisition System.

can be set-up for use very quickly and easily. The illustrated version (Fig. 9) is the very low-cost version of the system which uses a smartphone for the camera and controller—the total cost of the entire system is around $100, making it practically disposable. We have measured the precision of the scans and found them to be accurate to approximately 2/1000 inch (50 microns; approximately half the width of a human hair). We are continuing to refine our matching algorithms with help from the data from the Ur collection. Fig. 10 shows an example of a pair of virtual fragments scanned by our system. These two fragments were already known to match and were automatically paired by our joining algorithm. Results like this give us confidence that the scanning accuracy is perfectly adequate for finding and verifying virtual matches.

Fig. 10. Photogrammetrically acquired 3D computer models of Ur fragments (UET 6 748 and UET 6 759) (i) and (ii); automatically joined in (iii).

260

Erlend Gehlken, Tim Collins, Sandra Woolley, et al.

Conclusions and Further Work Joining the fragments from the ancient cities of Ur and Uruk will figuratively be tantamount to a new excavation: now it is not fragments that will be unearthed, as was the case in the past, but—as should be the case in the digital age—complete virtual tablets, which we can read as in an open book on mankind’s past. We plan to evolve the system into a sustainable environment for continued use by the Assyriological scholarly community. Our next aim: to find long distance joins of fragments in London and Philadelphia which can be verified electronically—this is an idea which until now would have been unthinkable. In ongoing work we aim to turn the automated matching process from proof-­ofconcept demonstration into a viable system for use on large virtual collections. Optimisation work is ongoing to increase the efficiency of the matching algorithm. More importantly, we are working on ways to identify which pairs of fragments are most likely to match and should, therefore, be tested with a higher priority. Statistical features of the broken surfaces and of the preserved writing (line distance, height of the signs, etc.) can also yield clues about what to look for in potential partners. We are very grateful to the British Museum for their assistance providing us access to the cuneiform fragments in the Ur collection. In order to develop and validate these techniques, a large collection of virtual fragments is needed for the source data. We are continuing to collect more samples all the time and are hoping, in the process, that our computation will find undiscov­ ered matches in the near future and that complete tablets will be automatically reconstructed.

Acknowledgments The cuneiform tablet fragments presented in Fig. 10 are held on study loan at the British Museum. The authors would like to thank the Trustees of the British Museum for permission to access the fragments and would also like to thank Birger Helgestad, Mathilde Touillon-Ricci, and Jonathan Taylor for their assistance which has made this work possible. The scanned W 18349 and Wy 777 fragments in Figs. 1, 2, 3 and 8 are published with the kind permission of the German Archaeological Institute, Berlin. The 3D printed models were produced using data from the 3D scanner of the Heidelberg Graduate School of Mathematical and Computational Methods for the Sciences at the Interdisciplinary Center for Scientific Computing, Heidelberg University. The authors also wish to acknowledge the support of the Leverhulme Trust for support of the progenitor project “A Collaborative Environment for Assisted 3D Reconstruction of Cuneiform Tablets.”

Bibliography Belenguer, C. S. and Vidal, E. V. 2012 Archaeological Fragment Characterization and 3D Reconstruction Based on Projective GPU Depth Maps. Pp. 275–282 in 18th International Conference on Virtual Systems and Multimedia, Milan, Italy. Piscataway, NJ: IEEE.

From Uruk to Ur: Automated Matching of Virtual Tablet Fragments

261

Blanco, J. L. 2010 A Tutorial on SE(3) Transformation Parameterizations and On-Manifold Optimization. University of Malaga, Tech. Report No. 012010. Ch’ng, E., Woolley, S. I., Hernandez-Munoz, L., Collins, T., Lewis, A., and Gehlken, E. 2014 The Development of a Collaborative Virtual Environment for 3D Reconstruction of Cuneiform Tablets. Pp. 35–42 in 2014 International Conference on Virtual Systems and Multimedia. Hong Kong: IEEE. Collins, T., Ch’ng, E., Gehlken, E., Woolley, S. I., Lewis, A., and Hernandez-Munoz, L. 2014 Computer-Assisted Reconstruction of Virtual Fragmented Cuneiform Tablets. Pp. 70–77 in 2014 International Conference on Virtual Systems and Multimedia. Hong Kong: IEEE. Gehlken, E., Collins, T., Woolley, S. I., and Ch’ng, E. 2014 Automated Joining of Cuneiform Tablet Fragments. NABU 2014/4: 119–120 no. 73. George, A. R. 2003 The Babylonian Gilgamesh Epic. Oxford: Oxford University Press. Helgestad, B. and Taylor, J. 2016 Ur of the Chaldees, A Virtual Vision of Woolley’s Excavations. See http://www. britishmuseum.org/research/research_projects/all_current_projects/ur_project. aspx Huang, Q. X., Flöry, S., Gelfand, N., Hofer, M., and Pottmann, H. 2006 Reassembling Fractured Objects by Geometric Matching. ACM Transactions on Graphics 25/3: 569–578. Lewis, A., Woolley, S., Ch’ng, E., and Gehlken, E. 2015 Observed Methods of Cuneiform Tablet Reconstruction in Virtual and Real World Environments. Journal of Archaeological Science 53: 156–165. Nelder, J. A. and Mead, R. 1965 A Simplex Method for Function Minimization. The Computer Journal 7/4: 308–313. O’Rourke, J. 1985 Finding Minimal Enclosing Boxes. International Journal of Computer & Information Sciences 14/3: 183–199. Papaioannou, G., Schreck, T., Andreadis, A., Mavridis, P., Gregor, R., Sipiran, I., and Vardis, K. 2017 From Reassembly to Object Completion: A Complete Systems Pipeline. ACM Journal on Computing and Cultural Heritage 10/2: 1–23. Rasheed, N. A. and Nordin, M. J. 2015 A Survey of Computer Methods in Reconstruction of 3D Archaeological Pottery Objects. International Journal of Advanced Research 3/3: 712–714. Vidal, E. V. and Belenguer, C. S. 2014 A Discrete Approach for Pairwise Matching of Archaeological Fragments. Journal on Computing and Cultural Heritage 7/3: 1–17 Wald, A. 1949 Statistical Decision Functions. The Annals of Mathematical Statistics: 165–205. Woolley, S. I., Ch’ng, E., Hernandez-Munoz, L., Gehlken, E., Collins, T., Nash, D., Lewis, A., and Hanes, L. 2017 A Collaborative Artefact Reconstruction Environment. Proceedings of the 31st British Computer Society Human Computer Interaction Conference. Sunderland: BCS Learning and Development.

262

Erlend Gehlken, Tim Collins, Sandra Woolley, et al.

Woolley, S. I., Flowers, N. J., Arvanitis, T. N., Livingstone, A., Davis, T. R., and Ellison, J. 2001 3D Capture, Representation and Manipulation of Cuneiform Tablets. Pp. 103– 110 in Three-Dimensional Image Capture and Applications IV, eds. B. Corner, J. Nurre, and R. Pargas. Bellingham, WA: SPIE.

The City of Ur: The Possibilities for Statistical Analysis on the Use of Space in Domestic Households Nicole Grunert

Christian-Albrects-Universität zu Kiel

Introduction From early on in archaeology, most studies focus toward an understanding of what sort of human behaviours took place in architectural spaces (Bagwell 2006). 1 Especially in Archaeology, studies focus on the supposition that the relationship between humans and the built environment is a reflexive one, where the organisation of built space is influenced by human behaviour and built space influences the human behaviour (Fig.1) (Lawrence and Low 1990; Rapoport 1969). 2 The second half of this relationship, space that is created by and for human actions, receives the most attention in this context and takes three main forms. The first kind of spatial study is the examination of room contents in order to determine activity areas or room functions that allow the identification of households. This in turn allows archaeologists to make inferences about the nature of prehistoric social organization (Bawden 1989; Donley-Reid 1990; Flannery 1976), and perhaps even its level of complexity (Kent 1993). The second kind of spatial study is the examination of the shape of the architectural space. Shape is usually argued to reflect the function (Hunter-Anderson 1977). The third kind of spatial study is the investigation of connections between spaces, their nature and how the built environment functions as a boundary that limits or enables certain kinds of interaction. Boundaries have two main functions in this context as they physically differentiate space by providing information about the function and/or structure of a space and secondly provide interpersonal regulation, encouraging and limiting both social and physical contact (Lavin 1981). 3 1. Bagwell focuses in her paper especially on the fact that production and its social complexity plays a huge role in archaeological analysis, foremost the increasing levels of specialisation production are related to increasing social complexity. The relations between the specialised production of architecture—the production of space—and social complexity are not yet fully understood. 2. Theoretical assumptions are based on ethnoarchaeological observations, but the analysis and comparison of architectural spaces is not possible by that. 3. Lavin proposes the concept of boundary as a central element in the specification of symbolic and regulatory processes in the built environment by using cognitive theories of person-environment

263

264

Nicole Grunert

Archaeological Evidence and Former Studies The two examples chosen for this paper are from the building complex in area AH, where Woolley’s excavations unearthed a series of buildings and streets that were part of the Old Babylonian city quarters about 150 meters south-east of the temenos (Woolley 1976). The fifty-two excavated buildings in this at least 7,500 m2 excavated area, unearthed before the war by the joint Expedition of the British Museum and the Museum of the University of Pennsylvania to Mesopotamia, show a remarkably good state of preservation. With walls still standing to a height of up to two meters and apart from the occasional changes to the architecture, the clear floorplans make the area of AH predestined for analysis with Space Syntax tools. As with most archaeological contexts and especially with old excavation data, there are always factors that have to be taken into account. For example, the room functions that Woolley assigned during Excavation are not always accurate, especially when it comes to the so-called lavatories. Also, this applies to the identified drains and sewers. Furthermore, working with the material and old excavation notes shows that the numbers of houses might have been mixed up at a certain point. 4 For the study conducted in this paper, only one of the mentioned problems is relevant: the problematic distinction of room functions. As always, the more information we get, the clearer the picture becomes. This applies also to the statistical analysis of floor plans. Even though some information has to be excluded from the analysis, Woolley’s observations can still be taken into account at a later point. Even if some of the data might be compromised, further studies on the architecture from Ur including the initial analysis by Woolley made it possible to reconstruct many aspects of daily life in Old Babylonian Ur. The studies focus mostly on the aforementioned three kinds of spatial study. The examination of room contents was used in order to determine activity areas or room functions that allowed the identification of households. He combined this with the second method of the examination of the shape of the architectural space to deduce the function further up to a point where he reconstructs second and third storeys (Woolley 1976: 26). 5 The third kind of spatial study, namely the investigation of connections between spaces, was used to interpret changes made to floorplans. Woolley even goes so far to speak of a generalised, ideal floorplan that people tried to follow, backed up by citing the translation of omen texts on features a house should have, which combines an ethnoarchaeological approach with the spatial analysis of the architectural context. Brusasco (1999) uses a different approach, using node graphs together with a building typology to define the general use of the buildings. 6 He compares the tree transactions. 4.  For the problems distinguishing certain House numbers within the publications and excavation documentation see https://www.penn.museum/blog/museum/ur-digitization-project-september-2013/ 5. Brusasco (1999: 87) argues together with other scholars that the archaeological evidence for second or third storeys are very scarce or even contradictory and the found stairs only demonstrate access to the rooftop. Ethnographic comparisons confirm that stairs do not indicate a second storey but only that the rooftop was used as additional living or working space. 6.  The node graphs are not used to mathematically analyse the built space but are used as permeability maps. The conducted series of t tests on accessibility, diffusion of control, floor area, and number of rooms has been carried out on the typology. It only shows a quantification of structural features captured within the typology. In Brusasco’s opinion the only factor allowing quantitative tests is the number of rooms on the sides of the courtyard. Therefore, the buildings are catalogued according to the

The City of Ur: The Possibilities for Statistical Analysis

265

against the ring permeability maps and infers that the tree-like shape of the single court buildings maximises the power and control on internal movements of people. Their relatively high permeability and small number of rings on the other hand underline the presence of a relatively stronger spatial relation with the outside. This means that the spatial solidarity of most of the Single court buildings is built on internal tension and constraint.

Theory and Method In most cases we are missing certain required information to analyse architectural remains, so the aforementioned first step in spatial studies by room contents is often not possible. This leaves us with the shape and the boundaries. Markedly Space Syntax Analysis, developed in modern Town/City and Building planning to analyse new and already existing structures, their relationship with each other and the human-building relationship enables us to analyse the social relationships in a city quarter based on the architectural remains even when other references and details are missing. In this paper, I will examine parts of the AH quarter in Ur, using the theoretical and practical tools offered by Space Syntax Theory. By analysing the connectivity and integration of areas in the built space environment, it is possible to generate a pattern of movement and locate hotspots for social con­ tact and interaction. It is also possible to allocate mathematical values to spaces in respect of their integration and control within the floorplan which makes it possible to compare spaces not only within one culture but on the overall picture as the shape of the space is not necessarily important but its position and connection is. Later on, the shape can be added to the analysis. Ferguson argues that graph analysis often does not attempt to incorporate a theory that explicitly relates the spatial structure of architecture to human society or behaviour (Ferguson 1996). 7 This is where the development of Space Syntax Theory comes into play. The concept of Space Syntax Theory was developed by Bill Hillier, Julienne Hanson and their colleagues at the Bartlett School of Architecture at University College London in the late 1970’s and early 1980’s as a tool to help urban planners simulate the likely social effects of their designs (Hillier et al. 1987; Hillier et al. 1989; Hillier 2007). The methods of Space Syntax allow architectural space to be represented and its pattern properties quantified, so that comparisons could be made between differently designed buildings or urban areas. The aim of this research was to develop an understanding of the way that spatial design and social function were related. The general idea behind all of it: Space can be broken down into components, analysed as network of choices, represented as maps and graphs to describe relative connectivity and integration of those spaces. 8 There are two fundamental propositions: First, space is no background to human activity but intrinsic to it. Second, space is first and foremost configurational. presence of rooms on one, two, three, and four sides of the courtyard. The results of his t tests on these variables suggest a clear pattern of significant variation among the types and a progressive increase in the values of accessibility, diffusion of control, floor area, and number of rooms. 7.  Ferguson gives in his analysis on Zuni Architecture a broad summary on architectural theory and practice. 8.  For a short summary and its application on Mesopotamian architecture see also the paper of Drennan about architecture in Mesopotamia (2010).

266

Nicole Grunert

Also, in general, there are two ways to use space: the conservative use, where space is being used to reflect patterns by the control of co-presence. This occurs mostly in domestic architecture. An exception is ritual architecture as well as buildings with a public and private function. The generative use, where space is used to create new patterns of co-presence and potentially new elements in social patterns occurs in modern public buildings. Seeing Space Syntax Theory from the technological side, there are three main options to represent a building in Space Syntax Analysis. Buildings can be represented as a node graph where every node represents a room/ space and every connection an entrance. The second option, a visibility graph, is represented as a grid of points, where all points are connected to all other points they can “see.” This way every space is categorised by their “intervisibility.” A third option is agent-based modelling, where the movement of individuals can be traced depending on characteristic factors for example by defining decision making in way finding processes. Both the second and the third options are not represented here. The opportunity Space Syntax presents are the following: it is possible to investigate regional housing culture by comparing the functions of spaces with their values. If different functions are statistically associated with different values, then there is a spatial meaning given to the idea of function. The spatial configuration of the layout can be said to have, within that culture, a social meaning. Area AH, even though in a mixed state of preservation and due to its long use stratigraphically very complex, shows manifold changes in the social organisation of households and neighbourhoods. Woolley refers to an unplanned city-layout and many buildings in Area AH show that the population fitted the available space to their needs. I decided for this paper to show, how domestic space can be analysed mathematical to make statistical analysis and modelling possible. These methods can show on a broader scale, over time for example, changes and constants that occurred in the use of space. By using node graphs to calculate the mathematical properties for each space, we are able to compare them to each other and similarities as well as differences become clear. The mathematical analysis of domestic building structures is conducted by using a software which was developed at the University of Oslo as part of a PhD project at the Oslo School of Architecture and Design named AGRAPH (Manum et al. 2005). By calculating the rooms as nodes and doorways as connections, the software calculates for every space within a building the parameters of Control Value, Total Depth, Mean Depth, Relative Asymmetry and the integration value for every space within the building. 9 For each calculation AGRAPH gives three output files, one shows the connectivity matrix, the second shows the matrix for internal distances. The third is the summary file which gives the needed values for statistical analysis and further space syntax calculations. Also, there exists a colour mode for the nodes to compare different floorplans. With the possibility to exclude nodes like the exterior space for 9.  The Control Values (CV) are found by letting each node give the total value of 1 equally distributed to its connected nodes. The Control Value of node n, CV(n), is the total value received by node n during this operation. Total Depth of a node n, TD(n), is the total of the shortest distances from node n to the other nodes in the systems, i.e. TD(n) is the total of line n (or column n) in the distance matrix. Mean Depth for a node n is the average depth (or average shortest distance) from node n to all the other nodes. The Relative Asymmetry (RA) describes the integration of a node by a value between (or equal to) 0 and 1, where a low value describes high integration. The integration value (i) is a parameter that (contrary to RA) describes integration by a high number when a node is highly integrated.

The City of Ur: The Possibilities for Statistical Analysis

267

certain analysis, it is possible to analyse only the living area within the build space. Using this method on different floorplans in the city of Ur, not only is it possible to compare different floorplans with each other but also the same floorplan with different stages in its existence and additions and subtracted parts.

Analysis and Interpretation The houses analysed in this paper only serve as a test series, as a thorough analysis of all floorplans at their different stages would take up substantially more room than this paper allows. Before starting to analyse the floorplans randomly, a typology has to be done to distinguish between domestic, partially domestic, and public buildings. For this I decided to use Brusasco’s typology but divided his types into those three far simpler classifications. The number of rooms of the individual buildings is only of secondary interest at first but should be considered later, if the difference is substantial. For this paper, only two domestic used buildings with one court will be considered to show the general principals of graph analysis and its possibilities for further statistical analysis. Even though, the labels of the rooms distributed by Woolley based on the features and small finds is a disputable factor, I will keep them to simplify the comparison. For further analysis, it must be kept in mind that certain designations have to be checked to compare different architectural complexes over time and space. For the analysis in this paper, the houses No. 3 and No. 4 Straight Street were chosen. The reason for that was because No. 4 shows all aspects of the ideal floorplan that Woolley suggested, while No. 3 shows the attempt to follow this floorplan whereas the builder had to fit the house into the available space. The analysis of the tree-shaped graph (Fig. 1) for house No. 4 shows an overall tree-shaped graph with a small ring structure in the back, where the courtyard is the main distributor for movement into the rooms of the house beyond this point. The kitchen as well as the chapel are in the farthest corners of the house and also the room identified as lavatory lies rather secluded. When we have a look on the computational results (Table 1) we see that the courtyard—as expected—has a high value in control (CV) and in the overall integration (i) with­in the floorplan. The value for relative Asymmetry (RA) is very low, and tends to go against zero, which is the desired value for a well-integrated central room within modern building planning with these means. 10 The kitchen and the chapel show Control Values (CV) below 0.5, which means they are not Fig. 1. Node graph for house No. 4 relevant at all for the main movement within the Straight Street. house and show a tendency to be used as special­ised 10.  The nearer the value goes against zero the better is the space cut out for being the main controlling room. Especially with more than one courtyard, there is the possibility that those spaces are not well integrated or fit into the relative Asymmetry while having high values in control.

268

Nicole Grunert

spaces. 11 The outside as a space shows the same values as the chapel and it is no wonder that that is the case. Both lie on the outskirts of our system and because of the distance created by the room that functions as lobby as well as the constellation with no direct line of sight the outside is not well integrated into the build space. Table 1. Computational values for house No. 4 Straight Street. Space

TDn

MDn

RA

i

CV

0

0

32

3,20

0,48

2,04

0,50

1

Lobby

23

2,30

0,38

3,46

1,20

2

Courtyard

16

1,60

0,13

7,50

3,16

3

3

25

2,50

0,33

3,00

0,20

4

Reception

19

1,90

0,20

5,00

1,53

5

Lavatory

28

2,80

0,40

2,50

0,33

6

Chapel

31

3,10

0,46

2,14

0,33

7

7

22

2,20

0,26

3,75

1,66

8

8

19

1,90

0,20

5,00

1,53

9

Kitchen

28

2,80

0,40

2,50

0,33

10

10

25

2,50

0,33

3,00

0,20

Min

16,00

1,60

0,13

2,04

0,20

Mean

24,36

2,43

0,31

3,62

1,00

Max

32,00

3,20

0,48

7,50

3,16

The floorplan of house No. 3 shows a few peculiarities. Woolley interpreted the floorplan as a single house, which was at a later point opened to No. 1 Old Street creating one big house complex. 12 For this article, I decided to choose Woolley’s interpretation of No. 3 as a singular house that was later opened and only use the floor plan of this earlier stage. Apart from that, the complex shows open spaces apart from the courtyard that are lying rather secluded and creating a small ring scheme in the overall treelike shape of the graph. Also, the floorplan was fit to the available space resulting in two not rectangular shaped rooms (a lobby and one room referred to by Woolley as a bedroom). House No. 3 is similar in size compared to house No. 4 with only three more build spaces, where two of them are the open yards flanking the house. Interpreting the tree-shaped graph (Fig. 2) shows that even though the floorplan is not following the ideal overall rectangular floor plan that Woolley imagined for the Old Babylonian houses, the connections are still similar to the ideal plan. The courtyard is the main distributor within the house lying at the centre of it. The kitchen in this case is slightly better integrated, as it serves as a connection to the yard spaces within the overall plan of the house. The chapel again lies far in the back, secluded from the rest of the house. Looking at 11.  This information is especially important when features are missing and the interpretation of a build space is not clear. 12.  Woolley 1976: 161. He describes that the backwall was rebuild and door added, but he makes no interpretation on the connection to No. 1 Old Street.

The City of Ur: The Possibilities for Statistical Analysis

269

the computational results (Table 2) we see again for the courtyard the ex­pect­ed high CV and i values that are expected within such a constellation. The kitchen space has a better CV than the kitchen in No. 4 due to the fact that it serves as a waypoint into the yards and is not a dead end. The chapel also has a better control value as it leads into another small room. The i value as well as the RA value show in this case only a slight difference in comparison to the chapel in No.4 which indicates that both rooms follow the same principle of meaning within the overall building plan. Comparing both tables with each other not just by the rooms but also by the mean values resulting for the whole floor plan we can see that both floor plans, though not looking the same and being difFig. 2. Node graph for house No. 3 ferent in their layout, do not necessarily diverge so Straight Street. much from each other. The mean control value for both houses has a value of 1.00. The difference in the integration value of 0.4 is not so much, showing that both houses show the same variances of integration in certain areas. As both houses show a low value in relative Asymmetry with 0.31 and 0.31 lying below 0.5 the overall Asymmetry is good, even though the value could be better by getting closer to zero. The mean total depth of the nodes (TDn) and the mean depth of the nodes (MDn) values differ slightly, which is attributed to the fact that house No. 3 has a few more build paces than house No. 4.

Conclusion Even though there are doubts on the interpretation of the use of rooms, the values of the space do not differ. The statistical analysis of Brusacso’s work (1989; 2004) takes into account the constellation of rooms around the courtyard and their numbers. In his later contribution, he uses Space Syntax Analysis to further substantiate his analysis on social complexity in Ur, by comparing it to Islamic architecture and the found text corpus. Apart from the possibilities to reconstruct social organisation and social complexity, graph analysis with Space Syntax tools offers the opportunity to use each build space for statistical comparisons for building features over time and space. It therefore not only makes it possible to reconstruct social complexity and general social organisation but also provides us with the means to see changes and trace them over time. Comparing these results then to earlier and later cultural groups can make it possible to highlight similarities or differences within social groups, helping to differentiate cultural groups not only by pottery tradition but also by their social organisation. The effect might be that a cultural group, defined by pottery types not necessarily shows any differences in their social complexity and organisation as often is assumed when speaking of different cultural groups. This relates to further topics as definitions of cultural groups solely based on small finds, interpretation of migration patterns, and changes in social organisation based on pottery especially in cultures without written sources.

270

Nicole Grunert

Table 2. Computational values for house No. 3 Straight Street. Space

TDn

MDn

RA

i

CV

0

0

45

3,46

0,41

2,43

0,50

1

Lobby

33

2,53

0,25

3,90

1,16

2

Courtyard

23

1,76

0,12

7,80

3,83

3

Lavatory

35

2,69

0,28

3,54

0,16

4

4

35

2,69

0,28

3,54

0,16

5

5

39

3,00

0,33

3,00

0,33

6

Reception

27

2,07

0,17

5,57

1,66

7

7

35

2,69

0,28

3,54

0,83

8

8

32

2,46

0,24

4,10

0,66

9

Kitchen

32

2,46

0,24

4,10

0,66

10

Chapel

45

3,46

0,41

2,43

1,50

11

11

57

4,38

0,56

1,77

0,50

12

12

41

3,15

0,35

2,78

1,00

13

Yard

41

3,15

0,35

2,78

1,00

Min

23,00

1,76

0,12

1,77

0,16

Mean

37,14

2,85

0,30

3,66

1,00

Max

57,00

4,38

0,56

7,80

3,83

Bibliography Bagwell, E. A. 2006 Specialization, Social Complexity and Vernacular Architecture: A Cross-cultural Study of Space Construction. Pp. 29–36 in Space and Spatial Analysis in Archaeology, ed. E. Robertson. Calgary: University of Calgary Press. Bawden, G. 1989 The Andean State as a State of Mind. Journal of Anthropological Research 45: 327–332. Brusasco, P. 1999 Family Archives and the Social Use of Space in Old Babylonian Houses at Ur. Mesopotamia 34–35: 3–173. 2004 Theory and Practice in the Study of Mesopotamian Domestic Space. Antiquity 78/299: 142–157. Donley-Reid, L. W. 1993 A Structuring Structure: The Swahili House. Pp. 114–126 in Domestic Architecture and the Use of Space: An Interdisciplinary Cross-cultural Study, ed. S. Kent. Cambridge: Cambridge University Press. Drennan, M. E. 2010 Architecture in Archaeology: An Examination of Domestic Space in Bronze Age Mesopotamia. Honors Scholar Theses. University of Connecticut. Ferguson, T. J. 1976 Historic Zuni Architecture and Society: An Archaeological Application of Space Syntax. Tuscon: University of Arizona Press.

The City of Ur: The Possibilities for Statistical Analysis

271

Flannery, K. V. 1976 The Early Mesoamerican Village. New York: Academic Press. Hillier, B. 2007 Space is the Machine: A Configurational Theory of Architecture. Cambridge: Cambridge University Press. Hillier, B. and Hanson, J. 1989 The Social Logic of Space. Cambridge: Cambridge University Press. Hillier, B., Hanson, J., and Graham, H. 1987 Ideas are in Things: An Application of the Space Syntax Method to Discovering House Genotypes. Environment and Planning B: Planning and Design 14/4: 363–385. Hunter-Anderson, R. L. 1977 A Theoretical Approach to the Study of House Form. PP. 287–315 in For Theory Building in Archaeology, ed. L. Binford. New York: Academic Press. Kent, S. 1993 Domestic Architecture and the Use of Space: An Interdisciplinary Cross-cultural Study. Cambridge: Cambridge University Press. Lavin, M. W. 1981 Boundaries in the Built Environment: Concepts and Examples. Man-Environment Systems 11/5–6: 195–206. Lawrence, D. L. and Low, S. M. 1990 The Built Environment and Spatial Form. Annual Review of Anthropology 19/1: 453–505. Manum, B., Rusten, E., and Benze, P. 2005 AGRAPH, Software for Drawing and Calculating Space Syntax Graphs. Pp. 97–101 in Proceedings of the 5th International Space Syntax Symposium, A. van Nes. West Lafayette, IN: Purdue University Press. Rapoport, A. 1969 House Form and Culture. Englewood Cliffs, NJ: Prentice-Hall. Woolley, L. 1976 The Old Babylonian Period. UE 7. Philadelphia: Trustees of the British Museum and of the University Museum, University of Pennsylvania, Philadelphia; British Museum Publications.

Trace Elements and Isotopes: The Origin of Gold from Ur from a Geochemical Point of View Moritz Jansen, 1 Andreas Hauptmann, 2 Sabine Klein,2 and Richard L. Zettler 3 Penn Museum and the Deutsches Bergbau-Museum

Introduction The analysis of archaeological artifacts made of gold is a challenging task given by the enormous cultural value. In a joint project of the German Mining Museum in Bochum, the Goethe University in Frankfurt and the University of Pennsylvania Museum in Philadelphia, metal and mineral based artifacts from Ur, including about seventy samples of gold artifacts, are investigated for provenance and tech­ nology (see Hauptmann et al. 2016). Within the framework of this cooperation, the following work is part of a PhD project on the Archaeometallurgy of Bronze Age gold in the Middle East. In the current study, the origin of the raw material for these gold artifacts will be discussed. The hundreds of gold artifacts from the Early Dynastic (ED) IIIa period are outstanding in third millennium Mesopotamia. The highly skilled manufacture of personal items and sculptures made of gold include various techniques like cast­ ing, hammering, chiseling, and soldering (Armbruster 2016). The fact that there are no natural sources of gold in Southern Mesopotamia prompts questions about the origin of the raw material used to create these artifacts. Gold had to be imported from other regions rich in metal deposits such as Anatolia, Iran, Afghanistan, Egypt or even India. The rich corpus of cuneiform tablets could help to identify possible trade relations, as they give an idea of possible origins due to named trading posts which potentially are closely related to the deposits. Some names can be assigned to geographical loca­ tions (following Maxwell-Hyslop 1977; Potts 1994; Potts 1997; Moorey 1999): Egypt cannot be identified in the cuneiform texts; hence, trade between Egypt and Meso­ 1. Center for the Analysis of Archaeological Materials, University of Pennsylvania Museum of Archaeology and Anthropology, 3260 South Street, Philadelphia, PA 19104, USA, [email protected] 2.  Forschungsstelle Archäologie und Materialwissenschaften, Deutsches Bergbau-Museum, Herner Str. 45, D-44787 Bochum, Germany 3.  University of Pennsylvania Museum of Archaeology and Anthropology, Near Eastern Section, 3260 South Street, Philadelphia, PA 19104, USA

273

274

Moritz Jansen, Andreas Hauptmann, Sabine Klein, & Richard L. Zettler

potamia can likely be excluded. The ancient Hah(h)um could indicate Southeastern Anatolia as a source for gold. Frequently, gold from Meluhha, present-day Indus Valley, is mentioned. Here, just as in Mesopotamia, gold deposits do not occur on the expansive alluvial plains. The sources of gold coming from Meluhha must be sought most likely in Afghanistan or other neighboring regions (Maxwell-Hyslop 1977). A detailed discussion of the textual evidence will be provided by Hauptmann et al. (2018). However, these tablets date slightly later than the ED period in Ur and nothing is known about the provenance of the earliest gold artifacts from the site. The gold artifacts of Ur present distinct kinds of composition, including deli­ berate alloying (La Niece 1995; Hauptmann and Klein 2016) which even can be identified in cuneiform texts at the end of the third millennium BCE (Hauptmann et al. 2018). This project seeks to define the geochemical fingerprint of gold artifacts from Ur in order to identify the origin of the gold through its chemical and isotopic composition. Ideally, this method could trace the gold back to its natural deposit(s). To define the geochemical fingerprint of the gold itself, non-deliberately alloyed gold artifacts need to be characterized. For deliberately alloyed gold artifacts, the alloy components copper and silver could obscure the distinctive signature of the gold source with their own trace element fingerprint. In this regard, it seems to be useful to analyze the artifacts of the headdresses with hair ribbons, flowers, diadems, earrings, and necklaces (Fig. 1) instead of artifacts that were more likely alloyed such as chisels, adzes, and daggers.

Fig. 1. Headdress of Queen Puabi (from Zettler and Horne 1998). For the geochemical characterization, jewelry artifacts from such headdresses were sampled to ensure the characterization of most likely not deliberately-alloyed gold (see text for discussion).

Trace Elements and Isotopes: The Origin of Gold from Ur

275

The focus of our analyses is on gold artifacts from the rich Early Dynastic royal tombs. Forty-four samples were taken from artifacts from the Great Death Pit (PG 1237), the so-called King’s Grave (PG 789) and other tombs. Ten samples came from artifacts in graves that Woolley dated to the Second Dynasty of Ur and the Sargonic Period. These graves are mixed in date (Buchanan 1954; Zettler and Horne 1998) and grouped here as Akkadian period to the Third Dynasty of Ur. Twelve additional artifacts are without stratified contextual information and are discussed separately.

Methods For our analytical work, we used two method compounds: first, the chemical composition with main, minor, and trace elements was determined by electron probe micro-analysis (EPMA) and laser ablation inductively coupled plasma mass spectrometry (LA-ICP-MS) analyzing solid samples mounted in polished sections. The use of the EPMA for the determination of main elements produces much more robust results than laser ablation quadrupole inductively coupled plasma mass spectrometry (LA-ICP-QMS) that has frequently been applied to gold artifacts in the last decade (e.g. by Ehser et al. 2011; Schlosser et al. 2012; Leusch et al. 2016). The use of sector field ICP-MS ensures a high sensitivity for trace elements. Second, the isotopic composition of selected elements lead, osmium and copper (Pb, Os, Cu) was measured by multi-collector inductively coupled plasma mass spectrometry (MC-ICP-MS), combined with femtosecond laser ablation (fsLA) for the determin­ ation of the osmium isotope composition of PGE mineral inclusions. For lead and copper isotope measurements, ten to twenty mg of sample material were dissolved and Pb was separated by ion exchange chromatography.

Geochemical Characterization Main and Minor Elements The main components of the samples of gold jewelry are gold, silver, and copper. Between fifteen and forty-five percent of silver and between one and six percent of copper was detected for the samples (Fig. 2). The Early Dynastic jewelry has the same composition as later Akkadian/Third Dynastic jewelry. Based on analysis of their main components, gold from the different periods could come from the same geological sources. It needs to be clarified whether the gold compositions with up to forty-five percent silver are natural or intentional alloys. It is state of research that in ancient times silver was mainly exploited by the use of silver-bearing lead minerals such as galena or cerussite. In a metallurgical process called cupellation, the silver is extracted by a selective oxidation of lead forming litharge while the silver as a noble metal remains unoxidized. The separation is never complete, and some of the silver is lost in the litharge, but also some lead ends up as trace element in the silver. Salzmann et al. (2016) found up to one and one-half percent of lead in the silver artifacts from Ur. Thus, one would expect a positive correlation between silver and lead for the gold artifacts if the silver content of the gold was intentionally added (Hauptmann et al. 2010). However, there is no correlation (Fig. 3): the silver of the gold artifacts does not seem to originate from lead ores and the composition of gold

276

Moritz Jansen, Andreas Hauptmann, Sabine Klein, & Richard L. Zettler

Fig. 2. Besides gold, the artifacts from Ur consist of major silver (Ag) and minor copper (Cu). There is a total overlap in the composition of artifacts from different periods. Elemental composition is analyzed by EPMA.

Fig. 3. Silver contents (Ag) and the trace element concentrations of lead (Pb) for the third millennium BCE gold artifacts from Ur. The line of the least squares regression does not show a relation between these two elements. Elemental composition is measured by EPMA (Ag) and LA-ICP-MS (Pb).

with varying contents of silver might reflect a natural alloy. In addition, we would expect an increased content of much more than 800 ppm lead if silver was added which had a similar composition to the artifacts characterized by Salzmann et al. (2016). For a general comparison with natural sources, the composition of gold, both primary (reef) and secondary (alluvial), of various occurrences in the Lesser Cauca­ sus is shown in Fig. 4. These deposits are unrelated to Ur, but form a broad database since they were systematically investigated in recent years by the authors (Jansen et al., forthcoming) and a second group from Mannheim and Halle (Wolf et al. 2011, 2013; Wolf and Kunze 2014). There is an overlap between the silver contents of the artifacts with silver in natural gold from this region. The content of silver of the artifacts is within the range of natural composition, adding to the argument for a natural alloy of gold with silver for the artifacts from Ur. But it is also evident that natural gold does not contain copper in the range of the gold artifacts from Ur. So, is this an intentional alloy? This question will be discussed below.

Trace Elements and Isotopes: The Origin of Gold from Ur

277

Fig. 4. The alloy components of the gold artifacts from Ur are compared to natural gold of the Lesser Caucasus (own data; Wolf and Kunze 2014). The variability of the silver content is comparable, but the natural gold does not show significant contents of copper. Note the logarithmic scale. Elemental compo­ sition is analyzed by EPMA, and by LA-ICP-MS for the data of Wolf and Kunze (2014).

Trace Elements Trace element patterns of gold artifacts used for provenance studies have been discussed in recent years through the introduction of sensitive laser ablation induc­ tively coupled plasma mass spectrometry (Guerra et al. 1999; Guerra and Calligaro 2004; Hauptmann and Klein 2009; Kovacs et al. 2009; Hauptmann et al. 2010; Ehser et al. 2011; Schlosser et al. 2012; Leusch et al. 2016). In the following section, the trace element patterns of the gold artifacts from Ur are presented. Following the definition in geochemistry, trace elements are elements with concentrations below 0.1 percent (given as ppm—parts per million) and which do not form own phases. In Fig. 5 the trace element pattern of the gold samples is plotted. There are significant concentrations of tin, platinum, and lead above 100 ppm.

Fig. 5. Trace element pattern of sixty-four artifacts from Ur. The solid dark grey line shows the mean of all artifacts, the black dashed lines show minimum and maximum of individual elements. Trace elements are determined by LA-ICP-MS. Concentrations are given in parts per million (ppm). Arrow symbols mark concentrations below detection limits.

278

Moritz Jansen, Andreas Hauptmann, Sabine Klein, & Richard L. Zettler

Because of the combination of tin and platinum, it is evident that the resources came from a placer deposit: 1.  Primary occurrences of tin minerals like cassiterite are formed in context of acidic igneous rocks such as granites as the alteration product greisen in their marginal zones through pneumatolysis. Greisen consist of quartz and white mica, whereas feldspar and biotite are absent, since these are the decomposed components. Greisen are enriched in elements such as tin, tungsten, fluorine, and lithium, crystallized as tourmaline, topaz, and wolf­ramite. Complex polymetallic mineralization with cassiterite, stannite, arsenopyrite and chalcopyrite can result from a subsequent hydrothermal overprint. The deposits of Cornwall have to be named as one of the larger examples of this type, but other examples can be found in, e.g., Afghanistan and Tajikistan. 2.  In contrast, platinum group minerals (PGM) are typically formed in (ultra-) basic rocks such as peridotites, serpentinites, and dunites. Fractionation of primitive mantle and partial melting leads to an enrichment and formation of PGM within such rocks.

After erosion of such protoliths, both tin and platinum minerals are co-deposited in placers. Subsequently, gold was panned from such placer deposits, and the gold-rich heavy mineral fraction bearing tin and platinum minerals was co-melted, leading to an enrichment of tin and platinum as trace element in artifact gold compared to natural free-gold particles in the concentrate. Compared to the composition of analysis from reef but also alluvial gold, elements like nickel, arsenic and bismuth are also enriched in the artifact gold (10–100 ppm). Only if there are inclusions of other minerals, the analysis of natural gold shows significant traces of other elements than silver and traces of copper (Jansen et al., forthcoming). Hence, the gold of the artifacts seems to have been intensively contaminated during metal­ lurgical processing. Washing gold never leads to pure gold concentrate. Different heavy minerals, which can never be separated completely through panning due to a very close density to gold, have influence on the trace element pattern of gold due to co-melting of the concentrate which increases trace element concentrations (Hauptmann et al. 2010). Thus, the trace element pattern of alluvial gold flakes differs from that of gold artifacts. By focusing on the dated artifacts, a very small variation in the trace element patterns of all the Early Dynastic artifacts is visible. And if we compare the later artifacts of the Akkadian period/Third Dynasty of Ur, these artifacts end up with the same pattern (Fig. 6). Only trace element concentrations of palladium, platinum, and bismuth are slightly higher for the younger artifacts compared to the Early Dynastic gold. That does not necessarily mean that different resources were used for the gold of the two periods, because natural variations also occur in this dimension, if separate locations within individual deposits were exploited. If we treat this as a plausible explanation for variations in trace elements, a stable use of the same occurrences over hundreds of years can be deduced, which implies a long-term exploitation of a placer deposit by several generations (“longue-durée phase,” see Stöllner 2003).

Trace Elements and Isotopes: The Origin of Gold from Ur

279

Fig. 6. Comparison of the trace elements between the Early Dynastic gold artifacts (given minimum and maximum as dashed black lines; n = 43) and the later Akkadian and Third Dynastic gold artifacts (given minimum and maximum as solid grey lines; n = 10). The patterns are the same. Trace elements are determined by LA-ICP-MS. Concentrations are given in parts per million (ppm). Arrow symbols mark concentrations below detection limits.

Isotope Compositions In addition to the chemical composition, isotope ratios of different elements were determined to define an extensive geochemical fingerprint. The lead (Pb) isotope composition provides information about the geological age of the deposits used in ancient times for metal exploitation. For the gold artifacts of Ur, a lead evolution model age of 0 to 200 Ma can be determined (Fig. 7). That means that the source of the lead in the gold artifacts was formed within the last 200 million years. We should keep in mind that the analysis of natural gold itself only reveals small traces of lead from free-gold grains of both reef and alluvial gold (Jansen et al., forthcoming). Potentially, the detectable amount of lead is rather introduced in secondary gold placer deposits by co-deposited components such as complex lead-containing copper minerals or galena grains and subsequent co-melting of such

Fig. 7. The lead isotope ratios of the gold artifacts from Ur plotted with the Stacey and Kramers lead evolution model (Stacey and Kramers 1975). The sources of the lead traces in the gold artifacts were formed in the last 200 Ma. Isotope ratios are measured by MC-ICP-MS and analytical uncertainty is given by the error bars as standard deviation (2σ).

280

Moritz Jansen, Andreas Hauptmann, Sabine Klein, & Richard L. Zettler

a concentrate. The lead that is measured for lead isotope composition in the gold thus does not originate from the same source as the primary gold. In consequence, the information, which is gained from Pb isotope analysis of gold objects leads to the provenance of the accumulated lead-bearing minerals rather than to the primary gold source. If the groups of different archaeological periods are compared, no significant differences are detectable in the Pb isotope compositions. That indicates that the sources of the gold of Ur may be continuous from Early Dynastic to Akkadian/Third Dynastic times. Both the chemical composition and Pb isotope ratios support this conclusion. Several copper-based artifacts from Ur dating between Early Dynastic and the Akkadian period were analyzed for their Pb isotope composition by Begemann and Schmitt-Strecker (2009). The dataset forms two clusters of Pb isotope ratios. Sur­ prisingly, gold artifacts from Ur, containing considerable copper amounts compared to natural gold (see again Fig. 4), have the same isotope composition as one of these clusters (Fig. 8). The question arises as to whether this is due to alloying of gold with copper from the same source as the copper-based artifacts originated from. In other words, could it be that the copper content of one to six percent in the gold

Fig. 8. Lead isotope ratios of gold artifacts compared to copper and copper-based artifacts from Ur which were analyzed by Begemann and Schmitt-Strecker (2009). The gold artifacts are compatible with one of the two clusters formed by the copper(-based) artifacts. Isotope ratios are measured by MC-­ICPMS and analytical uncertainty is given by the error bars as standard deviation (2σ).

Trace Elements and Isotopes: The Origin of Gold from Ur

281

objects is a reflection of intentional alloying? In this case, the lead traces in the gold could originate from the alloyed copper. Additionally, the lead concentration would positively correlate with the main component of the lead-introducing mineral. Most of the copper artifacts previously analyzed for Pb isotope ratios by Bege­ mann and Schmitt-Strecker (2009) were also analyzed for their chemical composi­ tion by Lutz and Pernicka (2004). We are able to calculate from these analyses an expected lead content for the gold objects under the assumption of deliberate alloy­ ing with this specific copper (Fig. 9, upper part). The result is that the calculated lead content in fact is much lower than the determined lead content in the artifacts and there is no positive correlation with increasing copper content. Thus, the lead trace in the gold artifacts does not originate from an alloyed copper. For the copper artifacts of Ur, arsenic was found to be a significant element amounting to several percent (Lutz and Pernicka 2004). To follow the argumen­ tation concerning the copper being intentionally added to the gold, we would in parallel expect an elevated arsenic content for the gold artifacts and again a positive correlation with copper. The trace element arsenic is not elevated in such quantities

Fig. 9. Hypothetical and measured lead and arsenic contents in the gold artifacts. Calculation based on the Neutron Activation Analyses of those copper-based artifacts from Ur, analyzed by Lutz and Pernicka (2004), which show consistent Pb isotope ratios to the gold artifacts (compare Fig. 8). Deter­ mined lead concentrations are higher and arsenic concentrations are lower than calculated, and there is no correlation between these and the copper content. Concentrations in the gold are determined by LA-ICP-MS.

282

Moritz Jansen, Andreas Hauptmann, Sabine Klein, & Richard L. Zettler

and just as observed for lead, neither there is a correlation of arsenic and copper (Fig. 9, lower part). To conclude this chain of argumentation, a contamination due to non-deliberate co-melting of placer gold with accessory lead-bearing heavy minerals like galena, pyromorphite, or cerussite and only minor amounts of arsenic-bearing minerals like arsenopyrite, both originating from the panned gold concentrate, is a more realistic explanation for the “metallurgical” origin of the lead and arsenic trace than inten­ tional alloying of gold with copper. At this point, an intermediate conclusion would result as follows: Since the ar­ tifact gold can be characterized as originating from alluvial gold (see above), the Pb isotope ratios reflect lead-bearing mineral sources of a region, eroded and deposited in placers, and not an individual ore occurrence and certainly not of the primary gold source. The eroded sources of the lead-bearing minerals have the same Pb isotope signature as a group of copper artifacts from Ur. The gold occurrences could be in proximity to the sources of these copper artifacts. So, what is the “metallurgi­ cal” origin of the copper content itself if we rule out intentional alloying for the gold artifacts? The identification of copper (Cu) isotope signatures allows a characterization of the types of minerals that were used to produce ancient copper (Klein et al. 2010). The Cu isotope signature of the gold from Ur shows a range of positive delta values (Fig. 10), which is typical for a source of oxidized minerals. Heavy mineral concentra­ tes can contain oxidized copper minerals like malachite or chrysocolla (Hauptmann et al. 2010). The copper isotope signature thus indicates that the copper in the gold from Ur originates from co-melting of a heavy mineral concentrate containing such minerals. The measured Cu isotope signature of the copper component in the gold artifacts is also different from that of copper-based artifacts from Ur, which tend to have lower δ65Cu values (see Salzmann et al. 2016), adding to the argument that the copper in the gold artifacts does not originate from intentional alloying with the same copper used for the copper-based objects.

Fig. 10. Copper isotopic composition of the gold artifacts from Ur. The positive δ65Cu values indicate oxidized copper minerals in the gold-bearing heavy mineral concentrates. Isotope ratios are measured by MC-ICP-MS.

Trace Elements and Isotopes: The Origin of Gold from Ur

283

PGE Mineral Inclusions A significant characteristic of the gold artifacts from Ur, first identified by Young (1972) and extensively studied by Meeks and Tite (1980), are numerous inclusions of a natural alloy component made of elements of the so-called Platinum Group (PGE). This group includes the six elements platinum (Pt), palladium (Pd), iridium (Ir), osmium (Os), ruthenium (Ru), and rhodium (Rh). In more than half of all samples from Ur, PGE mineral (PGM) inclusions were found to be present (Fig. 11). To cite another context just to give an impression, in only two samples of 120 prehistoric artifacts from a previous project in Georgia we could identify such inclusions (Jansen et al., forthcoming). The PGM inclusions in gold from Ur were already the focus of a previous article by the authors (Jansen et al. 2016). They consist of the iridium-like PGE osmium (Os), iridium (Ir) and ruthenium (Ru), and occasionally some platinum in addition.

Fig. 11. PGM inclusion of about fifty µm in diameter at center, surrounded by several smaller ones, identified in a gold sheet of a bead (B17590) from the King’s Grave (PG 789). These inclusions, consist­ ing of osmium, iridium, and ruthenium are typical for the gold artifacts from Ur. Scanning electron micrograph with backscatter detection.

The Os-Ir-R alloys are not soluble in gold. Metallurgical operations do not alter the chemical and Os isotope composition. Small inclusions of differing chemical com­ position found next to major PGM inclusions in goldwork of the Southern Urals and the Northern Black Sea region were recently interpreted as caused by metallurgical operations forming “secondary inclusions” (Zaykov et al. 2017). These “secondary inclusions” are distinct by a much lower Os content which is not consistent with Os-Ir-Ru alloys from deposits. It is questionable whether this observation is really related to metallurgical operations causing selective oxidation and dispersion of osmium in the gold melt as proposed by Zaykov et al. (2017). The “secondary inclu­ sions” are reported to comprise mainly ruthenium and iridium, two PGEs which are even more soluble in gold than osmium (see Zwicker 1998), so that rather this alloy than osmium would disperse in the gold melt. This type of inclusions would thus not be stable and observable as individual inclusions in a gold matrix, contra Zaykov

284

Moritz Jansen, Andreas Hauptmann, Sabine Klein, & Richard L. Zettler

et al. (2017). A long-term surface alteration of the PGE minerals due to weathering processes within the placer deposits might be more reasonable to explain these “secondary inclusions.” Our results of the investigation of PGM inclusions in the gold artifacts from Ur show that the chemical composition of these inclusions is compatible with PGE alloys originating from ophiolite complexes where they formed in chromite of ultra­ basic rocks (Jansen et al. 2016). In this geological context, Os-Ir alloys with high Ru content occur (Tolstykh et al. 2005). High Ru amounts were also determined for the inclusions in the gold artifacts from Ur. By calculating different geochronological models, a possible age of 290–610 Ma (Time of Rhenium Depletion—TRD) can be estimated for the crystallization of the alloys using their most frequent Os isotopic composition. The elevated content of platinum (Pt) and palladium (Pd), which is found with a correlation in the trace element pattern of the gold from Ur (Jansen et al. 2016), is not present as individual mineral inclusions. They originate from a second PGM source aside from the source for the Os-Ir-inclusions: Pt-Pd(-Fe) minerals, which in contrast to Os-Ir-Ru alloys are dissolved in the melt since Pt and Pd are soluble in gold. To summarize, the gold-rich heavy mineral concentrates contained natural OsIr-Ru alloys and Pt-Pd minerals, which behave differently in the concentrate during the co-melting process. Or-Ir-Ru alloys were not decomposed and are preserved as inclusions while Pt-Pd minerals were dissolved in the melt. Primary occurrences of both Os-Ir-Ru alloys and Pt-Pd minerals can be found within rocks of the same ophiolite complexes.

Implications for the Origin of the Gold With the geochemical characterization of the artifacts in hand, possible origins of the gold from Ur can be discussed. The Tethyan Eurasian Metallogenetic Belt (TEMB) is a geologic structure which hosts most of the ore deposits in the Middle East. It ranges from the Alps, Balkan, Anatolia, Iran, Afghanistan, and further East to the Himalayas. Comparing Pb isotope ratios of deposits of the TEMB, the following picture can be drawn: The Pb isotope composition of ore deposits from Anatolia and Iran fits with the gold artifacts from Ur (Fig. 12), but the respective ore deposits within the TEMB are generally difficult to distinguish (Begemann and Schmitt-Strecker 2009). Little is known about the Pb isotopic composition of deposits from Afghanistan. The rare datasets show some highly radiogenic Pb isotope compositions like the deposits of Mes Aynak (Begemann and Schmitt-Strecker 2009). There is an overlap with a few ore specimens, though additional samples are needed but not yet in sight. From the perspective of the Pb isotope composition, the general conclusion is that the gold artifacts from Ur fit well with the ore deposits of the TEMB. Besides ore deposits, the Tethyan Eurasian Metallogenetic Belt (TEMB) hosts the majority of the ophiolites (primary occurrences of PGM) in the Old World. Most of these ophiolites can be dated to the Mesozoic. The calculated formation date of the PGM minerals, which occur as inclusions in the gold from Ur is higher than the age of Mesozoic ophiolites. Thus, these ophiolites can be excluded as source for the PGM inclusions in the gold from Ur. This is also demonstrated by a direct comparison of

Trace Elements and Isotopes: The Origin of Gold from Ur

285

Fig. 12. Lead isotopic composition of the gold artifacts from Ur compared to ore deposits from Turkey (Wagner et al. 1986) and Iran (Mirnejad et al. 2011; Pernicka et al. 2011). The gold artifacts are com­ patible with this composition. Analytical uncertainty is given by the error bars as standard deviation (2σ).

Os isotope ratios from the Oman ophiolite, which is presently the best investigated Mesozoic example within the TEMB. Compared with the PGM inclusions in the Ur gold, the Oman ophiolite have deviate Os isotopic composition with higher ratios (Jansen et al. 2016). At first appearance, this observation seems to be contradictory to our interpretation of the Pb isotope ratios which are consistent with the deposits of the TEMB. We come back to this problem again. Older ophiolite complexes can be found in Egypt and Arabia. They were formed 700 to 900 Ma ago (Furnes et al. 2015). Cumulative there are gold occurrences in the Eastern Desert in close vicinity to these Proterozoic ophiolite complexes. They were exploited since the third millennium BCE (Klemm and Klemm 2013). The gold-silver ratio of the gold deposits of the Eastern Desert is similar to the artifacts from Ur, which reflect a natural composition (see above). In the south, the occurren­ ces from Nubia contain less silver, hence they can be excluded as source. Despite the above similarities with the Eastern Desert gold, all deposits of Egypt and Saudi Arabia can finally be excluded as sources for the gold from Ur from the Pb isotope point of view. If these deposits were exploited, the gold artifacts should have the same Pb isotopic composition as deposits from these regions. Instead, they are not compatible in Pb isotope ratios (e.g. Pb isotope data of the respective mining areas

286

Moritz Jansen, Andreas Hauptmann, Sabine Klein, & Richard L. Zettler

published by Doe and Rohrbough 1977; Sots-Gale and Gale 1981). We also excluded the Egyptian deposits through comparison of Os isotopes (Jansen et al. 2016). The Os isotope composition of the inclusions in the gold artifacts from Ur is different from that of the Os-Ir-minerals in the ophiolites of the Eastern Desert, the latter having lower ratios. In addition, the Eastern Desert ophiolites are geologically older than the calculated model age for the Ur gold. As mentioned before, it seems to be a contradictory fact that the source of the gold, respectively its lead trace, is consistent with deposits of the TEMB from a Pb isotope perspective, but its Mesozoic ophiolite complexes can be excluded as the source for the PGM inclusions. Potential ophiolites are only those which are older than Mesozoic (TEMB) and younger than Proterozoic ones (e.g. those of Egypt). A subsequent approach to the problem is a close look at very particular remnants of ophiolites: in the Middle East, some Paleozoic ophiolites occur, and the crystal­ lization age of the PGM inclusions fits their age. These are very rare, and hence a key to identifying the gold sources for Ur. Unfortunately, none of the Paleozoic ophiolites from the Middle East is yet characterized for Os isotope ratios for a direct comparison of the PGM inclusions in the Ur gold, hence we have to focus here on a mapping of potential occurrences (Fig. 13).

Fig. 13. Mapping of the Neoproterozoic and Paleozoic ophiolite relics in the Middle East which are consistent with the crystallization age (TRD) of the PGM inclusions in the gold from Ur. Ophiolites with gold deposits in the immediate vicinity are marked as black stars, while ophiolites without known gold deposits are marked as black circles. 1: Paleozoic ophiolites near placer deposits of Zarshuran in Takab (Iran); 2: Paleozoic ophiolites near placer deposits of Samti at the Amu Darya river in Takhar (Afghanistan). Some archaeological sites (white circles) and the lapis lazuli occurrence of Sar-i Sang (North-East Afghanistan) are also mapped (dark grey diamond). The numerous Mesozoic ophiolites are not listed since they are younger than the source of PGM inclusions from Ur.

For Anatolia and the Caucasus, there are no gold occurrences in proximity to Paleozoic ophiolites. In Iran, the largest gold reserve is close to Paleozoic ophiolites in Takab. There are very rich gold placers near Zarshuran (Fig. 13) with panning sites at Yaraziz, Yengikand, and Ghourouchay, which were also worked in pre-

Trace Elements and Isotopes: The Origin of Gold from Ur

287

modern times, and local mining activities go back at least to Parthian and Sassanian times (Momenzadeh et al. 2016). The earlier use of these alluvial occurrences could be reflected in the geochemical fingerprint of the artifacts from Ur, hence they are a potential candidate for the geological origin of the gold artifacts. Typical for the gold artifacts of Ur is the close association with lapis lazuli and carnelian. There are some analogies between the inventories of gold artifacts from the contemporaneous Harappan culture of the Indus Valley and the gold artifacts from Ur, including an association with gemstones, the shape of diadems, and the discoid beads which are found in Mesopotamia and beyond. Just a few analyses were done for the Harappan gold and focused on the composition of the main elements (Kenoyer and Miller 1999). Five artifacts from the site Harappa itself contain less silver (6–9 wt. %) than the Ur artifacts, while two artifacts from Lothal (thirty-four and forty-two wt. %) have the same silver content. Stylistics provides an additional hint, but must be considered with great care. While Miller (2000) identified some of the leaves from the headdresses of Puabi as poplar which also grows locally in Southern Mesopotamia, these leaves were interpreted by Tengberg et al. (2008) as the Indian Rosewood (Dalbergia sissoo Roxb.) which indicates an association to the East. The present-day distribution of this tree is in the Indus Valley, Baluchistan, and the Asian coastal area of the Gulf of Oman. As discussed in the introduction, the later cuneiform tablets also described gold deriving from Meluhha, identified as the present-day Lower Indus Valley. The geological situation in the Lower Indus Valley is similar to that of Southern Mesopotamia. Gold deposits are not present. The gold sources of the Harappan cul­ ture therefore were likely to have been in the North or North-West of the Upper Indus Valley, potentially even in Afghanistan or Eastern Iran. Cultural contact between Ur and Afghanistan is evident through the presence of numerous artifacts of lapis lazuli found in the Royal Tombs of Ur. As a mineralogical rarity, lapis lazuli occurs within the Middle East only in North-Eastern Afghanistan. In Afghanistan, some relics of Paleozoic ophiolites exist in Badakhshan. In proximity to the ophiolites, the richest alluvial gold deposits all over in Afghanistan can be found at the Amu Darya River near Samti in Northern Takhar (Fig. 13). The lapis lazuli deposit of Sar-e Sang, which supplied the entire Middle East in the Bronze Age is close to these placer deposits. There are also occurrences of tin placers in Western Badakhshan which are interesting in context with the early bronze objects in Ur. In combination with the above-mentioned suite of materials, this presents convincing evidence for the supply of a variety of raw materials from Afghanistan to Mesopotamia (Stech and Pigott 1986). Finally, the area of Amu Darya is of special interest, because also the Harappan trading post Shortugai was located there. Although it is hundreds of kilometers away from the Indus Valley, Shortugai has the typical inventory of a Ha­ rappan city with its architecture, ceramics, seals, copper related metallurgy, beads of gemstones, but also a few beads made of gold (Francfort 1989). Hence, Shortugai indicates a direct connection between the Indus Valley and a region of interest for gold and other raw materials. The geological setting of the region is consistent with the geochemical fingerprint of the gold artifacts from Ur, making the placer deposits of Samti a potential candidate for the geological origin of the gold artifacts from Ur.

288

Moritz Jansen, Andreas Hauptmann, Sabine Klein, & Richard L. Zettler

Conclusions The gold jewelry of headdresses from Ur analyzed in this study consists of a natural alloy of gold and silver originating from its gold source. In addition, it contains some copper, presumably from an oxidized mineral source. There are high concentrations of tin, platinum, lead, and elevated concentrations of other trace elements. Besides their silver content, natural free-gold alloys are very pure, and thus their trace element concentration is low, but artifact gold produced from goldrich concentrates such as from placer deposits or reef gold deposits has detectable amounts of various accompanying elements. This is due to contamination that happened naturally during the metallurgical processing of panned gold, which must be understood as a concentrate with relicts of distinct heavy minerals such as cop­ per oxides, galena, cassiterite, PGM, and other mineral grains. The metallurgical process is thus a co-melting operation, introducing various amounts of minor and trace elements and producing gold for artifacts that is different in composition than the primary gold source. The co-occurrence of tin and platinum in the trace element pattern as well as PGM inclusions are characteristic for artifacts from placer gold. There are no chronological developments detectable for this group of artifacts in the composition or recipe of the gold in terms of alloying components or the trace element pattern between Early Dynastic IIIa and Akkadian/Third Dynastic period of Ur. The mineral sources of the trace element lead were identified as lead-bearing heavy minerals, which are not older than 200 Ma. There are no changes in the Pb isotope composition between Early Dynastic IIIa and Akkadian/Third Dynastic period. This is readily explained by the exploitation of the same sources of gold over hundreds of years. The PGM inclusions are Os-Ir-Ru alloys, which are not soluble in gold. The PGE minerals originate from ophiolites, which were formed 290–610 Ma ago. In addition, Pt-Pd minerals were a component of the gold-rich heavy mineral concentrate, not visible as inclusions but found as trace element components in the gold. They were dissolved in the gold during metallurgical co-melting. The source of the gold from Ur is likely located within the Tethyan Eurasian Metallogenetic Belt, since the Pb isotope ratios are consistent with ore deposits from this orogenic structure. The PGM inclusions in the gold artifacts trace back to older ophiolite complexes within the TEMB. Possible origins of these are rare remnants of Paleozoic ophiolites, which are consistent with the crystallization age of the PGM inclusions. There are two locations that are potential candidates for the origin of the gold found at Ur: These two sources are the placer deposits at Takab in Iran and at Samti in Afghanistan. The geological settings of these occurrences are consistent with the geochemical fingerprint. These mineral sources should be urgently investigated in the future to compare the Pb isotopes of gold-bearing heavy mineral concentrates and Os isotope ratios of PGE minerals in local chromite oc­ currences. This will add most valuable information to the discussion and contribute significantly to the understanding of long-distance trade networks for raw materials in Bronze Age Mesopotamia.

Acknowledgments This study was funded by the German Science Foundation (Deutsche For­ schungsgemeinschaft) through a project on the analytical processing of the metals

Trace Elements and Isotopes: The Origin of Gold from Ur

289

from the Ur collection in Philadelphia. We thank the staff of the University of Pennsylvania Museum, especially Head Conservator Lynn Grant and Keeper of Collections Katherine Blanchard for the opportunity of sampling gold artifacts. We are gratefully for the support during sample preparation and measurements by Michael Bode, Heidi Höfer, Ingo Horn, Dirk Kirchner, Michael Krüger, Regina Kutz, Andreas Ludwig, Linda Marko, Sandra Morszeck, Hans-Michael Seitz, Melanie Sieber, and Georg Wange.

Bibliography Armbruster, B. 2016 Technological Aspects of Selected Gold Objects from Ur – Preliminary Results and Perspectives. Metalla 22: 113–135. Begemann, F. and Schmitt-Strecker, S. 2009 Über das frühe Kupfer Mesopotamiens. Extrait d’Iranica Antiqua 44: 1–45. Buchanan, B. 1954 The Date of the So-called Second Dynasty Graves of the Royal Cemetery of Ur. JAOS 74: 147–153. Doe, B. R. and Rohrbough, R. 1977 Lead Isotope Data Bank: 3,459 Samples and Analyses Cited. USGS Numbered Series Open-File Report 79-661. Reston: U.S. Geological Survey. Ehser, A., Borg, G., and Pernicka, E. 2011 Provenance of the Gold of the Early Bronze Age Nebra Disk, Central Germany. Geochemical Characterization of Natural Gold from Cornwall. European Journal of Mineralogy 23: 895–910. Francfort, H.-P. 1989 Fouilles de Shortugai. Recherches sur l’Asie Centrale Protohistorique, Memoires de la Mission Archeologique Francaise en Asie Centrale. Paris: Diffusion de Boccard. Furnes, H., Dilek, Y., and de Wit, M. 2015 Precambrian Greenstone Sequences Represent Different Ophiolite Types. Gondwana Research 27: 649–685. Guerra, M. F. and Calligaro, T. 2004 Gold Traces to Trace Gold. Journal of Archaeological Science 31: 1199–1208. Guerra, M. F., Sarthre, C.-O., Gondonneau, A., and Barrandon, J.-N. 1999 Precious Metals and Provenance Enquiries using LA-ICP-MS. Journal of Archaeological Science 26: 1101–1110. Hauptmann, A. and Klein, S. 2009 Bronze Age Gold in Southern Georgia. ArchéoSciences, revue d’archéométrie 33: 75–82. 2016 Golden Artifacts from the Royal Tombs of Ur, Mesopotamia. Metalla 22: 84–88. Hauptmann, A., Bendall, C., Brey, G., Japarize, I., Gambashize, I., Klein, S., Prange, M., and Stöllner, T. 2010 Gold in Georgien. Analytische Untersuchungen an Goldartefakten und an Natur­ gold aus dem Kaukasus und dem Transkaukasus. Pp. 139–160 in Von Majkop bis Trialeti. Gewinnung und Verbreitung von Metallen und Obsidian in Kaukasien im 4.-2. Jt. v. Chr., eds. S. Hansen, A. Hauptmann, I. Motzenbäcker, and E. Pernicka. Bonn: Habelt. Hauptmann, A., Klein, S., Paoletti, P., Zettler, R. L., and Jansen, M. 2018 Types of Gold, Types of Silver: The Composition of Noble Metal Artifacts Found in the Royal Tombs of Ur, Mesopotamia. ZA 108: 100–131.

290

Moritz Jansen, Andreas Hauptmann, Sabine Klein, & Richard L. Zettler

Hauptmann, A., Klein, S., Zettler, R. L., La Niece, S., Benzel, K., Armbruster, B., Jansen, M., and Salzmann, E. 2016 The Royal Tombs of Ur, Mesopotamia. New Investigations, New Results from the Examination of Metal Artifacts and other Archaeological Finds. Collection of Contributions from a Workshop at the Deutsches Bergbau-Museum. Metalla 22: 77–146. Jansen, M., Aulbach, S., Hauptmann, A., Höfer, H. E., Klein, S., Krüger, M., and Zettler, R. L. 2016 Platinum Group placer minerals in ancient Gold artifacts – Geochemistry and Osmium Isotopes of Inclusions in Early Bronze Age Gold from Ur/Mesopotamia. Journal of Archaeological Science 68: 12–23. Jansen, M., Hauptmann, A., Klein, S., Gambashidze, I., and Jalilov, B. fth. Geochemistry of Gold from the Prehistoric Mine of Sakdrisi and Transcaucasian Gold Artifacts between the 4th and 2nd Millennia BC. In On salt, Copper and Gold: The Origins of Early Mining and Metallurgy in the Caucasus, eds. C. Marro and T. Stöllner. Conference proceedings to be published in Les Travaux de la Maison de l’Orient. Kenoyer, J. M. and Miller, H.-L. 1999 Metal Technologies of the Indus Valley Tradition in Pakistan. Pp. 107–151 in The Archaeometallurgy of the Asian Old World, ed. V. C. Pigott. MASCA Research Papers in Science and Archaeology 16. Philadelphia: University Museum Univer­ sity of Pennsylvania. Klein, S., Brey, G. P., Durali-Müller, S., and Lahaye, Y. 2010 Characterisation of the Raw Metal Sources Used for the Production of Copper and Copper Based Objects with Copper Isotopes. Archaeological and Anthropological Sciences 2: 45–56. Klemm, R. and Klemm, D. 2013 Gold and Gold Mining in Ancient Egypt and Nubia – Geoarchaeology of the Ancient Gold Mining Sites in the Egyptian and Sudanese Eastern Deserts. Natural Science in Archaeology. Berlin, Heidelberg: Springer. Kovacs, R., Schlosser, S., Staub, S. P., Schmiderer, A., Pernicka, E., and Günther, D. 2009 Characterization of Calibration Materials for Trace Element Analysis and Fingerprint Studies of Gold using LA-ICP-MS. Journal of Analytical Atomic Spectrometry 24: 476–483. La Niece, S. 1995 Depletion Gilding from Third Millennium BC Ur. Iraq 57: 41–47. Leusch, V., Brauns, M., and Pernicka, E. 2016 Precise and Accurate Analysis of Gold Alloys: Varna, the Earliest Gold of Man­ kind – A Case Study. Pp. 95–113 in Recent Advances in Laser Ablation ICP-MS for Archaeology, eds. L. Dussubieux, M. Golitko, and B. Gratuze. Natural Science in Archaeology. Berlin, Heidelberg: Springer. Lutz, J. and Pernicka, E. 2004 Neutronenaktivierungsanalysen. Pp. 104–108 in Die Metallindustrie in Mesopotamien von den Anfängen bis zum 2. Jahrtausend v. Chr., eds. H. Hauptmann and E. Pernicka. Orient-Archäologie 3. Rahden/Westfalen: Verlag Marie Leidorf. Maxwell-Hyslop, K. R. 1977 Sources of Sumerian Gold: The Ur Goldwork from the Brotherton Library, Uni­ versity of Leeds. A Preliminary Report. Iraq 39: 83–86. Meeks, N. D. and Tite, M. S. 1980 The Analysis of Platinum-group Element Inclusions in Gold Antiquities. Journal of Archaeological Science 7: 267–275.

Trace Elements and Isotopes: The Origin of Gold from Ur

291

Miller, N. 2000 Plant Forms in Jewellery from the Royal Cemetery at Ur. Iraq 62: 149–155. Mirnejad, H., Simonetti, A., and Molasalehi, F. 2011 Pb Isotopic Compositions of Some Zn-Pb Deposits and Occurrences from UrumiehDokhtar and Sanadaj-Sirjan Zones in Iran. Ore Geology Reviews 39: 181–187. Momenzadeh, M., Nezafati, N., Sarraf, M. R., and Shabani, K. 2016 Ancient Gold-Mercury Mining in the Takht-e Soleyman Area, Northwest Iran. Metalla 22: 147–168. Moorey, P. R. S. 1999 Ancient Mesopotamian Materials and Industries: The Archaeological Evidence. Winona Lake: Eisenbrauns. Pernicka, E., Adam, K., Böhme, M., Hezarkhani, Z., Nezafati, N., Schreiner, M., Winterhol­ ler, B., Momenzadeh, M., and Vatandoust, A. 2011 Archaeometallurgical Researches at Arisman in Central Iran. Pp. 633–705 in Early mining and Metallurgy on the Central Iranian Plateau. Report on the First Five Years of Research of the Joint Iranian-German Research Project, eds. A. Vatandoust, H. Parzinger, and B. Helwing. Archäologie in Iran und Turan 9. Mainz am Rhein: Zabern. Potts, D. T. 1997 Mesopotamian Civilization: The Material Foundations. Ithaca: Cornell University Press. Potts, T. 1994 Mesopotamia and the East: An Archaeological and Historical Study of Foreign Relations ca. 3400–2000 B.C. Monograph 37. Oxford: University Committee for Archaeology. Salzmann, E., Klein, S., and Hauptmann, A. 2016 Analytical Investigations on Silver, Copper and the Earliest Tin Bronzes from Ur. Metalla 22: 141–145. Schlosser, S., Reinecke, A., Schwab, R., Pernicka, E., Sonetra, S., and Laychour, V. 2012 Early Cambodian Gold and Silver from Prohear. Composition, Trace Elements and Gilding. Journal of Archaeological Science 39: 2877–2887. Stacey, J. S. and Kramers, J. D. 1975 Approximation of Terrestrial Lead Isotope Evolution by a Two-Stage Model. Earth and Planetary Science Letters 26: 207–221. Stech, T. and Pigott, V. C. 1986 Metals Trade in Southwest Asia in the Third Millennium B.C. Iraq 48: 39–64. Stöllner, T. 2003 Mining and Economy – A Discussion of Spatial Organisations and Structures of Early Raw Material Exploitation. Pp. 415–446 in Man and Mining – Mensch und Bergbau. Studies in Honour of Gerd Weisgerber on Occasion of his 65th Birthday, eds. T. Stöllner, G. Körlin, G. Steffens, and J. Cierny. Der Anschnitt, Beiheft 16. Bochum: Deutsches Bergbau-Museum. Stos-Gale, Z. A. and Gale, N. H. 1981 Sources of Galena, Lead and Silver in Predynastic Egypt. Actes du XXme Symposium International d’Archéometrie, Revue d’Archéometrie, Supplement 1: 285–295. Tengberg, M., Potts, D. T., and Francfort, H.-P. 2008 The Golden Leaves of Ur. Antiquity 82: 925–936.

292

Moritz Jansen, Andreas Hauptmann, Sabine Klein, & Richard L. Zettler

Tolstykh, N. D., Sidorov, E. G., and Krivenko, A. P. 2005 Platinum-Group Element Placers Associated with Ural-Alaska Type Complexes. Pp. 113–143 in Exploration of Platinum-Group Elements Deposits, ed. J. E. Mungall. Short Course Series Volume 35. Quebec: Mineralogical Association of Canada. Wagner, G. A., Pernicka, E., Seeliger, T. G., Lorenz, I. B., Begemann, F., Schmitt-Strecker, S., Eibner, C., and Öztunali, Ö. 1986 Geochemische und isotopische Charakteristika früher Rohstoffquellen für Kup­ fer, Blei, Silber und Gold in der Türkei. Jahrbuch des Römisch-Germanischen Zentralmuseums 33: 723–752. Wolf, D., Borg, G., Meliksetian, K., Allenberg, A., Pernicka, E., Hovanissyan, A., and Kunze, R. 2013 Neue Quellen für altes Gold? Pp. 27–48 in Archäologie in Armenien II, eds. H. Meller and P. Avetisyan. Veröffentlichungen des Landesamtes für Denkmal­pflege und Archäologie für Vorgeschichte 67. Halle: Landesamt für Denkmalpflege und Archäologie Sachsen-Anhalt. Wolf, D., Borg, G., Pernicka, E., Meliksetian, K., Kunze, R., and Bobokhyan, A. 2011 Geoarchäologische Untersuchungen der Goldvorkommen von Sotk und Fioletovo, Armenien. Pp. 51–68 in Archäologie in Armenien, eds. H. Meller and P. Aveti­ syan. Veröffentlichungen des Landesamtes für Denkmalpflege und Archäologie für Vorgeschichte 64. Halle: Landesamt für Denkmalpflege und Archäologie Sachsen-Anhalt. Wolf, D. and Kunze, R. 2014 Gegharkunik – Neue Quellen für altes Gold aus Südkaukasien? Pp. 111–141 in Metalle der Macht – Frühes Gold und Silber / Metals of Power – Early Gold and Silver, eds. H. Meller, R. Risch, and E. Pernicka. Tagungen des Landesmuseums für Vorgeschichte Halle 11/I. Halle: Landesamt für Denkmalpflege und Archäolo­ gie Sachsen-Anhalt. Young, W. I. 1972 The Fabulous Gold of the Pactalos Valley. Bulletin of the Museum of Fine Arts 70.359: 5–13. Zaykov, V. V., Kotlyarov, V. A., Zaykova, E. V., and Melekestseva, I. Y. 2017 The Phenomenon of the Influence of Gold Melt on Microinclusions of Platinum Group Minerals in Ancient Gold Objects. Archaeometry 59: 96–104. Zettler, R. L. and Horne, L., eds. 1998 Treasures from the Royal Tombs of Ur. Philadelphia: University of Pennsylvania Museum of Archaeology and Anthropology. Zwicker, U. 1998 An Investigation of Inclusions of Platinum-Group Metals in Ancient Coinage. Metallurgy in Numismatics 4: 171–201.

Science and Technology: Using Ur-Online to Aid in Scientific Analysis Kyra Kaercher and Tessa de Alarcon Penn Museum

Ur of the Chaldees: A Virtual Vision of Woolley’s Excavations, a joint project between the Penn Museum and the British Museum, has led to the re-analysis of many objects from Ur. Ur-Online.org, the final product of this project, brings together modern and archival descriptions along with field records into one digital platform. As part of this project, conservators and researchers at both museums analyzed objects in order to gather modern information. At the Penn Museum this involved recording modern measurements and descriptions, taking new photographs, and conducting a condition assessment of all of the artifacts outlined in the proposal. Some objects were selected for further analysis. A petrographic study of clay objects, a portable X-ray fluorescence (PXRF) study of obsidian, Infrared Reflectance and X-rays of various materials, and photomicrographs of pseudomorphs have all been performed in the past three years. This paper explains these studies and their preliminary results, and shows how Ur-Online.org can be used to display these results. The scientific analysis of these materials can be used to refine what we know about Woolley’s excavations and the ancient city of Ur.

Introduction Ur of the Chaldees: A Virtual Vision of Woolley’s Excavations, was created with lead funding from the Leon Levy Foundation. 1 The project was carried out from 2013–2016, with teams at the British Museum and the Penn Museum completing analysis and description of the archives and objects from the excavations at Ur. Since 2016, the end product of the project, Ur-Online.org has been active, displaying the information that was gathered over the past three years. The website was conceived as a giant relational database bringing together primary sources, archival 1.  We would like to thank the other members of the Ur Project, Penn Museum Team: Co-Principle Investigators: Dr. Richard Zettler, and Dr. Steve Tinney, Project Manager: Dr. William Hafford, and Project Database Administrator and Developer: Sasha Renninger. We would also like to thank the Penn Museum Conservation Department: Head Conservator: Dr. Lynn Grant and Associate Conservator: Nina Owczarek, the Keeper of the Near East Section: Katy Blanchard, the Center for the Analysis of Ancient Materials Department: Laboratory Coordinator: Dr. Marie-Claude Boileau, and Archaemetallurgical Specialist: Moritz Jansen, the Keeper of Physical Anthropology Section: Dr. Janet Monge, and Penn Museum Consulting Scholar: Dr. Naomi Miller.

293

294

Kyra Kaercher and Tessa de Alarcon

documents, and modern descriptions of locations, people, media, and the objects. The Penn Museum project researcher, Kyra Kaercher, and the Penn Museum project conservator, Tessa de Alarcon, have performed multiple scientific projects on the objects, and decided how to best display the results of these projects online. The ruins of Ur are located in southern Iraq, about ten kilometers west of the modern town of Nasiriyeh and the Euphrates River. Ur was an important center for kingship and later kings often restored the ziggurat and temples. The importance is indicated by the lavish burials of the Royal Cemetery dating to the mid-third millennium and contains habitation stretching from around 5500 BCE to approximately 300 BCE. Major excavations occurred between 1922 and 1934 under the auspices of the British Museum and the Penn Museum and were directed by Sir Leonard Woolley. Woolley recorded a great deal of data on field note cards, in field photos, and in letters and reports he wrote to the museums. This information was used to create detailed publications over the next fifty years. Woolley wrote preliminary reports and newspaper articles every year and completed scholarly and popular books both during and after the excavation. Ur-Online is a database that brings together the archival information with the actual artifacts in one easily accessible spot. Since the original excavation was performed by the British Museum and the Penn Museum with permission of the Iraqi Government, the objects were spilt between the three partners. Half of the objects remained in Iraq, a quarter went to the British Museum and the remaining quarter went to the Penn Museum. Archives of each museum received some of the original documents sent from the field and duplicates of others, but none have a complete set. Ur-Online was created to make a modern, dynamic, open access, online publication of the Ur excavations that will preserve forever in digital formats the complete finds and records.

Ur-Online Digital humanities is a branch of research that uses information technology as a central part of its methodology for creating and/or processing data. Ur-Online was created as a research site to be entirely open source with the code stored on GitHub for anyone to utilize and build upon. All of the data is under Creative Commons licensing, and is linked to other open data sites such as Google API (Application Program Interface), VIAF (Virtual International Authority File), and CDLI (Cuneiform Digital Library Initiative). When we began the project, we realized that we needed to record and interlink four major areas: Objects, Con­ texts, Media, and People (Fig. 1). This would describe and connect the what, where, and who of the excavation, and adding when as it becomes available through research. The hows and whys are left up to the researcher. The initial object database Fig. 1. Diagram of the relationships between the entities of was created by transcribing all of Ur-Online. Woolley’s catalogue cards, which

Science and Technology: Using Ur-Online

295

give us a baseline of information on what was collected at Ur. These cards provide information on objects such as field descriptions and measurements, field photos, locations, and any other information that Woolley provided. The next step was acquiring modern descriptions, measurements, photographs, and conservation of selected objects at the British Museum and the Penn Museum. The selected objects, as outlined in the proposal, included those made of clay and metal, as well as the inscribed objects such as tablets. Like the objects, contexts started with Woolley’s catalogue cards as a baseline. This data was supplemented with Woolley’s field notes, which were transcribed using crowdsourcing technology. Finally, modern analysis of the contexts have also been added, including Woolley’s published analysis as well as more modern reanalysis. The context page then displays any sub-contexts, i.e. burials in a grave, as well as objects flagged as having been excavated in the context. Media includes all of Woolley’s field notations (cards and notes), photographs, correspondence, and publications, as well as archival documentation from both the British and the Penn Museums. These documents not only provide information on the excavation and Woolley’s thought processes, they also provide insight into the history of archaeology as a discipline, the period of big digs, and the formation of the modern nation of Iraq. These documents have been supplemented with notes from previous curators, unpublished manuscripts, and modern articles about the objects, contexts, or periods at Ur. Lastly, we have recorded general information on the people who were important to the excavations and are maintaining a bibliography of works by them and others concerning the ancient city. This information also provides insights into the history of archaeology, as well as the early days of the Penn Museum. The goal of Ur-Online is to bring together all this data, make it accessible to everyone, and create an online research tool that acts as a modern digital publication—lasting, accessible, search­ able, and continually utilized and updated. In creating this database researchers at both museums analyzed multiple objects in order to answer multiple research questions. The rest of the paper includes brief outlines of the research performed at Penn, as well as preliminary results of these projects.

Research for Ur-Online Five main projects were carried out from 2014 to 2016. The first identified the sources of obsidian that were used in objects at Ur. This study was carried out by Kevin Ennis, with help from conservator Nina Owczarek, and project re­searcher Kyra Kaercher. The second project analyzed the petrographic composition of clay objects from Ur. This study was undertaken by Kaercher with the help of Dr. Marie-Claude Boileau from the Center for the Analysis of Ancient Materials. The third project was the use of Infrared Reflectance Photography on objects from Ur undertaken by project conservator Tessa de Alarcon. The fourth was use of Xradiography on multiple types of objects, from metals to clay to understand technique of manufacture, led by de Alarcon and Boileau. The fifth and last project was the exam­ination and documentation of organic pseudomorphs on metal objects, undertaken by de Alarcon and Kaercher.

296

Kyra Kaercher and Tessa de Alarcon

UPM Obsidian Sourcing Study The UPM Obsidian Sourcing Study was an undergraduate project conceived and carried out by Kevin Ennis, with help from Conservator Nina Owczarek and project researcher Kyra Kaercher. The goal of the project was to identify the sources of obsidian that were used to construct objects at Ur. Portable X-ray Fluorescence (PXRF) was chosen for the elemental analysis of the samples. The results of the analysis were compared to published results from Gollu Dag, Bingol, Nemrut Dag I and IV (Anatolia) and Gutansar (Armenia). Using PXRF for obsidian sourcing has been contested as a way to help with sourcing as the results have been shown to vary when compared with Neutron Activation Analysis (NAA) (Cauvin et al. 1997; Frahm 2014; Rollinson 1993). PXRF only analyzes elements on the surface of the object, and only measures the portion of the sample directly in front of the sensor. However, PXRF is a non-destructive technique, does not require a nuclear reactor so is low cost, easy to use, and can identify the full spectrum of elements that can be found with NAA (Frahm 2014; Fitton 1997; Rollinson 1993). Because of these reasons, this project used PXRF. Using the Bruker Tracer III-SD machine, the samples were run at 40kV, a current of 30mA, for 180 seconds with a titanium/aluminum/copper filter, a standard procedure for testing obsidian. Owczarek and Kaercher oversaw this portion of the project. The spectrum was obtained using S1PXRF and the elements were identified and the resulting numerical values were saved using the Artax software. These numbers were then converted into quantitative elemental ppm based on calibration from Bruker and MURR (Missouri University Research Reactor) (Speakman 2012). Problems have been addressed with converting wt% into ppm, but for this test project Ennis and Kaercher decided to convert the spectra numbers into ppm for ease of comparing to the results from published sites (Frahm 2014, Fitton 1997; Rollinson 1993; Speakman 2012). Ennis then took these results and performed a standard cluster analysis on five elements (Rubidium, Strontium, Yttrium, Zirconium, and Niobium) that were identified based on their amounts and correlations with the published studies. Ennis and Kaercher graphed the ratios of sev­ eral of the elements to one another following pub­ lished studies of Renfrew et al. 1966, Khazee et al. 2014, and Chataigner and Gratuze 2014. Fig. 2 shows Rubidium vs Zirconium ppm and has three clusters. The first cluster (n=6), located on the bottom of the graph, matches with Ennis’s Cluster 2, Gollu Dag (square), and Fig. 2. Scatter plot of Rb vs. Zr ratios of obsidian elements showGutansar (triangle). The ing three distinct clusters second cluster (n=6),

Science and Technology: Using Ur-Online matches with Ennis’s Cluster 2 as well, Bingol (dash), and Nemrut Dag I (circle). The third cluster (n=54) matches with Ennis’s Cluster 1, and no sources. Fig. 3 shows Zirconium vs. Yttrium in ppm and has three clusters. The first cluster (n=4), located on the bottom of the graph, matches with Ennis’s Cluster 2, Gollu Dag (square). The second cluster (n=6), matches with Ennis’s Cluster 2, and Bingol (dash). The third cluster (n=54) matches with Ennis’s Cluster 1, and Nemrut Dag IV (circle).

297

Fig. 3. Scatter plot of Zr vs. Y ratios of obsidian elements showing three distinct clusters.

Fig. 4. Map showing locations of sites in analysis. Gollu Dag (cross), Bingol (polygon), Nemrut Dag I and IV (circle), Guntansar (triangle), Tepe Gawra (rectangle), and Ur (star).

298

Kyra Kaercher and Tessa de Alarcon

The statistical analysis performed by Ennis created three main clusters (Cluster 1 n=39; Cluster 2 n=11; Cluster 3 n=7). We then added in the three samples from Gawra, and five published results (Gollu Dag, Bingol, Nemrut Dag I and IV, and Gutansar). Based on the statistical analysis and the graphing of the elemental concentrations (Figs. 2 and 3), Cluster 1 matches most closely with Bingol, Nemrut Dag I, and IV, in Central Turkey. Cluster 2 matches most closely with the site of Gollu Dag in Western Turkey. Cluster 3 matches most closely with Gutansar with the high Yttrium and Zirconium values. However, the values of the other elements are not close enough to Gutansar to make a strong correlation. The high Yttrium and Zirconium values negate correlation with the other sites in Turkey. As of right now, the exact source of Cluster 3 is unknown. The three cores from Gawra fall within the range of samples in Cluster 1, matching the sources located in Central Turkey. From these very small, preliminary results it seems as if the obsidian is mainly coming from the Lake Van Region/Central Turkey, traveling down the Tigris to Gawra, and continuing on to Ur. A few samples are coming from Gollu Dag, travel­

Fig. 5. Results of searching “obsidian sourcing” on Ur-Online.org.

Science and Technology: Using Ur-Online

299

ing down the Euphrates, and then on to Ur. Trade routes of the prehistoric period also point to the Tigris and Euphrates being main trading routes. The obsidian data from Ur seems to back up this theory. With more testing of obsidian from Ur, and other sites along the Tigris and Euphrates, we can get a more complete picture of trade during the prehistoric period. When you search “obsidian sourcing” in the ur-online.org keyword box, the fifty-seven tested samples from Ur are returned. As seen in Fig. 5, the return page shows the objects, all with modern measurements and descriptions, and on the right hand side are various filters. Expanding the object type filter shows the different object types that were included in the study. The final report to the section will be attached to all the records. Showing this data online, in this manner will hopefully give a starting point for future research on obsidian at Ur and throughout the Middle East.

UPM Petrography Study The ceramic petrography project was a branch of a larger project studying ceram­ ics at Ur. The original project was established for the Center for Analysis of Ancient Materials course ANTH 221/521 Material World in Archaeological Sciences under Dr. Marie-Claude Boileau. This course incorporates three four-week modules, one on lithics, one on ceramics, and one on metals. The site of Ur was chosen as a case study for the ceramics portion of this class to illustrate the use of clay in the past and how scientific data informs our understanding of ancient cultures. The project sought to compare technological properties between different types of clay-based objects from the site of Ur. Kaercher’s independent study focused on resource procure­ ment and sought to understand the choice and procurement of raw materials used to create the objects, as well as analyzing the objects to see if the procurement of the material is different between different classes of object. The objects chosen for the study consisted of architectural ceramics (bricks, wall cones, drainpipes, and reliefs), containers (potsherds, and glazed potsherds), and small objects (sickles, figurines, sealings, and incense burners). These objects are from all time periods and contexts, giving a representative sampling of the clay based objects from Ur. Added to this was a sample of a sterile layer of soil from Ur, which Boileau hydrated, formed into briquettes, and fired at different tempera­ tures. All objects were thin sectioned by an outside laboratory by impregnating the samples with epoxy, attaching them to a slide, grinding the samples to .03 mm and placing a cover slip on top of the sample. To look at the choice and procurement of raw materials, Kaercher compared the mineralogy of the thin sections to the larger geological context. The project was carried out following the methodology outlined by Whitbread (1986; 1989; 1995), where the samples were placed into fabric groups based on similarities in microstructure, groundmass, and inclusions. Inclusions and the mineralogy give the most evidence for provenience (Quinn 2013: 117). Seven fabrics were identified with two fabrics being divided into three subgroups. All these objects were produced locally, using local sources and tempering agents. The main inclusions of all the samples were quartz, feldspars, and micas, which are some of the most common elements on earth. These minerals make up the Aeolian sands, as well as the sediments found on the flanks of the Mesopotamian plains (Jassim and Goff 2006: 193–194). Since these minerals are the most common

300

Kyra Kaercher and Tessa de Alarcon

on earth, it is impossible to match them with an exact formation, but all the min­ erals found in these fabrics match the minerals found in the geology of the region surrounding Ur. The metamorphic fragments are found in the basement sediments around Ur (Jassim and Goff 2006: 30). Igneous fragments are found in the alluvial fans as well as in the Dibdibba formation (al Naqib 1963: 44–45; Scholtz and Van 1960). Mudstones and other sedimentary rocks are found in the Fars formations, as well as the Euphrates Limestone Formation (al Naqib 1963: 43; Scholtz and Van 1960). The chert may come from the outcroppings of the Upper Fars formation (al Naqib 1963:42; Scholtz and Van 1960). The mineralogy of the samples also matches the mineralogy of the geological sample from Ur. The groundmass of all samples consists of secondary marine clay sources. Sec­ ondary clay sources consist of finer grained, more evenly sorted grains; more abundant quartz, feldspar, and mica; more organic matter; and are more homogenous in color than residual clay sources (Quinn 2013: 122). All samples match this description. Marine sedimentary clay sources are marked by calcareous clay, shells, and clasts of calcimudstones (Armstrong and Brasier 2005; Quinn 2013: 122). The samples that do not have calcareous clays, shells, or detrital micrite are all highly fired. Micrite begins to burn out around 700°C, and as such, is not apparent in highly fired ceramics (Quinn 2013). During most of the time Ur was inhabited, the Persian Gulf was between ten and thirty kilometers from the site (Lambeck 1996). Secondary clay sources tend to have a higher plasticity than residual sources, leading to a greater need to be tempered (Quinn 2013: 120). When we look at what objects were tempered with which inclusions, no patterns seem to appear. Possible mineral tempered samples include a vessel, drainpipe, painted wall cone, and a brick. Possible organic temper includes a sickle, grinder, wall cone, painted bowl, architectural fragment, vessel, trough, brick, and amphora. No tempered samples include a jar stopper, wall cone, and vessel. No patterns appear when we look at tempered samples versus time period either. Mineral tempered samples date to any period, with one object dating to the Uruk period and one dating to the Neo-Baby­ lonian period. Organic temper also dates to any period, with one object dating to the Ubaid period, one to the Uruk period, one to the Neo-Babylonian period, and one to the Persian period. No temper also dates to any period, with one object dating to the Uruk period and one to the Neo-Babylonian period. In this sample of nineteen objects, the potters are choosing local sources to create all the ceramic objects, regardless of period, final form, or use of object. These results are interesting, if not unexpected. A larger sample, or a sample focusing on one period or type of object will produce different, or more varied results. With the amount of ceramic material Woolley collected from Ur, spanning different time periods, contexts, and functions, this project should be expanded to include more ceramic material from Ur in order to check these results against a larger sample. When you search “petrographic study” in the ur-online.org keyword box, the nineteen objects from Ur are returned. As seen in Fig. 6, the return page shows the objects, all with modern measurements and descriptions, and on the right-hand side are various filters. Expanding the object type filter shows the different object types that were included in the study. The final report to the section will be attached to all the records. Showing this data online, in this manner will hopefully give a starting point for future research on ceramics at Ur and throughout the Middle East.

Science and Technology: Using Ur-Online

301

Fig. 6. Display of petrographic study of ceramic materials on Ur-Online.

Infrared Reflectance Photography Infrared reflectance (IR) photography is a technique where an image is captured in the infrared with a radiation source in the infrared. While two different set ups were used as part of the Ur Digitization Project, both methods utilized incandescent lights as the source for infrared radiation. Initially a Sabrant IR webcam with an 830 nm long pass filter was used for imaging objects. The images were captured with the webcam tethered to a laptop and free image grabbing software (Astra Image Webcam Video Grabber). This is an inexpensive method (total cost was less than thirty dollars with the filter being the most expensive component) for IR photography and allows for a quick review of multiple objects as it has live view of the IR images, however the image resolution is quite low. As a result, when the Penn conservation department purchased a modified camera this was used for capturing images as well. The modified camera is a Nikon DSLR (digital single lens reflex) D5200 modified full spectrum camera. The hot mirror filter was removed and Fig. 7. IR reflectance photographs showing markreplaced with a glass custom full spec- ings on Penn objects 30-12-548 and B17218.

302

Kyra Kaercher and Tessa de Alarcon

trum filter manufactured by LifePixel allowing it to capture in the full spectrum, UV-IR. An 830nm long pass filter was used over the camera during image capture. The image resolution with this camera is much higher than with the webcam. IR reflectance imaging is useful for rending pencil and some ink marking more legible. The pencil and many inks absorb infrared radiation making them appear darker in the IR image than they often appear under visual examination. When you search “infrared” in the uronline.org keyword box, the objects that have been analyzed are returned. The images appear at­tached to their object, under the analytic tab. Infrared reflectance imaging is often useful for documenting pencil and ink mark­ ings, particularly when they are no long­er leg­ible. These mater­ ials absorb infrared radiation and appear much darker in IR reflectance photos than under normal illumination and documentation. This technique was used on objects from Ur with illegible markings in the hopes of reconnect­ ing objects to disassociated field numbers as well as Fig. 8. Ur-Online display of infrared photographs. other contextual information written on the object.

X-Radiography With recent renovations to the conservation labs at the Penn Museum it became possible to X-ray objects in-house. This has been particularly useful for determin­ ing manufacturing technique, understanding the structure of some of the objects, and using the objects for teaching. The X-rays were all captured with a GE Eresco 65MF4 Tube and a Digital Capture plate (GE – DXR 250V). Images were captured using GE Rhythm RT and post processed in Rhythm Review. Rhythm Review is equipped with various proprietary post processing filters that can highlight features and details in an X-ray. Fig. 9 shows X-rays of a cosmetic kit from Ur. This image was captured at 130kV, 4mA for six seconds and has been post processed with the GE Flash filter. You can see here some of the tools contained within, which otherwise cannot be visualized.

Science and Technology: Using Ur-Online

303

Most of the kits from the Penn Museum were imaged, so that these objects can be better under­ stood and compared to one another. Inside each kit are four tools that are joined together by a ring. Before the X-ray images, we could see there were four instruments, Fig. 9. Copper cosmetic kit (30-12-336) and X-ray but had to use broken objects, or photograph. single instruments to theorize what instruments were con­tained in the reticule. From the X-ray, we can clearly see the kits are made of four instruments, tweezers, a blade, an ear-scoop, and a stiletto. Other cases we have X-rayed have shown different combinations of the instruments. Fig. 10 shows an X-ray of a silver hair comb from Ur with flower petal inlays. It is still contained in Fig. 10. Silver hair comb (30-12-683) in wax and X-ray the wax that Woolley used to block photograph. lift the object during excavation. The image was captured at 140kV, 2mA for six seconds and post processed with the GE Flash filter. The X-ray allows us to see the structure of the object without having to remove the wax or excavate the block. Fig. 11 shows a ceramic rattle of a bird or pig captured at 90kV, 2mA, for four seconds and post processed with the GE Flash filter. A group of these was X-rayed to confirm that they are in fact rattles, as none of them make any noise when moved at this time. The X-rays clearly show that the object is in fact a rattle and the number of pellets inside. Based on the texture and radiopacity the pellets appear to be ceramic as well. Fig. 12 shows two of a group of bowls that were X-rayed, where the method of manufacture was known and hence they were X-rayed to be used as teaching tools in

Fig. 11. Bird (31-16-968) and pig rattles (B15786) with X-ray photographs.

Fig. 12. Two bowls (31-16-325 and 31-43-591) and X-ray photographs showing different methods of manufacture.

304

Kyra Kaercher and Tessa de Alarcon

Fig. 13. Ur-Online display of X-ray photographs.

ANTH Material World in Archaeological Sciences. This provides not only a teaching tool, but also provides a known reference for expected features when examining X-rays to determine method of manufacture. Both X-radiographs were captured at 90kV, 2mA, for four seconds and post processed with the GE flash filter. X-radiography is often used to show methods of manufacture (Lang and Middleton 2005). In the case of the Uruk beveled rim bowl the pressure marks and irregularities in thickness are consistent with the object having been made through a combination of pinching and pressing into a press mold as proposed by Millard 1988. The other vessel shown here has clear evidence of rilling marks and is likely wheel thrown. When you search for “X-Radiography” in the ur-online.org keyword box, the objects that have undergone this analysis are returned. The X-rays are included with the objects in the analytics tab.

Organic Pseudomorphs Organic pseudomorphs are organic remains in contact with metal that have been preserved through a combination of the biocidal nature of metal corrosion (particularly copper corrosion) and through mineral replacement of the organic structures or through positive casts (Scott 2002: 72–77; Solazzo et al. 2014). The pseudomorphs were documented photographically with a macro lens on a Nikon DSLR

Science and Technology: Using Ur-Online

305

and with digital photomicrographs. Two different set ups were used for capturing the digital photomicrographs over the course of the project. Initially a Leica binocular microscope with a built-in digital camera was used. However, after the Center for the Analysis of Archaeological Materials (CAAM) purchased a Keyance VHX5000 digital microscope, this was used instead. This microscope is capable of both stitching and stacking images thereby producing a better depth of field and fully focused high-resolution images. Fig. 14 shows an example of textile pseudomorphs documented with a macro lens. This is a dagger covered in a textile. The raking light detail shows the weave pattern overall, which can be identified as an extended tabby pattern (Burnham 1980: 139). The photomicrographs captured with the digital microscope show the additional detail that the threads are S-twisted. Fig. 15 shows a pseudomorph of a wooden handle also documented with the 3D microscope. This was particularly helpful for this type of pseudomorph as it has made it possible to view larger portions of the cross-section in a single image. Images such as this can be used to identify wood species. Dr. Naomi Miller looked at this image and based on the structure determined that it is likely a hardwood, but could not narrow down to a species level of identification (Miller 2016). When you search for “pseudomorphs” in the ur-online.org keyword box, the objects that have undergone this analysis are returned. The pseudomorph images are included with the objects in the analytics tab, and are also identified there. Showing this data online, in this manner will hopefully give a starting point for future re­ search on pseudomorphs at Ur. Solazzo et al. (2014) have done species-level identification of animal fibers using peptide mass fingerprinting (PMF) of pseudomorphs. Mineralization is rarely complete and pseudomorphs are typically combination of mineral and preserved original organic: with increased degradation/deterioration

Fig. 14. Textile pseudomorphs from a dagger with a sheath (30-12-292) from Ur.

Fig. 15. Wood pseudomorph from axehead at Ur (30-12-268).

306

Kyra Kaercher and Tessa de Alarcon

Fig. 16. Display of pseudomorphs on Ur-Online.

of proteins over time. The goal of this research was the identification of intact protein markers in mineralized archaeological samples using matrix-assisted laser desorption/ionization time-of-flight mass spectrometry (MALDI-TOFF-MS). This technique identifies specific peptides after enzymatic digestion of samples from the core of a fiber. The objects with pseudomorphs from Ur at the Penn Museum are all flagged and searchable: a ready-made dataset for this or a similar type of analysis.

Conclusion The Ur Digitization Project set out to create an online database that is accessible to anyone, anywhere. As the analysis and gathering of data in the Penn Museum and British Museum occurred, modern measurements and descriptions were col­ lect­ed. As a side avenue, specific research was carried out by students, conservators, and professors on various materials. This research including obsidian sourcing, pet­ rographic analysis, infrared reflectance photographs, X-rays, and pseudomorphs have greatly increased our knowledge of Ur and southern Iraq in ancient times. Woolley and specialists at the time analyzed what they could, and their results were published in various volumes and reports. Since then, technology has advanced, allowing us to uncover more specific data about the objects, as well as Woolley’s excavations. Ur-Online.org is used to display this data to everyone, and will hopefully be utilized as a baseline for data collected at Ur. The scientific analysis of these materials can be used to refine what we know about Woolley’s excavations and the ancient city of Ur.

Science and Technology: Using Ur-Online

307

Bibliography Armstrong, H. and Brasier, M. 2005 Microfossils. Second Edition. Oxford: Blackwell Publishing. Bressy, C., Poupeau, C., and Yener, K. A. 2005 Cultural Interactions during the Ubaid and Halaf Periods: Tell Kurdu (Amuq Valley, Turkey) Obsidian Sourcing. Journal of Archaeological Science 32: 1560–1565. Burnham, D.K. 1980 Warp and Weft. Toronto: Royal Ontario Museum. Cauvin. M.-C., Keller, J., and Pernicka, E. 1997 Obsidian from Anatolian Sources in the Neolithic of the Middle Euphrates Region (Syria). Paleorient 23/1: 113–122. Chataigner, C. and Gratuze, B. 2014 New Data on the Exploitation of Obsidian in the Southern Caucasus (Armenia, Georgia) and Eastern Turkey, Part 2: Obsidian Procurement from the Upper Palaeolithic to the Late Bronze Age. Archaeometry 56: 48–69. Deer, W., Howie, R., and Zussman, J. 1992 An Introduction to Rock Forming Minerals. Second Edition. Essex: Pearson Education Limited. Frahm, E. 2014 Characterizing Obsidian Sources with Portable XRF: Accuracy, Reproducibility, and Field Relationships in a Case Study from Armenia. Journal of Archaeological Science 49: 105–125. Gungozdu, F. V. 2010 Obsidian, Trade, and Society in the Central Anatolian Neolithic. Master’s Thesis, Bilkent University. Jassim, S. and Geoff, J. 2006 Geology of Iraq. London: Geological Society of London. Jassim, S. Z., Hagopian, D. H., and al-Hashimi, H. A. J. 1986 Geological Map of Iraq. [Map] 1:1,000,000. Directorate General of Geological Survey and Mineral Investigation Khazee, M., Glascock, M. D., Masjedi, P., Khedemi Nadoshan, F., Soleimani Farsani, R., Delfan, M., Mansori, A., Sodaie, B., and Dolatyari, A. 2014 Sourcing the Obsidian of Prehistoric tools Found in Western Iran to Southeastern Turkey: A Case Study for the Sites of Eastern Chia Sabz and Chogha Ahovan. AnSt 64: 23–31. Lambeck, K. 1996 Shoreline Reconstructions for the Persian Gulf since the Last Glacial Maximum. Earth and Planetary Science Letters 142: 43–57. Lang, J. and Middleton, A. 2005 Radiography of Cultural Material. Oxford: Elsevier. Millard, A. R. 1988 The Bevelled-Rim Bowls: Their Purpose and Significance. Iraq 50: 49–57. Miller, N. 2016 University of Pennsylvania Museum Ethnobotanical Laboratory Note. Unpublished report. Penn Museum, Philadelphia. Naqib, K. M. al. 1963 Geology of the Arabian Peninsula: Southwestern Iraq. United States Government Printing Office. Nesse, W. 2004 Introduction to Optical Mineraology. Third Edition. Oxford: Oxford University Press.

308

Kyra Kaercher and Tessa de Alarcon

Quinn, P. 2013 Ceramic Petrography: The Interpretation of Archaeological Pottery and Related Artefacts in Thin Section. Oxford: Archaeopress. Renfrew, C., Dixon, J. E., and Cann, J. R. 1966 Obsidian and Early Cultural Contact in the Near East. Proceedings of the Prehistoric Society (New Series) 32: 30–72. Rice, P. 1987 Pottery Analysis: A Source Book. University of Chicago Press. Chicago. Scholle, P. and Ulmer-Scholle, D. 2006 A Color Guide to the Petrography of Carbonate Rocks: Grains, Textures, Porosity, Diagenesis. Tulsa: American Association of Petroleum Geologists. Scholtz, H. C. and Vann, G. 1960 Geological Map of Iraq. [Map] 1:1,000,000. Ministry of Development-Mineral Survey Project. Scott, D. A. 2002 Copper and Bronze in Art: Corrosion, Colorants, Conservation. Los Angeles: Getty Publications. Solazzo, C., Rogers, P. W., Weber, L., Beaubien, H. F., Wilson, J., and Collins, M. 2014 Species Identification by Peptide Mass Fingerprinting (PMF) in Fibre Products Preserved by Association with Copper-alloy Artefacts. Journal of Archaeological Science 49: 524–535. Spaargaren, F. A. 1987 Geological Map of Iraq and Southwestern Iran. [Map] 1:1,000,000. Robertson Research International Limited. Tucker, M. 2010 Sedimentary Petrology. Third Edition. Oxford: Blackwell Publishers. Vernon, R. 2004 A Practical Guide to Rock Microstructure. Cambridge: Cambridge University Press. Whitbread, I. K. 1986 The Characterization of Argillaceous Inclusions in Ceramic Thin Sections. Archaeometry 28/1: 79–88. 1989 A Proposal for the Systematic Description of Thin Sections Towards the Study of Ancient Ceramic Technology. Pp. 127–138 in Archaeometry: Proceedings of the 25th International Symposium, ed. Y. Maniatis. New York: Elsevier. 1995 Greek Transport Amphorae: A Petrological and Archaeological Study. Athens: British School at Athens. Wright, G. R. H. 2009 Ancient Building Technology, Volume 3: Construction. Technology and Change in History 12. Leiden: Brill.

Towards Archival Reconstruction of Ur III Cuneiform Tablets Yudong Liu and James Hearne Western Washington University

Introduction It is a longstanding problem of Assyriology that the global diaspora of primary sources has been conducted in an asystematic, even lawless, manner, resulting in the loss of their original historical setting. Since the second Iraq War, the looting of archeological sites has accelerated, exacerbating a problem that began in the nineteenth century when the tradition of careless looting began. Primary sources— cuneiform clay tablets—are without proper archaeological context, which means that their original location is unknown, unknown not only the physical place, but also the site type, what the tablet was found with or in relation to, etc. Cannon-­ Brookes (1994) refers to looted objects as being “culture orphans, which, torn from their contexts, remain forever dumb and virtually useless for scholarly purposes.” Monastersky (1990) estimates that looted objects have lost 95% of their value to tell us what was going on in the past. There is some hope for the recovery of information lost through the diaspora of tablets throughout the world in museums, private collections, and the black market. The vast majority of these tablets are records of financial transactions and they rec­ ord not only the agent and patient of the dealing but also a list of witnesses, as well as the name of the recording scribe. Since the parties and witnesses to the transaction were in the same room, they may be presumed to have known one another, giving rise to a kind of social network in which tablets serve as nodes and common names serve as links between them. Moreover, most tablets are precisely dated and often contain toponyms, names of localities relevant to the transaction recorded. These facts about tablet content raise the possibility that the original archaeological context might be recoverable by appeal to the (1) social network implied, together with (2) their precise dating and (3) geographical references. With this information, the original archives might be reassembled by associating tablets with significant names, dates, and identified toponyms overlap. By using the tablets as the basic data in a social network that connects names when they co-appear on a tablet, a graph of the named Ur III merchants can be built and the resulting topology—the shapes and clustering of the graph—may be sufficient to identify tablets as having a common archival origin. Note that we use the term “archive” advisedly. We do not 309

310

Yudong Liu and James Hearne

presume that all tablets resided in a physical archive. Our ultimate goal in using methods from social networking is to identify subsets of tablets that derive from a tightly linked community. The promise of this method is enhanced by the high quality of curation bestowed on the Ur III corpus in the Cuneiform Digital Library Initiative (CDLI: http://cdli. ucla.edu/) and Database of Neo-Sumerian Texts (BDTNS: http://bdtns.filol.csic.es/). Parsing dates and identifying toponyms is computationally quite straightforward. However, a primary impediment to building a social network is that personal names have not been systematically identified at the time of transliteration and are not otherwise easily recognized. Given the sheer size of these corpora, the identification of names must be automated, which entails the application of machine learning techniques that are not often found in the digital humanities. In addition, to maximize the extractible information for social networking purposes, what follows is an account of our now successful efforts to identify personal names in Ur III tablets. It serves as a case study addressing the immediate problem but also exemplifies the relevance of techniques from the artificial intelligence tradition in computing to the digital humanities. The presentation is intended not for specialists in computer science but for potential beneficiaries of it in other disciplines.

Overview In this paper, we will describe our named-entity recognition (NER) systems that automatically identify personal names in Sumerian cuneiform documents from the Ur III period. The advantage of unsupervised NER is that it requires minimal knowledge of the language in which names are written. This is particularly appropriate to a language such as Sumerian which is evidently not well understood. Experiments show that this system can identify over 90% of the personal names in over 50,000 tablets in the CDLI database. The system also renders a list of spelling rules and contextual rules with which it recognizes these names. The spelling rules are learned by the system to denote the common syllables that occur in names, and the contextual rules are learned to denote the context in which a personal name tends to occur. The supervised NER system learns to recognize new names by using the names that are already annotated in the database. Over 80% of the names in these tablets can be recognized through this system.

History of Research Research on the use of techniques developed in natural language processing and machine learning traditions to ancient languages is not abundant. By and large, the application of computer technology has been limited to electronic publishing and string searching. Such efforts applied to Sumerian are even more sparse. Tablan et al. (2006) described the creation of a tool for Sumerian linguistic analysis and corpus search applied to Sumerian literature. This work focused on the application of a morphological model for noun and verb recognition developed by Sumerologists. However, it did not extend beyond simple word and morpheme recognition. Jaworski (2008) developed an approach to the extraction of information from the same economic documents of concern to us by beginning with an ontology of the world presupposed by this corpus and in conjunction with syntax and semantic

Towards Archival Reconstruction of Ur III Cuneiform Tablets

311

analysis. This work claims that it supports research both into the Sumerian language, officials participating in the activities the tablets record and in document classification. The Machine Translation and Automated Analysis of Cuneiform Languages (MTAAC: https://cdli-gh.github.io/mtaac/) project aims to develop and apply computational methods to translate and analyze the contents of some 67,000 highly standardized administrative documents from southern Mesopotamia (ancient Iraq) from the twenty-first century BCE, and to make the tools and data open-source. Pagé-Perron et al. (2017) report their progress in this endeavor.

Unsupervised Sumerian Named-Entity Recognition System An unsupervised Sumerian Named-Entity Recognition (NER) system assumes minimal knowledge about the Sumerian language in which names are written. The knowledge that was utilized by the system is minimal in that it only involves three seed rules as follows: 1.  If the left context of a word is “giri3,” then this word is a personal name. 2.  If the left context of a word is “kiszib3,” then this word is a personal name. 3.  If the left context of a word is “mu-DU,” then this word is a personal name.

All three rules are “contextual rules,” given that each of them specifies the context for a personal name with the window size of 1 or -1 (the right word or the left word). Accordingly, a “spelling rule” specifies something about the actual text of a personal name. It is a sign sequence that can be either the full string of a name or is contained as a substring of the name. For example, “if a word contains ‘e2-kikken,’ then this word is a personal name” is a spelling rule. By applying the rule, the word “e2-kikken-ta” is recognized as a personal name. The unsupervised method we adopted is called the “Decision-Tree based CoTrain­ing (DL-CoTrain)” method, which was invented by Collins and Singer (1999). The major task of the system is to learn a decision list (that is, a list of contextual rules and spelling rules) to classify a word as a personal name. Initialized with the three contextual seed rules as shown above, the decision list is applied to the target Sumerian texts and a set of names are found and gathered. By retrieving the sub­strings of the collected names, and ranking the substrings based on the frequencies, a set of spelling rules with high confidence scores are generated. In the next iteration, the newly obtained spelling rules are applied to label the target Sumerian texts to get new names and new contextual rules are generated from these new names accordingly. In this alternating process, each iteration produces a new set of rules that are ranked by their strength. In our system, the top twenty rules with the highest strength are added to the decision list. This algorithm, if iterated 150 cycles, produces a decision list of over 2,000 rules and approximately 17,000 personal names in these Sumerian texts. Our experimental result shows that the DL-CoTrain based NER system can recover 92.5% of the personal names in the target Sumerian texts. Even though 44% of the names found by our NER system are not found in the original annotated texts, some preliminary investigation into these names by our language experts has suggested that some of the false positives are due to the application of a more conservative approach by

312

Yudong Liu and James Hearne

the annotators when they dealt with damaged signs in the tablets. Luo et al. (2015) gives more details of this work.

Supervised Sumerian Named-Entity Recognition System Compared to an unsupervised learning system, a supervised learning system requires (1) a fully annotated data set, which means a sufficient number of Sumerian texts already curated with personal names is needed; and (2) a good set of useful features should be proposed to capture the characteristics of the personal names that have been labeled in the texts. Thanks to the annotators of the CDLI and BDTNS databases, we do have a well-curated dataset available. Conventionally, it requires linguistic expertise to propose a set of useful features in such a context, especially for a language that has been long dead. To our knowledge, there is no previous work as such. Therefore, we have to propose a set of features ourselves. Fortunately, our previous work on unsupervised learning provides us with a list of decision rules from which we can get some idea about what contexts/spellings would constitute a good feature. In our supervised NER system, we proposed thirty-six features, the majority of which are either context-related features such as the word or its part-of-speech tag to the left or the right of the current word, or spelling features such as if the current word contains a certain sign or a certain part-of-speech tag (such as “profession”). Other features include the positional information about the current word in the current line, the repetitive occurrences of certain signs in the current word, and if the current word is the only word of the line. The following lists some samples of our features: •  if •  if •  if •  if •  if

left context is dumu (“child of ”) left context is a profession word starts with ur word starts with lu2word is alone on the line

Several machine learning algorithms were applied to this task, including Decision Tree Classifier, Gradient Boosting Classifier, Logistic Regression, Naïve Bayes, Stochastic Gradient Descent, Linear Support Vector Machine, and Random Forest Classifier. It happened that all of the classifiers produced similar results on both data sets. Interestingly, compared to the unsupervised learning results, the supervised system exhibits opposite behaviors, meaning that it can predict a name with high accuracy but is not competent enough to identify as many names as the unsupervised system does. Our experimental results show that 86% of the names recognized by the system are true names but 35% of the true names are missing from the result. This suggests that a better approach to combining these two systems would improve the performance of both systems. Liu et al. (2016) gives more details of this work.

Dating Tablets in the Garshana Corpus To maximize the information to be subjected to social network analysis, we also used the identification of name overlap in tablets to provide dates to tablets

Towards Archival Reconstruction of Ur III Cuneiform Tablets

313

in which scribes either failed to provide a date or ones in which the date has been physically eroded. This work also applies to the problem of date identification that arises when year-name abbreviation happens to converge to the same shorter form, as is the case with Shulgi Year 42 and Amar-Suen Year 6, both of which abbreviate to the same form. The Garshana corpus is a fully curated and annotated corpus of tables from the Ur III period provided to us by Manuel Molina of CSIC. It is fully curated in the sense that all place names, personal names, and dates are identified and normalized, making machine processing possible. Since 92% of the 1,488 tablets are properly dated, it provides a reasonable machine learning environment for testing dating techniques. The corpus is temporally circumscribed, covering a span of forty years; the dates in the corpus are sparse and only twelve unique years are present. Hearne et al. (2017) reports on an explorative effort to assign dates to undated tablets in the Garshana corpus. Two approaches were pursued: (1) a naive one that determined the date of a tablet by simply counting the name overlap with tablets of known year; and (2), invoking a collection of machine learning algorithms of established robustness. The naive method reports an accuracy of 45–55% and the machine learning algorithms achieved 78.8–84.15% accuracy. The experimental results show that the integration of the geographic names in the machine learning methods has greatly improved the performance. For future work, we would like to integrate more features, such as relations between persons and profession names. A graphical model that captures the relations between personal names across tablets could also potentially improve the performance. We also intend to extend this method to uncurated corpora of Sumerian texts toward the ultimate goal of exploring ancient social networks. This work will also ultimately be brought to bear in the task of reassembling tablets that were in the same ancient archive but are now distributed in museums and universities throughout the world.

Conclusion and Future Work Preliminary to making use of techniques from the social networking world, it has been necessary to automate the process of recognizing toponyms, dates, and personal names, of which the latter is the most challenging. We have now developed a computer program that applies to the entirety of the Ur III corpus as represented by the CDLI and the BDTNS. Our future work will involve developing a vast social network where personal names are linked by their occurrence on the same tablet, mediated by temporal plausibility. When this network is constructed, algorithms developed recently for the identification of quasi-cliques—identifying regions of the network that are tightly interlinked—will be applied to identify communities of persons active in the same historical time-slice, a temporal period that can be variously defined for different purposes. We thereby hope to recover some of the archeological context that has heretofore been lost owing to the dispersal of tablets throughout the world.

314

Yudong Liu and James Hearne

Bibliography Cannon-Brookes, P. 1994 Antiquities in the Market-place: Placing a Price on Documentation. Antiquity 68/295: 349–350. Collins, M. and Singer, Y. 1999 Unsupervised Models for Named Entity Classification. Pp. 100–110 in Proceedings of the Joint SIGDAT Conference on Empirical Methods in Natural Language Processing and Very Large Corpora. Hearne, J., Anderson, C., Liu, Y., Dixon, D., and Ritchie, D. 2017 Dating Tablets in the Garshana Corpus. Pp. 628–630 in Proceedings of the 30th International Florida Artificial Intelligence Research Society Conference. Jaworski, W. 2008 Contents Modeling of Neo-Sumerian Ur III Economic Text Corpus. Pp. 369–376 in Proceedings of the 22nd International Conference on Computational Linguistics. Liu, Y., Hearne, J., and Conrad, B. 2016 Recognizing Proper Name from UR III Texts through Supervised Learning. Pp. 534–540 in Proceedings of the 29th International Florida Artificial Intelligence Research Society Conference. Luo, L., Liu, Y., Hearne, J., and Burkhart, C. 2015 Unsupervised Sumerian Personal Name Recognition. Pp. 193–198 in Proceedings of the 28th International Florida Artificial Intelligence Research Society Conference. Monastersky, R. 1990 Fingerprints in the Sand: Federal Agents use Dirty Evidence against Archaeological Thieves. Science News 138/25: 392–394. Pagé-Perron, É., Sukhareva, M., Khait, I., and Chiarcos, C. 2017 Machine Translation and Automated Analysis of the Sumerian Language. Pp. 10–16 in Proceedings of the Joint SIGHUM Workshop on Computational Linguistics for Cultural Heritage, Social Sciences, Humanities and Literature, Association for Computational Linguistics Anthology. Tablan, V., Peters, W., Maynard, D., and Cunningham, H. 2006 Creating Tools for Morphological Analysis of Sumerian. Pp. 1762–1765 in Proceedings of the Fifth International Conference on Language Resources and Evaluation.

Excavating Ur in Children’s Literature Kevin M. McGeough and Elizabeth A. Galway University of Lethbridge

Introduction The sensation that surrounded Woolley’s excavations at Ur in the 1920s fundamentally transformed the popular reception of ancient Mesopotamia. The discoveries were in and of themselves remarkable, but Woolley’s particular skills at public outreach, his approach to storytelling as a means of excavation reporting, and his straightforward discussions of archaeological methods meant that much of the results of his work were readily adapted into other popular culture forms. This was certainly the case in regards to the presentation of Mesopotamia in children’s literature; the excavations at Ur transformed how child audiences were introduced to Mesopotamian culture. Prior to the Ur excavations, Biblical and Classical connections to Mesopotamia were paramount in children’s literature. 1 After Woolley’s excavations, outreach to children came to be oriented around iconic objects, narratives of human sacrifice, and class-based discussions of everyday life. Whiggish historical narratives about the triumph of liberalism in relation to despotism were easily supported by Woolley’s presentation of the cemetery at Ur. The auratic treasures of Ur, while evocative, were essentially empty signifiers allowing authors to create mean­ ing that was deemed appropriate for child readers, even if that meaning had more to do with twentieth-century British and American concerns than Mesopotamian culture. 2 By the time of Woolley’s excavations, Sumer had come to be understood as the “first civilization” in world history. The seeming chronological primacy of Sumer Authors’ note: This article is part of the authors’ larger project on the reception of archaeology and Near Eastern studies in children’s literature. This project is funded through an Insight Grant from the Social Sciences and Humanities Research Council of Canada. This article is dedicated to Erle Leichty, whose concern and care for graduate students at the University of Pennsylvania made such an important contribution to the field. 1. Even as late as 1921, in Hendrik van Loon’s account of Mesopotamia in The Story of Mankind, an influential children’s history book, Ur is discussed in relation to the story of the Hebrews, not in relation to Sumer (1921: 38). 2.  By auratic, we mean in the sense suggested by Walter Benjamin (1968: 224). However, Meskell (2004: 183) offers an important critique of “auratic” in reference to ancient objects; she argues that the domestication of ancient objects does not diminish their charisma for non-specialists as argued by Benjamin. Rather, the material engagement with ancient objects seems to elicit very distinct reactions of excitement, interest, and mystery.

315

316

Kevin M. McGeough and Elizabeth A. Galway

in treatments of “western civilization” meant that arguments regarding the origins of twentieth-century institutions and social practices made by authors of children’s literature were particularly powerful in naturalizing contemporary concerns and values.

Methods of Studying Children’s Literature Woolley’s knack for public outreach and his phenomenal discoveries at Ur are obviously well known and need not be reiterated here. However, it may be useful to make some brief comments about the approach taken here to studying children’s literature. As many scholars of children’s literature contend, as much (or more) can be learned about the preoccupations of adult society as can be learned about the lives of children from studying children’s literature. Children’s literature is profoundly influenced by adult concerns and motivations, and as a result, children’s authors can play a key role (intentional or otherwise) in shaping the views and attitudes of child readers. This is what Perry Nodelman (2008) terms “the hidden adult” in children’s literature, or what Jacqueline Rose (1984: 1–2) sees as “the impossible relation between adult and child” in children’s fiction, which she says “sets up a world in which the adult comes first (author, maker, giver) and the child comes after (reader, product, receiver), but where neither of them enter the space in between.” As Nodelman (2008: 4) points out, it is adults who write and publish books for children and the “actual purchasers of children’s books are and always have been, overwhelmingly, not children but parents, teachers, librarians: adults.” Children’s literature cannot be separated from adults, and thus is a revealing source of information on how adults view the world, how they view children, and how they may hope to influence their readers. Not all messages are directly intended by children’s authors. The nature of the genre is such that authors often tailor their works to mirror other similar books and to cover specific curricular elements that are established by other bodies. Nodelman (2008: 5) explains that the producers of children’s literature “make judgements about what to produce based not on what they believe will appeal to children but rather on what they believe adult consumers believe they know will appeal to children (or perhaps, what should appeal to them, or what they need to be taught).” The presentation of the past in these works by adults for children (but intended to be purchased by other adults for children) reflects values that these adults feel should be inculcated in the child audience. As Maurice (2015: 3) has argued, “ancient cultures are presented in ways that (consciously or sub-consciously) present ideological viewpoints that the young reader is expected to absorb.” This paper shall explore some of the ways that Woolley’s work manifests in different genres of children’s literature, beginning with retellings of Biblical and other ancient stories, then examining various examples of narrative fiction, and concluding with a comparison of different kinds of non-fiction works.

Bible Stories and Adaptations of Ancient Literature for Children In the didactic literature that dominated early nineteenth-century publishing for children, popular subjects included colloquial retellings of Biblical stories (Dalton 2016:11–14; Gold 2004: 1–5). This type of literature provides the clearest evidence

Excavating Ur in Children’s Literature

317

for how Woolley’s excavations transformed the reception of Mesopotamia for child audiences since there are many examples that predate his excavations. Before Woolley’s discoveries, little is usually made of Abraham’s origins in Ur. References to Mesopotamia are usually references to Nineveh, and after Layard’s excavations become popular knowledge, those discoveries are especially integrated into accounts of Jonah. Woolley’s discoveries at Ur brought a new level of popular recognition of the site of Ur that goes beyond reference to Abraham. A brief survey of some of the most popular children’s Bibles of the past twenty years demonstrates this kind of popular recognition. That is not to say that all children’s Bibles embrace Woolley’s results; for different reasons, many do not attempt any kind of historical contextualization or hold to an archaeological or veristic aesthetic. For the children’s Bibles that do attempt to situate the Old Testament within an ancient Near Eastern con­ text, however, Woolley’s discoveries are typically evoked. Vander Stichele and Pyper (2012: 1) observe that children’s Bibles are “often the first contact people have with the Bible, and as such they can shape their perception of its stories and characters at an early age.” 3 Just as importantly, they can shape a reader’s perception of place, and Woolley’s findings contribute to the image of the Near East that is presented in these Bibles. Indeed, as Gold (2004: 7) has argued: “Sacred texts reshaped for children provide an especially clear window on contemporary values, as they tend to stress prominent, widely agreed upon values.” 4 The Tower of Babel, for example, is often depicted as a variation of the reconstructed Ur ziggurat (Pulley 2005: 38). The modeling of the Tower of Babel after the ziggurat of Ur is even explicitly mentioned in The Children’s Illustrated Bible (Hastings 1994: 28–29), which also provides an ethnographic image of the manufacture of mudbricks, connecting Babel with contemporary Iraq. That same Bible also prints an image of the golden helmet and dagger found at Ur, linking them to Abraham and Lot implicitly (Hastings 1994: 32). Nothing meaningful is actually said about the relationship between Genesis 11 and these artifacts for there is no actual link. Rather, these images hint at a wealthy, militaristic, urban culture from which Abraham departs and so Woolley’s discoveries are used as a contrastive foil to the simple lives of the Hebrew patriarchs. As Vander Stichele and Pyper note (2012: 1) “images not only illustrate the events narrated in the biblical text, sometimes they even have a story of their own to tell.” The story that these images in The Children’s Illustrated Bible tell is that Biblical heroes are rural and simple and their enemies are urban and cosmopolitan, perhaps emphasizing such sentiments in contemporary North American theology, in which rural life is deemed morally or religiously superior to urban life. The artifacts discovered by Woolley provide evocative, material evidence of a site mentioned in scripture and emphasize the exotic wealth of the society, in keeping with the Biblical depiction. In The DK Illustrated Family Bible, Genesis 11–12 are similarly illustrated through Woolley’s finds (Costecalde 1997: 36–38). Again, these finds are used to make claims of historicity for the Biblical story and to emphasize the otherness of Abraham’s origins. They emphasize the chronological 3. Dalton (2016: 2–3) offers further discussion on the influential role of children’s materials on Biblical reception. Maurice (2015: 3) makes similar claims for children’s literature related to ancient Greece and Rome. 4. Weinlich (2015: 85) has made comparable arguments about how the repurposing of classical myth for children can also be highly ideologically charged.

318

Kevin M. McGeough and Elizabeth A. Galway

primacy of Mesopotamia while also emphasizing how Biblical values are the next step in an almost evolutionary sequence. Interestingly, despite the fact that critical Biblical scholarship does not typically date the stories of Abraham to the Early Dynastic Period, narratively later parts of the Bible are illustrated with artifacts from later periods. This creates a kind of pseudo-parallel chronological framework, that while nonsensical archaeologically, is readily adapted to certain denominational tendencies, an issue that we will touch on further later in this paper. Yet very often, Abraham and his family are explicitly not depicted as Mesopotamian; rather, they are depicted in nineteenth-century Bedouin garb, the type of clothing that has been established as “Hebrew” in popular receptions of the patriarchs since the mid-nineteenth century. 5 Here, the children’s Bibles mirror Biblical reception tradition more generally in which the Biblical patriarchs are imagined as analogous to nineteenth-century Bedouin (McGeough 2015: 88–103). Despite the Biblical description of Abraham’s birthplace as Mesopotamia, the patriarchs are marked as “other” and represented through different signifiers than Mesopotamian figures. Similar issues of identification and difference are apparent in Bible retellings aimed primarily at Jewish children, which Gold (2004: 175) argues attempt to inspire: “pride in the distance Jews had traveled from nomadic days to the present, and in their special contribution to the development of civilization.” Thus the patriarchs needed to be different from the urbanized Mesopotamians yet be depicted as having participated in that culture. Even strictly historical or archaeological non-fiction books still make the occasional reference to Abraham, although Biblical references make for contested educational materials. Christopher Fagg’s Exploring Lost Cities, first published in 1980, but reprinted frequently in many different collections, has a section on Woolley’s excavations called “Abraham’s Birthplace.” Fagg (1980: 4) comments on the importance of the Bible as a motivating influence on Woolley: “But if Ur had not been mentioned in the Bible, this ancient city might have been lost forever.” This claim reflects the kind of “hook” used to connect Woolley’s work with the Hebrew Bible. A genre related to these children’s Bibles are retellings of other ancient texts aimed at children. As with Bible stories, the presentation of Mesopotamian literature to child audiences usually entails some expurgation. For example, in Geraldine McCaughrean’s Gilgamesh the Hero (what she calls a free translation based on a number of texts), Ishtar’s proposition to Gilgamesh is more of a proposal, as she says (McCaughrean 2002: 32): “I love you king of Uruk. Marry me.” In terms of Woolley’s contributions to children’s reception, an interesting example is Karen Foster’s retelling of Adele Berlin’s translation of Enmerkar and Ensuḫkešdanna. As is apparent from the cover, which references the seated banqueting figures and their servants from the Standard of Ur, the aesthetics of the story are clearly rooted in the art discovered by Woolley. The entire book is illustrated in fashion evocative of the inlaid shell, lapis, and red limestone of the figural art found in the Royal Cemetery; Foster made a conscious effort to replicate this style but created her own illustrations. Although these are her own particular scenes, the composition is very successful and convincingly evokes this Sumerian style. This points to an issue that we seek to further explore in this project; are these children’s book written by 5.  See for example: Emmerson 2014: 30; Hastings 1994: 32–33; and Taylor 2000: 24.

Excavating Ur in Children’s Literature

319

scholars in the field more or less interesting to child audiences than those created by professional children’s authors? That is to say, do child audiences appreciate such attempts at verism or are these attempts more rhetorical flourishes aimed at convincing adult consumers of the historicity of the work?

Historical Fiction An ethos of historicity and accuracy is common in the presentation of historical fiction. Even if it is understood that much of the characters’ motivations in such works will be anachronistic, to make the literature compelling, the settings have to at least evoke historical locations. This is what Thaler (2003: 5) notes as the need “to give historical flavor to fiction that otherwise might lack it” or the need for readers to “recognize an historical reality.” Marjorie Cowley’s The Golden Bull: A Mesopotamian Adventure is a good example of children’s historical fiction that has been inspired by Woolley’s finds and, in some ways, by the kinds of thinking about artifacts that the University Museum at the University of Pennsylvania has pioneered. Cowley, herself trained at the Fowler Museum at UCLA, is a writer of children’s historical fiction of a variety of eras and she consulted numerous experts, especially conservators, for this book. In her author’s note (Cowley 2008: 203), Cowley mentions explicitly Woolley’s discoveries and she lists the objects he discovered that played a part in her story: wreaths, jewelry, a golden drinking straw, cylinder seals, the golden bull lyre (Cowley 2008: 203). Unlike many children’s authors (such as the authors of non-fiction discussed below), Cowley does not take Woolley’s attribution of “royal” to the grave goods unquestioningly but she explains that the fine artifacts were crafted by skilled craftsmen, like the characters featured in her story. Here is an American take on the finds from Ur, which de-emphasizes royal connections and highlights individual skill and entrepreneurship as driving engines of social complexity. The creation of the fine objects like those found in the Royal Cemetery is central to the narrative of the story. Cowley adopts a common practice in children’s literature whereby child characters are separated from their parents, thus en­abling experiences and adventures that would otherwise be unlikely. In this book, the separation is facilitated through reference to the Mesopotamian practices of debt slavery. The protagonist is a young boy, Jomar, who becomes apprenticed to a jew­ eler at Ur, a jeweler who is regularly commissioned to craft fine products for the temple of Nanna. Jomar’s sister, Zefa, is a naturally skilled musician with expertise in playing the lyre. These young characters are emblematic of what Thaler (2003: 3) describes as one of the foundations of historical fiction for children: The historical novel is based on an illusion, namely, the belief that history can be rewritten with the help of characters who have never existed and with whom the reader will be able to identify. This is even more true of historical fiction for young people—a revived literary genre which proposes to its readers heroes of their own age—since children are largely absent from the scene of history.

The story involves these rural children settling into their new lives in Ur and finding their own place within an economically complex civilization and interacting with institutions and objects known from the archaeological record. Here the Mesopo-

320

Kevin M. McGeough and Elizabeth A. Galway

tamian story mirrors the realities of Cowley’s contemporary readers, who will be expected to discover their own particular economic niche over the course of their education and fulfill a specific economic role in an increasingly urbanized American society. Having a chosen profession is presented as an integral component of the ancient children’s identities, just as it will presumably be for the readers of the book. The economic complexity of civilizations is clearly one of the main historical lessons in The Golden Bull. The story spends a good deal of time exploring issues of economic inequality, especially those that emerge within a standard redistribution model of city-state economics. It is rooted in older models of the Sumerian Temple economy, in which the temple is the driving force behind the development of eco­ nomic complexity in the city-state. 6 Cowley explicitly describes the Sumerian temple economy model that she bases the story on in her author’s note (Cowley 2008: 204). It is also readily apparent in the body of the narrative itself, and a slave named Qat-nu explains the economic organization of Ur to the children at one point (Cowley 2008: 29–30): “The temple granaries have great stores of food gathered from all the farmlands…The farmers do without so that the city of Ur may flourish. The temple takes a portion of food and animals from the farms to feed its officials, workers, and craftsmen…” Economic history is one of the major themes in explorations of ancient otherness for children; the presentation of a child reader’s own society as normative and desirable is made convincing by illustrating “unfairness” in other places and times. Interestingly, Cowley does not actually reference the cemetery of Ur or the human sacrifices identified by Woolley, which are typically invoked in children’s literature as means of exploring early social inequality. Rather, she focuses on who owns what products, the nature of specialized labor within the city, and how private institutions motivate economic growth. All of these messages about Sumerian economics mirror similar messages about American neoliberalism that are frequently imposed through children’s literature. These messages may not have been intended by the author but it is widely acknowledged that this genre is increasingly impacted by neoliberalism (Hade and Edmonson 2003). Woolley’s objects are central to the narrative. The first object that Jomar works on as an apprentice is a golden drinking straw. When he first delivers it to the Temple, he notices that the high priestess is wearing a crown like that Woolley associated with Pu-abi. The priestess’s adornments are described as follows (Cowley 2008: 91): “Large gold earrings shaped like crescent moons hung to her shoulders. Gold leaves wreathed her head, similar to the ones in the necklace he was string­ ing.” Yet what truly drives the narrative is the bull lyre. The golden elements of its construction are narrated in detail; as Jomar learns his craft, so does the child reader. For the master jeweler, the crafting of this object is not only a responsibility given to him by the temple as a main source of his economic prosperity, it also becomes a means for him and his wife to gain closure regarding the death of their child, when Zefa composes a funerary song for the dead child and performs it on the lyre. For Zefa, the lyre leads to her admission as a fulltime musician into the Temple of Nanna. For the elites of the Temple of Nanna, the desire for such a fine object leads them to offer substantial social and professional opportunities for the people involved in its crafting. Here is the kind of trickle down economic reasoning that has

6.  For critique of this model, see Foster 1981.

Excavating Ur in Children’s Literature

321

been familiar since Reagan’s era; the desire for luxury items by elites encourages economic growth for all sectors of society. There is also an interesting intersection with some of the other work done at the University Museum, well after Woolley’s days, for this book presents an object biography of the golden lyre, in an imagined sense, but rooted in scholarship surrounding it. 7 Similarly, the lyre seems to have its own agency, driving the actions of the narrative forward. At one point in the book, the lyre’s agency is made explicit when the master goldsmith explains why he uncovers the instrument at the beginning of the day and covers it again at night, even when not working on the musical part of it directly (Cowley 2008: 146): “I suppose it’s my way of saying that the workday begins when the lyre stands watch over our labors. When it withdraws under its cotton blanket, we also rest.” In The Golden Bull, Woolley’s discoveries are so charismatic that they have their own very powerful agency, an agency that is central to the narrative. It makes ancient Mesopotamia seem almost magical, which is very appealing to child readers, without actually having to deviate from the realistic fiction genre. This sense of magic is in keeping with Lucas’s (2003: xvi) claim that: “[t]he history that comes down to us from distant times is often perceived in a manner yet more simplified than is recent history, simplified but also perhaps magnified in grandeur.”

Non-Fiction Books The multiplicity of meanings and auratic powers of these objects are just as apparent in non-fiction treatments. 8 One might think that non-fiction accounts of Mesopotamia are the least interesting types of literature for scholars to deconstruct, as one might assume that these books simply provide factual but more simplified versions of scholarship, that while perhaps out of date, still reflect scientific thinking on the subject from some phase of Assyriology. To some extent this is true, but what has been an interesting discovery in our project so far has been how important Woolley’s own representations of discoveries at Ur have been in the reception of Sumer in non-fiction children’s materials. His evocative descriptions have been accepted mostly uncritically by children’s authors and despite the many years of research on Ur since, his readings are still dominant. This raises obvious concerns about how children’s literature, both fiction and non-fiction, can perpetuate outmoded and outdated ways of thinking about the past. As Stevenson (2003: 27) notes, historical fiction for children “seems either to be conveying orthodox history or to consider 7.  Perhaps the best-known result of this kind of consideration of artifacts at the University Museum is Kopytoff ’s important 1986 paper “The Cultural Biography of Things: Commoditization as Process.” 8.  To illustrate this, we have chosen to discuss Ur references in a number of types of non-fiction works. A so-called Atlas of Ancient Worlds (Crisp 2009) and the already mentioned Exploring Lost Cities (Fagg 1980), both present brief descriptions of Mesopotamia in the context of other ancient cultures. Elizabeth Scholl’s How Did They Do That? In Ancient Mesopotamia (2010) offers descriptions of daily life and institutions. Jane Shuter’s Excavating the Past: Mesopotamia (2007) is designed to explain how archaeological methods have been used to understand specific elements of Mesopotamian life. Eyewitness Mesopotamia (Steele 2007) is an illustrated reference book to Mesopotamia, the children’s equivalent of the “Oxford Companion” books that have proliferated in recent years. Allison Lassieur’s The Ancient World: Ancient Mesopotamia (2013) offers a topical account of Mesopotamia, almost in ethnographic form. It should be noted that McGeough (one of the authors of this article) worked as a consultant on Lassieur’s book.

322

Kevin M. McGeough and Elizabeth A. Galway

itself to be repairing history—in essence, creating a new orthodoxy while clinging firmly to the Rankean doctrine of history,” and the same can be said of non-fiction. Lucas (2003: xviii) asks if it is “the case that, in changing societies, nonfiction has an even greater capacity than fiction for becoming dated?” Woolley’s biographical presence is one of the key entry points for children’s authors. In most of the books we have surveyed, he is one of the main scholars whose personal history is emphasized, even overshadowing Austen Henry Layard. Part of what makes his biography so appealing to children’s authors is that he wrote so prolifically on archaeological methods and so he can be readily framed as an heroic scientist, a common figure in popular receptions of archaeology (McGeough 2006: 176–178). Many of the books focus more on Woolley and his excavations of Ur, rather than ancient Sumer, a conceit that is typical of popular presentations. In these works, the description of Mesopotamia cannot be separated from Woolley’s academic biography. Eyewitness Mesopotamia puts Woolley in two separate reference sections at the end of the book. He is listed as one of the important “Archaeologists and Scholars,” and along with Donny George and Layard is one of the three scholars whose image is provided (Steele 2007: 67). He is also referenced in the “A to Z of ancient sites,” although there he is named Frank Woolley (Steele 2007: 69). In Lassieur’s Ancient Mesopotamia (2013: 99), he is the only non-ancient personage mentioned in her “Biographies” section, listed along with Hammurabi, Sargon the Great, Gilgamesh, and Herodotus. In Shuter’s (2006: 45) archaeologically oriented Excavating the Past: Mesopotamia, Woolley is described as “the first archaeologist in Sumer to use care­ ful excavation in layers and detailed record-keeping while excavating.” Woolley’s scientific archaeological approach is also emphasized in the body of her text, where he is quoted as follows (Shuter 2006: 26): “Our aim was to understand the history, not to fill the museums with curious objects. The more rich the cemetery looked like being, the more important it was to leave it alone until we had provided a time frame for it from the rest of the site.” Shuter’s book is organized as a kind of history of archaeological methods, and as the book progresses, she showcases increasingly sophisticated archaeological methods in an evolutionary structure. Fagg’s (1980: 4) entry on Mesopotamia in his Exploring Lost Cities is framed as a narrative description of Woolley’s discoveries, rather than as a description of the site of Ur or Sumerian culture, a typical conceit of many popular archaeological reports, especially in National Geographic. Beyond Woolley’s own personality, what is striking in these books is how the auratic objects that Woolley discovered are used as empty or hollow signifiers. As Quinn (1994) argues, symbols and images from the ancient world point to an ancient reality but are empty enough of meaning that the appropriators of the images can use them to mean whatever they wish. Pictures of ancient artifacts are prominent in these children’s books, and are more immediately noticed, especially by child read­ ers, than the body of the text. The captions indicate how we are to understand the recontextualized artifacts and as is apparent, this recontextualization orients the reading of the artifacts in very specific ways, often in accordance with the authors’ specific interests. 9 As is clear, the materiality of these objects is invoked to give an 9.  Authors of these works do not always have input about captions or illustrations. They are hired to write the prose and the publisher uses an in-house illustration team who may or may not be attentive

Excavating Ur in Children’s Literature

323

empirical basis for a wide variety of claims about the past and differing ideological interests. Rarely do the authors acknowledge that there may be contested meanings to these artifacts. This raises questions about the author’s/narrator’s role as an authority figure in children’s literature who shapes the factual messages that are taken in by readers in a way that is not readily questioned by child readers. Rose (1984: 2) describes children’s fiction “as something of a soliciting, a chase, or even a seduction,” and one could argue that writers of non-fiction have an even greater ability to (Rose: 2) “take the child in” and lead them to accept what they are reading at face value. While many children’s scholars and Reader-Response critics have argued since Rose that young readers have agency and do not always passively accept what the author/narrator tells them, this remains an important issue for children’s literature scholars to consider, particularly in relation to the kind of historical, non-fiction work that we are examining here, where that which is presented as “fact” may be less likely to be queried by readers if it appears to have the weight of scholarship behind it. The most apparent example of this is the Standard of Ur. Scholars themselves debate the original function of this artifact and its art-historical readings. This allows children’s authors much intellectual room to invoke whatever is of interest to them. Eyewitness Mesopotamia (Steele 2007) prints images of the standard three times, including on its large, fold-out wallchart. The wallchart offers a typical narrative of growing economic complexity, in which rural people bring food to the city. Specific images on the standard are interpreted for young readers by lines pointing to captions, explaining socio-economic roles of different figures. Reading the standard as providing evidence for both Sumerian economics and warfare is not out of step with scholarly readings of the object and Shuter provides a similarly bifurcated reading of the artifact. She (Shuter 2006: 20) uses one side to explain how the rural areas around each city supplied the city, stating that it offers a scene of: “peacetime life, with farmers herding cattle, sheep, and goats, and fisherman with their catch.” The same side is reprinted later in the book (Shuter 2006: 39) as a means of showcasing ancient transportation equipment, the reed-framed backpacks. Shuter, as with the Eyewitness Mesopotamia example, also emphasizes the warfare element in a different section, describing the equipment of war but also the brutal treatment of prisoners (Shuter 2006: 16). An economic argument is also made in the Atlas of Ancient Worlds (Crisp 2009: 8), but there the “mosaic on a box” as it is described, is used to emphasize the competitive nature of Sumerian city-states, with that competition explicitly being over “land and resources.” The same economic argument is made in the interior of Eyewitness Mesopotamia, under the heading “Supplying the City,” where the author explains that the standard was more likely a musical instrument than a standard for use in battle (Steele 2007: 9). Similar interpretive presentation strategies are used in the same book to read the other side of the standard, describing the various war equipment depicted there (Steele 2007: 42). Lassieur (2013: 79) uses a detail from it to inform young readers that “Sumerian men did not always wear shirts,” a point that is likely of interest to her readership but is not one of the major scholarly concerns with the artifact. While these readings, at a surface level, do not seem very divergent, each reflects a slightly different economic/ideological argument made within the larger context of the work. to the authors’ wishes.

324

Kevin M. McGeough and Elizabeth A. Galway

These minor variations in artifact interpretations add up to more significant read­ ings when taken within the larger contexts of the works. Take for example the Atlas of Ancient Worlds’ treatment of the “ram caught in a thicket” in the context of the already discussed “mosaic on a box” and its larger commentary on the “First Cities of Sumer.” Young readers learn that the goat was (Crisp 2009: 9) “made of gold from Egypt and a blue stone called lapis lazuli from Afghanistan.” Indeed, the whole section on Sumer is a discussion of heroic Sumerian merchants, traveling to distant lands to acquire resources and therefore driving the origins of civilizations. This is not atypical of some scholarly reconstructions of the era but it is a very specific academic argument made into a generalized narrative about the past. In its privileged place in the Atlas, Sumerian merchants seem to be credited with creating the foundation for the rest of world history, an obviously pro-capitalist narrative. Shuter takes the ram in a thicket as evidence of remarkable Sumerian artisanship, again not an unusual reading of the artifact. Yet again she emphasizes the trade connection (Shuter 2006: 36): “This statue of a ram caught in a bush was made with gold leaf, lapis lazuli, ivory, and precious stones. These all had to be traded in from other places.” This is part of her section on jewelry but again the ram is invoked to reference the wider trade connections of ancient Sumer, as opposed to saying anything meaningful about jewelry per se. Sections on jewelry in these books offer information for child readers on daily life in Mesopotamia and perhaps that has been one of the major contributions that Woolley’s discoveries have made to the reception of Mesopotamia in popular culture generally, and in children’s literature specifically. Most of the non-fiction books surveyed here feature sections on jewelry and images of the finds made by Woolley, either in situ or in reconstruction. 10 Information about music is similarly present­ed through the lyre. 11 One of the items that Woolley discovered that has had the great­est afterlife in a variety of media aimed at children is the Game of Ur. In the non-fiction books surveyed here, some provide very specific rules for the games. 12 Other authors acknowledge that we do not really know how to play but offer some suggestions. 13 None reference Finkel’s 2007 article on the topic, which may reflect the lag between academic publication and popular reception. Related to these items is what is perhaps extremely interesting for child readers, the quasi-biography that emerges of Queen Pu-abi. Her clothing, her headdress, and even her cylinder seal are described for children. Presenting an individual, named woman allows history to be personified by an individual but, also allows discussion of what can be explicitly described as women’s history within what are often, in Mesopotamia’s case, fairly phallocentric narratives (Enheduanna provides another example of this). Here is an 10.  See for example: Scholl 2010: 40–42; Shuter 2006: 30; Steele 2010: 10–11. 11.  See for example: Scholl 2010: 43, and Steele 2010: 20, 27. 12.  Scholl (2010: 54) states: “The rules of the game as it was played in the Mesopotamian times are unknown, but historians believe it involved trying to get one’s pieces to the end of the board. It may be related to the game of backgammon.” 13.  Shuter (2006: 27) states: “The game was a race game for two players. The different patterns had different meanings and could be lucky or unlucky.” Steele (2007: 26) offers the most certainty when he writes: “The rules were similar to backgammon’s. A throw of the dice decided who went first. That player put a counter on his or her start square then threw the dice to see how many squares to move. The goal was to race all seven counters along a set path around the board, then bear them off at the finish. Landing on an opponent’s counter knocked it off the board so that it had to start again, but this could only happen when counters were moving along the central, shared row of squares.”

Excavating Ur in Children’s Literature

325

entry point for authors to challenge patriarchal presuppositions about history with an evidentiary basis.

Home Schooling and Christian Education The most extreme examples of how the results of Woolley’s excavations could function as empty signifiers can be found in textbooks written for the home schooling movement. Predominantly conservative Christian in orientation, these books use the materiality of archaeology as scientific evidence of scripture. What is perhaps most interesting in the case of Woolley’s work is how few of these books, which are explicitly creationist, actually reference Woolley’s infamous flood levels. Rather, Woolley’s finds are often enlisted as evidence for spurious claims related to chronology, especially in support of creationist ideas. Take John Ashton and David Down’s Unwrapping the Pharaohs as an example. 14 The subtitle, “How Egyptian Archaeology Confirms the Biblical Timeline” suggests the dominant theme of the book, which is filled with numerous illustrations and detailed finds from mainstream Egyptology put together into a nonsensical hodgepodge of chronological confusion. Woolley’s excavations are invoked here, as they often are in works from this community. We learn, for example, that the ziggurat of Ur is possible evidence that Djoser’s step pyramid is based on Mesopotamian design (Ashton and Down 2006: 16–18). There is an implication that the ziggurat, as reconstructed, was there at the time of Abraham, and that he may have brought this knowledge with him to Egypt (not an uncommon assumption in medieval scholarship or amongst esotericists), although the authors do explicitly state that they do not claim that Imhotep visited Mesopotamia. Woolley’s account from his Excavations at Ur of the funerary and sacrificial rites is quoted at great length, as further evidence of Egyptian-Mesopotamian connections and as part of a larger reframed chronology (Ashton and Down 2006: 17–18).

The Great Death Pit Of course, it is Woolley’s accounts of the human sacrifice in the royal cemetery and the Great Death Pit that are so evocative for all audiences, especially child read­ ers. Treatments like Christopher Fagg’s (1980: 4–5) provide an eerie and haunt­ing vision of the serving classes going willingly to their deaths for the sake of their royalty. Woolley’s romantic-gothic account of this is often merely rewritten, or, as with the home schooling text mentioned above, quoted at length. In these children’s books, the Great Death Pit works as part of a larger whiggish-style world history narrative. As with many approaches to dealing with the western tradition, the thrust of human history is often conceptualized as one where there has been a gradual recognition of liberty and the rights of the individual. One of the key messages that children learn about Mesopotamia is that the culture practiced human sacrifice and, even though this phenomenon is atypical of Mesopotamian civilization generally, this moment in the Early Dynastic period marks, for child audiences, the beginning of civilization and a period when the rights of the individual are not recognized. By “othering” ancient Sumer in this fashion, archaeology provides a foil 14.  The authors would like to thank James Linville for bringing this work to our attention.

326

Kevin M. McGeough and Elizabeth A. Galway

for contemporary North America and Europe and the starting point for a teleological trajectory that culminates in neo-liberalism.

Conclusions: Reversing the Hermeneutic Flow What makes the study of these children’s accounts of the ancient past so important is the power that historical treatments intended for children hold over those audiences. Nodelman (1990: 72) observes: The most significant fiction of history is the insistence that it is not fiction – that it does accurately represent the truth. Its success in denying its fictionality makes history particularly effective as propaganda: acceptance of its truthfulness allows readers to absorb the meanings and values it contains with less questioning or consideration than they might bring to that which is announced as invented fiction. It is this capability for propaganda that makes history intended for an audience of children so worthy of investigation.

When academic treatments of the past are transformed into genres intended for child audiences, much of that transformation involves crafting that academic work to convey specific messages that, because of the claims of historicity, can go unquestioned. In this paper, we have offered some preliminary thoughts on the impact that Woolley’s work at Ur and his representations of those works had on the reception of the ancient Near East in children’s literature. There is little evidence that Woolley’s narratives have been superseded in children’s outreach by more recent scholarly treatments of the data. This suggests that something is perhaps missing in archae­ ological work in public outreach. By reversing the hermeneutical flow and thinking about these presentations to children in the context of the discipline, we can gain insight into some of our professional practices. In terms of work as university and public educators, we can see how easily we fit our data to match pre-existing tropes about the past or established narratives of world history. Sometimes this may just involve lazy teaching but it may also suggest some of the ways that we are not reflective about how we think about Mesopotamia in relation to other disciplines or other bodies of knowledge. These results are not just important for understanding education, however. Looking at presentations aimed at children may also make us more aware of some of the unquestioned assumptions that inform our research. We can see the actual evidentiary basis for some of our grander assumptions in simplified form. Modeling our arguments in this fashion may lead to new ways of thinking about the relationship between evidence and the claims that are made surrounding it.

Bibliography Ashton, J. and Down, D.. 2006 Unwrapping the Pharaohs: How Egyptian Archaeology Confirms the Biblical Timeline. Green Forest, AR: Master Books. Benjamin, W. 1968 Illuminations. New York: Shocken Books. Costecalde, C. 1997 The DK Illustrated Family Bible. London: DK.

Excavating Ur in Children’s Literature

327

Cowley, M. 2008 The Golden Bull: A Mesopotamian Adventure. Watertown, MA: Charlesbridge. Crisp, P. 2009 Atlas of Ancient Worlds. New York: DK Publishing. Dalton, R. W. 2016 Children’s Bibles in America: A Reception History of the Story of Noah’s Ark in US Children’s Bibles. Scriptural Traces: Critical Perspectives on the Reception and Influence of the Bible 5. Library of the Hebrew Bible / Old Testament Studies 614. London: Bloomsbury. Emmerson, J. 2014 The Complete Illustrated Children’s Bible. Eugene, OR: Harvest House Publishers. Fagg, C. 1980 Exploring Lost Cities. Stevens Point Wisconsin: Piper Books Ltd. Finkel, I. 2007 On the Rules for the Royal Game of Ur. Pp. 16–38 in Ancient Board Games in Perspective. Papers from the 1990 British Museum Colloquium with Additional Contributions, ed. I. Finkel. London: The British Museum Press. Foster, B. 1981 A New Look at the Sumerian Temple State. JESHO 24: 225–241. Gold, P. S. 2004 Making the Bible Modern: Children’s Bibles and Jewish Education in TwentiethCentury America. Ithaca: Cornell University Press. Hade, D. and Edmondson, J. 2003 Children’s Book Publishing in Neoliberal Times. Language Arts 81.2: 135–144. Hastings, S. 1994 The Children’s Illustrated Bible. London: DK. Kopytoff, I. 1986 The Cultural Biography of Things: Commoditization as Process. Pp. 64–91 in The Social Life of Things: Commodities in Cultural Perspective, ed. A. Appadurai. Cambridge: Cambridge University Press. Lassieur, A. 2013 The Ancient World: Ancient Mesopotamia. New York: Scholastic Press. Loon, H. van. 1921 The Story of Mankind. New York: Boni & Liveright. Lucas, A. L. 2003 Introduction: The Past in the Present of Children’s Literature. Pp. xiii–xxi in The Presence of the Past in Children’s Literature, ed. A. L. Lucas. Westport, CT: Praeger. Maurice, L. 2015 Children, Greece and Rome: Heroes and Eagles. Pp. 1–14 in The Reception of Ancient Greece and Rome in Children’s Literature: Heroes and Eagles, ed. L. Maurice. Metaforms: Studies in the Reception of Classical Antiquity, Vol. 6. Leiden: Brill. McCaughrean, G. 2002 Gilgamesh the Hero. Grand Rapids, MI: Eerdmans Books for Young Reader. McGeough, K. M. 2006 Heroes, Mummies, and Treasure: Near Eastern Archaeology in the Movies. NEA 69 3–4: 174–185. 2015 The Ancient Near East in the Nineteenth Century, Appreciations and Appropriations II. Collecting, Constructing, and Curating. Hebrew Bible Monographs 68. Sheffield: Sheffield Phoenix Press.

328

Kevin M. McGeough and Elizabeth A. Galway

Meskell, L. 2004 Object Worlds in Ancient Egypt: Material Biographies Past and Present. New York: Berg. Nodelman, P. 1990 History as Fiction: The Story in Hendrik Willem van Loon’s Story of Mankind. The Lion and the Unicorn 14.1: 70–86. 2008 The Hidden Adult: Defining Children’s Literature. Baltimore: Johns Hopkins University Press. Pulley, K. 2005 The Beginner’s Bible: Timeless Children’s Stories. Grand Rapids, MI: Zonderkidz. Quinn, M. 1994 The Swastika: Constructing the Symbol. London: Routledge. Rose, J. 1984 The Case or Peter Pan or The Impossibility of Children’s Fiction. Philadelphia: University of Pennsylvania Press. Scholl, E. 2010 How’d They Do That? in…Ancient Mesopotamia. Hokessin, DE: Mitchell Lane Publishers. Shuter, J. 2006 Excavating the Past: Mesopotamia. Chicago: Heinemann Library. Steele, P. 2007 Eyewitness Mesopotamia. London: DK Publishing, Inc. Stevenson, D. 2003 Historical Friction: Shifting Ideas of Objective Reality in History and Fiction. Pp. 23–32 in The Presence of the Past in Children’s Literature, ed. A. L. Lucas. Westport, CT: Praeger. Taylor, K. 2000 A Child’s First Bible. Carol Stream, IL: Tyndale House Publishers. Thaler, D. 2003 Fiction Versus History: History’s Ghosts. Pp. 3–12 in The Presence of the Past in Children’s Literature, ed. A. L. Lucas. Westport, CT: Praeger. Vander Stichele, C. and Pyper, H. S. 2012 Text, Image, and Otherness in Children’s Bibles: What Is in the Picture? Atlanta: Society of Biblical Literature. Weinlich, B. 2015 The Metanarrative of Picture Books: ‘Reading’ Greek Myth for (and to) Children. Pp. 85–104 in The Reception of Ancient Greece and Rome in Children’s Literature: Heroes and Eagles, ed. L. Maurice. Metaforms: Studies in the Reception of Classical Antiquity Vol. 6. Leiden: Brill.

Aspekte der Sklaverei im altbabylonischen Ur: Untersuchungen zu den a- r u - a- Texten Christin Möllenbeck

Westfälische Wilhelms-Universität Münster

Einleitung Im Folgenden präsentiere ich einen Auszug aus meinen Forschungen zur institutionellen Sklaverei in Mesopotamien während der altbabylonischen Zeit. 1 Im Fokus stehen dabei die Texte aus Ur, insbesondere die a-ru-a-Texte, in denen Männer, Frauen, und Kinder als Weihgaben vom Tempel empfangen wurden. Diese Textgruppe wurde in der Wissenschaft bisher kaum beachtet oder gar bearbeitet, aber eine detaillierte Studie der altbabylonischen a-ru-a-Texte aus Ur kann hinsichtlich der Inanspruchnahme unfreier Arbeit im kultisch-religiösen Bereich lohnend sein.

Übersicht des Textmaterials zur Sklaverei im altbabylonischen Ur Im altbabylonischen Ur wurden Sklaven, Sklavinnen, und Sklavenkinder durch folgende Bezeichnungen definiert: (sag) urdu/úrdu (PN), (sag) gém e (PN), urdu-géme ḫi- a , sag munus PN, wardum, rēšum, und ṣuḫartum. Fehlen diese klaren und unmissverständlichen Begriffe, besteht die Möglichkeit eine Person aus dem Kontext des Textes (oder Archivs/Dossiers) heraus als unfrei bzw. gewaltunterworfen zu bezeichnen, wie beispielsweise in dem Mietvertrag UET 5 240. Dort heißt es:

1.  Die hier dargestellten Ergebnisse sind Teil meiner noch in Arbeit befindlichen Dissertation “Institutionelle Sklaverei – Die Inanspruchnahme unfreier Arbeit in den kultisch-religiösen und politischadministrativen Institutionen der altbabylonischen Zeit.” Sie ist Teil des Projekts “Im Dienste der Götter und des Königs – Politische, ökonomische und soziale Interaktion im Bereich kultisch-religiöser Institutionen in Mesopotamien im ausgehenden 3. und frühen 2. Jt. v. Chr” unter der Leitung von Prof. Dr. Hans Neumann am Exzellenzcluster “Religion und Politik” in Münster.

329

330

Christin Möllenbeck

Vs.

Ka-li-li m u- ni

1:

I

2:

á-bi nì m u- 1- k am

3:

4 g ín k ù- babbar

4:

k i A-at-ta-a

5:

I

6:

in- ḫun un

7:

šà á- bi nì m u- 1- k am

Den sogenannten Kalili, hat Imgur-Ea von Atta für ein Jahr für einen Lohn von 4 Schekel Silber gemietet.

Im-gur-é . a Von dem Lohn für ein Jahr wurde ein Schekel Silber empfangen.

8:

1 g ín k ù- babbar

u. Rd.

9:

šu- ba- t i- a

Rs.

10:

ig i Ga-mi-lum

Vor Gāmilum,

11:

I

30-a-ša-re-ed

Sîn-ašarēd

12:

I

lú- d ! ba- w à

Awīl-Bawa

13:

I

Im-li-ku

Imlikum

14:

i tu

g u 4 - si- sá u 4 - 30- k am

30. Ajaru (II)

15:

m u- 5- k am - ma

Rīm-Sîn 34

16:

ì- in-si- n[a]

o. Rd.

Der Mietling Kalili wird zwar nicht als Sklave bezeichnet, eine Selbstvermietung (stets ausgedrückt durch ki ra-ma-ni-šu “durch sich selbst”) 2 ist aber auszuschließen, sodass er eindeutig in einem Abhängigkeitsverhältnis zu seinem Vermieter Imgur-Ea stand. Ein Abhängigkeitsverhältnis aufgrund einer familiären

2.  Lautner 1936: 3.

Aspekte der Sklaverei im altbabylonischen Ur

331

Verbindung, also ein Vater-Sohn-Verhältnis von Vermieter und Mietling, würde durch die Angabe der Filiation zum Ausdruck kommen, wie Vergleichstexte zeigen. 3 Zusätzlich steht hinter dem Namen des Mietlings der Ausdruck mu-ni “sein Name,” der in den altbabylonischen Urkunden sehr häufig nach Sklavennamen notiert wurde. Anhand dieser Kriterien und den eindeutigen Bezeichnungen lassen sich schließlich 63 Texte, die auf die eine oder andere Weise auf Sklaven, Sklavinnen, oder Sklavenkinder hinweisen, identifizieren (Abb.1). Sie datieren in die Zeit der 1. Dynastie von Larsa ab Gungunum 7 (1926 v. Chr.) bis zur 1. Dynastie von Babylon mit Hammurapi und Samsuiluna (1743 v. Chr.) und umfassen eine Zeitspanne von etwa 183 Jahren. Die Texte verteilen sich wie folgt auf diverse Textgattungen (Abb. 2):

Die Erwähnungen in Briefen betreffen entlaufene Sklaven oder Sklaven als Kaufgegenstand; in Adoptions-, Erbteilungs-, und Mitgifturkunden sind sie Teil des notierten Vermögens. Die Schenkungen von Sklaven erfolgen als Ersatz für die Versorgung der Eltern der Schenkenden. Ferner werden Sklaven gekauft, vermietet oder freigelassen. In den Darlehensurkunden (davon vermutlich eine Schülertafel) werden Sklavinnen als Pfand gegeben. Auch wenn sich unter diesen Texten durchaus interessante und diskussionswürdige Belege bezüglich der Sklaverei befinden, die teilweise schon von Diakonoff in seiner Abhandlung “Ljudi goroda ura” (Moskau 1990) betrachtet wurden, werden sich die folgenden Ausführungen auf die a-ru-a-Texte beschränken, die zusammen mit weiteren administrativen Texten zum Verständnis der Sklaverei im kultisch-­ religiösen Bereich beitragen können.

3.  So zum Beispiel in Sippar: Goetze JCS 11 (1957) Nr. 14, YOS 13 486, VAS 8 46 (Vermieter=Vater), oder TLOB 30 (Vermieter=Bruder).

332

Christin Möllenbeck

Wie die Verteilung der Geschlechter (Abb. 2) bereits zeigt, nehmen die a-ru-a­ exte eine Sonderstellung ein: In allen Textgattungen dominiert die Anzahl der T Sklaven die der Sklavinnen und Sklavenkinder, außer in den a-ru-a-Texten. Dort dominieren Sklavenkinder mit insgesamt 10 Belegen gegenüber Sklaven und Sklavinnen. Diese 10 Belege sind darüber hinaus im Vergleich zu den Belegen für Sklavenkinder aus Ur insgesamt (nur 14) signifikant hoch.

Die a -r u - a - Texte Die Überreichung einer als a-ru-a bezeichneten Weihgabe an einen Tempel in Form von kostbaren Gegenständen, Tieren, und Personen ist eine seit präsargonischer Zeit 4 bekannte Tradition in Mesopotamien und wurde 1972 erstmals von Gelb im Zusammenhang mit der Publikation von zwei Ur III-zeitlichen administrativen Texten aus Lagaš diskutiert. 5 Er verstand das Prinzip der a-ru-a-Institution bezogen auf die Stiftung von Menschen wie folgt: Wohlhabende Bürger übergaben dem Tempel Personen aus ihrem Haushalt, die sie entbehren konnten. Verarmte Personen hingegen übergaben dem Tempel Familienmitglieder, die nicht mehr versorgt werden konnten, also größtenteils Frauen und Kinder. Gelb betonte dabei, dass es sich um eine quid pro quo-Vereinbarung handelte, bei der auf der einen Seite die Armen und Schwachen versorgt wurden—auch ein wichtiges Motiv in der mesopotamischen Königsideologie—und auf der anderen Seite der Tempel mit Personal versorgt wurde. Inwiefern seine Beobachtungen auf die altbabylonische Periode angewendet werden können, wird im weiteren Verlauf erörtert werden. Wie bereits eingangs erwähnt haben die a-ru-a-Texte aus Ur in der Assyriologie bisher kaum Beachtung gefunden. Marc van de Mieroop machte in seinem Artikel über “Gifts and Tithes to the Temples in Ur” darauf aufmerksam, dass diese Art der Weihung von Menschen an den Tempel existierte, sein Schwerpunkt galt jedoch der Abgrenzung der a-ru-a-Gaben—in Form von Menschen, Tieren, und kostbaren Wertgegenständen—von den profanen Händlerabgaben an den Tempel. 6 Von den altbabylonischen a-ru-a-Texten aus Ur liegt—bis auf eine Ausnahme 7— bisher keine Bearbeitung vor, sodass der Befund zunächst vorgestellt werden soll.

YOS 5 68 Vs.

Rs.

1:

genna g é m e A-ma-at- e n. zu

Kind, Sklavin, Amat-Sîn,

2:

a -r u- a A-ḫi-é - a

a - r u - a - Gabe des Ahī-Ea.

3:

ì -dab 5 I-din-dDa-mu

Übernommen: Iddin-Damu.

4:

i t u du 6 - k ù

Tašritu (XII)

5:

m u k a - zal- lu

d

⌈

⌉

ki

ba- g ul

Warad-Sîn 2

YOS 5 68 ist ein typischer Vertreter der a-ru-a-Texte. Er beginnt mit der Beschreibung des a-ru-a-Objektes, in diesem Fall der Sklavin Ammat-Sîn (durch die Bezeichnung genna wird deutlich, dass es sich noch um ein Kind handelte). Es 4.  Westenholz 1999: 61 Anm. 253. 5.  Gelb 1972: 1–32. 6.  Van De Mieroop 1989: 397–401. 7.  UET 5 564 (Archibab: Renger Corpus Berlin, http://pix.archibab.fr/4Dcgi/20562W7114.pdf).

Aspekte der Sklaverei im altbabylonischen Ur

333

folgen die Nennung des Stifters Aḫī-Ea und der Vermerk, dass Iddin-Damu die Sklavin übernommen hat. Der Text schließt mit dem Datum. Text-Nr.

Datum

a-ru-a

Stifter

ì-dab5

UET 5 564

---

d u m u Sîn-šeme

Ikūn-pî-Sîn

-

YOS 5 68

WS 2

1 g e nna g é m e PN

Aḫī-Ea

Iddin-Damu

YOS 5 72

WS 2

1 PN u r du

Nabi-Enlil, S. v. Sîn-abūšu

Ur-Nanna

dum u- m unus g e nna PN

Iddin-Damu

YOS 5 78

WS 2

g e nna g é m e PN

Gimillum, ašgab

Iddin-Damu

YOS 5 161

WS 2

g e nna dum u PN

Ḫunubtum, F. v. Sîn-gamil, lúmá-lá

Ningišzidagamil

YOS 5 74

WS 4

1 PN

Elali, S. v. Apil-aḫī

Sîn-abūšu

YOS 5 92

WS 4

1 PN

Ur-Amurrum

Iddin-Damu

YOS 5 67

WS 5

1 dum u- m unus P[N]

Sîn-muballiṭ

Illulu

1 dum u- m unus g e n n a PN PN dum u- m unus- a- n i

Ur-Nanna

Sîn-abūšu

1 PN g é m e

Enlil-mālik

Illulu

dum u [PN]

Ištar-ilum

Pirnidum

dum u P[N]

Nanna-palil

YOS 5 73

WS 5

YOS 5 77

WS 5

1 PN g é m e uš- bar x 1 P[N] u rd u

Imgur-Sîn, S. v. Narām-Sîn

Nūr-Ninšubur

YOS 5 71

WS 7

PN ur du

Šū-Ninšubur

Ku-Nanna

YOS 5 28

WS 8

1 g é m e PN m u- ni 1 g e nna PN dum u- m unus- a- ni

Sîn-rīm-Ur

nicht notiert

Mit Ausnahme von UET 5 564 stammen alle Urkunden, in denen eine Person dem Tempel in Form von a-ru-a geweiht wurde, aus Raubgrabungen und wurden 1919 von Ettalene M. Grice in YOS 5 publiziert. UET 5 564 ist der einzige Text von dem gesichert ist, dass er im Ganunmaḫ, 8 eine Verwaltungseinrichtung des Nanna-Ningal-Tempelkomplexes, aufbewahrt wurde. Renger datiert ihn aufgrund paläographischer und prosopographischer Indizien in die Regierungszeit des Sumu-El von Larsa. 9 Er ist also weitaus älter als die übrigen Texte, die in die ersten 8 Regierungsjahre Warad-Sîns datieren. Dank dieses Textes können zwei Schlussfolgerungen gezogen werden: Erstens kann eine Kontinuität der Stiftung von Menschen zwischen Sumu-El und Warad-Sîn angenommen werden, und zweitens dürften auch die a-ru-a-Texte aus YOS 5 im Ganunmaḫ-Disktrikt zu verorten sein. Darüber hinaus gibt es in UET 5 weitere a-ru-a-Texte, in denen Tiere und 8.  UET 5 564 (BM 131353) gehört zur Fundgruppe U8811 “from under pavement of E-nun-mah,” Woolley and Mallowan 1976: 232. 9.  Archibab: Renger Corpus Berlin (http://pix.archibab.fr/4Dcgi/20562W7114.pdf).

334

Christin Möllenbeck

kostbare Gegenstände dem Tempel geweiht wurden und eindeutig dem Ganunmah zugewiesen werden können. 10

Gestiftete Personen Die a-ru-a-Texte beginnen mit der Nennung der Person, die an den Tempel geweiht wird. UET 5 564

d u m u Sîn-šeme

YOS 5 28

1 g é m e PN m u- ni

d u mu [PN]

1 g e nna PN dum u- m unus- a- ni

d u mu P[N]

YOS 5 67

YOS 5 73

1 PN g é m e

1 dum u- m unus P[N]

YOS 5 74

1 PN

1 dum u- m unus g e nna PN

YOS 5 77

1 PN g é me u š - b a r x

PN dum u- m unus- a- ni

1 P[N] u rd u

YOS 5 68

1 g e nna g é m e PN

YOS 5 78

g e n n a g é me PN

YOS 5 71

PN ur du

YOS 5 92

1 PN

YOS 5 72

1 PN u r du

YOS 5 161

g e n n a d u mu PN

d um u- m unus g e nna PN

Fünf von den gestifteten Personen werden unmissverständlich als géme “Sklavin” bezeichnet (YOS 5 28, 68, 78, 73, und 77). Zwei von ihnen sind zusätzlich mit dem Vermerk genna gekennzeichnet, also als “Kind” ausgewiesen. In YOS 5 77 wird sogar deutlich, dass es sich um eine uš-bar “Weberin” handelte, eine für Sklavinnen typische Tätigkeit. In YOS 5 28 wurde eine Sklavin zusammen mit ihrer Tochter geweiht. Daneben gibt es eine Reihe von Kindern, die ohne Mutter oder Vater an den Tempel übergeben wurden (UET 5 564; YOS 5 67, 73, und 161). In den Texten werden sie häufig ohne Namen und nur mit entsprechender Filiation genannt. Schließlich wurden auch urdu “Sklaven” geweiht (YOS 5 71, 72, und 77). Daneben gibt es Personen, die nicht als Sklave bezeichnet und nur namentlich identifiziert werden (YOS 5 74 und 92). Die Tatsache, dass sie von einer anderen Person geweiht wurden, ist jedoch Hinweis genug, dass sie sich in einem Abhängigkeitsverhältnis zum Stifter befunden haben müssen.

Die Stifter Über das Motiv des Stifters geben die Texte keinerlei Auskunft: Befand er sich in einer wirtschaftlichen Notlage und konnte sich daher nicht mehr um seine Kinder oder seine Frau kümmern? Oder handelte es sich um prestigeträchtige Weihgaben als Zeichen der Pietät? Eine Kombination dieser beiden Möglichkeiten ist ebenfalls nicht auszuschließen. Über die Stifter selbst erfahren wir leider häufig nicht mehr als den Namen, manchmal auch die Filiation und zweimal den Beruf (YOS 5 161: lú má-lá; YOS 5 78: ašgab).

10.  S. Hinweise bei Van De Mieroop 1989: 397–401.

Aspekte der Sklaverei im altbabylonischen Ur

335

Empfänger der gestifteten Personen Ähnlich wie bei den Stiftern verhält es sich mit den Empfängern der als a-ru -a gestifteten Personen, da sie gleichermaßen nur namentlich erwähnt wurden. Dennoch ist es möglich, mehr über ihre Funktion hinsichtlich der Verwaltung von geweihten Personen zu erfahren. UET 5 564

-

YOS 5 73

Illulu

YOS 5 28

-

YOS 5 67

Illulu

YOS 5 74

Sîn-abūšu

Pirnidum

Sîn-abūšu

YOS 5 77

Nūr-Ninšubur

YOS 5 68

Iddin-Damu

YOS 5 78

Iddin-Damu

YOS 5 71

Ku-Nanna

YOS 5 92

Iddin-Damu

YOS 5 72

Ur-Nanna

YOS 5 161

Ningišzida-gamil

Iddin-Damu

Die Empfänger von als a-ru-a bezeichnetet Personen sind niemals als Empfänger von anderen als a-ru-a ausgewiesenen Gaben, wie Tieren oder kostbaren Gegenständen, belegt. Es ist davon auszugehen, dass sie für die Administration der geweihten Personen innerhalb des Tempels verantwortlich waren. Die Zuteilung einer a-ru-a-Gabe an diejenige Person, die unmittelbar Gebrauch von ihr machen konnten, ist auch in anderen Texten aus Ur belegt: So wurden beispielsweise Tiere direkt an den Schäfer oder den Kuhhirten übergeben und Musikinstrumente an einen Musiker. 11 Durch die Heranziehung weiterer administrativer Texte des Nanna-NingalTempelkomplexes kann diese Annahme bestätigt werden. Schon Van de Mieroop identifizierte Illulu (YOS 5 67 und 73) und Sîn-abūšu (YOS 5 67 und 74) als Empfänger von Dattelrationen für die “Truppe der Korbträger” (érin un-íl) 12 in YOS 5 12 aus dem Jahr Warad-Sîn 1. 13 Dieser Text deutet darauf hin, dass Illulu und Sîn-abūšu eine administrative Position hinsichtlich der Versorgung von Arbeitern innehatten. Weiterhin sollen drei administrative Texte aus Ur (YOS 5 80, 14 81, 15 und 102 16) herangezogen werden. YOS 5 80 und 81 datieren in die Jahre Sîn-iqīšam 3 und Warad-Sîn 2. Aufgrund der kurzen Regierungszeit Sîn-iqīšams und seines Nachfolgers Ṣillī-Adads liegen die Texte nur 6 Jahre auseinander und entstanden ungefähr gleichzeitig mit den hier besprochenen a-ru-a-Texten. Sie notieren Fehlbestände an Gersten-Mehl der ugula géme-uš-bar “Aufseher der Weberinnen,” darunter ein

11.  Van De Mieroop 1989: 398. 12.  Zu un-íl s. Sigrist 1979; Sigrist 1980; Steinkeller 2003: 44f. 13.  Van De Mieroop 1989: 398 Anm 19. 14. Archibab: Renger Corpus Berlin (http://pix.archibab.fr/4Dcgi/18696X8459.pdf); Verweis bei Butz, K., Der Nanna-Ningal-Tempelkomplex in Ur, in: WZKM 65/66 (1973–74) 35 Anm. 74. 15. Archibab: Renger Corpus Berlin (http://pix.archibab.fr/4Dcgi/18698T3308.pdf); Verweis bei Butz, K., Der Nanna-Ningal-Tempelkomplex in Ur, in: WZKM 65/66 (1973–74) 35 Anm. 74. 16.  S. Butz 1973–1974) 35ff. [Transliteration von ausgewählten Zeilen und Kommentar] und Archibab: Renger Corpus Berlin (http://pix.archibab.fr/4Dcgi/18704D1863.pdf).

336

Christin Möllenbeck

Ningišzida-gamil. 17 Eine Person gleichen Namens empfängt in dem a-ru-a-Text YOS 5 161 ein Kind. Es stellt sich nun die Frage, inwiefern der Ningišzida-gamil, “Aufseher der Weberinnen” mit dem Empfänger Ningišzida-gamil aus den a-ru-aTexten identifiziert werden kann. YOS 5 102, eine Liste mit Wollerträgen, gibt Aufschluss für diese Fragestellung. Sie ist nicht datiert, kann aber durch prosopographische Übereinstimmungen in die Zeit Sîn-iqīšam—Warad-Sîn datiert werden: Alle Aufseher aus YOS 5 80 und 81 sind auch in 102 notiert und als “Aufseher der Weberinnen” bezeichnet. In dieser Liste sind darüber hinaus neben Ningišzida-gamil zwei weitere “Aufseher der Web­ erinnen” notiert, die auch als Empfänger von einer als a-ru-a geweihten Person belegt sind: Sîn-abūšu (YOS 5 67 und 74) und Ku-Nanna (YOS 5 71). Die Tatsache, dass alle drei in einer Auflistung der “Aufseher der Weberinnen” vorkommen, die zeitgleich mit den a-ru-a-Texten entstanden ist, deutet auf ihre Funktion im Bereich der Verwaltung von Tempelpersonal hin. Siegelabdrücke auf Quittungen, die ebenfalls zeitgleich mit dem Empfang von geweihten Personen datieren (Warad-Sîn 2 und 5), geben weitere Hinweise auf die Funktion der Empfänger Ku-Nanna und Ur-Nanna. Ku-Nanna wird durch die Inschrift seines Siegels auf YOS 5 47 (Warad-Sîn 5), 51 (Warad-Sîn 2), 52 (Warad-Sîn 2), drei Quittungen über Schafe, als “a-ru-aSchreiber” ausgewiesen: 1: k ù - d n a n n a 2: d u b- s a r a -r u -a 3: d u mu PN 4: u r d u - é - d n a n n a -k e 4

“Ku-Nanna, a-ru-a-Schreiber, Sohn des PN, Diener des Nanna-Tempels.” Zu dem Empfänger Ur-Nanna gehört möglicherweise das Siegel: 1: u r - d n ann a 2: g u d u 4 - a b zu d n a n n a 3: d u mu k ù - d n i n -g a l 4: š a 13 - d ub -b a d n a n n a 5: u r d u ur d u - d e n .z u

“Ur-Nanna, gudapsûm-Priester des Nanna, Sohn des Ku-ningal, šandabakkum des Nanna, Diener des Warad-Sîn,” das ebenfalls auf den oben genannten Quittungen YOS 5 47, 51, und 52 abgerollt wurde und auch auf Quittungen gleichen Typs (YOS 5 46, 50, und 53, alle Warad-Sîn 2) erscheint. 18 Der Empfänger Ur-Nanna könnte folglich der gudapsûm-Priester und šandabakkum sein, dessen Archiv Dominique Charpin in seiner Abhandlung “Le Clergé d’Ur” diskutierte. 19

17.  YOS 5 80, Z. 9f. und 81, Z. 2. 18. Es gibt noch ein weiteres Siegel des Ur-Nanna: 1) u r- d n a n n a g u d u 4 a b z u 2) g á - d u b - b a d nanna 3) dumu k ù - n i ng a l . Es wurde auf einer Urkunde, die den Kauf eines Grundstücks vom Tempel dokumentiert, abgerollt (YOS 5 112, Warad-Sîn 9). 19.  Charpin 1986: 42–51.

Aspekte der Sklaverei im altbabylonischen Ur

337

Die Empfänger Pirnidum (YOS 5 73) und Illulu (YOS 5 67 und 73) erhalten außer­dem einen Sklaven und zwei Sklavinnen in YOS 5 42 aus dem Jahr Warad-Sîn 5: 20 A-na-d u t u -tak-la-ku ú r d u 2) ì -d a b 5 Pi-ir-ni-du-um “Ana-Šamaš-taklāku, Sklave, empfangen (von) Pirnidum.” 1) I

i š 8 - tár -tap-pa-[ti 21 gé ] m e 7) d u m u -m unus Ia-lí-x-[x x] ⌈úr du ⌉ 8) I Um-mi-[ṭaba-a]t g é m e 9) d u m u -m u n u s g e n n a d Ma-mi-tum-re-⌈me-ni⌉ 10) ì -dab 5 Il-lu-lu “Ištar-tapp[ātī, Skl]avin, Tochter des Iali-…, ⌈Sklave⌉, Ummī-[ṭāba]t, Sklavin, Tochter, Kind, der Mamitum-rē⌈mēnī⌉, empfangen (von) Illulu.” 6) I

Auch wenn die Sklaven und Sklavinnen hier zwar nicht als a-ru-a bezeichnet wurden, geht aus diesem Text deutlich hervor, dass Pirnidum und Illulu als Verantwortliche für unfreie Personen im Tempelkomplex agierten. Nach Einbeziehung der administrativen Urkunden kann also festgestellt werden, dass die Empfänger der als a-ru-a bezeichneten Personen auch für deren zukünftigen Aufenthalt im Temple verantwortlich waren.

Anlass der Weihung und Einsatz im Tempel Schließlich werden in einigen wenigen Urkunden Informationen über den Anlass der Weihung, oder vielleicht besser: das mögliche Motiv der Weihung notiert. In den drei relevanten Texten handelt es sich jedes Mal um einen kultischen Anlass. In dem a-ru-a-Text YOS 5 71 (Warad-Sîn 7) wird ein Sklave als Ersatz (ki-bé-gar-ra) für eine andere Person für das bur-sag des Nanna empfangen. Das bur-sag, bekannt ab der präsargonischen Zeit, 22 (im spätaltbabylonischen Uruk auch é-bur-sag) 23 sorgte für reguläre Opfergaben. Ein weiterer geweihter Sklave wird in YOS 5 72 (Warad-Sîn 2) für das bur-sag empfangen. An dieser Stelle sollen zwei weitere administrative Text (YOS 5 41 und 69, beide aus dem Jahr Sîn-iqīšam 4) herangezogen werden. Sie notieren zwar keine a-ru-a-­ Gaben, stehen aber in enger Verbindung mit den oben vorgestellten Texten. YOS 5 41 dokumentiert den Empfang zweier Männer, die mit Silber gekauft worden sind, von denen einer für das bur-sag übernommen wurde. 24 Die Tatsache, dass die beiden Personen für Silber gekauft worden sind, ist ein deutlicher Hinweis auf ihren Status als unfreie Person. In gleicherweise wird in YOS 5 69 ein Mann mit dem Wert von 19 Schekeln Silber nach dem Gewichtsstein des Šamaš für das bur-sag empfangen. 25 Die gekauften Personen werden von den aus den a-ru-a-­ 20.  Archibab: Renger Corpus Berlin (http://pix.archibab.fr/4Dcgi/18652L1717.pdf). 21.  Archibab: Renger Corpus Berlin (http://pix.archibab.fr/4Dcgi/18652L1717.pdf) ergänzt -tap-pa[at-tum], jedoch kein weiterer Beleg zur Schreibung dieses Namens gefunden. Zur Ergänzung -tap-pa[ti] vgl. FM 4 4 [T.315], i 20: e š 4 - t ár -t[ap-pa-ti] und FM 4 15 [M.16000], 54: e š 4 - tá r-tap-[pa-ti]. 22.  Seri 2013: 160 und Anm. 53. 23.  Seri 2013: 159–161. 24.  YOS 5 41: 1) Išu- d nin- m ar ki 2) k ù- t a sa 10 3) n a m b u r - s a g “Šū-Ninmar, für Silber gekauft für das bursa g.” 25. YOS 5 69: 1) I úr du- d e n. zu m u- ni 2) k ù- t a 19 g í n 3) n a 4 - d u tu 4) mu - tú m 5) d e n . z u -e-ri-baam 6) nam-ḫar-tí b ur - sag d n a n n a “Warad-Sîn, sein Name, sein Wert 19 Schekel Silber (nach) dem Gewichtstein des Šamaš, geliefert (von) Sîn-erībam. Empfangen für das Bursaggum-Opfer der Nanna.”

338

Christin Möllenbeck

Texten bekannten Empfängern Pirnidum und Ur-Nanna übernommen, womit erneut ein Hinweis auf deren Funktion als Verwalter von Arbeitskräften existiert. Ein weiterer Anlass wird in YOS 5 161 26 deutlich: ein Kind wird für die “Prozession des Schiffes der Erstlingsfrüchte” 27 geweiht. Ob das Kind eine Rolle bei der Prozession spielte, kann nicht gesagt werden. Es ist naheliegender, dass der Zeitpunkt ausschlaggebend für die Weihung war. Dies zeigen auch weitere a-ru -aTexte aus Ur, in denen beispielsweise Tiere für die “Prozession des Schiffes der Erstlingsfrüchte” geweiht wurden. 28

Weitere Belege für unfreie Personen im Nanna-Ningal-Tempel Schließlich lohnen sich einige Bemerkungen zu administrativen Texten aus Ur, die zwar keine a -ru-a-Weihgaben nennen, aber dennoch Aufschluss über die Verwaltung und den Einsatz unfreier Arbeitskräfte im Nanna-Ningal-Tempel geben können. In YOS 5 70 (Warad-Sîn 5) werden zwei Kinder von Ilum-pîšu geliefert und von Bur-Sîn für die Göttin Ningal übernommen. 29 Der Empfänger Bur-Sîn ist ähnlich wie andere Empfänger von geweihten Personen, auch in der Liste mit Wollerträgen YOS 5 102 als ugula géme uš-bar “Aufseher der Weberinnen” verzeichnet. YOS 5 76 (Sîn-iqīsam 5) ist der einzige Text, in dem die Tätigkeit eines Sklaven eindeutig formuliert wird: II-din-Èr-ra ì-dab 5 ur- d nanna nam a-bal-šè “Iddin-Erra, übernommen von Ur-Nanna. Für das Wasserschöpfen.” Dass es sich bei Iddin-Erra um eine unfreie Person handelte, ist nur durch den Kontext der bisher besprochenen Texte ersichtlich, da Ur-Nanna als Empfänger von geweihten Personen und als “Aufseher der Weberinnen” belegt ist.

Schlussbemerkung Die Untersuchung der a-ru-a-Texte unter Heranziehung weiterer administrativer Urkunden aus dem altbabylonischen Ur ermöglicht einen Zugang zum Verständnis des Erwerbs und des Einsatzes unfreier Arbeitskräfte in kultisch-religiösen Institutionen in dieser Stadt. Weiterhin wurde deutlich, dass die Tradition der Weihung von Menschen an den Tempel im frühen zweiten Jahrtausend v. Chr. fortgeführt wird, und zwar in ganz ähnlicher Weise, wie sie für das dritte Jahrtausend belegt ist. Gelbs Aussagen über die a-ru-a-Tradition der Ur III-Zeit können für die altbabylonische Zeit auch für Ur bestätigt werden. Es sind überwiegend Frauen und Kinder, die dem Tempel geweiht wurden. Die aus dem dritten Jahr26.  YOS 5 161: 1) g e nna dum u k al- dŠu-bu-l[a] 2) a - ru - a 3) g ì r m á - n e s a g - g á “Ein Kind, Sohn des Dān-Šūbula, Weihgabe für die Prozession des Schiffes der Erstlingsfrüchte.” 27.  Siehe hierzu van Dijk 1965: 21–24; Van De Mieroop 1989: 398; und Charpin 1986: 134. 28.  Van De Mieroop 1989: 398. 29. YOS 5 70: 1) IMu-ba-li-iṭ-tum 2) dumu-munus genna Ša-at-dNu-muš-da 3) d u mu - mu n u s - a - n i 4) ki-bé-gar-ra x [x] x 5) Idx-[x-x-x]-x 6) dum u- m unu s g e n n a A-x[-x-x]-d n i n - š u b u r 7) [d u mu - mu n ] us -a-ni 8) ù 10 gín k ù- babbar 9) d nin- g al- šé 10) m u - tú m d i n g i r- i n i m- š u 11) ì - d a b 5 Bur-d e n . z u “Muballiṭum, Tochter, Kind, des Šāt-Numušda, seine Tochter. Ersatz … ..., Tochter, Kind, des A...Ninšubur, seine Tochter und 10 Schekel Silber für Ningal. Geliefert (von) Ilum-pîšu. Empfangen (von) Būr-Sîn.”

Aspekte der Sklaverei im altbabylonischen Ur

339

tausend bekannte Weihung von Personen und Objekten in Form von a-ru-a ist in der altbabylonischen Zeit bisher nur aus Ur bekannt. Es verwundert nicht, dass ausgerechnet in Ur, dem Zentrum des Königreichs der 3. Dynastie von Ur, diese Tradition noch Anwendung finden konnte.

Bibliographie Butz, K. 1973/4 Konzentrationen wirtschaftlicher Macht im Königreich Larsa: Der NannaNingal-Tempelkomplex in Ur. WZKM 65/66: 1–58. Charpin, D. 1986 Le Clergé d’Ur au siècle d’Hammurabi (XIXe-XVIIIe siècle av. J.-C.). HEO 22. Paris: Librairie Droz. Dijk, J. van. 1965 Une insurrection générale au pays de Larša avant l’avènement de Nūradad. JCS 19: 1–25. Gelb, I. J. 1972 The Arua Institution. RA 66: 1–32. Lautner, J. G. 1936 Altbabylonische Personenmiete und Erntearbeiterverträge. SD 1. Leiden: Brill. Seri, A. 2013 The House of Prisoners. Slavery and State in Uruk during the Revolt against Samsu-iluna. SANER 2. Boston/Berlin: de Gruyter. Sigrist, M. 1979 Érin-un-íl. RA 73: 101–120. 1980 Érin-un-íl. RA 74: 11–28. Steinkeller, P. 2003 Archival Practices at Babylonia in the Third Millennium. Pp. 37–58 in Ancient Archives and Archival Traditions. Concepts of Record-Keeping in the Ancient World, ed. M. Brosius. Oxford: Oxford University Press. Van De Mieroop, M. 1989 Gifts and Tithes to the Temple in Ur. Pp. 397–401 in DUMU-E2-DUB-BA-A. Studies in Honor of Åke Sjöberg, eds. H. Behrens, D. Loding, and M. T. Roth. Occasional Publications of the Sameul Noel Kramer Fund 11. Philadelphia: Babylonian Section, University Museum. Westenholz, A. 1999 The Old Akkadian Period: History and Culture. Pp. 15–117 in Mesopotamien. Akkade-Zeit und Ur III-Zeit, eds. P. Attinger and M. Wäfler. OBO 160/3. Göttingen: Vandenhoeck and Ruprecht. Woolley, L. and Mallowan, M. 1976 The Old Babylonian Period. UE 7. London: Trustees of the British Museum and of the University Museum, University of Pennsylvania, Philadelphia; British Museum Publications.

Glyptic Art from the Ur III to the Šimaški Periods: Heritage and Overtaking of the Models Clélia Paladre

University of Paris

Introduction The purpose of this paper is to study the relations between Ur III and the dynasty of Šimaški at the turn of the second millennium BCE, through the lens of glyptic art. Ur III influences are clearly visible in the Šimaški glyptic. The final aim will be to discuss the adaptations of the Ur III models in the Šimaški glyptic, in order to understand which cultural and political realities are implied by such adaptations. I will study the evolution of the iconographic patterns from the Ur III to the Šimaški periods, focusing on the concept of cultural and political transition as re­ flected through the glyptic imagery, and I will explore when, why, and how the Šimaški officials engaged with this Mesopotamian “heritage.” All these consider­ ations are based on the assumption that the evolution of iconographic patterns should be understood as a reflection of social, cultural and political realities. As P. Amiet (1986: 210) has stated “l’étude des sceaux ne saurait se restreindre à la simple esthétique.” Before beginning, it is important to review the different “problematic” data that are related to this subject including terminology, political chronology and strati­ graphic data.

Terminology: One problematic term is “Elam.” This name has been used in so many ways to talk about the southwestern part of Iran that it is difficult to establish its clear mean­ing. Researchers use it as a synonym for many political entities, especially relevant here for Šimaški or Sukkalmah dynasties. 1 Following J. de Morgan (1896) who first defined “Elam” as a double-headed entity between highlands and lowlands, some define it as a real kingdom with political, cultural, and geographical realities, and with Susa and Tal-e Malyan as hearts of the territory. 2 If this definition can apply to the situation after the middle of the second millennium BCE, it is not 1.  See the recent synthesis about this problematic term by P. Michalowski (2008: 113) and D. T. Potts (2016: 1–13, 428–440). 2.  See Potts 2016 for a full overview.

341

342

Clélia Paladre

the case for the earlier periods. Beginning in the 1980s, scholars raised the issues concerning the definition of the geographical term. Researchers use it alterna­tively as having either a political meaning or a geographical one, very often in the same text. 3 But this ambiguity leads gradually to the consensus that “Elam” has a geo­ graphical mean­ing and designated the Iranian plateau and its piedmont (such as the Susiana) in the third millennium (Zadok 1987; Carter and Stolper 1984). P. Michalowki (2008: 112–113) recently proposed that “Elam” cannot be understood as a political entity before the Old Babylonian period. I agree with this definition. That is the reason why I prefer not to use “Elam” in this paper, except with reference to later periods. As M. Stolper (1982) established, and V. Scheil (1900; 1901) before him, “Elam” is a large geographic unity which included many political, cultural, geographical, and linguistic entities, and which was simplified to the extreme by the Mesopotamian vision. Susa and Tal-e Malyan were surely important sites for us (especially Susa here, since the majority of the seals are coming from there) but there were other important centers as well. The second problematic term is “Šimaški.” The meaning and, thus, the localiza­ tion of Šimaški are not settled. It has to be kept in mind that its earliest use refers to a geographic territory. 4 It is only after a period of time that this denomination was employed to designate the political formation. Thus, the factual geographical and political realities implied by the term “Šimaški” change through time. The heart of Šimaški (meaning the original land of Šimaški) is a debated subject. Some researchers consider the central and southern Zagros with its intermountain valleys as the heart of Šimaški (Hinz 1964; Kupper 1969; Stolper 1982; 1984; or Henrickson 1984) while some others prefer Kerman (Vallat 1980; 1985; 1991; and 1993), or the area in between Khuzestan and Fars (Cameron 1936; Lambert 1972; Amiet 1979; de Miroschedji 1980) or even recently Central Asia (Potts 2008; 2016: 143). For this paper I adopt the position that the original geographical land of Ši­ maški stands somewhere in between Khuzestan and Fars (Ram Ormuz), and/or in the western part of the Central Iranian Plateau (Fig. 1). 5 I join this opinion espe­ cially because this region shows a demographic decline during the Early Bronze Age period and then an increase during the last quarter of the third millennium (Carter and Stolper 1984: 135–136), precisely at the same moment as the emergence of the Šimaški power as recorded in the cuneiform texts. This core land was progressively extended beginning with the reign of the king (Kindattu) who contributed to the fall of Puzur Inšušinak (ca. 2110 BCE) to the rise of Ebarat I (ca. 2060 BCE).

3.  For example, F. Vallat (1980) defines Elam in the third millennium as a kind of political “fed­ eration” which becomes a geographical region occupied by the Persians and so called Persia by first millennium. P. de Miroschedji in his paper from 1980 speaks about “civilisation élamite” (1980: 129) and then defined “Elam” as a geographical territory just before speaking about “Grand Élam” as a kind of political federation (1980: 137, 139). 4.  The word “Šimaški” first appears at the very end of an inscription of Puzur Inšušinak which records his conquests in the Zagros. A king of the land of Šimaški is said to have paid tribute to him. See MDP 14 (1913: 7–16) and more recently Steinkeller 2014: 289, “in a single day he made (those lands) fall prostate at his feet; and, when the king of Šimaški came up (on learning about it), he seized the feet of Puzur-Inšušinak (in submission).” 5. For a recent and complete current state of the question, see Steinkeller 2014: 291–295. The conclusions obtained were already suggested by M. J. Steve in 1989, R. Zadok in 1991 and P. Steinkeller in 2007.

Glyptic Art from the Ur III to the Šimaški Periods

343

We have valuable information about the location of the state of Šimaški at the beginning of its expansion (so after the middle of the twenty-first century, during the reign of the Ur III dynasty) thanks to an inscription dated from Šu Sin 7 6 where six main districts (Zabšali, Šigriš, Nibulmat or Iapulmat, Alumiddatum, Garta or Karda, and Šatilu) and three peripheral zones (Azahar, Bulma, and Nususmar) are said to make up the Šimaški state (Stolper 1982: 45–46; Steinkeller 2007: 216; Potts 2016: 132). These localities can all be situated around the Central Zagros and Khuzestan, thus representing the first step of the expansion of Šimaški toward the northwest. The Išbi Erra royal hymn 7 allows us to know that the territory seems to extend farthest south, as far as Anšan in Fars 8 just after (and thanks to) the

Fig. 1. Localization of Šimaški from core to state (Paladre, map adapted from M. Sauvage Fig. 1). 6.  This text is in fact a copy of four different inscriptions mentioning the same military campaign. The more famous and quoted is the one copied from the base of the statue/stele of Šu Sin (see Kutscher 1989 and De Graef 2015: 294–295). 7.  This hymn narrates the war between Ibbi Sin, last king of the Ur III dynasty, and Kindattu, king of Šimaški. It also explains how Išbi Erra, king of Isin, finally defeated Kindattu and brought back peace in the region (Stolper 1984: 20). 8.  The city of “Pašime on the sea-coast” is mentioned. It was considered as the extreme southern extension of the Šimaški territory since it was identified on the Gulf coast. However, the recent excava­ tions in Tell Abu Sheeja, southwest of the Jebel Hamrin, revealed that it was in fact the ancient city of Pašime (Hussein et al. 2010). This revelation questions M. Stolper’s reading of the geographical situation

344

Clélia Paladre

fall of Ur III (Steinkeller 2014: 291). Unfortunately, no information is available for the remainder of the dynasty, but we can imagine that the territories remained essentially the same, at least until the reign of Ebarat II and the beginning of the dynasty of Sukkalmah, around 1980 BCE. To conclude and generally speaking, we can situate the state of Šimaški within the modern provinces of Kermanshah, Kurdistan, Hamadan, Luristan, and Khuze­ stan, and up until the border of Fars.

The Dynastic Succession There are a number of theories and interpretations about the dynastic suc­ cession of the Šimaškian kings. All of them depend on the interpretation of two different sources: the Royal King List of Susa 9 (RKL) and the “Genealogy of Šilhak Inšušinak.” 10 The RKL is dated to the Old Babylonian period and gives us the names of twelve Šimaški rulers (Fig. 2) together with the list of the twelve kings of Awan. This list is supported by some contemporary inscribed seals, silver vessels

Fig. 2. Royal king list of Susa and contemporary Mesopotamian rulers (Paladre). of Šimaški since Pašime is located farthest north (Stolper 1982: 45–46). Thus, this is not an axis northsouth - east-west as supposed by M. Stolper, but a double axis east-west that would be described. 9.  It was first published by V. Scheil (1931: 2–8) in RA 28 and then republished in MDP 23 (1932: IV). No precise indication of the original context was given except its Susian origin. 10.  It was published by F. W. König (1965: 110–115).

Glyptic Art from the Ur III to the Šimaški Periods

345

(the so-called gunagi), inscribed bricks, and tablets. 11 The “Genealogy of Šilhak Inšušinak” is dated to the twelfth century BCE and mentions old kings who built temples in Susa. 12 The relevance of this evidence is debated. While some researchers consider the RKL as a reliable historical source, such as M. Stolper (1982), F. Vallat (1996) and P. Steinkeller (2007; 2014), some others have questioned its historicity and prefer to use the “Genealogy of Šilhak Inšušinak,” such as J. J. Glassner (1996) and K. De Graef (2006). Moreover, the supporters of the RKL are divided into two opposite camps since some of them take it as a real genealogical list (Steinkeller 2014) while some others think that it should be interpreted with caution or with “small adaptations” (Stolper 1982; Vallat 1996; and Steinkeller 2007). 13 Two important discoveries must be mentioned here because they changed the debate drastically. Both of them were made by P. Steinkeller, the first one is the identification of the term LU.SU (see also LU.SU(.A) or LU.SUki) as a synonym of Šimaški (Steinkeller 1988). 14 The other (Steinkeller 2007) is an inscription on a gunagi vessel in which Idattu I is called son of Kindattu and grandson of Ebarat I. 15 Researchers who had first chosen to follow the “Genealogy of Šilhak Inšušinak,” changed their minds after the publication of this inscription. From then on, they all tried to manage the competing realities evoked in the “Genealogy” and in the RKL (De Graef 2015, Steve et al. 2002; and Glassner 2013). Therefore, the “Genealogy of Šilhak Inšušinak” is not a contemporary source of the period and logically may in­ clude some mistakes and incoherencies—e.g., Kindattu was the father of Idattu and not his grandson; Tan Ruhurater is known to be the husband of Makubi, daughter of Bilalama of Ešnunna, so, it is impossible that Kindattu, victor over the Ur III dynasty, could have lived after him; and finally Šilhaha is mentioned as the beloved son of Ebarat while at the same time he is identified as his “vizir” (Glassner 2013: 320). 16 In sum, we cannot now question the RKL. We decide to follow it even if the information must be treated with caution. Indeed, the current trend of the research itself considers it as more or less correct (Sallaberger and Schrakamp 2015: 25). 11.  For more precision, see Potts 2016, Table 5.5 which gathers the Šimaški kings mentioned in the cylinder seal legends, and Table 5.4, which gathers the attestations of Šimaški kings in the texts, and Desset 2018: 115–116 who lists the six vessels that give names of Šimaški kings. 12.  Among them, three old Šimaški kings are mentioned: Idattu I, his son Tan Ruhurater, and his grandson Kindattu. 13.  F. Vallat (1996) proposes to see the RKL as two different lists with associated rulers, one ruling on the lowlands, the other on the highlands. M. Stolper (1982; 1984) suggests that the first kings could have been contemporary while the last part of the list could have been a chronographic list of rulers. He is followed by P. Steinkeller who specifies and confirms this view in 2007 (before changing his mind in 2014) and by K. De Graef in 2008. In 2002, M. J. Steve et al. try to reconcile the RKL and the “Genealogy of Šilhak Inšušinak.” 14.  This is confirmed by two versions of the same text in the name of king Sin-iddinam, discovered in Emar. The term “Šimaški” in the first version is replaced by “LU.SU.KI.AN” in the second version. It has to be noted that some researchers disagree with that (Steve et al. 2002: 433). However, since then, a text from the Ur III period corroborates it since Girnamme is identified as “LU.SU.Aki” (Steinkeller 2007: 216). 15.  This inscription comes from two bronze vessels of unknown provenance (the so-called gunagi). One vessel was sold by Christie’s in London in 2001, the second one belongs to the Martin Schøyen Collection in Oslo. 16.  I prefer to put it aside and to consider it as an attestation of a propaganda campaign in which Šilhak Inšušinak’s scribes wrote the names of the old kings following a general lineage and not strictly chronologically (Steve et al. 2002: 436).

346

Clélia Paladre

Although the purpose of this paper is to study the glyptic art, an attempt to reconstruct the dynasty of Šimaški and its story is necessary in order to situate available glyptic evidence in time. This attempted reconstruction is summarized on Fig. 3. These historic and epigra­ phic debates are out of my spheres of competence but I will try to summarize the data in the following section. The first attestation of a king of Šimaški (and of the word “Šimaški” itself) is recorded on a fragmentary statue of Puzur Inšušinak, last ruler of the Awan dy­ nasty 17 (Steinkeller 2013: 302 and 2014: 288, contra Stolper 1984: 20, who sees the beginning of the dynasty around 2030 BCE and not before). This king 18 Fig. 3. Attempted reconstruction of the Šimaški dynasty (Paladre). is said to have paid tribute to Puzur Inšušinak. This reference establishes the importance and prestige of the Šimaški rulers as early as this time (André-Salvini 2006: 129; Steinkeller 2013: 302; and Steinkeller 2014: 288) and, thus, allows us to situate the beginning of this dynasty around 2100 BCE at least. Indeed, it is now recognized that the first kings of Šimaški assisted in the defeat of Puzur Inšušinak. An alliance between the first king of the Ur III dynasty, Ur Namma (2110–2093 BCE), the king of Lagaš, Gudea, 19 and the king of Šimaški ended the reign of Puzur Inšušinak and negotiated an agreement for sharing power 17.  Puzur Inšušinak was probably not truly king of Awan. Indeed, on the RKL of Susa, he is said to be the twelfth king of the Awan dynasty. The eleventh king, Hita, was a contemporary of Naram Sin, fourth ruler of the Akkad Empire, who reigned around 2261–2206 BCE. Thus, Hita ruled some decades (almost a century) before Puzur Inšušinak. He undoubtedly assumed this title as an old and mythic titulature, very useful to impose himself in the eastern lands (Steinkeller 2013: 296). 18.  It could be Girnamme (Steinkeller 2013: 302; 2014: 288; 2018: 195; and André-Salvini 2006: 129). However, the textual evidence attests to the contemporaneity of Girnamme, Ebarat I, and Tazitta around 2040–2030 BCE (Stolper 1982; Steinkeller 2007). P. Steinkeller (2014: 288) then suggests the existence of two Girnamme. 19. See recently, Pittman 2018 about the possibility of an alliance between Gudea and Puzur Inšušinak before this coalition. Indeed, Gudea could have been slightly earlier than Ur Namma (591–599).

Glyptic Art from the Ur III to the Šimaški Periods

347

over lands: the lowlands for the southern Mesopotamian dynasties and the highlands for the Šimaški dynasty (Steinkeller 2013: 303; 2014: 288; and 2018: 192). Whether this last division of power is true or not, the collapse of Puzur Inšušinak’s power allowed the Šimaški dynasty to begin its growth (Steinkeller 2018: 195). Moreover, the Šimaški dynasty can be considered as the direct successor of the Awan one (Stolper 1982: 49–50; André and Salvini 1989: 71; and Sallaberger and Schrakamp 2015: 24) and the reign of Puzur Inšušinak can be seen as a real “watershed event of the early Elamite history” (Steinkeller 2013: 303). 20 The development of the state of Šimaški is clearly a complex phenomenon. M. Stolper, in his fundamental paper of 1982, assumes the contemporaneity of the three first kings on the RKL list: Girnamme, Tazitta and Ebarat, on the basis of texts dated to Amar Sin and Šu Sin, around 2060–2030 BCE. He was followed by many researchers including P. Steinkeller (2007). But, in 2014, P. Steinkeller ques­ tioned this reconstruction by establishing the contemporaneity of the first king of Šimaški (Girnamme according to him) and Puzur Inšušinak, sometime around 2100 BCE. Hence, the Girnamme mentioned first on the RKL and the one mentioned together with Ebarat and Tazitta on the Ur III documents cannot be the same. Two different Girnamme existed. On the other hand, we are not certain that the RKL is comprehensive and we do not know the real purpose behind its creation (maybe it was meant to legitimate a specific branch of the dynasty). Even if it is strange to mention three contemporary kings on a royal king list while the rest of the kings on the list seem to be more or less consecutive, we have to remember that this list was a posterior creation. Thus, it is likely that the first three kings were on the list because of their importance in the original constitution of the state even if they did not literally rule the Šimaški state. This issue is still unresolved and here, I prefer to stay neutral and just consider the chronological ranges of the different figures mentioned on the RKL (summarized in Fig. 2) and the importance of the reign of two figures in the formation and establishment of the Šimaški state: Ebarat I and Kindattu. The pioneering works of M. Stolper (1982: 49–54; 1984: 20) and the continua­ tion of it by P. Steinkeller (2007; 2014; and, more recently, 2018: 195) have clearly established the importance of Ebarat I. He imposed himself on the other paramount rulers or chiefs of his time (Girnamme and Tazitta appear in texts from the Ur III period alongside him 21) while maintaining regular diplomatic contacts with the Ur III dynasty (Steinkeller 2007: 222; 2014: 287). Textual evidences suggest that Tal-e Malyan, ancient Anšan, was an ally 22 of the emerging state of Šimaški during the reign of Tazitta I, “man of Anšan” (Steinkeller 2007: 221–222), and truly became part of the Šimaški state from the end of the reign of Ebarat I when he placed his son Kindattu on the throne 23 (Steinkeller 2007: 224; 2014: 287). Ebarat took control of Susa during the first part of the reign of Ibbi Sin (from year 4 to 8 of his reign). 20.  Especially since he is the first political figure of southwestern Iran who achieved to unify the region, integrating Susiana (André-Salvini 2006: 130 and Steinkeller 2013: 303). 21.  Whether there is another earlier Girnamme or not. 22.  A vassal if we follow the works of P. Steinkeller (2007: 225–227; 2014: 287). 23. According to Steinkeller 2007, it is possible that the successive campaigns of Šulgi against Anšan helped Ebarat I in his power race (Steinkeller 2007: 227). However, P. Steinkeller mentioned in 2013 that campaigns against Kimaš and Hurti were the “only foreign expedition mentioned in Šulgi’s historical inscriptions” (Steinkeller 2013: 295).

348

Clélia Paladre

This control over the city was short lived since Ibbi Sin retook control of the city during his ninth year (Steinkeller 1989: 274–275; 2007: 223; and De Graef 2004: 107–108; 2005: 99, 105–106, 112–113; and 2015: 294). 24 In sum, it is during his reign and the beginning of the reign of his son, Kindattu, that Šimaški spread and became a powerful unified state that was in a position to lead the coalition which brought about the fall of Ur III state and even occupied Ur until Išbi Erra finally regained control of the city (Stolper 1982: 48; Steinkeller 2007: 228; 2018: 196). The end of the story of the Šimaški dynasty is, again, heavily debated. By con­ trast with M. Stolper (1982: 55–56) who thought that it was a peaceful transfer of power between two branches of the same extended ruling family or F. Vallat (1996: 315–316) who thought that the Šimaški kings ruled over the high lands while the early Sukkalmah ruled over Susa 25 and finally imposed themselves over the Šimaški kings, J. J. Glassner established convincingly in 2013 that the subsequent dynasty of Sukkalmah originated in a power struggle between the last Šimaški kings: Idattu II and Ebarat II. 26 The Šimaški power was weakened and Šilhaha, the Sukkalmah (vizir) of Ebarat II, took the opportunity to set up his own dynasty sometime after 1980 BCE. 27 From there, the Šimaški kings disappeared from the history of western Iran replaced by the new ruling dynasty: the Sukkalmah. If such a reconstruction seems complex, it can be explain by the fact that Ši­ maški was never a centralized power but rather was a pyramidal structure with different entities that were more or less autonomous and with “rulers” of various ages and ranks linked by marriages or family ties more than by political fealty or economic structures (Stolper 1982: 48). This unusual model explains why we have so many names and titles in uncertain relation to each other. We can assume that, at least in part, we are facing an organization where rulers were supported by their children who had been installed as “junior local ruler” (Stolper 1982: 48). This is made obvious through multiple examples of Šimaški kings who are mentioned along with their sons: the seal of Imazu (see Fig. 9b) mentions him as son of Kindattu and “king of Anšan” (however, he is not mentioned on the RKL), an inscription on a brick discovered in Susa mentions Tan Ruhurater as son of Idattu I and “ensi of Susa” (Potts 2016: Table 5.5). Later on, Tan Ruhurater will become king of Šimaški (the eighth one on the RKL). It is obvious that the future reigning princes would also have some important roles to play before ascending the official throne of Šimaški.

24.  Here again, it has to be noted that a debate exists. Twelve different texts from Susa mentioned Ebarat. They were published in 1932 in MDP 23 by V. Scheil (see texts 291–292: 162, texts 295–302, and 304–305: 163–165) who attributed it to Ebarat I. He was followed by M. Lambert in 1979, while W. Hinz (1973: 260) attributed it to Ebarat II. Moreover, M. Stolper argued that even if the reign of Ebarat was important, he never took control over Susa and no Šimaški documents from Susa itself did exist before Kindattu (Stolper 1982: 49). He was followed by M. J. Steve (1989: 14). In 2002, M. J. Steve et al. argued for an occupation of Susa by Ebarat I but as soon as the reign of Šu Sin, years 5 and 6 (Steve et al. 2002: 434). 25.  Illustrating by this the first attestation of the partition of the power between low and highlands. 26.  It could be an internal struggle between brothers, if we consider the Royal King List of Susa as an exact family tree. 27.  That could explain why he is considered as a reference by his successors with the title “ruhušak of Šilhaha” (sister’s son of Šilhaha).

Glyptic Art from the Ur III to the Šimaški Periods

349

The Stratigraphic Data (Fig. 4) Turning to the archaeological data, there is some existing at Susa, derived from the excavations of E. Carter and R. Ghirshman on the tell de la Ville Royale. R. Ghirshman excavated two levels in his Operation B: level VII uncovered the private house of Igibuni the scribe, which can be securely dated to the Ur III period. 28 This level shows traces of violent destruction that can be attributed either to the conquest of Ebarat I, or to an attempt by Ibbi Sin to recapture the city just after the conquest of Ebarat I (Gasche 1973: 12–13; and De Graef 2015: 290). Level VI is dated to the Šimaški dynasty (Steve, Gasche, and De Meyer 1980: 87–88). 29 E. Carter excavated an Ur III level: level 3 from the Operation I (Carter 1978: 198). 30 On the basis of this archaeological data, a level excavated by R. de Mecquenem in the 1920s in the middle of the southwest edge of the tell de la Ville Royale can be identified as a transition phase between Ur III and the Šimaški period (De Graef 2015: 294). 31

Fig. 4. Chrono-stratigraphy of Ur III and Šimaški phases on the sites of Susa and Tal-e Malyan (Paladre).

The data coming from Tal-e Malyan are more problematic. Two operations yielded material dated from the Kaftari period (2200–1600 BCE). Operation ABC level I has been published by W. Sumner (1972; 1974; 1976; and 1989). Unfortu­ nately, even if glyptic material is essential for this study, no stratigraphic data can be used since it is a trash deposit. Operation GHI yielded at least two stratigraphic levels for this period and seems to be an administrative building. But this time, it is not published. Despite the dates which fit with our period, D.T. Potts (2016: 143) assume that Ur III and Šimaški phases are not present because of the absence of cultural characteristic material. But data from GHI Operation will demonstrate its presence.

28.  A seal mentioning Šu Sin (Fig. 5a) and tablets mentioning Šu Sin and Ibbi Sin were discovered. 29.  A sealing mentioning Makubi, Tan Ruhurater’s wife, was discovered (see Fig. 7a; Ghirshman 1968: 6). 30.  Ur III seals were discovered. 31.  Tablets mentioning Ebarat I together with Ur III graves were identified.

350

Clélia Paladre

First Step: The Cohabitation This corresponds to the formative phase of the Šimaški state, with the first figures of the RKL (Girnamme to the fifth king: Lu (…) ak-luuh-haan), around the twenty-first century BCE. At that time, the dynasty of Šimaški is an emerging pow­ ­er with whom the Ur III kings must contend. They seem to have maintained good relations with the first rulers of the dynasty and especially with Ebarat I. 32 By the reign of Ibbi Sin, Ebarat I apparently sensed the weakening of the power of Ur III and he began to assert his own expansion (the takeover of Susa is a good example of this process) (Steinkeller 2007: 228 and De Graef 2015: 296). Even if we have extensive documentation coming from the Girsu archives, we do not have a full understanding of the level of independence of each entities/regions in the east. The Ur III dynasty was surely extended in the eastern territories, espe­ cially into the future lands of the state of Šimaški. Some of those were politically integrated into the peripheral areas of the empire (or ma-da in Sumerian) like Susa, Pašime, or AdamDUN (so Khuzestan in general) and formed a belt running parallel to the Zagros and the Tigris ranges. These territories were directly under the control of Ur III, paid taxes, and provided services and military support in case of war (Steinkeller 2018: 194). While other areas bordering the ma-da territory were totally independent (because too distant, too instable, or too powerful) and their relations with Ur III varied through time (see Steinkeller 1987 for the full study). Through raids (more than real conquests), 33 through its domination of Khuzestan as the gateway to the east (Steinkeller 2018: 194), through a policy of diplomatic marriages (Potts 2016: 126 and Steinkeller 2018: 195), 34 through close economic ties, and through the presence of Ur III soldiers, messengers and merchants, the Ur III state was present in these heterogeneous and independent lands of western Iran (Garfinkle 2013: 153, 157 and Potts 2016: 132). In sum, all of western Iran was directly in contact with the Ur III dynasty and was totally within its political and economic orbit, even though it was not politically integrated, apart, of course, from Susa and alluvial Khuzestan (Carter and Stolper 1984: 18; Garfinkle 2013: 162; and Potts 2016: 126). 32.  According to P. Steinkeller, Ebarat would have been a vassal or at least an ally of Ur III. He argued that Ebarat I remained faithful to his commitments with Ur III, at least until the reign of Ibbi Sin (Steinkeller 2007: 227; 2014: 287). However, two problems appear: 1- how can we explain the mili­ tary campaign launched by Šu Sin in his sixth year of reign against the lands of Simaški (see the full inscription in Kutscher 1989: 90–91)? 2- close ties already mentioned between Ebarat I and Anšan make his relationship with Ur III problematic since he would have “felt between two stools” (see note 22). P. Steinkeller (2014: 287) argued that the Ur III dynasty tolerated the domination of Anšan in recognition of his services and proposed to see Ebarat I as an ally of Šu Sin during his campaign in the east, or at least remained neutral, since Ebarat I does not appear in the text as an enemy or ruler of one of the cities/territories mentioned (Steinkeller 2007: 223; 2018: 196). More recently, he (2018: 194) argued for the transfer of Anšan under the authority of Ebarat in the years 34–35 of Šulgi, in front of the acts of resistance of the city. We remain skeptical about these assumptions and prefer to see an alternation of good and “moderate” phases in diplomatic relations. Moreover, no tangible proof argues for any power or influence of the Ur III kings on Anšan/Tal-e Malyan. 33.  This fact is especially true under the reign of Šu Sin with many mentions of Šimaški (with its main district Zabšali) and Anšan (Carter and Stolper 1984: 17). 34.  Marriages between Šulgi’s daughters and kings of Anšan, Pašime, or Marhaši are cited in the texts, but also between a daughter of Šu Sin and a king of Anšan and between a daughter of Ibbi Sin and a king of Zabšali (Potts 2016: 128–129).

Glyptic Art from the Ur III to the Šimaški Periods

351

The only glyptic evidence available for that period comes from Susa and it all belongs squarely within the Mesopotamian tradition. As already observed by P. Amiet (1972: 209), it is difficult to differentiate the local realizations from Ur III imports. The presentation scene (the characteristic model of the Ur III production) is predom­inant on seals. Heavily standardized, the seal imagery invariably shows the same scene: a worshipper lead by a goddess in front of a seated figure, under a crescent moon. This seated figure can be a king who is holding a cup sitting on a padded stool on a dais, or a deity (in most case a goddess) who is raising an arm sitting on a box-like or paneled throne sometimes on a dais (Collon 1987: 36). This is the case on a cylinder seal mentioning Šu Sin (Fig. 5a), on another one without inscription (Fig. 5b) or on the Hulal’s cylinder seal (Fig. 5c).

Fig. 5a. Cylinder seal in the name of a servant of Šu Sin, agate, Susa, De Morgan excavations, Louvre Museum (Sb 9284). Presentation scene to the king. The figure of the god Adad, standing on a dragon is a later addition (Amiet 1973: pl. VIII, fig. 39).

Fig. 5b. Cylinder seal, serpentine, Susa, Louvre Museum (Sb 1478). Presentation scene to the goddess (Paladre, see also Amiet 1972: pl. 154, fig. 1650).

Fig. 5c. Cylinder seal in the name of Hulal, jasper, Susa Ville Royale, De Mecquenem excava­ tions, Louvre Museum. Presentation scene to the goddess (Amiet 1972: pl. 154, fig. 1646). Fig. 5. Cylinder seals of the Ur III period from Susa.

It is likely that Ur III seals were distributed beyond Susa and Khuzestan and they were surely well known and attested in all of southwestern Iran. Indeed, the different diplomatic or military strategies put in place by Ur III to keep control of Khuzestan (a direct access to the highlands and its commercial roads) were sup­ ported by the administrative technologies (Carter and Stolper 1984: 144, 150). The diffusion of these Mesopotamian models was facilitated by the long term presence of Ur III in western Iran: Susa was already under the control of Ur III by the reign of

352

Clélia Paladre

Ur Namma, 35 Khuzestan was under the direct control of the Ur III state by the reign of Šulgi (2092–2045 BCE), and was truly considered as a province of the empire by the reign of Amar Sin (2044–2036 BCE) (Carter and Stolper 1984: 16–17). This conclusion is reinforced through the absence of an indigenous glyptic tradition in this zone. The reign of Ibbi Sin is marked by the loss of Susa, first at the beginning of its reign when it was taken by Ebarat I (Steinkeller 1989: 274–275; 2007: 223) and second at the end of his reign, when it was taken by Ebarat’s son, Kindattu (Stein­ keller 2007: 228). 36 The Ur III influence on the eastern regions, already fragile by the end of the reign of Šu Sin, collapsed under the reign of Ibbi Sin. In 2004 BCE, Šimaški people, led by Kindattu, in alliance with others, took control of Ur and brought about the end of the Ur III Dynasty.

Second Step: The Apogee of the Dynasty of Šimaški This period can be seen as a real apogee of the Šimaški dynasty. Indeed, it is at this point that we can be confident of its establishment as a real unified state spread over a wide territory. This phase goes from the reign of Kindattu (sixth king on the RKL) to the reign of Idattu II (tenth king on the RKL) and corresponds to a very short time, approximately thirty years, from ca. 2000 to 1980/1970 BCE. After the fall of Ur III, we learn from an Old Babylonian text (the Išbi Erra hymn) that Ibbi Sin was deported to Anšan while Kindattu occupied Ur (Carter and Stolper 1984: 19). Kindattu was finally expelled from Ur by his old ally Išbi Erra, a former vassal of Ibbi Sin who betrayed him and declared himself ruler of Isin (Potts 2016: 134). The vacuum left by the end of the Ur III dynasty led to a real power struggle in which Šimaški, Isin, Larsa, and others were important protagonists. Turning to the glyptic data, it is not so easy to interpret. First, because of the quasi absence of seals or sealings discovered in a stratigraphic context. 37 Second, because of the difficulty in identifying Šimaški seals. This fact is admitted by all researchers, consciously or unconsciously. If P. Amiet (1972: 209; 1986: 149) first noticed that this phase was strongly influenced by the “Neo-Sumerian cultural and artistic traditions” and humbly proposes a simple classification, with one group show­ing Neo Sumerian influences and one showing old Babylonian influences (1972: 209); J. Börker Klähn (1970) followed by U. Seidl (1990) unconsciously excludes the Šimaški production and start their studies at the beginning of the Sukkalmah period, revealing by there, the lack of confidence when approaching these data. Meanwhile, they all recognize the progressive detachment of the Šimaški glyptic from its Mesopotamian models (Amiet 1973: 12; Seidl 1990: 129; Ascalone 35. Even if no Ur Namma’s year name mentions locations in the east regions, we now know of the existence of two fragments of stone vessels in his name that are supposed to be booty from Susa (Marchesi 2013). 36.  K. De Graef mentioned also the recapture of Susa at the end of the reign of Ibbi Sin, but she suggests that it was accomplished by Idattu I and Tan Ruhurater rather than Kindattu (2015: 296). 37.  Except three sealings with the name of Tan Ruhurater, Makubi and Idattu II which are per­ fectly Mesopotamian in style and, thus, cannot help in the establishment of a glyptic classification for the period. Sealings in the name of Tan Ruhurater and Makubi were discovered on a floor from the level BV of the Ville Royale. The sealing with the name of Idattu II was discovered in a jar, under a floor from level AXIV of the Ville Royale (so re-used). For a recent state of the research about the stratigraphic information concerning the glyptic elements, see Neumann 2013: 83–85.

Glyptic Art from the Ur III to the Šimaški Periods

353

2010: 629; and Ascalone 2018: 632). The first researcher who tried to define the “elamite elements” was, of course, P. Amiet, in his paper from 1973. He recognized the impossibility of defining a real single group but started to define the details of the local production (apart from the “série populaire” and from the series influenced by Mesopotamian styles). It consists in a simplification of the composition, a lack of attributes (royal or divine), specific headdresses (as the so called “coiffe en visière”), specific attitudes (arms extended in front of a seated or standing figure), more elong­ ated figures, a lack of details, and a more schematic treatment of the figures with a clear preference for the hatching and the angular shapes (Amiet 1973: 13–14). This last point, often mentioned by the researchers as a mark of clumsiness of the craftsmen, can be seen, according to me, as a real aesthetic penchant in western Iran, but I will come back to that later. Very recently, E. Ascalone proposed a new classification (2010; 2011; and 2018). He defined a general group for the period (“Typo 2A” in his book from 2011, which corresponds to “Old Elamite II Early Phase” in his paper from 2018) and detailed it in his paper from 2010 (632). He summarized the general specificities identified by P. Amiet in 1973 and suggested an internal division into three series: one “elamite provincial” characterized by some features coming from the Iranian Plateau but expressed in the royal workshops of Susa (2010: 632, 635); one purely “elamite” cha­ racterized by the angular figures and iconographical originalities (2010: 633–634); and one “elamo-mesopotamian” where the iconographic features reflects an “elamite” origin but the quality of the style revealed a Mesopotamian influence (2010: 634). If the goal is laudable, I cannot accept the absence of chronological development and markers, the derogatory vocabulary used to speak about Iranian productions in comparison with Mesopotamian ones (or oriental productions in comparison with Susa ones), and the constant reference to the supposedly double nature of the “Elamite” world. Moreover, it is difficult to imagine that three sub-groups can be sustained when he assumes that the glyptic production of the Šimaški period is hea­ vily heterogeneous (2010: 632). 38 In fact, E. Ascalone fell in the same old trap with a “Mesopo-centrism” vision of the glyptic production hiding behind a “Suso-centrism” one, and with an anachronistic vision of the Western part of Iran, which is not yet the “Kingdom of Elam” as known during the second millennium BCE. In the light of these elements, we will continue to focus on the general tendencies identified by P. Amiet. Indeed, the heterogeneity of the glyptic production shows the trial-and-error exploration and cannot be denied since it is the first step to the path of “emancipation.” Thus, the adoption and adaptation of the Ur III technical procedures of man­ agement and administration by the Šimaški state (Carter and Stolper 1984: 144) is clearly manifest through its glyptic production. 39 Indeed, it is under a “Mesopota­ mian model” that most of the Šimaški glyptic production flourished. The craftsmen integrated their own concepts into the models inherited from Ur III. This is totally 38.  At this point, Ascalone followed P. Amiet’s statement (1973: 13–14). But he changed his mind in 2018 when he claimed the opposite idea by mentioning a production heavily homogeneous in style, iconography, and themes (2018: 635). 39.  It is heavily possible that the scribes and intellectuals in charge of the administration proce­ dures were still from a “Mesopotamian culture” at this time. See, for example, M. Lambert’s research where he suggests that during the reign of Idattu II Akkadian scribes were present at Susa as a kind of middle class in between the local elite and the rest of the population (1971: 220).

354

Clélia Paladre

logical since no glyptic tradition was attached to Šimaški people before. Whether it is small adaptations or totally new production, only few elements stand out to differentiate the Šimaški seals from their Ur III models. Two principal practices seem to coexist. Firstly, the scene can be copied comple­ tely, without any variations (some reuse of actual Ur III seals are attested, but the inscription is then erased and re-inscribed). The presentation scenes are of course the most obvious models (Carter and Stolper 1984: 148), like on these seals of the attendant Il-alsu (Fig. 6a), the scribe Šu[…] (Fig. 6b), or the royal guard Hun-ili (Fig. 6c).

Fig. 6a. Cylinder seal in the name of Il-alsu, intendant, lapis lazuli, Susa, Louvre Museum. Presentation scene to the goddess (Amiet 1972: pl. 155, fig. 1673).

Fig. 6b. Cylinder seal in the name of Šu[...], scribe, steatite, Susa, Louvre Museum. Presenta­ tion scene to the goddess. The worshipper is mis­ sing due to the state of preservation of the seal (Amiet 1972: pl. 155, fig. 1670).

Fig. 6c. Cylinder seal in the name of Hun-ili, royal guard, marble, Susa Ville Royale, De Mec­ quenem excavations, Louvre Museum. Presenta­ tion scene to the goddess (Amiet 1972: pl. 155, fig. 1665). Fig. 6. Examples of Ur III Models in the Šimaški Glyptic.

Some seals from the Isin-Larsa style can also be mentioned here. This glyptic production is contemporary with the Šimaški one and is also directly inspired by Ur III models, even if there are notable changes. A new kind of presentation scene exists with the leading goddess now standing behind the worshipper with both hands raised while the worshipper has his hands clasped. It is the case for the sealing of Aebanda, scribe of the queen Makubi (Fig. 7a). On another kind of scene

Glyptic Art from the Ur III to the Šimaški Periods

Fig. 7a. Sealing in the name of Aabanda, scribe and servant of the Queen Makubi, clay, Susa Ville Royale Operation B, Level VI, Louvre Museum. Presentation scene to the god (Amiet 1972: pl. 34, fig. 1676).

355

Fig. 7b. Sealing in the name of Kuk Inšušinak, messenger and servant of the King Idattu II, clay, Susa, Louvre Museum. Offering scene to the god (Amiet 1972: pl. 157, fig. 1678)

Fig. 7. Examples of Isin-Larsa models in the Šimaški glyptic.

a seated god is facing a goat bearer. 40 It is the case on the sealing of Kuk Inšušinak, messenger of Idattu II (Fig. 7b). 41 Secondly, we can identify the use of the Ur III models with some modifications. It has to be noted that here again, we have the attestation of the reuse of actual Ur III seals recut by Šimaški craftsmen. There can be small adaptations, as seen on the seal of Pinikir’s servant (Fig. 8a) where the leading goddess is replaced by a man and the protagonists are now wearing a large mop of hair or a hat. This last element appears frequently; for example, on the seals of Inzuzu (Fig. 8b), Ir-nanna (Fig. 8c), and Kuk-kalla (Fig. 8d). It seems to be an ethnic feature since we can find them in later periods (Scheil 1926: 38; Amiet 1973: 14; Seidl 1990: 131; Porada 1990: 174–175; and Ascalone 2010: 632). Or, it can be small iconographic adaptations which, nevertheless, imply a real “cultural revision.” We can see that through a very special composition: the “investiture ceremony scene.” The king is shown giving an “official weapon” to his servant as a gift for his good services. If the style and/or the composition are still Mesopotamian, the subject is purely original and shows an oriental practice that has nothing to do with Mesopotamia. Finally, the meaning changes completely just by the addition of a small iconographic feature: a specific weapon. This revision seems to occur as soon as the beginning of the “apogee phase” of the dynasty (ca. 2000 BCE). One has to be careful with the dating because of the lack of stratigraphic data for this kind of scene. But, there are two examples with inscriptions that help in this debate. The first one mentions Idattu II, the tenth ruler of the Šimaški dynasty (Fig. 9a). It figures a perfect Isin-Larsa presentation scene but with the addition of an axe. 42 The second one mentions Imazu, son of Kindattu (Fig. 9b). Kindattu is men­ tioned on the RKL of Susa as the sixth ruler of the dynasty, whereas Imazu is not 40.  The scene on this seal is described by P. Amiet as an “investiture ceremony” (1972: 217, fig. 1678). But after studying it in the Louvre Museum, we can be sure that it is a goat and not a weapon. 41.  This seal is a subject of debate. U. Seidl (1990) thought that Idattu I was here mentioned, while D. T. Potts (2016) thought it was Idattu II. Most of the scholars are still careful about it (Amiet 1972 and De Graef 2006). 42.  There is a good 3D parallel to this weapon with Attahušu’s axe (Amiet 1986: 276, fig. 84).

356

Clélia Paladre

Fig. 8a. Cylinder seal in the name of Te[...], servant of Pinikir, lapis lazuli, Susa Ville Royale, De Mecquenem excavations, Tehran Museum. Presentation scene with a leading god (Amiet 1972: pl. 155, fig. 1662).

Fig. 8b. Cylinder seal in the name of Inzuzu, stone, Susa, Louvre Museum. Presentation scene (Amiet 1972: pl. 155, fig. 1669).

Fig. 8c. Cylinder seal in the name of Ir-nanna, marble, Susa, Louvre Museum. Presentation scene to the king (Amiet 1972: pl. 158, fig. 1688).

Fig. 8d. Cylinder seal in the name of Kuk-kalla, servant of Šilhaha and King Ebarat II, hema­ tite, Susa, Louvre Museum. Recut cylinder seal from the Ur III period. Presentation scene to the king(?) (Amiet 1972: pl. 157, fig. 1685).

Fig. 8. Examples of adaptations of Ur III or Isin-Larsa models in the Šimaški glyptic.

Fig. 9a. Sealing in the name of Kuk Simut, scribe and servant of King Idattu II, clay, Susa, Louvre Museum (Paladre).

Fig. 9b. Sealing in the name of Imazu, son of Kin­ dattu, king of Anšan, clay tablet, Susa, Louvre Museum (Amiet 1972: pl. 157, fig. 1679).

Fig. 9. Examples of the “investiture ceremony scene” in the Šimaški glyptic.

Glyptic Art from the Ur III to the Šimaški Periods

357

mentioned. He could be a younger brother of Idattu I, son of Kindattu and seventh ruler of the dynasty. Whatever the family link uniting these individuals, they must be generally contemporaneous. It figures the king and his servant standing face to face and exchanging the weapon. P. Amiet mentioned it in 1972, as a “provincial” creation commenting the poverty of the style, the scarcity of the details and the simplicity of the thematic (210). In 1986 he added that this seal escapes from the Mesopotamian stylistic sphere (150). If the second conclusion is mostly true (clothes are smooth and not flounced, figures are thinner, and we can notice the presence of mop of hair and axe), both points must be discussed. The theme could, in fact, have been influenced by an Ur III composition. Indeed some rare attestations of two standing figures facing each other are known from the Ur III period (Porada and Buchanan 1948: 35; and Collon 1982: 156); 43 more precisely, an innaba seal in the name of Ibbi Sin shows the king facing an official who holds a stick element (Legrain 1951: fig. 438, pl. 29). It is interesting to note that innaba seals were offered by the king to his servant as a gift (Sollberger 1965a: 29; and Scheil 1925: 147–148). So, unlike what P. Amiet said in 1972, the simplicity of the composition hides a real and important step forward for the Šimaši glyptic, which can be dated as early as the very beginning of the third millennium BCE (see Figs. 2–3). 44 Moreover, this composition, not so common in Mesopotamia, seems to have been well received in western Iran since many examples are known that date from our period or slightly later in time. We can mention two stamp seals discovered in Operation ABC in Tal-e Malyan (Sumner 1974: 172, fig. 12 I and J) or two cylinder seals from Susa (Amiet 1972: fig. 1703, and 1705 pl. 159) or again a sealing discovered together with the sealing of Idattu II in a jar, under a floor from level AXIV of the tell de la Ville Royale (Amiet 1972: fig. 2328 pl. 195). If we have a look on all the examples mentioned here, it is striking to observe that the style appears more rude and “cursory” for the majority. In 1972, P. Amiet already noticed that this trend toward more “rigidity” was a clue to define the sty­ listic evolution of the “elamite style” (211). This is a fact already observed by the majority of the researchers but interpreted as a “provincial feature”, a mark of the difficulty of mastering this art. 45 I prefer to see it as a local preference that is about to find a climax with the elaboration of the “gouged style” production. Important to this discussion is that the majority of these seals belonged to im­ portant persons, members of the state organization (scribe, royal guard or messen­ ger), and were made with luxury materials (lapis lazuli, hematite, or marble). Thus, we can argue that Mesopotamian style and iconographic models were used by the Šimaški state for its own bureaucratic and administrative needs. The adaptation and stylistic changes that are distinctive of the local Šimaški glyptic production started, not with the common people, as asserted by P. Amiet with his “série populaire.” As E. Carter stated (1984: 144), it is obvious that the Šimaški dynasty used the technical procedures of management of their old ally or/and “master” Ur III rulers in order to stabilize and create a real independent state that succeeded in uniting 43. Examples are also known from the Isin-Larsa period (Porada and Buchanan 1948: 37; and Collon 1986: 60–61). 44.  On the basis of the approximate date of the Imazu’s reign, earliest attestation of this composi­ tion known. 45.  See for example, P. Amiet (1972: 210) who states that if it is rough, it comes from the periphery or very recently E. Ascalone (2010: 632).

358

Clélia Paladre

the territories of the low and highlands into one political entity. It is logical that they would have adapted the imagery as well as the administrative use of the seals to effect this unification. From this perspective, the Ur III state served as a model for Šimaški to compete in the race for the Ur III succession on the international scene. E. Carter (2014: 43) argued that the importance of women in the Šimaški system is a reflection of an attempt to legitimize their place in the international affairs, 46 while M. Stolper (1984: 22) argued that the deification of the kings was also adapted from Ur III as an effective strategy for achieving international recognition for the Šimaški kings. The utilization of year names by the first Šimaški kings (De Graef 2008: 81; 2015: 68–69) is another example of this geopolitical strategy. We can now propose to add to this list the continuation and adaptation of Ur III glyptic models to facilitate entry into geopolitical affairs. We do not see the Šimaški state as a pure copy of the Ur III state or as an entity that tried to replicate a pure Mesopotamian model. Such an interpretation is “Mesopotamian centric” when in fact Šimaški was a kind of neighboring entity that was trying to engage with the Mesopotamian center. Šimaški had a complex political structure, created on unusual bases without any comparisons to Mesopota­ mia and that explains why, to me, past researchers struggle to find some common features and historical analogies with the Ur III state. 47 Such an approach cannot work simply because Šimaški had a project of its own. They took what was useful and practical for their international aspirations, but also what fitted with their internal organization, which was fundamentally different from Mesopotamia. For W. Hinz (1972), Šimaški was a “federal state;” for P. Stein­ keller (1988), it is a “confederation;” for M. Stolper (1982), it is an “extensive and interregional Union;” and for D. T. Potts 48 (1999; 2016), it is a “fragmentary, flexible and fluctuant state.” All these theories are close and revolve around the same idea that Šimaški was a state composed by many political entities joined in a flexible structure thanks to effective diplomatic strategies instead of displays of power and to flexible and reinforced lines of authority that could include many players. Thus, the adaptation of the Ur III practices and seals makes clear sense. If you want to unify different cultures and ethnicities, you need a common model. It is useful to reuse some models that are well known by everybody in the region: the most familiar model was that of the Mesopotamian Ur III state. In sum, adapting Ur III seals was a perfect strategy for both internal and external political affairs; to ensure a place on the international scene, but also to join different political and cultural entities under a same model. So, when P. Amiet (1986: 149) talks about “une culture largement dépendante” or about kings who act “comme des potentats mésopotamiens,” I totally disagree.

46.  The union between the queen Makubi, daughter of Bilalama of Ešnunna and Tan Ruhurater, eighth king of Šimaški is a perfect example of this practice of diplomatic weddings. 47.  See, for example, Kutscher 1989, who considers Šimaški as an “empire.” 48.  His work is built on the basis of A. Southall’s theories (1952). In the second edition of his books he used the terminology “segmentary state” (Potts 2016: 145).

Glyptic Art from the Ur III to the Šimaški Periods

359

Third Step: The Emancipation According to me, the real and complete emancipation of the Šimaški glyptic was achieved with the creation of local workshop for the realization of a genuine production. This aim was only accomplished at the very end of the dynasty, around 1980 BCE, with the reign of Ebarat II. My argument is based on the so-called “Ebarat’s seal” (Fig. 10). It is frequently discussed. Some assume that it is the seal of the Ebarat’s I wife (Steve 1989; Steve et al. 2002; Steinkeller 2007; Carter 2014; and Daems 2018); some others assume, very recently, that this seal belongs to none of them but to an important court lady (Mofidi-Nasrabadi 2009: 43, 49 followed by Neumann 2013: 91); although, like W.G. Lambert (1979) and D. T. Potts (1999; 2016), I prefer to see it as the seal of the Ebarat’s II wife. Fig. 10. Cylinder seal in the name of Ebarat’s wife, Ebarat I is well known in the Ur chalcedony, unknown provenance, University of III texts, as a Šimaškian, a “Man of Durham. The king is seated on a platform, hold­ LÚ.SU.” 50 The only attestations of him ing tulips(?), and is surrounded by two squatting designated as a “King” come from Susa women, probably the queen and the queen mother. when he took control of the city. Thus, The inscription is partly erased but we can still we can suggest that it is only from read “[Eba]rat, king . [. . . . . ] [your/his]beloved this conquest and only in this context wife” (Porada 1990: pl. III, fig. 4; translation: Lam­ that he was considered as a “King.” bert 1979: 15). More­over, he did not participate in Šu Sin’s campaign, 51 and so cannot be con­ sider­ed as the “King of Šimaški.” On the contrary, we have many attestations that refer to Ebarat II as “king” (such as “King of Anšan and Susa”) and always in relation with some other Sukkalmah figures. 52 But, the inscription on the seal is clear: the Ebarat here quoted is “King of Šimaški.” A good parallel to this seal is found Fig. 11. Sealing in the name of Pala Iššan, king in the seal of Pala Iššan (Fig. 11), the of Susa, clay tablet, Susa, unknown museum. The potential second or third ruler of the seated king holds a vase with streams of water, Sukkalmah dynasty (see Fig. 3) 53 while the squatting queen faces him. A standing figure is behind him (Porada 1990: pl. III, fig. 3). (Glassner 2013: 327). The iconography 49.  He based his statement on the similarity of the inscription structure with another seal in the name of Kuk-kalla (see fig. 8d: “Ebarat, the king, Kuk-kalla son of Kuk-Šarum, servant of Šilhaha”). 50. A complete list of the Ebarat’s I mentions in the Ur III texts is given in Steinkeller 2007: 230–232. 51.  See more details above, note 32. 52.  A complete list of Ebarat’s II mentions is given in Potts 2016: 139, Table 5.5. 53.  The date of the reign of Pala Iššam is debated. The earliest studies dated it from the mid second millennium B.C.E (Scheil 1933: I–II; Cameron 1936: 229; and Hinz 1964: 150). It is now admitted that he was one of the first rulers of the dynasty just after Šilhaha and Ebarat II, so at the beginning of the

360

Clélia Paladre

of both seals is very close; thus, this seal can be more surely related to Ebarat II who is known as a contemporary of Šilhaha (De Graef 2015: 291). Finally, this Ebarat seal is clearly related to those of the so called “gouged style.” Many examples prove this relationship (Fig. 12), but the most significant seal here depicts a queen comparable to the ones on the “Ebarat’s seal,” together with a cult scene highly representative of the “gouged style” (Fig. 12c).

Fig. 12a. Cylinder seal, unknown material, unknown provenance, private collection. On the left side, a worshipper is facing the seated king. On the right side, the king is facing the squatting queen under a vine (Porada 1990: pl. III, fig. 5).

Fig. 12b. Cylinder seal, stone, unknown prov­ enance, Rosen Collection. A worshipper is facing the seated king. Behind the king, the squatting queen is shown under a vine (Porada 1990: pl. I, fig. 3a).

Fig. 12c. Cylinder seal, bitumen, Susa, Louvre Museum. A worshipper is facing the seated king. Behind the king, the squatting queen is shown under a vine (Paladre). Fig. 12. Examples of “gouged style” seals related to the “Ebarat’s seal.”

These “gouged style” seals are unfortunately not recorded from secure stratigra­ phic contexts but many elements allow a dating around the beginning of second mil­ lennium BCE. R. de Mecquenem mentioned “gouged style” seals in 1934 (233) and 1943 (129) with rare details except the fact that they were associated only with the sarcophagi tombs from the Donjon. 54 In 1986, P. Amiet detailed this establishment, second millennium BCE. But, theories about his exact place in the line of succession differ (Börker Klähn 1970: Tf. 85; de Miroschedji 1980: 134; Amiet 1986: 151; Vallat 2004; and Glassner 2013: 326–327). 54.  Contra E. Ascalone who gives mysterious details from the discovery context of these seals, mentioning R. de Mecquenem’s book from 1943 (2011: 203–205; 2013: 51; 2018: 636, 638). However, it appears that in 1943, R. de Mecquenem presents the seals discovered in sarcophagi in only two general figures: figure 95 for the seals supposedly dating from the Ur III to the Old Babylonian periods and figure 96 for seals supposedly dating from the Old Babylonian to Middle Elamite periods. The sarcophagi itself

Glyptic Art from the Ur III to the Šimaški Periods

361

mentioning that seals were present only in sarcophagi discovered below the level with the vaulted tombs (1986: 153, note 11). These vaulted tombs are attested in the level AXIV of the Ville Royale and, thus, the sarcophagi and the “gouged style” seals discovered by R. de Mecquenem can be correlated with level AXV from the Ville Royale. This is an early Sukkalmah level, directly following the Šimaški ones. It was excavated by R. Ghirshman and is clearly related to level BV and maybe with the end of level BVI (Steve and Gasche 1990: 16, 18, and 23). These seals are also coming from Operation ABC at Tal-e Malyan (see Fig. 4), and one example was discovered at Chogha Mish, on the surface of the High Mound (Kantor and Delougaz 1996: fig. C, pl. 5). The pottery and the architectural remains allow a dating of the occupation during the Sukkalmah period (Kantor and Delougaz 1996: 24). In sum, the inscription, the iconography, and the style lead us to state that “Ebarat’s seal” is the one of Ebarat’s II wife and has to be dated around 1980 BCE. Thus, it is a real “missing link” between the Šimaški seals and the Sukkalmah ones and proves that the Šimaški emancipation from the Ur III models at the extreme end of the period was initiated by Ebarat II (contra Ascalone 2010: 631–632, who attributes it to Kindattu). On the basis of this evidence, I propose to describe the genuine expressions of the “Šimaški culture” as seen in the glyptic art.

The “Gouged Style” These seals were first discovered in graves on the tell de la Ville Royale at Susa and defined by P. Amiet (1972: 239–242) as a “série populaire” due to the simplicity of style and material. During the 1970s, W. Sumner (1974: 172, fig. 12) discovered examples of this style at Tal-e Malyan. So, P. Amiet (1986: 153) proposed from this moment to call the style “anšanite” and to see its homeland in the Fars. But only few examples were found in Tal-e Malyan compared to Susa and the main material used (bitumen) cannot be found around Tal-e Malyan; rather, sources are well known around Susa. From there, I prefer to see its homeland in Susa or around, and using the appellation from J. Börker-Klähn “gouged style” (1970: 166, “Grabstilchelstil”). 55 The dating of this group differs according to researchers. The old studies work­ ing with a low chronology: 1700–1500/1400 BCE (Börker Klähn 1970: 175–177 and Seidl 1990: 130), while P. Amiet suggests a dating around 1900–1600/1500 BCE (1972: 240; 1986: 153, 158). The new tendency of research prefers a high chronology with a beginning around 2050/2000 BCE (Ascalone 2011: 203–204; 2013: 51; 2018: 636, 638; and Neumann 2013: 91). In any case, all the researchers agree on the longterm existence of this large group (almost 200 seals), but without no precision about possible internal evolutions. Very recently, E. Ascalone (2011: 206–207) suggested a chronological tri-partition of the group and added a fourth one (2018: 638) in 2018. However, he very quickly reduced the production to a simplified copy of a Mesopota­ mian model, marked by a mysterious “elamite” origin (Ascalone 2018: 636). are not precisely dated since they are presented in only one general figure, figure 64 on which all of the sarcophagi supposedly dating from the twenty-third to the seventeenth century BCE, are recorded with no other detail. 55.  Furthermore, she is the only one (followed by U. Seidl in 1990: 130) who defined this produc­ tion solely by its stylistic features without including iconographic elements or the material used. Her fundamental study was published in 1970 where she defined two different groups, the “Stillgruppe 1” and the “Stillgruppe 2” which are in fact two generations of the same production.

362

Clélia Paladre

According to me, some considerations allow to make the dating more precise. I have outlined previously the stratigraphic data from the Iranian world. Some others from the Mesopotamian world can also be mentioned here since this production is the only western Iranian one present in the west. Seals are attested mostly in IsinLarsa stratigraphic contexts: one in Lagash (Parrot 1954: 45) and three in Ešnunna (Frankfort 1955: fig. 689–691, pl. 64). 56 Some close ties are also visible between the “gouged style” group and the IsinLarsa (ca. 2000–1800 BCE), the Old Babylonian (2000–1595 BCE) and the Kültepe II production (ca. 1970–1835 BCE) (Börker Klähn 1970: 167; Amiet 1972: 241; and Ascalone 2011: 204; 2013: 51; and 2018: 638). 57 In the face of the stratigraphic and historical data and of possible stylistic comparisons, I propose to see the first development of the “gouged style” at the end of the Šimaški phase, developing into a highly recognizable style during the very beginning years of the Sukkalmah phase. Hence, its existence can be located on a short time period, from ca. 1980 to 1800/1700 BCE. Turning to the specifics of its production, we can observe an Ur III influence in the general organization of the scenes (a seated god facing a worshipper) or by many details (the cup balanced on fingertips, the presence of a padded stool, or simply the presence of a crescent moon). But the style and the composition of the “gouged style” seals are totally distinct. The composition is quite simple with the addition of the same juxtaposed elements and little or no inscription (Börker Klähn 1970: 167; Amiet 1972: 240; Neumann 2013: 86; and Ascalone 2018: 636). Indeed, the leading goddess is now absent and seems replaced by three types of elements: a snake (Fig. 13a), a goat (Fig. 13b), or a tree (Fig. 13c). Some exceptions exist but they are rare. One can also notice a rough and heavy engraving (ibid). This last point has to be emphasized since, according to me, this production seems to be the achievement of the more cursory style that I have been able to observe on the Ur III style seals from the prior periods (see above). Behind this apparent simplicity, the figures seem to illustrate themes of fertility and fecundity and to personify some concepts related to the nature. Thus, we are clearly facing more a cosmogony than a real mythology, far from the coded and defined Mesopotamian kings, gods, and goddesses. However, it is true to say that the main theme of the seals (royal or divine) is difficult to establish given the lack of attributes. I prefer to follow J. Börker Klähn and E. Porada, who identified the theme as mainly religious/cultic (Börker Klähn 1970: 167; E. Porada 1990: 174) 58; but it is not impossible, according to us, that some of it (the inscribed one for ex­ ample) could also represent a royal theme. 56.  It has to be noted that it is also attested from later period contexts: one seal was discovered in Nuzi in a twenty-fifth century BCE context (Porada 1946) and one seal was discovered in Larsa, in the palace of Nur Adad, so from an Amorrite context (Parrot 1954: 47). All these examples have to be treated as reuse items (Seidl 1990: 134, followed by Neumann 2013: 87). 57.  E. Ascalone mentioned the discovery of a Kültepe seal along with “gouged style” seals in Susa (2011: 204; 2013: 51; and 2018: 638). To prove it, he mentioned a paper published by V. Scheil in 1928 in RA 25. However, this reference is false, it is in fact in 1930 in RA 27 (this mistake was already made by P. Amiet in 1972: 241). Moreover, there is no attestation of a real Kültepe seal but just the mention of the similarity of a seal belonging to the “série populaire” (Amiet 1972: 241) with the Kültepe production. 58.  Contra Amiet 1972: 240, who prefers to see it as royal since there are no divine attributes. But he admits himself that this could also be justified by the absence of any attribute in “Elam”; and contra Ascalone 2013: 51, who links the entire corpus of the “gouged style” seals to the official power.

Glyptic Art from the Ur III to the Šimaški Periods

Fig. 13a. Cylinder seal, basalt, Susa Ville Royale, grave from the Nécropole du Donjon, Louvre Museum (Paladre).

363

Fig. 13b. Cylinder seal, bitumen, Susa, Louvre Museum (Amiet 1972: pl. 171, fig. 1920).

Fig. 13c. Cylinder seal, bitumen, Susa, Tehran Museum (Amiet 1972: pl. 168, fig. 1877). Fig. 13. Examples of “gouged style” seals.

The Place of Women This presentation of the “gouged style” and the presence of a queen on the “Ebarat seal” leads us to consider the importance of the place of women, which is clearly discernable in the Šimaški iconography at the end of the period and, then, in the Sukkalmah iconography. The place of women and the “gouged style” are closely linked through a subgroup of the “gouged style,” which shows only female iconography. For all researchers, it shows a stronger connection to the east. For P. Amiet (1986: 158–160), it is an Anšanite production; for M. J. Steve (1989: 17), it is linked to the Indus Valley; for E. Porada (1990: 176–177), it is linked to the Bactrian area; for G. Neumann (2013: 92), it is a production made by some Oxus peoples immigrants in Susa; and for E. Ascalone (2018: 638), it is an “eastern production” linked to the Oxus. Here again, the dating is debated. M.J. Steve (1989: 21–22) suggests a high chronology (ca. 2200–2000 B.C.E) without any convincing argument. G. Neumann (2013: 91) prefers a long-term use of these seals, with a dating around the twenty-­ first/twentieth and sixteeenth/fifteenth century BCE, but his argumentation is based on some epigraphic details that are not relevant to us. Finally, E. Ascalone (2018: 638) suggests a dating around 2100–1800 BCE, based on the similarity with gunagi vessels, Neo-Sumerian models, statues from the Oxus, and seals from southeast Iran. I prefer to see it as purely contemporaneous of the “gouged style” (ca. 1980–1800/1700 BCE), and, so, follow E. Porada (1990: 177), who suggests a dating around the twentieth and nineteenth century BCE, especially since no stra­ tigraphic or historic data and no relevant iconographic or stylistic similarities can be mentioned for the other possible periods.

364

Clélia Paladre

Turning to the meaning of such an iconography, it seems clear that with this kind of representation, it is the local foundation of the Šimaški culture that is visible. In form, the “importance of female figures…differentiates them [the cylinder seals] from the Mesopotamian repertory of Ur III and Old Babylonian cylinders in which women are rare and unimportant” (Porada 1990: 177). In substance, the omnipre­ sence of female iconography seems to reflect the existence of an official iconography which put forward the royal lineage and was surely the appanage of the official figures or of the royal family. 59 Indeed, even if diplomatic marriages were common throughout the Middle East, they were particularly important in southwestern Iran and reflect the necessity to keep the different political and cultural entities that composed the Šimaški state unified through the mechanism of diplomatic marriages and to establish legitimacy and secure alliances in the international sphere (Carter 2014: 49). It has to be noted that this emphasis on the female continues in the “Elamite line.” First, directly after Šimaški with the Sukkalmah dynasty and its glyptic pro­ duction (Porada 1990: pl. I, no. Ia). But also much later, during the “Middle Elamite Period,” as we can see on the Untaš Napiriša stele (Harper, Aruz, and Tallon 1992: 128, fig. 42), with the Napirasu statue (Harper, Aruz, and Tallon 1992: 133, fig. 83), or with the inscribed bead in the name of Šilhak Inšušinak where he is depicted handing the bead to his daughter Bar Uli (Sollberger 1965b: 31; Porada 1990: pl. III, fig. 1).

Conclusion I have tried here to describe and interpret the evolution of the glyptic models during the reign of the Šimaški rulers. I propose to see it as a logical evolution that begins with Mesopotamian models that were used and adapted in many ways, such as subject adaptations (with the “investiture ceremony scene”) or style adaptations (with the addition of details and/or a more cursory style), in order to finally create a new iconography and a new style (i.e. the “Ebarat wife’s seal” and the “gouged style”) at the end of the Šimaški period. Of course the Mesopotamian-style seals were still present and used, especially for internal and external political and eco­ nomic reasons (Fig. 14). By the end of the dynasty, the Šimaški rulers seem to have wanted a kind of emancipation from the old models in order to express their own nature and political systems. This resonates with the “secondary state theory” of M. Stolper (1982), who sees the Šimaški state as a creation under the pressure of Ur III aggressions. I propose to see the local Šimaški iconography as a creation emerging under the omnipresence of Ur III models. More than a “desire for emancipation,” we may have here an illustration of an “administrative reform” set up by the Šimaški rulers once the state was securely installed. Thus, the combination of new models and schemas were a necessary addition to the administrative tool kit in order to put forward the kings and their authority. In this process, the Šimaški state seems to have been an effective creator of novelties combined with old features, showing a real gift 59.  Contra Amiet 1986: 158–159; Steve 1989: 20; and Seidl 1990: 132, who suggest to see a religious thematic (for M. J. Steve the cup is the representation of the soma/haoma beverage while for P. Amiet it is the representation of the divine couple).

Glyptic Art from the Ur III to the Šimaški Periods

365

Fig. 14. Synthetic schema of the evolution of the glyptic models from the Ur III period to the Šimaški period (Paladre).

in adapt­ing and overtaking earlier models in order to support their own political purposes (internal or external). From 1980 BCE, the Sukkalmah dynasty took the Šimaški heritage, partly used it, 60 and began a phase of real apogee in the region. Thus, the Šimaški period can be seen as the real starting point for what becomes the “Elamite culture.” It can be seen as a cultural and iconographic laboratory for the next generation and, thus, it can be considered as a real cradle of the “elamite ethnogenesis.” It is interesting to mention that the two main iconographic novelties of this per­ iod (the female iconography and the “investiture ceremony scene”) show strong and deep parallels with the eastern regions (Kerman and Central Asia); especially the imagery of women, which seems to reflect a Bactrian and/or Magianian influence on the southwestern Iran as early as the reign of Puzur Inšušinak (Potts 2008). Future research may reveal some iconographic “borrowings” to the eastern regions by the people of Šimaški in order to create their own models exactly like they did with the Ur III ones.

Bibliography Amiet, P. 1972 Glyptique Susienne. Des origines à l’époque des perses achéménides. Cachets, sceaux-cylindres et empreintes antiques découverts à Suse de 1913 à 1967. Mé­ moires de la délégation archéologique en Iran 43. Paris: Librairie Orientaliste Paul Geuthner. 60. Both dynasties are fundamentally different and have their own identities. For example: the Sukkalmah dynasty has its homeland in Khuzestan where the population increases (Peyronel 2018: 210), they abandoned the utilization of year names (De Graef 2006: 68–69), etc.

366

Clélia Paladre

1973 Glyptique élamite, à propos de documents nouveaux. Arts Asiatiques 26: 3–64. 1979 Archaeological Discontinuity and Ethnic Duality in Elam. Antiquity 53: 195–204. 1986 L’âge des échanges inter-iraniens. 3500–1700 avant J.-C. Paris: Éditions de la Réunion des musées nationaux. André, B. and Salvini, M. 1989 Réflexions sur Puzur-Inšušinak. Iranica Antiqua 24: 55–72. André-Salvini, B. 2006 Puzur-Inšušinak. RlA 11/1–2: 129–131. Ascalone, E. 2010 Sigilli Paleoelamiti Dei Primi Secoli Del II Millennio A.C. I Codici Figurativi Della Glittica Paleoelamitta E Le Nuove Formule Di Espressione Dinastica Dei Sovrani Simashki. Pp. 629–639 in Proceedings of the 6th International Congress on the Archaeology of the Ancient Near East. May, 5th-10th 2008, “Sapienza” – Università di Roma, eds. P. Matthiae, F. Pinnock, L. Nigro, and N. Marchett. Wiesbaden: Harrossowitz Verlag. 2011 Glittica Elamita. Dalla metà del III alla metà del II millennio a.C. Sigilli a stampo, sigilli a cilindro e impronte rinvenute in Iran e provenienti da collezioni private e museali. Rome: “L’Erma” di Bretschneider. 2013 Sealing and Weighting: The Art of Power during the Sukkalmah Period. Iranian Archaeology 4: 50–57. 2018 Glyptic in the 4th-2nd millennium. Pp. 624–648 in The Elamite World, eds. J. Álvarez-Mon, G. P. Basello, and Y. Wicks. London and New York: Routledge, Taylor & Francis Group. Börker-Klähn, J. 1970 Untersuchungen zur altelamischen Archäologie. Ph.D. dissertation, Freie Univer­ sität Berlin. Cameron, G. G. 1936 History of Early Iran. Chicago/London: University of Chicago. Carter, E. 1978 Suse “Ville Royale”. Paléorient 4: 197–211. 2014 Royal Women in Elamite Art. Pp. 41–61 in Extraction & Control. Studies in Honor of Matthew W. Stolper, eds. M. Kozuh, W. Henkelman, C. E. Jones, and C. Woods. SAOC 68. Chicago: The Oriental Institute of the University of Chicago. Carter, E. and Stolper, M. 1984 Elam: Survey of Political History and Archaeology. Berkeley: University of Cali­ fornia Press. Collon, D. 1982 Catalogue of the Western Asiatic Seals in the British Museum. Cylinder Seals II. Akkadian, Post Akkadian and Ur III Periods. London: British Museum Press. 1986 Catalogue of the Western Asiatic Seals in the British Museum. Cylinder Seals III. Isin/Larsa and Old Babylonian Periods. London: British Museum Press. 1987 First Impressions. Cylinder Seals in the Ancient Near East. London: British Mu­ seum Publications. Daems, A. 2018 Women of Elam. Pp. 763–780 in The Elamite World, eds. J. Álvarez-Mon, G. P. Basello, and Y. Wicks. London and New York: Routledge, Taylor & Francis Group. De Graef, K. 2004 Les Noms d’année du roi simaškéen Ebarat I. Akkadica 125: 107–108. 2005 Les Archives d’Igibuni: les documents Ur III du chantier B à Suse. Mémoire de la Délégation archéologique en Iran 54. Gand: Université de Gand.

Glyptic Art from the Ur III to the Šimaški Periods

367

2006 De la Dynastie Simashki Au Sukkalmahat: Les Documents Fin PE IIb – Début PE III Du Chantier B à Suse. Ville Royale De Suse. Mémoire de la Délégation archéologique en Iran 55. Gand: Université de Gand. 2008 Annus Simaškensis. L’usage des noms d’année pendant la période Simaškéenne (ca. 1930–1880 av. notre ère) à Suse. Iranica Antiqua 43: 67–87. 2015 Part II: Chronology and Historical Geography of the 3rd Millennium. 9. Susa in the late 3rd Millennium: From a Mesopotamian Colony to an Independent State (MC 2110–1980). Pp. 289–296 in ARCANE Vol. 3. History & Philology, eds. W. Sallaberger and I. Schrakamp. Turnhout: Brepols. Desset, F. 2018 Nine Linear Elamite Texts Inscribed on Silber “Gunagi” Vessels (X, Y, Z, F’, H’, I’, J’, K’ and L’): New Data on Linear Elamite Writing and the History of the Sukkalmah Dynasty. Iran 56: 105–143. Garfinkle, S. 2013 The Third Dynasty of Ur and the Limits of State Power. Pp. 153–167 in From the 21st Century B.C. to the 21st Century A.D. Proceedings of the International Conference on Sumerian Studies Held in Madrid 22–24 July 2010, eds. S. Garfinkle and M. Molina. Winona Lake, IN: Eisenbrauns. Gasche, H. 1973 La Potterie élamite du deuxième millénaire a.C. Mémoires de la mission archéologique en Iran 47. Leiden/Paris: Brill/Geuthner. Ghirshman, R. 1968 Suse au tournant du IIIe au IIe millénaire avant notre ère. Arts asiatiques 17: 3–44. Glassner, J. J. 1996 Les Dynasties d’Awan et de Šimaški. NABU 1996/1: 25–29 no. 34. 2013 Les Premiers Sukkalmah et les Derniers Rois de Simaški. Pp. 319–328 in Susa and Elam. Archaeological, Philological, Historical and Geographical Perspectives. Proceedings of the International Congress held at Ghent University, December 14–17, 2009, eds. K. De Graef and J. Tavernier. Leiden/Boston: Brill. Harper, P., Aruz, J., and Tallon, F. 1992 Royal City of Susa: Ancient Near Eastern Treasures in the Louvre. New York: The Metropolitan Museum of Art. Henrickson, R. C. 1984 Šimaški and Central Western Iran: The Archaeological Evidence. ZA 74: 98–122. Hinz, W. 1964 Das Reich Elam. Stuttgart: W. Kohlhammer. 1972 The Lost World of Elam. London: Sidgwick and Jackson. 1973 Persia c. 1800–1550 B.C. Pp 256–288 in The Cambridge Ancient History 2/1, 3rd Edition, eds. I. E. S. Edwards, et al. Cambridge: Cambridge University Press. Hussein, A. M., Hamza, H. A., Thaher, A. K., Kadhum, S. J., Hashem, M., Taha, H. M., Altaweel, M. R., and Studevent-Hickman, B. 2010 Tell Abu Sheeja/Ancient Pašime. Report on the First Season of Excavations, 2007. Akkadica 131: 47–65. Kantor, H. J., and Delougaz, P. 1996 Choga Mish, Vol. 1. The First Five Seasons, 1961–1971. OIP 101. Chicago: Orien­ tal Institute. König, F. W. 1965 Die elamischen Königsinschriften. AfO Beih. 16. Graz: Im Selbstverlage des Herausgebers. Kupper, J.-R. 1969 Le Pays de Simaški. Iraq 31: 24–27.

368

Clélia Paladre

Kutscher, R. 1989 The Brockmon Tablets at the University of Haifa: The Royal Inscriptions. Haifa: Haifa University Press. Lambert, M. 1971 Investiture de fonctionnaires en Elam. Journal Asiatique 259: 217–221. 1972 Hutélutush-Inshushinak et le pays d’Anzan. RA 66: 61–76. 1979 Le prince de Suse Ilish-Mani et l’Élam de Naramsin à Ibi-sîn. Journal Asiatique 267: 11–40. Lambert, W. G. 1979 Near Eastern Seals in the Gulbenkian Museum of Oriental Art, University of Durham. Iraq 41: 1–46. Legrain, L. 1951 Seal Cylinders. UE 10. London: Trustees of the British Museum and of the Uni­ versity Museum, University of Pennsylvania, Philadelphia. Marchesi, G. 2013 Ur-Nammâ(k)’s Conquest of Susa. Pp. 285–291 in Susa and Elam. Archaeological, Philological, Historical and Geographical Perspectives. Proceedings of the International Congress Held at Ghent University, December 14–17, 2009, eds. K. De Graef and J. Tavernier. Leiden/Boston: Brill. Mecquenem, R. de, Contenau, G., Pfister, R., and Belaiew, N. 1943 Archéologie Susienne. Mémoires de la Délégation archéologique en Iran 29. Paris: Presses Universitaires de France. Michalowski, P. 2008 Observations on “Elamites” and “Elam” in Ur III Times. Pp. 109–123 in On The Third Dynasty of Ur. Studies in Honor of Marcel Sigrist, ed. P. Michalowski. Boston: American Schools of Oriental Research. Miroschedji, P. de. 1980 Le dieu élamite Napirisha. RA 74: 129–143. Mofidi-Nasrabadi, B. 2009 Aspekte der Herrschaft und der Herrscherdarstellungen in Elam im 2. Jt. v. Chr. AOAT 356. Münster: Ugarit-Verlag. Morgan, J. de. 1896 Mission scientifique en Perse. Recherches archéologiques IV, 1ère partie. Mission scientifique en Perse 1–4, 1894–1905. Paris: E. Leroux. Neumann, G. 2013 Elams Kulturkontakte mit seinen Nachbarn im Spiegel der Glyptik des 2. Jahrtausends v. Chr. Pp. 83–128 in Susa and Elam. Archaeological, Philological, Historical and Geographical Perspectives. Proceedings of the International Congress held at Ghent University, December 14–17, 2009, eds. K. De Graef and J. Tavernier. Leiden/Boston: Brill. Parrot, A. 1954 Glyptique mésopotamienne. Fouilles de Lagash (Tello) et de Larsa (Senkereh). Paris: Geuthner. Peyronel, L. 2018 The Old Elamite Period. Pp. 203–231 in The Elamite World, eds. J. Álvarez-Mon, G. P. Basello, and Y. Wicks. London and New York: Routledge, Taylor & Francis Group. Pittman, H. 2018 Origins of Monumental Sculpture in Elam: Two Case Studies. Pp. 585–601 in The Elamite World, eds. J. Álvarez-Mon, G. P. Basello, and Y. Wicks. London and New York: Routledge, Taylor & Francis Group.

Glyptic Art from the Ur III to the Šimaški Periods

369

Porada, E. 1946 The Origin of Winnirke’s Cylinder Seal. JNES 5: 257–259. 1990 More Seals of the Time of the Sukkalmah. RA 84: 171–181. Porada, E. and Buchanan, B. 1948 Corpus of Ancient Near Eastern Seals in North American Collections I, the Collection of the Pierpont Morgan Library. The Bollingen Series 14. Washington: Pantheon Books. Potts, D. T. 1999 The Archaeology of Elam. Formation and Transformation of an Ancient State. Cambridge: Cambridge University Press. 2008 Puzur-Inšušinak and the Oxus Civilization (BMAC): Reflections on Šimaški and the Geo-political Landscape of Iran and Central Asia in the Ur III Period. ZA 98: 165–194. 2016 The Archaeology of Elam. Formation and Transformation of an Ancient Iranian State. 2nd revised ed. Cambridge: Cambridge University Press. Sallaberger, W. and Schrakamp, I. 2015 Part I: Philological Data for a Historical Chronology of Mesopotamia in the 3rd Millennium. 2. Sources I: King Lists and Related Texts. Pp. 13–32 in ARCANE Vol. 3. History & Philology, eds. W. Sallaberger and I. Schrakamp. Turnhout: Brepols. Scheil, V. 1900 Textes élamites-sémitiques (Première série). MDP 2. Paris: E. Leroux. 1901 Textes élamites-anzanites (Première série). MDP 3. Paris: E. Leroux. 1913 Textes élamites-sémitiques (Cinquième série). MDP 14. Paris: E. Leroux. 1925 Passim. RA 22: 141–162. 1926 Raptim. RA 23: 35–48. 1931 Dynasties élamites d’Awan et de Šimaš. RA 28: 1–8. 1932 Actes juridiques susiens. MDP 23. Paris: E. Leroux. 1933 Actes juridiques susiens; inscriptions des Achéménides. MDP 24. Paris: E. Leroux. Seidl, U. 1990 Altelamische Siegel. Pp. 129–135 in Mélanges Jean Perrot, ed. F. Vallat. Paris: Editions Recherche sur les Civilisations. Sollberger, E. 1965a Three Ur-Dynasty Documents. JCS 19: 26–30. 1965b A New Inscription of Šilhak-Inšušinak. JCS 19: 31–32. Southall, A. 1956 Alur Society. Cambridge: Heffer and Sons. Steinkeller, P. 1987 The Administrative and Economic Organization of the Ur III State: The Core and the Periphery. Pp. 19–41 in The Organization of Power: Aspects of Bureaucracy in the Ancient Near East, eds. McG. Gibson and R. D. Biggs. SAOC 46. Chicago: Oriental Institute. 1988 On the Identity of the Toponym LÚ.SU(.A). JAOS 108: 197–202. 1989 Sale Documents of the Ur-III-Period. Stuttgart: Franz Steiner Verlag Wiesbaden GMBH. 2007 New Light on Šimaški and Its Rulers. ZA 97: 215–232. 2013 Puzur-Inšušinak at Susa: A Pivotal Episode of Early Elamite History Recon­ sidered. Pp. 291–317 in Susa and Elam. Archaeological, Philological, Historical and Geographical Perspectives. Proceedings of the International Congress held at Ghent University, December 14–17, 2009, eds. K. De Graef and J. Tavernier. Leiden/Boston: Brill.

370

Clélia Paladre

2014 On the Dynasty of Simashki: Twenty Years (or so) After. Pp. 287–296 in Extraction & Control. Studies in Honor of Matthew W. Stolper, eds. M. Kozuh, W. Hen­ kelman, C. E. Jones, and C. Woods. SAOC 68. Chicago: The Oriental Institute of the University of Chicago. 2018 The Birth of Elam in History. Pp. 177–202 in The Elamite World, eds. J. ÁlvarezMon, G. P. Basello, and Y. Wicks. London and New York: Routledge, Taylor & Francis Group. Steve, M. J. 1989 Des sceaux cylindres de Simashkis? RA 83/1: 13–26. Steve, M. J. and Gasche, H. 1990 Le Tell de l’Apadana avant les Achéménides: Contribution à la topographie de Suse. Pp. 15–60 in Mélanges Jean Perrot, ed. F. Vallat. Paris : Éditions Recherche sur les Civilisations. Steve, M. J., Gasche, H., and De Meyer, L. 1980 La Susiane au Deuxième millénaire: à propos d’une interprétation des fouilles de Suse. Iranica Antiqua 15: 49–160. Steve, M. J., Vallat, F., Gasche, H., Jullien, C., and Jullien, F. 2002 s.v. “Suse”. Pp. 359–652 in Supplément au Dictionnaire de la Bible, Fascicule 73. Paris: Létouzey et Ané. Stolper, M. 1982 On the Dynasty of Šimaški and the Early Sukkalmahs. ZA 72: 42–67. Sumner, W. 1972 Tall-i Malyan. Iran 10: 176–177. 1974 Excavations at Tall-i Malyan 1971–1972. Iran 12: 155–180. 1976 Excavations at Tall-i Malyan 1974 Iran 14: 103–115. 1989 Anshan in the Kaftari Phase: Patterns of Settlement and Land Use. Pp. 135–161 in Archaeologia Iranica et Orientalis – Miscellanea in Honorem Louis Vanden Berghe Volume I, eds. L. De Meyer and E. Haerinck. Leuven: Peeters. Vallat, F. 1980 Suse et l’Elam. Paris: Recherche sur les grandes civilisations. 1985 Éléments de géographie élamite (Résumé). Paléorient 11: 49–54. 1991 La Géographie de l’Elam d’après quelques textes mésopotamiens. Pp. 11–21 in Mésopotamie et Elam, eds. L. De Meyer and H. Gasche. History and Environment Occasional Publications 1. Ghent: University of Ghent. 1993 Les Noms géographiques des sources suso-élamites. RGTC 11. Wiesbaden: L. Reichert. 1996 L’Élam à l’époque paléo-babylonienne et ses relations avec la Mésopotamie. Pp. 297–319 in Amurru: actes du colloque international (Paris, mai 1993); 1: Mari, Ébla et les Hourrites, dix ans de travaux, première partie, ed. J.-M. Durand. Paris: Éditions Recherche sur les Civilisations. 2004 Le cylindre de Hute-kazan et la chronologie des premiers sukkalmah. Akkadica 125/2: 135–140. Zadok, R. 1987 Peoples from the Iranian Plateau in Babylonia during the Second Millennium BC. Iran 25: 1–26. 1991 Elamite Onomastics. Studi Epigrafici e Linguistici sul Vicino Oriente 8: 225–237.

The Status of Real Estate in Neo-Babylonian Ur: The Case of the Gallābu Family Olga V. Popova

Institute of Linguistics of the Russian Academy of Sciences Among the Neo-Babylonian texts of the first millennium BCE found at Ur is one particular group: the family archive of the descendants of Gallābu, “the Barber’s archive” (Jursa 2005: 133–134). The archive was found during the excavations conducted by L. Woolley in Ur 1 in house number six (Woolley and Mallowan 1962: 48) and published in copies in 1949 by H. H. Figulla (UET 4). The Gallābu archive consists of forty-five legal and administrative texts. 2 Some texts are not related to the family prosopographically and often can be considered retroacts; they are contracts concerning the history of the property before its acquisition by the Gallābu family. 3 Chronologically the Gallābu archive extends for a period of 260 years: from year twenty-nine of Nebuchadnezzar II (576/575 BCE) until year seven of Philip III Arrhidaeus (316 BCE). One can reconstruct the family tree of the Gallābu family and reckon up nine generations including “Generation 0,” which refers to Hašdiya, a member of the family who appears only as an ancestral reference. 4 The Gallābu archive goes from the Neo-Babylonian period, covers all of the Achaemenid period, and ends at the beginning of the Hellenistic period, and in this sense it represents a unique case of such a long continuous private archive. We can observe the family activities during three different political regimes and state that it is only the admin­ istrative systems that change and not the social reality. The status of the family remains the same: they are members of the urban elite, which keep their home and properties, continue to hold agricultural land granted by the Crown, and fulfill prestigious functions in the temple of the god Sîn. 1.  The city of Ur was excavated by L. Woolley in 1922–1934. 2.  The Gallābu archive is the archive of the series U.17243. The series consists of fifty-one items. Eight tablets remain unpublished because they are useless fragments: U.17243 25, 29, 30, 34, 41–42, 45, and 47. Two texts, which clearly belong to the archive, have no inventory number in the series U.17243: these are texts UET 4 24 and 60. Thus, there are forty-five texts in the archive. 3.  Certainly, UET 4 12, 14, 16, and 19 can be considered as retroacts. J. Oelsner (2006: 83) proposes to consider as retroacts also UET 4 18, 29, and 30. I propose to include also UET 4 34 in this category. 4.  My reconstruction of these nine (0-VIII) generations of the Gallābu family appeared in my PhD thesis defended in 2018 (in preparation for publication). Earlier, reconstructions were made by J. Oelsner (2006: 80–83) and G. van Driel (1987: 178). The concept of “generation 0” was introduced by J. Oelsner (2006: 80 n. 3).

371

372

Olga V. Popova

The Gallābu are not too large of a family. They manage their heritage as much as possible in community, especially land and houses, and establish some of their social assets by belonging to the staff of the temple. With the profits or income allowed by their implantation in such a large organization, they buy land outside the town as soon as it is possible. Thus, one group of the Gallābu documents deals with the real estates of the Gallābu family. This is the houses and land belonging to them. In this article, I will concentrate on the land of the Gallābu. The archive has eighteen documents concerning land property: UET 4 5, 12, 14, 16–19, 41–44, 52–53, 60, 106, 193–194, and 205. These texts extend from year eight of Cyrus the Great (531 BCE) until year seven of Philip III Arrhidaeus (316 BCE) and permit one to observe the creation of the family’s patrimony. The agricultural land of the Gallābu family consists of orchards with palm-trees and of arable lands mentioned in some management contracts. When we consider the creation of the Gallābu’s patrimony, we see that one of the most remarkable points is the status of that agricultural land. We find several private property titles along with their retroacts and the cadastral characterization of these lands, but some indications, on the other hand, refer to the same orchards as the property of a religious institution, the temple of Sîn at Ur, in the beginning. Then, during the second part of the Achaemenid period, we find the royal admin­ istration in charge of managing the fiefs called bīt qašti, “bow estate.” 5 I propose to examine whether and how the documents of the Gallābu archive related to the management of the family’s patrimony enable us to reconsider the changing status of land properties in Neo-Babylonian Ur.

Land Properties of the Gallābu Family By means of toponymy and prosopography, a study of these eighteen texts allows us to distinguish three agricultural areas belonging to the family: 1.  The land mentioned in the texts UET 4 12, UET 4 16 (two retroacts), UET 4 17 (via prosopographical supposition), UET 4 52, UET 4 53, and UET 4 205, as located “on the banks of the canal Ḫarru-ša-Bēl-iddin in the ugāru-area 6 of the tamirtu Šarīgu, in the region of Ur” (aḫi nār Ḫarri-ša-Bēl-iddin ugār tamirtu Šarīgu pīḫāt Uri). 7 2.  The land located “on the banks of the canal Gugallu in the ugāru-area of the tamirtu Makkē, 8 in the region of Ur” (aḫi nār Gugallu ugār tamirtu Makkē pīḫāt Uri) and mentioned in the texts UET 4 193, UET 4 53, and UET 4 106. 3.  A group of several agricultural lands, located “on the banks of the canal Ḫarru-ina-aḫi-Uri in the ugāru-area of the town Madumē, in the region of Ur” (aḫi nār Ḫarri-ina-aḫi-Uri ugār āl Madumē pīḫāt Uri) in the texts 5.  Bīt qašti are lands of Crown property which were attributed to the communities which cultivated these lands as family plots. The holder of this land was to provide the Crown with a warrior in case of need and to pay an ilku-tax to the King. These lands were consequently called bīt qašti, ‘bow estate,’ when landholder procured an archer, bīt sisī, ‘horse estate,’ or bīt narkabti, ‘chariot estate.’ 6.  The term ugāru refers to an agricultural zone located near the watercourse. 7.  For all the attestations see RGTC 8: 354. 8.  The toponym is not indicated in RGTC 8. H. H. Figulla cites it as Makku (UET 4 59). The copy is not very clear whether the name of the tamirtu is Makkē or Bakkē; one can suppose the etymological connection with the Akk. baqqu “gnat.”

The Status of Real Estate in Neo-Babylonian Ur

373

UET 4 41, UET 4 42 and “on the banks of the canal Ḫarru-ina-aḫi-Uri in the ugāru-area of the tamirtu Mamar, in the region of Ur” (aḫi nār Ḫarriina-aḫi-Uri ugār tamirtu Mamar pīḫāt Uri) in UET 4 44. 9

All the locations are in the region of Ur, outside the city, but their exact locations are unknown.

First Area: Land on the Banks of the Canal Ḫarru-ša-Bēl-iddin In year twenty-five of Darius I (UET 4 17), Iqīšāya/Sîn-iddin//Gallābu and Šamaš-iddin/Sîn-iddin//Gallābu of Generation III (early fifth century), buy an orchard with date-palms, that is located inside the town in front of the Lugal-girra Gate (ana meḫret abul Lugal-girra ša qereb Uri). The land’s size is about 2,077 m² and the price is twenty shekels of silver. Thereafter, this orchard is not mentioned again. But a prosopographical study of the neighbors allows one to link this text with UET 4 16, dating to the reign of Cyrus—one of the neighbors in UET 4 17, Ummu/Rēmūt-Gula//Etellu, seems to be the brother of Sîn-zēr-līšir/Rēmūt-Gula// Etellu who purchases the land during the reign of Cyrus in UET 4 16—and suppose that this date-palm orchard could be related to another orchard, located on the canal Ḫarru-ša-Bēl-iddin. This land appears in UET 4 52 during year five of Xerxes at the time of the division of inheritance between Iqīšāya/Sîn-iddin//Gallābu and Šamaš-iddin/ Sîn-iddin//Gallābu of Generation III. This document refers to an orchard and an arable land located on the banks of the canal Ḫarru-ša-Bēl-iddin in an ugāru-area of the tamirtu Šarīgu. It remains difficult to draw the exact plan of the whole because the orientations are not given, but we find that each of the two brothers inherits a field located between two bigger plots of land in Šarīgu with common neighbors. The parts owned by the brothers are as follows: the first part of Iqīšāya measures 13,400 m², the second part of Iqīšāya measures 5,838 m², the first part of Šamaš-iddin measures 10,000 m², the second part of Šamaš-iddin (neighboring the canal Ḫarru-ša-Šumāya 10) measures 8,000 m². The first piece of the land has a size of about 1.4 ha and the second of about 2.6 ha. The first brother receives 1.92 ha, and the second 1.80 ha. The allotment of the shares is then not equivalent to the actual distribution of the plots of land. The shares are probably equal, but the dimensions remain approximate. In terms of Babylonian measurements, the orchard cited represents about three kurru of land: it amounts to the average real estate for a family’s patrimony. This property, or rather this set of properties, was purchased much earlier and we have several previous property titles (retroacts), which date from year eight of Cyrus (UET 4 16) and year seventeen of Darius I (UET 4 12). A part of this land mentioned in UET 4 52 adjoins the Gallābu’s land called qaqqar qašti “bow land” (UET 4 52: 37). 11 Since the other mentions of this land present it as a private prop­ 9.  For all attestation of the canal Ḫarru-ina-aḫi-Uri see TGTC 8: 353; for Madumē RGTC 8: 216; for Mamar RGTC 8: 219. 10.  The canal Ḫarru-ša-Šumaya is a hapax. Judging by the name, it presents the second example of a canal due to a personal initiative, as the canal Ḫarru-ša-Bēl-iddin. 11.  The most common term for “bow estate” is bīt qašti. The appellation qaqqar qašti, literally “bow land,” is probably just a rarer and earlier synonym (see for instance in YOS 19 70 dated to the reign of Nabonidus [Jursa 2010: 200 n. 1140]).

374

Olga V. Popova

erty, it permits one to suppose that the Gallābu family had in this location their private lands as well as lands given to it for a military obligation. The same land is also mentioned in two texts dating to the time of Generation V (last quarter of the fifth century): UET 4 205, a transaction document, and UET 4 53. The last one is a document of division of inheritance (one house and two lots of land) during year seven of Darius II. This is a division between Nidinti-Sîn/ Sîn-ahhē-iddin//Gallābu, Sîn-iddin/Sîn-ahhē-iddin//Gallābu and Sîn-balāssu-iqbi/ Sîn-ahhē-iddin//Gallābu of Generation V. The area of one brother’s (Sîn-balāssuiqbi/Sîn-ahhē-iddin//Gallābu) part of this land is 1,500 m². This part is small compared with the 1.92 ha land in this area that belonged to his grandfather Iqīšāya/ Sîn-iddin//Gallābu of Generation III. This is probably due to different land divisions within the family, but it is difficult to judge the ensemble because the other measure­ ments are not preserved in the tablet. To sum up, this land appears in the documents of generations III and V and in two retroacts. We have several private property titles along with their retroacts, but in UET 4 52 a part of probably the same orchard is defined as a qaqqar qašti “bow land.” Thus, in this location the Gallābu family has a patrimony of several plots of land and a part of this land appears as a land of the royal service.

Second Area: Land on the Banks of the Canal Gugallu The second property is formed on the banks of the canal Gugallu in an ugāruarea of the tamirtu Makkē, in the region of Ur, 12 and is mentioned in UET 4 193, UET 4 53, and UET 4 106. For the first time, this land is mentioned in UET 4 193, which dates to year thirteen of Artaxerxes I, and represents a precision of the division of inheritance, which occurred previously during the year twenty-one of Xerxes. Here appears Ninazu-uballiṭ/Iqīšāya//Gallābu, Balāṭu/Iqīšāya//Gallābu, Bulluṭ/Iqīšāya//Gallābu, and Sîn-aḫḫē-iddin/Iqīšāya//Gallābu of Generation IV of the Gallābu family (middle of the fifth century). The text deals with a date-palm orchard and arable land. The condition of the tablet does not allow one to determine the size of the property. The next mention of this land is during year seven of Darius II in the above-­ mentioned text UET 4 53, another document of the inheritance division. Sînbalāssu-­iqbi/Sîn-aḫḫē-iddin//Gallābu of Generation V, the one who got 1,500 m² on the banks of the canal Ḫarru-ša-Bēl-iddin, receives one part of the date-palm orchard on the banks of the canal Gugallu in the ugāru-area of the tamirtu Makkē, in the region of Ur. The area of his part is 4,250 m². This land appears in these two documents as a private property. However, the text UET 4 106, which concerns an agricultural tax, šibšu, and is dated to year forty-two of Artaxerxes II, shows that one plot of agricultural land in the same region is a bīt qašti owned by a group of bēlū qašti among whom are Kuṣurēa/ Sîn-aḫḫē-uballiṭ//Gallābu and Anapiēa/Sîn-iddin//Gallābu of Generation VI (first half of the fourth century). Thus, this land appears in the documents of generations IV, V, and VI. In the first documents (UET 4 193 and 53), it appears as a private property and is not mentioned as bīt qašti. One can suppose that either the land changed status and 12.  RGTC 8 371; see also UET 4 47 for the canal Gugallu and UET 4 106 for the ugāru-area of the tamirtu Makkē.

The Status of Real Estate in Neo-Babylonian Ur

375

became a bīt qašti by the time of the reign of Artaxerxes II (UET 4 106) as result of state confiscation, or that the family acquired or received a bīt qašti next to their own lands, as in the case of the lands on the banks of the canal Ḫarru-ša-Bēl-iddin.

Third Area: Land on the Banks of the Canal Ḫarru-ina-aḫi-Uri The third group is formed by several plots of agricultural land, located on the banks of the canal Ḫarru-ina-aḫi-Uri in a town called Madumē (UET 4 41, UET 4 42) and in the ugāru-area of the tamirtu Mamar (UET 4 44). In these documents we observe a group of individuals managing and leasing bīt qašti lands located in the town Madumē 13 (UET 4 41, UET 4 42) and another in the ugāru-area of the tamirtu Mamar (UET 4 44). These documents concern day-to-day agricultural management. Between year twenty-seven and year forty-two of Artaxerxes II, several agricultural lands, located in a town called Madumē and in the tamirtu Mamar, were rented by a group of co-owners, including members of the Gallābu family of Generation VI (Kuṣurēa/Sîn-aḫḫē-uballiṭ//Gallābu and Anapiēa/Sîn-iddin//Gallābu) who held them as a feudal tenure called bīt qašti. The tenants who rented the land did not use it ana sūti, i.e. for a share of the crops, but ana mandatti, i.e. for the payment of a fixed annual rent in silver instead of grain. Interestingly, these lands are qualified as bīt qašti but at the same time as makkūr Sîn “property of Sîn” (níg.ga d30 é lúban ša PN). We can deduce that the temple lands were taken over by the Crown and turned into bīt qašti given into private hands (van Driel 1987: 166). In some way, this was a taxation imposed on temple lands by the Crown. This plot of land appears only during Generation VI of the Gallābu family (first half of the fourth century) and only as a bīt qašti of someone and a makkūr Sîn. The text UET 4 42 demonstrates clearly how that system functioned. The bīt qašti is registered to the name of a certain Ša-Sîn-bāni/Sîn-kaṣir, but he is ina pān of the group of seven bēlū qašti (holders of the bow land), probably with a certain Aḫu-šunu/Šamaš-iddin in charge of it. This group does not cultivate the land themselves (indeed, the archive presents no documents concerning the commerce of dates or barley by the Gallābu) but subleases it for ten shekels a year to a certain Sîn-iddin/ Sîn-aḫḫē-ušallim by mandattu-contracts. 14 We can see that they are not mentioned as the owners in the title of these lands (they are only holders of bīt qašti), neither are they the occupants registered on this land, but they are an operating society responsible for finding a cultivator for a certain sum of money. We can observe that at Ur the group of holders of the bow-land could be formed by either family members, managing the adjoining lots together, or neighbors or socio-professional groups cooperating between themselves. They operate as owners in the sense that they can buy, purchase, leave as inheritance, and lease these lands, but they always have some obligations to fulfill to the Crown, which reserves an eminent property right on these lands.

13.  Another mention of a land on the banks of the canal Ḫarru-ina-aḫi-uri in the town of Madumē is found in UET 4 32, dated to the reign of Šamaš-šum-ukīn (middle of the seventh century BCE). 14.  The term mandattu commonly refers to a “quitrent,” the payment regularly paid by the slaves to their master in return for the permission to work independently. The masters in these cases had a regular income instead of the direct benefit of the slave’s labor (Jursa 2010: 682, 699). In texts from Ur, the term refers to a fixed annual rent.

376

Olga V. Popova

Conclusion 1.  The texts of the Gallābu archive permit one to distinguish two consecutive distributions of bīt qašti in this region. We can date the first one to the reign of Darius I, and here UET 4 60 is of help. It is about an agreement concerning the obligations imposed on agricultural stubble land retained as a bīt qašti. Sîn-aḫḫē-uballiṭ/Balāṭu//Gallābu and Sîn-ušallim/Balāṭu// Gallābu of Generation V accept binding obligations on two bīt qašti previously accepted by Iqīšāya/Sîn-iddin//Gallābu and Šamaš-iddin/Sîn-iddin// Gallābu of Generation III. This document dates back to year one of Artaxerxes II and does not mention the exact location of the land, but provides a clear reference to the first distribution into private hands of bīt qašti in this region, which dates back to Generation III, the reign of Darius I (even if the system had existed before). The second attribution probably takes place in the time of Generation VI, the reign of Artaxerxes II. Thus, the lot of lands on the banks of the canal Ḫarru-ina-aḫi-Uri, characterized as bīt qašti, appears only in Generation VI, and it shows perhaps a new attribution of bow estates in the region. This is also confirmed by text UET 4 106 where in year forty-two of Artaxerxes II one bīt qašti appears in the area of the second lot of lands located on the banks of the canal Gugallu. 2.  One can observe a significant expansion of the Gallābu holdings until Generation III of the family (early fifth century). After that, the Gallābu have a “conservative” property management strategy, using the term of M. Jursa (2010: 180), i.e. the family probably manages and maintains their estate without buying a lot of new lands (at least we have no indication of this in the documents). The major changes at this stage are caused by inheritance divisions. Apparently, the Gallābu’s holdings were limited to three gardens, i.e. the Gallābu were an average family in high urban society. 3.  These three plots of land of the Gallābu patrimony show that in the Gallābu archive there are patrimonial lands and bīt qašti land, the latter probably of religious origin (lands under the control of the temple transformed to land under royal control). All this is managed in the extended family framework (we rarely have only one holder), or in a socio-professional association that subleases these lands by mandattu-contracts. This joint management is probably connected with the cost of operating equipment, which results in the shared use of as much equipment as possible. 4.  The agricultural land of the Gallābu family consists mainly of orchards with palm-trees (and a smaller amount of arable land) which were purchased before the reign of Xerxes, and that are mentioned later in some management contracts. On the one hand, we find several private property titles concerning these lands, but, on the other hand, we also find some indications, that refer to the same orchards as the property of a religious institution, the temple of Sîn at Ur. Thus, in the beginning the land belonged mostly to the temple of Sîn, which granted its use to some members of the family in accordance with prebendary functions performed by them in the temple, and the Gallābu ensured the exploitation of this land in a wider family framework. However, during the second part of the Achaemenid period, we find the royal administration in charge of managing the bīt qašti held by the Gallābu. 5.  The temple lands were probably taken over by the Crown, turned into bīt qašti land, and given into private hands. The transformation of the temple land into bīt qašti land should not be considered as a change of owner, be-

The Status of Real Estate in Neo-Babylonian Ur

377

cause the land is still characterized as makkūr Sîn “property of Sîn.” It was a de facto taxation imposed on the temple land by the Crown. 6.  We can conclude that there are thus two parallel property rights. There did exist a private sphere of real estate with proper contracts in the first millennium BCE Babylonia. But some of these lands were subject to an eminent property right, according to which a large organization such as an urban temple was in charge of the land. Thus, it seems that the land could somehow, probably by creation of bīt qašti land by the Crown, move from the institutional sphere into private hands, provided that these private owners remained liable for some specific obligations.

Bibliography Beaulieu, P.-A. 2000 Legal and Administrative Texts from the Reign of Nabonidus. YOS 19. New Haven: Yale University Press. Driel, G. van. 1987 Continuity or Decay in the Late Achaemenid Period: Evidence from Southern Mesopotamia. Pp. 159–181 in Achaemenid History I: Sources, Structures and Synthesis, eds. A. Kuhrt and H. Sancisi-Weedenburg. Leiden: Nederlands Instituut voor het Nabije Oosten. Figulla, H. H. 1949 Business Documents of the New-Babylonian Period. UET 4. London: Trustees of the British Museum and of the University Museum, University of Pennsylvania, Philadelphia. Jursa, M. 2005 Neo-Babylonian Legal and Administrative Documents. Typology, Contents, and Archives. GMTR 1. Münster: Ugarit-Verlag. 2010 Aspects of the Economic History of Babylonia in the First Millennium BC. AOAT 377. Münster: Ugarit-Verlag. Oelsner, J. 2006 Zu spätbabylonischen Urkunden aus Ur und dem Archiv der Familie Gallābu ‘Barbier.’ Pp. 75–87 in Recht gestern und heute. Festschrift zum 85. Geburtstag von Richard Haase, eds. J. Hengstl and U. Sick. Wiesbaden: Harrassowitz. Popova, O. V. 2018 Étude d’une archive d’une famille de notables de la ville d’Ur du VIe au IVe siècle av. J.-C. L’archive des Gallābu. Ph.D. dissertation, Université Paris 1 PanthéonSorbonne, Paris. Woolley, L. and Mallowan, M. 1962 The Neo-Babylonian and Persian Periods. UE 9. London: Trustees of the British Museum and of the University Museum, University of Pennsylvania, Philadelphia. Zadok, R. 1985 Geographical Names According to New- and Late-Babylonian Texts. RGTC 8. Wiesbaden: L. Reichert.

Palace and “Private” Sectors in Old Babylonian Ur Mitchell S. Rothman Penn Museum

Introduction My topic for this paper is the state. What is its nature, and how do developed states like those of the Isin-Larsa and Old Babylonian periods differ from archaic states? As a corollary to these more general issues, I am asking whether the terms “private” and “elite” are heuristically useful in our attempts to make sense of early states. My data set comes from the second millennium BCE states of the southern alluvium of Mesopotamia, especially Ur at the time of the last Isin-Larsa dynasties and Hammurabi’s conquest and incorporation of the region into his empire. I will be highlighting one key datum, the message content of seals or glyphs as a representation of the mental map of how various players were integrated into ancient states.

The State The understanding of state by many neo-evolutionists and historians is one that tends to emphasize the administrative and political core of the state (Wallace 1971; Wright and Johnson 1995; Feinman and Marcus 1998). An alternate view presented by Julian Steward (1955) and more recently by Anne Porter (2012) is that the state is the whole society, of which the centralized institutions are just one, albeit an important part. It makes sense that those who wanted to understand its origins would focus on the functioning of newly evolved central institutions, but understanding that alone would be like saying one should study only the city center to understand urban systems. The state is by definition a heterogeneous society, fundamentally different from more homogeneous communally or kinship organized societies. Julian Steward (1955) emphasized that means of integration of all the different constituencies of the state, therefore, was as important as documenting their central institutions’ strategic efforts. If one thinks of the challenges of the Old Babylonian Mari state, for example, integration of the many mobile pastoral groups was as important as the control of the city that defined it (Porter 2012).

379

380

Mitchell S. Rothman

The Mesopotamian State As this developmental process is evolutionary, starting in pre-state times is useful for this exploration. The state or actually many kinds of states in ancient Mesopotamia tell a story of origin and elaboration. At places like prehistoric Tepe Gawra, centralized storage and communal efforts at production and long-distance exchange had created a need for coordinating the larger populace of the piedmont and organizing access to many kinds of goods from tools to wheat, cloth, and exotic materials already in the late fifth millennium BCE (Rothman 2002). By phase XI of level XI/XA at 3900 BCE seals or glyphs (more on that term in “Language of Glyphs” below) and especially sealings were concentrated in workshop, administrative, and temple, that is, non-domestic contexts (Fig. 1). Along with economic and political interactions, temple worship integrated the residents of the site and the larger polity (Rothman 2002). By Gawra’s last prehistoric level VIII at about 3600 BCE a system of centralized distribution by those in what I called the comptroller’s house had evolved in a small center now with few private houses (Fig. 2). Most researchers would concur that by 3500 BCE the first state-level organi­ zations emerged in the southern alluvium of Mesopotamia. Cities like Uruk-Warka encompassed central institutions within the Eanna district, but also required crafts persons and those who fed the city with plant and animal foods, as well as those

Fig. 1. Tepe Gawra: Distributions of seals and glyphs, Phase XI, Level XI/XA (after Rothman 2002: Figure 5.80).

Palace and “Private” Sectors in Old Babylonian Ur

381

Fig. 2. Tepe Gawra: Distributions of seals and glyphs, Phase VIIIB, Level VIII (after Rothman 2002: Figure 5.35).

who provided labor and raw materials for craft production and making symbols for status signification. By 3000 BCE, this state system had elaborated further. For controlling access and more essentially for auditing transactions tablets and writing became an important technology. One gets the impression from the tablets that the central institutions were very much inward looking and somewhat isolated from the larger populous. This and the use of early tablets by central institutions led scholars like Falkenstein (1974) to propose a kind of monolithic, theocratic state in the Early Dynastic period. However, Oppenheim (1974: 111) and others cited texts describing a city assembly, which was independent of the palace or temple. Perhaps this is why the palace and temple’s financing evolved not primarily through tribute, as other states had, but through ownership of their own great agricultural estates. By the end of the third millennium BCE the nature of the state had elaborated further. As territorial states and empires evolved under Sargon and later Hammurabi, the necessity of integrating even more diverse groups and larger territorial expanses changed the strategies of the palace and temple institutions, and the practices of many other groups within the state. The ownership of the great estates continued, while the involvement of the central institutions into non-palace affairs, especially in conquered lands, increased. The archive of Šamaš-ḥazir, Hammurabi’s land registrar for his southern conquests including Ur, shows the king’s interest in land-­ ownership outside of the palace realms (Rothman 1994a). At the same time, the long

382

Mitchell S. Rothman

Fig. 3. Distribution of grain fields and date orchards Old Babylonian Larsa/Ur (after Rothman 1994a; Figure 5).

debate over whether the Old Babylonian merchants (tamkārum) were private entrepreneurs or palace dependents (Leemans 1950; Charpin 2015) reveals a problem with the commonly held bi-modal model of the state as two sectors: “private” and palace/temple. For example, in my research on agricultural decision-making at Old Babylonian Larsa from an archive including Ur (Rothman 1994a), the texts suggest that the so-called private sector emphasized the more reliable date crop over grains (Fig. 3). The palace and temple tried to get as much return as possible through industrial-like grain production, pushing the land beyond its limits and ultimately destroying many fields through salinization of the soil (Jacobsen 1982). Yet many of these private landowners received land grants from the temples and the palace, as Stone documented at Nippur (Stone 1987) and I at Larsa and Ur (Rothman 1994a), and were part of a system of providing labor for large estates (Ellis 1976). I am hypothesizing here that the Old Babylonian state had changed dramatically in the number of different constituencies (increasing heterogeneity) and the way it was integrated from its earlier forms. Rather than the archaic state, in which central institutions contrasted sharply from the larger pre-existing social networks, the Old Babylonian state was a much more highly integrated society. How can we measure these changes? I propose here to show that glyptique is one, only one, of many datasets that may clarify this changing integration.

The Language of Glyphs To make my case here, I am taking a leaf from earlier studies like those of Winter (1987), Pittman (1993, 1994), Scott (2005), Charvát (2005), and others that there are patterns in the subjects and the organization of seal designs that represent a kind of symbolic language comprehensible to the largely illiterate population of

Palace and “Private” Sectors in Old Babylonian Ur

383

ancient Mesopotamia. I am asserting that this language and its messages correlate with changes in state integrative structures. A good way to start, as I did some years ago (Rothman 1994b), is to borrow from Dell Hymes’ Ethnography of Speaking methodology (Hymes 1962, 1964). He proposed that not only the words, but the speech act itself conveyed important cultural information. Hymes argued that the interpretive significance of speech acts depends on: the audience(s), and the social identities of sender(s) and receiver(s). They vary according to (1) the code (how the message is conveyed), (2) what topics are discussed, (3) the channel (through what means messages are conveyed: written, spoken in person, drums, phone, text), and (4) the messages’ form. The social setting (formal, informal) and context (the broader social structures at play in relations of, for example, ethnic groups, genders, friends, kin, authorities versus subjects, different occupational groups, etc.) also structure the speech act. I think the same applies to unspoken symbols as well. To interpret designs out of the context of their findspot, their time, and their larger social structural context limits and possibly distorts what information we can glean. A seemingly similar design’s meaning and social utility from Tepe Gawra at 3900 BCE and from Ur at 1800 BCE are necessar­ ily different in the code, the cultural context, the social identities of senders and receivers, and therefore in their meaning. The image of an important personage hunting would not be the same, unless one thinks the roles of the Neo-Assyrian and Uruk (LC4/5) kings were the same, and the need to present themselves publicly in particular ways was unchanged. Further, as Denham (2018) argues, most early “seals” probably were never used to seal anything. They were amulets or identifiers of group affiliation or belief. Hence, I think “glyph” is a better term for the category. The functions of glyphs are those of an amulet, a seal, an identifier, or simply for decorative display; most likely they represented a combination of those functions. Glyphs become seals when they are applied to clay locks, pots, or tablets (Fig. 4). For example, the clay sealings at Sabi Abyad, one of the earliest sites with them, had a very limited number of designs (Akkermans and Duistermaat 1996). The fact of sealing was more important, perhaps, than the identity or authorization of the sealer, represented by its design. Theoretically, the seal design represented a small personal lock rather than an administrative control device. Like many of the designs in this pre-state society, individual or institutional identities were much less important than group association. The audience for sealings was personal and the heterogeneity of senders and receivers’ roles were limited; Fig. 4. Seals and tablets. again, most glyphs were not for

384

Mitchell S. Rothman

sealing (Fig. 5). Parenthetically, because of the difference between glyphs uses as seals and other things, clay sealings (always impressed locks or tags) may need to be a separate category in analysis from glyphs, which may or may not be control devices. The use of glyphs for sealing and, as many of us argue, for controlling access to raw mater­ials or goods, grew early in the Late Chalcolithic. The societal context was one in which the centralization of some activities, hierarchical authority structures, workshop production of goods, differential social status, formal temple religion, and established trading networks over a large geographical area had evolved (Rothman 2007). Some of the glyphs were control mechanisms and auditing devices, there­ fore administrative. The distinction between public and private was not so clear; they still could be personal or, as Pittman (1993) argues, about the good transferred as much as the institution. I am increasingly dubious that the seals showing some economic activity were simply marking the product Fig. 5. Prehistoric stamp seals (von Wickede 1990: drawcontrolled. Why not have a symbol ings 208–223; © Profil Verlag). of the item and not the complexity of other elements of design? If later Ur III practice applies— again, we must demonstrate this and not simply assume the compar­ ison is a literal equiv­ alent—occupational groups were responsible for providing their products to various institutions (Molina 2016). The seal represents the occupational association. Individuals within the occupational groups would still want to signify their identity. I am also not sure how many non-central institu­ tion­ al (“private”) transactions needed to be locked or audited, but Fig. 6. Tepe Gawra sealings, Phase XI (after Rothman within the occupational group such 1994a: Figure 5). own­ership might be important.

Palace and “Private” Sectors in Old Babylonian Ur

385

As early as Gawra Phase XI (3900 BCE) the topics of designs appear to differentiate among activities: those sealings associated with administration and craft production functions emphasized predation scenes; those associated with religion showed domestic animals (Rothman 1994b; Fig. 6). At the same time the glyph found in the temple was of snakes and two animals, perhaps lions. It was a simple language, but to my mind one that showed the development of new mental map of social organization. The great variety of seal designs suggest that they may have been more like signatures recognizable to individuals who knew many of the players. Stamp seals, which present the whole image, were suc­ceed­ed by cylinder seals. They were developed later to increase their information potential (Nissen 1977) as the social context expanded geographically and in population size, the audiences became more heterogeneous, and there was less face-to-face contact. The message code of the image in the Early Dynastic Period, as Scott writes (2005), in many ways ran parallel to the written script (Fig. 7). She found that symbols were put together like logograms of the Uruk IV writing system. Sets of symbols became homogenized (“images repeatedly portrayed the same way, then paired with other symbols into new messages”: “an eagle hovering over a calf (representing a mag­ ical aspect of Early Dynastic society at Ur), a man milking a cow (representing productive aspects of the economy), or a lion mauling a caprid (another representation of socio-religious force)” (Scott 2005: 204). This is less of a signature and more of socially based code. Fig. 7. Uruk IV glyphs (courtesy of Sarah Scott). By the Akkadian period and even more clearly during the Ur III dynasty, the most bureaucratic system known in ancient Mesopotamia, formulaic narratives—for example, the presentation scene, the predation scene, the banquet scene, a worship scene, etc.—became coherent, set symbolic narratives, as opposed to re-combinations of symbols. The whole narrative frame was in a sense homogen­ ized. Further, as Winter (1987) argues, the standard narrative in the Ur III period represented the institution (Fig. 8). The king or a temple each had its Fig. 8. Ur III Seals on tablets (after Winter 1987).

386

Mitchell S. Rothman

own narrative identifier, now including text, which were given to each subordinate identifying them as a member of the institution and at some specific rank. That rank was often represented by small changes in the formulaic narrative image, like distance from the king or god. However, the practice in the Ur III was to make sure that the text and the image of the leader was impressed. Often the rest of the image was not. The idea that these public glyphs represented a signature is less likely. So, interpreting the language, the message, sender, audience, channel, code and topic requires the analyst to look at the social setting or context. In the remainder of this paper I will look at the glyphs of Isin-Larsa and Old Babylonian period Ur, not as a whole but in their social context to see if their messaging and its relationship to ancient mental maps of social organization can be teased out.

Old Babylonian Ur The Ur excavations revealed a series of public buildings as well as neighborhoods within a walled area (Woolley and Mallowan 1976; Fig. 9). The population within the wall contained what one might think of as the central institutions and groups with greater access to economic ben­ efits. Although a true picture of the Old Babylonian society must extend beyond these two organizational groupings, this is what we have. The central institution was represented by the Gipar-ku, the southern part of the Nanna temple and storehouse complex, as well as houses and graves in the residential neighborhoods AH, EM, and EH (the last full of stratigraphic and dating prob­ lems). Texts from the neighborhoods imply that residents of the EM neighborhood were dependents of the Nanna Temple, AH of theoretically more “private” landowners and entrepreneurs (van der Mieroop 1992; Charpin 1986). These AH families produced bread, including bread for the central institutions, fish, and other products. They also made silver loans, rented boats, etc. EH according to Woolley was inside the Temenos wall and therefore Fig. 9. Site of Ur, Old Babylonian period (after Woolley and part of the sacred sanctuary. Mallowan 1976: Figure 116).

Palace and “Private” Sectors in Old Babylonian Ur

387

If my theory of integration is correct, the different sites should present different social contexts: Gipar-ku glyphs and sealings should be of a limited number of formulaic narratives. EM residents should be using closely related ones. However, AH residents should exhibit symbols and visual narratives distinct from the central institutions. At the same time their integration into the larger state should include some of the formulaic designs associated with the temple. The sample is seventy-four glyphs and ten sealings, not bad as a number, but very hard to put in secure proveniences, and images for all the glyphs are not readily available (Table 1). Table 1. Sample size of analysis. Locus

Glyphs

Sealings

Gipar-ku

7

9

AH

30

1

AH Graves

13

0

EM

14

0

EM Graves

6

0

EH

13

0

EH Graves

1

0

Total

74

10

What is evident is that of the narrative scenes in these glyphs most fall into a very few categories: the presentation scene, ones in which the participants worship a bush, or combat scenes with wild animals (according to Collon (1987), where the key figure is sitting, it is Isin-Larsa period, Old Babylonian when all are standing) (Fig. 10). By far the most common is the presentation scene. The combat scene is the second most popular. In AH and EM, people placed a version of this scene in graves. The characters and their style of dress appear quite similar. Details changed in the Worship and Combat scenes. Some, however, were extremely similar between, for example, EM and Gipar-ku, and excavators recovered only one from a grave. As predicted, all the glyphs in EH and EM were of scenes shared with the Gipar-ku (Figs. 11, 12). Also, as predicted, AH yielded a variety of designs and forms different than those of the Gipar-ku in addition to the formulaic ones (Fig. 13). Especially interesting are the stamped glyphs, one from an unidentified provenience, the other from the Hendursag Chapel, 1 Straight Street. My guess would be that these are glyphs not used for sealing, as their unusual form appears more decorative.

Fig. 10. Most common scenes (drawing by M. S. Rothman).

388

Mitchell S. Rothman

Fig. 11. Old Babylonian presentation scenes from Ur (drawn from Legrain 1951, Woolley and Mallowan 1951, and the Ur archives).

Palace and “Private” Sectors in Old Babylonian Ur

389

Fig. 12. Worship and combat Scenes, Old Babylonian Ur (drawn from Legrain 1951, Woolley and Mallowan 1951, and the Ur archives).

Mitchell S. Rothman

390

How can we explain such seeming similarity in narrative scenes, shared by “private” citizens and central institutions? A clear factor is the writing (Table 2). The pattern evident with writing is too distinct to be random. Even given the problems with excavating and recording, most glyphs and sealings from the Gipar-ku had inscriptions and the image, whereas very few of the glyphs from AH had inscriptions. The inscriptions in the Gipar-ku glyphs are almost all either royal seals—one “given by the king”—and mention administrative workers, dub-sar or gà-dub-ba, scribe and warehouse manager. The use of the glyph as seal suggests that someone had to be able to read it. The majority of these Gipar-ku inscriptions are on sealings, presumably locked items imported from palace sources. The inscriptions in AH are only personal names with no mention of administrative titles. They are on presentation scenes or two text-only sealings. Two are from graves. Perhaps, they mark people who had regular interactions with the temple or palace. However, the audiences for those with the common narrative design but no text probably were different than those of the central institutions. Table 2. Counts of glyphs with text or image by provenience. Text?

Gipar-ku

EM

EM Graves

EH

AH

AH Graves

Image Only

0

2

2

6

21

7

Text Only

1

1

0

0

2

0

Text & Image

11

0

3

6

3

3

Unclear

3

0

0

0

1

1

This pattern raises a little discussed question: from where do the glyphs come? Who made them? Does one have the glyph maker do the design as one wants? Does one get a generic workshop-made one and add some detail? Did those who did not receive an institutional seal buy one with a common scene? If the glyphs are from a common set of narratives, how much individual meaning can we ascribe to them? They become general cultural icons like the Old Baby­lonian goddess molds rather than identifiers of individuals or offices without writing. Collon (1987: 103a) certainly suggests this: “Often the stones were ready-cut with standard subjects and the prospective owner would select the seal he wanted and would then have it en­ graved with an inscription and, perhaps, with appropriate filling motifs.” A new text dis­cov­ered by Charpin Fig. 13 Other scenes, Old Babylonian Ur (drawn from Legrain 1951, Woolley and Mallowan 1951, and the Ur archives). (2016) suggests exactly

Palace and “Private” Sectors in Old Babylonian Ur

391

that, as Yasmah-Addu’s secretary at Mari instructs a servant to buy a seal from a kuṭimmum, a carver. There is clear evidence of seal making workshops at sites like nearby Sippar in the Old Babylonian period (Werr 1986). Contexts of finds matter in their interpretation as well. For example, new excavations at Ur have yielded a number of new glyphs from graves (Hafford, personal communication). Many are not drilled through or complete, suggesting they were never used as seals and represent perhaps the religious beliefs encoded in the form­ ulaic design.

Discussion and Conclusion This analysis is obviously not the last word. New excavations in AH are produc­ ing more artifacts, which will presumably have more secure proveniences, there may be more sealings on tablets I have not been able to find, and, of course, more texts are being translated all the time. However, at the core of understanding the Old Babylonian state and its development, I would argue, are the senders and receivers. We have a tendency to use familiar terms like “private” and “elite” to describe them, but as the analysis above suggests, this clear demarcation may not be sufficient to understand the social structure of ancient Mesopotamia. In a single paper at the Rencontre, the speaker referred to the military elite, economic elite, and governing elite. What makes one elite? Is it those with the greatest authority in large institutions? Are some of the clearly wealthy residents of AH with connections to the central institutions elite? How far up does one need to be in the hierarchy to be elite? Are palace elites the same or different than economic elites? How do they interact? The same applies to “private” sectors. Are they persons who have no interaction or dependence on the palace/temple sector? At Old Babylonian Ur this is clearly not the case. The residents of AH are probably more elite, I would surmise, than farmers in agricultural villages or pastoralists in nomadic camps, but how so? It is easy to apply terms from modern state systems that I doubt typified the second millennium BCE. If one wants to use the terms elite or private at least one has to define them in terms of the societal structures of the time. Although this could be the topic of a whole other paper, one suggestion is to construct the ancient structure as social identities. Everyone in a society has identities with different functions, requiring different kinds of relationships, and associated with different statuses (Goodenough 1965). One can be a father/mother, a son/daughter, a boss/employee, a leader/ citizen, a slave/master, a landless laborer, a craftsperson, a husband, or a wife at the same time. One’s social role is the whole collection of one’s social identities. The homogeneity of kinship-related societies is reflected in the fact that one social identity defines a person’s total role. As societies become more heterogeneous, the number of separately identifiable identities increases. The residents of AH had identities, probably of lower status as suppliers to the Nanna Temple. Yet, they received various kinds of subsidies and payments from the Nanna Temple. They were not dependent workers of the temple, because they also controlled their own capital in the form of entrepreneurial “businesses” to produce foodstuffs for a wider audience than the temple, and to give loans in silver and boats presumably to others not fully integrated into the temple system. The use of glyphs as seals was certainly tied to the social identities of the players in this system. I suspect it was limited to those

392

Mitchell S. Rothman

with an identity in one of the central institutions, but also those non-dependents of the central institutions who controlled some means of production or political role. I would bet, however, that even landless laborers had glyphs as amulets. They did not need them for seals, however. To make sense of the messages contained in these glyphs researchers need to understand the organizations in which they participated and the social identities they had that would require or permit them to use glyphs as seals. In regard to those messages, clear patterns in the Old Bablyonian Ur glyph designs do exist, as discussed above. If one uses the seal for sealing, as a control mechanism, in a large hierarchical institution like Gipar-ku, having a scribe who could read them is key. The image acts as a symbol of the institution, but the same glyph can appear in non-centralized, “private” contexts. The image on the seal may, on the other hand, have little to do with control, even with sealing clay. In the end, analysis is complex. To fully understand its nuances, such analysis requires a correlation of textual, glyptique, and artifactual information from well-provenienced and representatively collected and well-recorded excavation squares (see Keith 1999 for another leg of this kind of analysis). At present, having reviewed the fieldnotes, that is not possible for Old Babylonian Ur. What we can say certainly indicates the power to understand complex social interactions and mental models and the kinds of information provided by glyptique.

Bibliography Akkermans, P. and Duistermaat, K. 1996 Of Storage and Nomads. The Sealings from Late Neolithic, Sabi Abyad, Syria. Paléorient 22/2: 17–44. Charpin, D. 1986 Le Clergé d’Ur au siècle d’Hammurabi. Geneva: Librarie Droz. 2015 Gods, Kings, and Merchants. Leuven: Peeters. 2016 Un sceau gravé et inscrit sur commande d’après une lettre inédite des archives royale de Mari. Pp. 87–97 in Mille et une empreintes. Un Alsacien en Orient. Mélanges en l’honneur du 65e anniversaire de Dominique Beyer, eds. J. Patrier, P. Quenet, and P. Butterlin. Subartu 36. Turnhout: Brepols. Charvát, P. 2005 The Iconography of Pristine Statehood. Prague: Karolinum. Collon, D. 1987 First Impressions: Cylinder Seals in the Ancient Near East. Chicago: The University of Chicago Press. Denham, S. 2018 Late Neolithic and Early Chalcolithic Stamp Seals in the British Museum. London: British Museum Press. Ellis, M. 1976 Agriculture and the State in Ancient Mesopotamia. Occasional Publications of the Babylonian Fund 1. Philadelphia: Babylonian Fund, Penn Museum. Falkenstein, A. 1974 The Sumerian Temple City. Monographs on the Ancient Near East 1/1. Malibu: Undena Publications. Feinman, G. and Marcus, J. 1998 The Archaic State. Santa Fe: SAR Press.

Palace and “Private” Sectors in Old Babylonian Ur

393

Goodenough, W. 1965 Rethinking ‘Status’ and ‘Role’. Pp. 1–24 in The Relevance of Models for Social Anthropology, ed. M. Banton. A.S.A. Monographs 1. London: Tavistock. Hymes, D. 1962 The Ethnography of Speaking. Pp. 15–52 in Anthropology and Human Behavior, eds. T. Gladwin and W. Sturtevant. Anthropological Society of Washington DC. 1964 The Ethnography of Communication. American Anthropologist 66/2: 1–35. Jacobsen, T. 1982 Salinity and Irrigation Agriculture in Antiquity. BM 14. Malibu: Undena Publications. Keith, K. E. 1999 Cities, Neighborhoods, and Houses: Urban Spatial Organization in Old Baby­ lonian Mesopotamia. Ph.D. dissertation, University of Michigan, Ann Arbor. Leemans, W. F. 1950 The Old Babylonian Merchant: His Business and His Social Position. SD 3. Leiden: Brill. Legrain, L. 1951 Seal Cylinders. UE 10. London: Trustees of the British Museum and of the University Museum, University of Pennsylvania, Philadelphia. Molina, M. 2016 The Routes to Ur: Communications between the Capital City and the Core Provin­ces of the Ur III State. Unpublished paper given at the 62nd Rencontre Assyriologique Internationale, Philadelphia, 2016. Nissen, H. 1977 Aspects of the Development of Early Cylinder Seals. Pp. 15–24 in Seals and Sealings in the Ancient Near East, eds McG. Gibson and R. D. Biggs. BM 6. Malibu: Undena Publications. Oppenheim, A. L. 1977 Ancient Mesopotamia: Portrait of a Dead Civilization. Chicago: University of Chicago. Pittman, H. 1993 Pictures of an Administration: The Late Uruk Scribe at Work. Pp. 235–246 in Between the Rivers and Over the Mountains, eds. M. Frangipane, H. Hauptmann, M. Liverani, P. Matthiae, and M. Mellinck. Rome: Università di Roma. 1994 Towards an Understanding of the Role of Glyptic Imagery in the Administrative Systems of Proto-literate Greater Mesopotamia. Pp. 177–204 in Archives Before Writing, eds. P. Ferioli, E. Fiandra, G. Fisore, and M. Frangipane. Rome: Universitá di Roma. Porter, A. 2012 Mobile Pastoralism and the Formation of Near Eastern Civilizations. Cambridge: Cambridge University Press. Rothman, M. S. 1994a Palace and Private Agricultural Decision-making in the Early 2nd Millennium B.C. City-State of Larsa, Iraq. Pp.149–167 in The Economic Anthropology of the State, ed. E. Brumfiel. Monographs in Economic Anthropology 11. Lanham: University Press of America. 1994b Seal and Sealing Findspots, Design, Audience and Function. Pp. 97–121 in Archives Before Writing, eds. P. Ferioli, E. Fiandra, G. Fisore, and M. Frangipane. Rome: Universitá di Roma. 2002 Tepe Gawra. the Evolution of a Small, Prehistoric Center in Northern Iraq. Philadelphia: The University of Pennsylvania Museum Publications.

394

Mitchell S. Rothman

2007 The Archaeology of Early Administrative Systems in Mesopotamia. Pp. 235–254 in Settlement and Society: Essays Dedicated to Robert McCormick Adams, ed. E. Stone. Los Angeles: Cotsen Institute of Archaeology, UCLA. Scott, S. 2005 Figure, Symbol, and Sign: Semiotics and Function of Early Dynastic I Cylinder Seal Imagery from Ur. Ph.D. dissertation. University of Pennsylvania. Steward, J. 1955 Theory of Culture Change. Bloomington: University of Indiana Press. Stone, E. 1987 Nippur Neighborhoods. SAOC 44. Chicago: The Oriental Institute of the University of Chicago. Van De Mieroop, M. 1992 Society and Enterprise in Old Babylonian Ur. BBVO 12. Berlin: Dietrich Reimer Verlag. Von Wickede, A. 1991 Prähistorische Stempelglyptik in Vorderasien. München: Profil Verlag. Wallace, A. F. C. 1971 Administrative Forms of Social Organization. McCaleb Module in Anthropology 9. Reading, MA: Addison-Wesley. Werr, L. al-Gailani. 1986 The Sippar Workshop of Seal Engraving. American Journal of Archaeology 90/4: 461–463. Winter, I. J. 1987 Legitimation of Authority Through Image and Legend: Seals Belonging to Officials in the Administrative Bureaucracy of the Ur III State. Pp. 69–93 in The Organization of Power, eds. McG. Gibson and R. D. Biggs. SAOC 46. Chicago: The Oriental Institute of the University of Chicago. Woolley, L. and Mallowan, M. 1976 The Old Babylonian Period. UE 7. London: Trustees of the British Museum and of the University Museum, University of Pennsylvania, Philadelphia; British Museum Publications. 1975 Population, Exchange and Early State Formation in South Western Iran. American Anthropologist 77: 267–291.

Materiality and Mark-Making: Sealing and Community in Early Dynastic Ur Sarah J. Scott Wagner College

Introduction The topic of this paper is the relationship between image-making and writing during the early third millennium BCE in Mesopotamia. I am exploring specifically a group of incised sealings from the Seal Impression Strata (SIS) at Ur. The larger corpus of sealings was the topic of my dissertation, completed at University of Pennsylvania in 2005. In that work, I categorized the sealings according to the visual elements impressed with seals: figural or abstract (Figs. 1a, 1b). This paper looks at a third category of marks present on the sealings: incised visual elements (Fig. 1c). All three visual elements extant on the Ur sealings were used together in a visual communicative system that predated the extensive use of cuneiform for the recording of grammatical language. 1 My intention is to present these incised sealings as an essential component for the understanding of a trifold visual system. Through the lens of materiality-based methodologies these sealings allow us to more deeply understand how the image-based system is related to the social context. It is not my intention to present a reading of the images, symbols, or signs carried by these sealings, but rather to identify these clay sealings and the marks they carry as agentively powerful objects, thus arguing for their worth as an essential social tool of early fourth millennium Mesopotamia.

Materiality, Objects, and Humans: Theoretical Underpinnings “Wisdom has been accredited to those who claim that materiality represents the merely apparent behind which lies that which is real” (Miller 2005: 1). In the case of religion, this much has been true, as we now believe, as early as the Neolithic in the Near East. A statue, although made of material, in fact possesses the reality of the 1. Cuneiform was certainly employed during the early Early Dynastic periods, at Ur as well as other places, but is used in very limited ways for the recording of economic transactions. It is only in the following period, with the emergence of the Fara texts, for example, where writing records historic and religious narratives.

395

396

Sarah J. Scott

deity. The medium of stone or clay allows for the manifestation of the divine, which ultimately gives the statue its functionality. The “wisdom” to which Miller refers is the knowledge of the user of that material, their knowledge of the underlying idea which is made apparent through the material. As an anthropologist and scholar of material culture, Miller asserts that the tendency to focus on materiality is a trait of humanity; this tendency is ubiquitous. In addition to our application of materiality as a heuristic device in religion, it is a driving force of economics; the transformation of the material world in production (Marxism), the accumulation of material, and ultimately the identities we create for ourselves through production (Capitalism). As humans, we approach materiality and engage with a material world in order to make the world what we believe it should be (Miller 2005: 2). As objects constructed of a meaningful medium as well as objects that embody economic and divine power, sealings can be considered as agentive objects. Stemming specifically from the Bourdieu work, Outline of a Theory of Practice (1977), material objects are the source for the socialization of humans. Variables such as categories of objects, placement of them within a working space, and the way they are ordered belie other orders (or systems) such as gender or social hierarchy. Objects, then, can help us to understand the nature of human thought at any point and time. As Appadurai (2015) articulates, the material object functions as the mediator of the social collective. A material object as mediator also requires the capability of such an object to possess agency. Gell (1998) and Latour (2007) consider agency on multiple levels: according to Gell, agency lies in the ascription of action humans imprint on objects, while Latour prefers a larger network of interaction between objects to be the cause of agentive action. This paper will consider sealings as objects capable of illuminating the nature of human thought and the ordering of the world in the early third millennium BCE. The analysis of the “artifacts” presented here will address three specific components of their materiality. First, clay itself was an agentive medium; second, sealings are agentive objects that acted as mediators of communication through “object entanglement” (practice theory), and third, the images contained on the sealings allowed for the object to participate in a system of “object pluralism,” where certain related attributes (iconography, in this case) allowed for communication across multiple social groups. This final area of analysis posits that an understanding of the impressed image can only be approached through the material that carries it, and the actions through which the impressions are used or exchanged. 2 Ultimately, what I argue is that through these materiality-based approaches, the Ur sealings provide a window into the pre-Early Dynastic III period where administrators developed a visual system for communication that was legible by multiple groups. By the Early Dynastic III period the sign-based system of cuneiform writing becomes distinct from the image-based system of sealing.

2. Recently Appadurai (2015) has critiqued Latour (2007) and other proponents of the material turn, and he proposes that a focus on “mediants” and “mediation” permits more historically sensitive analyses of the formation of diverse social collectives entangling people, places, and things. At the same time, archaeological research is an inherently mediated enterprise, for interpretation relies on the traces and material signs of past practices. Interpretations based on the material turn include the applications of theories of entanglement, network, assemblage, or indexicality, some of which are applicable to the Ur sealings.

Materiality and Mark-Making

397

I offer the idea that the visual vocabulary carried on the incised sealings is related to that found on the abstract and figural seal-impressed imagery, as well as the Archaic Ur tablets and even on other media from across the Early Dynastic horizon. Although an actual translation of the incised or impressed signs may not be possible, I argue the visual similarities across the variable types of marks on the sealings suggest they functioned within the same context of administration as the figural imagery. At the same time, figural seal imagery and larger-scale works of art with similar figural iconography functioned to communicate related ideas across similar semantic categories as the incised and abstract sealings. I will first present a visual analysis of the marks and images on the impressions and related large-scale works of art, highlighting those that bridge multiple artifact types and media. I will then present new work on the archaeological context of the SIS and the sealings’ relationship to the Archaic texts in order to establish the context of use for these objects. Lastly, I will touch upon ways we might understand how such imagery and the material objects that carried this imagery may have had implications for those who produced, viewed, used, and lived with it—thus linking the functionality of the incised and impressed imagery to its material nature.

The Data-Set: Incised “Signs,” Impressed “Signs,” Impressed “Images” A small portion of the Ur sealings bear incised markings (Fig. 1c), which, to my knowledge, have not been adequately discussed in relation to the other admini­ strative tools found in the SIS. 3 These incised marks are distinct from the figural or abstract visual element types both of which were applied to the clay by the rolling of cylinder seals. 4 This subset of incised sealings makes up about 16% of the total number of approximately 1,200 sealings, 980 of which date to the early part of the Early Dynastic period. These incised sealings largely date to the Early Dynastic I period, and the majority are jar stoppers. 5 Although difficult to interpret, some carry marks that are visually similar to proto-cuneiform signs indicating concepts such as commodities, institutions, and labor. Another subset of the SIS sealings bears cylinder seal impressions depicting abstract elements (Fig. 1b). Abstract Seal-impressed container and door sealings carry about 200 unique images, and of that group seventy-one contain archaized Sumerian signs 6 representing a wide variety of semantic categories, such as commo3.  Work presented at this RAI by Petr Charvat, Giacomo Benati, and Camille Lecompte was instrumental in elucidating the archaeological and textual context of these objects. Their work too, seeks to understand the place of the sealings within a broader social context. 4.  Within this group of impressions, there are 460 unique images, bridging these two categories distinct in style and iconography. 5.  Most incised sealings come from a similar fill context hence I suggest they are from one administrative period. 6. The fact that these signs are found in both systems (proto-cuneiform and glyptic imagery) is significant because it provides one bit of evidence that they were part of the same system. A further consideration of the signs used on the Ur seals is their physical form as it relates to the evolution of the proto-cuneiform script. The signs found in Early Dynastic I tablets (documents) are linear and impressed with the wedge-shaped section of a stylus. They are abstracted from the earlier Uruk IV-III phases of the script. This earlier script, composed of signs literally drawn on the clay with the pointed end of a stylus, is a more representational form, and dates to the periods preceding Early Dynastic I. It is

398

Sarah J. Scott

dities, fields, architectural structures, institutions, 7 or the “cities list,” some of which have been discussed by Jacobsen (1957), Michalowski (2003), Steinkeller (2002), Matthews (1993), and Yoffee (1995). Other abstract elements such as the EDINNU and LEGCROSS are not iconographically similar to proto-cuneiform signs, but are similar in style and are found in context with the sign-based elements. 8 A third subset of the sealings bear cylinder seal impressions of a figural nature (Fig. 1a). These impressions contain more highly sculptural elements depicting iconography relating to human or animal activities such as herding, presentation or dairy scenes, procession, combat, and banqueting. 9 These three subsets of markings, although distinct, were all being used within a larger system of record keeping. There are certain elements that either semantically or formally relate to particular categories of information relevant to a temple econ­ omy. These categories include commodities, institutions or geographic locations, or particular components of administration. Below, I will present just a few examples of links between the incised, abstract, and figural markings on the sealings, followed by brief references to other related larger objects. Commodities is one category possibly indicated through incised markings and abstract impressions. As noted first by Legrain (1936) and most recently by Lecompte (2016), incised sealings contain marks possibly related to signs for textiles (or plow, GADA) (Fig. 2). Dairy products may also be referenced through an incised therefore interesting to note that the signs found in the seal impressions in this corpus, although they do date to the Early Dynastic I period, exhibit an archaizing form of the proto-cuneiform script. The sign list used in this study, therefore, is of the Uruk IV phase of the script. The sign renderings used in the database have been taken from ZATU (Green and Nissen 1987), and rough suggestions for values for the signs have been compiled from various other sources. Transliteration and translation of signs in the Ur glyptic corpus are based on a study of both Ur III and archaic Sumerian. Here, I have restricted the study to uncontested examples, most of which are discussed by Englund (1998). The Archaic Texts from Ur evidence a sign-form linked to both the Archaic Uruk texts (earlier) and the Fara texts (later) (as discussed at this RAI by Benati and Lecompte). In fact, the argument made by Benati and Lecompte that the Ur Archaic Texts are transitional between the Uruk and Fara text periods supports my argument that the entire corpus of written and impressed clay artifacts reflects a moment of diversion between written and figural visual systems. Of the thirty-six identifiable signs eleven occur in more than five seals, many discussed by Szarzynska (1992). 7.  A detailed list of the sign elements found on the seals can be found in Scott 2007, and include various signs related to administrative concerns of temple administration, as well as the so-called “City Seals.” The latter have a tradition reaching back to the Uruk and Jemdet Nasr periods and have been explained as the visualization of an inter-city cooperative organization (Kengir League). The exact nature of this organization is still debated (see Postgate 1986; Jacobsen 1957; Steinkeller 2002; Matthews 1993; Yoffee 1995). Comperanda for such imagery has been found at other sites such as Fara, Al Hiba and most recently Konar Sandal South, in Iran. Redistribution on an inter-city level, some say they acknowledge “an ideology of self-conscious cultural similarity” (Yoffee 1995) while others claim them to be purely semiotic function—the routing of the actual goods (Michalowski 2003). Steinkeller suggests it was an amphictyonic organization that originally started out at Uruk (slaves given to that city) and ultimately moved to Nippur. 8.  Most of these seal images contain such abstract elements arranged in free-floating compositions, and frequently contain mixtures of symbol (non-sign related) and sign elements. 9.  Distinct themes represent distinct categories of administration linkable to the identity of either the agents involved or the commodities. Distinction between and inclusion within social groups, as articulated also through larger, more “public” objects, indicate there was a tightening and increased control over time through the imagery. The categories of administration and agents involved can become more visible to us through a look at the tablets.

Materiality and Mark-Making

399

A1. Drawing of sealing, Ur SIS. After Legrain 1936, cat. no. 318.

B1. Drawing of sealing, Ur SIS. After Matthews 1993, cat. no. 1.

C1. Drawing of sealing, Ur SIS. After Legrain 1936, cat. no. 80.

A2. Photo of sealing, Ur SIS. Author’s photo, Penn Museum, Philadelphia, 33.35.410.

B2. Photo of sealing, Ur SIS. Author’s photo, Penn Museum, Philadelphia, 33.35.293.

C2. Photo of sealing, Ur SIS. Author’s photo, Penn Museum, Philadelphia, 31.16.680.

Fig. 1. Ur sealings.

A1. Drawing of sealing, Ur SIS. After Legrain 1936, cat. no. 106.

A3. Drawing of sealing, Ur SIS. After Legrain 1936, cat. no. 116.

B1. Drawing of sealing, Ur SIS. After Matthews 1993, cat. no. 118.

A2. Drawing of sealing, Ur SIS. After Legrain 1936, cat. no. 82.

A4. Drawing of sealing, Ur SIS. After Legrain 1936, cat. no. 65.

B2. Photo of sealing, Ur SIS. Author’s photo, Penn Museum, Philadelphia, 31.16.653.

Fig. 2. Ur Sealings.

400

Sarah J. Scott

mark perhaps identifiable as a byre (Fig. 3). 10 Figural seal images depict the herding of animals from which these very commodities are produced, A. Drawing of sealing, as well as depictions of byres (Fig. 4). Ur SIS. After Legrain 1936, cat. no 118. Large-scale temple inlay and plaques depict similar imagery, while other free-standing sculptural evidence, such as the hero-figure from Tell Agrab depicts such vessels presented as offerings B1. Drawing of sealing, C1. Drawing of sealing, (Fig. 5). Ur SIS. After Matthews Ur SIS. After Matthews As suggested by Matthews and 1993, cat. no. 69. 1993, cat. no. 32. others (1993: 41–46) institutions or locations may be indicated through multiple marks (Fig. 6). The UB sign appears frequently in both figural and abstract sealing imagery. Another institution or location possibly indicated though incised marks include the city of Eridu, through the NUN sign. It too occurs frequently on the abstract seal-impressed sealings both as a rel­ C2. Photo of sealing, atively isolated element, as well as in B2. Photo of sealing, more formal lists of city names. Another Ur SIS. Author’s photo, Ur SIS. Author’s photo, Penn Museum, PhilaPenn Museum, Philamark and some with cross-lines bear delphia, 33.35.423. delphia, 31.16.670. resemblance to the UNUG sign (Uruk/ Fig. 3. Ur Sealings. great temple). 11 Marks similar to E2 also occur. Both UNUG and E2 appear frequently on the abstract seal-impressed sealings. 12 Additionally some of these institutions are mentioned (E2 GAL [palace]) in a number of the Ur Archaic texts (Lecompte et al. 2016: 23). Figural counterparts to the incised and abstract markings discussed above can be found in depictions of temples. Temples are frequently depicted in the figural seals: milking and procession or ritual activities are depicted in and around temple facades (Fig. 7). Animals in procession, recumbent bovines, and herding are found sometimes with an eagle hovering in a protective gesture. The hovering eagle provides a direct link with an incised example. Such scenes are found in many other media spanning the Early Dynastic Horizon, often associated with large-scale temple contexts. Labor and field administration are topics frequently referred to in the Ur Archaic texts, and some of the incised marks may relate to this arena (Fig. 9). Some bear marks similar to the DUN sign, as identified by Legrain (1936). A mark rem­ iniscent of the DU sign appears multiple times, once within divided quadrants, 10.  Here, according to Legrain’s translation, there are signs indicating “butter of royal quality from the byre” (ia áb lugal bár) 929, although Burrows suggests it is a proper name. The concentric square and cross mark indicates the “byre.” 11.  Read by Legrain as temen (1936). 12. Sometimes these signs appear in columned and cases arrangements with a great variety of other sign types and at other times in more freely arranged compositions.

Materiality and Mark-Making

401

A1. Drawing of sealing, Ur SIS. After Legrain 1936, cat. no. 191.

A2. Drawing of sealing, Ur SIS. After Legrain 1936, cat. no. 314.

A3. Drawing of sealing, Ur SIS. After Legrain 1936, cat. no. 308.

A4. Photo of sealing, Ur SIS. Author’s photo, Penn Museum, Philadelphia, 31.16.681.

B1. Drawing of sealing, Ur SIS. After Legrain 1936, cat. no. 338.

B2. Drawing of sealing, Ur SIS. After Legrain 1936, cat. no. 337.

B3. Drawing of sealing, Ur SIS. After Legrain 1936, cat. no. 349.

B4. Photo of sealing, Ur SIS. Author’s photo, Penn Museum, Philadelphia, 33.35.332.

Fig. 4. Ur Sealings.

A. Temple frieze with inlay of cows, Tell B. Nude male holding vase C. Animal herd plaque, KhaAl-’Ubaid. Image from Penn Musuem, on head, Tell Agrab Shara fadje Temple Oval. Drawing Philadelphia, B15880. Temple. Image from Oriental after Boese 1971, CT4. Institute, Registration Number A18067, Accession Number 2111. Fig. 5. Early Dynastic Sculpture.

402

A1. Drawing of sealing, Ur SIS. After Legrain 1936, cat. no. 105.

Sarah J. Scott

B1. Drawing of sealing, A4. Photo of sealing, Ur SIS. Author’s photo, Ur SIS. After Matthews Penn Museum, Philadel- 1993, cat. no. 70. phia, 33.35.408.

A2. Photo of sealing, A5. Drawing of sealing, Ur SIS. Author’s photo, Ur SIS. After Matthews Penn Museum, Philadel- 1993, cat. no. 7. phia, 33.35.265.

A3. Drawing of sealing, Ur SIS. After Matthews 1993, cat. no. 39.

D1. Drawing of sealing, Ur SIS. After Legrain 1936, cat. no. 84.

B2. Photo of sealing, Ur SIS. Author’s photo, British Museum, London, 1930.12.13.410.

D2. Drawing of sealing, Ur SIS. After Matthews 1993, cat. no. 114.

B3. Drawing of sealing, A6. Photo of sealing, Ur SIS. Author’s photo, Ur SIS. After Matthews Penn Museum, Philadel- 1993, cat. no. 11. phia, 31.16.614.

D3. Drawing of sealing, Ur SIS. After Matthews 1993, cat. no. 1.

C. Drawing of sealing, Ur SIS. After Legrain 1936, cat. no. 73.

Fig. 6. Ur sealings.

Materiality and Mark-Making

A1. Drawing of sealing, Ur SIS. After Legrain 1936, cat. no. 207.

A2. Drawing of sealing, Ur SIS. After Matthews 1993, cat. no. 48.

A3. Drawing of sealing, Ur SIS. After Legrain 1936, cat. no. 349.

B1. Drawing of sealing, Ur SIS. After Legrain 1936, cat. no. 329.

403

C1. Drawing of sealing, Ur SIS. After Legrain 1936, cat. no. 475.

B2. Drawing of sealing, Ur SIS. After Legrain 1936, cat. no. 378.

B3. Drawing of sealing, Ur SIS. After Legrain 1936, cat. no. 387.

C2. Photo of sealing, Ur SIS. Author’s photo, Penn Museum, Philadelphia, 33.35.275.

Fig. 7. Ur Sealings.

A. The Imdugud relief, Tell Al’Ubaid. Image from the British Museum, London, 114308.

B. Stone mace head, Tello. Image from the British Museum, London, 23287.

D. Plaque with a libation scene, Ur. Image from the British Museum, London, 120850.

C. The Uruk trough, Uruk. Image from the British Museum, London, 120000.

E. Wall plaque with libation scenes, Ur. Image from the British Museum, London, 118561. Fig. 8. Early Dynastic sculpture.

404

Sarah J. Scott

A1. Drawing of sealing, Ur SIS. After Legrain 1936, cat. no. 119.

B1. Drawing of sealing, Ur SIS. After Legrain 1936, cat. no. 85.

A2. Photo of sealing, Ur SIS. Author’s Photo, Penn Museum, Philadelphia, 31.16.642.

B2. Photo of sealing, B4. Photo of sealing, Ur SIS. Author’s photo, Ur SIS. Author’s photo, Penn Museum, Philadel- Penn Museum, Philadelphia, 33.35.266. phia, 33.35.267. Fig. 9. Ur sealings.

A1. Drawing of sealing, Ur SIS. After Legrain 1936, cat. no. 474.

B1. Drawing of sealing, Ur SIS. After Matthews 1993, cat. no. 10.

B3. Drawing of sealing, Ur SIS. After Legrain 1936, cat. no. 79.

C1. Drawing after Matthews 1993, cat. no. 118.

C2. Photo of sealing, Ur SIS. Author’s photo, Penn Museum, Philadelphia, 31.16.653.

A2. Photo of sealing, Ur SIS. Author’s photo, Penn Museum, Philadelphia, 33.35.270.

B2. Photo of sealing, B3. Drawing of sealing, Ur SIS. Author’s photo, Ur SIS. After Matthews Penn Museum, Philadel1993, cat. no. 22. phia, 31.16.640. Fig. 10. Ur sealings.

Materiality and Mark-Making

A1. Drawing of Sealing, Ur SIS. After Legrain 1936, cat. no. 87.

A2. Drawing of Sealing, Ur SIS. After Legrain 1936, cat. no. 97.

405

A3. Drawing of Sealing, Ur SIS. After Matthews 1993, cat. no. 91.

Fig. 11. Ur Sealings.

A1. Drawing of sealing, Ur SIS. After Legrain 1936, cat. no. 110.

B1. Drawing of sealing, Ur SIS. After Matthews 1993, cat. no. 9.

C1. Drawing of sealing, Ur SIS. After Matthews 1993, cat. no. 3.

A2. Photo of sealing, Ur SIS. Author’s photo, British Museum, London, 1930.12.13; 413.

B2. Photo of sealing, Ur SIS. Author’s photo, Penn Museum, Philadelphia, 31.16.630.

C2. Photo of sealing, Ur SIS. Author’s photo, British Museum, London, 1930.12.13.453.

Fig. 12. Ur Sealings.

A. Drawing of sealing, Ur SIS. After Legrain 1936, cat. no. 169.

B. Drawing of sealing, Ur SIS. After Legrain 1936, cat. no. 381.

Fig. 13. Ur sealings.

406

Sarah J. Scott

and again as a more discrete symbol combined with other signs. Such markings also occur on scribal training tablets, as discussed below. A related elongated form with central circles appears on incised sealings, which I call the legcross, and has a seal-impressed version found on both abstract and figural impressions, as does the DU sign (Fig. 10). Other incised sealings carry marks similar to GAN2 signs (Fig. 11). In addition to the GAN2 sign, the edinnu symbol (Fig. 12) may allude to field cultivation or administration, as discussed by both Matthews (1993: 41–50) and Legrain (1936: 39). Even the tiny notchings inside the borders are similar to the seal-impressed Edinnu elements. The organization, payment, and administration of human labor and fields are the primary concerns of the Archaic tablets. Thus these signs retain a visual link to such economic arenas. Figural counterparts to such economic concerns may be found in scenes depict­ ing procession toward temple facades, offering scenes, or banquet imagery, as they may relate to ration disbursement or consumption. Such imagery is found on Ur figural seal imagery (Figs. 7 and 13) and larger scale objects from a variety of sites (Figs. 8 and 14). One particular plaque from Tell Asmar depicts a clear link between field cultivation and banqueting.

A. Plaque with banquet scene, Tell Agrab Shara Temple. Image from Oriental Institute, Registration Number A18073, Accession Number 2111, Field Number Ag.35:668.

B. Plaque with banquet scene, Tell Agrab Shara Temple. Drawing after Boese 1971, AG1.

C. Plaque with plowing and procession scenes, Tell Asmar Square Temple. Drawing after Boese 1971, AS1.

D. Plaque with banquet and recumbent bovines, Tell Asmar Abu Temple. Drawing after Boese 1971, AS3.

Fig. 14. Early Dynastic sculpture.

The quadrant mark is yet another incised mark that occurs with frequency (Fig. 15). It takes a form similar to a pinwheel element found in the abstract and, by extension, the figural sealings. Occurring in various versions, the pinwheel mark and abstract element provide an iconographic link to one of the most common figural themes found in the Ur corpus—the contest scene. Lion and bovid combatants appear, sometimes accompanied by a hovering eagle and/or human actors in varying number and design. Contest scenes and the pinwheel motif are among the most common figural and abstract elements in the corpus, and have highest occurrence of duplicate sealings. Such contest images can be compared to Early Dynastic temple plaques, and even images of the hero grappling with animals (Fig. 16). I suggest these images may be related to temple fields or labor administration, as impressions

Materiality and Mark-Making

407

A1. Drawing of sealing, Ur SIS. After Legrain 1936, cat. no. 78.

A3. Drawing of sealing, Ur SIS. After Legrain 1936, cat. no. 77.

A2. Photo of sealing, Ur SIS. Author’s photo, Penn Museum, Philadelphia, 33.35.278.

A4. Photo of sealing, Ur SIS. Author’s photo, Penn Museum, Philadelphia, 33.35.263.

B1. Drawing of sealing, Ur SIS. After Matthews 1993, cat. no. 57.

B3. Drawing of sealing, Ur SIS. After Legrain 1936, cat. no. 274.

B5. Drawing of sealing, Ur SIS. After Legrain 1936, cat. no. 256.

B2. Photo of sealing, Ur SIS. Author’s photo, Penn Museum, Philadelphia, 33.35.273.

B4. Photo of sealing, Ur SIS. Author’s photo, Penn Museum, Philadelphia, 31.16.682.

B6. Photo of sealing, Ur SIS. Author’s photo, Penn Museum, Philadelphia, 33.35.452.

Fig. 15. Ur sealings.

408

Sarah J. Scott

A. Plaque with contest scene, possibly from Fara. Drawing after Boese 1971, F4.

B. Plaque with wrestlers/contest scene, Khafadje Nintu Temple. Drawing after Boese 1971, CN2.

C. Limestone cup, likely from Uruk. Image from the British Museum, London, 118465. Licensed under Creative Commons Attribution-NonCommercial-ShareAlike 4.0 International (CC BY-NC-SA 4.0) license. Fig. 16. Early Dynastic sculpture.

bearing this imagery are found in SIS contexts with tablets recording similar types of transactions. Additionally, Donald Hansen drew similar conclusions regarding the archaeological context of contest scene impressions from Nippur (Hansen 1971: 48). Imagery of the contest and heroes grappling with rampant animals surely is a statement about control over resources. 13

Constructing Context: Ur Seal Imagery and Archaeology This formal analysis of markings on the Ur SIS sealings illustrates the variety of image types. Much of this variety can be subjected to semantic groupings which were likely used for particular communicative purposes. I will discuss some of the recent work on the archaeological context of the sealings in order to illustrate the 13. Is the contest scene a form of display by the emerging elite? Rather than seek a particular identification, it has been suggested the hero grappling with animals may embody emerging concepts of kingship (Costello 2010).

Materiality and Mark-Making

409

chaine operatoire of these objects. It is through an understanding of this component of the sealings function that the meaning of the imagery may be understood against the larger social backdrop of early fourth millennium BCE Mesopotamia. Based on Benati’s recent re-assessment of the depositional sequence of strata, the sealings I have discussed above are likely from an Early Dynastic I landfill and subsequent surface fill. 14 A high percentage of door sealings from one loci (RC4c; F. 1011), and higher percentage of mobile container sealings from another (SIS 5/4, RC4a/b), suggests there was large-scale warehousing at one place and moderate mobilization of containers and some warehousing at another. Since both sets of sealings were discarded in the same context, it is likely they were all being produced, used, and discarded by one large and tightly controlled bureaucracy that was receiving goods, storing them in a central storehouse, and then redistributing them. The archaic tablets suggest these goods were being allocated as wages, salaries, or rations. 15 The picture that emerges suggests sealing imagery related to both the loci of the sealing process, as well as a change in practice over time. Sealings bearing figural impressions more frequently sealed mobile containers, while the abstract sealings more frequently sealed rooms. 16 Additionally, door-peg sealings came from a differ­ ent location (Benati 2015). Over time there was an increase in duplicate sealings, and an increase in the use of jar stoppers (Benati 2015: 18). I have also determined there was an increase in the use of abstract seals over time. The increased use of door-peg sealings in the later part of the Early Dynastic I period was likely a result of an economic change from a kin-based household model to a larger temple-based model. How do the categories of imagery then relate to the places and times of the administrators? It seems that that figural seals were being used at Ur from the Jemdat Nasr through the early Early Dynastic I period, followed by an explosion of seal use in the later Early Dynastic I period, a hallmark of which was the use of figural and incised sealings used in one loci and abstract sealings being used to seal rooms at another. Certain figural themes may also have been related to particular administrative loci. For example, figural imagery depicting combat, heraldic eagles, and herding are found in very high numbers from particular fill loci, while those from what Benati has identified as the Administrative Quarter seem to be strictly abstract in nature, very small in quantity, and accompanied by a large number of tablets. Even though certain distribution patterns can be identified, it is clear that fig­ural, abstract, and incised sealings were all being used alongside each other, 17 14.  Benati’s recent analysis of the Ur SIS and associated finds and architecture alongside unpublished excavation records suggests that a depositional sequence can be identified (Benati 2015). Based on his conclusions, there are multiple fill levels and building sequences from the Jemdat Nasr to Early Dynastic IIIa periods. Jemdet Nasr period dump (RC 1, RC 1-2), an early Early Dynastic I landfill context (RC 2-3), a later Early Dynastic I Landfill (RC 4a-b; SIS 5/4) and surface preparation deposit (RC 4c; F. 1011) for a new building, and an even later Early Dynastic I-II maybe IIIa building (RC 4d—two building phases) and associated rubbish tips (RC 4d-e). 15.  This has been determined by Lecompte’s reassessment of the Archaic Tablets (2013: 8). 16.  This data comes from my own assessment of the figural vs. abstract sealings. 17.  There are also examples of duplicate sealings found across loci. Some suggest that the presence of duplicate sealings across loci indicates the movement of the seal user across space; however there is no reason to discount the possibility that it was the seal impressions were moved to various loci in order

410

Sarah J. Scott

while cuneiform documents too, with increasing frequency, were being used within the same administration. 18 Over the course of time, the roles of sealer and scribe di­verged; the archaeological context at Ur supports this change in practice. One function of the abstract and incised sealings was as a communicative tool for the conveying of ideas about the commodities to the bookkeepers. Once the Administrative Quarter was built, it is likely sealings were used in different locations. We see a transition from an office where sealing, scribal training, and some bookkeeping is happening to a moment when only bookkeeping is occurring in an isolated context.

Ur Archaic Texts The sealings were part of an administrative toolkit which also included tablets. Burrows, and more recently Lecompte, have identified specific semantic tablet types present in the Ur corpus; such texts are considered to be transitional between the Uruk Archaic and Fara texts. Semantic categories from Ur include lexical lists (god lists, cultic personnel lists, city lists). A number of tablets from Ur contain more traditional signs/textual documentation and drawn, or incised motifs bearing figural similarity to the incised sealings, such as UET 2 299, a Lu2 A text, with an animal leg and hoof drawn on the obverse. 19 A god list, as well as school and practice tablets, and cereal accounts illustrate similar practices. 20 In fact, a number of the signs (such as še-ninda, UB, LU2) found on these tablets are almost identical to those on the incised sealings. 21 Other Archaic Ur texts suggest that staple items such as bread, beer, dough, and vessels containing liquids were being distributed, frequently as rations. The signs that appear on the abstract sealings include signs related to these types of commodities. It is possible therefore that the imagery on these sealings indicates to be recorded elsewhere, hence it would have been possible that the seal image imparted information about the container or commodity itself. 18.  Although there are many examples of seal images that can help us to further understand the function of these technologies, one in particular is useful to highlight. It depicts the Edinnu, a pinwheel, UB, signs for Larsa (U4 AB) and Ur (URI3 + AB), and vessels as filler elements. This seal was used multiple times as evidenced in twenty-eight sealings, of which all diagnostic examples were door pegs. The sealings were found in the construction fill AND administrative quarter context, indicating it may have been an essential visual formula for large-scale storage perhaps at multiple locations. 19.  Other such texts include: UET 2 253 (U. 14180), UET 2 264 (U. 14718), UET 2 299 (U. 14895), UET 2 53 (U. 12520), UET 2 51 (U. 12516), UET 2 55 (U12521), UET 2 99 (U. 12832), UET 2 157 (U. 12896), UET 2 232 (U. 13004), and UET 2 233 (U. 13088), the latter with a seal impression as well. 20.  Many of these early accounts from Jemdat Nasr and Ur that deal with the cultivation, processing, and distribution of cereals include the signs that maintain an iconographic relationship to the objects of reference. This is a point that scholars have emphasized as being typical for early accounting practices (Englund 1991). 21. Lecompte, Benati, and Clevenstine’s paper from this conference, “Non-Administrative Documents …” presents a detailed look at the distinctions between the incised sealings and non-traditional Archaic Ur texts. There are many insightful examples that indicate the scribes were drawing on their tablets. In their words: “Legrain already noticed that some of the symbols incised on the sealings have cuneiform significance, such as those presented here (MUNU4, SAL), and can even tally with personal names (for instance, here, Lugal-x-NI and giri16-x-sal). A systematic study of these inscriptions—still to be accomplished—most certainly connected to economic operations related to storage and mobilization of staple products, will shed more light on this peculiar administrative custom and its relationship with the use of cylinder seals. In the meantime, we can hypothesize that literate operators, possibly scribes, were involved in the operation of sealing containers and storerooms in ED I Ur.”

Materiality and Mark-Making

411

for what parties or what events these foodstuffs would be intended. 22 For example, texts from the later “Archaic Room” that deal with beer distribution may record wages for workers for the AB, workers associated with the fields of the NANNA sacred household, and messengers and even workers that were sent to Uruk (Benati and Lecompte 2016: 12). These commodities, activities, and institutions as mentioned in the texts have visual complements found in the seal imagery. The high number of references in the tablets to the administration of the fields of NANNA likely find counterparts in the high instance of EDINNU and GAN2 motifs on the seals. Clearly, the hands of the scribes and those doing the sealing were the same. These tablets’ context is also useful for our understanding of the chain operatoire, as they indicate that scribal training and sealing of mobile goods were going on in the same spaces. The physical existence of these incised sealings within the body of SIS clay artifacts is essential in establishing the relationship between seal-impressed imagery and writing. Additionally, the temporal location of these sealings and the Ur Archaic texts establishes this early Early Dynastic context at Ur as being distinct from the Uruk texts prior and the Fara traditions of the following period. It is a moment when the relationship between images and proto-cuneiform is diverging.

Materiality, Objects, and Humans: Theoretical Applications for the Ur Sealings I propose that during the time of the generation of the SIS there is a fundamental change in how administrators and consumers were using and absorbing images and writing. The incised sealings can be considered as a link between image-based and sign-based image systems, and they can help us reconstruct the process of record keeping and commodity distribution. It is likely, as imagery found on the sealings and forms of writing used on the sealings and tablets diverged, the resultant two systems were meant to be understood by different audiences; we can approach such an understanding through an examination of how the medium of clay functioned as a matrix for image systems. First, clay was an agentive medium; sealings as clay objects could make things happen. Clay as a material possessed a particular power. Objects need not be figural to exercise agency (Gell 1998); the materiality of sealings would have been cause for effective participation in Mesopotamian society (Meskell 2005; Miller 2005). As a ubiquitous and easily obtainable material, it played a role in the making of vessels, shaping of plaques and sculptures, and embellishment of architecture. 23 22.  The process of sealing and record keeping and the many ways this process might have played out has been the topic of many a scholar’s research (particularly Dittman 1986; Rahmstorf 2012). Steinkeller (2004) in general has proposed a two-tiered operational chain in the administration process. Benati (2015) has identified such a process in his re-evaluation of the archaeology at Ur, where the first order tasks such as the sealing and unsealing of storage areas and accounting and sorting of inventories, and then secondly the bookkeeping of official records. Specifically in Benati and Lecompte’s reassessment of the materials from SIS 5/4 and F.1001, these operations seem to have been going on side by side. Additionally, the texts suggest these foodstuffs were being distributed by high-ranking officials such as overseers (nu-banda), temple cooks (engiz), and temple administrators (sagga), and they were possibly being distributed to specialized crafts men such as smiths (simug) and builders (sidum), and that they were attached to waged workers working specifically for the institution (Benati and Lecompte 2016: 12). 23.  It is referenced frequently in literary and historical texts in omens and incantations (Graff 2014: 371).

412

Sarah J. Scott

Textual sources present clay as the material of human creation; goddesses used clay to fashion the first humans. Further, molds and kilns used to shape and bake clay were sometimes referred to as birthing wombs. Bricks, which were made of clay, also may feature in literary analogies for placental material. Clay bricks may have acted a substitute, or double for newborn babies during the first week of birth (Graff 2014: 381). Sensory processes would have been at play while handling a sealing. Holding wet clay in one’s hand, placing it on a container and rolling or impressing a seal, or pushing a stylus through the clay, would have been an experience that became inexorably linked to the administrator’s physical body and ultimately their identity. Additionally, the feeling of the bumps on the clay, or scratchings of the stylus point, even olfactory and auditory stimulation would have become part of the sensory experience of controlling and using the goods contained therein. For the Ur material, we have seen that there were multiple accounting devices, all of which were made with clay. The powers of clay as a medium linked the devices through its status as an agentive matrix. Whether it was the sealer, scribe (or both!), ration distributor, or consumer, all experienced a physical participation and thus connection to the chaine operatoire. Second, sealings are agentive objects that acted as communicative mediators through object entanglement. As discussed above, the process of sealing and opening a commodity was likely enacted by multiple individuals. As a container and the sealing passed through the hands of the sealer, the distributor, was viewed by a consumer receiving rations, recorded by yet another administrator, and discarded, there were multiple opportunities for meaning to be mapped onto the clay object. As has been shown through the application of object entanglement and practice theory during the Uruk period (Ross 2014) the process of creating meaning through imagery on clay sealings was likely transferred to protocuneiform signs. Through a technological process involving multiple parties competent in “reading” seal-impressed imagery and early signs, the first lexical lists were created. People had to work together to develop these (ultimately) universally accepted signs. Two important points based on Ross’s work would still hold true, I argue, at Ur for this early part of the Early Dynastic period: Not only did scribes and seal-cutters have similar working knowledge of visual vocabularies, but it was based on knowledge co-opted from two social spheres: producers and consumers. Third, the sealings participated in a system of “object pluralism,” where the imagery contained on them allowed for communication across multiple social groups. Because the imagery was extant on clay, certain functionalities (actions through which the sealings were used or exchanged) were inherent in the sealing (as above), but the image may have had multiple meanings based on the visual and/ or textual literacy of the viewer. Additionally, the image meaning may also have possessed additional object pluralism when present on another medium (such as large-scale sculpture). The sealings were intended to be viewed and analyzed. Our analysis today would be different than that of the third millennium BCE audiences. We might attempt to “read” them as a text as an Assyriologist—or analyze the visual imagery as an art historian—but still, these experiences around the objects would be distinct from those of the scribe or consumer of antiquity. In fact, I would argue that the textual reading/figural analysis binary may have been much more fluid in the sealings’ ori-

Materiality and Mark-Making

413

ginal context. It is likely that the very process of seeing imagery and text together in ancient Mesopotamia is not so straightforward as we might think today (Crawford, 2014). In fact, this is a moment, it has been argued by Suter (2014), Woods (2010), Rothmann (2016) and others, that we can read Mesopotamian images as if they were a text, and I would argue that some of these early texts can be understood as image—specifically when considering the incised and abstract sealings from Ur. The sharing of the image, through the material of clay and their functionalities allowed for multiple viewing experiences and interpretations. Objects carry ideas through their materiality and are then transferred to the next individual with which the object interacts. We can accept that the administrator would have conceived of the sealing’s agency, and that the recipient of goods or rations would have a different understanding of the sealings’ affective qualities. It is likely sealings were created and removed in temple or ritual settings. Many objects in ancient Mesopotamia were ritually consecrated, and were believed to be animate, possessing the same affective powers as humans (Porter 2010; Bahrani 2002). The agency of the seal image likely possessed multiple levels of affectiveness given that the images also appeared on larger-scale objects (as mentioned in the above section). The meaning of the eagle hovering over bovines is multivalent when appearing on a clay sealings or large copper temple sculpture, but when being seen in both contexts, that image’s agency was certainly compounded in the mind of the viewers. The subsequent life of the sealing’s impression continued to affect and be seen by others outside their immediate context of creation. The sealing was both object and image. Even if the sealing was not seen by non-temple personnel, the imagery it carried was. Through larger-scale objects, imagery was visually accessible to laborers and other segments of the society outside scribal circles. Also relevant in these contexts is the ability for the temple, or sculptures, to then emit affective properties to the community, and the fact that social institutions may exercise agency (Winter 2007). Such individuals, as representing the larger population of early Dynastic Mesopotamia, would have seen larger-scale temple imagery upon occasions such as ration dispersal, festivals, or mass banqueting. We are understanding new ways that imagery in the ancient Near East may have been a conduit for the divine and that the viewing of certain forms may have heightened spiritual awareness (Shafer 2016; Stein 2016). The encounter with the divine, as we are now coming to understand it, was very much based on the agency of cult statuary (Evans 2012; Dick 2005). Therefore, if the agentive quality of the image encountered first from a monumental object, then perhaps sealing imagery viewed afterward imparted even greater value to a commodity or disbursement. Such a phenomenon of “commodity branding,” must have been utilized during the early Early Dynastic period at Ur. 24 The creation of the image in the mind’s eye, the impressing or making physical that image, the functioning of the sealing, and the subsequent viewing of that sealing by multiple parties was all part of a recursive act that helped to create an understanding across multiple social groups in early Mesopotamia.

24.  This would have been a large-scale process that we certainly see in chain of operations at Ur (sealings from Ur were dry, indicating multiple days between the time of packaging and redistribution [Benati 2015: 8]).

414

Sarah J. Scott

Conclusion What I have traced here, through the imagery found on the early Early Dynastic sealings from Ur, is a fluidity of vocabulary between visual elements that are in­scribed, incised, and seal-impressed, that represent a range of elements from sculpted, figural objects to written semiotically coded icons. This range of elements through the seal’s visual message and its presence on functional, living objects allowed for multiple persons to identify with and create their own identities within a diverse society. As roles developed, administrators would have been further re­ moved from subsistence work, but would have kept a link with such processes and the populations involved through their image-making and writing tasks. The processing, depositing, storing, redistribution, and use of goods during this period was happening more frequently outside kin-based social networks. The main foci of the institution from which these sealings came were administering staple goods and cultivated land. Staples were accepted and distributed, land was allotted and administered, labor was organized, and animal husbandry was carried out. There was a huge diversity of roles involved in these processes. The viewing of the range of imagery at Ur, and its de-coding and understanding by each individual in the society would have served to promote both the goods themselves, but also those distributing the goods. 25 I also suggest that such image-based branding would have promoted the producer and record keeper as well, bringing all of society into an understanding of and appreciation for their individual roles. We cannot ever know the specifics of how each individual was affected by these images and thus how their lives were lived out based on these artifacts, but it is essential that we consider the individual’s encounter with the object, and that each may have had a unique experience and understanding around the imagery and the clay sealing. The trajectory of symbol making in ancient Mesopotamia follows from imagebased systems of sealing to sign-based, and ultimately grammatically understood system of writing. The mid-fourth to late third millennium BCE period represents a crucible of activity along this trajectory. The sealings from Ur can in fact help us to understand how human agents acting within this time period took part in it. Specifically, the incised marks exist within a state of symbolic limbo, carrying concepts that relate to a newly emergent writing system but being used within a larger image-based system.

Bibliography Appadurai, A. 2015 Mediants, Materiality, Normativity. Public Culture 27/2: 221–237. Bahrani, Z. 2002 Performativity and the Image: Narrative, Representation, and the Uruk Vase. Pp. 119–171 in Gender Through Time in the Ancient Near East, ed. D. Bolger. Lanham, MD: Altamira Press. 25.  Sherratt (2004) and Stein’s (1998) work suggest that sealings like those found in the contexts at Ur were used as branding or quality stamps, and that the viewing of them would have acted as semiotic tools for promoting not only the goods themselves, but those parties distributing them. Benati suggests that the incised and abstract seal images were being used inside a hyper-bureaucratic system, as they would have only been “legible” to the administrators within that system (2015).

Materiality and Mark-Making

415

Benati, G. 2015 Re-modeling Political Economy in Early 3rd Millennium BC Mesopotamia: Patterns of Socio-Economic Organization in Archaic Ur (Tell al-Muqayyar, Iraq). Cuneiform Digital Library Journal 2015:2 See http://www.cdli.ucla.edu/pubs/ cdlj/2015/cdlj2015_002.html. Benati, G. and Lecompte, C. 2016 New Light on the Early Archives from Ur: The «Ancient Room» Tablet Hoard. Pp. 13–30 in Proceedings of the 9th International Congress on the Archaeology of the Ancient Near East, 9–13 June 2014, Basel, Volume 3, eds. R. A. Stucky, O. Kaelin, and H.-P. Mathys. Wiesbaden: Harrassowitz Verlag. Bourdieu, P. 1977 Outline of a Theory of Practice. Cambridge, U.K: Cambridge University Press. Charvát, P. 2016 Signs from Silence: The Beginning of History of the Sumerian Ur. Unpublished paper given at the 62nd Rencontre Assyriologique Internationale, Philadelphia, 2016. Costello, S. K. 2010 The Mesopotamian ‘Nude Hero’: Contexts and Interpretations. Pp. 25–35 in The Master of Animals in Old World Iconography, eds. D. B. Counts and B. Arnold. Budapest: Archaeolingua. Crawford, C.D. 2014 Relating Image and Word in Ancient Mesopotamia. Pp. 241–264 in Critical Approaches to Ancient Near Eastern Art, eds. M. H. Feldman and B. A. Brown. Berlin: de Gruyter. Dick, M.B. 2005 The Mesopotamian Cult Statue: A Sacramental Encounter with Divinity. Pp. 43–67 in Cult Image and Divine Representation in the Ancient Near East, ed. N. H. Walls. Boston, MA: American Schools of Oriental Research. Dittmann, R. 1986 Seals, Sealings and Tablets: Thoughts on the Changing Patterns of Administrative Control from the Late-Uruk to Proto-Elamite Period at Susa. Pp. 332–366 in Ğamdat Nasr. Period or Regional Style? Papers given at a Symposium held in Tübingen November 1983. Beihefte zum Tubinger Atlas des Vorderen Orients. Reihe B 62, eds. U. Finkbeiner and W. Rollig. Wiesbaden: Reichert Verlag. Englund, R. K. 1998 Texts from the Late Uruk Period. Pp. 15–217 in Mesopotamien. Späturuk-Zeit und Frühdynastische Zeit, eds. P. Attinger and M. Wäfler. OBO 160/1. Freiburg: Universitätsverlag Freiburg Schweiz Vandenhoeck & Ruprecht Göttingen. Evans, J. M. 2012 The Lives of Sumerian Sculpture. Cambridge: Cambridge University Press. Gell, A. 1998 Art and Agency: An Anthropological Theory. Oxford: Oxford University Press. Graff, S. 2014 Sexuality, Reproduction and Gender in Terracotta Plaques from the Late Third– Early Second Millennia BCE. Pp. 371–390 in Critical Approaches to Ancient Near Eastern Art, eds. M. H. Feldman and B. A. Brown. Berlin: de Gruyter. Green, M. and Nissen, H. 1987 Zeichenliste der archaischen Texte aus Uruk. ATU 2. Berlin: Mann. Hansen, D. P. 1971 Some Early Dynastic I Sealings from Nippur. Pp. 47–54 in Studies Presented to George M.A. Hanfmann, ed. D. G. Mitten. Cambridge, MA: P. von Zabern.

416

Sarah J. Scott

Jacobsen, T. 1957 Early Political Development in Mesopotamia. ZA 52: 91–140. Latour, B. 2007 Can We Get Our Materialism Back, Please? Isis 98/1: 138–142. Lecompte, C. 2013 Archaic Tablets and Fragments from Ur (ATFU) from L. Woolley’s Excavations at the Royal Cemetery. Nisaba 25. Messina: DiCam. Lecompte, C., Benati, G., and Clevenstine, E. 2016 “Non-administrative Documents from Archaic Ur and from the Early Dynastic I-II Period.” Unpublished paper given at the 62nd Rencontre Assyriologique Internationale, Philadelphia, 2016. Legrain, L. 1936 Archaic Seal Impressions. UE 3. Oxford: Trustees of the British Museum and of the University Museum, University of Pennsylvania, Philadelphia; Oxford University Press. Matthews, R. J. 1993 Cities, Seals, and Writing: Archaic Seal Impressions from Jemdet Nasr and Ur. Berlin: Gebr. Mann Verlag. Meskell, L. 2005 Archaeologies of Materiality. Oxford: Blackwell. Michalowski, P. 2003 The Earliest Scholastic Tradition. Pp. 450–478 in Art of the First Cities: The Third Millennium BC from the Mediterranean to the Indus, ed. J. Aruz. New York: The Metropolitan Museum of Art. Miller, D. 2005 Materiality: An Introduction. Pp. 1–50 in Materiality, ed. D. Miller. Durham, NC: Duke University Press. Porter, A. 2010 Akkad and Agency, Archaeology, and Annals: Considering Power and Intent in the Third-Millennium BCE Mesopotamia. Pp. 166–80 in Agency and Identity in the Ancient Near East, eds. S. R. Steadman and J. C. Ross. London: Equinox. Postgate, J. N. 1986 The Transition from the Uruk to the Early Dynastic: Continuities and Discontinuities in the Record of Settlement. Pp. 90–106 in Ğamdat Nasr. Period or Regional Style? Papers given at a Symposium held in Tübingen November 1983. Beihefte zum Tubinger Atlas des Vorderen Orients. Reihe B 62, eds. U. Finkbeiner and W. Rollig. Wiesbaden: Reichert Verlag. Rahmstorf, L. 2012 Control Mechanisms in Mesopotamia, the Indus Valley, the Aegean and Central Europe, c. 2600–2000 BC, and the Question of Social Power in Early Complex Societies. Pp. 311–326 in Beyond Elites: Alternative to Hierarchical Systems in Modelling Social Formations. International Conference at the Ruhr-University Bochum, Germany, October 22–24, 2009. Bonn: Habelt. Ross, J. C. 2014 Art’s Role in the Origins of Writing: The Seal-Carver, the Scribe, and the Earliest Lexical Texts. Pp. 295–318 in Critical Approaches to Ancient Near Eastern Art, eds. M. H. Feldman and B. A. Brown. Berlin: de Gruyter. Rothman, M. 2016 “State and Private Sectors in Old Babylonian Ur.” Unpublished paper given at the 62nd Rencontre Assyriologique Internationale, Philadelphia, 2016.

Materiality and Mark-Making

417

Scott, S. J. 2005 Figure, Symbol, and Sign: Semiotics and Function of Early Dynastic I Cylinder Seal Imagery from Ur. Ph.D. dissertation, University of Pennsylvania. Shafer, A. 2016 “Topographical Peripheries in the Assyrian Center: Abstract Ornament and the Sacred in Architecture.” Unpublished paper given at the 62nd Rencontre Assyriologique Internationale, Philadelphia, 2016. Sherratt, A. 2004 Material Resources, Capital, and Power: The Coevolution of Society and Culture. Pp. 79–104 in Archaeological Perspectives on Political Economies, eds. G. Feinman and L. M. Nicholas. Salt Lake City: University of Utah Press. Stein, D. 2016 Reflections on the Banquet Scene: An Interdisciplinary Approach to Thematic Investigations. Unpublished paper given at the 62nd Rencontre Assyriologique Internationale, Philadelphia, 2016. Stein, G. J. 1998 Heterogeneity, Power, and Political Economy: Some Current Research Issues in the Archaeology of Old World Complex Societies. Journal of Archaeological Research 6: 1–44. Steinkeller, P. 2002 Archaic City Seals and the Question of Early Babylonian Unity. Pp. 249–257 in Riches in Hidden Places: Ancient Near Eastern Studies in Memory of Thorkild Jacobsen, ed. T. Abusch. Winona Lake, IN: Eisenbrauns. 2004 The Function of Written Documentation in the Administrative Praxis of Early Babylonia. Pp. 65–88 in Creating Economic Order: Record-keeping, Standardization and the Development of Accounting in the Ancient Near East: A Colloquium Held at the British Museum, November 2000, eds. M. Hudson and C. Wunsch. Bethesda, MD: CDL Press. Suter, C. E. 2014 Human, Divine, or Both? The Uruk Vase and the Problem of Ambiguity in Early Mesopotamian Visual Arts. Pp. 545–568 in Critical Approaches to Ancient Near Eastern Art, eds. M. H. Feldman and B. A. Brown. Berlin: de Gruyter. Szarzynska, K. 1992 Names of Temples in the Archaic Texts from Uruk. ASJ 14: 269–287. Winter, I. J. 2007 Agency Marked, Agency Ascribed: The Affective Object in Ancient Mesopotamia. Pp. 42–69 in Art’s Agency and Art History, eds. R. Osborne and J. Tanner. Malden, MA: Wiley. Woods, C. 2010 The Earliest Mesopotamian Writing. Pp. 33–50 in Visible Language: Inventions of Writing in the Ancient Middle East and Beyond, ed. C. Woods. Oriental Institute Museum Publications 32. Chicago: Oriental Institute. Yoffee, N. 1995 Political Economy in Early Mesopotamian States. Annual Review of Anthropology 24: 281–311.

Eclipse of a King: A New Interpretation of the Bull-Headed Lyre JoAnn Scurlock Chicago, IL

Among the most spectacular of finds in the tombs at Ur are musical instruments whose original purpose and relevance to context remain something of a mystery. Particularly intriguing is the sideboard of a bull-headed lyre that shows itself being played by an equid as a part of one of four scenes representing anthropomorphized animals engaged in apparent feasting activity. According to the Penn Museum Royal Cemetery Website 1, this is a representation of a royal funeral beginning with a nude hero grappling bulls to represent royal control over nature. A hyena (?) presents the meat and the lion pours the drinks while equids and a bear provide musical accompaniment. In the last stage of the ritual, the deceased meets the scorpion man at the entrance to the Netherworld, so in all we have the “human cycle of the king’s control of nature,” so life, followed by death as represented by the funeral and entry to the other world, all of this presided over by the god Šamaš whose head is allegedly the bull on the lyre. From where I sit, this is theory-generated drivel. The so-called nude hero is actually a Laḫmu, and his struggle with the bull has nothing to do with control over nature by the king or anyone else. 2 It was the boatman Siluši and the gatekeeper Bidu, and not the scorpion man, whom ordinary mortals expected to encounter on the way to the Netherworld, and Šamaš, although handling cases that arise between the dead and the living, was not in the admissions department. 3 In our field, we all too often try to solve conundrums or to interpret mysterious objects or to make sense of ambiguous or uncommunicative data by immediate recourse to theory. The advantages to this strategy are many, including rapidity of produced results, guaranteed acceptance at least by fellow proponents of the same theory, and fast track to hiring and promotion. The main casualty is the truth. As with torture, of which prematurely applied theory is essentially a form, it is not a proof procedure but a means by which the torturer seeks to convince himself, and his victim, of the truth of the accusations against him. What is needed is a return to the Euclidian hypotheses upon which our Aristotelian theories should have Author’s note: This paper has particularly benefited from the input of a practicing astronomer, Ed Guinan. 1.  https://www.penn.museum/sites/iraq/?page _id=58. 2.  For details, see Scurlock forthcoming a. 3.  For details, see Scurlock 1995: 1886–1888.

419

420

JoAnn Scurlock

been based. In short, there is no substitution for the down and dirty, frustrating, and lengthy process of getting rid of preconceived notions and laboriously sifting through various possible explanations for observed phenomena, both learning to recognize red flags that indicate failure to fit the data and actually rejecting those hypotheses that elicit them. Even with perfect fits, there is still room for refutation if a better hypothesis can be proposed. I hasten to add that this is not to say that you are ever left with nothing to show for your efforts. Even without a perfect fit, the best available hypothesis is to be retained, if only until a better hypothesis can be suggested. A good start was made in this direction by F. A. M. Wiggermann (1995: 79–83) who noted the presence of the scorpion-man on the bottom panel. Since these were the guardians of sunrise and sunset passed by Gilgamesh on his journey to Ut-­napištim, and since these are also mentioned in the text accompanying the Babylonian Map of the World, he suggests that we are dealing with phantasmata imagined at the edges of the known world to whom the tomb inhabitant is to present the lyre as a gift. Unfortunately, most of the figures shown on the lyre are real animals and not phantasmata; neither are the animals in question invariably either wild or danger­ ous. Thus, the available data, such as they are, do not support the hypothesis, but it must stand unless a better hypothesis is on offer. I propose to provide precisely this better hypothesis. Let us begin by seeing whether there is not something about the fuller context of the lyres that would help us in our search. One of the most spectacular finds from the tombs at Ur is the famous statue of a goat entangled in a tree. It is hard for a Biblically informed observer such as myself to look at this statue without thinking darkly of scapegoats and rams caught in thickets. In support of this instinct, it would seem that the goats involved prominently in the Ištar and Dumuzi rituals represent Dumuzi as an offering to the goddess. 4 This is in a context of Dumuzi, Ištar’s substitute to Ereškigal, taking the ills of participants in the ritual away with him to the Netherworld like so many sheep (Farber 1977: 152–153, lines 171–188). Since Dumuzi is directly associated with human kingship in ancient Mesopotamia, 5 a promising hypothesis is that the Ur burials represent the performance of a substitute king (šar puḫi) ritual. These are well known from the Neo-Assyrian period but traceable back to at least the Isin-Larsa period (the infamous incident of Erra-imitti and Enlil-bani), by which time it seems to have been fully institution­ alized. By this hypothesis, the statue of the goat caught in the thicket is an image of the substitute-king in this role, Dumuzi’s goat sacrificed to Inanna, the tree 6 whose branches enclose him in loving embrace. The suggestion itself is by no means new to this author (Scurlock 1995: 1885) but, since we are treating this as a hypothesis and not a theory, we need to test it by seeing whether there is some discernible difference between an ordinary royal burial and one for a substitute king and not just any difference, but one that would show up in the archaeological record. Neither is it sufficient to have such archaeologically discernible differences only in theory, they should at some point other than the 4.  This is the only obvious reason for the specific offering to Ištar of the goat’s heart (Farber 1977: 64–67, lines 20 and 29–30). 5.  For more on this subject, see Scurlock 2013. 6.  On the nexus between the Assyrian sacred tree, Mt. Ebiḫ and Inanna, see Scurlock forthcoming a.

Eclipse of a King

421

Ur tombs have appeared visibly in the archaeological record and/or been directly referenced by a text describing such a burial. We are in luck, having not only parts of actual rituals for royal burials on the one hand and substitute kings on the other, but references to the performance of substitute king rituals in letters reporting to the king, and even the remains of actual royal burials, some of them intact. What we learn, mostly from archaeological evidence, is that Assyrian kings were never buried with their queens, let alone any attendants (Andrae 1977 [1938]: 194–201 and ill. 1787–180). The queens, too, were intended to be buried alone, the rare exception to the rule being a Ruth and Naomi burial at Nimrud of a mother-in-law and daughter-in law of West Semitic origin. 7 Indeed, Assyrian queens and palace women were not even buried in the same city as their husbands, fathers, and brothers. By contrast, one letter reassures Esarhaddon that the substitute king has gone to his fate and been buried and adds that he was accompanied by his queen who died and was bejewelled, displayed, buried, and bewailed alongside her husband as a substitute for the king and for Šamaš-šumukin and to cancel all the bad portents (Parpola 1993 no. 352: 5–17). There is thus a significant difference between the burial customs associated with normal royal and substitute king burials in the same (Neo-Assyrian) period. Moreover, the simultaneous burial of hundreds of persons unrelated to the tomb owner, who were dressed in elaborate garments, and then probably drugged before being killed with an axe blow to the skull (Baasdgaard 2011), 8 distinguishes the Ur tombs from all other attested burials in Mesopotamia of any period. It seems to me that, in a culture where it is exceedingly rare for non-family members to be buried with family members in the same tomb and then typically the odd servant, Arabian Nights style, across the door, the performance of the šar puḫi ritual provides the best available hypothesis to explain the Ur tombs. We need, however, to hold off on the champagne. At this point, it is by no means obvious that, even if our hypothesis of the lyre’s context in the burial of a substitute king is correct, the lyres have anything to do with the substitute king ritual as such. Maybe every royal burial from its time period had a harp with this decoration on it and, since the substitute was supposed to be a king, he got a lyre. Maybe every king had a David who played for him and since, in substitute king rituals, the attendants got buried, the harp simply came along with. Since this is a hypothesis and not a theory, we may not simply leave this issue hanging; instead, we need to refine our hypothesis. We know that the substitute king ritual was, in principle at least, performed to take away an evil omen affecting the king that came in the form of an eclipse of the sun or moon during which the planet Jupiter was not visible and the direction of whose shadow portended evil for the Mesopotamian king and not enemy kings. 9 Thus, if we can establish some 7.  For the latest on these tombs and their contents, see Curtis 2008 and Hussein 2016. 8.  I would dispute Baadsgaard’s claim that the bodies were either heated before burial or treated with mercury sulphide. Both of these stigmata were almost certainly produced by the archaeological team. The research team based their findings on analysis of the skulls upon which fragile headdresses and helmets were transported to the museum for separation and conservation. It was the practice of the archaeologists to dig a pit around the headdresses and helmets to be recovered and to fill this pit with boiling wax, which eventually hardened, allowing safe removal from the ground. It so happens that sealing wax is red due to its high content of mercury sulphide. 9.  For details on the rules and attested mistakes made in applying them, see Ambos 2006.

422

JoAnn Scurlock

kind of relationship between harps and eclipses, we may at least hypothesize that there is some substitute king ritual-specific reason for the presence of lyres in our tombs. As luck would have it, we have a reasonably well preserved lunar eclipse ritual which does indeed prominently feature a harp not, curiously, being played by human hands but set out at the beginning and removed at the end of the ritual as a central and essential feature of it. The ritual is, unfortunately, from the Seleucid period. However, the fact that one of the recitations associated with it is a clearly updated version of a Sumerian original known from an Old Babylonian copy, about which we shall have more to say presently, allows us to proceed, with caution, to the analysis of this ritual to see if any enlightenment can be gleaned from it. I am speaking of the lunar eclipse ritual edited by Linssen, namely BRM 4 6 and its parallel BM 134701. More precisely, there are four rituals, two of which appear on each tablet. The second ritual of BRM 4 6 (BRM 4 6: 42′–55′ [Linssen 2004: 307–308]) duplicates the first ritual of BM 134701 (BM 134701 obv. 1′–16′ [Linssen 2004: 308]). These two rituals are undeniably directed at an eclipse, since there is specific mention of how far the moon is eclipsed at various points in the ritual when specific recitations are to be performed (BRM 4 6: 48′–53′ and BM 134701 obv. 7′–14′ [Linssen 2004: 308]). Curiously, the first (BRM 4 6: 1′–41′ [Linssen 2004: 306–307]) and fourth (BM 134701 obv. 17′–rev. 22′ [Linssen 2004: 308–309]) rituals do not involve harps and seem to contemplate either a two day ritual (BRM 4 6: 31′ [Linssen 2004: 307]) or a “day and night” eclipsing (BM 134701 rev. 20′ [Linssen 2004: 309]), the latter impossible for an actual eclipse. One might, in any case, wonder why there should be four rituals for eclipses, since on average only one of these will have been seen every year. The answer to this conundrum is the fact that there is such a thing as an eclipse season of a little over a month. Eclipses occurring in one season are very likely to be followed after a hiatus of five or six months by another eclipse. What this means is that, if you saw an eclipse in the Spring, you were likely to see another in the Fall and vice versa. The situation was similar with Summer and Winter. You thus needed at least two separate rituals, one for Spring/Fall and one for Summer/Winter to cover all bases. In addition, an eclipse in Fall was complicated by the fact that the Akkadian term “eclipse” also confusingly (for us) refers to points in the ancient Mesopotamian solar-lunar calendar in which the Sun, Moon, etc. were imagined as being disadvantaged. 10 In the Hellenistic period, these and other points were taken up with modifications and transformed into the dignities or houses of Horoscopic Astrology. Month 8 of the Mesopotamian calendar, right in the center of the Fall, was assumed to contain one of these disadvantage points for the moon, what we might term a “lunar depression” but for which they used a term which, in other contexts, clearly refers to a lunar eclipse. If an actual lunar eclipse were to occur in Fall, then, it would be particularly dangerous, and require more attention than was usually expended to deal with the lunar depression when occurring by itself. Similarly, an eclipse occurring in the Spring could hardly be exorcised without reference to the terrible dust storms that caused the Spring in Syria to be characterized as the earthquake season in the British Admiralty handbook (Admiralty

10.  For example, in KAR 178 rev. 4–5 (Labat 1939: 120–21), the 23rd of Tašrītu (the autumnal equinox) is called an “eclipse of Šamaš.”

Eclipse of a King

423

Handbook 4/II 105–106). 11 These are not, of course, actual earthquakes but refer to the ability of these winds to collapse walls and even to destroy buildings (de Moor 1971: 173–76). I still remember quite distinctly from my childhood photographs published in Life Magazine: a very daring photographer had captured on film a caravansary being reduced literally to dust by a whirlwind. A glance at the lunar eclipse rituals allows us to pick out which ritual is Spring, which Fall, and which the remaining two seasons. The first ritual (BRM 4 6: 1′–41′ [Linssen 2004: 306–07]) and the fourth ritual (BM 134701 obv. 17′–rev. 22′ [Linssen 2004: 308–09]) share specific features associated with that time of the year when, as MUL.APIN puts it, the proportion of water used in the water clock for daytime and nighttime is “3 minas, 3 minas” (I iii 2 [Hunger and Pingree 1989: 43]). These features are the sealing closed and reopening of gates (BRM 4 6: 31′; BM 134701 obv. 18′–21′, rev. 1′–9′ [Linssen 2004: 307, 309]), the use of a black and white thread of equal lengths—used to seal the doors—to indicate day and night of equal length BRM 4 6: 31′–32′; BM 134701 obv. 18′–23′ [Linssen 2004: 307, 309]) and recitations mentioning the Pleiades (BRM 4 6: 34′–36′; BM 134701 rev. 10′–22′ [Linssen 2004: 307, 309]) which, as was known to ancient agriculturalists since the Stone Age, appeared every year in the Spring around harvest time and disappeared in the Fall at planting time. What will have been the Spring ritual is not merely the first in line, but also makes specific mention of terrible dust storms (BRM 4 6: 3′–13′ [Linssen 2004: 306]). Moreover, the now harvested fields and dry watercourses are to receive funerary offerings (BRM 4 6: 19′–20′ [Linssen 2004: 306]). The fourth ritual is fragmentary, but enough survives to clinch this ritual as appropriate to the Fall. The presence of the king (BM 134701 obv. 17′, 19′, 21′ [Linssen 2004: 308–09]) would indicate this in and of itself, since his akītu took place in that season, but the Pleiades recitation (BM 134701 rev. 10′–22′ [Linssen 2004: 309]) is fortunately similar to that appear­ ing in the series Uttuku Lemnūtu. The Uttuku Lemnūtu version is the so-called lunar eclipse myth. According to John Wee (2014), this myth was concocted by the king to avoid giving a pretext for the “treason and regime change” that was a likely result of “long held ideas about the lunar eclipse as divine judgment” for the king’s “personal guilt and failure”. This is another sterling example of putting the theory cart before the evidentiary horse. The “long held ideas” were that it was whole countries that were threatened by the eclipse and the popular reaction, as evidenced in a report to the king from Babylonia (Parpola 1993 no. 352), was a demand for the proper performance of all requisite apotropaic rituals. Mesopotamian kings were not in the habit of concocting myths and, if they had been, the Mesopotamian public was not so gullible as we are today. Finally, the Utukku Lemnūtu passage is not actually a lunar eclipse myth; neither is it of comparatively recent origin. The Old Babylonian Sumerian original was an etiological myth that recounted the origin of the equinox, in which the constellation Scorpio was imagined as a mount for a martial Ištar ganging up with her brother Šamaš and the Pleiades in an attack on the moon (van Dijk 1998). Scorpio corresponds to the eighth month in the Mesopotamian calendar, so a “lunar depression.” This is presumably the explanation of the otherwise inexplicable reference in the fourth ritual to a “day 11.  Similar seasonal dust storms occur in Southern Iraq, as described to me by M. Gibson.

424

JoAnn Scurlock

and night” eclipse when the moon did not “sit on his royal seat” (BM 134701 rev. 20′ [Linssen 2004: 309]). By process of elimination, the second and third lunar eclipse rituals, the ones that concern themselves exclusively with eclipses, must have covered Summer and Winter occurrences. To note is the fact that harps feature only in these two rituals (BRM 4 6: 43′ and BM 134701 obv. 2′ [Linssen 2004: 307, 308]), and not in the Spring and Fall versions. Also significant is the ordering of the seasons. We should expect Spring, Summer, Fall, and Winter, and yet, Fall is clearly last and not third in line as it should be. There was nobody in ancient Mesopotamia of any period who did not know the correct ordering of the seasons. The only obvious explanation, then, for this anom­ alous ordering: Spring, then Summer, followed by Winter and then Fall, was the desire to create a magic circle within the text. Magic circles have potent protective powers but, quite apart from this, a time-connected magic circle would have the effect of keeping time moving, thus ensuring the continued survival of mankind as well as of the plants and animals upon which he depended for food. We, too, are concerned with entropy, but we are more pessimistic about our ability to do anything about it. We are now beginning to hum, since the decorative panel on the Bull-headed lyre, and indeed on all of the harps found in the so-called Royal Tombs has four sections. Of these, only the two center panels on the Bull-headed lyre show harps. Not only this, but the lyre actually shows itself in one of the panels. Since, however, this is a hypothesis and not a theory, we need to devise a test. If there is any connection between the seasonally varied lunar eclipse ritual and our Bull-headed lyre, then the four panels must somehow represent the four seasons, and in the following order: Spring, Summer, Winter, and finally Fall. Taking a leaf from Wiggermann’s earlier hypothesis, and connecting the figures to the phantasmata of the Babylonian Map of the World yields a further hypothesis, namely that, like the phantasma of the Babylonian Map of the World, the figures on the lyre represent constellations. A first glance is most promising. Besides the Scorpion-man = Scorpio on the fourth panel, we have a Lion = Leo on the second panel. However, since this is a hypothesis and not a theory, we cannot stop here. We need all the figures to be identifiable with known constellations and these constellations as gathered together in their respective panels must represent four distinct seasons of the year, and in the same order of presentation as in the lunar eclipse ritual. This seems a very tall order. However, a quick consultation with MUL.APIN allows us not only to identify the remaining figures but to associate each panel with a specific season of the year, and in the requisite order. The first panel depicts a Laḫmu gate-guardian of the Abzu micromanaging an obstinate Taurus 12 whose appearance, in Sumerian times, marked the Vernal Equinox. 13 This obviously represents the Spring. The second panel shows Leo dang­ ling Crater to represent Summer. 14 With Leo and his wine jar and dipper we have 12.  For details, see Scurlock, forthcoming a. 13.  The Bull of Heaven’s jaw was expected to appear by the 20th of Ayyaru according to I ii 39 (Hunger and Pingree 1989: 40). The windy season at the beginning of the year was defined as the 1st of Addaru to the 30th of Ayyaru—III Gap A 1–2 (Hunger and Pingree 1989: 88). 14.  MUL.APIN takes the hot season (summer) to be from the 1st of Simānu to the 30th of Abu—II Gap A 3–4 (Hunger and Pingree 1989: 88). Lion was expected to be visible by the 15th of Du’ūzu and its

Eclipse of a King

425

also a Fox with a dagger in his belt, the constellation of the god of Summer plagues, Erra. 15 The seven black dots of the Pleiades are shown on the dagger which is stuck in Erra’s belt. 16 He carries a rectangular object piled high with disarticulated animal parts, about which more presently. This is conventionally interpreted as a table. However, we are hardly in ignorance on the subject of ancient Mesopotamian furniture and this “table” resembles no ancient Mesopotamian table or stand that I know of. Given that it is apparently stringed, we are probably safe in assuming that we are looking at another musical instrument. If this sequence were in normal calendar order, Fall should be next but this is instead represented by the fourth and last panel which shows Scorpio to represent the Fall. 17 Accompanying him is the Goat-Star of Gula carrying two wine cups, a constellation known to have received offerings as part of the rites of the akītu of the king 18 that occurred in the seventh month of the Mesopotamian calendar. By process of elimination, the remaining, third frame will have represented Winter. On this panel, three animals are represented. Two of them appear to be horses, and both are seated, playing instruments. One plucks a model of the Bull-headed lyre and, facing him, a second equid wields a claw-like instrument in his hand poised as if to strike a rectangular object placed in his lap. Standing over the little horse, a bear-like animal grasps the harp’s sounding board. According to MUL.APIN, the Horse Star stands to the left of Panther-Nergal which is meant to mark Winter. 19 We may, then, presume that the equids are meant to be horses. The Panther is more difficult since the representation seems instead to be a bear, although comparable panels on other harps clearly show leopards. Either a mistake was made or, more interestingly, the intent was to show a con­ stellation known in antiquity as the Bear. MUL.APIN refers to this constellation as the Wagon, which might suggest either a quite early date for this representation or a non-Mesopotamian artist, or both. 20 In sum, we have accounted for all the figures on the Bull-headed lyre, the four panels and the order of presentation and our hypothesis is looking very strong indeed. However, we have not one but at least two different musical instruments to account for, our lyre in panel three and one or possibly two more instruments in panels two and three. Since this is a hypothesis and not a theory, the musical instrument(s), which are not our harp cannot simply be left hanging.

breast (King) by the 5th of Abu—I ii 42–44 (Hunger and Pingree 1989: 41–42). 15.  According to the Hymn to Ḫendursaga 78, 80, the first of the seven is like a fox dragging his tail, the third like a raven. For a recent French translation, see Attinger 2011. For a transliteration of the Sumerian, see Black 1999. 16.  These dots are hard to see on most photographs; see the description in Hansen 1998: 56. 17.  The Scorpion’s appearance was expected on the 5th of Araḫšamna with its Breast and the GoatStar (of Gula) visible by the 15th—I iii 3–4 (Hunger and Pingree 1989: 44). The windy season between summer and winter was defined as stretching from the 1st of Ulūlu to the 30th of Araḫšamna—III Gap A 5–6 (Hunger and Pingree 1989: 89). 18.  For more on this ritual, see Scurlock 2019. 19.  I i 28–30 (Hunger and Pingree 1989: 26–27). The Panther was expected on the 15th of Kislīmu—I iii 5–6 (Hunger and Pingree: 44–45), the cold season (winter) being defined as the 1st of Kislīmu to the 30th of Šabāṭu—III Gap A 7 (Hunger and Pingree 1989: 89). 20.  Art historians generally see similarities with Elamite art. For a discussion of the Bull-headed Lyre from this perspective, see Hansen 1998: 53–57.

426

JoAnn Scurlock

One of these mystery instruments is the object attended by Mr. Fox in the second (Summer) panel. Making our life considerably simpler, let us suppose that we have only two instruments, the Bull-headed lyre and another harp shown twice, once being carried about with offerings on top in panel 2 (Summer) and another actually being played in panel 3 (Winter). Both Summer and Winter in Mesopotamia were associated with visits by the dead to the living. As we also know, the dead were summoned up by the playing of a BALAG-instrument, which was, depending on whom you ask, either a harp or a drum (Kilmer 1995/7: 463–64). The obvious solution to this problem is to see in the BALAG-instrument a very old form of harp that was not plucked but beaten. In both Japan and Medieval England, the one-stringed Catalpa Bow (Japan) and the Psalter (England) were credited with the power to allow the dead to speak. 21 From the appearance of the instrument on the Bull-headed lyre, a hammered dulcimer also comes to mind as a possibility. Unfortunately, this is not the instrument in which we are actually interested, although it does allow us to eliminate the possibility of some sort of necromantic function as a use for our lyre. So what is left? Numerous laments have come down to us from ancient Mesopotamia and there is an entire sequence that spans the Winter months. As I understand it, the ritual, which goes back to at least the Old Babylonian period, was originally designed to explain and expel Winter darkness as well as to avoid major political catastrophes such as that which occurred during the reign of Samsuiluna, the alleged author of the lamentation series. 22 What was supposed to be happening was that Anu and Enlil were engaged in a total war of mutual annihilation to determine which of them would rule heaven and earth. The idea of the laments was to calm down both parties before heaven and earth and all those living in it got ground up. These laments culminate on the 5th of Nisannu when a bull representing Anu was burned in a pit to signal his final defeat by Enlil and an end to the dangers represented by the conflict. 23 The presence of the bull’s head on the lyre would, by this understanding, specifically represent Anu and point directly to Winter as the season to exorcise the dangers of which the Bull-headed lyre was manufactured. In sum, we have a perfect match between the lunar eclipse rituals and our Lyre’s sounding board, and one that further explains the peculiar order of seasons as represented in both text and image. In conclusion, we have arrived at what could be a usable theory, namely that this lyre in the Ur cemetery was used to lament a substitute king who took away the evils portended by a royally dangerous eclipse that occurred in Winter as, I would argue, is specifically represented in the iconography of its sounding board. It would follow that the burials in the cemetery can be used as archaeological evidence for performed ritual, namely an early form of the already substantially known substitute king ritual. What is more, since the number of solar or lunar eclipses during which the planet Jupiter was not present and whose shadow will have portended 21.  For Japan, see Blacker 1975, especially 106–07. There is a rather gruesome old English folksong the conceit of which is that a psalter made from the bones of a murdered girl, when played, gave the victim voice and allowed her to pin the guilt on her ex-lover. 22.  Over the course of time, cosmic events were updated, eventually pitting Marduk against Anu as part of the Babylonian New Years’ Festival. A full analysis of this fascinating ritual has yet to be written. For an edition of one of the key laments, see Kutscher 1975. For the schedule for the performance of laments, see Parpola 2017 no. 12. 23.  For details on this sacrifice, see Scurlock forthcoming b.

Eclipse of a King

427

evil for southern Mesopotamia as opposed to other regions and that will have been visible in Winter at Ur will have been relatively restricted, we may find it possible to assign one or more potential absolute dates to one of the burials in the cemetery. But that is a job for one far more versed in astronomy than I.

Bibliography Admiralty 1919 A Handbook of Asia Minor Prepared on Behalf of the [British] Admiralty and War Office. London: Naval Staff Intelligence Department. Ambos, K. 2006 Unverstandene Rahmungen-interkulturelle und intrakulturelle Missverständnisse bei der Durchführung des Mesopotamischen Ersatzkönigsrituals. Pp. 77–86 in Rituale in Bewegung: Rahmungs- und Reflexivitätsprozesse in Kulturen der Gegenwart, eds. H. Jugaberie and J. Weinhold. Berlin: Lit Verlag. Andrae, W. 1977 Das wiedererstandene Assur. München: C.H. Beck. Attinger, P. 2011 Hendursaga A (4.06.1). See https://cdli.ucla.edu/dl/pdf/P473743.pdf. Baadsgaard, A., Monge, J., Cox, S., and Zettler, R. L. 2011 Human Sacrifice and Intentional Corpse Preservation in the Royal Cemetery of Ur. Antiquity 85/327: 27–42. Black, J. 1999 Hendursaga A (4.06.1). See http://etcsl.orinst.ox.ac.uk/cgi-bin/etcsl.cgi?text=c.4.0 6.1&display=Crit&charenc=gcirc#. Blacker, C. 1975 The Catalpa Bow: A Study of Shamanistic Practices in Japan. London: Allen and Unwin. Curtis, J. E. 2008 New Light on Nimrud. Proceedings of the Nimrud Conference 11th–13th March 2002, eds. J. E. Curtis, H. McCall, D. Collon, and L. al-Gailani Werr. Exeter: Meeks and Middleton. Dijk, J. J. A. van. 1998 Inanna raubt den ‘Grossen Himmel’. Ein Mythos. Pp. 9–38 in Festschrift für Rykle Borger zu seinem 65. Geburtstag am 24, Mai 1994, ed. S. M. Maul. CM 10. Leiden: Styx. Farber, W. 1977 Beschwörungsrituale an Ištar und Dumuzi. Wiesbaden: Franz Steiner. Hansen, D. P. 1998 Art of the Royal Tombs of Ur: A Brief Introduction. Pp. 43–72 in Treasures from the Royal Tombs of Ur, eds. R. L. Zettler and L. Horne. Philadelphia: University of Pennsylvania Museum of Archaeology and Anthropology. Hunger, H. and Pingree, D. 1989 MUL.APIN: An Astronomical Compendium in Cuneiform. AfO Beih. 24. Horn: Ferdinand Berge & Söhne. Hussein, M. 2016 Nimrud: The Queens’ Tombs. Chicago: The Oriental Institute. Kilmer, A. D. 1995/7 Musik A.I in RlA 8/5–8: 463–482. Kutscher, R. 1975 On Angry Sea: The History of a Sumerian Congregational Lament. YNER 6. New Haven: Yale University Press.

428

JoAnn Scurlock

Labat, R. 1939 Hémerologies et Ménologies d’Assur. Paris: Librairie d’Amerique et d’Orient. Linssen, M. J. H. 2004 The Cults of Uruk and Babylon: The Temple Ritual Texts as Evidence for Hellenistic Cult Practices. CM 25. Leiden: Brill-Styx. Moor, J. C. de. 1971 The Seasonal Pattern in the Ugaritic Myth of Baʿlu According to the Version of Ilimilku. AOAT 16. Neukirchen-Vluyn: Butzon & Bercker Kevelaer. Parpola, S. 1993 Letters from Assyrian and Babylonian Scholars. SAA 10. Helsinki: Helsinki University Press. 2017 Assyrian Royal Ritual and Cultic Texts. SAA 20. Helsinki: Helsinki University Press. Scurlock, J. A. 1995 Death and the Afterlife in Mesopotamian Thought. Pp. 1886–1893 in Civilizations of the Ancient Near East, ed. J. Sasson. New York: Scribners. 2013 Images of Tammuz: The Intersection of Death, Divinity and Royal Authority in Ancient Mesopotamia. Pp. 151–82 in Experiencing Power, Generating Authority: Cosmos and Politics in the Ideology of Kingship in Ancient Egypt and Mesopotamia, ed. P. Jones. Philadelphia: University of Pennsylvania. 2019 Feasting in the Garden of God: Ramat Rahel and the Origins of Rosh Hashanah and Yom Kippur? Pp. 44–74 in What Difference Does Time Make?, eds. J. Scurlock and R. Beal. Archaeopress Ancient Near Eastern Archaeology Series 6. Summertown, OX: Archaeopress. fth. a World Encircling River. In Irrigation in Early States: New Directions, ed. S. Rost. OIS 13. Chicago: The Oriental Institute. fth. b Assyria and Babylon in the Oracles Against the Nations Tradition: The Death of a King (Is 14:5–20). JAOS. Wee, J. 2014 Grieving with the Moon: Pantheon and Politics in the Lunar Eclipse. JANER 14: 29–67. Wiggerman, F. A. M. 1995 Extensions of and Contradictions to Dr. Porada’s Lecture. Pp. 77–154 in Man and Images in the Ancient Near East, ed. E. Porada. London: Moyer Bell.

Local and Imported Religion at Ur Late in the Reign of Shulgi Tonia M. Sharlach

Oklahoma State University Ur must have been a magnificent place. Though clearly thriving in the Early Dynastic period, Ur’s heyday may well have been during the Third Dynasty of Ur, when it served as the dynastic seat to one of the region’s superpowers. In addition to its geopolitical importance, Ur was also the major southern seaport for Babylonia. Extensive archaeological work at Ur has of course revealed traces of its former grandeur—the sacred precinct with its ziggurat and many other shrines and buildings, the burial places of at least some of the kings of the Third Dynasty, as well as fragments of monumental artworks (Woolley and Moorey 1982: 156; Molina 2015: 359). Ur, as the seat of the Third Dynasty, was undoubtedly like Washington or New York today—a place where things of importance happened, a magnet for people from all over, a hub. Yet, from a textual perspective, Ur is almost a blank book right up to the reign of Ibbi-Sin. Documents from Ur from the first half of Shulgi’s reign are almost nonexistent (Molina 2015: 357). Documents start to appear from about S26 on but are very few and far between until S36. And Widell has noted a drop-off in even this evidence from about S40 onward (Widell 2003: 97). We do not have the Ur palace archives for the reign of Shulgi or the other Ur III kings. We do not have the accounts of Shulgi’s kitchens or the en-priestess’ fields or the army rosters. We do not even have any evidence for sea-going merchants of Ur III. What we do have from the reign of Shulgi are some slave sales over a period of more than twenty years, interestingly often involving persons from Adab, 1 a few 1.  The slave sales with Adab involvement from Ur include in chronological order: UET 3 19, dating to S29, in which a female slave was sold and one Aakalla, some sort of functionary (literally the EGIR) of Ur-Ašgi the énsi of Adab sealed the tablet; UET 3 14, dating to S42, a male slave was sold and one witness, also a sealer of the tablet, was a captain (nu-bànda) from Adab named Ur-ašar; UET 3 15, dating to S47, in which a woman slave was sold. After all the witnesses, the transaction is said to be šu, literally in the hands of, Habalukke, the énsi of Adab. UET 3 18 dates to AS2, documenting the sale of another female slave said to be šu, in the hands of, Habalukke, the énsi of Adab. Further note UET 3 9 (no date but the seal bears a dedication to Shulgi), now badly damaged but possibly also to be considered in this set of tablets. To the best of my knowledge, no other provinces are attested in similar texts. Why officials from Adab were involved in slave sales in Ur over a twenty year period is not clear to me; these tablets surely deserve to be studied in their own right.

429

430

Tonia M. Sharlach

letter orders, and tablet basket tags. 2 We are more in luck for the latter half of the reign of Ibbi-Sin, as a craft archive has survived from those years. Widell estimated that seventy-seven percent of the Ur texts with dates came from Ibbi-Sin’s reign (Widell 2003: 96). So to look textually at Ur in Shulgi’s reign, particularly the last twenty years of Shulgi’s reign, we must look in other archives. The crown administration at Drehem, ancient Puzrish-Dagan, is the most useful and obvious source here, as Sigrist for instance pointed out (Sigrist 1992: 381). One of the most helpful archives with regard to Ur is the archive associated with the woman Shulgi-simti, one of Shulgi’s many consorts. Her archive is attested from roughly S32 to the death of Shulgi in S48 (Weiershäuser 2008: 31; earlier treatments include among others Sallaberger 1993: 18 ff., Sigrist 1992: 222–223). We will be mining this archive for evidence about religious observances at Ur. Because these observances are not exactly what we might expect from Ur, the usual explanation has been that we are looking at imports or foreign things and that the likely source of these foreign things was Shulgi’s wife, Shulgi-simti herself (e.g. Sallaberger 1993: 19; Weiershäuser 2008: 32). But when we examine the evidence, I am not sure it is strong enough to support these assumptions. In order to evaluate, we must look in a little more detail at the archive of Shulgi-simti itself. Tablets from the archive of Shulgi-simti began to be discovered in about 1909 and, on the basis of its find date, month names and other criteria, as Sigrist put it, “No writer has questioned Drehem as the place of origin of the archive” (Sigrist 1995: 21). It is well-known, however, that the archive often details observances that occurred in Ur and other locations (Sigrist 1992: 237–244; Sallaberger 1993: 20–21). It has already been soundly established by Sallaberger and others that the extant Shulgi-simti tablets recording cultic expenditures do not fall into a regular pattern but tend more towards chaos, as a large number of the cultic events seem to have occurred on an ad-hoc basis (Sallaberger 1993: 42; Weiershäuser 2008: 46). But, as a number of scholars including Sallaberger, Sigrist and Weiershäuser have aptly remarked, two goddesses, named Belat-Shuhnir and Belat-Terraban, ap­peared frequently at Ur. 3 They have a prominent place in this archive and seem to be central to the existence of the foundation (Sigrist 1992: 237; Sallaberger 1993: 18ff; Weiershäuser 2008: 34). But this is really a quartet of goddesses, as Jones and Snyder (1961: 204) noted many years ago. Belat-Shuhnir and Belat-Terraban are generally found together with two other deities, known as martial forms of Ishtar, Annunitum and Ulmashitum, as we see below. TRU 287 (S39.7–8.00) obv. 2 máš níta 2 míáš-gàr 2 u8 2 udu níta 2.  Letter orders are represented for example by Widell 2003 Text Two p. 117 and UET 3 6 (undated). Basket tags include UET 3 1058 (S38) and UET 3 54 (from S46 to AS2). 3.  Because the theme of the volume is Ur, I am focusing on cultic matters that took place there. It is of course the case that documents from Shulgi-simti’s archive also sometimes show that this pair of goddesses, Belat-Shuhnir and Belat-Terraban, could be worshipped elsewhere.

Local and Imported Religion at Ur Late in the Reign of Shulgi

431

iti ezem-dšul-gi 4 u8 2 udu níta 2 míáš-gàr iti šu-eš-ša rev. sá-dug4 dbe-la-at-suh-nir dbe-la-at-dar-ra-ba-an an-nu-ni-tum ù dul-ma-[ši-tum]

Ulmashitum was also known as Ishtar of Akkade, named for her temple there, the é-ul-maš (Sigrist 1992: 238; Sallaberger 1993: 198). Annunitum appears to be very similar in nature, but is best attested currently from Sippar, though she had a wider diaspora (Gödecken 1973). 4 Over and over again in the Shulgi-simti archive, these four goddesses appear together, almost in the same breath, as it were. I provide just two of the many extant examples of this quartet. Note that in the second text, a festival called elūnum is mentioned. CUSAS 16 289 (S39 months 2–4) 5 obv. 4 udu niga 2 máš 2 sila4 iti šeš-da-kú 2 u8 2 ud5 3 máš 1 sila4 iti u5-bí-kú 6 udu níta rev. [x] máš iti ki-siki-dnin-a-zu sá-dug4 dbe-la-at-suh-[nir] ù dbe-la-at-dar-ra-ba!-an an-nu-ni-tum ù dul-ma-ši-tum iti 3-kam … MVN 18 67 (S41 month 2) obv. 2 udu niga an-nu-ni-tum 1 máš-gal dul-ma-ši-tum 2 máš-gal dbe-la-at-suh-nir ù dbe-la-at-dar-ra-ba-an 4.  Annunitum may have been particularly associated with Sippar-Amnanum, where, confusingly, she had a temple also named Ulmash in OB times, according to Westenholz and Asher-Greve (2013: 265). See also Gödecken 1973. Other explanations for Annunitum can be found. Sigrist (1992: 238) for instance defined that deity as the “nom féminin d’An, une divinité guerrière, devenue indépendante d’Ištar.” 5.  This tablet currently belongs to the Columbia University Library.

432

Tonia M. Sharlach rev. è-lu-núm šag4 uri5ki-ma gìri ma-šum iti-ta ud-20 ba-ra-zal …. zi-ga á-bí-lí-a…

Were These Goddesses Foreign? As just discussed, Ulmashitum’s home was Akkade and Annunitum’s was likely Sippar, though both attained “überregionale Bedeutung” in Sargonic times, as Sallaberger noted (Sallaberger 1993: 198). Though Akkade remains to be located with certainty, a north Babylonian location seems very likely. The homes of Belat-­ Shuhnir, the Lady of Shuhnir, and Belat-Terraban, the Lady of Terraban, are more difficult to pin down. Trying to find out exactly where Shuhnir and Terraban are has not, as yet, met with any clear consensus; and it would not necessarily help us understand what kind of goddesses they were anyway. They may be local forms of Ishtar, but they do not appear to be local to Ur. A more fruitful line of approach has been to trace where Belat-Shuhnir and Belat-­ Terraban were worshipped. Jacobsen, Whiting, Sallaberger and others showed the divine pair Belat-Shuhnir and Belat-Terraban were worshipped in the Eshnunna/ Diyala area. 6 The most convincing evidence for this link comes from Eshnunna texts recording a major offering, which included Belat-Shuhnir and Belat-Terraban, and the seal inscription of the Eshnunna ruler, Shu-iliya (a contemporary of Ishbi-Erra) who refers to Tishpak, Belat-Shuhnir, and Belat-Terraban (Reichel 2008: 138). Therefore, when we find Belat-Shuhnir and Belat-Terraban worshipped at Ur, I am quite comfortable in saying that they appear to have antecedents in the Diyala area. That is, they were imports. We should keep in mind that the Diyala was well and truly within the core of the Ur III state, one of the heartland provinces, so we are talking about an import from another area within the state, not about anything “foreign.” 7 (I would repeat that we are always leaving out the other two goddesses, Annunitum and Ulmashitum, about whom I will have more to say later.) If we want to know more details about how these deities were worshipped in Ur, we are limited by the nature of our texts. The offering lists from the Shulgi-simti archive, as laconic as they are, do often add terms to describe the rituals, and as a number of scholars have also remarked that at Ur, these tended to be Akkadian terms such as elūnum, girrānum, nabrium, and so on (Sigrist 1992: 237; Sallaberger 1999: 259). The ritual terms elūnum, girrānum, and so on are not unknown—they do appear to have relatives in the northern territories in the Third Millennium— but they are, as Sigrist commented, notable for their Akkadian/Semitic character (Sigrist 1992: 237–238). This idea can be traced back to an article Schneider published in 1933, in which he saw the veneration of deities like Belat-Shuhnir and Belat-Terraban as evidence for a trend towards the greater Semitization of Mesopotamian culture (Schneider 1933: 13–23).

6.  Jacobsen 1940: 143; Sallaberger 1993: 19; see also Frayne 1997: 435–437. 7.  On Eshnunna’s core status, see Reichel 2008: 138. It might be possible to argue that it was a borderland.

Local and Imported Religion at Ur Late in the Reign of Shulgi

433

If one has the preconception that southern Babylonia at this date was all Sumerian, then it would be quite natural to deem any Akkadian terms intrusive, that is, imports from the “Semitic north.” And we do see this in the historiography on this topic. 8 It is not impossible that northern Babylonian deities were brought down to Ur along with their associated rites. But here I think we should be careful. As recent studies on archives like Garshana and Uru-Sagrig have shown, there was a very considerable Akkadian element thriving in southern Babylonia at this time, perhaps particularly in royal circles. It is not impossible that the Sumerian language was spoken less frequently than Akkadian at this date (Sallaberger 2011). With these ideas in mind, when we turn back to the Semitic festival names, do we really need to seek for outside origins? Is everything Semitic by definition “foreign” and “intrusive?” Turning to the Ur III evidence, we do not find that there is an exclusive link between these Akkadian festivals and the Shulgi-simti archive (pace Sallaberger 1993: 202). That is to say, the Akkadian festival names do occur in the Shulgi-simti archive but not only there. In the interests of time, let us take just one of the festival names, the elūnum. I am not attempting to determine the nature of these observances—that may be impossible on the present evidence 9—and I am not arguing that Shulgi-simti’s foundation did not use such names to describe rituals for the quartet of goddesses in Ur—but I do not feel that it is accurate to state that these Semitic festivals were unique to Shulgi-simti. To begin with, there is evidence—admittedly only a single tablet—that the elūnum rites for Belat-Shuhnir and Belat-Terrraban were being practiced in S28, prior to the foundation of the Shulgi-simti foundation (insofar as we can tell, that foundation began operations in S32) (Sallaberger 1993: 18). This is Princeton 2 134 (S28 month 1), which tells us that four sheep were expended for the è-lu-núm é dbela-at-suh-ner, for the elūnum celebrations in the temple/shrine of Belat-Shuhnir. We can be fairly certain that it was not just an early Shulgi-simti tablet as the officials named can be identified as working not for Shulgi-simti but for other roy­al households (the official Ahima was never a bureau chief of Shulgi-simti’s, as discussed in Sharlach 2017: 204). It is difficult to make a cogent logical argument that Shulgi-simti was responsible for bringing both the elūnum and the goddesses to Ur when both—even the physical building of the shrine—are attested prior to her foundation’s start date. Other Ur III royal women performed elūnum rites for deities, often goddesses. As early as S33, there is a reference to an elūnum of the goddess Geshtin-anna DIRI-tum in an Umma tablet, MVN 15 162. 10 Another one of Shulgi’s wives, Geme-Ninlilla, used two oxen for an elūnum rite in Shulgi year 41 according to YOS 4 240, as D’Agostino already pointed out in 1998 (D’Agostino 1998: 1, n. 3). A little later in the Ur III period, Abi-simti performed an elūnum of Inanna Haburitum in AS 5 according to MVN 20 31 (D’Agostino 1999: 1–5; Weiershäuser 2008: 135–136). Similarly, also in the reign of Amar-Sin, the princess ME-Ištaran did an elūnum 8.  To give just one recent example, Weiershäuser 2008: 69. 9.  There have of course been speculations: for instance, Sigrist (1992: 237) connected elūnum to the Semitic month name Ellul. 10.  The tablet MVN 15 162 (S33 month 8 in the Umma calendar) is from Umma and records expenditures of reeds, mats and such, mainly in Ur.

434

Tonia M. Sharlach

and received one ox and two sheep on that occasion, but the deity for whom she performed the rite is not recorded (MVN 11 144). Another text from Garshana dated also to Amar-Sin’s fifth year documents another elūnum celebration which ME-Ištaran performed; this text specifies that the celebration was for Haburitum and occurred in Puzrish-Dagan. 11 Other elūnum rites for goddesses in the later Ur III and early Isin-Larsa period can also be found. A text from Puzrish-Dagan that may date to early Ibbi-Sin refers to an è-lu-núm dnun-gal. 12 In the reign of Išbi-Erra, we find records of the expenditure of an ox-hide for the è-lu-núm dinanna (BIN 9 331, Ishbi-Erra years 7–8). Most of the evidence collected above could be used to suggest that the elūnum was particularly suited to the veneration of goddesses. But the elūnum rite could also be linked to male deities. Garshana texts link an elūnum (spelled slightly differently, that is, e-lu-núm) to Nergal in at least four cases dating between Shu-Sin 6 and IS3. 13 Still later, the elūnum was still being celebrated in Old Babylonian Mari in the reign of Zimri-Lim. Two letters republished by Sasson show that when vassals of Zimri-Lim celebrated the elūnum to local goddesses in their home cities, namely Andarig and Karana, portions of the meat offering were conveyed by messenger up to the king in Mari. One letter reads, “I have just celebrated the elūnum of Ishtar of Qaţţṭṭara. Herewith is your portion” (ARM 28 174, Sasson 2015: 249, n. 44). The second letter is more detailed, “I have written earlier to say that I shall offer elūnum sacrifice to Ishtar of Andarig, who protects your life and mine. I am now conveying my brother’s share of the meat” (ARM 28 169, Sasson 2015: 249, n. 44). Lafont, following Kupper, went so far as to describe the elūnum as “une sorte de rite de communion” (Lafont 1999: 63); Cohen proposed that it “was an observance at which one partook of special offerings and sacrifices… in communion with deities” (Cohen 2015: 277). The term “communion” is perhaps anachronistic and confusing. There does seem however to be an element of sharing consecrated food with others who were not physically present at the sacrifice. By Old Babylonian times, at least in certain locales, the elūnum was celebrated regularly in the early part of the year and the term could in fact be used as the month name for month two in certain local calendars (Sallaberger 1993: 202; Cohen 2015: 277–282). The OB evidence for the elūnum celebrations stretches from Ur as far north as Leilan (D’Agostino 1998: 5 and n. 14). To me this suggests a widespread rite used for many deities—more a type of worship that was common throughout numerous periods and places than a rite specific to a particular city or a particular cult. If one were to scrutinize the evidence for another religious celebration with an Akkadian name, the girrānum, which some have also linked to Shulgi-simti (e.g. Sallaberger 1993: 47, 200; Weiershäuser 2008: 57), the same pattern seems to appear. While there is a solid chunk of evidence that Shulgi-simti’s foundation performed

11.  This is CUSAS 6 1582 (CDLI P429467), dating to AS 5 month 3. Here a wider variety of commodities were used, including metal vessels, cloth, baskets of apples, beer, and also livestock. It is not impossible that MVN 11 144 and the Garshana text record the same elūnum celebration, but it could be that multiple ones were performed for different deities. 12.  Nisaba 8 92 iii and iv. 13.  CUSAS 3 19, 90, 980, and 1032.

Local and Imported Religion at Ur Late in the Reign of Shulgi

435

such ceremonies, believed to have been lamentations (Weiershäuser 2008: 57, 61), 14 the girrānum was certainly not limited to her lifetime, being attested currently from AS2–SS1 15 and in several locations other than Shulgi-simti’s residence (in Umma according to the undated tablet UTI 6 3639 and Uruk according to TIM 6 8, a tablet dating to SS1.07.24). I think it is quite possible that festivals like the elūnum and the girrānum had a long pedigree in places like the cosmopolitan seaport of Ur, and that we simply lack the textual evidence at present that would show us this longue durée (remember that we only have about thirty documents from Ur from the whole of the Sargonic period, Molina 2015: 359). This is speculation, but we should exercise restraint before assuming everything with a non-Sumerian name was an import or that all imports (if they were such) date to the reign of Shulgi. Thus, perhaps we can agree that the presence of the Semitic festival names in Ur may demonstrate a commonality of religious expression over space and time rather than an intrusive import. Turning now to some of the goddesses attested regularly in the Shulgi-simti archive, including Belat-Shuhnir and Belat-Terraban, along with Annunitum and Ulmashitum, the situation appears a little different. These goddesses were not widespread or long-lived. It is quite possible that they had a homeland in northern Babylonia or the adjacent Diyala region, as we shall see momentarily, and it is quite likely that the cult of these goddesses was not native to Ur. How, when, or why these goddesses were imported is quite unknown. Now I am not claiming that Shulgi-simti did not have a particular affection for this quartet of goddesses and I am not claiming that she had no choices in the matter. It seems very logical to say that she did choose in particular to venerate this quartet of goddesses most of all. Since Belat-Shuhnir and Belat-Terraban appear frequently in the Shulgi-simti archive, some have tried to imagine a scenario in which Shulgi-simti had very close, almost exclusive links to these two goddesses, so that two corollaries are posited: first, that their north Babylonian/Diyala origins show Shulgi-simti’s own place of origin (in the Diyala) and second, that Shulgi-simti brought the goddesses down to Ur and that their worship in Ur continued at the court only as a sort of a hangover from her days. In this scenario, Belat-Shuhnir and Belat-Terraban are then “foreign” gods imported by the foreign woman. This view is quite widely held, for example by Sallaberger: “Die dem babylonischen Pantheon fremden Göttinnen Bēlat-šuhnir und Bēlat-Deraban müssen daher von Šulgisimtum nach Ur gebracht worden sein” (Sallaberger 1993: 19). More recently, we can see the same position espoused by Weiershäuser: “Šulgi-simtī kam wahrscheinlich aus Ešnunna, worauf die beiden von ihr besonders verehrten Göttinnen Bēlat-šuhnir und Bēlat-Deraban weisen….Diese Göttinnen brachte Šulgi-simtī mit nach Ur. Mit einer dynastichen Verbindung zu Ešnunna konnte Šulgi das strategisch wichtige Dijala-Gebiet … sichern” (Weiershäuser 2008: 32; a similar view to be found also in Frayne 1997: 170). Sometimes this hypothesis is presented as though it were simply fact, as by J. Westenholz: “We do know that she (Shulgi-simti) brought her personal god14. To give just a few examples, the Shulgi-simti archive includes the following evidence for girrānum celebrations: MVN 18 56 (S35 month 10), MVN 3 153 (S37 month 6), OIP 115 62 (S40 month 11), TRU 274 (S41 month 1). Many other attestations exist. 15. The girrānum is attested in AS2 according to the evidence of MVN 13 715, AS9 according to the evidence of Durand’s DoCu EPHE 306 and SS1 according to TIM 6 8.

436

Tonia M. Sharlach

desses Bēlat-šuhner and Bēlat-terraban from her home in Ešnunna in the Diyala region ....” 16 I would reiterate the point that there is a logical flaw in the argument. If a person worships four goddesses from different places, why do only two of those goddesses determine origin? By leaving Annunitum and Ulmashitum out of our narrative, no one seems to argue that Shulgi-simti’s worship suggests she had personal links to Akkade or Sippar. However, there may be more fundamental problems with the argument, namely, the assumption that the homelands of deities worshipped can tell us about Shulgi-simti’s homeland at all. I would argue that we should perhaps pause for a moment to consider what we know and what we think we know. What does appear to me to be clear is that the quartet of goddesses who most frequently appear in the Shulgi-simti archive, namely, Belat-Shuhnir and Belat-Terraban, Annunitum and Ulmashitum, had homelands in northern Babylonia and the Diyala rather than at Ur. It is not clear to me that Shulgi-simti had any exclusive links with this quartet, which appears both before her foundation and after it, in the Third Dynasty of Ur. It is not clear to me that, even if one did argue that she viewed this quartet as her “favorite” or her “personal” goddesses, that one could use this as evidence of where she was born. It is also not clear to me how “foreign” is an accurate description. Let us move through these premises one by one.

Can One Tell a Woman’s Origin from Her Favorite Goddesses? There are very good parallels for royal wives picking a form of Ishtar and venerating that form in particular. Abi-simti is well known to have had a special place in her heart for Ishtar of Zabalam (Weiershäuser 2008: 135) and Zimri-Lim’s wife Shibtu displayed a particular veneration for Ishtar of Tuba (Ziegler 1999: 40). Abi-simti did not come from Zabalam and Shibtu did not come from Tuba. These are incontrovertible facts. So I am very uncomfortable asserting that Shulgi-simti came from the Diyala on the basis of her fondness for Belat-Shunir and Belat-Terraban and Annunitum and Ulmashitum. I think the examples of Abi-simti and Shibtu ought to caution us that the royal women did not choose their favorite goddesses on the basis of their hometown. And to what extent are these goddesses in fact “foreign?” Annunitum and Ulmashitum are clearly Babylonian. Belat-Shuhnir and Belat-Terraban seem to belong in the Diyala region. Eshnunna was however a core province, which paid bala and had governors, just like Umma or Marad (Reichel 2008). It was not a foreign principality and in my opinion, one cannot talk about “dynastic marriages” in this context. As for Shulgi-simti herself, there is zero evidence as to her place of origin. She could have been local, or she could have come from abroad: we simply do not know.

16.  Westenholz 1997: 60. Many similar statements in secondary scholarship are at least phrased in such a way that the reader can see some uncertainty; for example, Bahrani tells us that the worship of Belat-Shuhnir and Belat-Terraban “seems to have been brought by Shulgi-simti from her home region” (Bahrani 2001: 108), though even here we are indulging in circular argumentation.

Local and Imported Religion at Ur Late in the Reign of Shulgi

437

The only reason to place her origins in the Diyala is her alleged promotion of Belat-­ Shuhnir and Belat-Terraban.

Was There an Exclusive Link Between Shulgi-simti and Belat-Shuhnir and Belat-Terraban? But can we really be sure that the veneration of Belat-Shuhnir and Belat-­ Terraban were so closely linked to one single woman? The worship of Belat-Shuhnir does in fact appear earlier then Shulgi-simti in the Ur III period. Puzur-­Inshushinak dedicated an inscription to Belat-Shuhnir (Steinkeller 2012: 296). It seems one of Shulgi’s earlier wives, Geme-Sin, also made offerings to Belat-Shuhnir and Belat-Terraban in S29. 17 The courtier Babati, the brother of Shu-Sin’s mother, was heaped with high ranks and titles, according to his seal: among these titles and honors was a priesthood of Belat-Shuhnir and Belat-Terraban (Whiting 1976). After Shulgi-simti’s death, we do not have an archive that shows any contin­ uity with the type of work her foundation did. But there are a handful of tablets from the reign of Amar-Sin and Shu-Sin that show us the continued veneration of Belat-Shuhnir and Belat-Terraban, particularly associated with the royal family and the king, and these observances clustered around the middle of month 10 (Sallaberger: 150). Supposing that Shulgi-simti had brought Belat-Shuhnir and Belat-­ Terraban into religious prominence at Ur (which is not to my mind at all clear), I would expect that if the goddesses were to continue to receive offerings after the death of Shulgi-simti, we would see a general continuation of what had gone before. That is, we would expect to see continued veneration in Ur along with the other two goddesses, Ulmashitum and Annunitum. As the evidence collected below indicates, however, we see a very different set-up. Belat-Shuhnir and Belat-Terraban do not generally appear together with the other two Ishtar-forms anymore; they instead appear now to be part of the official state cult and thus are some of the deities whose livestock needs were met by official bala payments. It is not unusual for Belat-Shuhnir and Belat-Terraban to occur along with netherworld-type deities in these later documents.

Belat-Shuhnir and Belat-Terraban’s Worship Late in the Ur III Period By the reigns of Amar-Sin and Shu-Sin, funding for the worship of Belat-Shuhnir and Belat-Terraban came not from a cobbled-together donation system or royal women’s foundations, but rather from the official state funds via the bala system. This is shown by two tablets. These are AUCT 2 97, whose date is not preserved but which cannot be earlier than Shu-Sin 1, and MVN 3 344, dated to AS 3.9.16. Looking first at the earlier tablet, MVN 3 344, we can see that it recorded the expenditure of oxen and sheep described as zi-ga bala ud-16-kam, “expended for the bala on the sixteenth day.” The tablet is not completely preserved, but we can see a number of deities who received sacrificial animals. The temples or shrines of Nin-sun, Allatum and Meslamtaea all received animals, as did Belat-Shuhnir and Belat-Terraban 17.  OIP 115 18: expenditure of Shu-kubum, who worked for Geme-Sin in S29.6.

438

Tonia M. Sharlach

under the authority of a scribe named Ur-gipar. This tablet, by the way, is part of the collection of the Free Library in Philadelphia. We shall see that in the next tablet, Allatum, Meslamtaea, and the bala all recur. AUCT 2 97 is a long list of livestock allocations paid for out of a province’s bala (unfortunately the text is broken and we do not know which province). Most of the livestock went to divinities, but there were also some sent to the kitchen, the ki-a-nag of Ur-Namma and the like. As for the gods, one group is described as sízkur šag4 é-gal, and these deities included Nin-gal, Nin-egal, Enki, Martu, Dumuzi, Gestin-anna, and the dlamma lugal. There were also clusters of deities in the temple of Ninsun and yet more in the temple of Gula. On the reverse (i 9–18) of AUCT 2 97 are yet more deities receiving offerings: Allatum (a netherworld deity), Ishara, Belat-Nagar, Geshtin-anna DIRI-tum, Meslamtaea (a plague god), an armed Ishtar (Inanna gištukul), and then Belat-Shuhnir and Belat-Terraban. It would be imprudent to generalize too much on the basis of two tablets that may be ten years or so apart. Still, what we see here is not just a continuation of what Shulgi-simti’s foundation was doing. Her foundation used donations from individuals to underwrite sacrifices in Ur to a quartet of goddesses; here we see official funds (identified as part of the provincial taxation system) for the two goddesses, who now are to be found not with north Babylonian Ishtar-forms but with other circles of deities. Another tablet, AUCT 2 366 is not a bala tablet but does once again show us that the financial obligations for the cults of these two deities was considered a royal obligation and so expenditures on their behalf were categorized as zi-ga lugal, royal expenditures. This tablet dates to the end of the reign of Amar-Sin (AS8.11.06). Certain tablets tell us little other than that the cults of these two goddesses continued. For example, AUCT 3 390 (dating to AS3.10.00) tells us that four fattened sheep (4 udu niga) were a regular delivery (sá-dug4) for this divine pair. Other tablets are more informative, such as Princeton 2 448, dating to SS9.10.17. The date should be noted, as we know that the next king, Ibbi-Sin, was crowned in that very month (Sigrist 1992: 389). This sealed tablet shows us that Belat-Shuhnir and Belat-Terraban continued to be honored in the palace in Ur, here with a banquet (kaš-dé-a). It is not improbable that the veneration of these deities in Ur continued throughout the reign of Shu-Sin, as the tablet Nisaba 8 77, dating to SS1.10.20 shows us that expenditures of livestock in Ur were used for Belat-Shuhnir, Belat-­ Terraban, as well as numerous other deities and physical features like the gu-za (throne) of Sulgi and Amar-Sin. Another tablet, published by Sigrist in the Levine Festschrift, also seems to date to this same time, near the death of Shu-Sin (Sigrist 1999: 132 ff). In a section that appears to concern the funerary rites for Shu-Sin himself, Belat-Shuhnir occurs, along with Inanna gištukul (Sigrist 1999: 146). Yet more data is provided by the document published as SET 57, dating to IS2.10.09. Here, the bala of Ur-mes, the provincial governor of Uru-sagrig, included sheep for rites at the ki-a-nag of Ur-Namma, Nin-sun, Lugal-banda, and numerous other deities, including Belat-Shuhnir and Belat-Terraban, as well as the dlamma lugal. This evidence indicates that there was a royal interest in the cults of Belat-­ Shuhnir and Belat-Terraban, also of course for Annunitum and Ulmashitum, that continued well into the Third Dynasty of Ur. That is to say, clearly Shulgi-simti’s foundation had a special regard for this quartet of goddesses, but veneration for them was not limited to her exclusively. I am not sure that the evidence is strong

Local and Imported Religion at Ur Late in the Reign of Shulgi

439

enough to posit that it was Shulgi-simti who caused them to be brought down to Ur, nor can their worship later in Puzrish-Dagan in court circles be linked directly to her. It does not seem to me to be warranted to use the appearance of this quartet of goddesses in ritual expenditure texts as evidence for an alleged dynastic marriage or as evidence Shulgi-simti alone venerated them as an example of personal piety. As for the cultic activities at Ur in the latter reign of Shulgi, it is not easy to reconcile what we might expect to find with what is in fact documented on tablets such as those of Shulgi-simti’s archive. The archaeological evidence shows us the magnificent, traditional state religion, and that was no doubt an important element of the religious life of Ur at that time. But texts, not texts from Ur itself, but from Puzrish-Dagan, help us fill in some of the blanks, showing a diversity even at the court itself. Religious rituals with Akkadian names were certainly conducted regularly at Ur, and at least some deities from northern Babylonia received veneration there. Shulgi-simti was a major proponent in such rites, but they were not limited to her and the evidence does not link them exclusively to her. Religion at Ur late in the reign of Shulgi, then, may have been complex and multi-layered, as one might expect in a city that for centuries had served as a regional hub.

Bibliography Bahrani, Z. 2001 Women of Babylon. London: Routledge. Cohen, M. 1993 The Cultic Calendars of the Ancient Near East. Bethesda: CDL Press. 2015 Festivals and Calendars of the Ancient Near East. Bethesda: CDL Press. D’Agostino, F. 1998 Ein neuer Text über Abī-simtī und das Elūnum-Fest in Puzriš-Dagān. ZA 88: 1–5. Frayne, D. 1997 Ur III Period (2112–2004 BC). RIME 3/2. Toronto: University of Toronto Press. Gödecken, K. 1973 Bemerkungen zur Göttin Annunītum. UF 5: 141–163. Jones, T. B. and Snyder, J. 1961 Sumerian Economic Texts from the Third Ur Dynasty. Minneapolis: University of Minnesota. Lafont, B. 1999 Sacrifices et Rituels a Mari et dans la Bible. RA 93: 57–77. Molina, M. 2015 Ur. RlA 14/5–6: 355–361. Reichel, C. 2008 The King is Dead, Long Live the King. Pp. 133–155 in Religion and Power: Divine Kingship in the Ancient World and Beyond, ed. N. Brisch. OIS 4. Chicago: Oriental Institute. Sallaberger, W. 1993 Der Kultische Kalender der Ur III-Zeit. UAVA 7. Berlin: de Gruyter. 1999 Ur III-Zeit. Pp. 121–414 in Mesopotamien: Akkade-Zeit und Ur III-Zeit, eds. W. Sallaberger and A. Westenholz. OBO 160/3. Freiburg: Universitätsverlag. 2011 Sumerian Language Use at Garšana. Pp. 335–372 in Garšana Studies, ed. D. I. Owen. CUSAS 6. Bethesda: CDL.

440

Tonia M. Sharlach

Sasson, J. 2015 From the Mari Archives: An Anthology of Old Babylonian Letters. Winona Lake, IN: Eisenbrauns. Schneider, N. 1933 Das Götterpaar dBe-la-at-múš-nir ù dBe-la-at-dír-ra-ba-an. Pp. 13–23 in Keilschriftliche Miscellanea. AnOr 6. Rome: Pontificio Istituto Biblico. Sigrist, M. 1992 Drehem. Bethesda: CDL Press. 1995 Neo-Sumerian Texts from the Royal Ontario Museum: The Administration at Drehem. Bethesda: CDL Press. Steinkeller, P. 2012 Puzur-Inšušinak at Susa: A Pivotal Episode of Early Elamite History Reconsidered. Pp. 293–307 in Susa and Elam: Archaeological, Philological, Historical and Geographical Perspectives. Proceedings of the International Congress Held at Ghent University, December 14–17, 2009, eds. K. De Graef and J. Tavernier. MDP 58. Leiden and Boston: Brill. Waetzoldt, H. 1991 Review of M. Gibson and R. Biggs, The Organization of Power. JAOS 111: 637–64. Weiershäuser, F. 2008 Die Königlichen Frauen der III. Dynastie von Ur. Göttingen: Universitätsverlag. Westenholz, J.G. 1997 Nanaya: Lady of Mystery. Pp. 57–84 in Sumerian Gods and their Representations, eds. I. Finkel and M. Geller. CM 7. Groningen: Styx. Westenholz, J. G. and Asher-Greve, J. 2013 Goddesses in Context: On Divine Powers, Roles, Relationships and Gender in Mesopotamian Textual and Visual Sources. OBO 259. Fribourg: Academic Press. Whiting, R. 1976 Tiš-atal of Nineveh and Babati, Uncle of Šu-Sin. JCS 28: 173–182. Woolley, C. L. and Moorey, P. R. S. 1982 Ur ‘of the Chaldees’. Ithaca: Cornell University Press. Ziegler, N. 1999 Le Harem de Zimrî-Lim. FM 4. Paris: SEPOA.

The Third Millennium Banquet Scene at Ur and Its Archaeological Correlates Diana L. Stein

Birkbeck, University of London

Introduction The banquet scene is a recurrent theme in ancient Near Eastern art. From its oldest known depiction on painted pottery of the Halaf period to large-scale representations on relief panels of the neo-Assyrian period, the theme comes and goes in and out of fashion as the subject of decoration on a wide variety of artifacts from Anatolia to Iran. There have been many attempts at interpretation. Most begin in the third millennium BCE with the cylinder seals from the Royal Cemetery at Ur and look to the textual record for corroborative evidence of a festive occasion that accounts for the theme’s many manifestations through space and time. Suggestions range from ritual to secular events and from specific to generic ones, 1 but they all revolve around the consumption of food and drink. 2 This underlying premise, based largely on the evidence from Ur, serves as a starting point for comparative studies on the banquet scene. 3 It also constitutes the basis for recent contributions to cross-cultural inquiries into the politics of feasting, 4 and that subject, in turn, has prompted a targeted search for evidence of feasting in the archaeological record. 5 There is no question that commensal festivities of various kinds occurred regularly and marked the public and private lives of the elite not just during the ED III 1.  Early suggestions include a funerary meal attended by the deceased (Weber 1920: 108–109; Conteneau 1931: 629–630), the Sacred Marriage (Frankfort 1934: 12 n. 3; 1939a: 43–48; and 1939b: 77–78), and the New Year Festival (Moortgat 1949: 19–21, 53–79; and 1954). Others favor a less specific meaning and propose festivities in general (Frankfort 1939b: 77; Amiet 1961: 121–130). For a comprehensive review of early research on the banquet scene see Selz 1983: 1–13. See also Draycott and Stamatopoulou 2016. 2.  Several scholars follow Michalowski (1994: 29 n. 6) in pointing to the Sumerian term for banquet, KI.KAS.NINDA, which literally means “place of beer and bread.” See, for example, Pollock 2003: 24 and Selz 2004: 190 with reference to depictions of people drinking though tubes from a common vessel. 3.  Dentzer 1982; Collon 1992; Pinnock 1994; Joannès 1996; Ziffer 2005. 4.  Schmandt-Besserat 2001; Pollock 2003, 2012; Renette 2014; O’Connor 2015. 5.  For example, Helwing 2003. The author begins by citing the banquet scenes of the third millennium BCE, which she accepts as evidence for the existence of elaborate feasts, “where both food and alcoholic beverage were provided in abundance.” This premise colors her evaluation of vessels and food remains from Tell Sabi Abyad, Degirmentepe, and Arslantepe.

441

442

Diana L. Stein

period in Mesopotamia but throughout the history and area of the ancient world. The textual record provides ample evidence of real and mythical banquets and feasts 6 that were held in honor of all manner of occasions including royal investitures, military victories, and political alliances in addition to monthly festivals celebrating the multiple gods of the pantheon. 7 Important though banquets and feasting were in the socio-economic and political life of the ancient Near East, 8 I question their relevance to the banquet scene and suggest that this iconic Near Eastern theme belongs in another cross-cultural debate of an altogether different nature. Following a brief review of the main arguments for a revised reading of the banquet scene and its archaeological background, which I have presented else­ where, 9 this paper focuses on the data from the Royal Cemetery at Ur and what it contributes to our understanding of funerary remains. The resulting picture alters our perspective on several secondary themes and motifs that were difficult to reconcile with a literal interpretation of the banquet scene. It also sits more comfortably with the banquet scene’s many manifestations on objects from different periods and places.

The Banquet Scene Reviewed Recent studies on the banquet scene refer to the dissertation of Gudrun Selz, which was published in 1983 and is the only systematic analysis of the theme avail­ able to date. Her investigation spans the Early Dynastic and Akkadian periods (ca. 3000–2159 BCE) and offers a comprehensive survey of the theme within greater Mesopotamia. The assembled corpus of 615 entries is made up of provenanced and unprovenanced material and includes votive plaques, sculpture in the round, a selection of inlays and inlaid objects, as well as painted pottery sherds, a stela, and part of a woodcarving. The bulk of the corpus consists of glyptic, however, and Licia Romano has recently re-examined the Early Dynastic seals and sealings that come from excavated sites. 10 6.  In defining these two terms, most scholars cite modern dictionaries or recent ethnographic studies on the subject (Dietler 1996: 89; Dietler and Hayden 2001; Bray 2003; Gero 2003). “Banquet” denotes a formal dinner, usually ceremonial. “Feast” suggests a more inclusive, neutral event (Renette 2014: 69). 7.  Vanstiphout 1992; Bottéro 1994: 11; Michalowski 1994; Cohen 1996: 14; Charpin 2013. Because, as some scholars have pointed out (e.g., Sallaberger 1993; Cohen 1993) most of these references postdate the Early Dynastic period, the current trend has been to focus on banquets mentioned in ED IIIB records such as the archive of the “House of the Women” (É.MUNUS) in Girsu (Schmandt-Besserat 2001: 397; Selz 2004: 188–189; Romano 2012: 270; Renette 2014: 67–68, 75–76). 8.  On this subject see Schmandt-Besserat 2001: 391; Pollock 2003: 21–27 and Renette 2014 as well as Bunimowitz and Greenberg 2004; Benz 2006; Benz and Wächtler 2006; Feldman 2014; Lewis 2014; and Emberling 2015. 9.  Stein 2020. The main arguments for a revised interpretation of the banquet scene were summarized as a case study in an article devoted to ancient Near Eastern glyptic. The present paper provides more detail and adopts a wider perspective, considering the full range of artifacts, contexts and imagery associated with the banquet scene, while reflecting on its particular relevance to the Royal Cemetery at Ur. 10. The corpus of 615 entries includes 529 seals and seal impressions (86%), thirty-eight votive plaques (6%), thirty statues in the round (5%), fourteen inlays and inlaid objects (2%), and two painted pottery sherds (.3%). More than a third of the 615 entries is unprovenanced. Of the ca. 352 seals and sealings that date to the ED period, ca. 200* come from twenty-one excavated sites, but in several cases the date of their context is still subject to debate. *Romano (2015: 291–292) mentions 300 excavated examples, but the number of seals in her Table 1 totals 207.

The Third Millennium Banquet Scene at Ur

443

The traditional interpretation of the banquet scene is based on two mistaken assumptions. The first derives from an uncritical adoption of Selz’s division of the theme into three main types. 11 The second hinges on a mistaken impression that the banquet scene is quintessentially Mesopotamian 12 and best represented by a small group of well-preserved seals from the Royal Cemetery of Ur. 13 Closer inspection of Selz’s subgroups reveals that there are, in fact, only two main types: people drinking through tubes from a common vessel and people drink­ ing from cups. The third type, defined by the presence of a food-laden table, is by Selz’s own account comparatively rare. In addition, she observes, the table seldom occurs alone, unaccompanied by a drinker of one type or the other,  14 and it is also worth mentioning that people hardly, if ever, eat from it. The so-called banquet scene thus appears to revolve around the act of drinking, and the food-laden table is not an essential component of the scene. 15 Most scholars agree that the drink was alcoholic. The vessels are likely to have contained beer, perhaps even a specific type called kaš2-su-ra (“strained beer”) that was ingested by means of a straw. 16 The cup may have contained another variety of beer or some kind of wine. 17 How large were these drinking events and was there a competitive angle as several scholars now suggest? 18 A quick scan through Selz’s catalogue of “banquet scenes” reveals that the majority of the 615 entries depicts only one or two seated participants. According to Romano’s survey of excavated seals and seal impres­ sions, single drinkers outnumber multiple drinkers in the Khabur/Jezirah region and are also common throughout the Early Dynastic period in the northern part of lower Mesopotamia. Scenes depicting two or more participants are clustered in south-central Mesopotamia with the majority dating to the ED IIIA period and coming from the Royal Cemetery at Ur. 19 It is these more exceptional seals—which measure double the average in height and diameter—that often include additional components of the drinking ceremony. Besides the food-laden table, they might show musicians, dancers, acrobats, and attendants. Similar expanded versions of the drinking ceremony usually occur on larger artifacts such as votive plaques and 11.  For example, Collon 1986: 93 and 1992; Ermidoro 2016: 221–223. 12.  For example, Pinnock 1994, 17; Ziffer 2005: 133; Zajdowski 2013: 2; Renette 2014: 68. 13.  For example, Schmandt-Besserat 2001: 397; Selz 2004: 188–189; Romano 2012: 270; Renette 2014: 77–78 with fig. 1. 14.  Selz 1983: 436–437. See also 462: “da das Trinken für fast alle Bankettszenen konstitutiver Akt ist….” 15.  For similar assessment of basic theme see Hansen 1963: 161. 16.  See Selz 1983: 447–448 and Romano 2012: 273 with reference to Powell 1994 and Hornsey 2003: 86. “Strained beer” contains emmer in the form of “imgaga” in addition to malt “munu” and “bappir” (Damerow 2012: §4.7). Straws filtered out the barley husks and stalks, most of which would float on the drink’s surface. Imbibing beer through tubes or straws also accelerates the rate of intoxication. Although straws are rarely found in the archaeological record, bone and/or metal strainer tips attached to the bottom of straws have been found at a number of sites. These could be used for all kinds of beverages, including wine. (Homan 2004: 86). 17.  Collon (1992: 24) suggested wine or date wine, pointing out that evidence for wine does not exist in Mesopotamia before the second millennium BCE. However, there is a reference to “wine” in the Early Dynastic hymn to the Sun God (Michalowski 1994: 32), and the earliest known winery now dates to the fourth millennium BCE. See n. 25 for earlier evidence. 18.  For example, Pinnock 1994: 24; Pollock 2003: 21–27; Benz and Wächtler 2006; Renette 2014: 78. 19. Romano 2015: 295. Even at Ur, nearly half of the “banquet scenes” feature only two seated participants.

444

Diana L. Stein

the Standard of Ur. Rather than a lavish display of conspicuous consumption, we seem to be dealing with the depiction of a carefully staged, ritual drinking ceremony that was sponsored by and for an affluent elite. 20 The figures have been described as stock and impersonal, the setting appears to be irrelevant, and the act of drinking seems to be conducted in a formal and prescribed manner. 21 The uniform character of the core drinking scene is difficult to reconcile with the notion of competitive feasting, although references to status and wealth may appear in the secondary imagery (see below). Informative though the “banquet scenes” on seals from the Royal Cemetery at Ur may be, they do not represent the norm, nor does this theme originate in Mesopotamia during the third millennium BCE; this was not the place or time of its origin. Representations of one or two people drinking from vessels or cups date back to the sixth millennium BCE and come from the Mesopotamian periphery. Examples include a painted Halaf vase, several Ubaid stamp seals, and an Uruk cylinder seal (Fig. 1). 22 The surreal appearance of the drinkers on these seals is taken as an indication of the ecstatic nature of the ritual. To quote Frank Hole, “The human form, in stylized posture, is the first convincing evidence of humans acting a role that we think of today as namash (= shaman)… who is thought to be endowed with the ability to communicate with, and influence the behavior of, supernatural forces” (2010: 234). Whereas Hole dates the emergence of elaborate ritual to the fifth millennium BCE and claims that this marks the beginning of a tradition that came to epitomize Mesopotamian religious expression, I would argue that these early scenes of ritual drinking and its ecstatic nature are nothing new. They illustrate one aspect of an age-old ritual ceremony that is attested in the archaeological record from as early as the PPNA period.

The Archaeological Background Representations of surreal human figures occur at Göbekli Tepe near Urfa, where an estimated twenty megalithic circular enclosures covered the site during its earlier PPNA phase (Layer III). All of the excavated enclosures are con­struct­ed alike, with huge T-shaped pillars interspersed with walls and stone benches surrounding two even taller pillars in the middle. Indications of shoulders, arms, hands and elements of clothing point to the anthropomorphic character of these pillars that were also decorated with animal motifs such as foxes, snakes, scorpions, and the like. As the excavators have pointed out, these abstract, impersonal, but distinctly anthropomorphic, T-shaped beings clearly belong to another transcendent sphere (Dietrich et al. 2012: 679). Various kinds of meetings and rituals are likely to have occurred inside these megalithic enclosures, but the prominence of predatory animals and carrion birds, the odd headless figure, disembodied head and human-bird hybrids—all of which have mortuary connotations—suggest that burial rites fea-

20. The term “banquet scene” was introduced by Contenau (1922: 109–110). It was one of several proposed labels, including “drinking scene” (“Trinkszene”) and “symposion/symposium” (Selz 1983: 1–11). The latter was adopted more recently by Reade 1995 and Romano 2015, but like “banquet” and “feast,” this term implies a large convivial gathering of people. 21.  Pollock 2003: 25; Renette 2014: 79. 22.  Hijara 1978: fig. 1; Amiet 1980: nos. 42 and 122; Behm-Blanke 1993: fig. 1.1; Hole 2010: 227–243.

The Third Millennium Banquet Scene at Ur

445

Fig. 1a. Drinking ritual with contest, dancing, and symbols. Arpachiyah, Burial G2, ca. 5200– 4500 BCE.

Fig. 1b. Drinking ritual with large jar. Tepe Gawra, Level XII, ca. 4500–3800 BCE.

Fig. 1c. Drinking ritual with ecstatic aspect. Susa I, ca. 4500–3800 BCE.

Fig. 1d. Drinking ritual with animals. Arslantepe VIA, ca. 3400–3200 BCE.

Fig. 1. Earliest depictions of the “banquet scene.”

Fig. 2. Painted and impressed wall decoration. Palace Corridor, Arslantepe VIA, ca. 3400–3200 BCE.

446

Diana L. Stein

tured among them. 23 Alcohol played a key role in these events. Some of the large limestone basins that were permanently fixed inside these megalithic enclo­sures contained evidence of oxalate. 24 With capacities of up to 160 liters, these basins would have provided the assembled group of up to thirty people with plenty of beer (or malt). Alcohol and funerary rites are also linked at Körtik Tepe, a PPNA site near Batman in southeastern Turkey, where ceremonies performed at the grave­side involved wine and appear to have had an ecstatic aspect (McGovern 2009). 25 Inter­ estingly, the earliest known winery at Arenı-1 in Armenia (ca. 4000 BCE) is also connected with burials. According to the excavators, the winery supplied alcohol for funerary and sacrificial rites not for feasting, as was originally assumed (Maugh 2011 and Areshan et al. 2011: 68). Decorated, non-domestic enclosures exist at other PPN sites as well. 26 The iconography, elements of which survive into the Pottery Neolithic period and beyond, is also widespread and points to the existence a common system of symbolic communication that revolves around the universal experience of death and altered states of consciousness. 27 At a number of these ritual sites, large amounts of animal bones have been found. In some cases, such as Göbekli Tepe and the burial site of Zawi Chemi-Shanidar in northwest Iran, where the majority of bones belong to the wings of carrion birds, we get a hint of costumes and the theatrical nature of these early ecstatic rites. 28 The remainder of the faunal debris can be attributed either to feasting or to animal sacrifice. By the late Neolithic Halaf period ritual practice seems to be in the hands of an elite at certain sites. 29 Evidence of this development is also apparent at the late Ubaid period site of Değirmentepe on the Upper Turkish Euphrates, where most of the elite tripartite buildings in Layer 7 show signs of having served both as domestic houses and as venues for ceremonial gatherings. The gatherings took place in the main halls, most of which were equipped with ovens, podiums, and tables. The presence of seals and sealings suggests the recording of goods delivered in portable containers. Evidence of ritual activity (feasts or sacrifice) exists in the form of numerous rubbish pits filled with animal bones, 30 and this impression is reinforced by the discarded bowls left over from liquid libations or drinks and by the mural decorations consisting of V-shaped symbols and sun- and tree-motifs. Another possible example of ritual performance involving liquids comes from a public building of the Late Uruk period at Arslantepe VII in the Malatya plain. 23.  While no burials have yet been located at the site, human bones found in the back filling of the enclosures suggest that some form of burial or excarnation took place at the hilltop site (Schmidt 2010: 243). 24. All six of these large basins were found in PPNB contexts but fragments of similar vessels appear in all strata (Dietrich et al. 2012: 687). 25.  Before this discovery, the oldest evidence for wine was detected in a Neolithic (mid sixth millennium BCE) jar from Hajji Firuz Tepe in northern Iran (McGovern et al. 1996). Traces of wine and beer found at Godin Tepe date to the late fourth millennium BCE (Badler 2000). On the oldest brewery dated to the eleventh millennium BCE, see Liu et al. 2018. 26.  Examples include Çayönu, Nevali Çorı, Hallan Çemi, Nemrik, Qermez Dere, Mureybet I, Jerf el-Ahmar, Tell ‘Abr 3, and Tell Qarmel. See Dietrich et al. 2012: 675 for references. 27.  Morenz and Schmidt 2009; Dietrich et al. 2012: 679; Notroff et al. 2015: 40. 28.  Solecki et al. 2004; Schmidt 2010; Notroff et al. 2015: 42. 29.  For example, Arpachiyah (Stein 2019). 30.  Helwing 2003: 71–74.

The Third Millennium Banquet Scene at Ur

447

The main hall (A900) of Building XXIX (Temple C), a bipartite structure that stood raised on a platform overlooking the settlement, was decorated with niches and several layers of red and black wall paintings. One motif painted onto the northern wall resembles liquid flowing from a vase. The podium in the center of the hall shows traces of fire, and fragments of more than 1,100 mass-produced bowls covered the surrounding area. Similar bowls were found stacked ready for use in a side room (A932), along with broken clay sealings from portable containers that had been opened on delivery. Rather than evidence of feasts and ceremonial redistribution, as has been proposed, 31 we could be looking at the remains of ritual drinking ceremonies and the sacrificed goods donated by the participating clientele. If, along with Helwing (2003: 76–77), we assume that the bowl fragments originate from a single ceremony, then the event was indeed huge, involving hundreds of people. But the hall, which measures approximately 7 x 20 meters, was too small to accommodate such a large gathering of people. Alternatively, if people consumed more than one drink or the bowls were left to accumulate, the number attending any one event would have been more limited. The temple complex of Arslantepe VIA, which dates to the end of the fourth millennium BCE, contains much the same assemblage. Large numbers of seals and sealings were found in storerooms and retrieved from dumps in which they appear to have been deposited in discrete groups. There is no correspondence between the sealings from these two locations. Whereas those discarded in dumps belonged to inhouse administrators, the container sealings recovered from the storerooms entered the complex with goods delivered from outside. 32 Some of these latter sealings are believed have been in use for only a short span of time (Ferioli and Fiandra 1983: 468). This observation and the conspicuous lack of large quantities of cereal led Frangipane to conclude that long-term storing of agricultural surplus was not a primary concern (1997: 69). Instead, the combined evidence points to a quick turn­ around of processed foods. This would be consistent with the accumulated faunal debris. The choice cuts of meat from wild game found discarded in the main halls as well as in dumps may have been sacrificed and possibly consumed shortly after delivery. Vine residues discovered in storeroom A340 raise the possibility of alcohol consumption, and nearby mural decorations consisting of stamped and painted patterns and abstract motifs suggest the performance of ecstatic rites that now take place within the controlled and exclusive environment of a palace-temple complex (Fig. 2). In light of the cumulative evidence for ritual drinking ceremonies during the late fourth millennium BCE near the northern limits of greater Mesopotamia, one ought perhaps to reconsider the evidence at contemporary temple complexes further south. The Level IVb sanctuary of Eanna at Uruk comprises several monumental temple and administrative buildings elevated on a platform. This platform is accessed by means of a columned portico and the “Cone Mosaic Courtyard.” The walls and columns surrounding this courtyard were decorated with geometric patterns in red, black and white. Thousands of colored baked clay cones were arranged in a variety of designs including zigzags, lozenges, triangles, and diagonal bands (Fig. 3). 103.

31.  Helwing 2003: 76 with reference to Frangipane 2001: 3 and 2002: 124. See also D’Anna 2012:

32. Contra Helwing, who maintains that the sealings “prove clearly that the exchange of goods occurred” (2003: 78).

448

Diana L. Stein

Traditionally compared with the weave of reed matting and elaborate reed houses, these designs are also reminiscent of trance-induced imagery. 33 Entoptic patterns of this kind – that resemble some of the temple decoration at Arlsantepe VIA – would certainly be consistent with the religious function of the “Cone Mosaic Courtyard” and its adjacent buildings.

Fig. 3. Reconstruction of red, white, and black cone mosaic columns. Eanna IVB, Uruk, ca. 3300–3000 BCE.

Entering the Early Dynastic period in southern Mesopotamia, the evidence for ritual drinking continues. At Tell Asmar, for example, large stacks of solid-footed drinking vessels were found in a storeroom of the Abu Temple (Archaic Shrine III), and heaps of similar vessels lay strewn around the courtyards of Sin Temples V and VI at Khafaje. The courtyard of Sin Temple V also contained two circular kilns, which recall the brewery kilns in the temple at Lagash (al-Hiba). Evidently, alcohol (beer) played a key role in the rituals of both cities. 34 Drinking ceremonies were also an important part of contemporary funerary rites. Several of the graves excavated at Abu Salabikh, and Tell Kheit Qasim, for example, contained hundreds of drink­ ing vessels. 35 And finally at Ur, where some of the royal tombs contained dozens of people, many held a cup in their hand. What they consumed remains a mystery, but Woolley believed it was an intoxicant of some kind. 36

The Drinking Ceremony at Ur As a group, the thirty-four exceptional seals from the Royal Cemetery at Ur are particularly revealing when viewed against the wider corpus of Early Dynastic “banquet scenes” (Table 1). Apart from the additional iconographic details which 33.  Stein 2006, 2009, 2017 and 2019 with reference to Lewis-Williams and the “Three Stages of Trance” (TST) model. 34.  Selz 2004: 193–194; Romano 2012: 270–272 with reference to brewery kilns in the courtyard of the Sin Temple V at Khafaje and Hansen’s identification of a temple brewery at al-Hiba (1978: 82–83). See also Homan 2004: 84–85, “beer had an integral role in religious observance in the ancient Near East. In addition to Ninkasi, several other deities were linked to beer and beer production, including Siris, Dumuzi, Enlil, Inanna, Hathor, Menqet, Dionysus and Ceres…. Many temples even had their own brewers.” 35.  Grave nos. 80, 205, and 234 at Abu Salabikh (Postgate 1977: 286 and 1984: 103; Matthews and Postgate 1987: 100; Martin, Postgate, and Moon 1985). See also Wright 1969: 83 and Forest 1983: 136. 36.  Woolley 1982: 72–75 with reference to PG 1237.

The Third Millennium Banquet Scene at Ur

449

they include, most of these seals come from a recognizable context, several are inscribed with the seal owner’s name, and a few can be associated with a particular profession or rank. Together this information clarifies a number of points concern­ ing the nature of the ceremony depicted, its setting, the people involved, and their dress and equipment. Table 1. Tabulation of drinking ritual on thirty-four seals from the Royal Cemetery at Ur. No. and Sex of Clients

Selz 1983 cat. no.

Seal Material

Associated Person(s)

Inscription

2: m?/m

No. 225

white calcite

female dancers

PG 1237

Mesanepada

2: m/f

No. 254

gold/bitumen shell caps

female

PG 1054

Ninbanda, queen, wife 1 of Meskalamdug

2: m/f

No. 256

limestone

-- --

Space for legend

RC

2: m/f

No. 257

lapis lazuli

?

Hekunsig nin-dingir d pa-bìl

PG 580

A’anepada, king, s. of Mesanepada

2: f/m

No. 260

lapis lazuli

female

PG 1237

Mesanepada

2: m/m

No. 261

?

female

2: ?/m

No. 282

lapis lazuli

male

PG 1625

2: m/?

No. 294

lapis lazuli

?

PG 1847

2: m/f

No. 297

lapis lazuli

female

PG 381

2: m/f

No. 298

lapis lazuli

female

PG 1130

2: f/f

No. 302

lapis lazuli

?

PG 55

2: m/f

No. 320

shell

female

PG 357

3: ?/f/?

No. 230

lapis lazuli

?

PG 89 nearby

3: f/f f

No. 245

lapis lazuli

female

space for legend

PG 800

Puabi, queen, wife 2 of Meskalamdug

3: f/f m

No. 246

shell

male

ga-den-ki

PG 1236

A’imdugud, king

3: f/m m

No. 251

marble

?

3: m/m/f

No. 255

lapis lazuli

female harp player

3: f/f/f

No. 281

lapis lazuli

3: m/f/m

No. 321

lapis lazuli

space for legend

Findspot

Principal Occupant (Reade 2001)

PG 1749

Ur III grave?

PG 1774 dumu-kisal daughter of the court

PG 1237

Mesanepada

female

PG 1237

Mesanepada

male

PG 1312, in front of body

with female ornaments

450

Diana L. Stein

No. and Sex of Clients

Selz 1983 cat. no.

Seal Material

Associated Person(s)

Inscription

3: m/m/f

No. 325

lapis lazuli

female

PG 789

4: f/f m/f

No. 227

lapis lazuli

male

PG 31

4: f/m m/f

No. 228

lapis lazuli

female

PG 1315, on chest

4: f/f f/f

No. 229

lapis lazuli

female

PG 1130, beside body

4: f/m m/m

No. 231

lapis lazuli

female

4: m/f m/f

No. 248

lapis lazuli

male

PG 221

4: m/?/? ?

No. 253

marble

male

PG 37

5: m/? f/f/?

No. 233

lapis lazuli

male

PG 156

5: f/m/m m/f

No. 236

lapis lazuli

female

5: f/m/m f/f

No. 241

lapis lazuli

female

PG 15

6: m/f/m f/m/f

No. 238

shell

?

PG 1332

6: f/m/f f/m/f

No. 239

lapis lazuli

?

a-bará-ge son of the true divine chamber?

PG 800, at the end of the wooden chest

6: m/m/m m/m/m

No. 240

lapis lazuli

female?

space for legend

PG 337

Queen, wife of Akalamdug

7: ?/m/f/? f/m/m

No. 243

lapis lazuli

female

space for legend

PG 789

Meskalamdug, king, s. of Urpabilsag

7: f/m/m m/m/m/m

No. 244

lapis lazuli

male

Puabi nin

space for legend

Findspot

PG 800, on upper right arm

PG 800

Principal Occupant (Reade 2001) Meskalamdug, king, s. of Urpabilsag

Puabi, queen, wife of Meskalamdug

Puabi, queen, wife of Meskalamdug

Akalamdug, king, s. of Meskalamdug

PG 1750

1. The Clientele Based on the assembled corpus of Early Dynastic drinking scenes, the ceremony was open to both male and female clients. Although the number of seated figures attending any one event was normally as little as one or two, it could be as high

The Third Millennium Banquet Scene at Ur

451

as seven. 37 Gatherings usually consist of couples, but family groups and groups of single sex drinkers occur as well. 38 Where the sex of the clientele is identifiable, women outnumber men.  39 The cylinder seals, votive plaques, and statues of people holding cups provide several clues to the identity and social standing of these drinkers. Where the owner of the object is identified by inscription, we may assume that he or she features as the principal figure in the scene. The votive plaque of Urnanshe is one example, and several other inscribed plaques, statues, and seals fall into this category as well. Mostly, the inscriptions refer to prominent members of the community such as rulers, courtiers, priests, and priestesses, 40 but they also include professional performers. 41 In the absence of an inscription, information on the status of the clients, who took part in ritual drinking ceremonies, can be gleaned from the material and findspot of the objects they commissioned. Most of the cylinder seals from the Royal Cemetery at Ur are made of expensive, imported materials such as lapis lazuli and shell. One of these seals belonged to a harpist, a lady of the court, and several accompanied the principal occupant of the royal tombs. 42 The votive plaques and statues that commemorate the drinking ceremony belonged to a similarly privileged elite, as many of these objects were found in the vicinity of monumental buildings and temples, places that were inaccessible to the general public.

2. The Attendant Staff and Personnel The number of people attending to the seated figures varies depending on the size of the object in question. Not all standing attendants are necessarily sub­servient

37.  See Table 1: Two participants (12x), three participants (8x), four participants (6x), five and six participants (3x), and seven participants (2x). 38.  We see pairs of women (Selz 1983: no. 302) or men (ibid. nos. 225 and 261). Women also occur in groups of three (ibid. nos. 245, 281) and four (ibid. no. 229). Men sometimes appear in groups of six (ibid. no. 240) and six/seven (Standard of Ur). The votive plaque of Urnanshe, ruler of Lagash, provides an example of a family group. 39.  As several scholars have noted, it is often difficult to identify the gender of the participants in “banquet scenes” on Early Dynastic seals (see Asher-Greve 1985: 111–114; Pollock 2003: 22; Romano 2015: 293). Rathje, in his study of the RC seals from Ur, observed that most were associated with female burials (1977: 25–32; Selz 1983: 460). Moorey advocates caution about jumping to conclusions (1977: 35–36 and 39). Of the thirty-four seals depicting the “banquet scene”, seventeen to eighteen accompanied a female skeleton, eight accompanied a male, and in the remaining eight cases, the sex is unknown (see Table 1). In the case of the Early Dynastic votive statues, only four out of thirty are male: Selz 1983: nos. 105 (Agrab), 199 (Khafaje), 441 (Mari), and 442 (unprov.). Two represent couples: ibid. nos. 200 (Khafaje) and 443 (Susa). In addition to Selz’s group of thirty statues in the round, one might also add some of the statues found buried in the Abu Temple at Eshnunna (Tell Asmar), which likewise hold cups (Evans 2014: 652–653). 40.  For example, Selz 1983: nos. 203(?), 204, 231, 239, 246, 255, and 257. 41.  Among the women, who owned or were associated with drinking scene seals, we find a group of dancers and a harpist (see Selz 1983: nos. 225 and 255, respectively). As performers in the ceremony, they too might have been entitled to use this exclusive theme on their personal seals. 42.  In the case of the harpist, the seal inscription identifies her as a lady of the court. According to Woolley, the lyre players in PG 789 were “the most richly adorned” of all in the tomb (1934: 65). On the subject of who was buried in the Royal Cemetery see Moorey 1977; Pollock 1991; Reade 2001; Marchesi 2004; Vidale 2011. Another extended version of the “banquet scene” with possible links to temple staff comes a sealing found in the Early Dynastic I levels of the Inanna temple at Nippur (see Zettler 2011: fig. 3).

452

Diana L. Stein

to the seated drinkers, as is often assumed. 43 While this characterization may apply to those holding towels and flywhisks or fans, the other attendants appear to be professionals. All rituals, especially those involving the consumption of medicinal brews, were carefully monitored by trained specialists. This is apparent from later magicomedical texts as well as from illustrations of magical rites. 44 The attendant figure, who stands before the seated client in the expanded versions of the ritual-drinking scene, probably fulfilled this role. 45 Usually there is one hierophant per seated figure, but sometimes this person attends to several clients. The hierophant can be male or female and often they are paired with clients of the same sex. On one example, a votive plaque from Nippur (Inanna temple VIII), the female attendant may even be a priestess. 46 This suggests that in other cases too, the lead attendant is likely to be a qualified religious practitioner, someone equal or superior in status to the customers they guide through a drinking ceremony. Among the other professionals attending the drinking ceremony are musicians, singers, chanters, dancers, and acrobats. They all contributed in some way to the surreal experience of the seated clientele and were probably affiliated with the temple household. These highly trained artists are known to have performed at burials and commemorative festivals, where music and wrestling were the norm. 47 Their presence in the royal tombs is noteworthy, as is the fact that at least two of them owned seals engraved with the drinking scene, while others held elaborately decorated musical instruments that we see represented in the drinking scene (Fig. 4). 48 Indeed, one of these instruments, a bull-headed lyre, was found lying beside

Fig. 4a. Painted scarlet ware. Khafaje?

Fig. 4b. Votive plaque. Inanna Temple VIIB, Nippur.

Fig. 4. Bull-headed lyres in scenes of the drinking ritual. 43.  Frankfort refers to them as “butlers and servants” (1939b: 77–78). See also Pinnock 1994: 16; Pollock 2003: 22; Costello 2017. 44.  For example, in scenes of exorcism such the famous Pazuzu plaque and several cylinder seals. See Wiggermann 2000: 219 n. 11 for list of Lamashtu plaques. 45.  Selz 1983: 460 calls these particular attendants “Mundschenk.” 46.  Selz 1983: no. 75 and Stein 2020, fig. 8.3 (IM 66154). 47.  According to one possible reading of difficult lines in the epic poem “Gilgamesh’s Dream and Death” (Selz 2004: 190). 48.  The bull-headed lyre is so far only encountered in scenes of the ritual drinking ceremony. It appears on seals of the ED I–II period (Amiet 1980: no. 1201 = Selz 1983: no. 37), a painted Scarlet Ware fragment that may come from Khafaje (Selz 1983: no. 38), a number of ED III seals (Amiet 1980: nos. 1193 and 1194 = Selz 1983: nos. 254 and 255), a votive plaque (Hansen 1963: pl. VI, [7N133–134] from Level VIIB of the Inanna Temple, Nippur) and the Standard of Ur.

The Third Millennium Banquet Scene at Ur

453

the female musician, whose lapis lazuli seal portrays a woman playing an identical instrument at a ritual drinking ceremony. 49

3. Ceremonial Costumes and Equipment There are several close parallels between the contents of Woolley’s sixteen royal tombs and the accessories depicted in extended versions of the ritual drinking scene, and some of them appear to be exclusive to this ceremony. Among the common items are harps and lyres. Harps played a significant role in funerary ritual according to later Sumerian texts. 50 Lyres probably feature in many contexts as well, but the bull-headed lyre is distinctive and appears to have a particular association with the ritual drinking ceremony. As mentioned above, one example was found in situ beside a person, whose seal suggests that she played this instrument during such a ceremony. Another example, the Great Lyre from PG 789, 51 has a decorative front panel that contains an illuminating self-reference. It portrays a funerary ritual, complete with an offering table laden with choice cuts of meat, jar and pouring vessel, a sistrum, and a bull-headed lyre that resembles the one on which it is depicted (Fig. 5a). 52 The animal actors allude to another world, as does the scorpion man below, a figure known from the Epic of Gilgamesh as the guardian of the entrance to the Underworld. The key to opening the doors to the Underworld to allow the spirit of the deceased individual(s) to cross over seems to be contained in the two tumblers that have just been filled from a jar in the lower right-hand corner and are being proffered by the upright goat standing beside it. Identical tumblers are seen held by the seated drinkers on votive plaques, inlay musical instruments and some seals (Fig. 5b). They are also a distinctive accessory of several votive statues found in temples and shrines. In the archaeological record, similar vessels have been found stacked ready for use in temple storerooms and deposited in the royal tombs at Ur. 53 They have been likened to a vessel called ku-li, which signifies “tumbler of friendship.” 54 In any case, the small size and shape of the tumblers and solid-footed goblets that are often found in Early Dynastic burials make them unsuited for containing anything but liquids. As Susan Pollack points out (2003: 31), the solid-footed goblets have a capacity of less than 0.4 liters and neither vessel was suited for setting down or walking about when full. Whatever they contained, it is likely to have been consumed on the spot. The tumblers from the royal tombs are made of gold and silver (Fig. 5c–d). The rosette that decorates 49.  See n. 42. For similar observation see Baadsgaard, Monge, and Zettler 2012: 152. 50. A later reference to “mourning harps” in connection with Shu-Sin’s funerary ritual gives an indication of their importance in this context (Katz 2007: 177). 51.  Woolley 1934: pl. 105; Hansen 1998: 53–57). Also relevant is the sound box of a smaller lyre from PG 1332 (Woolley 1934: pl. 116, U. 12435). 52.  The food and objects that were put into the grave suggest that the spirit was still in the corpse at the time of interment (Katz 2007: 172). 53.  Pollock 2003: 31 and n. 34 above. Not all the bodies in the royal tombs were provisioned with cups or bowls, and it has been suggested that only those who were should be identified as active participants in the ceremony (Baadsgaard, Monge, and Zettler 2012: 138, 152). Perhaps these were the ones involved in the final drinking ceremony that took place inside the tomb. 54.  Selz 2004: 203. The ku-li was intended for drinking beer. If this identification of the tumbler is correct, the vessels and cups represented in “banquet scenes” may refer to two kinds of beer. During the Early Dynastic period, at least five types of beer are attested. They are described as golden, dark, sweet dark, red and strained (Peter Damerow 2012: 5–8). See n. 16 above.

454

Diana L. Stein

Fig. 5a. Great lyre, front panel of sound box. PG 789, Royal Cemetery at Ur.

Fig. 5b. Inlay, front panel of lyre. PG 1332, Royal Cemetery at Ur.

Fig. 5c. Electrum tumbler.

Fig. 5d. Electrum tumbler base. Fig. 5. Ritual equipment.

the base of these tumblers and the band of herringbone design above a double zigzag that encircles the rim are telltale references to entoptic imagery and the cultic function they were designed to fulfill. 55 Rosettes and flowers feature prominently in the decoration of costumes and furniture recovered from the royal tombs. The ceremonial dress of the female attendants included an elaborate headdress that was topped by multi-petalled flowers made of gold and lapis lazuli (Fig. 6a). 56 According to Gebhard Selz, these flowers bear a striking resemblance to the fruit stands of atropa belladonna or atropa mandragora, plants that have a long history of use in ecstatic rituals as aphrodisiacs and as poison. 57 Their alkaloid content also causes a dilation of the pupils, which he notes is a characteristic feature of Early Dynastic votive statues, many of whom hold 55.  See G. Selz, who identifies the zigzag with the cuneiform sign for “water” and the herringbone band with the sign for “barley” (2004: 200). A double-entendre is possible. Rosettes are a prominent feature of the so-called Glazed Steatite seals, whose decoration seems to have been largely inspired by mental imagery (Stein 2017: 517–520). As a symbol of the luminous light at the end of the tunnel, the rosette’s association with divinity or “the beginning of life,” as Selz suggests (2004: 201), is possible. 56. For example, the sixty-four female attendants in Death Pit 1237. For possible examples of super­natural heads or headdresses on seals see Selz 1983: nos. 93, 114(?) 119, 162, and 193. For an example of an animal costume see Amiet 1980: no. 843 from Fara, ED I–II. 57.  Selz 2004: 194 with n. 36.

The Third Millennium Banquet Scene at Ur

Fig. 6a. Headdress of lady attendant. PG 800, Royal Cemetery at Ur.

Fig. 6b. Marble cylinder seal. Umma?

455

Fig. 6c. Table support. PG 1237, Royal Cemetery at Ur.

Fig. 6. Ritual costumes and furniture.

Fig. 7a. Votive plaque. Temple Oval, Khafaje.

Fig. 7b. Votive plaque. Shara Temple, Tell Agrab.

Fig. 7c. Votive plaque. Basel, Erlenmeyer.

Fig. 7. Supplementary motifs alluding to status and wealth.

Fig. 8. Votive plaque. Sin Temple IX, Khafaje.

456

Diana L. Stein

tumblers and are included in Selz’s catalogue of “banquet scenes.” 58 Floral rosettes also decorate items of furniture that seem to be restricted to ritual contexts. The offering table seen in expanded versions of the drinking scene sometimes rests on an elaborate stand in the form of an upright bull or goat (Fig. 6b). 59 Two examples of such a stand were found in PG 1237, and both incorporate a tree with branches ending in floral rosettes (Fig. 6c). 60 The pair of goats nibbling from trees echoes the age-old iconographic motif of goats flanking trees. If my assessment of that motif is correct and it refers to an hallucinogenic plant (Stein 2009 and 2017), then the goat-stands may provide yet another reference to the ritual use of stimulants. Plant motifs, it should be noted, feature prominently in third millennium BCE drinking scenes, both as a secondary motif in two-dimensional illustrations and as an accessory in the hands of seated statues. 61

4. The Secondary Imagery At its most basic, the so-called banquet scene consists of one or two drinking figures. This abbreviated form of the theme sufficed as proof of attendance at a ceremony that was exclusive and took place behind closed doors. Additional motifs that shed light on the nature of this ceremony can be found on larger seals and votive objects. Many of these motifs were difficult to explain in relation to banquets and feasts, but in the context of the drinking ceremony, they make perfect sense. In fact, most of the secondary imagery falls into one of four categories of information relating to the drinking ceremony. 1.  Some of the motifs allude to the occasion or circumstance that prompted a performance of the drinking ceremony. In addition to the hunting expeditions, military campaigns and building projects, reasons for seeking divine assistance or giving thanks might include personal issues such impotence. 62 A number of votive plaques and other commemorative objects feature visual references to a person’s wealth or social standing (chariots, boats, food, and offering-bearers, etc.) by way of demonstrating his/her ability to finance a

58.  Large eyes and dilated pupils are also diagnostic of people, who meditate with open eyes. The implications are much the same; namely, that the subject has transcended and is in the presence of the divine. On the subject of staring eyes, Winter (2000: 36) observes “in addition to the focus of attention, what we are seeing in those eyes is literally the physical manifestation of the individuals struck by the u6-di response to the deity before him/her, a deity whose awe-inspiring nature is then reflected in the wide-eyed stare.” These same meditating eyes are the defining feature of the Uruk-period eye idols from Brak. Like several of the Early Dynastic votive statues, the eye idols sometimes come in pairs or family groups. So too do the PPNB lime-plaster statues from Ain Ghassal (Jordan) that also have prominent eyes and were buried in hoards. Could they all belong to the same ongoing tradition? 59.  For example, Moortgat 1940: no. 144 (Berlin VA 3878) from Umma(?) and a votive plaque from the Inanna temple, Level VIII at Nippur (IM 66154 = Selz 1983: no. 75). 60.  See also Moortgat 1940: 96 and Woolley UE 2: p. 265, re pls. 87–90. By contrast, Winter suggests that the rams might have formed part of the frame of musical instruments (1987: n. 33). 61.  See Stein 2017: 525–526, fig. 10–11. Selz 1983: 454 with n. 3 provides a list of statues of seated drinkers who hold a cup in one hand and a date cluster or branch (palm fond?) in the other. Strong palm wine derived from the sap of various palm trees including date palms, can be highly intoxicating. 62.  Later medical and magical texts refer to the treatment of infertility, and in one case, the impotent patient was given an hallucinogenic drink to induce dreams of a sexual encounter while he slept in the temple (Haas 2003:121, X.2.2–3).

The Third Millennium Banquet Scene at Ur

457

private drinking ceremony. The delivered goods were no doubt intended for display, sacrifice, and consumption (Fig. 7). 63 2.  Other motifs refer to the ceremony itself. This category includes the foodladen table, an altar, and all attendant staff and personnel: hierophants, musicians, chanters, acrobats, and servants (Fig. 4b, Fig. 7, and Fig. 8). To the extent that sexual acts can be part of ecstatic rituals, scenes of this nature could fall into this category as well. 64 3.  The third category of supplementary motifs refers to various sources of intoxication. This might include plants, trees, date clusters, fish, 65 frogs, scorpions, snakes, or anything else that causes delirium and that in combination with various dramaturgical techniques, helps the subject to loosen his grip on reality, focus his attention on the present, and become more receptive to episodes of heightened emotional experience (Fig. 9 and Fig. 10a). 66 4.  The fourth category of motifs relates to the trance experience itself. The patterns, distortions, mythical beings, and surreal situations record the subject’s mental visions and encounters (Fig. 10). These images echo the decoration—often abstract or bizarre—that decorate the walls of prehistoric temples where similar rituals appear to have taken place. 67 I suspect that the mythical contest scene belongs in this category (Stein forthcoming).

Fig. 9a. Plant.

Fig. 9b. Fish.

Fig. 9c. Fish.

Fig. 9d. Fish.

Fig. 9e. Scorpion.

Fig. 9f. Plant and scorpion.

Fig. 9. Secondary motifs referring to source of intoxication. 63.  For example, Selz 1983: nos. 59, 60, and 106; and Amiet 1980: no. 1223 (Khafaje). 64. For examples see Amiet 1980: nos. 845, 850; Boese 1971: AS 4; and Frankfort 1955: pl. 53 no. 599 = Amiet 1980: no. 1203). On the association between women and alcohol in sexual scenes, see Michalowski 1994: 38; Joffe 1998; Mazzoni 2002. 65.  Near Eastern fish known to induce hallucinations include the Sarpa salpa, also known as the dreamfish, which is found off the coast of Israel and causes LSD-like symptoms when eaten (http://www. readcube.com/articles/10.1111/j.1095-8649.1989.tb03342.x?locale=en) and the dusky spinefoot, which lives in the Red Sea and the Persian Gulf, and can cause “neurological symptoms.” The Mesopotamian association of fish with Ea, god of Wisdom, may stem from this source of hallucinogen. 66.  See Stein 2017: Figs. 10–11; Selz, nos. 28, 37, 375, 411, 469; Hansen 1963: pl. V (plaque 7N408 from the Inanna Temple, Level VIII at Nippur). 67.  For further examples of this category of secondary motifs see Amiet 1980: nos. 830, 832, 1054, 1055, 1056, 1060. See also Stein forthcoming: fig. 3.

458

Diana L. Stein

5. Tweaking the Tomb Tableau The number of cross-references between the so-called banquet scenes and the contents of the royal tombs is remarkable. They share the same elite clientele, servants, skilled staff, elaborate costumes, and specially designed equipment. The remnants of cooked food in the tombs, and depiction of food-laden tables in the “banquet scenes,” and the presence in both of tumblers, musicians, chanters, and dancers together allude to a staged, multi-sensory production. It is tempting to conclude that the scenario discovered in the royal tombs mirrors the scene engraved on several associated seals, and in fact, this has been suggested before. But whereas others have assumed that we are looking at a funerary feast, 68 I propose instead that we are dealing with a ceremony of a more general kind, one that involves an hallucinogenic beverage and seeks contact with the spirit world. In the context of a funeral, this drinking ceremony may have occurred in private or at the graveside, but before the general feasting and wailing events that preceded the burial accord­ ing to Early Dynastic III texts from Lagash. 69 This would accord with what most likely occurred at other, non-funerary occasions, where the drinking ceremony was performed as well (see below). As some of the royal tombs attest, the outcome of the ceremonial drinking ritual could be lethal and sometimes resulted in the death and mass burial of numerous attendant staff and personnel. Perhaps these particular deaths had something to do with the unusual size of the event and the need to prevent details of its secret proceedings from reaching the public. 70 Recent radiographs, CT scans, and forensic analyses of select skeletons from the royal tombs reveal that some of the victims succumbed to blows to the head while others underwent embalming and preservation. 71 This suggests that the performance of the drinking ceremony took place outside the death pits, probably behind closed doors. The transfer of the bodies to the tombs and their careful arrangement may have been intended to replicate the ritual at death. 72

The Longue Durée The archaeological record suggests that from the PPNA period alcohol played a key role in the ritual process of obtaining altered states. 73 Much of the archaeological evidence links this process to rites performed at burials and within religi68.  For example, Pollock 2003 and 2012; Cohen 2005; Baadsgaard, Monge, and Zettler 2012. Cohen observes that none of the tombs were equipped for feasting except in the narrowest sense, that is, the consumption of food and drink (2005: 92). 69.  Jagersma 2007: 291–294; Katz 2007: 174–182. 70.  On the secrecy surrounding esoteric knowledge see Huffmon 1992: 479; Hunger 1968: 12–14; Geus and Geller 2014: 3–4. 71.  Baadsgaard, Monge, and Zettler 2012. The authors stress that the mass burials took place as part of “a single coherent set of funerary events” (ibid. 137). 72.  A variation on Porter 2012, who suggests that the victims were meant to depict ritual at death. 73.  See Guerra-Doce 2015 for archaeological evidence attesting to the widespread use of psychoactive substances to induce altered states of consciousness (ASC) from as early as prehistoric times. Interestingly, the list of four categories of archaeological data on which this study relies includes “artistic depictions of mood-altering plant species and drinking scenes/artistic depictions inspired by ASC” (ibid. 94). At Ebla, a powerful city-state of the mid-third millennium BCE in Syria, traces of opium poppy and other hallucinogenic plants were found in a herbal kitchen at the heart of palace G. Contemporary texts refer to priests taking an unknown ritual beverage, so some of the palace-produced drugs may have had

The Third Millennium Banquet Scene at Ur

459

ous buildings. In the case of burials, the ritual consumption of alcohol helped to facilitate the journey of the deceased and to establish contact with the spiritual world. Both these objectives lie at the heart of Sumerian funerary rituals, which were meant to liberate the soul from the body and take care of the corpse. 74 Traces of similar funerary practices are encountered throughout the ancient Near East. In the Caucasus, braziers containing hemp seeds were found in third millennium BCE graves. 75 Funerary vessels of a certain shape and type point to the consumption of alcohol and other psychoactive substances as part of the mortuary ritual in Late Bronze Age Cyprus. 76 And contemporary Late Bronze Age texts from Ugarit refer to an elite religious ritual, called marzaeah, that involves drinking excessive amounts of alcohol in order to commune with ancestral spirits. 77 This particular role of the drinking ceremony accounts for the theme’s survival and recurrent illustration on a wide range of objects associated with death and burial, including the twelfth century BCE sarcophagus of Ahiram from Byblos, the funerary stele of the Syro-Hittite peri­od from southeastern Anatolia, and the bronze situlae from Early Iron Age Iran. 78 The seeking of supernatural contact, advice, and powers was not confined to burial rites. It occurred in other situations as well. Since Palaeolithic times, we have evidence for the performance of such rituals in connection with the hunt, 79 and there is ample textual and material evidence from greater Mesopotamia attesting to the need for divine support in all manner of situations, including military maneuvers and building projects as well as in the treatment of personal, inter-personal, and political issues. 80 As a performance designed to facilitate a spiritual encounter, the drinking ceremony was staged for different reasons. It marked official occasions such as the laying of building foundations (votive plaque of Urnanshe) or a military victory (Standard of Ur), but it also occurred at the request of wealthy individuals. This may explain the many statues, plaques, seals, ivory boxes, and other decorative objects that were made to commemorate the experience. It was personal, exclusive, and denoted access to the spiritual world. 81 Whereas the drinking ceremony was usually a private affair, the feasting and festivities that followed are likely to have been more public. 82

Concluding Remarks Man’s relation to the spirit world is a theme that dominates the art of the ancient Near East from Neolithic times. Early depictions of the drinking ceremony usually center on the village “shaman,” who performs the ritual. This focus shifts to a ceremonial usage (Peyronel et al. 2014 with n. 110 and reference to textual evidence for ceremonial use of hallucinogenic beverages at Mari in the second millennium BCE) 74.  Katz 2007: 167. 75.  Sherratt 1991: 51–57 and 1995: 27. 76.  Collard 2012: 24–29. 77.  Pope 1972: 190–193; Armstrong 1998: 87–93; McLaughlin 2001. 78.  Porada 1973; Rehm 2004; Pinnock 1994; Calmeyer 1973. 79.  For example, representations and models of animals that are ritually pierced. See inter alia Formozow 1963 and Betts 1987. 80.  See for example, Marchetti 2009; Annus 2010; Rutz 2014; Koch 2015. 81.  On the question of who participated in the drinking ceremony, see also Stein 2017 and 2020. 82.  See Stein 2020 for comparison between the drinking ceremony with its associated festivities and the performance of the Eleusinian mysteries in Classical Greece.

460

Diana L. Stein

Fig. 10a. Abstract design.

Fig. 10b. Geometric design.

Fig. 10c. Undulating pattern.

Fig. 10d. Symbols.

Fig. 10e. Undulating pattern.

Fig. 10f. Contest scene.

Fig. 10g. Contest scene.

Fig. 10h. Contest scene.

Fig. 10i. Hunt/contest scene.

Fig. 10j. Distorted figures.

Fig. 10. Secondary motifs reflecting experience of altered mental states.

The Third Millennium Banquet Scene at Ur

461

the principal participants during the third millennium BCE in Mesopotamia, when an emergent ruling class seizes upon this aspect of the ceremony as an emblem of its privileged position in society. The evolution of this theme over the course of the Early Dynastic and Akkadian periods (as witnessed mainly on seals and sealings) reveals subtle but significant changes both in the presentation of the religious experience and in the dynamics between the temple and palace institutions. Several scholars have remarked upon the fact that divinities and divine symbols do not appear in “banquet scenes” until later in the third millennium BCE. 83 Indeed, up until that time, religious experience was portrayed from a personal perspective. We see ritual dancing, drinking and/or the associated mental imagery. From the ED IIIB period onwards, divine participants in anthropomorphic form become more frequent, and by the time the ritual-drinking scene gives way to the presentation scene, the identity of the cup-holder is either royal or divine. These changes mark a pivotal point in the evolution of Near Eastern culture, when Mesopotamia puts its own stamp on a common heritage and takes it in a new direction. No longer is the experience of altered states expressed in terms that are individual to the seal owner. It is now interpreted in a manner that reflects and legitimates a new political order. Members of the elite, who are entitled to an official seal, are henceforth divided into two groups. One is led directly into the presence of the divine, while the other is presented to its earthly mediator, the king. 84 Momentous though the implications of these changes are, the adapted imagery continues to serve the same purpose. In life it identifies the owner of a seal, plaque, or statue as a member of a privileged group that has a direct or indirect line to divinity, and in death it documents a prior relationship with god and perhaps also the right to a higher order of existence in the afterlife.

List of Illustrations Fig. 1. Earliest Depictions of the “Banquet Scene.” Fig. 1a. Painted Vase: Drinking Ritual with Contest and Dancing. Arpachiyah (Iraq). Halaf bowl from ritual burial G2, tholos area, Late Halaf, ca. 5200–4500 BCE (a) exterior (b) interior (c) base. Hijara 1978: fig. 1 Fig. 1b. Stamp Seal Impression: Drinking Ritual. Tepe Gawra, Level XII, Ubaid, ca. 4500–3800 BCE Amiet 1980: no. 42 Fig. 1c. Stamp Seal Impression: Drinking Ritual with Ecstatic Aspect. Susa I Ubaid, ca. 4500–3800 BCE Amiet 1980: no. 122

83.  Frankfort 1939b: 77; Selz 1983: 445. 84. Winter 1986; Zajdowski 2013; Renette 2014. In the latter two cases, read “drinking scene” instead of “banquet” and “feast.” In common with Zajdowski, I would argue that the “presentation scene” cannot be understood without prior consideration of the “banquet scene.”

462

Diana L. Stein

Fig. 1d. Cylinder Seal Impression: Drinking Ritual With Animals. Arslantepe VIA Uruk, ca. 3400–3200 BCE Behm-Blanke 1993: fig. 1.1 Fig. 2. Fig. 2. Painted and Impressed Wall Decoration. Arslantepe VIA, Palace corridor Uruk, ca. 3400–3200 BCE Palumbi 2010: fig. VII.1a © MAIAO, Sapienza University of Rome Fig. 3. Fig. 3. Reconstruction of Red, White, and Black Cone-Mosaic Columns. Eanna IVb, Uruk Uruk, ca. 3300–3000 BCE © Staatliche Museen zu Berlin-Vorderasiatisches Museum, Foto: Olaf M. Teßmer Fig. 4. Bull-Headed Lyres in Scenes of the Drinking Ritual. Fig. 4a. Painted Scarlet Ware Fragment. Khafaje(?) Early Dynastic II, ca. 2700–2500 BCE Selz 1983: no. 38 Fig. 4b. Votive Plaque. Inanna Temple VIIB, Nippur Early Dynastic IIIa, ca. 2550–2400 BCE Hansen 1963: pl. VI (7N133-134) Fig. 5. Ritual Equipment. Fig. 5a. Front Panel of the Great Lyre. PG 789, Royal Cem. Ur (U.10556) Early Dynastic IIIa, ca. 2550–2400 BCE © Penn Museum, B17694 Fig. 5b. Inlay Front Panel of Lyre. PG 1332, Royal Cem., Ur (U.12435) Early Dynastic IIIa, ca. 2550–2400 BCE © Penn Museum, 30-12-484 Fig. 5c–d. Electrum Tumbler and Base. PG 800, Royal Cem., Ur (U.10453) Early Dynastic IIIa, ca. 2550–2400 BCE © Penn Museum, B17691 Fig. 6. Ritual Costumes and Furniture. Fig. 6a. Headdress of Attendant. PG 800, Royal Cem., Ur Early Dynastic IIIa, ca. 2550–2400 BCE Woolley 1982: 172

The Third Millennium Banquet Scene at Ur Fig. 6b. Marble Cylinder Seal. Umma(?), Berlin VA 3878 Early Dynastic III, ca. 2550–2250 BCE © Staatliche Museen zu Berlin-Vorderasiatisches Museum, Foto: Olaf M. Teßmer Fig. 6c. Table Support. PG 1237, Royal Cem., Ur Early Dynastic IIIa, ca. 2550–2400 BCE © Penn Museum, 30-12-702 Fig. 7. Secondary Motifs Alluding to Occasion, Status, and Wealth. Fig. 7a. Votive Plaque. Oval Temple, Khafaje Early Dynastic III, ca. 2550–2250 BCE Amiet 1980: no. 1223 Fig. 7b. Votive Plaque. Shara Temple, Tell Agrab Oriental Institute, Chicago (A.18073) Early Dynastic II, ca. 2700–2550 BCE Boese 1971: Tf. I,2 Fig. 7c. Votive Plaque. Unprovenanced Erlenmeyer Collection, Basel Boese 1971: Tf XXXVIII Fig. 8. Secondary Motifs that Expand on the Drinking Ritual. Fig. 8. Votive Plaque. Khafaje, Sin Temple IX Oriental Institute, Chicago (A12417) Early Dynastic II–III, ca. 2700–2550 BCE Amiet 1980: no. 1222 (Frankfort 1939a: pl. 105, no. 185) Fig. 9. Secondary Motifs Referring to Source of Intoxication. Fig. 9a. Cylinder Seal Impression. Ishtar Temple, Mari Early Dynastic Amiet 1980: no. 1168 Fig. 9b. Fragment of Votive Plaque. Inanna Temple VIII, Nippur Early Dynastic II–III, ca. 2700–2250 BCE Boese 1971: Tf. XVI,1 (Hansen 1963: pl. V [7N408]) Fig. 9c. Cylinder Seal Impression. PG 1749, Ur Royal Cem., Ur Early Dynastic IIIa, ca. 2550–2400 BCE Amiet 1980: no. 1190 (Woolley 1934: 200, no. 102) Fig. 9d. Cylinder Seal Impression. SIS 4, Ur Early Dynastic II–III, ca. 2700–2550 BCE Amiet 1980: no. 836 (Legrain 1936: no. 381)

463

464

Diana L. Stein

Fig. 9e. Cylinder Seal Impression. Fara Early Dynastic II, ca. 2700–2550 BCE Amiet 1980: no. 1169 (Heinrich and Andrae 1931: Tf. 63n) Fig. 9f. Cylinder Seal Impression. Ishtar Temple, Mari Early Dynastic II–III, ca. 2700–2250 BCE Amiet 1980: no. 1174 (Parrot 1956: pl. LXVI, 567) Fig. 10. Secondary Motifs Reflecting Experience of Altered Mental States. Fig. 10a. Cylinder Seal Impression: Abstract Design. SIS 4–5, Ur (U. 18397, 18400, 18413) Early Dynastic II, ca. 2700–2550 BCE Amiet 1980: no. 832 (Legrain 1936: no. 169) Fig. 10b. Cylinder Seal Impression: Geometric Pattern. Assur Early Dynastic, ca. 2900–2550 B.C Amiet 1980: no. 1054 (Moortgat 1940: no. 142) Fig. 10c. Cylinder Seal Impression: Undulating Pattern. Iraq Museum (IM 21090) Early Dynastic II, ca. 2700–2550 BCE Amiet 1980: no. 1055 Fig. 10d. Cylinder Seal Impression: Symbols. Iraq Museum (IM 5631) Early Dynastic II, ca. 2700–2550 BCE Amiet 1980: no. 1056 Fig. 10e. Cylinder Seal Impression: Undulating Pattern. Khafaje Early Dynastic II, ca. 2700–2550 BCE Amiet 1980: no. 1060 (Frankfort 1955: 416, Khafaje III.278) Fig. 10f. Cylinder Seal Impression: Contest Scene. Unprovenanced (Bibliothèque Nationale) Early Dynastic II–III, ca. 2700–2250 BCE Amiet 1980: no. 1179 Fig. 10g. Cylinder Seal Impression: Contest Scene. Unprovenanced (Pierpont Morgan Library) Early Dynastic II–III, ca. 2700–2250 BCE Amiet 1980: no. 1178 Fig. 10h. Cylinder Seal Impression: Contest Scene. Unprovenanced Early Dynastic, ca. 2900–2250 BCE Amiet 1980: no. 1180

The Third Millennium Banquet Scene at Ur

465

Fig. 10i. Votive Plaque: Hunt/Contest Scene. Ninhursag Temple, Susa Early Dynastic II, ca. 2700–2550 BCE Boese 1971: Tf. XXIV, 2 Fig. 10j. Cylinder Seal Impressions: Distorted Figures. SIS 4–5, Ur Early Dynastic II, ca. 2700–2550 BCE Amiet 1980: no. 840 (Legrain 1936: no. 377)

Bibliography Amiet, P. 1961 La Glyptique mésopotamienne archaïque. Second Edition (1980). Paris: CNRS. Annus, A. 2010 Divination and Interpretation of Signs in the Ancient World. OIS 6. Chicago: The Oriental Institute. Areshian, G. E. et al. 2011 Wine and Death: The 2010 Excavations Season at the Arenı-1 Cave Complex, Armenia. Research in Action, Backdirt: 65–70. Armstrong, D. E. 1998 Alcohol and Altered States in Ancestor Veneration Rituals of Zhou Dynasty China and Iron Age Palestine. Lewiston, NY: Edwin Mellen Press. Asher-Greve, J. M. 1985 Frauen in altsumerischer Zeit. Bibliotheca Mesopotamica 18. Malibu: Undena Press. Baadsgaard, A., Monge, J., and Zettler, R. L. 2012 Bludgeoned, Burned and Beautified: Reevaluating Mortuary Practices in the Royal Cemetery of Ur. Pp. 125–158 in Sacred Killing. The Archaeology of Sacrifice in the Ancient Near East, eds. A. M. Porter and G. M. Schwartz. Winona Lake, Indiana: Eisenbrauns. Badler, V. 2000 The Dregs of Civilisation: 5000 Year Old Wine and Beer Residues from Godin Tepe, Iran. BCSMS 35: 49–56. Behm-Blanke, M. R. 1993 Glyptische Traditionen beiderseits des Ost-Taurus im ausgehenden 4. und frühen 3. Jahrtausends v. Chr. Pp. 247–255 in Between the Rivers and Over the Mountains: Archaeologia Anatolica et Mesopotamica Alba Palmieri dedicate, eds. M. Frangipane, H. Hauptmann, M. Liverani, P. Matthiae, and M. Mellink. Rome: La Sapienza. Benz, M. 2006 Zur Bedeuting von Festen während der Neolithisierung im Vorderen Orient. Ethnographisch-Achäologische Zeitschrift 47: 439–462. Benz, M. and Wächtler, N. 2006 Von der Integration zur Distinktion: Die feiernde Elite der Frühdynastischen Zeit. Ethnographisch-Archäologische Zeitschrift 47: 463–483. Betts, A. V. G. 1987 The Hunter’s Perspective: 7th Millennium BC Rock Carvings from Eastern Jordan. World Archaeology 19: 214–225. Boese, J. 1971 Altmesopotamische Weihplatten. Eine sumerische Denkmalsgattungen des 3. Jahrtausend v. Chr. UAVA 6. Berlin/New York: Walter de Gruyter.

466

Diana L. Stein

Bottéro, J. 1994 Boisson, banquet et vie sociale en Mésopotamie. Pp. 3–13 in Drinking in Ancient Societies. History and Culture of Drinks in the Ancient Near East. Papers of a Symposium held in Rome, May 17th–19th 1990, ed. L. Milano. Padova: Sargon srl. Bray, T. L. 2003 The Archaeology and Politics of Food and Feasting in Early States and Empires. New York: Kluwer Academic/Plenum. 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. Calmeyer, P. 1973 Reliefbronzen in babylonischem Stil. Eine westiranische Werkstatt des 10. Jahrhunderts v. Chr. Bayrische Akademie der Wissenschaften, Philosophischhistorische Klasse, Abhandlungeen NF 73. Munich: Verlag der Bayerischen Akademie der Wissenschaften. Charpin, D. 2013 Les usages politiques des banquets d’après les archives mésopotamiennes du début du IIe millenaires av. J.-C. Pp. 31–52 in Le Banquet du monarque dans le mond antique: Table des homes, eds. C. Grandjean, C. Hugonoit, and B. Lion. Rennes: Presses Universitaires François-Rabelais. Cohen, A. C. 2005 Death, Rituals, Ideology and the Development of Early Mesopotamian Kingship. Leiden: Brill. Cohen, M. E. 1993 The Cultic Calendars of the Ancient Near East. Bethesda: CDL Press. 1996 The Sun, the Moon, and the City of Ur. Pp. 7–20 in Religion and Politics in the Ancient Near East, ed. A. Berlin. Bethesda: University Press of Maryland. Collard, D. 2012 Drinking with the Dead: Psychoactive Consumption in Cypriote Bronze Age Mortuary Ritual. Pp. 23–32 in Food and Drink in Archaeology 3: University of Nottingham Postgraduate Conference 2009, eds. D. Collard, J. Morris, and E. Perego. Totnes: Prospect Books. Collon, D. 1986 Review of Selz 1983. AfO 33: 92–98. 1992 Banquets in the Art of the Ancient Near East. Pp. 23–30 in Banquets d’Orient, ed. R. Gyselen. Res Orientales vol. IV. Bures S/Y: Groupe pour l’Etude de la Civilisation du Moyen-Orient. Contenau, G. 1922 La Glyptique Syro-Hititte. Paris: Paul Geuthner. 1931 Manuel d’achéologie orientale II. Paris: Auguste Picard. Costello, S. K. 2018 Rematerializing the Early Dynastic Banquet Seal. Pp. 68–80 in Seals and Sealings in the Ancient World: Case Studies from the Near East, Egypt, Aegean and South Asia, eds. M. Ameri, S. K. Costello, G. Jamison, and S. J. Scott. Cambridge: Cambridge University Press. Damerow, P. 2012 Sumerian Beer. The Origins of Brewing Technology in Mesopotamia. Cuneiform Digital Library Journal (2): 1–20.

The Third Millennium Banquet Scene at Ur

467

D’Anna, M. B. 2012 Between Inclusion and Exclusion: Feasting and Redistribution of Meals at Late Chalcolithic Arslantepe (Malatya, Turkey). Pp. 97–123 in Between Feasts and Daily Meals: Toward an Archaeology of Commensal Spaces, ed. S. Pollock. Berlin: Exzellenzcluster 264 Topoi. Dentzer, J.-M. 1982 Le motif du banquet couché dans le Proche-Orient et le monde grec du VIIIe au IVe siècle avant J.-C. Bibliothèque des Ecoles françaises d’Athènes et de Rome 246. Paris: Diffusion de Boccard. Dietler, M. 1996 Feasts and Commensal Politics in the Political Economy: Food, Power and Status in Prehistoric Europe. Pp. 87–125 in Food and the Status Quest: An Interdisciplinary Perspective, eds. P. Wiessner and W. Schiefenhövel. Providence: Berghahn Books. Dietler, M. and Hayden, B. 2001 Feasts: Archaeological and Ethnographic Perspectives on Food, Politics, and Power. Washington, D.C.: Smithsonian Institution Press. Dietrich, O. et al. 2012 The Role of Cult and Feasting in the Emergence of Neolithic Communities. New Evidence from Göbekli Tepe, South-eastern Turkey. Antiquity 86: 674–695. Draycott, C. M. and Stamatopoulou, M. 2016 Dining & Death. Interdisciplinary Perspectives on the ‘Funerary Banquet’ in Ancient Art, Burial and Belief. Colloquia Antiqua 16. Leuven: Peeters Press. Emberling, G. 2015 Structures of Authority: Feasting and Political Practice in the Earliest Mesopotamian States. Pp. 34–59 in Social Theory in Archaeology and Ancient History: The Present and Future of Counternarratives, ed. G. Emberling. Cambridge: Cambridge University Press. Ermidoro, S. 2015 Commensality and Ceremonial Meals in the Neo-Assyrian Period. Antichristica 8, Studi orientali 3. Venice: Edizioni Ca’Foscari. Evans, J. M. 2014 The Tell Asmar Hoard and Rituals of Early Dynastic Sculpture. Pp. 645–665 in Critical Approaches to Ancient Near Eastern Art, eds. B. A. Brown and M. H. Feldman. Boston/Berlin: Walter de Gruyter. Feldman, M. 2014 Religious, Communal, and Political Feasting in the Ancient Near East. Pp. 63–68 in Remembrance of Me: Feasting with the Dead in the Ancient Middle East, eds. V. R. Hermann and J. D. Schloen. Exhibition Catalogue. Oriental Institute Museum Publications 37. Chicago: The Oriental Institute of the University of Chicago. Ferioli, P. and Enrica, F. 1983 Clay Sealings from Arslantepe VIA: Administration and Bureaucracy. Pp. 455–509 in Perspectives on Protourbanization in Eastern Anatolia: Arslantepe (Malatya). An Interim Report on the 1975–83 Campaigns, ed. M. Frangipane and A. Palmieri. Origini 12/2. Rome: La Sapienza. Forest, J.-D. 1983 Les pratiques funéraires en Mésopotamie du cinquième millenaire au début du troisième. Editions Recherche sur les Civilisations, Mémoire n. 19. Paris: A. D. P. F. Formozov, A. A. 1963 The Petroglyphs of Kobystan and their Chronology. Rivista di Sienze Prehistor­ ische 18: 91–115.

468

Diana L. Stein

Frangipane, M. 1997 A 4th-Millennium Temple/Palace Complex at Arslantepe-Malatya. North-south Relations and the Formation of Early State Societies in the Northern Regions of Greater Mesopotamia. Paléorient 23/1: 43–73. 2001 The Transition Between Two Opposing Forms of Power at Arslantepe (Malatya) at the Beginning of the 3rd Millennium BC. TÜBA-AR, Türkish Akademy of Sciences Journal of Archaeology 4: 1–24. 2002 Non-Uruk Developments and Uruk-Linked Features on the Northern Borders of Greater Mesopotamia. Pp. 123–148 in Artefacts of Complexity. Tracking the Uruk in the Near East (Iraq Archaeological Reports 5), eds. J. N. Postgate and S. Campbell. Warminster: Aris and Phillips Ltd. Frankfort, H. 1934 Gods and Myths on Sargonid Seals. Iraq 1: 2–29. 1939a Sculpture of the Third Millennium B.C. from Tell Asmar and Khafājah. OIP 44. Chicago: University of Chicago. 1939b Cylinder Seals. A Documentary Essay on the Art and Religion of the Ancient Near East. London: Macmillan. 1955 Stratified Seals from the Diyala Region. OIP 72. Chicago: University of Chicago. Gero, J. M. 2003 Feasting and the Practice of Stately Manners. Pp. 285–288 in The Archaeology and Politics of Food and Feasting in Early States and Empires, ed. T. L. Bray. New York: Kluwer Academic/Plenum. Geus, K. and Geller, M. 2014 Esoteric Knowledge in Antiquity. Berlin: MPIWG. Guerra-Doce, E. 2015 Psychoactive Substances in Prehistoric Times: Examining the Archaeological Evidence. Time and Mind 8/1: 91–112. Haas, V. 2003 Materia Magica et Medica Hethitica. Ein Beitrag zur Heilkunde im Alten Orient. Berlin: Walter de Gruyter. Hansen, D. 1963 New Votive Plaques from Nippur. JNES 22/3: 145–166. 1978 Al-Hiba. A Summary of Four Seasons of Excavation: 1968–1976. Sumer 34: 72–85. 1998 Art of the Royal Tombs of Ur: A Brief Introduction. Pp. 43–74 in Treasures from the Royal Tombs of Ur, eds. R. L. Zettler and L. Horne. Philadelphia: University of Pennsylvania Museum of Anthropology and Archaeology. Heinrich, E. and Andrae, W. 1931 Fara. Ergebnisse der Ausgrabungen der Deutschen Orient-Gesellschaft in Fara und Abu Hatab. Berlin: Staatliche Museen zu Berlin. Helwing, B. 2003 Feasts as a Social Dynamic in Prehistoric Western Asia – Three Case Studies from Syria and Anatolia. Paléorient 29/2: 63–86. Hijara, I. 1978 Three New Graves at Arpachiyah. World Archaeology 10/2: 125–128. Hole, F. 2010 A Monumental Failure: The Collapse of Susa. Pp. 227–243 in Beyond the Ubaid: Transformation and Integration in the Late Prehistoric Societies of the Middle East, eds. R. Carter and G. Philip. Chicago: Oriental Institute of the University of Chicago. Homan, M. M. 2004 Beer and Its Drinkers: An Ancient Near Eastern Love Story. NEA 67/2: 84–95.

The Third Millennium Banquet Scene at Ur

469

Hornsey, I. S. 2003 A History of Beer and Brewing. Cambridge: Royal Society of Chemistry. Huffmon, H. B. 1992 Prophecy (ANE). Pp. 477–482 in Anchor Bible Dictionary V, ed. D. N. Freedman. New York: Doubleday. Hunger, H. 1968 Babylonische und assyrische Kolophone. AOAT 2. Neukirchen-Vluyn: Verlag Butzon & Berker Kevelaer. Jagersma, B. 2007 The Calendar of the Funerary Cult in Ancient Lagash. BiOr 64: 289–308. Joannès, J. 1996 La fonction sociale du banquet dans le premières civilisations. Pp. 47–59 in Histoire de l’alimentation, eds. J.-L. Flandrin and M. Montanari. Paris: Fayard. Joffe, A. 1998 Alcohol and Social Complexity in Ancient Western Asia. Current Anthropology 39: 297–322. Katz, D. 2007 Sumerian Funerary Rituals in Context. Pp. 167–188 in Performing Death. Social Analyses of Funerary Traditions in the Ancient Near East and Mediterranean, ed. N. Lanieri. OIS 3. Chicago: The Oriental Institute of the University of Chicago. Koch, U. S. 2015 Mesopotamian Divination Texts: Conversing with the Gods. Sources from the First Millennium BCE. GMTR 7. Münster: Ugarit-Verlag. Legrain, L. 1936 Archaic Seal Impressions. UE 3. Oxford: Trustees of the British Museum and of the University Museum, University of Pennsylvania, Philadelphia; Oxford University Press. Lewis, T. J. 2014 Feasts for the Dead and Ancestor Veneration in Levantine Traditions. Pp. 69–74 in Remembrance of Me: Feasting with the Dead in the Ancient Middle East, eds. V. R. Hermann and J. D. Schloen. Exhibition Catalogue. Oriental Institute Museum Publications 37. Chicago: The Oriental Institute of the University of Chicago. Liu, L., Wang, J., Rosenberg, D., Zhao, H., Lengyel, G., and Nadel, D. 2018 Fermented Beverage and Food Storage in 13,000 y-old Stone Mortars at Raqefet Cave, Israel: Investigating Natufian Ritual Feasting. Journal of Archaeological Science: Reports 21: 783–793. Marchesi, G. 2004 Who Was Buried in the Royal Tombs of Ur? The Epigraphic and Textual Data. Or. NS 73: 153–197. Marchetti, N. 2009 Divination at Ebla during the Old Syrian Period: The Archaeological Evidence. Pp. 279–295 in Exploring the Longue Durée: Essays in Honor of Lawrence E. Stager, ed. D. Schloen. Winona Lake, Indiana: Eisenbrauns. Martin, H., Postgate, J. N. and Moon, J. 1985 Graves 1 to 99. Abu Salabikh Excavations, vol. II. London: British School of Archaeology in Iraq. Matthews, R. J. and Postgate, J. N. 1987 Excavations at Abu Salabikh, 1985–86. Iraq 49: 91–119. Maugh, T. H., II. 2011 Ancient Winery Found in Armenia. Los Angeles Times, January 11. See https://www.latimes.com/archives/la-xpm-2011-jan-11-la-sci-ancient-winery20110111-story.html.

470

Diana L. Stein

Mazzoni, S. 2002 The Squatting Woman: Between Fertility and Eroticism. Pp. 367–377 in Sex and Gender in the Ancient Near East: Proceedings of the 47th Rencontre Assyriologique Internationale, Helsinki July 2–6, 2001, eds. S. Parpola and R. M. Whiting. Comptes Rendus Rencontre Assyriologique Internationale 47. Helsinki: NeoAssyrian Text Corpus Project. McGovern, P. E. 2009 Uncorking the Past: The Quest for Wine, Beer and Other Alcoholic Beverages. Berkeley and London: University of California Press. McGovern, P. E., Glusker, D. L., Exner, L. J., and Voigt, M. M. 1996 Neolithic Resinated Wine. Nature 381: 480–81. McLaughlin, J. L. 2001 The Marzeah in the Prophetic Literature, References and Allusions in Light of the Extra-Biblical Evidence. Leiden: Brill. Michalowski, P. 1994 The Drinking Gods: Alcohol in Mesopotamian Ritual and Mythology. Pp. 27–44 in Drinking in Ancient Societies. History and Culture of Drinks in the Ancient Near East. Papers of a Symposium held in Rome, May 17th–19th 1990, ed. by L. Milano. Padova: Sargon srl. Moorey, R. 1977 What Do We Know About the People Buried in the Royal Cemetery? Expedition 20/1: 24–40. Moortgat, A. 1940 Vorderasiatische Rollsiegel. Berlin: Gebr. Mann Verlag. 1949 Tammuz. Der Unsterbichkeitsglaube in der altorientalische Bildkunst. Berlin: Walter de Gruyter. 1954 Der Bilderzyklus des Tammuz. Pp. 18–41 in Compte Rendu de la troisième Rencontre Assyriologique Internatalionale organisée à Leiden du 28 juin au 4 juillet 1952. NINO 1. Leiden: Brill. Reprinted in: Anton Moortgat, Kleine Schriften zur Vorderasiatischen Altertumskunde, Damascus: Tartolls 1990, 407–446. Morenz, L. D. and Schmidt, K. 2009 Große Reliefpfeiler und kleine Zeichentäfelchen. Ein frühneolitisches Zeichensystem in Obermesopotamien. Lingua Aegyptia 8: 13–31. Notroff, J., Dietrich, O., Peters, J., Pöllath, N. and Köksal-Schmidt, Ç. 2015 What Modern Lifestyles Owe to Neolithic Feasts: The Early Mountain Sanctuary at Göbekli Tepe and the Onset of Food-Production. Actual Archaeology 15: 32–49. O’Connor, K. 2015 The Never-Ending Feast: The Anthropology and Archaeology of Feasting. London/ New York: Bloomsbury Academic. Palumbi, G. 2010 Pastoral Models and Centralised Animal Husbandry. The Case of Arslantepe. Pp. 149-63 in Economic Centralisation in Formative States. The Archaeological Reconstruction of the Economic System in 4th millennium Arslantepe, ed. M. Frangipane. Studi di Preistoria Orientale 3. Rome: Sapienza University of Rome. Parrot, A. 1956 Mission archéologique de Mari V.I: Le temple d’Ishtar. Institut français d’archéologie de Beyrouth. Bibliothèque archéologique et historique tome 65. Paris: P. Geuthner. Peyronel, L., Vacca, A., and Wachter-Sarkady, C. 2014 Food and Drink Preparation at Ebla, Syria. Food & History 12/3: 3–38.

The Third Millennium Banquet Scene at Ur

471

Pinnock, F. 1994 Considerations on the ‘Banquet Theme’ in the Figurative Art of Mesopotamia and Syria. Pp. 15–26 in Drinking in Ancient Societies. A History and Culture of Drinks in the Ancient Near East. Papers of a Symposium held in Rome, May 17th–19th 1990, ed. L. Milano. Padova: Sargon srl. Pollock, S. 1991 Of Priestesses, Princes and Poor Relations: The Dead in the Royal Cemetery of Ur. Cambridge Archaeological Journal 1/2: 171–89. 2003 Feasts, Funerals, and Fast Food in Early Mesopotamian States. Pp. 17–38 in The Archaeology and Politics of Food and Feasting in Early States and Empires, ed. T. L. Bray. New York: Kluwer Academic/Plenum Publishers. 2012 Between Feasts and Daily Meals: Towards an Archaeology of Commensal Spaces. Berlin: Exzellenzcluster 264 Topoi. Pope, M. H. 1972 A Divine Banquet at Ugarit. Pp. 170–203 in The Use of the Old Testament in the New and Other Essays. Studies in Honor of William Franklin Stinespring, ed. J. M. Efird. Durham: Duke University Press. Porada, E. 1973 Notes on the Sarcophagus of Ahiram. JANES 5: 355–372. Porter, A. M. 2012 Mortal Mirrors: Creating Kin through Human Sacrifice in Third Millennium Syro-Mesopotamia. Pp. 191–215 in Sacred Killing. The Archaeology of Sacrifice in the Ancient Near East, eds. A. M. Porter and G. M. Schwartz. Winona Lake, IN: Eisenbrauns. Postgate, J. N. 1977 Excavations at Abu Salabikh, 1976. Iraq 39: 269–99. 1984 Excavations at Abu Salabikh, 1983. Iraq 46: 95–113. Powell, M. A. 1994 Metron Ariston: Measure as a Tool for Studying Beer in Ancient Mesopotamia. Pp. 91–127 in Drinking in Ancient Societies. History and Culture of Drinks in the Ancient Near East. Papers of a Symposium held in Rome, May 17th–19th 1990, ed. L. Milano. Padova: Sargon srl. Rathje, W. J. 1977 New Tricks for Old Seals: A Progress Report. Pp. 25–32 in Seals and Sealing in the Ancient Near East, eds. McG. Gibson and R. D. Biggs. Bibliotheca Mesopotamica 6. Malibu: Undena Publications. Reade, J. E. 1995 The Symposion in Ancient Mesopotamia: Archaeological Evidence. Pp. 35–56 in In Vino Veritas, eds. O. Murray and M. Tecusan. London: British School at Rome. 2001 Assyrian King-Lists, the Royal Tombs of Ur, and Indus Origins. JNES 60: 1–29. Rehm, E. 2004 Dynastensarkophage mit szenischen Reliefs aus Byblos und Zypern Teil I.1. Der Ahiram-Sarkophag. Forschungen zur phönizisch-punischen und zyprischen Plastik II.1. Mainz: Philipp von Zabern. Renette, S. 2014 Feasts on Many Occasions: Diversity in Mesopotamian Banquet Scenes during the Early Dynastic Period. Pp. 61–86 in Feasting in the Archaeology and Texts of the Bible and the Ancient Near East, eds. P. Altmann and J. Fu. Winona Lake, IN: Eisenbrauns.

472

Diana L. Stein

Romano, L. 2012 Banqueting in a Temple. Pp. 269–281 in Proceedings of the 7th International Congress on the Archaeology of the Ancient Near East: 12th April – 16th April, the British Museum and UCL, London, eds. R. Matthews and J. Curtis. Wiesbaden: Harrassowitz. 2015 Holding the Cup: Evolution of Symposium and Banquet Scenes in the Early Dynastic Period. Pp. 289–302 in The Long Way to a Historiography of the Early Dynastic Period(s), eds. R. Dittmann and G. Selz. Altertumskinde des Vorderen Orients, Bd. 15) Münster: Ugarit Verlag. Rutz, M. T. 2014 The Archaeology of Mesopotamian Extispicy: Modeling Divination in the Old Babylonian Period. Pp. 97–120 in Archaeologies of Text, eds. M. T. Rutz and M. M. Kersel. Oxford/Philadelphia: Oxbow Books. Sallaberger, W. 1993 Der kultische Kalender der Ur III-Zeit. Berlin: Walter de Gruyter. Schmandt-Besserat, D. 2001 Feasting in the Ancient Near East. Pp. 391–403 in Feasts: Archaeological and Ethnographic Perspectives on Food, Politics, and Power, eds. M. Dietler and B. Hayden. Washington DC: Smithsonian Institution Press. Schmidt, K. 2010 Göbekli Tepe – the Stone Age Sanctuaries. New Results of Ongoing Excavations with a Special Focus on Sculptures and High Reliefs. Documenta Praehistorica 37: 239–256. Selz, G. 2004 Early Dynastic Vessels in ‘Ritual’ Contexts. WZKM 94: 185–223. Selz, G. 1983 Die Banketteszene: Entwicklung eines “überzeitlichen” Bildmotivs in Mesopotamien von der frühdynastischen bis zur Akkad-Zeit, Teil I-II. FAOS 2. Wiesbaden: Franz Steiner Verlag. Sherratt, A. 1991 Sacred and Profane Substances: The Ritual Use of Narcotics in Later Neolithic Europe. Pp. 50–64 in Sacred and Profane: Proceedings of a Conference on Archaeology, Ritual and Religion, eds. P. Garwood, D. Jennings, R. Skeates, and J. Toms. Oxford University Committee for Archaeology Monographs 32. Oxford: Oxford University Committee for Archaeology. 1995 Alcohol and Its Alternatives: Symbol and Substance in Pre-Industrial Cultures. Pp. 11–46 in Consuming Habits: Drugs in History and Anthropology, eds. J. Goodman, P. E. Lovejoy, and A. Sherratt. London/New York: Routledge. Solecki, R. S., Solecki, R. L., and Agelarakis, A. P. 2004 The Proto-Neolithic Cemetery in Shanadar Cave. College Station, TX: Texas A&M University Press. Stein, D. L. 2006 Palaeolithic Iconography on Bronze Age Seals from the Mesopotamian Periphery? Pp. 19–34 in The Iconography of Cylinder Seals, ed. P. Taylor. London: Warburg Institute. 2009 Winged Disks and Sacred Trees at Nuzi: An Altered Perspective on Two Imperial Motifs. Pp. 573–603 in Studies on the Civilization and Culture of Nuzi and the Hurrians 18, eds. D. I. Owen and G. Wilhelm. Bethesda, Maryland: CDL Press. 2017 The Role of Stimulants in Early Near Eastern Society: Insights through Artifacts and Texts. Pp. 507–533 in At the Dawn of History. Ancient Near Eastern Studies in Honour of J. Nicholas Postgate, eds. Y. Heffron, A. B. Stone, and M. J. Worthington. Winona Lake, Indiana: Eisenbrauns.

The Third Millennium Banquet Scene at Ur

473

2019 Architecture and Acoustic Resonance: The ‘Tholoi’ at Arpachiyah Reconsidered in the Context of a Wider Neolithic Horizon. Pp. 125–148 in Representing the Senses in the Ancient Near East: Between Text and Image, eds. A. Hawthorn and A.-C. Rendu-Loisel. University Park, PA: Eisenbrauns. 2020 The Canon of Ancient Near Eastern Glyptic on a Roll: Leaps, Hurdles and Goals. Pp. 172–194 in Testing the Canon of Ancient Near Eastern Art and Archaeology, eds. A. R. Gansell and A. Schafer. Oxford: Oxford University Press. fth. A New Angle on the Contest Scene: Exploring Its Context on 3rd Millennium B.C. Seals and Sealings. In Proceedings of the 11th ICAANE in Munich, 14–17 April 2018, eds. A. Otto, et al. Wiesbaden: Harrassowitz Verlag. Vanstiphout, H. L. J. 1992 The Banquet Scene in the Mesopotamian Debate Poems. Pp. 9–21 in Banquets d’Orient, ed. R. Gyselen. Res Orientales 4. Bures-sur-Yvette: Groupe pour l’étude de la civilisation du Moyen-Orient. Vidale, M. 2011 PG 1237, Royal Cemetery of Ur: Patterns in Death. Cambridge Archaeological Journal 21/3: 427–51. Weber, O. 1920 Altorientalische Siegelbilder I und II. Der Alte Orient 17–18. Leipzig. Wiggermann, F. A. M. 2000 Lamaštu, Daughter of Anu. A Profile. Pp. 217–252 in Birth in Babylonia and the Bible: Its Mediterranean Setting, ed. M. Stol. CM 14. Gron­ingen: Styx. Winter, I. J. 1986 The King and the Cup: Iconography of the Royal Presentation Scene on Ur III Seals. Pp. 253–268 in Insight Through Images. Studies in Honor of Edith Porada, eds. M. Kelly-Buccellati, P. Mattiae, and M. Van Loon. Bibliotheca Mesopotamica 21. Malibu: Undena Publications. 1987 Women in Public: The Disk of Enheduanna, The Beginning of the Office of ENPriestess, and the Weight of Visual Evidence.” Pp. 189–201 in La Femme dans le proche orient antique. Proceedings of the Rencontre Assyriologique Internationale, Paris, July 1986, ed. J. M. Durand. Paris: Editions Recherche sur les Civilisations. 2000 The Eyes Have It: Votive Statuary, Gilgamesh’s Axe, and Cathected Viewing in the Ancient Near East. Pp. 22–44 in Visuality Before and Beyond the Renaissance, ed. R. Nelson. Cambridge: Cambridge University Press. Woolley, C. L. 1934 The Royal Cemetery. UE 2. London: Trustees of the British Museum and of the University Museum, University of Pennsylvania, Philadelphia. 1982 Ur ‘of the Chaldees,’ revised and updated by P. R. S. Moorey. London: The Herbert Press. Wright, H. T. 1969 The Administration of Rural Production in an Early Mesopotamian Town. Anthropological Papers 38. Ann Arbor: University of Michigan Museum of Anthropology. Zajdowski, K. 2013 Transformation of the Mesopotamian Banquet Scene into the Presentation Scene in the Early Dynastic, Akkadian and Ur III Periods. Akkadica 134: 1–16. Zettler, R. 2011 Banqueting and Music: An Early Dynastic I Sealing from Nippur. Pp. 275–285 in Strings and Threads: A Celebration of the Work of Anne Draffkorn Kilmer, eds. W. Heimpel and G. Szabó. Winona Lake, IN: Eisenbauns.

474

Diana L. Stein

Ziffer, I. 2005 From Acemhöyük to Megiddo. The Banquet Scene in the Art of the Levant in the Second Millennium BCE. Journal of the Institute of Tel Aviv University 32/2: 133–167.

New Excavations at Ur Elizabeth C. Stone, Paul Zimansky, Katheryn Twiss, Michael Charles, and Melina Seabrook Stony Brook University and the University of Oxford

Ur is the Mesopotamian site with the best preserved and published architectural remains. There, as elsewhere, public buildings dating to the Ur III period (Woolley 1974) and Isin-Larsa domestic areas (Woolley and Mallowan 1976) have been the most easily accessible for archaeologists and looters alike, resulting in a cuneiform record dominated by Ur III public archives and Isin-Larsa private documents. This discontinuity in the textual records from these two time periods has led some to interpret the Ur III as a very top down society. For example, the Assyriologist Robert Englund (2012) has described much of the Ur III population as “compar­ able to chattel slaves.” This stands in contrast to the prevailing view of Isin-Larsa society where it is thought that the larger population enjoyed considerable agency (e.g. Charpin 1980, 1986; Van De Mieroop 1992). We have argued that rather than reflecting a dramatic transformation of Mesopotamian society, these differences are more likely the result of differential preservation of the sources of public and private documents due to the differential erosion of site surfaces dating to these two time periods. Domestic populations generate considerable amounts of trash, much of which appears to have been discarded in nearby streets and open areas, but this was not the case for public buildings. When most archaeological sites in the south were abandoned in the early to mid-second millennium (Stone 2002: 80–81) they were subject to sheet erosion from dust-laden desert winds. The high density of sherd-rich trash associated with private housing would have generated an erosion-resistant cap which was not present in areas with public buildings. Private Ur III contracts do exist, but were only recovered when the University of Pennsylvania tunneled deep into Nippur, and the administrative documents similar to Ur III archives were associated with those early second millennium public buildings at Ur that resisted erosion thanks to their unusually extensive use of baked brick. These data suggest that the differences in both the textual and architectural records from these two time periods are likely more the result of differential erosional processes between public and private buildings than evidence for a dramatic change in social organization. While the temples, palaces, and administrative buildings that have been the focus of much Mesopotamian excavation are the ubiquitous reflection of the institutions that defined Mesopotamian state society, it is from the investigation of domestic areas that the architecture, burials, private documents, 475

476

Elizabeth C. Stone, Paul Zimansky, Katheryn Twiss, et al.

artifacts, and plant and animal remains speak to the status of the larger population. It is for this reason that the planned three season project at Ur focuses on investigating private housing dating to both the Isin-Larsa and the Ur III periods. Given the extensive excavations carried out by Woolley in early second millennium BCE domestic areas, and his suggestion that he quite often dug down until he reached the Ur III period (Woolley and Mallowan 1976: 14), our research plan comprised excavations beneath his areas to obtain Ur III material and beside them to access a comparable sample of Isin-Larsa domestic remains. This is still our plan, although it is now clear that Woolley rarely penetrated beneath levels dating to the late IsinLarsa/early Old Babylonian period. Over the course of a ten-week field season between October and December 2015, funded by the National Geographic Society and a generous private donor, we excavated in four areas (Fig. 1). Two were aimed at penetrating down to Ur III levels within areas already dug by Woolley: in 1 Baker’s Square and in the Niche Lane area. Two new areas outside his excavations were also investigated with the aim of recovering comparable data from the late Isin-Larsa to early Old Babylonian periods. One was to the southwest of Area AH within the courtyard of Woolley’s Neo-Babylonian House 7, and the other to the north, near where the ancient river cut Ur into two parts. All excavations followed natural stratigraphy, with arbitrary levels used within deep deposits. All were subject to an intensive program of flotation for the recovery of plant and microfaunal remains. Unfortunately, the unavailability of mesh in either Iraq or Greece prevented the sieving of deposits, so the faunal remains recovered from the flotation samples will be used to determine how much Fig. 1. Excavation areas (imagery from Google Earth). was missed from the hand-picked samples from excavation.

Exploring Areas Previously Investigated by Woolley When we dug within Woolley’s old excavation areas, we quickly realized that he had not gone down as far as his publications suggest; indeed, in neither area had he excavated beneath the uppermost floor level. The result was that he recovered only a small percentage of the burials associated with these houses. In 1 Baker’s Square (Fig. 2), we encountered late Isin-Larsa to early Old Babylonian burials

New Excavations at Ur

477

beneath the floors of all four spaces investigated. Woolley had already identified the brick tomb in our central space and sug­ gested that he had looked through its doorway and found it to be empty. We removed its roof and confirmed that it had been almost com­plete­ly cleared. But he had only dug down in front of the doorway, leaving unexcavated the extensive evidence for ritual activities outside the tomb. This included reed matting along its edges Fig. 2. View of 1 Baker’s Square. and a number of pits and bowls with offerings in its vicinity. More burials were found beneath the floors in the two rooms to the west; infant burials in paired bowls and child burials in pots in one, an adult inhumation in the next room. Yet another adult interment was found in the room to the east. Thus, all excavated spaces included burials of one kind or another. Since the latter, Woolley’s Room 4, was the least disturbed by burials, we se­ lect­ed this as the space where we could explore lower levels. We first encountered earlier Isin-Larsa material associated with mud-brick walls which were under and slightly offset from the baked brick of the latest occupation. Beneath this we were in Ur III levels based on the ceramics, but the walls no longer defined the area. Thus, we were able to recover a sample of Ur III ceramics and plant and animal remains, but these lack good contextual data. The objects recovered from the upper levels of this room are typical of Isin-Larsa domestic areas (Figs. 3 and 4). Grave goods mostly consisted of beads, and a steatite bowl more at home in the Gulf shares its decoration with two examples recovered from graves by Woolley (U.15470 and U.16723; Woolley and Mallowan 1976: pl. 100,

Fig. 3. Objects from 1 Baker’s Square.

Fig. 4. Objects from 1 Baker’s Square.

478

Elizabeth C. Stone, Paul Zimansky, Katheryn Twiss, et al.

Stone Vessels 1 and 6), one from the Niche Lane area (Woolley and Mallowan 1976: 31), and the other from near the city wall (Woolley and Mallowan 1976: 234). Other objects include a terracotta Humbaba mask, two cylinder seals and a substantial bronze chisel. Dominque Charpin (see his contribution in this volume), tells us that the four cuneiform tablets which were recovered from this building were all accounts; three dated to the Isin-Larsa/Old Babylonian period; one came from deeper levels and is a gold account dated to Shulgi 43. Our second probe beneath Woolley’s excavations was where Niche Lane took a zig-zag path (Fig. 5). Here we placed a ten by twelve meter excavation area which sampled the lane itself, along with Room 2 of 2 Niche Lane, Room 5 of 1 Niche Lane, and Room 7 of 1 Boundary Fig. 5. View of Niche Lane. Street. The latter had a number of inhumations beneath the floor and perhaps served as the “chapel” for that building. As was the case for Building 1, these upper levels yielded both ceramics and objects which are largely comparable to those found by Woolley, and indeed in late Isin-Larsa levels at other sites. Stone beads and bone rings, together Fig. 6. Sealings from Niche Lane. with cylinder seals, were used for decor­ ation. Terracottas included part of a Humbaba mask, figurines, a model bed, and model chariot wheels. Objects of copper-bronze were mostly decorative items from the graves. In addition to cuneiform tablets, two tablets with only sealings were also recovered (Fig. 6). We excavated three spaces to a depth of some three meters, penetrating through Ur III levels and into Akkadian remains within Niche Lane itself, in the process observing the complex patterns of rebuilding and reworking evident in the architecture. As was the case in Square 1, we found that the defining baked-brick architecture was built above an earlier mud-brick phase, although these walls did not continue down on all sides to the base of our excavations. When we reached Ur III levels this was made obvious by the presence of a drain partly beneath the higher wall. Nearby we recovered perhaps our most exotic item, a piece of black wood (Fig. 7), most likely ebony given how unusual it is for the remains of uncharred wood to be preserved in a Mesopotamian site. Since ebony is native to both Africa and India it is tempting to connect this with Ur III texts recording the importation of black wood from Meluhha, despite its absence from the lexical lists of the time (Moorey 1999: 352).

New Excavations at Ur

479

As we penetrated deeper below the top of the drain we found a small archive of Akkadian tablets, miraculously during the visit by our epigrapher Dominque Charpin. These include accounts dealing with cattle, fish, gold, silver, and the measuring of fields (see Charpin, this vol­ume). They significantly increase what we know about the Akkadian presence at Ur. Other early objects include weights (including one set of three) and objects of stone and copper-bronze (Fig 7). These two excavations have added to our understanding of Woolley’s work at Ur, demon­ strating his high level of competence in identifying floors and architecture while also indicating that in most instances his work stopped at the highest floor, with the result that he misFig. 7. Objects from Niche Lane. sed the majority of the underfloor burial deposits. Indeed, his so-called chapels were by no means the only loci of burial. In Area 1, burials were found beneath the floors of all three of the excavated rooms that surrounded the “chapel,” and in Area 2 in all rooms other than the street itself.

Excavating New Areas We also investigated two areas that had not been sampled by Woolley. Area 3 was located immediately to the south of the channel which separated Area AH from the main mound (Fig. 1). In the uppermost layers we encountered a bewildering variety of late graves ranging from brick tombs, bathtub and keyhole coffins, simple inhumations, as well as the odd Kassite pit (Fig. 8). These were dug into an area which has close parallels with the architectural patterns recorded by Woolley (Figs. 9 and 10). In the north and west the architecture was preserved extraordinarily well. The baked brick walls with their mud-brick caps rose to a height greater than 1.5 meters, and were accompanied by solid brick staircases. It remains unclear whether the stairs simply provided access to the roof or to an upper storey (as argued by Woolley), but if it were only to the roof, the high preservation of the walls surely indicate that the roof itself must also have been walled, a pattern which is common in modern Iraq, including at our Mashkan-shapir dig house located in the town of Shomeli. These provide a private but cooler place to sleep in the heat the Iraqi summer. The architecture which we uncovered also differs from all of the houses excavated by Woolley in Fig. 8. Late graves in Area 3.

480

Elizabeth C. Stone, Paul Zimansky, Katheryn Twiss, et al.

Fig. 9. Plan of Area 3 architecture.

Fig. 10. View of Area 3.

that it has not one but two stairways leading to a single roof or upper storey (Fig. 9). It is possible that the smaller, spiral staircase in the south was superseded at a later time by the more direct stairway in the north. The latter was certainly better preserved, and had an intact supporting arch beneath it. However, the more ruinous character of the smaller stairway was largely due to its disturbance by later burials. The room whose roof was accessed by the two staircases was exceptionally well preserved. In addition to the pavement at the base, we identified three clay floors, one at floor level, one fifty cm above the tile pavement and the other in between. The top and bottom floors were accompanied by traces of reed doors giving access to the courtyard; the middle floor preserved a door which would once have closed a back doorway which was later bricked up (Fig. 9). The upper floor exhibited signs of burning, as indeed did all areas of this square at this level. Since this represents the last occupation (other than the burials) in this area, we would agree with Woolley and Mallowan that this almost certainly represents the destruction of Ur by Samsuiluna of Babylon (1976: 13). This room opened into a largish paved courtyard which, like the internal space, was overlaid by later clay floors. The last of these was covered with date palm wood, heavily burned presumably due to Samsuiluna’s conquest of Ur. The underlying baked-brick pavement shows no discontinuity between inside and outside areas, suggesting that the entire complex was built as a single unit. This courtyard gave access to the second stairway and perhaps to three additional rooms. Here some of the walls have been robbed out and are only visible as a break in the paving of the floor, making it difficult to be certain that we know where the doors were. Two of these rooms have troughs within them. When we encountered them, we expected them to be graves, but found them to be empty on excavation. The robbing of the baked bricks makes it somewhat difficult to determine how many houses are represented here. The two spaces with drains suggest at least two since almost all the AH houses have only one courtyard. Moreover, the brickwork at the base of the stairway in the south makes it highly unlikely that there was a door in the now robbed out part of its western wall. The objects recovered from the upper levels of this square reflect the date of the later pits, burials and the like. More typical Isin-Larsa terracottas and this

New Excavations at Ur

481

unusual stele fragment were recovered from the main occupation layers along with shell and copper/bronze jewelry (Figs. 11–12). A sharpening stone and the one cylinder seal found here reflect the late Isin-Larsa date of this building, although the two cuneiform tablets are Ur III in date. This is a very encouraging area for our broad­er aim to investigate domestic areas dating to both the Isin-Larsa and the Ur III period, since the extensive robbing of the walls provides us with broad areas where we can penetrate beneath the floors and expect to find contiguous areas of Ur III occupation.

Fig. 11. Objects from Area 3.

Fig. 12. Objects from Area 3.

Preliminary results from the analysis by Katheryn Twiss and Melina Seabrook of the macrofaunal remains recovered from the deposits associated with the occupation of the building indicate little difference between the Ur III and Isin-Larsa samples and exhibit a predominance of sheep and sheep-sized animals, a fair amount of cattle and some pig. These data are decidedly preliminary and have yet to include the fish and bird bones recovered through flotation. By contrast, the plant remains analyzed by Michael Charles indicate significant differences between the two time periods. The dominance of barley in the Ur III samples is clear, and contrasts with the plant remains from Isin-Larsa levels, which were more varied and generally had lower percentages of barley. An exception is the one sample from Area 3 which more closely resembles the Ur III pattern seen in samples from Area 2. It must be stressed here that we are looking here at a tiny sample of the plant and animal remains collected, but these do suggest that when the analyses are completed, these will provide important data on the resources available to the residential population at Ur. We also did some research into the large expanse of the site to the west and south of Area AH. We had hoped that we could use magnetic gradiometry to recover the patterns of sub-surface architectural traces of early second millennium baked-brick walls since in many instances the Neo-Babylonian houses excavated by Woolley were so badly eroded that he was unable to identify doorways. Moreover, since baked brick foundations were not common at that time, they were unlikely to show up clearly in the imagery. Indeed, the last surmise proved correct; although our data did record the orientation of the Neo-Babylonian walls, we did not recover clear architectural patterns, but nor did we recover data on the Old Babylonian

482

Elizabeth C. Stone, Paul Zimansky, Katheryn Twiss, et al.

occupation. The reason was made clear by our excavations in Area 4, a 5 x 10 meter trench located within the courtyard of Woolley’s Neo-Babylonian House 7, some twenty meters to the west of Area AH. Here instead of encountering early second millennium architecture near the surface, we had to dig down oneand-a-half meters through numerous later pits before we found our first significant architecture which, like that found in our other excavations, dates Fig. 13. Architecture in Area 4. to the early second millennium BCE (Fig. 13). Architectural traces were only recovered in the eastern end of this 5 x 10 excavation and it is likely that the western area served as a courtyard. The remains from this excavation differ in important ways from those found elsewhere. Whereas the walls in Squares 1, 2, and 3 were generally around eighty cm in width and had baked-brick walls preserved for up to one-and-a-half me- Fig. 14. Late objects from Area 4. ters, those in Square 4 were only fifty cm wide, and the uppermost level had no more than three courses of baked brick before they were carried up in mud brick, although a partial lower wall had eleven courses of baked brick. The finds from Square 4 were also impoverished. The upper levels had no architecture but yielded a number of horse and rider figurines, crude grinding stones and a tiny jar presum­ably for perfume, all dating to the first mil­ lennium BCE (Fig. 14). No objects were associated with the early second millennium levels. Our excavations into Isin-Larsa/Old Babylonian levels were too limited to provide a representative sample of plant and animal remains, so we cannot say whether the impoverishment in both architecture and objects when compared with the other excavated areas was accompanied by differences in diet. There can be varying interpretations of these data: The top of the architecture in Square 4 was below the base of the walls of Woolley’s AH excavations located only twenty meters away which suggests that the AH area was built on higher ground. It is therefore possible that the slope between them might have reflected a boundary between neighborhoods separating the lower Area 4 from AH. Not necessarily incompatible with this interpretation would be the possibility that the greater overburden of Kassite and Neo-Babylonian remains in Area 4 preserved slightly later architecture which was largely eroded away in AH, namely the rebuilding of Ur following Samsuiluna’s conquest. We know that this must have happened since there are rare texts dated to the couple of years following this attack. The latter interpretation for Area 4 might be strengthened by the line of burning that divides the well-constructed earlier levels which we have scarcely sampled from the more scrappy remains higher up.

New Excavations at Ur

483

Conclusions Interpretations of the data recovered from our 2015 excavations include the possibility that different neighborhoods, here marked by changes in elevation, also reflected economic differences between their residents; perhaps even class differences, even though such a pattern has not been observed between other excavated neighborhoods at either Ur or Nippur. Since we are not informed of how Woolley chose his areas for excavation, it may well be that they were selected because of the wealth of their surface finds. Another issue to be considered are differences in site formation processes. Area 4 has the only excavated early second millennium architecture at Ur dug by us to be buried beneath extensive Kassite and Neo-Babylonian levels. The Ur excavations did produce a few tablets dated to a few years following the destruction of much of the city by Samsuiluna of Babylon, an indication that not all of the city was abandoned immediately. It is therefore possible either that while the uppermost early second millennium levels were removed by erosion elsewhere— whether in Area AH or in Kassim Rawi’s unpublished excavations at the far west of the South Mound—they were preserved in Square 4, or that Square 4 was an area where settlement continued following the conquest of the city. This season’s work has provided data on Ur’s occupation from Akkadian through earliest Old Babylonian levels. Next season we will continue our work especially in Areas 3 and 4. We will also be joined by Adelheid Otto, who will explore new areas, and by Emily Hammer, who will conduct a survey of the site.

Bibliography Charpin, D. 1980 Archives familiales et propriéte privée en Babylonie ancienne: étude des documents de “Tell Sifr.” Geneva: Librairie Droz. 1986 Le clergé d’Ur au siècle d’Hammurabi (XIXe-XVIIIe siècles av. J.-C.). HEO 22. Geneva: Librairie Droz. Englund, R. K. 2012 “Equivalency Values and the Command Economy of the Ur III Period in Mesopotamia,” Pp. 427–458 in The Construction of Value in the Ancient World, eds. J. Papadopoulos and G. Urton. Los Angeles: Cotsen Institute of Archaeology. McCown, D. E. and Haines, R. C. 1967 Nippur I. Temple of Enlil, Scribal Quarter, and Soundings. OIP 78. Chicago: University of Chicago Press. Moorey, P. R. S. 1999 Ancient Mesopotamian Materials and Industries: The Archaeological Evidence. Winona Lake, IN: Eisenbrauns. Stone, Elizabeth C. 2002 “The Ur III-Old Babylonian Transition: An Archaeological Perspective.” Iraq 64: 79–84. Van De Mieroop, M. 1992 Society and Enterprise in Old Babylonian Ur. BBVO 12. Berlin: Dietrich Reimer Verlag. Woolley, C. L. 1974 The Buildings of the Third Dynasty. UE 6. London: Trustees of the British Museum and of the University Museum, University of Pennsylvania, Philadelphia.

484

Elizabeth C. Stone, Paul Zimansky, Katheryn Twiss, et al.

Woolley, C. L. and Mallowan, M. 1976 The Old Babylonian Period. UE 7. London: Trustees of the British Museum and of the University Museum, University of Pennsylvania, Philadelphia; British Museum Publications.

Ur and Other Cities in Some Sumerian and Akkadian Personal Names: The Pre-Sargonic and Sargonic Periods Jan Tavernier

Université catholique de Louvain

1. Introduction The study of Ancient Near Eastern onomastics is an important as well as a fascinating field of research, mainly because of the information yielded by personal names (anthroponyms) and place names (toponyms) on the history and culture of the Ancient Near East. Anthroponyms are one of the few sources for the study of religion on the individual level, while toponyms may give us information on the settlements of various population groups. 1 This paper aims at giving more attention to the relation between place names (toponyms) and personal names (anthroponyms). This relation can be assessed in two ways: anthroponyms in toponyms and toponyms in anthroponyms. The first situation is well known. Many toponyms in the Ancient Near East (and elsewhere) are based on a personal name. Perhaps the most famous type of such a structure is the toponyms composed of bīt “house,” followed by a personal name (Bīt-PN). This form of names, attested from the Kassite period onwards, occurs massively in first millennium Babylonia. In Achaemenid Iran too, many place names are constructed upon or even identical with a personal name (Tavernier 2006). The second situation, toponyms in anthroponyms, has received less scholarly attention. Stamm in his Die Akkadische Namengebung (1939) does mention the practice, but does not go deeper into it. Limet in his L’Anthroponymie sumérienne (1968) gives more attention to it, but still one could have expected much more. It Author’s note: This research has been funded by the Interuniversity Attraction Poles Programme initiated by the Belgian Science Policy Office (IAP VII/14: “Greater Mesopotamia: Reconstruction of its Environment and History”). I owe my sincere thanks to Thomas Balke, Anne Goddeeris, and Bram Jagersma for their valuable comments. Abbreviations are cited according to the systems used in the CDLI website (http://cdli.ox.ac.uk/ wiki/doku.php?id=abbreviations_for_assyriology). Another abbreviation is CUNES (Department of Near Eastern Studies, Cornell University). 1. A nice example are the Anatolian proper names. In Anatolia, proper names with an Anatolian origin are still abundantly attested in the Greco-Roman period and nicely corroborate the continuation of Anatolian culture under Greek and Roman domination.

485

486

Jan Tavernier

is therefore the intention of this paper to make up a typology of personal names containing a place name and this for the names of the Pre-Sargonic (c. 2800–2300 BCE) and the Sargonic periods (c. 2300–2100 BCE). This type of names is labelled here “toponymical anthroponyms.” Not only in Mesopotamia, but also in other Ancient Near Eastern civilisations personal names may be formed based on toponyms. It must immediately be noted, however, that this practice is relatively rare. The exception is Anatolia, where various names are constructed with toponyms, e.g. Alashaili, Halpaziti, Halpaili, Halpassi, Halpashulupi, Hattusaziti, Hattusili, Nerikkaili, Ninuwaziti, etc. (Laroche 1952: 71–76; id. 1966: 247–279; Kinlaw 1967: 351–352, 357; Répertoire onomastique by Marie-Claude Trémouille 2) and in Iranian names the element Arya-“Iranian” occurs regularly, e.g. in Old Persian Aryāramna- “Who creates peace for the Iranians” (cf. Tavernier 2007: 14 no. 1.2.6) or in the reconstructed name *Aryabāma- “Having the lustre of the Iranians” (cf. Tavernier 2007: 115 no. 4.2.122), but apart from this element no geographical indications occur in Old Iranian names. Also in WestSemitic names the pattern is rare and all names are actually gentilic names (e.g. ʾKy “Akkoite,” Bwzy “Buzite,” Kšy “Cushite,” Mṣry “Egyptian,” Prsy “Persian,” and Ṣry “Tyrian”; cf. Noth 1928: 232 and Benz 1972: 238–239, 335–338). In Hurrian such anthroponyms also occur, but again not very frequently (cf. Gelb, Purves and MacRae 1943: 205 [Arrapḫa] and 234 [Matka]; Kinlaw 1967: 160, 168, 170).

2. The Sumerian Names of the Pre-Sargonic Period In Mesopotamia, personal names containing place names already appear in the earliest periods. The oldest attestations come from Ur and occur in texts dating to about 2,800 BCE. Examples are Lú-Ud.nunki “The man from Adab” (UET 2 93 rev. i 9) and Urim-pa.a (UET 2 99 i 4), whose meaning is not really clear. From then on, the number of this type of names will steadily grow, but they will never be massively present in the Mesopotamian onomastic material.

2.1. Ethnonyms The two least complex forms of anthroponyms containing toponyms are the hypo­coristics (cf. infra) and the ethnonyms, names that do not indicate more than the city or the region from which the name-bearer originates. It must be noted that it is not always clear whether these names are really personal names or simply ethnonyms. When in a text the spelling Lú Adabki occurs, this may very well be a personal name meaning “The man from Adab,” but it can as well be an ethnonym referring to a person with another, specific personal name, which does not necessarily have to be mentioned in the text (e.g. Ur-dUtu Lú Adabki “Ur-Utu, the man from Adab”). Here the context appears to be very important. A phrase ĝìr Lú-Ma-riki clearly indicates that a personal name is intended and must accordingly be translated “Under the responsibility of Lu-Mari” and not “Under the responsibility of a man from Mari.” Limet (1968: 37–38, 66) distinguishes three types of ethnonyms and their origin: 1.  The scribe of a city uses the expression to denote and identify the individual as a stranger. This ethnonym then became his personal name, but was prob-

2.  To be consulted at https://www.hethport.uni-wuerzburg.de/hetonom/ONOMASTIdata.html.

Ur and Other Cities

487

ably not his original name. To this category belong the names with a distant land or city (e.g. Lú-Urua and Lú-Uruaz, both being towns in Elam). 2.  The name was attributed to a new immigrant by the inhabitants of the city he arrived in. Here too distant lands are involved. In my view, however, this category largely coincides with the previous one, so both could be taken together. 3.  The parents of a person wanted to express their allegiance to and respect for the city they came from and thus chose the name themselves. To this category belong the names with Mesopotamian cities, e.g. Lú-Unugki “The man from Uruk.”

In the Pre-Sargonic period, the only type of Sumerian ethnonym is Lú-GN. The feminine counterpart Munus-GN “The woman from GN” is not found in this period. Examples are: 1.  Lú-Áb-za-niki “The man from Abzanu”: ECTJ 35 ii 8 (from Nippur). 2.  Lú-A.ḫaki “The man from Kuara”: ECTJ 38 iii 3 (from Nippur). 3.  Lú-Eréški “The man from Eresh”: ECTJ 40 iii 1 (from Nippur). 4.  Lú-Giški.ùḫ “The man from Umma”: FTP 78 ii 3 (from Shuruppak). 5.  Lú-Inki “The man from Isin”: IM 42084 3 ii′ 2′ (unknown provenance). 6.  Lú-Kin-nir “The man from Kinnir”: FTP 34a obv. 2′ (from Shuruppak). 7.  Lú-Kiški “The man from Kish”: ECTJ 38 ii 8 (from Nippur). 8.  Lú-Már-daki “The man from Marad”: ECTJ 35 ii 3; ECTJ 67 iii 9; OSP 1 38 iii 1. All attestations come from Nippur. 9.  Lú-Su.kur.ruki “The man from Shuruppak”: ECTJ 8 iv 1; ECTJ 35 ii 7; ECTJ 67 vii 11. All attestations come from Nippur. 10.  Lú-Ud.nunki “The man from Adab”: FTP 78 ii 1 (from Shuruppak); UET 2 93 rev. i 9 (from Ur). 11.  Lú-Ud.unuki “The man from Larsa”: ECTJ 32 iii 1 (from Nippur). 12.  Lú-Unugki “The man from Uruk”: ECTJ 94: 4 (from Nippur). 13.  Lú-Uru-a “The man from Urua”: TSŠ 131 rev. i 2 (from Shuruppak). 14.  Lú-Uru-azki “The man from Uruaz 4”: ITT 1 1195 rev. 3 (from Girsu/Lagash). 15.  Lú-Uru-saĝ-rig7ki “The man from Urusagrig”: ECTJ 40 ii 2 (from Nippur); FTP 78 i 1 (from Shuruppak).

2.2. Genitival Names The second type is genitival names, frequently attested in Sumerian and Akkadian onomastics. It must in any case be noted that the names cited here may all be hypocoristics of a longer name, the verb being omitted. Amar-Isin may thus be a hypocoristic from a name like “The calf of Isin has done this or that” or “The calf has come to Isin.” 1.  Amar-GN “Calf of GN”: because of the many occurrences of the name type Amar-DN (cf. Di Vito 1993: 23, 25 5), it may be suggested that the toponyms represent the divinities of the cities mentioned. 3.  Published by Edzard (1959: 23 and pl. 2), who, however, believes in an Akkadian reading Awīlumkīn, as a Sumerian reading makes no sense in his eyes. 4.  Uruaz was a city situated in Elam, possibly not far from Urua (Edzard, Farber, and Sollberger 1977: 181). 5.  This name type is less frequently attested in the Sargonic period. Some examples may be found in Di Vito 1993: 45.

488

Jan Tavernier A.  Amar-Ḫa.a(ki) “Calf of Ku’ara”: DP 31 i 3, ii 9; DP 130 ii 15; DP 135 ix 3; DP 594 ii 2; FTP 96 v 4; HSS 3 9 vi 9; HSS 3 10 vi 8; HSS 3 11 vi 2; MVN 3 8 vi′ 2′; Nik. 1 13 iv 11; TSA 7 iii 3. All texts, except FTP 96 (from Shuruppak), come from Girsu/Lagash. B.  Amar-In “Calf of Isin”: ECTJ 54 ii 1 (from Nippur). C.  Amar-Kišxgána-tenûki “Calf of Girid”: CT 50 33: 133; CT 50 34 x 4′; CT 50 37: 30; CT 50 38 iv 5; CT 50 39 iv 12; DP 112 ii 9; DP 116 xi 8; DP 117–118 xi 1; DP 119 iv 2, viii 5; DP 121 viii 7, ix 6, x 12; DP 130 v 12, vi 8, viii 7; DP 132 ii 4; DP 133 ii 8; DP 137 v 11; DP 145 ii 11; DP 146 i 4; DP 147 i 3; DP 149 v 2; FAOS 15/1 545 ii 2, iii 4; Nik. 1 5 iii 1, 5; Nik. 1 149 i 5; Or. 9–13 110 (VAT 4641 v), 148 (VAT 4660 v), 230 (VAT 4664 iv); Or. 21 62 (VAT 4813 ii), 65 (VAT 4900 i); RTC 18 vii 10; RTC 48 i 2; RTC 66 iv 8; VAS 14 99 ii 2. All texts come from Girsu/Lagash. D.  Amar-Urum9ki “Calf of Urum”: CUSAS 23 2 obv. ii 3 (from the Umma region). 2.  Me-GN “The divine forces of GN” A.  Me-Kèški “The divine forces of Kesh”: CUSAS 33 120 obv. iii 8–9 (from Umma). 3.  Šeš-GN “The brother of/in GN” A.  Šeš-An-eden-na “The brother of/in An’eden”: DP 229 iv 11; HSS 3 20 ii 15; HSS 3 21 iv 9; HSS 3 22 iv 13; HSS 3 23 v 13; Nik. 1 1 iii 24; Nik. 1 8 iv 15; TSA 11 iii 24. All attestations come from Girsu/Lagash. 4.  Ur-GN “Hound/man(?) of GN” (cf. Di Vito 1993: 116–117): because of the many occurrences of the name type Ur-DN (cf. Di Vito 1993: 23–25, 30–31), it may be suggested that the toponyms of this type represent the divinities of the cities mentioned. A.  Ur-An-ta-sur-ra “Hound/man(?) of Antasurra”: CTNMC 1 ii 5; Nik. 1 108 iv 2; Nik. 1 140 iii 6; Nik. 1 141 ii 6. All attestations come from Girsu/Lagash. B.  Ur-Gú-eden-na “Hound/man(?) of Gu’eden”: DP 138 iii 10; Or. 26 37 (VAT 4681: 11). Both texts come from Girsu/Lagash. C.  Ur-Inki “Hound/man(?) of Isin”: ECTJ 24 ii 2; OSP 1 95: 12. Both texts come from Nippur. D.  Ur-Kèš(ki) “Hound/man(?) of Kesh”: CUSAS 26 68 vii 2 (from Adab); MC 4 10: 3 (unknown provenance). E.  Ur-Kin-nir 6 “Hound/man(?) of Kinunir”: FTP 109 obv. i 9; SRU 6 vi 2; TSŠ 570 vi 3 (written Ur-Nir-kin); YOS 1 1: 1. All texts come from Shuruppak. F.  Ur-Ki-sal4-la “Hound/man(?) of Kisalla”: Nik. 1 84 ii 5 (from Girsu/ Lagash). G.  Ur-Kiš “Hound/man(?) of Kish”: DP 468 i 2 (from Girsu/Lagash). H.  Ur-Pa5-sír-ra “Hound/man(?) of Pasirra”: Nik. 1 164 iii 2; Nik. 1 180 i 2. Both texts come from Girsu/Lagash. I.  Ur-Su.kur.ruki “Hound/man(?) of Shuruppak”: UET 2 Suppl. 2 iii 4 (from Ur). J.  Ur-Ti-ra-áš “Hound/man(?) of Tirash”: Or. NS 42 236 iv 37 (from Girsu/ Lagash). 6.  Cf. Lambert (1971) on this name.

Ur and Other Cities

489

5.  Ur-saĝ-GN “The hero of GN” A.  Ur-saĝ-A.kib.nunki “The hero of Akibnun”: ECTJ 56 i 4 (from Nippur). B.  Ur-saĝ-Kèš “The hero of Kesh”: CTMMA 1 2: 3 (from Nippur); CUSAS 26 69 i 6 (from Adab). C.  Ur-saĝ-Uruxgána-tenû “The hero of Urub” 7: DP 591 iv 8 (from Girsu/ Lagash). 6.  Finally, a double genitive occurs in two anthroponyms: A.  Nin-é-Unug-ga “The Lady of the temple of Uruk”: DP 112 i 18 (from Girsu/Lagash). B.  Nin-é-Unug-ga-nir-ĝal “The authoritative Lady of the temple of Uruk”: BIN 8 344: 10; CT 50 34: 22; CTNMC 4 ii 4; DP 229 ii 4; HSS 3 21 ii 1; HSS 3 22 i 10; HSS 3 23 viii 10′, 15′; HSS 3 24 vii 5; Nik. 1 1 i 18; Nik. 1 15 ii 1; Or. 43/44 69 (VAT 4612 vii, viii); TSA 11 1: 19, 13, rev. i 2. All texts come from Girsu/Lagash.

2.3. Predicative Names The third type is names with a predicative relation. Here the place name is something and is situated in initial position. The various attested examples are: 1.  GN-ama “GN is the mother” A.  Ab.mùš-ama “Zabalam is the mother”: BIN 8 116: 17 (from Adab). 2.  GN-a-nun “GN has/is princely offspring” A.  Kèš-a-nun “Kesh has/is princely offspring” 8: CUSAS 26 68 ix 3 (from Adab). 3.  GN-du-du “GN is the shelter (= place of entering)” A.  Kiški-du-du “Kish is the shelter” 9: TSŠ 936 ii 2 (from Shuruppak). 4. GN-ki-du10 “GN is a good place” A.  Ab.mùš-ki-du10 “Zabalam is a good place”: BIN 8 237: 19 (from Girsu/ Lagash); CUSAS 33 157 obv. 2 (from Umma); CUSAS 33 195 obv. v 1 (from Umma); TCBI 1 11 rev. i 2, ii 1 (from Adab). B.  Ğír-su-ki-du10 “Girsu is a good place”: DP 134 ii 14; DP 135 ix 6; DP 326 iii 1; Nik. 1 53 ii 10; Nik. 1 125 i 5; HSS 3 9 vi 10; HSS 3 10 vi 9; HSS 3 11 vi 3; MVN 3 8 vi′ 3′; RTC 61 iii 3; TSA 7 iii 7; TSA 47 iv 4. Not surprisingly, all these texts come from Girsu/Lagash. C.  Giš-kúšuxkaskal-ki-du10 “Giš-kúšuxkaskal is a good place”: CUSAS 33 285 obv. 1–2 (from Adab). D.  Ğišša-ki-du10 “Gišša is a good place”: CUSAS 23 32 obv. ii 5; CUSAS 33 56 obv. i 10; CUSAS 33 224 obv. ii 6; RA 34 178: 2; TSA 10 ix 5. All texts, except TSA 10 (from Girsu/Lagash), come from Umma. 10

7.  See Selz (2013: 319) for the reading of this toponym. 8.  Th. Balke, pers. comm. 17/10/2017. 9.  Th. Balke, pers. comm. 17/10/2017. 10.  Giššakidu was an Early Dynastic king of Umma. In all likelihood, he was a contemporary of Enanatum II and Enemtarzid of Lagash (Sollberger 1969; Bauer 1998: 493).

490

Jan Tavernier E.  Gù-ab-ba-ki-du10 “Gu’aba is a good place”: BIN 8 370: 7; DP 135 iv 157; DP 191 v 3; DP 590 ix 7; Nik. 1 19 rev. ii 5. All texts come from Girsu/ Lagash. F.  Ḫa.a-ki-du10 “Ku’ara is a good place”: DP 195 ix 5′; Nik. 1 307 iii 6; Or. 16 28 (VAT 4450 ii). All three texts come from Girsu/Lagash. G.  In-ki-du10 “Isin is a good place”: CUSAS 20 349 obv. ii 3 (from Adab); NTSŠ 152 ii′ 7 (from Shuruppak); TSŠ 230 iii 9′ (from Shuruppak). H.  Kèš-ki-du10 “Kesh is a good place”: CUSAS 33 256 obv. iii 1 (from Umma). I.  Ki-an-ki-du10 “Kian is a good place”: CUSAS 26 90: 9 (from Adab). J.  Ki-nu-nir-ki-du10 “Kinunir is a good place”: RA 64 15: 4 (from Luristan). K.  Nina-ki-du10 “Nina is a good place”: DP 135 xi 7; DP 137 ii 3; DP 216 ii 4; TSA 57 ii 8. All texts come from Girsu/Lagash. L.  Saḫar-ki-du10 “Saḫar is a good place”: Nik. 1 160 ii 6 (from Girsu/ Lagash). M.  Urim-ki-du10 “Ur is a good place”: UET 2 310 iii 4′; Suppl. 14 rev. i 5. Both texts come from Ur. N.  Uruxgána-tenû-ki-du10 “Urub is a good place”: Nik. 1 3 rev. iii 11′; Nik. 1 230 iv 1; TSA 7 xi 12. All attestations come from Girsu/Lagash. 5. GN-zi-ĝu10 “GN is my life” A.  Urim-zi-ĝu10 “Ur is my life”: UET 2 310 rev. i′ 3′ (from Ur). 6.  Type 2 and 3 are one time combined, in the name En-Kinunir-ki-du10 “The Lord of Kinunir, the good place” (DP 129 v 5; RTC 17 vii 7; both attestations come from Girsu/Lagash).

2.4. Complex Names The fourth type of toponymical anthroponyms consists of more complex names. These names consist of at least three elements and are the least frequently attested category. The Pre-Sargonic examples are: 1.  Amar-Ḫa.a-du “The calf of Ku’ara goes”: BIN 8 26: 15; BIN 8 27: 6; BIN 8 37: 36 (from Adab). 2.  Amar-Nina.tag-si “The calf of Sirara fills”: UET 2 338 ii 3′ (from Ur). 3.  Kèš-pa-è “Kesh makes appearance”: CUSAS 26 69 vi 2 (from Adab). 4.  Lugal-Nunki-še “The king (has come) to Eridu”: DP 135 i 8 (from Girsu/ Lagash). 5.  Me-Kul-ab4ki-ta “From the divine forces of Kulaba”: DP 54 xi 4; VAS 14 74 x 8. Both texts come from Girsu/Lagash. 6.  Me-Ud.má.nina.tag-du “The divine forces of Sirara go”: DP 57 vi 7; DP 223 vii 8′, ix 4′. Both texts come from Girsu/Lagash. 7.  Me-Ud.má.nina.tag-ta “From the divine forces of Sirara”: Or. 2 41 (VAT 4875 iii 4) (from Girsu/Lagash). 8.  Nin-Uruxa-i3-du10 “The lady of Urua is good” or “The lady, who made it good for Urua” 11: FAOS 15/1 19 viii 10 (from Girsu/Lagash). 9.  Sipa-Šir.bur.laki-ki-aĝ “The shepherd who loves Lagash”: DP 113 x 14; DP 114 xi 4; HSS 3 15 vii 11; HSS 3 16 viii 1; HSS 3 17 rev. ii 12; MVN 3 2 x 3; Nik. 1 2 rev. i 10′; Nik. 1 9 viii 4; Or. 34/35 46 (VAT 4415 x); Or. 34/35 50 (VAT 4628 ix). All attestations come from Girsu/Lagash. 11.  B. Jagersma, pers. comm. 14/10/2017; Th. Balke, pers. comm. 17/10/2017.

Ur and Other Cities

491

10.  Sipa-Šir.bur.laki-e-ki-aĝ “The shepherd who loves Lagash”: CT 50 36 xi 22; DP 115 xi 8; HSS 3 18 ix 8′; TSA 14 rev. i 10; TSA 15 rev. ii 11. All attestations come from Girsu/Lagash. 11.  Šir.bur.laki-ĝìri-na “Who sets Lagash under his feet”: DP 47 v 2 (from Girsu/ Lagash). Probably a shortened form of *šir.bur.laki-ĝìri-na-sè/gub (Th. Balke, pers. comm. 17/10/2017). 12.  Ud.má.nin[a].tag-pa-è “Sirara makes appearance”: SRU 117 ii 6 (from Girsu/ Lagash).

2.5. Hypocoristics Within the group of hypocoristic toponymic anthroponyms, rather frequently attested in the textual material, two types may be distinguished: 1.  Names identical to the toponym: A.  En.lílki “Nippur”: DP 33 ii 3′ (from Girsu/Lagash). B.  Ti-ra-áš “Tirash”: RTC 7 i 4; RTC 8 i 4. Both texts come from Girsu/ Lagash. C.  Ud.nunki “Adab”: MVN 3 67 ii 2 (from Nippur). D.  Uruxa “Urua”: TSŠ 40 iii 4′; TSŠ 453a iii′ 1′. Both attestations come from Shuruppak. 2.  Names composed of the toponym and a prefix or suffix: A.  Ḫé-Kixgána-tenûki “May Girid”: AnOr 2 41: 16; DP 57 vii 13; VAS 14 172 iii 6. All texts come from Girsu/Lagash. B.  Kèš-ta “From Kesh”: BIN 8 347: 24; DP 140 ii 4; DP 482 v 1; VAS 14 72 vii 6; VAS 14 79 i 6. This name, thus far only attested in texts from Girsu/Lagash, may be the abbreviation of a name such as “(The king has departed) from Kesh.” C.  Ti-ra-áš-šè “To Tirash”: DP 163 iv 10 (from Girsu/Lagash). This name may be the abbreviation of a name such as “(The king has come) to Tirash.” D.  Ud.nunki-bé “By Adab”: CUSAS 11 63 rev.1; CUSAS 26 67 i 3; CUSAS 26 74 ii 3, 8, iii 2, 7, v 4, 7; CUSAS 26 76 iv 6; TCBI 1 14 obv. ii 3, 8, iii 2, 7, rev. ii 4. All texts come from Adab. E.  Ud.nunki-ke4 “To Adab”: MVN 3 105 iii 2 (unknown provenance). 3.  Names composed of a noun and a toponym, but most likely not a genitival name: all names belong to a type Lugal-GN “The king - GN” and lack any verbal form. This type could accordingly be translated “The king (has come to) GN.” A.  Lugal-En.lilki “King – Nippur”: BIN 8 166: 21 (from Nippur). B.  Lugal-Kèš “King – Kesh”: DP 116 x 3; DP 117 v 17, ix 6; DP 118 vi 3, 8, ix 10; HSS 3 27 vi 2; Nik. 1 16 v 1, 11, rev. iii 3; Nik. 1 20 vi 3′; Or. 16 21 (VAT 4861 v); Or. 16 32 (VAT 4726 ii); Or. 16 39 (VAT 4856 v,vii); Or. 16 44 (VAT 4703 ii); VAS 14 41 ii 1. All texts come from Girsu/Lagash. C.  Lugal-Kiš “King – Kish”: DP 301 ii 1; DP 305 iii 2; HSS 3 17 rev. ii 4; RTC 35 v 2; TSA 14 rev. i 2; TSA 15 rev. ii 3. All attestations come from Girsu/Lagash. D.  Lugal-Kur.nigín.ruki “King – Kur-nigin-ru”: PBS 9/1 2 iv 6 (from Girsu/ Lagash).

492

Jan Tavernier E.  Lugal-Šir.bur.laki “King – Girsu/Lagash”: BIN 8 370: 16; DP 114 x 13; DP 115 xi 1; DP 135 xi 9; DP 139 i 4; DP 177 viii 2; HSS 3 17 rev. ii 5; HSS 3 28 rev. ii 7; TSA 14 rev. i 3; TSA 15 rev. ii 4; TSA 47 iii 3. All attestations come from Girsu/Lagash. F.  Lugal-Urim “King – Ur”: ECTJ 6 i 4; ECTJ 38 i 9. Both texts come from Nippur.

3. The Sumerian Names of the Sargonic Period The next section deals with the Sumerian names attested in texts dating from the Sargonic period. For these names, the same typology as the one drawn up for the Pre-Sargonic names can be used.

3.1. Ethnonyms 1.  Lú-GN “The man from GN” A.  Lú-Abxḫaki “The man from Nina”: MVN 6 89 rev. 4–5; MVN 6 379 rev. 3; MVN 6 443 obv. 3; RTC 186: 9. All attestations come from Girsu/ Lagash. B.  Lú-dA-ga-d[èki] “The man from the divine Akkad”: CUNES 00-04-060 rev. 5 (from Adab). Cf. Such-Gutiérrez (2005: 2, n.14) on this name. C.  Lú-En.lílki “The man from Nippur”: ITT 2 2942 (from Girsu/Lagash). D.  Lú-Giš.ùḫki “The man from Umma”: ITT 2 5885 iii 5′; MVN 6 378 obv. 4. Both attestations come from Girsu/Lagash. E.  Lú-Ḫu-riki “The man from Ḫuri”: MVN 7 306: 2 (from Girsu/Lagash). F.  Lú-Nunki “The man from Eridu”: MVN 6 41 rev. 3′ (from Girsu/Lagash). G.  Lú-Šir.bur.laki “The man from Girsu/Lagash”: ITT 2 2942; ITT 2 4513; MVN 6 102 rev. 3′; MVN 6 423 obv. 2, rev. 6. All attestations come from Girsu/Lagash. H.  Lú-Ud.nunki “The man from Adab”: CUSAS 19 64 obv. 5 (from Adab); CUSAS 27 26 obv. 5; CUSAS 27 27 rev. 7 (from Mashkan-ili-Akkade); OIP 14 193 rev. 6 (from Adab). I.  Lú-Uruxgána-tenû “The man from Urub”: MVN 6 543 obv. i 1, ii 16, rev. iii 3 (from Girsu/Lagash). 2.  Munus-GN “The woman from GN” A.  Munus-Urimki “The woman from Ur”: RTC 248 rev. 3′ (from Girsu/ Lagash).

3.2. Genitival Names 12 The second type is the genitival names, which, as was the case for the PreSargonic names, possibly are hypocoristics of longer names: 1.  Géme-GN “The female servant of GN”

12.  Edzard, Farber, and Sollberger (1977: 161) mention a Sargonic personal name Ìr-Tu-tu (OAIC 4: 12), in which they tend to see the toponym Tutu. The name would thus mean “Servant of the city of Tutu.” Nevertheless, it is rather plausible to see the divine name Tutu here and to have a name ArádTutu “Servant of Tutu”, a common name type in Mesopotamia. An additional argument for this is that Tutu is frequently written without his divine determinative (Richter 2014: 241).

Ur and Other Cities A.  Géme-Šir.bur.laki “The female servant of Girsu/Lagash”: CT 50 88: 5 (from Girsu/Lagash). 2.  Me-GN “The divine forces of GN” A.  Me-Ki-te4 “The divine forces of Kite”: OAIC 33: 25 (from the Diyala River region). B.  Me-ù-Kèš “The divine forces of Kesh”: CUSAS 27 76 obv. 5 (from Mashkan-ili-Akkade). C.  Me-Unugki “The divine forces of Uruk”: MC 4 57: 2 (unknown provenance). 3.  Nin-GN “The Lady of GN” A.  Nin-Kèški “The lady of Kesh”: CUSAS 27 41 iv 5; CUSAS 27 42 iii 14; CUSAS 27 102 obv. 10 (from Mashkan-ili-Akkade). B.  Nin-Ab.mùški “The lady of Zabalam”: CUSAS 27 148 obv. 9 (from Mashkan-ili-Akkade). 4.  Puzur-GN “The protection of GN” A.  Puzur4-Kèški “The protection of Kesh”: MAD 1 237 rev. 8 (from Tutub). 5.  Saĝ-GN “The head of GN” A.  Saĝ-Kèš “The head of Kesh”: CUSAS 23 155 obv. iii 17 (unknown provenance). 6.  Ur-GN “The hound/man(?) of GN” (cf. Di Vito 1993: 34–35, 48–49 for this type of name). A.  Ur-Kèški “The hound/man(?) of Kesh”: CCL D14 (seal; unknown provenance); CUSAS 23 91 rev. ii 6 (from Adab); MAD 1 135 rev. 8′ (from Eshnunna); OIP 14 192: 6 (from Adab); OIP 104 40 C ix 25, x 17, xi 15 (unknown provenance). B.  Ur-Már-da “The hound/man(?) of Marad”: OIP 104 40 C iii 5, 13, ix 24 (unknown provenance). C.  Ur-Saĝ-ùbki “The hound/man(?) of Sagub”: MVN 7 416 obv. 7 (from Girsu/Lagash). D.  Ur-Tum-al6 “The hound/man(?) of Tummal”: OSP 2 93 iv 7; OSP 2 98 v 2; OSP 2 99 vi 4; OSP 2 104 v 2. All attestations come from Nippur. 7.  Ur-saĝ-GN “The hero of GN” A.  Ur-saĝ-A.kib!.nunki “The hero of Akibnun”: ECTJ 24 ii 2; ECTJ 56 i 4; OSP 1 31 i 3′. Both attestations come from Nippur. B.  Ur-saĝ-A-ga-dèki “The hero of Akkade”: CUSAS 27 148 obv. 9 (from Mashkan-ili-Akkade); MCS 9 239: 2 (unknown provenance); RTC 102 i′ 5′ (from Girsu/Lagash). C.  Ur-saĝ-Kèš “The hero of Kesh”: CUSAS 23 125 obv. 2 (from Adab). D.  Ur-saĝ-Ud.kib.nunki “The hero of Sippar”: OIP 104 40 C vi 7, 13 (unknown provenance).

3.3. Predicative Names 1.  GN-ki-du10 “GN is a good place” A.  A-ga-dèki-ki-du10 “Akkade is a good place”: OSP 2 68: 7 (from Nippur).

493

494

Jan Tavernier B.  Giš.ùḫ-ki-du10 “Umma is a good place”: Nik. 2 84 rev. 3 (from Girsu/ Lagash). C.  Im-ki-du10 “Karkar is a good place”: CUSAS 23 106 obv. 5 (written Imki; from Adab); CUSAS 23 196 rev.4 (unknown provenance). D.  In-ki-du10 “Isin is a good place”: PBS 9/1 50 rev. 7 (from Nippur). E.  Šir.bur.laki-ki-du10 “Girsu/Lagash is a good place”: ITT 2 2948 (from Girsu/Lagash). F.  Ud.nunki-ki-du10 “Adab is a good place”: OSP 1 102 ii 4 (from Nippur); TCBI 1 92 obv. 3, rev. 4 (written Ud.nunki-ki; from Adab). 2.  GN-ki-ĝal-la “GN is a big place” A.  Ud.nunki-ki-ĝal-la “Adab is a big place”: TCBI 1 26 obv. ii 7 (from Adab).

3.4. Complex Names 1.  Ab.mùš-ir-nun “The (city) of Zabalam has a princely scent” 13: CUSAS 26 162 obv. 1 (from Umma). 2.  En.lílki-da-lu “With (= thanks to/by) Nippur – we are numerous” 14: OSP 1 138 ii 1 (from Nippur). 3.  Lugal-Ti-ra-áš-šè “The king (has come) to Tirash”: CT 50 106: 56; ITT 2 2856; ITT 2 3030; ITT 2 5795. All texts come from Girsu/Lagash. 4.  Lugal-Urimki-e “The king (is) in Ur”: TCBI 1 227 obv. 4 (from Adab). 5.  Me-Unugki-ge-si “The divine forces that fill Uruk”: CUSAS 23 82 obv. i 2–3; CUSAS 33 281 obv. ii 6, 10. Both attestations come from Adab. 6.  Nin-Kèš-e “The lady (is) in Kesh”: CUSAS 27 42 iii 14; CUSAS 27 102 obv. 10 (both from Mashkan-ili-Akkade). 7.  Ti-Unugki-e “The life (is) in Uruk”: ITT 4 7055 (from Girsu/Lagash).

3.5. Hypocoristics Finally, the fifth type is again the hypocoristic anthroponyms, with their subcategories: 1.  Identical to the toponym A.  É-igi-ilki: ITT 2 4556; ITT 2 4580; ITT 2 4637. All three texts come from Girsu/Lagash. B.  En.lílki: MVN 3 61 iii 3′ (unknown provenance). C.  Ki-nu-nirki: CT 50 51: 6′; CT 50 182: 10. Both texts come from Girsu/ Lagash. D.  Már-da: TCBI 2 I-50 obv. ii 2 (from Adab). E.  Ab.mùš: BIN 8 243: 14 (unknown provenance). 2.  Names composed of the toponym and a suffix A.  Ĝír-s[uki-t]a “From Girsu/Lagash”: ITT 2 3050 rev. 2 (from Girsu/ Lagash). B.  Possibly the name in CUSAS 23 70 ob. ii 6 (⌈Kèški-x⌉) must be read Kèški-ta. 13.  Th. Balke, pers. comm. 17/10/2017. The element ir-nun also occurs in names such as É-ir-nun and Lugal-ir-nun. 14.  Th. Balke, pers. comm. 17/10/2017. Cf. the Pre-Sargonic royal name Lugal-da-lu, borne by a king of Adab (cf. Frayne 2008: 23 and Marchesi and Marchetti 2011: 173).

Ur and Other Cities

495

C.  Šir.bur.laki-e “To Girsu/Lagash”: ITT 2 4475 (from Girsu/Lagash). D.  Ud.nunki-bé “In Adab”: CUSAS 11 84 rev. i 2; CUSAS 11 151 obv. ii 1; CUSAS 19 173 obv. 4′; CUSAS 26 94 rev. 6; CUSAS 26 118: 6; MVN 3 28 iv 3′; TCBI 1 26 obv. ii 5; TCBI 1 64 obv. iv 6; TCBI 1 78 obv. 6; TCBI 1 79 rev. 6′; TCBI 1 96 obv. 2; TCBI 1 125 obv. i 4; TCBI 1 224: 5. All texts, except MVN 3 28 iv 3′ (unknown provenance), come from Adab. 3.  Names composed of a noun and a toponym A.  Lugal-GN: i. Lugal-En.lílki “King – Nippur”: BRM 3 26: 4; ITT 2 4635. Both texts come from Girsu/Lagash. ii. Lugal-Kèški “King – Kesh”: CUSAS 27 32 ii 20 (from Mashkan-iliAkkade); OIP 14 124: 2 (from Adab); TCBI 1 179 obv. 3 (from Adab). iii.  Lugal-Ki-en-gi: “King – Sumer”: ITT 2 2940 (from Girsu/Lagash). iv. Lugal-Urim “King – Ur”: CUSAS 23 164 obv. 2 (unknown provenance).

4. Pre-Sargonic and Sargonic Akkadian Anthroponyms Next to the Sumerian ones, there are also some Akkadian names attested in both periods. It should however not surprise us that those names occur especially in the Old Akkadian period, when Akkadian became the principal administrative language. The only examples from the Pre-Sargonic period are Iš-qi-Mari (RA 31 140 no. 174: 1; from Mari) “Mari has become elevated” and Ud.nunki-bí-tum “Adab is the house” (CUSAS 33 289 obv. iv 2; from Adab). With regard to the toponymic anthroponyms of the Sargonic period, one can see the following types: 1.  Ethnonyms: A.  Plain ethnonyms i. Akkade’um “The Akkadian” (spelled A-ga-dè-um and A-ga-ti-um: HSS 10 44: 1; HSS 10 48: 9; HSS 10 94: 10; HSS 10 105 i 3; HSS 10 108: 18; HSS 10 112: 6; HSS 10 133 vi 8; HSS 10 157 i 6; HSS 10 161: 16; HSS 10 212: Le.Ed.; all from Nuzi). ii.  Elamitum “The Elamite woman” (spelled E-la-mi-tùm; OAIC 9: 10; from the Diyala River region). iii. Subaritum “The Subarite woman” (spelled Śu-ba-rí-tum; MAD 1 292: 3; MAD 1 306: 3; MAD 1 331: 2; from Eshnunna). iv.  Subari’um “The Subarean” (spelled Śu-ba-rí-um; MAD 5 98 rev. i 6′; OAIC 29: 2; from Kish and the Diyala River region). B.  Composed ethnonyms i. Šu-A-wa-alki “The one from Awal”: HSS 10 185 iv 11 (from Nuzi). ii. Šu-Šir.bur.laki-e “The one from Girsu/Lagash”: CT 50 83: 4 (from Girsu/Lagash). iii. Šu-Ti-bar “The one from Tibar”: HSS 10 108: 11, 23 (from Nuzi); MAD 1 163 ix 21 (from Eshnunna).

496

Jan Tavernier 2.  Genitival names: A.  Dan-Ti-bar “The strong one of Tibar”: HSS 10 185 ii 14 (from Nuzi). B.  Ú-ṭa-at-Kèški “The barley of Kesh”(?): CUSAS 27 116 ii 14; CUSAS 27 157 rev. 2; CUSAS 27 163 rev. 13; CUSAS 27 187 obv. 10; CUSAS 27 196 rev. 8 (from Mashkan-ili-Akkade). 3.  Complex names: Ipḫur-Kiš “Kiš has gathered” (spelled Ip-ḫur-Kiški; MAD 1 172: 5; from Eshnunna) and Ummī-Gasur “My mother is Gasur” 15 (spelled Ama-Ga-sag; OAIC 20: 2; from the Diyala River region; cf. Gelb 1955: 242). 4.  Hypocoristic names: A.  Amurrum (spelled A-mu-ur-ru-um; BIN 8 142: 5; unknown provenance). B.  Names composed of a divine name and a toponym: Eš4-tár-A-wa-alki “Ishtar – Awal” (HSS 10 153 i 6 [from Nuzi]).

5. Some Concluding Remarks Some concluding remarks may be made with regard to this category of names and their typology. 1.  Not many personal names have a toponym in them. 2.  The female character of cities is confirmed by names such as Ummī-Gasur and Zabalam-ama (cf. Stamm 1939: 92). 3.  There is in general no link between the origin of the name and the toponym attested in the name. Nevertheless, two remarks can be formulated: (1) names containing the element Girsu/Lagash only occur in texts from that region; (2) names containing the element Nippur only occur in texts from Girsu/Lagash and Nippur. 4.  Interestingly, the place names often appear in the same role as temples or gods. Name types such as Lú-GN and Ur-GN are extremely frequently attested as Lú-DN and Ur-DN: e.g. Lú-Ningirsu (Limet 1968: 485), Ur-Šara (Limet 1968: 560), etc. On the other hand, there is no exclusive context for the use of toponyms in anthroponyms. 5.  Only one divinity (Ishtar) appears in the names studied here. 6.  In general, there are no differences in name-giving practices between the Pre-Sargonic and the Sargonic periods, except the larger presence of Akkadian elements in the latter period. A Sargonic accent is also that the Sargonic names show little variation in predicative names, when compared to their Pre-Sargonic counterparts 7.  According to Limet (1968: 173), some of these names refer to processions in which lugal refers to a king or a god (epithet). As is generally known, the king participated in such processions as representative of the city’s tutelary divinity (Falkenstein 1956: 329). Examples are the Pre-Sargonic names Lugal-Eriduki-šè “The king (has come) to Eridu” and Lugal-GN-ta “The king (has departed) from GN” and its hypocoristic GN-ta (Kèški-ta).

15.  Ancient name of Nuzi.

Ur and Other Cities

497

Table 1: Relation between origin of the texts and the toponymic anthroponyms (Pre-Sargonic period).

Origin

Toponym

Example(s)

Adab

Adab

Adab-be

Giš-kúšuxkaskal

Giš-kúšuxkaskal-ki-du

Isin

Isin-ki-du

Kesh

Keš-a-nun, Ur-Keš, Ur-saĝ-Keš, Keš-pa-e

Kian

Kian-ki-du

Ku’ara

Amar-Ku’ara

Zabalam

Zabalam-ama, Zabalam-ki-du

An’eden

Šeš-An’eden

Antasurra

Ur-Antasurra

Eridu

Lugal-Eridu-še

Girid

Amar-Girid, Ḫe-Girid

Girsu/Lagash

Girsu-ki-du, Lagaš-ĝìri-na, Lugal-Lagaš, Sipa-Lagaš-ki-aĝ, Sipa-Lagaš-e-ki-aĝ

Gišša

Gišša-ki-du

Gu’aba

Gu’aba-ki-du

Gu’eden

Ur-Gu’eden

Kesh

Keš-ta, Lugal-Keš

Kinunir

En-Kinunir-ki-du

Kisala

Ur-Kisala

Kish

Lugal-Kiš, Ur-Kiš

Kuara

Amar-Ku’ara, Ku’ara-ki-du

Kulaba

Me-Kulaba-ta

Kur-nigin-ru

Lugal-Kur-nigin-ru

Nina

Nina-ki-du

Nippur

Nippur

Pasirra

Ur-Pasirra

Saḫar

Saḫar-ki-du

Sirara

Me-Sirara-du, Me-Sirara-ta, Sirara-pa-e

Tirash

Ur-Tiraš, Tiraš, Tiraš-še

Urua

Nin-Uruxa-i-du

Uruaz

Lu-Uruaz

Urub

Ur-saĝ-Urub, Urub-ki-du

Uruk

Nin-e-Uruk, Nin-e-Uruk-nirĝal

Zabalam

Zabalam-ki-du

Kinunir

Kinunir-ki-du

Girsu/Lagash

Luristan

498

Jan Tavernier

Origin

Toponym

Example(s)

Nippur

Abzanu

Lu-Abzanu

Adab

Adab

Akibnun

Ur-saĝ-Akibnun

Eresh

Lu-Ereš

Isin

Amar-Isin, Ur-Isin

Kesh

Ur-saĝ-Keš

Kish

Lu-Kiš

Ku’ara

Lu-Ku’ara

Larsa

Lu-Larsa

Marad

Lu-Marad

Nippur

Lugal-Nippur

Shuruppak

Lu-Šuruppak

Ur

Lugal-Ur

Uruk

Lu-Uruk

Urusagrig

Lu-Urusagrig

Adab

Lu-Adab

Isin

Isin-ki-du

Kinunir

Lu-Kinunir, Ur-Kinunir

Kish

Kiš-du-du

Ku’ara

Amar-Ku’ara

Umma

Lu-Umma

Urua

Lu-Urua, Urua

Urusagrig

Lu-Urusagrig

Gišša

Gišša-ki-du

Kesh

Keš-ki-du, Me-Keš

Urum

Amar-Urum

Zabalam

Zabalam-ki-du

Adab

Adab-ke

Isin

Lu-Isin

Kesh

Ur-Keš

Adab

Lu-Adab

Sirara

Amar-Sirara-si

Shuruppak

Ur-Šuruppak

Ur

Ur-ki-du, Ur-pa.a, Ur-zi-ĝu

Shuruppak

Umma

Unknown provenance

Ur

Ur and Other Cities

499

Table 2: Relation between origin of the texts and the toponymic anthroponyms (Sargonic period).

Origin

Toponym

Example(s)

Adab

Adab

Adab-be, Adab-ki-du, Adab-ki-gal-la, Lu-Adab

Akkade

Lu-Akkade

Kesh

Lugal-Keš, Ur-Keš, Ur-saĝ-Keš

Marad

Marad

Ur

Lugal-Ur-e

Uruk

Me-Uruk-ge-si

Diyala River region

Kite

Me-Kite

Eshnunna

Kesh

Ur-Keš

Girsu/Lagash

Akkade

Ur-saĝ-Akkade

E’igi-il

E’igi-il

Eridu

Lu-Eridu

Girsu/Lagash

Geme-Lagaš, Girsu-ta, Lagaš-e, Lagaš-kidu, Lu-Lagaš

Ḫuri

Lu-Ḫuri

Kinunir

Kinunir

Nina

Lu-Nina

Nippur

Lu-Nippur, Lugal-Nippur

Sagub

Ur-Saĝub

Sumer

Lugal-Kiengi

Tiraš

Lugal-Tiraš-e

Umma

Lu-Umma, Umma-ki-du

Ur

Munus-Ur

Urub

Lu-Urub

Uruk

Ti-Uruk-e

Adab

Lu-Adab

Akkade

Ur-saĝ-Akkade

Kesh

Lugal-Keš, Me-u-Keš, Nin-Keš, Nin-Keš-e

Zabalam

Nin-Zabalam

Adab

Adab-ki-du

Akibnun

Ur-saĝ-Akibnun

Akkade

Akkade-ki-du

Isin

Isin-ki-du

Nippur

Nippur-da-lu

Tummal

Ur-Tummal

Tutub

Kesh

Puzur-Keš

Umma

Zabalam

Zabalam-ir-nun

Unknown provenance

Adab

Adab-be

Mashkan-ili-Akkade

Nippur

500

Origin

Jan Tavernier

Toponym

Example(s)

Akkade

Ur-saĝ-Akkade

Karkar

Karkar-ki-du

Kesh

Saĝ-Keš, Ur-Keš

Marad

Ur-Marad

Nippur

Nippur

Sippar

Ur-saĝ-Sippar

Ur

Lugal-Ur

Uruk

Me-Uruk

Zabalam

Zabalam

Table 3: Relation between origin of the texts and the toponymic anthroponyms (Akkadian names).

Origin

Toponym

Example(s)

Adab

Adab

Adab-bitum

Diyala River region

Elam

Elamitum

Gasur

Ummī-Gasur

Subaru

Šubari’um

Kish

Ipḫur-Kiš

Subaru

Śubarítum

Tibar

Šu-Tibar

Girsu/Lagash

Girsu/Lagash

Šu-Lagash

Kish

Subaru

Šubari’um

Mari

Mari

Išqi-Mari

Mashkan-ili-Akkade

Kesh

Uṭṭat-Keš

Nuzi

Akkade

Akkade’um

Awal

Ištar-Awal, Šu-Awal

Tibar

Dan-Tibar, Šu-Tibar

Amurru

Amurrum

Eshnunna

Unknown provenance

Bibliography Bauer, J. 1998 Der vorsargonische Abschnitt der mesopotamischen Geschichte. Pp. 431–585 in Mesopotamien. Späturuk-Zeit und Frühdynastische Zeit, eds. J. Bauer, R. K. Eng­ lund, and M. Krebernik. OBO 160/1. Freiburg: Universitätsverlag. Benz, F. L. 1972 Personal Names in the Phoenician and Punic Inscriptions. A Catalog, Grammatical Study and Glossary of Elements. Studia Pohl 8. Rome: Biblical Institute Press. Di Vito, R. 1993 Studies in Third Millennium Sumerian and Akkadian Personal Names. Studia Pohl: Series Maior 16. Rome: Pontificio Istituto Biblico. Edzard, D. O. 1959 Königsinschriften Des Iraq Museums II. Sumer 15: 19–28.

Ur and Other Cities

501

Edzard, D. O., Farber, G., and Sollberger, E. 1977 Die Orts- und Gewässernamen der präsargonischen und sargonischen Zeit. RGTC 1. Wiesbaden: Dr. Ludwig Reichert Verlag. Falkenstein, A. 1956 Die neusumerischen Gerichtsurkunden, vol. 2. Bayerische Akademie der Wissenschaften. Phil.-hist. Klasse. Abhandlungen N.F. 40. München: Beck. Frayne, D. 2008 Presargonic Period (2700–2350 BC). RIME 1. Toronto: University of Toronto Press. Gelb, I. J. 1955 Old Akkadian Inscriptions in Chicago Natural History Museum. Texts of Legal and Business Interest. Fieldiana. Anthropology 44/2. Chicago: Chicago Natural History Museum. Gelb, I. J., Purves, P. M., and MacRae, A. A. 1943 Nuzi Personal Names. OIP 57. Chicago: The University of Chicago Press. Kinlaw, D. F. 1967 A Study of the Personal Names in the Akkadian Texts from Ugarit. Ph.D. dissertation, Brandeis University. Lambert, M. 1971 Ur-kin-nir. RA 65: 94. Laroche, E. 1952 Recueil d’onomastique Hittite. Paris: Klincksieck. 1966 Les Noms des Hittites. Études linguistiques 4. Paris: Klincksieck. Limet, H. 1968 L’Anthroponymie sumérienne dans les documents de la 3e dynastie d’Ur. Bibliothèque de la Faculté de philosophie et lettres de l’Université de Liège 180. Paris: Les Belles Lettres. Marchesi, G. and Marchetti, N. 2011 Royal Statuary of Early Dynastic Mesopotamia. MC 14. Winona Lake, IN: Eisenbrauns. Noth, M. 1928 Die israelitischen Personennamen im Rahmen der gemeinsemitischen Namengebung. Beiträge zur Wissenschaft vom Alten und Neuen Testament 3/10. Stuttgart: Kohlhammer. Richter, T. 2014 Tutu. RlA 14/3–4: 241–242. Selz, G. 2013 On Some Mesopotamian Early Dynastic Toponyms. Pp. 317–326 in Florilegium Aegyptiacum – Eine wissnschaftliche Blütenlese von Schülern und Freunden für Helmut Satzinger zum 75. Geburtstag am 21. Jänner 2013, eds. J. Budka, R. Gundacker, G. Pieke, and H. Satzinger. Göttinger Miszellen. Beihefte 14. Göttingen: Seminar für Ägyptologie und Koptologie der Universität Göttingen. Sollberger, E. 1969 Giššakidu. RlA 3/6: 404. Stamm, J. J. 1939 Die akkadische Namengebung. MVAG 44. Leipzig: Hinrichs. Such-Gutiérrez, M. 2005 “Untersuchungen zum Pantheon von Adab im 3. Jt.” AfO 5: 1–44. Tavernier, J. 2006 Iranian Toponyms in the Elamite Fortification Archive. Beiträge zur Namenforschung 41: 371–397.

502

Jan Tavernier

2007 Iranica in the Achaemenid Period (ca. 550–330 B.C.). Lexicon of Old Iranian Proper Names and Loanwords, Attested in Non-Iranian Texts. OLA 158. Leuven: Peeters.

Implications of Intertextuality: Erra and Išum and the Lamentation Over the Destruction of Sumer and Ur Selena Wisnom

The Queen’s College, University of Oxford Interconnectedness is one of the defining characteristics of Mesopotamian scholarship. Every cuneiform sign has a wide semiotic range, omens are interpreted through word play and puns, and all such connections were thought to reflect fun­ damental truths about the universe. 1 As such, intertextuality is deeply embedded in the Mesopotamian worldview, and interpreting a text depends on grasping its surrounding networks of meaning. This is particularly true for literary texts, which are full of puns, word plays, and allusions to other poems. When we understand the significance of literary allusions, new layers of meaning are unlocked which can yield surprising new insights into the text, as well as reinforce already prevalent themes. In this article I intend to show how taking account of intertextuality can open up surprising avenues of interpretation and push the limits of our understanding of cuneiform literature. In particular, I propose that there may be an intertextual connection between Erra and Išum and the Lamentation over the Destruction of Sumer and Ur (henceforth LSU). My suggestion is a controversial one—that Erra and Išum alludes to this particular city lament, or an Akkadian one based on it, despite the fact that no manuscripts of the Sumerian city laments survive after the Old Babylonian period. 2 Resemblances between the two poems go beyond what is likely to be coincidence or generic convention, and are supported by other hints that the city lament genre could have survived into the first millennium, either in a Sumerian or Akkadian form. Although it cannot be definitively proven that LSU survived into the first millennium we should not discount the possibility alto­ gether that it could have done, and there are legitimate grounds at least to question this assumption. Ultimately my aim is to show how much can be gained not only Author’s note: I would like to thank Uri Gabbay, Dina Katz, and Alan Lenzi for their comments on this paper, and Andrew George for his suggestions at an earlier stage. Responsibility for the article and its arguments is mine alone. 1.  E.g. the diviner’s manual explicitly sets out the connection between signs in heaven and earth (edition Oppenheim 1974, lines 37–42). For the influence of divination on the commentarial tradition see Frahm 2011: 20–23. 2.  Frauke Weiershäuser has come to a similar conclusion independently, which will be discussed in her forthcoming edition of Erra and Išum.

503

504

Selena Wisnom

from noticing intertextual links but also from interpreting them, and how specific connections can constitute important pieces of evidence for how the Mesopotamians thought about their world.

Intertextuality The term “intertextuality” refers to a system where texts relate to each other. It encompasses all types of text within it, and the different ways they can relate, whether by deliberate allusion, quotation, echoing of a concept, the use of stock phrases, or even casual similarities. Intertextuality is really a property of language itself: in any language, culture, or literary system, there are bound to be phrases, topics, figures of speech that recur, and which are the building blocks we use to create new ones. This, at least, is closer to what the term originally meant when it was coined by Julia Kristeva in 1966, as a philosophical statement about the nature of language (see Kristeva 1980). 3 The most common use of the word “intertextual” today, however, is simply for connections “between texts,” and that, in a basic sense, is how I use it. If intertextuality is the larger system, allusion is more specific. When we say that a line of poem X alludes to poem Y we are pointing to one specific intertextual connection. By saying this, however, we implicitly make the poem a bearer of in­ tention, an issue which has been the subject of much debate since Barthes’ famous lecture on the death of the author (published in Barthes 1977). Of course, we cannot know exactly what was in “the author’s mind” but this need not stop us making interpretations of a text as a coherent entity. The study of ancient texts is, after all, an attempt to work out what they mean, which inevitably comes back to intentions of one kind or another. Literary critics have since reclaimed intention as a useful concept, though with the proviso that the author’s ultimate intentions are not the sole or highest aim of criticism. Umberto Eco speaks of intentio operis, “the intention of the work,” which is essentially to draw sensible conclusions about what the text is doing from the text itself, without invoking any biographical information about its author (Eco 1992: 66). 4 In Classics, Heath (2002: 63–64) has argued that this is not to ask about un­ knowable psychological states, but rather to enquire into the text as the product of purposive behaviour. What the debates about authorial intention in literary theory have done for us is to show that arguments about whether or not an allusion is intended are really just about how strong the resemblance between the two texts really is—how likely it is that a connection is there, and that an educated audience of the time would also perceive it (Fowler 1997: 15). If there is enough evidence in the text itself for an interpretation to be reasonable and likely, then we are justified in making it. What matters is that our enquiries remain empirical and focused on the evidence of the text (Lyne 1994; Fowler 2000: 112). Furthermore, secondary interpretations are just as valid as authorial intentions, and what later Mesopota­ mians could have seen in a text is just as useful to us as what the scribe might have originally envisioned. 3.  See Seri 2014: 89–91 for a brief overview of the term’s history, and a discussion of intertextual­ ity in Enūma eliš. Intertextuality in Anzû, Enūma eliš, and Erra and Išum is treated in detail in my forthcoming book. 4.  That is, not to commit the so-called “intentional fallacy” (Wimsatt and Beardsley 1946).

Implications of Intertextuality

505

Allusions can in fact help us understand more about the text’s original pur­ pose—about the intention of the work. For example, the references that Enūma eliš makes to Anzû, first pointed out by Lambert (1986), establish Marduk’s dominance over Ninurta as part of a sustained ideological programme that runs throughout the text (Katz 2011). These allusions—and others—are so frequent and pronounced that it is uncontroversial to say that one of the purposes of the poem is to demonstrate Marduk’s superiority over other gods (as e.g. Lambert 2013: 248). Allusions can, then, be important signs of other agendas in the text, and can constitute evidence in their own right if they are identified and interpreted. Allusions are often part of a deliberate strategy to create meaning as they impli­ citly ask us to compare the current text with the one being alluded to. Resemblances can of course occur by coincidence as well as by design, such as in the use of stock phrases, shared passages, and generic conventions. However, we should not neces­ sarily exclude these from being meaningful. In fact, it can be said that the prevalence of generic language and topoi can actually enable allusive play. Each redeployment of a formula will have its own individual meaning due to its use in a different context, and yet will resonate with its occurrences elsewhere in the tradition. Pucci (1987: 29) argues that repetitive language in fact makes allusion inevitable, as it results in “an incessant play of sameness and difference” that encourages us to reflect on such comparisons. It is also possible for one formula to become famously associated with one particular instance of its use, and thus this will inevitably be recalled (Hinds 1998: 34–47). Thus, arguments about whether or not a resemblance is “deliberate” or coincidental are, in fact, not so important after all: meaning is constructed at the point of reception as well as at the point of creation, in the mind of the audience as well as of the author. This is as true for ancient audiences as modern ones, for other Mesopotamians could also have had varying interpretations. What matters is the strength of the connection, rather than outright proof that it was intended. Not only is such proof impossible to obtain from ancient Mesopotamia, where the authors have been dead for thousands of years, but the interpretation of modern literary texts does not require it either, even when the authors are alive and available to be asked. This is a point of encouragement for the study of our ancient texts, for it means that deeper interpretations are still possible despite the fact that we know little about the original circumstances of their composition.

Erra and Išum and LSU Erra and Išum is a poem of national disaster that laments the fall of Babylon as it is torn apart by enemy invasion and civil war. It was composed during the early first millennium, though a precise date is elusive. 5 Thematically it shares a great 5.  Attempts to date the poem have been based on finding historical parallels to the events it de­ scribes. Sutean invasions occurred in the eleventh century (Oppenheim 1977: 267–268), causing damage that was only restored in the ninth, and Lambert (1958: 397–400) suggests that Erra was composed as part of a celebration of this revival, similarly to how Enūma eliš may have been composed after the retrieval of the statue of Marduk from Elam. Others favour the tumultuous eighth century (Cole, 1994) based on disturbances in Uruk (von Soden 1971: 255), including the loss of Ištar’s statue (Beaulieu 2001). A Neo-Assyrian date has been proposed by Gössmann (1955: 89) and Franke (2014), although an inscription of Sennacherib probably alludes to a line from the poem thus providing a terminus ante quem (Weissert 1997: 197).

506

Selena Wisnom

deal with the Sumerian city laments. The six basic themes of the genre 6 are all present to a remarkable degree: destruction of the city, assignment of responsibility, abandonment of the city by its god, restoration, return of the god, and presentation of a prayer (Green 1975: 295). Marduk himself utters a lament for his city in IV 40–44, which Oppenheim (1977: 268) characterised as taking up an old Sumerian literary tradition. Erra develops these themes in its own way and fuses them with elements of Akkadian heroic poetry and its own narrative concerns. 7 The style of Erra, too, has been compared with the Sumerian city laments, in that the story is largely conveyed through speeches (George 2013: 47). However, Erra shares specific similarities with LSU that are not found in the other city laments or other more contemporary texts. Remarkably, Erra mirrors LSU in structure at a detailed level. When the plot events of this lamentation are analysed kirugu-by-kirugu 8 and compared with those of Erra tablet-by-tablet, sev­ eral correspondences appear: Table 1. A structural comparision of the LSU and Erra LSU

Erra

Kirugu 1

The gods’ intention to destroy Sumer is narrated and they begin to carry it out.

Tablet I

Kirugu 2

Begins with the god Zababa aban­ doning his dwelling (Kiš).

Tablet II Begins with Marduk rising from (fragmentary) his dwelling.

The gods of various cities in Sumer abandon their cities and there is lamentation. Kirugu 3

The destruction reaches Ur. Nanna beseeches Enlil, describing the city’s plight and pleading for its restitution.

Erra announces his intent to wreak destruction, spurred on by the Sebēttu, and starts to put his plan into action by unseating Marduk.

The gods in assembly are con­ cerned about Erra’s anger (fragmentary). Tablet III

Erra begins to carry out his destruction. Išum appeals to Erra, describing the plight of Babylon and rebuk­ ing him for it. Erra replies that the gods have made their decision and implies that Išum is foolish for opposing it. Išum responds. Erra responds (lost). Išum responds.

6.  I follow Green (1975) and Vanstiphout (1986a) in taking the city laments together as a genre for critical purposes. 7.  Cf. George 2007: 13, noting that Erra and the city laments express the same myth. 8.  Plot analysis of LSU based on that of Michalowski 1989.

Implications of Intertextuality LSU Kirugu 4

Enlil replies that Ur’s fate is de­ creed and Nanna should not waste his time weeping.

Erra Tablet IV

Nanna pleads with Enlil again.

Most, if not all, of the tablet is a speech of Išum, aiming to per­ suade Erra to desist.

This time he relents and declares that Ur shall be restored.

This time he relents.

Nanna returns to his city.

The destruction is turned back on Ur’s enemies. Ur’s prosperity is restored. Prayer that it may continue.

Išum’s speech continues. The main description of the destruction of Babylon, and other cities. In Uruk there is a corrupt governor and cultic activity has ceased.

The main description of the destruc­ tion of Ur. Includes the cessation of justice and cult activity.

Kirugu 5

507

The destruction is turned back on Babylon’s enemies. Tablet V

Erra returns to his dwelling in Emeslam. Erra declares Babylon is to become prosperous again.

Both poems begin by announcing a divine intention to wreak destruction: An, Enlil, Enki and Ninmah in LSU, versus the Seven and Erra against Babylon. This is not the case in the other city laments: the Nippur Lament and Eridu Lament open with the destruction already taking place, the Ur Lament with a list of divine abandonments, and the Uruk Lament with the gods conspiring to create a monster. Kirugu 2 of LSU sees various gods abandoning their cities. In Erra, Marduk abandons his city at the beginning of tablet II (line 1). Not all the laments put the same emphasis on divine abandonment: in the Nippur Lament it is only briefly stated (line 88), and in the Eridu Lament the gods specifically do not abandon the city (kirugu 6.7′= 13′ = 19′ = 24′). In kirugu 3 and tablet III the destruction hits the cities of Ur and Babylon. Both poems contain long descriptions of the devastation at this point. In LSU, Nanna-Suen laments the destruction and pleads with Enlil to reverse it. In the Akkadian poem, it is Išum who pleads with Erra to cease the violence, his speeches describing the damage done. The main description of the destruction of Ur and Babylon takes place in kirugu 4 and tablet IV respectively. The cessation of justice and cultic activity are import­ ant in both. After further pleading, the god responsible finally relents and turns back the destruction upon the city’s enemies. Nanna returns to his city in kirugu 5, Erra returns to his dwelling in tablet V, and prosperity is restored in both. These similarities will be discussed in detail below. Most of the events take place in the same order and in the same tablet in Erra as they do in the corresponding kirugu of LSU. This is striking because it implies that Erra does not draw on lamentation in a general way or on the city-lament genre as a whole, but on this composition in particular or another closely based on it. The kirugu/tablet division is actually found on two of the Ur witnesses (UET 6/2,

508

Selena Wisnom

132 and UET 6/2, 133) (Dahl 2011: 76). LSU is the only lamentation to have five kirugus: the Ur Lament has eleven, the Eridu Lament has at least eight, the Uruk Lament has at least twelve, and the Nippur Lament has twelve (Michalowski 1989: 6). There is no standard format or length for Akkadian heroic poetry, but the other models that influence Erra are quite different: Anzû has three, Enūma Eliš has seven, Atraḫasīs three in the OB version and at least six in the Late Babylonian version, and Gilgameš twelve (or eleven if tablet XII is not counted as part of the main story). There is thus no Akkadian precedent for Erra’s arrangement over five tablets. This relationship has two important implications. The first is the possibility that a text such as LSU could have been known in the first millennium, a thousand years later than prevailing opinion holds. While this cannot be proven outright, the extensive similarities between LSU and Erra may be a clue to a possibility that has so far been overlooked. The second is the extra meaning that these connections bring to Erra, and the new interpretations they allow us to make. This latter aspect is where the potential of intertextuality as an interpretive tool comes into its own. Strikingly, the Mesopotamians themselves drew a connection between Erra and the destruction of Ur. A scholarly text from Hellenistic Uruk, W.20030, 7, lists renowned ancient kings together with their associated apkallu or ummânu (edition Lenzi 2008). 9 The text begins with seven pairs of antediluvian kings and sages before moving on to name scholars known to us from historical periods as the ummânū along with their associated kings. The historical list begins with Sîn-lēqi-unninni, the author of Gilgameš (line 12, paired with Bilgames), and in line 13 Kabti-ilānī-Marduk is paired with Ibbi-Sîn, king of Ur who features in LSU, weeping in his palace and is led off to the land of Elam in chains: [ina tar-ṣi mdbilga]-⸢meš3⸣ šarri(LUGAL) Idsîn(30)-lēqi(TI)-unninni(ER2) lu2um-man-nu [ina tar-ṣi i]-⸢bi⸣-dsîn(30) šarri(LUGAL) mkabti(IDIM)-il-dmarduk(ŠU2) lu2um-man-nu In the time of Gilgameš, the king, Sîn-lēqi-unninni was scholar. In the time of Ibbi-Sîn, the king, Kabti-il-Marduk was scholar.

The Uruk List of Kings and Sages lines 12–13 (Lenzi 2008: 140–143)

The function of this text is to provide a mythological “genealogy” for the scholars of Uruk by connecting them with their antediluvian counterparts and earlier famous kings (Lenzi 2008: 161–162). The connection between the kings and their sages is not always clear. In the case of Sîn-lēqi-unninni, however, it is obvious that he was not contemporary with king Bilgames but rather wrote about him. Perhaps then we can deduce a similar connection between Kabti-ilānī-Marduk and Ibbi-Sîn: that this scholar was thought to be writing “about” this king in some way. It is notable in this regard that all of the post-diluvian sages in this list were well-known authors of literary texts (for further analysis see Helle, 2018: 222–224). 10 9.  See also Schaudig’s forthcoming book on the Ibbi-Sîn disaster for an analysis of this pairing. 10. Listed after Kabti-ilānī-Marduk in line 14 is Sidu, to whom was attributed authorship of a series of 35 texts in the first millennium, mostly proverbs (Finkel 1986; Frahm 2010: 168–176). Sidu is

Implications of Intertextuality

509

Here, then, is another association between the fall of Babylon and the fall of Ur, and it was made in the first millennium. 11 A connection existed between Erra and Ibbi-Sîn, and so interpreting connections between Erra and LSU may not be so implausible either.

Other Possibilities At first we might think that the balaĝ tradition would be a more likely model for Erra, since balaĝs are also laments for destruction and are extensively attested in the first millennium. The balaĝs are closely related to the city laments in style and form, even sharing whole passages (Cohen 1988: 34 ff.; cf. Löhnert 2009: 3–10). 12 They share the theme of the destruction of cities and temples; both make allu­sions to sheepfolds and cattle-pens, and both express the idea that the gods, though capricious, can be appeased. However, there are also significant differences between balaĝs and city laments. 13 The balaĝs only refer obliquely to destruction and do not tell stories in a narrative form. 14 Furthermore, the historiographical aspect of the city laments is absent from these ritual laments (Black 1991: 31), which have been generalised to fit a number of different occasions. This historical dimension is crucial to Erra, as the events it describes are more specific than those of the balaĝs. Moreover, the balaĝs lack the element of restoration, an aspect that is found in all the city laments (Tinney 1996: 51), and in Erra. The centrality of the date palm image in both Erra and LSU is another key consideration, and is not found as an image of destruction in any extant ritual laments. The balaĝs, then, are not by themselves sufficient as a source of inspiration for these particular similarities. It might also be posited that the similarities can be explained by common folktale story structures, i.e., that all stories about the destruction of cities would inevitably paired with Išbi-Erra, a king who is mentioned in proverb collection 3, many of whose proverbs feature in the series of Sidu (Frahm 2010: 176). Line 15 associates two scholars with the king Abi-ešuḫ: Gimil-Gula and Taqiš-Gula, both of whom are mentioned in the Catalogue of Texts and Authors. Gimil-Gula is said to be responsible for Marduk, Revered Lord, the Luxuriant One of Heaven, Who […], while Taqiš-Gula is assigned six different compositions, including the Exaltation of Inanna (Lambert 1962: 66, vi 7–8; 64, iv 6–9). Esagil-kīn-apli is named next, the compiler of several series of scholarly texts including Sakkikū and Alamdimmû (line 16, pairing lost). Esagil-kīn-ubba follows, associated with both Adad-apla-iddina and Nebuchadnezzar (lines 17 and 18): he is probably to be equated with Saggil-kīna-ubbib, author of The Babylonian Theodicy (van Dijk 1962: 51; Oshima 2014: 123–124). Last in the list (other than the enigmatic Nikarchos) is Ahiqar, paired with Esarhaddon, and ascribed as the author of a collection of proverbs (Lindenberger 1983). 11. A tradition associating Ibbi-Sîn with disaster survived into the first millennium in other sources, which have been collected and analysed by Schaudig in a forthcoming book. The significance of the omen tradition is also discussed there. It is important to note that Ibbi-Sîn was not the only king linked with the downfall of empires, as Šarkališarri (Koch-Westenholz 2000: no. 26: 19) and Ku-ba’u in K 766 (Leichty 1970: 8) are as well. 12.  See also Gabbay 2014: 13–14 on connections between emesal prayers and other genres. 13.  Gabbay (2014: 13) notes the arguments on both sides and concludes that the ritual laments are more likely to have preceded the city laments, though the question of the original relationship between the genres and which came into being first is of less concern in relation to Erra. 14.  Eršemma 92 comes closest, in which Enlil commands Nergal to destroy the land (Gabbay 2015: 245–251), but this alludes to a story rather than telling the story itself. Uru amirabi consisted of twentyone tablets, of which we only have portions, and so it is possible that it had more narrative content than we currently realise. The emesal prayers dealing with Inanna and Dumuzi contain more narrative details than those concerned with destruction (e.g. Edena usagake), conveyed in the form of direct speech.

510

Selena Wisnom

follow a similar trajectory. However, if this is the case, why are the Sumerian city laments so different from each other? They share the same general subject matter, but vary significantly in structure and detail (Michalowski 1989: 5–6; Tinney 1996: 20). For example, the destruction is caused either by a storm or invading enemies, or both, but both the details and the positioning in the narrative structure are differ­ ent: it is only in LSU that the storm is turned back on the invaders (Tinney 1996: 20), and in Erra (IV 131–150). The resemblances are more specific than what we would expect from a common generic background, and I have taken care to compare them with the other city laments at each step to control for this. If these similarities in structure, detail, and the structural placement of that detail are down to inherent generic patterns, then we would find them in the other Sumerian city laments too. However, this is not the case. The same is true for any assumed common background behind these texts: it is always possible that a now lost oral tradition could have accounted for these similarities, but this is ultimately unprovable. Here again the number of specific details in the comparisons weights the evidence towards a more direct connection. Furthermore, it is not only on the level of structure that the two poems are compar­ able: the key image of the date palm as a metaphor for the destruction of the city occurs in both, and will be discussed further below. The combination of similarities in both structure and detail makes it less likely that either are due to coincidence, for the details pin down the comparison and make it more specific. Lenzi (2015) has used a similar structural approach to show that Ludlul imitates the structure of Akkadian incantation prayers. Like Erra, Ludlul is a complex and multi-faceted text that draws on and alludes to a number of different compositions, not only incantation prayers, yet the interweaving of reference to this particular genre is still marked and perceptible. Lenzi compares the positions of certain elements in the text with the expectations of the genre of incantation prayers, showing that they are remarkably aligned. The similarities go beyond what can be reasonably expected to be coincidence, or even a common background (Lenzi 2015: 81). Here too it is the specificity of the comparisons which makes a strong case for borrowing; not only in the overall structure of the texts, but also in their thematic and lexical details (Lenzi 2015: 70–93). At the same time, the similarities do not only show a dependence but constitute literary allusion that produces new meaning. Lenzi interprets Ludlul as an exhortation of thanksgiving; alluding to the type of prayers that would be made by a person while still suffering thus underscores the message that restoration and healing is possible (2015: 97). Again, it is not only the similarities that are important, but what they reveal about the meaning of the poem. This kind of structural comparison can, then, be a useful tool for revealing significant connections between texts, and not only in the case of Erra and LSU.

The Continuity of a Tradition The lack of city lament manuscripts after the OB period may only be an accident of survival. Many Old Babylonian Sumerian literary compositions do survive in first millennium copies: Lugal-e, An-gin7, Enki and Ninmaḫ, Enlil and Ninlil, Enlil and Sud, The Return of Lugalbanda, the creation poem KAR 4, The Ballad of Early Rulers, The Sumerian King List, as well as many balaĝs and eršemmas. Sumerian incantations widely used in the first millennium also have early roots: Udug-hul

Implications of Intertextuality

511

goes all the way back to the Old Akkadian period (Geller 2016: 5). Tablet XII of Gilgameš is a translation into Akkadian of the Sumerian poem Gilgameš, Enkidu, and the Netherworld. Other Akkadian poems are adaptations of Sumerian originals: Ištar’s Descent (from Inana’s Descent) and Adapa, whose Sumerian version only recently became known (Cavigneaux 2014). The flood story is another such example (Civil in Lambert and Millard 1969: 138–145). These versions of Sumerian texts have indeed become part of Akkadian literature. However, it is not only purely mythological texts that are passed down to the first millennium, but ones purporting to be historical also. The so-called “Dynastic Chronicle” or “Babylonian Royal Chronicle” (Glassner 2005: 126–135) is a bilingual text listing the lengths of reigns of Babylonian kings up to the middle of the first millennium. It begins with a mythological introduction describing the descent of kingship from heaven to Eridu and giving the reigns of antediluvian rulers: this clearly goes back to the Sumerian King List (Lambert 1973: 273), and has in fact been categorized as a bilingual copy of it (Finkel 1980: 70). Jacobsen (1939: 11) even included it in his composite edition of the Sumerian King List as ms. K. The copies of the chronicle (four) are from Ashurbanipal’s library and are in both Neo-Assyrian and Neo-Babylonian script. There is evidence, then, that Sumerian historical tra­ ditions survived long after the Old Babylonian period and well into the time when Erra was composed. Sumerian texts were also adapted into Akkadian. For example, a tradition of Akkadian lamentations based on Sumerian models in the Old Babylonian period is implied by a list of eight compositions designated as laments for Bēlet-ilī, six of them with Akkadian titles (Löhnert 2009: 13). An Akkadian translation of kirugus two and perhaps three of the balaĝ uru 2 am 3 -ma-ir-ra-bi is also known from this period (UET 6/2 403; Wassermann 2003; Wassermann and Gabbay 2005). This portion of text corresponds to the lament of Inanna over her destroyed city. It is very unusual for an Akkadian translation to appear without the Sumerian text (it is never the case for Lugal-e, for example), and tantalisingly suggests that the city lament as a stand-alone piece may have crossed over into the Akkadian tradition. The lament for a city is, at least, clearly the part of the text that was of interest to whoever copied it, and it stands alone as an independent literary composition. Other kinds of laments are known in Akkadian from the first millennium: one for a woman dead in childbirth (Livingstone 1989: 37–38; George 2010), one for Dumuzi (Livingstone 1989: 39–41) and one in the manner of a lament for Dumuzi (Lambert 1983: 211–215). This last text has been compared with Erra (Foster 2005: 952) since it is primarily about the effects of war, specifically the laments of women for their dead husbands. That the author of the primary publication suggested it had been originally inspired by the destruction of Sumer and was re-copied when Babylon was under foreign domination in the Seleucid period (Pinches 1901: 196–199) reflects the similarity of its language to Sumerian laments. As Lambert (1983: 213) remarked, “Laments over Tammuz are otherwise a Sumerian genre, so the language itself is a matter of interest.” We have here the most likely example of a late Akkadian text based on Sumerian prototypes, with all evidence for the intervening period lost. It is therefore also possible that Erra was based on a later Akkadian composition, extremely similar to LSU in theme, structure, and plot, rather than LSU itself. However, since no such text survives we must work with the original Sumerian and let the results speak for themselves.

512

Selena Wisnom

Akkadian Traditions of the Destruction of Cities Further continuity of a tradition of the destruction of cities can be seen in Akkadian texts. A bilingual inscription of Nebuchadnezzar I (RIMB 2 B.2.4.8–9) describes Marduk abandoning his city in language extremely similar to that of the city laments, showing that this genre had not completely died out (Dalley 2005: 278; Löhnert 2009: 17). The inscription was composed in the twelfth century BCE and copied all the way into the seventh, as it was found in Ashurbanipal’s library. One particularly striking example is K 3200, an overlooked fragment of what looks like a city lament in Akkadian. 15 This small tablet piece from Ashurbanipal’s library consists of twenty-six lines, fifteen of which are more or less complete. The fragment was published as an appendix to Thompson’s edition of Gilgameš (1930: pl. 59), where it was called “The Oppression of Erech?” The text concerns a siege of Uruk and contains topoi familiar from the Sumerian city laments: for three years the enemy have encircled it, the inhabitants suffer, and guardian spirits abandon the city while the gods discuss the situation. The suffering of the inhabitants and livestock are familiar from Sumerian laments, though the details here are especially self-destructive. A boatman sinks his own boat (line 4), a tavern keeper shatters her own pot stands (line 6), and in the animal realm jennies trample their own foals and cows spurn their own calves (lines 7–8). The closest parallel for this behaviour is in the Uruk Lament where the cowherds overturn their own cattle pens and shepherds burn their own sheepfolds (2.14′–15′). However, these acts of self-destruction in K 3200 do not seem to occur elsewhere in the Sumerian texts. There we more usually find desertion in these cases—family members turn away from each other rather than actively causing harm to each other (e.g. LSU 95–96). The closest parallel is in fact Erra, where the inhabitants of Babylon torch their own sanctuaries (IV 14). K 3200 seems to be push­ ing the topoi of laments to the extreme, showing the people turning on them­selves through images that in other cases direct the violence onto others. Variation can also be seen in the way that the protective spirits leave the city; this also happens in other laments (e.g. Uruk Lament 2.21′–22′) but not “through the water pipes” as here (ina nu-un-ṣa-ba-a-ti, line 14). 16 What this shows is that city laments cannot be considered a homogenous group that simply repeat the same images in the same ways, as even this Akkadian frag­ ment is quite different in detail from the other surviving compositions. It demon­ strates the variety inherent within the genre: a story of a city’s destruction does not inevitably have the same features as any other. The resemblances between Erra and LSU therefore should not immediately be put down to coincidence. 15.  I thank Andrew George for bringing this text to my attention. 16. It is of course possible that the images arose independently or were drawn from the ritual laments. For example, the sinking boat does not appear in any of the surviving lines of the city laments, but does occur in at least three balaĝs (though the details are different, and they are not sunk by the boatman himself), e.g. Udam ki amus line 170: [ d il ] m u n ki s u g - g a ma 2 s u 3 - a - me n 3 , ‘I am (a merchant from) Dilmun whose ship has sunk in the swamp’ (Cohen, 1988: 148); Immal gudede line c+230: gi š ma 2 gab a-ri -a-ni g iš m a 2 su 3 ( : ) su- [ am 3 ] , e-lip-pu im-hu-ru-šu ṭi-bi-tu, ‘The ship that comes to meet him is a sunken ship, (Cohen 1988: 617); Edena usaĝake e+119: tu r- tu r- b i g i š ma 2 s u 3 - s u 3 i n - n u 2 , ṣe-eh-he-ru-tu-šu ina gišeleppi(MA2) ṭi-bi-ti3 ṣa-lu4, ‘Its young ones are lying within sunken ships’ (Cohen 1988: 689); OB version of Edena usaĝake c+78: d i 4 - d i 4 - l a 2 ma 2 s u 3 - s u 3 n u - me - e n - n a , ‘ Are you not the one who sank the boats (with) the children?’ (Cohen 1988: 673).

Implications of Intertextuality

513

Sumerian and Akkadian traditions of the destructions of cities need not be mutu­ ally exclusive. Erra draws on a vast number of different texts and traditions, 17 both Sumerian and Akkadian, and could well incorporate both. In short, in the light of these other texts, it is not so unlikely that the city laments could have survived into the first millennium, either in the Sumerian originals or in Akkadian adaptations.

The Parallels in Detail LSU opens by announcing the gods’ intended destruction of Ur in fifty-four lines. This opening is unique among the city laments, which almost all begin with the destruction already taking place. 18 Erra also begins, after the short prologue, with Erra announcing his destructive intention (I 6–9). 19 There is a delay in action while Erra hesitates (I 15–20), but the Seven are now introduced to spur him on (I 23 ff.). They utter a long speech (I 46–91) encouraging Erra to act on his impulses. Išum makes his first address to Erra, who tells him to be quiet and lists his destructive qualities in a speech of 18 lines (I 106–123). All of this is part of the announcement of the devastation to come. Kirugu 2 of LSU begins with the god Zababa abandoning his dwelling in his city of Kiš: d za-ba 4 -ba 4 ki-tuš [ki-aĝ 2 -ĝa 2 -ni giri 3 kur 2 ba-ra-an-dab 5 ], “Zababa [took an unfamiliar path away from his beloved] dwelling,” (116). 20 This is the first in a list of gods abandoning their cities, followed by their lamentation. The second tablet of Erra begins with Marduk abandoning his dwelling: ⸢it⸣-bi-ma i!-na šubti-šu2 a-⸢šar⸣ [la a-ri], “he rose from his dwelling, an [unapproachable] place” (II 1). Unlike Zababa, he is not the first in a sequence to do so—in Erra this is the only abandonment, but it is a crucial one upon which the plot turns. The other cities in Babylonia, like the others in Sumer, do suffer devastation, but this is to be related later in tablet IV, for the focus is primarily on Babylon and on setting up the conditions for the destruction to take place. In the Eridu Lament the gods destroy their cities but specifically do not abandon them (kirugu 6.7′= 13′ = 19′ = 24′). In the Nippur Lament Enlil’s abandonment of the city is only briefly stated in line 88. In the Uruk Lament all the important gods of the city abandon it (2.21′–25′). The Ur Lament begins with a list of abandonments, similarly to LSU, but in a different place in the narrative structure (1–36). 21 LSU begins kirugu 2 by listing the deser­ tion of the gods of the cities of Sumer, but when it comes to Ur in kirugu 3, its patron deity and thus the most important deity, Nanna, is the focus of the narrative. The effect of an allusion to LSU, then, is to put Marduk in the role of these lesser deities who leave their dwellings early on in the poem, abandoning hope at an early stage. In kirugu 3 the destruction hits Ur, and Enlil inflicts famine and treachery on the city (296–299 ff.). In Erra it is probably in tablet III that Erra begins his 17.  For example see Cooley 2008 on Erra’s relationship to the celestial omen series Enūma Anu Enlil. 18. The Uruk Lament opens by briefly announcing the gods’ decision to destroy Uruk (I 1–8, fragmentary), but the narrative quickly moves on to the creation of their monster (I 9–12). 19. The speaker of these lines has been debated. I follow Müller’s (1995) interpretation, which solves the difficulties raised by other identifications. See also George 2013: 51–52. For previous attribu­ tions see Weiershäuser 2010: 359–364, with further literature. 20.  Text based on Michalowski 1989. 21.  Ningal stays behind to plead on Ur’s behalf, but unlike in LSU and Erra there is no dialogue between the lamenting deity and the god responsible, and she receives no answer.

514

Selena Wisnom

onslaught; we have heard extensively about Erra’s plans for destruction, but we do not know what has taken place until we hear Išum’s account (III C.58 ff.). Cagni suggests it was narrated in the lacunae in tablet III, or that it was never described outright and we only hear of it through Išum’s speeches (1969: 208–210). Either way, the destruction must happen at this point in the narrative. Again, Erra is in a role structurally corresponding to that of Enlil, emphasising the power he enjoys. 22 Išum is distressed by Erra’s excessive violence and demands an explanation. In the course of the conversation, he catalogues its effects in another description of the destruction. This corresponds structurally to the plea of Nanna in kirugu 3 of LSU. Nanna makes two pleas to Enlil, the first unsuccessful (341–356), the second successful (451–458). Išum has one long conversation with Erra spanning III C.1–IV 127. Here he describes the catastrophe striking. In LSU this is given by the narrator between the pleas of Nanna, but here they are spoken by Išum; thus the main event occurs in a structurally similar place in both poems. Both Nanna and Išum remind their interlocutor what he has done in the hope that this will move him to desist. Nanna’s first plea emphasises that with the peo­ ple destroyed, there is no one left to bring offerings to Enlil, and thus he is acting against himself (341–347). Išum’s tactic is rather to describe all the suffering Erra has caused. He expresses care for the Babylonians, exclaiming uʼ3-a nišī(UG3)meš-a, “Ah, my people!” at III C.30, a cry similar to those uttered by city deities in both city and ritual laments. 23 Thus, although Išum is never explicitly called a city-god of Babylon, he is placed in the role of one, interceding to protect it. At the beginning of kirugu 4, Enlil replies to Nanna’s plea by telling him not to waste his time weeping for Ur; its appointed time has come and he should give up hope (361–370). This is expressed in legalistic language: di-til-la inim pu-uḫ 2 ru-um-ma-ka šu gi 4 -gi 4 nu-ĝal 2 / inim du 11 -ga an d en-lil 2 -la 2 -ka šu bal-e nu-zu, “The case, the word of the assembly cannot be reversed, / it is not known to overturn the word spoken by An and Enlil” (364–365). Enlil’s tone is aggressive: [dumu]-⸢ĝu 10 ⸣ dumu gi 7 ni 2 -za-bi-me-en er 2 -ra -na-bi-me-en, “My [son], you are a Sumerian in your own right, why are you in tears?” (362a) and d nanna-ĝu 10 na-an-kuš 2 -kuš 2 -u 3 -de 3 iri ki -zu e 3 -bar-ra-ab “My Nanna, do not exhaust yourself, leave your city!” (370). Towards the end of tablet III (C.40–65) Erra responds to Išum’s speech by telling him that the decision has been made collectively by the gods: ša2 di2-gi3-gi3 ṭe3-en-šu-nu ti-de-ma ša2 da-nun-na-ki mi3-lik šu2-un, “You know the decision of the Igigi, the counsel of the Anunnaki” (III C.40). Opposing it is futile—Erra asks Išum: min3-su ki-i la mu-de-e ta-ta-me at-ta “Why are you talking like one who is foolish?” (III C.42). He implies that Marduk too has resolved on this: ki-i ša2 a-mat dMarduk(AMAR.UTU) la ti-du-u ta-mal-li-kan-ni ia-a-ši, “You counsel me as if you do not know Marduk’s word” (III C.43), and since he has left, the situation is irremediable: šar(LUGAL) ilāni(DINGIR)meš ina šub22.  Cf. Cagni 1969: 221, who suggests that Erra compares himself with Marduk here. 23.  E.g. Ningal in the Ur Lament 247 and 248, a u ru 2 - ĝ u 10 i m - m e “’Oh my city!’ she says, ‘Oh my house!’ she says”, and passim. In LSU the refrain is a i ri g u l -l a e 2 g u l - l a - ĝ u 10 , “Oh the destroyed city, my destroyed house!” (lines 118, 122, 126, 135, 138 etc.). See also Nammu in E turgin Niginam a u ru 2 ĝu 1 0 im-me “ ‘Oh my city!’ she says” (Cohen 1988: 77), or Inanna in an Uruhulake (Cohen 1988: 650–667, lines 1, 2, and 8) which is also the incipit of a balaĝ to Inanna (Cohen 1988: 642–649). In the Uruhulake to Inanna the lament takes a more personal tone as she also cries a a m a - ĝ u 10 “Oh my mother!” and a dum u-ĝu 1 0 -u “Oh my child!” (line 9). Cf. also Uru amirabi lines 62–68 (Cohen 1988: 542–543).

Implications of Intertextuality

515

ti-šu2 it-ti-bi / mātāti(KUR.KUR) nap-ḫar-ši-na i-ku-na mi-i-na, “the king of the gods has risen from his dwelling! / What is left of all the lands?” (III C.44–5). The hopelessness of the situation is again underscored: qa-bal ilī(DINGIR) u amēli(LU2) ip-paṭ-ṭa-⸢ra⸣-[ma] a-na ra-ka-si iš-ši-ṭa, “the bond between god and man is undone, difficult to tie (again)!” (III C.49). The replies of both Enlil and Erra indicate that the destruction has been decreed and therefore it is irrational to oppose it. The responses occur at slightly different points in the kirugu/tablet schema but at the same point structurally. Their contents are parallel and with similar tones. Enlil’s decision to destroy Ur results from a cruel but principled decision that the dynasty has reigned too long. 24 Erra invokes the same idea here by referring to the word of Marduk, but he is twisting it, for he obtained Marduk’s word by trickery. Kirugu 4 describes the destruction of Ur after Nanna abandons it (379–448): the main attack of the invading enemies, the hunger ravaging the people, the de­ struction of the palace, and the wrecking of the sanctuary. Throughout, the people lament their dire situation. In tablet IV of Erra comes the main description of the destruction of Babylon: Erra aims arrows at the walls and kills the gate-keeper, Babylonians attack their own city, and the governor orders his troops to kill and loot indiscriminately (IV 6–35). The similarities are not only structural, but are also found in specific details. There is one image that stands out: in Marduk’s lament for Babylon, he compares his city to a date palm: u’3-a bābilu(TIN.TIR)ki ša2 ki-ma gišgišimmari(GIŠIMMAR) qim-ma-tu2 u2-ša2-aš2-ri-ḫu-u2-ma ub-bi-lu-šu2 ša2-a-[ru] u’3-a bābilu(TIN.TIR)ki ša2 ki-ma gišterinni(ŠE.U3.SUH5) še-im u2-ma-al-lu-šu2-ma la aš2-bu-u2 la-lu-šu2 Ah, Babylon, whose crown I made as splendid as a palm tree, but the wind has. carried away! Ah, Babylon, which I filled with seed like a (date)-cone, but of whose delights I could not have enough! Erra IV 40–41

The image of the date palm also is used in kirugu 4 of LSU, describing the felling of the statue of Nin-e’iga (412–417): ⸢gešimmar⸣-gin7 gu2-guru5ru ba-ab-du11 teš2-bi ba-ra-an-kad4 gešimmar uruda nig2 kal-ga a2 nam-ur-sag-ga2 u2 numun2-gin7 ba-bu u2numun2 ba-ze2 ur2 -ba ti mi-ni-ib-bala sag sahar-ra ki ba-ni-ib-us2 lu2 zi-zi la-ba-tuku ĝeš ze2-na-bi ha-ba-⸢an-guru5⸣-uš sag šu bi2-in-hu-hu-uz ĝeš a2 zu2-lum-ma-bi pu2 ul-ul ba-ra-an-bu-bu-de3-eš [ĝeš] ĝeš

It was pruned like a date palm, the (pieces) were collected, Date Palm, Mighty Thing of Copper: Right Arm of Heroism,

24. See Chen 2013: 223–225 for variation among the city laments regarding the reason for the destruction.

516

Selena Wisnom The rushes (=hair) were torn out like weeded rushes: the roots (=feet) were turned sideways. The head (=crown) was in the dust: there was no one who could raise.it. They cut off the central ribs, the head (=crown) was burnt through, They tore out its date cluster, an everlasting well LSU 412a–417, translation following Dahl (2011)

“Date palm” is listed as one of the statue’s names in 412, and it is used as an extended metaphor for its destruction. This comes at the climax of the poem and is its greatest horror, as is the destruction of Babylon narrated on tablet IV of Erra. As one of the mainstays of the Babylonian economy, the date palm is an image of prosperity. 25 However, the felling of date palms as an image of destruction is not attested outside of these two poems. 26 This uniqueness is conspicuous. At the same point structurally, we have a similar image used to express a similar highly emotional moment: the culmination of a city’s destruction. The implication may be that the ruin of Babylon is sacrilegious, compared to the demolishing of a divine statue. The image of the felled date palm occurs in two other places in LSU leading up to this climax: at 195 where it is said that the throne in one of Nanna’s temples: ĝiš nimbar-gin 7 gu 2 -guru 5 ba-ab-du 11 , “was cut down like a date palm” – again the same image used in the same context. The other instance is in line 241 where the inhabitants cry: ĝiš nimbar-gin 7 šu nu-du 11 -ga-me a-na-aš mu-e-ze 2 er-ze 2 -re-ne, “Like date palms that have not been tended, why are we being cut down?”—the same image used to express this time the destruction of the people. Not only is this particular image unique to these two poems but it is particularly strong in LSU, which further reinforces the likelihood of a direct connection in Erra. In kirugu 4, as a result of the destroyed buildings and people, judicial admin­ istration and cult activity cease in Ur (438–448). This may be a topos, as there is also mention of it in the Uruk Lament in the second kirugu (A.2, A.4), although the surrounding lines are so fragmentary it is impossible to know how much emphasis it was accorded. What is important is that the topos occurs in the same place in Erra as it does in LSU, in kirugu 4 and tablet IV, following the main episode of destruction. Since this theme is more important in Erra than in LSU it occurs sev­ eral times rather than in one block. First, the people’s revolt against the governor 25.  E.g. in the Babylonian Theodicy (Oshima 2013) line 56: gi-šim-ma-ru iṣ-ṣi meš-re-e a-ḫi aq-⸢ru⸣, “Date palm, tree of wealth, my esteemed brother.” “Tree of riches” is a synonym for date palm in Ur5-ra Ḫubullu III 273, where 274: gi š- n i g 2 - t uk and g iš. mu . n i g 2 . t u k are equated with gi-šim-ma-ru. See Streck 2004 for the many uses of the date palm. 26. Sargon includes it among the destruction wreaked on campaign: gišgišimmaru(GIŠIMMAR) tuk-lat-su-nu giškirâti(KIRI6)meš bal-ti na-gi-šu2-nu ak-šiṭ, “I cut down the date palms they rely upon (and) orchards, the pride of their region” (Fuchs 1994: 149, l. 289); however it is part of a general description of devastation rather than carrying the same symbolic weight. Date palms appear in omens portending the destruction of Larsa and Der in Šumma ālu tablet II (omens 31 and 32) but it is the fact that they cannot be seen inside the city which gives them their meaning. Symbols of abundance not being visible inside the city probably correlates with its ill fortune. Six date palm omens also appear in the list of prodigies portending the fall of Akkad (Guinan 2002: 35–40). On the one hand, their ill-boding significance is explicable by a pattern of abnormal growth. On the other, it may well be significant that these portents were allegedly seen during the reign of Ibbi-Sîn: the name is written cryptographically as I⸢babbar⸣-aše-em-me-íb-bi, not only using Ašim-babbar as a synonym for Sîn but also writing the name backwards (Schaudig forthcoming). Thus date palms are again associated with the destruction of Ibbi-Sîn’s city.

Implications of Intertextuality

517

of Babylon, followed by the governor himself who incites his troops against his own people; second, the governor of Uruk; third, the inhabitants of Dēr, who: kit-ta ⸢u2maš⸣-ši-ra-ma iṣ-ba-ta pa-rik-ta, “forgot righteousness and took up violence” (IV 73), for which their god Ištarān forsakes them. Tablet IV is the climax of the poem, and the emphasis of this theme here makes it clear that the brutality of civil strife is the most intense sorrow of the whole poem. Kirugu 4 ends with Enlil relenting and declaring that Ur shall be restored. Kirugu 5 begins with the destructive storms that blew on Sumer turning back to their source. LSU is the only city-lament where this happens, saying: u 4 ki-engi-ra ba-e-zal-la kur-re ḫe 2 -eb-zal, “The storm that passed over Sumer indeed passed over the foreign land 27” (486). Lines 488–490 specify which lands: Tidnum, Gutium, and Anšan—those of the enemy invaders. Famine too has been inflicted upon the enemies just as it was upon Ur (492). Likewise at the end of Išum’s speech at the end of tablet IV, Erra relents, decrees that the destruction shall be turned back on the enemies of Babylonia and that an Akkadian (ak-ka-du-u2, IV 136) will rise to destroy and rule them all (IV 131–136). This goes one step further than LSU, or any of the city-laments, since LSU ends with prayers of abundance for Ur and does not express hopes of conquering its enemies. If anything, the sentiment is the opposite in LSU, expressing the wish that: an-ki niĝin 2 -na uĝ 3 ⸢saĝ si 3 ⸣-ga, “in the whole universe the people be cared for” (513). The theme of the Akkadians dominating their enemies comes up again during Erra’s instructions for Akkad’s restitution at V 27–30. The element of revenge pre­ sent in LSU (and not in the other city-laments) is thus extended to include ruling over the invaders, particularly specifying the hauling in of booty at V 30 and tribute (biltu) at V 35. However, although they have the semblance of topoi that we might expect in the other city laments, the other blessings of restitution are found only in LSU (493–512). The multiplication of the people (Erra V 25, LSU 509, 512), the safety of the roads (Erra V 26, LSU 495), and the abundance of the fields and waters (Erra V 33–34, 37, LSU 499–500) are absent from the Ur Lament, which prays for a restoration but is not specific about the details. The Nippur Lament refers to the rebuilding of houses, the sprouting of seeds, and the multiplication of animals (252–258) but not the multiplication of people, and is primarily concerned with resumption of cult. The situation is the same in the Uruk Lament, and the part that would have mentioned restoration is lost in the Eridu Lament. Erra returns to his seat at the beginning of tablet V: ul-tu der3-ra i-nu-ḫu ir-mu-u2 šu-bat-su, “When Erra had calmed down and occupied his dwelling,” (V 1). Nanna had returned to his city at the end of kirugu 4, immediately after the favourable reply from Enlil (475–477a), and the turning back of the destruction happens after this (483–491, the beginning of kirugu 5). In Erra these two events have been re­versed, presumably since although it is Išum who carries out this destruction, Erra is the abstract force that causes it and so cannot be appeased until it has been carried out. Thus established in his peaceful dwelling, he announces the regeneration of Akkad (V 25–38).

27.  Var. KK: kur-kur-ra “lands.”

518

Selena Wisnom

Interpretation Invoking LSU rather than one of the other city laments has ideological impli­ cations. Green observes that LSU identifies Ur with its long dynastic reign as the target for the gods’ anger, whereas in the Uruk Lament, for example, it is directed towards all mankind (1984: 254). By using LSU as a model rather than a different city lament, Erra may be comparing the fallen power of Babylon and the cities in its domain with the famous dynasty of Ur. This lays claim to an identity of power and splendour, rather than, for example, the religious centrality that would stem from a comparison with the Nippur Lament and Nippur. The allusion to this lament may also give a theological explanation for Babylon’s downfall, a contrast with the fate of Ur. In Erra there is no decision that, like Ur, Babylon’s appointed reign had simply come to an end. Erra’s ravaging of Babylonia is based not on a decision that its hegemony had lasted too long, but on the desire for personal recognition. Erra thus uses the departure from its model, the omission of any explanation, to give its own: the destruction is not the inevitable result of a fixed and necessary decree, but of out-of-control forces that need to be calmed. The cause of this is complex and unclear, particularly the balance of human and divine responsibility, since there has been neglect on both sides. Marduk is depicted as a weak ruler who is easily manipulated and unable to protect his people, but there are hints that the Babylonians themselves have neglected his cult. There are no easy answers to the question of culpability in this worldview. While the Ur Lament and Eridu Lament rebuke their city gods for having aban­ doned them, LSU is different in that Nanna remains loyal to his city and pleads with Enlil on its behalf (Green 1984: 254). This is a key similarity between LSU and Erra, for by making Išum plead with Erra, the poem casts Išum in the role of benevolent city god and Erra in the role of apathetic and hostile chief god. It is also significant that Erra occupies the role of Enlil and not Marduk; Erra usurps Marduk’s primacy and power elsewhere in the poem, 28 and this is another allusion that reinforces the idea. Most striking of all is the image of the destroyed date palm used in both poems at a heightened emotional moment to express the horror of the city’s destruction. In Erra the date palm symbolises the riches of Babylon: when the wind carries its fronds away, the city’s fortunes are taken with it. In LSU the breaking up of the date palm is a metaphor for the destruction of the statue of Nin-e’iga, goddess of milk and cheese, perhaps adorned with fertility symbols 29—aspects of abundance. No other examples of date palms used in metaphors of destruction are so far known, which makes an association between these two poems even stronger. Comparing the destruction of Babylon to the demolition of a divine statue implies that the ruin of Babylon is sacrilegious, an insult to the gods themselves. We should also remember that Marduk’s statue itself has been neglected by the Babylonians: its refurbishment was the reason for his absence from the city in tablet II. Perhaps, then, Marduk could also be referring to the destruction of his own statue by using this metaphor: a self-reflexive statement that he knows he is finished. At the end of 28.  It is notable that Erra is the one to return to his dwelling in Babylon rather than Marduk, who fades out of the narrative and is not heard of again (Machinist 2005: 48). 29.  According to Dahl’s (2011: 59) interpretation of lines 416–417.

Implications of Intertextuality

519

the poem it is Erra who controls Babylon, not Marduk (Machinist 2005: 221). Unlike the statue of Nin-e’iga, which was systematically dismantled and its destruction explained by some inscrutable divine plan, in Erra there is no such plan: the forces of chaos have prevailed. However, Babylon’s fall in fortune is only temporary: while Ur’s days as a ruling empire truly were over by the first millennium, Erra prophesies that Babylon will rise again. An Akkadian will arise to slay and rule all Babylon’s enemies (IV 136), and the governors of all the cities will haul their tribute into Šuanna (V 35). This is further brought out by the way that both poems uniquely turn the destruction back on their enemies: while other city laments also end with restoration, Erra takes this aspect of retaliation and pushes it further, repeatedly speaking of dominance and tribute. Even the renewed produce of the sea and fields are both described as biltu (V 33–34), and Erra’s instructions end by culminating with the command that the governors of all other cities must defer to the provisioner of Esagil (V 38). Here is another contrast: although in LSU Enlil states that Ur was not given an eternal reign (366), Babylon still has more days of predominance ahead. There are differences between the structures of the poems, too, of course. It is evident that much more happens in tablet I of Erra than in kirugu 1 of LSU, for example, but this is because Erra is not simply a lament for Babylon within the genre of city-laments, but a heroic narrative fused with the structure and imagery of a city-lament, among other things. Indeed, it is because of this combination that while it is clear that tablet IV of the poem draws heavily on laments, their influence upon the other tablets is less obvious. A structural approach enables us to see this more clearly to begin with, and then to inspect the details.

Conclusion In summary, I hope to have shown specific resemblances in structure and detail between Erra and the Lamentation over the Destruction of Sumer and Ur that may well go beyond coincidence, and that they can be meaningfully interpreted. Erra and Išum is a poem that memorialises the fall of Babylon from its former splendour, and invokes an ancient example as a comparison: the fall of Ur and its famous em­ pire. Such a comparison emphasises the loss of Babylon’s power and might, and the loss of Marduk’s power as a now ineffectual city god, powerless in the face of Erra’s chaos, and replaced by Išum, who represents a new heroic ideal. 30 Despite the lack of manuscripts of this Sumerian text in the first millennium, absence of evidence is not necessarily evidence of absence. The Uruk List of Kings and Sages implies that Hellenistic scholars at least saw a connection between the author of Erra and the fall of Ur, and other surviving texts show that the continuity of city laments into this later time is not so unlikely as we may have assumed. This supports the evidence inherent in the text of Erra itself: the intertextual connections that are both numerous and meaningful. The implication is that later Mesopotamians may have known much more of their own literary history than we currently recognise. What implications does this have for cuneiform scholarship more widely? In this case, intertextuality has given us an important clue about what kinds of texts may have been linked in Mesopotamian culture, links we would not otherwise have 30.  See George 2013: 62 for the differing ethics of war that Erra and Išum each represent.

520

Selena Wisnom

considered. It is another piece of evidence that we can use to reconstruct their ways of thinking. The fundamental interconnectedness of Mesopotamian scholarship means that it is particularly suited to intertextual interpretation, and links between different genres of scholarship can reveal a great deal more about their meaning. As well as pointing out that the connections are there, we can also say what difference they make to our understanding of the texts. Not only can this show us what other texts the scholars may have been reading, but it can give us a deeper insight into how they interpreted the world. Babylonian scholarship is inherently intertextual: by investigating this fully, we stand a better chance of seeing things their way.

Bibliography Barthes, R. 1977 The Death of the Author. Pp. 142–148 in Image-Music-Text 5–6. New York: Hill and Wang. Beaulieu, P. 2001 The Abduction of Ištar from the Eanna Temple. Pp. 29–40 in Proceedings of the XLVe Rencontre Assyriologique Internationale, Part 1: Historiography in the Cuneiform World, eds. T. Abusch et al. Bethesda, MD: CDL Press. Black, J. 1991 Eme-sal Cult Songs. Pp. 23–36 in Velles paraules: Ancient Near Eastern Studies in Honor of Miguel Civil on the Occasion of his Sixty-fifth Birthday, eds. P. Michalowski, P. Steinkeller, E. C. Stone, and R. L. Zettler. Barcelona: Editorial AUSA. Cagni, L. 1969 L’epopea di Erra. Studi semitici 34. Rome: Instituto di studi del vicino oriente, Università di Roma. Cavigneaux, A. 2014 Une version sumérienne de la légende d’Adapa (Textes de Tell Haddad X). ZA 104/1: 1–41. Chen, Y. S. 2013 The Primeval Flood Catastrophe. Oxford: Oxford University Press. Cohen, M. E. 1988 The Canonical Lamentations of Mesopotamia. Potomac, MD: CDL Press. Cole, S. 1994 The Crimes and Sacrileges of Nabû-šuma-iškun. ZA 84: 220–252. Cooley, J. 2008 ”I want to dim the brilliance of Šulpae!” Mesopotamian Celestial Divination and the Poem of Erra and Išum. Iraq 70: 179–188. Dahl, J. L. 2011 The Statue of Nine-iga. Pp. 67–77 in Akkade is King: A Collection of Papers by Friends and Colleagues Presented to Aage Westenholz on the Occasion of His 70th Birthday 15th of May 2009, eds. G. Barjamovic, J. Dahl, U. S. Koch, W. Som­ merfeld, J. G. Westenholz, and A. Westenholz. Leiden: Nederlands Instituut voor het Nabije Oosten. Dalley, S. 2005 The Language of Destruction and its Interpretation. BaM 36: 275–285.

Implications of Intertextuality

521

Dijk, J. van. 1962 Die Inschriftenfunde. Pp. 39–62 in XVIII. vorläufiger Bericht über die von dem Deutschen Archäologischen Institut und der Deutschen Orient-Gesellschaft aus Mitteln der Deutschen Forschungsgemeinschaft unternommenen Ausgrabungen in Uruk-Warka, ed. H. Lenzen. Abhandlungen der Deutschen Orientgesellschaft 7. Berlin: Gebr. Mann. Eco, U. 1992 Overinterpreting Texts. Pp. 45–66 in Interpretation and Overinterpretation, eds. U. Eco, R. Rorty, J. Culler, and C. Brooke-Rose. Cambridge: Cambridge University Press. Finkel, I. 1980 Bilingual Chronicle Fragments. JCS 32: 65–75. 1986 On the Series of Sidu. ZA 76: 250–253. Foster, B. R. 2005 Before the Muses: An Anthology of Akkadian Literature. Bethseda, MD: CDL Press. Foucault, M. 1977 What is an Author? Pp. 113–138 in Language, Counter-memory, Practice - Selected Essays and Interviews. Oxford: Basil Blackwell. Fowler, D. 1997 On the Shoulders of Giants: Intertextuality and Classical Studies. Materiali E Discussioni 39: 13–34. 2000 Roman Constructions: Readings in Postmodern Latin. Oxford: Oxford University Press. Frahm, E. 2010 The Latest Sumerian Proverbs. Pp. 155–184 in Opening the Tablet Box: Near Eastern Studies in Honor of Benjamin R. Foster, eds. S. C. Melville and A. L. Slotsky. Culture and History of the Ancient Near East 42. Leiden: Brill. 2011 Babylonian and Assyrian Text Commentaries: Origins of Interpretation. GMTR 5. Münster: Ugarit Verlag. Franke, S. 2014 Der Zorn Marduks, Erras und Sanheribs. Zu Datierung und Funktion von “Erra und Išum.” Pp. 315–328 in Krieg und Frieden im Alten Vorderasien. 52e Rencontre Assyriologique Internationale, International Congress of Assyriology and Near Eastern Archaeology, Münster, 17.–21. Juli 2006, eds. H. Neumann et al. AOAT 401. Münster: Ugarit-Verlag. Fuchs, A. 1994 Die Inschriften Sargons II. aus Khorsabad. Göttingen: Cuvillier Verlag. Gabbay, U. 2014 Pacifying the Hearts of the Gods: Sumerian Emesal Prayers of the First Millennium BC. Wiesbaden: Harrassowitz Verlag. 2015 The Eršema Prayers of the First Millennium BC. Heidelberg: Harrassowitz Verlag. Geller, M. 2016 Healing Magic and Evil Demons: Canonical Udug-hul Incantations. Boston; Berlin: De Gruyter. George, A. R. 2007 The Epic of Gilgamesh: Thoughts on Genre and Meaning. Pp. 37–66 in Gilgamesh and the World of Assyria. Proceedings of the Conference Held at the Mandelbaum House, the University of Sydney, 21–23 July 2004, eds. J. Azize and N. Weeks. Leuven: Peeters.

522

Selena Wisnom

2010 The Assyrian Elegy: Form and Meaning. Pp. 203–216 in Opening the Tablet Box: Near Eastern Studies in Honor of Benjamin R. Foster. Culture and History of the Ancient Near East 42. Leiden: Brill. 2013 The Poem of Erra and Ishum: A Babylonian Poet’s View of War. Pp. 39–72 in Warfare and Poetry in the Middle East, ed. H. Kennedy. London: I.B. Tauris. Glassner, J.-J. 2005 Mesopotamian Chronicles. Writings from the Ancient World 19. Atlanta: Society of Biblical Literature. Gössmann, F. 1955 Das Era-Epos. Würzburg: Augustinus Verlag. Green, M. W. 1975 Eridu in Sumerian Literature. Ph.D. dissertation, University of Chicago. 1984 The Uruk Lament. JAOS 104/2: 253–279. Guinan, A. 2002 A Severed Head Laughed: Stories of Divinatory Interpretation. Pp. 7–40 in Magic and Divination in the Ancient World, eds. L. Ciraolo and J. Seidel. Ancient Magic and Divination 2. Leiden; Boston: Brill-Styx. Heath, M. 2002 Interpreting Classical Texts. London: Duckworth. Helle, S. 2018 The Role of Authors in the “Uruk List of Kings and Sages”: Canonization and Cultural Contact. JNES 77/2: 219–234. Hinds, S. 1998 Allusion and Intertext: Dynamics of Appropriation in Roman Poetry. Cambridge: Cambridge University Press. Jacobsen, T. 1939 The Sumerian King List. AS 11. Chicago: University of Chicago Press. 1941 Review of Lamentation over the Destruction of Ur by S. N. Kramer. AJSL 58/2: 219–224. Katz, D. 2011 Reconstructing Babylon: Recycling Mythological Traditions Towards a New Theology. Pp. 123–134 in Babylon: Wissenskultur in Orient und Okzident, eds. E. Cancik-Kirschbaum, M. van Ess, and J. Marzahn. Berlin: de Gruyter. Koch-Westenholz, U. 2000 Babylonian Liver Omens: The Chapters Manzāzu, Padānu, and Pān Takālti of the Babylonian Extispicy Series Mainly from Aššurbanipal’s Library. Cophenhagen: The Carsten Niebuhr Institute of Near Eastern Studies, University of Copen­ hagen; Museum Tusculanum Press. Kristeva, J. 1980 Desire in Language. A Semiotic Approach to Literature and Art. New York: Co­ lumbia University Press. Lambert, W. G. 1958 Review of Gössmann “Das Era-Epos.” AfO 18: 395–401. 1962 A Catalogue of Texts and Authors. JCS 16/3: 59–77. 1973 Studies in Nergal. BiOr 30: 355–362. 1983 A Neo-Babylonian Tammuz Lament. JAOS 103/1: 211–215. 1986 Ninurta Mythology in the Babylonian Epic of Creation. Pp. 55–60 in Keilschriftliche Literaturen. Ausgewählte Vorträge der XXXII. Rencontre assyriologique internationale Münster, 8.–12.7.1985, eds. K. Hecker and W. Sommerfeld. BBVO 6. Berlin: D. Reimer. 2013 Babylonian Creation Myths. Mesopotamian Civilizations 16. Winona Lake, IN: Eisenbrauns.

Implications of Intertextuality

523

Lambert, W. G. and Millard, A. R. 1969 Atra-ḫasīs: The Babylonian Story of the Flood. Oxford: Clarendon. Leichty, E. 1970 The Omen Series Šumma izbu. TCS 4. Locust Valley, NY: J. J. Augustin. Lenzi, A. 2008 The Uruk List of Kings and Sage and Late Mesopotamian Scholarship. JANER 8/2: 137–169. 2015 The Language of Akkadian Prayers in Ludlul and its Significance within and beyond Mesopotamia. Pp. 67–106 in Mesopotamia in the Ancient World: Impact, Continuities, Parallels. Proceedings of the Seventh Symposium of the Melammu Project Held in Obergurgl, Austria, November 4–8, eds. R. Rollinger and E. Van Dongen. Münster: Ugarit Verlag. Lindenberger, J. M. 1983 The Aramaic Proverbs of Ahiqar. Baltimore: Johns Hopkins University Press. Livingstone, A. 1989 Court Poetry and Literary Miscellanea. SAA 3. Helsinki: Helsinki University Press. Löhnert, A. 2009 “Wie die Sonne tritt heraus!”: Eine Klage zum Auszug Enlils mit einer Untersuch­ ung zu Komposition und Tradition sumerischer Klagelieder in altbabylonischer Zeit. AOAT 385. Münster: Ugarit Verlag. Lyne, R. O. A. M. 1994 Vergil’s Aeneid: Subversion by Intertextuality. Greece and Rome 41: 187–204. Machinist, P. 2005 Order and Disorder: Some Mesopotamian Reflections. Pp. 31–61 in Genesis and Regeneration: Essays on Conceptions of Origins, ed. S. Shaked. Jerusalem: Israel Academy of Sciences and Humanities. Michalowski, P. 1989 The Lamentation over the Destruction of Sumer and Ur. MC 1. Winona Lake, Indiana: Eisenbrauns. Müller, G. G. W. 1995 Wer spricht? Betrachtungen zu “Išum und Erra.” Pp. 349–360 in Vom Alten Orient zum Alten Testament. Festschrift für Wolfram Freiherrn von Soden zum 85. Geburtstag am 19. Juni 1993, eds. O. Loretz and M. Dietrich. AOAT 240. Neukirchen-Vluyn: Neukirchener Verlag. Oppenheim, A. L. 1974 A Babylonian Diviner’s Manual. JNES 33/2: 197–220. 1977 Ancient Mesopotamia: Portrait of a Dead Civilization (Revised ed.). Chicago: University of Chicago Press. Oshima, T. 2013 The Babylonian Theodicy. SAACT 9. Winona Lake, Indiana: Eisenbrauns. 2014 Babylonian Poems of Pious Sufferers: Ludlul Bēl Nēmeqi and the Babylonian Theodicy. Orientalische Religionen in der Antike 14. Tübingen: Mohr Siebeck. Pinches, T. G. 1901 Assyriological Gleanings II. PSBA 23: 188–210. Pucci, P. 1987 Odysseus Polutropos: Intertextual Readings in the Odyssey and Iliad. Ithaca: Cornell University Press. Schaudig, H. fth. Explaining Disaster: Tradition and Transformation of the “Catastrophe of IbbiSîn” in Babylonian Literature. Dubsar 13. Münster, Zaphon.

524

Selena Wisnom

Seri, A. 2014 Borrowings to Create Anew: Intertextuality in the Babylonian Poem of “Creation” (Enūma Eliš). JAOS 134/1: 89–106. Soden, W. von. 1971 Etemenanki vor Asarhaddon. UF 3: 253–263. Streck, M. P. 2004 Dattelpalme und Tamariske in Mesopotamien nach dem akkadischen Streitge­ spräch. ZA 94/2: 250–290. Thompson, R. C. 1930 The Epic of Gilgamish: Text, Transliteration, and Notes. Oxford: Oxford Univer­ sity Press. Tinney, S. 1996 The Nippur Lament: Royal Rhetoric and Divine Legitimation in the Reign of Išme-Dagan of Isin (1953–1935 B.C.). OPSNKF 16. Philadelphia: Samuel Noah Kramer Fund. Vanstiphout, H. 1986 Some Thoughts on Genre. Pp. 1–17 in Keilschriftliche Literaturen. Ausgewählte Vorträge der XXXII. Rencontre Assyriologique Internationale Münster, 8.–12.7.1985, eds. K. Hecker and W. Sommerfeld. BBVO 6. Berlin: D. Reimer. Wassermann, N. 2003 A Forgotten Old-Babylonian Lament over a City’s Destruction: UET 6/2, 403 and its Possible Literary Content (in Hebrew). Eretz-Israel 27: 127–132. Wassermann, N. and Gabbay, U. 2005 Literatures in Contact: The balaĝ úru àm-ma-ir-ra-bi and its Akkadian Transla­ tion UET 6/2, 403. JCS 57: 69–83. Weiershäuser, F. 2010 Weiser Išum, der du den Götern vorangehst. Pp. 351–376 in Von Göttern und Menschen, Beiträge zu Literatur und Geschichte des Alten Orients. Festschrift für Brigitte Groneberg, eds. D. Shehata, F. Weiershäuser, and K. V Zand. CM 41. Leiden: Brill. Weissert, E. 1997 Creating a Political Climate: Literary Allusions to Enūma Eliš in Sennacherib’s Account of the Battle of Halule. Pp. 191–202 in Assyrien im Wandel der Zeiten. XXXIXe Rencontre Assyriologique Internationale, Heidelberg 6.–10. Juli 1992, eds. H. Waetzoldt and H. Hauptmann. Heidelberger Studien zum Alten Orient 6. Heidelberg: Heidelberger Orientverlag. Wimsatt, W. K. and Beardsley, M. C. 1946 The Intentional Fallacy. Sewanee Review 54: 468–488.

Was the Karzida of Ur’s Akītu Festival at Tell Sakhariya? Paul Zimansky

Stony Brook University In the canonical collection of forty temple hymns purportedly composed by the high priestess Enheduanna, five of the compositions are devoted to temples in and around Ur. The last of this sub-group, Number 12, praises the temple of Nanna in Ga’eš as the Karzida, or “good wharf.” 1 This particular hymn’s Ur III context rules out a direct connection with Enheduanna, but does not challenge the Karzida’s ranking among the most resplendent temples of the Sumero-Akkadian world of the late third millennium. Shrine, great sanctuary?, founded at a cattle-pen, ‘Small’ city, . . . . . . of Suen Karzida, your interior is a . . . . place, your foundation is holy and clean, Shrine, your Gipar is founded in purity, Your door is (of) strong copper, set up at a great place, Cattle-pen (filled with) the lowing (of the cows), like a young bull you . . . the horn, Your prince, the lord of heaven, standing in the . . . . ., At noon (like the sun) radiating . . . . . . . . . ., O Karzida, he, Ašimbabbar, has placed the house upon your . . . . has taken his place on your dais. The house of Nanna in Gaeš. 2

The Karzida’s primary claim to distinction is that it was locus of the earliest recorded Akītu house and played a key role in the Akītu festival of Nanna, the most important, greatest, and longest festival at Ur in the time of the Third Dynasty (Sallaberger 1993: 183). There were actually two festivals so named, a lesser one associated with the vernal equinox and the new year; and the greater one which gave its name to the seventh month and was associated with the autumnal equinox. Akītu festivals would later be celebrated at other sites in honor of other gods, but the moon god’s Akītu was unparalleled when Ur reigned supreme at the end of the third millennium.

1.  Enheduanna is currently very much alive in the popular press as the first named author of a literary work. A collection of her hymns for the non-specialist is available in Meador 2010. 2.  Sjöberg and Bergmann 1969: 26.

525

526

Paul Zimansky

The seminal discussion of this topic was Adam Falkenstein’s article, “Akiti-Fest und Akiti-Haus” published in the Friedrich Festschrift in 1959. The subject has subsequently been elaborated upon by both Sallaberger (1993) and Cohen (1993; 2015) in their studies of Mesopotamian cultic calendars. At the beginning of the festivities, the moon god left his sanctuary in Ur and moved by barge through the countryside to the Karzida, where various sacrifices and rituals were performed for him over the course of several days. He then returned with great celebration to Ur. Allusions to structures and other material associations of the Karzida and Ga’eš are scattered through cuneiform sources of the Ur III period. 3 The physical relationship between the Temple of Nanna, the Karzida, the Akītu house, and Ga’eš is not entirely clear and these all may be components of a single large complex. The temple of Nanna was probably the original core, and a priestess of Nanna separate from the well-known line following Enheduanna in Ur must have presided there. In his ninth year, Amar-Suen recorded the creation of a Gipar, a residence for this priestess, which he said had not existed there previously. A palace in which the king took up residence during the festivals is also attested. The Akītu house, where the god stayed had a courtyard of some sort where rituals were performed before a podium or barag, and storehouses for receiving local produce. This important cultic site with all its trappings clearly lay in the countryside near Ur, but its precise location is unknown. In 1994 François Carroué, discussing Sumerian geography relevant to the Iturungal watercourse, published a sketch map tentatively locating Ga’eš a little more than six kilometers east-northeast of Ur (Carroué 1993: 63). Since no ancient source gives specific information on the distance or direction from Ur other than that the Karzida was located on a canal within a day’s journey this is only a guess, and indeed Carroué marked it as such with a question mark. By coincidence, however, it is remarkably close to Tell Sakhariya (30º 58’ 33.84” N by 46º 08’ 28.36” E, elevation approximately five meters above sea level), 4 where Elizabeth Stone and I conducted a brief field project in the winter of 2011–2012 (Fig 1). The argument of this paper is that the archaeological evidence we uncovered there is indeed consistent with locating the Karzida at Tell Sakhariya. We were not, at the time, looking for anything like the Karzida and it is only subsequently that we have come the conclusion that Ga’eš and Tell Sakhariya are probably one and the same. Two years ago, we presented a paper at the 9th International Conference on Near Eastern Archaeology in Basel, suggesting this rather tentatively 5 assuming there would be more excavations in the future to either solidify or undermine our argument. It is now clear, however, that these are not going to take place, at least in our professional lifetime. A prison to which the Iraqi government consigns its most dangerous malefactors has expanded to block access to the site, and we have now turned our attention to excavating at Ur itself. For this reason, and because the topic is of direct relevance to the theme of this Rencontre,

3.  For a list of references, see Edzard and Farber 1974: 50–51. 4.  This locational information is taken from Google Earth. In the image publicly available as I write, dated to June 21, 2010, Tell Sakhariya is barely perceptible. This image predates the ones we used at the time of our excavations, presumably substituted to avoid providing sensitive details on the area of the prison, which has expanded significantly in recent years.

Was the Karzida of Ur’s Akītu Festivalat Tell Sakhariya?

527

Fig. 1. Google Earth image of Ur and Tell Sakhariya, captured June 21, 2010, before our project was initiated. The prison to the north of Sakhariya has expanded considerably in recent years.

I thought it might be worth revisiting the arguments, and reconsider their merits now that we have had more time to think about them. An explanation of how we came to work here is in order. In 2011 Elizabeth Stone and I were looking for a village whose archaeological remains we might compare with the urban assemblage at Ur. The idea was that we might learn something about settlement in the early second millennium by comparing house forms and their artifactual inventories from a smaller site with their analogs from the well-known urban neighborhoods Woolley had excavated ninety years ago. In July we travelled to Nasiriya with Abdulamir Hamdani (Al-Dafer), 6 where we were welcomed by the Deputy Governor of Dhi Qar Province, who did much to facilitate our survey. Our first candidate for a village site, chosen on the basis of satellite photos and surface ceramics as reported to us, was the small mound of Madhthuriya, seven kilometers to the northwest of Ur. There we found no Old Babylonian pottery and some much later sherds on the surface. We also realized the logistics of an expedition would be difficult here and immediately started looking for another candidate, using Henry Wright’s survey of the early 1960s as our guide (Wright 1981). On paper, Tell Abu Barura Shimal—“Father of Sheep Droppings, North”—did not look any more prom­ ising than its name would suggest. Wright had assigned it to the Kassite period 6.  Abdulamir has been instrumental in facilitating our work in southern Iraq both in this project and in subsequent expeditions. At this point in his career he was on leave from his position in the Nasiriyah Museum and enrolled in the Interdisciplinary Program in Anthropological Sciences at Stony Brook University, from which he received his PhD in 2016.

528

Paul Zimansky

and noted a “recent Qual’a ruin” at its summit (Wright 1981: 339, catalog no. 20). However, a brief visit to this site on July 7 suggested that it might be what we were looking for, and a second, longer one on July 11 after we had failed to discover any other accessible Old Babylonian village sites in the immediate environs of Ur, deepened our enthusiasm. Granting that the edges of the tell were not easy to define precisely given its shallow slope, the size was in the appropriate range for a village: 170 by 120 meters by Wright’s calculation and somewhat larger, roughly 250 by 200 meters in our total station readings. Its two highest areas, separated by a shallow saddle, were about three-and-a-half meters above plain level. Although classic Kassite pottery was indeed found on the summit, many Old Babylonian sherds were also in evidence, particularly around the edges of the mound. Wright’s “Qual’a” ruin was nowhere to be seen, although there were some trenches and metal debris from recent military activity. Most significantly, on the first visit Abdulamir discovered a large inscribed clay nail (Fig. 2), which we were later to learn dated to Rim-Sin year 15 (Fig. 3). 7 On our return to the United States, Elizabeth Stone studied satellite imagery and identified faint traces of a square, fort-like building on the mound’s summit, so we assumed that excavations would first uncover a late Kassite occupation and below that there would be remains of a larger occupation dating to the early second millennium. Support from the State Board of Antiquities and Heritage, for which we are most grateful, enabled us to secure an excavation permit, and we were able to conduct a five-week field season over the winter academic recess, beginning on December 15, 2011. As Wright’s name for the tell, which was no longer known locally, was an embarrassment, we changed it “Sakhariya.” 8 We began work with a surface survey collecting objects and diagnostic ceramics. These confirmed our initial impression that the site dated to both Old

Fig. 2. Abdulamir Hamdani at Tell Sakhariya with Rim-Sin clay nail at time of discovery, 7 July 2011.

Fig. 3. Clay nail of Rim Sin, surface find at Tell Sakhariya.

7.  I would like to thank Stephanie Dalley for reading what could be read on this nail from our photos when we returned to Oxford later that summer. 8.  Participants in the expedition included Elizabeth Stone and Paul Zimansky, co-directors, Irene Winter, Stephanie Rost, John MacGinnis, Demetrious Brellas, and Abdulamir Hamdani. Saleem Khalaf (Director of Research SBAH), Ali Khadhem (Director of Antiquities for Dhi Qar Governorate), and Wasa Abdulsahib of the Nasiriyah Museum were the Iraqi representatives for the fieldwork. The project was funded by the National Geographic Society and a private donor.

Was the Karzida of Ur’s Akītu Festivalat Tell Sakhariya?

529

Babylonian and Kassite periods, and in addition we found Ur III materials. The artifacts found in this survey were generally unexciting and showed insignificant patterns. We did, however, find fragments of four stamped bricks. These inscriptions were too incomplete to assign to a specific ruler, but with the Rim-Sin nail found the summer before, collectively affirmed erstwhile royal interest in the site. Our magnetic gradiometry survey Fig. 4. Ur Ziggurat from Tell Sakhariya. (Fig. 4) failed to detect any sub-surface architecture, and we gave up on it after a few days. Once we started digging it became clear that there was little architecture to detect. Tell Sakhariya is the most peculiar archaeological site I have encountered in a career of more than forty years of excavating in and around Mesopotamia. Most of its mass, as it survives today, is made up hard packed clean clay. We encountered this quite close to the modern surface, and never found the bottom of it, despite coring down four meters to a depth of about one meter below plain level. The conclusion that this was an artificial mound seems inescapable, and the complete absence of sherds, bones, or other living debris suggests this packing was meant to be ritually pure. Just enough ceramics and cuneiform fragments are found on and beside this core to suggest that the site was inhabited at some minimal level from time to time in the third and second millennia BCE. There was, to be sure, some stratigraphy near the surface, generally in the form of thin layers of late Kassite ceramics and debris left by somewhat earlier tran­s­ ient populations who perhaps camped on the site intermittently in the mid-second millennium (Fig. 5). There were also late Kassite burials. One neonate was buried in a jar with a typical late Kassite goblet base used as its stopper. In another area we found a shallow a burial consisting only of a skull and a tibia accompanied by three late Kassite vessels. Also prob­ ably late Kassite were two large kilns, the best preserved of these was cut into the ground with access from the above. The trans­ient populations left a series of ashy Fig. 5. Contour plan of Sakhariya with areas of soundings marked. pits–one of which

530

Paul Zimansky

contained an entire cow. They also appear to have made use of baked bricks robbed from earlier structures, but we found no substantial buildings. The object inventory of these superficial levels was quite mundane. There were a number of items in stone—grind­ing stones, cuboids, and the like—including one well-made balance weight. All the spindle whorls were simple pierced sherds, rather than the conical stones in vogue at Mashkan-shapir and other Old Babylonian urban sites. Sakhariya produced nearly as many chipped stone tools—including a number of cores—as it did fragments of bronze. Sakhariya is unusual for a second millennium Mesopotamian site in the near absence of terracotta figurines and plaques. Items of personal adornment were comparatively common, as one would expect in a mobile population. We found three very small pieces of gold and a group of even smaller silver objects was recovered through flotation as we looked for floral and faunal remains. Floral and faunal evidence from these late phases suggests a summer occupation by a marsh-adapted population. There was a small amount of barley but lots of wild plants almost certainly derived from animal dung, and Scorpio—the raw material for reed mats and the huts made from them. Bones of sheep, goats, cattle, donkeys, pigs, and small equids are among the domesticates, and gazelles were also present. 9 All in all, it would appear that Sakhariya, in this phase, was simply a place nomads passed through seasonally, taking advantage of the high ground. This high ground, of course, came from the deliberately created mud platform, which we found almost everywhere we dug in the central part of the site. It was so hard it was almost impossible to excavate with our small picks and lacked any trace of pottery or other artifactual material. We encountered this packing at exactly the same elevation in our two initial trenches which were fifty-five meters apart (Fig. 6). In the westernmost of these (N36) we gave up after digging through a little more than a meter of this sterile packing. In trench O42 to the east, it was equally clean and hard, but we discovered a baked-brick platform within it, associated with some ceramics and animal bones. But this was an anomaly. We cored several meters below this and came up with only clean mud. The edge of this clay mass was encoun­tered a few meters further east within the next trench, but time prevented us from exploring this edge to any depth. Perhaps contemporary was a broad area of clean water-lain silt with many fish bones, some of which represented whole fish apparently trapped between layers of sediment. At the base of this material we found a well-made drain or channel of baked brick whose base was so level that we could not figure out which direction Fig. 6. Sounding O42. Workmen in center are clearing water was meant to flow through it. It Kassite levels, with kiln at right. Deep sounding on would have been quite a feat to keep a left is clean clay. 9.  For a detailed study of the faunal remains, see Twiss 2017.

Was the Karzida of Ur’s Akītu Festivalat Tell Sakhariya?

531

large fish pond full of water at two meters above plain level, especially considering the aridity and high temperatures of Iraqi summers, but somebody seems to have been doing it. The inscriptions we found in excavation were consistent with what we found on the surface in dating to the Ur III and Old Babylonian periods, although none was found in its original context. These included a complete Ur-Namma brick inscription describing the construction of a barag—a pedestal or podium—and a garden. Two clay cones were found in eroded material near the edges of the mound. We also found two Sumerian tablets among the early Kassite materials along with re-used old Babylonian bricks. One of these is very poorly preserved, and doesn’t tell us much. The other is a quite well-preserved receipt of copper utensils dated to year 28 of Shulgi. The reason it is so well preserved is that it was baked in antiquity and came from inside a late Old Babylonian or Kassite drinking vessel. In general, these inscriptions are not what one would expect to find in domestic contexts and are prima facie evidence of royal building activity. The royal inscriptions, clay platforms—even the putative fish pond—all indicate that Sakhariya was an important place in the Ur III to Isin-Larsa periods. There is simply too much of this material to support any notions that it might have been brought here sporadically from somewhere else, by design or chance. There are baked bricks of the sizes that one would expect in an Isin-Larsa site, but no surviving Isin-Larsa structures and no living debris to indicate that a substantial population lived here in this or previous periods. It would seem that Sakhariya was once a much taller mound covering a smaller area. After the Isin-Larsa period, the upper parts of the mound eroded. Indeed, we find sloping layers of material washed down from above in our smaller trenches around the edges of the site. With such a modest and ambiguous inventory of finds, there is obviously no iron-clad case to be made that Tell Sakhariya was the location of the Karzida, but I do feel that the combination of what we found and what we did not find is strongly suggestive. Let me review the major points on location, chronol­ ogy, and character of the material evidence. That Ga’eš was near Ur has not been disputed in recent literature and no other tell that we could find in the vicinity fit into the range of its assumed size and date as well as Sakhariya. From the summit of Tell Sakhariya the ziggurat at Ur is clearly visible 6.45 km away to the west-southwest. It lies in an agricultural zone, crisscrossed by scars of ancient and modern canals in an area that could be easily reached by boat from Ur Textual evidence for the Karzida is heavily weighted toward Ur III, but shows the site remained important for cultic purposes until the Old Babylonian Period. The last specific reference to Ga’eš is on a cone in which Sin-iddinam claims to have been born there. The archaeological remains at Tell Sakhariya, specifically the fragmentary inscriptions, support its importance to royalty in the Ur III and Isin-Larsa periods, and show no trace of mundane domestic occupation and burials. Afterwards, in the Sealand and Kassite periods, the circumstances are reversed: there is no evidence of royal or cultic interest and the mound was used for transient residence, scattered burials, and ceramic manufacture. In short, the site shifts from a pure place to a decidedly impure one at roughly the same time that Ga’eš disap­ pears as a sacred location.

532

Paul Zimansky

The clean clay packing that made up the bulk of the mound would make sense for ritually pure foundations for sacred buildings, even if the buildings themselves are gone. They certainly don’t make sense for a tell of the usual kind, where living debris of villagers would create a very different stratigraphic record. Not many people lived here in the Ur III period, but somebody put a lot of effort into creating a mound in this rural area. When the Karzida ceased to be the home of the Akītu festival and its public structures were gone, the site remained as high ground in a thinly populated marshy environment. Those who made use of it showed little appreciation for the erstwhile glories of Nanna’s rural abode. In some ways, I am not particularly enthusiastic about identifying Tell Sakhariya with the Karzida. Accepting my arguments leads to a depressing conclusion: we are never going to excavate the splendid temple to which the hymn refers. It may have survived the destruction of Ur in the famous lament, but in the centuries that followed, the Karzida eroded away. I hope I am wrong.

Bibliography Carroué, F. 1993 Etudes de Géographie et de Topographic Sumériennes III. L’Iturungal et le Sud Sumérien. ASJ 15: 11–69. Cohen, M. 1993 The Cultic Calendars of the Ancient Near East. Bethesda, MD: CDL Press. 2015 Festivals and Calendars of the Ancient Near East. Bethesda, MD: CDL Press. Edzard, D. O. and Farber, G. 1974 Repertoire géographique des texts cuneiforms, 2: Die Orts- und Gewässername der Zeit der 3. Dynastie von Ur. Wiesbaden: Harrassowitz. Falkenstein, A. 1959 Akiti-Fest und Akiti-Festhaus. Pp. 147–182 in Festschrift Johannes Friedrich zum 65. Geburtstag am 27 August 1958 gewidmet, eds. R. von Kienle, et al. Heidelberg: C. Winter. Meador, B. 2010 Princess, Priestess, Poet: The Sumerian Temple Hymns of Enheduanna. Austin: University of Texas Press. Sallaberger, W. 1993 Der Kultische Kalender der Ur III-Zeit Teil I. UAVA 7/1. Berlin: Walter de Gruyter. Sjöberg, Å.W. and Bergmann, E. 1969 The Collection of the Sumerian Temple Hymns. TCS 3. Locust Valley, NY: J.J. Augustin. Twiss, K. T. 2017 Animals of the Sealands: Ceremonial Activities in the Southern Mesopotamian “Dark Age.” Iraq 79: 257–267. Wright, H. 1981 The Southern Margins of Sumer: Archaeological Survey and the Area of Eridu and Ur. Pp. 323–345 in Heartland of Cities: Surveys of Ancient Settlement and Land Use on the Central Floodplain of the Euphrates, ed. R. McCormick Adams. Chicago: University of Chicago Press. Zimansky, P. and Stone, E. 2016 Tell Sakhariya and Gaeš. Pp. 57–66 in Proceedings of the 9th International Congress on the Archaeology of the Ancient Near East (ICAANE), Basel 2014, Vol. 3. Wiesbaden: Otto Harrassowitz.