The First Ninety Years: A Sumerian Celebration in Honor of Miguel Civil 9781501503696, 9781501511738

Citation preview

The First Ninety Years

Studies in Ancient Near Eastern Records

General Editor: Gonzalo Rubio Editors: Nicole Brisch, Petra Goedegebuure, Markus Hilgert, Amélie Kuhrt, Peter Machinist, Piotr Michalowski, Cécile Michel, Beate Pongratz-Leisten, D. T. Potts, Kim Ryholt

Volume 12

The First Ninety Years A Sumerian Celebration in Honor of Miguel Civil

Edited by Lluís Feliu, Fumi Karahashi and Gonzalo Rubio

ISBN 978-1-5015-1173-8 e-ISBN (PDF) 978-1-5015-0369-6 e-ISBN (EPUB) 978-1-5015-0367-2 ISSN 2161-4415 Library of Congress Cataloging-in-Publication Data A CIP catalog record for this book has been applied for at the Library of Congress. Bibliographic information published by the Deutsche Nationalbibliothek The Deutsche Nationalbibliothek lists this publication in the Deutsche Nationalbibliografie; detailed bibliographic data are available on the Internet at http://dnb.dnb.de. © 2017 Walter de Gruyter Inc., Boston/Berlin Typesetting: Meta Systems Publishing & Printservices GmbH, Wustermark Printing and binding: CPI books GmbH, Leck ♾ Printed on acid-free paper Printed in Germany www.degruyter.com

E tants llibres són estats fets e compilats de gestes e històries antigues, que no seria suficient l’enteniment humà compendre e retenir aquelles. Joanot Martorell, Tirant lo Blanc, Pròleg.

Miguel Civil in his office at the Oriental Institute.

MS 3370, courtesy of the Schøyen Collection, Oslo and London

1 2 3 4 5 6

lú dub-sar nu-da-ki-a eme-gi giš-ta nu-gá-gá-e ša tup-šar-r[u]-tám ˹la˺i-ra-˹mu˺ ˹a˺-na šu-me-ri-im uz-nam ú-la i-š[a-k]a-an

One who does not love the scribal art will not pay attention to Sumerian.

Andrew George

Contents Andrew George, Edition of MS 3370 Preface

viii

xiii

Bibliography of Miguel Civil (1991–2016) Tabula Gratulatoria

xvii

xxi

Nicole Brisch “As I write this letter to you”: An Attempt at the Letter from Nanna-manšum 1 to Rīm-Sîn? Antoine Cavigneaux À tâtons dans le noir : À la recherche du sens de ḫabātum Jerrold S. Cooper “Enlil and Namzitara” Reconsidered

12

37

M. Erica Couto-Ferreira ‘Let me be your canal’: some thoughts on agricultural landscape and female 54 bodies in Sumero-Akkadian sources Franco D’Agostino The Eridu Project (AMEr) and a Singular Brick-Inscription of Amar-Suena 70 from Abū Šahrain Paul Delnero The Silences of the Scribes, Pt. II: An Unfinished Enlil Lament 80 from Nippur Gertrud Farber A New Seal of the Scribe Ur-Nanše from Girsu Lluís Feliu and Adelina Millet Albà 110 Texts and Workers

103

x

Contents

Agnès Garcia-Ventura Emesal studies today: a preliminary assessment

145

Fumi Karahashi Royal Nurses and Midwives in Presargonic Lagaš Texts

159

Jacob Klein The Hymn to the Ekur: Its Literary Structure and Cultic Background

172

Bertrand Lafont Game of Thrones: the Years when Šu-Sin Succeeded Amar-Suen 189 in the Kingdom of Ur Piotr Michalowski Literary Journeys from Babylonia to Assyria: Second Millennium Copies 205 of a Bilingual Poem Concerning Ninurta Manuel Molina and Piotr Steinkeller New Data on Garšana and the Border Zone between Umma 231 and Girsu/Lagaš David I. Owen A New Silver Balanced Merchant Account from Umma

250

Jeremiah Peterson A Middle Babylonian Sumerian Fragment of the Adapa Myth from Nippur and an Overview of the Middle Babylonian Sumerian Literary Corpus at 262 Nippur Gonzalo Rubio Sumerian Temples and Arabian Horses: On Sumerian e2-gal

284

Gebhard J. Selz To Carry Coals to Newcastle or Observations Concerning Sign Formation 300 as an Early Concept in Mesopotamian “Commentaries” Marcel Sigrist, Uri Gabbay and Mark Avila Cuneiform Tablets and Other Inscribed Objects from Collections 311 in Jerusalem

Contents

Claudia E. Suter On Images, Visibility, and Agency of Early Mesopotamian Royal Women 337 Niek Veldhuis Words and Grammar: Two Old Babylonian Lists

363

Lorenzo Verderame The Seven Attendants of Hendursaĝa: A study of animal symbolism 396 in Mesopotamian cultures Christopher Woods The Abacus in Mesopotamia: Considerations from a Comparative 416 Perspective Index of texts and compositions discussed Index of words and terms discussed Index of personal names discussed

487 492

479

xi

Preface This volume intends to be not just another Festschrift in the traditional manner of anniversary volumes. It is true that its publication will coincide with Miguel Civil’s birthday. It is also true that the contributors and the editors want to celebrate Civil’s invaluable contributions to Sumerology and his kindness towards his colleagues.1 Nevertheless, what sets this Festschrift apart from others is that Civil’s ninetieth birthday happens to coincide with the first ninety years of Sumerology as a distinct field within Assyriology and Ancient Near Eastern studies. In many regards, the scholar and the field have become inextricably linked. In 1923, Arno Poebel published his Grundzüge der sumerische Grammatik, a work that signaled Sumerology’s coming of age and set the basis for successive approaches to the Sumerian language and its various textual corpora. Three years later, in 1926, Miguel Civil was born. That same year marks the publication of the second volume of Fossey’s Manuel d’assyriologie, which still remains the most exhaustive compilation of different shapes of cuneiform signs from all periods.2 A half century later, Borger would refer to Civil as “der beste Kenner der sumerischen Schrift.”3 For the last six decades, Civil has been a central figure in our field. His command of the lexical corpus is proverbial, as attested in the volumes he published in the series Materials for the Sumerian Lexicon and in his work as co-editor of the Chicago Assyrian Dictionary. Moreover, it is difficult to think of any other scholar, in this or any other field, whose range and depth can be compared to Civil’s, from his first publication (1960) on Sumerian medical prescriptions to his many contributions on matters as diverse as grammar, literature, agriculture, economic documents, royal inscriptions, and so forth. Civil moves back and forth, from philology to linguistics, from material culture to literature, while enriching Ancient Near Eastern studies with his intellectual passion and rigor. One cannot help but be in awe of Civil’s legendary capacity to join dispersed fragments of a tablet and to recognize different manuscripts of the same composition. Likewise, it is amazing to note that in

1 It is no accident that this is the second volume devoted to honor him: see P. Michalowski, P. Steinkeller, E.C.Stone & R. L. Zettler, Velles paraules: Ancient Near Eastern Studies in Honor of Miguel Civil on the Occasion of his Sixty-Fifth Birthday (AuOr 9. Sabadell: Ausa, 1991). 2 Charles Fossey, Manuel d’assyriologie, II: Évolution des cuneiformes (Paris: Louis Conard, 1926). 3 Rykle Borger, Assyrisch-babylonische Zeichenliste (AOAT 33. Neukirchen-Vluyn: Neukirchener, 1978), p. VII. https://doi.org/10.1515/9781501503696-205

xiv

Preface

the last quarter of a century since the publication of his previous Festschrift, Civil’s prodigious scholarly output has not only continued, but in some regard has actually increased, especially since he retired from the Oriental Institute. Recent years have seen the appearance of The Farmer’s Instructions (1994), The Early Dynastic Practical Vocabulary A: Archaic HAR-ra A (2008), The Lexical Texts in the Schøyen Collection (2010), the edition of the UrNamma law collection (2011), and the edition and study of AD-GI4, a major Sumerian composition whose very genre puzzled many scholars for years (2013). In late 2015, Civil finished his overview of lexicography for the Handbook of Ancient Mesopotamia, to be published by De Gruyter, an insightful and thoughtful distillation of a lifetime studying the subject. Although known to most by the Castilian version of his name (Miguel Civil), Miquel Civil i Desveus was born in Sabadell, near Barcelona, on May 7, 1926. After the Spanish Civil War, he and his younger brother joined the famous Abbey of Montserrat, where he received a well-rounded education and excelled in two very different subjects, Greek and Mathematics, a fact that foreshadowed his multiple intellectual interests in later life. The museum of the Abbey (Museu Bíblic de l’Abadia de Montserrat) houses a collection of over a thousand cuneiform tablets. With that treasure trove at hand and the magnificent library of the monastery, founded almost a millennium ago, Civil taught himself the basics of Akkadian, Sumerian, cuneiform, and Mesopotamian studies. In 1947, a young Civil met Nikolaus Schneider when the latter came to visit the Abbey. Civil was eager to know who was the best Assyriologist at the time, particularly after the passing of François Thureau-Dangin (1944). Schneider’s answer was rather prophetic: Benno Landsberger. Little could Civil have known then that a few years later he would become Landsberger’s assistant in Chicago. In 1955, Civil left the Abbey and moved to Paris. He was escaping the provincial nature of the cultural and political environment of Catalonia and Spain in the 1950s, but he was not yet sure what to do in life. His main choices were to pursue graduate studies, or to become a painter or a writer. In his early days in Paris, he held a number of jobs, from unloading trucks to working as a house painter and an elevator operator. He even had the more glamorous task of working for a film studio, where he was employed to hire extras. He eventually took a class with Jean Nougayrol, which decided his vocation and led him to pursue a doctoral degree at the École Pratique des Hautes Études. In addition to Nougayrol, his professors in Paris included Raymond Jestin and René Labat. Still, Civil was planning on going back to Catalonia after obtaining his degree, and had decided he might become a highschool teacher. Fortunately, he then met Samuel Noah Kramer. Kramer had

Preface

xv

tried to hire Jean Bottéro, then at the CNRS, as an assistant professor at the University of Pennsylvania, but Bottéro declined the invitation because he did not want to move to the United States. Bottéro did recommend a very promising graduate student, however, who he thought might be interested in that position. After meeting Kramer in Paris, Civil arrived in the United States in December, 1958. In Philadelphia, Civil shone because of his prodigious photographic memory, a natural gift that enabled him to join fragments of broken tablets, which were often housed at museums in different countries. In 1961, he met Landsberger while the latter was making collations in Philadelphia. Thus started a relation that would eventually lead to Civil’s move to Chicago, where he joined the faculty of the Oriental Institute as an associate professor in 1963. This happened two years before he actually received his doctorate in Paris, where he defended his dissertation, “Le débat sumérien entre la houe et l’araire,” in 1965. Civil would later return to the École Pratique as directeur d’études associé during the academic year 1969–1970. The fact that Civil ended up in Chicago is a good illustration of his pragmatic and endearing mindset. He actually had two offers to leave Philadelphia: one from Chicago and another one from Berkeley. After visiting Berkeley, he realized that life would be so fun in California that he would not be able to do much work, so he chose Chicago precisely for a climate that calls for life indoors and its inherent lack of distractions – as he says, “there are not even any mountains to ski or climb.” Moreover, after WWII, with the arrival of many scholars from Europe, Chicago had become the most important center for Near Eastern Studies. It was in Chicago that Civil’s scholarship flourished for the next half century, as epigrapher of the Nippur Expedition to Iraq, as co-editor of CAD, as the author of several MSL volumes, and even as a pioneer in the use of computers in the Humanities. In recent years, Civil has taught several courses at the University of Barcelona, which bestowed upon him an Honorary Doctorate in 2000. After over four decades of teaching and research in the United States, now a new generation of Catalan and Spanish Assyriologists has also been able to benefit from Civil’s erudition and from his engagement with the world of Ancient Mesopotamia as a whole, a scholarly endeavor that encompasses all realms of human experience, from farming to poetry. It is in the context of his ground-breaking contributions and achievements and in the light of his kindness and generosity that we contributors and editors dedicate this volume to Miguel Civil, as a sign of our unwavering gratitude and admiration. We wish to celebrate Civil’s long and fruitful career, which overlaps with the history of our very field. At ninety, Civil

xvi

Preface

remains an intellectual engine and an immeasurable force in Ancient Near Eastern studies.4 Lluís Feliu University of Barcelona [email protected]

Fumi Karahashi Chuo University, Tokyo [email protected]

Gonzalo Rubio Pennsylvania State University [email protected]

4 This introduction draws from the biographical profile by Lluís Feliu, “Miquel Civil: d’exiliat cultural a sumeròleg,” Afers: Fulls de recerca i pensament 85 (2016): 631–663.

Bibliography of Miguel Civil 5 (1991–2016) 1991

1992 1993

1994

1995 1996 1997

1998

1999

Sumerian and Akkadian Lexicography. Pp. 1682–86 in Wörterbucher − Dictionaries – Dictionnaires, 5.2, ed. Franz Josef Hausmann et al. Berlin: De Gruyter. Education. Pp. 301–305 in The Anchor Bible Dictionary, 2, ed. David Noel Freedman. New York: Doubleday. On Mesopotamian Jails and Their Lady Warden. Pp. 72–78 in The Tablet and the Scroll. Near Eastern Studies in Honor of William W. Hallo, ed. Mark E. Cohen, Daniel C. Snell & David B. Weisberg. Bethesda, Md.: CDL Press. Sumerian Poetry. Pp. 1233–1234 in The New Princeton Encyclopedia of Poetry and Poetics, ed. Alex Preminger & T. V. F. Brogan. Princeton: Princeton University Press. The Farmer’s Instructions. A Sumerian Agricultural Manual. AuOr Suppl. 5. Barcelona: Ausa. Sumerian [Linguistics]. Pp. 76–87 in History of Linguistics, I: The Eastern Traditions of Linguistics, ed. Giulio Lepschy. London: Longman. Ancient Mesopotamian Lexicography. P. 2305–14 in Civilizations of the Ancient Near East, 4, ed. Jack M. Sasson. New York: Scribner. Literary Text About Ur-Namma. AuOr 14: 163–67. Sin-iddinam in Emar and SU.A = Šimaški. N.A.B.U: 41. The Instructions of King Ur-Ninurta: a New Fragment. AuOr 15: 43–53. A new Lipit-Eštar inscription. N.A.B.U.: 98. Sumerian. Pp. 92–95 in The Oxford Encyclopedia of Archaeology in the Near East, 5, ed. Eric M. Meyers. Oxford: Oxford University Press. HAR-ra = ḫubullu: Tablet X dug = karpatu. Pp. 129–159, tb. 7–9 in Walther Sallaberger, Der babylonische Töpfer und seine Gefäße nach Urkunden altsumerischer bis altbabylonischer Zeit sowie lexikalischen und literarischen Zeugnissen, Mesopotamian History and Environement. Memoirs 3, Ghent: University of Ghent. Bilingual Teaching. Pp. 1–7 in Festschrift für Rykle Borger zu seinem 65. Geburtstag am 24. Mai 1994: tikip santakki mala bašmu …, ed. Stefan M. Maul. CM 10. Groningen: Styx. ‘Adamdun,’ the Hippopotamus, and the Crocodile. JCS 50: 11–14. Correction to N.A.B.U. 1997/98. N.A.B.U: 57. Of Reed Fences and Furrows. Pp. 259–64 in Landwirtschaft im Alten Orient. Ausgewählte Vorträge der XLI. Rencontre Assyriologique Internationale, Berlin, 4.–8. 7. 1994, ed. Horst Klengel & Johannes Renger. BBVO 18. Berlin: D. Reimer. An Ebla incantation against insomnia and the Semiticization of Sumerian. Notes on ARET 5 8b and 9. (With Gonzalo Rubio.) Or. n.s. 68: 254–66. Reading Gilgameš. AuOr 17/18: 179–189.

5 For a list of Miguel Civil’s publications prior to 1991, see his 65th-birthday Festschrift (Velles Paraules) in AuOr 9 (1991): 283–290. https://doi.org/10.1515/9781501503696-206

xviii 2000

2001

2002

2003

2004 2005

2006

2007 2008

Bibliography of Miguel Civil

Una creació mesopotàmica: els primers diccionaris. Pp. 23–30 in Solemne investidura del doctor honoris causa al professor Miquel Civil, ed. Gregorio del Olmo Lete. Barcelona: Universitat de Barcelona. http://www.ub.edu/ubtv/en/video/honoris-causa-miquel-civil From the Epistolary of the Edubba. Pp. 105–18 in Wisdom, Gods and Literature. Studies in Assyriology in Honour of W. G. Lambert, ed. Andrew R. George & Irving L. Finkel. Winona Lake, Ind.: Eisenbrauns. Modal Prefixes. ASJ 22: 29–42 (publ. 2005). El arte de escuchar voces lejanas. Pp. 11–17 in De la Estepa al Mediterráneo. Actas del Ier Congreso de Arqueología e Historia Antigua del Oriente Próximo, Barcelona, 3–5 de Abril de 2000, ed. Juan-Luis Montero Fenollós, Jordi Vidal Palomino & Felip Masó Ferrer. Monografies Eridu 1. Barcelona: Eridu. Esbós de gramàtica sumèria. Barcelona: Privately distributed. Remarques sobre Sîn-iddinam 6. AuOr 20: 245–47. The Forerunners of Marû and Ḫamṭu in Old Babylonian. Pp. 63–71 in Riches Hidden in Secret Places. Ancient Near Eastern Studies in Memory of Thorkild Jacobsen, ed. Tzvi Abusch. Winona Lake, Ind.: Eisenbrauns. Of Bows and Arrows. JCS 55: 49–54. Reading Gilgameš II: Gilgameš and Ḫuwawa. Pp. 77–86 in Literatur, Politik und Recht in Mesopotamien. Festschrift für Claus Wilcke. Edited by Walther Sallaberger, Konrad Volk & Anette Zgoll. Orientalia Biblica et Christiana 14. Wiesbaden: Harrassowitz. The Series diri = (w)atru. MSL 15. Roma: Pontificium Institutum Biblicum. Nos. 45–53: Syllabaries A and B. Pp. 213–21 in Literary and Scholastic Texts of the First Millennium B.C., ed. Ira Spar & Wilfred G. Lambert. Cuneiform Text in the Metropolitan Museum of Art, 2. New York: The Metropolitan Museum of Art. Nos. 55–60: Texts from the Series Urra = ḫubullu Pp. 230–44 in Literary and Scholastic Texts of the First Millennium B.C., ed. Ira Spar & Wilfred G. Lambert. Cuneiform Text in the Metropolitan Museum of Art, 2. New York: The Metropolitan Museum of Art. No. 61: Late Babylonian Grammatical Text. Pp. 244–47 In: in Literary and Scholastic Texts of the First Millennium B.C., ed. Ira Spar & Wilfred G. Lambert. Cuneiform Text in the Metropolitan Museum of Art, 2. New York: The Metropolitan Museum of Art. be5/pe-en-zé-er = biṣṣūru. Pp. 55–61 in If a Man Builds a Joyful House: Assyriological Studies in Honor of Erle Verdun Leichty, ed. Ann K. Guinan et al. CM 31. Leiden: Brill. The Song of the Millstone. Pp. 121–38 in Šapal tibnim mû illakū. Studies Presented to Joaquín Sanmartín on the Occasion of His 65 th Birthday, ed. Gregorio del Olmo Lete et al. AuOr Suppl. 22. Barcelona: Ausa. Early Semitic Loanwords in Sumerian. Pp. 11–33 in Studies Presented to Robert D. Biggs, June 4, 2004, ed. Martha T. Roth et al. AS 27. Chicago: The Oriental Institute. The Early Dynastic Practical Vocabulary A (Archaic HAR-ra A). ARES 3. Rome: Missione archeologica italiana in Siria. A Sumerian Connective Particle and Its Possible Semitic Counterparts. AuOr 26: 7–15. An Agricultural Account from Umma. Pp. 35–43 in On the Third Dynasty of Ur. Studies in Honor of Marcel Sigrist, ed. Piotr Michalowski. JCS Suppl. 1. Boston: ASOR.

Bibliography of Miguel Civil

2009

xix

The Mesopotamian Lexical Lists: Authors and Commentaries. Pp. 63–69 in Reconstructing a Distant Past: Ancient Near Eastern Essays in Tribute to Jorge R. Silva Castillo, ed. Diego A. Barreyra Fracaroli & Gregorio del Olmo Lete. AuOr Suppl. 25. Barcelona: Ausa. 2010 The Lexical Texts in the Schøyen Collection. CUSAS 12. Bethesda, Md.: CDL Press. Sumerian Compound Verbs: Class II. Pp. 523–33 in Language in the Ancient Near East. Proceedings of the 53e Rencontre Assyriologique Internationale 1, ed. Leonid Kogan et al. Babel und Bibel 4/1–2. Winona Lake, Ind.: Eisenbrauns. 2011 The Law Collection of Ur-Namma. Pp. 221–310 in Andrew R. George, Cuneiform Royal Inscriptions and Related Texts in the Schøyen Collection. CUSAS 17. Bethesda, Md.: CDL Press. 2013 Remarks on ad-gi4 (A.K.A. “Archaic Word List C” or “Tribute”). JCS 65: 13–68. Ur III as a Linguistic Watershed. Pp. 3–17 in From the 21 st Century b.c. to the 21 st Century a.d.: Proceedings of the International Conference on Neo-Sumerian Studies Held in Madrid, 22–24 July 2010, ed. Steven J. Garfinkle & Manuel Molina. Winona Lake, Ind.: Eisenbrauns. In Press On Some Ur III Scribal Exercises, in From Sherds to Landscapes: Studies on the Ancient Near East in Honor of McGuire Gibson, ed. Mark Altawell & Carrie Hritz. SAOC. Chicago: The Oriental Institute of the University of Chicago. Forth- Lexicography. In Handbook of Ancient Mesopotamia, I–II, ed. Gonzalo Rubio. coming Berlin: De Gruyter.

Tabula Gratulatoria Tzvi Abusch, Waltham, Mass. Guillermo Algaze, San Diego Alfonso Archi, Rome Pascal Attinger, Bern Richard E. Averbeck, Deerfield, Ill. Pedro Azara, Barcelona Thomas E. Balke, Heidelberg Richard H. Beal, Chicago Paul-Alain Beaulieu, Toronto Gary Beckman, Ann Arbor, Mich. Juan A. Belmonte Marín, Albacete Maria Giovanna Biga, Rome Robert D. Biggs, Chicago Barbara Böck, Madrid Walter Bodine, New Haven, Conn. Marco Bonechi, Rome John A. Brinkman, Chicago Giorgio Buccellati, Los Angeles Amalia Catagnoti, Florence Jordi Del Cerro, Barcelona Mark E. Cohen, Bethesda, Md. Steven W. Cole, Evaston, Ill. C. Jay Crisostomo, Ann Arbor, Mich. Esteban Cruañas, Barcelona Stephanie Dalley, Oxford Eva von Dassow, Minneapolis Fred M. Donner, Chicago Maria deJ. Ellis, Philadelphia Richard S. Ellis, Lower Merion, Pa. Mario Fales, Udine Walter Farber, Chicago Irving Finkel, London Daniel Foxvog, Berkeley Eckart Frahm, New Haven, Conn. Grant Frame, Philadelphia Pelio Fronzaroli, Florence Alhena Gadotti, Towson, Md. Steven Garfinkle, Bellingham, Wash. Markham J. Geller, Berlin McGuire Gibson, Chicago Jean-Jacques Glassner, Paris Anne Goddeeris, Leuven Petra M. Goedegebuure, Chicago Gene Gragg, Chicago Ann Guinan, Philadelphia https://doi.org/10.1515/9781501503696-207

Nils Heeßel, Würzburg Markus Hilgert, Berlin Wayne Horowitz, Jerusalem Theo van den Hout, Chicago Fabienne Huber Vulliet, Bernex, Geneva John Huehnegrad, Austin, Tex. Hermann Hunger, Vienna Bram Jagersma, Leiden Janet H. Johnson, Chicago Charles E. Jones, State College, Pa. Michael Jursa, Vienna Josué J. Justel, Alcalá de Henares Dina Katz, Leiden Karlheinz Kessler, Würzburg Alexandra Kleinerman, Ithaca, N.Y. Renee Kovacs, Corte Madera, Calif. Manfred Krebernik, Jena Theo J. H. Krispijn, Leiden Sylvie Lackenbacher, Paris Giovanni-Battista Lanfranchi, Padua Jacob Lauinger, Baltimore Erle Leichty, Philadelphia Karel van Lerberghe, Leuven Brigitte Lion, Paris Mario Liverani, Rome Jaume Llop, Berlin Remco de Maaijer, Dronten Peter Machinist, Cambridge, Mass. Tohru Maeda, Tokyo Maynard P. Maidman, Toronto Gianni Marchesi, Bologna Ignacio Márquez Rowe, Madrid Joachim Marzahn, Berlin Paolo Matthiae, Rome Stefan M. Maul, Heidelberg Vera Meyer-Laurin, Berlin Cécile Michel, Paris Marc Van De Mieroop, New York Lucio Milano, Venice Salvatore F. Monaco, Rome Wakaha Mori, Kyoto Hans Neumann, Münster Hans J. Nissen, Hainfeld Juan Carlos Oliva Mompeán, Ciudad Real Gregorio del Olmo Lete, Barcelona

xxii

Tabula Gratulatoria

Paola Paoletti, Munich Dennis G. Pardee, Chicago Simo Parpola, Helsinki Susanne Paulus, Chicago Laurie E. Pearce, Berkeley Ilan Peled, Leiden Holly Pittman, Philadelphia Simonetta Ponchia, Verona Beate Pongratz-Leisten, New York J. N. Postgate, Cambridge D. T. Potts, New York Hervé Reculeau, Chicago Clemens Reichel, Toronto Rocío Da Riva, Barcelona Francesca Rochberg, Berkeley Robert Rollinger, Innsbruck Martha T. Roth, Chicago Matthew Rutz, Providence, R.I. Walther Sallaberger, Munich Joaquín Sanmartín, Barcelona Jack M. Sasson, Chapel Hill, N.C. JoAnn Scurlock, Chicago Ursula Seidl, Munich Andrea Seri, Rosario, Argentina Tonia Sharlach, Stillwater, Okla. Daniel C. Snell, Norman, Okla. Wilfred van Soldt, Leiden Walter Sommerfeld, Marburg Ira Spar, New York Matthew W. Stolper, Chicago

Elizabeth C. Stone, Stony Brook, N.Y. Philippe Talon, Brussels Michel Tanret, Ghent Jon Taylor, London Javier Teixidor, Paris Jonathan Tenney, Ithaca, N.Y. Jeffrey Tigay, Philadelphia Steve Tinney, Philadelphia Maria Vittoria Tonietti, Florence Jordi Vidal, Barcelona Konrad Volk, Tübingen Caroline Waerzeggers, Leiden Hartmut Waetzoldt, Heidelberg Ronald Wallenfels, New York Nathan Wasserman, Jerusalem Chikako E. Watanabe, Suita, Osaka Matthew W. Waters, Eau Claire, Wis. Wilfred G. E. Watson, Northumberland John Z. Wee, Chicago Margot Stout Whiting, Helsinki Robert Whiting, Helsinki Claus Wilcke, Munich Gernot Wilhelm, Würzburg Irene Winter, Cambridge, Mass. Cornelia Wunsch, Hirschbach Shigeo Yamada, Tsukuba, Ibaraki Norman Yoffee, Ann Arbor, Mich. Annette Zgoll, Göttingen Paul Zimansky, Stony Brook, N.Y.

Nicole Brisch

“As I write this letter to you”: An Attempt at the Letter from Nanna-manšum to Rīm-Sîn? Literary letters have recently experienced renewed interest from the scholarly community. As is well known, it is hard to overestimate Miguel Civil’s contributions to Sumerian literature. As has been the case with almost every topic related to Sumerian, he has made a visionary contribution to the subject matter at hand. Regarding a group of Sumerian literary letters that belonged to the Old Babylonian school curriculum at Nippur he argued convincingly that they were not “real-life letters mentioning students and schools, but artificial didactic texts.” (Civil 2000: 107). It is a great honor to dedicate this brief contribution to him on this happy occasion, knowing full well that the honoree will be able to significantly improve the attempts presented below. Literary letters from ancient Mesopotamia are commonly divided into two groups, those written in prose, and those that are poetry (Hallo 1968; Michalowski 2011: 28), though the boundaries between these two categories can sometimes be slightly blurred (Michalowski 2011: 28–29). The famous Royal Correspondence of Ur belongs to the former category, whereas the so-called “letter-prayers” or “letters of petition” belong to the latter. Michalowski (2011: 14–34) recently discussed the history of literary letters as part of his in-depth and thought-provoking study of historical, literary, and political aspects of the royal correspondence of Ur, therefore there is no need to repeat his findings here.1 “Letter-prayers” or “letters of petition” were composed in a poetic mode of writing. For a long time the study of Sumerian (and, to an extent, Akkadian) literature has focused on the identification of genres. Though literary letters are easily identified as a generic category, the difference in writing modes has received less attention. Yet, the question of writing modes seems rather significant, especially pertaining to literary letters, and I would like to suggest that the time has come to begin deconstructing genres in Sumerian literature and

1 Kleinerman (2014) explored mainly aspects of the role that literary letters played as part of the Old Babylonian school curriculum. Acknowledgements: I would like to express my gratitude to Jakob Dahl and Klaus Wagensonner for making a photograph of this tablet available online and for directing me to it. https://doi.org/10.1515/9781501503696-001

2

Nicole Brisch

concentrate on writing modes instead. I hope to explore this issue in a larger study on Sumerian “letter-prayers” and their intertextual connections to other genres in the Sumerian and Akkadian literary corpus, so as go beyond the boundaries that are set by the “law of genre,” to borrow Derrida’s phrase (1980).2 Such a deconstruction could be useful in order to take away focus from questions of literary genres, their form and function, and instead explore commonalities and connections within the Mesopotamian literary corpus. The edition of the letter from Nannamanšum presented here has to be viewed as extremely preliminary and may turn out to be considerably incorrect in places. The tablet is housed at the Ashmolean museum and was first published by Langdon (BE 31 no. 7) and then re-published by Gurney as part of his volume on Sumerian literary texts in the Ashmolean Museum (OECT 5 = Gurney 1976). A first, partly incomplete edition was provided by Langdon (1914: 21– 25). The tablet was already fragmentary at that time (1914: 21), and its condition seems not to have improved over time, probably due to the poor quality of the clay from which the tablet was made. Michalowski (1980–83: 56) already noted certain similarities between this letter and the letter of princess Ninšatapada to Rīm-Sîn, leading him to propose that this Nannamanšum letter was probably also addressed to the same king, a suggestion that is not unlikely, though has to await confirmation if and when other manuscripts are discovered. It is unclear how long the letter was; only the obverse of the tablet is preserved, the reverse shows only faint traces of signs in the lower portion of the reverse, yet it is possible that the reverse contained a different composition. The text is quite unusual and one wishes it had been better preserved. In particular the passages towards the end of the obverse appear to make use of interesting phraseology, unfortunately often unintelligible due to the poor state of preservation. Langdon, BE 31, no. 7 = OECT 5, no. 31 (Ashm 1913–0318) Letter from Nannamanšum to Rīm-Sîn (?) Transliteration obv.  1 [lugal-mu-ra u3-na-a-dug4]  2 [d?]˹ri˺-[im-dsîn(en.zu) (?) …] x e? x [x x (x)]  3 den-lil2 x […] x gaba-˹gi4?˺ nu-mu-u[n-tuku?]  4 ˹nam?˺-nir?-[ĝal2?] x x [… dir]i?-ga-ni-a da-ri2? šu? i-ni-in-x-x

2 It is unclear, though, if such a thing is possible.

An Attempt at the Letter from Nanna-manšum to Rı ̄m-Sîn?

 5  6  7  8  9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42

3

na-d[e5] e2-kur-[ra] en gal [dn]in-urta a2-tah-a-ni šu!? i-ni-in-ĝar-ra? d˹ en-ki˺ ˹ĝeštug2˺ daĝal ki? [x] gar x ma x šu gal ˹mi?˺-[ni-]-˹du7?˺ d nanna? en gal ud-sakar? ˹an? ki?˺ bi?(-)e-ne-ne x x x x x x [x] nam-ti-la u3-[l]uh? niĝ2-zi niĝ2-˹si-sa2˺ ˹šu?˺ zi? mu-˹na?˺-[x]-˹ĝar˺-a x dingir-re-e-ne igi(-)˹kar2?˺ ˹nidba?˺-bi me-bi x […] u4-˹da˺ gub šu sikil šu ku3-ga ni2-tuku-a x x […] x [e]n3 tar-tar niĝ2-˹ak-ak˺ ˹ĝiš-hur?˺-ra ga?-na-x[…] [x] ˹bi? KA?˺ ˹hul2?-la?˺ ˹igi?!˺ mu-un-du8 ˹ĝiri3?˺ x […] x ur? x gal-lake4? [lu2?] gu2-du3-a-ni šu-ni-še3 ˹mu?˺-u[n?-…] x x-˹si?˺ ˹uĝ3?˺ erim2-˹e?˺ ˹iri?˺ ˹sa? ga?˺ x ˹BU?˺ [x (x)] ĝiri3-bi si x mu pad3-˹da˺ ˹d?nin?-urta?˺ x ĝidru? ˹nam??˺-[lugal?-la?] x x x x-an-[x] ˹ezen?˺ gal-gal gu2? x-˹ĝin7?˺ x ba-an-dim2?-e ˹inim den-lil2-la2-ta˺ ˹ki?˺ bad-ra2 igi-bi nu-ĝar-ra ˹niĝ2?˺-x x-me-ne ba ni2? ba?-an-x-(x)-ne eš3? ˹e2?-x˺-še3? še? x bala mu-ni-in-x x ba-ĝar x mi-in-il2? dumu tu-da en d[n]ergal-la ka x niĝ2-zi niĝ2-si-sa2 lugal-ĝu10-ra u3-na-a-dug4 Id nanna-ma-˹an-šum2 dub-sar˺ ˹dumu˺ x x ˹ba?˺ x lu2 uri2ki-ma-ke4 arad2-zu na-ab-be2-a ˹ad?-da?˺-ĝu10? arad2? gi-na-˹zu?˺ [g̃iš?] ˹mu?-un?˺-ši-in-tuku-a lu2 kur2-ra mu-˹ni?˺ nu-mu-˹un?˺-˹pad3?-de3?˺ x x [dingir? saĝ?]-du-gani-me-[en] u4 igi-bi igi mu-e-du8 […] x za [e]n3? tar ˹lu2˺-igi-du gal-˹an-zu?˺ […] ˹niĝ2?˺ x x ˹gi˺-me-en en3-bi tar-re-[de3?] […] mu?-ni-dug4?-ga? zu-zu-de3 saĝĝa? kala-ga mu-e-ni-x x ˹gu3?-bi?˺ ˹teš2˺-a si3-ga-de3 ˹mu?˺ x ˹šu˺ ˹ad?˺ ˹nu?˺ x ul? x x bi? mu-ni-in-e3 lu2-lirum ma-da šu mu-ni-x x (x) ˹ka?˺ a2 kala-ga zu2 mu-e-˹keš2?˺ x x ˹a2/da?˺ ru x ˹ĝal2 bi mu ĝeš-gaba?˺-a-ke4?-eš […] x ni? x ˹daddag?-ga?-ta˺ x x lu2 x x [… ur?-saĝ?] gal nam-tar-tar-re im-mi-in-x […] [… ur?]-saĝ ˹mah?-e?˺ mu?-e-da-an-ra? […] […] ˹ma?˺ x ˹bi2-in-ak?˺ ˹teš2-bi˺ mu-x-x-[…] […] ĝal2 i?-ni?-in?-lu?-ga? dumu? mah-e ˹bi2˺-x x […] x e-ne ˹mu?-e?-ni?-gar?˺ ˹saĝ?˺ ˹mah?˺-bi ˹šu˺ mi-ni-in-tu-[tu?] […] x-kin? mi? x mu nam-ti-la-ka ne niĝ2-al-di lugal-mu-ra […] ˹ba-da-an-gub?˺-be2 en-e-ne-am3 ba-ni-ib-˹gu2?-be2?˺ […] x ba-ĝen? x ˹ta?˺ ša3 gi-na-zu e-ne-bi [x x] da šu ad-da-ĝu10 šu mu-un-e3 a2-bad ˹x˺ ˹šu? zi?˺-ga? ˹im?-x?˺ [x]-zu im-ri-a ka nam-ti-zu?-še3? ˹ka?˺ im-mi-in-lah4

4

Nicole Brisch

43 44 45 46 Rev.

˹u3˺-na-a-dug4 mu-ra-ab-gub-ba ša3 lugal-ĝu10?? ˹he2-em-ma-sed??˺-e / ša3 x-gin7 x x ga-mu-na-ab-du11 lugal den-lil2-le kur-kur-ra mu-ni bi2-in-sa4?-a u3-mu-na(-)mu-x x x ra? x x x ne-zu-a me-en x x ˹nu?-mu?˺-x […] x x-˹me?-en?˺ only illegible traces in the lower portion of the tablet.

Translation  1 [To my king say:]  2 [To Rīm-Sîn (?) …]  3 Enlil …, he [has] no rival.  4 Authority (?) …, in his surpassing … forever (?) …  5 The adviser of the Ekur, the great lord Ninurta, provided his help.  6 Enki made broad wisdom … completely perfect,  7 Nanna, the great lord, the crescent of heaven and earth (?), …  8 bestowed [breath (?) and] life and the scepter of righteousness and justice on you/him.  9 the … of the gods, the …, their food offerings and their rites (?) … 10 Who is daily at service, (who) … with pure hands, with sacred hands, piously (?). 11 Who takes care of (good) deeds and plans (?), … 12 … a joyful sound (?) … 13 His enemies he? [filled?] into his hand. 14 The enemy (?) … 15 Who was named by Ninurta (?), … the scepter of kingship (?) …, 16 Who creates (?) great festivals like? … on the command of Enlil, 17 At a far away place (?), which no one can see (?), … 18 In (?) the sanctuary of … 19 to my king, the son, born of lord Nergal, the … of righteousness and justice, 20 say: 21 Nanna-manšum, the scribe, the son of …, a citizen of Ur, 22 your servant says: 23 My father, your faithful servant, who has (always) been obedient (?), 24 So that he not choose another person (?), … you are (?) his creator (?). 25 When you look at me (?) … take care (of me) (?)! 26 The foremost one, the wise one, you are … 27 To take care, to teach …, which you command (?),

An Attempt at the Letter from Nanna-manšum to Rı ̄m-Sîn?

28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 Rev. …

5

You … the mighty chief administrator (?), to make … speak with one voice, … Athlete (?), you gathered the land, which …, with a mighty arm (?). … … […] great [warrior (?)], who determines the destinies, … … … … … … years of life (?) … a request for my king (?), … you set up (?), they (?) … … … my father escaped (?), protection … Your … to (?) the family brought news on your behalf (?). As I write this letter to you, may the heart of my king calm down (?) / like a … heart let me speak … (?). The king, whom the Enlil of all the lands has named, (?) … You are (?) …

Commentary 1 The reconstruction of this line is by no means clear. It is based on the first line of the letter from Ninšatapada to Rīm-Sîn (for the latest edition see Brisch 2007: 246–61 with further references), but frequently such restorations turn out to have been incorrect as we do not always credit ancient scribes with having been inventive. 2 The reconstruction of Rīm-Sîn’s name here is also tentative and an educated guess based on line 19 (see below). The remainder of this line is too broken for a reconstruction. 3 It is unclear whether this line consists of one or two phrases; if Enlil is the subject of the finite verb at the end, one could imagine the verb to have been šum2 “to give” and one could translate something like “Enlil … gave you no rival. Alternatively, there could be two phrases in this line, the first one describing something that Enlil bestowed upon the king, the second describing that Rīm-Sîn had no rival. A restoration with the verb tuku is overall more likely. The topos of

6

4

5

6

8

9

Nicole Brisch

“having no rival” is a well-established epithet in royal and divine hymnography. The appearance of a finite verbal form here in the hymnic, introductory part of the letter is unusual. Langdon’s copy (1914, pl. 8) shows more signs in the beginning of the line, though whether the copy is reliable is unclear. If Langdon copied the signs accurately, a restoration to nam-nir-ĝal2 would be impossible here. According to Langdon’s copy one would have to read namnir nam-du10 ša3 … in the beginning of the line. The sign ni after i in the verbal form is clearly recognizable on the photograph of the tablet now on CDLI: http://www.cdli.ucla.edu/ P345812. The epithet na-de5 e2-kur-ra “adviser of the Ekur” occurs in connection with Nuska in the hymn Išme-Dagan Q segment A line 10 (Sjöberg 1973, no. 7 line 10 = Black et al 1998–2006, ETCSL 2.5.4.17) and in the Temple Hymns line 51 (Sjöberg and Bergmann 1969: 19). It is also an epithet of (most likely) Ninurta in the royal hymn Būr-Sîn A line 8, an adab of Ninurta that is preserved on a Sammeltafel (Sjöberg 1976: 418–419 line 124 = Black et al. 1998–2006, ETCSL 2.5.7.1 line 8). According to Langdon’s copy the beginning of the line reads den-ki. The signs can be recognized neither in Gurney’s copy nor in the recent photograph. Based on Gurney’s copy the signs following the break could be read niĝ2-nam!-ma, though the photograph does not support such a reading. The first sign in this line is difficult to reconstruct. One could suggest a restoration as ĝišgu-za “throne” after Rīm-Sîn C l. 19 (ĝišgu-za namti-la-ka gal-bi hu-mu-ra-ab-dur2 “On the throne of life he shall seat you grandly” [Brisch 2007: 201]), though there is not enough space in the beginning of this line for such a restoration and it would not fit well with the remainder of the line. Langdon’s copy shows the sign nam followed by a question mark, though nam nam-ti-la would be strange. Another possibility could be to restore zi nam-ti-la “breath and life” here (compare the royal inscription dedicated on behalf of Ammiṣaduqa lines 16–17: zi nam-ti-la in-na-ba-a “(and) granted him breath and life” [Frayne 1990: 429 = RIME 4.3. 10. 2001]). The remnants of the sign after mu in the line drawing in OECT 5, 31 look like na, though one would expect a second person dative prefix ra here. The other letter of petition to Rīm-Sîn, written by Ninšatapada, also uses second person forms sparingly. I cannot make any suggestions for the first sign of this line.

An Attempt at the Letter from Nanna-manšum to Rı ̄m-Sîn?

10

12

13

14

15

16

7

Compare Rīm-Sîn E l. 7: ša-ra-da-gub šu sikil-la ak-a-ni niĝ2-nam mi-ni-in-dug3-dug3 “He stands ready to serve you, and when he performed the purification he made everything good.” (Brisch 2007: 212–13). Though the line is not an exact parallel, it is possible that the meaning was similar. The end of the line is too broken to attempt a restoration. For the phrase u4-da gub “who serves daily” see Tinney 1996: 160 ad 175. The current state of preservation of this line appears worse than at the time when the line drawing was made. For ka hul2-la “a joyful sound” see Civil 1996: 163, col. II l. 6′ (an Ur-Namma literary fragment = Ur-Namma I). The use of the third person forms here is strange, but unfortunately the text is too fragmentary to fully understand the context here. The sign that looks like ud is more likely to be part of the sign un, and a prefix /nu/ would not make sense in this context. The last sign could possibly be a si, yet on the photograph on CDLI it is difficult to see, because the sign is on the edge of the tablet and the surface is fragmentary. The first sign could be lu2 or gu2, though the remaining wedges, at least as copied by Gurney and Kramer, would conform to neither of these options. Other texts related to Rīm-Sîn only mention having been named by An (Rīm-Sîn C lines 1 and 2) or An and Enlil (Rīm-Sîn D 6) or Enlil (Rīm-Sîn G 11). Being named by Ninurta is otherwise unattested in Rīm-Sîn texts to my knowledge. However, compare the letter from Ninšatapada to Rīm-Sîn l. 3, which describes having been chosen by Ninurta: en gal dnin-urta-ke4 zi-de3-eš pa3-da “who was truly chosen by the god Ninurta” (Brisch 2007: 247). The signs after Ninurta’s name are extremely hard to decipher. According to Gurney’s copy, there could be a repetition of the sign pa but I find it difficult to make sense of ˹sag3-sag3-ge-de3˺ in this context, which is, admittedly, broken. The suggestion to read ĝidru here is owing to the possibility that kings are often bestowed a scepter as a mark of divine legitimation to rule; the transliteration presented here is based on the photograph rather than Gurney’s copy, although the tablet seems to be in worse shape now than at the time when Gurney copied it. The middle of the line is quite broken, so even tentative suggestions are difficult. The signs ˹inim den-lil2-la2-ta˺, if read correctly, are strange at the end of the line.

8

Nicole Brisch

18

Unfortunately, there is very little I can make of this line. The temple name, if it was one, following the eš3, does not seem to be one of the usual suspects. The name of Nanna-manšum’s father is hard to decipher due to surface damage. It does not seem to resemble the patronym of the Nannamanšum, son of Ilšu-muballiṭ, who wrote the letter to Ninisinna (see already Michalowski 1980–83: 56). Langdon read the patronym abgiš-in, a reading that seems difficult to reconcile with the traces on Gurney’s copy or the photograph of the tablet. I assume these lines contain flattery of the king before launching into the actual petition of the sender, fragmentary though they are. The first two signs in Gurney’s copy look more like ab-ba, though Langdon’s copy shows clearly the signs ad-da, which would also fit better with the signs as visible today. Especially the second sign would be a bit too long for a ba. The reconstruction of this line is by no means clear. The sign saĝ is not visible anymore but was transliterated by Langdon (1914: 24 l. 25), though it is unclear how reliable this transliteration is. The phrase dingir saĝ-du-ga-ni/a occurs frequently in royal inscriptions of RīmSîn and Kudurmabuk and is commonly translated as “the god who engendered him” or the like (see Hallo and van Dijk 1968: 66). The end of the line is unclear, too little is preserved to allow for a restoration. Repeating the nominal part of the compound verb is the kind of redundancy that is known from other Larsa texts and shows that Sumerian compound verbs were not properly understood anymore (see Brisch 2007: 105–106), though in most extant cases the nominal part appears both before and after the verbal prefixes, which is not the case here (provided that the reading of this line is correct). Or could this be a non-finite form of en3 – tar at the end of the line? Langdon (1914: 24) read the first sign of this line dub and Gurney’s copy appears to confirm this suggestion. However, the following two signs are clearly kala and ga. The adjective kala-ga occurs together with dub in just one other place, a bal-bal-e song of Ningišzida line 7 (van Dijk 1960: 81–107 = Black et al. 1998–2006, ETCSL 4.19.1), in which the reading and translation are problematic. Van Dijk (1960: 81–82) read and translated the line in the Ningišzida hymn: “zag3-ga2 dub-kala-ga!-ba ni3-zi KA keš2-da” “Der an die rechtsgültige Tafel der Registrierung die Gerechtigkeit heftet.” The reading zag3-ga2 in the beginning is highly unlikely, because /zag/ “side, etc.” never ends in a nasalized /ĝ/, which the sign ĝa2 always

21

23–40? 23

24

25

28

An Attempt at the Letter from Nanna-manšum to Rı ̄m-Sîn?

29

30 31–32 33 34

37 39

41

9

indicates. Black et al. read and translated: “saĝ5-ĝa2 dub kala-ga-ba niĝ2-zid zu2 keše2-da” “checker of tablets, who secures justice …” (Black et al. 1998–2006, no. 4.19.1). A spelling of saĝĝa “temple administrator” or the like with the signs saĝ5-ĝa2, to my knowledge, not attested otherwise. The line in the Ningišzida hymn is too uncertain to help out with our line here. However, the remnants of the first sign in the Nannamanšum letter would fit the sign saĝĝa (see Mittermayer 2006: 84, sign no. 212, Sonderformen). The phrase “mighty temple administrator” is otherwise unattested, though the professional title does occur as a divine title (see, for example, in the hymn Rīm-Sîn B, a hymn to Haya, divine spouse of Nidaba, l. 14, where Haya is called the saĝĝa ban3da “junior chief administrator”). This type of re-using and deliberate altering of traditional phrases is typical for the, sometimes convoluted, Larsa style of literary Sumerian. Nevertheless, the suggesstions presented here may turn out to be incorrect if and when new manuscripts of this letter are found. Too much of this line is too fragmentary to dare a translation. I cannot reconcile Langdon’s transliteration “… túg sūb … dSíg zur-zu-ǔr er-bi mu-ni-in-è” with the remaining sign forms. The final verbal form, well recognizable from Gurney’s copy, is nowadays almost impossible to recognize. The word lirum here is unusual, if indeed this letter was addressed to Rīm-Sîn, because he was never praised for his physical strength. These lines were already too fragmentary for transliteration and translation in the first edition (see Langdon 1914: 24). Several gods are referred to as “determining the destiny” (An, Enlil, Utu). I cannot suggest a reconstruction for the verbal base here. The suggestion to restore ur-saĝ here was first made by Langdon (1914: 24). Too little of the line is intelligible to understand the context. For the verb šu-tu-tu see, for example, the royal inscription by WaradSîn (RIME 4.3.19 line 14 = Frayne 1990: 238). Could en-e-ne here stand for e-ne-ne? The verb is not clear, perhaps a syllabic spelling of gub? Alternatively, one could think of a compound verb, in which the nominal part was incorporated into the verbal stem (for other such cases see Brisch 2007: 105–106), for example, the verb gu2-ĝal2 “to submit.” The signs after a2 bad are fairly uncertain. Gurney’s copy indicates a pi following the bad, yet the photograph shows that the sign could

10

42

43 44–46

Nicole Brisch

equally be a damaged igi. The sign afterwards is hardly legible. The possibility to read pi-la as a syllabic spelling for pel would not make sense in this context. I would expect something to the effect of “… my father escaped (and is now?) under your protection (?)”. For a2-bad see van Dijk 1960: 21 ad 25. For ka-de6 Akkadian pû abālu “to bring news” see Civil 1976: 92 and Tinney 1996: 167 ad 217. The meaning “to bring news” could fit in this context. On the meaning of gub as “to write” see Brisch 2007: 88–89 and Robson 2007. The transliteration of these lines is largely based on Gurney’s copy. The photograph shows far less information.

Bibliography Black, Jeremy.A., Graham Cunningham, Jarle Ebeling, Esther Flückiger-Hawker, Eleanor Robson, John Taylor, and Gabor Zólyomi. The Electronic Text Corpus of Sumerian Literature. (http://etcsl.orinst.ox.ac.uk/), Oxford 1998–2006. Brisch, Nicole. 2007. Tradition and the Poetics of Innovation: Sumerian Court Literature of the Larsa Dynasty (c. 2003–1763 BCE). AOAT 339. Münster: Ugarit-Verlag. Civil, Miguel. 1976. The Song of the Plowing Oxen. Pp. 83–95 in Kramer Anniversary Volume: Cuneiform Studies in Honor of Samuel Noah Kramer, ed. Barry Eichler. AOAT 25. Kevelaer: Butzon & Bercker. Civil, Miguel. 1996. Literary Text about Ur-Namma. Aula Orientalis 14: 163–67. Civil, Miguel. 2000. From the Epistolary of the Edubba. Pp. 105–11 in Wisdom, Gods and Literature. Studies in Assyriology in Honour of W. G. Lambert, ed. Andrew R. George, Irving L. Finkel. Winona Lake, Indiana: Eisenbrauns. Derrida, Jacques. 1980. The Law of Genre. Critical Inquiry: 55–81. van Dijk, Jan A. A. 1960. Sumerische Götterlieder, II. Teil. Heidelberg: Carl Winter Universitätsverlag. Frayne, Douglas. 1990. Old Babylonian Period (2003–1595 B.C.). RIME, 4. Toronto: The University of Toronto Press. Hallo, William W. 1968. Individual Prayer in Sumerian: The Continuity of a Tradition. JAOS 88: 71–89. Hallo, William W. and Jan A. A. van Dijk. 1968. The Exaltation of Inanna. Yale Near Eastern Researches, 3. New Haven: Yale University Press. Kleinerman, Alexandra. 2014. Education in Early 2nd Millennium BC Babylonia. The Sumerian Epistolary Miscellanea. CM, 42. Leiden: Brill. Langdon, Stephen. 1914. Historical and Religious Texts from the Temple Library of Nippur. The Babylonian Expedition, series A, vol. 31. Munich. Michalowski, Piotr. 1980–83. Königsbriefe. RlA 6: 51–59. Michalowski, Piotr. 2011. The Correspondence of the Kings of Ur. An Epistolary History of an Ancient Mesopotamian Kingdom. Mesopotamian Civilizations, 15. Winona Lake, Indiana: Eisenbrauns.

An Attempt at the Letter from Nanna-manšum to Rı ̄m-Sîn?

11

Robson, Eleanor. 2007. Gendered Literacy and Numeracy in the Sumerian Literary Corpus. Pp. 215–48 in Analysing Literary Sumerian, corpus-based approaches, ed. Jarle Ebeling and Graham Cunningham. London: Equinox. Sjöberg, Åke and Eugen Bergmann. 1969. The Collection of the Sumerian Temple Hymns. TCS, 3. Locust Valley, New York: Augustinus Publishers. Tinney, Steve. 1996. The Nippur Lament. Royal Rhetoric and Divine Legitimation in the Reign of Išme-Dagan of Isin (1953–1935 B.C.). Occasional Publications of the Samuel Noah Kramer Fund, 16. Philadelphia, PA: The University of Pennsylvania Museum.

Antoine Cavigneaux

À tâtons dans le noir : À la recherche du sens de ḫabātum À Miguel Civil, à qui j’ai emprunté (aḫbut) l’idée

À la parution du CAD Ḫ (1956), F. R. Kraus fut intrigué par les cinq verbes ḫabātum (ḫabātum A–E plus un comme variante de abātum) admis par les éditeurs; et de nouveau à la parution du fascicule du AHw (1962), qui en posait six (I à VI), dont un (ḫabātum V) considéré comme une forme alternative à ḫapātum et un autre (ḫabātum VI) comme variante de abātum. Ses questions s’exprimèrent dans un article de la Revue d’Assyriologie où Kraus (1975) livra ses réflexions, ses corrections aux classements lexicographiques des deux dictionnaires, et ses conclusions : il réduit à la fin CAD ḫabātum A–E et AHw ḫabātum I–IV à deux lexèmes : 1. ḫabātum (a/u), subdivisé en 1-I ‹ piller, dépouiller › (rauben) 1-II ‹ prêter › (leihen) et 1-III ‹ aller, se rendre (à un endroit) › (sich begeben). 2. ḫabātum ou ḫapātum (i/i) ‹ être/devenir important/omnipotent/écrasant › ou un sens proche (etwa « gewichtig/übermächtig sein/werden »). Méthodique, précis, comme dans tous ses travaux, Kraus analyse subtilement les passages cruciaux et refuse la spéculation, sauf, mais avec précautions, pour la combinaison des diverses significations et pour la recherche d’un possible sens originel. Lire cet article issu d’un commentaire à deux dictionnaires, est un exercice assez ardu pour lequel il faut non seulement avoir sous les yeux à la fois CAD et AHw, mais effectuer les correspondances entre les citations de Kraus et les publications plus récentes, souvent de la main de Kraus lui-même. 1

Je ne me risquerais pas à reprendre la réflexion après Kraus si je n’avais pas été intrigué à mon tour par de nouvelles attestations qui me semblaient toujours irréductibles aux significations admises par lui. Je ne chercherai pas à documenter exhaustivement une fois de plus les emplois, ni en détail les citations qu’on peut retrouver aisément dans les dictionnaires ou dans l’article de

1 Ainsi LFBD Nr. 15 = AbB 10 : 15 etc. Acknowledgements: Je remercie B. Kouwenberg, qui, sans adhérer à toutes mes conclusions, m’a apporté de précieuses suggestions. https://doi.org/10.1515/9781501503696-002

À tâtons dans le noir : À la recherche du sens de ḫabā tum

13

Kraus, mais j’essaierai, à l’aide des exemples les plus parlants, de structurer les emplois qui ont pour la plupart été déjà reconnus par les chercheurs et de montrer leur interrelation.

‹ Piller, capturer › (ḫubtam ḫabātum) Pour ḫabātum (a/u) 1-I ‹ rauben ›, ‹ plunder ›, personne ne dispute. C’est le premier choix des deux dictionnaires. C’est un verbe transitif bi-relationnel, avec pour sujet le pilleur et pour objet en général la proie, chose, bête ou homme (ḫubtum), mais pas à la fois la victime et le bien qu’on lui arrache; ḫbt 1-I peut aussi être employé absolument (Kraus 1975 : 35). En toute rigueur on ne peut donc le traduire par ‹ voler ›, verbe tri-relationnel impliquant un voleur, un objet et une victime; en outre le vol n’implique pas nécessairement la violence, tandis qu’un concept inhérent à ḫbt est la violence brutale, ne reconnaissant la personne, si elle est impliquée, que comme butin ou comme proie. Quand l’objet de ḫbt est une personne, on peut hésiter entre ‹ enlever › et ‹ dépouiller ›, comme dans nakru ina āl pāṭija iḫabbatanni « l’ennemi m’enlèvera dans une de mes villes frontiers » (Koch 2005 : 177, l. 71), où ‹ enlever › paraît bien plus plausible que ‹ dépouiller ›, mais le sens pourrait peut-être aussi être un peu élargi « me fera violence » ou « m’infligera un dommage, des pertes ». Normalement ḫabtāku devrait signifier « je suis enlevé », victime d’un rapt, et non d’un vol, mais la différence est ténue et il y a des exceptions : awīlum ḫabtum est « la victime d’un vol avec violence » (CH § 23 et ci-dessous ex. 16a). 2 La constante ‹ dommage avec violence › demeure. Même si on devait l’élargir au sens ‹ arracher à quelqu ’un par la force un bien ou la liberté », le sens général paraît incontestable, tant les exemples sont nombreux et clairs. Pourtant il serait illusoire de le considérer comme primitif. On verra qu’il est issu d’une métaphore, ce qui ne pouvait ressortir des articles de dictionnaires, et qu’il s’inscrit dans un réseau de relations sémantiques qui le lie étroitement aux autres emplois. Le sens Kraus 1-II ‹ prêter ›, avec son réciproque ‹ emprunter ›, mieux documenté et interprété par Kraus que par les dictionnaires, est à première vue un

2 L’expression ḫabtāk(u) « au secours ! » fait partie de la langue courante, comme le montrent les nombreux exemples cités par F. Joannès, ARM 26/2, p. 299 et Durand (1993), qui rendent librement l’expression par « quel scandale ! ». On entend encore très bien le sens littéral dans Eidem 2011 : 374, l. 9′′, où c’est un cri poussé par des prisonniers! Il me semble qu’on peut rattacher aussi à cet emploi šumma ana eqlim urrad lu ḫabitkum « s’il descend pour revendiquer le champ, qu’il soit ton prisonnier! », c’est à dire « fais-en ce que tu veux! » (de l’adversaire, et non du champ) PBS 7 : 11 = AbB 11 : 11, 22.

14

Antoine Cavigneaux

degré plus faible du sens 1-I ‹ piller ›. Néanmoins la distinction n’est pas toujours nette : obtient-on quelque chose par violence ou par voie légale? Disons déjà ici qu’on peut aussi considérer l’emploi 1-II à part, du moins dans son sens strict, quasi-juridique, de ‹ prêter/emprunter ›, qui est vraisemblablement dénominal (< ḫubuttātum). Le sens 1-III ‹ s. begeben › ne concorde pas avec les deux premiers et doit être soumis à un nouvel examen aussi bien pour sa définition propre que pour ses relations aux autres significations.

ḫabātum ‹ battre, fouler, piétiner › On ouvre une nouvelle perspective si on part d’un emploi nouveau, reconnu par M. Civil dans une version bilingue des Géorgiques sumériens (exemples 1a et b), et qui suppose pour ḫabātum un sens ‹ battre, fouler, piétiner › (dreschen, zertrampeln), qu’on retrouve dans un hymne à Ishtar (ex. 2): (1) a. alpī ina šu-úḫ-bu-ti-ka // gu4-gu4 šub-bu-da-zu-ne « quand tu fais battre (les épis) aux bœufs // sum. quand tu fais tomber les bœufs?)3 » Georg. bil. iv 9 (Civil 1994 : 44) b. šēp alpī ejam šu-úḫ-bi-it // ĝìri gu4-gu4-da ba-ra-gaz « fais battre l’orge par les sabots des bœufs » (le sum. diffère) Georg. bil. iv 10 (Civil 1994 : 95 et 107 n. 123).4

3 Le sumérien šub est ambigu : ‹ faire tomber › semble signifier «quand tu les mets sur l’aire » ou peut-être « quand tu les fais jouer de leur poids pour écraser les épis ». Comme dans ce témoin l’akkadien est primaire, on peut soupçonner – horresco referens – que c’est la proximité entre šuḫbut (akk.) et šubud (sum.), qui motive cette équivalence, qui semble unique! Comme le remarque Civil (1994 : 45), l’auteur force, dans les additions, la syntaxe sumérienne dans le moule de l’akkadien; il pourrait aussi avoir forgé ad hoc des monstres lexicaux. Comparer munu4 ḫu-bu-uš dug4-ab : buqlam ḫubuš « broie le malt » TIM 9 : 88, 11. 4 Tout comme dans la citation précédente (ex. 1a), l’akkadien n’est pas cohérent avec le sumérien, qui semble dire « les pattes des bœufs ne doivent pas être cassées » ou « elle (l’orge) est écrasée par les pattes des bœufs ». La correspondance n’est donc pas syntaxique mais purement lexicale (entre gaz et ḫabātu, encore une innovation, peut-être inspirée par sa-gaz : ḫabbātum, mais qui reste dans le champ sémantique de gaz ‹ écraser ›). S’agit-il de précautions à prendre pour ménager les animaux au travail, éviter que les bœufs ne se bousculent sur la surface instable, ou de les laisser jouer librement sur l’aire, une sorte d’équivalent du précepte biblique « tu ne muselleras pas le bœuf en train de battre » (Dt 25, 4) ? Mais c’est peu probable, car ici l’akkadien est primaire (donc porteur du sens originel) et le sumérien secondaire.

À tâtons dans le noir : À la recherche du sens de ḫabā tum

15

(2) sapāḫ šubti u šu-uḫ-bu-ut libnāti « démolir la maison et faire (?) fouler les briques » (c’est à dire fouler l’argile pour faire les briques) Ištar Louvre I 16 (Groneberg 1997 : 23).5 Ces trois exemples sont au thème Š, ce qui est attendu dans les deux premiers, où il y a un deuxième sujet (les bœufs), moins dans le troisième, où il n’y en a pas. Le thème Š peut en l’occurrence être motivé par la recherche de l’allitération avec sapāḫ šubti, peut-être aussi par la réticence à se représenter Ištar accomplissant ce travail servile et particulièrement salissant. On retrouve le thème Š avec un sens métaphorique à l’époque mB: (3) ušaḫbit (ú-šaḫ-bit) mēsīšun « (l’ennemi araméo-sutéen) foula aux pieds leurs institutions sacrées (de Nippur et Sippar) » RIMB 2 : 73 l. 12 (SimbarŠipak), où une traduction causative (*fit fouler) serait forcée. Je ne connais pas d’attestation claire de ḫabātum au thème G avec le sens de ‹ battre, dépiquer le blé ›, qui se dit en akkadien plutôt diāšum.6 S’il avait été usuel à l’époque aB, on aurait pu l’attendre dans la liste lú-azlag : ašlaku (MSL 12 : 163 l. 176 sqq.), mais il y semble absent, tout comme diāšum il est vrai, peut-être parce que cette activité est plutôt typiquement associée aux animaux. L’équivalence lexicale (4) a. ĝìri—sil (pour si-il) : ḫabātum Antagal A 112 ‹ faire un mouvement ondulé, répété, saccadé du pied › (sum.) : ḫabātum (akk.), pourrait cependant être invoquée pour un sens tel que ‹ piétiner ›, mais l’expression sumérienne ne m’est pas connue en contexte, sauf, dans un texte scolaire, *ĝìri si-il-la-ab ki nu-bulug-ga dib-ba-ab šēpam purus Zarrartam šūtiq « …, passe en rase campagne (sum. « par un terrain sans

5 Groneberg s’inspire déjà de Civil pour traduire ‹ Niedertreten der Ziegeln ›, ce qui me paraît exact. Le sens concret reste ambigu : il pourrait s’agir d’un sens analogue à celui du verbe spḫ, dans le champ sémantique de la destruction, ce qui a sans doute poussé Lambert (1999–2000 : 275) à traduire ‹ to let the bricks be plundered ›, mais c’est invraisemblable, car les briques sèches exposées aux intempéries ne se laissent guère récupérer; peut-être aussi ‹ faire évacuer les briques › (de façon à ne laisser subsister aucune trace de la maison détruite; voir infra pour ce sens de ḫabātu). Mais on pourrait aussi avoir un sens opposé (construction), conformément à la personnalité d’Inana, pétrie de contraste et de contradiction. C’est celui qui nous paraît le meilleur; peu importe si c’est l’argile de la maison démolie qui va servir pour les briques de la nouvelle ou non. 6 Le verbe diāšum peut être aussi employé métaphoriquement pour ‹ écraser, piller systématiquement › (un pays), tout comme le correspondant hébreu dūš, voir les dictionnaires.

16

Antoine Cavigneaux

bornes ») » (MSLSS 1 : 89, l. 2), mais ĝìri est une conjecture moderne (l’original a má), et le sens n’est pas clair; il n’est pas sûr que les deux phrases soient connectées sémantiquement.7 b. On retrouve peut-être la même expression dans une autre équivalence lexicale : [x][x]-gi-ir-si-iltar : me-e ḫa-b[a?-tum (?)] Erim III c1 (MSL 17 : 50) (et pour la copie Gesche 2000 : 346) ‹ piétiner l’eau (mélangée à l’argile) › (?). Il y a cependant un emploi en contexte plus évocateur, comme me le rappelle B. Kouwenberg, dans le problème mathématique Haddad 104 : (5) 20 alli ḫabātim 20 alli labānim 10 alli balālim, Sumer 43 : 204 iv 16–17; cf. aussi 20 ḫabātam 20 labānam 10 balālam luput « prends 20 ḫabātu, 20 mouler, 10 mélanger » iv 29. Dans ces textes étudiés depuis par Friberg (2001 : 110) et Robson (1999 : 74–76 et 81–82), les coefficients (igi.gub) 20 et 10 sont donnés pour les normes de production (ici à peu près le sens technique de allu8) de trois activités participant de la fabrication de briques; même si elles ne sont pas données dans l’ordre chronologique, comme le note Robson, on peut les interpréter de la façon suivante : balālum = faire le mélange (d’eau, d’argile et de paille); ḫabātum = fouler (le mélange); labānum = mouler (les briques). C’est ainsi à peu près que J. Friberg comprenait ḫabātum, qu’il rend par ‹ crushing ›, tandis que E. Robson s’en tenait à ‹ perhaps to dig › des premiers éditeurs, Al-Rawi et Roaf 1984 : 211. La traduction n’a pas d’incidence sur l’interprétation mathématique, mais il paraît bien plus vraisemblable d’y voir le sens mis en lumière par Civil.9 Il est peut-être envisageable de rattacher encore à ce sens deux passages, l’un transitif, l’autre intransitif:

7 Il s’agit semble-t-il d’ordres donnés à un soldat. Woods 2006 : 114, rend par « bloque l’accès ! » (cut off the approach !), suivant le sens normal de šēpam parāsum. On attend quelque chose comme « marche! »; on pourrait penser aussi à ‹ aller et venir › ou ‹ battre la semelle ›, comme font les sentinelles ce qui pourrait justifier le má du sumérien, par la métaphore de la navette qui va d’une rive à l’autre (?) ou le roulis du bateau (?). Quant à l’akk. Zarrartum, il faut peutêtre le comprendre comme ‹ hostile › (de sarrāru). 8 Voir Robson 1999 : 75 ‹ daily rate ›. 9 Noter aussi accessoirement le coefficient 25 ša sig4 šaḫāṭim (Robson 1999 : 74 et 77). Le verbe šaḫāṭum (a/u) ‹ démouler (?) › (à l’origine expulser la brique du moule), est en assyrien le terme correspondant à labānum, voir Radner 1997–98 : 159–61; comme le montre Postgate 2006), le même verbe pouvait se dire aussi de l’épluchage (dehusking) des céréales. Voir encore George 1995 : 178 n. 34 (lire šḫṭ!). On a donc encore un verbe qui, pour des raisons tout à fait différentes, s’applique aussi bien au travail de la brique qu’à celui des céréales.

À tâtons dans le noir : À la recherche du sens de ḫabā tum

17

(6) a. i niḫbussū (ni-iḫ-bu-us-sú-˹ú?˺) « (avant que l’ennemi ne progresse encore), voulons-nous l’écraser? » Charpin 1992 : 98, 24,10 mais, comme le suggère déjà Charpin, la forme pourrait se rattacher aussi à ḫbṣ II (AHw : 303), de sens proche et qui se dit aussi de la meule à main, ou à ḫbš. b. eṭlūtum iḫbutū maḫarka Gilg aB III 153 = iv 18 ‹ les jeunes gens battirent des pieds, trépignèrent devant toi › (pour t’encourager au combat?).11 Signalons enfin, comme le suggérait déjà Civil 1994 : 107 n. 123, déjà ici l’emploi du statif dans maškanu ḫabit, que nous traduisons plus bas « l’aire est vidée » (ex. 13 a et b), mais qui pouvait être originellement « l’aire est battue (fertig gedroschen) », puisque vider l’aire c’est finir le battage. Le sens ‹ fouler, battre › est celui de l’arabe ḫbṭ et de l’hébreu ḥbṭ (aussi en syriaque), si bien qu’on pourrait songer à opposer deux racines akk. ḫbt et ḫbṭ. Mais le seul exemple potentiellement dirimant utilise pour consonne finale+i le signe TI, qui ne plaide pas pour une telle opposition. Par ailleurs ḥbt/ḫbt manque en arabe et en hébreu. On peut en déduire que la consonne initiale est originellement le troisième terme (glottalisé) de la série uvulaire ġ/ḫ/ḫʾ,12 et que la consonne finale (ṭ) s’est dissimilée comme celle de *ḍabāṭum > ṣabātum, et pour les mêmes raisons. En arabe irakien le verbe ḫbṭ signifie plutôt ‹ mélanger ›,13 mais particulièrement ‹ souiller avec de la boue › (correspondant en quelque sorte au balālum de l’ex. 5); l’eau boueuse, non filtrée qu’on amène aux jardins est māy ḫābuṭ (< ḫābiṭ).

ḫabātum ‹ arracher ›, ‹ emprunter ›, ou ‹ se procurer › de l’orge ? Même en faisant abstraction du sens ‹ fouler ›, quand on rencontre ḫabātum au thème G avec l’orge comme objet, le sens est parfois délicat à cerner, comme dans l’exemple suivant, déjà longuement commenté par Kraus (1975 : 38) : 10 Je dois cette référence aussi à B. Kouwenberg. 11 La traduction de George ‹ fled › (George 2003 : 201 et 569 avec justification p. 211 et renvoi à George 1991) est tout aussi plausible, mais les variantes tardives iḫ-bu-tu et iḫ-mu-uṭ-[ṭu] ‹ se hâtèrent (trépignèrent d’impatience)› ne plaident pas pour ʾbt ‹ fuir ›, puisqu’elles confirment le ḫ. Si mon interprétation est correcte, il s’agit des jeunes d’Uruk. 12 Voir Voigt 1992 : 43–44, pour cette série qui n’apparaît (graphiquement) complète dans aucune langue sémitique. 13 Peut-être avec influence de ḫlṭ, dont c’est le sens propre, et de ḫrbṭ « jeter la confusion, aller de travers », non classique, mais très employé dans la langue populaire.

18

Antoine Cavigneaux

(7) êm ašar kīma ḫabātim aḫabbat « je me procurerai de l’orge par tous les moyens » ARM 27 : 1 rev. 28–29. Pour l’interprétation de ce passage on a hésité entre plusieurs interprétations : 1) « je prendrai par la force » (ich werde mit Gewalt nehmen, Aro) 2) « j’emprunterai » (Birot), 3) « je raflerai tout ce que je pourrai d’orge » (Finet), 3a) « je me procurerai » (Kraus ‹ sich verschaffen ›) ou 4) « j’irai où il faudra mais j’aurai le grain » (Durand).14 Comme l’observe Kraus, la traduction 1 est trop forte. La traduction 2 est possible, mais explique mal la figure étymologique, qui doit avoir un contenu expressif, et est aussi mal adaptée à la situation concrète que la traduction 1. Les traductions 3 et 4 me paraissent toucher d’assez près le sens; grammaticalement seules 1, 2 et 3 sont correctes; toutes supposent pour le verbe ḫabātum un sens ‹ littéral › (1) ou des dérivations sémantiques plausibles et aisées à suivre : ‹ emprunter › suppose le verbe dénominal issu de ḫubuttātum ‹ zinsloses Darlehen ›; ‹ rafler › suppose un affaiblissement du sens ‹ dérober, piller ›; ‹ aller › suppose qu’on a à faire à ḫabātum comme verbe de mouvement.15 Kraus (1975 : 38) a recours à une ‹ vox media › ‹ s. verschaffen › (se procurer). Il semble être question d’orge obtenue à grand peine, d’où ma traduction, qui s’inspire du ‹ s. verschaffen › de Kraus, particulièrement adéquat, car il implique l’idée d’effort pour parvenir à son but. On pourrait aussi essayer ‹ besorgen › (qui peut, secondairement, prendre la nuance ‹ voler ›). Kraus reconnaît les cas où il ne saurait être question de violence et il apporte un exemple assez clair où il est question d’une forme de contrainte, mais légalement justifiée et non-violente (Selbsthilfe): (8) iḫbutma uštaddinanni « il m’a oblige (il a fait pression sur moi) à payer » (AbB 3 : 8, 7–9). Dans ce cas au moins, dans la progression linéaire de la violence qui va de ‹ se procurer › à ‹ extorquer ›, la traduction ‹ prêt › est exclue. Kraus aurait pu lui-même ajouter cet exemple à ses Koppelungen, Abt. D « Modifikatoren, welche auf die vom Hauptprädikat beschriebene Handlung bezügliche Angaben über Impulse, Bemühungen, Befähigung und Motive ihres Subjekts beibringen » sous le § 17, avec des modificateurs comme anāḫum « faire … à grand peine », šutapšuqum « s’échiner à … » et ramānšu šaqāšum « se tuer à … » (Kraus 1997 : 22–24).

14 Birot (1993 : 41), dans son commentaire ARM 27, envisage les deux premières traductions; J. M. Durand, LAPO 17, p. 542–545 défend la quatrième, interprétant ḫabātum comme verbe de mouvement (ḫabātum ša alāki). 15 C’est l’interprétation philologiquement la moins plausible, car ḫabātum ne signifie pas ‹ aller › sans autre, comme j’essaie de montrer plus bas. Cependant le sens obtenu est à peu près le même que la traduction 3 !

À tâtons dans le noir : À la recherche du sens de ḫabā tum

19

(9) ana sūqim lūṣi mannu ša ejam ana ukullê bītika ni-ḫa-ab-pa-ta-am « même si je descendais dans la rue (au marché?), de qui est-ce que nous pourrions arriver à obtenir de l’orge pour la subsistance de ta maisonnée? » (TCL 18 : 110 = AbB 14 : 164). Cette construction gauche que relève Kraus (1975 : 37) rappelle le français populaire « qui pourrions-nous bien taper pour avoir de l’orge? ». L’auteure de la phrase semble avoir hésité entre deux objets pour son verbe : l’orge et un éventuel fournisseur, et, dans ce cas éclairant aussi, les notions de don, de prêt, d’achat ou d’extorsion n’entrent pas en ligne de compte : il est tout simplement impossible d’obtenir de l’orge! La même chose vaut pour ejum ana ḫabātim ul ibašši AbB 10 : 41, 19 : il n’y a pas d’orge sur le marché; s’il y en avait, on pourrait l’emprunter, mais c’est impossible. La situation concrète est toujours défavorable. Le verbe ḫabātum a souvent un objet interne (ḫubtam ḫabātum); dans l’emploi particulier ‹ chercher à obtenir ›, on peut avoir un complément à la fois interne (pour la sémantique) et adverbial (pour la syntaxe) : tous les traducteurs ont essayé de rendre ašar kīma ḫabātim en exprimant l’urgence, le caractère impérieux du besoin à satisfaire. On a encore : (10) kīma taḫabbatī ḫubtī « obtiens comme tu pourras, débrouille-toi pour obtenir le plus possible » AbB 11 : 40, 17–18. Il est envisageable d’en faire un critère pour distinguer ‹ s. verschaffen › de ‹ leihen ›, comme dans (11) inūma ḫabātum taḫabbatuma « quand tu pourras te procurer … » (Kraus « wenn du eine Anleihe aufnehmen kannst … » semble trop limité) AbB 10 : 15, 34. L’éventail est large entre ‹ obtenir par une demande plus ou moins instante › et ‹ arracher par force ›, les gradations et les nuances nombreuses, et on peine à croire qu’un seul mot puisse les recouvrir toutes. À la recherche d’une Grundbedeutung, Kraus, après avoir pensé à ‹ wegführen › (Kraus 1975 : 36), ‹ wegtragen › (Kraus 1975 : 39), ajoute ‹ losmachen ›, ‹ abtrennen ›, ‹ absondern › (Kraus 1975 : 40). Cependant l’intensité variable de la violence inhérente au verbe s’explique si on reconnaît que le concept fondamental est celui de besoin, de difficulté : l’emploi de ḫbt implique qu’on cherche, dans des circonstances difficiles, à obtenir quelque chose au prix d’efforts plus ou moins grands. Cependant ḫabātum ‹ rauben › ne se situe pas simplement à l’extrémité d’une ligne qui commencerait à ‹ emprunter ›. La violence brutale de ‹ piller › fait abstrac-

20

Antoine Cavigneaux

tion de la personne et contraste avec ‹ emprunter › qui la considère par définition. ‹ Rauben › n’est envisageable que dans les cas extrêmes, désespérés, et à cause de l’impact d’un autre sens du verbe (‹ ravager › et ‹ envahir militairement ›, cf. infra); la solution normale pour obtenir quelque chose dans une situation économico-sociale paisible est évidemment l’emprunt au sens large du terme. Kraus aurait donc dû garder ‹ s. besorgen › comme un sens de ḫabātum, ‹ se procurer comme on peut, se débrouiller pour obtenir ›. Quant au mot ḫubuttātum ‹ prêt ›, il peut théoriquement s’expliquer aussi bien par une dérivation de ‹ battre › que de ‹ voler, arracher par force › (comme ḫubullu de ḫabālu). Du sens de ‹ battre ›, on passe en aisément au sens ‹ obtenir par pression, par insistance, par force, puis par violence ›, donc ‹ dérober ›. Mais une dérivation sémantique directe de ‹ battre › est aussi possible, en passant par l’idée de pression (de la nécessité), d’urgence, dont nous venons de souligner le rôle essentiel pour comprendre ḫbt. Le sens ‹ emprunter/prêter › s’en laisse dériver, mais est difficile à cerner précisément dans son sens technique hors des contextes juridiques. Une dérivation de ‹ voler › serait plausible seulement si on se plaçait du point de vue du créancier, ce qui est absolument invraisemblable, l’urgence est a priori du côté de l’emprunteur. Même quand on emploie le verbe au sens strict de ‹ prêter, remettre ›, c’est l’idée de pression économique, de besoin, qui détermine l’attitude du prêteur potentiel aussi bien que de l’emprunteur, comme le montre par exemple : (12) mammanni ana mammannim ul iḫabbati « personne ne prête à personne » LFBD 4 = AbB 10 : 4, 30–31 (Kraus 1975 : 37), c’est à dire que la situation économique et sociale est très tendue et que ceux qui ont du blé le stockent pour leur propre besoin; la spécification ‹ prêter › est superflue et on pourrait traduire tout aussi bien : « personne ne cède rien à personne », « il n’y a plus d’échange de biens ». ‹ Emprunter › suppose une recherche et une demande adressée à un prêteur, ou la rencontre de circonstances favorables permettant d’échapper au besoin.16 En français aussi, le verbe emprunter peut signifier ‹ recevoir en prêt › mais aussi ‹ prendre › dans un sens plus large; une langue peut même emprunter (un mot, un concept qui lui sont inconnus) à une autre; l’idée fondamentale est qu’on prend quelque chose dont on manque et dont on a besoin (on emprunte un bus, une route …), c’est le contexte qui détermine le sens sans qu’il y ait jamais d’ambiguïté. On peut préciser le sens économicojuridique en disant ‹ prendre/contracter un emprunt ›. Le verbe arabe a

16 Le correspondant sum. eš-dé-a pourrait être une forme verbale figée avec infixe terminatif, avec le sens ‹ (orge) versée pour compenser (un manque) ›.

À tâtons dans le noir : À la recherche du sens de ḫabā tum

21

connu un développement sémantique comparable à l’akkadien : ḫbṭ I et VIII (+ acc de personne) ‹ demander un service à qqn › aussi bien que le réciproque ‹ qqn rend ce service au demandeur, lui accorde ce bienfait › (biḫayrin); les dictionnaires (déjà Al-Khalil, au VIIIème siècle, et les compilations plus récentes) précisent ‹ demander sans formalité, directement, à quelqu’un avec qui on n’a ni relation ni obligation », qui rappelle fortement l’akkadien ḫabātum/ḫubuttātum ‹ obtenir une aide dans l’urgence, là où on peut ›.17

Nettoyer l’aire? Un autre cas limite est ḫabātum employé avec maškanu comme objet; CAD A2(b) donne deux exemples; dans les deux cas la traduction ‹ vider › ou ‹ piller › est envisageable : (13) a. maškanī ḫabitma « (tandis que suis à Babylone) mon aire a-t-elle été vidée? » AbB 11, 108, l. 21 et annûm maškanu ḫabit l. 27 (cité CAD M/1 : 369b s. v. maškanu). Le texte est difficile. L’auteur de la lettre se demande si son aire est vide, mais n’en est pas sûr. Il pense avoir encore de l’orge en sa possession, puisqu’il demande à son correspondant de la faire battre, vanner et trier (l. 37 sq.) en son absence. Je suppose donc qu’avant son départ il n’avait pas eu le temps de battre et qu’il demande si le travail a bien été accompli, donc si l’aire est propre. b. maškanātēšunu aḫbut « je vidai/pillai leurs aires » AfO 3 : 154 obv. 11 (Assur-dan II). Grammaticalement ‹ piller › est possible, mais ‹ vider, déblayer, nettoyer › est plus vraisemblable en contexte pacifique. En contexte militaire, la différence importe peu. L’expression fait songer à l’image qu’on a dans l’évangile de Luc « il tient la pelle à vanner pour nettoyer son aire » (Luc 3, 17). Le sens de ‹ vider ›, ‹ dégager ›, ‹ débarrasser ›, ‹ nettoyer › convient aussi à: (14) ḫābit ṣibittim « qui vide la prison » littéralement ‹ qui débloque l’emprisonnement › (c’est à dire accomplit une sorte de mīšarum en libérant les per-

17 Les lexicographes arabes le rattachent au sens ‹ secouer les feuilles d’un arbuste › (qu’on obtient en les faisant tomber, comme on obtient un cadeau grâce à une demande) ou au sens ‹ aller à l’aveuglette (à la recherche d’un donateur) ›, voir infra.

22

Antoine Cavigneaux

sonnes contraintes à la prison pour dette),18 une épithète d’Iddin-Nunum de Dēr, André-Salvini et Salvini 1997 : 39, l. 7. (15) bāb araḫtim ḫabātum ḫubtā (ḫu-ub-ta) « dégagez/déblayez/désengorgez à fond la bonde de l’Araḫtum », une formule pour aider l’accouchement dans une incantation paléo-ass (Michel 2004 : 396, 12–14 // Kouwenberg et Fincke 2012–13 : 145); on peut imaginer que ces mots accompagnaient un massage du ventre de la parturiente. Dans le cas d’une femme en dystocie, si on part du sens ḫbt ‹ piétiner (l’argile) ›, la métaphore se révèle peut-être encore plus riche d’associations : en effet le mélange d’argile, d’eau et de matières végétales dont on fait les briques s’appelle aĝarin/ agarinnu ‹ mélange, moût, matrice ›, un mélange que divers processus, fusion ou fermentation, transforment en un nouveau produit, le mot s’appliquant en outre à son contenant : l’argile est aĝarin de l’humanité, matière première de sa substance.19 Du sens de ‹ battre › (le blé ou l’argile) se dérive aisément un sens ‹ extraire, obtenir avec difficulté, obtenir par pression, par force ›, d’où une évolution dans deux directions différentes : ‹ dégager, vider, débarrasser › ou ‹ faire le maximum pour obtenir quelque chose dont on a besoin ›, avec diverses nuances liées à l’intensité plus ou moins forte de la nécessité : ‹ emprunter ›, ‹ obtenir à force d’insistance, par l’usage de la force », à la rigueur jusqu’à l’extrême limite ‹ extorquer, ravir, dépouiller ›, ce dernier emploi étant conditionné par le sens suivant, car un autre trait sémantique détermine encore le sens pris par ḫabātum.

Le piétinement sourd 20 … ḫabātum comme verbe de mouvement (ḫ. ša alāki) Dans les textes historiques ou épistolaires qui nous sont restés, mais particulièrement dans les apodoses divinatoires, qui font presque toujours allusion à des 18 Comme il ressort clairement des lignes suivantes Salvini et Salvini 1997 : 8–12 : lā ḫābil awīlim, muštēšir ḫablim, u ḫabiltim, šākin, mīšarim « qui ne lèse nul citoyen, qui fait droit à l’opprimé et à l’opprimée, qui établit l’ordre juste », avec un effet de contraste entre ḫbt et ḫbl, phonétiquement proches et dont les champs sémantiques se recouvrent partiellement. 19 Noter que les filles d’Anu sont, comme des maçons, équipées de houes (marru) et de paniers (tapšukku). Pour aGarin voir Cavigneaux et Al-Rawi 1995 : 195; Sallaberger 1994 : 544; récemment Sallaberger 2012 : 311; Heimpel 2005 : 239, Jaques 2015 : 138–41. Dans le mot sumérien aGarin, la consonne ouvrant la deuxième syllabe est incertaine (ĝ/g ?). 20 Le piétinement sourd des légions en marche (J.-M. de Heredia, La Trebbia).

À tâtons dans le noir : À la recherche du sens de ḫabā tum

23

situations critiques, ḫabātum peut signifier clairement ‹ ravir, piller ›. Le statif ou le thème N peuvent donc avoir le sens passif, comme dans: (16) a. ana mīnim ḫabit « pourquoi a-t-il été victime d’un pillage? » AS 22, 24, 12–13 (lettre paléo-bab. ancienne); b. awīlum ana šubat ḫabbāti īšširma ul iḫḫabbat « l’homme ira droit sur une embuscade de brigands, mais ne sera pas dévalisé (ou : enlevé) » (Koch-Westenholz 2000 : 59.66; // ḫabbātū iḫabbatūšu « les brigands le dévaliseront (ou : l’enlèveront) » à la ligne précédente); c. naḫbutum ittaḫbat « (si un homme) est capturé et enlevé » Codex Ešn A II 40 (§ 29). Mais le verbe est le plus souvent employé absolument. CAD traduit en ce cas par ‹ faire une incursion, une razzia › (CAD A (2) b to raid; D (1) move across, make an incursion, a razzia into enemy territory). On pourrait traduire en français ‹ faire marche, faire campagne, marcher › (au sens militaire). Concrètement il y a presque toujours un rapport avec le pillage, mais, comme le souligne Kraus (1975 : 39 sq.), il est difficile d’imaginer un objet sous-entendu (comme les moissons, les biens …), qui pourrait expliquer ce sens. Cependant on dit aussi en français ‹ battre la campagne ›, tout comme ḫabātum un terme militaire (et cynégétique). L’équivalence lexicale sar : min (ḫabātu) ša alāki (Antagal A 114) pourrait se traduire ‹ faire une battue › (sar est aussi kuššudu ‹ chasser ›). (17) nakru ina ri’īti ana kašād (kur) būli (máš.anše) i-ḫab-bat ‹ l’ennemi sillonnera la pâture pour attraper du bétail › (Koch-Westenholz 2000 : 27.18). L’exemple est instructif, montrant que ḫabātum, employé ici apparemment au thème G,21 est bien ici verbe de mouvement, il ne s’agit pas d’un verbe transitif signifiant ‹ ravir, voler ›. En outre, dans cet emploi très commun dans les apodoses divinatoires, c’est le thème N, intransitif et ingressif, qui est de loin le plus fréquent :22 21 Avec i- et non iḫ- à l’initiale, comme on peut vérifier sur la photo pl. XLIII. 22 Kouwenberg 2010 : 299, et déjà Kraus (1975 : 39) mettent même en doute l’emploi du thème G. Les exemples aB sont en effet rares : deux apodoses divinatoires déjà relevées par Kraus : i[ ?-]ḫa-ba-at (RA 40 : 91, 37) (mais les hachures esquissées par Nougayrol dans la copie, RA 38 : 84, laissent place à l’incertitude); i-ḫa-ba-tam (RA 27 : 142, 10) n’est pas signifiant, car le texte n’écrit pas le redoublement des consonnes, sauf devant les suffixes enclitiques (pronoms possessifs et -ma) et dans quelques cas exceptionnels; cependant quelques passages font pencher la balance pour le thème G : (lāma ana aḫ Purattim) i-ḫa-ab-ba-tu (Syria 19 : 122, 42); i-ḫaab-ba-a[t] (ARM 28 : 95, 36); [a]na libbi māt … lā a-ḫa-ab-ba-tu (Eidem 2011 : 411, 3 sq.); peut-

24

Antoine Cavigneaux

(18) ana lā naḫbut nakrim « pour éviter que l’ennemi ne fasse des ravages » ARM 26/1 : 142, 7. Ou encore (19) ākiltum lā kâttum ana mātika iḫḫabbatam « un fléau malvenu (dont tu te serais volontiers passé)23 envahira ton pays » YOS 10 : 25 rev. 50 :24 et j’ajouterai quelques exemples que j’emprunte (aḫbut !) à Koch-Westenholz 2000 : (20) (nakrum) ana mātija iḫḫabbat” (l’ennemi) marchera contre mon pays”; ina mātija iḫḫabbatma išatam inaddi « il déferlera sur mon pays et y mettra le feu »; iḫḫabbatamma ejam imaḫḫaṣ « il marchera contre moi et détruira les récoltes »; iḫḫabbatamma iddâk/addâk « il marchera contre moi, mais sera battu (ou : je serai battu) ».25 Le sens n’est évidemment pas limité aux apodoses divinatoires, mais se retrouve dans de multiples exemples : (21) ab

ana māt Qabrâ aḫḫabitma mātam šâti ebūrša amḫaṣ « Je marchai contre le pays de Qabrâ, et ce pays j’en détruisis la moisson » (RA 7 : 153, Šamši-Adad); PN iḫḫabtamma ebūr NG imḫaṣ « PN s’est mis en campagne contre moi et a détruit la moisson de … » Eidem et Læssøe 2001 : 36, 4, où – tout comme dans la littérature divinatoire juste citée – ‹ moisson › n’est pas l’objet du verbe, et interdit de traduire sans autre par ‹ piller ›.

Dans tous ces exemples, tous tirés de contextes politico-militaires, le sens est à peu près ‹ parcourir un pays en le ravageant ›. Quand le roi Atamru dit au régent élamite ana mātišu naḫbit « va … vers son pays ! » (ARM 6 : 51, 11 // 52, 12), il s’agit évidemment d’une randonnée destructrice. Si on veut respecter la construction grammaticale, il est difficile de ne pas avoir recours à une périphrase ‹ aller à tel endroit pour détruire ›, mais on pourrait souvent traduire

être i niḫbussu[m], s’il faut, avec Charpin, lire ainsi ex. 6a. On peut envisager que ḫbt G soit dénominal ‹ faire le pillard › (ḫabbātum). 23 Ou peut-être « un fléau inconnu » (qui ne vient pas de chez toi). 24 Le parallèle ākiltum la kâttum iṭeḫḫiakkum karêka ugammar « un fléau malvenu surviendra chez toi et anéantira tes silos » YOS 10 : 44, 57, utilise le verbe ṭeḫûm au sens négatif ‹ venir importuner, attaquer › qu’il a quand le sujet est un démon ou une maladie. 25 Koch-Westenholz 2000 : 88 v 22, 64.61–64; et d’autres exemples dans l’index p. 503 s. v. ḫabātu ‹ to rob ›.

À tâtons dans le noir : À la recherche du sens de ḫabā tum

25

simplement par ‹ envahir ›, même si le verbe français est transitif. Le sujet est presque toujours collectif, même quand il s’agit d’un singulier, la présence d’une armée est implicite : aḫ(ḫ)abbat ‹ je partirai en campagne ›. Exceptionnellement cet emploi plutôt militaire peut s’employer métaphoriquement en contexte civil : mīnum ša taḫḫabtuma eqlam ša ON ša ina temmenim lā kunnukum šipra teppešu « comme oses-tu venir en soudard et travailler un champ qui ne t’est pas attribué dans le cadastre? » (Kupper 1959 : 35, D 31, l. 5);26 la traduction de Kupper « tu as commis une usurpation » me semble toucher à peu près le sens. L’emploi du thème N interdit de traduire sans autre par ‹ piller › et l’exaction est beaucoup plus forte que dans ex. 8. Cependant, sans passer par une métaphore militaire ou un sens juridique, on pourrait donner ici à ḫbt N la nuance ‹ piétiner maladroitement ›, « tu viens avec tes gros sabots, (effaçant nos plans bien tracés), … ». On sent encore la marque du sens originel dans l’apodose divinatoire: (22) šēp nakri (gìri kúr) ana mātija iḫḫabbata kimin « le ‹ pied › de l’ennemi marchera contre mon pays » (Koch 2005 : 412–13, l. 4). Même s’il faut lire : « ‹ pied › de l’ennemi; il marchera … », ce qui est improbable, car il s’agit d’une apodose, il y a un rapport concret sensible entre le pied et le piétinement des envahisseurs. En général, comme signalé dans CAD Š/2 : 306a, il y a une relation systématique entre la marque omineuse du ‹ pied › et l’idée d’arrivée ou d’attaque dans l’apodose. On pourrait prendre le français ‹ razzier, faire une incursion › pour rendre compte du sens transitif ‹ dérober › et du sens intransitif ‹ faire mouvement ›, mais d’une part le verbe en cet emploi est très clairement intransitif, d’autre part la nuance hostile est moins celle d’attaque soudaine avec une troupe légère (plutôt šiḫṭu, šaḫāṭu) que le déferlement de colonnes armées laissant derrière elles pillage et destruction. En hébreu aussi le verbe ḥbṭ est employé

26 Kupper rend temmen(n)um par ‹ document officiel ›. Normalement temen est un terme d’arpentage désignant un quadrilatère régulier posé pour pouvoir calculer les surfaces de champs de forme géométrique complexe. Ce quadrilatère est marqué au sol par des piquets aux angles, tout comme le plan architectural d’un bâtiment tracé au sol, ou le cadre du métier à tisser (voir Dunham 1986 : 32–62). Je proposerais donc ‹ cadastre ›. Le sens posé par J.-M. Durand ‹ un quartier d’habitation › (apud Birot, ARM 27, p. 111 et LAPO 16, p. 82–83) pourrait s’expliquer s’il s’agit d’un bâtiment construit sur un plan régulier, comme le secteur sud-ouest du palais de Mari (pièces 86 à 105), qui était sans doute réservé aux serviteurs mâles. Cf. encore Ziegler 2007 : 122. N.B. Chez Ziegler 2007 : 237, l. 12 te-em-ma-na n’est sans doute pas une variante, mais plutôt un incipit sum. (< *ta im-ma-an-ak?).

26

Antoine Cavigneaux

avec le sens concret de ‹ battre le blé › (comme dans Ruth 2,17) ou métaphoriquement : (le Seigneur) « fera les batteries (jaḥboṭ) du cours du fleuve jusqu’au ruisseau d’Égypte » (Isaïe 27, 12),27 c’est à dire marchera d’un pas destructeur, fera le grand nettoyage de l’Euphrate à l’Égypte, avec à peu près le sens qu’a ḫabātum N dans les apodoses divinatoires. On pourrait songer à une évolution sémantique rappelant le français ‹ presser ›, ‹ presser le pas, se presser › (sich beeilen), mais en akkadien le verbe semble a bien plutôt le sens ‹ piétiner, avancer lourdement, péniblement › (angl. plodding), une marche lourde, laborieuse, pénible ou destructrice, comme le montre encore l’emploi suivant.

Voyage dans la nuit On notera avec Kraus que ḫabātum a/u ‹ wandern › est transitif, comme peut l’être alākum; le contexte est très fréquemment, mais pas obligatoirement, militaire : *ḫa-bi-it ḫuršānī « qui piétine les montagnes », une erreur de lecture (lire qer-bé-et), doit disparaître des dictionnaires. Néanmoins il subsiste au moins trois des exemples donnés par AHw : (23) (Sargon) ša ekletam iḫbutu « qui marcha à l’aveugle dans le noir » (// ša ekletam illiku); l’expression est un topos qu’on retrouve dans liḫbut ekletam « qu’il (Anzu) titube dans l’obscurité ! » Anzu Suse Aa, 65 (Vogelzang 1988 : 98) (24) tušaḫbatanni appara ša makî u dulli ‹ tu me fais patauger (dans) un marais de privation et de labeur‘ (MDP 18 : 250, 9–11)28 On entrevoit, de loin, ici encore très concrètement l’image des ouvriers pataugeant dans le limon à briques. Plus abstraitement, l’objet de ḫabātum peut-être le travail, les difficultés, la peine elle-même, comme dans le ‹ Juste souffrant › aB :

27 La métaphore se réfère sans aucun doute au travail du grain, le texte poursuit « et vous, vous serez glanés un à un, enfants d’Israël », opposant l’action massive et destructrice du battage au soin délicat apporté à recueillir chaque grain tombé à terre. 28 Transcrit et traduit dans Bottéro, 1982 : 395 « après m ’avoir fait arpenter un marécage de privation(s) et de peine(s) »; la traduction plus tardive (Steve et Gasche 1996 : 334) « tu me fais emprunter un marécage de misère et de malheur » semble s’appuyer sur ḫbt. ‹ leihen ›, ce qui serait artificiel.

À tâtons dans le noir : À la recherche du sens de ḫabā tum

27

(25) bēliššu dulli iḫbutu imannu « he recounts to his lord the toil he has gone through » (Lambert 1987 : 190, 10). Le sens commun à ces trois exemples n’est pas ‹ aller ›, ni même ‹ marcher ›, mais ‹ marcher d’un pas lourd › (all. stapfen, angl. slog), ou ‹ progresser d’une démarche hésitante, incertaine › (all. torkeln; angl. blunder). Encore une fois les dictionnaires arabes nous fournissent une analogie frappante : ḫbṭ I ‹ marcher sans direction, à l’aveuglette ›; qui s’applique aussi métaphoriquement à quelqu’un qui se conduit sans réflexion, particulièrement yaḫbiṭu ḫabṭa ʿašwā’a « il agit au hasard ». Cette expression, ou encore ḫabaṭa hallayla, ou haẓẓalmāʾa ‹ battre la nuit, ou l’obscurité ›, également attesté, sonnent presque comme une traduction de l’akk. ex. 23. Encore aujourd’hui en arabe d’Irak ḫabbaṭa (ḫbṭ II) = ‹ marcher à tâtons ›.

Ḫubbutu (ḫbt au thème D) Le thème D est rare dans les textes. On l’a cependant dans la formule de serment [ana māt A]pim lā nuḫabba[tu] « nous n’envahirons jamais le pays d’Apum » (Eidem 2011 : 355, traité 1a 4′); le contexte est mal préservé, mais ce texte semble employer ḫbt D avec le sens de ‹ marcher contre › qui est en général (mais peut-être pas exclusivement) celui de ḫbt N. Comme les textes de Tell Leilan emploient surtout ḫbt G avec le sens de ‹ capturer › (traité 4 iv 3′; traité 6 m 5′), il se pourrait que l’opposition N/G soit remplacée à Tell Leilan par D/ G. Cependant la lettre 44 (Eidem 2011 : 115 sq.) emploie clairement ḫbt G avec le sens ‹ capturer, enlever › (l. 13), l’acteur étant un ḫābitānum ‹ kidnappeur › (l. 19), et par contraste ḫbt D avec le même sens, mais augmenté de la notion de pluralité : uḫabbatū « ils enlèvent (des gens de Yamutbal et les vendent) » (l. 8), le sujet étant en ce cas des muḫabbitū ‹ kidnappeurs professionnels, kidnappeurs de masse › (l. 4). De ce sens pluriel relève aussi le passif Dt uḫtabbatā « elles (les filles?) seront enlevées » (YOS 10 : 33 v 32). Un pluriel est sans doute aussi impliqué dans ṣabit-ma mārī Apim uḫabbat « il s’est mis29 à enlever systématiquement les gens d’Apum » (ARM 28 : 95, 35). Dans ce même texte le contraste avec ḫbt G de la ligne suivante est difficile à expliquer : ša kīma išerru u ušamšû maḫrêm-ma iḫabbat « du matin au soir il ne pense d’abord qu’à brigander » (ARM 28 : 95, 36). Dans ce dernier cas le verbe est employé absolument,

29 Ou : « il est décidé ». La conjecture de J.-M. Durand *ḫa!-bi-it-ma, LAPO 18, p. 155, est arbitraire.

28

Antoine Cavigneaux

donc peut-être avec un sens un peu plus large ‹ piller, brigander ›, comprenant toutes les facettes du métier de pillard.30

ḫabbātum ‹ manouvrier, piéton ›? Il existe une catégorie de travailleurs agricoles qualifiée de ḫabbātum, que AHw distingue de ḫabbātum ‹ voleur › (dérivé de ḫabātum I ‹ rauben ›) et rattache à son ḫabātum III ‹ wandern ›. J.-M. Durand (1992 : 106 et n. 71) dérive, comme von Soden, ce substantif du verbe de mouvement et traduit d’abord en 1992 ‹ travailleurs itinérants louant leurs services de lieu en lieu ›; puis en 1998 ‹ gitans ›, ‹ travailleurs en déplacement › (LAPO 17 : 31 et 419; de ḫabātum ‹ circuler ›).31 Eidem (1991 : 127) avait déjà remarqué qu’outre le sens de ‹ pillard ›, ḫabbātum désigne aussi une activité structurée, socialement reconnue, même s’il s’agit de main d’œuvre non qualifiée. Les travailleurs (toujours au pluriel), ainsi qu’un waklum šu ḫabbātī (pa šu ḫa-ba-ti TH 02-T192 [inédit]), sont déjà attestés dans les textes de Mari de type šakkanakku-tardif. On trouve encore le ḫabbātum (sa.gaz) dans une liste néo-assyrienne comme profession ou activité entre ‹ puiseur d’eau » (a.bala) et ‹ moissonneur › (še.kin.kud).32 Les attestations assez fréquentes de ‹ chefs de section de ḫabbātū › encore à l’époque aB, comme dans Tell el-Rimah 267,33 montrent qu’il ne s’agit pas d’une exception, mais d’un statut bien établi; la lettre AbB 7, 11634 est presque un paradigme pour l’usage du verbe ḫabātum et de ses dérivés : elle est adressée à un wakil ḫabbātī (ugula ḫa-ba-ti), par quelqu’un qui a été victime d’un vol (l. 8 ḫabbātīja aṣbat « j’ai attrapé mes voleurs ») et emploie plusieurs formes dérivées de la racine ḫbt. Sur ce texte et wakil ḫabbātī voir l’aperçu que donne Wilcke 1992 : 63 sq. Les troupes de ḫabbātum devaient être dures à tenir, mais étaient jusqu’à

30 Noter accessoirement dans cette lettre le contaste avec šḫṭ Gtn ‹ faire des incursions répétées › : ana kajjantim ištanaḫḫiṭanni, ištanaḫḫiṭū (l. 7 et 44). 31 Comm. ad 456 (= Dossin 1973 : 279) rev. 9′ : lúḫa-ab-ba-tum ù lúki-na-aḫ-nimmeš « des ḫabbātum et des gens de Kanaan ». 32 Voir MSL 12, p. 240 v 21–23. 33 Sur les ḫabbātum dans les textes voir le point sur la question chez Stol 2004 : 798 sq. Eidem (2011 : 20) explique leur apparition massive dans le nord de la Mésopotamie et l’importance qu’ils prennent comme facteur militaire et social par la ruine des états-cités qui contrôlaient la région et le vide politique qui s’ensuivit. Il penche pour la traduction ‹ mercenaire ›. 34 Avec un long commentaire de Kraus, AbB 7, p. 95, note ad 116a. L’importance de ce texte pour la définition de ḫabātum est relevée aussi par Eidem (2011 : 19 n. 27) et par Stol (2004 : 799).

À tâtons dans le noir : À la recherche du sens de ḫabā tum

29

un certain point intégrées dans le potentiel de main d’œuvre civile ou militaire, peut-être parfois, mais pas nécessairement, des forçats.35 Pour ces travailleurs, même si c’étaient parfois des brutes incontrôlables, aussi bien la dérivation de ‹ voleur › que celle de ‹ vagabond, rôdeur › paraissent forcées. Bien plus plausible : ‹ batteur (qui fait les batteries) › > ‹ qui fait les tâches laborieuses, les travaux, les marches pénibles › ! L’équivalence lexicale nārāru (aide, auxiliaire) : ḫabbātum (Malku IV 204) pourrait faire allusion à des équipes de supplétifs, mobilisées pour les gros travaux urgents ou la guerre, les piétons, les fantassins, la piétaille.

La force irrésistible et la lumière éclatante Le verbe plus rare et avec le vocalisme en i, qu’on a dans (26) liḫbit râmī « que mon amour écrase », l’emporte (sur tout) von Soden 1950 : 164 i 11; on pourrait peut-être dire aussi ‹ éclate › (comme éclate la bale du blé), pour exprimer la force irrésistible de l’amour, – peut aussi se rattacher sans peine au ḫbt ‹ écraser ›; en effet ‹ être puissant, prédominant ›, c’est aussi ‹ être écrasant ›, v. Soden pose une autre racine, avec une sourde (ḫpt). M. Held y voyait une variante kbt/ḫbt, une hypothèse que récusait Kraus, mais qui néanmoins reste digne de considération, dans la mesure où la proximité sémantique entre kbt et ḫbt a pu exercer une influence sur le changement de classe vocalique du verbe (passant de a/u à i). Le mot semble réservé à la poésie lyrique: (27) […] dašpu ḫitabbut (ḫi-ta-bu-ut) elīja « le doux … exerce une pression incessante sur moi, m’oppresse sans cesse » YOS 11 : 24, i 3.36 C’est ce sens qui semble apparaître aussi dans l’onomastique (Stamm 1939 : 318 sq.; Kraus 1975 : 33) :

35 « Einsatz Festgenommener unter staatlicher Aufsicht » C. Wilcke 1992 : 64. 36 Hecker 1989 : 748 « Süsse [Liebe zu dir] hält mich dauernd überwältigt » doit être assez près de l’original; W. G. Lambert apud Foster 2005 : 162–164 corrige : [ga-ar]-da-mu iḫ!-ta-buut; Sigrist et Westenholz 2008 : 679 et Streck et Wasserman (2016 : 4.1.2.2) sur le site SEAL de Leipzig consulté le 12.II.2016, proposent [dUTU ši]-it-mu-rum ḫi-ta-pu-ut e-li-ja « the vigo[rous sun] keeps overcoming me ».

30

Antoine Cavigneaux

Erra-ḫābit « Erra est écrasant, tyrannique », pourrait signifier aussi « Erra exige beaucoup » ou « fait grand ravage ».37 Dans les noms Ḫabit-Sīn, Ḫabit-nūršu, on pourrait peut-être traduire ḫabit par ‹ est éblouissant › pour suggérer l’éclat insupportable de la lumière divine. Pour interpréter Itti-ilīja-aḫbut, Stamm et CAD ḫabātu B (1b) avaient recours au dénominatif de ḫubuttātum « J’ai obtenu un prêt de mon dieu », ce qui est trop spécifique. On peut aussi envisager un sens « J’ai parcouru un dur chemin avec (l’aide, la permission de) mon dieu », qui n’est pas absolument impossible, mais pour lequel je manque de parallèles. Normalement les noms en itti-ND sont positifs (Itti-ili-aḫdu …). On peut donc choisir la possibilité : « À force (de demandes insistantes) j’ai obtenu de mon dieu (un fils) ». Si on pose le verbe ḫpt de vSod = ḫbt 2 de Kraus, on obtient aussi le sens plausible « Grâce à mon dieu, j’ai eu le dessus »; mais à cause du vocalisme en u (à cette époque on attend i) on le récusera, avec Kraus (1975 : 33). Les exemples amenés depuis par Westenholz 1987 : 34, commentaire au no 8, révèlent cependant que le NP paléo-akk. Iḫbut-alsin « Il l’a emporté sur elles, il les a écrasées (les sœurs) » correspond au paléo-bab. Iḫbit-elšunu « Il l’a emporté sur eux »; le verbe a donc pu changer de classe vocalique vers le tournant du IIIème au IIème millénaire.38 Apparemment c’est surtout pour le sens ‹ übermächtig s./w. › que le changement de classe va de pair avec l’intransitivité plus nette. Plus tard le verbe tendra à passer entièrement à la classe i, comme nombre de verbes akkadiens.39

Cas douteux Il restera toujours des passages difficiles à expliquer. Je ne reviendrai ici que sur enūma taḫbilu Idiglat ubbal enūma ta-ḫab-bi-tu4 šamû ēzibūka « Quand tu commets une exaction, le Tigre emporte …; quand tu …, les pluies t’abandonnent » (Lambert 1960 : 253, 1–4). Tout comme à Kraus (1975 : 34), cette sen-

37 Pour ce nom, Hilgert (2002 : 210 n. 168) laisse le choix entre ḫbt et ḫpt. Ce nom rappelle Ilum-ḫābil ‹ Le dieu est ravisseur › (de vies) et Erra-ḫābiš, non mentionné par les dictionnaires, mais reconnu par Stol (1991 : 195), et dont le sens est plus obscur (Erra met en pièces?). 38 De la même façon, le verbe rsb, dans le sens ‹ faire une erreur › élucidé par Charpin 1995 : 43–55 (et déjà Kraus, AbB 10, p. 175, note g ad no 192), est dans la Ablautklasse a/u, tandis que rsb « frapper » passe à la classe i. Le lien sémantique est sans doute la nuance ‹ (cogner) à tort et à travers, (donner) un mauvais coup, faire une faute de frappe ›. 39 Voir Kouwenberg 2010 : 77.

À tâtons dans le noir : À la recherche du sens de ḫabā tum

31

tence proverbiale me semblait, au delà du parallélisme ḫbl/ḫbt et Tigre/pluie, inexplicable. CAD H 5a supposait que l’objet implicite est un champ. En effet un contexte agricole semble plausible. Cependant il me semble maintenant qu’on peut aussi interpréter en fonction du sens habituel de ḫbt i/i que nous venons de voir : « quand tu commets une exaction, le Tigre emporte (cela peut passer avec le courant); mais quand tu exagères (tu exerces une trop forte pression), les pluies t’abandonnent (plus une goutte pour faire passer le crime, tu ne peux plus réussir) »; trop d’injustice tue l’injustice, une sorte de mise en garde !40

Conclusion Au terme de cette laborieuse marche, je poserai l’hypothèse qu’il y a un seul et unique ḫabātum, dont la sémantique a connu une histoire riche et des ramifications complexes. Il me semble qu’on peut, en simplifiant, résumer ainsi l’évolution : D’un sens originel ‹ battre, écraser, fouler, piétiner ›, qui n’est plus guère employé pour l’orge qu’au thème Š, mais encore au thème simple pour le foulage de la terre à bâtir, se développent de nouvelles significations. Les développements suivent essentiellement quatre directions : 1. (ḫabātu ša alāki) marcher lourdement, laborieusement; peiner à la tâche (> ḫabbātu ‹ journalier, manouvrier, piéton ›) ; dans le domaine militaire ; battre la contrée; se mettre en marche, faire campagne 2. obtenir par pression a) obtenir avec peine, chercher à obtenir avec insistance; se faire prêter/ consentir à prêter (> ḫubuttātu ‹ prêt obtenu dans l’urgence › > ḫabātu ‹ emprunter/prêter ›) b) arracher par la force (ḫabātu ša šalāli) (> ḫabbātu ‹ brigand ›) 3. extraire les grains et nettoyer l’aire a) vider, dégager b) piller, enlever (ḫabātu ša šalāli) (> ḫabbātu ‹ brigand ›)41 4. ‹ manifester une puissance écrasante ›, surtout dans la poésie et l’onomastique, et qui passe à la classe i.

40 L’interprétation de Lambert (šamû = ‹ les dieux ›) me semble aller trop loin, mais ne changerait pas le sens. 41 Et aussi les dérivations ḫābitānum ‹ ravisseur, kidnappeur › et ḫābitānūtum ‹ le statut de personne victime d’un enlèvement ›, voir Eidem 2011 : 21.

32

Antoine Cavigneaux

Le sens de ‹ prêt › (ḫubuttātum) se dérive du sens 2 et est lié lui-même au verbe ḫabātu au double sens spécifique d’‹ emprunter/prêter › reconnu par Kraus, et qu’il est parfois à peine possible de distinguer d’un sens ‹ obtenir quelque chose dont on manque ›, sans doute parce que, pour les Babyloniens, la notion de nécessité ou de contrainte lui était encore inhérente. Ces multiples développements, en partie parallèles, laissent place à de nombreuses nuances et à quelques ambiguïtés. La frontière entre les emplois transitifs (2 et 3) et intransitifs (1 et 4) est très floue, et les interconnexions entre les divers sens sont nombreuses : ‹ faire campagne › (1) implique ‹ piller › et recoupe donc partiellement aussi bien le sens ‹ obtenir avec peine › (2) que le sens ‹ vider › (3); l’équivalence lexicale ḫabātu ša šalāli pourrait faire référence aussi bien à l’emploi militaire ‹ marcher › (1) qu’au sens ‹ piller › (3b); il n’est pas toujours aisé de distinguer un ḫabbātu ‹ pillard › d’un ḫabbātu en quête de travail. Ces interférences sont le principal obstacle à un classement rationnel des divers emplois, qui me paraît cependant possible dans l’ensemble. Le lexique arabe fournit des analogies qui ne confirment pas les dérivations sémantiques proposées ici, mais les rendent plausibles. Celles que j’ai signalées me semblent assez suggestives ; d’autres seraient encore envisageables, mais moins évidentes et plus contestables.42 L’absence dans les dictionnaires arabes du sens ‹ piller ›, – alors que le sens ‹ emprunter où on peut › est présent, – est tout aussi caractéristique et suggère que ‹ piller › est un développement propre à l’akkadien, issu du sens ‹ extraire par force ›, issu lui-même de ‹ battre le blé (ou la glaise) ›, lui aussi absent des dictionnaires arabes. Il est vrai que, conditionnés par le milieu sociologique des tribus du désert où ils cherchaient leurs sources et par leur obsession de l’arabité, les lexicographes arabes étaient plus enclins à noter pour ḫbṭ le piétinement caractéristique de la chamelle, ou le puits abîmé par le piétinement des animaux (ḫabīṭun), que le travail des paysans ignorants de l’arabe. On notera que le concept de force, de contrainte exercée ou subie, est commun à tous les emplois : la force exercée pour extraire le grain, la fatigue des longues marches, la violence des pillards qui parcourent le pays, la nécessité qui pousse quelqu’un à chercher la nourriture, et les efforts nécessaires pour l’obtenir, l’éclat irrésistible du dieu … On comprend aisément comment deux activités concrètes, particulièrement pénibles, mais essentielles pour l’économie babylonienne – le battage des blés et la production des briques – sont à l’origine de nombreux développements métaphoriques. Les Babyloniens qui

42 Le sens ‹ prédominer › de ḫbt i/i (ḫbt 2 de Kraus) pourrait aussi être rapproché des sens arabes ‹ terrasser, frapper d’épilepsie, posséder › (action d’un démon, du destin sur un homme); ‹ être frappé de torpeur ou de maladie › (I actif et passif).

À tâtons dans le noir : À la recherche du sens de ḫabā tum

33

usaient du verbe ḫabātum n’avaient pas devant les yeux les aires de blé ou les mares de boue à briques, pas plus que nous ne voyons la mer quand nous parlons de démarrer (partir avec un véhicule ou le faire partir) ou de louvoyer (biaiser au lieu d’aller droit au but); il semble que dans la production métaphorique, ce soit le piétinement de l’argile qui l’emporte sur celui des céréales, néanmoins il y a encore une sorte d’affinité élective entre le verbe ḫbt et l’orge, qui transparaît dans de nombreux emplois dérivés, comme si le sens originel, le battage de la récolte par les hommes ou les animaux, avait laissé une empreinte ineffaçable : ce que pillent et incendient les soudards, ce qu’essaient d’obtenir les gens dans la difficulté, c’est presque toujours du grain; et les armées qui ravagent le pays transforment les vergers et les champs ondoyants en un désert plat et nu, comme l’aire à la fin des batteries.

Bibliographie et abréviations LAPO = Littératures Anciennes du Proche-Orient Al-Rawi, Farouk N. H. et Michael Roaf. 1984. Ten Old Babylonian Mathematical Problems from Tell Haddad, Himrin. Sumer 43: 175–218. André-Salvini, Béatrice et Mirjo Salvini. 1997. Ein König von Dēr. AoF 24: 39–43. Birot, Maurice. 1993. Correspondance des gouverneurs de Qaṭṭunân. ARM 27. Paris: Editions Recherche sur les Civilisations. Bottéro, Jean. 1982. Les inscriptions cunéiformes funéraires. Pp. 373–406 in La mort, les morts dans les sociétés anciennes, ed. Gherardo Gnoli et al. Cambridge University Press: Cambridge/New Tork. Cavigneaux, Antoine et Farouk N. H. Al-Rawi. 1995. Textes Magiques de Tell Haddad (Textes de Tell Haddad II). Troisième partie. ZA 85: 169–220. Charpin, Dominique. 1992. De la Vallee du Tigre au « Triangle du Habur »: Un engrenage geopolitique. Pp. 97–102 dans Jean-Marie Durand, Recherches en Haute Mesopotamie: Tell Mohammed Diyab, Campagnes 1990 et 1991. Mémoires de NABU 2. Paris: Société pour l’Étude du Proche-Orient Ancien. Charpin, Dominique. 1995. « Lies natürlich … » À propos des erreurs de scribes dans les lettres de Mari. Pp. 43–55 dans Vom Alten Orient zum Alten Testament: Festschrift fü r Wolfram Freiherrn von Soden zum 85. Geburtstag am 19. Juni 1993, ed. Manfried Dietrich et al. AOAT 240. Münster : Neukirchener Verlag und Verlag Butzon & Bercker Kevelaer. Civil, Miguel. 1994. The Farmer’s instructions. Sabadell: Editorial Ausa. Dossin, George. 1973. Une mention de cananéens dans une lettre de Mari. Syria 50: 277–82. Dunham, Sally. 1986. Sumerian Words for Foundation. RA 80: 31–64. Durand, Jean-Marie. 1992. Unité et diversités au Proche-Orient à l’époque amorrite. Pp. 97– 128 dans La circulation des biens, des personnes et des idé es dans le Proche-Orient ancien: Actes de la XXXVIII e Rencontre Assyriologique Internationale (Paris, 8–10 juillet 1991), ed. Dominique Charpin et al. Paris: Éditions Recherche sur les Civilisations. Durand, Jean-Marie. Vindicatio in libertatem. NABU 1993/25.

34

Antoine Cavigneaux

Durand, Jean-Marie. 1998. Les documents épistolaires du palais de Mari 2. LAPO 17. Paris: Les Éditions du Cerf. Eidem, Jesper. 1991. The Tell Leilan Archives 1987. RA 85: 109–35. Eidem, Jesper. 2011. The Royal Archives from Tell Leilan: Old Babylonian Letters and Treaties from the Lower Town Palace East. Leiden: Nederlands Instituut voor het Nabije Oosten. Eidem, Jesper et Jørgen Læssøe. 2001. The Shemshara Archives 1. The Letters. Kobenhavn: The Royal Danish Academy of Sciences and Letters. Foster, Benjamin R. 2005. Before the Muses. An Anthology of Akkadian Literature. Third Edition. CDL Press: Bethesda, Maryland. Friberg, Joran. 2001. Bricks and Mud in Metro-Mathematical Cuneiform Texts. Pp. 61–153 in Changing Views on Ancient Near Eastern Mathematics: From a Workshop Jointly Organized by Altorientalisches Seminar, Freie Universitat Berlin; Seminar fur Vorderasiatische Altertumskunde, Freie Universitat Berlin; Max Planck Institute for Human Development and Education, Berlin, ed. J. Høyrup et P. Damerow. BBVO 19. Berlin: Dietrich Reimer Verlag. George, Andrew R. 1991. Seven Words. NABU 1991/19. George, Andrew R. 1995. The Bricks of E-sagil. Iraq 57: 173–97. George, Andrew R. 2003. The Babylonian Gilgamesh Epic : Introduction, Critical Edition and Cuneiform Texts, Vol. I. Oxford: Oxford University Press. Gesche, Petra D. 2000. Schulunterricht in Babylonien im ersten Jahrtausend v. Chr. AOAT 275. Münster: Ugarit-Verlag. Groneberg, Brigitte R. M. 1997. Lob der Ištar: Gebet und Ritual an die altbabylonische Venusgöttin. CM 8. Groningen: Styx Publications. Hecker, Karl. 1989. Akkadische Hymnen und Gebete. Pp. 718–83 dans Lieder und Gebete I, ed. Otto Kaiser. TUAT II/5. Gütersloh: Gütersloher Verlagshaus Gerd Mohn. Heimpel, Wolfgang. 2009. Workers and Construction Work at Garšana. CUSAS 5. Bethesda, Maryland: CDL Press. Hilgert, Markus. 2002. Akkadisch in der Ur III-Zeit. Münster: Rhema. Jaques, Margaret. 2015. Mon dieu qu’ai-je fait?: Les diĝir-ša-dab(5)-ba et la piété privée en Mésopotamie. OBO 273. Fribourg: Academic Press; Göttingen: Vandenhoeck & Ruprecht. Joannès, Francis. 1988. Menaces sur Karanâ. Pp. 283–303 in Dominique Charpin et al. Archives Épistolaires de Mari I/2. Archives Royales de Mari XXVI/2. Paris: Editions Recherche sur les Civilisations. Koch, Ulla S. 2005. Secrets of Extispicy: The Chapter Multābiltu of the Babylonian Extispicy Series and Niṣirti bārûti Texts Mainly from Aššurbanipal’s Library. AOAT 326. Münster: Ugarit-Verlag. Koch-Westenholz, Ulla S. 2000. Babylonian Liver Omens: The Chapters Manzāzu, Padānu, and Pān tākalti of the Babylonian Extispicy Series Mainly from Aššurbanipal’s Library. Carsten Niebuhr Institute Publications 25. Copenhagen: Museum Tusculanum Press. Kouwenberg, Bert and Jeanette C. Fincke. 2012–13. A « New » Old Assyrian Incantation. JEOL 44: 141–46. Kouwenberg, Norbertus Johannes Cornelis. 2010. The Akkadian Verb and Its Semitic Background. Languages of the Ancient Near East 2. Winona Lake: Eisenbrauns. Kraus, Fritz Rudolf. 1975. Akkadische Wörter und Ausdrücke, IX. RA 69: 31–40. Kraus, Fritz Rudolf. 1984. Sonderformen akkadischer Parataxe: die Koppelungen. Mededelingen der Koninklijke Nederlandse Akademie van Wetenschappen, Afd. Letterkunde. Amsterdam: North-Holland Publishing Company. Kupper, Jean-Robert. 1959. Lettres de Kiš (suite et fin). RA 53: 19–38.

À tâtons dans le noir : À la recherche du sens de ḫabā tum

35

Lambert, Wilfred George. 1960. Babylonian Wisdom Literature. Oxford: Oxford University Press. Lambert, Wilfred George. 1987. A Further Attempt at the Babylonian ‹ Man and His God. › Pp. 187–202 in Language, Literature, and History: Philological and Historical Studies Presented to Erica Reiner, ed. Francesca Rochberg-Halton. AOS 67. New Haven, Connecticut : American Oriental Society. Lambert, Wilfred George. 1999–2000. Compte-rendu Brigitte R. M. Groneberg, Lob der Ištar: Gebet und Ritual an die altbabylonische Venusgöttin. AfO 46–47: 274–77. Michel, Cécile. 2004. Deux incantations paléo-assyriennes: Une nouvelle incantation pour accompganer la naissance. Pp. 395–420 dans Assyria and Beyond: Studies Presented to Mogens Trolle Larsen, ed. Jan G. Dercksen. PIHANS 100. Leiden: Nederlands HistorischArchaeologisch Instituut in het Nabije Oosten. Postgate. J. Nicholas. 2006. riḫṣu, ḫiṣnu and šiḫṭu, but not bulgur. NABU 2006/12. Radner, Karen. 1997–98. Zur Bedeutung von šahāṭu im Neuassyrischen: « Ziegel herstellen » order « Ziegel glasieren »? AfO 44–45: 157–61. Al-Rawi, Farouk N. H. et Michael Roaf. 1984. Ten Old Babylonian Mathematical Problems from Tell Haddad, Himrin. Sumer 43: 175–218. Robson, Eleanor. 1999. Mesopotamian Mathematics, 2100–1600 BC: Technical Constants in Bureaucracy and Education. OECT 14. Oxford: Oxford University Press. Sallaberger, Walther. 1994. Compte-rendu Watson, Catalogue of cuneiform tablets in Birmingham City Museum. 2. OLZ 89: 538–46. Sallaberger, Walther. 2012. Bierbrauen in Versen: Eine neue Edition und Interpretation der Ninkasi-Hymne. Pp. 291–328 dans Altorientalische Studien zu Ehren von Pascal Attinger: mu-ni u4 ul-li2-a-aš-ĝa2-ĝa2-de3, ed. Catherine Mittermayer and Sabine Ecklin. OBO 256. Fribourg: Academic Press; Göttingen: Vandenhoeck & Ruprecht. Sigrist, Marcel et Joan Goodnick Westenholz. 2008. The Love Poem of Rīm-Sîn and Nanaya. Pp. 667–704 dans Birkat Shalom. Studies in the Bible, Ancient Near Eastern Literature, and Postbiblical Judaism Presented to Shalom M. Paul on the Occasion of His Seventieth Birthday, Vol. Two, ed. Chaim Cohen et al. Winona Lake, Indiana: Eisenbrauns. Stamm, Johann J. 1939. Die akkadische Namengebung. MVAG 44. Leipzig: J. C. Hinrichs Verlag. Steve, M.-J. et H. Gasche. 1996. « L’accès à l’au-delà. » Pp. 344–45 in Collectanea Orientalia. Histoire, arts de l’espace et industrie de la terre: Études offertes en hommage à Agnès Spycket, ed. H. Gasche et B. Hrouda. CPOA 3. Neuchâtel: Recherches et publications. Stol, Marten. 1991. Old Babylonian Personal Names. SEL 8: 191–212. Stol, Marten. 2004. Wirtschaft und Gesellschaft in Altbabylonischer Zeit. Pp. 643–975 dans Mesopotamien. Die altbabylonische Zeit. OBO 160/4. Fribourg: Academic Press; Göttingen: Vandenhoeck & Ruprecht. Streck, Michael P. et Nathan Wasserman, Sources of Early Akkadian Literature. A Text Corpus of Babylonian and Assyrian Literary Texts from the 3 rd and 2 nd Millennia BCE. http:// www.seal.uni-leipzig.de/. Vogelzang, M. E. 1988. Bin šar dadmē: Edition and Analysis of the Akkadian Anzu Poem. Groningen: Styx Publications. Voigt, Rainer. 1992. Die Lateralreihe /ś ṣ́ ź/ im Semitischen. ZDMG 142: 37–52. von Soden, Wolfram. 1950. Ein Zwiegespräch Ḫammurabis mit einer Frau (Altbabylonische Dialektdichtungen Nr. 2). ZA 49: 151–94. Westenholz, Aage. 1987. Old Sumerian and Old Akkadian Texts in Philadelphia 2. Carsten Niebuhr Institute Publications 3. Copenhagen: Museum Tusculanum Press.

36

Antoine Cavigneaux

Wilcke, Claus. 1992. Diebe, Räuber und Mörder. Pp. 53–78 dans Aussenseiter und Randgruppen, ed. Volkert Haas. Xenia 32. Konstanz: Universitätsverlag Konstnaz. Woods, Christopher. 2006. Bilingualism, Scribal Learning, and the Death of Sumerian. Pp. 91–120 dans Margins of Writing, Origins of Culture, ed. Seth L. Sanders. OIS 2. Chicago: The Oriental Institute. Ziegler, Nele. 2007. Les Musique et les musiciens d’après les archives royales de Mari. FM 9. Paris: Société pour l’Étude du Proche-Orient Ancien.

Jerrold S. Cooper

“Enlil and Namzitara” Reconsidered Four decades ago, Miguel Civil (1977) published “the brief, but fortunately complete tale” (Civil 1977: 65) of “Enlil and Namzitara.” In that publication, Civil deployed all the virtues ascribed to him by Samuel Noah Kramer (1991): “a sharp eye and a photographic memory” that made him “a master at identifying” Sumerian literary fragments, “making ‘joins’ whenever possible, and assigning them to the right compositions;” a “special flair and talent for matters technological;” and “an affinity for, and attraction to, lexicographical problems and details” which made him “the unchallenged master” of Sumerian lexicography. To Kramer’s concluding encomium – “original, innovative, creative, extraordinarily productive, and (justifiably) self-confident” – one only need add warm, congenial, and extraordinarily generous. The initial publication of a Sumerian literary text is never the last word on that composition, and a number of scholars made small improvements to our understanding of “Namzitara” before it was re-edited together with the later bilingual version from Emar (published only after Civil’s first edition; see Civil 1989: 7) by Bendt Alster in 2005 (327–38).1 More recently Yoram Cohen (2010, 2013) and I (Cooper 2011) have returned to the composition, Cohen more interested in the wisdom aspects of the text, especially in its Late Bronze Syrian iteration,2 whereas I focused on the Old Babylonian version and its puns, as well as the interpretation of a recurring Sumerian phrase as nam-mu tar-ra “Bless me!” rather than nam mu-tar-ra “the one who determines destinies.” Wilfred Lambert (2013: 286–87) offered a different take on that phrase in his explication of our tale’s Enmešara episode, for which see below. Yet after all has been said and done, Miguel’s succinct description of the composition forty years ago remains absolutely on target: [“Enlil and Namzitara”] is a Sumerian variation of a well-known theme, attested in the folklore of many cultures: a) D (= a deity, a holy man, or a semi-divine hero) in disguise meets H (= a man), b) H recognizes the true identity of D, and c) D rewards H with a blessing, material goods, or spiritual powers (Civil 1977: 65).

What follows is a presentation of the manuscripts and text with translation of Old Babylonian “Enlil and Namzitara,” an examination of the composition’s

1 See the bibliography in Alster 2005: 327; Cohen 2013: 163. 2 In addition to the Emar version a fragment of the Akkadian translation has turned up at Ugarit (Arnaud 2007: 141–42, pl. xxii no. 47). https://doi.org/10.1515/9781501503696-003

38

Jerrold S. Cooper

origin and role in the Nippur curriculum, as well as a discussion of certain aspects of its interpretation that remain controversial. More comprehensive manuscript details and textual matrices for the composition will be found at the end. Miguel Civil’s edition of “Enlil and Namzitara” utilized seven manuscripts, and those remain the only witnesses to the OB text. Three of those (C, D, E) are type II school tablets according to Civil’s typology (e.g., 1979: 5; 1995: 2308). The other four are on compilation tablets3 together with compositions that Miguel has dubbed the “Lisina-group:” the “Tale of Lisina,” “a short letter (?) of ca. 14 lines,” the composition “Nothing is of Value”, “Enlil and Namzitara,” and “a short composition … very incompletely preserved” (Civil 1977: 67). Of these five, only our composition and “Nothing of Value” (Alster 2005: 266–87) have been edited.4 Five of the tablets containing “Enlil and Namzitara” are short-line tablets.5 Ms. A has ten very narrow columns; in our composition, many of the lines of text as found in mss. B and F are spread over two and even three (1, 6, 10, 15, 17, 27) independent, that is, non-indented, lines. Line 18 is spread over four lines, and in 10, A breaks a short line in the middle of a verbal prefix chain. Mss. C, D, E and G also have short lines to varying degrees, but none with the consistency of A, and there is only one instance among them of a “normal” line spread over three short lines (G: 15). The lineation used below follows B and F.

Eclectic text  1 nam-zi-tar-ra den-lil2 mu-zal-le  2 inim in-na-an-du11  3 me-ta-am3 nam-zi-tar-ra  4 e2 den-lil2-la2-ta  5 bala gub-ba-mu bi2-silim-ma-am3  6 ki gudu4-e-ne-ka udu-bi-da i3-gub-bu-nam  7 e2-mu-še3 al-du-un  8 nam-mu-un-gub-be2-en  9 giri3-mu u4 ul4-ul4 106 a-ba-me-en za-e-me-en lu2 en3 mu-e-tar-re 3 For compilation tablets, see Kleinerman 2011: 57–74; Worthington 2008: 625–27. The German Sammeltafel is often translated into English as “collective tablet,” but “compilation tablet” is a less ambiguous rendering. 4 The Lisina tale is discussed briefly in Michalowski 1987–90: 32–33. 5 See Tinney’s (1995: 9–14) preliminary description of short line tablets. 6 See the mss. variants to this line in the textual matrices.

“Enlil and Namzitara” Reconsidered

11 127 13 14 15 168 17 18 19 20 21 22 23 24 25 26 27

39

ga2-e den-lil2-me-en d en-lil2-le igi-ni mu-ni-in-gi4 ugamušen-aš u3-mu-ni-in-ku4 gu3 al-de2-de2-e ugamušen nu-me-en den-lil2-me-en nam-mu tar-ra ga2-e den-lil2-me-en a-gin7 bi2-zu u4 den-me-šar2-ra šeš ad-da-zu še29-da-a nam-den-lil2 ba-e-de6-a u4-ne-en-gin7 nam ga-zu-e-še ku3 he2-tuku za he2-tuku gu4 he2-tuku udu he2-tuku u4 nam-lu2-lu7 al-ku-nu nig2-tuku-zu me-še3 e-tum3-ma d en-lil2-me-en nam-mu tar-ra a-ba-am3 mu-zu-um nam-zi-tar-ra mu-mu-um mu-zu-gin7 nam-zu he2-tar-re e2 lugal-za-ka e3-a ibila-zu e2-ga2 si-sa2-e he2-en-dib-dib-be2-ne

Translation9  1 Namzitara was passing by Enlil, (and)  2 He (Enlil) spoke to him:  3 “Where (are you coming) from, Namzitara?”  4 [N:] “From the temple of Enlil –  5 “I have completed my turn of service there;  6 “I serve with the gudu4-priests and their sheep, (but now)  7 “I’m going home.  8 “Don’t stand in the way!  9 “I’m in a hurry! 10 “(And anyway,) who are you to be asking me questions?” 11 [E:] “I am Enlil!” 12 (Then) Enlil changed his appearance, (and) 13 After he had turned into a raven (uga),10

7 Mss. A and D conflate lines 12 and 13, skipping from igi in 12 to u3 (igi.dib) in 13, then going back to 13’s beginning. The conflation must have first occurred in a ms. that did not omit the u3- in 13, as do D and E. 8 Ms. B omits this line, probably because of the den-lil2-me-en both here and in the preceding line. 9 For the many still unresolved difficulties, see Cooper 2011: 41–42. 10 Mss. A and D conflate 12 and 13: After Enlil had turned into a raven.

40

Jerrold S. Cooper

14 15 16 17 18 19 20 21 22 23 24 25 26 27

He (Enlil) was making bird calls. [N:] “You’re no raven! You’re Enlil! Bless me!”11 [E:] “How did you know that I’m Enlil?” [N:] “When Enmešara, your father’s brother, was captured, “You carried off Enlilship saying, ‘As of this day (u4), I shall (ga-) assign destinies/blessings!’” [E:] “You may acquire precious metals, you may acquire jewels, you may acquire herds, you may acquire flocks, (but when) “The time of a human being has come, “What good will your possessions be?”12 [N:] “You’re Enlil! Bless me!” [E:] “What 13 is your name?” [N:] “Namzitara (‘given a good/reliable destiny/blessing’)14 is my name.” [E:] “May you be blessed according to your name: “Be outstanding15 in the temple of your lord,16 (and) “May your heirs pass through my temple on a regular basis!”

Enlil’s Blessing There is general agreement that Namzitara earns Enlil’s blessing in lines 26–27 because he has seen through Enlil’s avian disguise (18), remembering the syllables in Enlil’s words quoted in l. 18 that combine to form the word uga “raven”. Only W. G. Lambert (2013: 286–87; 1989: 509) denied this, pointing out, quite logically, that Enlil had already identified himself as such in 11. But logic is not a necessary constitutive principle in this composition when it comes to identification: Enlil recognizes Namzitara by name in line 3, but asks his name

11 For the imperative nam-mu tar-ra and the translation “Bless me!,” see Cooper 2011. 12 Lit. “Where will your possessions be brought?” In other words, you can’t take it with you. 13 Lit. “who” (a-ba rather than the expected a-na), and though preserved in only one ms., it apparently was transmitted as such to Emar (Civil 1989: 7). 14 The name also has a meaning even more specific to this text: “Given a good/reliable temple position.” Prebends in OB were designated by nam + the position’s function, so nam-gudu4, nam-i3-du8, etc. 15 See Cooper 2011: 41. 16 Or possibly “in the house of your king;” see Cooper 2011: 41.

“Enlil and Namzitara” Reconsidered

41

in 23.17 In a recent article and book, Y. Cohen (2010; 2013) proposed that the bit of wisdom offered here in lines 20–21 (and expanded in the Late Bronze Syrian versions) is spoken not by Enlil, as usually supposed, but by Namzitara, that is, like Lugalbanda rejecting the initial blessings of the Anzu-bird, Namzitara is holding out for something he considers more valuable. According to Cohen, when Enlil offers a prosaic blessing of riches (OB 19), Namzitara rejects them on the grounds that men die and then their riches are of no use (OB 20– 21). He again demands a (real) blessing, and Enlil complies by granting him a hereditary prebend in Enlil’s temple, a blessing that will survive Namzitara and grace his descendants. For this to work, Cohen must assume that nig2-tukuzu in l. 21 means “the possessions I (Namzitara) get from you” rather than simply “your possessions” and that the Late Bronze Syrian version’s equivalent (B 17′) ku3-babbar-zu etc. = kasapka etc. must be understood not as Enlil addressing Namzitara (“your silver” etc.) but as Namzitara rejecting “the silver etc. I get from you (Enlil)” (Cohen 2013: 159). This is very awkward, as is, despite the parallels Cohen adduces (2010: 95–96), the notion of lowly Namzitara lecturing the chief of the pantheon. I (Cooper 2011) prefer to see Enlil attempting to wriggle out of having to bless Namzitara, or, perhaps, testing Namzitara to see if he is just trying to get rich, and telling Namzitara that a conventional blessing of riches is pointless (the vanity theme) because humans are in any case mortal. Namzitara insists (22) on receiving the blessing he is due, which will be a heritable prebend in Enlil’s temple (27), a blessing that will persist even after his death. The best argument for Cohen’s position is in the Ugarit Akkadian fragment’s translation of the Late Bronze Syrian Sumerian B 18′ me-še3 al-tum3 as [ayyik]â ˹alqe˺ anāku,18 which Cohen (2013: 155) understands as Namzitara asking where he could take the possessions Enlil might bless him with. But Arnaud’s interpretation (2007: 141) is equally plausible: Enlil is asking to what end he might give Namzitara those possessions, since, as the Late Bronze Age text continues, human life is limited.

17 Lambert was using Civil’s original lineation, in which l. 16 begins nam-mu-tar-ra, but, in fact, there is no ms. that begins the line with that phrase, and the omission of 16 by ms. B is easier to explain if that line begins as I have it here. 18 From Arnaud’s copy (2007: 256 no. 47) the alqe is far from certain, although on p. 140 he transliterates it without any brackets whatsoever. The anāku, however, is perfectly clear.

42

Jerrold S. Cooper

Namzitara’s prebend Miguel Civil (1977: 65) recognized that the tale he was publishing told “how Namzitara’s family … gained possession of a prebend, one of the most important sources of familial wealth in Old Babylonian Nippur.” In lines 4–6, Namzitara explains that he is coming from Enlil’s temple, where he has been serving “with the gudu4-priests and their sheep,” and in 27, Enlil proclaims that Namzitara’s heirs will be a regular presence in his (Enlil’s) temple. However, there were no gudu4-priests at Enlil’s temple Ekur in OB Nippur: it is the nu-eš3priest, who “ist im wesentlichen auf das Ekur Enlils in Nippur beschränkt … wo er dem gudu4-Kultp. anderer Heiligtümer entspricht.” (Sallaberger and Huber Vulliet 2005: 630). Moreover, the gudu4-priests “brachten vor allem unblutige Opfer dar.” There is no indication that they dealt with “blutige Opfer” (Renger 1969: 162), the sheep of our l. 6. These are things anyone in OB Nippur would have known, and it may be that, in addition to the many amusing puns laced throughout the composition, the entire premise of the story is a big joke. However, there are records from OB Nippur of a gudu4 prebend handed down through the family of Mannum-mēšu-liṣṣur (Wiebke 2015) said to be for a specific manifestation of Enlil called Enlil-urumah-anki, who was worshipped in Ekurigigal (Richter 2004: 44 n. 198), which was not just a part of Enlil’s temple Ekur, but a separate structure with a courtyard, storehouse, its own gate and a variety of cult installations (Richter 2004: 43–45; Michalowski 2006: 157–59). So it may be, rather, that a student or scholar in OB Nippur would have known that a reference to gudu4 at “the temple of Enlil” situated Namzitara not in Ekur, but in the Ekurigigal, a temple with links to both the curricular use of the Namzitara tale (see below) and to the Enmešara myth invoked in lines 17–18.

Enlil and Enmešara In our lines 17–18, Namzitara recalls that when Enlil’s uncle Enmešara was captured, Enlil took for himself the “Enlilship,” that is, the leadership of the pantheon that enabled him, Enlil, to assign fates and functions (nam). Lambert (2013: 281–98) has edited a first millennium myth that related how Marduk defeated and captured Enmešara, and subsequently reorganized the pantheon. Unfortunately, the preserved manuscripts begin only after Enmešara has been captured, and are only fragmentary as well. Therefore, the details of the battle between Marduk and Enmešara remain obscure, as does, necessarily, the ear-

“Enlil and Namzitara” Reconsidered

43

lier story of the conflict between Enlil and Enmešara that lies behind the first millennium myth. As Lambert (2013: 284) notes, “the best-attested characteristic of Enmešarra is that he was a primaeval god who held power before the gods worshipped by the historical Sumerians and Babylonians.” For this reason, I prefer to see the battle between Enlil/Marduk and Enmešara as a succession myth, rather than, with Lambert (2013: 287), as a story of a rebellion put down. Lambert insists that the “Enlilship” in our composition “would hardly have been called ‘Enlilship’ if it had first, and legally, belonged to any god other than Enlil.” But Lambert himself goes on to describe how, in Enūma Eliš Tiamat gives “Anuship” to Qingu, long before it becomes Anu’s much later in the story. Also, the name en-me-šar2-ra “lord of the myriad cosmic powers” implies that the god once legitimately controlled those powers (pace Lambert 2013: 288), before they passed into the possession of Enlil, and later Marduk. If, in fact, Namzitara was serving in the Ekurigigal and not the Ekur itself (see above), then his evocation of Enmešara and the succession myth takes on added significance. A first millennium cultic commentary (OECT 11: 69) reports that Enlil married Šuziana, the daughter of Enmešara, and ina libbi ekurigigal dingir ilmassima “the god had intercourse with her inside the Ekurigigal” (Gurney 1989: 27). Šuziana was venerated at Nippur (Richter 2004: 86–89; Such-Gutiérrez 2003, I: 299–302),19 and the collection of Sumerian temple hymns already calls her dam ban3-da a-a den-lil2-la2 “secondary wife of father Enlil.” Since our text calls Enmešara Enlil’s paternal uncle (17), if the first millennium report that Šuziana is Enmešara’s daughter holds for earlier periods, Enlil has taken his patrilateral parallel cousin (father’s brother’s daughter) as a bride, a choice which remains preferred today in the Middle East. Again, assuming the first millennium report reflects earlier traditions, Enlil enjoyed his honeymoon with Šuziana in the very temple where Namzitara served as gudu4-priest. Perhaps this is why Namzitara was so familiar with Enlil’s history with Enmešara, and was thus able to see through his disguise. He remembered Enlil’s words after he took over Enmešara’s powers – u4 … ga-zu “now I will

19 Šuziana’s cult in Nippur figures in the post-Old Babylonian comic story “Why do You Curse Me?” (Foster 2005: 937–38 with bibliography) In that composition, a doctor from Isin arrives in Nippur to collect his fee from a patient who resides there, and is told that the patient is not home, but has gone to present an offering the temple of his god, Šuziana. The joke here must be due to the equation of Šuziana with Gula, the goddess of healing and titulary deity of Isin, in later texts (Krebernik 2011–13: 377–79). The hilarity is compounded by both the office held by the physician, sanga of Gula, and the injury for which he treated the patient, a dog bite. Gula was traditionally associated with and symbolized by a dog.

44

Jerrold S. Cooper

assign” (18) – and reinterpreted them, as Miguel Civil (1977: 67) told us, to read uga zu “recognize the raven.”20

Namzitara and the Nippur Curriculum In his 1995 discussion of “Ancient Mesopotamian Lexicography,” Miguel Civil (1995: 2308) describes his Type II class of cuneiform tablets containing lexical lists: Type II is limited to the period of the Old Babylonian schools. Such tablets contain divergent material on each of its two sides. To the left of the flat side (II/1) there is a carefully written lexical passage extracted from a fuller list, apparently the work of an instructor, while to the right the passage is copied by a student. On the convex side (II/2) of a Type II tablet, there is a multicolumn excerpt from a longer list.

These tablets “predominate in Nippur, … are less common at other sites and are almost unknown at Ur.” (Tinney 1998: 45) Three of our “Enlil and Namzitara” manuscripts are II/1 tablets (C, D and E), preserving only the left-hand side.21 Type II tablets belong to the elementary level of scribal education, and usually contain lexical texts, the lists of signs and values that are learned by a beginner as well as more advanced lists, or the model contracts and proverbs that constitute the next step in the curriculum.22 “Enlil and Namzitara” is one of the very few literary texts to appear on Type II tablets (Veldhuis 1997: 65–66),23 and thus stands near the very beginning of a student’s initiation into the corpus of Sumerian literature.24 Steve Tinney has noted that the short-line format of mss. A, C, D, E and G is typical for texts “used early in the course of scribal training.” While unavoidable perhaps for Type II/1 tablets (C, D and E), the use of short lines on other multi-column tablets (A and G) is likely because the student, having learned

20 The precise significance of the raven in our text remains mysterious. The bird is sometimes associated with Enlil, but with other gods as well (Veldhuis 2004: 299–301; Weszeli 2007: 210– 13). P. Michalowski reminds me, also, that in Wilcke’s (1996) reading of the “Figure aux plumes,” Enlil’s son Ningirsu is identified with a raven. 21 Because the student’s work on the right-hand side was repeatedly rubbed smooth and rewritten, that side of the tablet grew ever thinner and eventually broke or was broken off, so that many Type II tablets preserve only the left-hand side (Tinney 1998: 45; Veldhuis 1997: 32). 22 For the Nippur elementary curriculum, see Veldhuis 2014: 204–12. 23 To Veldhuis’s list, add, for Lipit-Eštar B, Peterson 2011: No. 176. 24 For the literary curriculum, see Delnero 2010; Kleinerman 2011: 75–81.

“Enlil and Namzitara” Reconsidered

45

the composition on Type II/1 tablets, “laid out” his tablet “in a manner … reminiscent of the type II texts.” (Tinney 1995: 11). Another feature of our mss. that derives from the position of “Enlil and Namzitara” early in the literary curriculum is the complete absence of the ligature an+en in mss. A, C, F and G. In D, the two signs are clearly separate in 11 and 17, but in 12 and 15 the ligature must have been present because the preserved end of the en indicates that it stands at the beginning of the line (so no room for a separate an), and it was probably present in 16 and 18 as well. It is noteworthy that C and D are both Type II/1, so models theoretically inscribed by an instructor.25 The other Type II/1 tablet, E, does use the ligature (although the bit of the tablet preserved has only one possibility, as is the case for C as well). Neither A nor G use any determinative before the name Enmešara; D has an unligatured determinative there, as does B, which otherwise always uses the ligature with Enlil. When our composition does not occur on Type II/1 tablets, it is found on compilation tablets26 (A, B, F and G) with one or more other short compositions. Most literary compilations fall into two groups: literary letters and associated miscellany, and wisdom compositions. According to A. Kleinerman (2011: 58) “there is very little overlap among these … groups,” and, looking at her list of so-called wisdom compilations (2011: 65), it is clear that even among these, certain compositions tend to group together, and not all are “wisdom,” no matter how that term may be defined. This is certainly true of Civil’s Lisinagroup (see above), which is named after its longest composition, a lamentation of the goddess Lisina over her dead son. There may be a tenuous connection between Kleinerman’s “Sumerian Epistolary Miscellany” and our composition. Four of the twenty-two compositions Kleinerman’s epistolary compilations are not letters; the longest and best known of the four is the Tumal Text (Kleinerman 2011: 139–143; Michalowski 2006), a fictional history of the sanctuary of Ninlil at Tumal, south of Nippur, featuring father-son pairs of Sumerian kings, of which the father builds part of Enlil’s temple complex at Nippur, and the son “made Tumal resplendent” (pa bi2-i-e3) and led Ninlil thither. The final line of the text reads, simply “Išbi-Erra built Ekurigigal, the storehouse of Enlil.” That is, this staple of the Nippur schools concludes its history of the sanctuaries of Nippur’s chief god and goddess with the construction of Ekurigigal, the very temple where Namzitara may have served (see above). “Enlil and Namzitara” is set in Nippur and the name Namzitara itself was borne by an Ur III governor of Nippur from the distinguished Urmeme family.27 25 The misshapen ba in C’s rendering of our l. 5 suggests that perhaps a more advanced student or teaching assistant was responsible for the model, and not the instructor himself. 26 For the term, see above, n. 3. 27 As pointed out to me by P. Michalowski. For the Urmeme family, see Zettler 1984.

46

Jerrold S. Cooper

A. Kleinerman (2008) has discussed the use of the names of historical Nippur personages of the Ur III period in the literary Nippur correspondence and related texts that were studied by students in Old Babylonian schools. She cited Veldhuis’s (2004: 40) perceptive remarks on the relationship between OB Sumerian literary compositions and specific cities. In her words, “As scribal themes were associated with Nippur, so, for example, epic heroes were associated with Uruk.” Although there is nothing specifically scribal about our composition, it was among the first texts an aspiring scribe learned in Nippur, and hence fitting that the name of an eminent Nippurian of centuries past was appropriated for the human protagonist.

Manuscripts28 A

B

C

D

3N-T326 (IM 58427; fig. 1) + 3N-T360 (A 30218; P274955) rev. Ten narrow very short-line columns (five per side) containing the entire Lisina group, “Enlil and Namzitara” beginning in 3N-T360 ix 8 (toward the top of the rev.) and continuing onto 3N-T326 for the rest of col. ix, returning back to 3N-T360 at the top of col. x, and ending on 3N-T326 x 8. The tablet’s contents are discussed by Alster 2005: 275–276, with photos of the Nippur casts there on pls. 36–39. The CDLI image of 3N-T360 is of the actual tablet; the photo of the cast of 3N-T326 in fig. 1 here is courtesy G. Rubio. CBS 4605 (P260904) A single-column tablet containing all of “Enlil and Namzitara” through the upper reverse, followed by the beginning of Civil’s text d). The last clear line begins dumu-sag lugal and is then very worn, as is the reverse in general. The lower two-thirds of the reverse looks as if it were erased, or else is just very abraded. The text of our composition is clear through line 18; the ends of 19–20 are quite unsure, based on the traces on broken lower right edge of the obverse. Line 23 is assumed from the single zu preserved on the lower edge, and 24–27 on reverse are quite uncertain. N 5149 (P229658) The top of a Type II/1 tablet, preserving lines 2–9 of our composition. The only short line occurs in 6, an exceptionally long line. N 5909 (P230162) Type II/1 (so Civil, but not certain), preserving lines 6–18, breaking lines only for the very long lines 6 and 15.

28 Images of all mss. except 3N-T326 can be found online at CDLI under the P-numbers indicated.

“Enlil and Namzitara” Reconsidered

47

Fig. 1: Photo of the cast of 3N-T326 (IM 58427), obverse and reverse, at the Oriental Institute, University of Chicago (G. Rubio).

48 E F

G

Jerrold S. Cooper

CBS 7917+N 4784 (P230161) Type II/1 fragment preserving 10–14. Short lines at 10 and 12. UM 29-16-79A (P256684) Upper right fragment of a two-column tablet, with the Lisina lament on the obverse, and lines 19–27 of “Enlil and Namzitara” on the reverse, ending at the bottom of col. iii. According to Civil, the tablet contained the entire Lisina-group, but what is visible of col. iv seems empty except for a clear bar in the middle of the column, so the entire group may not have been present. Alster 2005: 267 mentions 79B, “a tiny chip,” yet UM 29-16-79B (P256685) seems to be a complete Ur III tablet. CBS 6924+N 3097 (P264351) Obverse (?) fragment of compilation tablet with three or more columns per side. The right-hand column contains “Enlil and Namzitara” 15–18 in shortline format; to the left is nig2-nam nu-kal version D (Alster 2005: 276–9). The entire tablet is transliterated separately by Alster 2005: 281, with photo on pl. 34.29

New Textual Matrices for Namzitara (+=sign present, –=sign omitted, .=sign partly present, o=entirely broken, x=unidentified sign, *=determinative not in ligature with en, |=new line) 1 A B

nam-zi-tar-ra d.en-lil2 mu-zal-le + + + + | +*+ + | + + + o . + . . + + + + .

A B C

inim + o .

A B C

me-ta-am3 nam-zi-tar-ra + + + + + + + o . + + + + + + + + + + + +

2

3

in-na-an-du11 + + – + + + – +! + + + +

29 Alster seems to have learned of CBS 6924 late in the preparation of his manuscript, and didn’t realize that it was joined with the same tablet (N 3097) that he used in the same volume for his edition of “Enlil and Namzitara.” His transliteration of the joined tablet on p. 281 is imprecise.

“Enlil and Namzitara” Reconsidered

A B C

e2-d.en-lil2-la2-ta + +*+ + + + o o + + + x-kam + +*+ + + +

A B C

bala + + +

4

5

6

gub-ba-mu bi2-silim-ma-am3 + + + | . . . . + + + + + + – + +! + . + + –

ki gudu4-e-ne-ka udu-bi-da i3-gub-bu-nam A + . + . . | + + de3 | + + . x B + + + + ke4! + + + + + + + C + + + + + + + + | + + . + D o o . + + + . o | o + + o

7 A B C D

e2-mu-še3 + . + + + + + + + o + +

A B C D

nam-mu-un-gub-be2-en + + + + + + + + + + + + + + + + + . . + + + + .

A B C D

giri3-mu . + + + . . + +

A B D E

(see individual mss.) a-[b]a-am3 za-e-me-en|lu2 en3 mu-e-|ši-tar-r[e-e]n a-ba-me-˹en za?˺-e-me-˹en lu2˺ en3 mu-tar-re a-ba-me-en za-e lu2 en3 mu-e-tar-re [a-ba]-me-en za-e | [l]u2 en3 mu-e-tar-re

8

9

10

al-du-un |+ . + + + + + + + + + o

u4 ul4-ul4 + + + + + + . . . + + .

49

50

11

Jerrold S. Cooper

ga2-e A + + B + . D + + E . +

d

d

+* . +* +

en-lil2-me-en . . + + . + + + + + . + + + + +

12

en-lil2-le A +* + + + B + + + + D . . + + E o . + +

13

ugamušen-aš A + + +| B + + + D + + še3 E + + še3

igi-ni mu-ni-in-gi4 |– – – – – – + + + + + . – – – – – – | . – . . + +

u3-mu-ni-in-ku4 . + + + + + + + + o – + + . . – + + – +

A B D E

gu3 al-de2-de2-e . + + + – + + + + o + + + + . + + . + +

A B D G

ugamušen nu-me-en d en-lil2-me-en nam-mu tar-ra . + + + +|o . + + + | o . + + + + + + + + + + o o o o o o . + + + . | . + . . . + + + . o o o o o | +* . o o o | + + o o

A B D G

ga2-e d en-lil2-me-en a-gin7 bi2-zu o o +*+ + – – | o o + + (omits) . + . o . + + + + i3 o + + +*. o o o | + + . o

14

15

16

“Enlil and Namzitara” Reconsidered

17 A B D G

A B D G

nam + + o .

A B F

ku3 he2-tuku + + + . . . + + +

A B F

u4 nam-lu2-lu7 + + + + o o o o + + + +

18

19

20

21

d

u4 + + . +

– +* +* –

en-me- šar2-ra šeš ad-da-zu LU2xKAR2-da-a . . + + | + + + + | LU2xŠE3 + + + + d + + + + . o o + o + + . + + a + + . o o + + . o o o o o | + x o

d

+* + o .*

en-lil2 ba-e-de6-a u4-ne-en-gin7 nam ga-zu-e-še + + | + + + . | + + + + + | + + ++ + + + . + + + . o o o o + .! . . . . . o o o o o o o o o o o . o o o o o o o o o o o o o o

za + o +

he2-tuku gu4 he2-tuku udu he2-tuku + + | + + + + + + o o o o o o o . . o o o o o o o

lu

al-ku-nu + + + o o .? + + o

(see individual mss.) A [nig2-tu]ku-zu me-še3 e-tum3-ma e-D[U?-?] F nig2-tuku-zu me

d

22

en-lil2-me-en nam-mu tar-ra A oo o + + | o o . + F +*+ + + + + + + .

23

a-ba-am3 mu-zu-um A o o o o o + B o o o o + – F o + + + o o

51

52

Jerrold S. Cooper

24 A B F

nam-zi-tar-ra mu-mu-um . . . o | . . + o o o o o + – . . . + + + +

A B F

he2-tar-re mu-zu-gin7 nam-zu . . + + + |[nam?]30 + + + o o o o o . . . + + + + + + + +

25

26

e2 lugal-za-ka e3-a A o o . + + + B (illegible) F + + + + + +

27

ibila-zu e2-ga2 si-sa2-e A o . | o o + + +| B o o . . . . – F + + + + + + –

he2-en-dib-dib-be2-ne o o + + + . . . x o o + + + + + +

Bibliography Alster, Bendt. 2005. Wisdom of Ancient Sumer. Bethesda, MD: CDL Press. Arnaud, Daniel. 2007. Corpus des textes de bibliothèquede Ras Shamra-Ougarit (1936–2000) en sumérien, babylonien et assyrien. Aula Orientalis-Supplementa 23. SabadellBarcelona: Editorial Ausa. Civil, Miguel. 1977. Enlil and Namzitara. AfO 25: 65–71. Civil, Miguel. 1979. Ea A = nâqu, Aa A = nâqu, with their Forerunners and Related Texts. Materials for the Sumerian Lexicon 14. Rome: Pontificium Institutum Biblicum. Civil, Miguel. 1989. The Texts from Meskene-Emar. AuOr 7: 5–25. Civil, Miguel. 1995. Ancient Mesopotamian Lexicography. Pp. 2305–14 in Civilizations of the Ancient Near East, vol. 4, ed. Jack M. Sasson. New York: Charles Scribner’s Sons. Cohen, Yoram. 2010. ‘Enlil and Namzitarra’: The Emar and Ugarit Manuscripts and a New Understanding of the ‘Vanity Theme’ Speech. RA 104: 87–97. Cohen, Yoram. 2013. Wisdom from the Late Bronze Age. Writings from the Ancient World 34. Atlanta, GA: Society of Biblical Literature.

30 Ms. A never indents lines, thus there should be a sign at the broken beginning of the line, preceding he2-.

“Enlil and Namzitara” Reconsidered

53

Cooper, Jerrold S. 2011. Puns and Prebends: The Tale of Enlil and Namzitara. Pp. 39–43 in Strings and Threads: A Celebration of the Work of Anne Draffkorn Kilmer, ed. Wolfgang Heimpel and Gabriela Frantz-Szabo. Winona Lake, IN: Eisenbrauns. Delnero, Paul. 2010. Sumerian Literary Catalogues and the Scribal Curriculum. Zeitschrift für Assyriologie 100: 32–55. Foster, Benjamin R. 2005. Before the Muses: An Antholology of Akkadian Literature. 3 rd ed. Bethesda, MD: CDL Press. Gurney, O. R. 1989. Literary and Miscellaneous Texts in the Ashmolean Museum. OECT 11. Oxford: Clarendon Press. Kleinerman, Alexandra. 2008. Money, Memory and Miscellany: Understanding the Presence of Historical Elites in Early Second Millennium Pedagogical Composition[s]. Kaskal 5: 173–86. Kleinerman, Alexandra. 2011. Education in Early 2nd Millennium BC Babylonia: The Sumerian Epistolary Miscellany. Cuneiform Monographs 42. Leiden: Brill. Kramer, Samuel N. 1991. Miguel Civil: An Appreciation. AuOr 9: 11–13. Krebernik, M. 2011–13. Šuzi-ana. RlA 13: 377–79. Lambert, Wilfred G. 1989. A New Interpretation of Enlil and Namzitarra. Or 58: 508–509. Lambert, Wilfred G. 2013. Babylonian Creation Myths. MC 16. Winona Lake, IN: Eisenbrauns. Meinhold, Wiebke. 2015. Das Vermögen der Familie des Mannum-mēšu-liṣṣur. ZA 105: 7–29. Michalowski, Piotr. 1987–90. Lisin. RlA 7: 32–33. Michalowski, Piotr. 2006. The Strange History of Tumal. Pp. 145–65 in Approaches to Sumerian Literature. Studies in Honour of Stip (H. L. J. Vanstiphout), ed. Piotr Michalowski and Niek Veldhuis. Leiden: Brill. Peterson, Jeremiah. 2011. Sumerian Literary Fragments in the University Museum, Philadelphia. BPOA 9. Madrid: CSIC. Richter, Thomas. 2004. Untersuchungen zu den lokalen Panthea Süd- und Mittelbabyloniens in altbabyonischer Zeit. 2nd ed. AOAT 257. Münster: Ugarit-Verlag. Sallaberger, Walther and F. Huber Vulliet. 2005. Priester. A. I. Mesopotamien. RlA 10: 617– 40. Such-Guitiérez, Marcos. 2003 Beiträge zum Pantheon von Nippur im 3. Jahrtausend. 2 vols. Materiali per il Vocabulario Sumerico 9. Rome: Università degli Studi di Roma “La Sapienza.” Dipartimento di Studi Orientali. Tinney, Steve. 1995. On the Poetry for King Išme-Dagan. OLZ 90: 5–26. Tinney, Steve. 1998. Texts, Tablets, and Teaching: Scribal Education in Nippur and Ur. Expedition 40/2: 40–50. Veldhuis, Niek. 1997. Elementary Education at Nippur. Ph.D. diss., Groningen University. Veldhuis, Niek. 2004. Religion, Literature, and Scholarship: The Sumerian Composition “Nanše and the Birds.” Leiden: Brill. Veldhuis, Niek. 2014. History of the Cuneiform Lexical Tradition. Guides to the Mesopotamian Textual Record 6. Münster: Ugarit-Verlag. Weszeli, M. 2007. Rabe(nvögel). RlA 11: 210–13. Wilcke, Claus. 1996. Die Inschrift der “Figure aux plumes” – ein frühes Werk sumerischer Dichtkunst. Pp. 669–74 in Beiträge zur Kulturgeschichte Vorderasiens. Festschrift für Rainer Michael Boehmer, ed. Uwe Finkbeiner et al. Mainz: Von Zabern. Worthington, M. 2008. Sammemtafel. A. Mesopotamien. RlA 11: 625–27. Zettler, Richard. 1984. The Genealogy of the House of Ur-Me-me: a Second Look. AfO 31: 1–14.

M. Erica Couto-Ferreira

‘Let me be your canal’: some thoughts on agricultural landscape and female bodies in Sumero-Akkadian sources The vocabulary of love and sex has attracted much attention in Assyriology, and over the past few decades a large number of studies have been devoted to both the edition of texts, and the discussion of lexical items, figurative language, and representation of sexual matters. Within this trend, the female body in cuneiform sources and visual representation has also been scrutinized.1 If we stick to Mesopotamian cuneiform records, it becomes clear that the female body was envisioned in a number of different ways, depending on the aim and specificities of particular texts. Thus, there is the pathological body healing texts described, but also the sensual body of love songs and erotic compositions, the flawless body of cult and ritual performance, and the body of old age that becomes an object of derision. In each of these cases, the female body is re-created by way of emphasizing a particular set of qualities or characteristics instead of others. Hair, make-up, adornment, and the description of the female external genitals are central elements of the erotic body, while malfunction of body parts and alteration of physiological activities constitute the ground the pathological body is built upon. Aim and audience, therefore, determine what bodily image a text intends to provide. When it comes to elicit the realm of bodily pleasure, of allure and sexual proximity, Sumerian sources usually turn to agricultural and “domesticated vegetation” figurative language to describe it. Taking two Sumerian passages focusing on Inanna’s genitals as a starting point, it is the object of this article to explore, however briefly, the imagery of women’s bodies in terms of agricultural landscape. The results of this research are dedicated to my Doktorvater Miquel Civil in his 90th birthday, in admiration and gratefulness for his teachings and constant support. I have tried to approach topics he has worked on over the

1 The bibliography produced on the topic is too vast to be quoted in full. For a recent treatment of sexuality in ancient Mesopotamia, with an updated reading list, see Wiggermann (2010). Acknowledgements: This research is carried out within the project “Medical Systems in Transition. The Case of the Ancient Near East”, which is funded by the Cluster of Excellence “Asia and Europe in a Global Context,” University of Heidelberg. https://doi.org/10.1515/9781501503696-004

Agricultural Landscape and Female Bodies

55

years (lexicography, agriculture), making some hints to other fields that had attracted his interest, such as Ethnography and Linguistic Anthropology.

1 A goddess body The well-known passage A Balbale to Inanna as Nanaya (Inanna H), segment A, 21–26 makes use of an erotic vocabulary that focuses on an agricultural presentation of Inanna − Nanaya’s genitals, equating them with a canal, a field, a pastureland, and a furrow: id2 na-an-ba-al-le id2-zu he2-me-en a-ša3 na-an-ur11-ru a-ša3-zu he2-me-en mu-un-gar3 ki-duru5 na-an-kig2-kig2-e [ze2]-˹ba˺ kal-la-mu ki-duru5-zu he2-am3 [x (x)]-e ab-sin2-zu he2-am3 x tur-tur-me aš2-zu he2-am3 Do not dig a canal, let me be your canal. Do not plough a field, let me be your field. Farmer, do not search for a wet place; my precious sweet, let this be your wet place, … let this be your furrow, … let this be your curse! (Inanna H, segment A: 21–26).2 A second Sumerian text, Ni 9602, first edited by Kramer (1963), is extremely interesting for our case, since the range of realities Inannas’s genitals are compared to gets richer. gal4-la-ni šir3-ra mi-ni-[ib2-ra-ra] gal4-la-ni gar-ra? (var. šir3-ra) ne-en kak-x [x-a] si-gin7 (var. si-ge8-en) gišmar-gal-e k[eš2-da-mu] ma2-an-na (var. gišma2-an-n[a ]) ne-en eš2-la2-[a x x (x)] u4-sakar-gibil-gin7 hi-li gu[r3-ru-mu] kislah ne-en edin-na šub?-[ba-mu]

2 Transliteration and translation from eTCSL 4.08.08, with changes, accessed 27 March 2015; see Sjöberg (1977) for the editio princeps and different witnesses to the composition.

56

M. Erica Couto-Ferreira

a-ša3 uzmušen ne-en uzmušen dur2-[ra-mu] a-ša3-an-na ne-en a ma-ra-mu ma-a gal4-la-mu du6-du8-du8-a a ma--ra ki-sikil-men3 a-ba-a ur11-ru-a-bi gal4-la-mu ki-duru5 a ma-ra3 ga-ša-an-gen gu4 a-ba-a bi2-ib2-gub-be2 [Praises] her vulva in song: ‘(My) vulva that which …, Which is at[tached] to a large cart like a horn; [My] boat of heaven, that which is fastened with ropes, Like the new moon is full of allure; M[y] threshing floor, that which lays down (?) in the steppe, M[y] field of ducks, where the ducks sit; My high field, that which is watered, My vulva, a watered, opened elevation. I, the maiden, who will plow it? My vulva, the wet and watered ground, I, the young lady, who will station there an ox? (Dumuzi-Inanna P: ii 17–27; translation adapted from Sefati 1998: 220 and 224–25).4 In both cases, the primary object of the metaphor is Inanna’s vulva (gal4-la); while secondary objects her genitals are likened to include the canal, different kinds of fields, and other man-made landscape features, together with other realities such as the new moon and the horn of a wagon. The analysis of women’s bodies in terms of fields is not new, and has been dealt with in Assyriology and in academia in general.5 As previously noted, digging canals6 (id2 ba-al, id2 dun) and ploughing fields are metaphors referring to the sexual act.7 The action of the penis penetrating the vagina is evoked through the 3 This is the Emesal form for a gar-ra, in equivalence to Akkadian a.meš raḫāšu “to move or set the waters in motion,” and a.meš šaḫātu “to wash, rinse,” in Landsberger (1956: 33 l. 67 and 69), quoted in Sefati (1998: 232). 4 The passage is also dealt with in Lambert (1987: 27). A different translation of line 22 is provided by ePSD: “this waste land abandoned in the desert”. 5 See Stol (2000: 1–2) and Stol (2008: 137–38) for Mesopotamian, Biblical and Ancient Near Eastern examples of women’s bodies as fields to be ploughed. 6 For occurrences of the term ba-al “to dig” in agricultural contexts, see Civil (1994: 111–12). For canal-digging as a metaphor for sexual intercourse, as well as for the representation of female sexual organs as canals, see Tinney (1999: 35–36). 7 See Jacobsen (1976: 45) for the equation between sex and the act of ploughing the land. For similar expressions from Latin contexts, see Adams (1982: 154–55).

Agricultural Landscape and Female Bodies

57

images of hole-making (digging), and of the tooth (zu2) of the plough entering the furrow or breaking the soil, in ways that tend to emphasize the receptive nature of female genitals as well as their hollow physiognomy. Nonetheless, these similes can also be taken as extended metaphors that implicitly refer to a number of aspects of the female body, sexuality, and even the domestication of spaces in terms of socio-economic values. Following the philosophical statements that “only some of a metaphor’s primary subjects and some of its secondary subjects are explicitly referred to by any verbal expression contained therein”, and that “listeners must work the others out for themselves. In this respect, every metaphor leaves something implicit”,8 we will try to dig up some of those implicit elements that serve to link female genitals to fields and watercourses.9

2 Watercourses: id2 Fields and especially watercourses (gu2 id2 “riverbank”, ambar “marsh”) are favourite settings for sexual encounters, as narratives such as Enlil and Ninlil and Enki and Ninhursaga show. In Enlil and Ninlil, Ninlil bathes in the river despite her mother’s advice not to do it, and this brings Enlil’s sexual arousal as a consequence (Enlil and Ninlil, 23–29). In Enki and Ninhursaga, 65–66, Enki is presented excavating ditches and penetrating reed thickets with his penis, after which he will proceed to have intercourse with Ninhursaga as a sort of development of his phallic activities: giš3-a-ni eg2-a ba-an-ši-in-dun-e giš3-a-ni gi-a gir5-gir5-e ba-an-ši-gir5-gir5-e (Enki) was digging his phallus into the dykes, Plunging his phallus into the reedbeds (Enki and Ninhursaga: 65–66).10

8 Stanford Encyclopedia of Philosophy sub Metaphor, http://plato.stanford.edu/entries/ metaphor/, accessed 25 April 2015. 9 Westenholz (1992), in analyzing the metaphorical language of love, notes that gardens and orchards are places where sexual intercourse takes place, but they also become spaces that make possible the description of the lovers’ bodies. Metaphors, therefore, allow a multiplicity of layers of interpretation. 10 See eTCSL 1.1.1, accessed 27 March 2015. For a different translation, see Attinger (1984: 15).

58

M. Erica Couto-Ferreira

The preference for water-based sexual metaphors may be based on a number of reasons. As already stated, one is linked to the equation between canal digging and male penetration. Another reason may lie in the specificities of the vocabulary of water and irrigation: as Tinney has pointed out, there is a clear play between the concepts “water” and “semen”, both of which are intended by the Sumerian sign a (Tinney 1999: 36, who also notes how Enki fills the Tigris by ejaculating in it). To fill the canal with good water would be equivalent to fill the vagina with good semen. Parallels, from a terminological viewpoint, between the irrigation of fields and the impregnation of the female body are also found in Enki and Ninhursaga. The expression a dug4, which is a technical term employed in texts reporting irrigation from canal water, is applied to Enki in the act of ejaculating in Ninhursaga’s insides.11 The description of the sexual encounter between the two divine figures occurs, in fact, after a narrative praising the abundance of water in the city of Dilmun, so that the sexual passage may be considered a continuation of the previous discourse and can be inscribed within the water imagery characteristic of Enki (Espak 2010: 161–168 and 174– 184, with previous bibliography). d

en-ki-ke4 a ddam-gal-nun-na-ka-ni mi-ni-in-dug4 nin-hur-sag-ga2-ke4 a ša3-ga ba-ni-in-ri a ša3-ga šu ba-ni-in-ti a den-ki-ka3-ka d

Enki distributed his semen (lit. irrigated) destined for Damgalnuna. He poured semen into Ninhursaga’s insides, And she seized the semen in the insides, the semen of Enki (Enki and Ninhursaga: 72–74). On the other hand, the imagery of canals and canal work also turns up in childbirth contexts. An Old Assyrian incantation to help a woman in labor makes reference to the “daughters of the šassūrātu, seven and seven” intervening with shovels and earth baskets to open the gate of the parturient:

11 For the Sumerian vocabulary of irrigation, see Civil (1994: 68–69). ePSD gives šaqû ša eqli as the Akkadian reading of a dug4. See also CAD Š/2 sub šaqû B, pp. 24–28 “to give to drink, to water animals: to libate; to water fields, gardens, to irrigate”, esp. 3, pp. 26–28 for the irrigation of fields; Salonen (1968: 212–31) for the lexicon of watering and irrigation; Finkel (1980: 44 and n. 11) for the explanation of a dug4 as makāru ša mê “to irrigate, to flood” (CAD M/1 s. v. A, pp. 125–26, with the final commentary “The term makāru denotes the flooding of an entire field, whereas šaqû seems to refer to a method of irrigation which uses small ditches or furrows”), and šiqītu “watering, irrigation” (CAD Š/3 s. v., pp. 95–96).

Agricultural Landscape and Female Bodies

59

a-ra-aḫ-tum ar-ḫa-at (…) lá-áš-pur ú lu-wa-e-er a-me-er!-ú-at ša-sú-ra-tim 7 ù 7-ma: ma-ri-ki-/na ù ta-áp-šu-kà-tí-ki-na le-qé-a-nim-ma: ba-áb a-ra-aḫ-tim ḫa-ba-tum12 ḫu-ub-ta The Araḫtum canal 13 is quick. (…) Whom, you may say, should I send to the daughters of the birth goddesses, seven, indeed seven they are, with the orders: ‘Take your spades and your baskets! Forcefully open the gate of the cow!’? (Kt 90/k, 178: 2 and 8–14 in Michel 2004: 396; see also SEAL 5.2.2.2.) Spades and baskets are instruments of labor closely related to agriculture and canal work. In fact, the Akkadian terms marru “shovel” and tupšikku “earth basket”, which are the tools the daughters of the šassūrātu are exhorted to take in Kt 90/k, 178, become in Atra-ḫasīs the implements characterizing divine, and then human, toil.14 Both marru and tupšikku frequently appear together in contexts of building work and agriculture,15 and therefore they can be understood as instruments for taming and bringing order to an environment that is frequently hostile towards men. In a Neo-Assyrian ritual to stop abnormal bleeding in a female patient, the pathological state is conceptualized in an incantation using the imagery of rivers and canals. The procedure aims at holding back the excessive flow, and,

12 CAD H sub ḫabātu D, p. 12 “to move across, make and incursion, a razzia into enemy territory”. 13 For Araḫtum, a branch of the Euphrates, see George (1992: 351–56). 14 See Atra-hasīs I 337 for the gods burning their spades and baskets (or their picks and spades, allī marrī) used in construction and canal digging (Lambert and Millard 1969: 64–65, and pp. 46–47 lines 65–66 for the Old Babylonian recension; also George 1996: 158, lines 57′–58′). 15 CAD M/1 sub marru, pp. 287–90, especially meaning b, pp. 288–89 for examples of its specific use (agriculture, making of bricks, digging of canals, etc.); CAD T sub tupšikku, pp. 476–79 “hod (for carrying bricks), basket (for carrying earth).” See also Michel (2004: 407– 408).

60

M. Erica Couto-Ferreira

in relation to this, the female body is envisioned as a flooded field.16 As Steinert (2013: 10) puts it: “For instance, the incantation ‘Her blood is a carnelian river’ (nār sāmti damša) in BAM 235: 10–16 // BAM 236 rev. 1′–9′ from Aššur), describes the woman’s haemorrhage as ‘a carnelian river, a carnelian canal (atappu)’, the inside of her body as a meadow (tamirtu) which has to be dammed up (sekēru). In the same way, the incantation incipit partially preserved in the Neo-Babylonian tablet K. 263+10934: 42 (a collection of recipes against gynaecological haemorrhage from Nineveh) speaks of the woman’s canals, which need to be blocked”. On a metaphorical level, the “rebellious”, pathological body of a woman experiencing abnormal bleeding or a difficult childbirth can also be brought back to normal through intervention, which, in the case of Kt 90/k, 178, is equaled to the work of cleaning canals. If we move to the realm of sexuality within the institution of marriage, we may argue that the female body is equally domesticated, possessed, and brought to fruition when irrigated, ploughed, and cultivated.17 The vocabulary of landscape control can also be applied to the dominion of the body, since both realities, agricultural and hydraulic landscape, on the one hand, and bodies, whether diseased or not, on the other, can be intervened upon in order to command them.18 The ways in which the vocabulary of fields is employed in love compositions, as we are about to see, seem to reinforce this notion.

3 Fields and agricultural landscapes 3.1 a-ša3 a-ša3 is a general term for “field” and its association with sexual intercourse has already been noted above with regard to the expression “ploughing the

16 The verb sekēru pertains to the technical lexicon of water management (CAD S sub sekēru A, pp. 210–13), as well as tamirtu (CAD T s. v., pp. 119–122 “a type of agricultural, especially irrigated, land”, with final bibliography), which is suitable for crop production. 17 In fact, the act of ploughing only makes sense when planting of the seed and harvesting follows (see, for example, Išme-Dagan and Enlil’s chariot: 85–88 in Civil 1968: 5–6). See King (1994: 38–39) for the Greek distinction between “labor” or “work” (ponos, aiming at the production of legitimate offspring) and “play” (paidia, indicating pleasure) in sexual encounters (King 1994: 39). 18 See Bastien (1985) for an ethnographic case study of how human bodily physiology is explained in terms of topography and landscape. Interestingly enough, among the Qollahuayas the body metaphor serves also to explain socio-environmental relations.

Agricultural Landscape and Female Bodies

61

field”.19 The comparison between female genitals and fields, on the other hand, is common to many agriculture-based societies (see, for example, Boggione and Casalegno 2004: 22–23 sub Arare and Aratro).

3.2 kislah As for the term kislah,20 it alludes to the “threshing floor” (Civil 1994: 93), although the translation “uncultivated land”, “abandoned ground” is also possible in some contexts.21 In fact, the latter translation has been preferred by most scholars dealing with Dumuzi-Inanna P (see, for instance, Jacobsen 1976: 45; Sefati 1998: 224). Despite being tempting to see in the equation a reference to Inanna’s vulva as a field or plot of land that has yet to be ploughed, the metaphor construction may be grounded on other elements or, at least, recall further metaphorical relations. The kislah can be filled with water (Enki and Ninhursaga, 155),22 and since water metaphors play a substantial part in love and sexual figurative language, it may be argued the metaphor is built upon this fact. However, we could also be facing an indirect metaphor, where the act of threshing, especially by way of flailing the grain, would be equated with sexual intercourse.23 The repetitive rhythmic movements employed both in threshing the grain and in sexual relations, and the contact of the threshing

19 For the equation penis = seeder plough, see Lambert (1982); Lambert (1987: 29–30); Livingstone (1991: 6). 20 On the term kislah, see Goetze (1936), who proposed the translation “threshing floor”; Steinkeller (1989: 123–25 and n. 370 and 371). 21 See the Akkadian equivalents maškanu “threshing floor”, nidûtu “waste land”, terīqtu “uncultivated land”, turbalû “bare ground”. Goetze already distinguished two readings of the same set of signs (KI.UD), namely kislah for “threshing floor” and ki-gal8 “uncultivated field” (Goetze 1936: 151). 22 “Enki filled the dykes with water, he filled the canal with water, he filled the kislah with water” (Enki and Ninhursaga: 153–55). ePSD proposes in this case the translation “fallows”. 23 On threshing procedures and instruments in Sumerian texts, see Civil (1994: 95). See also Steinkeller (1990); Anderson (2004). Three are the threshing techniques reported by Civil: še bad(-ra2) “sledging”; še us2 “treading” (by animals); and še (giš) ra(-ah/an) “beating” with flails. Because of beating (of flails or animals legs on the cereal) and/or friction (of the threshing sledge against the cereal) lie at the core of these activities, they are suitable to be used as secondary objects in metaphors regarding sexual activity. For examples of “beating” to refer to the sexual act in Latin texts, see Adams (1982: 145–49). For examples taken from literary Italian, see Boggione and Casalegno (2004: 50–51) sub Battere. English literature from the 16th century offers a number of examples where “threshing floor” and “flail” are employed in reference to the vagina and the penis, respectively (Williams 1994: 11–13 sub Agricultural imagery; pp. 498–99 sub Flail; p. 1384 sub Thresh).

62

M. Erica Couto-Ferreira

sledge with the grain that remembers that of flesh against flesh, make analogical relations between the two activities possible. However, while digging, ploughing and mounting metaphors in allusion to sexual intercourse are well attested in cuneiform sources, beating, as far as textual evidence goes, doesn’t seem to have been taken as secondary object for the construction of metaphors. An exception to this is LKA 102 rev. 10–12 (Biggs 1967: 42), a ša2.zi.ga incantation where the precative of maḫāṣu (limḫaṣ) “may you hit/strike” is quoted together with other verbs of sexual congress such as rakābu (lirkab “may you mount”). Nonetheless, a particular Sumerian example from the Inanna and Dumuzi love songs may provide some hints on this direction, since forms of the verb “to beat” are utilized. The relevant passage pertains to Inanna and Dumuzi A (Sefati 1998: 120–27). It takes the form of a dialogue between Inanna and his brother Utu around the different phases of the processing of flax (gu) and the weaving of a garment (gada-mah, usually interpreted in this context as a reference to the bridal bed sheets). The terms employed to describe the preparation of flax are “to beat”, “to spin”, “to twine”, “to warp”, “to weave” and “to bleach”.24 When Inanna inquires who will process the flax for her, Utu answers that he will bring it to her beaten, spun, twined, and so on. Once the cloth is ready, when the goddess asks “who will lie down (nu2) with me?”, his brother reveals the identity of the bridegroom, Ama-ušumgal-ana. The composition has been differently interpreted by scholars. Thorkild Jacobsen understands the narrative of flax processing from raw material to the woven cloth as Utu’s subtle way to make Inanna understand that “binding arrangements [have been made] for her to marry Ama-ushumgal-anna” (Jacobsen 1987: 13); while Gwendolyn Leick takes it as a metaphor of the “ripening” of a nubile girl: “The linen sheets can also be understood as a metaphor for the girl herself. She is transformed from ‘a lovely young thing’, the green shoot, to the bathed and anointed, beautifully bedecked bride, who, like the sheets, lies on the bed in expectation of her groom. (…) The final purpose of the sheets and all their elaborate stages of manufacture is the wedding bed.” (Leick 1994: 81). In the light of cross-cultural and cross-linguistic parallels, however, it might be argued the composition also deploys a sexual meaning. On the specific, terms for “beating”, “weaving”, and so on could be alluding in this context to the physical act of sexual intercourse, or even to a sort of initiation in

24 The text makes use of the following vocabulary: ri (instead of ra) “to beat”, nu-nu “to spin”, tab “to twine”, ze2-ze2 “to warp/twist”, tag-tag “to weave” and dan3-dan3 “to bleach, clean”. For an analysis of the cultivation of flax and the production of linen in ancient Mesopotamia, see Waetzoldt (1980–83).

Agricultural Landscape and Female Bodies

63

lovemaking.25 Note in this regard that Jacobsen understands Inanna’s inquiries on who will process the flax as an expression of her youth and ignorance, an ignorance and inexperience that could also apply very well to the knowledge of the flesh. On the whole, Inanna and Dumuzi A might also account for sexual union, on the one hand, and marriage and social union, on the other, which are codified and explained in terms of cloth production.26

3.3 ki-duru5 ki-duru5, Akkadian ruṭibtu, names a “damp ground, irrigable land”, “wet spots”.27 Authors agree it means a plot of land that has been conveniently inundated before ploughing can take place.28 In the specific context of Inanna H, Sjöberg already noted that it functioned “as an expression for the vagina” (Sjöberg 1977: 24, commentary to line 22). There is a manifest intentionality in both Inanna H and Ni 9602 in insisting on the notion of humidity, which seems to hint at welcoming genitals and sexual fluids. Besides, the fact that ki-duru5 alludes to a field ready to be ploughed reinforces the semantic association between “ploughing the field” and “having sexual intercourse” that has been pointed out so far. The ki-duru5 is also mentioned in relation to sheep and pasture in The Shepherd and the Farmer, 66–68 (Sefati 1998: 330–31 and 333), and in one instance sheep are said to graze in the ki-duru5 (The Shepherd and the Farmer, 76, in Sefati 1998: 328): udu-zu u2-ki-duru5 he2-em-mi-gu7 “let your sheep

25 For examples of the vocabulary of wool textile manufacture applied to sexual intercourse, see Boggione and Casalegno (2004: 50) Battere la lana sub Battere 2; p. 93 sub Cardare. 26 See Scheid and Svenbro (1996: 53–82, esp. 63–75) for the chlaina (wedding cloak) representing sexual union and marriage, thus coupling at both physical and social levels, in ancient Greek sources. The authors discuss a number of examples that point to the bed sheets and the wedding cloak being one and the same thing, thus allowing textile imagery to provide further correlations between the immediacy of sexual intercourse, and the solidity of marital bonds (Scheid and Svenbro 1996: 61–62). It is also interesting to note that the bridegroom was expected to provide the wedding cloth to the bride (Scheid and Svenbro 1996: 63–66). We might wonder, therefore, whether Inanna and Dumuzi A might allude to a similar situation where Inanna’s future bridegroom is expected to provide the bridal cloth. For a number of different interpretations on this fact, see Ackerman (2008: 1 n. 2), with previous bibliography. 27 ePSD sub kislah, accessed 2 March 2015; Civil (1994: 69). See also CAD R sub ruṭibtu, pp. 437–38. 28 On the gan2-ki-duru5, Yamamoto states the “gan2-ki-a designates the state of the field soaked in water immediately before plowing” (Yamamoto 1980: 172–73); similarly, “a plot where preparatory flooding had already been performed” (Maekawa 1984: 73).

64

M. Erica Couto-Ferreira

eat grass on the moist land”. The similitude between genitals and wet spots, therefore, bring to mind a thicker net of multi-layered associations: the luxuriant pasture talks about the agreeability of a pleasant place that allows abundance and the proliferation of flocks, but it also establishes similarities between grass and pubic hair.29 What’s more, the fact that the ki-duru5 is presented as a setting propitious for the grazing of animals seems to address an implicit metaphoric relation, namely the linking of eating and food consumption with the performance of the sexual act.30

3.4 ab-sin2 The term ab-sin2, Akkadian abšinnu, šer’u “furrow”, goes further deep into the semantic relations between fields, agriculture, and female genitals.31 There is a formal association between the opening of the vulva and the shape of the furrow and, as previously noted, both can be subjected to ploughing. On the other hand, a link between the sprouting of vegetables and the birth of human beings, and therefore between soil productivity and female fertility, can also be adduced, so that, from the viewpoint of metaphor construction, the field is to the woman’s body what the furrow is to the vulva.32 Significantly, first millennium lexical list Antagal provides the equation ab-sin2 = šà-sur-rum “womb” (Antagal B 88), which is integrated into a section listing terms related to the concept “mother” (ummu, bantu, and agarinnu, in Antagal B 85–87, Cavigneaux 1985: 192).33 Both the furrow and the womb receive the seed, hatching it until the final product comes into being. This group of terms is placed in the list between those regarding fatherhood (Antagal B 80–84)34 and offspring terminology (Antagal B 89–92): 29 Jacobsen (1975: 82–83; 1976: 45) interprets most of the secondary objects Inanna’s vulva is compared to as allusions to her pubic hair. 30 The topic is far too vast to be dealt with in full in this article. A study on the relations between food consumption and sexual performance is being undertaken by the author. 31 For furrow terminology within Sumerian agricultural contexts, see Civil (1994: 173–76). For the Akkadian equivalents, see CAD Š/2 sub šer’u, pp. 327–330; CAD A/1 sub absinnu, p. 65. 32 For mythological examples of human beings and creatures being born through emersion from the earth (Theogony of Dunnu, The Myth of the Hoe, and so on), see Pettinato (1971: 30– 39 and 49–53). For a recent edition of the Theogony of Dunnu, see Lambert (2013: 387–95). 33 For bantu, a cognate from banû “to create”, see CAD B s. v., pp. 80–81; CAD A/1 sub agarinnu 2, p. 146; CAD Š/2 sub šassūru A, pp. 145–46 and AHw s. v., p. 15 “Mutterleib, Bassin”. Cf. Malku I 122–124 ba-an-tu, a-ga-ri-in-nu, šá-as-su-ri = um-mu (Hrůša 2010: 38). 34 Interestingly enough, line 83 reads ninda2 = it-tu-˹ú˺ “seed funnel”, which is quoted among other terms related to agricultural works and implements that evoke male sexual and reproductive functions. Note that (gud)ninda also means “breeding bull”.

Agricultural Landscape and Female Bodies

ama gul.šid amaa-ga-ri-in tun3 ab-sin2

um-mu ba-an-tum a-ga-rin-nu šà-sur-rum

65

“mother” “mother” “womb, mother” “womb, mother goddess”.

4 Other metaphors Other secondary items Inanna’s genitals are equated to further insist on landscape, farming and irrigation elements. The “field of ducks” (a-ša3 uzmušen, in Dumuzi-Inanna P ii 23) seems to insist on the watery element of female genitalia by referring to birds inhabiting lacustrine and marsh environments.35 It is unclear to me what other semantic correlations can be built upon this image. With regard to du6 (Dumuzi-Inanna P ii 25), it is a general term for “elevation”, but in the particular context of love compositions, its use points once again towards the specialized vocabulary of irrigation. More specifically, it is used in The Farmer’s Instructions to designate a feature of watering landscapes that allows the water to flow: “(When you have to prepare a field for irrigation,) inspect the levee, canals, and elevations which have to be opened (e pa5 du6 du8-u3-de3)” (see Civil 1994: 68, commentary to line 3; he also discusses du8 in the sense of “to open a water course”). The hypothesis seems to be confirmed by the expression a ma-ra = a gar-ra (Dumuzi-Inanna P ii 24, 25 and 27), which means “to irrigate” and occurs in farming contexts (Civil 1994: 68). The lexica of irrigation and sexuality seem to overlap once again. Outside the realm of agricultural metaphors, two other elements emerge from the two Sumerian fragments our analysis has begun with. The comparison with a horn attached to a wagon36 (Dumuzi-Inanna P ii 19, perhaps in reference to the tongue or shaft animals were harnessed or yoked to) represents the vulva as a sort of figurehead both preceding and almost crowning the female body; or, as the foremost part of the female body, the one that comes first. Last but not least, the comparison of the vulva with both a boat (ma2-an-na) and the new moon (u4-sakar-gibil) is probably based on formal similitudes. The thin nail-like portion of the moon that can be seen during this phase brings to

35 CAD U sub ūsu A, p. 282 “duck”. None of the examples quoted in CAD suggest a relation with sexual imagery; see Veldhuis (2004: 303) for uzmušen. Jacobsen (1975: 82–83; 1976: 45) offers a different interpretation. 36 gišmar / eriqqu “wagon, cart” (CAD E s. v., pp. 296–97). See Salonen (1951: 81–136) for chariots and their parts. For a different interpretation, see Jacobsen 1975: 82–83; 1976: 45.

66

M. Erica Couto-Ferreira

mind both the curvature of the labia and the opening of the vulva, just as the hull of a boat does, notwithstanding the fact that the crescent of the moon is explicitly represented and described as a boat in cuneiform sources (Verderame 2002: 37 n. 59 and p. 116 § V.3.2.4, with previous bibliography). On the other hand, u4-sakar-gibil is also employed in literary examples where the appearance of the new moon is used in comparisons that express wonder and beauty. In this sense, it becomes clear that female pudenda, when mentioned in compositions dealing with love and sex, can be compared to a broad number of realities than transcend the semantic field of agriculture.

5 Conclusions Images and analogies can be created following different patterns: formal similarities (furrow – labia or opening of the vulva; new moon and boat – vulva); “functional” attributes (creative and formative capacities of the womb are equated with brick molding, the baking of ceramics, and the melting of metal);37 physiological characteristics (humidity); cultural values (allure, pleasantness); graphic associations (word and orthographic play based on principles of cuneiform writing). The richness of this referential system, therefore, allow for multiple layers of interpretation of each metaphor. In the case study here provided, the reading of Inanna’s body is strongly informed by landscape and surroundings, more specifically, by the domesticated environment. Images present the female genitals as a landscape subjected to anthropization, based on man-made spaces that need to be managed and constantly cared for so as to ensure productivity. These metaphors culturally inscribe themselves in a context of a fragile and unstable ecosystem that demands constant human intervention. Note, for example, the centrality landscape domestication plays in The Debate of Hoe and Plough. The diatribe between the two implements concentrates in their respective capacities to make crops and fruit trees grow, as well as in their usefulness (or not) in building up houses, maintaining the irrigation system, and in general, in bringing abundance to the people. It is water management, artificial irrigation, and land exploitation what guarantees life and social reproduction.38 In this socioeco37 Probably because of their focus on seduction, courtship, and lovemaking, the Dumuzi and Inanna love songs leave out those images that emphasize the shaping capacities of the womb. For an in-deep treatment of the latter, see, for instance, Arnaud (1996). 38 For a literary account of a time where agriculture and irrigation did not yet exist, and the consequences this fact brought to the land, see, for instance The rulers of Lagaš in eTCSL 2.1.2, accessed 27 March 2015.

Agricultural Landscape and Female Bodies

67

nomic context, we may argue that female genitals and bodies, when described in contexts of courtship and marriage, are often seen as landscapes to govern and to be made fruitful, even when in the case of the Dumuzi and Inanna songs no explicit reference to the production of offspring is made. Besides, the analogy between female genitals and fields / watercourses carries further implications: it is also revealing of social roles of men as cultivators, managers, and even possessors of land and/or canals,39 and it also tells us something on the functions attributed to the male in sexual intercourse and reproduction. Although most images employed in love compositions present sexual intercourse as an act of pleasure, by using a whole range of references to watering and agriculture, they implicitly emphasize a reproductive discourse that contrasts with the immediacy and non-procreative profile of the images employed in, for example, the ša2.zi.ga texts (Biggs 1967). To sum up, the hypotheses presented so far don’t aim at invalidating previous interpretations on body metaphors and sexuality. Quite on the contrary, this article intends to reflect upon the multiple levels figurative language work at.

Bibliography Ackerman, Susan. 2008. Asherah, the West Semitic Goddess of Spinning and Weaving? JNES 67: 1–30. Adams, James N. 1982. The Latin Sexual Vocabulary. London: Duckworth. Anderson, Patricia C., Jacques Chabot, and Annelou van Gijn. 2004. The Functional Riddle of ‘Glossy’ Canaanean Blades and the Near Eastern Threshing Sledge. Journal of Mediterranean Archaeology 17: 87–130. Arnaud, Daniel. 1996. Le foetus et les dieux au Proche-Orient sémitique ancien. Naissance de la théorie épigénétique. RHR 213: 123–142. Attinger, Pascal. 1984. Enki et Ninḫursaĝa. ZA 74: 1–52. Bastien, Joseph W. 1985. Qollahuaya-Andean Body Concepts: A Topographical-Hydraulic Model of Physiology. American Anthropologist NS 87: 595–611. Biggs, Robert. 1967. ŠÀ.ZI.GA. Ancient Mesopotamian Potency Incantations. TCS 2. Locust Valley, NY: Augustin. Boggione, Valter and Giovanni Casalegno. 2004. Dizionario del lessico erotico. Torino: UTET. Cavigneaux, Antoine, Hans G. Güterbock, and Martha T. Roth. 1985. The Series Erim-huš = anantu and An-ta-gál = šaqû. MSL 17. Roma: Pontificio Istituto Biblico.

39 The owner of a field should plow and cultivate it. According to the Code of Hammurabi § 30, the landlord or tenant who doesn’t work his field may lose it in the hands of anyone taking care of it. For a number of examples, see CAD E sub errēšu, pp. 304–306 “tenant farmer, cultivator”.

68

M. Erica Couto-Ferreira

Civil, Miguel. 1968. Išme-Dagan and Enlil’s Chariot. JAOS 88: 3–14 Civil, Miguel. 1994. The Farmer’s Instructions: A Sumerian Agricultural Manual. AuOr Suppl. 5. Sabadell: Ausa. Cooper, Jerrold S. 1989. Enki’s Member: Eros and Irrigation in Sumerian Literature. Pp. 87– 89 in DUMU-E2-DUB-BA-A: studies in honor of Åke W. Sjöberg, ed. Hermann Behrens Philadelphia: University Museum. Espak, Peeter. 2010. The God Enki in Sumerian Royal Ideology and Mythology. Tartu: Tartu University Press. Finkel, Irving L. 1980. The Crescent Fertile. AfO 27: 37–52. George, Andrew R. 1992. Babylonian Topographical Texts. OLA 40. Leuven: Dep. Oriëntalistiek / Peeters. George, Andrew R. and Farouk N. H. al-Rawi. 1996. Tablets from the Sippar Library VI. Atraḫasīs. Iraq 58: 147–190. Goetze, Albrecht. 1936. The Meaning of Sumerian kislaḫ and Its Akkadian Equivalents. AJSL 52: 143–59. Hrůša, Ivan. 2010. Die akkadische Synonymenliste malku = šarru : Eine Textedition mit Übersetzung und Kommentar. AOAT 50. Münster: Ugarit-Verlag. Jacobsen, Thorkild. 1975. Religious Drama in Ancient Mesopotamia. Pp. 65–97 in Unity and Diversity: Essays in the History, Literature and Religion of the Ancient Near East, ed. Hans Goedicke. Baltimore: Johns Hopkins University Press. Jacobsen, Thorkild. 1976. The Treasures of Darkness. A History of Mesopotamian Religion. New Haven: Yale University Press. Jacobsen, Thorkild. 1987. The Harps that Once … Sumerian Poetry in Translation. New Haven–London: Yale University Press. King, Helen. 1994. Sowing the Field: Greek and Roman Sexology. Pp. 29–46 in Sexual Knowledge, Sexual Science: the History of Attitudes to Sexuality, ed. Roy Porter and Mikulas Teich. Cambridge: Cambridge University Press. Kramer, Samuel N. 1963. Cuneiform Studies and the History of Literature: The Sumerian Sacred Marriage Texts. PAPS 107: 485–527. Lambert, Wilfred G. 1982. Sum. nínda = Akk. ittû ‘father’. RA 76: 94. Lambert, Wilfred G. 1987. Devotion: The Languages of Religion and Love. Pp. 21–36 in Figurative Language in the Ancient Near East, ed. M. Mindlin London: School of Oriental and African Studies, University of London. Lambert, Wilfred G. 2013. Babylonian Creation Myths, Mesopotamian Civilizations. Winona Lake: Eisenbrauns. Lambert, Wilfred G. and Alan R. Millard. 1969. Atra-ḫasīs. The Babylonian Story of the Flood. Oxford: Clarendon Press. Landsberger, Benno. 1956. Emesal-Vocabulary (Series dimir-dingir-ilum). Pp. 1–44 in MSL IV. Roma: Pontificium Institutum Biblicum. Leick, Gwendolyn. 1994. Sex and Eroticism in Mesopotamian Literature. London: Routledge. Livingstone, Alasdair. 1991. An Enigmatic Line in a Mystical / Mythological Explanatory Work as Agriculture Myth. NABU 1991: 5–6. Maekawa, Kazuya. 1984. Cereal Cultivation in the Ur III Period. BSA 1: 73–96. Michel, Cecile. 2004. Deux incantations paléo-assyriennes. Une nouvelle incantation pour faciliter la naissance. Pp. 395–420 in Assyria and Beyond : Studies Presented to Mogens Trolle Larsen, ed. Jan G. Dercksen. PIHANS 100. Leiden: Nederlands Instituut voor het Nabije Oosten.

Agricultural Landscape and Female Bodies

69

Pettinato, Giovanni. 1971. Das altorientalische Menschenbild und die sumerischen und akkadischen Schöpfungsmythen. AHAW 1971/1. Heidelberg: Winter. Römer, Willem H. P. 1969. Einige Beobachtungen zur Göttin Nini(n)sina auf Grund von Quellen der Ur III-Zeit und der altbabylonischen Periode. Pp. 279–305 in Lišān mitḫurti. Festschrift Wolfram Freiherr von Soden zum 19 VI. 1968 gewidmet von Schülern und Mitarbeitern, ed. Manfried Dietrich and Wolfgang Röllig. AOAT 1. Kevelaer / NeukirchenVluyn: Verlag Butzon & Bercker / Neukichener Verlag. Salonen, Armas. 1951. Die Landfahrzeuge des alten Mesopotamien, nach sumerischakkadischen Quellen. Helsinki: Suomalainen Tiedeakatemia. Salonen, Armas. 1968. Agricultura Mesopotamica, nach sumerisch-akkadischen Quellen. Helsinki: Suomalainen Tiedeakatemia. Scheid, John and Jesper Svenbro. 1996. The Craft of Zeus: Myths of Weaving and Fabric, Revealing antiquity. Cambridge: Harvard University Press. Sefati, Yitschak. 1998. Love Songs in Sumerian Literature : Critical Edition of the DumuziInanna Songs. Rāmat-Gan: Bar-Ilan University Press. Sjöberg, Åke W. 1977. Miscellaneous Sumerian Texts, II. JCS 29: 3–45. Steinert, Ulrike. 2013. Fluids, Rivers, and Vessels: Metaphors and Body Concepts in Mesopotamian Gynaecological Texts. JMC 22: 1–23. Steinkeller, Piotr. 1989. Sale Documents of the Ur III Period. FAOS 17. Stuttgart: SteinerVerlag. Steinkeller, Piotr. 1990. Threshing Implements in Ancient Mesopotamia: Cuneiform sources. Iraq 52: 19–23. Stol, Marten. 1988. Old Babylonian Fields. BSA 4: 173–81. Stol, Marten. 2000. Birth in Babylonia and the Bible. Its Mediterranean Setting. CM 14. Groningen: Styx. Stol, Marten. 2008. Embryology in Babylonia and the Bible. Pp. 137–55 in Imagining the Fetus the Unborn in Myth, Religion, and Culture, eds. Jane Maria Law and Vanessa R. Sasson. Oxford−New York: Oxford University Press. Tinney, Steve. 1999. Ur-Namma the Canal-Digger: Context, Continuity and Change in Sumerian Literature. JCS 51: 31–54. Veldhuis, Niek. 2004. Religion, Literature, and Scholarship: The Sumerian composition Nanše and the birds with a catalogue of Sumerian. CM 22. Boston: Brill. Verderame, Lorenzo. 2002. Le Tavole I–VI della serie astrologica Enūma Anu Enlil. Nisaba 2. Messina: Di.Sc.A.M. Waetzoldt, Hartmut. 1980–83. Leinen (Flachs). RlA 6: 583–94. Westenholz, Joan G. 1992. Metaphorical Language in the Poetry of Love in the Ancient Near East. Pp. 381–87 in La circulation des biens, des personnes et des idées dans le ProcheOrient ancien, eds. Dominique Charpin and Francis Joannès. Paris: Ed. Recherche sur les Civilisations. Wiggermann, Frans A. M. 2010. Sexualität (sexuality) A. In Mesopotamien. RlA 12: 410–26. Williams, Gordon. 1994. A Dictionary of Sexual Language and Imagery in Shakespearean and Stuart Literature. 3 vols. London: Athlone Press. Yamamoto, Shigeru. 1980. The Agricultural Year in Pre-Sargonic Girsu-Lagash I. ASJ 2: 169– 87.

Franco D’Agostino

The Eridu Project (AMEr) and a Singular Brick-Inscription of AmarSuena from Abū Šahrain In collaboration with Salvatore Monaco

§ 1 In 1991, during my Doctorate, I had the wonderful opportunity to be hosted as visiting student at the Oriental Institute of the University of Chicago for several months.1 I still have in my heart the beautiful memories of my period in that Institution, and the kind and warm helpfulness of all the staff there, from M. Stolper, who had the patience to give advices to a very unripe student of Neo-Babylonian, to E. Reiner, R. Biggs, G. and W. Farber, J. Brinkman, G. Gragg … And, of course, Miguel Civil. My Sumerian interests made me look for him very often in those months, with questions and requests of help (to him I owe, for instance, the Maori grammar I bought in Chicago in order to “rethink” my way to think the [Sumerian] grammar, as he advised me to do), and it is a pleasure for me to acknowledge to him here how much I learnt from him and from his very unique and innovative way to look at the Sumerian language and literature. Although I could enjoy of his mastery only for an (unfortunately too) short time, I hope he will accept this brief paper as a (surely too) small token of my respect and affection for him, notwithstanding the fact that my activities in the last years are much more archaeological than philological: this contribution will focus on a new, for me absolutely thrilling, archaeological project.2 And on some philology too … § 2 In August 2015 I received the permit of excavating the city of Eridu (Abū Šahrayn) by the Iraqi Authorities of the Ministry of Tourism and Antiquities and of the State Board of Antiquities and Heritage (SBAH), and the project will be carried on at Sapienza, University of Rome, by an Italian-Iraqi team

1 At that time I was working on my dissertation on Nabonidus, a task somehow imposed by my Doktorvater, G. Pettinato, who told me that I had to follow the “German” system (the only one he knew): MA Dissertation on 3rd Millennium Sumerian, Ph.D. on something else … I admit today that this attitude helped me very much to manage documents (or try to) of very different taxonomical and historical value. 2 I am Director, together with Dr Licia Romano, of the Iraqi-Italian excavations at Abu Tbeirah since 2010 (see e.g., D’Agostino and Romano 2014); on the new project on Eridu see below at §§ 2–3. https://doi.org/10.1515/9781501503696-005

Brick-Inscription of Amar-Suena from Eridu

71

Fig. 1: View from the North of the Ziggurat and the E’abzu complex (all pictures taken in October 2014), which encompasses an area of 600 × 500 m. You can see the sand deposits that cover the surface of the Tell during the violent summer sand-storms, a climatic characteristic that will make it very laborious the work of excavation at the beginning of each season. In the foregreound the traces of the ED Palace (Mound 2).

(Fig. 1).3 This will be a new and fascinating adventure, which will resume the archaeological activities on the 7 mounds (plus Usalla) of the oldest southern city of Mesopotamia after 65 years of interruption.4

3 HE Dr L. Sumaisem, the former Minister of Tourism and Antiquities and Dr A. Kamil, former Director of Excavations, must be warmly thanked for their convinced support to our project for Eridu. I want to acknowledge here the important and enthusiastic help for our work in Southern Iraq of HE Dr Habeeb Mohammed Hadi Ali Al-Sadr, Iraqi Ambassador to the Holy See, and of HE S. Barzani, Iraqi Ambassador to Italy. 4 Two studies on the history of the excavations at Eridu from Taylor to Fuad Safar are foreseen as a preparation to the digging activities – we are planning the first campaign for the period of Autumn 2016.

72

Franco D’Agostino

As is known, the first activities in Eridu were carried out by Esq. J. G. Taylor in 1854,5 followed ca. 65 years later by R. Campbell Thompson in 19186 and by H. R. Hall in the next year.7 But the most important diggings and soundings in Abū Šahrayn and the surrounding mounds were undertaken by Fuad Safar and Mohammed Ali Mustafa, with the cooperation of Seton Lloyd for the first two seasons, between 1946 and 1949.8 The wonderful results of the three campaigns have been published in Safar et all. 1981, itself a wonderful book of synthesis, and, since then, no other direct information from the city and its remains has been made known. It is worthwhile to stress here that, beyond the textual data, all what we know on Eridu derives from diggings, which, taken all together (1854–1949), lasted for less than 15 months.9

5 On this interesting figure of adventurer and man of culture, see Sollberger 1972, where some letters of Sir H. Rawlinson to him and sketches drawn by Taylor’s hand of Eridu’s remains can be found; his names are often abbreviated as J.E., instead of the correct J.G., because of an error in the Taylor 1855 paper (not corrected by the Author) and often repeated (Sollberger 1972: 129 n. 2); see also Lloyd 1974: 130–34, with the following comment: “Nevertheless, it is now possible to say that, with the limited means at his disposal, and in his total ignorance regarding the implications of what he was finding, he accomplished all that could be expected: also that his observations were intelligent and logical”. New documentary material that Dr J. Gattai, grandnephew of Mr Taylor, put generously at my disposal will be presented in the article cited in the preceding footnote. On the excavations at Eridu before 1946, see also Safar et al. 1981: 34–35, where Lloyd 1974 is repeated almost verbatim. 6 See Campbell Thompson 1920 and Lloyd 1974: 134–36, with the comment: “It is perhaps unfair to criticize the methods of these early archaeologists, when large-scale excavations under improved conditions, have thrown so much light on aspects of the ruins which were naturally incomprehensible to them. Yet, in the case of Campbell Thompson, one may perhaps be permitted some regret that the expenditure of so much labour and energy should have produced almost nothing that Taylor had not already found, or that could not have been deduced from a further examination of the mound’s surface,” though stressing further that Campbell Thompson’s geological and geographical observations, together with the synthesis on Eridu in the written tradition of Mesopotamia (Campbell Thompson 1920: 106–109) are useful contributions. 7 Hall 1923; 1925; 1930; Lloyd 1974: 136 ff. H. R. Hall is the first to hypothesize that Eridu was abandoned after the end of the 3rd millennium and only survived afterwards as a holy city. 8 See Safar et al. 1981: 41–48, for the general results of these three digging activities. 9 A visit to the remains was paid by J. Curtis, together with the then Director of Excavations (now Deputy Minister of Tourism and Antiquities), my good friend Dr. Qais H. R., Dr. M. Ali, then Vice-Director of the Iraq Museum, and Dr A. Al-Hamdani, who was at the time the Superintendent of Dhi Qar Antiquities, and other experts in order to look “for evidence of looting, and when it might have occurred; for evidence of military damage, whether before or after the coalition invasion in March 2003; and for evidence of damage deriving from nearly thirty years of neglect” – see Curtis et al. 2008: 218–20 (and note what is said on p. 217: “… we did not arrive at the site [Eridu, ed.n.] until 11.00 a.m. by which time it was uncomfortably hot and there was a strong wind blowing, affecting visibility and impairing our ability to work effi-

Brick-Inscription of Amar-Suena from Eridu

73

§ 3 Beyond the importance of Eridu as one of the most significant political and cultural centers of Mesopotamia between Proto-History and Early Dynastic, the choice of Abū Šahrain and the surrounding mounds as goal of a new project of excavation and research is due to the necessity of building up a new plan for the development of this archaeological area. In August 2015, in fact, a request with the proposal of entering in the Tentative World Heritage List of UNESCO the extraordinary Marshlands and Wetlands of Southern Iraq has been put forward by the Republic of Iraq, and to this request the proposal of inserting Ur, Eridu and Uruk in the same List has been added as well, based on the close cultural relations these cities held with the Marshlands (this is true especially for Eridu) since their very first foundation.10 The Project AMEr (Italian-Iraqi Archaeological Mission at Eridu) has accordingly been created, aimed at not only prepairing the archaeological work on the various mounds of the sites, but also at building up a plan for the harmonious development of the huge archaeological area involved by Eridu according to the UNESCO standard. Beyond the obvious necessity of restarting and expanding the works on Mounds 1 and 2 and on the Obeid Cemetery, the most thoroughly excavated and (more or less) known areas of the site, the task of AMEr is to investigate the mounds, which have not been so far dug or where only minor soundings have been done, which can be found on the immense surface of the tell.11 This second part of the project is especially important,

ciently on the site”) – a report on the present situation at Eridu is being prepared by the present writer. As for the tablets, which G. Pettinato and S. Chiodi claimed to have found in Eridu (500 documents of all the ED Periods), see Biggin and Lawler 2006; Curtis et al. 2008: 219–20. 10 The task of writing the file of approval to be delivered to UNESCO has been assigned to the Arab Regional Center for the World Heritage (ARC-WH), housed in Manama (Bahrein). The present writer has been working as expert for the cultural and historical aspects of both the cities of Eridu and Ur (for the Italian work at Ur in the last three years,see some information in https://www.academia.edu/7273009/Ur_Maintenance_Project; in Fall 2015 the last phase, devoted to the Ziqqurat, has been started) inside the Committee created in order to prepare the file (see http://arcwh.org/merging-natural-and-cultural-management-plans-iraqi-marshlands for more general information). 11 See Safar et al. 1981: 30 ff.: of the other mounds beyond 1 and 2, soundings have been dug on Mound 3 (300 × 150 m, h. 2,5 m; 2.2 km SSW of Abū Šahrain; “At its summit a trench was sunk where pottery of Isin-Larsa was encountered”), Mound 4 (600 × 300 m, no h. given; 2,5 km SW of Abū Šahrain; 5 trenches dug with “generally Kassite pottery, among which long beakers with solid base”), and Mound 5 (500 × 300, h. 3 m; 1,5 km SE of Abū Šahrain; 4 trenches dug, “which fixed the date of the settlement to Neo-Babylonian and Achaemenid periods”), while Mound 6 (300 × 200 m, h. 2 m; ca. 2,5 km SW of Abū Šahrain), 7 (400 × 200 m, h. 1,5 m; ca. 3 km E of Abū Šahrain; for both mounds, see the map, loc.cit., p. 31) and 8 (Usalla or ’Usaila; see loc.cit., p. 32) have never been investigated.

74

Franco D’Agostino

because it will shed light on the life of Eridu after its floruit in the IV and III millennia BC and will offer new information on the situation, both ecological and cultural, of the city during its period of “reduction” (itself a trait that should be confirmed, according to my point of view).12 § 4 During the stay in Southern Iraq in October 2014, for the excavations at Abu Tbeirah, a first visit to the tell of Eridu has been done.13 The main purpose of the two-days survey, which was limited to the mounds 1 (Ziqqurat Sacred Area) and 2 (Early Dynastic Palace), was to assess the present status of the remains. It must be stressed that the logistics is quite complicate as far as the diggings of Eridu are concerned: it lays in a depressed area 15 miles south-west from Ur (25 miles from Nāṣiriyyah) that can be reached only with a 4-Wheel Drive car because the last 4 miles are unpaved; it is impossible to arrive to the remains for days in case of rain – it is why we decided to work in the already hot months of April and May, when, admittedly, the reverse is true because of sandstorms. No village can be found in the area, apart from some Bedu tents pitched in the surrounding area especially in winter.14 The scope of this survey was to ascertain the safety conditions of the site, given the information of some newspapers in Iraq that the area was at danger because of remnants of war:15 fortunately, the two visits to the tell did not provide any evidence of hazardous material on its surface (Figs. 2 and 3).16

12 The first campaign is foreseen for Fall 2016 and will be devoted mainly to a general topographical survey of the area of the entire city, including the mounds around the real Abū Šahrain. 13 I want to acknowledge here the patience and helpfulness of the Co-Director of the mission at Abu Tbeirah, Dr Licia Romano, who was of invaluable help in organizing the logistics for the brief survey during the complicated and stressing period of an open excavation. 14 The feelings of the visitors of Eridu today are dramatically identical with the ones visitors had almost 100 years ago, according to the (quite literary) description of the site made by Sir R. Campbell Thompson in 1918: “From the zigurrat as far as the eye can see there is naught but awful solitude; you look down on sombre desert which encircles you for miles. Northwards lie the flat lands, yellow in April and unrelieved except for sparse arabesques of salt spreading like mares’-tails in a breezy sky, while afar, just visible as a little pimple in the mornings but blotted out in the afternoon haze, is the temple-tower of Muqaiyar …” (Campbell Thompson 1920: 105). 15 See Curtis et al. 2008: 219–20. 16 Insurance on the lack of such a danger on the tell was also given to us by the Directorate of the State Board of Antiquities and Heritage in Baghdad. To the survey the following people participated: M. Ramazzotti, V. Davidović, S. Monaco and A. Savelli.

Brick-Inscription of Amar-Suena from Eridu

75

Fig. 2: At the bottom of the ziqqurat, a sign warns visitors of the danger of ERW found on Tell Abū Šahrain, which was the scene of violent clashes. In the picture, one can see the remains of a military vehicle destroyed during that bloody period.

Fig. 3: The picture shows the watch-tower, now torn, built by the Italian Carabinieri in 2003, when they created a network of protection of the most important sites of Dhi Qar, including Eridu.

76

Franco D’Agostino

§ 5 Now, among the debris scattered all over the site, there were, especially on Mound 1, fragments of inscribed bricks.17 A few of them were taken to the house of Ur and there photographed for later study.18 They all belong to king Amar-Suena and have the standard 13 lines inscription very well known from Eridu and elsewhere mentioning the building of the E2-Abzu (Fig. 4).19 But two of them, both with 10 lines preserved, Brick A 1–10 and Brick B 4–13 respectively, present a particularity, which singles them out from the other inscriptions.20 The reconstructed text reads as follows (Fig. 4):  1  2  3  4  5  6  7  8  9 10 11 12 13

d

Amar-dSuena En-lil2-le Nibruki-a mu pa3-da saĝ-us2 e2 dEn-lil2-/la2!(LE)-ka lugal kal-ga lugal Ur/i5(ŠEŠ./AB)ki-ma lugal an ub-da / limmu2-bake4 d En-ki lugal ki-aĝ2-/ĝa2-ni-ir Abzu ki-aĝ2-/ĝa2-ni mu-na-an-du3 d

Amar-Suena, (whose) name Enlil in Nippur has pronounced, supporter of the temple of Enlil, the strong king, the king of Ur, the king of the four regions (of the Earth), to Enki, his beloved Master, his beloved Abzu he built for him.

17 See Curtis et al. 2008: 219: “there is evidence of erosion across the site, especially in two deep gulleys (of Mound 1, ed.n.) …, which has revealed inscribed bricks.” 18 We had the permit of taking samples from Eridu’s surface delivered by Dr A. Kamil, then Director of Excavations, and on behalf of the Superintendent of Ur and Eridu, Mr. Ali KadhemGhanim, a representative of SBAH came to visit the tell with us. However, the material was not given an identification number and all the objects were taken back to Eridu after photographing and studying. 19 See Frayne 1997: 260 ff. (15), where 55 exemplars are listed; Tyborowski 2001: 90; and in the text below. As is well known, the construction of the temple of Enki in Eridu (but not labeled according to its name E2-Abzu) has been started by king Ur-Namma, following what he states in his 6 lines standard brick-inscription found at Eridu (see Safar et al. 1981: 64–65, 228–29, no. 1; RIME3/2.1.1.10; Frayne 1997: 11 [e]) and on the stone fragment IM 52374 (Safar et al. 1981: 65; RIME3/2.1.1.31; see also RIME3/2.1.1.32) – the different epithets of the two inscriptions do not necessarily hint to a different period of their formulation, see D’Agostino and Pomponio in press (3.6.7). Eridu, together with Ku’ara, was most probably part of the core of the still in nuce Ur III Dynasty. 20 Autoptic analysis makes clear that the two bricks were inscribed with the same mold (see Fig. 4).

Brick-Inscription of Amar-Suena from Eridu

77

The presence of new bricks in Eridu with inscription of King Amar-Suena is, of course, no surprise, given the importance this king bestowed to Enki and Eridu.21 But the curious mistake in l. 6 is completely new: the wrong sign /le/, instead of the right /la2/, itself never appearing in the text of the standard 13 lines inscription, is unexpected and not to be found in the brick-inscriptions so far known, at the best of my knowledge. The error is much probably due to a kind of dittography, given the presence of the same sign four lines above.22 Another unique characteristic of this inscription is the final verbal form, where the presence of the sign /an/ is not attested in any other variant of it.23 The existence of this new variant to the standard 13 lines brick-inscription mentioning the construction of the E2-Abzu brings the number of molds prepared by the Chancellery of Ur for this event at least to five:24

Variant 1 d

Variant 2 ki

l. 2: En-lil2-le

l. 2: Nibru -a

l. 3: Nibruki-a

l. 3: dEn-lil2-le

Variant 3

Variant 4

Variant 5

l. 9: lugal an ubda limmu2ba-ke4

l. 9: lugal an ubda limmu2ba-ka

l.  6: e2dEn-lil2-/LE-ka

l. 13: mu-na-an-du3

For sure the high number of wooden molds created for remembering the building of the E2-Abzu must be understood through the importance king Amar-Suena bestowed to this event and its memory. Hints to this can be found both in the contemporary evidence as well as in later literary and omen tradition (see fn. 21 above). Among the many other important information, the new project of excavation will perhaps offer in the future a clue to understand the reasons for this attitude of the third king of the Ur III Dynasty.

21 See Frayne 1997: 236 ff.; for the later literary tradition on Amar-Suena, see also Zólyómy 2000. 22 Too scanty is the evidence for assuming a sort of Höhrfehler, with the corollary of the phonetic implications involved (see in general Jagersma 2010 [7.2 ff.]). It would be nice to think that the complex and beautiful /le/ sign has been incised because of aesthetic reasons (much like the sign /â/ in the French word théâtre …). 23 The infix -an- is quite rare in the Amar-Suena’s inscriptions, in accordance with the graphic rules of the period; it is to be found only four times: RIME3/2.1.3.5, l. 17 (gub); RIME3/ 2.1.3.6, l. 17 (du3); RIME3/2.1.3.9, l. 27 (e3); RIME3/2.1.3.13, l. 15 (dim2). 24 See Frayne 1997: 262, for these variants.

78

Franco D’Agostino

Brick A

Brick B

Graphic Reconstruction A+B

Fig. 4: The two Bricks A and B containing the variant al ll. 6 and 13 (Pictures and graphic reconstruction of Salvatore Monaco).

Bibliography Biggin, S. and Lawler A. 2006. Iraq Antiquities Find Sparks Controversy. Science 312: 173. Campbell Thompson, R. 1920. The British Museum Excavations at Abu Shahrain in Mesopotamia in 1918. Archaeologia 70: 101–44. Curtis, John et al. 2008. An Assessment of Archaeological Sites in June 2008: An Iraqi-British Project. Iraq 70: 215–37. D’Agostino, Franco and Licia Romano. 2014. Excavations at Abu Tbeirah, Southern Iraq. Pp. 189–204 in Proceedings of the 8 th International Congress on the Archaeology of the Ancient Near East, 30 April-4 May 2012, University of Warsaw, 2, ed. Piotr Bielinski et al. Wiesbaden: Harrassowitz.

Brick-Inscription of Amar-Suena from Eridu

79

D’Agostino, Franco and Francesco Pomponio. In press.The Ur III Period. In A Handbook of Ancient Mesopotamia, ed. Gonzalo Rubio. Berlin: De Gruyter. Frayne, Douglas. 1997. Ur III Period (2112–2004 BC). RIME 3/2. Toronto: University of Toronto Press. Hall, H. R. 1923.Ur and Eridu: The British Museum Excavations of 1919. The Journal of Egyptian Archaeology 9: 177–95. Hall, H. R. 1925. The Excavations of 1919 at Ur, el-’Obeid, and Eridu, and the History of Early Babylonia. Man 25: 1–7. Hall, H. R. 1930. A Season’s Work at Ur, Al-’Ubaid, Abu Shahrain (Eridu), and elsewhere. Being an unofficial account of the British Museum Archaeological Mission to Babylonia, 1919, London: Methuen. Jagersma, Abraham H. 2010. A Descriptive Grammar of Sumerian. Ph.D. diss. Leiden University. Lloyd, Seton. 1974. Abu Shahrein: A Memorandum. Iraq 36: 129–38. Safar, Fuad et al. 1981. Eridu. Baghdad: Ministry of Culture and Information. Sollberger, Edmond. 1972. Mr. Taylor in Chaldaea. Anatolian Studies 22 (Fs. Seton Lloyd): 129–39. Taylor, J. G. 1855. Notes on Abu Shahrein and Tel el Lahm. JRAS 15: 404–15. Tyborowski, W. 2001. Brick Inscriptions from Private Collections in Wielkopolska. N.A.B.U. no. 94: 89–91. Zólyomi, Gábor. 2000. Amar-Suena and Enki’s Temple. N.A.B.U. no. 51: 56–58.

Paul Delnero

The Silences of the Scribes, Pt. II: An Unfinished Enlil Lament from Nippur I still remember, with unforgettable clarity, the transformative experience of encountering the work of Miguel Civil for the first time. Reading his “Les limites de l’information textuelle” (Civil 1980), which had been assigned for a seminar on Mesopotamian archaeology, only a few days into graduate school, and learning that the Mesopotamian textual record probably concealed more than it revealed about the civilization I had come to study, changed in a fundamental way what I had always assumed, and never thought to question, ancient texts could tell us about the past. Far from rejecting the value of philology, however, the approach Civil proposed for using Mesopotamian texts as historical sources, begins, but does not end with establishing the literal meaning of a text as closely as possible through sound philological analysis. To determine what a text means, it is not only necessarily to understand what the scribes wrote, but also to be attentive to what they did not write: What is the source of the message? For whom was it intended? From what perspective was the message intended to be viewed? What modes of expression does the text employ and how does it relate to other texts in which similar modes of expression can be found? There is hardly an aspect of the Mesopotamian textual record that Miguel Civil has not immeasurably advanced our understanding of and this early encounter was to be the first of many to change profoundly what I thought Sumerian texts were capable of revealing about the time and place in which they were compiled. In the spirit of listening to “the silences of the scribes” (the title of Miguel Civil’s presidential address to the American Oriental Society in Toronto in 2001), it is a great pleasure to offer an edition of a short, unfinished Sumerian lament, and some preliminary remarks about textual transmission it raises, to a scholar whose contributions to Sumerian lexicography, paleography, grammar, and literature, and the study of Mesopotamian social, administrative, and legal history, among many other subjects, have moved the field of Assyriology well beyond its limits, into the 21 st century and beyond. Knowing how and why Mesopotamian texts were transmitted over time, and from one place to another, is of fundamental importance not only to reconstructing the history of Sumerian literary works, and their cultural significance at different periods and places, but also to addressing questions pertaining to Sumerian grammar and its diachronic development, the role of writing and copying in knowledge production and the formation of cultural identities, and https://doi.org/10.1515/9781501503696-006

The Silences of the Scribes, Pt. II: An Unfinished Enlil Lament from Nippur

81

the scribal practices that contributed to the preservation and eventual standardization of texts, in some instances for as long as nearly two millennia. In studying the transmission of Mesopotamian texts, and of Sumerian literary works in particular, one question that looms especially large is the effect of orality on written transmission. When the content of texts is fixed, the available source texts for producing a new copy are accurate, and great value is placed on reproducing every detail of the source text as accurately as possible, manuscript genealogies can be reconstructed, and many textual variants can be explained as one of a number of scribal errors that can be expected to occur during the copying process. But when these conditions are not met, as is almost always the case with literary and scholarly texts from early Mesopotamia, the possible role of orality in the transmission process must be taken into account, and a whole set of additional, more complex questions must be considered. Which texts might have been composed orally before being put into writing, and how might the original oral version(s) have differed from the later written version(s)? What changes might the content of these texts have undergone as they were performed or recited at different places and times? How, if it at all, could the changes introduced during oral transmission have influenced the written content of the compositions when they were copied in writing from one instance to another? Were there separate oral and written versions of the same composition in circulation at the same time, and if so, what influence, if any, did the oral and written versions have on each other? Or more broadly: what was the interface between orality and literacy in early Mesopotamia, and what differences might there have been in the nature of this relationship for texts of different types at different periods and places? As anyone familiar with the work of Miguel Civil knows, many of his most profound insights are buried in footnotes, or appear, without fanfare and wasted words, in a single sentence or short paragraph. One such observation captures the complexity orality introduces to textual transmission, and calls attention to the caution that should be exercised when reconstructing manuscript genealogies, which is seldom heeded in discussions of the relationship between the later and (very often hypothetical) earlier versions of Mesopotamian texts: “The existence side by side of a long, living version and a short written one would create a constant feeding and counterfeeding of textual variants. Without getting into details, it is easy to see how this could complicate in unexpected, and undetectable ways the genealogy of manuscripts.” (Civil 1999: 189).

This observation, which is included in a discussion of the Sumerian literary composition, Gilgamesh and Agga, was intended to apply specifically to the

82

Paul Delnero

large group of Sumerian hymns, narratives, and other literary genres that were copied by apprentice scribes during the Old Babylonian Period as part of their training to read and write Sumerian, which can be classified more broadly as “curricular texts”.1 In addition to this group of texts, many of which were originally composed at an earlier date, but are known primarily, and often exclusively from early second millennium school copies, which only rarely contain substantially different versions of the same composition, however, there is another group of literary texts, which were actively performed as hymns and liturgies in cultic rituals, for which the complicating effects of orality are even more pronounced. One type of texts that were both performed and copied in writing is the Sumerian laments, known as Balags and Eršemmas.2 Many of the factors that contribute to complicating the relationship between the oral and written version of Sumerian laments have already been noted by Civil, and are summarized in the following observations from his edition of the 10 th tablet of the Sumerian lament, uru2 am3-ma-i-ra-bi: “The reconstruction of the so-called ‘liturgical’ series, i.e., the balag-poems, is a notoriously difficult task. One reason is structural: these poems are composed of textual blocks that are repeated within the same balag, in other balags, and in eršemma compositions (see JCS 28 [1976] 72 ff.) making it very easy to combine fragments in an incorrect sequence. A second reason is the complicated textual transmission: a balag can be preserved partly in a tablet of the OB period and partly in NA or NB sources. The text is modified through the centuries by weak links of oral tradition or poor copies by inexpert scribes.” (Civil 1983: 45)

The reference to the discussion of “textual blocks” refers to an earlier article (Civil 1976), in which Civil presents the following observations on lines from a lament, which he calls “Enlil, the Merchant,” a text which is similar to the lament presented: “The text [Enlil, the Merchant] belongs to a period and literary genre in which new compositions are created almost exclusively as multiple combinations of more or less stan-

1 For the distinction between curricular and non-curricular Sumerian literary compositions, see Tinney 2011 and Delnero 2015. 2 The secondary literature pertaining to Balags and Eršemmas is vast. In addition to the primary editions of many of the known Balags in Cohen 1988 and Eršemmas in Cohen 1981, for a more recent discussion of Sumerian laments see Gabbay 2014 and Löhnert 2009, with extensive references to previous literature. Two older, comprehensive studies which contain invaluable editions and discussions of Sumerian laments that continue to provide a substantial basis for further studies are Krecher (1966) and Volk (1989), though there are also many others, including the innumerable earlier contributions to the subject by Jacobsen and Kramer.

The Silences of the Scribes, Pt. II: An Unfinished Enlil Lament from Nippur

83

dardized blocks a few lines long, which may appear in several compositions at the same time.” (Civil 1976: 72)

In summary, as noted in these comments, the reconstruction of the content of laments is complicated by: 1. The recurrence of the same passages, in identical or modified form, in multiple laments, which makes it difficult, and often impossible, to determine, on the basis of the passage alone, which composition the passage is from in the absence of any other features indicative of which composition the source contains (such as the incipit by which the composition was identified in antiquity, a colophon in which the composition is identified, or multiple sections that are known to occur in the same sequence in a particular lament). 2. Substantial differences between the earlier, Old Babylonian version and the later, first millennium version of the “same” composition, which are also often impossible to reconstruct, since the complete version of both the earlier and the later version of the text are not preserved in any of the existing sources. 3. Insufficient ties, particularly in later periods, between the oral performance of laments, and the written copies of the same laments. 4. Poor or inexpert copying. All four of these factors are amply evident in CBS 10417, an unfinished Old Babylonian copy of a Sumerian lament about Enlil, edited in this study. CBS 10417 is the lower half of a multi-column tablet, with what remains of 10–12 lines from near the ends of the first three columns on the obverse. The reverse of the tablet, which has the column rulings for three further columns, however, is completely uninscribed, and the presence of columns with rulings but without text is a clear indication that the scribe stopped writing after probably no more than three or four columns before completing the copy. Judging from the tapering of the left edge, which tapers down from the upper to the lower edge of the preserved part of the tablet, slightly more than the first half of each column is missing, but no more than five to ten lines are missing at the end of each column, indicating that when the tablet was complete, there were approximately 30 to 35 lines per column. Since the tablet breaks off at its thickest point, near the middle of the third column, it would also appear that the tablet is broken near the middle of the obverse, suggesting that there were probably two more columns after the broken third column. If this reconstruction is correct, then the complete tablet would have had five columns per side, with 30– 35 lines per column, and would have thus contained between 300–350 lines, had it been copied in its entirety.

84

Paul Delnero

There are three conceivable reasons why the scribe might have stopped writing before completing the copy: the act of copying was interrupted and never continued; the scribe did not need to finish copying the text, since only the lines copied were needed; or the copy had failed and a new tablet was needed to complete the copy correctly. While it is of course not possible to rule out completely that the scribe was interrupted, there are at least six other Old Babylonian copies of laments, which like CBS 10417, have also been left unfinished: – VAT 1367 (VS 2: 95), a single column tablet containing lines from a lament to Enlil, edited by Krecher (1967: 32–65), in which the scribe erased the last line on the reverse and then stopped writing, leaving the rest of the reverse uninscribed, before completing the copy. – VAT 1421 (VS 10: 179), a small, roughly copied, single-column tablet, with numerous poorly written signs and erasures, containing a short extract from a lament, in which the scribe stopped writing after six lines on the reverse, leaving the rest of the tablet uninscribed. – VAT 1557 (VS 2: 71), a single-column tablet, with lines from a lament similar to the Eršemma dutu e3-ma-ra, in which the scribe stopped writing in the middle of the reverse, before completing the tablet. – PRAK B 389, a single column tablet containing five lines from the beginning of the Balag uru2 am3-ma-i-ra-bi, after which the scribe stopped copying, leaving the rest of the obverse and the reverse uninscribed. – W 17259 m (AUWE 23: 127), a single-column tablet, containing four lines which partially duplicate lines from the beginning of VAT 3514 (VS 10: 154 − a source with lines that also occur in the 1 st millennium version of the Balag an-na e-lum-e), before the scribe stopped writing, leaving the rest of the obverse uninscribed. – CBS 10085 (BE 30/1: 11), a multi-column tablet, with two columns per side, containing a section of a lament in which the scribe stopped writing in the middle of the first column and left the rest of column i, as well as all of column ii, and columns iii and iv on the reverse uninscribed. Three of these sources (VAT 1421, PRAK B 389, and W 17259m) are almost certainly scribal exercise tablets, or tablets with hastily copied extracts that were not intended to be archived or kept for practical use. PRAK B 389 and W 17529m contain only 4–5 lines, and are too short to have been useful for any other purpose than practice, or ephemeral use, and VAT 1421, in addition to being short, is extremely poorly copied, and contains numerous erasures, indicating that it was copied by an inexpert, and probably apprentice scribe. VAT 1367 and CBS 10085, by contrast, contain longer, more carefully copied passages,

The Silences of the Scribes, Pt. II: An Unfinished Enlil Lament from Nippur

85

and are less likely to be scribal exercises; and VAT 1557 belongs to a distinct group of phonetically written sources, which appear to be have been copied to aid in performance,3 and to have been left unfinished because only a shorter passage from the text was needed. Of all of the unfinished tablets, CBS 10417 bears the closest resemblance to CBS 10085, which is the only tablet of the six that is a not a single-column tablet, and to VAT 1367, which is the only other tablet which contains a longer passage of text. CBS 10417, like these two tablets, is written in a more careful script than would be expected in an exercise tablet and contains a text that is too long to be a scribal exercise. But unlike CBS 10085 and VAT 1367, which are cleanly and correctly copied from beginning to end, CBS 10417 contains three lines that have been completely or almost completely erased, and numerous instances in which individual signs or groups of signs have been intentionally or unintentionally damaged or effaced.4 Since the source is otherwise carefully written, and like VAT 1367 and CBS 10085, is unlikely to have been copied as a scribal exercise, it is therefore probable that scribe stopped copying because mistakes had been made that were too substantial to eliminate without necessitating erasures and changes that would compromise the quality and readability of the completed copy and/ or involve redoing much of the copying work that had already been done. Because two of the lines that are fully effaced occur toward the beginning of the copy, in the middle of the first column, the scribe may have written these lines in the wrong sequence on the tablet, and realized the error only after completing three or more columns on the obverse, deciding then, after a failed attempt to undo the mistake by erasing the misplaced lines, to stop copying instead of continuing to produce a flawed and damaged copy. Moreover, if the misplaced lines had incorrectly altered the sequence of all the lines that came after them, it may have been impossible to continue copying without erasing all of the lines after the mistake that the scribe had already copied, making it easier to begin the work again on another tablet. CBS 10417 contains a lament addressed to the god Enlil. The tablet is written consistently in Emesal, the register of the Sumerian language in which the two mains types of lament, Balags and Eršemmas, are always written. Furthermore, since there would have been 300 or more lines in the completed copy

3 For VAT 1557, and the other tablets in this group, and their possible use in performance, see Delnero 2015. 4 Examples of damaged or erased signs include “da” in obv. ii line 5′, and the sequence of signs “la2” and “a” (the first sign in the copula -am3, written a.an) in the second part of the same line; “ra” in the second kur-ra-am3 at the end of col. ii line 8′; and “re” after kur2 in the second part of line 5′ in col. iii.

86

Paul Delnero

and there is a dividing line in the middle of the second column, which indicates the presence of multiple sections, and Eršemmas consist of only a single section and are rarely longer than 100 lines, it is likely that the lament on the tablet is a Balag. That this lament was addressed to Enlil can be inferred from the occurrence of the name of Enlil, in its Emesal form dmu-ul-lil2, in col. i 4′, along with the name of his city, Nippur, in col. ii 7′ and 9′, and the occurrence of epithets that are associated exclusively with this deity in col. i 6′ (u3-mu-un ka-na-aĝ2-ĝa2 ‘lord of the land’) and col. iii 1′–3′ (see commentary to these lines below). The passage that is preserved in col. iii of the tablet is duplicated in BM 81014 + BM 132267 (CT 58: 39A) col. i 3′–8′, another Old Babylonian source which also contains multiple sections of a lament to Enlil, further increasing the likelihood that the lament in CBS 10417 is also from a Balag addressed to Enlil. Enlil is one of the main deities who is addressed in Balags, and is the focus of no less than nine of the Balags listed in the first millennium catalogue 4 R2 53 (dutu-gin3 e3-ta, u4-dam ki am3-us2, e-lum di-da-ra, e-ne-eĝ3-ĝa2-ni i-lu i-lu, an-na e-lum-e, e-lum gu4-sun2, am-e bara2-an-na-ra, zi-bu-u3-um zi-bu-u3-um, and a-ab-ba hu-luh-ha), at least five of which are also known from Old Babylonian duplicates which can be securely identified as being copies of these texts (dutu-gin3 e3-ta, u4-dam ki am3-us2, e-lum gu4-sun2, zi-buu3-um zi-bu-u3-um, and a-ab-ba hu-luh-ha).5 Although the content of CBS 10417 cannot be directly attributed to any of the known Enlil Balags, it is possible that it contains lines from one of the many missing or incomplete sections or ki-ru-gu2s (the Sumerian rubric used to identify the individual sections of Balags) in one of these Balags, or another, unidentified Balag that is similar in content to one or more of these compositions. While the existence of an additional copy of an Enlil Balag is not unusual, CBS 10417 is of particular interest because the source itself is from Nippur, Enlil’s city, and the religious center of Mesopotamia during the Old Babylonian Period, when the tablet was copied. It has been noted that most of the known sources for Sumerian laments from the Old Babylonian Period were compiled at northern Mesopotamian cities like Kish and Sippar, implying that there are significantly fewer sources from southern Mesopotamia, particularly at Nippur, but also further south at cities like Ur and Uruk.6 But the number of preserved Old Babylonian sources from

5 For a recent, comprehensive discussion of Enlil Balags, with editions of the Old Babylonian and 1 st millennium versions of the Balags dutu-gin7 e3-ta and zi-bu-u3-um zi-bu-u3-um, see Löhnert 2009. 6 Compare, more recently, for example, Löhnert (2009: 19): “Zur geographischen Verteilung in der altbabylonischen Zeit läßt sich feststellen, daß die Emesal-Klagelieder hauptsächlich aus den Fundorten Nordbabyloniens stammen, wobei die Verbreitung durch einzelne Belege

The Silences of the Scribes, Pt. II: An Unfinished Enlil Lament from Nippur

87

Nippur for Emesal laments is not nearly as insignificant as this view could be seen to imply. In addition to CBS 10417, there are at least 43 published Old Babylonian sources for Emesal laments from Nippur,7 and around 30 further unpublished sources from the same period and place.8 With the prominence of Enlil, the chief deity of Nippur and the Sumerian pantheon during the third and early second millennium, in the theology of the Sumerian lament tradition it should not come as a surprise that laments were also compiled and performed at Nippur, and not only at the northern Mesopotamian centers with which the tradition tends to be associated. But the identification of CBS 10417 and at least 70 other Old Babylonian copies of laments from the city provide further evidence that the lament tradition was not confined to the north during this period. Even though the text in CBS 10417 is unfinished, and the top halves and the ends of the three inscribed columns on the obverse is broken, it is still possible to reconstruct at least some of its content and structure from parallels to the preserved lines in other laments, and to anticipate how the text might have continued had the scribe completed the copy on the basis of BM 81014 + BM 132267 (CT 58: 39A), which contains sections from the end of what is proba-

bis in den tiefen Süden hinein und auch außerhalb Babyloniens von Mari im Westen bis in das Dijāla-Gebiet im Osten gesichert ist.” 7 The forty-three published Old Babylonian copies of laments from Nippur comprise: 1. CBS 2218 + (Löhnert 2009: 144 + pl. II); 2. CBS 2227 (BE 30/1: 10); 3. CBS 2249 (HAV 13); 4. CBS 2281 (HAV 14); 5. CBS 3656 (BE 30/1: 2); 6. CBS 6890 (PBS 10/2: 17); 7. CBS 6891 (SEM 91) + CBS 6897 (STVC 83); 8. CBS 7285 (STVC 85); 9. CBS 7990 (STVC 122); 10. CBS 8547 (STVC 39) +; 11. CBS 10084 (BE 30/1: 7) +; 12. CBS 10085 (BE 30/1: 11); 13. CBS 10885 (HAV 3); 14. CBS 11151 (BE 30/1: 9); 15. CBS 11178 (STVC 32); 16. CBS 11326 (BE 30/1: 5); 17. CBS 11328 (BE 30/1: 8); 18. CBS 11330 (BE 30/1: 6); 19. CBS 11344 (HAV 15); 20. CBS 11349 (STVC 30); 21. CBS 11359 (PBS 1/ 1, 8; Löhnert 2009: pl. III); 22. CBS 11393 + (BE 30/1: 1); 23. CBS 11932 (MBI 5); 24. CBS 15204 (PBS 10/4: 5); 25. N 3073 + N 3312 (Peterson 2012b: n.44); 26. N 3390 (ASJ 18: 31–32); 27. N 4210 (Löhnert 2009: 116–17 + pl. X); 28. N 5662 (Marchesi 2006, pl. I); 29. N 6471 + N 7401 (Alster 1992) +; 30. UM 29-13-246 + N 6399 (Peterson 2012a: n.8); 31. UM 29-15-368 + UM 29-15-443 (ASJ 18: 33 n.4); 32. Ni 1577 (BE 31: 19); 33. Ni 2723 (BE 31: 17); 34. Ni 2327 (BE 31: 16); 35. Ni 2405 (SLTN 110); 36. Ni 2412 (SRT 46); 37. Ni 4357 (ISET I: pl. 246); 38. Ni 9798 (ISET I: pl. 185); 39. HS 1431 (TUM 4: 84) +; 40. HS 1453c + (TUM 4: 50); 41. HS 1494 (TUM 3: 26); 42. HS 1507 (TUM 4: 56); and 43. 2 N-T 226 (Civil 1983: 52–53). 8 The unpublished Old Babylonian sources from Nippur for Sumerian laments are listed in an unpublished catalogue compiled by Jeremiah Peterson, which he has kindly made available. The existence of some of these sources had also been noted by Miguel Civil, Jeremy Black, Steve Tinney, Anne Löhnert, and others who have worked on Sumerian laments from Nippur housed in the University Museum in Philadelphia, but Peterson’s catalogue extends the number of unpublished sources significantly, and includes a substantial number of fragments that he has identified and joined to both published and unpublished sources.

88

Paul Delnero

bly a similar composition. The content of CBS 10417 can be reconstructed as follows: – First section or Kirugu (preserved on the tablet) − col. i 1′−col. ii 4′: A goddess laments to Enlil, in the first person, about the destruction of Enlil’s temple, the Ekur, and the devastating effect its destruction has on the people who see and enter it. The invocation of a cow coming to lament the loss of its calf (col. i 8′–12′), which is an image typically associated with Inana (see commentary to col. i 8′ below), suggests that Inana is delivering the lament; but the reference to “my city”, referring to Nippur, may also be an indication that the lament is being delivered by Enlil’s wife, Ninlil and not Inana. The first preserved part of the lament (col. i 3′–7′) contains a description, with no known parallels, of the ancestors of Enlil bestowing Nippur to him in the distant past, and is presumably intended to serve as reminder to Enlil of his duty to protect and preserve the city which had been entrusted to him by his ancestors. The section concludes, after a break of 20–25 lines, with a bitter complaint about how the city has been taken over by thieves (col. ii 2′), followed by a description of the wailing and tearful lamenting that overcome anyone who approaches and enters the destroyed temple (col. ii 3′–4′). If the reconstruction of the tablet and this section of the text are correct, the first section of the text would have consisted of approximately 50 lines, with around 15 lines missing from the broken beginning of the section, and another 20–25 missing in the break between near the end of col. i and the lines from the end of the section after the break in col. ii. Determining whether the first section on the tablet was also the first Kirugu of the Balag, and not from a later Kirugu in the lament, however, is more difficult. Since there is not a single known source from the Old Babylonian Period or the first millennium which contains a complete Balag, with all of its sections, it is not possible to reconstruct with certainty what the standard structure of a typical Balag might have been. But if NCBT 688 – an Old Babylonian source containing the first four Kirugus of the Balag uru2 am3-ma-i-ra-bi in their entirety (source C in Cohen’s edition of the lament; see Cohen 1988: 541–45 and 587–90 for a transliteration and translation of this source) – is representative of a Balag with a typical beginning, the 1 st person lament of the goddess Inana does not begin until the second Kirugu, 67 lines into the text, and is preceded by a long, 60-line Kirugu, in which the destruction of the city is invoked in the third person, suggesting that the first Kirugu in CBS 10417, which contains a 1 st person lament by a goddess, may have also have been preceded by one or more Kirugus. If this is the case, then CBS 10417 might have also begun with a section from later in the composition, and not at the beginning of the text.

The Silences of the Scribes, Pt. II: An Unfinished Enlil Lament from Nippur

–

89

Second section (col. ii 5′-col. 9’f.): This section of the lament begins (in col. ii 5′) with the incipit: e2 zi-da uri3 gul-la-bi / uru2 lil2-la2-am3 uru2 lil2la2-am3 ‘The true house, its destroyed standard. The city is deserted, the city is deserted!’. The beginning of this incipit closely parallels the first line of the fourth Kirugu of the Balag a-še-er ĝi6-ta, as reconstructed by Black (1985), and occurs in identical form at the end of CBS 12674, an unpublished Old Babylonian fragment in which only this line, which occurs at the end of this tablet, is preserved, where it may be the catchline to the next section of a lament. The section continues (in col. ii 6′–9′), before the column breaks off, with another lament delivered in the first person, presumably by the same goddess (Ninlil or Inana), in which the goddess laments that Nippur has become deserted and its inhabitants have been forced to live in distant, hostile places. Nearly all of this passage is directly parallel to CBS 2218 + 2334 + N 3335 + N 3369 + N 4226 [source N3 in Löhnert’s (2009) edition on the Balag dutu-gin7 e3-ta; see Löhnert 2009: 144 for a description of the source and pl. II for a copy of the reverse of the tablet] obv. 8′–10′, which contains another Old Babylonian lament addressed to Enlil. When the text resumes (in col. iii 1′–9′), after a break of another 20–25 lines between the end of col. ii and the first half of col. iii, Enlil is being accused directly of neglecting to protect his city sufficiently. The lines in this section, which also occur in BM 81014 + BM 132267 col. i 1′–6′ and a number of other texts (see the reconstruction of the content of CBS 14017 and the commentary to the source below for a discussion of these parallels) include a description of Enlil staring at the ground instead of watching over his city (col. iii 2′) and handing over the young of the ewe and the goat to a hostile place or enemy instead of ensuring their protection (col. iii 4′–5′). After this section the column breaks off and the remainder of the preserved part of the tablet (on the reverse) is uninscribed.

If it is assumed that BM 81014+ – which begins with a section of a lament that has a passage (in col. i 3′–8′) that is identical to col. iii 1′–6′ of CBS 10417, and is preceded is by a break at the beginning of the column which is similar in length to the number of lines that would have preceded this passage in the second section of CBS 10417 – continued with sections that are similar to what would have followed in CBS 10417 had the scribe completed the copy, the broken and unfinished sections of CBS 10417 could have had something resembling the following structure: – The second section (as reconstructed from col. ii 5″–6″ of BM 81014+, which are followed by a dividing line marking the end of the section) might

90

–

–

–

Paul Delnero

have ended with description of the churns of the land no longer producing milk and burnt offerings ceasing to be made in the temples. A third section (as reconstructed on the basis of the second section of BM 81014+ in col. ii 7″–iii 11″ f.), in which Enlil continued to be addressed directly, but is praised for his destructive power and implored not to complete the annihilation of the land which he had begun, in a passage with lines that are closely parallel to BM 132095 (CT 42: 1) obv. 34–37 = BM 81014+ iii 6″–9″, a Neo-Babylonian tablet containing lines from the Balag a-ab-ba hu-luh-ha (see Kutscher 1975 and Cohen 1988: 374–400), and lines from the “Enlil, the Merchant” section of the first millennium version of the Balag dutu-gin7 e3-ta as preserved in K. 4613 (4 R 11) rev. 6–9 and 39–47 = MMA 86.11.182 (CNMMA 2: 3) rev. 1′–4′ = BM 81014+ col. iii 6–9 and 6″–10″, and numerous other sources with the same passage. A fourth section (in the preserved portions of BM 81014+ rev. columns iv and v), with parallels in the Old Babylonian sources PRAK C 99 obv. 3′– 8′ = BM 81014+ rev. iv 2′–7′; VAT 607 (VS 2: 11) + AO 3924 (TCL 15: 2)10 iv 9′–15′ and rev v 1′–11′ = BM 81014+ rev. col. iv 1′–7′ and col. v 4–12; and CBS 497 (PBS 10/2: 12) + VAT 1334 (VS 2: 12) + VAT 1357 (VS 2: 16)11 rev. col. iv 1′–2′ and col. v 1′–2′ = BM 81014+ rev. col. v 1–2 and 11–12, in which the first person lament to Enlil continues. A fifth, and final section (in the preserved section of BM 81014+ rev. col. vi 1–5, the last two lines of which are duplicated in CBS 497+ rev. vi 1′–2′, which continues with the remainder of the same section), in which Enlil is implored to restore the temples and shrines in his land.

As can be inferred from the description of CBS 10417 and the reconstruction of its content, this source possesses all the features that characterize the corpus of Sumerian laments noted by Civil. The tablet itself, while not poorly written, was inexpertly copied, and the scribe probably had to abandon the task of completing the copy as a result of mistakes made in the completed sections of the composition. Nearly two-thirds of the lines and passages in the source, including col. ii 2′, col. ii 5′–9′, and col. iii 1′–6′, are duplicated in at least one, and frequently more than one other lament. In the one instance in which a longer passage from this source is repeated in a source from the first millen-

9 Parallel source T in Cohen’s (1988) edition of the Balag e-lum gu4-sun2. 10 Source C in Cohen’s (1988) edition of e-lum gu4-sun2. 11 Parallel source J for the Balag uru2 hul-a-ke4 of Gula; see Cohen 1988: 258–60 and 264– 68 for a transliteration of this source.

The Silences of the Scribes, Pt. II: An Unfinished Enlil Lament from Nippur

91

nium,12 the passage occurs in a completely different context in a first millennium lament, obscuring how – and whether – the lament in CBS 10417 was transmitted into the first millennium. Lastly, the sheer range of parallels between lines and passages in this source with lines and passages in sources for similar and completely different laments, and conversely, the absence of a single source which has all or most of the same sections of the text in the same sequence, are clear indications that the content of individual laments – at least during the Old Babylonian Period – was more open to adaptation and modification for oral performance than it was fixed and standardized in a written tradition. For all of these reasons, and particularly this last, the process by which laments were transmitted in writing over time in ancient Mesopotamia should be seen as distinct from that of other types of literary compositions, whose transmission process is less variable and subject to a constant interplay of additional factors that complicate the relationship between different versions of the text (oral and written) at different places and times. Furthermore, in the absence of identical, or nearly identical written copies of the “same” lament, it may be more productive to view individual laments, even when they are identified by the same incipit, not as stable texts with consistent content, but instead as “variations on a theme”, which share a certain more or less fixed set of lines, passages, and themes, but whose content was also subject to constant modification and change each time the lament was copied or performed. Edition of CBS 10417 < > = a sign that is partially broken > = sign or signs that have been deliberately erased obv. col. i 1′ 2′ 3′

beginning of the column broken še […] / uru2-ĝu10-uš -[a dmu-ul-lil2] u3-mu- -[na-aĝ2-ĝa2] / /

12 Col. iii 1′–6′, a passage which occurs in almost identical form in the first millennium sources for the Balag mu-tin nu-nuz dim2-ma (Cohen 1988: 222–52), K. 2003+ (4 R2 28 4 = Cohen Source B) obv. 35-rev. 3 and VAT 245+ (SBH 46 = Cohen Source L) rev. 20′–16′.

92

Paul Delnero

 4′ a-a dmu-ul-lil2  5′ -ki dnin-ki-e / saĝ-eš2 bi2-rig7!(pa-hub2)-rig7!(.)-eš  6′ u3-mu-un ka-na-aĝ2-ĝa2  7′ den-ul dnin-ul- / saĝ-eš2 bi2-rig7!(pa-hub2)-rig7!(pa-hub2)--eš  8′ ab2 er2-ra mu-e-ši-in-  9′ u3-mu-un-ra ab2 er2-ra / mu-e-ši-in-ĝen 10′ (.ši)-ra mu-un-ĝen 11′ -me- mu-- 12′ [er2-ra] -un-ĝen 13′ […] rest of column broken col. ii  1′  2′  3′  4′  5′  6′  7′  8′  9′ 10′

beginning of column broken

-ga me nu-me-a- i-bi2-bi ir2-re / mu-lu mi-ni-in-ku4-ku4 ša3-bi a-še-er-re / mu-lu im-ta-an-e3 single dividing line on tablet e2 zi- uri3 -la- / uru2 lil2--(.an) uru2 lil2la2-am3 lil2-la2-am3 -- uru2-ĝu10 (.lil2) / lil2-la2- -- me-e uru2-ĝa2 dam-bi ki kur2-ra-am3 / dumu-bi ki kur2--am3 nibruki- -bi kur2-ra-am3 / dumu- - […] […] rest of column broken

col. iii  1′  2′  3′  4′  5′  6′  7′  8′

beginning of column broken --? [u6 di-de3 nu-kuš2-u3] gu2- -[al-la nu-gi4-gi4] ša3-zu bal-bal [en-še3 i3-kuš2-u3] u8 sila4 -[da] / kur2-re ba-e--[eĝ3] ud5 maš2 zi-[da] / kur2- ba--[ze2-eĝ3] tuku-? [eĝ3-ĝa2-ni] / mu-[un-e-til] ? ? […] / […] e2(-)? […] / ? […]

The Silences of the Scribes, Pt. II: An Unfinished Enlil Lament from Nippur

93

 9′ […] rest of column broken rev. preserved part of reverse divided into three columns that were left uninscribed Translation obv. col. i beginning of the column broken (break of approximately 10–15 lines)  1′ … / toward my city ?  2′ The lord of /  3′ /  4′ (To) father Enlil  5′ Enki and Ninki gave as a gift.  6′ (To) the lord of the land  7′ Enul and Ninul gave as a gift.  8′ The calf, lamenting, came to you.  9′ To you, the lord, the calf, came lamenting. 10′ Lamenting, it came 11′ To the personal god, 12′ [Lamenting], it came. 13′ … (break of approximately 5 lines) col. ii  1′  2′  3′  4′  5′  6′  7′  8′

(break of 10–15 lines)

My robbed temple without mes! Lamenting, one enters before it (the temple). Wailing, one comes out from inside it. single dividing line on tablet True house, its standard destroyed − the city is desolate, the city is desolate! It is desolate, it is desolate! My city, Nippur! It is desolate! It is desolate! I am in my city (while) its wife is in a foreign place, its son is in a foreign place!

94

Paul Delnero

 9′ (I am) in my Nippur (while) its wife is in a foreign place, its son is in a foreign place! 10′ … (break of approximately 5 lines) col. iii  1′  2′  3′  4′  5′  6′  7′  8′  9′

(break of 10–15 lines) [Your] eyes do not tire [from keeping watch]. Your neck does [come back from] staring at the ground. Your heart, turning about [- when will it cease]? The true ewe and lamb are given to the enemy. The true goat and kid are given to the enemy. [You have put an end] to the property [of those] with [property]. … My/true(?) temple … … rest of column broken

Commentary col. i 1′: The sign after še has been completely erased, but since the word še-eb ‘brickwork’ occurs frequently at the beginnings of lines in laments, particularly in litanies involving temples and shrines, it is not unlikely that the scribe had originally written a line that began with še-eb before later erasing the entire first part of the line. col. i 2′: The scribe has intentionally erased the entire second part of the sign “ka” and had probably deliberately erased everything after it as well. The entire second part of the line has also been deliberately, and very thoroughly, erased. Since the erasure in the second part of this line, like the erasure in the second part of the next line, continues into the beginning of the adjacent line in the next column, and nothing was written over the erasure after the surface of the tablet in this part of the line had been completely effaced, it is very probable that the erasure was made only after the scribe had finished inscribing the first and second column. The writing of the erased verbal form, saĝ-eš2 bi2-rig7rig7-ge-eš, is almost certainly an error influenced by the occurrence of the same verbal form in the second part of line 5′. While it is conceivable that An and Uraš were the intended agents of the verb that were written and then erased in the first

The Silences of the Scribes, Pt. II: An Unfinished Enlil Lament from Nippur

col. i 3′:

95

part of the line, since Enki and Ninki, who are the agents of the same verb in l. 5′, usually head the litanies listing the ancestors of Enlil, it is more likely that the scribe had written the first half of the line incorrectly, necessitating the erasure. Wherever it appears, u3-mu-un ka-na-aĝ2-ĝa2, is an epithet of Enlil, the deity who is being addressed throughout the lament. Since this epithet usually follows the invocation of Enlil as a-a dmu-ul-lil2 (see, for example, Kirugu 1 of the Balag dutu-gin7 e3-ta ll. 54– 61 = Löhnert 2009: 186–87), the error might have been writing the lines with a-a dmu-ul-lil2 and u3-mu-un ka-na-aĝ2-ĝa2 in the wrong sequence. The mention of cities being given as a gift by or to Enlil also occurs in VAT 617 (VS 2: 2) col. iv 32′–33′ (lines 23–24 in Cohen’s [1981: 108–109] edition of Eršemma 166.2), in the fourth Eršemma on the tablet, which is an Eršemma to Bau. These lines read: 32′ uru2-ĝu10 a-a-ĝu10 saĝ-e-eš mu-ni-rig7 33′ gir-suki dmu-ul-lil2-la2 mi2 zi mu-un-du11 My father gave my city (to me) as a gift. He truly cared for the Girsu of Enlil.

col. i 5′

If these lines are from the direct speech of Bau – the lady of Lagash, introduced in the preceding line – then it is likely that Bau is appealing to Enlil to spare her city, Girsu, by reminding him that he had once given over his beloved city to her care. A similar tactic is probably being employed in this passage as well, except here the goddess is going even further back in the remote past to remind Enlil of when his ancestors had given the city over to him. The occurrence of Enki and Ninki, followed in these lines by Enul and Ninul in l. 7′, is consistent with the standard order of god litanies which list the “ancestors of Enlil”. These lists almost always begin with Enlil and his wife Ninlil and then continue with, Enki-Ninki, and Enul and Ninul, which are followed by Endašurum-Nindašurum and other divine couples, except in VAT 607 (VS 2: 11) + col. vii 1 f., a source for e-lum gu4-sun2, in which the divine pair An and Uraš precede Enki and Ninki near the head of the list.13

13 For a description of litanies beginning with the ancestors of Enlil, with references to additional sources with this list of deities, see Gabbay 2014: 56–58. See also Lambert 2008: 27–31

96 col. i 8′

Paul Delnero

The motif of the lamenting cow is common in Sumerian laments. Notable examples include CBS 6890 (PBS 10/2: 17), the main source for the Balag ab2-gin7 gu3 de2-de2 (Cohen 1988: 533–35) col. i 1, which reads: ab2-gin7 gu3 de2-de2 ga-ba-ra-e3 edin-še3 ga-ba-ra-e3 ‘Like a cow calling out, I want to come out, I want to come out to the steppe’. In this composition, and nearly every instance in which the motif occurs, the image of the lamenting cow crying out for its lost calf refers to Inana. Further examples include VAT 611 (VS 2: 26) +, a source for the Balag edin-na u2-saĝ-ĝa2-ke4 (Cohen 1988: 668–703), col. iii 4–5 (duplicated in BM 98396 = CT 58: 5 ll. 25–26): 4 5

ama-gan ab2 amar-ra gu3 nam-me i-bi2-zu gar-ra-am3-ma ab2 amar-ra gu3 nu-gi4-gi4 i-bi2-zu gar-ra-am3-ma

The birth-giving mother, the cow, should not low for the calf − set your face toward me (instead)! Cow, do not call out to the calf − set your face toward me!14

col. ii 2′

And also the Balag a-še-er ĝi6-ta (Black 1985) l. 326: ab2-e edinna-na er2 gig mu-un-ma-ma- ‘The cow laments bitterly in its steppe’, a line which is followed by a passage in which the same verbal form is repeated to describe Inana lamenting. The frequency with which this motif occurs in association with Inana is a clear indication that Inana, or Enlil’s wife Ninlil, is also being referred to in this passage. This line is directly parallel to L.1492 (ISET I: 221–22) = Eršemma no.32 (Cohen 1981: 65–69), an Eršemma of Inana, which has a litany of Inana which begins in ll. 40–41 (= rev.10–11) with the lines: 40 mu-gig an-na men3 e2 [la]-ga me nu-me-a[-ĝu10] 41 ga-ša-an-e2-an-na men3 uru2 la-ga me nu-me-a[-ĝu10] I am the mistress of heaven − (woe) my robbed temple without mes! I am Inana − (woe) my robbed city without mes!

and 2013: 405–17 for a detailed discussion of the significance of the ancestors of Enlil in these litanies. 14 See also Katz 2003: 312, who translates these lines: “The birth-giving mother, cow, do not low for the calf, set your face toward me! Cow, do not cry to the calf, set your face to me!”.

The Silences of the Scribes, Pt. II: An Unfinished Enlil Lament from Nippur

col. ii 3′

97

Cohen bases his restoration of the ends of these lines on K.58, a source for the Balag a-še-er ĝi6-ta (= Black 1985: 147–48) obv. 16–17 which has the same phrase: [e2] la-ga me nu-me-a-ĝu10 // [uru2] la-ga me nu-me-a-ĝu10. Similar lines also occur in H 2, an Old Babylonian source from Me-Turan for the Balag uru2 am3ma-i-ra-bi, ll. 147–48, cited by Black (1985: 69). The passage in K.58 with these lines is followed by a section of the same Balag which begins with the line, […] -bi-ke4 zi-da uri3 gul-la-bi er2-ra-gin7 u4 mu-ni-ib2-zal, a line which is very similar to the first line of the next section in this source (CBS 10417). Although not directly parallel, a passage in which the profoundly negative effect the sight of a devastated temple has on one who passes in and out of it occurs near the end of a section of another Enlil Balag, in BM 96691 (CT 36: pl. 42–43) rev. col. vi 16–18: 16 mu-lu hul2-la-ni nu-mu-ni-in-dib-be2 17 mu-lu li-ba-ni nu-mu-ni-in-dib-dib-be2 18 mu-lu hul2-la-ni nu-mu-un-ta-e3-a One who is happy does not pass through (the destroyed temple). One who is rejoicing does not pass through it. And one who is happy does not come forth from it.

col. ii 5′–9′

For a discussion of a similar pair of lines that occur in other laments, see Krecher 1966: 93–96 and Löhnert 2009: 279. The first line of this section of the text occurs in the only two preserved lines in CBS 12674, where they occur at the end of the tablet, possibly as the catch-line to the next section of a lament. The entire beginning of this section of the text is also closely parallel to CBS 2218+15 obv. col. i 1–10, which reads:  1  2  3  4  5  6

[…] / [(x)] e2 uru2 -[la2-am3] lil2-la2-am3 -[la2-am3] lil2-la2-am3 -[la2-am3] lil2-la2-am3 -[la2-am3] lil2-la2-am3 -[la2-am3] lil2-la2-am3 -[la2-am3]

15 Source N3 in Löhnert’s edition of dutu-gin7 e3-ta; see Löhnert 2009: 144 for a description of the source and pl. II for a copy of the reverse of the tablet.

98

Paul Delnero

 7  8  9 10

col. iii 1′–6′

lil2-la2-am3 -[la2-am3] lil2-la2-am3 -[la2-am3] me-e uru2-ĝa2 [du5-mu-bi ki] - nibruki- [dam du5-mu-bi ki] kur2-ra

[…] the temple, the city − it is desolate! It is desolate, it is desolate! It is desolate, it is desolate! It is desolate, it is desolate! It is desolate, it is desolate! It is desolate, it is desolate! It is desolate, it is desolate! It is desolate, it is desolate! I am in my city (while) its wife and son are in a foreign place! (I am) in my Nippur (while) its wife and son are in a foreign place! These line are duplicated in BM 81014 + 132267 (CT 58: 39A), an Old Babylonian source with a lament to Enlil (see the reconstruction of the content of CBS 10417 above for a more detailed discussion of BM 81014+ and its relation to this source), col. i 3′–8′. Parallels to these lines also occur at the end of a Kirugu of the first millennium version of the Balag mu-tin nu-nuz dim2-ma (Cohen 1988: 222–52) in K.2003 + 3466 (4 R2 28), obv. 35-rev.4 and VAT 245+ (SBH 46) rev. 20′–26′. These lines, which are written in a partially phonetic orthography in BM 81014+ col. i 3′–8′, read (with parallels): i-bi2-zu!(ba) u3 di-de nu-kuš2!-šu = i-bi2-zu u6 di-de3 nu-kuš2-u3 (SBH 46 rev. 20′ = 4 R2 28 obv. 35) 4′ gu2-zu ki ma-al-e nu-gi-gi = gu2-zu ki ma-al-e nu-gi4-gi4 (SBH 46 rev. 21′ = 4 R2 28 obv. 36) 5′ ša3-zu bal-bal en-še3!(tug2) kuš2-u3 = ša3-zu bal-bal en-še3 i3-kuš2-u3 (SBH 46 rev. 22′ = 4 R2 28 obv. 37) 6′ u8 sila4 zi-da kur-re ba-e-ze-eĝ3 = u8 sila4 zi-da kur2-re ba-an-ze2-eĝ3 (SBH 46 rev. 23′ = 4 R2 28 rev. 1) 7′ ud5 maš2! zi-da kur-re ba-e-ze2-eĝ3 = ud5 maš2 zi-da kur2-re [[ba-an-ze2-eĝ3]] (SBH 46 rev. 24′ = 4 R2 28 rev. 2) 8′ an-tuku-e a-da-ni nu-te-te-en = eĝ3 tuku-da eĝ3-ĝa2-ni mu-un-e-til (SBH 46 rev. 25′ = 4 R2 28 rev. 3) 3′

The Silences of the Scribes, Pt. II: An Unfinished Enlil Lament from Nippur

99

col. iii 3′–4′ These two lines, as they occur in 4 R2 28 and SBH 46, and other sources, are translated by Krecher (1966: 222 n.640): “dein (Enlils) Auge wird nicht müde zu spähen // dein Nacken bleibt immer gebeugt (wörtlich: wendet sich nicht, zur Erde hin zu sein).” col. iii 6′–9′ The parallel source BM 81014+ col. i 9′–10′ continues with the following two lines, before the remainder of the column breaks off:  9′ e-lum mu-nu10 u4-de3 i-bi2-zu nu / kuš2-u3 10′ dmu-ul-lil2-le tug2 gal-gal šed9(a.muš3!.di) / bi-gam!(nu) i-bi2-zu nu-kuš2-u3 = 4 R2 28 rev. 4: dmu-ul-lil2 tug2 gal-gal-la šedx(muš3.a.di)de3 ba-an-gam i-bi2-zu nu-kuš2-u3 Honored one, your eyes do not tire of … the cattle herder. Enlil, your eyes do not tire of laying low the freezing ones in thick garments. The reconstruction of these lines in CBS 10417 is uncertain, but the presence of a clear e2 at the beginning of col. iii 8′ is a certain indication that the content of CBS 10417 diverges from the content of this passage in BM 81014+ and the parallel first millennium sources for mu-tin nu-nuz dim2-ma, in which the Kirugu with this passage ends with the line dmu-ul-lil2 tug2 gal-gal-la šedxde3 ba-an-gam i-bi2-zu nu-kuš2-u3. It is thus like that CBS 10417 and BM 81014+ contained similar, but not identical laments to Enlil.

100

Paul Delnero

Fig. 1a: CBS10417 obv.

The Silences of the Scribes, Pt. II: An Unfinished Enlil Lament from Nippur

Fig. 1b: CBS10417 rev.

101

102

Paul Delnero

Bibliography Alster, Bendt. 1992. The Manchester Tammuz. ASJ 14: 1–46. Black, Jeremy. 1985. A-še-er Gi6-ta, a Balag of Inana. ASJ 7: 11–87. Civil, Miguel. 1976. Enlil, the Merchant. Notes to CT 15, 10. JCS 28: 72–81. Civil, Miguel. 1980. Les limites de l’information textuelle. Pp. 225–32 in L’archéologie de l’Iraq du debut de l’époque néolitique à 333 avant notre ère: Perspectives et limites de l’interprétation anthropologique des documents. Ed. Marie-Therese Barrelet. Paris: CNRS. Civil, Miguel. 1983. The 10 th Tablet of uru2 am3-ma-ir-ra-bi. AuOr 1: 45–54. Civil, Miguel. 1999. Reading Gilgameš. AuOr 17: 179–89. Cohen, Mark. 1981. Sumerian Hymnology: The Eršemma. HUCA Supplements 2. Cincinnati: Hebrew Union College. Cohen, Mark. 1988. The Canonical Lamentations of Ancient Mesopotamia. Potomac: Capital Decisions Limited. Delnero, Paul. 2015. Texts and Performance: The Materiality and Function of the Sumerian Liturgical Corpus. Pp. 86–117 in Texts and Contexts: The Circulation and Transmission of Cuneiform Texts in Social Space. Studies in Ancient Near Eastern Records 9. Ed. Paul Delnero and Jacob Lauinger. Boston/Berlin: Walter de Gruyter. Gabbay, Uri. 2014. Pacifying the Hearts of the Gods: Sumerian Emesal Prayers of the First Millennium BC. Heidelberger Emesal-Studien 1. Wiesbaden: Harrassowitz Verlag. Katz, Dina. 2003. The Image of the Netherworld in the Sumerian Sources. Bethesda, MD: CDL Press. Krecher, Joachim. 1966. Sumerische Kultlyrik. Wiesbaden: Otto Harrassowitz. Krecher, Joachim. 1967. Die sumerische Texte in “syllabischer Orthographie”. ZA 66: 16–65. Kutscher, Raphael. 1975. Oh Angry Sea (a-ab-ba hu-luh-ha): The History of a Sumerian Congregational Lament. YNER 6. New Haven: Yale University Press. Lambert, Wilfred G. 2008. Mesopotamian Creation Stories. Pp. 15–59 in Imagining Creation. IJS Studies in Judaica 5. Ed. Markham Geller and Mineke Schipper. Leiden/Boston: Brill. Lambert, Wilfred G. 2013. Babylonian Creation Myths. MC 20. Winona Lake, Indiana: Eisenbrauns. Löhnert, Anne. 2009. “Wie die Sonne tritt heraus”: Eine Klage zum Auszug Enlils mit einer Untersuchung zu Komposition und Tradition sumerischer Klagelieder in altbabylonischer Zeit. AOAT 365. Münster: Ugarit-Verlag. Marchesi, Gianni. 2006. LUMMA in the Onomasticon and Literature of Ancient Mesopotamia. History of the Ancient Near East − Studies 10. Padova: S.A.R.G.O.N. Editrice e Libreria. Peterson, Jeremiah. 2012a. The Banishment of Inana. N.A.B.U. 2012 n.8: 9–10. Peterson, Jeremiah. 2012b. A Fragmentary Sumerian Context Involving the Removal of Clothing and Accoutrements. N.A.B.U. 2012–3 n.44: 56–57. Tinney, Steve. 2011. Tablets of Schools and Scholars: A Portrait of the Old Babylonian Corpus. Pp. 577–96 in The Oxford Handbook of Cuneiform Culture, eds. Karen Radner and Eleanor Robson. Oxford: Oxford University Press. Volk, Konrad. 1989. Die Balaĝ-Komposition Uru2 am3-ma-ir-ra-bi: Rekonstruktion und Bearbeitung der Tafeln 18 (19’ff.), 19, 20 und 21 der späten, kanonischen Version. FAOS 18. Stuttgart: Franz Steiner Verlag.

Gertrud Farber

A New Seal of the Scribe Ur-Nanše from Girsu During the 1930s and 40s Mr. Milton S. Yondorf of Chicago acquired a small collection of Near Eastern artifacts and cuneiform tablets, many of which were at the time expertized by members of the Oriental Institute, especially I. J. Gelb. The collection then disappeared only to resurface again a few years ago here in Chicago at an antiques’ dealer who graciously agreed to the publication of the complete collection, which my husband Walter and myself are planning to finish in the near future. A little text labelled Yondorf 24 especially caught my attention. It is a small unopened case tablet that was sealed several times on the obverse, reverse and the edges, mainly with the text portion of the seal. The picture portion, an introduction scene, however, was – probably deliberately – rolled over the edges only, except for the seated figure which is visible also in the right sealing on the reverse. The complete scene can best be seen on the lower edge: a seated god(dess) to the right facing three standing figures to the left, of which the first, the introducing god(dess), is looking back at the second, probably the king who is to be introduced to the deity (see Fig. 1). The text comes from Girsu and is a receipt for food supplies for workmen from the 5 th month of Šulgi 46. Yondorf 24, case (see Figs. 2–3) obv. 1) 33.1.0 še gur lugal še šuku-ra eren2-na

5) rev. 1)

33 gur 1 nigida grain (measured) in royal gur , grain as sustenance for workers, (seal inscription rolled twice next to one another) for zid2-ka-delivery zid2-ka-še3! ki ku-li muhaldim-ta from Ku-li, the cook. Seal of Ur-dNanše, son of Ur-li. kišib ur-dnanše dumu ur-li From the big granary. ŋa2-nun-mah-ta The month munu4-gu7 (5 th month). iti munu4-gu7 (Measured) according to the gur gur i3-pa3-da-ta (measurement) of I3-pa3-da. (seal inscription again rolled twice) The overseer is Lugal-lu2-sa6-sa6. nu-banda3 lugal-lu2-sa6-sa6 Year in which Kimaš was destroyed mu ki-maški ba-hul (= Š 46).

https://doi.org/10.1515/9781501503696-007

104

Gertrud Farber

Yondorf 24, seal, reconstructed from several impressions (see Fig. 4) ur-dnanše Ur-dNanše, dub-sar the scribe, dumu ur-li son of Ur-li, ga2-dub-ba high accounting official. Trying to put the personal names of the text into some context, I found two texts that are very similar to Yondorf 24. These are the tablet MVN 12: 26, which is almost identical with the Yondorf text, and the tablet and case of MVN 12: 179. MVN 12: 26 obv. 1) 12.0.0 še gur lugal

5) rev. 1)

5)

še eren2-na zid2-ka-še3 ŋa2-nun-mah-ta ki ku-li-ta ur-dda-mu šu ba-ti gur i3-pa3-da-ta nu-banda3 lugal-lu2-sa6-sa6 iti munu4-gu7 mu ki-maški ba-hul

12 gur grain, (measured) in royal gur, grain for workers for zid2-ka from the great granary from Ku-li Ur-dDa-mu has received. (Measured) with the gur (measurement) of I3-pa3-da. The overseer is Lugal-lu2-sa6-sa6. The month munu4-gu7 (5 th month), year in which Kimaš was destroyed (= Š 46).

This tablet has not been preserved as a case tablet and has no sealings. Instead of the phrase kišib PN that we have on the Yondorf 24 case, MVN 12: 26 has PN šu ba-ti. If this tablet actually had a case, this would probably have had the phrase kišib PN instead, or both phrases, and would have been sealed by the recipient (šu ba-ti). MVN 12: 179 Another similar text is MVN 12: 179. In this case both the sealed envelope and the tablet are preserved. It dates from one year later. The tablet has both phrases kišib PN1 and PN2 šu ba-ti, while the case is sealed by the recipient PN2. That means that in this text kišib PN on the tablet does not refer to the person who actually sealed the envelope and at the same time was the recipient of the grain. I therefore have translated kišib differently on tablet

A New Seal of the Scribe Ur-Nanše from Girsu

105

and case. The same might apply to our Yondorf case and the tablet that is still enclosed in it. But obviously, we can not be sure as long as the case remains unopened. MVN 12: 179, tablet obv. 1) 33.3.4 še gur lugal še zid2-ka-še3 še niŋ2-gal2-la kišib ur-dnanše dumu ur-li rev. 1)

5)

u3 še ur-e2-ninnu ki na-di-ta ur-ŋar šu ba-ti (blank) mu us2-sa ki-maški ba-hul

MVN 12: 179, case 33.3.4 še gur lugal [še zid2]-˹ka-še3˺ kišib ur-ŋar dumu lugal-gaba ki na-di-ta mu us2-sa ki-maški ba-hul

MVN 12: 179, seal ur-ŋar dub-sar [dumu] lugal-[gaba]

33.3.4 gur grain (measured) in royal gur, grain for zid2-ka, grain (from) the existing supply – transaction confirmation by Ur-dNanše, son of Ur-li – and grain of Ur-e2-ninnu from Na-di Ur-ŋar has received. Year following the year in which Kimaš was destroyed (= Šulgi 47)

33.3.4 gur grain (measured) in royal gur (seal) [grain for zid2]-ka – seal of Ur-ŋar, son of Lugal-gaba – from Na-di. Year following the year in which Kimaš was destroyed (= Šulgi 47)

Ur-ŋar, the scribe, [son] of Lugal-[gaba]

While the transaction confirmation (kišib) mentioned in the tablet is that of Ur-dNanše, the seal itself (also kišib) on the envelope is that of Ur-ŋar, the recipient of the grain. Notes to the Yondorf Text Obv. line 3 The term (še) zid2-ka-še3 is not uncommon. It occurs exclusively in texts from Lagaš/Girsu, frequently together with la2-i3 su-ga “payment of balance” or

106

Gertrud Farber

niŋ2-ŋal2-la “existing supply.” For a discussion of the phrase še3(zid2).ka see Such-Gutiérrez (2003: 394–410) who defines the term as originally meaning a kind of flour that in Ur III changed to mean some kind of contribution or delivery of flour. Obv. line 4 The PN could be read as ku-li-ŋu10 or as ku-li muhaldim. Both ku-li and ku-li-ŋu10 occur in Lagaš texts of this period. In the above quoted parallel text MVN 12: 26 the PN, however, is just ku-li. Obv. line 5 (and seal) Ur-Nanše who sealed the tablet is most probably the recipient of the grain. If the tablet inside the case contains the phrase kišib PN, this PN is probably somebody else, while Ur-Nanše should be named on the tablet as the recipient (šu ba-ti). See the discussion of MVN 12: 179 above. Ur-dNanše, son of Ur-LI according to BDTNS and Verderame (2008: 233 n. 9) until now was attested from Š 47 to IS 1. Our text is one year before that period and therefore now is the earliest known reference for our Ur-dNanše. Rev. line 2 A charming and grammatically correct interpretation of this line would be that the measuring device had been lost and the grain could only be weighed “after the gur had been found.” This, however, is not likely since we have other references for gur PN-ta. The ones I could find are all from Girsu or Umma and are all about the delivery or rations of grain or seeds. Interesting are the references pointed out to me by Manuel Molina where the grain is measured in gur zabar PN(-ta), probably a bronze measuring vessel, which seemingly was in the care of a certain person; the references are: ASJ 3: 164 no. 137 obv. 2: gur zabar lu2-dnin-su4-tum-zi; BCT 2: 164 obv. 2: gur zabar lugal-ku3-ga-ni-ta; MVN 11: 12 obv. 2: gur zabar lu2-dnin-sun2-zi; PPAC 5: 1120 obv. 2: gur zabar lu2-dnin-giškim-zi-ta. Maybe there existed several gur lugal measuring vessels, some of which were made of bronze, which were kept in the care of different people and had not exactly the same capacity, which made it necessary to state which one was used. In our text the scribe seemingly forgot at first to define which vessel was used and then had to add it after he had already written the month name.

A New Seal of the Scribe Ur-Nanše from Girsu

Seal

107

ga2-dub-ba or pisaŋ-dub-ba is a high accountant, probably corresponding to Akkadian šandabakku. It might therefore be read ša13-dub-ba. For references for ur-dnanše dumu ur-li see above ad obv. line 4; also Verderame (2008: 233–34) who depicts a seal of Urd Nanše from the year AS 5 (BM 12561) that is not identical with our seal from Š 46. While three figures are standing opposite the deity on our seal, there are only two on the seal shown by Verderame. Another seal of Ur-dNanše is ITT pl. IV: 10037. This, however, shows no scene. Unfortunately we do not know its date and I do not know where C. Fischer’s dating of this object into year IS 2 comes from (see Fischer 1997: 125 n. 152); I also cannot see any swimming bird as described by Fischer under the legend of the BM seal on the excellent photos of Verderame. In any case, the seal impressed on our Yondorf tablet is a new seal of Urd Nanše and dates one year earlier than the one(s) known until now.

108

Gertrud Farber

Fig. 1: Yondorf 24, lower edge: the sealing.

Fig. 2: Yondorf 24, obverse.

Fig. 3: Yondorf 24, reverse.

A New Seal of the Scribe Ur-Nanše from Girsu

109

Fig. 4: Yondorf 24, reconstructed seal.

Bibliography Fischer, Claudia. 1997. Siegelabrollungen im British Museum auf Ur-III-zeitlichen Tafeln aus der Provinz Lagaš. BaM 28: 97–183. Such-Gutiérrez, Marcos. 2003. Überlegungen zu der sumerischen Zeichen-Gruppe šè.ka. Sefarad 63: 394–410. Verderame, Lorenzo. 2008. Rassam’s Activities at Tello (1879) and the Earliest Acquisition of Neo-Sumerian Tablets in the British Museum. Pp. 231–44 in On the Third Dynasty of Ur: Studies in Honor of Marcel Sigrist, ed. Piotr Michalowski. JCS Suppl. Series 1. Boston: American Schools of Oriental Research. Internet: BDTNS: http://bdts.filol.csic.es/index.php?reg=si (last visited April 2015)

Lluís Feliu and Adelina Millet Albà

Texts and Workers In this article we publish an Ur III administrative tablet and a study of it in relation to three other texts with almost identical content. We hope that this modest contribution may please Miquel Civil on the occasion of his 90th birthday.1

1 Four Texts and Several Editions The tablet published here was purchased at auction in London in 2012 and is currently in a private collection in Barcelona. The text records the monthly ration to various people grouped by profession. At first sight we can observe that the text is almost a twin of Nisaba 15/2 668 (a tablet published in Feliu 2014) currently held in a private collection in Madrid and purchased in the antiques market in Barcelona. Both tablets were purchased in London at the same auction and in the same lot. The first option was to identify this new tablet with one of the tablets published by Owen 2013 in transliteration only (Nisaba 15/2 1031 or 1032), but the transliteration of our tablet and the broken passages do not fit the transliteration provided by Owen. Although the content is very similar, in fact almost identical in some passages, differences in the order of the people quoted on the text and, especially differences in the broken passages militate against an identification of both texts. Consequently we can say that there are four texts with a very similar content and a very close chronology. For ease of quotation we will quote the texts as follows: Text A: Tablet published in this article. Text B: Nisaba 15/2 668 = Feliu 2014 = CDLI P453995 = BDTNS 193119. Text C: Nisaba 15/2 1031 = CDLI P454184 = BDTNS 193292. Text D: Nisaba 15/2 1032 = CDLI P454185 = BDTNS 193293. The transliterations of texts B, C and D published by Owen show several peculiarities. The basis of Owen’s transliterations was the handwritten transliteration

1 We would like to express our gratitude to M. Molina for his comments and suggestions and to W. G. E. Watson who was kind enough to correct the English. We should stress that any faults and errors are our own responsibility. L. Feliu wrote §§ 1, 2, 3 and 5; A. Millet wrote § 4. Note that we use numerical indexes also in the transliteration of syllabograms in Semitic names, as opposed to accents, since these are otherwise Sumerian texts and the alternation of conventions within the same line could generate confusion. https://doi.org/10.1515/9781501503696-008

Texts and Workers

111

by Lambert.2 In checking the transliteration provided by Owen against the original tablets, it is quite clear that Lambert’s transliteration is the product of an initial sight reading, a provisional and rapid first appraisal of the texts, with two or three tablets simultaneously on the desk. Consequently, the transliterations are full of inconsistences, small errors, changes in the order of the lines and even the omission of some lines. On the other hand, there are some mistakes in Owen’s transliterations due to the difficulties in transcribing Lambert’s handwritten transliterations.3 Two examples of each kind of mistake: Nisaba 15/2 668 iii 97 reads: ninda 0.0.3 da-ḫi-im-ti, but it should read: ninda 0.0.3 d ˹ a- šar2˺-im-di3 (see copy in Feliu 2014: 232 iii 7). Nisaba 15/2 1032 i 16 reads: [0.0.n.] maḫ-ri?-du, but texts A i 16 and B i 15 have: aš-da-ḫuda.

The first example is a clear mistake in choosing the reading of one sign (ḫi/ šar2) and also a misreading of Lambert’s handwritten transliteration (da/da). The second example is the typical confusion of similar signs in group signs: maḫ/aš-da and ri/ḫu. Obviously, the reading of personal names favours this kind of mistake due to lack of context.

2 Chronology and Format of the Texts Tablet A has the same format and measurements as text B, with much the same content as B, C and D. Probably the format of C and D is also similar to A and B because all the texts are four-column tablets. Text B is the only tablet with the date preserved: Ibbi-Suen 1. Unfortunately, the date on text A is missing but probably it was written shortly before text B because in A vii 10 a person named zu-lu-ti-ia receives a ration of bread whereas B vii 22 records the same person as dead (uš2). The relative chronology of C and D is difficult to establish. Texts A iv 21, C iv 47 and D iv 156 record Aḫušuni as absent/a fugitive (zah3) from the team of reed-cutters (lu2-gi-zi) but in text B iv 36 he receives 30 litres of bread as a member of the team of garden workers (si12-a u3 du3-atar). On the other hand, A vii 9 records Engar-zi as absent/a fugitive, but in B vii 20 he receives 4 litres of barley.

2 B = Unpubl. ms. WL 900 (BDTNS 193119); C= Unpubl. ms. WL YYY (BDTNS 193292); D = Unpubl. ms. WL ? (BDTNS 193293). 3 To appreciate the difficulties in reading Lambert’s handwriting, see Lambert 2014.

112

Lluís Feliu and Adelina Millet Albà

In spite of the similar format of the tablets, it is clear that tablets A and B were written by different scribes. There are clear calligraphic differences in both texts. See, for example, the sign bu:

Fig. 1: Text A (I 30).

Fig. 2: Text B (I 32).

There are also clear differences in the orthography of the texts. Texts A and C use the plene writing for the third person singular possessive pronoun {ani} (dumu-munus-a-ni; dumu-nita-a-ni), whereas texts B and D always use the contracted form (dumu-munus-ni; dumu-nita-ni). Texts A and C do not write geme2 before uš-bar (A ii 26; C ii 27), whereas text B and D do (B ii 30; D ii 67). There are also differences in the spelling of some personal names, for example: am3-ma-a (A ii 28; C ii 29; D ii 69); a-ma-a (B ii 32). nu-ur2-um (A iv 25; C ii 53) nu-ur2-ru-um (B iv 43; D iv 162). [i]-din-dsuen (A vi 14); i-di3-dsuen (B vi 25).4 ša-ru-še-ni (A vii 10) ša-ru-uš-še-ni (B vii 23).

3 Reconstruction of the Teams of Workers The workers listed in the texts are people of a low social standing (Steinkeller 2013: 367). Apart from the people of the household (ša3 e2-a), three of the four first teams of workers are professions traditionally assigned to women:5 weavers ((geme2) uš-bar), grinders ((geme2)-ar3-ra) and sesame pressers (ĝiš-i3 sur-sur). These three groups of workers usually occur together in the ration list of a single complex/household (Lafont 2013: § 1). The other professions found in the lists are virtually identical in the four texts and the people regis-

4 On these writings, see Hilgert 2002: 57 and 295. 5 Waetzoldt 1987: 121. For an alternative view of the traditional interpretation of Waetzoldt concerning the gender of workers in the textile industry see Garcia-Ventura 2014; see also Lafont 2013.

Texts and Workers

113

tered in each section are very similar and occasionally, identical. The amount of the monthly rations is standard: 50/60 litres of barley or bread for an adult male and 30/40 litres for an adult female. Children and young people receive from 10 to 25 litres. The texts record some sections with several people described as zah3. This term is usually interpreted as “fugitive”,6 however, some of those who receive a ration in text B are described as zah3 in text A. Therefore, text A would certainly be earlier than text B, if they were fugitives, probably people captured and made to work with the others. On the other hand, we can interpret zah3 as “absent”,7 with no reason for absence given. There is a high percentage of absentees in the teams formed by male workers, particularly craftsmen (gašam) and gendarmes (aga3-us2), a tendency also observed in the documentation of XIV–XIII century Nippur, where almost 91 % of fugitives/absentees are male (Tenney 2011: 108). In the following section we reconstruct the different teams of workers recorded in the four texts. We note the texts where the name is recorded (A, B, C or D). We also state whether the name is broken or damaged and the certainty of the reconstruction. When there is a discrepancy between our reading and the reading of the proper names in the original editions of the texts, the original reading is quoted in a footnote. The teams of workers are listed in the same sequence as the texts. Weavers ((geme2) uš-bar) A i 1–ii 26; B i 1–ii 30; C [i 1]–ii 27; D i 1–ii 67: um-mi-a ([A]B[C?][D]); um-mi-du10 ([A]B[C?][D]); ḫa-al-ni (AB[C?]D); ki-ni-ib-ši (AB[C?]D); ṣ]a-limtum (AB[C?]D); nin-e2-mah-e (AB[C?]D); na-ab-ri2-tum (AB[C?]D); eš18-tar2-tu-kul2- (AB[C?]D); nin-e-i3-sa6 (AB[C?][D?]); šu-mu-uš-tum ([A]B[C?]D8); ereš-diĝir-ĝu10 (AB[C?]˹D˺); ze2-lu-a (AB[C?]˹D˺9); aš-da-ḫu-da (AB[C?]D10); ama-ge-na (AB[C?]D); da-du-ra (AB[C?]D11); a-ga-tum (AB[C?]D); im-ti-dam (AB[C?]D); ad-du-en-na (AB[C?]D); geme2-dištaran (AB[C?]D); ama-ge-na (AB[C?]D); geme2-e11-e (AB[C?]D); qa2-du-ša-num2 (AB[C?]D); en-ni-lu (AB[C?]D12); ši-nu-nu-tum (AB[C?][D]); ku-ba-at (AB[C?]˹D˺13); ta2-din-eš18-tar2 (AB[C?][D]);

6 On the term zah3 “fugitive” see Monaco 1986: 4–6; Steinkeller 1989: 69; Lafont 2013: § 2. 7 See the OB texts with receipts of silver qualified as šam2 zah3 or ku3-babbar zah3 (van Koppen 2001: 215 and n. 12). It is quite clear that in this context zah3 refers to an absentee because it would be not normal to assign silver to fugitives. However, see the original publication of the text, which translates saĝ arad2 zah3 as “entlaufenen Sklaven” (Sanati-Müller 1990: 194). 8 D i 11: […] x-nu-ru-uš. 9 D i 15: […] at-ba?-a. 10 D i 16: […] maḫ-ri?-du. 11 D i 18: […] du-du-um. 12 D i 26: […] en-ni-e. 13 D i 18: […] x-di-at.

114

Lluís Feliu and Adelina Millet Albà

ga-bu-a-ti (AB[C?]˹D˺); iš-me-a-ni (AB[C?]˹D˺); šu-dan-ni (AB[C?][D?]); ama-ge-na (AB[C?][D?]); a-za-at-al-la (AB[C?]D); geme2-ĝišgigir (AB[C?]); a-ša-ti-nu-ri (AB[C?]); ša-at-i3li2 (AB[C?]˹D˺); u2-[x]-ni-il (˹A˺˹B˺[C?]˹D?˺); ˹x˺-da-a (˹A˺˹B˺[C?][D]); […]-˹šu?-zi?˺ (A); u2-nazi (˹A˺[B][C?]D); u2-zi-li (˹A˺[B][C?]D); ta/da-sal-la (AD); a-zi-um-mi ([A][B][C?]D); ba-n]i-sa3 (˹A˺B[C?]D); eš18-tar2-bi2-tum (˹A˺B[C?]D); ut-tum (AB[C]D); saĝ-nin-e-i3-zu (˹A˺B[C?]D14); nin9-kal-la (AB[C?]D); u-bar-tum (AB[C?]D); za-ne-a (AB[C?]D15); bil-lu (AB[C?]D); ur-dnungal (A[C?]D)16; du-mu-uq-še-ni (ABC); a-li2-a-ḫi (ABC); um-mi-nu-nu (ABC); ama-ge-na (ABC); ga-na-za (ABCD17); ba-la-la-a (ABCD); nin-e2-ku3-ta (ABCD18); nin-e2-ku3-ta min (ABCD19); u3-ba (ABCD); geme2-diškur (ABCD); PI-ru-ru-tum (AB20CD); ṣi-lu-uš-du10 (ABCD); ša-at-i3-li2 (B); nin-me-du (B); geme2-dnanna (B); me-dtišpak (B); e-ni-ša-ag (B).

Weavers receive a ration of 30 litres of barley. The unnamed children receive 10 or 15 litres and the named children 15 or 20 litres. Exceptionally, geme2ĝiš gigir receives 40 litres in text B i 37 (the amount of barley is broken in text A) and also du-mu-uq-še-ni receives 50 litres in A ii 13 and B ii 12. Text B II 25– 29 records five supplementary weavers not recorded in the other texts. People from a temple/household?21 (ša3 ˹e2˺-a) A ii 27–iii 14; B ii 31–iii 19; C ii 28–iii 30; D ii 68–iii 98: geme2-dutu (A CD); am3-ma-a (ABCD); a-ga-ti-a (˹A˺BCD); šu-im-mi-ni (˹A˺BD22); ar-bi2-tum (˹A˺B˹D˺); eš18-tar2-ra-bi2-at (˹A˺BD); eš18-tar2-um-mi (˹A˺BD); dšud suen-nu-ri (˹A˺BD); iš-me-a-ni (˹A˺BD); ša-ru-uš-še-ni (˹A˺BD23); a-pil-nu-ri ([A]B24D); enni-ma-ad (˹A˺BD); en-ni-ma-ad (˹A˺˹B˺D); u3-ša-ak-ni-il (˹A˺˹B˺25D26); ki-za-na-d[a] (ABD); a-na-eš18-tar2-ta2-ak-la-at (ABD); lu-ba-li (˹A˺BD); mi-na-ar-ni (˹A˺27BD); am3-ma-a-(a) (AD); da-[ša]r2-im-di3 (B); eš18-tar2-puzur4-ri (˹A˺BD); dšu-dsuen-tu-ri (ABD); dšu-dsuen-im-ti (ABD); ša-lim-ni-rum (BD); dna-na-a-nu-ri (ABD); me-diškur (ABD); zi-mu-dar (ABD28); ša-at-dna-na-a (ABD); mi-ni-mi (ABD).

14 D ii 49: saĝ-nin-e-i3-zi 15 D ii 52: za-qar-e2-a. Text B shows a clear ne. 16 C ii 10: ib-na-[…], bad reading of Ur-Nungal. 17 D ii 59: ga-na-a-la. 18 D ii 61: nin-e2-ku3-ga. 19 D ii 62: nin-e2-ku3-ga. 20 B ii 23: ši-ru-ru-tum. 21 Possibly we can read ša3 e2-duru5; if so it would be a hamlet with a threshing floor (ki-sur12), some houses and possibly a small temple (Steinkeller 2007: 191). 22 D ii71; šu-im-mi-ir. 23 D ii 79 reads a-zi-ni but any of the other three texts have the same name. It could be an incorrect reading of ša-ru-uš-še-ni? 24 B ii 41: a-˹x˺-[-n]u?-ri. 25 B iii 1: z[a-x-š]a-ak-ni-il. 26 D iii 82: u3-ša-ak-ni. 27 A iii 2: mi-n[i-a]r-ni. 28 D iii 94: zi-nu-um.

Texts and Workers

115

The people of the household receive between 20 to 35 litres of barley. Some people receive 30 litres of bread (ninda), and children get 10 litres of bread. Text B is an exception and records a supplementary or compensatory ration of dates (zu2-lum); more specifically, Amaya receives 60 litres of barley in A (ii 27); C (ii 29) and D (ii 69), but receives 30 litres of barley plus a supplement of 20 litres of dates in text B (ii 32). a-pil-nu-ri receives 30 litres of barley in text A (ii 36) and D (iii 80) but receives 20 litres of barley plus 10 litres of dates in text B (ii 41). ša-at-dna-na-a receives 30 litres of barley in text A? (iii 11) and D (iii 95) but text B (iii 15) records 25 litres of barley plus 5 litres of dates. This complementary payment of dates may be due to the scarcity of barley in the granary during that month to make up the normal amount of food (Waetzoldt 1987: 125). Grinders ((geme2)-ar3-ra) (10 names) A iii 15–iii 20; B iii 20–iii 30; C iii 31–[…]; D iii 99– iii 110: a-da-ga (ABC29D); ad-du-zu(m) (ABC30D); geme2-dsuen (ABD); en-˹x-x-x˺ (BD31); gu-na-na (ABC32D); dšara2-ba-zi-ge (BD); a/a2-zu-zu (BD); e-li2/la-ša-ru-um (AB˹C˺D) ša-ati3-li2 (D).

Grinders receive 30 litres of barley in the four texts, with 10 litres for children. Texts A and C record exactly the same five individuals, but texts B and D record the same team plus three children of ad-du-zu(m) and a/a2-zu-zu. For that reason it is reasonable to think that line iii 35 of text C is the last line of the section listing grinders, despite the rest of the column being broken. Sesame pressers (ĝiš-i3 sur-sur) A iii 21–iii 25; B iii 31–iii 35; C: iii […]; D iii 111–iii 115: gibil-la (˹A˺BD); ša-at-nu-nu (ABD); ša-at-da-ti (˹A˺BD); a-li2-a-bi (˹A˺BD).

This section is identical in the three texts that preserve it (text C is broken). The ration is also constant in all three texts: 30 litres of barley; the exception is gibil-la, who receives 15 litres in text B and in text D is recorded without a ration, together with ša-at-nu-nu. Fullers (azlag7) A iii 26–IV 9; B iii 36–IV 23; C iii […]; D iii 116–IV 141: ba-lu-ša (ABD); al-la-ša-ru-um (AB); šul-gi-tu-ri (AB); ku-ša-ti (ABD33); šu-i3-li2 (ABD); šu-ti-ru-um (ABD); na-mu-ra (ABD34); uq-ni-tum (ABD); ur-dšul-pa-e3 (ABD); ša-lim-be-li2 (ABD); ni-ba-aš-ḫa-

29 30 31 32 33 34

C iii 31: a-da-lal3. C iii 32: mal-x-du-ud. D iii 101: en-tukumbi?. C iii 35: geme2-na-na. D iv 123 LU-šu-ti. D iv 126: na-mu-um.

116

Lluís Feliu and Adelina Millet Albà

ti (ABD35); lu2-dsuen (ABD); ša-at-er3-ra (ABD); en-um-i3-li2 (ABD); en-um-i3-li2 (A); lu2d nanna (ABD); lu2-ša-lim (BD); ba-at-ri (BD); nu -uḫ2-im-eš18-tar2 (B); dsuen-[…] (D).

The rations for the team of fullers are more variable. These rations range from 30 to 50 litres of barley. Children receive 10 or 15 litres. Specifically, ur-dšulpa-e3 receives 60 litres (D iv 129), but does not receive any ration in text A (iii 38). He receives 40 litres in text B but he also receives a supplementary amount of 20 litres of dates (B iv 8). In this section we can observe that the equivalence in litres between barley and dates is almost always one to one: lu2-dsuen receives 40 litres of barley in text A (iv 3) and D (iv 135) but he receives 30 litres of barley plus 10 litres of dates in B IV 15.36 Team of garden workers (si12-a (u3 du3-a-tar) / dumu a-ru-a)37 A A iv 10–iv 19; B iv 24– iv 37; C iii […]-iv 45; D iv 142–iv 154: lugal-he2-ĝal2 (AB˹C˺D); an-lu-sa6 (ABD38) ur-lugal (AB˹C39˺D); di-ib-la-ba (BD40); šu-da-ta (AC); šu-da-ni-a (B)41; dnin-šubur-zi-ĝu10 (ABC); šu-ma-ma (ABCD); diškur-kal (ABCD); ur-dnanše (ABCD); li-bur-be-li2 (B); ur-dba-ba6 (B); a-ḫu-šu-ni (B); lu2-diĝir-ra (D); ur-dnin-a-zu (D); lu2-uš-gi-na (D); nam-tar-ib2-gu-ul (D).

This team of workers, denoted by si12-a (u3 du3-a-tar) in texts A, B and D, is denoted as dumu a-ru-a in text C iv 45. Anyone denoted as a-ru-a is a “dedicated one” or “one given as a dedicatory gift” (Maaijer and Jagersma 1997: 283; Gelb 1972). According to BDTNS, there is no other reference to dumu a-ru-a in Ur III texts and according to CDLI, there is only one reference in an Old Akkadian text (OSP 1 71). There are donations of people as a-ru-a to the temples, often recorded as war booty (nam-ra-ak) (Sigrist 1992: 136).42 In spite of that, this reference to dumu a-ru-a in text C, one of the texts transcribed by Lambert and edited by Owen, is not certain. The transliteration could be the product of

35 D iv 134: ni-ba-aš-zi-ti-me. 36 The sons and daughters of Šulgi-tūrī receive 10 litres of barley in text A (iii 29–30); on the other hand, his two daughters receive 5 litres of barley and 5 litres of dates in text B iii 40–41 (see corrections in the appendix). 37 On this team workers, see Feliu 2014: 222 nt. 7; on si12-a, see also Greco 2015: 47–49 (‘water drawer’); Focke 2015: 822–26 (‘Kultivateur’). Concerning du3-a-tar workers, see also Greco 2015: 45 (‘arborists’); Focke 2015: 829–30 (“Wasserregler”). 38 D iv 143: dšara2-sa6. 39 C iv 38: x-lugal-x. 40 D iv 145: i-id-la-ba. 41 šu-da-ta and šu-da-ni-a possibly refer to the same person, who always receives high quantities of barley and is always followed by dnin-šubur-zi-ĝu10 who always receives 15 litres of barley. Neither of them is recorded in text D. 42 For the influence of the war on the Ur III economy see Garfinkle 2014; Hebenstreit 2014. On YOS 4 67 (CDLI P142131) and its duplicates made by forgers, see Pomponio 2012: 647–48.

Texts and Workers

117

a rapid reading by Lambert, a common occurrence in the transcription of texts B, C, and D published in Nisaba 15. This group of workers receives the standard quantity of 60 litres of barley for an adult. Text B records complementary amounts of dates, as usual. Young workers receive 15 litres. Exceptionally, ur-lugal receives 60 litres of barley in A, C and D, but only 5 litres of barley plus 5 litres of dates in text B (iv 35). Also interesting is the case of diškur-kal, who receives rations in texts A, B and D, and is recorded as absent/a fugitive in text C. Exceptionally, ur-dba-ba6 and a-ḫu-šu-ni are recorded in this section of workers in text B, although they are recorded as lu2-gi-zi in texts A, C and D. Reed-cutters (lu2-gi-zi)43 A iv 21–iv 26; B iv 38–iv 42; C iv 46–iv 52; D iv 155–iv 161: ur-dba-ba6 (ACD); a-ḫu-šu-ni (ACD); dšul-gi-ha-ma-ti (ABCD); nar-zu-mu (˹A˺BC44D45) šuer3-ra (˹A˺B˹C˺D46); lugal-ku3-zu (˹A˺B[C?]D).

The rations of these workers are quite regular. dšul-gi-ḫa-ma-ti is always the person who receives the highest amount of rations: 60 litres; šu-er3-ra and lugal-ku3-zu receive 50 litres each and ur-dba-ba6 and nar-zu-mu 40 litres. a-ḫu-šu-ni is absent/a fugitive in texts A, C and D, but he receives a ration as a si12-a u3 du3-a-tar in text B. Boatmen ((lu2-)ma2-lah5) A iv 27–IV 34; B iv 43–v 5; C iv 53–iv 60; D iv 162–v 169: nu-ur2(ru)-um (A˹B˺CD); ur-diškur (A˹C˺D); a-a-lu2-du10 (A[B]C47D48); lugal-kiri3-zal (A[B]C49[D]); d iškur-an-dul3 (A[B]C); ur-dḫal-mu-tum2 (A[B]50CD).

The team of boatmen is also quite regular in the four texts and also they receive a constant amount of rations. nu-ur2-(ru)-um and ur-dḫal-mu-tum2 receive 60 litres of barley, ur-diškur is given 50 litres and the rest of the team, sons of urd iškur, receives 20 or 15 litres. Text B is broken in this section except for the two first lines, which record the usual compensation of dates in this text.

43 These were people who cut and transported the gi-zi reed used as fodder for sheep during the summer (Civil 1987: 44–45; Civil 1987b: 44 on the gi-zi reed, see Klein 1990: 117–18; Waetzoldt 1992: 129–30). 44 C iv 49: i3-li2-[…]. 45 D iv 158: lu-[…]. 46 D 159: u-er3-ra. 47 C iv 55: a-a-lu2-x. 48 D iv 164: a-a-bi2-du10. 49 C iv 56: lugal-ki-zal. 50 B v 4: ˹ama˺-d[…].

118

Lluís Feliu and Adelina Millet Albà

Cooks (muhaldim) A iv 35–v 5; B v 6–v 20; C iv 61–v 74; D v 170–v 183: zu-lu-um (AB51C52[D?]); a-ba (AB53C54[D?]); dnin-šubur-an-dul3 (A˹B˺C[D?]); dšul-gi-i3-li2 (A˹B?˺C[D?]); ur-ki-ku3-ga (A[B?]C[D?]) nu-ur2-dsuen (A[B?]C[D?]); lu2-du10-ga (ABC[D?]); ḫula-al (A˹B˺C[D?]); šu-na-wi-ir (AB˹C˺D); geme2-den-lil2 (B daughter of zu-lu-um).

The group of cooks is constant and is assigned very stable rations in the four texts. zu-lu-um receives 60 litres of barley; lu2-du10-ga earns 50 litres and dšulgi-i3-li2, ur-ki-ku3-ga and ḫu-la-al each receive 30 litres. Children receive 10 or 15 litres. Text B, as usual, records a smaller amount of barley made up by the equivalent in dates. dnin-šubur-an-dul3 and šu-na-wi-ir are absentees/fugitives. Brewers (lu2-lunga3) A v 6–v 13; B v 20–v 27b; C v 75–v 82; D v 184–v 191: ma-aš2 (AB55[C?]D56); šu-ma-ma (ABCD57); dnin-šubur-i3-sa6 (A˹B˺˹B˺D); i-id-ra-bi2 (AB˹C˺D); urd šakkan2 (AB˹C˺D); li-bur-ni-aš (A˹B˺CD); nin-dub-sar (A[B]CD).

The team of brewers is constant in the four tablets. ur-dšakkan2 earns 60 litres,58 li-bur-ni-aš receives 40 litres and the other members of the team receive 30 litres, except for dnin-šubur-i3-sa6 who earns 20 litres. Leather-workers (ašgab) A v 14–v 26; B v 28- v 40; C v 83–v 94; D v 192–v 204: bi2-za-ti (A[B]CD); zi-za-num2 (ABCD); puzur4-er3-ra (ABCD) ip-qu2-ša (A˹B˺C59D); lu2-diĝir-ra (˹A˺B60C61D); er3-re-eb (ABCD62); lu2-du10-ga (AB˹C˺D); dnanše-kiri3-kal ([A][B][C]D); išma-an-še-en (A˹B˺C[D]); arad2-dlamma (A[B]C63[D]).

The small team of leather-workers is also quite consistent. They receive from 60 to 30 litres regularly. Children receive 10 litres. From a total of twelve people in this section, five are recorded as absentees/fugitives.

51 B v 6: ba-lu-˹x-x˺. 52 C iv 61: šu-lu-um. 53 B v 7: ˹x˺-ba-˹lu˺?. 54 C iv 62: a-zu. 55 B v 21: ur-˹lugal?˺ correct ma-x 56 D v 184: ma-šum. 57 D v 185: šu-ga?-ga?. 58 D v 188 records 0.0.1 to Ur-Šakkan, but probably the transliteration in Nisaba 15 is an error in transcribing Lambert’s manuscript. We assume a transliteration 0.1.0, as in the other three tablets. 59 C v 86: ib-ba-ša. 60 B v 32: ˹lu2˺-d˹šara2?˺. 61 C v 87: lu2-x-x. 62 D v 198: i-ri-ib. 63 C v 93: arad2-dšara2.

Texts and Workers

119

Craftsmen (gašam) A v 27–v 38; B v 41–vi 7; C v 95–vi 106; D v 205–v [221]: ur-niĝarĝar A[B]C64[D]); ri-iṣ-diĝir (A[B]C); i-re-eb (“silversmith” A[B]C); a-da-lal3 (“metalworker” A[B]C); a-lu-lu (“stone-cutter” (zadim) A[B?]C); aḫ-dam-i3-li2 (“rope-maker” A[B?]C); puzur4-e2-a (A[B?]C); an.gal.lu.kal (A[B?]C65); i3-li2-ir-ti (A[B?]C); e-pi.ni (A˹B˺66C).

In this generic section of “craftsmen” there is a range of professions, some with their speciality recorded in the text. ur-niĝarĝar “senior potter” (bahar3 bur-šuma); i-re-eb “silversmith” (ku3-dim2); a-da-lal3 “metalworker” (simug); a-lu-lu “stone-cutter” (zadim) and aḫ-dam-i3-li2 “rope-maker” (tug2-du8). Also noteworthy is the large number of absentees/fugitives recorded in this section or persons simply recorded in the section without any ration assigned. In fact, the only people who receive rations in the four texts are Ur-niĝar (60 litres), two sons of puzur4-e2-a (15 litres) and e-pi.ni (30 litres). Gendarmes (aga3-us2) A v 29–vi 24; B vi 8–VI […]; C v 107–vi 137; D vi 222–vi 255: a-budu10 (ABC67D); nu-ur2-diškur (ABC[D]); be-li2-du10 (AB68C[D]); u-˹bar˺-um (AB69[C][D]); šu-diškur (ABC[D]); en-um-i3-li2 (˹A˺B[C][D]); puzur4-dšara2 (AB[C][D]); puzur4-eš18-tar2 (AB[C][D]); i3-li2-aš-ra-ni (˹A˺B[C][D]); dšara2-an.dul3 (˹A˺B[C][D]); nu-ur2-eš18-tar2 (˹A˺B[C][D]); dutu-a-bi ([A]B[C][D]); ˹a˺-[p]il-la-ti ([A]˹B˺[C]D); ˹x˺-ri-šum ([A]B[C]); dšud suen-kal (˹A˺B[C]D); a-pi-la-ša (˹A˺B[C]D) dšu- dsuen-i3-li2 (AB[C]D); i-din-dsuen (AB[C]D); ki-lu-ba-ni-ib (A˹B?˺70[C]D); dutu-niĝ2-sig5 (A˹B˺71C72D73); a-da-lal3 (A[B?]CD]); arad2-diĝirra (A[B?]CD74); ur-si-ĝar (A[B?]C75D); dšara2-ba-zi-ge (A[B?]CD76); i-ib-na-da (A[B?]CD77); a-bu-du10 (A[B?]CD); ur-dšul-pa-e3 (A[B?]CD); e-PI-NI (B).

Text B preserves almost all this section, text A also preserves almost all this section and texts C and D preserve the end of the section. The total number of lines in these four texts is almost identical and for that reason we can conclude that the total amount of gendarmes in this section is constant in the four texts. At first glance, the number of absentees/fugitives of this section seems surpris-

64 65 66 67 68 69 70 71 72 73 74 75 76 77

No profession is provided in the transliteration of text C V 95. C v 102: dal-lu?-dan. B vi 6: ˹šu-dx˺. C vi 137: a-dim4-aš-kam. B vi 10: š[u]-˹i3˺-˹li2˺. B vi 11: [u]-bar-˹tum˺. B vi 29: ki-s[ar-…]. B vi 29: a possible variant of the name u4-de3-niĝ2-sig5. C vi 129: dšamaš-šub?-ši. D vi 247: dutu-ĝar-saga. D vi 249: arad2-dlu?. C vi 132: ur-e-[x]. D vi 251: dlu-ba-zi-gim6. D vi 252: i-id-na-da.

120

Lluís Feliu and Adelina Millet Albà

ing. Out of 29 names, 7 are absentees/fugitives. The amount of rations ranges from 60 litres to 10 litres for a daughter. Many people receive average salaries of 40, 30 and 20 litres. There is a certain “gradation” in the salaries and the preserved passages shown a regular amount of barley to each person. Animal drivers (kir4-dab5) A vi 25–vi 29; B […]; C vi 138–vi 142; D vi 256-[…]: ur-dḫal-mutum2 (A[B?]C78D); ur2-ra-ni (A[B?]CD); nu-bad-ma-ti-am3 (A[B?]C79[D?]).

Both this and the following section of musicians are unfortunately very damaged in texts B and D. Texts A and C are very homogeneous in this sections and show exactly the same people with the same rations. Musicians (nar) A vi 30–vi 39; B […]; C vi 143–VI 152; D […]: lu2-diĝir-ra (mušen-du3) (A˹B80˺C[D?]); inim-ma-a (A[B?]C81[D?]); a.ĝiš.ki (A[B?]C[D?]); ur-dšul-pa-e3 (A[B?]C[D?]); nu-ḫi-diĝir (A[B?]C[D?]); šu-la-lum (A[B?]C[D?]); a-li2-a-ḫi (A[B?]C[D?]); ša-at-er3-ra (A[B?]C[D?]); ša-at-dšu-šum (A[B?]C[D?]).

The section listing musicians is almost identical in text A and C, except for two people who receive 30 litres of barley in A but only 20 litres in text C. Unfortunately, text B is damaged in this section. Traces of writing in B vi 45–46 may indicate the reading of the first two names of this section, but the reading is doubtful. Ox-drivers (ša3-gu4)82 B […]-vii 11: ˹x˺-da-da ([A?]B[C?][D?]); šu-dsuen-i3-li2 ([A?]B[C?][D?]); pu3-˹x˺-zi ([A?]B[C?][D?]); lu2-ib-gal ([A?]B[C?][D?]); ([A?]B[C?][D?]).

This section appears as an exception because it is only recorded in text B. Unfortunately, this section is damaged in that text and seems to be missing from texts A and C (and it is completely missing from text D). Unskilled workers and miscellaneous section: A vi 40–vii 27; B vii 12–vii 30; C vi 153–vii 185; D […]:ib-ši-ka (A[B?]C83[D?]); […]˹x˺-na (A); ze2-ra-am (˹A˺B[C?][D?]); a-bu-um-diĝir (˹A˺B[C?][D?]); ul-lu-ki-ib-ri (AB[C?][D?]); ur-niĝarĝar i3-ra2-ra2 (AB[C?][D?]); Maltster (AB[C] [D?]); engar-zi (AB[C?][D?]); a-ba-dšara2-gen7, zu-lu-ti-a (AB[C?][D?]); ša-ru-še-ni (AB[C?] [D?]); ḫu-nu-um (AB84[C?][D?]); šu-eš18-tar2 (AB[C?][D?]); ḫu-zu-tum (AB[C?][D?]); ku-bu-ul-

78 79 80 81 82 83 84

C vi 138: dur-dpap-mu-ra2. C vi 141: nu-BE-x-ti-am3. B vi 44: lu2-˹diĝir-ra˺. C vi 144: ka-ba-a. On ša3-du4, see Studevent-Hickman 2006: 29 n. 48. C vi 153: it-ši-ka. B vii 24: ḫu-nu-uruda.

Texts and Workers

121

tum (AB85C86[D?]); zu-zu lu2-tir (AC[D?]) lu2-dinana (A[B?]C87[D?]); PI-ru-ni-du8 (AB88C89[D?]); a-ḫi-ṣi2-im-ti (A˹B˺C[D?]); geme2-dnin-mug (A[B?]C90[D?]); nin-me-a-DU (A[B?]C91[D?]); ṣi-il-diškur (A[B?][C][D?]); a-tu (ABC[D?]); inim-dšara2 (ABC[D?]).

This section is preserved only in texts A and B, while text C preserves the first names of the section and the last thirteen lines. According to Owen’s edition, the number of lines of this section of text C matches the number of lines in texts A and B. Therefore, we can predict the same personnel in this section of text C. This last section of rations mostly lists people without a profession in a miscellaneous section of unskilled workers. Only ur-niĝarĝar “perfumer” (i3ra2-ra2), an unnamed maltster (munu4-mu2) and zu-zu “forester” (lu2-tir) are recorded with a particular profession. The ration for each worker is standard in each text. Adults receive 60 litres of barley and children 10 or 15 litres. There are four absentees/fugitives. Engar-zi is recorded as absent in text A, but in text B he is recorded as receiving 40 litres of barley. After the miscellaneous section of workers and without either a blank or a double line, texts A, B and C (this section is missing from D) record amounts of food. Text A is the best preserved text in this section. In first place we find barley and bread as “expenditure from the table” (A vii 30: ĝišbanšur-ta ziga). Then the text records barley, bread, semolina (dabin) and aromatics (šim). Then we find a record of “regular offerings” (sa2-du11) to Nin-šubur and Šauška.92 According to BDTNS this text is the first occurrence of both deities together in the same text. Column viii (badly damaged or missing from the four texts) records totals and the date.

4 Classification of the Proper Names Each of the four texts studied in this paper contains practically the same proper names, and we could almost say that they are duplicates. Aside from slight

85 B vii 24: lu-bu-ul-tum. 86 C vii 173: ba-ni-bu-ul-tum. 87 C vii 176: lu2-dnana. 88 B vii 31: […] i3-du8. 89 C vii 177: wi-ru i3-du8. 90 C vii 179: geme2-dnuska. 91 C vii 181: ddam-a-du. 92 The goddess Šauška in scarcely attested in Ur III texts. Her cult is documented in Drehem and Umma. According to Wilcke, Tiāmat-bāšti, the second lukur of Šu-Suen, was responsible for introducing her cult to the Ur III court (Trémouille 2009).

122

Lluís Feliu and Adelina Millet Albà

differences, they are almost exactly the same, but not actually identical. The proper names are arranged in various columns on the obverse and reverse of each tablet. A high percentage of the proper names that appear in the texts are of Semitic origin and can be identified as either east or west Semitic. In fact, about two thirds of the proper names have a Semitic origin. We observe that there is also a significant percentage of proper names of Sumerian origin, about sixty in total. It is important to note that some hybrid names (i.e. Sumerian and Semitic) are classified as Semitic, when at least one element of such a compound name is Semitic, so we assume that they should be read as Semitic. A small number of names may be Hurrian, slightly more than ten. Finally, another small group of names is of unknown origin. Because of the bad state of conservation of the texts in many places, some proper names are incomplete or are very difficult to interpret. In any case, among the incomplete or broken names, the same ratio between Semitic, Sumerian, Hurrian and unknown personal names can be seen as in the complete proper names. The great majority of the names are Semitic, but some may be Sumerian, Hurrian or of unknown origin.

4.1 The Semitic Personal Names As usual, when analysing the Semitic personal names we have to face the general problem that in many cases it is very difficult to establish the Semitic language in question. Sometimes it is hard to determine whether a Semitic personal name is west Semitic or east Semitic. This is because the same basic vocabulary is used to form personal names. This situation can also be observed, in fact, in most of the textual collections dated to the end of the third millennium and the second millennium. We have to admit that it is very annoying the impossibility of establishing precisely whether those names are east or west Semitic. It occurs when the Semitic elements used to form a personal name are common to both branches of the Semitic language family. On the contrary, in many occasions it is very easy to identify the personal names as east or west Semitic, because the elements used can be identified without any doubt. In that case, if we find personal names such as Abum-ilum or Eštar-Ummī, we have to admit that there is no way to establish whether they are west or east Semitic. The reason is clear: the common terms used in those personal names, and others are known as “cultural words,” which are common to many Semitic languages. For certain personal names, however, there is clear evi-

Texts and Workers

123

dence that they should be classified as west or east Semitic. For example, Eštartukultī or Puzur-dŠara, are undoubtedly east Semitic personal names.93 In research involving anthropology, population, ancient societies and the like, the possibility of classifying proper names can be very useful to establish the origins of a population, or the incidence of a given population in a certain place at a given time. But, even if the linguistic origin of a given proper name can be determined, this does not necessarily imply the ethnicity of the bearer. As everyone knows, even from experience, there are several reasons for giving personal names, including the latest fashion. Normally the decision is made by the person’s parents or superior, but sometimes the choice is personal. As is well known, in the ancient Near East, any individual could change his or her name several times for a variety of reasons.

4.1.1 Theophoric Personal Names A significant percentage of the Semitic proper names are theophoric. There may be many reasons for divine names used in the formation of personal names, in particular, the birthplace of the bearer. In many cases the most important divinities of the pantheon of the place of origin are used to form personal names. In those cases, the place of origin of a document can be established. But the personal piety of the bearer of the name, or of his or her family, counts when giving a name. Ada, Eštar and Šara are the most frequent divinities in the personal names that appear in the texts studied in this article. Less frequent are Sîn, Ea, Erra, Nanaya and Šamaš. The god Šušum occurs once. The frequent use of ilu (or diĝir, in Sumerian), as a theophoric element in proper names is interesting as it may refer to a personal god.

4.1.2 Non-Theophoric Personal Names The non-theophoric Semitic personal names are normally descriptive. Alternatively, they may refer to the individuals related to the bearer of the name: aḫu or abu, for example.

93 In other documentations, but is not the case of the texts object of this paper, there exist undoubtedly west Semitic personal names, for exemple, Aqba-ahum or Hammu-rabi, in Mari.

124

Lluís Feliu and Adelina Millet Albà

4.2 The Sumerian Personal Names In the texts concerned in this article we find an important number of Sumerian personal names, as noted above. Considering the date of the texts – the Ur III period – it would be quite normal to find so many personal names expressed in Sumerian. We also find Semitic personal names with a logographic element, which were therefore classified as Semitic, possibly Akkadian. There is no rule concerning which element is written logographically. Sometimes it is the theophoric element, which would not be unusual, but occasionally it is the nontheophoric element. Here we present some examples: Rīš-diĝir = Rīš-ilu Ṣiluš-du10 = Ṣiluš-ṭāb d Utu-abī = Šamaš-abī

4.2.1 Theophoric Personal Names It is very interesting to note that Sumerian theophoric personal names are very frequent in these lists. There are more than thirty theophoric personal names, out of a total of about sixty Sumerian personal names. This means that half the Sumerian personal names are theophoric. The following divinities are mentioned: dBa-ba6, dEnlil, dInana, dIškur, dIštaran, dLamma, dNanna, dNanše, d Nin-šubur, dNungal, dSu’en, dŠakan, dŠara, dŠulpa’e, dTišpak, dUtu.

4.2.2 Non-Theophoric Personal Names The non-theophoric personal names mainly refer to the king, the lord (possibly a reference to the king), or the lady (perhaps referring to an anonymous goddess).

4.3 Hurrian Personal names The incidence of Hurrian personal names is usually very high in documents dated to the second millennium BC, particularly in the Old Babylonian period. However, it seems that this is not the case here. In the texts concerned in this paper, there are scarcely more than ten personal names that can be identified as Hurrian with certainty. There are two possible explanations for this. First, the Hurrian population was not very sig-

Texts and Workers

125

nificant in Mesopotamia in the Ur III period. Second, the tradition of giving Hurrian personal names was not widespread. In fact, both these causes could be related.

4.4 List of the Personal Names Semitic Personal Names Abu-ṭāb A v 39; vi 22, B vi 8, C vi 135, D vi 222, vi 253. Abum-ilum A vii 5, B vii 15. Adad-dan A iv 17; B iv 32, C iv 43; D iv 148. Adaga? A iii 15, B iii 20 (he has three sons En-xxx, dŠara-ba-zi-ge, and a third unnamed son), C iii 31, D iii 99. f Adduzu A iii 16, B iii 21, C iii 32, D iii 100. f Agatiya A ii 29, B ii 33, D ii 70. f Agatum Ai 19, B i 19, D i 19. Aḫdam-Ilī A v 32, C v 100. Aḫī-ṣimtī A vii 20, B vii ˹32˺, C vii 178. Aḫušuni A iv 21, B iv 36, C iv 47, D iv 156. f Ali-abī A iii 24, B iii 34, D iii 114. f Ali-aḫī (daughter of fDummuq-šeni, sister of fUmmī-Nunu) A ii 14, vi 36, B ii 13, C ii 15, vi 149 f Alla-šarrum A iii 27; B iii 37. f Ama’a A ii 28, iii 3, B ii 32. C ii 29, D ii 69. f Ana-Eštar-taklat A ii 41, B iii 4, D iii 84. Apilaša A vi 12, B vi 25, D vi 243. Apillati B vi 22. Apil-nūrī A ii [37]; B ii 41, D iii 80. f Arbitum A ii 31, B ii 35, D ii 72. fd Ašar-imti B iii 7. f Ašati-nūrī A i 38, B i 38. f Ašdaḫuda? Ai 16, B i 17, D i 6. Awīl-šālim B iv 19, D iv 140 f Azat-alla A ii 30, B i 36, D i 38. f Azi-ummī B i 46, D ii 45. f Azuzu B iii 27, D iii 109. f Baluša A iii 26, B iii 36, D iii 116. f Banisa A ii 1, B ii 1, D ii 46. Batri B iv 20, D iv 138. Bēlī-ṭāb A v 41, B vi 10, C v 109. Billu (has one son and one daughter unnamed) A ii 8, B ii 8, D ii 53. Bizati A v 14, B v 28, C v 83, D v 192. f DaDUra Ai 18, B i 21, D i 18. f Eli-šarrum A iii 27, B iii 29, D iii 106. f Ennilu (she has an unnamed son) A i 26, B i 26, D i 26. Ennum-ilī A iv 5, 7, vi 1, B iv 17 (son of fŠāt-Erra), vi 15, D iv 137.

126

Lluís Feliu and Adelina Millet Albà

Errēb A v 20, 29, B v 34, C v 89, 97, D v 198. Errēšum B vi 23. f Eštar-bītum A ii 2, B ii 2, D ii 47. f Eštar-puzurri A iii 4, B iii 8, D iii 88. f Eštar-rabiat A ii 32, B ii 36, D ii 73. f Eštar-tukultī Ai 10, B i 10, D i 10. f Eštar-ummī A ii 33, B ii 37, D ii 74. f Gabuati A i 31, B i 32, D i 32. f Gunana B iii 28, A iii 19, D iii 107. f ḪalNI A i 3, B i 3 (mother of fKinibši and one unnamed daughter), D i 3. Ḫulal A v 3, B v 17, C v 72. Ḫunum A vii 12, B vii 24. f Ḫuzutum A vii 14, B vii 26. d Iʾib-laba (son of Lugal-he2-ĝal) B iv 27, D iv 145. Ibnada A vi 21, C vi 134, D vi 252. Ibšika (?) A vi 40, C vi 153. Iddin-Sîn A vi 14, B vi 28, C vi 127, D vi 245. Idrabi ? A v 9, B v 24, C v 78, D v 187. f Ilī-ašrani A vi 4, B vi 18. Ilī-irti A v 35, C v 103. f Imdīdam A i 20, B i 19, D i 20. Ipquša A v 17, B v 31, C v 86 (father of Lu2-dŠara2 and f[…i]rib), D v 195. f Išmeani A ii 35, B i 33, ii 39, D i 33, ii 76. Kilubanib A vi 15, B vi 29, C vi 128, D vi 246. f Kinibši A i 4, B i 4, D i 4. f Kizanada A ii 40, B iii 2, D iii 83. f Kubat A i 28, B i 29, D i 28. f Kubultum A vii 15, B vii 27, C vii 173. f KUšati A iii 32, B iv 2, D iv 123. Libūr-bēlī B iv 33. Libūr-niāš A v 11, B v 26, C v 80, D v 189. f Lubali A iii 1, B iii 5, D iii 85. Minarni (Miniarni) A iii 2, B iii 6, D iii 86. f Nabritum A i 9, B i 9, D i 9. Namura (son of Šū-ilī) A iii 35, B iv 5, D iv 126. f Nanaya-nūrī A iii 8, B iii 13, D iii 92. Nibašḫati A iv 2, B iv 14, D iv 134. Nūḫ-ilum: A vi 34, C vi 147. Nūḫim-Eštar B iv 22. Nūr-Adad: A v 40, B vi 9, C v 108. Nūr-Eštar A vi 6, B vi 20. Nūr-Sîn A iv 42, C iv 68. Nūrum A iv 27, B iv 43, (father of Ur-diškur, A-a-lu2-du10, Lugal-kiri3-zal and diškur-an-dul3), C iv 53, D iv 162. f pirurutum A ii 24, B ii 23, C ii 25, D ii 65. Puzur-Ea A v 33, C v 101. Puzur-Erra A v 16, B v 30, C v 85, D v 194. Puzur-Eštar A vi 3, B vi 17.

Texts and Workers

Puzur-Šara A vi 2, B vi 16. f Qadušanum Ai 25, B i 25, D i 25. Rīṣ-ilim A v 28, C v 96. f Ṣalimtum A i 7, B i 7, D i 7. Ṣil-Adad A vii 25, C vii [183]. f Ṣilluš-ṭāb A ii 25, B ii 24, C ii 26, D ii 66. Šamaš-abī B vi 21, D vi 239. Šalim-bēlī A iv 1, B iv 13, D iv 133. Šalim-nīrum (has an unnamed daughter) B iii 11, D iii 91. f Šatdati A iii 23, B iii 33, D iii 113. f Šāt-Erra A iv 4 vi 37; B iv 16, C vi 150, D iv 136. f Šāt-ilī A i 39, B i 39, ii 25, D i 39, iii 108. f Šāt-dNanaya (she has an unnamed son) A iii 11, B iii 16, D iii 95. f Šāt-Nunu A iii 22, B iii 32, D iii 112. f Šāt-dŠušum A vi 38, C vi 151. Simat-Adad: A iii 9, B iii 14, D iii 93. f Simat-Tišpak B ii 28. f Šinūnūtum A i 27, B i 28. Šū-Adad A v 43, B vi 12 (son of Ubartum) m Šudanni/Šudata (son of Išmeani) A i 33, B i 34, D i 33. Šudaniya B iv 29. Šudata A iv 14, C iv 40. Šū-Erra A iv 24, B iv 40, C iv 50, D iv 159. Šū-Eštar A vii 13, B vii 25. Šū-ilī A iii 33, B iv 3, D iv 124. f Šū-immini A ii 30, B ii 34, D ii 71. Šulalum A vi 35, C vi 148. d Šulgi-ḫamati A iv 22, B iv 38, D iv 157. Šulgi-ilī A iv 40, C iv 66. Šulgi-tūrī (has an unnamed son and two unnamed daughters) A iii 28, B iii 38. Šumama A iv 16, v 7, B iv 31, v 22, C iv 42, v 76, D iv 147, v 185. f Šumuštum (she has an unnamed daughter) A i 11, B i 11 D 11. Šū-nāwir A v 4, B v 19, C v 73, D v 182. Šunium B vii 19. f Šū-Sîn-imdī A iii 6, B iii 10, (she has an unnamed daughter). Šū-Sîn-ilī A vi 13, B vi 26, vii 8, C vi 126, D vi 244. Šū-Sîn-dan: A vi 11, B vi 24. f Šū-Sîn-nūrī A ii 34, B ii 38, D ii 75. f Šū-Sîn-tūrī A iii 5, B iii 9, D iii 89. Šutirum (son of Šu-ilī) A iii 34, B iv 4, D iv 125. f Tadin-Eštar (she has an unnamed daughter) A i 29, B i 30. f Ubartum B ii 6, vi 11 (mother of Šū-Adad and of an unnamed daughter), D ii 51. Ullu-kibrī A vii 6, B vii 16. f Ummī-Nunu (daughter of fDummuq-šeni, sister of fAli-aḫī) A ii 15, B ii 14, C ii 16. f Ummi-ṭāb A i [2], B i 2, D i 2. f Ummiya A i [1], B i 1, D i 1. Uqnitum (son of Šu-ilī) A iii 36, B iv 6, D iv 127. f Uttum A ii 3, B ii 3, D ii 48.

127

128

Lluís Feliu and Adelina Millet Albà

f

Ušaknil? A ii 39, B iii 1; D iii 82. Uzili? A i 44, B i [45]. f Zelua? A i 15, B i 15, D i 15. Zimudar A iii 10, B iii 15, D iii 94. Zizanum A v 15, B v 29, C v 84, D v 193. Zulum A iv 35 (father of Aba and of an unnamed daughter), B v 6, C iv 61. Zulutiya A vii 10, B vii 22. Zuzu A vii 17, C vii 175. f

Sumerian Personal Names A-a-lu2-du10 A iv 29 (son of Nurrum and brother of UR-dIŠKUR, Lugal-kiri3-zal and diškur-andul3), C iv 55, D iv 164. A-ba ?? (son of Zulum, he has an unnamed daughter) A iv 36, B v 7, C iv 62. A-da-lal3 A v 30, vi 17, C iii 31, v 98, vi 130, D vi 248. A-ba-dŠara-gen7 B vii 21. a.ĝiš.ki A vi 32. f A-tu A vii 26, B vii 29, C vii 184. f Ama-ge-na A i 17, 34, ii 17, B i 16, 23, 35, ii 16, C ii 18, D i 7, 23. Am3-ma-a A ii 28, B ii 32 (a-ma-a), C ii 29, D ii 69, iii 87. AN-lu-sa6 (son of Lugal-he2-gal2) A iv 11, B iv 25, D iv 143. Arad2-diĝir-ra A vi 18, C vi 131, D vi 249. Arad2-dLamma A v 25, C v 93. Engar-zi A vii 9, B vii 20. f Ereš-diĝir-ĝu10 A i 14, B i 14, D i 14. f Geme2-e11-e (daughter of Amagena) A i 24, B i 24, D i 24. f Geme2-dEnlil (daughter of Balu[xx]) B v 8. Geme2-ĝišgigir A i 37, B i 37. f Geme2-dIškur A ii 23, B ii 22, C ii 24, D ii 64. f Geme2-dIštaran A i 22, B i 22, D i 22. f Geme2-dNanna B ii 27. Geme2-dNin-mug A vii 21, C vii 179. f Geme2-dSîn A iii 17, B iii 26, C iii 33, D iii 105. f Geme2-dUtu A ii 27, B ii 31, C ii 28, D ii 68. f Gibil-la A iii 21, B, iii 31, D iii 111. Inim-ma-a A vi 31, C vi 144. f Inim-dŠara A vii 27, B vii 30, C vii 185. d Iškur-an-dul3 (son of Nurrum and brother of ur-diškur, Lugal-kiri3-zal and A-a-lu2-du10), A iv 31 C iv 57, D iv [166]. Lu2-diĝir-ra (has an unnamed son) A v 18; vi 30, B v 32, C v 87, vi 143, D iv 150, v 196. Lu2-du10-ga A v 2, 22, B v 18; v 36, C v 71, [90], D v 200. Lu2-ib-gal B vii 10. Lu2-dInana A vii 18, B vii 14, C vii 176. Lu2-dNanna A iv 8, B iv 21, D iv 139. Lu2-dSîn A iv 3, B iv 15, D iv 135. Lu2-uš-gi-na D iv 152. Lugal-he2-gal2 A iv 10, B iv 24, C iv 36, D iv 142. Lugal-kiri3-zal A iv 30 (son of Nurrum and brother of ur-diškur, A-a-lu2-du10 and diškur-andul3), C iv 56, D iv [165].

Texts and Workers

129

Lugal-ku3-zu-a A iv 25, B iv 41, D iv 160. Ma-aš2 A v 6, C v [75], D v 184. Nam-tar-ib2-gu-ul D iv 153. d Nanše-kiri3-kal D v 201. Nin-dub-sar A v 12, B v [27], C v 81, D v 190. f Nin-e-i3-sa6 A i 13, B i 13, D i [13]. f Nin-e2-ku3-ta A ii 20, 21, B ii 19, 20, C ii 21, 22, D ii 61, 62. f Nin-e2-mah-e A i 8, B i 8, D i 8. f Nin9-kal-la A ii 5, B ii 5, D ii 50. Nin-me-a-du A vii 23, C vii 181. f Nin-me-du B ii 26. d Nin-šubur-an-dul3 A iv 39, C iv 65. d Nin-šubur-i3-sa6 A v 8, B v 23, C iv 37, v 77, D v 186 d Nin-šubur-zi-gu10 A iv 15, B iv 30, C iv 41 Nu-bad-ma-ti-am3 A vi 28, C vi 141. Saĝ-nin-e-i3-zu A ii 4, B ii 4, D ii 49. d Šara2-an-dul3: A vi 5, B vi 19. d Šara2-ba-zi-ge A vi 20, B iii 23, C vi 133, D iii 102, vi 251. d Šulgi-ha-ma-ti A iv 22, B iv 38, C iv 48, D iv 157. f U3-ba A ii 22, B ii 21, C ii 23, D ii 63. Ur-dBa-ba6 A iv 20, B iv 35, C iv 46, D iv 155. Ur-dḫal-mu-tum2 A iv 33, vi 25, B v 4; C iv 59, vi 138, D v 168, vi 256. Ur-dIškur (son of Nurrum and brother of A-a-lu2-du10, Lugal-ka-zal anddiškur-an-dul3) A iv 28, C iv 54, D iv 163. Ur-ki-ku3-ga A iv 41, C iv 67. Ur-lugal (son of Lugal-he2-gal2) A iv 12, B iv 26, v 21, C iv 38, D iv 144. Ur-dNanše (son of A-ru-a-me) A iv 18, B iv 34, C iv 44, D iv 149. Ur-dNin-a-zu D iv 151. Ur-niĝar A v 27; vii 7, C v 95. Ur-dNungal (son of Billu) A ii 9, C ii 10, D ii 54. Ur2-ra-ni (has an unnamed daughter) A vi 26, C vi 139, D vi 257. Ur-si-ĝar A vi 19, C vi 132, D vi 250. Ur-dŠakkan2 A v 10, B v 25, C v 79, D v 188. Ur-dŠul-pa-e3 (has an unnamed son and two unnamed daughters) A vi 23; 33, B iv 8, C vi 136, 146, D iv 12994, vi 254. d Utu-niĝ2-sig5 A vi 16, B vi 30, C vi 129, D vi 247. Hurrian Personal Names f Ad-du-en-na A i 21, B i 20, D i 21. f Du-mu-uq-še-ni (mother of fAli-Aḫī and fUmmī-Nunu) A ii 13, B ii 12; C ii 14. f E-ni-ša-ag B ii 29. f En-ni-ma-ad A ii 38, B iii 3, D iii 81. Iš-ma-an-še-en A v 24, B v 38, C v 92. f Ša-ru-(uš)-še-ni A ii 35; vii 11, B ii 40; vii 23, D iii 79?.

94 In Text B he has one son and two daughters.

130

Lluís Feliu and Adelina Millet Albà

f

Unazi A i 43, B i [44], D ii 42. Other A-lu-lu A v 31, C v 99. an.gal.lu.kal A v 34, C v 102. f Ba-la-la-a A ii 19, B ii 18, C ii 20, D ii 60. Da/ta-sal-la A i 45, D ii 44. E.pi.ni A v 37, B vi 6, 27, C v 105, D vi 240. f Ga-na-za A ii 17, B ii 17, C ii 19, D ii 59. Mi-ni-mi A iii 13, B iii 18, D iii 97. Nar-zu-mu A iv 23, B iv 39, D iv 158. pi-ru-ni.du8 A vii 19, C vii 177. Ut-tum A ii 3; B ii 3; D ii 48. U2-zi-li A i 44, B i [45], D ii 43. Za-ne-a A ii 7, B ii 7, D ii 52. Broken Personal Names AN[…] A vi 10. AN-X-DU D vi 241. Ba-lu-˹x-x˺ B v 6. E-la-[…] C iii 34. En-˹x-x-x˺ Text B iii 22 (son of fAduzzu, brother of dŠara-ba-zi-ge), Text D iii 101.95 d Ga-˹x-x˺ B v 37. Ḫu-l[a-x] B v 17. Ilī-[…] C iv 49. d Nin[…] B v 11. PU3-˹x˺-zi B vii 9. Sîn-[…] D iii 117. ˹Šu-x-x˺ B vi 5. d Š[ul-gi…] B v 12. […]-da-a-(a) A i 41, B i 41, D i [41]. ˹x˺-da-da B vii 7. […]-ni-il B i 40. f […i]-ri-ib B v 34 (daughter of Ipquša).

95 En-tukumbi?

Texts and Workers

131

5 Edition of Text A Measurements: 149 × 110 × 25 mm. Four-column tablet, almost complete, with the surface flaking in some sections of the obverse and reverse. Obverse i 1.

i

5.

i

10.

i

15.

i

20.

i

25.

i

30.

[… še-ba um-mi-a] [… um-mi-du10] [… ḫa-a]l-ni [… ki-n]i-[i]b-ši [… d]umu-[mu]nus-˹a˺-ni [dumu-ni]-me [x.x.x ṣ]a-lim-tum [… nin]-˹e2˺-mah-˹e˺ [0.0].˹3˺ na-[ab]-ri2-[tu]m? […] eš18-˹tar2-tu˺-kul2- [… šu-mu-uš]-˹tum˺ [dumu-munus-ni] [0.0].˹3˺ nin˺-[e]-˹i3˺-sa6 […] ereš-diĝir-ĝu10 ˹ninda˺ ˹0.0.3˺ ze2-˹lu˺-a ˹0.0.3˺ aš-da-ḫu-da ˹0.0.3˺ [x?] ama-ge-na [… d]a-du-ra ˹0.0.3?˺ ˹a˺-ga-tum [ninda?]˹0.0.3?˺ [i]m-di3-dam […] ad-du-en-na [… ge]me2-dištaran [… a]ma-ge-na […] ge[me2]- ˹e11˺-e dumu-ni ˹0.0.3?˺ [q]a2-du-ša-num2 ˹x˺ en-ni-lu ˹0.0.1˺.5 (sila3) [š]i-nu-nu-tum 0.0.3.˹5(sila3)˺ ku-ba-at 0.0.3.˹5? (sila3)˺ [t]a2-din-eš18-tar2 0.0.1.5 (sila3) [du]mu-munus-ni 0.0.3 [… g]a-bu-a-ti ˹0.0.3˺ […i]š-me-a-ni [… šu-d]an-ni dumu-a-ni ˹0.0.3˺ ama-ge-na

132

i

i

i

ii

ii

ii

ii

ii

ii

Lluís Feliu and Adelina Millet Albà

35.

[… du]mu-munus-ni […]˹a˺-za-[a]t-al-la [… gem]e2-ĝišgigir […].˹5 (sila3)˺ ˹a˺-ša-ti-nu-ri […] š[a]-˹at-i3-li2˺ 40. […] ˹x˺ ú-[…] […] ˹x˺-˹da?˺-˹a˺-a […]-˹šu?-zi?˺ […] ˹u2˺-˹na˺-zi […] ˹x˺ u2-˹zi˺-li 45. […] ˹x-x-x˺ ta-sal-la L.E. [x.x].2.1 […] 1. [… ba-n]i-sa3 […]˹eš18-tar2-bi2-tum˺ ˹0.0.3˺ [ut]-˹tum˺ ˹0.0.2˺.5 (sila3) ˹saĝ-nin-e-i3-zu˺ 5. zah3 ˹nin9-kal-la˺ 0.0.2.5 (sila3) u-bar-t[um] zah3 za-n[e-e2]-˹a˺ 0.0.2 ˹bil-lu˺ 0.0.1 ur-d n[un-ga]l 10. 0.0.1 dumu-nita-˹ni˺ 0.0.1 dumu-munus-a-ni dumu-ni-me 0.0.5 du-mu-uq-še-ni 0.0.3 a-li2-a-ḫi 15. 0.0.2 um-mi-nu-nu dumu-ni-me --- ama-ge-na 0.0.2 ga-na-za zah3 ba-la-la-a 20. 0.0.1.5 (sila3) nin-e2-ku3-ta --- nin!(geme2)-e2-ku3-ta min min 0.0.3 u3-ba 0.0.2 geme2-diškur --- pi-ru-ru-tum 25. 0.0.1.5 (sila3) ṣi-lu-uš-du10 (Blank)

Texts and Workers

ii

30.

ii

35.

ii

40. L.E.

iii

1.

iii

5.

iii

10.

iii

15.

iii

20.

uš-bar-me 0.0.2 geme2-dutu 0.1.0 am3-ma-˹a˺ ninda 0.0.3 (sila3) ˹a-ga-ti-a˺ 0.0.1.5 (sila3) šu-i[m]-m[i-ni] 0.0.3 a[r]-b[i2-tum] ninda ˹0.0.3˺ ˹eš18-tar2˺-˹ra˺-[bi2-at] 0.0.3 ˹eš18-tar2˺-[um-mi] ˹0.0.x˺ ˹d˺š[u-dsuen-nu-ri] ˹ninda˺ ˹0.0.3?˺ ˹iš-me˺-[a-ni] ˹0.0.3?˺ [ša]-˹ru˺-˹še˺-ni ˹0.0.3?˺ [a-pil-nu-ri] ninda 0.0.3 ˹en˺-ni- ˹ma-ad˺ ˹ninda 0.0.3˺ ˹u3˺-ša-˹ak˺- ˹ni˺-˹il˺ ˹0.0.2.5˺ (sila3) ki-za-na-d[a] 0.0.2 a-na-eš18-tar2-[t]a2-ak-/˹la˺-at 2.1.4.5 sila3 (še-gur) 0.2.3 ninda [x-(x)]-˹x-x˺ ˹0.0.4˺ ˹lu-ba˺-li 0.0.2 mi-n[i-a]r-ni ˹ninda 0.0.3˺ ˹am3˺-[m]a-a-a ˹ninda 0.0.3?˺ ˹eš18-tar2˺-[pu]zur4-ri ˹ninda 0.0.3?˺ ˹dšu˺-dsuen-tu-/ri ˹ninda 0.0.3?˺ ˹dšu˺-dsuen-im-dì ˹0.0.1?˺ ˹dumu˺-[mun]us-ni […] ˹dna˺-na-a-nu-ri ˹0.0.2˺ ˹me˺-diškur ˹0.0.2?˺ zi-mu-dar ˹0.0.3?˺ ˹ša-at˺-dna-na-/a 0.0.1 d[umu]-nita-a-ni --- [m]i-ni-mi ša3-˹e2˺-a-me zah3 a-da-ga 0.0.3 ad-du-˹zu˺ 0.0.3 geme2-d˹suen˺ 0.0.3 ˹e-la-ša-ru-um˺ 0.0.3 gu-na-na ˹ar3˺-ra-me 0.0.3 ˹gibil-la˺

133

134

Lluís Feliu and Adelina Millet Albà

iii

25.

iii

30.

iii

35.

iii

40. L.E.

iv

1.

iv

5.

iv

10.

iv

15.

0.0.3 ˹ša-at˺-nu-nu ˹0.0.3˺ [ša-at-da]-˹ti˺ […] [a-l]i2-a-˹bi˺ ĝiš-˹i3˺ [sur-s]ur-me ˹0.0.3˺ [ba]-˹lu-ša˺ 0.0.1.˹5˺ (sila3) [al-l]a-ša-ru-[um] […] [d][š]ul-g[i-t]u-/ri ˹0.0.1˺ [dumu-ni]ta-a-ni […] d[umu-mun]us-a-ni [dumu-ni]-me […]˹ku˺-ša-ti […] ˹šu-i3˺-li2 ˹0.0.1.5˺ (sila3) [š]u-ti-ru-um ˹0.0.1.5˺ n[a-m]u-ra ˹0.0.1˺ [u]q-ni-tum du[mu]-ni-me --- ur-dšul-pa-e3 […] dumu-nita-a-ni --- dumu-munus-a-ni ˹dumu˺-ni-me 2.1.1.x ˹sila3˺ (še-gur) 0.2.˹x˺ ˹ninda˺ 0.0.4 ša-lim-be-li2 0.0.3 ni-ba-aš-ḫa-ti 0.0.4 lu2-dsuen ninda 0.0.3 ša-at-er3-ra 0.0.5 en-um-i3-li2 --- dumu-nita-a-ni 0.0.4 en-um-i3-li2 0.0.1 lu2-dnanna azlag7 -me 0.1.0 lugal-ḫe2-˹ĝal2˺ 0.0.1.5 (sila3) an-lu-sa6 0.0.1 ur-lugal dumu-ni-me 0.1.0 šu-da-ta 0.0.1.5 (sila3) dnin-šubur-zi-ĝu10 0.1.0 šu-ma-ma ninda 0.1.0 diškur-kal zah3 ur-dnanše

Texts and Workers

iv

20.

iv

25.

iv

30.

iv

35.

iv

40.

L.E.

Reverse v 1.

v

5.

v

10.

si12-a-me ninda 0.0.4 ur-dba-ba6 ugula šaḫ2 zah3 a-ḫu-šu-ni 0.1.0 dšul-gi-ḫa-ma-ti 0.0.4 [na]r-z[u-m]u 0.0.5 šu-e[r3-ra] 0.0.5 lu[gal-ku3]-zu lu2-gi-zi-me 0.1.0 nu-ur2-um 0.0.5 ur-diškur 0.0.2 a-a-lu2-du10 0.0.2 lugal-kiri3-zal 0.0.1.5 (sila3) diškur-an-dul3 dumu-ni-me 0.1.0 ur-d-mu-tum2 ma2-lah5-me 0.1.0 zu-lu-um 0.0.1.5 (sila3) a-ba 0.0.1 dumu-munus-a-ni dumu-ni-me zah3 dnin-šubur-an-dul3 ninda 0.0.3 dšul-gi-i3-li2 0.0.3 ur-ki-ku3-ga 0.0.1 nu-ur2-dsuen 3.1.2 gur 0.˹2˺.4 ninda

0.0.1 dumu-munus-a-ni 0.0.5 lu2-du10-ga 0.0.3 ḫu-la-al […]˹šu-na˺-wi-ir [muhaldi]m-me ˹0.0.3˺ ma-aš2 ˹0.0.x˺ šu-ma-ma […] ˹dnin-šubur˺-i3-sa6 0.0.3 i-id-ra-bi2 0.1.0 ur-dšakkan2 0.0.4 ˹li˺-bur-ni-aš 0.0.3 nin-dub-sar

135

136

Lluís Feliu and Adelina Millet Albà

v

15.

v

20.

v

25.

v

30.

v

35.

v

40.

v

45. L.E.

vi

1.

lu2-lunga3-me 0.0.4 bi2-za-ti 0.0.4 zi-za-num2 0.0.3 puzur4-er3-ra ˹0.1.0˺ ip-qu2-ša 0.1.0 lu2-di[ĝir-ra] ˹0.0.1˺ [du]mu-nita-a-ni z[ah3] [e]r3-re-eb d[umu-munus]-˹a˺-ni ˹x˺ l[u2-du10-g]a zah3 d˹x˺-[…] zah3 iš-˹ma˺-an-še-en zah3 arad2-dlam[ma] ašgab-me 0.1.0 ur-niĝarĝar baḫar3 bur-šu-ma zah3 ri-iṣ-diĝir --- i-re-eb ku3-dim2 --- a-da-lal3 simug zah3 a-lu-lu zadim zah3 aḫ-dam-i3-li2 tug2-du8 zah3 puzur4-e2-a ninda 0.0.1.5 (sila3) an.gal.lu.kal ninda 0.0.1.5 (sila3) i3-li2-ir-ti dumu-ni-me 0.0.3 e-pi.ni nagar gašam-me --- a-bu-du10 ˹zah3 nu-ur2˺-diškur ˹0.0.1.5 (sila3) ˹be-li2˺- du10 […] u-˹bar˺-um […] ˹šu˺-diškur [0.0.1 dumu-munus]-˹a˺-ni [dumu-ni]-me ˹2?.2?.3.5˺ (sila3) ˹gur˺? 0.1.0 ninda […] [en-u]m-i3-li2 […] [puzur4]-˹dšara2˺ […] [puz]ur4-eš18-tar2 […] ˹i3-li2˺-/[aš-r]a-ni

Texts and Workers

vi

5.

vi

10.

vi

15.

vi

20.

vi

25.

vi

30.

vi

35.

vi

40.

[…] ˹dšara2-an-dul3˺ […] [nu]-˹ur2-eš18-tar2˺ ˹zah3˺ […] ˹zah3˺ […] ˹zah3˺ […] ˹zah3˺ AN-[…] zah3 d[šu-dsue]n-˹kal˺ zah3 a-p[i-l]a-[š]a zah3 dšu- dsuen-i3-/li2 0.0.2.5 (sila3) [i]-din-dsuen 0.0.2 ki-lu-ba-n[i-i]b ninda 0.0.3 dutu-niĝ2-˹sig5˺ 0.0.3 a-da-˹lal3˺ 0.0.1.5 (sila3) arad2-diĝir-˹ra˺ ninda 0.0.1 ur-si-[ĝar] ninda 0.0.1.5 (sila3) dšara2-ba-˹zi/-ge˺ 0.0.4 i-ib-na-da 0.0.2 a-bu-du10 --- ur-dšul-pa-e396 aga3-us2-me 0.0.5 ur-dḫal-˹mu-tum2˺ 0.1.0 ur2-ra-ni 0.0.1 dumu-munus-a-ni 0.0.2.5 (sila3) nu-bad-ma-ti-am3 kir4-dab5-me 0.0.4 lu2-diĝir-ra mušen-du3 0.1.0 inim-ma-a zah3 a.ĝiš.ki --- ur-dšul-pa-e3 ninda 0.0.3 nu-ḫi-diĝir ninda 0.0.3 šu-la-lum --- a-li2-a-ḫi --- ša-at-èr3-ra --- ša-at-dšu-šum nar-me ˹ninda˺ ˹0.0.5˺ ˹ib˺-ši-ka […]˹x˺-na

96 Erasure before ur.

137

138

Lluís Feliu and Adelina Millet Albà

L.E. vii

1.

vii

5.

vii

10.

vii

15.

vii

20.

vii

25.

vii

30.

[…]˹x-x˺-NI […] […] ˹2.2.0.5?˺ ˹sila3 gur?˺ ˹0.2.3.5˺ ˹sila3 ninda˺ […] […] 0.1.0 ˹ze2-ra˺-[am] ˹x˺ ˹x˺ […] ˹zah3˺ a-b[u-um-diĝir] ˹zah3˺ ˹ul-lu-ki-ib˺-[ri] 0.1.0 ur-˹niĝar˺ĝar i3-/˹ra2-ra2˺ ˹0.0.x˺ ˹munu4-mu2˺ zah3 ˹engar-zi˺ ninda ˹0.0.x˺ zu-[lu]-˹ti-a˺ --- ša-ru-še-ni […] ḫu-nu-um ˹0.0.x˺ šu-eš18-tar2 […] ḫu-zu-tum 0.0.1 ku-bu-ul-tum dumu-ni-me 0.1.0 zu-zu lu2-tir 0.0.2 lu2-dinana --- pi-ru-ni-du8 --- a-ḫi-ṣi2-im-ti --- geme2-dnin-mug dumu-nita-a-ni 0.0.1.5 (sila3) nin-me-a-du97 dumu-ni-me 0.0.1 ṣi-il-diškur 0.0.2 a-tu ninda 0.0.1.5 (sila3) inim-dšara2 --- 0.4.0.5 sila3 še --- 0.0.3.5 sila3 ninda 0.0.3.5 (sila3) ĝišbanšur-ta zi-ga 0.2.0 dumu e2-a 2.0.0 dabin gur

97 Erasure before nin.

Texts and Workers

vii

vii

viii

viii

viii

viii

viii

2.3.2 še gur 1.0.0 gu2 šim sig5 35. 2.0.0 gu2 šim gen še šim dugutul2-še3 ˹x˺-˹x˺ munus --- [da]bin gur 1.˹x˺.˹x˺.˹x˺ ˹dabin gur˺ ˹1˺.[x.x.x] ˹x˺ ˹x˺ 40. ˹1˺.[x.x.x] ˹x˺ ˹x˺ […] […] […] L.E. […] […] 1. […] ˹x˺ a ˹x˺-ti 0.0.3 še bar-ra ˹x˺ ˹x˺ 0.˹x˺.3 ˹še˺ kaš ˹sa2?-du11?˺/ d˹nin˺-šubur 0.0.1 zid2 0.0.1.eša d ša-u-˹ša˺ 5. 0.0.0.5 sila3 zid2 0.0.0.5 sila3 ˹eša˺ ˹x˺ ˹x˺ ˹x˺ ˹x˺ ˹x˺ ˹x˺ ˹x˺ [x x] ˹x˺ […] 10. 5.3.2 gur 0.0.3 ninda (Blank) šu-niĝin2 14.3.5 /gur ˹šu-niĝin2˺ 2?.0.1.5 sila3 zid2 /gur ˹šu-niĝin2˺ [x.x.x].5 sila3 eša /gur ˹šu-niĝin2˺ [x].˹2˺.3.5 (sila3) ninda gur 15. ˹šu-niĝin2˺ […] gun2 šim sig5 ˹šu-niĝin2˺ 1+[x] gun2 6 ma-na ˹šu-niĝin2˺ [… ši]m gen (Blank) […] ˹x˺ 20. […] ˹x˺ […] ˹x˺ […]

139

140

Lluís Feliu and Adelina Millet Albà

6 Appendix. Corrections to Text B (= AuOr 22: 222–30). ii iii iii v v v v v v v v v v v v v v vi vi vi vi vi vi vi vi vi vi vii vii vii viii

41. 40. 41. 4. 5. 6. 7. 10. 21. 23. 27b. 32. 34 40. 41. 42. 43. 6. 10 30. 31. 40. 41. 42. 43. 44. 45. 24. 27. 28. 19.

0.0.2 (še) 0.0.1? (zu2-lum) a-˹pil˺-[-n]u?-ri 5 (sila3 še) 5 (sila3 zu2-lum) dumu-munus-ni 5 (sila3 še) 5 (sila3 zu2-lum) dumu-mu[nus] min-kam ˹0.0.3?˺ (še) 0.0.2 (zu2-lum) ˹ur˺-d[…] ˹ma2˺ lah5˺-[me] ˹0.0.3˺ (še) 0.0.2 (zu2-lum) zu-lu-˹um˺ ˹0.0.1˺.5 (sila3) ˹a˺-ba du[mu-ni-me] 0.0.2 (še) 0.0.1 (zu2-lum) ma-˹x˺ ninda 0.0.2 dnin-šu[bur-i3-sa6] [lu2-lunga3-me] […] ˹lu2˺-diĝir-˹ra˺ [… er3]-˹re-eb˺ ˹lu2˺-˹ašgab˺-˹me˺ [ur-niĝarĝar lunga3] [zah3 ri-iṣ-diĝir] [i-re-eb ku3-d]im2 ˹x˺ 5 5 ˹ e-pi-ni˺ 0.0.1.5 (sila3) ˹be˺-˹li2-du10˺ ninda 0.0.3 u4-d[e3-niĝ2-sig5] 0.0.2.1 (sila3) ˹a˺-d[a?…] [… ur2]-˹ra˺-ni [dumu-munus]-ni […] [kir4-dab5-me] […] lu2-˹diĝir-ra˺ […] 5 ˹inim-ma-a˺ zah3 ḫu-nu-um 0.0.1 ku-bu-ul-tum ˹dumu˺-ni-me ˹šu-niĝin2˺ 23 […]

Texts and Workers

Fig. 1: Obverse.

141

142

Lluís Feliu and Adelina Millet Albà

Fig. 2: Reverse.

Texts and Workers

143

Bibliography Civil, Miguel. 1987a. Feeding Dumuzi’s Sheep: The Lexicon as a Source of Literary Inspiration. Pp. 37–55 in Language, Literature, and History: Philological and Historical Studies Presented to Erica Reiner, ed. Francesca Rochberg. AOS 67. New Haven: American Oriental Society. Civil, Miguel. 1987b. Ur III Bureaucracy: quantitative aspects. Pp. 34–44 in The Organization of Power. Aspects of Bureaucracy in the Ancient Near East, ed. McGuire Gibson and Robert D. Biggs. Chicago: The Oriental Institute of the University of Chicago. Feliu, Lluís. 2014. A New Edition of NISABA 15/2 668. AuOr 32: 221–33. Focke, Karen. 2015. Der Garten in neusumerischer Zeit. AOAT 53. Münster: Ugarit-Verlag. Garcia-Ventura, Agnès. 2014. Weaving in Ur III Mesopotamia: Womens Work? Pp. 135–40 in Textile Trade and Distribution in Antiquity. Textilhandel und -distribution in der Antike, ed. Kerstin Dross-Krüpe. Wiesbaden: Harrassowitz. Garfinkle, Steven J. 2014. The Economy of Warfare in Southern Iraq at the End of the Third Millennium BC. Pp. 353–62 in Krieg und Frieden im Alten Vorderasien. 52e Rencontre Assyriologique Internationale. International Congress of Assyriology and Near Eastern Archaeology, Münster, 17.−21. Juli 2006, ed. Hans Neumann AOAT 401. Münster: UgaritVerlag. Gelb, Ignace J. 1972. The Arua Institution. RA 66: 1–32. Greco, Angela. 2015. Garden Administration in the Ĝirsu Province during the Neo-Sumerian Period. BPOA 12. Madrid: Consejo Superior de Investigaciones Científicas. Hebenstreit, Laurent. 2014. The Sumerian Spoils of War During Ur III. Pp. 373–80 in Krieg und Frieden im Alten Vorderasien. 52e Rencontre Assyriologique Internationale. International Congress of Assyriology and Near Eastern Archaeology, Münster, 17.− 21. Juli 2006, ed. Hans Neumann AOAT 401. Münster: Ugarit-Verlag. Hilgert, Markus. 2002. Akkadisch in der Ur III-Zeit, Münster: Rhema. Klein, Jacob. 1990. Šulgi and Išme-Dagan: Originality and Dependence in Sumerian Royal Hymnology. Pp. 65–136 in Bar-Ilan Studies in Assyriology dedicated to Pinḥas Artzik, ed. Jacob Klein and Aaron Jacob Skaist. Ramat Gan: Bar-Ilan University Press. Van Koppen, Frans. 2001. Sweeping the court and locking the gate: the palace of Sippirṣērim. Pp. 211–24 in Veenhof Anniversary Volume − Studies Presented to Klaas R. Veenhof on the Occasion of his Sixty-Fifth Birthday, ed. Wilfred H. van Soldt. PIHANS 89. Leiden: Nederlands Instituut voor het Nabije Oosten. Lafont, Bertrand. 2013. State employment of women during the Ur III period. REFEMA. Available at: http://refema.hypotheses.org/976. Lambert, Wilfred G. 2014. The Notebooks of W. G. Lambert. Available at: http://oracc. museum.upenn.edu/contrib/lambert/. De Maaijer, Remco and Jagersma, Bram. 1997. Review of ed. Sjöberg. The Sumerian Dictionary of the University Museum of the University of Pennsylvania. Vol. 1 A Part I. AfO 44/45: 277–88. Monaco, Salvatore. 1986. Parametri e qualificatori nei testi economici della Terza Dinastia di Ur. OrAnt 25: 1–20. Owen, David. 2013. Cuneiform Texts Primarily from Iri-Saĝrig / Āl-Šarrākī and the History of the Ur III Period. Nisaba 15. Bethesda, MD: CDL Press. Pomponio, Francesco. 2012. Un motivo per cui le tavolette amministrative neo-sumeriche sono così numerose. Pp. 637–52 in Leggo! Studies Presented to Frederick Mario Fales

144

Lluís Feliu and Adelina Millet Albà

on the Occasion of His 65 th Birthday, ed. Giovanni B. Lanfranchi et al. Wiesbaden: Harrassowitz. Sanati-Müller, Shirin. 1990. Texte aus dem Sînkāšid-Palast. Dritter Teil. Metalltexte. BagM 21: 131–213. Sigrist, Marcel. 1992. Drehem. Bethesda, MD: CDL Press. Steinkeller, Piotr. 2007. City and Countryside in Third-Millennium Southern Babylonia. Pp. 185–211 in Settlement and Society: Essays Dedicated to Robert McCormick Adams, ed. Elizabeth C. Stone. Los Angeles: Cotsen Institute of Archaeology University of California − The Oriental Institute of the University of Chicago. Steinkeller, Piotr. 2013. Corvée Labor in Ur III Times. Pp. 347–424 in From the 21 st Century B.C. to the 21 st Century A.D.: Proceedings of the International Conference on NeoSumerian Studies Held in Madrid, 22–24 July 2010, ed. Steven J. Garfinkle and Manuel Molina. Winona Lake IN: Eisenbrauns. Steinkeller, Piotr. 1989. Sale Documents of the Ur-III-Period. FAOS 17. Stuttgart: Franz Steiner Verlag. Studevent-Hickman, Benjamin. 2006. The Organization of Manual Labor in Ur III Babylonia, Ann Arbor: UMI. Tenney, Jonathan S. 2011. Life at the Bottom of Babylonian Society: servile laborers at Nippur in the 14 th and 13 th centuries, B.C. CHANE 51. Leiden/Boston: Brill. Trémouille, Marie-Claude. 2009. Šauška, Šawuška. RlA 12: 99–103. Waetzoldt, Hartmut. 1992. “Rohr” und dessen Verwendungsweisen anhand der neusumerischen Texte aus Umma. BSA 6: 125–45. Waetzoldt, Hartmut. 1987. Compensation of Craft Workers and Officials in the Ur III Period. Pp. 117–41 in Labor in the Ancient Near East, ed. Marvin A. Powell. AOS 68. New Haven: American Oriental Society.

Agnès Garcia-Ventura

Emesal studies today: a preliminary assessment Since the end of the 19th century, several proposals have been put forward to define the use and context of Emesal, a variant of the Sumerian language which has been characterized variously as a literary genre, a dialect, or a sociolinguistic register. These proposals all suggest different definitions. However, almost all of them discuss the same point, either to support it or to refute it: that is, whether or not Emesal can be labeled a “women’s tongue” or “women’s language”, and why. So the debate on Emesal, in one way or another, always links language and gender. In this paper I present some of the recent contributions to this debate, concentrating on the ones put forward over the last decade or so in Sumerian grammars and in general presentations of Sumerian language or literature published from 2001 to 2014. In doing so I aim, first, to take stock of the debate on the nature of Emesal, and thus to continue the task embarked upon by Manfred Schretter and Gordon Whittaker. In 1990 Schretter published his Habilitation, the most complete and exhaustive state of the art of Emesal studies so far. Some years later, from the perspective of Linguistic Anthropology, Whittaker (2002) discussed Schretter’s proposals and completed the picture by adding more recent references. The present article is a first attempt to update the state of the art of Emesal studies. Second, I intend to outline certain common features that link Emesal studies and gender studies, laying the foundations for future research on the topic. In my view, certain assumptions regarding sex and gender underlie most of these proposals, though these assumptions are hardly ever stated explicitly. In examining these recent publications I hope to highlight the main trends of this debate inside Emesal studies, trying to identify their points of convergence and difference with regard to the interplay between language and gender. I defend that exploring what has been said about Emesal, and bearing in mind the con-

Acknowledgements: In this paper I present some results of the research project on Emesal studies I have been conducting since October 2014 at the Sapienza, Università degli Studi di Roma (Italy), funded through a Postdoctoral scholarship awarded by the Beatriu de Pinós programme, with the support of the Secretariat for Universities and Research of the Ministry of Economy and Knowledge of the Government of Catalonia. I wish to thank Jordi Vidal for his valuable comments. Obviously all remaining errors and omissions are my own responsibility. https://doi.org/10.1515/9781501503696-009

146

Agnès Garcia-Ventura

tributions of gender studies, can guide us towards a better understanding of the topic. Third, I will conclude with a few words stressing the potential of Emesal studies inside the development of Assyriology in the coming years. I will stress the need for historiographical analyses, in addition to work on the primary sources, in order to enrich this field of study. As a former student of Miguel Civil, it is an honor to be able to present this overview of the current state of Emesal studies. It is a modest contribution to the historiography of Sumerology, a discipline deeply indebted to Civil’s research. I am delighted to offer this piece as a mark of my respect and admiration for him both as a scholar and as a person.

1 The research on Emesal (2001–2014 AD): Sumerian grammars and reference works Between 2001 and 2014, several Sumerian grammars have been published, along with a number of general entries about the Sumerian language and literature in reference works. This constitutes a significant volume of production in comparison with previous years, as noted by Claus Wilcke in one of the opening lectures of the 53rd Rencontre Assyriologique Internationale held in 2007 in Moscow and St. Petersburg (Wilcke 2010: 5, note 5 for a survey). On the one hand, since almost all scholars who have tried to summarize the current state of research into the Sumerian language, grammar, and literature mention Emesal, an examination of their work is a good starting-point for my preliminary assessment of the current state of Emesal studies. On the other hand, as all these grammars and summaries have been written by scholars from different academic traditions, it is especially interesting to consider which specific aspects each one chooses to highlight.1

1.1 Emesal in Sumerian grammars Two Sumerian grammars published in the period under consideration here briefly discuss Emesal as a variant of Sumerian that links a specific use of the

1 For a presentation of these different traditions in the study of the Sumerian language since Poebel (1923) to current grammars, mainly represented by the German and by the US school respectively, see Feliu 2012. In addition, Wilcke alludes indirectly to these two main “schools”

Emesal studies today: a preliminary assessment

147

language with gender. These are the works by Dietz Otto Edzard (2003) and by Abraham Jagersma (2010). Edzard devotes a short chapter of his grammar to Emesal (Edzard 2003: 171–172 = chapter 16), opening it with a definition of what Emesal is and a note on main users of what he characterizes as: “a sociolinguistic variety of Sumerian attested for the speech of women or goddesses and of the «cantor» (gala) […] Note: The fact that the «cantor» used a form of speech otherwise attested for female persons has caused much speculation: Was the gala a eunuch or an effeminate male? See the discussion in Schretter 1990, 124–36. It is, however, well known that the modern counter-tenor or even altus is sung by sexually normal male persons.” (Edzard 2003: 171)

Some years later, Jagersma (2010) read his Dissertation titled A Descriptive Grammar of Sumerian at the Universiteit Leiden. He devoted some paragraphs to Emesal in the introduction to his work (Jagersma 2010: 8–9), in a discussion of Sumerian dialects: “This is the Sumerian name (eme-sal «thin tongue») of a dialect documented in certain literary and cultic texts dating from the Old Babylonian period or later. Some texts are fully in Emesal, among them the texts used by lamentation priests. Other texts display a shift between Emesal and standard Sumerian sections. In such texts, it is primarily the direct speech of goddesses that is in Emesal. […] What we call a text in Emesal is in practice a standard Sumerian text containing a larger or smaller number of Emesal morphemes.” (Jagersma 2010: 8)

The opening lecture by Claus Wilcke at the 53rd Rencontre was also published in 2010. Though not formally a grammar as such, but a paper published in a proceedings volume, it is a detailed survey entitled Sumerian: What We Know and What We Want to Know. In this comprehensive summary of about 70 pages, Wilcke examines Emesal in the section devoted to “Dialects, phonology, orthography” (Wilcke 2010: 8–11). He agrees with Edzard’s characterization of Emesal as a “sociolinguistic variety” and proposes that several arguments and sources “suggest a social, gender related context in which speaking eme-sal was obligate for women and the gala. […] I assume it to have originated as a gender specific variety, its cultic use being a secondary development” (Wilcke 2010: 9–10).2

in his paper in his defense of several arguments, and especially in his presentation as “a philologist, not a linguist […] A pupil of Falkenstein’s” (Wilcke 2010: 5). 2 For similar arguments, see Krispijn 2005: 155–56.

148

Agnès Garcia-Ventura

1.2 Emesal in general surveys of Sumerian language and literature The entries or articles devoted to the Sumerian language and/or literature in reference works have multiplied in recent years.3 A significant number of handbooks, companions and encyclopedias have been published as scholars and publishing houses have attempted to provide authoritative syntheses on specialized subjects. Here I will refer to the contributions of six scholars:4 Bertrand Lafont (2001), Piotr Michalowski (2004), Gonzalo Rubio (2005a, 2005b, 2007 and 2009),5 Gábor Zólyomi (2006), Jeremy Black (2007) and Graham Cunningham (2013).6 Lafont is the author of the entry “Emesal” in the Dictionnaire de la civilisation mesopotamiénne, edited by Francis Joannès in 2001. The first point to note is the existence of this entry as such, and not as a section of entries like “Sumerian”, “Sumerian language” or “Sumerian literature”, for instance, as in the

3 Civil, the scholar honored in this volume, also wrote some general presentations of the Sumerian language in several contexts. Here I highlight his entry “Sumerian” for the twovolume History of Linguistics, edited by Giulio Lepschy, published in 1994 in English (the original Italian version was published in 1990, see final bibliography). In that publication, Civil mentioned Emesal and described it as follows: “The term for ‘Sumerian’ is eme-gir7 (‘tongue’ + an adjective of uncertain meaning possibly related to the name of the country, ke-en-gi) and the women’s dialect is known as eme-sal (‘fine or thin tongue’)”; (Civil 1994: 76) // “The women’s dialect is used by women speakers, regardless of the sex of the hearer, and in one of the genres of religious poetry. The differences between female and male speech consist in changes in the articulation of certain sounds and in lexical alternations.” (Civil 1994: 85, note 1). Besides general presentations like this one, Civil continued to work with research issues related with Emesal, among others, as shown by his edition of “The Song of the Millstone”, an Emesal proverb (Civil 2006). 4 During this period, the volume containing the entry “Sumer, Sumerisch” (Cooper 2013) was published in one of the main reference works of ancient Near Eastern studies, the Reallexikon der Assyriologie und Vorderasiatischen Archäologie. However, as Emesal is not discussed in this entry, I do not include it in the list of general entries and publications. 5 I list their proposals according to their year of publication. However, in the case of scholars with more than one general entry published in this period, as is the case of Gonzalo Rubio, I will discuss all the proposals together taking the year of publication of the first one as my reference point. Moreover, for Rubio and for Michalowski, I will refer not just to their general entries, but also to a couple of their papers that are relevant to the debate on Emesal that I deal with here (Michalowski 2006 and Rubio 2001). 6 Cunningham is also the author of another general presentation of Sumerian, more specifically the entry “Sumerian” in the multi-volume Encyclopedia of Language and Linguistics, published in 2006. However, the author does not mention Emesal in this entry, and so here I will only discuss his 2013 paper.

Emesal studies today: a preliminary assessment

149

case of the other ones listed below. Second, Lafont (2001: 281) characterized Emesal as “une sorte de dialecte du sumérien”. As regards the interplay between language and gender, he presented the debate on the interpretation of the term Emesal as follows: “Emesal signifié probablement « langue raffinée », sal étant l’équivalent de l’akkadien raqqu, « fin », alors que eme-gir15 signifie «langue noble». Il n’existe cependant pas de consensus sur la question de savoir de que représente réellement l’emesal par rapport au sumérien courant: s’agit-il d’un dialecte lié a des particularismes locaux, d’un dialecte propre au monde des femmes et des efféminés, ou encore d’un dialecte à caractère poétique adapté au récitatif des mélopées et hymnes sacrés? On en discute encore.” (Lafont 2001: 281)

In his survey of “Sumerian” in The Cambridge Encyclopedia of the World’s Ancient Languages (2004) Michalowski also devoted some lines to a discussion of the meaning of the term Emesal and the valences of sal. He claimed that the meaning “thin” instead of “woman” gives us the clue to understanding the nature of the term. Consequently, he held that it refers to pronunciation, rather than to an exclusive use of this variety of language by women. He summarized the matter as follows, while rejecting certain proposals which had been made during the debate on Emesal studies: “The main dialect distinction in Sumerian, as reflected in native terminology, is between eme-gir17 (EG) and eme-sal (ES). The former seems to be the native term for what we could call Standard Literary Sumerian. The latter is restricted to ritual texts, primarily those used by lamentation priests (gala) − and to the direct speech of certain goddesses and their messengers in literary texts, although the same goddesses speak fluent ‘Standard Sumerian’ in other compositions. On the basis of false etymology, and misunderstandings of the distribution of Emesal, it has been often called a ‘women’s tongue’, leading some to invoke unnecessary ethnographic analogies. Likewise, it has been claimed that the gala priest and the divine messengers were eunuchs (e.g. Boisson 1992: 434), although there is no evidence for castration, human or divine, in ancient Sumer.” (Michalowski 2004: 23)

On the other hand, in a paper on gala published in 2006, Michalowski added some remarks to this general description.7 First, he noted that “dialect” is an imprecise term for defining Emesal, and proposed “mode of elocution” (Michalowski 2006: 49; cf. Wilcke 2010: 8–9 for opposition to this point of view).

7 In the volume of the Journal of Cuneiform Studies in which this paper was published (volume 58, year 2006), a paper by Jerrold S. Cooper was also published on the same topic. Michalowski refers to this paper to support some of his proposals and to add nuances, and even to disagree with his own 2004 publication mentioned above.

150

Agnès Garcia-Ventura

Second, at the center of the characterization of Emesal he places its close interplay with “life-changing events”: “[…] we can now appreciate that its origins, as well as its continuing function, lie not in a regional dialect or a social register of a living language but in a complex nexus of sociolinguistic facts linked to a constellation of cultic practices that involve basic life-changing events, namely, love and death.” (Michalowski 2006: 49)

The next scholar in our review, Gonzalo Rubio, is the most prolific author of the period in so far as general presentations of the Sumerian language and literature are concerned. In only five years, four of his survey contributions were published in the following compendia and reference works: A Companion to the Ancient Near East (2005a), Encyclopedia of Linguistics (2005b), Morphologies of Asia and Africa (2007) and From an Antique Land: An Introduction to Ancient Near Eastern Literature (2009).8 Like most of the scholars mentioned so far, Rubio discusses the contexts in which Emesal is attested, the possible meanings and readings of the word Emesal itself and the term gala, since, as we have already noted, it seems that these gala, besides women, were the main users of Emesal. Taking several arguments into account, Rubio wonders whether Emesal might be characterized as a “Sumerian Genderlect or Genrelect” (Rubio 2005b: 1050), concluding that “[…] emesal may have originally stemmed from an actual regional dialect or from the particular dialect of a certain group (such as a genderlect). However, as we have it attested, in most cases, the occurrence of emesal forms may be determined mostly by the genre of the text, rather than by the gender of the fictional speaker or even the performer.” (Rubio 2009: 32)

Gábor Zólyomi, one of the regular contributors to the Electronic Text Corpus of Sumerian Literature,9 is the author of the entry “Sumerisch” in the volume Sprachen des Alten Orients, edited by Michael P. Streck (2006). In this publication, Zólyomi characterizes Emesal as a literary variant and as a women’s dialect: “Eine Besonderheit des Sumerischen ist die ausschließlich literarische Varietät mit Namen eme-sal, was wahrscheinlich ‚feine Sprache‘ bedeutet. […] Wahrscheinlich hat sie

8 Rubio discusses Emesal in the following pages in these publications: 2005a: 81–82; 2005b: 1050; 2007: 1369–1370; 2009: 31–32). Cf. Rubio 2001: 269–71, an article-review in which the author presents a state of the art of Emesal studies summarizing the main references discussed by Yitschak Sefati (1998). 9 http://etcsl.orinst.ox.ac.uk/ (accessed April 2015).

Emesal studies today: a preliminary assessment

151

ihren Ursprung in einer Form des gesprochenen Sumerischen. Eine ihrer Gebrauchsweisen in der gesprochenen Sprache war ziemlich sicher die eines Frauendialektes; doch für welche historische Periode und welche Regionen dies gilt, läßt sich nicht mehr feststellen, weil zu Beginn des 2. Jahrtausends (als diese Varietät erstmals durch Texte bezeugt ist) dieser Gebrauch durch eine Spezialisierung auf bestimmte religiöse und poetische Textgenres und Kontexte abgelöst wird (einschließlich – jedoch ohne alleinige Beschränkung auf – der literarischen Wiedergabe von Frauenrede).” (Zólyomi 2006: 14)

Just one year later, 2007, the proceedings of a workshop held in 2003 at the British School of Archaeology in Iraq under the title Languages of Iraq, Ancient and Modern were edited and published by J. Nicholas Postgate. In his piece on the Sumerian language, Jeremy Black describes Emesal as follows: “The principal linguistic variation is the ‘women’s language’ Emesal. […] Emesal was used in literature to characterise the speech of women, and in the religious cult songs of the (male) gala singers.” (Black 2007: 23–24)

Finally we should mention Graham Cunningham’s article titled “The Sumerian Language” in the volume The Sumerian World edited by Harriet Crawford in 2013. Cunningham also discusses the possible readings of sal, and the meaning of the Sumerian term eme-sal, characterized as a “Sumerian genre”: “Emesal and Emegir differ occasionally in their lexicon and morphology but more often in the distribution of their consonants. In addition to the cultic laments, Emesal is particularly attested in the speech of the female protagonist in songs of love and marriage. These genre associations, and the fact that the sign used to write sal depicts the pubic triangle (its other values include munus ‘woman’ and gal4 ‘vulva’), have encouraged many scholars to suggest a correlation between Emesal and female speech. However, establishing such a correlation, or indeed one between Emesal and any spoken variety of the language, remains difficult.” (Cunningham 2013: 100–101)

2 The interplay between language and gender in the characterization of Emesal Most of the summaries mentioned so far discuss coincident issues, even though each of them provides nuanced or even discordant points of view. In this section I concentrate on three of these issues and their variants. First, all of them must face the elusive question of what Emesal actually is: a mode of elocution; a genre; a register or variant (linguistic or literary variant); or a dialect associated with a specific geographical area or human group (with the possibilities

152

Agnès Garcia-Ventura

of defining it as “sociolect” or as “genderlect”), or with a specific literary genre (“genrelect”).10 Second, Emesal is often presented in opposition to Emegir or as a complement to it. Consequently, the way this relationship is described affects the way Emesal itself is characterized. In some cases, its relationship is defined as hierarchical (Emegir as main or standard Sumerian versus Emesal as a variant of the standard and therefore secondary). In others, Emegir and Emesal are both presented as dialects or as variants of Sumerian, and little attention is paid to their potential hierarchical relationship. Third, all the overviews discuss the interplay between language and gender through a series of themes, all of them linked to the main research question of identifying the users of Emesal. Here I highlight three of these themes: the relationship between Emegir and Emesal mentioned above; the role and identity of the gala; and the possible readings and interpretations of the term emesal itself, and more particularly of the sign sal that forms the second part of the word. Let us begin with the relationship between Emegir and Emesal. When they are presented as a dichotomy and, consequently when they are defined through their opposition, Emegir is often presented as the variant, register or dialect used by men and potentially by all other possible speakers but women. Emesal in turn is presented as associated only with women. Here again two possible interpretations of Emesal arise: either as an option for female speakers, or as mandatory and/or restricted to female speakers. In any case, despite the nuances, the binary contrast of Emegir-masculine-general versus Emesalfeminine-particular reproduces a pattern which has been widely discussed and contested from the sphere of gender studies, more specifically from feminist epistemologies.11 For this reason, the analysis of this binary contrast seems to me a good avenue for further research applying gender studies perspectives as well as insights from the branch of linguistics that deals with the relationship between language and gender.12 10 To the several options presented in the previous section (§ 1) it is also worth adding the ones expressed by two of the main scholars of the study of Emesal, whom we mention in the last section (§ 3): Uri Gabbay and Anne Löhnert. Gabbay (2014: 1) presents Emesal as a “register of Sumerian”, while Löhnert names it as a “sociolect” (Löhnert 2008: 423) and as a “phonologische Sprachvarietät” (Löhnert 2014: 210). 11 Donna Haraway, Sandra Harding and Nancy Hartsock are some of the main scholars engaged in feminist epistemologies. For an overview of the main contributions from this trend, see Harding 2004. For a classical reference on the binary contrast men-general versus womenparticular, its nuances and its potential effects at different levels, see Hartsock 1983. 12 For a brief presentation of this field of study and an overview of its main contributions, see Bucholtz and Hall 2006.

Emesal studies today: a preliminary assessment

153

Moving on now to the second widely discussed issue regarding the interplay between language and gender, it is worth mentioning the case of the gala, who are usually presented as male lamentation singers or male priests. Those who see the use of Emesal as restricted to women, or mandatory for them, are obliged to discuss the sexual identity of these gala; they must explain why these men used a language that was allegedly for women. Some scholars suggest that the gala might have been effeminate males or eunuchs13 (two labels which are widely contested by gender studies, and deserve further discussion).14 In this regard, the primary sources provide us with little or no information on the gala’s sexual preferences and physical conditions. On the other hand, other scholars, as we have also seen, contend that the gala used Emesal because of the cultic context in which they worked or because of the literary genre linked to them, regardless of their sexual preferences or physical nature. Finally, as regards readings and interpretations of the term eme-sal, the two main options are “fine” or “thin tongue” on the one hand and “women’s tongue” or “women’s language” on the other. Currently there is a broad consensus favoring the former option. However, despite this consensus and even though the reading of sal as “woman” as the clue to interpret the term emesal has been contested and dismissed since the beginning of the 20th century,15 we have seen that allusions to Emesal as “women’s tongue”, “women’s language” or “Frauensprache” are not infrequent, at least to explain a certain stage of its genesis and development. Again, the insistence on this translation is something to be contextualized and analyzed critically from the viewpoint of gender studies – not just from a philological perspective, as has been the case so far.

3 To conclude: gender, historiography, and the vitality of Emesal studies Through the excerpts and references compiled in the first section (§ 1) and the summary offered in the previous section (§ 2), I have tried to highlight one of 13 For opposing views on this issue, see Michalowski 2004 and 2006, and Gabbay 2014: 67– 68, with previous references. Contra gala as eunuchs, see Michalowski 2004: 23 (§ 1). For gala as eunuchs, see Gabbay: “My own opinion is that at least up to the Sargonic Period, the gala may have been a hermaphrodite or eunuch” (Gabbay 2014: 67). 14 On the value of feminist research for assessing the use of certain labels and categories of analysis and their potential androcentrism, see Spencer-Wood 2006: 60–61. 15 See for example Delitzsch 1914: 21, footnote 1: “Eine Deutung möchte ich weder für emesal noch für eme-te-ná schon wagen. Nur das steht fest, dass eme-sal unmöglich ‘Weiber-’

154

Agnès Garcia-Ventura

the main differences between the various conceptions of Emesal: that is, that while some of them consider that it was the gender of the speakers that dictated the use of this variant of the language, others consider gender as one of a number of factors to take into account, rather than the only one or the central one. For those who consider it as a crucial element, the research on Emesal aims to explain why and how women (and all those considered from this perspective as not conforming to a certain hegemonic ideal of masculinity, like the gala) use a variant of the language distinct from the one presented as standard (Emegir), characterized in turn as the one used by men. However, those who consider that the use of Emesal by women (and by gala) is just one of several factors to take into account, the genre of the literary text or even the context of use should also be considered in order to explain why we find Emesal in some texts and not in others. The latter scholars point out that we also find women speaking Emegir in the texts and that there is no clear evidence for questioning the potential hegemonic masculinity of the gala. The debate on the centrality and uniqueness of one factor of analysis to explain a certain phenomenon is not peculiar to Emesal studies, but is one of the concerns of the theoretical approaches that come under the heading of gender studies. Indeed, decentering (together with deconstruction) was one of the strategies applied by the feminist epistemologies during the 1980s and the 1990s, clearly influenced by proposals from Postmodernism (Mann and Huffman 2005). In doing so, the aim was to determine whether or not decentering certain factors of analysis (previously considered as central) might modify the results of the research. A classic example was the questioning of the centrality of the sex- or gender-based division to explain the organization of work. To examine this idea, one of the alternatives put forward was the idea of intersectionality, a proposal which decentered sex and gender as the key factors influencing the organization of work and took into consideration other factors like age or rank. In fact, it has been shown that applying intersectionality instead of the sex- or gender-based division of work might change the results of our research, or at least might help us to confer on them more complexity.16 I wonder if it is the application of a similar strategy of decentering to Emesal studies (explicitly or implicitly, consciously or unconsciously) that has helped to bring out some of the nuances and the variety shown in the studies and syntheses outlined here. To answer this question, I think we now need

oder “Frauensprache” übersetzt werden darf, weil sumer. sal überhaupt nicht ‘Weib’ bedeutet (siehe Glossar). Das ‘Weib’ heisst gême oder ki-el.” 16 Garcia-Ventura: forthcoming (with previous references).

Emesal studies today: a preliminary assessment

155

historiographical studies which can allow us to contextualize the proposals published so far.17 In other words, I defend the need for work on secondary sources, as historiographical analyses do, in addition to work on the primary sources in order to enrich this field of study. Regarding the work on primary sources it is worth noting that during the period here under consideration (2001–2014) not only the syntheses on Sumerian language and literature multiplied dramatically, but also Emesal studies themselves as demonstrated by recent PhD research and the launch of the new series “Heidelberger Emesal-Studien”. Among recent and current PhD research, the most significant contributions have been made by Uri Gabbay, Anne Löhnert and Jana Matuszak. These scholars are likely to be in the vanguard of Emesal studies in the coming years.18 Jana Matuszak, PhD candidate at the Universität Tübingen, is working on the Sumerian composition known as “A dialogue between two women B”. Anne Löhnert (Ludwig-Maximilians-Universität München) and Uri Gabbay (Hebrew University of Jerusalem) read their dissertations in 2006 and 2007 respectively. Both studies have since been published as monographs, in Löhnert’s case in the AOAT19 series (Löhnert 2009) and in Gabbay’s case as the first volume of the series Heidelberger Emesal-Studien (Gabbay 2014).20 As I noted in the above paragraph, the launch of this new series devoted specifically to Emesal studies, edited by Stefan M. Maul, is excellent news for this area of Assyriology. At this point, then, it seems that Emesal studies are in full swing as several scholars and now a specialized series, are devoting their efforts to the edition and study of texts written in Emesal from the second and first millennia BCE. One hopes very much that this research will discuss the nature of Emesal itself, since we have seen that this is still one of the basic issues to be clarified. If this happens, I think that one of the crucial topics to be discussed is the centrality of the relationship between the use of Emesal and women. In this

17 On the purpose and usefulness of historiography in Assyriology, see Vidal 2015, especially pp. 26–27, with previous references. 18 All were invited speakers at the “Emesal-Kolloquium” held at the Universität Innsbruck, March 27th 2014, to celebrate Manfred Schretter’s 70th birthday. See the full program at: http:// www.uibk.ac.at/alte-geschichte-orient/links/tagungen/emesal-kolloquium.pdf (accessed April 2015). I wish to thank Uri Gabbay and Anne Löhnert for providing me with, or helping me to find, copies of some of their recent publications. 19 Alter Orient und Altes Testament, published by Ugarit-Verlag. 20 The research of both Löhnert (2008, 2009, 2014) and Gabbay (2011, 2013 and 2014), focuses on the philological as well as the ritual and cultic aspects of the texts written in Emesal. Both also deal with the performers of these texts, discussing, among other things, the role of the abovementioned gala.

156

Agnès Garcia-Ventura

respect, as I have suggested here, the contributions of gender studies may turn out to be particularly fruitful.

Bibliography Black, Jeremy A. 2007. Sumerian. Pp. 4–30 in Languages of Iraq, Ancient and Modern, ed. J. Nicholas Postgate. Cambridge: British School of Archaeology in Iraq. Bucholtz, Mary and Kira Hall. 2006. Gender, Sexuality and Language. Pp. 756–758 in Encyclopedia of Language and Linguistics, vol. 4, ed. Keith Brown. Oxford: Elsevier. Civil, Miguel. 1990. Sumerico. Pp. 96–105 in Storia della linguistica, vol. 1, ed. Giulio Lepschy. Bologna: il Mulino. Civil, Miguel. 1994. Sumerian. Pp. 76–87 in History of Linguistics, vol. 1, ed. Giulio Lepschy. London−New York: Longman. Civil, Miguel. 2006. The Song of the Millstone. Pp. 121–38 in Šapal tibnim mû illakū. Studies Presented to Joaquín Sanmartín on the Occasion of His 65 th Birthday, ed. Gregorio Del Olmo Lete, Lluís Feliu, and Adelina Millet Albà. Aula Orientalis − Supplementa 22. Sabadell: Ausa. Cooper, Jerrold S. 2006. Genre, Gender, and the Sumerian Lamentation. JCS 58: 39–47. Cooper, Jerrold S. 2013. Sumer, Sumerisch. RlA 13: 290–97. Cunningham, Graham. 2006. Sumerian. Pp. 271–74 in Encyclopedia of Language and Linguistics, vol. 12, ed. Keith Brown. Oxford: Elsevier. Cunningham, Graham. 2013. The Sumerian Language. Pp. 95–110 in The Sumerian World, ed. Harriet Crawford. London−New York: Routledge. Delitzsch, Friedrich. 1914. Grundzüge der Sumerischen Grammatik. Leipzig: J. C. Hinrichs’sche Buchhandlung. Edzard, Dietz Otto. 2003. Sumerian Grammar. Handbuch der Orientalistik 71. Leiden−Boston: Brill. Feliu, Lluís. 2012. El estudio del sumerio en el siglo XXI. Entre la filología y la lingüística. Pp. 102–10 in Séptimo Centenario de los Estudios Orientales en Salamanca, ed. Ana Agud, Alberto Cantera et al. Estudios Filológicos 337. Salamanca: Ediciones Universidad de Salamanca. Gabbay, Uri. 2011. Lamentful Proverbs or Proverbial Laments? Intertextual Connections between Sumerian Proverbs and Emesal Laments. JCS 63: 51–64. Gabbay, Uri. 2013. The Performance of Emesal Prayers within the Regular Temple Cult: Content and Ritual Setting. Pp. 103–21 in Temple im Alten Orient. 7. Internationales Colloquium der Deutschen Orient-Gesellschaft 11.−13. Oktober 2009, München, ed. Kai Kaniuth. Colloquien der Deutschen Orient-Gesellschaft 7. Wiesbaden: Harrassowitz Verlag. Gabbay, Uri. 2014. Pacifying the Hearts of the Gods. Sumerian Emesal Prayers of the First Millennium BC. Heidelberger Emesal-Studien 1. Heidelberg: Harrassowitz Verlag. Garcia-Ventura, Agnès. Forthcoming. The sex-based division of work versus intersectionality: some strategies for engendering the Ur III textile work force. The Role of Women in Work and Society. Proceedings of the final Conference of the REFEMA project (Nanterre, 5–7 November 2014), ed. Brigitte Lion and Cécile Michel. Studies in Ancient Near Eastern Records 13. Berlin−New York: de Gruyter.

Emesal studies today: a preliminary assessment

157

Harding, Sandra. 2004. The Feminist Standpoint Theory Reader. Intellectual and Political Controversies. London−New York: Routledge. Hartsock, Nancy. 1983. The Feminist Standpoint: Developing the Ground for a Specifically Feminist Historical Materialism. Pp. 283–310 in Discovering Reality: Feminist Perspectives on Epistemology, Metaphysics, Methodology, and Philosophy of Science, ed. Sandra Harding and Merrill B. Hintikka. Boston: Ridel. Jagersma, Abraham. 2010. A Descriptive Grammar of Sumerian. Leiden: PhD, Leiden Universiteit. Krispijn, Theo J. H. 2005. The Change of Official Sumerian in the City-State of Lagaš. Pp. 153– 75 in The Study of Diachronic and Synchronic Variation in Sumerian. Special Volume in Honor of Professor Mamoru Yoshikawa, 1. (= ASJ 22), ed. Jeremy Black and Gábor Zólyomi. Hiroshima: Middle Eastern Culture Center in Japan. Lafont, Bertrand. 2001. Emesal. Pp. 281–82 in Dictionnaire de la civilisation mésopotamienne, ed. Francis Joannès. Paris: Éditions Robert Lafont. Löhnert, Anne. 2008. Scribes and Singers of Emesal Lamentations in Ancient Mesopotamia in the Second Millennium BCE. Pp. 421–47 in Papers on Ancient Literatures: Greece, Rome and the Near East. Proceedings of the «Advanced Seminar on the Humanities», Venice International University 2004–2005, ed. Ettore Cingano and Lucio Milano. Padova: SARGON editrice. Löhnert, Anne. 2009. «Wie die Sonne tritt heraus!» Eine Klage zum Auszug Enlils mit einer Untersuching zu Komposition und Tradition sumerischer Klagelieder in altbabylonischer Zeit. AOAT 365. Münster: Ugarit-Verlag. Löhnert, Anne. 2014. Was reden die da? Sumerisch und Emesal zwischen Alltag und Sakralität. WO 44: 190–212. Mann, Susan Archer and Douglas J. Huffman. 2005. The Decentering of Second Wave Feminism and the Rise of the Third Wave. Science & Society 69/1: 56–91. Michalowski, Piotr. 2004. Sumerian. Pp. 19–59 in The Cambridge Encyclopedia of the World’s Ancient Languages, ed. Roger D. Woodard. Cambridge: Cambridge University Press. Michalowski, Piotr. 2006. Love or Death? Observations on the Role of the Gala in Ur III Ceremonial Life. JCS 58: 49–61. Rubio, Gonzalo. 2001. Inanna and Dumuzi: a Sumerian Love Story. JAOS 121/2: 268–74. Rubio, Gonzalo. 2005a. The Languages of the Ancient Near East. Pp. 79–94 in A Companion to the Ancient Near East, ed. Daniel C. Snell. London: Blackwell. Rubio, Gonzalo. 2005b. Sumerian. Pp. 1045–51 in Encyclopedia of Linguistics, ed. Philipp Strazny. New York−Oxon: Fitzroy Dearborn, Taylor & Francis Group. Rubio, Gonzalo. 2007. Sumerian Morphologie. Pp. 1327–79 in Morphologies of Asia and Africa, ed. Alan S. Kaye. Winona Lake, Indiana: Eisenbrauns. Rubio, Gonzalo. 2009. Sumerian Literature. Pp. 11–75 in From an Antique Land. An Introduction to Ancient Near Eastern Literature, ed. Carl S. Ehrlich. Lanham, MD: Rowman & Littlefield. Schretter, Manfred K. 1990. Emesal-Studien. Sprach- und Literaturgeschichtliche Untersuchungen zur sogenannten Frauensprache des Sumerischen. Innsbrucker Beiträge zur Kulturwissenschaft − Sonderheft 69. Innsbruck: Verlag des Instituts für Sprachwissenschaft der Universität Innsbruck. Spencer-Wood, Suzanne M. 2006. Feminist Theory and Gender Research in Historical Archaeology. Pp. 59–104 in Handbook of Gender in Archaeology, ed. Sarah Milledge Nelson. Lanham MD: AltaMira Press.

158

Agnès Garcia-Ventura

Vidal, Jordi. 2015. Reflexiones historiográficas sobre el Orientalismo Antiguo. Pp. 25–36 in Descubriendo el Antiguo Oriente. Pioneros y arqueólogos de Mesopotamia y Egipto a finales del s. XIX y principios del s. XX, ed. Rocío Da Riva and Jordi Vidal. Barcelona: Bellaterra arqueología. Whittaker, Gordon. 2002. Linguistic Anthropology and the Study of Emesal as (a) Women’s Language. Pp. 633–44 in Sex and Gender in the Ancient Near East. Proceedings to RAI 47, ed. Simo Parpola and Robert M. Whiting. Helsinki: The Neo-Assyrian Text Corpus Project. Wilcke, Claus. 2010. Sumerian: What We Know and What We Want to Know. Pp. 5–76 in Language in the Ancient Near East. Proceedings of the 53e Rencontre Assyriologique Internationale, ed. Leonid Kogan, Natalia V. Koslova et al. Babel und Bibel 4/1.Winona Lake, Indiana: Eisenbrauns. Zólyomi, Gábor. 2006. Sumerisch. Pp. 11–43 in Sprachen des Alten Orients, ed. Michael P. Streck. Darmstadt: Wissenschaftliche Buchgesellschaft.

Fumi Karahashi

Royal Nurses and Midwives in Presargonic Lagaš Texts amb tot l’afecte

Miguel Civil (2011) published, from the Schøhen Collection, a clay cylinder that is dated to the Ur III period based on the script. The cylinder is inscribed with the laws of Ur-Namma and contains new paragraphs not known until then, including some concerning women. The timing of the publication could not have been better for my research because it coincided with the period in which I started studying the Presargonic Lagaš material with a focus on women. Among the new woman-related laws of Ur-Namma, of special interest for me has been the one that stipulates the wages of the hired wet-nurse with a midwife function (see 1.2 below). This prompted me to review the nurses and midwives who appear in Presargonic Lagaš texts, with the aim of examining their socio-economic status. In writing this paper I would like to celebrate with all his family, friends, and colleagues the day when Miguel was born ninety years ago, destined to become one of the greatest Sumerologists.

1 Functions of um-me, um-me-da, and nu-gig The Presargonic Lagaš E2-mi2 corpus testifies to a variety of female professionals and menials (Asher-Greve 1985: 161–63; Prentice 2010: 31–64).1 Among them was a group of female personnel called ar3-tu munus who worked for the ruling families of Lagaš.2 The job category or title of these personnel was not specified in the documents, with the exception of um-me, um-me-da, and nu-gig.

1 On Presargonic texts and chronology, most recently Sallaberger and Schrakamp 2015: 67– 74. 2 The term ar3-tu munus has been interpreted and translated in different ways: e.g., de Genouillac 1909: xxviii; Deimel 1931: 110; Bauer 1967: 87; Selz 1993: 204, 209–11 ad 7: 3; Selz 1995: Acknowledgements: I thank Frederick W. Knobloch for helpful discussion and comments. All flaws and errors are of course mine alone. Abbreviations follow CDLI. Others are as follows: E = Enentarzi; L = Lugalanda; UL = Urukagina Lugal; the number following the letter(s) is the regnal year. https://doi.org/10.1515/9781501503696-010

160

Fumi Karahashi

1.1 um-me “Wet-Nurse” and um-me-da “Nursemaid” Deimel (1929: 126–27) interpreted the um-me as “Amme” or “Geburtshelferin” and compared it with um-me-ga-lal = mušēniqtu “wet-nurse.” As for the umme-da, he translated “Kindermädchen,” an assistant of the um-me, and suggested that the -da might be identified with the commutative postposition. Steinkeller (1981: 88–90) has done a detailed study on these two words and concluded that um-me was “wet-nurse” and um-me-da was equivalent to the Akkadian tārītu “nursemaid.” Steinkeller also paid attention to the confusion of um-me and um-me-da.3 With regard to a possible derivative relationship between these two words, he was much more cautious than Deimel. Linguistic analysis apart, these were surely “two different, but related, professions” as Civil (2011: 283) has pointed out.

1.2 nu-gig “Midwife” The term nu-gig (Akkadian qadištu) has been translated “hierodule,” “cult/ sacred prostitute,” “harlot,” and the like in spite of the fact that these women were often connected with childbirth and midwifery and that there is no evidence that they were prostitutes (CAD Q 1982: 50; Gruber 1986: 146; Goodnick Westenholz 1989: 260; Assante 1998; 2003: 32).4 The connection with childbirth and nu-gig is now confirmed by § E2 of the Ur-Namma Laws published by Miguel Civil, which concerns women whose occupation was to take care of the children of other people. Civil’s (2011: 251) composite text and translation read as follows: tukum-bi dumu lu2-ra lu2 ga i3-ni-gu7 mu 3-a še-bi 6 gur sik2-ni 30 ma-na i3-ni 30 sila3 nig2 nam-nu-gig-kam um-[me]-da hun-ga2 mu a2-ni 1 gin2-am3. If someone nurses a man’s child, her barley will be six gur, her wool thirty mina, and her oil thirty sila, for three years. It is part of the nugig-functions. The yearly fee of a hired wet-nurse will be one shekel.

Civil (2011: 284) paraphrases the clause “[i]t is part of the nugig-functions” (nig2 nam-nu-gig-kam) as “including the nugig function” and adds comments 75; Beld 2002: 130; Prentice 2010: 31; Asher-Greve 2013: 367; I will discuss the women so designated elsewhere. 3 It reminds us of the somewhat blurred usage of the word nutrix in epitaphs from the city of Rome: this word may mean wet-nurse or nursemaid, although wet-nurse was the primarily meaning of nutrix and assa nutrix could be used to designate a dry-nurse (Bradley 1986: 202). 4 For the process by which these women came to be viewed as prostitutes, see Assante 2003.

Royal Nurses and Midwives in Presargonic Lagaš Texts

161

that the wet-nurse is hired not only for lactation (i.e., as a wet-nurse)5 but also for the preparations for childbirth and assistance in delivery (i.e., as a midwife).6 In giving the unequivocal translation “midwife” to the word nu-gig, Civil (2011: 281 n. 134) explains that “[t]he attribution of the title nugig to Inanna is part of her aggregation of the duties and rights of all female deities” and rejects the “traditional, but unwarranted, translation” of nugig that implies cultic activities or “sacred prostitution.”7 From the context (ga-gu7) it is clear that in the Ur-Namma Laws the word um-me-da is meant to refer to a wet-nurse rather than a nursemaid. This might have been a case of confusion of these two words (see 1.1 above).

2 Nurses and Midwives in the Presargonic Lagaš In the Presargonic Lagaš corpus the majority of um-me wet-nurses and umme-da nursemaids are attested among the female personnel caring for Lugalanda’s daughters (Asher-Greve 1985: 152). The um-me wet-nurse Al-tuš and um-me-da nursemaid Geme2-id2-edin-na cared for Geme2-dNanše, the elder daughter, and the um-me wet-nurse Nin-guru7 and um-me-da nursemaid Gan-ezem cared for Munus-sa6-ga, the younger daughter (Table 1). A text recording the eleventh barley ration of L4.11b contains solely Geme2d Nanše’s personnel (DP 160 rev. iii 3–iv 3: še-ba geme2-dumu / Geme2d Nanše / dumu Lugal-an-da / ensi2 Lagaški-ka), from which we learn that Munus-sa6-ga had not been born yet. This is confirmed by other documents, which testify that a daughter (i.e., Munus-sa6-ga) was born during the month of the festival of the goddess Baba (= the twelfth month) of L4.8 The suggestion 5 Wet-nurses were employed for breast-feeding, the period of which was approximately three years in the ancient Near East (Gruber 1989: 63 and 76; Stol 2000: 181; Biggs 2000: 6). 6 In the well-known biblical story of Moses both midwife and wet-nurse appear: the Hebrew midwives’ excuse for not killing the newborn baby boys was that Hebrew women, being strong, gave birth before the midwives arrived (Exodus 1: 15–21); and Pharaoh’s daughter hired a Hebrew wet-nurse for the baby that she “drew out of the water” (Exodus 2: 6–10). For the fee for wet-nursing in Mesopotamia, see Stol 2000: 181–82. 7 Thus, Mesanepada’s seal inscription, Mes-an-ne2-pa3-da lugal Kiški dam nu-gig (RIME1.13.5.2; Pittman 2013: 332 fig. 16.15), must mean “Mesanepada, spouse of the Nugig (= Inanna)”; cf. Frayne’s (2008: 392) translation, “spouse of the nugig priestess.” 8 DP 218 rev. iii 6–iv 2: iti ezem-dBa-ba6-ka / Bara2-nam-tar-ra / dam Lugal-an-da / ensi2 / Lagaški-ka-ke4 / dumu-munus i3-tu-da-a “in the month of the festival of the goddess Baba when Baranamtara, wife of Lugalanda, ruler of Lagaš, gave birth to a daughter”; also Nik 1: 209 obv. ii 1–4. For the month cycle, see Cohen 1993: 38; Beld 2002: 107–17, 203. On the reading of Baba, see Rubio 2010: 35–39.

162

Fumi Karahashi

Tab. 1: Female personnel (ar3-tu munus) working for Lugalanda’s daughters (Geme2-dNanše and Munus-sa6-ga) and the amount of their barley rations. Text (date)

Geme2-dNanše

Munus-sa6-ga

DP 160 (L4.11b)

72 sila Al-tuš um-me 36 sila Geme2-id2-edin-na um-me-da 24 sila dInanna-ama-mu 24 sila A-teš2-mu

Ø

BIN 8: 23 (date broken) 72 sila Al-tuš um-me RTC 53 (L5.8b) 36 sila Geme2-id2-edin-na um-me-da 24 sila dInanna-ama-mu 24 sila A-teš2-mu

72 sila Nin-guru7 um-me 36 sila Gan-ezem um-me-da 24 sila Geme2-ganun

72 sila Al-tuš um-me 36 sila Geme2-id2-edin-na um-me-da 24 sila dInanna-ama-mu 24 sila A-teš2-mu 24 sila Ama-ra

72 sila Nin-guru7 um-me 36 sila Gan-ezem um-me-da 24 sila Geme2-ganun 24 sila Ama-teš2-mu

DP 157 (L6.9b) VS 25: 37 (L6.10b) VS 25: 14 (L6.11b)

made by Stol (2000: 181–92) that nursemaids who were supposed to take care of a weaned child may have already been active when the child was still being nursed is borne out in the case of Munus-sa6-ga. Text RTC 53 (see Table 1), which was written when Munus-sa6-ga was one year and eight months old and still being wet-nursed, lists both an um-me wet-nurse and an um-me-da nursemaid. As pointed out by Civil (2011: 283), “there is no information in these texts about their duties, only that the second of the two always receives half the grain ration of the first”9; an exception is the reference to an activity of an Ur III royal nursemaid (see 4.5 below). Nurses and midwives also occur in the so-called “pure milk and pure malt” (ga-ku3 munu4-ku3) texts (Table 2). In addition to the um-me wet-nurse Al-tuš (see Table 1), we learn of the existence of another um-me-da nursemaid named Geme2-e2-dam and a nu-gig midwife named Geme2-šu.10 The latter 9 For comparison of the rations of wet-nurses and nursemaids in Mari of the second millennium BCE, see Lion and Michel forthcoming. 10 A cylinder seal depicting a supine woman on a bed with a moon and a star above her and a scorpion underneath the bed and a kneeling person touching the woman’s feet (Asher-Greve 1985, Pl. XXX, No. 593; Westenholz 1999: 73 fig. 7) has been variously interpreted as a dream interpretation scene (Asher-Greve 1985: 113–14; 1987) and as a woman giving birth with a midwife at work (Ulla Koch-Westenholz, cited in Westenholz 1999: 172 n. 342). The typical position for giving birth in ancient Mesopotamia, however, was not horizontal but “crouching or squatting” (Stol 2000: 123; see also n. 86 referring to Drower’s ethnological description of Iraqi women giving birth this way); for sig4-tu-tu “brick(s) of birth,” see Civil 2011: 282 with n. 139

Royal Nurses and Midwives in Presargonic Lagaš Texts

163

Tab. 2: Nurses and midwives in the “pure milk and pure malt” texts. Text (Date)

DP 226 (L4); VS 14: 173 (L4); DP 132 (L5)

DP 133 (UL1); TSA 5 (UL2)

Name

Al-tuš um-me Geme2-e2-dam um-me-da

Geme2-šu nu-gig

should be identified with Geme2-šu-ga-lam-ma, who is named as a nu-gig midwife in other texts. In a text apparently dating to the reign of Lugalanda, Geme2-šu-ga-lam-ma is preceded by a woman named Ama-ša6 (VS 14: 106 = AWL 124 obv. ii 2–4), who seems to have been another nu-gig midwife. Two texts, one dated to UL2 and the other dated to UL3, which were classified as ziz2-ba ušur3 nam-dumu “emmer rations of the friends of royal children”11 mentioned certain Gan-ezem and designated her as nu-gig (DP 128 rev. i 5–6 and DP 129 rev. i 7–8). A question arises as to the identity of these two women named Gan-ezem: one was a nu-gig midwife and the other was an um-me-da nursemaid (Table 1). There seems to be a strong likelihood that these are references to the same individual. If this is indeed the case, it would mean that Gan-ezem could perform both nu-gig and um-me-da-functions and be designated by either job title. This reminds us the aforementioned § E2 of the Ur-Namma Laws, in which the um-me-da was supposed to perform nu-gig functions (Civil 2011: 284). Note, however, that the term um-me-da seems to refer to different professions in the Presargonic Lagaš and the Ur-Namma Laws.

3 Their Socio-Economic Status As we have already seen, the um-me wet-nurses Al-tuš and Nin-guru7 always received 72 sila of barley rations, which was twice the rations of the um-meda nursemaids Geme2-id2-edin-na and Gan-ezem (Deimel 1929: 126). Compared with other female personnel working on behalf of the royal daughters who

and also Stol 2000: 118–22. Cf. glyptic images in which a male and female are on a bed with a scorpion underneath and a man standing at the feet (Boehmer 1965: 121, Tafel LVIII, No. 691; Mazzoni 2002: 372 fig. 3.1) or a nude female is on a bed with a scorpion underneath (Boehmer 1965, Tafel LVIII, No. 690 = Mazzoni 2002: 372 fig. 3.3). While Boehmer (1965: 121 and 190) views the images as representing sacred marriage, Assante (2003: 15 fig. 1) interprets the standing man as an attendant manipulating the woman’s feet. 11 Cf. ušur3, described as “a group of seemingly important people” (Prentice 2010: 177) and “hetairai?” (Powell 1974: 401).

164

Fumi Karahashi

receive only 24 sila each (Table 1), the wet-nurses were clearly better off. It is particularly noteworthy that the rations given to the um-me wet-nurses (72 sila) were the same as those of E2-mete, the leader of the female personnel (Maekawa 1973–74: 104 n. 36) and Nin-uru-da-kuš2, the woman who was second to E2-mete in rank.12 In an Adab text (OIP 14: 65 obv. i 1–3, iii 4–rev. i 2), an um-me wet-nurse named Al-tuš appears as a borrower of a large quantity of barley: 10.0.0 še gur-sag-gal2 / Al-tuš / um-me / … / Amar-ezem / [ur-nu]-ke4 / [ur5]-še3 ba-šum2 “Amar-ezem … gave the um-me wet-nurse Al-tuš 10 gur of barley as loan” (Bauer 1975). This person might have been the aforementioned um-me wet-nurse Al-tuš of our Lagaš documents (Steinkeller 1981: 89). Regardless of her identity, this document testifies to the economic activities of an um-me wet-nurse. No documents mentioning landholding by an um-me wet-nurse or um-me-da nursemaid are preserved in the E2-mi2 corpus. Yet there are documents suggesting that nurses were allotted subsistence land. In one such document dating to E2, classified as še-ba ziz2-ba lu2-igi-nigin2 šub-lugal-ke4-ne “barley and emmer rations to the people of igi-nigin2 and šub-lugal,” an um-me-da nurse named Nam-šita was mentioned: 0.0.4 še 0.0.2 ziz2 / Namšita / um-me-da (BIN 8: 347 rev. iii 6–8). Maekawa (1973–74: 110) has pointed out that all of the šub-lugal and a large part of the lu2-igi-nigin2 were subsistence landholders (lu2-šuku-dab5-ba).13 Nam-šita must have belonged to the group of lu2-igi-nigin2 (see below). This is not direct evidence but might imply that she was allotted subsistence land. The female personnel who participated in the “pure milk and pure malt” ceremony (ga-ku3 munu4-ku3) were labeled lu2-igi-nigin2. Among them were, as seen above (Table 2), the um-me-da nurse Geme-e2-dam, the um-me wetnurse Al-tuš, and the nu-gig midwife Geme2-šu.14 The latter must be identified with the nu-gig midwife Geme2-šu-ga-lam-ma, who was allotted 6 iku (ca. 2.2 ha) in UL1 (VS 25: 70 rev. iv 4–5). The size of her field matches that of the aforementioned E2-mete (Karahashi forthcoming). She is also known to have received five mina of wool (VS 14: 106 obv. ii 3–4).

12 VS 25: 71 obv. vi 4–5 (L6.9b); RA 71: 102 obv. vi 7–8 (L6.12b). 13 Maekawa (1973–74: 109–14) takes BIN 8: 347 as the forerunner of še-ba lu2-šuku-dab5-ba ration lists (Deimel’s Type I). 14 Asher-Greve (1985: 158 with n. 111) understood the nu-gig Geme2-šu to be the wife of the chief scribe of Nanše. In fact, however, the nu-gig Geme2-šu was the giver and the wife of the chief scribe of Nanše was the receiver in a ceremony (DP 133 obv. iv 4–7; TSA 5 obv. iii 12–13, iv 1–2). The marital status of Geme2-šu is unknown to us.

Royal Nurses and Midwives in Presargonic Lagaš Texts

165

The nu-gig midwife Gan-ezem and a few others received 36 sila each on the occasion of emmer distribution in UL2, when many others received half that amount (18 sila), and only one received 72 sila (DP 128).15 An unnamed nu-gig was mentioned with 36 sila of barley and 36 sila of emmer in VS 27: 33 dated to L6. As is clear from the above, the information gathered on the Presargonic Lagaš nurses and midwives is rather scanty and the resulting picture fuzzy. In contrast to the cases discussed in section 4 below, neither cylinder seals belonging to a nurse or mid-wife nor seal impressions have been found among the Presargonic Lagaš material. Baranamtarra, wife of Lugalanda, is the only Lagaš woman whose seal is known to us (Orthmann 1975: 239, fig. 44b). The aforementioned documents, however, suggest that they enjoyed a relatively higher socio-economic standing than other regular female personnel, to the extent to that some were on the level of the important female servant E2-mete.

4 Prominent Nurses and Midwives in the Second Half of the Third Millennium BCE With the aim of shedding light on the Presargonic Lagaš material, the following section surveys the attested nurses and midwives of kings or important persons from the second half of the third millennium BCE, who are chronologically close to the women discussed in this paper.

4.1 Nippur From ED IIIb Nippur a ša3-zu “midwife” who dedicated a bowl is known. The broken inscription reads: (beginning lost) x [x]-na-na / ša3-zu / a mu-ru (Goetze 1970: 42; Braun-Holzinger 1991: 132 G97).16 She was an “independent” 15 The nu-gig Gan-ezem was believed to be the wife of a nu-banda3-official (Asher-Greve 1985: 158 with n. 111), but these were two different individuals. This misinterpretation is repeated in Goodnick-Westenholz 1989: 258. As is the case with Geme2-šu (see n. 14), we do not know this nu-gig’s family situation. 16 Civil (2011: 284) points out the possibility of “a conflict of attributions with the nu-gig” as can be seen in a co-occurrence of nu-gig and ša3-zu in a record of barley rations from Fara (ED IIIa), EDATS 6 (WF 74): ten women were designated as ša3-zu (rev. i 3–14) and five women as nu-gig (rev. v 1–6) – in the “total” section of the document (rev. vii 2–6) five occupation titles including nu-gig and ša3-zu occur but unfortunately their respective totals are not preserved. Cf. RTC 12 (SRU 62), which probably originated from Fara and deals with the giving of

166

Fumi Karahashi

woman, to use Asher-Greve’s (2006: 57) term for women who were of non-royal rank, working professionally, and possessing substantial economic means. Many, if not all, elite nurses and midwives might have fit this profile.17

4.2 Ebla At Ebla, the families of kings and prime ministers employed wet-nurses, and in the last period of the city there were fourteen of them (Biga 1997: 42–43). These women, together with other female court personnel, were mentioned in the administrative accounts of the issuing of food (e.g., Davidović 1987), textiles, and metals, and they are believed to have remained in the households until their death (Biga 2000: 64–65).18 Among the Eblaite wet-nurses the most prominent was Gisadu (Gi-sa/ša-du // Ki-sa/ša-du), the “wet-nurse of the king” (ga-du8 en), who served Dusigu, wife of Irkab-Damu, king of Ebla in the first half of the 24th century BCE (Biga 1997: 35).19 Gisadu had at least two sons of her own: one is named Bu-ma-na-im, who probably lived with the royal offspring in the palace (Biga 2000: 69 n. 21), and the other named Gi-da-na-im, who received the purification gift at the death of his mother (Biga 2000: 78– 79, texts [63] and [75]).

a house and a male individual. It lists six witnesses, five of whom were women, including a ša3-zu (obv. iv 3–4). 17 Cf. the stone statue of the Egyptian woman Sit Snefru (Twelfth Dynasty), found at Adana in Asia Minor: it is supposed that before leaving home to accompany the family of an Egyptian official dispatched to this remote region, she had the statue made to be placed in her tomb (Hayes 1953: 215 fig. 132). This also informs us of her substantial economic means. For other Egyptian nurses, see Robin 1993: 73 fig. 24, which shows the nurse of Amenhotep II, and Robin 1993: 76 fig. 25, which shows a nurse standing behind the man who dedicated the stela. 18 Biga (2000: 80) made the suggestion that some Eblaite nurses were dancers at a different point in their life. Another instance that might show the close relationship between nurses and musicians is the Fara text EDATS 6 (WF 74), mentioned above in n. 16, where gala, geme2kar-kid3, and sa-har occur together with nu-gig and ša3-zu (rev. vii 2–6). Pomponio (1986: 65–66) discussed a certain case in Fara in which a woman was narx “musician” in one text and geme2-kar-kid3 in another. Further, narx “musician” and sa-har were interchangeable, and sa-har and geme2-kar-kid3 were also interchangeable (Pomponio and Visicato 1994: 244–45). In the New Kingdom period in Egypt the wife of a chief priest or high-ranking official was in charge of female musical troupes of the temple, and one such woman named Merit was also nurse to a daughter of Thutmose IV (Robin 1993: 148–49). 19 For the synchronism between the Mesopotamian and Syrian dynasties, see the chronological table in Archi 1996: 228.

Royal Nurses and Midwives in Presargonic Lagaš Texts

167

4.3 The Wet-Nurse Takunai A cylinder seal of Takunai (Da-gu-na) was first published by Nougayrol (1960: 209–14, fig. 2) and then by Lambert (1987; 1988). This is “one of the very best quality in lapis lazuli,” according to Westenholz (1999: 72–73, fig. 8b), who differs from Nougayrol and Lambert in reading the seal’s inscription as follows: “Timmuzi, the abarakkatum: Takunai, her daughter’s wet-nurse (ama.ga.gu7).” The image depicts the scene in which the goddess Lama leads a woman towards the goddess Ninhursag seated on the left. The woman is followed by another woman with a pail (Suter 2008: 19, 52 fig. 17). Westenholz (1999: 72) and Asher-Greve (2006: 67) identify the second woman with Takunai, while Suter (2008: 19) takes the first as Takunai and the second as anonymous.

4.4 Urkesh Excavations at Tell Mozan, the ancient city of Urkesh, have yielded a seal that belonged to a nurse named Za-me-na, who served the queen Uqnitum: the legend reads “Seal of Za-me-na, the nurse (um-me-da) of Uqnitum” (Buccellati and Kelly-Buccellati 1995–96: 22; 1998: 197–200). Not only the inscribed legend but also the iconography of the seal explicitly linked its owner to the Urkesh queen. Budin (2011: 187) offers the following analysis: Uqnitum appears with the standard long braid with braid ornament, seated upon a stool, and … holding a small child upon her lap. Behind her stands a hairdresser braiding the queen’s hair. Before the queen and child pair stands Zamena, who has one hand on the child’s knee, while she hold the child’s wrist with her other hand … This is the scene of familiarity and intimacy. The nurse plays with the child’s hand while in the presence of the queen during her toilette.

She concludes, “Zamena’s seal most assuredly broadcasts the nurse’s intimacy with the royal family, a high honor.”20

4.5 Ur III In the Ur III documents several “royal nursemaids” um-me-da lugal are attested (Molina 2012: 396–97). The career of one such woman named Ku8-ba20 En passant, it should be compared with another seal from Urkeš that belonged to the female cook (whose name is unintelligible) of the same queen. It has been shown that the seals of the queen, the nurse Zamena, and the female cook were all cut in the same workshop (Kelly-Buccellati 1998: 42–47).

168

Fumi Karahashi

tum, who was mentioned in Amar-Suen 4 (CTMMA 1: 17 obv. ii 23–24) and who possibly later became the queen of Šu-Sin (Sigrist 1986: 185; Weiershäuser 2008: 154–55), may exemplify the intimate relationship between the nurses and the royal members. Ki-na-at-e2-a, the um-me-da nursemaid of Taddin-Eštar, Amar-Suen’s daughter, was recorded as taking animals to the gods for an offering on behalf of Taddin-Eštar during her illness (BPOA 7: 2721 obv. i 1–4; Civil 2011: 283). Note that some Ur III nurses are attested in relation with landholding. An um-me-da nurse (“wet-nurse” in Michalowski’s translation) or a woman named Ummeda had four bur (72 iku = 26 ha) of land (TCS 1: 253 = Michalowski 1993: 102, no. 196). Another um-me-ga wet-nurse was given one bur (18 iku = 6.5 ha)21 of subsistence land (šuku) (SAT 3: 2157 obv. 5). A cylinder seal owned by Nin-kal-la, who was designated as midwife (ša3-zu) of the goddess Baba, is preserved (Asher-Greve 2013: 367, 372, 373 fig. 18.7).

Concluding Remark Presargonic Lagaš documents provide rather limited information on nurses and midwives and do not allow us to get a clear picture. Nevertheless they point to the relatively high socio-economic status of these women as is suggested by larger rations, subsistence landholding, and membership in the group of lu2igi-nigin2. Although we cannot be certain whether they enjoyed a privileged situation similar to that of their prominent colleagues in other cities, their situation might not have been much different.

Bibliography Archi, Alfonso. 1996. Chronologie relative des archives d’Ébla. Pp. 11–28 in Amurru 1: Mari, Ébla et les Hourrites: dix ans de travaux, ed. Jean-Marie Durand. Paris: Éditions Recherche sur les Civilisations. Asher-Greve, Julia M. 1985. Frauen in altsumerischer Zeit. Bibliotheca Mesopotamica 18.Malibu: Undena Publications. Asher-Greve, Julia M. 1987. The Oldest Female Oneiromancer. Pp. 27–32 in La femmedans le proche-orient antique, ed. Jean-Marie Durand. CRRAI 33. Paris: Éditions de Recherche sur les Civilisations.

21 The number was kindly checked by Manuel Molina (BDTNS 050340).

Royal Nurses and Midwives in Presargonic Lagaš Texts

169

Asher-Greve, Julia M. 2006. ‘Golden Age’ of Women? Status and Gender in Third Millennium Sumerian and Akkadian Art. Pp.41–81 in Images and Gender: Contributions to the Hermeneutics of Reading Ancient Art, ed. Silvia Schroer. OBO220. Fribourg: Academic Press; Göttingen: Vandenhoeck & Ruprecht. Asher-Greve, Julia M. 2013. Women and Agency: A Survey from Late Uruk to the End of Ur III. Pp. 359–77 inThe Sumerian World, ed. Harriet Crawford. London: Routledge. Assante, Julia. 1998. The kar.kid / ḫarimtu, Prostitute or Single Woman? A Reconsideration of the Evidence. UF 30: 5–96. Assante, Julia. 2003. From Whores to Hierodules: The Historiographic Invention of Mesopotamian Female Sex Professionals. Pp. 13–47 in Ancient Art and Its Historiography, ed. Alice A. Donohue and Mark D. Fullerton. Cambridge: Cambridge University Press. Bauer, Josef. 1967. Altsumerische Wirtschaftstexte aus Lagash. Inaugural-Dissertation zur Erlangung der Doktorwürde der Philosophischen Fakultät der Julius-MaximiliansUniversität zu Würzburg. Bauer, Josef. 1975. Darlehensurkunden aus Girsu. JESHO 17: 189–218. Beld, Scott G. 2002. The Queen of Lagash: Ritual Economy in a Sumerian State. Ph.D. diss., The University of Michigan. Biga, Maria G. 1997. Les nourrices et les enfants à Ebla. Ktema 22: 35–44. Biga, Maria G. 2000. Wet-Nurses at Ebla: A Prosopographical Study. Vicino Oriente 12: 59– 88. Biggs, Robert D. 2000. Conception, Contraception, and Abortion in Ancient Mesopotamia. Pp. 1–14 in Wisdom, Gods and Literature: Studies in Assyriology in Honour of W. G. Lambert, ed. Andrew R. George and Irving L. Finkel. Winona Lake, IN: Eisenbrauns. Boehmer, Rainer M. 1965. Die Entwicklung der Glyptik während der Akkad-Zeit. Berlin: Walter de Gruyter. Bradley, Keith R. 1986. Wet-Nursing at Rome: A Study in Social Relations. Pp. 201–29 in The Family in Ancient Rome: New Perspectives, ed. Beryl Rawson. Ithaca, NY: Cornell University Press. Braun-Holzinger, Eva A. 1991. Mesopotamische Weihgaben der frühdynastischen bis altbabylonischen Zeit. Heidelberg: Heidelberger Orientverlag. Buccellati, Giorgio and Marilyn Kelly-Buccellati. 1995–96. The Royal Storehouse of Urkesh: The Glyptic Evidence from the Southwestern Wing. AfO 42–43: 1–32. Buccellati, Giorgio and Marilyn Kelly-Buccellati. 1998. The Courtiers of the Queen of Urkesh: Glyptic Evidence from the Western Wing of the Royal Storehouse AK. Pp. 195–216 in Subartu IV/2, ed. Marc Lebeau. Brussels: Brepols. Budin, Stephanie L. 2011. Images of Woman and Child from the Bronze Age: Reconsidering Fertility, Maternity, and Gender in the Ancient World. Cambridge: Cambridge University Press. Charpin, Dominique. 1977. Un nouveau compte de rations présargonique. RA 71: 97–105. Civil, Miguel. 2011. The Law Collection of Ur-Namma. Pp. 221–86 in Cuneiform Royal Inscriptions and Related Texts in the Schøyen Collection, ed. Andrew George. CUSAS 17. Bethesda, MD: CDL Press. Cohen, Mark E. 1993. The Cultic Calendars of the Ancient Near East. Bethesda, Maryland: CDL Press. Davidović, Vesna. 1987. The Women’s Ration System in Ebla. Oriens Antiquus 26: 299–307. Deimel, P. Anton. 1929. Die Lohnlisten aus der Zeit Urukaginas und seines Vorgängers (Fortsetzung). Orientalia Series Prior 43–44: 1–131.

170

Fumi Karahashi

Deimel, P. Anton. 1931. Šumerische Tempelwirtschaft zur Zeit Urukaginas und seiner Vorgänger: Abschluss der Einzelstudien und Zusammenfassung der Hauptresultate. Analecta Orientalia 2. Rome: Pontificio Istituto Biblico. Frayne, Douglas R. 2008. Presargonic Period (2700–2350 BC). RIME 1. Toronto: University of Toronto Press. Genouillac, Henri de. 1909. Tablettes sumériennes archaiques. Paris: Paul Geuthner. Goetze, Albrecht. 1970. Early Dynastic Dedication Inscriptions from Nippur. JCS 23: 39–56. Goodnick Westenholz, Joan. 1989. Qĕdēšā, Qadištu, and Sacred Prostitution in Mesopotamia. The Harvard Theological Review 83: 245–65. Gruber, Mayer I. 1986. Hebrew Qĕdēšāh and her Canaanite and Akkadian Cognates. UF 18: 133–48. Gruber, Mayer I. 1989. Breast-Feeding Practices in Biblical and in Old Babylonian Mesopotamia. JANES 19: 61–83. Hayes, William C. 1953. The Scepters of Egypt: A Background for the Study of the Egyptian Antiquities in the Metropolitan Museum of Art. Part I: From the Earliest Times to the End of the Middle Kingdom. Greenwich, CT: The Metropolitan Museum of Art. Karahashi, Fumi. Forthcoming. Women and Land in the Presargonic Lagaš Corpus. Kelly-Buccellati, Marilyn. 1998. The Workshops of Urkesh. Pp. 35–50 in Urkesh and the Hurrians: Studies in Honor of Lloyd Costen, ed. Giorgio Buccellati and Marilyn KellyBuccellati. (Urkesh/Mozan Studies 3). Bibliotheca Mesopotamia 26. Malibu: Undena Publications. Lambert, Wilfred G. 1987. Hurrian Names on a Seal? Oriens Antiquus 26: 13–16. Lambert, Wilfred G. 1988. Daguna’s Seal Again. N.A.B.U. 1988/83. Lion, Brigitte and Cécile Michel. Forthcoming. Les métiers féminins dans les palais et grands domaines de Syrie et haute Mésopotamie au début du IIe millénaire av. J.-C. Maekawa, Kazuya. 1973–74. The Development of the é-mí in Lagash during Early Dynastic III. Mesopotamia 8–9: 77–144. Mazzoni, Stefania. 2002. The squatting Woman: Between Fertility and Eroticism. Pp. 367–77 in Sex and Gender in the Ancient Near East: Proceedings of the 47 th Rencontre Assyriologique Internationale, Helsinki, July 2–6, 2001, ed. Simo Parpola and Robert M. Whiting. Helsinki: The Neo-Assyrian Text Project. Michalowski, Piotr. 1993. Letters from Early Mesopotamia. Atlanta, GA: Scholars Press. Molina, Manuel. 2012. Taddin-Eštar. RlA 13: 396–97. Nougayrol, Jean. 1960. Documents du Habur. Syria 37: 206–14. Orthmann, Winfried. 1975. Der alte Orient. Berlin: Propyläen Verlag. Pittman, Holly. 2013. Seals and Sealings in the Sumerian World. Pp. 319–41 in The Sumerian World, ed. Harriet Crawford. London: Routledge. Pomponio, Francesco. 1986. géme-kar-kìd: The Sumerian Word for “Prostitute.” Oikumene 5: 63–66. Pomponio, Francesco and Giuseppe Visicato. 1994. Early Dynastic Administrative Tablets of Šuruppak. Naples: Istituto Universitario Orientale di Napoli. Powell, M. A., Jr. 1974. Graphic Criteria for Dating in the Old Babylonian Period. Orientalia 43: 398–403. Prentice, Rosemary. 2010. The Exchange of Goods and Services in Pre-Sargonic Lagash. AOAT 368. Münster: Ugarit-Verlag. Robins, Gay. 1993. Women in Ancient Egypt. London: British Museum Press. Rubio, Gonzalo. 2010. Reading Sumerian Names, I: Ensuhkešdanna and Baba. JCS 62: 29– 43.

Royal Nurses and Midwives in Presargonic Lagaš Texts

171

Sallaberger, Walther and Ingo Schrakamp. 2015. The Presargonic Period. Pp. 67–84 in History and Philology, ed. Walther Sallaberger and Ingo Schrakamp. ARCANE 3. Turnhout: Brepols. Selz, Gebhard. 1993. Altsumerische Verwaltungstexte aus Lagaš, Teil 2: Altsumerische Wirtschaftsurkunden aus amerikanischen Sammlungen, 1. Abschnitt. FAOS 15/2–1. Stuttgart: Franz Steiner Verlag. Selz, Gebhard. 1995. Untersuchungen zur Götterwelt des altsumerischen Stadtstaates. Occasional Publications of the Samuel Noah Kramer Fund 13. Philadelphia: The University of Pennsylvania Museum. Sigrist, Marcel. 1986. Kubatum. RA 80: 185. Steinkeller, Piotr. 1981. More on the Ur III Royal Wives. ASJ 3: 77–92. Stol, Marten. 2000. Birth in Babylonia and the Bible: Its Mediterranean Setting. CM 14. Groningen: Styx Publications. Suter, Claudia E. 2008. Who are the Women in Mesopotamian Art from ca. 2334–1763 BCE? KASKAL5: 1–55. Weiershäuser, Frauke. 2008. Die königlichen Frauen der III. Dynastie von Ur. Göttinger Beiträge zum Alten Orient, Band 1. Göttingen: Universitätverlag Göttingen. Westenholz, Aage. 1999. The Old Akkadian Period: History and Culture. Pp. 17–117 in Mesopotamien Akkade-Zeit und Ur III-Zeit, ed. Pascal Attinger and Markus Wäfler. OBO 160/3. Freiburg: Universitätsverlag; Göttingen: Vandenhoeck & Ruprecht.

Jacob Klein

The Hymn to the Ekur: Its Literary Structure and Cultic Background Contrary to the other genuine temple hymns, composed during the third millennium BCE,1 this is a relatively short poem,2 provided with liturgical rubrics, typical of the majority of Ur III and Isin divine and royal hymns, which were used periodically in the temple cult. The hymn, which was first edited by Kramer (1957) nearly sixty years ago, had been preserved only in a single copy from the OB period.3 Since Kramer’s publication, it had not been subjected to any other thorough study.4 According to the native liturgical rubrics, the hymn seems to consist of three major parts, each concluding with a one-line antiphon (ĝiš-gi4-ĝál): sagida (ll. 1–30), saĝara (ll. 31–55), and a third part whose rubric is missing (ll. 56–71).5 This three-fold liturgical structure is somewhat obscured by the atypical rubric ki-ru-gú, which is inserted into the second, saĝara part, of the hymn (l. 42).6 Since the present, single copy of the hymn lacks a generic 1 For the three Old-Sumerian temple hymns (the so-called Zà-mì Hymns, the Keš Temple Hymn, and “the Collection of Sumerian Temple Hymns”), see my forthcoming article: “Temple Hymns in Early Sumerian Literature: A Literary-Historical Overview” (to be published in the Mordechai Cogan Festschrift). 2 Not counting the cultic rubrics, the hymn contains altogether 65 lines. 3 UM 29-16-51 (Photos: Kramer 1957: 96; CDLI P256659; cf. below Appendix, Fig. 1). 4 For a revised (electronic) transliteration and translation, see ETCSL 4.80.4. 5 In spite of the missing rubric, this third part also concludes with an antiphon (l. 70). 6 Actually the line reads: ki-«ud»-ru-gú-dam (“it is the kirugu”), clearly a scribal error. To the best of my knowledge, Sumerian hymns provided with the rubrics sagida and saĝara never contain the rubric kirugu, and vice versa. Accordingly, I suspect that a former version of this hymn, which was divided into two or more parts by the rubric kirugu, was sometimes revised and modified, with the original rubric displaced by the rubrics sagida, saĝara and an unknown third rubric. Acknowledgements: I dedicate this study to Miguel Civil, “The Treasurer of Sumerian Literature,” who provided us with the key to the innumerable sources of Sumerian literature. The temple hymn restudied here appeared in his “Catalogue of Sumerian Literary Texts” as “4.84 EkH”; cf. n. 4 above. I am grateful to Yitschak Sefati for reading a draft of this article, and providing important comments and criticism. Editorial note: This contribution does not follow the conventions employed to transliterate Sumerian in this volume (mostly in the use of accents instead of numerical indexes). In order to avoid potential replacement errors, however, the editors decided to leave the Sumerian as submitted. https://doi.org/10.1515/9781501503696-011

The Hymn to the Ekur: Its Literary Structure and Cultic Background

173

subscript, its original type cannot be determined. It may have been a tigi, like the building and dedication hymn Urnamma B.7 Just like the liturgical type and structure of this temple hymn is obscure, its content and literary structure are also very complicated, and therefore it is difficult to determine its cultic background. On the basis of the liturgical rubrics, as well as various literary considerations, the hymn may be divided into four poetic units: A Praise of the Various Buildings of the Ekur Complex (ll. 1–30).8 B “For him who declared … the house comes forth like the daylight” (ll. 31– 42).9 C A Repeated Praise of the Ekur Complex (ll. 43–55).10 D Praise of Ninurta as Worthy of the Status of the First Born of Enlil (ll. 56– 70).11 In the following discussion, I will describe in detail the content and literary structure of this short hymn, as I understand it, and I will make an attempt to determine its complicated cultic-historical background.

7 For Urnamma B see Klein 1989: 44–56; Flückiger-Hawker 1999: 183–203 (ETCSL 2.4.1.2). Interestingly, the above 70 lines long tigi hymn, also dedicated to Enlil and the Ekur, contains only two parts: sagida (ll. 1–38) and saĝara (ll. 40–71), with no antiphons, just like its predecessor Gudea A. To the best of my knowledge, adab hymns with the above rubrics, each provided with an antiphon (ĝiš-gi4-ĝál) appear first time in the Šulgi corpus (see Šulgi G and Sulgi Q). However, our hymn could hardly be an adab type composition, because this generic subscript is characteristic only of royal hymns, which as a rule consist of a sagida, saĝara and a (three or four line) urubi section. For the rubrics sagida, sagara, urubi, barsud and šabatuk in Sumerian hymns, see Klein 1981: 26, 44 f. n. 94; Wilcke 1975: 290 ff. Table of tigi. Klein points out the simple structure of the Šulgi hymns (sagida, saĝara and urubi) as against the more complex structure of the Isin hymns. A survey of the hymns ending in u18-ru12-bi-im indicates that, as a rule, all of them are adab-hymns, the urubi contains only 3–4 lines, and never ends with an antiphon. An exception to this rule is Rimsin B (presumably and adab, containing an urubi and an antiphon, but no other rubrics!). Note, also the divine hymn Nanna L, a šir-nam-gala, which consists of 2 antiphons without preceding rubrics, and a kirugu, ending with an urubi. 8 This section is marked by the rubric sa-gíd-da (l. 28) and contains a one line antiphon (l. 29). 9 This section is marked by the rubric ki-«ud»-ru-gú-dam (l. 42) and contains no antiphon. 10 This section is marked by the rubric sa-ĝar-ra (l. 53) and contains a one line antiphon (l. 54). 11 This section is not marked by any rubric, but contains one line antiphon (l. 69). The cultic rubric was probably omitted by scribal error (cf. the empty line 68b in our transliteration below).

174

Jacob Klein

A Praise of the Various Buildings of the Ekur Complex (ll. 1–27) This section is introduced by a three-line prologue: é gal kur-ra-àm gal é den-líl-lá kur-ra-àm gal é dnin-líl-lá kur-ra-àm gal

The great house is as great as a mountain. The house of Enlil is as great as a mountain. The house of Ninlil is as great as a mountain.

The prologue is followed by a panoramic list of all the more important buildings of the temple complex of Enlil (ll. 4–13) and Ninlil (ll. 15–27),12 each of them endowed by the recurring praise kur-ra-àm gal “is as great as a mountain.”13 Both the expression é gal and the recurring metaphor kur-ra-àm-gal are obvious word-plays on the well known epithet of Enlil: kur-gal.14 This section ends with the obscure antiphon (l. 29): 12 The section praising the temple complex of Ninlil is again introduced by the sentence: é dnin-líl-lá kur-ra-àm gal (= l. 3). 13 The following edifices of the Ekur are listed in the subsection, belonging to Enlil: é itimama, é ud nu-zu, é ká mah, é ká silim-ma, kisal den-líl-lá, hur-saĝ-galam-ma (= the Ekur ziggurat), ká gal-di kug, ká še nu-kud, ub-šu-unken-na, ĝá-ĝiš-šú-a (= cella of Ninlil in the Ki’ur). The following edifices are listed in the subsection belonging to Ninlil: ká-innam-ra, é-itid-da-buru14, kisal é-gal-mah, é-itid-da-buru14, en-túm-gal-zu, in-nam-gídda-zu, ká dsuen, du6-kug, a-šag4 é-dìm-ma, an-né-ĝar-ra, aš-te ki sikil, é-tilla2-mah, ĝáĝiš apin-na. It is not clear whether the above list follows the order of importance (i.e., from the most important building/locale to the least important one), or a certain topographical order. Furthermore, at least some of the gate edifices in the list seem to belong to the city at large, rather than to the temple complex (for a graphic reconstruction of the Ekur and the Ki’ur, see Gibson, Hansen, and Zettler 1998–2001: 551 fig. 2; for a reconstruction of the city plan of Nippur, with some of its gates, see Gibson, Hansen, and Zettler 1998–2001: 560 fig. 10). It is not the purpose of this article to discuss the location and the function of each of these buildings, and document them in Sumerian literature and various other genres (for the most important edifices see George 1993: 63 ff. sub “Gazetteer”). However, it is interesting to note that all these buildings, without exception and in the same order, are listed in some of the OB and later balaĝ lamentations, which are dedicated to Enlil and the Ekur; see e.g., Utugin Eta 16–52 (CLAM 97–98); zi-bu-ù-um zi-bu-ù-um 13–52 (Cohen 1988: 348–50). Noteworthy is also the reference to the (é) itima and the é ud nu-zu in the Curse of Agade 129, as well as to the two gates ká še nu-kud-da and ká silim-ma in the Curse of Agade 123–26; and even more interesting is the reference to the five architectural terms ká mah, ká gal, ká silim-ma, hur-saĝgalam-ma and ká še nu-kud-da in Urnamma B 22 (for hur-saĝ-galam-ma, see further Urnamma B 29–32). 14 This epithet of Enlil appears already in the prologue of the archaic Zà-mì Hymns from Abu Ṣalābīkh (cf. Biggs 1974: 46). See especially the Enlil Hymn 35–40 (the introductory lines to the praise of the Ekur): uru á-dam kug den-líl-lá-kam / nibruki èš ki áĝ a-a kur gal-la-kam / barag nam-hé é-kur é za-gìn sahar-ta ša-ba-ra-an-íl / hur-saĝ íl-la-gin7 ki sikil-la bí-in-

The Hymn to the Ekur: Its Literary Structure and Cultic Background

bí-in-dug4 an-šè

175

He declared: “Heavenward!”15

B “For him who declared … the house comes forth like the daylight” (ll. 31–42) The entire Ekur temple complex, which was described in the first section as a great (cosmic) mountain, shines forth for Ninurta (and/or to Nanna), who declared that he belongs to the temple (or who praised the temple).16 This obscure section is beset with the following grammatical and exegetical difficulties: a) Who is the subject of the recurring verbal form bí-in-dug4ga-ra, and what is the precise meaning of the verb dug4 in this context? Kramer assumes that the verb means “to command,” and that the subject of this verb is Enlil, translating “for him who had commanded … (the house rose like the sun).” Another possible interpretation could be to assume that the verb means “to declare” (i.e., to praise), and the subject of the sentence is someone other than Enlil, probably Ninurta, who is mentioned by his name in l. 39; and perhaps also Nanna, who may be alluded to by one of his epithets dumu-nunna “the princely son” in l. 41. b) What is the nature, the substance of the relevant command/declaration? Kramer seems to assume that the nature (i.e., the direct object) of this command is a hypothetical sentence like é-e an-šè saĝ hé-ni-ni-ib-íl-e “Let the house raise high its head to heaven.”17 Accordingly,

mú / nun-bi kur gal a-a den-líl / é-kur èš mah-a barag-ba dúr bí-in-ĝar “In the city, the holy settlement of Enlil, / in Nibru, the beloved shrine of father Great Mountain, / he has made the dais of abundance, the E-kur, / the shining temple, rise from the soil; / he has made it grow on pure land as high as a towering mountain. / Its prince, the Great Mountain, Father Enlil, / has taken his seat on the dais of the E-kur, the lofty shrine.” Otherwise, the epithet kur-gal (a-a) d en-líl is documented by ETCSL in Sumerian literary texts no less than 65 times. 15 So ETCSL, with the tacit assumption that the subject of the sentence is someone other than Enlil, probably Ninurta, who is mentioned for the first time in l. 39 (and/or Nanna who may be referred to in l. 41 by the epithet dumu-nun-na). Kramer translates: “He (=Enlil?) commanded: ‘Towards heaven’–”; in his commentary he surmises that an-šè may be an abbreviation for é-e an-šè saĝ hé-ni-ib-íl-e “Let the house raise high its head to heaven,” anticipating the refrain of the fourth section: é-e me gal-la saĝ mi-ni-íb-íl (ll. 43 and passim). 16 Kramer assumes that the indirect object, marked with the dative -ra refers to Enlil, who uttered a command (bí-in-dug4-ga-ra). However, with all probability, the dative refers to Ninurta (and/or Nanna), who is (or are) the subject of the third and fourth section, and not to Enlil. For a detailed discussion of this problem, see below. 17 Reconstructed on the basis of the former antiphon (l. 29) and the refrain ll. 43 and passim.

176

Jacob Klein

he interprets the preceding nominal phrases as locative adverbs. Thus, e.g., he translates lú kur-ra bí-in-dug4-ga-ra (l. 33) “For him who in the mountain had commanded …” and so on. Another possibility is to assume that the phrases preceding the recurring verbal form constitute its direct object, each to be understood as genitives without the regens (which is only implied). Accordingly, lú kur-ra bí-in-dug4-ga-ra (l. 33) may be translated “(For him who declared that) he is of the mountain,” and so on. c) The last three lines (39– 41) seem to be the most problematic part of this section: Firstly, the scribe seems to have omitted erroneously the head noun é in ll. 39 and 41.18 Secondly, the mention of “house of Ninurta/the princely son” is an unexpected turn in the poem, which until this point spoke of the “house of Enlil” and the “house of Ninlil” only.19 Thirdly, one wonders, whether the phrase “the princely son” (41) is just an epithet of Ninurta, or perhaps it is an epithet of Nanna, who may also be mentioned later by the epithet dumu nun é-kur-ra (l. 67).20 This section of our hymn ends with the cultic rubric ki-«ud»-ru-gú-dam “(This) is the Kirugu,” which exceptionally is not followed by an antiphon.

C A Repeated Praise of the Ekur Complex (ll. 43–54): The poet resumes his praise of the entire Ekur, referring to it by the two all inclusive terms: é “house” and kisal “courtyard” (both of Enlil and Ninlil). At this time each of these entities are described by the extended refrain: me gal-la saĝ mi-ni-ib-íl šag4-bi-ta kur šim ĝišerin-na-kam “in accordance with the great me-s it raised its head high, in its midst is a mountain of aromatic cedars.” This refrain is obviously the resumption and elaborated extension of the refrain kur-ra-àm gal of section A (ll. 1 ff.), as well as of the enigmatic

18 Restored both by Kramer and ETCSL. 19 Another, secondary problem, is the question whether the “house of Ninurta” refers to a mere ‘guest-chamber’ within the Ekur complex for this god, or perhaps it is identical with his well known temple é-šu-me-ša4 in the city (for Ninurta as the official city god of Nippur and his temple, the Ešumeša, see George, 1993, p. 147 sub No. 1065; Streck, RLA 9, p. 519, sub #13.1 “Kultorte”). 20 So Kramer. For dumu nun-na and dumu nun é-kur-ra as epithets of Nanna-Suen, see the references collected by Sjöberg, 1960, p. 142, comment to l. 2 (cf. also Hall 1985, pp. 646 f., sub 7. dumu-nun). For é dumu nun-na as the name of a temple of Sîn in Uruk see George 1993: 80 (No. 214). Otherwise we find the epithet dumu nun-na applied to other deities, such as Šara (TH 309); Numušda (Sîn-iqīšam A 9); Ninĝišzida (Ninĝišzida D 5); and Šuziana (TH 83).

The Hymn to the Ekur: Its Literary Structure and Cultic Background

177

phrase an-šè in the antiphon of that same section (l. 29). In simple terms, the poet tells us that the Ekur temple complex resembles to a towering cosmic mountain of aromatic cedars.21 This section is concluded by a damaged antiphon, partially restored by Kramer as [é…..] àm-da-húl-la (l. 54) “The [house…] in which he rejoiced with them.” Kramer probably assumed that the subject of the verb is Enlil, and that the dative prefix -da- refers to the two junior deities, Ninurta and Nanna, with whom he presumably rejoiced in the temple. However, traces in the beginning of the line seem to point to an en sign, and restoring é seems to be excluded. But, even if we restore ˹den˺?-[líl?] in the beginning of the line,22 it remains unclear who rejoices over whom: Enlil over his offspring(s), the offspring(s) over Enlil, or perhaps Enlil rejoices over his towering temple.23 Thus the context and the meaning of the antiphon remain obscure.24

D Praise of Ninurta as Worthy of the Status of the First Born of Enlil (ll. 56–70) In this last section, the poet seems to praise Ninurta, who confessed to belong to the Ekur (or praised it) in the second part of the hymn (ll. 31–41). Ninurta is presented here as the heroic offspring of Enlil and Ninlil, being worthy of the

21 The likening of a Sumerian temple to a towering mountain, is a very common literary topos; cf. Išme-Dagan W 36 (referring to Nippur); TH 66 (referring to the Ešumeša); The Keš temple hymn 8 and passim. The expression “in accordance with the great me-s it raised its head high” (me gal-la saĝ mi-ni-ib-íl) is attested in similar context only in Gudea Cyl B vi 8 and xvi 3 (referring to the Eninnu). The likening of a temple to an “aromatic cedar mountain” (kur šim ĝišerin-na) is alluded to already by the short hymn to Nanna of Ur in the Zà-mì Hymns from Abū-Ṣalābīkh, which reads: urim2 kur šim lugal dnanna zà-mì (ll. 35–36). I could not find this expression elsewhere in temple hymns, although it is quite frequent in other literary compositions. 22 The traces almost clearly point to the bottom of an en, and there is enough space in the lacuna for a DN (˹den˺-[líl] or ˹den˺.[zu]) and one more medium size sign (see further n. 24 below). 23 For húl with the comitative prefix, see Gragg 1973: 62; Thomsen 1984: 397. See especially Nanna’s Journey to Nippur 319 dsuen-ra den-líl mu-un-da-húl mí zid na-mu-un-ne “Enlil rejoiced over Suen and spoke kindly”; Enki and the World Order 262 den-ki-da den-líl muun-da-húl nibruki [giri17-zal-àm] “Enlil rejoiced over Enki, and Nibru was happy.” 24 One could tentatively restore in the damaged line ˹den˺?-[líl?-da?] àm-da-húl-la “He who rejoiced with/over Enlil,” assuming that the reference is to Ninurta, who is the subject of the following section. However, it is difficult to fit the non-finite verbal form into the context.

178

Jacob Klein

status of the first born heir or crown-prince of Enlil the king.25 The poet refers to Ninurta by his name, and by his major filial epithets, repeating the refrain: é nam-dumu zid-da lugal den-líl-ra túm-ma-àm “in the true house of youth, is worthy of Enlil the king.”26 Surprisingly, in addition to the two neutral epithets,27 Ninurta is endowed with two epithets, which otherwise are regularly applied to Nanna: en daš-ím-babbar (l. 65) “Lord AŠimbabbar”28 and dumu nun é-kur-ra (l. 67) “the princely son of the E-kur.”29 25 Literally é nam-dumu zid-da lugal den-líl-ra túm-ma-àm “worthy of Enlil the king in the true house of youth.” For nam-dumu in a literary text, see Nanna E 8 nam-dumu nun gal-zu den-lil2-le nam-men! šu mi-ri-in-du7 “and Enlil has perfected for you your great and noble filial status and lordship.” Otherwise, the term nam-dumu in the meaning “status of a (crown-)prince” or “princeship,” is common in OB royal inscriptions, see Edzard 1958: 74 with n. 357 (appearing in expressions like ká nam-dumu, bàd nam-dumu and uru-nam-dumunaki). As to é nam-dumu, usually referring to the estate of the king’s or the ruler’s son(s) or child(ren), it is common in ED IIIb and Ur III royal and administrative documents; cf. e.g., Nik 1: 227 obv. ii 3 é nam-dumu-ta; SANTAG 7: 108 obv. 6–7 3 ˹gu4˺ 1 áb níĝ-gur11 é namdumu ensi2 nibruki; RIME 1.9.9.1 vii 9–10 and passim é nam-dumu ašag nam-dumu-ke4 “The estate of the (ruler’s) child(ren) and the fields of the (ruler’s) child(ren).” 26 For Ninurta as the first-born of Enlil and Ninlil see recently Streck 1998–2001: 513 sub § 4 “Familie und Hofstaat.” For his temple at Nippur, the é-šu-me-ša4, see TH No. 5, ll. 61–76; George 1993: 147 s. v. A most significant reference to Ninurta in the “Sumerian Temple Hymns,” from the present point of view, is piriĝ kur-gal-e tu-da (l. 73). 27 He is “the lord, the hero (en ur-) of the Ekur” (l. 61), the offspring (dumu-ka) of Ninlil and Enlil (ll. 59 and 63). For dumu-ka, a somewhat multivalent kinship term, see already RIME 1.9.5.17 i 13, where Enmetena is called dumu-ka ur-dnanše (“the descendant of Ur-Nanše”). In the Gudea Temple Hymn 1342 (Cyl B xxiii 18), Ninĝišzida is referred to as dumu-ka an-nakam, which is translated variously “he is the grandson of An” (Edzard) and “he is the greatgrandson of An” (Sjöberg). For extensive discussions of this term see Sjöberg 1960: 42 n. 4; Sjöberg 1967: 209–211. Sjöberg (ibid.) notes that emendation of dumu-ka to dumu-*saĝ (“firstborn”) in our hymn is excluded by the photo. Indeed, a careful collation of the photo indicates an absolutely clear dumu-ka in l. 59; and a slightly damaged dumu-ka! in line 63 (the expected wedge in the front box of the ka sign was obliterated by a crude vertical gash, made either by an ancient scribe, or by a modern skeptic Sumerologist, who seems to have been dissatisfied by the reading dumu-ka in his context). Note finally, that in TCL 16: 86 obv. 5 (Šulgi F 70) Nanna is referred to as dumu-ka an-kug-ga “grandson of An” (so Sjöberg 1967; see also Hall 1985: 645 and 729; Lämmerhirt 2012: 57). 28 For the reading, meaning and theology of this epithet, see discussion in Hall 1985: 42–46; Krebernik 1993–97: 362 f. sub Mondgott I. § 2. 3. They suggest the alternative reading dil-ímbabbar and the meaning “The-Shining-Solitary-Runner.” 29 For dumu nun é-kur-ra applied to Nanna-Suen elsewhere, see the fragmentary passage Šulgi D 47–50 dumu nun é-kur-[ra] […] / lugal má-gur8 [kug an-né dirig-ga] / nanna en […] / dsuen […]. For refernces to the short form dumu-nun(-na) in connection to this deity, see Sjöberg 1960: 142, comment to l. 2; see especially Nanna E 15; Šulgi G 17; Nanna’s Journey to Nippur 246; Lament for Sumer and Ur 29. For occasional references by this epithet to other deities, see Sjöberg 1960: 146 n. 2. See especially TH 309, where Šara, Inanna’s son, is referred

The Hymn to the Ekur: Its Literary Structure and Cultic Background

179

Kramer (followed by Sjöberg), assumes that in ll. 65–68 in our hymn the poet refers to Nanna-Suen, who as a major son of Enlil and Ninlil, is celebrated here as having had a chamber of honor in the Ekur temple complex, along with Ninurta.30 This interpretation of section D of our hymn, however, raises two questions: (a) If the poet wanted to speak here of Nanna-Suen, why did he not refer to him by one of his genuine proper names Nanna or Suen? (b) If the section is praising both Ninurta and Nanna as the favorite sons of Enlil, why does its antiphone (l. 70) say den-líl-lá lú šag4-ga-na “He is the favorite one of Enlil,” referring only to a single junior deity?31 A reasonable solution to these inconsistencies could be to assume that the present text does not represent the original version of the Hymn to the Ekur, but a modified, somewhat corrupted version of the original. As a working hypothesis, we may posit that originally this hymn was composed during the reign of one of the early kings of the Ur III dynasty, and the young heroic deity, who was described in sections B and D as the offspring of Enlil and Ninlil, “worthy of Enlil the king in the true house of youth” was Nanna-Suen.32 The hymn may have been composed in the time of Urnamma, for the celebration of the dedication of the newly rebuilt Ekur, presumably when a statue of Nanna-Suen was installed into a shrine (é nam-dumu) built especially for him, in proximity to the shrine of Enlil or Ninlil.33 Alternatively, the installment of Nanna’s statue to the Ekur could have taken place at the beginning of the reign of Šulgi, on the day of his coronation in Nippur, or on any other suitable

to as dumu nun nu-gig-ga; and TH 148, where Ningublaga seems to be referred to as dumunun. 30 See Kramer’s (1957: 202) comment; Sjöberg 1960: 142, comment to l. 2. 31 This rarely attested poetic expression is applied in Sumerian cultic poetry only to Ninurta and Nuska. See Ninurta C 55 [a]-a-ĝu10 den-líl-le lú šag4-ga-na-me-en; Nuska B 14 dnuska d en-líl-lá lú šag4-ga-na-me-en. For somewhat similar expressions see lú ˹šà?-ab? ˺-[ĝá] “the man of my heart” (restored in Dumuzi-Inana F 40); and the DN dšul-šà-ga-na. See further Šulgi D 13 šul-gi dnin-líl-lá ki áĝ šag4-ga-na “Šulgi, the beloved of Ninlil’s heart”; Ninurta’s Journey to Eridu Segment D 15–16 [nam]-˹mah˺-zu den-líl-lá níĝ šag4-ga-na-ka [dnin-urta] nam-mah˺-zu den-líl-lá níĝ šag4-ga-na-ka “your grandeur is a pleasure to Enlil; Ninurta, your grandeur is a pleasure to Enlil” (cf. also Ninurta’s Journey to Eridu ll. 17–20). 32 With all probability, Nanna was elevated to the status of Enlil’s firstborn by the Ur III theologians, whose intention was to ignore or refute the older Lagaš theology, which attributed this status Ninĝirsu. As has long been recognized, the older Lagaš theology is reflected especially in the two Ninurta myths, “Ninurta’s Return to Nippur” and “Ninurta’s Exploits” (see recently discussion by Klein 206: 295–98). 33 For the tigi-hymn, Urnamma B, which commemorates the building and dedication of the Ekur by Urnamma, see Klein 1989: 44–56; Flückiger-Hawker 1999: 183–203 (ETCSL 2.4.1.2).

180

Jacob Klein

occasion.34 The present form of our temple-hymn may be a rewritten, adapted version of the Ur III one, commissioned by one of the kings of the Isin dynasty, who decided to restore the senior status of Ninurta as the first-born of Enlil, by removing Nanna’s statue from the é nam-dumu, installing instead Ninurta’s statue into this (or another) shrine. The purpose of this supposed revision and adaptation could be mainly political: To demonstrate the demise of the religious and political supremacy of Ur, after its collapse, and thus to assert the supremacy of the kingdom which inherited it. Or else, the motive behind this cultic and literary reform could be to win the support of the priesthood of Nippur in general or the priesthood of Ninurta in particular, in the new political power that came to dominate the old and prestigious religious center of the land.35 A preliminary, statistical, survey of the royal hymns with prayers to Nanna and Ninurta in the Ur III and Early OB periods seems to support this hypothesis.36 Such a survey indicates that the Ur III kings commissioned far more royal hymns and prayers dedicated to Nanna-Suen than to Ninurta, at the ratio of 9 : 2.37 On the other hand, the Isin kings clearly preferred Ninurta over Nanna-Suen, as a deity to pray to; the ratio between extant prayers to Ninurta and Nanna in the Larsa corpus is 6 : 1.38 34 For another hymn, presumably commemorating the installation of a divine statue into a temple by Šulgi, see Šulgi O 1–52 (Klein 1976: 272 and 286, comment to ll. 29–33). 35 This hypothesis is suggested here with utmost reservations. One could alternatively posit an original hypothetical version of the hymn, which was dedicated to the installation of the statues of both minor deities, Ninurta and Nanna into the Ekur. Admittedly, this alternative hypothesis, suggested already by Kramer, could explain the rather vague epithet “offspring (dumu-ka) of Enlil,” applied here to both Ninurta and Nanna, rather than the more common and lucid epithet “first-born (dumu-saĝ) of Enlil,” which would sound contradictory in such a context. However, it is unlikely that one and the same hymn would refer to two minor deities as being worthy for the é nam-dumu, unless we assume that this term refers to “the house of the (royal) children” (i.e., the house of the princes). Equally improbable would be to assume that the original version of the hymn was dedicated to the installation of a statue of Ninurta, the present version being an adaptation to the elevation of Nann-Suen to his status; for in that case the oblique and corrupted allusions to the latter would be inexplicable. 36 The following survey is based on sections 2.3–2.6 of the ETCSL catalogue. 37 The distribution of royal hymns/prayers to Nanna-Suen in the Ur III period is: Urnamma (2), Šulgi (2 – Šulgi F and Šulgi X which end with a zà-mí doxology to AŠimbabbar), Šu-Sin (1), Ibbi-Sin (4). The distribution of royal prayers to Ninurta, in the same period is: Šulgi (1), Šu-Sin (1). It is interesting to note, that the kings of Larsa, who most of the time controlled only Ur, commissioned all of their prayers to Nanna (as against 6 prayers to this deity, no prayer to Ninurta survived in the Larsa corpus). 38 While we find only one royal prayer to Nanna in the Larsa corpus (Išme-Dagan M), there are 6 prayers to Ninurta in this corpus, according to the following distribution: Išme-Dagan (3), Lipit-Eštar (1), Ur-Ninurta (1), Būr-Suen (1). As to the genuine divine hymns, which survived in copies from the OB period, there are 19 hymns dedicated to Nanna, as against 8 hymns

The Hymn to the Ekur: Its Literary Structure and Cultic Background

181

This is not the only case of a temple hymn which was revised and adapted for political or religious purposes. Already C. Wilcke (2006) has pointed out that the Keš Temple Hymn, originally a local composition, was most probably reworked so as to adapt it for the national Enlil theology. Another example for the adaptation of an old temple hymn to a new theological system is the addition of a hymn dedicated to the Ehursaĝ of Šulgi, to the ‘Collection of Sumerian Temple Hymns’.39 We also know of a Dumuzi-Inana love-song, which has been preserved in two versions: An older version, in which Dumuzi figures as the royal lover and provider of Inana, and a later version in which it is Išmedagan of Isin, who fulfills this role, with his name replacing that of Dumuzi.40 Another, even closer, parallel to the present case, is the myth Enlil and Ninlil.41 As Annette Zgoll (2013) convincingly demonstrated, the present version of this myth is an adaptation and revision of an older, ED version, by the poets of Urnamma of Ur, for the purpose of commemorating or inaugurating his grand canal-digging and irrigation project. For that purpose, the poet has rewritten the myth replacing in it three ancient, minor deities with three major divine figures of the Ur III period: Enlil, Ninlil and Nanna. But what is even more important to the hypothesis presented above as to our temple hymn is the observation, that when the ancient poets revised and rewritten an old myth or poem, they often did it somewhat carelessly, leaving in the new version contradictions and inconsistencies, which betray the complex history of their composition.42 Perhaps, it is not a fortuity that three of the above four cases of adaptations of cultic literature are connected to the Ur III period, in one way or another.

dedicated to Ninurta (see ETCSL, sections 4.13 and 4.27 respectively). If we assume that most of these hymns were composed in the Ur III period, this ratio is reasonable. However, the composition of these hymns cannot be dated with any certainty. 39 Hymn No. 9 (ll. 119–134). Cf. the colophon dah-hu-um é-hur-saĝ šul-gi uri2ki-ma “Supplement. Ehursaĝ Šulgi of Ur.” 40 The poem under discussion is A Balbale to Inana (ETCSL 4.08.a) and Išmedagan J (ETCSL 2.5.4.10). For a separate edition of the two versions see Klein 1998. 41 Editio princeps: H. Behrens 1978 (ETCSL 1.2.1). 42 See especially Zgoll 2013: 93 and 103 f.

182

Jacob Klein

Appendix The Hymn to the Ekur: Transliteration and Translation 43 A Praise of the Various Buildings of the Ekur Complex  1  2  3  4  5

é gal é den-líl-lá é dnin-líl-lá é itimama é ud nu-zu

kur-ra-àm gal kur-ra-àm gal kur-ra-àm gal kur-ra-˹àm˺ gal kur-ra-˹àm˺ gal

 6

é ká mah

kur-ra-àm gal

 7

é ká silim-ma

kur-ra-àm gal

 8

kisal den-líl-lá

kur-ra-àm gal

 9 10

hur-saĝ-galam-ma ká gal-di kug

kur-ra-àm gal kur-ra-àm gal

11

ká še nu-kud

kur-ra-àm gal

12 13

ub-šu-unken-na ĝá-ĝiš-šú-a

kur-ra-àm gal kur-ra-àm gal

14 15 16 17

é dnin-líl-lá ká-in-nam-ra é-itid-da-buru14 kisal é-gal-mah

kur-ra-àm gal kur-ra-àm gal kur-ra-àm gal kur-ra-àm gal

18 19 20 21 22

é-itid-da-buru14 /mah en-túm-gal-zu in-nam-gíd-da-zu ká dsuen du6!(iti)-kug ki kug

«kur-ra-àm gal» kur-ra-àm gal kur-ra-àm gal kur-ra-àm gal kur-ra-àm gal kur-ra-àm gal

23 24 25

a-šag4 é-dìm-ma/ an-né-ĝar-ra aš-te ki sikil

kur-ra-àm gal kur-ra-àm gal kur-ra-àm gal

The great house is as great as a mountain. The house of Enlil is as great as a mountain. The house of Ninlil is as great as a mountain. The bedchamber is as great as a mountain. The house which knows no daylight is as great as a mountain. The house at the Lofty Gate is as great as a mountain. The house at the Gate of Well-being is as great as a mountain. The courtyard of Enlil is as great as a mountain. The Ḫursaĝ-galama is as great as a mountain. The holy Renowned Gate is as great as a mountain. The Gate of Incessant Grain-Supplies is as great as a mountain. The Ubšu-unkena is as great as a mountain. The Ĝa-ĝiš-šua is as great as a mountain. The house of Ninlil is as great as a mountain. The Innamra Gate is as great as a mountain. The E-itida-buru is as great as a mountain. The courtyard of the Egal-maḫ is as great as a mountain. The lofty E-itida-buru is as great as a mountain. The Entum-galzu is as great as a mountain. The Innam-gidazu is as great as a mountain. The Suen Gate is as great as a mountain. The Du-kug, the holy place, is as great as a mountain. The field of E-dima is as great as a mountain. The Ane-ĝara is as great as a mountain. The Ašte, the pure place, is as great as a mountain.

43 The revised transliteration of the hymn, offered below, is based on a collation of a high quality photograph found in the library of the S. N. Kramer Institute of Assyriology, at Bar-Ilan University (cf. Fig. 1 below).

The Hymn to the Ekur: Its Literary Structure and Cultic Background

26 27 28

é-tilla2-mah/ ĝá-ĝišapin-na sa-gíd-da-àm

29 30

bí-in-dug4 an-šè ĝiš-gi4-ĝál-bi-im

kur-ra-àm gal kur-ra-àm gal

183

The E-tilla-mah is as great as a mountain. The Ĝa-apina is as great as a mountain. (This is the) Sa-gida. He declared: “Heavenward!” (This is) its ĝišgiĝal.

B “For him who declared … the house comes forth like the daylight” 31

bí-in-dug4-ga-ra/ bí-in-dug4-ga-ra

32 33

é ud-dam ša-mu-na-è lú kur-ra bí-in-dug4-ga-ra

34 35

é ud-dam ša-mu-na-è é den-líl-ka/ bí-in-dug4-ga-ra

36 37

é ud-dam ša-mu-na-è é dnin-líl-ka/ bí-in-dug4-ga-ra

38 39

é ud-dam ša-mu-na-è dnin-urta-ka/ bí-in-dug4-ga-ra

40 41

é ud-dam ša-mu-na-è dumu nun-na-ka bí-in-dug4-ga-ra

42

ki-«ud»-ru-gú-dam

For him who declared it, for him who declared it, the house comes forth like the daylight. For him who declared that he is of the mountain, the house comes forth like the daylight. For him who declared that he is of the house of Enlil, the house comes forth like the daylight. For him who declared that he is of the house of Ninlil, the house comes forth like the daylight. For him who declared that he is of the house of Ninurta, the house comes forth like the daylight, for him who declared that he is of the house of the princely son. (This) is the Kirugu.

C A Repeated Praise of the Ekur Complex 43 44 45 46 47 48 49 50

é-e me gal-la/ saĝ mi-ni-íb-íl šag4-bi-ta kur šim ĝišerin-na-kam é den-líl-lá-ke4 me gal-la/ saĝ mi-ni-ib-íl šag4-bi-ta kur šim ĝišerin-na-kam é dnin-líl-lá-ke4 me gal-la/ saĝ mi-ni-ib-íl šag4-bi-ta kur šim ĝišerin-na-kam kisal den-líl-lá-ke4 me gal-la/ saĝ mi-ni-ib-íl šag4-bi-ta kur šim ĝišerin-na-kam

The house towers high in full grandeur; in its midst is a mountain of aromatic cedars. The house of Enlil towers high in full grandeur; in its midst is a mountain of aromatic cedars. The house of Ninlil towers high in full grandeur; in its midst is a mountain of aromatic cedars. The courtyard of Enlil towers high in full grandeur; in its midst is a mountain of aromatic cedars.

184

Jacob Klein

52 53

kisal dnin!(en)-líl-lá-ke4 me gal-la/ ˹saĝ mi˺-ni-ib-íl ˹šag4˺-[bi-ta kur šim] ˹ĝiš˺erin-na-kam [sa-ĝar]-ra-àm

The courtyard of Ninlil towers high in full grandeur; in its midst is a mountain of aromatic cedars. (This) is the [sa-ĝa]ra.

54 55

˹den˺?-[líl? (x)] àm-da-húl-la ĝiš-gi4-ĝál-bi-im

……… who rejoiced over it/him. (This) is its ĝišgiĝal.

51

D Praise of Ninurta as Worthy of the Status of the First Born of Enlil 56

Its master, in the true house of youth is worthy of Enlil the king. The hero Ninurta, in the true house of youth, is worthy of Enlil the king. The offspring of Ninlil, in the true house of youth, is worthy of Enlil the king. The lord, the hero of the E-kur, in the true house of youth, is worthy of Enlil the king. The offspring of Enlil in the true house of youth, is worthy of Enlil the king. Lord Ašimbabbar, in the true house of youth,

68

lugal-bi é nam-dumu zid-da/ lugal d en-líl-ra túm-ma-àm ur-saĝ dnin-urta/ é nam-dumu zid-da lugal den-líl-˹ra˺ túm-ma-àm dumu-ka dnin-líl-lá/ é nam-dumu zid-da lugal den-líl-ra túm-ma-àm en ur- é-kur-ra/ é nam-dumu [zid-da] lugal den-líl-˹ra túm˺-ma-àm dumu-ka! den-líl-lá/ é nam-dumu zid-da lugal den-líl-ra túm-ma-àm en daš-ím-babbar/ é nam-˹dumu˺ zid-da lugal den-líl-ra túm-ma-àm dumu nun é-kur-ra/ é nam-˹dumu˺ zid-da lugal den-líl-ra túm-ma-àm

68a



69

d

He is the favourite of Enlil.

70

ĝiš-gi4-ĝál-bi-im

57 58 59 60 61 62 63 64 65 66 67

en-líl-lá lú šag4-ga-na

is worthy of Enlil the king. The princely son of the E-kur, in the true house of youth, is worthy of Enlil the king.

(This) is its ĝišgiĝal.

The Hymn to the Ekur: Its Literary Structure and Cultic Background

Fig. 1: UM 29-16-51 obverse.

185

186

Jacob Klein

Fig. 2: UM 29-16-51 reverse.

The Hymn to the Ekur: Its Literary Structure and Cultic Background

187

Bibliography Behrens, Hermann. 1978. Enlil und Ninlil. Ein sumerischer Mythos aus Nippur. Studia Pohl Series Major 8. Rome: Biblical Institute Press. Biggs, Robert D. 1974. Inscriptions from Tell Abū Ṣalābīkh. OIP 99. Chicago: University of Chicago. Cohen, M. E. 1988. The Canonical Lamentations of Ancient Mesopotamia I–II. Potomac, MD: CDL. Edzard, Dietz O. 1957. Die “Zweite Zwischenzeit” Babyloniens, Wiesbaden: Harrassowitz. Flückiger-Hawker, Esther. 1999. Urnamma of Ur in Sumerian Literary Tradition. OBO 166. Friboug: University Press; Göttingen: Vanderhoeck & Ruprecht. George, Andrew R. 1993. House Most High: The Temples of Ancient Mesopotamia. Mesopotamian Civilizations 5. Winona Lake, IN: Eisenbrauns. Gibson, McGuire, Donald P. Hansen, and Richard L. Zettler. 1998–2001. Nippur. B. RlA 9: 546–65. Gragg, Gene B. 1973. Sumerian Dimensional Infixes. AOATS 5. Neukirchen-Vluyn: Neukirchener Verlag. Hall, Mark G. 1985. A Study of the Sumerian Moon-God Nanna/Suen. Ph.D. dissertation, University of Pennsylvania. Klein, Jacob. 1976. Šulgi and Gilgameš: The Two BrotherPeers. Pp. 271–92 in Kramer Anniversary Volume, ed. Barry I. Eichler et al. AOAT 25. Neukirchen-Vluyn: Neukirchener Verlag. Klein, Jacob. 1981. Three Šulgi Hymns: Sumerian Royal Hymns Glorifying King Šulgi of Ur. Ramat Gan: BarIlan University Press. Klein, Jacob. 1989. Building and Dedication Hymns in Sumerian Literature. Acta Sumerologica 11: 27–67. Klein, Jacob. 1998. The Sweet Chant of the Churn: A Revised Edition of Išmedagan J. Pp. 205–22 in dubsar anta-men: Studien zur Altorientalistik: Festschrift für Willem H. Ph. Römer zur Vollendung seines 70. Lebensjahres mit Beiträgen von Freuden, Schülern und Kollegen, ed. Manfried Dietrich and Oswald Loretz. AOAT 253. Münster: Ugarit-Verlag. Klein, Jacob. 2001. The Genealogy of Nanna-Suen and Its Historical Background. Pp. 279–301 in Historiography in the Cuneiform World (= CRRAI 45), ed. Tzvi Abusch Bethesda, MD: CDL Press. Kramer, Samuel Noah. 1957. Hymn to the Ekur. Pp. 95–102 in Scritti in onore di Giuseppe Furlani, ed. Giuseppe Furlani. Rivista degli Studi Orientali 32. Roma: Giovanni Bardi. Krebernik, Manfred. 1993–97. Mondgott. A. I. In Mesopotamien. RlA 8: 360–69. Lämmerhirt, Kai. 2012. Die Sumerische Königshymne Šulgi F . Texte und Materialen der Hilprecht Collection 9. Wiesbaden: Harrassowitz. Sjöberg, Ǻke. 1960. Der Mondgott Nann-Suen in der sumerischen Überlieferung. Stockholm: Almqvist & Wiksell. Sjöberg, Ǻke. 1967. Zu einigen Verwandtschaftsbezeichungen im Sumerischen. Pp. 201–31 in Heidelberger Studien zum alten Orient. Wiesbaden: Harrassowitz. Streck, Michael P. 1998–2001. Ninurta/Ninĝirsu. A. I. In Mesopotamien. RlA 9: 512–22. Thomsen, Marie-Loise. 1984. The Sumerian Language, Copenhagen: Akdemisk Forlag. Wilcke, Claus. 1975. Formale Gesichtspunkte in der sumerischen Literatur. Pp. 205–316 in Sumerological Studies in Honor of Thorkild Jacobsen on His Seventieth Birthday June 7, 1974, ed. Syephen Lirberman. AS 20. Chicago: The University of Chicago Press.

188

Jacob Klein

Wilcke, Claus. 2006. Die Hymne auf das Heiligtum Keš: Zu Struktur und “Gattung” einer altsumerischen Dichtung und zu ihrer Literaturtheorie. Pp. 201–37 in Approaches to Sumerian Literature: Studies in Honour of Stip (H. L. J. Vanstiphout), ed. Piotr Michalowski and Niek Veldhuis. Cuneiform Monographs 35. Leiden: Brill. Zgoll, Annette. 2013. “Fundamente des Lebens: Vom Potential altorientalischer Mythen.” Pp. 79–107 in Arbeit an Mythos: Leistung und Grenze des Mythos in Antike und Gegenwart, ed. Annette Zgoll and Reinhard G. Kratz. Tübingen: Mohr Siebeck.

Bertrand Lafont

Game of Thrones: the Years when Šu-Sin Succeeded Amar-Suen in the Kingdom of Ur As a contribution to this volume in honor of Miguel Civil, which I am happy to join, I propose to reassess a file of political history concerning the time of the Third Dynasty of Ur,1 a subject to which Civil has contributed so much. Following an investigation initiated by H. Waetzoldt (1970–71), I tried in 1994 to answer a double question (Lafont 1994)2: What happened at the head of the Ur kingdom at the time of the transmission of power between kings Amar-Suen (AS) and Šu-Sin (ŠS), the third and fourth kings of the Ur III “Dynasty”? And how are we to interpret the many anomalies or puzzling data that we observe in various groups of administrative records from this brief period? Several scholars have pointed out that disturbances seem to have marked the three or four years between the end of the reign of Amar-Suen and the beginning of that of his successor Šu-Sin.3 It appears that some of the kingdom’s elites were reluctant to recognize the authority of the new sovereign. Over twenty years after my earlier survey, and despite the publication of much new data, we must admit that we still do not quite understand what really happened at the time of this succession, for all its clearly troubled context. Thus, for F. Pomponio (2013: 231), “these chronological anomalies are the most thrilling challenge to the scholarship of the Neo-Sumerian period,” while P. Michalowski (2013b: 310 n. 105) considers that “the various changes, reforms, and general instability during Amar-Sin’s reign require a full investigation.” An important point must however be stressed from the outset: the uncertainty of the last years of Amar-Suen’s reign is most likely rooted in conditions from a decade earlier with his accession to the throne of Šulgi, the king who

1 I am very grateful to Dan Fleming, Jacob Dahl, Bob Englund, Manuel Molina and Walther Sallaberger who read a draft of this manuscript and offered helpful comments. Needless to say, they are not committed by the ideas and conclusions proposed here, especially as these are based on some hypothetical scenarios. “P numbers” of documents mentioned in this article are id numbers of the CDLI database at . 2 It was at a time when there was no CDLI nor BDTNS …! 3 See among others Maekawa 1996a: 127; Sallaberger 1999: 166–67; more recently, Dahl 2004: 131; Michalowski 2013a; 2013b; Pomponio 2013. https://doi.org/10.1515/9781501503696-012

190

Bertrand Lafont

may have been murdered after ruling for 48 years.4 Those events unfortunately remain unclear themselves, and our ignorance is reinforced by the fact that we still do not know with certainty what kinship linked Šulgi to his three successors. While several hypotheses have been advanced, none can be confidently affirmed:5 1. The most widely accepted is consistent with the “paradigm” of the Sumerian King List:6 from Ur-Namma down to Ibbi-Sin a hereditary dynastic succession from father to son would have prevailed (“father-son” pattern). Šulgi and Taram-Uram would be Amar-Suen’s parents; Amar-Suen and Abi-simti would be Šu-Sin’s parents;7 Šu-Sin and Kubatum would be IbbiSin’s parents.8 2. Several arguments have nevertheless been advanced in favor of the idea that the three successors of Šulgi were all his sons, and were therefore brothers. Some texts offer a few clues in that direction and thus, for L. Vacin (2011: 62–63 n. 134), “together, this new data seems to imply that all Šulgi’s successors were his sons, albeit probably from two different mothers, i.e. Amar-Suena from one queen and Šu-Suen and Ibbi-Suen, together with their sister, Simat-Ištaran, children of another queen.”9 3. A third scenario has been recently proposed by P. Michalowski (2013b: 316– 17), which might explain the difficulties encountered by Amar-Suen to emerge: he would not have been Šulgi’s son, unlike his two successors, but his nephew, son of the Mari princess Taram-Uram and an unknown brother or half brother of Šulgi. More generally, certain features of Amar-Suen’s reign, when compared to other kings of the Dynasty, have been highlighted on several occasions: lack of hymnography related to him, an unfavorable reputation conveyed in the later scholarly tradition, etc.10 The odd lack of any trace of Amar-Suen in the documentation from Šulgi’s period also raises questions: was he in exile or far from

4 According to P. Michalowski (2013b: 293), one scenario is that “Šulgi was assassinated and for political reasons two of his consorts [Šulgi-simti and Geme-Ninlila] were murdered with him”; see also ibidem: 315. 5 See Dahl 2004; Vacin 2011; Michalowski 2013b: 302 and 314. 6 See Michalowski 2013b: 288–89. 7 A soon to be published votive inscription (information courtesy of D. I. Owen) confirms explicitly that Queen Abi-simti is the mother of Šu-Sin. 8 See Michalowski 2013b: 288; Vacin 2011. On the relationship between Šulgi, Amar-Suen and Šu-Sin, see also most recently Sallaberger 2012; Wu 2012. 9 See also Dahl 2004; 2007. 10 Michalowski 1977; Pomponio 1990; Sallaberger 1999: 167.

When Šu-Sin Succeeded Amar-Suen

191

the major centers of the kingdom at that time? Or did he bear a different name, prior to the one that would become his “throne name”?11 In any case, it has been suspected that Amar-Suen was not, after Šulgi’s death, the first awaited heir. Be that as it may, all these questions surely have a connection with the events that took place in the last years of this king’s reign. From this starting point, let us try to put in order all available information that reveal changes, peculiarities or anomalies in the ordinary course of the affairs of the kingdom during the time in question. However, it is beyond the scope of this brief contribution to elaborate all changes in assignment for the many royal servants, which occurred here and there throughout the kingdom. Below then, in an order both geographical and chronological, are the main available data:

1 Entire Kingdom AS6

AS7 AS7–8

The beginning of the military campaign led by King Amar-Suen against Huhnuri, an Iranian place between Susa and Anšan.12 According to P. Michalowski (2013b: 309–11), Amar-Suen himself took leadership of his army, and during his long absence, the prince Šu-Sin left Der (Tell ‘Aqar near modern Badra, on the eastern marches of the kingdom), where he resided as a military chief, to return to the centers of power. See also below the reference to Huhnuri related to the ensi of Umma in AS7/iv. From this date, Amar-Suen’s official royal title changes: he is called lugal kal-ga, “strong king,” instead of nita kal-ga.13 As noted by L. Allred (2013: 119), “so many governors changed during Amar-Suen’s reign, particularly during the final years of his reign.” The turnover of governors at that time concerned notably Umma, Girsu, Kazallu, Kiš, Šuruppak, Irisagrig, Babylon, Ešnunna.14 Many of these new ensis had no direct links with their predecessors, nor do they seem to have had any deep ties to the local area, familial or otherwise. Were they appointed by Amar-Suen and

11 Sallaberger 1999: 163; Wu 2012; Michalowski 2013b: 296. 12 Close to the present Ramhormoz region, in southwestern Iran. See Nasrabadi 2005. 13 Sallaberger 1999: 166 and 180. But this would have to be checked again carefully, especially as information comes mainly from seal inscriptions. 14 See Allred 2013: 122–23.

192

AS9/ii

ŠS2–3

Bertrand Lafont

his followers, or by Šu-Sin and his new supporters? Were these changes the visible result of a competition between two groups?15 At this point, it seems certain that Amar-Suen is dead, since mortuary offerings are now assigned to him.16 I still think that the Girsu texts (see below under 4. Girsu) allow us to assert that Amar-Suen was considered – at least by some in the kingdom – dead or no longer able to govern as early as the eighth year of his reign and perhaps even before, although texts continued to be dated with his year names. But, contrary to what we know for the other rulers of the Dynasty, how are we to explain the dearth of information on the demise of this king? A period of strong affirmation of the new king’s power. Some of the changes imposed by the preceding king are removed and many of Šu-Sin’s decisions seem to be a reaction against those of the previous period. An important reorganization of the administrative system and of the Reichskalender is undertaken (Sallaberger 1999: 167 and 170).

2 Ur AS7/x

P. Steinkeller (2008: 187 and n. 4) has shown that, “sometime during the tenth month of Amar-Suen’s seventh regnal year, an important event took place at Ur. A number, or perhaps even all, of the generals of the realm had gathered in the city to swear a loyalty oath to the king. (…) This is the only documented instance of high state officials taking a loyalty oath in Ur III times.”17 We can note

15 Could this idea of “two groups” be linked to the situation described by M. Molina in a forthcoming article (paper read at the RAI 59 in Ghent in 2013), where he highlights the two figures of the sukkalmah and the zabardab and their struggle for power during the period that concerns us? He thus insists on “the growing political figure of Arad-Nanna [linked especially to Girsu and Irisagrig] and his movements at that time, and the decline of the zabar-dab5.” I thank Manuel for having sent me his paper prior to publication. 16 Or more precisely: “to his throne” (Nisaba 30: 50 [P130517]). See Sallaberger 1993: 147; 1999: 167 and n. 157; Dahl 2010: 89. See also the three important texts PDT 1, 541 [P125957], SAT 2: 1142 [P144342], and BPOA 7: 2072 [P290581], respectively dated to AS9/iv, AS9/v and ŠS1, celebrating the a-bu-um damar-dsuen, a festival that has a clearly funerary character. See Sallaberger 1993: 205–207; Steinkeller 2008: 188. 17 The text is in Steinkeller 2008: 185–86 [P200514].

When Šu-Sin Succeeded Amar-Suen

AS8/xi

193

that there must have been around 1,000 in attendance of the meal offered on the occasion of the oath ceremony. Enthronement of a new high priestess (en) of Nanna at Gaeš, following the death of the previous office holder (Enagazi’anna). The interesting point is that the name of this high priestess, enthroned between AS8/xi and AS9/I (Frayne 1997: 241), is En-Amar-Suenaraki’ag-Anna, “En:Amar-Suen-is-the-beloved-of-An,” even though it was the time of Amar-Suen’s demise.

3 Umma AS6

Seals of some royal servants begin to be dedicated to the “deified king Šu-Sin.”18 Did the Umma administration consider this a period of co-regency (Waetzoldt 1970–71)19 or were they cautious concerning some uncertainty as to who was the head of the kingdom? AS6/vii Emergence of a new temple estate named e2 damar-dsuen20 (to exist until ŠS3) and appearance of the new Umma month iti ezemd amar-dsuen (in use until ŠS5). Did this celebrate a dead king or a living one? AS7/iv An Umma text is perhaps linked to the end of the military campaign against Huhnuri.21 AS8/viii A new governor (ensi2) is appointed at Umma: Ayakalla replaces Ur-Lisi. He starts using two seals at the same time, one dedicated to Amar-Suen, the other to Šu-Sin.22 Assets of his predecessor UrLisi are confiscated and perhaps Ur-Lisi is even killed.23 Furthermore, M. Molina (in press) has shown that, from then, accusations are launched against some provincial officers at Umma, who were in charge before AS8: they are dismissed and condemned. AS9–ŠS1 During these two years not one messenger text is written at Umma whereas these texts are particularly numerous before and after this

18 The oldest is rolled on the tablet UTI 3, 1845 [P139864]. See Sallaberger 1999: 166; Allred 2010. 19 See Sallaberger 1999: 166. 20 MVN 5, 57 [P114277], etc. 21 BPOA 2, 2681 [P340515]: “1 zamirītum-weapon for Dadda[…], the son of Buzu, man of Huhnuri, when he went to the ensi2 of Umma to hand over the barley loan (še ur5-ra).” 22 Archi and Pomponio 1995: 15; Maekawa 1996a: 127; Dahl 2007: 63–67. 23 Maekawa 1996a; 1997; Heimpel 1997; Dahl 2007: 27 n. 110; Stepien 2012: 29 n. 56.

194

Bertrand Lafont

period.24 This situation is the reverse of the one attested at Girsu and Irisagrig, where there are many messenger texts between AS9 and ŠS1. For exactly the same period, there is a gap among the livestock accounts.25 An explanation for such an odd situation has still to be found.

4 Girsu AS3–7

AS7/xi

AS7/xi

In AS3, Ur-Lama, governor of Girsu, is eliminated along with the members of his family, and is replaced in the governorship by Nanna-zišagal, then Šara-kam and, finally, from AS7 on, by the sukkalmah Arad-Nanna.26 Between AS3 and ŠS3, Namhani’s official estate (e2 nam-ha-ni)27 is renamed “Amar-Suen’s estate” (e2 damard suen) (Sallaberger 1999: 167). This e2 Namhani returns with its original name after ŠS2 (Michalowski 2013a: 192). Furthermore, for several years there is in Girsu an intensification of commemorations of the former ensis of Lagash, particularly the deified Gudea. According to P. Michalowski (2013a: 188–96; 2013b: 310), Girsu was distinguished at that time by some “signs of instability and perhaps quiet resistance to the central authorities signaled by the celebration of local rulers of the past.” Arad-Nanna, sukkalmah, is appointed as new governor at Girsu.28 Arad-Nanna having married a daughter of Amar-Suen,29 this should have secured his loyalty to this king. Yet in spite of this, he is appointed at the time when Šu-Sin takes over the throne (he keeps his position of ensi of Lagaš throughout the entire reign of this king and even afterward). Here begins an exceptional period of regular shipments of heavy cargoes from Girsu, linked to the bala duty. Tens of boats bring shipments until AS9/xi (during months iii-vi and x-xii of year AS8, these departures are almost daily).30 The boats are loaded with

24 Di Pasquale 2002; Pomponio and Notizia 2006: 180–81; Notizia 2008; 2009: 101 n. 224. 25 See notably Heimpel 1993: 158, or several charts in Stepien 1996: 72, 78, 79, and 85. 26 Maekawa 1996a: 121–23; 1996b: 177–78; Michalowski 2013a: 191. 27 From the name of the former ensi2 of the Lagaš II Dynasty, still worshiped locally. 28 Maekawa 1996a: 122 n. 28; Ouyang 2013: 34 n. 53. 29 See Dahl 2004: 130–31. 30 Sharlach 2004: 86–90, 335–36; Lafont 2010: 168–74 (see the list of the texts on p. 169 and the synthetic chart on p. 173 of this article).

When Šu-Sin Succeeded Amar-Suen

AS8/iii

AS8/iv

195

personnel, food, furniture and various utensils. Texts mention the cities of Ur, Nippur, Uruk, Girsu, Isin, Sagdana and Huhnuri. People mentioned include: the king (lugal), the sukkalmah, several šaginas, and certain VIPs such as Dukra, Naplanum, Nanna-kam, Ilalum, and Babati.31 In my opinion, “l’embarquement et le départ régulier de Girsu, pendant toute cette période, de plusieurs centaines de bateaux chargés d’hommes de troupe, de personnels spécialisés, de bétail, de victuailles les plus diverses, de mobilier, de vaisselle, d’ustensiles variés, ne peuvent être intervenus que dans le cadre d’un événement exceptionnel” (Lafont 2010: 174). In addition, this “boat file” is definitely linked to the contemporaneous dossier of personnel assignments described below. I still think that the main destination of these barges was Nippur and that all these simultaneous events were maybe directly or indirectly linked to the demise (or infirmity/incapacity?) of Amar-Suen (Lafont 1994: 111– 12; see now n. 52 below). About twenty texts on personnel management are written between AS8/iii and ŠS1/ix, notably concerning assignments to the funeral chapels (e2/ki-a-nag) of Ur-Namma, and Šulgi, plus Amar-Suen in some texts (these dated from AS9/xi).32 This dossier is surely linked to the shipments question (see above), and it is contemporaneous with the oaths sworn by the new sangas of the main institutional households of the Girsu province (see below).33 In RA 88, I gave the reasons why I thought these deliveries could have been related to the funerary rites of Amar-Suen, but I was probably wrong (for a new hypothesis concerning these texts, see below n. 52). Appointment of several new sangas at the head of the main institutional households of the Lagaš province; they all have to take an oath before the sukkalmah, who is also the new governor of the

31 For important persons such as Naplanum and Babati, see recently Michalowski 2011: 108– 21, and 147–49. For Ilalum, see below under 5. Irisagrig. 32 See Lafont 1994: 107–10, and since then notably Uchitel 1996; Sallaberger 1999: 310–13; Sharlach 2004: 95–98, 338–40; Allred 2008. Texts of that group and that period range in date from AS8/iii to ŠS1/ix. The complete text file is currently composed of 20 texts: ITT 2: 970 [P110840]; 3503 [P110952]; TCTI 1: 914 [P110784]; 949 [P110819]; TCTI 2: 2796 [P132390]; TCTI 3: 5652 [P416106]; 6429 [P416107]; UDT 41 [P136172]; HLC 175 (pl. 104) [P110050]; RA 10: 66, pl. IV no. 100 [P127609]; ASJ 18: 223–28 [P102671, P102673, P102675]; Fs Sigrist: 14–18 [P102666, P102667, P102668, P102669, P102670, P381724]; PPAC 5: 723 [P377959]. 33 This idea of interlinked files was the starting point of my RA article over twenty years ago (Lafont 1994).

196

Bertrand Lafont

province. At the same time and probably linked to these appointments, a systematic census of the full staff of these households is undertaken between AS8 and ŠS1. One related text has the feature of being composed of a tablet dated to ŠS1, and an envelope dated to … AS8!34 AS9–ŠS1 Unlike the situation attested at Umma (see above), a high proportion of Girsu messenger texts are dated to the period AS9–ŠS1. In contrast, administrative texts of the še-ur5-ra category disappear at Lagaš during the same period, while they are common before and after this period.35

5 Irisagrig AS7/iii AS7/xi

The dating of the many messenger texts from Irisagrig starts from that time (Owen 2013b: 132). While Ur-mes is known as governor of Irisagrig from AS3 down to IS9 (i.e. during a quarter century), there is a break in AS7–8, when two other individuals, Ilalum and Dadani, are mentioned as governors (Owen 2013a: 53).36 Ur-mes is back in office in AS9/i (Nisaba 15/2: 146 [P453668]). Note further that there is a peak in AS7–9 of the number of dated texts from Irisagrig, compared to the immediately previous and following years (Owen 2013a: 60).

6 Nippur AS9

The three important Nippur offices of ugula e2 dinanna, šabra d inanna, and ensi2 nibruki were lost by the members of the Urmeme family during the reign of Amar-Suen: these offices return to the family’s control at the end of that reign.37

34 ITT 2: 907 [P110777]: see Lafont 1994: 104–106. The staff census tablets are currently five: P110735; P110777; P110960; P111014; P111159; the oath tablets are six (with the new one published in Lafont 2010: 167–68): P110914; P111022; P111179; P115922; P133386; P416105. For these six texts, see also now the note of P. Steinkeller in NABU 2015/61. 35 Notizia and Pomponio 2008: 82; Pomponio 2013: 231. 36 Note that Ilalum appears in the “boat file” of Girsu above. 37 See Zettler 1992: 177–213; Dahl 2004: 131; the chart in Dahl 2007: 25–26.

When Šu-Sin Succeeded Amar-Suen

197

7 Drehem AS6/xii

AS7–8

AS9

From that date, the royal guard usually composed of aga3-us2 is no longer mentioned in the Drehem texts. In its place appears a group called “gar3-du damar-dsuen” for a period of just over two years ending in AS9/ii; then the aga3-us2 reappear when Amar-Suen is already dead.38 Several changes are attested at Drehem. According to M. Sigrist (1992: 273), “une crise politique grave a dû se produire au début de l’année AS7. Elle s’est traduite par le remplacement des gendarmes aga3-us2 par les gar3-du. (…) Et durant les cinq premiers mois d’AS8, Lugalamarku remplace Abbašaga au bureau central de Drehem. De plus, un important renouvellement de personnel a lieu à cette date dans le bureau central de Drehem.”39 From AS8/v on, we observe a resumption of the “deliveries to the king” (mu-du lugal) which had been suspended since AS5 (Sigrist 1992: 280).40 At the end of the reign of Amar-Suen, an intense administrative activity is attested at Drehem: most tablets of inventories and summary accounts date back to that time.41

For now, it is difficult to go beyond this simple statement of successive facts, although they could be used to develop many various scenarios. Some additional issues and questions, however, remain unsolved: 1. The chronological ordering of all texts dated to the period between AS8 and ŠS1 continues to be problematic. However, the equation AS8 = AS9 = ŠS1 that I had proposed in RA 88 as a hypothesis (on the basis in particular of the text ITT 2: 907 [P110777], see below and n. 34) cannot be maintained and should probably be abandoned.42 Generally speaking, there are too many arguments to show that years AS8, AS9 and ŠS1 are fully differentiated in the administrative records: sufficient here is to mention the summary year list from AS1 to IS3 on the tablet CUSAS 17: 101 [P250592], or the many balanced accounts that perfectly differentiate these years.43 Several 38 Sigrist 1992: 273; the chart in Allred 2006: 97–100. 39 See also Sigrist 1992: 283 and 291. 40 See also Tsouparopoulou 2013: “Most changes in the officialdom took place during the three-year period Amar-Suen 5–8, apparently a most troublesome time for this king.” (§ 2.2.7). 41 See Lafont 1994: 112–13 n. 44. 42 But see the attempts of P. Notizia and others (see above n. 24). 43 See for example YOS 4: 313 [P142377] or ITT 2: 3772 [P110978] that explicitly states that there were seven full years between AS7 and ŠS4, and many other texts with complete year

198

Bertrand Lafont

“quirks” nevertheless remain to be explained, perhaps related to the attempts of some administrators to reorganize differently the calendar of that period. – The Girsu text ITT 2+TCTI 1: 907 [P110777] is composed of a tablet dated to ŠS1 and its envelope dated to AS8. – An unusual year name is found on a Nippur tablet (NATN 462 [P121160]): mu en eriduki u3 en ga-eški ba-hun (= AS8+9). Are two different year names conflated here into a single one?44 – Texts dated with “mu us2-sa AS9-year-name” are numerous and occur until ŠS1/ix (MVN 13: 222 [P116994]), while Amar-Suen has been dead for at least eighteen months. – At Umma, during Amar-Suen’s reign, why is the month name ezemd amar-dsuen mentioned only in texts of AS6, AS7 and AS8, with a peak in AS7 and a complete gap in AS9 (min3-eš3 is used in its place)? Then this month name of Amar-Suen reappears, in each of the years ŠS1, ŠS2 and ŠS5.45 – When trying to sort in chronological order so many text groups in the administrative archives of that period, especially at Girsu, Umma, or Drehem, recurrent anomalies and disruptions appear for years AS8 and AS9. Thus for example, when having a look at the charts of W. Sallaberger’s (1993) book on Ur III cultic calendars, we observe that the ešeš festival is well attested every year between Š44 and ŠS3, except for the period AS8-ŠS2 (UAVA 7/2: 12–13, Table 6). The situation is more or less the same at Umma for the sa2-du11 dšara2 (UAVA 7/2: 47, Table 23) and for other festivals (UAVA 7/2: 50–52, Table 25 and 27), or at Drehem for the geškiri6-mah festival (UAVA 7/2: 63, Table 31). And we just saw a similar situation for the running accounts of livestock from Umma (see above n. 25).

names between AS8 and ŠS2 (AAS 81 [P100068]; AnOr 7: 316 [P101611], etc.). Otherwise, R. Englund kindly informs me that, through the request of the regular expression /AmarSuen.+Shu-Suen.01/ in “dates referenced” of the CDLI search window, we can pick up 42 instances of records, most of them from Umma, spanning both rulers (AS and ŠS). On the other hand, for this multi-year accounts and to explain some quirks, P. Steinkeller (private communication) suggests to take into account the practice of ante-dating tablets, with new tablets re-introduced later by the scribes to the existing archives, to update the existing documentation. 44 But an interpretation like “mu AS8 u3 [mu] AS9” would also be possible, with a mu sign written in the broken part of the edge. 45 SAT 3: 1189 [P144389]; OrSP 18, pl. 14 no. 39 [P124883]; Nisaba 3: 123 [P201753].

When Šu-Sin Succeeded Amar-Suen

2.

3.

199

During the period that concerns us, there is a real problem with the “proliferation” of intercalary months in local calendars: why was it necessary, between AS9 and ŠS2, to add annually a supplementary month to the Umma and Drehem calendars? Four intercalary months were added in ten years at Irisagrig, Nippur and Girsu; nine intercalary months in ten years at Umma; seven intercalary months in ten years at Drehem, without any synchronization between these various calendars.46 It is difficult to explain so many “diri” month names during this period, as normal computation would require only one supplementary month roughly every three years to fit the desired synchronism with the seasons. It seems that during AmarSuen’s reign a substantial dislocation accumulated in the local calendars, and that the reforms undertaken in ŠS2–3 were intended to restore seasonal congruence and administrative harmony. How are we to explain the “deification” of Šu-Sin so early in the reign of Amar-Suen? Here is some evidence: – There is a proper name Ur-dšu-dsuen-an-dul3!(= ka) as early as AS3 (Nisaba 24: 37 rev. ii 10 [P375997]; kindly collated for me in June 2015 at the BM by M. Molina). – Statues of dŠu-dsuen are attested as early as AS6/i (Nisaba 5: 231 [BDTNS 071308]). – As early as AS6 several seals are dedicated to the “divine king Šu-Sin” (dšu-dsuen lugal, UTI 3: 1848 [P139864], etc.). – dŠu-dsuen is among a list of deities who receive offerings at Irisagrig in AS7/xii (Nisaba 15: 56 [P453616]). – There may be an allusion to the “crown of divine Šu-Sin” (aga dšud suen-na) at Nippur as early as AS8/viii (TMH NF 1–2: 225 [P134536]).

Would it be therefore possible that Šu-Sin was, after Šulgi’s death, the real prospective king,47 but that, too young to rule, he had to wait for some years before he could be enthroned? 4. If there were murders, rivalry and competition between Ur-Namma’s various successors, why was there no post mortem effort to obliterate the memory of the dead king when each new one came to the throne? All the deceased former sovereigns seem to have been permanently worshiped in the same way, starting immediately with the accession of the successor.

46 See chart 6 in Owen 2013a: 68; Owen 2013b: 91. See also Wu 2002. 47 He was therefore probably a son of Šulgi and Abi-Simti. See above n. 7; Sallaberger 2012; Wu 2012; Owen 2000, where the proper name dŠu-Sin-walid-dŠulgi (“Divine-Šu-Sin-engendered-by-divine-Šulgi”) is found on a tablet dated to ŠS4 (to be published in Nisaba 30).

200

Bertrand Lafont

Thus, there was no damnatio memoriae for Šulgi, neither for Amar-Suen: as soon as Šu-Sin ascended the throne, funerary offerings were delivered to his predecessor,48 and personnel were assigned to the funeral chapels (e2) of Ur-Namma, Šulgi and Amar-Suen.49 The two e2 damar-dsuen established at Umma and Girsu still exist three years after Šu-Sin’s accession and the month name ezem damar-dsuen is used at Umma until ŠS5. And a few years later, early in the reign of Ibbi-Sin, we observe that the last king of Ur continues to honor in his capital the memory of his four predecessors.50 As noted by P. Michalowski (2013a: 187), “Ur III royal writings offer not a single allusion to any dissident …” In conclusion, here is a brief summary, in chronological order, of some of the main events described above, that are precisely dated: AS6 AS6/xii AS7/x AS7/xi

AS7–9 AS8/iv

AS8/viii AS9/ii AS9

Beginning of the military campaign against Huhnuri. First occurrence of the gar3-du at Drehem. First occurrences of some seals dedicated to the “divine king Šu-Sin”. All the generals of the realm are gathered at Ur to swear a loyalty oath. Arad-Nanna, the sukkalmah, is appointed as ensi of Girsu. At Irisgarig, for a bit more than one year, there are two new ensis, but in AS9/i, their predecessor is back in office. Peak of administrative activity and of messenger texts at Irisagrig. New sangas at the head of the main institutional households of Lagaš; they have to take an oath before the sukkalmah. Almost daily shipments of tens of boats from Girsu to Nippur in the frame of the bala, and assignment of personnel from Girsu, notably for worship of the former deceased kings. A new ensi is appointed at Umma, who starts using two different seals, one dedicated to Amar-Suen, the other to Šu-Sin. At that date, it is quite sure that Amar-Suen is dead. At Drehem, end of the gar3-du. At the main offices in Nippur, return of members of the Ur-meme’s family.

48 SAT 3: 1197 [P144397] dated to ŠS1/iv. 49 Above n. 32. 50 OrSP 47–49: 465 [P125354], and mention of the ki-a-nag lugal 4-ba in UET 3: 242 [P136559], dated to IS7.

When Šu-Sin Succeeded Amar-Suen

201

Can this chronological chart help us to understand what happened in the kingdom of Ur at this tumultuous time? According to P. Michalowski (2013b: 311), “Šu-Sin returned to Sumer from his post at Der (…) towards the end or in the aftermath of the Huhnuri campaign of Amar-Sin’s sixth year, when the king was probably absent from his land.” From that moment on, a long period of indecision and disturbance started all across the country. Could this be attributed to the fact that Amar-Suen was far from home and that his absence lasted longer than expected? He may have been seriously injured, captured, or killed in action, without any possibility, in Sumer, to get precise information on his real situation. The kingdom would then have experienced a prolonged period of uncertainty. However, there are several clues that lead us to consider another scenario in which Amar-Suen did return to Sumer. The drama seems then to have played out precisely during the two years and three months between AS6/xii and AS9/ii, when the gar3-du damar-dsuen were present at Drehem, exactly the period when many administrative archives display significant anomalies and instability. This could correspond to a time when Amar-Suen was back in his kingdom but no longer able to rule effectively. Did he return to Nippur seriously wounded or ill from the campaign against Huhnuri?51 Some may have considered it certain that the king would not recover or would not be able to return to power, and thus immediately recognized his successor, while others preferred to wait before rallying.52 There is no reason to envision fratricidal conflict or competition between the two kings. In any case, (the young?) ŠuSin began gradually to exercise power from that time. By AS9/ii at the latest, Amar-Suen had died, but it still took several months before Šu-Sin could impose his authority definitively.

51 Note that at Drehem, a tablet dated AS8/x (AUCT 2: 200 [P104018]) mentions the delivery of “six mattresses for the king’s bed” (6 ša3-tuku5 gešna2 lugal). The king nevertheless continued to move (or to be moved!) during this period. In AS8/iv, he (was) embarked well on a boat to Šuruppak (PDT 1: 461 [P125877]), and a bit later (AS8/vi) he traveled with his gar3-du to Uruk (Ontario 1: 81 [P124494]), then to Ur (CTNMC 5 [P108736]). But as we have seen, nothing is known about his demise. 52 Would it be possible that the twenty Girsu texts for personnel assignments (Lafont 1994: 107–109 and above n. 32) and the thirty texts for boat charters (Lafont 1994: 110–11 and above n. 30), that are divided between AS8 and ŠS1 (mostly from AS8), and that mention the worship of the former deceased kings (to whom Amar-Suen is finally added), could have been linked to rituals, ceremonies and musters pending Amar-Suen’s possible healing, before he finally died?

202

Bertrand Lafont

Bibliography Allred, Lance B. 2006. Cooks and kitchens: Centralized food production in Late III rd Millennium Mesopotamia. Unpublished Ph.D. Dissertation. Johns Hopkins, Baltimore. Allred, Lance B. 2008. Labor Assignments from the City of Girsu. Pp. 11–19 in On the Third Dynasty of Ur: Studies in Honor of Marcel Sigrist, ed. Piotr Michalowski. The JCS Supplemental Series 1. Boston: ASOR. Allred, Lance B. 2010. More Šu-Suen Seals During the Reign of Amar-Suen. CDLN 2010:3. Allred, Lance B. 2013. The Tenure of Provincial Governors: Some Observations. Pp. 115–24 in From the 21 st Century B.C. to the 21 st Century A. D. Proceedings of the International Conference on Sumerian Studies Held in Madrid 22–24 July 2010, ed. Steven J. Garfinkle and Manuel Molina. Winona Lake: Eisenbrauns. Archi, Alfonso, Francesco Pomponio, and Giovanni Bergamini. 1995. Testi Cuneiformi Neosumerici da Umma, Parte I, Catalogo del Museo Egizio di Torino, Serie seconda VIII. Torino: Soprintendenza al Museo delle Antichita’ Egizie. Dahl, Jacob L. 2004. The Quest for Eternity. Pp. 117–36 in Assyria and Beyond. Studies presented to Mogens Trolle Larsen, ed. Jan G. Dercksen. PIHANS 100. Leiden: NINO. Dahl, Jacob L. 2007. The Ruling Family of Ur III Umma: A Prosopographical Analysis of an Elite Family in Southern Iraq 4000 Years Ago. PIHANS 108. Leiden: NINO. Dahl, Jacob L. 2010. Naming Ur III Years. Pp. 85–93 in Why Should Someone Who knows Something Conceal it? Cuneiform Studies in Honor of David I. Owen on His 70 th Birthday, ed. A. Kleirnermann, and J. M. Sasson. Bethesda, MD: CDL Press. Di Pasquale, Emanuela. 2002. La fine del regno di Amar-Suena e i “testi dei messaggeri” di Umma. NABU 2002/18. Frayne, Douglas R. 1997. Ur III Period (2112–2004 BC). RIME 3/2. Toronto : The University of Toronto. Heimpel, Wolfgang. 1993. Zu den Bezeichnungen von Schafen und Ziegen in den Drehemund Ummatexten. Pp. 115–60 in Domestic Animals of Mesopotamia, Part I, ed. J. N. Postgate, and M. A. Powell. Bulletin of Sumerian Agriculture VII. Cambridge: Sumerian Agriculture Group. Heimpel, Wolfgang. 1997. Disposition of Households of Officials in Ur III and Mari. ASJ 19: 63–82. Lafont, Bertrand. 1994. L’avènement de Shu-Sin. RA 88: 97–119. Lafont, Bertrand. 2010. Sur quelques dossiers des archives de Girsu. Pp. 167–79 in Why Should Someone Who Knows Something Conceal It? Cuneiform Studies in Honor of David I. Owen on His 70 th Birthday, eds. A. Kleirnermann, and J. M. Sasson. Bethesda, MD: CDL Press. Maekawa, Kazuya. 1996a. Confiscation of Private Properties in the Ur III Period: A Study of é-dul-la and níg-GA. ASJ 18: 103–68. Maekawa, Kazuya. 1996b. The Governor’s Family and the ‘Temple Households’ in Ur III Girsu. Pp. 171–79 in House and Households in Ancient Mesopotamia: Papers Read at the 40 e Rencontre Assyriologique Internationale, Leiden July 5–8, 1993, ed. Klaas R. Veenhof. PIHANS 78. Leiden: NINO. Maekawa, Kazuya. 1997. Confiscation of Private Properties in the Ur III Period: A Study of é-dul-la and níg-GA (2). Supplement. ASJ 19: 273–91. Michalowski, Piotr. 1977. Amar-Su’ena and the Historical Tradition. Pp. 155–57 in Essays on the Ancient Near East in Memory of Jacob Joel Finkelstein, ed. Maria de J. Ellis. Hamden: Archon.

When Šu-Sin Succeeded Amar-Suen

203

Michalowski, Piotr. 2011. The Correspondence of the Kings of Ur, Mesopotamian Civilizations 15. Winona Lake, IN: Eisenbrauns. Michalowski, Piotr. 2013a. The Third Dynasty of Ur and the Limits of State Power in Early Mesopotamia. Pp. 169–206 in From the 21 st Century B.C. to the 21 st Century A.D.: Proceedings of the International Conference on Sumerian Studies Held in Madrid 22–24 July 2010, ed. Steven J. Garfinkle and Manuel Molina. Winona Lake, IN: Eisenbrauns. Michalowski, Piotr. 2013b. Of Bears and Men. Thoughts on the End of Šulgi’s Reign and on the Ensuing Succession. Pp. 285–320 in Literature as Politics, Politics as Literature: Essays on the Ancient Near East in Honor of Peter Machinist, ed. David S. Vanderhooft and Abraham Winitzer. Winona Lake, IN: Eisenbrauns. Molina, Manuel. In press. Cases on Malpractice by Provincial Officers at Umma. Nasrabadi, Behzad Mofidi. 2005. Eine Steininschrift des Amar-Suena aus Tappeh Bormi (Iran). ZA 95: 161–71. Notizia, Palmiro. 2009. I testi dei messaggeri da Girsu-Lagaš della Terza Dinastia di Ur, Nisaba 22. Messina: Di.Sc.A.M. Notizia, Palmiro and Francesco Pomponio. 2008. Sui messenger texts e sui testi-še-ur5-ra di Umma e di Girsu. NABU 2008/60. Ouyang, Xiaoli. 2013. Monetary Role of Silver and Its Administration in Mesopotamia during the Ur III Period (c. 2112–2004 BCE): A Case Study of the Umma Province, BPOA 11. Madrid: CSIC. Owen, David I. 2000. On the Patronymy of Shu-Suen. NABU 2000/82: 90. Owen, David I. 2013a. Cuneiform Texts Primarily from Iri-Saĝ rig/Āl-Šarrākī and the History of the Ur III Period, Nisaba 15/1. Bethesda, MD: CDL Press. Owen, David I. 2013b. The Archive of Iri-Saĝrig / Āl-Šarrākī. Pp. 89–103 in From the 21 st Century B.C. to the 21 st Century A. D. Proceedings of the International Conference on Sumerian Studies Held in Madrid 22–24 July 2010, ed. Steven J. Garfinkle and Manuel Molina. Winona Lake, IN: Eisenbrauns. Pomponio, Francesco. 1990. Le Sventure di Amar-Suena. SEL 7: 1–14. Pomponio, Francesco. 2013. The Ur III Administration: Workers, Messengers, and Sons. Pp. 221–32 in From the 21 st Century B.C. to the 21 st Century A.D.: Proceedings of the International Conference on Sumerian Studies Held in Madrid 22–24 July 2010, ed. Steven J. Garfinkle and M. Molina. Winona Lake, IN: Eisenbrauns. Pomponio, Francesco and Palmiro Notizia. 2006. I messenger texts: la più numerosa categoria di testi nella documentazione neosumerica. Pp. 175–90 in L’ufficio e il documento. I luoghi, i modi, gli strumenti dell’amministrazione in Egitto e nel Vicino Oriente antico, ed. Clelia Mora and Patrizia Piacentini. Quaderni di Acme 83. Milano: CISALPINO Istituto Editoriale Universitario. Sallaberger, Walther. 1993. Der Kultische Kalender der Ur III-Zeit, UAVA 7/1 und 7/2. Berlin: De Gruyter. Sallaberger, Walther. 2012. “Šulgi” and “Šu-Suen von Ur”. RlA 13: 270–80 and 362–65. Sallaberger, Walther and Åge Westenholz. 1999. Mesopotamien. Akkade-Zeit und Ur III-Zeit. Orbis Biblicus et Orientalis 160/3. Freiburg & Göttingen: Universitätsverlag. Sharlach, Tonia M. 2004. Provincial Taxation and the Ur III State, CM 26. Leiden: Brill/Styx. Sigrist, Marcel. 1992. Drehem. Bethesda, MD: CDL Press. Steinkeller, Piotr. 2008. Joys of Cooking in Ur III Babylonia. Pp. 185–92 in On the Third Dynasty of Ur. Studies in Honor of Marcel Sigrist, ed. Piotr Michalowski. The JCS Supplemental Series 1. Boston: ASOR.

204

Bertrand Lafont

Stepien, Marek. 1996. Animal Husbandry in the Ancient Near East. A Prosopographic Study of Third-Millenium Umma. Bethesda, MD: CDL Press. Stepien, Marek. 2012. The Economic Status of Governors in Ur III Times: An Example of the Governor of Umma. JCS 64: 17–30. Tsouparopoulou, Christina. 2013. A Reconstruction of the Puzriš-Dagan Central Livestock Agency. CDLJ 2013:2. Uchitel, Alexander. 1996. Erín-èš-didli (II): Pattern of Conscription and Work Assignment during the Years AS 8–ŠS 1. ASJ 18: 217–28. Vacin, Ludek. 2011. Šulgi of Ur: Life, Deeds, Ideology and Legacy of a Mesopotamian Ruler as Reflected Primarily in Literary Texts. Unpublished Ph.D. Dissertation, University of London. Waetzoldt, Hartmut. 1970–71. Thronfolger auch Mitregent? Mesopotamia 5–6: 321–23. Wu, Yuhong. 2002. The calender synchronization and intercalary months in Umma, PuzrišDagan, Nippur, Lagaš and Ur during the Ur III period. JAC 17: 113–34. Wu, Yuhong. 2012. The Identifications of Šulgi-simti, Wife of Šulgi, with Abi-simti, Mother of Amar-Sin and Šu-Sin, and of Ur-Sin, the Crown Prince, with Amar-Sin. JAC 27: 1–27. Zettler, Richard L. 1992. The Ur III Temple of Inanna at Nippur, Berliner Beiträge zum Vorderen Orient 11. Berlin: Dietrich Reimer Verlag.

Piotr Michalowski

Literary Journeys from Babylonia to Assyria: Second Millennium Copies of a Bilingual Poem Concerning Ninurta For much of my adult life, Miguel Civil has been a mentor, colleague and dear friend, whose humane wide-ranging, erudite, and intellectually penetrating approach to scholarship continues to serve as a model for what one should strive to be. I can only hope that he will find this modest contribution in his honor of some interest. Over the last few decades the study of Sumerian literature has focused primarily, although not exclusively, on the abundant materials unearthed from Old Babylonian teaching establishments, with special attention to the numerous examples that come from the city of Nippur. This is in no small part due to the often-unacknowledged role of Miguel Civil’s magisterial catalog of Sumerian literary texts, which even though it remains unpublished, has served as the basis for the information utilized by the Oxford University based Electronic Text Corpus of Sumerian Literature (ETCSL).1 Much of this early second millennium literary material was preserved as a result of dramatic events that took place during the reign of the Babylonian king Samsu-iluna (ca. 1750–1712 BC), although some of the surviving Old Babylonian documentation from Uruk and elsewhere is somewhat older. It is now also evident that small communities of people lived in Nippur and other southern Babylonian cities after Samsuiluna’s reign (George 2009: 136–42; Van Lerberghe and Voet 2009: 1–7) and that literary activity continued in the ensuing centuries, albeit conducted by small numbers of scholars and teachers, even while literary and scholarly teaching activities continued unabated in northern cities such as Sippar, Babylon and Kish.2 The political, military and cultural aspects of the transition from Late Old Babylonian times to the Middle Babylonian period are slowly becoming better articulated in the documentary record, although much of this remains opaque

1 http://etcsl.orinst.ox.ac.uk/. 2 The complex matters of the posited depopulation of southern Mesopotamia in the wake events from Samsu-iluna’s reign are subject to much revision and it is becoming clear that the area was not deserted in a manner posited only a few years ago and that some of our analysis has taken social changes as traces of abandonment. The incomplete and sometimes contradictory evidence on these issues is discussed by Tenney (in press a & b). https://doi.org/10.1515/9781501503696-013

206

Piotr Michalowski

(Van Koppen 2010). The ensuing Middle Babylonian period has been largely viewed as a time when undocumented changes took place in Mesopotamian literary repertoires but scholars have had to rely in large part on literature exported to and adapted in regions outside of Mesopotamia proper, documented in archives of the Hittite capital at Hattuša, from Tell el-Amarna in Egypt and from Levantine Ugarit and Emar. Recent attention to Middle Babylonian literary activity in Mesopotamia proper, in Sumerian and Akkadian as well as in the form of bilingual texts, has revealed a good degree of continuity between Old and Middle Babylonian traditions, accompanied by new developments, by the pruning of the tradition and addition of new compositions. The surveys, new text exemplars and identifications by Veldhuis (2000), Sassmannshausen (2008) and Viano (2009), to which we now must add the informative study of Peterson (this volume) have changed the way in which we view the transmission of Sumerian learning during Middle Babylonian times. The new scholarly attention to this period is also enriched by Niek Veldhuis’ new on-line project that aims to present a Corpus of Kassite Sumerian Texts (CKST).3 Important Middle Babylonian texts excavated at Babylon itself remain unpublished (Pedersén 2005, 2011) and Alexa Bartelmus (n. d.) will soon make some of these available. As stressed by Peterson, it is now apparent that many more elements of the older traditions were still known during this period than had been suspected; moreover, the Kassite period scribes were particularly interested in certain deities, including Ištar and Ninurta. His succinct study of the issue describes well the state of our knowledge on the matter and will not be recapitulated here. While these new studies have provided much new insight into the literary repertoire of second millennium Mesopotamia, we are still very much in the dark concerning the arrest and movement of various strands of the traditions. Two historical moments stand out with particular force: the reinvigoration of copying and writing Sumerian during the reign of the fourteenth century Babylonian ruler Kurigalzu I and his successors (Clayden 1996, Veldhuis 2008, Bartelmus 2010, Bartelmus and Schmitt 2014: 88) and the destruction of Babylon by the forces of the Assyrian king Tukulti-Ninurta I in 1225 BC with the ensuing plunder of the city with booty taken back to Assyria that included literary cuneiform tablets and even possibly inscribed monuments (Lambert 1957–58: 44– 45,4 Bartelmus and Schmitt 2014). The rule of Kurigalzu I was not only a time 3 http://oracc.museum.upenn.edu/ckst/. 4 The broken passage in the Tukulti-Ninurta Epic mentions, among other texts taken back from Babylon, the repertoire of the exorcist (ašitūtu), of the diviner (barūtu) and ritual laments (eršahunga’s).

Literary Journeys from Babylonia to Assyria

207

of cultural revival, but was also a singular moment in the ideological landscape of Kassite kingship because some later rulers imagined him as the founder of their lineage (Brinkman 1969: 326, Paulus 2014: 69). Likewise, the reign of Tukulti-Ninurta I was not only characterized by military expansion, but also by literary creativity in Sumerian as well as Akkadian (Machinist 1976, Lambert 1976); indeed this may have been part of a broader cultural program aimed at shifting the concept of Babylon as the axis point of the universe to Assur and Assyria (Maul 1997: 122). Here I present an edition of two post-Old Babylonian manuscripts of a bilingual poem concerning the god Ninurta. The text, if this is indeed one composition, remains fragmentary, but is known from two versions, one from Middle Assyrian Assur and the other one from Kassite Nippur. The exact find spots of both tablets are unknown.5

Sources As = VAT 10610 (KAR 119, Lambert [1960] pl. 32, fig. 1).6 Ni = N 3462 (P278493, fig. 2).7 The Assur text was transliterated and translated in modern times by van Dijk (1953: 114–18) and Lambert (1960: 118–20); these editions served as the basis for Foster’s (1993: 616–17) modern English translation. The text is bilingual, with the Akkadian text in somewhat smaller hand under the Sumerian, it is

5 A good number of people have generously aided my work on these tablets. I am grateful to Steve Tinney, current associate curator of the Babylonian Section of the Penn Museum, and the late Åke W. Sjöberg, the former curator, for permission to publish N 3462 (and to Steve for sending me his own photographs). I must thank Jeremiah Peterson for kindly providing his photographs of this tablet for publication in fig. 2 (and for comments on a draft of this article); to Stefan Maul, who collated the Assur source on an unpublished photograph; to Juliane Eule for collation of a passage cited in the commentary and measurements of As; to Philip Jones for measurements of Ni, and to Markus Hilgert, director, and especially Airun Gutow of the museum photo archive for arranging photography in the Vorderasiatische Museum, Berlin; to Olaf M. Teßmer for taking the excellent photos; and to Martin Lutz, deputy director of the Museum, for official permission to publish them here. Finally, I must thank Alexa Bartelmus and Jonathan Tenney for comments on a final draft and for generously providing me with copies of their unpublished manuscripts. I should note that I only had access to Veldhuis (2014) after the basic text of this article was already completed. 6 58.42 mm × 73.66 mm × 22.86 mm (height/width/depth). 7 62 mm × 58 mm × 28 mm (height/width/depth).

208

Piotr Michalowski

broken on the top and bottom and thus it is impossible to establish how many lines may be missing and the relationship between the passages on the two sides of the tablet are equally difficult to gauge. Although Lambert marked the two sections as “Obverse” and “Reverse,” Ebeling’s copy in KAR indicates that he was unclear about which side was the obverse and which the reverse. Indeed, van Dijk (1953: 116) proposed, with due caution, that the “obverse” and “reverse” should be reversed to make better sense, but collation of the tablet and the evidence from the new Nippur manuscript provide evidence that van Dijk’s arrangement may not be preferable. It is difficult to recover the find spot of the tablet in the city of Assur. Weidner (1952–53: 200) noted that the Assur number was lost but listed it as part of his reconstructed “Library of Tukulti-Ninurta I,” and it has been cited as such in much subsequent literature. I have not been able to trace any further information on the tablet on any of the more recent writings on the Assur archives; presumably it was found in one of the Middle or Neo-Assyrian tablet collections succinctly described by Pedersén (1998: 83–84), possibly brought to Assur around 1300 BCE, half a century or so before the Assyrian raid on Babylon. As can be readily ascertained from the photographs published here, the tablet was written in a decidedly archaizing cuneiform hand. The format is interlinear; the Akkadian translations are written in an ever so slightly smaller hand and some of these lines are indented. Lambert (1960: 118) followed Weidner (1952–53: 200), who thought that it had been imported from Babylon in Middle Babylonian times, or was a local copy of such a tablet. He even went so far as to suggest that it might have been part of the loot obtained by TukultiNinurta’s armies when they plundered tablets from the city. The fragment from Nippur seems to be the only duplicate currently known. I came across it over three decades ago, but then discovered that it had already been identified by Miguel Civil, who listed it in his unpublished catalog of Sumerian literary texts under the label “4.278, Ninurta I.” Over the years I have had many opportunities to approach Miguel with what I thought was a new discovery only have him mutter – through teeth clutching pipe – “very nice, but we already knew that.” And so it goes, because he really does know everything. Like its Assur counterpart, the Nippur piece is bilingual, of similar layout, and is frustratingly broken so that it only adds a small increase to our knowledge of the composition or compositions that it bears. The dating is uncertain and is based on paleography alone, but it is most likely Middle Babylonian, or perhaps even early Neo-Babylonian, written with archaizing features.8

8 The existence of the Penn tablet – as noted in Civil’s catalog – has since been signaled by Viano (2009: 120) and Peterson (this volume), as well as in the relevant online materials.

Literary Journeys from Babylonia to Assyria

209

The Akkadian translation on the obverse is interlinear in slightly smaller script, with a marked empty space breaking up each line in two. This is a format well attested in Middle Babylonian bilingual literary tablets, last described by Peterson (this volume). Because the beginnings of all the lines are broken off it is impossible to ascertain if the Akkadian lines were indented as well. The tablet N 3462 is broken on top and at the bottom; the preserved text on the obverse probably begins halfway to the break, which is roughly in the middle of the tablet.9 I therefore assume that roughly ten lines are missing at the beginning and twenty more or so missing at the bottom. Thus the complete text may have had close to forty lines on the obverse. The reverse is a more complicated matter. Unlike the bilingual obverse, the preserved lines on the reverse are in Sumerian only. Most important, this section addresses Ninurta directly, while Part A on the obverse references the deity in the third person. It is therefore impossible to ascertain whether we are dealing with one composition or two. The Nippur piece, although it adds only small amounts of new material, helps to clarify the structure of section A. However the text may have begun, it includes a third person hymnic passage in honor of Ninurta. Then follows a section of ethical transgressions; the only remote thematic parallels to this are the Old Babylonian “proverbs” that describe various human foibles that are niĝ2-gig of Ninurta, Utu, Sin, Inana and someone’s personal god.10 Hallo (1985) surveyed these passages and proposed the word “taboo” as a rendering of niĝ2-gig, an interpretation questioned by Geller (1990a). Of the various meanings proposed by the latter, “anathema” probably best fits the context of these “proverbs.” Section B includes a description of a procession of Ninurta into the city of Nippur, where he was welcomed with music and athletic performances. He entered through the most northern of the western gates to the city, the one named Gate of the Unclean (abula u2-zug-e-ne), and then moved on through the streets to his temple Ešumeša, where he was greeted by his beloved spouse, presumably Nin-Nibru, whose name would have been in the first or second line after the final break. The text is quite specific here, providing topographical information on the route taken by Ninurta as he made his way into the city and to his temple Ešumeša.11 The gate named here was one of the main gates of Nippur (Komoróczy 1976. Behrens 1978: 171–173), clearly marked on the Middle

9 The scribe was not in the best of form. There are some infelicities, notably a-ra-am-šu in A 4′ and kar-ar-ṣí in A 5′. 10 I should note that J. Peterson also noted these analogies in his thoughts on this text. 11 On this whole passage see Steinert (2014: 152).

210

Piotr Michalowski

Babylonian “map of Nippur.”12 The gate opened onto a quay on the Euphrates and therefore Ninurta and his entourage must have arrived at the city by boat; therefore this cannot refer to the poetic narrative of Angim that has the god entering Nippur on a chariot. For an ancient reader, this description of Ninurta’s entry into the city would have contrasted strikingly with the one he would have known from Angim, which was filled with ominous descriptions of the hero’s martial prowess and marked by dangerous tension between the god and his father Enlil. It has even been suggested that the procession described in the tablet edited here was part of complex Akītu celebrations in the second month of the year that commemorated Ninurta’s return from battles in the east recounted in Lugale (Annus 2002: 62–64, Emelianov 2013: 286). The lynchpin of this discussion is a passage in the Middle Babylonian Astrolabe B, wherein the second month of Ayaru is described as the month of Ningirsu; here the divine name harks back to the original name of the hero of Lugale, according to Horowitz (2010: 184). But the wording of this entry describes Ninurta/Ningirsu in his role as farmer, not in his martial guise (Cohen 1993: 310). The much later OECT 11, 69+70, an expository text that provides mythological explanations for elements of the cultic year in Nippur, provides some details on the second month Akītu festivities in that city. George (1990: 157–160), who was first to recognize the character and implications of this composition, observed that it provided evidence that the 15th day of the month was thought to commemorate the moment when Ninurta came home to the Ešumeša from his victories abroad. Moreover, he noted that just earlier, on the 18th day, the cultically impure women, the musuk-ka-ti, went out, “because Ninurta entered the Ešumeša in his anger,” as the text explains. George (1990: 158) went further and wrote: “no doubt the ritual procession of these ladies gave its name to the city gate of Nippur known as the Gate of the Impure,” and proceeds to cite our text, among others. While it is tempting to link our text with these celebrations in the manner of authors cited above, I am reluctant to do so. Granted that the associations between the going out of “impure women” and the entry of Ninurta through the Gate of the Impure are suggestive, the atmosphere described in of our text, with Ninurta proceeding in joy is different from the rituals in which the god is always angry and wrathful. The sources on these Akītu celebrations come from different times and places and it is difficult to paint a coherent picture of a specific set of rituals practiced at particular moments in time so there is little to suggest that the text under consideration here is linked to any of them.

12 See Maic (1976), Oelsner and Stein (2011) and the overlay of the reconstructed version over a topographical map of the city in Gibson (1993: 12), based on calculations by M. Civil.

Literary Journeys from Babylonia to Assyria

211

Perhaps more to the point, the topographical details of Ninurta’s entry into Nippur and the procession to his temple Ešumeša described in Part B of the composition under consideration here may have resonated in a particular manner in the city of Assur, where manuscript A was discovered. The prominence of Ninurta in Middle Babylonian times is well known; he was much celebrated in literature and art in both Babylonia and Assyria (Hallo 1981, Annus 2001, 2002, Streck 2001, Pongratz-Leisten 2001: 224–30, Collon 2006). Scribes and scholars of the second and first millennia were familiar with his role as the main protagonist of two major literary texts, originally in Sumerian and then in bilingual form, Lugale (van Dijk 1983) and Angim (Cooper 1978), but he also had an important role in various royal and temple cults.13 Kurigalzu I built a major temple for Ninurta, E-saĝ-diĝir/dimmir-ene in his newly established capital of Dur-Kurigalzu in a new sector that was part of the expansion beyond the older city of Parsa (Nashef 1983), which underlay the new city. This temple was feted in the already mentioned Sumerian Emesal composition in honor of Ninurta (KAR 97). Surprisingly, two copies of a dedicatory inscription now reveal there was apparently a second Ninurta temple or chapel by the same name in Nippur, wherein an unnamed enemy, using allies drawn from Der, committed a massacre during the reign of Kurigalzu II, half a century or so after the first king of that name had restored many of the old cult sites of Babylonia and built Dur-Kurigalzu (George 2011, 2012). As observed by Maul (1997: 121, 1998: 191–92) and George (1990: 157), among others, the theological equation of the god Assur with Enlil led to the identification of Assur as a new Nippur with many of the processional streets and temples using the same names as in the old Sumerian city. Thus Ninurta’s role as Enlil’s son conferred him particular prominence and his temple in Assur was named Ešumeša, as in Nippur. In this context it is instructive to observe that Enlil, and by extention his wife and son, was also singularly worshipped in Kassite Babylonia and when Kurigalzu I in the fourteenth century built his new residence at Dur-Kurigalzu, the main ziggurat belonged to Enlil, perhaps already linked to the identification of the newly elevated Marduk with the god of Nippur, which would not be surprising given that there is now evidence for the celebration of the Akītu festival of Marduk in Nippur and Babylon as early as the thirteenth century (Tenney 2016). These parallel rival developments in Babylonia and Assyria are undoubtedly important for the understanding of the role of Enlil, Ninurta and Nippur in the literary traditions of the later half of

13 Both of these major compositions had resonance in other text types, e.g. the balaĝ that began nir-ĝal2 lu2 e3-ne that includes a generically unusual hymn of praise to Ninurta and allusions to Lugale and Angim; see Maul (2005: 31).

212

Piotr Michalowski

the second millennium and the concomitant tensions that must have accompanied the competing claims of Babylonia and Assyria over the imagined visions of early Mesopotamian cultural achievements. It is difficult to find criteria that would be helpful in establishing the date of composition of the text or texts represented by these two tablets. The fact that no earlier versions have been recovered to date has led to the suggestion that it was perhaps a Middle Babylonian period composition (e.g. Peterson, this volume). The one characteristic that stands out is the consistent use of rare words, attested mainly in the Post-OB lexical tradition and in OB literary compositions that were known in later times, as I point out in the commentary below. This may be a clue to the date of composition of the Ninurta tablets, but such reliance on lexical material in the construction of poetry is attested already during the Old Babylonian period, as perceptibly observed by Civil (1987) and more recently elaborated by George (2009: 107–11) in a discussion of what he describes as Academic Sumerian. The date of composition is important for understanding the history of the literary Sumerian language because there is little evidence that a substantial amount of new poetic works in the main “dialect” were created after the Old Babylonian period. To be sure, older texts were copied, provided with Akkadian translations and sometimes reworked, such as the bilingual version of Ninisina’s Journey to Nippur, documented by two Middle Assyrian manuscripts from Assur and a monolingual version from Old Babylonian Nippur (Wagensonner 2008) or the Ninurta hymn fragment from Neo-Assyrian Nineveh that is likewise known from an Old Babylonian tablet from Nippur (Cohen 1975a). But outside of the lexical tradition, incantations and a growing number of Babylonian royal inscriptions (Bartelmus 2006, Paulus 2011) as well as scientific texts such as Astrolabe B (Horowitz 2010), new mid-second millennium Sumerian writings utilized both the standard version, but also, increasingly, the Emesal liturgical version of the language. When a new Ninurta hymn celebrating his temple in Dur-Kurigalzu was composed (KAR 97; Ebeling 1917: 75–78; see Falkenstein 1953: 2; Viano 2009: 118), it was not in monolingual poetic Sumerian, but in Emesal alongside an Akkadian version, which may have served as the original form of the text. The same was true when bilingual prayers for Tukulti-Ninurta I were composed in Assyria (KAR 128 [+] 129, Viano 2009: 124–25).14 According to Lambert (1976: 86), the scribe who copied or com14 Lambert (1976: 86) cites them as KAR 118 and 119; see Hallo (1981: 254); Yamada (1994: 12 n. 4). In all likelihood this is more than a simple error; Lambert undoubtedly also had in mind Ninisina’s Journey to Nippur (KAR 118/119 [Cohen 1975b: 609–11, Wagensonner 2008, with OB duplicate from Nippur]), written by the same scribe. On these poems, see now Veldhuis (2014: 347–53).

Literary Journeys from Babylonia to Assyria

213

posed this royal text may also have worked on another bilingual, a first person composition in the name of the Assyrian conqueror of Babylon – a rare datable post-Old Babylonian composition in the standard Sumerian dialect (BM 98496, see below). The choice of the non-liturgical version of the language in the TukultiNinurta text published by Lambert was doubtless prompted by the historical, quasi-chronicle-like contents that required scholarly prose. This Middle Assyrian bilingual was written or copied with the two languages in parallel columns, similar to many such texts from Kassite Babylonia. The author used a highly learned middle second millennium form of main dialect Sumerian, derived from lexical texts and other sources, as well as the literary Babylonian version of Akkadian to provide gravitas and decorum to novel Assyrian dynastic claims, relying on older political notions such as the bala, the divinely ordained time of earthly imperial dominion (Michalowski: forthcoming). In his edition of this fascinating text, Lambert surmised that it may have been composed by a scribe from Babylon in service to the Assyrian monarch and, moreover, that the same person may have also composed the Babylonian dialect Tukulti-Ninurta Epic. If so, the rhetorical and linguistic complexity of the bilingual may mirror the intricate literary structure of the “Epic” and its use of traditional Babylonian language mixed with a few Assyrianisms and its structural indebtedness to older Akkadian as well as Sumerian models, described so well by Peter Machinist (1976) forty years ago. The bilingual standard dialect Ninisina’s Journey to Nippur, Lugale and Angim, not to mention various lexical and other scholarly texts, were copied by three brothers in Assur named Marduk-balassu-ereš, Bel-aḫḫa-iddina and Sin-šuma-iddina (Wagensonner 2011). Moreover, Geller (1990b: 212) personally compared the Lugale and Angim tablets (which were written by Mardukbalassu-ereš) with BM 98496, confirming Lambert’s hypothesis that the Tukulti-Ninurta bilingual was from the same hand. Geller concluded that the text may have been composed during the reign of Tiglath-Pileser I (1113–1076 BC), but we now know that the brothers made the copies during the reigns of Ninurta-apil-ekur (1181–1169) and Aššur-dan (1168–1133), half a century or so before the reign of Tiglath-Pileser I (Wagensonner 2011: 650). They did not compose any of this material but only made library copies and were perhaps involved in some form of redactional activities. Moreover, as Veldhuis (2014: 341) noted in a survey of the philological activities of this family, “eleven out of the twenty-three tablets have demonstrable Nippur connection.” In light of this, one could well imagine that the Assur source of the Ninurta composition was brought up from Nippur, not Babylon, as some have suggested. As both Lambert and Viano (2009: 125) have observed, the Sumerian of these texts, and of other Kassite period and Middle Assyrian bilinguals is diffi-

214

Piotr Michalowski

cult and idiosyncratic, to say the least. There are a few main dialect texts known only from late bilingual versions, mostly proverbs, but as Frahm (2010: 170) has noted, it is possible that the post-Old Babylonian traditions derived from a variety of older sources, not just from the school rooms of Nippur that are so familiar to us. The same holds true for MA versions of Angim and Lugale, which do not derive from any known OB manuscripts, even if the colophons indicate they were copied from Nippur originals (Geller 1990b: 211). The colophons of the two MA manuscripts of Ninisina’s Journey to Nippur specify them as copies of originals from Nippur and Babylon (Wagensonner 2008: 286; 2011: 653); other texts written by the three brothers also specify Nippur or Babylon as the source. Therefore one should not be too concerned about the fact that we currently have no older manuscripts of the Ninurta composition(s) edited here; the good Sumerian and the other issues raised above strongly suggest that we are not dealing with new Middle Babylonian or Assyrian material but with copies and redactions of older southern Mesopotamian literature. But when no such information is available, it is more difficult to establish the date of composition of a bilingual that is attested only in a Middle Babylonian source. The best-known example of such a case is the difficult Nippur tablet PBS 1/1, 11, treated not long ago by Joan Westenholz (2005), that includes poetic Sumerian and Akkadian paraphrases in parallel columns and appears to mention the Ur III ruler Šulgi. Although usually cited as a Middle Babylonian creation, Westenholz made a case for dating its origins to Old Babylonian times, even though the Sumerian is difficult and full of unusual forms, an opinion also shared by Peterson (this volume). Veldhuis, however, considers it to have been a rare example of “a new creation” from the Kassite period.15 The intricacies of textual transmission from Old Babylonian to later second millennium times are well illustrated in the case of the menological first part of Astrolabe B.16 The earliest witness is a somewhat imperfect monolingual Sumerian Kassite period tablet from Babylon (VAS 24, 120, Horowitz 2010) with abbreviations and pronunciation glosses; the second is in the opening section of an Assur bilingual manuscript (KAV 218), from the hand of Marduk-balassuereš, mentioned above. It cannot be ascertained at present if this scientific tradition derives from the Old Babylonian period or whether it was a later Babylonian innovation (Horowitz 1998: 157). If the composition is post-Old Babylonian, it would provide one more example of a Kassite period poetic text composed in the standard version of Sumerian. Even though the Babylon text is fragmentary, it is obvious that the two manuscripts differ substantially and

15 http://oracc.museum.upenn.edu/ckst/introduction/index.html. 16 See in general Horowitz (2013); for editions, see Casaburi (2003) and Horowitz (2014).

Literary Journeys from Babylonia to Assyria

215

bear witness to the complexities of textual redaction in the latter part of the second millennium and to the use of non-Emesal Sumerian in scientific texts. The present edition relies on my old transliteration and notes on the Nippur text as well as photographs kindly provided by Jeremiah Peterson (fig. 2), by Steve Tinney and a picture posted on CDLI. I have not been able to collate the Assur tablet in person in Berlin, but Stefan Maul was kind enough to check my transliteration against an old photograph, and subsequently I was able to thoroughly collate the text myself on new high-resolution photos, which are published below (fig. 1).

216

Piotr Michalowski

Obv.

Left edge

Rev.

Right edge

Fig. 1: VAT 10610 (photograph by Olaf M. Teßmer, © Staatliche Museen zu BerlinVorderasiatisches Museum).

Literary Journeys from Babylonia to Assyria

Obv.

Rev.

Fig. 2: N 3462 (photograph by Jeremiah Peterson).

217

218

Piotr Michalowski

Part A. Obverse of Ni and obverse of As 1′.

Ni 1′

[ ]xx[ ] [ ] x te-es-l[i-ti ] ]xx[ ] 2′. Ni 2′ [ ] du-ri2-š[e3 [ ] a-na da-˹ri šu˺-lu-ta mu x-x [ ] 3′. As 1′ [ ] ˹an-gin7˺ […] [ ] x x ˹ṭa-bu˺-[(u) šá] pár-ṣú-šu ˹ki-ma˺ [… ] Ni 3′ [ du10] an-gin7 me si-ig [ ] ˹ṭa˺-a-bu-u ša pa-ar-ṣú-šu ki-ma ša-mé-e el-lum 4′. As 2′ ˹lu2 dam lu2-da˺ na2-a nam-tag-ga ˹dugud˺-[(da)-…] ra-ḫu-ú aš-ti a-wi-lim a-ra-an-šu kab-[tu4] Ni 4′ [ ] nam-tag-ga dugud-da-˹ni?˺ [ ] ša a-ra-am-šu kab-tu4 5′. As 3′ lu2 niĝ2-nu-ĝar-ra ka-bal-e lu2 eme sig-ga ˹gu7˺-[gu7] mu-ta-mu-ú nu-ul-la-a-ti a-kil kar-˹ṣí˺ Ni 5′ [ ] lu2 eme sig-ga gu7-gu7 [ ] a-kil karar-ṣí 6′ As 4′ lu2 gaba-ri eĝir-ra-ni šu hul bi2-in-du3-a ša ar-ki mi-iḫ-ri-šu ú-ba-an li-mut-ti i-tar-ra-˹ṣú˺ Ni 6′ [ ] šu hul bi2-in-du3-a [ ] ú-ba-an ˹li˺-mu-ut-tim i-tar-ra-ṣú 7′. As 5′ lu2 du11-du11-ga nu-me-a im-ri-a-še3 mu-un-šub-ba šá la qá-bi-tam el a-ḫi i-na-ad-du-ú Ni 7′ [ ] im-ri-a-še3 mu-un-šub-ba [ ] i-na-du-u 8′. As 6′ lu2 lu2-mašda2-e nam-gu2 bi2-in-ak-a šá muš-ke-na i-ḫa-ab-bi-lu4 Ni 8′ [ ] nam-gu2 bi2-in-du3-a [ ] i-˹ḫa˺-bi-˹lu˺ 9′. As 7′ lu2 nam-kala-ga si-ga sag-e-eš2 bi2-in-˹rig7-ga˺ en-ša ana da-an-ni i-šar-˹ra-ku˺ Ni 9′ [ ] sag-˹e˺-eš bi2-in-˹rig7˺ [ ] ˹i˺-[šar-ra-k]u 10′. As 8′ a-ša-an-gar3-ra dumu uru-na-ka a2 ˹ib2-ta-an˺-x-x šá ina ta-aš-gir-ti dumu.uru-šu ú-š[a- ] Ni 10′ [ ]xx

Literary Journeys from Babylonia to Assyria

219

aš-dah a-ša3-ga tab-ba-na-ka še ˹zar˺17[-(maš) ba-da-an-kar/zuh?-a] ar-da-du šá ina a.ša3 tap-pi-šu [ ] ] 12′. As 10′ x x x x x ˹ki˺ x šid? x la na[m-bi2-ib[ ]x x[ ] 13′. As 11′ [ ] x[ ] 11′.

As 9′

1′. 2′. 3′.

… [who accepts] su[pplications …] … for ever … votive offering(s) … [Whoever] sleeps with (another) man’s wife--grievous is his offence/ punishment. 5′. Whoever spreads malicious words is a slanderer; 6′. Whoever points at his fellow behind his back with evil intent; 7′. Whoever impugns unspeakable words to his brother; 8′. Whoever oppresses the poor person 9′. And hands over the weak to the strong; 10′. Whoever …’s a (fellow) city citizen with treachery; 11′. A thief, who [takes/steals] grain stacks from his neighbor’s field, 12′. … 13′. … 2′.

4′–5′. 7′.

8′.

10′.

ana dāri may be a clue to post-MB origin or redaction of the Nippur manuscript. The end of line is difficult to read; with caution I interpret part of what remains as šūlûtu, “votive offering.” On these lines and the relationship between them and the lexical tradition, see Böck (1999: 58–59). The Akkadian aḫḫū, “brothers,” as an equivalent of Sumerian im-ria, “family, kin group, etc.,” is unique; on im-ri-a and its Akkadian renditions see Sjöberg 1967: 208–209, Krispijn 2001: 257. If this line is parallel but not contrasting with the following one, the use of mašda2=muškēnu with meaning “poor, destitute,” follows postOB usage. This line is cited, from both sources, by Civil (2013: 38 w. n. 67), who draws attention to “du3 instead of ak, a typical OB Akkadianism.” Both the Sumerian a-ša-an-gar3-ra and Akkadian tašgirtu are rare words; see PSD A/I: 167–68 and CAD T: 286. The Sumerian term occurs only three other times in literary contexts, in (a) Ininšagura: 157, (b)

17 It is difficult to discern the inscribed sign, but it is most probably šum2.

220

11′.

Piotr Michalowski

Enlil and Sud: 21 and (c) Letter from Amar-Sin to Šulgi 1: 10.18 Both (a) and (b) are attested in post-OB versions. Aside from Old Babylonian grammatical texts, the later lexical attestations of the word are derived from (a); see Michalowski 1998. The end of the line is difficult to restore. The authors of PSD A/II: 51 proposed a2 ib2-˹ta˺-a[n-dar?-ra?], from a2…dar, “to cheat someone.” This is an attractive proposal, but it must be kept in mind that the verb a2…dar is hardly attested outside of Ur III dispute resolutions; in later lexical texts it is equated with Akkadian daṣû, which is not attested in the Š stem and is therefore unlikely here. The Nippur text is very damaged at this point and has only the very upper portion of a penultimate sign (which looks like du) and the very end of a final one, which could be E or a different sign that ends with one vertical over another one, such as ˹du-e˺ or ˹dag-dag˺. aš-dah is the rare Sumerian word aš2-dah, “criminal” (Krecher 1970: 218). The Akkadian ardadu occurs only a few times, exclusively in MB and later lexical texts and commentaries. In Commentary A to Aa 16 (MSL 14: 324, l. 42): lagabe-rim = a-a-bi = ar-da-tú. In the first millennium synonym list Malku = šarru 1: 90 (Hrůša 2010: 36) it is one of a number of words explained as sarru, “dishonest, criminal,” and in the corresponding entry in the Murgud source SpTU 2, 53 o I 55′: [aš2]-dah = ardadu = min(sarrum) (Crisostomo 2014: 373) the Sumerian is added. From Middle Babylonian times, there is an interesting entry in an earlier lexical text, in a Boghazköy manuscript of Erimhuš: [a]š2dah = ar-ta-ti-el-lu = ka-aš-ta-an-za (Erimhuš Bogh. A i 44 [MSL 17, 104]). Attinger (1993: 446 n. 1212) views the first explanation as the Akkadian plant name art/dadillu, probably a loan word, a mistake based on the sound similarity. This is possible, but it does not explain the Hittite translation “famine.” Perhaps the origin of the latter misperception at root involves Akkadian arurtu, “drought,” “famine,” falsely associated with arāru, “to curse” (because of Sumerian aš2). Both the Sumerian and Akkadian words appear to be post-OB neologisms or borrowings. In OB Izi 1 185–186 (Crisostomo 2014: 225):

18 (a) Sjöberg (1975: 192), (b) Civil (1983: 51); Attinger (1993: 571–72) and (c) Michalowski (2011: 392), respectively. Note that in (b) the Neo-Assyrian bilingual source from Sultantepe shows that this rare term could be easily misunderstood. In the OB version the goddess Sud warns Enlil that “others have tried to deceive [my mother] and made her angry,” but the late version renders the Sumerian a-ša-an-gar3…ak with pseudo-etymologizing ú-šam-ga-ru, “others will try to get my mother to give her (marriage) consent.”

221

Literary Journeys from Babylonia to Assyria

aš-dah (or just dah) is glossed as rēṣūtum, “help, support,” and arratum, “curse,” respectively. There is obviously some confusion between aš-dah and aš2-bal. zar-(maš) is the literary term for “grain sheave/stack.” As observed by Civil (1994: 91), “the Sumerian terminology for sheaves and stacks is at times confusing.” The term zar (lagabךe/šum2) takes the place of zar3(šum2), known from Ur III Umma, but also of garadin, used in other places, which survives in literary texts as well (Civil 1994: 91). The term zar3, Semiticized as za-ru12-um was already present in the Ebla Sign List 91 (Archi 1987: 97). There is also a similar sign combination known from late lexical texts, written as lagarךe/šum2, that is glossed as lugud4 in Ea III 21 (MSL 14: 303) = raḫiṣu, “trampling down” (so CAD R: 76); maškanu ša tibni, “threshing floor for straw,”19 and as su-hi-rin/rum in Ea III 22a (MSL 14: 304) and Sb Voc. I Rec. B 240 (MSL 3: 116) = tarammu, “grain pile” (the sign does not occur in PEa). Nevertheless, zar-(maš) is not a logogram, as evidenced by Akkadian sarmāšu (CAD S: 177, “an agricultural worker”). The connections between the inscribed lagab and lagar signs, as well as the complex agricultural and animal husbandry connotations of the latter should prove fruitful, but would be out of place here. For MB textual references to people stealing grain stacks, see CAD S: 284 sub sarru A. The line is too damaged for easy restoration.

12′.

Part B. Reverse of As 1′. 2′.

[ [

] -˹i(-)ka˺-[ [ urudu šem3 ˹a2-la2˺ ĝ[iš20balag-di mu-r]a-an-d[u12-du12-ne] ˹ḫa-al˺21-ḫal-la-tu a-˹lu˺-[ú tim-bu-ut-t]u iz-za-am-mu-[ru-ku]

] ]

19 The sign also occurs in the Middle Assyrian unpublished paleographic list VAT 9523 with broken unclear Akkadian (x-gu-u), cited by Landsberger 1968: 135, and in the later Practical Vocabulary from Assur ll. 44–45 (followed by meš), equated with parūgu and tarammu (Landsberger and Gurney 1957–1958: 328). 20 Lambert’s copy (confirmed by collation) has the head of a lower horizontal at the break (not in Ebeling’s copy) that could be the beginning of ĝiš. 21 It seems that the scribe may have erased the first two signs.

222

Piotr Michalowski

 3′. sizkur2 lugal-la gud niga ud[u niga] mu-ra-an-ga[z-gaz-ne] ni-iq šar-rim.meš niga.meš u[du.meš niga-meš] up-tál-la-q[ú-ku]  4′. a2 tuku-bi gešpu2 ˹lirum˺-ma mu-ra-an-ra-r[a-e-n]e22 eṭ-lu-tu be-el e-mu-qí ina ú-ma-ši u a-ba-ri im-taḫ-ḫa-ṣ[ú-ni-ik-ka]  5′. dumu nibruki ildu2-ildu2-ba he2-ĝal2-ta u4 m[u-un-zal-zal-e-n]e dumu.meš ni-ip-pu-ru ina il-la-ti-šu-nu ḫi-in-gál-la uš-[ta-bar-ru]  6′. šir3-zu un sag gi6-ga me-teš2 im-i-i-[ne] zi-im-ri-ka ni-šu ṣal-mat qaq-qa-di ut-ta--`-a-d[a]  7′. ki-bi-ta igi-zu ĝar-ra-[zu-ne] iš-tu aš-ri šu-a-tu4 pa-ni-ka ina ša-ka-ni-[ka]  8′. abula u2-zug2 bar-šeĝ3-ĝa2-bi ku4-ra-[zu-ne] ina a-bu-ul ú-suk-ki šar-bi-iš ina e-re-bi-i-k[a]  9′. sila dagal abula u2-zug2 asil3-la2 ĝal2-la dib-be2-da-zu-[ne] ina re-bit a-bu-ul ú-suk-ki šá re-šá-ti ma-la-a-at ina ba-i-k[a] 10′. ˹e2˺-šu-me-ša4 e2 an ki-da la2-a bal-e-da-zu-[ne] [ana é-šu]-˹me˺-ša4 e2 šá ana ane u kitim tar-ṣú ina e-˹re˺-[bi-(i)-ka] 11′. [nitala]m ˹ki˺-aga2-zu igi la2-e-d[a-zu-ne] [ḫi-ir-ti na-r]a-˹am-ti-ka ina˺ n[a-ṭa-li-ka] Rest of column broken. 1′. 2′. 3′. 4′. 5′. 6′. 7′. 8′. 9′.

… Small and large drums play for you … They slaughter fattened cattle and sheep for you as a royal sacrifice, Strong lads fight (for you) with vigor and power, The citizens of Nippur, clan by clan, spend the day in plenty, The black headed people sing songs in praise of you. As soon as you direct your glance to this place, Enter like a cold draft through the Gate of the Unclean, Promenade down the broad street leading from the Gate of the Unclean that is filled with jubilation, 10′. Cross over into Ešumeša, the temple that stretches from the upper to the lower regions, 11′. Cast your eyes on your beloved s[pouse (Nin-Nibru/Šarrat-Nippuri)], Rest of column broken.

22 The end of the sign in the Sumerian of this and the following line is very far to the right on the edge of the tablet. Either the signs were quite spread out, or there was more writing the break, perhaps glosses.

Literary Journeys from Babylonia to Assyria

223

2′. The Mari archives have revealed that the a2-la2 = alû was a very large drum indeed, because it took sixteen men to transport one to Aleppo; see Ziegler 2007: 75. The gišbalag-di = timbuttu equation appears to be a post-OB one; see Shehata 2009: 94 with n. 504. It is usually translated as a “harp” with a question mark, but I know of no information that would support such a supposition. Here, it is in the company of percussion instruments. Mesopotamian string instruments had little carrying power and would hardly be heard in outdoor settings, unless massed together as in the Elamite orchestra depicted in the Assurbanipal relief from the Southwest Palace in Nineveh (Barnett et al. pl. 319). For this reason I would continue to maintain that most of the cultic instruments used in public ceremonies, such balaĝ, adab, tigi, šem3, a2-la2 etc. were of the percussion class. I understand that there is contradictory evidence suggesting that some of these instrument names such as tigi, for example, may have at some time designated string instruments, but for the most part the evidence is fairly clear. 4′. The words ĝešpu2 and lirum form a synonymic pair in hendiadys, always in the same order (e.g., Marriage of Martu 68, Death of Gilgameš N2 7′, N1 v 9). Notably, the pair occurs in the roughly contemporary Middle Babylonian Astrolabe B: in KAV 218 (from Assur, hand of Marduk-balassuereš), obv. ii the Sumerian is guruš ĝešpu2 lirum-ma ka2-ne-ne while the slightly older VS 24, 120: 11 (from Babylon) has guruš kalag ga-n[e-ne], possibly an abbreviation (.kalag ga [for ka2] etc.). The equivalent Akkadian line in KAV 218, l. ii 15 reads ú-ma-áš-ú-ba-ri, with crasis and without ina, as if it were one word (u-ma[š] u a-ba-ri in the NeoAssyrian K.2920+, l. obv. 28′, Reiner and Pingree 1981: 62). For the line, see Reiner and Pingree 1981: 62, 81, CAD U: 97, Casaburi 2003: 35–36. 8′. The adverb šarbiš is a hapax and the underlying noun šarbu, “rainy season” (CAD S/II, 60) is one of the many words in this text that are rarely attested in Akkadian. 11′. Previous editors and translators have restored the temple name e2-šu-meša4 at the beginning of this line, but parallels from other Ninurta texts suggest that he cast his eyes not on the building, but on his beloved wife; see Cooper 1978: 137. Biggs (2001: 477) thought it uncertain if this goddess was still worshipped in Kassite times, but as Alexa Bartelmus kindly reminds me, there are MB votive and dedicatory inscriptions to Nin-Nibru from Dur-Kurigalzu and elsewhere (see Bartelmus 2006: 29–30). As observed by Attinger (2014: 67), in OB the verb igi…la2 means “to glance/look at with envy,” but in the meaning “to look at” is only frequent in later texts and in lexical lists.

224

Piotr Michalowski

Part C. reverse of Ni 1′. 2′. 3′. 4′. 5′.

[ ]x-˹ab?-be2?˺ [ ]x-˹ĝa2?˺-za [ ] x x x ˹bi2?˺-in-zal-zal-e-ne [ ] x ˹ba/pa˺-e3 [ur-saĝ ga]l ˹dnin- urta za3-mi2˺-zu ˹du10˺-ga

Faint impressions of ends of signs in breaks below that must be traces of a colophon. 5′.

Note that there is no subscript, but there are good reasons to believe that this was a šir3-gid2-da, a Sumerian poetic text label that is difficult to define, as discussed, most recently by Rubio (2009: 25, 27, 66) and Shehata (2009: 274–278). The final line is typical of some, but not all šir3-gid2-da songs in honor of Ninurta; it closes both Angim and Lugal-e, but not the OB Ninurta hymn A (Sjöberg 1973: 117) or Ninurta’s Journey to Eridu (likewise a šir3-gid2-da; Riesman 1971: 4). Instead, these end with nir-ĝal2 a-a ugu-na za3-mi2-zu du10-ga, “respected son of the father who begot him, your praise is sweet.” Cooper (1978: 4) observed that “all šir3-gid2-da texts end with DN za3-mi-zu du10-gaam3 ‘DN, your praise is sweet!’ or a variant thereof, and this is, in fact, the only textual feature that all known šir3-gid2-da’s have in common.” Now that more examples of poems labeled in this fashion are known, it appears that some have za3-mi2-zu du10-ga-(am3), while others use the seemingly homophonic za3-mi2 du11-ga. The collection of Ninurta hymns TMH IV 49+88, copied by Wilcke (1976: 75) includes (ii′ 1′–6′) a fragmentary composition that ends with (5′–6′): za3-mi2 du11-ga d![nin-urta] šir3-gid2-da d[nin-urta]

Compare the end of the šir3-gid2-da Nuska B: 79 (Sjöberg 1977: 28): za3-mi2 du11-ga kingal dnuska, “Praise be to nobleman Nuska,” but the šir3-gid2-da Nuska A: iv 24 (Goetze 1950: 139) has kingal dnuska za3-mi2-zu du10-˹ga-am3˺, “O nobleman Nuska, your praise is sweet!” A variant of this is now known from an Old Babylonian šir3-gid2-da to Utu, attested in copies from Me-Turan, Susa and possibly Sippar (mentioned also in a Nippur literary catalog) that ends (Cavigneaux 2009: 10, l. 71, MeTuran source): šul dutu en dumu dnin-gal-ke4 mi2 du11-ga ka-tar-zu ga-sil,

Literary Journeys from Babylonia to Assyria

225

“O (ever)-youthful Utu, Ningal’s cherished son, of your glory I do sing!” and this seems to be a clever, if somewhat radical, reinterpretation of the standard ending element, exchanging the only two Sumerian lexical items that share the morpho-grapheme mi2 (/mim?/). One could propose that the omission of before mi2 is a mistake and the line was really “Praise be to (ever)youthful Utu, Ningal’s son, of your glory I do sing!,” but the scribe of this tablet seems too good for such a blunder. The only other witness to this line, VS 10, 212: (possibly Sippar), does not provide a solution because it is broken at the crucial point ([…m]i2 du11-ga etc. [collated by Juliane Eule]).

Bibliography Annus, Amar. 2001. Ninurta and the Son of Man. Pp. 7–17 in Mythology and Mythologies. Methodological Approaches to Intercultural Influences: Proceedings of the Second Annual Symposium of the Assyrian and Babylonian Intellectual Heritage Project. Held in Paris, France, October 4–7, 1999, ed. Robert W. Whiting. Helsinki: The Neo-Assyrian Text Corpus Project. Annus, Amar. 2002. The God Ninurta in the Mythology and Royal Ideology of Ancient Mesopotamia. SAAS 14. Helsinki: The Neo-Assyrian Text Corpus Project. Archi, Alfonso. 1987. The “Sign-list” from Ebla. Pp. 91–113 in Eblaitica: Essays on the Ebla Archives and Eblaite Language I, ed. Cyrus H. Gordon, Gary A. Rendsburg, and Nathan H. Winter. Winona Lake: Eisenbrauns. Attinger, Pascal. 1993. Eléments de linguistique sumérienne: La construction de du11 /e/di “dire.” OBO Sonderband. Göttingen: Vandenhoeck und Ruprecht. Attinger, Pascal. 2014. Nabu-Enlil-Īterpīša (ANL 7). N.A.B.U. 2014/2 65–68. Barnett, Richard David, Erika Bleibtreu, Geoffrey Turner, and Dominique Collon. 1998. Sculptures from the Southwest palace of Sennacherib at Nineveh. London: British Museum Press. Bartelmus, Alexa. 2005. Das Sumerische der Königsinschriften der Kassitendynastie Orthographie, Lexikon, Idiomatik. MA thesis, Ludwig-Maximilians-Universität München. Bartelmus, Alexa. 2010. Restoring the Past. A Historical Analysis of the Royal Temple Building Inscriptions from the Kassite Period. KASKAL 7: 143–71. Bartelmus, Alexa. n. d. The Role of Babylon in Babylonian Scribal Education. In Cultures and Societies in the Middle Euphrates and Habur Areas in the Second Millennium BC, vol. 1: Scribal Education and Scribal Traditions. Studia Chaburensia. Wiesbaden: Harrassowitz. Bartelmus, Alexa and Aaron Schmitt. 2014. Beutestücke aus Babylonien in Assyrien. Überlegungen zu einer neuen Weihinschrift Kurigalzus I. aus Assur. ZA 104: 74–90. Behrens, Hermann. 1978. Enlil und Ninlil. Ein sumerischer Mythos von Nippur. StPoSM 8. Rome: Pontificio Instituto Biblico. Biggs, Robert D. 2001. Nin-Nibru. RlA 9: 476–77. Böck, Barbara. 1999. Homo Mesopotamicus. Pp. 53–68 in Munuscula Mesopotamica: Festschrift für Johannes Renger, ed. Barbara Böck, Eva C. Cancik-Kirschbaum, and Thomas Richter. AOAT 267. Münster: Ugarit-Verlag. Brinkman, J. A. 1969. Ur: The Kassite Period and the Period of the Assyrian Kings. OrNS 38: 310–48.

226

Piotr Michalowski

Casaburi, Maria Cristina. 2003. Tre-stelle-per-ciascun(-mese): l’Astrolabio B: edizione filologica. Annali Supplemento 93. Naples: Università degli studi di Napoli “L’Orientale.” Cavigneaux, Antoine. 2009. Deux hymnes sumériens à Utu. Pp. 3–18 in Et il y eut un esprit dans l’Homme. Jean Botté ro et la Mé sopotamie, ed. Xavier Faivre, Brigitte Lion, and Cécile Michel. Paris: De Boccard. Civil, Miguel. 1983. Enlil and Ninlil: The Marriage of Sud. JAOS 103: 43–66. Civil, Miguel. 1987. Feeding Dumuzi’s sheep: The Lexicon as a Source of Literary Inspiration. Pp. 37–55 in Language, Literature, and History: Philological and Historical Studies presented to Erica Reiner, ed. Francesca Rochberg. AOS 67. New Haven: American Oriental Society. Civil, Miguel. 1994. The Farmer’s Instructions: A Sumerian Agricultural Manual. Aula Orientalis Supplementa 5. Sabadell: Editorial Ausa. Civil, Miguel. 2013. Remarks On ad-gi4 (a.k.a. “Archaic Word List C” or “Tribute”). JCS 65: 13–67. Clayden T. 1996. Kurigalzu I and the Restoration of Babylonia. Iraq 58: 109–21. Cohen, Mark E. 1975a. ur.sag.me.šar2.ur4. A Širnamšubba of Ninurta. WO 8: 22–36. Cohen, Mark E. 1975b. The Incantation-Hymn: Incantation or Hymn? JAOS 95: 592–611. Collon, Dominique. 2006. The Iconography of Ninurta. Pp. 100–109 in The Iconography of Cylinder Seals, ed. Paul Taylor. Warburg Institute Colloquia 9. London: Warburg Institute. Cooper, Jerrold S. 1978. The Return of Ninurta to Nippur. AnOr 52. Rome: Pontifical Institute. Crisostomo, C. Jay. 2014. Bilingual Education and Innovations in Scholarship: The Old Babylonian Word List Izi. Ph.D. dissertation, University of California, Berkeley. van Dijk, J. J. A. 1953. La sagesse suméro-accadienne: recherches sur les genres littéraires des textes sapientiaux. Leiden: Brill. van Dijk, J. J. A. 1983. LUGAL UD ME-LÁM-bi NIR-ĜÁL: Le récit épique et didactique des Travaux de Ninurta, du Déluge et de la nouvelle Création. Leiden: Brill Ebeling, Erich. 1918. Quellen zur Kenntnis der babylonischen Religion I. MVAG 23/I. Leipzig: Hinrichs. Emelianov, V. V. 2013. К истории шумер. (m)uzug: орфография, семантика, этимология (On the History of Sumerian (m)uzug: Orthography, Semantics and Etymology of the Word, in Russian). Индоевропейское языкознание и классическая филология (Indoevropejskoe jazykoznanie i klassicheskaja filologija) 17: 282–89. Falkenstein, Adam. 1953. Zur Chronologie der sumerischen Literatur. MDOG 85: 1–13. Foster, Benjamin F. 1993. Before the Muses: An Anthology of Akkadian Literature. Bethesda, MD: CDL. Frahm, Eckart. 2010. The Latest Sumerian Proverbs, Pp. 155–84 in Opening the Tablet Box: Near Eastern Studies in Honor of Benjamin R. Foster, ed. Sarah C. Melville and Alice L. Slotsky. CHANE 42. Leiden: Brill. Geller, Mark J. 1990a. Taboo in Mesopotamia: A Review Article. JCS 42: 105–17. Geller, Mark J. 1990b. Astronomy and Authorship. BSOAS 53: 209–13. Gibson, McGuire. 1993. Nippur, Sacred City of Enlil, supreme god of Sumer and Akkad. Al-Rafidan 14: 1–18. Goetze, Albrecht. 1950. Texts and Fragments. JCS 4: 137–40. George, Andrew. 1990. Review of O. R. Gurney, Literary and Miscellaneous Texts in the Ashmolean Museum (OECT 11), ZA 80: 155–60. George, Andrew. 2009. Babylonian Literary Texts in the Schøyen Collection. CUSAS 10. Bethesda: CDL.

Literary Journeys from Babylonia to Assyria

227

George, Andrew. 2011. Kurigalzu II. No. 61. Pp. 117–18 in Cuneiform Royal Inscriptions and Related Texts in the Schøyen Collection, ed. A. George. CUSAS 17. Bethesda: CDL Press. George, Andrew. 2012. CUSAS 17 no. 61. NABU 2012.4: 99–100. Hallo, William W. 1981. Review of Cooper 1978. JAOS 101: 253–57. Hallo, William W. 1985. Biblical Abominations and Sumerian Taboos. JQR 76: 21–40. Horowitz, Wayne. 1998. Mesopotamian Cosmic Geography. MC 8. Winona Lake: Eisenbrauns. Horowitz, Wayne. 2010. VAT 17081: A Forerunner to the Menology of Astrolabe B. Pp. 183– 188 in Studies Presented in Honour of Veysel Donbaz dub.sar.é.dub.ba.a: Veysel Donbaz’a Sunulan Yazilar, ed. Şevket Dönmez. Istanbul: Ege. Horowitz, Wayne. 2013. The Astrolabes: An Exercise in Transmission, Canonicity, and ParaCanonicity. Pp. 273–87 in Between Text and Text, International Symposium on Intertextuality in Ancient Near Eastern, Ancient Mediterranean, and Early Medieval Literatures, ed. Michaela Bauks, Wayne Horowitz, and Armin Lange. Journal of Ancient Judaism, Supplements 6. Gottinhiem: Vandenhoeck & Ruprecht. Horowitz, Wayne. 2014. The Three Stars Each: The Astrolabes and Related Texts. Archiv für Orientforschung Beiheft 33. Vienna: Institut für Orientalistik der Universität Wien. Hrůša, Ivan. 2010. Die akkadische Synonymenliste malku-šarru. Eine Textedition mit Übersetzung und Kommentar. AOAT 50. Münster: Ugarit-Verlag. Komoróczy, Géza. 1976. Zum Namen eines Stadttores von Nippur. AoF 4: 341–45. Krecher, Joachim. 1970. Review of H. Otten and W. von Soden, Das akkadisch-hethitische Vokabular KBo I 44 + KBo XIII. Wiesbaden: Harrassowitz, 1968, in ZA 60: 217–21. Krispijn, Theo J. H. 2001. The Sumerian Lexeme *urum, a Lexico-Etymological Approach. Pp. 251–61 in Veenhof Anniversary Volume: Studies Presented to Klaas R. Veenhof on the Occasion of his Sixty-fifth Birthday, ed. Wilfred H. van Soldt. Uitgaven van het Nederlands Historisch-Archaeologisch Instituut te Istanbul, 89. Leiden: Nederlands Instituut voor het Nabije Oosten. Lambert, Wilfred G. 1960. Babylonian Wisdom Literature. Oxford: Clarendon Press. Lambert, Wilfred G. 1957–58. Three Unpublished Fragments of the Tukulti-Ninurta Epic. AfO 18: 38–51. Lambert, Wilfred G. 1976. Tukulti-Ninurta I and the Assyrian King List. Iraq 38: 85–94. Landsberger, Benno. 1928. Review of F. Nötscher, Ellil in Sumer und Akkad. DLZ 49: 1221–23. Landsberger, Benno. 1968. The Third Tablet of the Series Ea A Nâqu. JAOS 88: 133–47. Landsberger, B. and O. R. Gurney. 1957–1958. Practical Vocabulary of Assur. AfO 18: 328–41. Machinist, Peter. 1976. Literature as Politics-Tukulti-Ninurta Epic and The Bible. CBQ 38: 455–82. Maul, Stefan M. 1997. Die altorientalische Hauptstadt – Abbild und Nabel der Welt. Pp. 109– 214 in Die orientalische Stadt: Kontinuität, Wandel, Bruch. 1. Internationales Colloquium der Deutschen Orient-Gesellschaft 9.−10. Mai 1996 in Halle/Saale, ed. Gernot Wilhelm. CDOG 1. Saarbrücken: Saarbrücker Druckerei und Verlag. Maul, Stefan M. 1998. Marduk, Nabu und der assyrische Enlil. Die Geschichte eines sumerischen Šu’ilas. Pp. 159–97 in Festschrift für Rykle Borger zu seinem 65. Geburtstag am 24. Mai 1994: tikip santakki mala bašmu, ed. Stefan M. Maul. CM 10. Leiden: STYX. Maul, Stefan M. 2005. Bilingual (Sumerian-Akkadian) Hymns from the Seleucid-Arsacid Period. Pp. 11–116 in Literary and Scholastic Texts from the First Millennium B.C., ed. Ira Sparr and Wilfred G. Lambert. CTMMA 2. New York: The Metropolitan Museum of Art. Mayer, Werner, R. 1992. Ein Hymnus auf Ninurta als Helfer in der Not. OrNS 61: 17–57. Mayer, Werner, R. 2005. Das Gebet des Eingeweideschauers an Ninurta. OrNS 74: 51–56.

228

Piotr Michalowski

Michalowski, Piotr. 1998. Literature as a Source of Lexical Inspiration: Some Notes on a Hymn to the Goddess Inana. Pp. 65–73 in Written on Clay and Stone: Ancient Near Eastern Studies Presented to Krystyna Szarzyńska on the Occasion of her 80th Birthday, ed. Jan Braun Warsaw: Agade. Michalowski, Piotr. 2011. The Correspondence of the Kings of Ur An Epistolary History of an Ancient Mesopotamian Kingdom. MC 15. Winona Lake: Eisenbrauns. Michalowski, Piotr. Forthcoming. The Domestication of Stranger Kings: Making History by List in Ancient Mesopotamia. In proceedings from the conference Historical Consciousness and Historiography (3000 BC–AD 600), Merton College, University of Oxford, 17–19 September 2014. Nashef, Khalid. 1983. Nochmals PARsā. RA 77: 169–74. Oelsner, Joachim and Peter Stein. 2011. Der Stadtplan von Nippur (HS 197). AfO 52: 104–16. Paulus, Susanne. 2011. Sumerische Inschriften des ausgehenden 2. und der ersten Hälfte des 1. Jahrtausends v. Chr. Pp. 17–22 in Grab-, Sarg-, Bau- und Votivinschriften, ed. B. Janowski and D. Schwemer. TUAT/NF 6. Gütersloh: Gütersloher Verlagshaus. Paulus, Susanne. 2014. Babylonien in der 2. Hälfte des 2. Jts. v. Chr. – (K)ein Imperium? Ein Überblick über Geschichte und Struktur des mittelbabylonischen Reiches (ca. 1500– 1000 B.C.). Pp. 65–100 in Imperien und Reiche der Weltgeschichte: Epochenü bergreifende und globalhistorische Vergleiche, ed. Michael Gehler and Robert Rollinger. Wiesbaden: Harrassowitz. Pedersén, Olof. 1998. Archives and Libraries in the Ancient Near East 1500–300 B.C. Bethesda: CDL. Pedersén, Olof. 2005. Archive und Bibliotheken in Babylon: Die Tontafeln der Grabung Robert Koldeweys 1899–1917. Abhandlungen der Deutschen Orient-Gesellschaft 25. Saarbrücken: Saarländische Druckerei und Verlag. Pedersén, Olof. 2011. Excavated and Unexcavated Libraries in Babylon. Pp. 47–67 in Babylon: Wissenskultur in Orient und Okzident, ed. Eva Cancik-Kirschbaum, Margarete van Ess, and Joachim Marzahn. Berlin Studies of the Ancient World 1. Berlin: de Gruyter. Peterson, Jeremiah. This volume. A Middle Babylonian Sumerian Fragment of the Adapa Myth from Nippur and an Overview of the Middle Babylonian Sumerian Literary Corpus at Nippur. Pongratz-Leisten, Beate. 2001. The Other and the Enemy in the Mesopotamian Conception of the World. Pp. 195–231 in Mythology and Mythologies: Methodological Approaches to Intercultural Influences. Proceedings of the Second Annual Symposium of the Assyrian and Babylonian Intellectual Heritage Project. Held in Paris, France, October 4–7, 1999, ed. Robert M. Whiting. Melammu Symposia 2. Helsinki: The Neo-Assyrian Text Corpus Project. Reiner, Erica and David Pingree. 1981. Babylonian Planetary Omens: Part Two. Enūma Anu Enlil, Tablets 50–51. BiMes 2/2. Malibu: Undena. Reisman, Daniel. 1971. Ninurta’s Journey to Eridu. JCS 24: 3–10. Rubio, Gonzalo. 2009. Sumerian Literature. Pp. 11–75 in From an Antique Land: An Introduction to Ancient Near Eastern Literature, ed. Carl S. Ehrlich. Lanham: Rowman and Littlefield. Sassmannshausen, L. 2008. Babylonische Schriftkultur des 2. Jahrtausends v. Chr. in den Nachbarländern und im östlichen Mittelmeerraum. AuOr 26: 263–93. Shehata, Dahlia. 2009. Musiker und ihr vokales Repertoire. Untersuchungen zu Inhalt und Organisation von Musikerberufen und Liedgattungen in altbabylonischer Zeit. Göttinger Schriften zum Alten Orient, 3. Göttingen: Universitätsverlag Göttingen.

Literary Journeys from Babylonia to Assyria

229

Sjöberg, Åke W. 1967. Zu einigen Verwandtschaftsbezeichnungen im Sumerischen. Pp. 201– 31 in Heidelberger Studien zum Alten Orient, ed. D. O. Edzard. Wiesbaden: Harrassowitz. Sjöberg, Åke W. 1973. Hymn to Numušda with a Prayer for King Sîniqīšam of Larsa and a Hymn to Ninurta. OrSuec 22: 107–21. Sjöberg, Åke W. 1975. in-nin ša3-gur4-ra: A Hymn to the Goddess Inanna by the en-Priestess Enḫeduanna. ZA 65: 161–253. Sjöberg, Åke W. 1976. Hymns to Ninurta with Prayers to Šūsîn of Ur and Būrsîn of Isin. Pp. 422–26 in Kramer Anniversary Volume. Cuneiform Studies in Honor of Samuel Noah Kramer, ed. Barry Eichler, Jane W. Heimerdinger, and Åke W. Sjöberg. AOAT 25. Neukirchen-Vluyn: Butzon & Bercker Kevelaer. Sjöberg, Åke W. 1977. Miscellaneous Sumerian Texts, II. JCS 29: 3–45. Steinert, Ulrike. 2011. Akkadian Terms for Streets and the Topography of Mesopotamian Cities. AoF 38 : 309–47. Steinert, Ulrike. 2014. City Streets: Reflections on Urban Society in the Cuneiform Sources of the Second and First Millennium BCE. Pp. 123–69 in The Fabric of Cities: Aspects of Urbanism, Urban Topography and Society in Mesopotamia, Greece and Rome, ed. Natalie N. May and Ulrike Steinert. Culture and History of the Ancient Near East 68. Leiden: Brill. Streck, M. 2001. Ninurta/Ningirsu. A. I. In Mesopotamien. RlA 9: 512–22. Tenney, Jonathan S. 2016. The Elevation of Marduk Revisited: Festivals and Sacrifice at Nippur in the High Kassite Period. JCS 68: 153–80. Tenney, Jonathan S. In press a. Uruk in Southern Babylonia under the Kassite Kings. Proceedings of the conference: Uruk − Altorientalische Metropole und Kulturzentrum 8. Internationales Colloquium der Deutschen Orient-Gesellschaft am 25. 4. 2013 und 26. 4. 2013 in Berlin. Tenney, Jonathan S. In press b. The Middle Babylonian Period. In A Handbook of Ancient Mesopotamia, ed. G. Rubio. Berlin: de Gruyter. Van Koppen, Frans. 2010. The Old to Middle Babylonian Transition: History and Chronology of the Mesopotamian Dark Age. Ägypten und Levante 20: 453–64. Van Lerberghe, Karel and Gabriella Voet. 2009. A Late Old Babylonian Temple Archive from Dūr-Abiešuḫ. CUSAS 8. Bethesda, MD: CDL. Veldhuis, Niek. 2000. Kassite Exercises: Literary and Lexical Extracts. JCS 52: 67–94. Veldhuis, Niek. 2008. Kurigalzu’s Statue Inscription. JCS 60: 28–51. Veldhuis, Niek. 2014. History of the Cuneiform Lexical Tradition. GMTR 6. Münster: UgaritVerlag. Viano, Maurizio. 2009. La tradizione letteraria sumerica negli archivi siro-anatolici durante il Tardo Bronzo. PhD dissertation, Università degili studi di Trieste. Yamada, Shigeo. 1994. The Editorial History of the Assyrian King List. ZA 84: 11–37. Wagensonner, Klaus. 2008. Nin-Isina(k)’s Journey to Nippur Reconsidered. WZKM 98: 277– 94. Wagensonner, Klaus. 2011. A Scribal Family and its Orthographic Peculiarities. On the Scientific Work of a Royal Scribe and his Sons. Pp. 645–701 in The Empirical Dimension of Ancient Near Eastern Studies / Die empirische Dimension altorientalischer Forschungen, ed. Gebhard J. Selz and Klaus Wagensonner. WOO 6. Vienna: LIT. Weidner, Ernst. 1952–53. Die Bibliothek Tiglatpilesers I. AfO 16: 197–215. Westenholz, Joan G. 2005. Sing a Song for Šulgi. Pp. 343–373 in An Experienced Scribe Who Neglects Nothing: Ancient Near Eastern Studies in Honor of Jacob Klein, ed. Yitzhak Sefati Bethesda: CDL.

230

Piotr Michalowski

Wilcke, Claus. 1976. Kollationen zu den sumerischen literarischen Texten aus Nippur in der Hilprecht-Sammlung, Jena. ASAW 65/4. Berlin: Akademie-Verlag. Ziegler, Nele. 2007. Les Musiciens et la musique d’après les archives de Mari. FM 9. Paris: SEPOA.

Manuel Molina and Piotr Steinkeller

New Data on Garšana and the Border Zone between Umma and Girsu/Lagaš 1 The Location of Garšana The recent publication of ca. 1,500 texts stemming from the ancient town of Garšana,1 done in an exemplary fashion by David I. Owen (CUSAS 3 and 4), has generated a great deal of interest among Sumerologists. Given the fact that these tablets had become available as a result of looting operations, whose place remains unknown, one of the issues that attracted their attention was the question of Garšana’s precise location. Garšana is known to have belonged to the Umma province, but Wolfgang Heimpel (2009: 7–9; 2011) and Piotr Steinkeller (2007: 188; 2011, 2013) have disagreed on where exactly it should be placed. While Heimpel tried to situate Garšana in the northwestern section of the province, near the cities of Umma and Zabalam, Steinkeller argued for a location near the border with the Girsu/ Lagaš province, i.e., in Umma’s eastern or southeastern part. Among the arguments given by these two scholars in support of their views, the textual data on the duration of trips between various localities, the topographical relationships between certain places, and some considerations of political nature have played the fundamental role. In this study, which we offer to Miguel as a token of our friendship and great admiration for his extraordinary scholarship, we present a set of new data that, in our view, clearly point to the eastern option as the most plausible one for Garšana’s location. The most important among those data is an unpublished tablet from Garšana, which forms part of The Green Collection, Oklahoma City. The tablet in question reads as follows (photographs available at BDTNS under 193423, http://bdtns.filol.csic.es/193423): Green 2013–3152  1 3 guruš sag-t[ag]  2 2 guruš a2 ⅔-ta

3 full-wage workers (and) 2 ⅔-wage workers, during 1 day,

1 The reading of this toponym most likely is Nig2-ša(-an)-naki. See Steinkeller 2012. 2 We are deeply greatful to Lance Allred, the Curator of the Cuneifrom Section of The Green Collection, who gave us the permission to publish this tablet and provided us with its photographs. Our warm thanks also go to Marcel Sirgist and Tohru Ozaki, who had told us about the existence of this tablet in the first place, and shared with us their preliminary transliteration of this document. https://doi.org/10.1515/9781501503696-014

232

Manuel Molina and Piotr Steinkeller

 3  4  5  6

ud 1-še3 Gar-ša-an-naki-ta An-za-gar3 hi-a3 Ummaki-še3 ma2 gid2-da u3 ma2 dirig-ga

towed and floated the boats from Garšana to the main? Anzagar of Umma.

 7  8  9 10 11 12

5 guruš sag-tag 2 guruš a2 ⅔-ta ud 3-še3 An-za-gar3-Ummaki-ta Gir2-suki-še3 ma2 dirig-ga Gir2-suki-ta Gar-ša-an-na-˹še3˺ ma2 gid2-da

5 full-wage workers (and) 2 ⅔wage workers, during three days, flotated the boats from AnzagarUmma to Girsu, (and) towed the boats from Girsu to Garšana.

7 guruš sag-tag 2 guruš a2 ⅔-ta 3 guruš a2 ½-ta ud 7-še3 Gar-ša-an-naki-ta U3-dag-gaki-še3 ma2 dirig-ga U3-dag-gaki-ta gi u3 gi-zi Gar-ša-an-naki-še3 la2-a

7 full-wage workers, 2 ⅔-wage workers, (and) 3 ½-wage workers, during 7 days, floated the boats from Garšana to Udaga, (and) hauled reeds and fresh reeds from Udaga to Garšana.

 9

1 guruš ud 1-še3 Dar-ra-um-še3

1 worker, during 1 day, (travelled) to Darra’um.

10 11 12 13

ki Puzur4-a-ku-um-ta ba-zi iti ezem-an-na mu ma2-gur8 mah ba-ab-du8

Expended from Puzur-akum.

13 r. 1  2  3  4  5  6  7  8

Month xi. ŠS8

The interpretation of this text hinges on the relationship between Garšana and the other toponyms mentioned therein, which are Anzagar-Umma, Darra’um, and Udaga. We assume that Anzagar-Umma is identical with the Anzagar fre-

3 The signs hi-a, usually read hi-a and meaning “sundry, various,” defy explanation in this context. Since the same toponym is written simply An-za-gar3-Ummaki in line 10, it conceivably is a scribal error. An alternative solution would be to read these signs as šar2-a, “great, important.” See šar2 = rabû, “great” (CAD R, p. 38 lexical section of rabû A). If so, this desigantion would mean: “the great/main Anzagar of Umma.” As a matter of fact, this particular anza-gar3 happened to be the main “fortified village” of the Umma province (see below).

New Data on Garšana and the Border Zone between Umma and Girsu/Lagaš

233

quently appearing in Umma texts, which belonged to the Umma province, and was situated on the banks of the Girsu/Namhani canal (Steinkeller 2011: 382), upstream from Girsu (UTI 4: 2401).4 Although the Umma and Garšana sources name a number of other toponyms that contain the element an-za-gar3 “fortified village,” such as An-za-gar3-giškiri6-Zabalam4ki, An-za-gar3-a-šag4-La2mah, An-za-gar3-Da-da, and An-za3-gar3-Nig2-ul-pa-e3, the name of Anzagar-Umma itself clearly points to a border area, which, as it will be shown below, has to be the eastern frontier of Umma with the Girsu/Lagaš province. Two other texts (CUSAS 3: 553; AAICAB 1/3: Bod. S 160) likewise demonstrate a close relationship between Anzagar, Garšana, and Eduru-lumah, the last being an important topographic point for our discussion, whose location will be considered below. As for Darra’um, recorded in rev. 9, this settlement is documented only in two other texts, both stemming from Garšana (CUSAS 3: 462; 1097). Possibly, therefore, Darra’um was a neighbor of Garšana, but this point must be left open for now. The third toponym mentioned in Green 2013–15, Udaga, is of particular importance for the understanding of this text. We discuss it in detail in the immediately following section.

2 The Town of Udaga and the Udaga Canal Udaga was a royal town, which belonged to the Girsu/Lagaš province (MVN 12: 464; 466; TUT 160 obv. iv 24′, 29′; SAT 1: 420 rev. i 16). Udaga had a field that was partly in the Girsu/Lagaš province (NATN 447 and 451) and partly in the Umma province, very close to the Gu’edena field, and not faraway from the Ušgida field (MVN 13: 312). The section of the field that belonged to the Umma province was cultivated by members of Umma’s royal sector, as shown by BPOA 1: 852 (AS3), a text recording a court decision regarding the repayment of barley lost due to the flooding of the field; the decision was formally made by the governor of Umma, but it was actually supervised by the crown (with Ur-Nanše acting as a royal deputy), while the plot managers belonged to the royal sector as well (Lugal-ka, plot manager of the zabar-dab5, was among them). This explains why field works and yields from the Udaga field are not attested in the texts from the provincial archives of Umma. Likewise, BPOA 1:

4 UTI 4: 2401: 3 guruš ud ˹6?˺-[še3], An-za-gar3 id2 Gir2-˹su˺ki-ta, Gir2-suki-še3 ma2 dirig˹ga˺, ma2 ba-al-la ˹u3˺, ma2 su3 gur-ra.

234

Manuel Molina and Piotr Steinkeller

852 shows that the Udaga field was adjacent to the field of Nin-hedu, the wife of the sukkal-mah, which is in accordance with the location of Udaga in the Girsu/Lagaš province. On the other hand, the management of the Udaga field in the Girsu zone was under the responsibility of Ur-Meme, an administrator (šabra) of royal households located in the border area between Umma and Girsu/Lagaš during the reign of Šu-Suen (see below). Texts from Garšana record frequent travels of messengers to Udaga (CUSAS 3: 43, etc.), as well as the supplies of reeds (CUSAS 3: 189; 205) and barley (CUSAS 3: 185) from Udaga to Garšana, which is in agreement with Udaga’s belonging to a network of royal towns and rural estates. The records of trips between Garšana and Udaga show that these two localities were connected through a waterway that was most probably the Udaga canal (see below), with Garšana being situated upstream from Udaga (CUSAS 3: 185; Green 2013–315). The distance between them cannot be established with certainty, although a round trip by foot could be covered in one day. This is stated in CUSAS 3: 553 (see below), an account that, given its context, does not record man-days, but an overland trip that took about one day. Other texts record longer trips, taking between seven to ten days (supply of reeds: Green 2013–315; supplies of barley: CUSAS 3: 261; 287; 288; 289; 290; 414). These accounts cover not only the round trip (cf. Green 2013–315), but also the days spent in Udaga while the workers in question performed various other tasks. In fact, the workers listed in these texts usually not only towed the boats, but they also served in their professional capacities, such as reed workers, leather workers, fullers, felters, female weavers, female millers, and female oil-pressers. These work assignments also show how closely the economic activities of Udaga and Garšana were intertwined. In the case of Green 2013– 315, the transported reeds were probably intended for the construction or maintenance of flood-protection structures, operations that were typically performed in March/April (Rost 2015: 56), a period that fits well the date of our text (iti ezem-an-na = February/March). As is strongly indicated by Green 2013–315, Garšana and Udaga lay on the same waterway, which, in all probablity, was the Udaga canal. The Umma documentation shows that Udaga was one of the largest canals of the Umma province, possibly the main artificial waterway that branched from the Tigris. This canal is likely mentioned in MVN 10: 105, an incompletely preserved text describing the excavation or dredging of a canal that branched from the Tigris: […] id2-˹Idigna˺-ta 705 nindan gid2 id2 ba-al-la (lines 1′–3′). The canal in question was very long; the preserved sections of MVN 10: 105 record 991 nindan = 5,946 m of an excavated/dredged waterway. That the canal in question

New Data on Garšana and the Border Zone between Umma and Girsu/Lagaš

235

was the Udaga is indicated by the fact that one of the points along the waterway described in MVN 10: 105 was giššinig ki-su7 dŠul-pa-e3, “the tamarisk grove at the threshing floor of (the field of) Šulpa’e” (rev. 3′). As we discuss it below (see also map 1), the field of Šulpa’e was situated near Apišal, precisely in the area where the Udaga is expected to have flown. On the other hand, the “tamarisk grove” associated with that threshing floor may be connected with the kab2-kud giššinig, “canal regulator of the tamarisk grove,” which likewise was situated in the same general area.5 Importantly, this canal regulator is mentioned together with that of Inim-ma-an (Nik. 2: 114; SAT 3: 1654), the latter toponym also appearing in MVN 10: 105 rev. 16′ (˹x˺ dNin-ur4-ra igi ga2udu Inim-ma-an). It is independently known that Inim-ma-an was situated in the area of Apišal, near a village called E2-duru5-A-bu3 or simply A-bu3.6 Apart from the sheer number of the Udaga’s attestations (1027), its importance may be deduced, for example, from the large quantities of reed bundles needed to repair a barrage (kun-zi-da) controlling the Udaga’s water level (UTI 5: 3499), very similar to the quantity required for a similar barrage situated in the Tigris (SAT 2: 323) (cf. Rost 2015: 103). On the other hand, the fact that the Udaga canal had at least three separate barrages, which were situated “opposite the village of Eduru-Ṣilašu” (Nisaba 24: 10 rev. v 21; cf. Rost 2015: 106), “opposite Anzagar-Nigulpa’e” (Nisaba 24: 10 rev. vi 12–13), and near the village of Eduru-lumah (CUSAS 3: 553) respectively, is also an indication of its importance and considerable length. The maintenance of these barrages was the responsibility of the Umma provincial administration (MCS 3, p. 92 BM 113089; SACT 2: 26, CDLJ 2003/1 1 v 7; MVN 18: 404; 689, etc.). As demonstrated by our sources, both the inlet (ka) and the outlet (kun) of the Udaga were located within the borders of the Umma province. The maintenance of the canal inlet, where the nakabtum was found (BPOA 7: 1893; MVN 14: 312; BPOA 7: 1721; BPOA 2: 2545), likewise was the responsibility of the

5 See further YBC 3882 = SAT 2: 292 (collations kindly provided by Agnete Lassen), which reads: 3600+600+300 sa gi / kab2-kud giššinig E2-dag-ga-a (ii 7′–8′); 1200+300 sa gi / gu2 nigin2 a-ša3 dŠara2-ka E2-dag-ga-a (ii 5′–6′). If E2-dag-ga (which is not otherwise attested) is a variant spelling of U3-dag-ga, these examples would conclusively place the “tamarisk grove” on the Udaga canal. Note in this regard that the PN E2-dag-ga, rarely attested in Ur III sources (SAT 2: 1051 and 1053; Nisaba 23: 7 and 33), could also be considered a variant spelling of the well known PN U3-dag-ga. 6 See, especially, RA 12, p. 49: a-ša3 A-bu3 Inim-ma-an. In turn, the Eduru-Abu and its field were situated near the nakabtum establishment (Nik. 2; 141; BPOA 1: 918; UTI 4: 2836), which belonged to this region as well (see below). 7 According to the data collected in BDTNS. This figure includes the attestations of the town Udaga, excluding those of the PN Udaga.

236

Manuel Molina and Piotr Steinkeller

Map 1: Border zone between Umma and Girsu/Lagaš.

provincial administration (UCP 9/2/1: 44; MCS 3, p. 84 BM 113032; Princeton 1: 509). The inlet was in the vicinity of a village called Eduru-Ṣilašu (MCS 6, p. 82 HSM 7165), directly connected with Apišal (UTI 4: 2881; MCS 6, p. 82 HSM 7165), downstream but not too far from it (UTI 4: 2881). The ancient Tigris showed a marked anastomosing configuration in the area of Apišal, with multiple branches that made it more prone to flooding. This might explain the concentration of major flood-protection works in that region (Rost 2015: 241). Thus, in close proximity to a barrage on the Udaga (probably the one near EduruṢilašu), several other barrages are documented: on the Sisa, Ubada, and E-anše canals (Hom. Lenoble no. 44; MVN 16: 1593; MVN 18: 404; SAT 3: 1657; UTI 5: 3499, etc.; cf. Rost 2015: 98 n. 64). Because of the complicated nature of this geomorphological context, it is difficult to say whether the Girsu/Namhani canal branched from the Udaga canal or one of the Tigris branches, but any of the alternatives would fit well the expeditions described in our texts. Texts frequently qualify the Udaga canal by the adjective “old” (sumun) (UTI 3: 1910, etc.), which suggests that, at some point in time, the bed of the canal was rehabilitated and extended further south. The “Old Udaga” canal terminated in the area of Eduru-lumah (UTI 4: 2372), a village located on the east bank of the canal, very close to Anzagar-Umma. In that locality, a barrage

New Data on Garšana and the Border Zone between Umma and Girsu/Lagaš

237

and a flow-divider (kab2-kud), situated at the outlet of the Udaga, supplied irrigation water for the fields of Eduru-lumah. This location is also mentioned in CUSAS 3: 553, for which see below. Another canal taking off from the Udaga (most probably from its west bank) in that region was the Šara-pada, which provided water for the Šara-pada field (Nisaba 24: 10 rev. ii 9; SAT 3: 1733). Another topographic feature found in the same general area was the Agam (SAT 3: 1733), a type of lake used to drain flood and excess irrigation waters (Civil 1994: 130 f.; Rost 2015: 164 f.). The Udaga canal flowed further south for a few kilometers, more or less along the border between the Girsu/Lagaš and Umma provinces, down to its outlet, which was likewise located within the limits of the Umma province, close to the Agartur field (MVN 18: 644). Like its inlet and all the barrages, the outlet of the Udaga canal was maintained by the provincial administration of Umma (UTI 4: 2585; BPOA 7: 1831; Nisaba 23: 10). In consideration of these data, the trip described in Green 2013–315 may be reconstructed as follows: from Garšana the boat was towed upstream on the Udaga canal to the junction with the Girsu/Namhani canal; from there, it was floated down on the Girsu/Namhani canal, via Anzagar-Umma, to Girsu. It appears that the boat was loaded at Girsu (possibly with barley), and then, following the opposite route, it was brought back to Garšana. Probably the same towers traveled subsequently downstream from Garšana to Udaga, hauled reeds back to Garšana, and worked there for a few days on a floodprotection project. Similar trips are described in CUSAS 3: 553,8 a text that provides aditional evidence on this route: obv. i. 1

[n guruš ud n-še3]

n workers tied sacks during n days.

2

[kuša-g]a2-l[a2] keš2-ra2

3 4 5

[n]+1? guruš ud 1-še3 [GA]R-ša-an-naki-ta U3-dag-gaki-še3 / gen-na

n+1 workers walked from Garšana to Udaga during 1 day.

6  7  8

6 ⅔ guruš arad2 e2-a 14 guruš hun-ga2 1 ma2 10.0.0 gur 0.0.2

6⅔ palace workers, 14 hired workers, 1 10-gur boat (its wage is) 20 sila, 1 8-gur boat (its rental

8 This tablet was collated from the photographs kindly supplied by Laura W. Johnson-Kelly. The text has been partially discussed by Heimpel 2009: 309 f.

238

Manuel Molina and Piotr Steinkeller

 9 10 11

1 ma2 8.0.0 gur 0.0.1 5 sila3 [1] ma2 5.0.0 gur 5 sila3 ˹ud˺ 3-še3

fee is) 15 sila, [1] 5-gur boat (its rental fee is) 5 sila, for 3 days.9

12 13 14

[…] ˹x.x.x˺ [g]ur ˹gar? x˺ [x (x)] An-za-gar3-ta kun-zi-da id2 / a-gar3 us2-sa a-šag4 / lu2-mah-še3 ib2-ga6 u3 ma2-a si-ga [u]s2-bi 36 nindan

From […] of Anzagar to the barrage of the canal at the agricultural tract, next to the (Eduru-)Lumah field, they carried (the barley) and then loaded it on boats. The distance was of 216 meters.

[ku]n-zi-da / An-za-gar3-ta […] (3 lines lost) […] ma2 še gid2-[da … ] us2-bi 290 [nindan]

From the barrage of Anzagar … they towed the boats with the barley. The distance was of 1,740 meters.

22 ½ guruš ud-[1-še3] a-ga2-la2 keš2-[ra2]

22 ½ workers tied sacks during [1] day.

22 ½ guruš ud-1-[še3] Gar-ša-an-naki-[ta] U3-dag-gaki-[še3 gen-na]

22 ½ workers walked [from] Garšana [to] Udaga during 1 day.

29 10 gin2 g[uruš] 6 ma2 15.0.0 gur [0.0.2 5 sila3] 10 ma2 10.0.0 gur [0.0.2.] ud 2-[še3] A-pi4-sal4-[laki-ta] U3-dag-[gaki-še3] ma2 su3 [dirig-ga?] še ma2-[a si-ga]

29 ⅙ workers [floated down?] empty boats [from] Apišal [to] Udaga: 6 15-gur boats (whose rental fee is) 25 sila each and 10 10-gur boats (whose rental fee is) 20 sila each, during 2 days, and they loaded them with barley.

15 16 17 17 ii. 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16

kuš

9 Wages might be calculated on the basis of CUSAS 3: 561 (cf. Heimpel 2009: 122): 6⅔ guruš arad2 e2-a × 2 sila3/day × 3 days = 40 sila3 15 guruš hun-ga2 × 8 sila3/day × 3 days = 360 sila3 1 10-gur boat × 20 sila3 × 3 days = 60 sila3 1 8-gur boat × 15 sila3 × 3 days = 45 sila3 1 5-gur boat × 5 sila3 × 3 days = 15 sila3 Total: 520 sila3.

New Data on Garšana and the Border Zone between Umma and Girsu/Lagaš

17 18 19 20

rev. i. 1′ 2′ 3′ 4′ 5′ 6′ 7′ 8′ ii. 1′

239

U3-dag-[gaki-ta] kun-zi-[da] A-pi4-[sal4-laki-še3] ma2 še [gid2-da] […] (3 lines lost)

They [towed] the boats [from] Udaga [to] the weir of Apišal …

[…] (5 lines lost) 4 m[a2 x.0.0 gur …] 1 m[a2 x.0.0 gur …] ud [n-še3] še ma2-a [si-ga] kun-zi-d[a (…)] / A-pi4-sal4-laki[ta] Gar-ša-an-naki-[še3] ma2 še [g]id2-da

… 4 n-gur boats, 1 n-gur boat. They loaded the boats. The boats with the barley were towed [from] the weir of Apišal [to] Garšana.

[ …] ˹x x˺ […] (rest of col. lost) […] (ca. 4 lines lost) 0.1.0 8 s[ila3 š]e / kaš sa10 ma2lah5-/e-ne

2′

6.1.1 4 sila3 / še gur

3′ 4′ 5′

še U3-dag-ga[ki-ta] [Gar]-ša-an-na[ki-še3] de6-[a]

6′

[gi]ri3 diškur.[illat]

7′ 8′

[it]i zahx-da-[gu7] mu d[Šu-dSuen …]

… 68 sila of barley (and) beer, purchased for the boatmen. 6 gur and 74 sila of barley, transported [from] Udaga [to] Garšana.

Under the responsibility of Adadillat. Month iii. ŠS …

Unfortunately, the interpretation of this text is not free of difficulties. The main problem is the fact that the endings of lines in columns obv. ii and rev. i are missing, which makes it difficult confidently to reconstruct the directions of trips. Still, it appears certain that CUSAS 3: 553 describes four separate operations, which were conducted along the same waterway: a) A group of workers walked from Garšana to Udaga and filled there sacks with barley. They then transported the boats (with barley) via Anzagar to an unspecified location (possibly Girsu). It appears that the boats had to

240

Manuel Molina and Piotr Steinkeller

be unloaded – and then loaded again – at the barrage of Anzagar (obv. i 1– ii 3). b) A group of workers walked from Garšana to Udaga and filled there sacks with barley. Another group of workers brought empty boats to Udaga from Apišal. Subsequently, they loaded the boats with barley and transported them back to Apišal (obv. ii 4–20). c) A number of boats were loaded with barley, to be then transported from Apišal to Garšana (rev. i 1′–ii 1′).10 d) A volume of barley was carried on foot from Udaga to Garšana (rev. ii 2′– 5′).

3 The Settlement of Dusabara The political and economic reality behind these operations will be better understood if one considers the fact that, during the reign of Šu-Suen, the fields surrounding Udaga belonged to the household of Lugal-kugzu, whose properties were administered by Ur-Meme. As it will be seen below, Lugalkugzu most probably belonged to the royal entourage, and Ur-Meme managed his household using Dusabara (Du6-sa-bar-ra(ki)), a rural settlement, as a storage and distribution center. Its grain-storing facilities (i3-dub, guru7) are frequently mentioned in the surviving texts. Dusabara was located somewhere between Girsu and the ancient course of the Tigris, a short distance away from the border with the Umma province. This can be deduced from the following data: a) The field of Šulpa’e was controlled from Dusabara by Ur-Meme, who ultilized the šuku plots of their holders as a means of repaying their personal debts (TMH NF 1/2: 247; 249; 250; 253; 254; NATN 748; PDT 2: 932; 933; cf. Steinkeller 2002: 122 f.). The field of Šulpa’e, not faraway from Apišal (RA 79: 30 no. 23), was located along the border between the Umma and Lagaš provinces, and was therefore administered both by the governor of Girsu (MVN 11: 90, Š33) and by the provincial administration of Umma (passim). b) Dusabara was located close to Eduru-Ninazida (PPAC 5: 619; Nisaba 7: 12), a hamlet with a grain silo and a field. Eduru-Ninazida lay on the Tigris

10 Since the second operation involved the shipping of barley to Apišal, it is strange that the barley should now be transported from Apišal to Garšana. This difficulty could be removed if one reconstructs A-pi4-sal4-laki-[še3] / Gar-ša-an-naki-[ta] in rev. i 5′–6′. But this would violate the pattern by which directions usually are expressed.

New Data on Garšana and the Border Zone between Umma and Girsu/Lagaš

241

(TCTI 1: 742 rev. i 5), not faraway from Girsu (Nisaba 7: 12) and close to Apišal, possibly on the confluence with the Nadu’a-igidu-canal (ASJ 13: 224 no. 70; 225 no. 71). c) Dusabara was situated in the Girsu area (PPAC 5: 984). Documents mentioning Dusabara can be divided into two gropus, which provide an interesting diachronic view of its history: a) The first group consists of some fifteen texts coming from Girsu dated from Š45 (TUT 177) or Š46 (MVN 10: 128) to AS1. Texts from this group record deliveries of barley mostly from Bazi, an official who worked for the provincial administration (his seal, dedicated to the gorvernor Ur-Lamma, was rolled on CBT 3: BM 27084 and MVN 6: 257). The activities of Bazi are mainly attested from Š35 (MVN 7: 175) to AS2 (ASJ 14: 241 no. 89), although he occasionally appears also in the sources from AS7 to AS9. Bazi was responsible for the delivery of loans of barley to eren2 people (MVN 10: 128; MVN 12: 245; Nisaba 10: 87; 74; AAICAB 1/3: Bod. A 37 obv. iii 3), most probably royal dependents, since the deliveries were frequently received or supervised by members of the military (Nisaba 10: 61–62; PPAC 5: 1117; Nisaba 17: 81). Cereals were also supplied by Bazi (and occasionally by other officials, such as Lugal-zuluhu), from the silo of Dusabara, as barley allotments and for other purposes (zi3-ka-še3 [MVN 12: 84; MVN 12: 102], sa2-du11 še-ba [MVN 12: 100], šag4-gal he2-dab5 [MVN 9: 6], še-ba a-bala du3-a-kud, še-numun [BM 13002A unpubl.; PPAC 5: 964], še-ba didli [MTBM 327], etc. [PPAC 5: 619; CT 10, pl. 38 BM 15296]). During those years, goods stored at the silo of Dusabara were administered together with those from other silos of the same area, namely the silos of Eduru-Ninazida (PPAC 5: 619), the city of Girsu, Eduru-Inanna, Kirizuhara, Sipa-dari, and the Sugan field, as shown by PPAC 5: 984, a balanced account of barley belonging to Hala-Lamma, most probably the daughter of the governor Lu-kirizal (RIM E3/2.1. 2. 2012). This text, dated to Š47, records the expenditures of barley from the above-mentioned silos for the governor of the province and for a certain Dudu, no doubt identical with the šabra of the household of Ningirsu (cf. Borrelli 2014: 140 f.). b) In the second group of texts, dated between ŠS4 and ŠS5 and apparently stemming from Nippur, Dusabara likewise appears as a rural center with storage facilities. During this period, one of the chief officials in charge of Dusabara was Ur-Meme, also known for his role as administrator (šabra) of the households of Lugal-kugzu, Amar-Suen, Nin-munuszida, and ŠatŠu-Suen (NATN 59; TMH NF 1/2 171; 316). Part of the activities of Ur-Meme at Dusabara involved his private business activities, as shown by some

242

Manuel Molina and Piotr Steinkeller

fifty documents recording transactions related to loans privately made by himself (Steinkeller 2001b: 54; 2002: 122 f.). Therefore, it appears that UrMeme took advantage of a pre-existening distribution network, based at Dusabara, which had earlier been managed by the provincial administration, and from which loans of barley and other types of expenditures were being issued since the time of Šulgi. Besides, Dusabara retained its role of a storage and redistribution center, from where cereals were sent to other localities such as Uruk or Ur (NATN 461; TMH NF 1/2 131; 293). It is possible that Ur-Meme was administatively responsible for that kind of activity as well, but no proof of this exists. During both periods, the Dusabara center was associated with similar agricultural areas. This can be deduced, for example, from the fact that Bazi was responsible for the Aballa field (BPOA 1: 207; CT 7, pl. 14 BM 12945; MTBM 329; BM 94502, unpubl. letter-order), whose administration fell under the authority of the provincial administration and the sanga’s of the temple households at least since Š33 (CT 9, pl. 17 BM 12917; MVN 6: 539; ASJ 8: 113 no. 30; PPAC 5: 605; MVN 2: 8). Later on, the Aballa field became part of the household of Lugal-kugzu (ŠS1–5: JCS 24: 163 no. 68; JCS 54: 6 no. 41; NATN 444; 447; 451; TMH NF 1/2 88; 126), who very likely belonged to the royal entourage. The field of Aballa was located within the district of Girsu (CT 5, pl. 36 obv. ii 10), in the vicinity of the Ukunuti field (SNAT 511), and therefore very close to the border with the Umma province (cf. Steinkeller 2013: 305). The overall size of the Aballa field was around 140 bur3 1 eše3 4 iku, roughly 910 ha = 9 km2 (Borrelli 2013: 153).11 A similar land-tenancy shift occurred in the case of other fields and properties situated in the area of Dusabara that had earlier been managed by temple households and the provincial administration. Thus, the household of Lugalkugzu controlled also the fields of Kiri-zuhara, Sagub, Udaga (NATN 447; 451), Lu-Haia, Nag-esira, Sugan, and Sipa-dari (NATN 35; 104; 727; 739; 568; PDT 2: 921+), extensive gardens and orchards (NATN 568, etc.), and herds of sheep and cattle. Most of these properties (if not all of them) were managed by UrMeme, whose duties are particularly well described in two pisan-dub-ba texts (TMH NF 1/2: 316; NATN 451): he was supposed to provide fodder for draught animals, seed for sowing, and wages for the hirelings performing hoeing and

11 The distance (Farmer’s Instructions 7.06) from the Aballa field to the field of Hazi was of 4 da-na (43.2 km), and from Barasiga to the Aballa field the distance was of 2 da-na (21.6 km). For the time being Hazi cannot be located. Barasiga, probably identical with Barasiga of Hurim (RTC 399; Fs. Sjöberg: 61 rev. ii 8′, etc.), belonged to the district of Gu’abba.

New Data on Garšana and the Border Zone between Umma and Girsu/Lagaš

243

weeding; barley allotments for gardeners and orphans; wool, oil and date allotments for shepherds and gardeners; wages for the conscripts and hirelings involved in the maintenance of irrigation canals; and the allotments of subsistence land. It appears that Ur-Meme was the owner of a seal that reads: Ur-Me-me, dub-sar, dumu Ur-dSamanx(bu.nun.še.še3)-[ka?] (NATN 455; 757; TMH NF 1/2: 148; cf. TMH NF 1/2: 82+, etc.), which is also found on a tablet from the Aradmu archive (no. 59, to be published by Studevent-Hickman), in a small locality in the vicinity of Nippur. In this text Ur-Meme bears the title of šabra, which makes the identification very likely. This occurrence and the fact that most of the texts dealing with Ur-Meme’s activities were recorded in the museums as coming from Nippur suggest that Ur-Meme lived and/or had his private archive there. The dating of his texts with the Reichskalender (rather than with the Girsu month names) is explained by the fact that, as pertaining to the royal economy, they did not belong to the provincial archives. The identity of Lugal-kugzu is more difficult to ascertain. He no doubt was a member of the royal circle, but no explicit links with the royal family are traceable. He may tentatively be identified as the important chief archivist of the Puzriš-Dagān treasure archive (cf. Sallaberger 1999: 246 f.; Paoletti 2012: 106–109), who at least once hosted the king (AUCT 1: 793), but there are no clear data supporting this identification. Apart from the household of Lugal-kugzu, Ur-Meme also managed the estates of Nin-munuszida, Šat-Šu-Suen, and Amar-Suen (NATN 35; 59; 739; PDT 2: 921+; TMH NF 1/2: 171). These four households shared at least some of the above-mentioned fields, i.e., Kiri-zuhara, Sagub, Udaga, Lu-Haia, Nag-esira, Sugan, and Sipa-dari. Šat-Šu-Suen was an ereš-dingir priestess of Šu-Suen (JAOS 126: 166–67 KM 89100; CT 32, pl. 12 BM 103436 obv. ii 6), whose name was occasionally abbreviated as Šat-Suen (NATN 59; TCS 1: 237). Conceivably, she could be identical with Šat-Suen, daughter of Šulgi (BPOA 7: 2668, etc.). Concerning Ninmunuszida, because of the proximity of their respective households, she probably was a relative of Šat-Suen and/or Lugal-kugzu, but our texts do not provide further details about her identity. Finally, it is highly likely that the estate of Amar-Suen (NATN 35; TMH NF 1/2: 171) was identical with the important household of Namhani (Maekawa 1986: 96 f.), which was situated along the eastern border of the Umma province (Steinkeller 2011: 381 f.). In fact, these rural royal estates and Dusabara were all located within the same broad area, which extended from the ancient course of the Tigris southwards along the border between the Umma and Girsu/Lagaš provinces, not faraway from the city of Girsu. This is also the area where Garšana, a royal

244

Manuel Molina and Piotr Steinkeller

settlement on the other side of the border, evidently was situated. Texts show that during the second half of Šulgi’s reign, until Aradmu assumed the governorhip of the Girsu/Lagaš province (by the end of AS7 at the latest, but probably considerably earlier), this zone was administered by temple households and the provincial administration. It appears that it was around that time that the chancellor and the new governor instituted a policy by which large areas of agricultural land were taken away from the provincial domain, to be distributed among the various types of royal dependents.12 The area most affected by this new policy apparently was the stretch of land extending along the border with Umma, which directly adjoined the town of Garšana and its surrounding terrirories. This development no doubt was part of a larger strategic scheme, since Garšana not only was the biggest royal settlement in the Umma province, but it also was governed by the same Aradmu, being economically supported from Girsu (Steinkeller 2011: 376 f.). Since Zabalam likewise was a royal settlement, it is not surprising that it too shared a close rlationship with Garšana, both economically and administratively. However, against Heimpel,13 there are no textual data indicating in any way a geographical proximity between these two towns. 12 Similar developments took place, likewise in the beginning of Amar-Suen’s reign, in the province of Umma. See Steinkeller forthcoming. 13 The distance from Garšana to Zabalam is deduced by Heimpel (2011) from CUSAS 3: 258 and 246, where one-day trips by foot are apparently recorded: 4 geme2 ud 1-še3, nig2-gu2-na Zabalam4[ki]-še3 d[e6-a] (CUSAS 3: 258); 2 guruš ud 1-[še3], gi Zabalam4[ki-še3] de6-[a] (CUSAS 3: 246). Heimpel’s interpretation appears to be incorrect: first, because man-days are most probably meant in these two texts (Steinkeller 2011: 389); second, because the verb de6 does not necessarily imply a transportation by foot; and third, because texts do not specify what kind of operation did these workers perform. The same may be said regarding Garšana’s geographical position vis-à-vis Karkar, which likewise was a royal settlement. Heimpel (2011: 153) estimates that the distance between Garšana and Karkar was covered on foot in one day on the basis of CUSAS 3: 884 and 237. As Steinkeller (2011: 388) wrote, the days recorded in these texts were abstract notations (mandays), and Heimpel accepted the plausibility of this argument for CUSAS 3: 884. He nevertheless rejected it on the basis of CUSAS 3: 237, where workers were recorded as receiving full wage (sag-tag), ⅔ wage and ⅓ wage, notations that would not be compatible with the expression n guruš ud 1-še3 understood as “n man-days.” Nevertheless, a photo of the latter text (CDLI P324424) shows that the transliteration of ud 1-še3 for line 5 of the tablet was a misreading in CUSAS 3: 237, being ud 2-še3, or more probably ud 3-še3, the correct reading: 9 guruš sag-tag, 1 guruš a2 ⅔, 1 guruš a2 ⅓, azlag7-me-eš2, ud 3(or 2)-še3, še E2-kab-baki, [Kar]karki-ta, [Ga]r-š[a-a]n-n[a]˹ki˺-še3 de6-a. “9 full-wage workers, 1 ⅔-wage worker, 1 ⅓wage worker. They are fullers. During 3 (or 2) days they brought the barley of E(duru)-Kabba from Karkar to Garšana.” (CUSAS 3: 237 = CDLI P324424) On the other hand, it is not entirely clear that a trip by foot is being recorded in these texts. CUSAS 3: 884 records in fact a transportation of grain on foot, but it could also refer to

245

New Data on Garšana and the Border Zone between Umma and Girsu/Lagaš

From a geopolitical point of view, it is clear that there was no neat cut between the provinces of Umma and Girsu/Lagaš. Large fields extended over the border and were managed by the administrations of both provinces. This was demonstrated by Steinkeller for the Ukunuti field (2011), which laid to the north-northeast of Anzagar, and for the Ušgida field (2013), located to the south of the Ukunuti field. The same can now be said also about the Šulpa’e and the Udaga fields. The situation of Garšana, close to the Ušgida field and upstream from Udaga, no doubt was similar. As concerns Garšana’s exact location, all that can be said at this point is that it could be identical with Tell Baridiyah or any other site located in that general area (Steinkeller 2013: 306–307): the

a transportation from the quay of Garšana to the storehouse, as in CUSAS 3: 235 (see below): 1 guruš ud 1-še3 še ga6-g[a2], 3 guruš ud 1-še3, še E2-kab-[baki], Karkarki-t[a (…)], Gar-šaan-naki-[še3], ib2-ga6. “1 worker during 1 day (= 1 man-day) carried barley. 3 workers during 1 day (= 3 man-days) carried the barley of E(duru)-Kabba, (on the occasion of a trip?) from Karkar to Garšana.” (CUSAS 3: 884) Except for this text, trips are recorded as being made by boats towed from Garšana to Karkar, and abstract notations (n workers ud 1-še3) are used (CUSAS 3: 235; 884). There is nevertheless one exception that needs to be explained. CUSAS 3: 236 reads: 52 guruš ugula Ba-zi, 24 guruš ugula Za-la-a, šu-nigin2 76 guruš, šag4-bi-ta, 76 guruš, 4 ma2 10.0.0 gur-ta, Gar-ša-an-nakita, Karkarki-še3 gid2-da, ud 1-am3 ma2 gar-ra, ud ⅔-am3 ma2 dirig-ga, gu-kilib-bi 647 ½. “52 workers (= man-days) whose foreman is Bazi, 24 workers (= man-days) whose foreman is Zala’a; total 76 workers (= man-days). Out of them: 76 workers (= man-days) towed 4 boats of 10 gur of capacity each from Garšana to Karkar, loaded the boats at full wage, floated the boats at ⅔ wage. The bales (of shok they cutted?, loaded and transported) were 647 ½.” In Heimpel’s view (2009: 300 f.), this text complements CUSAS 3: 235, which reads: 10 gu-kilib giš.u kiši17, a2 guruš-a 4 gu-kilib-ta, a2-bi 2 ½ ud 1-še3, a-šag4 Karkarki-še3 giš.u kiši17 kudra2, 4 guruš ud 1-še3 Gar-ša-an-naki-ta, Karkarki-še3 ma2 gid2-da, 4 guruš ud 1-še3 m[a2 ga]r?-ra, 4 guruš ud 1-še3, Kar[karki-ta (…)], Gar-ša-an-naki-še3, 4 guruš ud 1-še3, kar-ta ga2-nun-še3 ga6-g[a2], ugula Šu-dDumu-zi. “10 bales of shok, the wage of a worker is for 4 bales each, the wage is for 2 ½ days. (The workers went) to Karkar (and) cut shok. 4 workers during 1 day (= 4 man-days) towed the boat(s) from Garšana to Karkar. 4 workers during 1 day (= 4 man-days) loaded? the boat(s). 4 workers during 1 day (= 4 man-days) (floated de boat) from Karkar to Garšana. 4 workers during 1 day (= 4 man-days) carried (the shok) from the quay to the storehouse.” CUSAS 3: 236 shows an unusual formulation. Heimpel interprets it as “a plan of operation drawn up by Adad-tillati,” in which the expression ud ⅔-am3 ma2 dirig-ga denoted a trip from Karkar to Garšana that took ⅔ of a day. Be it a “plan of operation” or a real account of work credited to two foremen (which seems to be more plausible), man-days were probably also accounted in this text. Thus, besides the anomaly of not specifying the duration of the trip from Garšana to Karkar, 76 workers towing four 10-gur boats would make a total of 19 workers per boat, which seems to involve too many people. It would be better to assume that 76 man-days due by two foremen were wholly satisfied on the basis of full wages for towing the boats from Garšana to Karkar and loading them, and of ⅔ wages for floating the boats back to Garšana. The operation resulted in the transportation of 647 ½ bales of shok to Garšana. 2

2

246

Manuel Molina and Piotr Steinkeller

stretch of land extending a few kilometers northward along the Udaga canal, and not too faraway from the Girsu/Namhani canal.14

14 The proximity of Garšana to the Girsu/Namhani canal was proved by Steinkeller (2011) on the basis of OrSP 47/49: 382 (see Steinkeller 1987), where the Umma forests appear grouped in three big sections controlled by three foremen, reflecting their geographical distribution within the Umma province: 11 12 13 14 15 16 17 18 19 20

E2-lugal Kar-ra Ma-sar Tu-ru-da Nag-suki Gir13-giški Ka-e3 Bala-a-ti-im-ku.ku ki-sur-ra Adabki Zabalamki

Šeškala " " " " " " " " "

21 22 23 24 25a 25b 26 27 28 29 30

Id2-gal-la tir Gar-ša-na-kaki tir eren2 Gar-ša-na-kaki tir Šag4-nin9-ti-na tir gaba-ri Gar-ša-na-kaki tir ma2-pad-dirig gu2 id2 Nam-ha-ni tir Uku2-nu-ti gu2 id2 dŠul-gi-he2-gal2 tir uru-bu3-ra tir A-du10-ga-nigin tir Nig2-su4-da tir Maš:gan2ki

Ur-Šara " " " " " " " " " "

Heimpel (2011: 154–56) has questioned the validity of Steinkeller’s hypothesis aducing the existence of a double ruling between 25a and 25b, verified after collation. This blank line would have corresponded to the person in charge of forest 25b, who for some reason was not recorded. Therefore – Heimpel argued – forests 25a and 25b were not conflated into a single one and their proximity would not be proven. This would also be supported by BPOA 2: 2685, where the forests are listed in a different order. Nevertheless, the existence of a blank line between the name of forests 25a and 25b does not affect Steinkeller’s position: the fact that the verbal form i3-dab5 was written after the forest name 25b would indicate that the latter was under the responsibility of the same family as forest 25a, and therefore the two forests were conceivably in the same area. They have been identified by Steinkeller (2011: 389) as the forests on the opposite sides of the Girsu/Namhani canal recorded in BPOA 2: 2685 (tir gu2 id2 Nam-ha-ni gu2 a2 2-a-bi). On the other hand, for its structure, purpose, and contents, OrSP 47/49: 382 was clearly arranged on a geographical basis (Steinkeller 1987: 76 f.), while BPOA 2: 2685, a list of allotment fields held by Umma foresters, was written with a different motivation and with a different structure. BPOA 2: 2685 (BM 114689) was collated from the photographs kindly provided by Enrique Jiménez.

New Data on Garšana and the Border Zone between Umma and Girsu/Lagaš

Fig. 1a: Green Collection, Tablet 2013-315 obv.

247

248

Manuel Molina and Piotr Steinkeller

Fig. 1b: Green Collection, Tablet 2013-315 rev.

New Data on Garšana and the Border Zone between Umma and Girsu/Lagaš

249

Bibliography Borrelli, Noemi. 2013. Managing the Land: Agricultural Administration in the Province of Ĝirsu/Lagaš During the Ur III Period. Ph. D. Diss. Università degli Studi di Napoli “L’Orientale.” Heimpel, Wolfgang. 2009. Workers and Construction Work at Garšana. CUSAS 5. Bethesda: CDL Press. Heimpel, Wolfgang. 2011. On the Location of the Forests of Garšana. Pp. 153–59 in Garšana Studies, ed. David I. Owen. CUSAS 6. Bethesda, MD: CDL. Maekawa, Kazuya. 1986. The Agricultural Texts of Ur III Lagash of the British Museum (IV). Zinbun 21: 91–157. Paoletti, Paola. 2012. Der König und sein Kreis. Das staatliche Schatzarchiv der III. Dynastie von Ur. BPOA 10. Madrid: CSIC. Rost, Stephanie. 2015. Watercourse Management and Political Centralization in ThirdMillennium B. C. Southern Mesopotamia: A Case Study of the Umma Province of the Ur III Period (2112–2004 B.C.). Ph. D. Diss. Stony Brook University. Sallaberger, Walther. 1999. Ur III-Zeit. Pp. 119–390 in Mesopotamien. Akkade-Zeit und Ur III-Zeit, ed. Pascal Attinger and Markus Wäfler. OBO 160/3. Freiburg Schweiz: Universitätsverlag / Göttingen: Vandenhoeck & Ruprecht. Steinkeller, Piotr. 1987. The Foresters of Umma: Toward a Definition of Ur III Labor. Pp. 73– 115 in Labor in the Ancient Near East, ed. Marvin A. Powell. AOS 68. New Haven, CT: AOS. Steinkeller, Piotr. 2001a. The Ur III Period. Pp. 47–62 in Security for Debt in Ancient Near Eastern Law, ed. Raymond Westbrook and Richard Jasnow. CHANES 9. Leiden: Brill. Steinkeller, Piotr. 2001b. New Light on the Hydrology and Topography of Southern Babylonia in the Third Millennium. ZA 91: 22–84. Steinkeller, Piotr. 2002. Money-Lending Practices in Ur III Babylonia: The Issue of Economic Motivation. Pp. 109–37 in Debt and Economic Renewal in the Ancient Near East: A Colloquium Held at Columbia University (November 1988), vol. III, ed. Michael Hudson and Marc Van De Mieroop. Bethesda, MD: CDL. Steinkeller, Piotr. 2007. City and Countryside in Third-Millennium Southern Babylonia. Pp. 185–211 in Settlement and Society: Essays Dedicated to Robert McCormick Adams, ed. Elizabeth C. Stone. Los Angeles: UCLA / Chicago: The University of Chicago. Steinkeller, Piotr. 2011. On the Location of the Town of Garšana. Pp. 373–90 in Garšana Studies, ed. David I. Owen. CUSAS 6. Bethesda, MD: CDL. Steinkeller, Piotr. 2012. More on the Reading of the Toponym Garšana. NABU 2012/42. Steinkeller, Piotr. 2013. The Umma Field Ušgida and the Question of Garšana’s Location. Pp. 295–308 in Beyond Hatti: A Tribute to Gary Beckman, ed. Billie J. Collins and Piotr Michalowski. Atlanta, GA: Lockwood. Steinkeller, Piotr. Forthcoming. An Estimate of the Population of the City of Umma in Ur III Times. In At the Dawn of History: Ancient Near Estern Studies in Honour of J. N. Postgate, ed. Y. Heffron et al. Winona Lake, IN: Eisenbrauns.

David I. Owen

A New Silver Balanced Merchant Account from Umma For Miguel Civil on his entry into his ninth decade, in friendship and with admiration

Silver balanced merchant accounts were the subject of pioneering studies by John B. Curtis and William W. Hallo (1959) and later by Daniel Snell (1982). Subsequent publications by Hans Neuman (1987), D’Agostino and Pomponio (2004), Claus Wilcke (2007), Steven J. Garfinkle (2012), D’Agostino and Pomponio (2014), among others, and the recent comprehensive study on the use of silver at Umma by Xiaoli Ouyang (2013), have each added significant new data to, and interpretations of, the use of silver in the economy of the Third Dynasty of Ur. Furthermore, the role of entrepreneurs and merchants, who utilized silver as a means of exchange during the Third Dynasty of Ur, has been clarified substantially by both Garfinkle and Ouyang. The addition below to the small corpus of silver balanced merchant accounts adds to the data1 for those merchants and other officials involved with the use of silver in the trade of aromatics, “luxury” goods, metals and stones, agricultural products, etc. and follows the conclusions offered by Snell on the values of specific commodities now brought up-to-date by Ouyang. Unfortunately, the first (I) and last columns (VI) of the new text are missing and, as a result, both the indication of the type of account, presumably a balanced account (niĝ2-kas7-aka), and its colophon and date are lost. However, its format and contents indicate clearly that it is a silver balanced merchant account with an Umma provenance. Unfortunately, as with virtually all of the published Umma silver balanced merchant accounts, the tablet lacks an excavated context. Nevertheless, in spite of recent assertions to the contrary,2 texts without archaeological context continue to contribute much to our understanding of ancient Mesopotamian culture and particularly to our understanding and reconstruction of its economic system(s).3 1 See also BM 110126 recently published in al-Rawi, Gorello and Notizia, Nisaba 26 (2013), no. 2 and the study by D’Agostino and Pomponio 2014. 2 Hanson 2008, passim. 3 I have addressed repeatedly (Owen 2013: 335–56 and in my Prefaces to CUSAS volumes) the baseless assertions that texts without excavated context have no value and should not be published. Up until the 1970 declaration by UNESCO, the utilization of unprovenanced texts and artifacts was widely accepted as an integral component of all research and publication and provided much, if not most, of our knowledge of the history, literature, and culture of https://doi.org/10.1515/9781501503696-015

A New Silver Balanced Merchant Account from Umma

251

Only Lugal-niĝlagare, mentioned in this fragmentary account, is attested as a merchant in the published Umma archival sources and especially in the silver balanced merchant accounts. Given the fact that the aromatics and the other commodities mentioned in this text are usually associated with merchant accounts it is likely that some of the other individuals might be identified with otherwise well-known merchants. In any case, all of the individuals mentioned in the text are known also from silver merchant accounts discussed by Snell (1982), Neumann (1987), and Ouyang (2013). Merchant’s price equivalency, balanced account 125 × 65 mm Umma provenance date lost Col. I. Lost 4 II.  1.  2.  3.  4.  5.

[n …] [ku3-bi n] gin2 [5 ĝešp]eš3 še-er-gu5 [ku3-b]i igi-6-ĝal2 6 še [kišib3] lu2-dnin-šubur6

Mesopotamia. It continues to do so today (note in particular the seminal contributions via the CUSAS series) in spite of the efforts of a vocal and influential minority particularly, but not exclusively, in the United States, to suppress or otherwise ignore or impede such publication and research. 4 This tablet, now in a European private collection, was acquired along with notes by the late Wilfred Lambert. It is published with the kind permission of its current owner. According to Lambert’s notes the tablet likely contained three columns on each side. The numbering of the columns follows his observations. This publication is based entirely on a series of detailed photos provided by the owner and not on an examination of the original tablet. Its format can be compared with the three-column texts copied in Snell 1982: plates xv-xx, nos. 9–11. 5 “A string of figs.” Lu-Nin-Šubur, who apparently deals with fruits, dealt with the same item in Grégoire 1996: Ashm. 1924–0667 rev. ii 7–8 (AS 5/xi/-). Restored amount based on value of 4 ĝešp]eš3 še-er-gu ku3-bi igi-4-ĝal2 3 še, ibid., Ash. 1924–0667 rev. ii 7. See also the following note and Ouyang 2013, pp. 134–135 and 283. 6 The name Lu-Nin-Šubur is relatively common at Umma. A single, questionably restored reference suggests that he was a merchant in AS 8/vii/- (AnOr 1: 134:5 [= BDTNS 007702 = CDLI P101125). Another lone reference indicates a man by that name is the son of Lugal-niĝlagare (AnOr 7: 374 ii′ 10′) who is likely the same person as the merchant by that name (see note 7). See also Snell 1982, p. 249 s.v. But the evidence is hardly conclusive.

252

 6.  7.  8.  9. 10. 11. 12. 13. 14. 15.

David I. Owen

[1.2.5].4 sila3 u2gamun27 [ku3-b]i 3 ¹⁄₃ gin2 12 / še [kišib3 lugal-niĝ2-lagar-e8 [n ĝeš]u3-suh5 ĝešmi-ri2-/za9 [ku3-b]i igi-4-ĝal2 [kišib3] lu2-ur4-ša3-ga10 [n g]in2 šim-gan211 [ku3-b]i BLANK [kišib3] ur-dnun-gal 12 [n ma]-na šim(KWU-752)13

7 “Cumin,” for which see Steinkeller and Postgate 1992: 77 and Ouyang 2013: 133 sub 5.3.H. It was understood formerly only as “(a resin),” Snell 1982: 237 s. v., and “(a plant),” Owen 2013: 415 s. v. Lugal-niĝlagare deals with this commodity also in TCL 5 5680 rev. iii 16–17 (ŠS 2/-/-). On the basis of Sigrist and Ozaki 2009, 2664: 1–2 (0.2.4. u gamun2 ku3-bi 6 ½ gin2), 1 sila3 of u gamun2 = ~2.5 še. The restoration here is thus approximate. 8 Lugal-niĝlagare is attested as a merchant from Š 46/i/- (NYPL 335: 4) to ŠS 2/-/- (MVN 16: 685 5). He is the son of Nabasa (Sigrist and Ozaki 2009, 1995: 8, n. d.) and has a son Lu-duga (Ozaki and Sigrist 2006, 710:5, ŠS 2/v/4) who is also a gendarme of the governor according to his seal: lu2-du10-ga / aga3-us2-ensi2 / dumu lugal-niĝ2-lagar-e (UTI 4: 2823 seal, ŠS 2/vi/-). Note that in BIN 5: 77, 5–6, he dispenses ²⁄₃ sila3 u gamun2. See also Snell 1982, p. 246 s.v. 9 “Fir for (boat punting) poles,” for which see Snell 1982, p. 174 s. v. and “pine for (boat punting) poles,” Owen 2013: 395 and 416–17 s. v. Heimpel 2011: 103–09, §§ 3.2–3.2.3 translates ĝeš u3-suh5 ĝešmi-ri2-za as “plank pines.” See also Ouyang 2013: 288–290. The values for ĝešu3suh5 ĝešmi-ri2-za presumably vary according to their respective lengths; e.g., Ozaki and Sigrist 2006, 1877 obv. ii 8 and 14 and reflect quantities between 30 and 60 poles/planks per shekel. 10 The name, Lu-uršaga, is attested frequently from AS 2/-/- (Schneider, OrSP 47–49 [1930] 324 ii 4) to IS 3/-/- (MVN 20: 70, 2), the former where he deals in reeds, the latter where he deals in textiles, perhaps different individuals. Lu-uršaga dub-sar dumu Lugal-he2-ĝal2 musar-dšara2 deals in reeds but may be the same individual, son of Lugal-he2-ĝal2 who (i.e. the father) is a nagar tug2-du8 in OrSp 47–49: 324 i 19 (AS 2/-/-). A Lu-uršaga deals also in textiles but given the same names of the fathers, they both may be the same individual. In any case, he is the only one in our text who does not occur in any of the texts discussed in Snell 1982 and now only briefly in Ouyang 2013: 110, nor does he appear among those receiving timber products compiled by Ouyang 2013: 288–91. 11 See Brunke and Sallaberger 2010, p. 50 s. v., “a bad smelling aromatic (“schlechter Duft”),” contra Snell 1982: 232 s.v., “(a metal).” No value is assigned here so it is not possible to restore a quantity. Ordinarily the value is 1 sila3 = 5 še according to Snell 1982, plate 17 10 obv. ii 20– 21, 0.0.1.2 sila3 šim-gan2 ku3-bi 1 gin2. In all but one references to šim-gan2 the quantities are always in sila3, the only exception being Snell 1982, plate 17 10 obv. ii 32–33, 10 gin2 šimgan2 ku3-bi 3 še. 12 See Snell 1982: 253 s. v. Ur-Nungal was probably the scribe (Ouyang 2013: 69) and son of the archivist, Ur-Šara. (Ouyang 2013: 90 n. 223). 13 The price of šim appears to vary perhaps based on its quality. Cf. Snell 1991: 123 obv. iii 13, 2 ma-na šim ku3-bi ²⁄₃ gin2 and 123 obv. v 5–6, 2 ²⁄₃ ma-na šim ku3-bi ⁵⁄₆ gin2 10 še. 2

2

2

A New Silver Balanced Merchant Account from Umma

16. 17. 18. 19. 20. 21. 22. 23. 24. 25.

253

[ku3-b]i ¹⁄₃ gin2 12 še [n si]la3 še-li14 [ku3-b]i igi-6-ĝal2 6 še [n ma]-na gi15 [ku3-b]i 10-la2-1 še [kišib3? x]-bi-kam16 [1 ¹⁄₄ ma]-na eren17 [ku3-bi] 22 ½ še [kišib3 m]aš2-šu-gid2-gid218 BLANK SPACE

Lower Edge 26. [n] gin2 igi-6-ĝal2 22 ½ še

14 “Juniper berry / pine seeds (“aromatic”), Owen, 2013: 410 s. v. and Brunke and Sallaberger 2010: 51 s. v. Classed as a resin in Ouyang 2013: 132 sub 5.3.G. The price of šim appears to vary perhaps based on its quality. E.g. Snell 1982: plate 15 09 obv. ii 5–6, 1 sila3 še-li ku3-bi ½ gin2 and 15 09 obv. iii 2–3, 6 ½ sila3 še-li ku3-bi 3 gin2 igi-4-ĝal2, Snell and Lager 1991: 123 rev. i 23–24, 3 ²⁄₃ sila3 še-li ku3-bi ¹⁄₂ gín 20 še. 15 According to Brunke and Sallaberger 2010: 50 s. v., gi appears to be the equivalent of gi-i3 and gi-du10-ga, “sweet (smelling?) reed.” See also Ouyang 2013: 295–97. The value of gi appears to fluctuate, perhaps by quality. See, e.g. Snell and Lager 1991: 123 rev. ii 11–12, 7 mana gi ku3-bi ¹⁄₃ gin2 3 še, Sigrist 1984: 144 7–8, 5 ma-na gi ku3-bi igi-6-ĝal2, Sigrist and Ozaki 2009: 200 7–8, 1 ma-na gi ku3-bi 6 še, etc. 16 I am unable to restore any personal name associated with the balanced accounts. The only personal name known to me that ends with -bi-kam is šeš-kal-la dumu uru?-ki-bi-kam in the fragmentary Ĝirsu text, TCTI 2 2816 obv. v 4. But the break in our tablet appears to preclude this restoration. 17 “Cedar.” See Brunke and Sallaberger 2010: 49 s. v. for a discussion of the sign form and its meaning. It appears here without a ĝeš determinative. The value of eren varies perhaps by quality; e.g., 1 ma-na eren ku3-bi 18 še is found in Snell and Lager 1991: 123 obv. ii 7–8, obv. iv 8–9, but 18 ma-na eren ku3-bi 1 ²⁄₃ gin2 24 še, and rev. i 3–4, 15 ma-na eren ku3-bi 1 ²⁄₃ gin2. 18 The only “diviner” known by name from Umma is La-ma-ša (MCS 8: p. 96 BM 111750:2, ŠS 3/-/-). It is likely he is the “diviner” involved here. Note that in the BM text, he receives 5 minas of wool and is associated with Lu-kala in a text sealed by A’akala, governor of Umma. See Snell 1982: 260 s.v. and Ouyang 2013: 133 n. 574 where she points out that “some of the resin withdrawals went to the offering for the diviner.” This would be only the third such reference.

254

III.  1.  2.  3.  4.  5.  6.  7.  8.  9. 10. 11. 12. 13. 14. 15. 16. 17. 18. 19.

David I. Owen

ERASED SPACE 3 ½ gin2 16 še ku3 / niĝ2-sa10-ma du6-ku3-ga19 kišib3 lugal-niĝ2-lagar-e 3 ¹⁄₃ ma-na šu-/ur2-me20 ku3-bi ¹⁄₃ gin2 3 ¹⁄₃ ma-na ad2?21 ku3-bi ¹⁄₃ gin2 3 ¹⁄₃ ma-na gi ku3-bi igi-6-ĝal2 3 ¹⁄₃ ma-na šim22 ku3-bi 1 ¹⁄₃ gin2 3 ¹⁄₃ sila3 še-li23 ku3-bi ²⁄₃ gin2 0.0.1.3 ¹⁄₃ sila3 u2gamun2 ku3-bi ½ gin25 še 0.0.2. esir2-e2-a24 ku3-bi 24 še DOUBLE LINE niĝ2-dab5 ĝanun-lugal 25

19 “Silver for acquired merchandise,” ((erworbenes) “Kaufgut,” Wilcke 2007: 267) for the “cultic place (of Šara)” (Snell 1982: 89–90). Note that Lugal-niĝlagare certified transactions for various goods belonging to the “cultic place” in the large, Umma account, Grégoire 1996, Ashm. 1924–0667 rev. ii: 6 (AS 5/xi/-): niĝ2-dab5 du6-ku3-ga kišib3 lugal-niĝ2-lagar-e. In the following year, AS 6/-/- (TCS 361: 7), Lugal-niĝlagare was again involved with certifying goods for the “cultic place,” and once more two years later in ŠS 1/vii/- (Englund 1992: 99 no. 1 15– 16) he again certified goods for the “cultic place.” Sallaberger 1993: 253–54, suggests that the du6-ku3-ga was a cultic place dedicated to Enlil in Nippur. See also Ouyang 2013: 126 n. 473. 20 “Cypress,” for which see Brunke and Sallaberger 2010: 49 s. v. 21 “Myrtle,” for which see Brunke and Sallaberger 2010: 50 s. v. 22 “Aromatic substance.” The sign is clearly differentiated from KWU-759. See note 35 below.

23 “Juniper berry or pine nut.” See Brunke and Sallaberger 2010: 51 s. v. 24 “Wet bitumen,” for which see Stol 2012, passim and Ouyang 2013: 130 n. 526–132 referring to Heimpel 2009 as “bitumen of Ea” and not “house bitumen.” 25 The ĝanun-lugal is otherwise unattested but cf. ĝa2-nun e2-lugal (BPOA 1: 0652 4, etc.), ĝa2-nun gaba e2-lugal (BPOA 6: 4 2, etc.), and ĝa2-nun kar-lugal (UTI 6: 3821 5) that are well attested. The sign read ĝanun here is not entirely clear but the reading is likely as is its equation with the ĝa2-nun e2-lugal.

A New Silver Balanced Merchant Account from Umma

255

Lower Edge 20. 7 gin2 14 še Rev. IV 1. 101 gu2 esir2-had226 2. ku3-bi 10 gin2 18 še 3. kišib3 lugal-e-ba-an-sa627 4. 17 ma-na 2 ½ gin2 /uruda 5. ku3-bi 12 gin2 6. ⁵⁄₆ ma-na 5 gin2 su-/gan28 7. ku3-bi 1 gin2-la2-15 še 8. kišib3 lu2-den-lil2-la229 9. 0.0.2. 2 sila3 al-la-ha-ru30 10. ku3-bi 1 gin2 11. kišib3 a-kal-la ašgab31 12. 4 LUM-ba32 13. ku3-bi ¹⁄₃ gin2 14. 5 gin2 u2za-gul 33 15. ku3-bi 5 gin2 16. kišib3 dšara2-kam34 18. 1 gu2 im-babbar235

26 “Dry bitumen.” See M. Stol 2012, passim, and Ouyang 2013: 130–32 sub 5.3.F. Lugal-ebansa (scribe, son of Ur-Ištaran) deals with bitumen in BPOA 1: 812 4 (AS 9/-/-), Fish 1939: 38 6 (AS 9/-/-), and YOS 18: 123 rev. iii 24–28 (AS 9/-/-). 27 See Snell 1982: 246 s. v. Lugal-ebansa also delivers barley in Santag 6: 119 15 (AS 3/x/-); discussed in Ouyang 2013: 111–16. 28 For su-gan, “(a metal),” see Snell 1982: 232 s.v and “expensive additive in metallurgy,” Ouyang 2013: 127–28, 253–55. 29 See Neuman 1987: 115–18. He is possibly “a top administrator in the metal bureau in Umma,” Ouyang 2013: 127. Lu-Enlila generally receives metals and metal supplies from the Umma merchants. 30 “Mineral tanning agent,” Ouyang 2013: 138, Kleinerman and Owen 2009: 15 s. v. al-lu-haru-um and Owen 2013: 361 s. v. 31 Akala, the leatherworker, is attested frequently in the Umma archives from Š 46/-/- (AAS: 72:2) to IS 3/xi/- (DoCuEPHE: 24:5). See Snell 1982: 242 s. v. and Ouyang 2013: 89, 114, 134, and 138. 32 Possibly to be read sig4-ba, “turtle shell?,” for which see Snell 1982: 231 s.v. 33 za-gul when preceded by na4 is “carnelian.” But here it is preceded by the u2-determinative for plants. I know of no parallels. 34 Šara-kam is known also from merchant accounts. See Snell 1982: 251 s. v. 35 For gypsum, see Owen 2013: 385 n, 675, M. Stol 2012: passim, and Snell 1982: 224 s.v.

256

David I. Owen

19. ku3-bi 9 še 20. 0.0.1. naĝa-si-e336 21. ku3-bi 1 še 22. kišib3 lu2-kal-la37 23. 6 ²⁄₃ ma-na šim(KWU-752) 24. ku3-bi ²⁄₃ gin2 [n? še?] 25. 5 ma-na za-ba-[lum]38 26. ku3-bi ½ [gin2] 27. 5 ma-n[a …] 28. k[u3-bi n] (rest of col. IV lost) Col. V  1. 0.1?.0.39 bulugx (KWU-759 = šim×uh3)40  2. [ku3-bi] ¹⁄₃ gin2  3. [n ma]-na ni-gi4-/tum41  4. [ku3-b]i igi-6-ĝal2 še  5. [6 ma]-na šim-IM42  6. ku3-bi 18 še

36 “Sprouted (alkaline plant),” for which see Snell 1982: 231 s. v. 37 Lu-kala may be the same individual as the ĝiri3-official in Snell 1982: plate xxiv, no. 13 21 (ŠS 6/-/-). See also Snell 1982: 248 s. v. 38 “Juniper.” See Brunke and Sallaberger 2010: 49 s. v. 39 The reading and interpretation of the numerical sign, presumably written over an erasure, are reasonably certain. However, bulugx is known from only six texts in which the quantities are all in gu2 or ma-na (Brunke and Sallaberger 2010: 41, MVN 15: 127, Nisaba 15/1: 538 and 899, Santag 6: 119 and 373). Thus the quantity designation is either unique or erroneous.

40 For the reading of KWU-759, as bulugx/mulugx see Brunke and Sallaberger 2010: 50 s. v., and Ouyang 2013: 132 sub 5.3.G. The sign is clearly differentiated from the šim sign (note 18 above) and the internal uh3 (earlier misread as gu) is clear.

41 See Brunke and Sallaberger 2010: 50 s. v. ni-gi4-ib2. For a value of 6 ²⁄₃ ma-na = 80 še, see Snell 1982: plate 17 10 obv. ii 16. 42 For “(a resin),” see Brunke and Sallaberger 2010: 50 s. v., and Snell 1982: 235 s.v. Restored amount based on value of 5 ma-na = 20 še, Snell 1982: plate 17 10 obv. ii 6.

A New Silver Balanced Merchant Account from Umma

257

 7. [1] ma-na šim-hi43  8. [ku3-b]i 9 še  9. [1 ¹⁄₂ m]a-na tam2-še-lum44 10. [ku3]-bi 18 še  9. [¹⁄₃ si]la3 še-li45 10. [ku3]-bi igi-6-ĝal2 10-še 11. [5+?] sila3 šim-gan246 12. [ku3]-bi ¹⁄₃ gin2 7 še 13. [3? s]ila3 gam-gam-ma47 14. [ku3]-bi 12 še 15. [5 ¹⁄₃ s]ila3 gu4-ku-ru48 16. [ku3]-bi 8 še 17. BLANK SPACE 18. [5 si]la3 šim-hi-a49 19. [ku3-bi] ¹⁄₂ gin2 20. [n sila3] im-babbar2 21. [ku3-bi n] še /[. .]-x (rest of col. V lost) (col. VI lost)

43 For “(a resin),” and the reading with hi and not du10, see Owen 2013: 412 n. 783 and the remarks of Brunke and Sallaberger 2010: 50 s. v. See also Snell 1982: 234 s.v. Restored amount of šim-hi is based on the value indicated in Sigrist and Ozaki 2009: 200 obv. 11, 1 ma-na šimhi ku3-bi 10-la2-1 še. 44 Restored amount based on value of 2 ma-na = 12 še, Snell 1982: plate 17 10 obv. ii 5. 45 Restored amount based on value of 1 sila3 = 20 še, Snell 1982: plate 15 09 obv. ii 5. 46 Restored amount based on Snell 1982: plate 17 10 obv. ii 20, 0.0.1.2 sila3 šim-gan2 ku3-bi 1 gin2, and plate 19 11 obv. ii 11–12, 6 sila3 šim-gan2 ku3-bi ½ gin2, i.e. 1 sila3 = 5 še. At this price the restoration should be 5 ²⁄₅ sila2. 47 For the aromatic, “(terebinth),” see Brunke and Sallaberger 2010: 51 s. v., Owen 2013: 412 n. 780, and Snell 1982: 233–34 s.v and plate 17 10 ii 22–23, 0.0.1.6 sila3 šim gam.gam-ma ku3bi ½ gin2 6 še, for value and basis of estimated quantity. 48 For gukuru, “(an aromatic product from a tree),” see Brunke and Sallaberger 2010: 51 s. v., and Owen 2013: 412 n. 782. Restored amount based on value of 1 sila3 = 1 ½ še, Snell and Lager1991: 123 obv. v 22, rev. i 29. 49 Restored amount based on value of 1 sila3 = 6 še, Snell and Lager 1991: 123 obv. ii 13. šimhi-a may be a general designation for “mixed aromatics” or a specific aromatic to be differentiated from šim-hi.

258

Fig. 1: Obv.

David I. Owen

A New Silver Balanced Merchant Account from Umma

Fig. 2: Rev.

259

260

David I. Owen

Bibliography Brunke, Hagan and Walther Sallaberger. 2010. Aromata für Duftöl. Pp. 41–74 in Why Should Someone Who Knows Something Conceal it? Cuneiform Studies in Honor of David I. Owen on his 70 th Birthday, eds. Alexandra Kleinerman and Jack M. Sasson. Bethesda: CDL Press. Curtis, John B. and William W. Hallo. 1959. Money and Merchants in Ur III. HUCA 30: 103–39. D’Agostino, Franco and Francesco Pomponio. 2004. Due bilanci di entrata e di uscita di argento da Umma. ZA 94: 172–207. D’Agostino, Franco and Francesco Pomponio. 2014. A Third Balanced Account of Silver from Neo-Sumerian Umma, SEL 31: 1–24. Englund, Robert K. 1992. Ur III Sundries. ASJ 14: 77–102. Fish, Thomas. 1939. Some Sumerian Tablets of the Third Dynasty of Ur. JRAS 1939: 29–39. Garfinkel, Steven J. 2012. Entrepreneurs and Enterprise in Early Mesopotamia. A Study of Three Archives from the Third Dynasty of UR (2112–2004 B.C.E.). CUSAS 22. Bethesda: CDL Press. Grégoire, Jean-Pierre. 1996. Contribution à l’Histoire Sociale, Économique, Politique et Culturelle du Proche Orient Ancien. Archives Administratives et Inscriptions Cunéiformes de l’Ashmolean Museum et de la Bodleian Collection d’Oxford (AAICAB) I. Les Sources 1. Paris: Geuthner. Hanson, Katharyn. 2008. Why Does Archaeological Context Matter? Pp. 45–49 in Catastrophe: The Looting and Destruction of Iraq’s Past, ed. Geoff Emberling and Katharyn Hanson. Chicago: The Oriental Institute Museum of the University of Chicago. Heimpel, Wolfgang. 2009. The Location of Madga. JCS 61: 25–61. Heimpel, Wolfgang. 2011. Twenty-Eight Trees Growing in Sumer. Pp. 75–152 in Garšana Studies, ed. David I. Owen. CUSAS 6. Bethesda: CDL Press. Kleinerman, Alexandra and David I. Owen. 2009. Analytical Concordance to the Garšana Archives. CUSAS 4. Bethesda: CDL Press. Neumann, Hans. 1987. Handwerk in Mesopotamien: Untersuchungen zu seiner Organisation in the der Zeit der Ur III. Dynastie von Ur. Berlin: Akademie-Verlag. Ouyang, Xiaoli. 2013. Monetary Role of Silver and Its Administration in Mesopotamia During the Ur III Period (c. 2112–2004 BCE): A Case Study of the Umma Province. BPOA 11. Madrid: Consejo Superior de Investigaciones Científicas. Owen, David I. 2013. Cuneiform Texts Primarily from Iri-Saĝrig/Al-Šarrākī and the History of the Ur III Period, Vol. 1: Commentary and Indexes. NISABA 15/1. Bethesda: CDL Press. Ozaki, Tohru and Marcel Sigrist. 2006. Ur III administrative Tablets from the British Museum. Part One. Biblioteca del Próximo Oriente Antiguo 1. Madrid: Consejo Superior de Investigaciones Científicas. Sallaberger, Walther. 1993. Der kultische Kalender der Ur III-Zeit. Teil 1. Berlin: Walter de Gruyter. Sigrist, Marcel. 1984. Neo-Sumerian Account Texts in the Horn Archaeological Museum. Andrews University Cuneiform Texts 1. Berrien Springs: Andrews University Press. Sigrist, Marcel and Tohru Ozaki. 2009. Neo-Sumerian Administrative Tablets from the Yale Babylonian Collection. Part One. Biblioteca del Próximo Oriente Antiguo 6–7. Madrid: Consejo Superior de Investigaciones Científicas. Snell, Daniel. 1982. Ledgers and Prices, Early Mesopotamian Merchant Accounts. YNER 8. New Haven: Yale University Press.

A New Silver Balanced Merchant Account from Umma

261

Snell, Daniel and C. Lager. 1991. Economic Texts from Sumer. YOS 18. New Haven: Yale University Press. Steinkeller, Piotr and Nicolas Postgate. 1992. Third Millennium Legal and Administrative Texts in the Iraq Museum, Baghdad. Winona Lake: Eisenbrauns. Stol, Marten. 2012. Bitumen in Ancient Mesopotamia. The Textual Evidence, BiOr 69: 48–60. Wilcke, Claus. 2007, Der Kauf von Gütern durch den “staatlichen” Haushalt der Provinz Umma zur Zeit der III. Dynastie von Ur: Ein Beitrag zu “Markt und Arbeit im alten Orient am Ende des 3. Jahrtausends vor Christus,” JCS Sup. 1: 61–285.

Jeremiah Peterson

A Middle Babylonian Sumerian Fragment of the Adapa Myth from Nippur and an Overview of the Middle Babylonian Sumerian Literary Corpus at Nippur It is a singular honor to contribute to a volume honoring Miguel Civil, without whom it is difficult to imagine where exactly the field of Sumerology would be. The following article comes from my survey of the tablet collection in the University Museum in Philadelphia, an endeavor which he was kind enough to encourage me with, and whose unpublished catalog of Sumerian manuscripts, not to mention his many wonderful publications, were fundamental to obtaining results. The small surface fragment N 3662 is the far left edge of what was most likely a two or multi-column tablet. It is unclear if this fragment belonged to the obverse or reverse. The ductus and format all are suggestive of a Middle Babylonian date. This meager fragment preserves only Sumerian, but it could have belonged to either a unilingual Sumerian text or either a fully bilingual or partially glossed bilingual text. Examples of all of these configurations are attested among the smattering of non-extract Sumerian literary manuscripts from Middle Babylonian Nippur (see below). N 3662 preserves a small passage of the Adapa myth rendered in Sumerian. Along with the long-known Akkadian version of the Adapa myth first attested at MB Amarna, a unilingual Sumerian Old Babylonian version of the Adapa myth from Meturan, along with a fragment from Nippur, has recently been edited by Cavigneaux (2014)1 that is significantly different than yet still highly evocative of the Akkadian version. As such, in the current absence of identified extant pre-Middle Babylonian Akkadian exemplars of the text, it is tempting to understand it as an originally Sumerian text, although the inconsistent quality of the Meturan version prompts Cavigneaux to leave the question open (Cavi-

1 See also Cavigneaux and Al-Rawi 1993: 92–93, 95, Alster 1997: 424, Izreel 2001: 7 and Milstein 2015. Acknowledgements: I would like to thank John Brinkman, Jay Crisostomo, Piotr Michalowski, and Matthew Rutz for their extensive input relating to the content of this article, as well as Grant Frame for his responses to numerous queries about the collection in the University Museum, all of which were invaluable. https://doi.org/10.1515/9781501503696-016

A Middle Babylonian Sumerian Fragment of the Adapa Myth from Nippur

263

gneaux 2014: 36). Thus, we are presented with a situation that is somewhat analogous to the complex relationships between the Sumerian myth Inana’s Descent and its Akkadian counterpart Ishtar’s Descent, the former of which is also attested at MB Nippur in a glossed extract fragment published by Veldhuis,2 as well as the Sumerian Gilgamesh cycle and the Akkadian Gilgamesh epic, where the Akkadian version retains many elements of the basic story in common, but with extensive elision, recombination, and new content. The Sumerian version contained in N 3662 was roughly contemporary to the Akkadian version from MB Amarna, which is understood to be a Babylonian import.3 Thus, it verifies the occurrence of coeval independent Sumerian and Akkadian versions of the same text, which is a relatively rare but not unknown phenomenon among extant tablet finds.4 What remains of it is more reflective of the Akkadian version of the myth than the OB Sumerian version, as it preserves the passage where Adapa explains his actions against the South Wind directly to An, which is not contained in the OB Sumerian version,5 and An is named simply as king as opposed to the king of the gods, which may indicate that this version, like the Akkadian version, makes no mention of the divine assembly. Thus, this fragment further suggests that the revision of the Adapa myth may not have necessarily been a strictly cross-linguistic phenomenon. The quality of the Sumerian in N 3662 appears to be rather dubious, at least to judge from this meager sample. The possessive suffix occurs before the genitive without reflecting the vowel /a/ of the genitive morpheme, the grapheme -ke4- seems to be used indiscriminately,6 and the finite form of the compound verb gu3-de2 appears to begin line i′ 5′ in what would be a strange syntactic position. Some degree of degradation of the overall quality of Sumerian is not surprising for a Sumerian manuscript from the Middle Babylonian period, however, and it does appear that the finite verbs [mu]-un-ši-bar-[…]

2 Veldhuis 2000: 75, fig. 10. This fragment is evocative of the OB version, but it seems to abbreviate the OB version (possibly due to elipsis), as Inana’s epithet-filled address to Ninšubur is largely done away with, and possibly the list of her accoutrements as well (for lines 1–2 read […] i?-im?-gen (gloss illegible to me) […] (= OB line 26?), […-nin]dan? eš2 gana2 zagin3 […] (= OB line 25)). 3 See Veldhuis 2014: 302 n. 707. 4 For examples, see Wasserman and Gabbay 2005: 74–75. 5 For this divergence, see the discussion of Milstein 2015: 34. 6 Potential examples of the spurious use of the grapheme -ke4 from MB Nippur literary manuscripts include CBS 10475 iii 18 (= dEn-lil2-su3-ra2-še3 119) and N 3498 3′ (= Lipit-Ištar and the Plow 30). For examples of this practice from MB Ugarit and OB Meturan, see Viano 2008–2009: 79–80, 88–89, who suggests that the grapheme -ke4 may have been intended to render only the genitive.

264

Jeremiah Peterson

and [mu]-un-na-an-de2-e closely reflect their likely parallels in the OB Meturan version, lines 161 and 173, respectively. The transliteration of N 3662 advanced here is provisional. All restorations given below, which draw upon the OB Sumerian and Akkadian versions of the myth as well as some independent potential parallels, are obviously speculative.

Fig.:: N 3662

i(′?) 1′ = Adapa Fragment B 45′) [… igi mu]-˹un?˺-ši-in-bar-[… zu2 al/na?-bir9-bir9] 2′ = ?) […] ˹x˺ an-na-ke4-˹e˺-˹ne˺ […] 3′ = 46′) [A-da?]-˹ab˺-˹ba˺ igi An lugal-la-˹ke4 ˺ […] 4′ = 47′-48′?) [gu3] ˹mu˺-un-na-an-de2-e […] 5′ = 48′) [A-da?]-ab-ba tum9u18-lu [pa-bi a-na-aš? mu-e-haš] 6′ = 49′) [A-da-a]b-ba An lugal-˹la˺-˹ke4? ˺ […] 7′ = 49′-50′) [lugal?-g]u10 ˹e2?˺ lugal-gu10-˹ke4˺ ˹ku6? ˺ […] 8′ = 51′) ˹ab?˺-˹ba? ˺ mu-˹e?˺-bar […] 9′ = 52′) [tum9u18-lu ] ˹in? ˺-ri-˹a˺ […-su-su?]

A Middle Babylonian Sumerian Fragment of the Adapa Myth from Nippur

265

This passage echoes the Akkadian Adapa myth as preserved in fragment B 45′52′,7 where Adapa enters heaven and speaks with An (overlapping content with the fragmentary Sumerian version is marked on bold face): 45′) 46′) 47′) 48′) 49′) 50′) 51′) 52′)

d

Dumu-zi dGiš-zid-da a-ḫa-mi-iš ip-pa-al-su-ma iṣ-ṣe-né-eḫ-ḫu DIŠA-da-pa a-na pa-ni dA-ni šar-ri i-na qé-re-bi-šu i-mu-ur-šu-ma dA-nu il-si-ma al-ka DIŠA-da-pa am-mi-ni ša šu-ú-ti ka-ap-pa-ša te-e-eš-bi-ir DIŠA-da-pa dA-na ip-pa-al be-lí a-na bi-it be-lí-ia i-na qá-a-ab-la-at ta-am-ti nu-ni a-ba-ar ta-am-ta i-na mé-še-li in-ši-il-ma šu-ú-tu i-zi-qá-am-ma ia-a-ši uṭ-ṭe-eb-ba-an-ni “Dumuzi and (Nin)gišzida looked at each other and laughed. (When) Adapa, before An the king was in his presence, An saw him and spoke: “Come here, Adapa! Why did you break the south wind’s wing?” Adapa answered An: “My lord, For the house of my lord (Enki), in the midst of the sea I was catching fish, and when (my journey/boat?) was halfway through the sea(?) The South Wind blew and sank me”

2′) The genitival phrase … an-na-ke4-e-ne “the … of An/heaven (pl.) cannot be immediately reconciled with the Akkadian version. Perhaps this genitive phrase involves the gate of An/heaven that is mentioned earlier in the analogous context of the Akkadian version, Fragment B lines 17′, 19′, and 20′, and the OB Sumerian version 153–54. 3′) Given the lack of a ruling below it, this line appears to have been indented. The enigmatic name or epithet Adapa has been revisited in detail by Cavigneaux (2014: 36–37), who compares the spelling a-da-ba in the OB Meturan version8 to the professional, possibly musician, designation a-daba from pre-Sargonic Lagash and the song/instrument designation a-da-

7 Talon 1990: 56, Izreel 2001: 18. 8 The orthography a-da-ab/a-da-be2 is also understood by Alster to possibly reflect a Sumerian or Sumerianized orthography of Adapa in Proverb Collection 11.70 (PBS 12 29+ and duplicate 3N-T 914 ff =A 33454: see Alster 1997: 195, 424), but this is not certain from context: note the uncertainty expressed in PSD A I: 53.

266

Jeremiah Peterson

ab.9 He considers the spelling a-da-pa3, which is attested in the Old Babylonian period in the Nippur forerunner to Udughul III 59 (Ni 623+Ni 2320 ii 24; see Geller 1985: 22), to be a secondary etymology. In N 3662, Adapa seems to be rendered as [a-da]-ab-ba, to judge from the available space in the break, which gives some pause as to the correct restoration of the a-da-[…] of the OB Nippur fragment. The spelling with a reduplicated b is partially reminiscent of the late etymology u3-tu-ab-ba (= adapu) “born of the sea”(?). The role of the sea in the plot of the Adapa myth perhaps lent itself to such a folk etymology, if such were the case here, perhaps the graphemes a-da were interpreted as the lexeme a-da “fight, contest,” although its derivations otherwise involve a number, a human title, or deixis (see Civil 1987: 18; Attinger 1993: 418–19). 6′) If a parallel to both the Akkadian and OB Sumerian versions obtains here, the genitival phrase “wing of the South Wind” was rendered as an anticipatory genitive, followed by an interrogative in the OB Sumerian version line 174 (see also lines 112, 130, 176). 9′) The verbal root is clearly spelled with the bar sign, which is not easy to straightforwardly reconcile with its apparent correspondent in the Akkadian version, mašalu “to be similar, equal” as well as “to be half” and the cognate adverbial phrase i-na mi-še-li, which has elicited several minimally paralleled interpretations. However, bar does acquire a meaning of mišlu in post-OB contexts,10 perhaps as an effective allograph of sa9 “half.” The two similar signs, which differ only when the placement of the horizontal component is carefully distinguished, are interchanged with each other in some paleographic environments. The lexeme sa9 is only rarely utilized as a verbal root, with the approximate sense of “to reach the middle”,11 but one might compare Lugalbanda Hurrim 75: kaskal mu-un-sa9 kaskal muun-sa9-ba “he had (made it through) half the journey, after he had (made it through) half the journey.” It seems possible that this sentence is describing Adapa’s trip (with his boat as his subject/agent in the Akkadian version?), with the approximate meaning “I had (traversed) half of the sea.” It is not necessarily the case that the South Wind is the subject of the sentence in the Akkadian version, as has been assumed, since it is only explicitly named in the next sentence. 9 In late tradition, the instrument term /adab/ is also rendered with the spelling urudua-dapa3, perhaps as a folk etymology for the term (mur-gud to ur5-ra 11: 193 (MSL 7: 153), see Krispjin 1990: 70: 3–4 and Attinger 1993: 416 n. 1116). 10 The verb bar and the root mašālu(m) are also associated in the entry šu-bar-ra = [mi]-išla-a-nu-um, “half shares” of bilingual Nigga B 234 (MSL 13: 121). 11 For this verb, see Attinger and Krebernik 2005: 62–63.

A Middle Babylonian Sumerian Fragment of the Adapa Myth from Nippur

267

10′) The association of Sumerian ri with Akkadian zâqu(m) is well attested. For this line, compare perhaps the phrasing of Hendursaŋ Hymn 228′ (Attinger and Krebernik 2005: 48), which I tentatively base my restoration on: ambar-ra tum9ulu3 ri-a-bi gišma2 la-ba-an-da-su-su “The southern wind that blows in the marshes cannot sink (his) boat.”

Sumerian Literature and Literary Manuscripts at Nippur in the Middle Babylonian Period The Sumerian literature of the Middle Babylonian period has been treated from a variety of angles, with notable contributions including those of Cooper (1978: 32, 36–39), Veldhuis (2000, 2006), Viano (2008–2009: 134–47), Delnero (2012: 104), and Michalowski (this volume). A new window into the Sumerian literature of this period will be available with the publication of Alexa Bartelmus’ dissertation, which treats the MB school texts from Babylon, including a detailed examination of text typology. The current treatment here focuses on the corpus of non-extract texts from Kassite Nippur, to which the Adapa fragment edited above is to be considered part of. Due to the vagaries of find context that pervade the pre-WWII Nippur excavations, it is impossible to determine an exact date for the Middle Babylonian literary corpus at Nippur.12 However, the identification of similar features among a small group of manuscripts allows for the provisional establishment of a textual corpus. Unfortunately, these mostly fragmentary manuscripts preserve no scribal colophons or dates.13 The non-extract exemplars, many of which are either fully bilingual or otherwise glossed, are distinguishable by ductus and format. They are frequently bilingual, rendered more specifically in what Krecher (1976–80: 124–25) refers to as interlinear type 1a, with the Sumerian and Akkadian occurring within the same ruled line, with the Sumerian occurring at the top, in a generally larger script. The Sumerian line is often divided in two and justified to the left and right. A sub-column formed by the second halves of the lines is frequently utilized, which is sometimes established by a formal ruling. Occasionally, this column is used to mark an indent. The Akkadian is entered below the Sumerian, and the translation is typically incomplete, with only select lines rendered in Akkadian or sporadic glossing

12 See the discussion of Delnero 2012: 103. 13 A broken colophon is preserved in the MB ur5-ra 10b exemplar CBS 11889 (see Veldhuis 2014: 249 and n. 547).

268

Jeremiah Peterson

in Akkadian in smaller script within the same line underneath the Sumerian it pertains, which is analogous to the predominant method of glossing in Old Babylonian literary texts. There are also numerous examples of Krecher’s type 3 interlinear bilinguals, with the Sumerian occurring in the left column and the Akkadian in the adjacent right column. In contrast to the type 1a format, the type 3 format seems to necessitate a more complete Akkadian translation, although in the case of the Inana and An exemplar CBS 3832, several Akkadian lines are left blank. CBS 13905, a partially bilingual collective of Udughul incantations that I have previously mistaken to be Old Babylonian (Peterson 2007: 50 n. 118; 2013: 1), is a rather unique example in that it seems to be a hybridization of interlinear formats: a sporadic Akkadian translation is advanced in the same size script as the Sumerian in an adjacent column that also functions as the middle ruling for Sumerian in lines that are not translated. A potential link between the extract and non-extract manuscripts is presented by UM 29-16-35 (Veldhuis 2000: 75, fig. 10), which contains a substantial extract of Inana’s Descent (9+ lines) on one side of the tablet in type 1a format with sporadic glosses and an extract of An = Anum 5 on the reverse.14 In contrast to prevailing OB Nippur formats, formal column rulings are frequently used to establish the middle of the line and to establish a small empty margin between the end of the first column and the beginning of the second column. Rulings on MB literary texts are frequently distinguishable by their quality and their depth, which is often quite shallow and rounded in comparison to OB rulings. Some diagnostic sign forms whose presence allows for differentiation from OB manuscripts have been identified in Middle Babylonian scholarly texts, including the kur, ne, ka, ru, kug, sikil, and hur signs,15. Other signs can be posited as well, such as the ub, munus, and he2 signs. The use of these sign forms is not entirely consistent, as they can occur in this corpus along with good OB forms,16 thus representing an admixture of innovation and conservatism that is frequently encountered in Babylonian scripts. The underlying

14 Another potential example would be N 1789, a left corner fragment with an unidentified literary extract of 3(+) lines on one side (see MSL SS 1: 75 and further below) and an unidentified lexical extract consisting of entries with initial ki- on the other. This piece raises the question of whether some of the surface fragments listed below actually reflect this style of extract. Due to the rarity of proof for this configuration, however, I have elected to provisionally list them as literary texts only. 15 See Cooper 1978: 32, Geller 1985: 7, Civil 1997: 44, Veldhuis 2000: 70, Goodnick Westenholz 2005: 345, Rutz 2006: 72, Peterson 2010: 574 n. 6, Peterson 2011: 154 and n. 5, Viano 2012: 43, Delnero 2012: 103, and Veldhuis 2014: 166. 16 Note the comments of Rutz 2006: 72.

A Middle Babylonian Sumerian Fragment of the Adapa Myth from Nippur

269

character of incision, readily recognizable yet difficult to put into words, is also generally quite distinct from OB hands. MB Nippur literary texts are highly comparable to non-extract MB Nippur lexical texts17 in terms of paleography and ruling conventions, as well as some MB Nippur Akkadian scholarly texts18 and Kassite administrative texts. The following list coincides extensively but not entirely with the list compiled by Viano (2008–2009: 128–29). A few of the texts listed by Viano, however, are probably to be dated OB instead.19 N 3395, which contains unparalleled bilingual proverbs, and PBS 1/1: 11, a bilingual that mentions Šulgi,20 Lu-Nanna,21 and Ur-Gatumduga, have been identified as Middle Babylonian.22 However, none of the sign forms in these two manuscripts are incompatible with the repertoire of literary hands at OB Nippur (generally understood to date no later than the reign of Samsuiluna), and the ductus and format are not suggestive of a Middle Babylonian date. The OB identification of PBS 1/1: 11 by Goodnick Westenholz (2005: 353; see also Michalowski in this volume),23 which she makes according to its paleography24 and its Akkadian orthography and

17 For a comprehensive catalog, see Veldhuis 2014: 248–49. 18 Examples have been provisionally listed, for example, by Rutz 2006: 72 n. 49, 76 n. 92, Sassmanshausen 2008: 270–271, and Rutz 2013: 222–23 and n. 118, 231). Especially notable is the dated extispicy report CBS 13517 (Brinkman 1976: 116, Kraus 1985: 145–50), dated to the twenty-first year of Burna-buriaš II. 19 These include CBS 8547 (now joined to N 3465 and N 6898), and CBS 10295. CBS 13990 is likely of Kassite date, but it appears to be a fragment of a cultic itinerary. The unprovenienced manuscript MS 2291 is more likely to be the doctored top half of a Old Babylonian im-gid2-da extract text. I was unable to locate the exemplar CBS 9899, a manuscript of Two Women B (Dialogue 5) reported by Viano, in the Babylonian Section for confirmation. 20 Another unusual OB text involving Šulgi and a musical performance is UET 6/3: 522 (contra Shaffer and Ludwig 2006: 15 and Sassmanshausen 2008: 272, the ductus is not Kassite), which is only partially intelligible to me. The incipit of this composition is listed in UET 6/2: 196 rev. 5′ (Shaffer 2000). 21 The Lu2-dNanna who is prominent in later tradition as a famous apkallu and the alleged author of the Etana myth (for discussion and further bibliography, see Goodnick Westenholz 2005: 346 n. 14, Schwemer 2015: 213) occurs in the mostly broken obverse, line 15 (abgal AŠLu2d Nanna …). 22 Van Dijk 1998: 12, Veldhuis 2000: 73, Veldhuis 2008: 31 and n. 11, Veldhuis 2014: 263–66. The orthographic features and content pointed out by Veldhuis do not strike me as decisive evidence that this must be post-OB. Note that the rare word abbunnuna (ud.mud.nun.(na.ki)) is also attested in an OB exemplar of the unedited curricular personal name list Ur-me (Wilson 2008: no. 57 rev. iii 18). The word spelled murub2 is reminiscent of Dumuzi-Inana J 33 and 36 (Alster 1985: 223), where murub4 seems to be a technical musical term (see Kilmer 1992: 106). 23 See also Krecher 1976–80: 127. 24 The interchange between the simple and complex forms of the na sign is commonplace in literary manuscripts from OB Nippur.

270

Jeremiah Peterson

the tentative attribution of N 3395 to OB by Lambert (1960: 272) are more likely. These two texts, along with the Lamašaga hymn CBS 10986 (Sjöberg 1974: 161– 63; Krecher 1976–80: 127), the Emesal lament UM 29-13-596+N 4188,25 and the OB versions of ED Proverbs BT 926 and UM 29-15-174 (Civil and Biggs 1966: 5– 6; Alster 1991–92: 2), constitute rare examples of non-extract OB Nippur literary bilinguals, all of which are Krecher type 3. Thus, it may be the case that occurrences of this format in Middle Babylonian contexts are more conservative than those advanced in Krecher type 1a format, although it could also be argued that type 1a was the logical outgrowth of the predominant glossing convention at OB Nippur that was still practiced at MB Nippur, which was entering the Akkadian gloss below the Sumerian word in a smaller script. A direct connection with the Old Babylonian period is suggested by the nature of the Akkadian translations, as in some instances, as noted by George in conjunction with the Lipit-Ištar A bilingual fragment CBS 3558+ (George 2012: 370) both distinctly OB and MB dialectical features coexist, suggesting a multi-phase translation that may have utilized previous OB glosses. It should be noted as well that interlinear bilingual texts from OB Sippar (for which see, for example, Michalowski 2011: 45), MB Babylon, as well as Middle Assyrian Assur have notable similarities in their format to interlinear bilingual texts from Nippur. The following is a list of non-extract Middle Babylonian Sumerian literary manuscripts from Nippur that I am currently aware of, many of which were first identified and/or published by Professor Civil. Due to the importance of ductus and format in making this identification, I have primarily restricted my identifications to manuscripts I have seen. More exemplars can undoubtedly be identified, particularly in the National Museum in Istanbul. Digital images of all of these manuscripts except from those in Istanbul are available at CDLI.

25 Unpublished: quoted by Civil 2008: 100. 26 Klein 2004. This manuscript also includes a syllabic rendering of the Sumerian text.

Šulgi B Poem of Early Rulers, duplicate to RS 25.130 19′–31′, and nig2-nam nu-kal-la (Alster version B), at least

Counsels of Wisdom

Šulgi O? En-lil2-su3-ra2-še3 An-gim

d

Alster 2005: 266, 282, 323–326, Finkel 1986: 252, Civil 1989: 7, Viano 2008–2009: 218, Rutz 2013: 268–269, n. 233

Cooper 1978: 32, 41–43, 46–52, 54, Rutz 2006: 72 n. 45, Viano 2008–2009: 129, Viano 2012a: 42–43, Delnero 2012: 104 and n. 44, Wagensonner 2014: 1 Civil 1997: 43, Alster 2005: 226

Klein 1976: 272 and n. 9, 274, 284

Delnero 2006: 891–892, 896–897, 1914, Delnero 2012: 103–104, Viano 2008–2009: 129, Peterson 2011: no. 170, George 2012: 369–370 van Dijk 1998: 12 f., Viano 2008–2009: 129 Sjöberg 1975: 166, 169 Unpublished Unpublished Delnero 2006: 1221–1224, 2109, Delnero 2012: 103–104

Editions and Additional Literature

27 Thus line 13′: … lugal u2-a-bi he2-til3 … “May …, its(?) king and provider, live.” 28 Rev. col. ii 14′ ([…]-la-tum).

CBS 10900 CBS 10903 (UF 42: 2) CBS 11153 (Viano 2012a: 43)(+?) N 6286 (Cooper 1978: pl. 14) CBS 11945 (Cavigneaux 1996: 18–21) CBS 13509 (BPOA 9: 132) CBS 13777 (Alster 2005: pl. 32–33)

Lipit-Ištar A

CBS 3558 (JCS 31, p. 226)+?Ni 9696 (ISET I, p. 109)(+?)Ni 4557 (ISET I, p. 109) CBS 3832 (JCS 31, p. 225) CBS 7946 (ZA 65, p. 173) CBS 10070 CBS 10433 CBS 10475

Inana and An in-nin šag4 gur-ra Unidentified, minimal Unidentified royal hymn?27 d En-lil2-su3-ra2-še3

Text

Museum Number and Image Publication

1a

3

3 Unclear Unclear 1a Unilingual (one gloss preserved)28 1a 1a 3

1a

Format

Uncertain

Uncertain

Uncertain 2? 2

Uncertain Uncertain Uncertain Uncertain 2

Uncertain

Number of Columns per Side

A Middle Babylonian Sumerian Fragment of the Adapa Myth from Nippur

271

Udughul incantation, reverse column i parallels Udughul 3 Forerunner 39– 45 and Udughul 3: 67 f. (Geller 1985: 22, Geller 2007: 103) Unidentified 29 en2-e2-nu-ru incantation with a provision for purifying the body after an attack by a lion(?) Unidentified, mentions Nippur and the “herald(ship)” of the city(nimgir iri/nagirutu IRI) in-nin-šag4-gur-ra

two other unidentified compositions in-nin šag4 gur-ra

Text

Michalowski 1998: 70 n. 17, Veldhuis 2000: 75 n. 23, Viano 2008–2009: 128

Unpublished

Unpublished Unpublished

Sjöberg 1975: 165, 168, Krecher 1976–1980: 128, Cooper 1978: 164, Viano 2008–2009: 128 Peterson 2007: 50 n. 118, Peterson 2013: 1

Editions and Additional Literature

29 Among the meager contents is a repeated refrain ending in … e3-ma-ra = … ṣi-i “come out to me.”

CBS 15203

CBS 15143?

CBS 14075? CBS 15080

CBS 13905

CBS 13860 (ZA 65, p. 168)

Museum Number and Image Publication

(continued)

1a

1a

1a 1a

Hybridized (see above)

3

Format

1?

Unclear

Unclear 1

2

Uncertain

Number of Columns per Side

272 Jeremiah Peterson

30 Listed incorrectly there as UM 29-13-523.

N 3662 (see above) N 6126 (BPOA 9: 13) Ni 2676+Ni 2997+Ni 4017+Ni 4018 (Geller 1985: pl. 5–6) Ni 13227 UM 29-13-419A (AuOr 15, p. 53) +Ni 9734 (ISET 2: 26) UM 29-13-513 (JCS 28: 93–94)

Šulgi O? Instructions of Ur-Ninurta, Counsels of Wisdom(?) nin-me-šar2-ra

Enlil and Ninlil Unidentified Unidentified Enlil and Ninlil Ninurta I (duplicate to KAR 119 (Lambert 1960: 118– 120)) Lipit-Ištar and the Plow Lipit-Ištar and the Plow Passage with unparalleled antediluvian kinglist Adapa Enlil and Ninlil Udughul incantations

N 2243 (Behrens 1978: pl. V) N 2306 N 2875+N 4113 N 3455 (BPOA 9: 12) N 3462

N 3495 N 3498 (BPOA 9: 181) N 3514? (AuOr 26, p. 258)

ki dUtu prayer Unidentified Enlil and Ninlil Enlil and Sud

HS 1512 (ZA 91, p. 244) N 1634 N 1747 (Behrens 1978: pl. XVI) N 2203 (JAOS 103, p. 47)

Klein 1976: 272, 285 Civil 1997: 43, Alster 2005: 225, Viano 2008–2009: 129 Delnero 2006: 1071–1072, 2021,30 Delnero 2012: 103–104

Geller 1985: 4, 7, Viano 2008–2009: 135–36

Viano 2008–209: 128

Viano 2008–2009: 118, 120, Civil 2013: 38 n. 67

Behrens 1978: 8, Green 1982: 341 Unpublished Unpublished

Krebernik 2001, Bonechi and Alaura 2012: 17 n. 69 Unpublished Behrens 1978: 11, Green 1982: 341 Civil 1983: 47, Viano 2008–2009: 129

Unclear 3? 2

Unilingual

Unclear Unclear 4

Unclear Unclear Unclear

Unclear Unclear Unclear Unclear 1

1 Unclear Unclear Unclear

1a Unilingual

Unclear Unclear 1a (partial)

1a 1a Unclear

1a Unclear 1a Unclear 1a

1a Unclear Unclear Unclear

A Middle Babylonian Sumerian Fragment of the Adapa Myth from Nippur

273

Song of the Plowing Oxen

UM 29-13-560(+)N 3196(+)N 3529 UM 29-15-399+Ni 9734 (ISET 2: 26)

Lipit-Ištar and the Plow

Text

Museum Number and Image Publication

(continued) Format

Civil 1976: 86, Cooper 1978: 164, Viano 2008–2009: 3 128 Civil 1976: 84 n. 3, Cooper 1978: 164, Viano 2008– 3 2009: 128

Editions and Additional Literature

2?

Unclear

Number of Columns per Side

274 Jeremiah Peterson

A Middle Babylonian Sumerian Fragment of the Adapa Myth from Nippur

275

When combined with the contents of the school exercises, these texts reflect a considerable cross section of the OB Sumerian literary corpus in what is a miniscule amount of manuscripts in comparison to the OB finds at Nippur. Ultimately, one can suspect that Sumerian texts attested in the Old Babylonian period that survived into peripheral MB,31 MA, NA and NB contexts were in circulation in southern Mesopotamia in the Middle Babylonian period as well, although for peripheral MB Sumerian, such an assumption is complicated by the fact that it has been observed that Northern Mesopotamia may have been prominently involved in the dissemination of Sumerian literature to the periphery in the Middle Babylonian period, perhaps as a result of greater geographic proximity and political stability at the beginning of the period.32 Later tradition mentions Kassite kings and scholars, and the enigmatic series Si-du3, which is known until the very end of the cuneiform era, seems to have had a specific link to Nippur.33 This could be understood to indicate, albeit partially as an argument from silence, that a substantial amount of the literary corpus was still known to the inhabitants of Middle Babylonian Nippur,34 and that the intervening period of upheaval at the site of Nippur in the late Old Babylonian and early Middle Babylonian periods did not exert a halting effect on the ability to perpetuate the Sumerian literary tradition, with its preservation perhaps sponsored it part by the cities of Northern Mesopotamia35 and the First Sealand Dynasty.36 How-

31 See Viano 2008–2009, and for a recent overview, see Fincke 2012. 32 See in detail Viano 2015, who focuses on similarities between syllabic orthographies, and the discussion of Michalowski 1995: 50–51 specifically involving Nergal D, which is attested in OB North Mesopotamia (perhaps Sippar), MB Hattusa, and NA Nineveh, but is not currently attested at OB Southern Mesopotamia. For the likelihood that Sumerian compositions attested in post-MB contexts, including Enlil and Sud, Enki and Ninmah, Lugale, Angim, Lugalbanda, Farmer’s Instructions (known from its citation in Si-du3), Lugal-ibila to Lugal-nesag, Sin-iddinam to Utu, a tigi song for Nintur/Aruru, (see, for example, the lists compiled by Michalowski 1987: 38–39 and Viano 2008–2009: 143), and various proverbs, were also present in Kassite Nippur, see, for example, the remarks of Veldhuis 2000: 81. The assignment of Kuyunjik accession numbers to completely unrelated Old Babylonian manuscripts in the British Museum (see, for example, Hallo 1976: 192, n. 9, Civil 1994: 10), has understandably led to some confusion. 33 For the series Si-du3, see Lambert 1962: 72, Finkel 1986, Frahm 2010: 168–76, Rutz 2013: 269. It is possible that the Si-du3 dub-sar “Sidu, the scribe” of SEpM 14 line 8 (Kleinerman 2011: 154) is reflective of this individual, see Hallo 1977: 67 and the less enthusiastic opinion of Kleinerman 2011: 46. Given that the scholar Zu-zu in the previous line is celebrated elsewhere, it is certainly possible that this is the famous Si-du3. 34 See Veldhuis 2014: 261–62. 35 See, for example, Michalowski 1995: 2287–88. 36 Scholarly texts attributable to this dynasty have been recently published (George 2007, George 2013, and Gabbay 2014).

276

Jeremiah Peterson

ever, it is overly simplistic to assume there was an insular local tradition in an era where there is contemporary evidence for the importation of a manuscript of solar omens to Nippur all the way from Susa (Rutz 2006). The evidence of the identified extant manuscripts, as misrepresentative of the whole as they may well be, could be taken to suggest that a particular cultural interest was maintained in myths and extra-cultic hymns of the deities Enlil, Ninurta,37 the two most important deities at Nippur, and Inana, for whom the underlying goals of Inana’s Descent and the highly fragmentary Inana and An, namely, Inana’s acquisition of the netherworld and heaven and etiologies for the seasons and the shortening of the length of the day, respectively, are natural complements to each other. There seems to have been an interest in the historical figures of Šulgi,38 Lipit-Ištar, and Ur-Ninurta,39 as well as texts belonging to the category of “wisdom literature,” the latter of which invites comparison to the apparent thrust of Si-du3 as suggested by its identified components. To judge from its post-OB distribution, some of the E2-dub-ba-a literature, such as E2-dub-ba-a E, which incorporates content from E2-dub-ba-a A, E2-dub-ba-a C, and Dialogue 3 (Enki-ma-an-šum2 and Gir3-ni-i3-sag9) (see Civil 1987, 2000: 113), as well as the literary letter Lugal-ibila to Lugal-nesag (Civil 2000: 109–18), was likely to have been present at MB Nippur. The extract fragments N 1789 obv. and N 6064 preserve contexts that may have involved the scribal school,40 although I cannot place them further. The perpetuation of Lipit-Ištar and the Plow41 and the Song of the Plowing Oxen (and perhaps the

37 See the observations of Cooper 1978: 11, Veldhuis 2000: 82 and Viano 2008–2009: 146. 38 The manuscript of Šulgi A from Babylon VS 24: 43, which has not been connected to controlled excavation (see Delnero 2006: 57 n. 156, to his variant analysis see on pg. 728 one can add the syllabic/plene spelling an-nu-ub-da for an-ub-da for line 4), also seems to have an MB ductus. Note as well the post-OB (NB?) landscape extract N 1751+N 4523, which appears to be a divergent version of Šulgi N that makes extensive use of the ditto marker ki.min in the text. The tigi incipit en-ni ni2 gu-ru an-ki-a me x […] listed by the MA song catalog KAR 158 iii 5′ may reflect a tigi for Nergal (see Wilcke 1976: 290) that may have originally composed for Šulgi as an adab composition (who may have been written out of the latter incarnation of the text, see Peterson 2015). 39 Note as well the occurrence of the incipit of the adab of An for this ruler (Ur-Ninurta E) in the MA catalog KAR 158 iii 36′ (Hallo 1976: 192, Wilcke 1976: 266). 40 See MSL SS 1: 75, where Civil describes N 6064 as “a small fragment of an unidentified MB or NB Edubba text. These fragments mention a series of obscure grammatical(?) designations (see MSL SS 1: 75, Civil 2000: 108). N 1789 may involve translation (thus obv. 1: eme-gir15 -uriki galam […] “Sumerian, Akkadian, the … “language”). The closest extant literary reflexes of these fragmentary lines come from the post-OB “Examenstexte” A and B (A line 18 (from CT 58: 64 obv. 5″, not used by Sjöberg 1975), B iv 1). 41 The incipit of this composition is currently unknown.

A Middle Babylonian Sumerian Fragment of the Adapa Myth from Nippur

277

Farmer’s Instructions, given its occurrence in an MB manuscript from Babylon as well as the series of Si-du3) could be understood to reflect particular interest in texts involving agriculture and perhaps more specifically the royal festival in the month of itiGu4-si-su/Ayyaru. Three Decad compositions, dEn-lil2-su3ra2-še3, Nin-me-šar2-ra, and Lipit-Ištar A, are attested in this group, although this may be more of a coincidence conditioned by other factors than the perpetuation of this curriculum-based textual group into the Middle Babylonian period. The Adapa myth has many plot similarities to the fragmentary myth Inana and An, including a conflict between the protagonist and An and the role of a fisherman in a boat that is assailed by the southern wind. There are numerous indications that the content of these texts changed in both minor and major ways, ranging from the particularly conservative reproduction of d En-lil2-su3-ra2-še3 to the Sumerian Adapa fragment edited above, which may have been substantially different from the Old Babylonian Sumerian version. A fascinating Sumerian text centered around Ninurta, possibly a šir3-gid2-da composition, found at MB Nippur (N 3462) and MA Assur (KAR 119) is treated in detail by Michalowski in this volume. It is not yet attested in the Old Babylonian period and utilizes some post-OB neologisms: thus, it is possible that it was original to the Middle Babylonian period. The incantation manuscripts, although very sparsely evidenced, demonstrates a link between OB traditions with later ritual series such as Utukkū lemnūtu42 and Bīt Rimki.43 The perpetuation of certain textual traditions, such as the cultic songs bearing indigenous labels would be particularly doubtful were it not for the extensive listing of Sumerian cultic songs, including tigi and adab compositions, in the Middle Assyrian catalog KAR 158,44 the attestation of a šir3-nam-šub of Ninisina known from OB Nippur at MA Assur,45 and the attestation of a tigi song of Nintur/Aruru (BL 95+) in Ashurbanipal’s library, tellingly described as originating from Nippur in the later copy.46 Metcalf notes

42 See the discussion of Geller 1985: 7–8. 43 See the discussion of Krebernik 2001: 241–42. 44 See, for example, Hallo 1963: 169, Wilcke 1976: 255, 263–64, Hallo 1976: 169, Shehata 2009: 254, 293 and note 1687, Michalowski 2010: 213–14, 216, and Metcalf 2015: 79–80, with further citations. 45 Edited by Cohen 1975: 609–11. See also Hallo 1989: 239–40, Viano 2008–2009: 118–19. TMH 3: 53+4: 53, a catalog of various cultic songs, was identified as MB by Kramer and Bernhardt 1956–1957: 391 but has since been identified as OB by Tinney (apud Vanstiphout 2003: 10). 46 See Hallo 1989: 240–45, Black 2005: 53. The late version was re-labeled as a tigi of Ninurta, a seemingly incompatible label that is reminiscent of other Sumerian textual labels in Assurbanipal’s library (see, for example, Black 1985: 11–12). Metcalf (2015: 79 n. 2) suggests that it reflects a simple error for the original dNin-tur5.

278

Jeremiah Peterson

that the bulk of the tigi and adab compositions in KAR 158 have not been identified and takes this as an indication that a substantial portion of these corpora is not recovered. As discussed recently by Gabbay, the more fluid liturgical corpus of the gala priest is only very rarely attested in Mesopotamia during the Middle Babylonian period,47 yet it nevertheless must have been steadfastly maintained, given its ample attestation in the Old Babylonian period and re-appearance in force in later periods. A significant number of unidentified literary passages remain in both the list of non-extract exemplars given here and in Veldhuis’ catalog of extract exemplars, whose identification would obviously further clarify the picture. It is difficult to pinpoint shifts in the cultural role that Sumerian literature experienced in this period, although a few limited observations are possible. As Veldhuis has demonstrated, excerpts of Sumerian literary texts were utilized in what must have been a very elementary curricular context, to judge from the inexactness of the hands and the textual paucity of the format, analogous to the use of Sumerian literary passages in elementary tablet formats at Old Babylonian Nippur. The cultural role of the large literary manuscripts, which certainly could reflect library texts, is less clear. A certain degree of antiquarianism, a relative constant throughout the cuneiform tradition, undoubtedly prompted the authorship of these sources. A particularly illustrative example of a text from MB Nippur that demonstrates such an intellectual role is not a literary text, but rather a manuscript of the personal name list Inana-teš2 PBS 11/1: 7 (see Peterson 2011b: 249 and n. 14). The presence of breakage notation and the rather fanciful selective Akkadian translations demonstrates that this text was no longer in routine circulation in the same city that had utilized it intensively in the OB scribal curriculum only a few centuries earlier, and had assumed a completely different intellectual role. Given that the use of this personal name list may have been confined to the OB Nippur curriculum and the typical find contexts of school texts indicating haphazard disposal,48 one wonders if a Kassite scholar simply stumbled upon a broken OB curricular text in the course of his/her daily activities. This situation can be contrasted with the role of Proverb Collection 2, also extensively utilized in the OB scribal curriculum, which was also at least partially perpetuated in the Middle Babylo-

47 Gabbay 2014: 146–47. Veldhuis (2000: 81, 2014: 262) observes that some Kassite extracts may in fact quote from the corpus of the gala priest. Given the perpetuation of the major proverb collections in this period, it is notable as well that there is a definite reflex of this corpus in some Sumerian proverbs, for which see Gabbay 2011. 48 See the overview of Taylor and Cartwright 2011: 316–18.

A Middle Babylonian Sumerian Fragment of the Adapa Myth from Nippur

279

nian period, finding its way into the MB scribal curriculum and surviving all the way into Si-du3, which lasted until the very end of the cuneiform era.49 Could manuscripts like these be taken as evidence that a few particularly antiquarian libraries were in circulation during this period that may have preserved an even wider sample of the earlier Sumerian literary corpus?50

Bibliography Alster, Bendt. 1985. Geštinanna as Singer and the Chorus of Uruk and Zabalam. JCS 37: 219–28. Alster, Bendt. 1991–92. Early Dynastic Proverbs and other Contributions to the Study of Literary Texts from Abū Ṣalābīkh. AfO 38–39: 1–51. Alster, Bendt. 1997. Proverbs of Ancient Sumer. Bethesda: CDL Press. Alster, Bendt. 2005. Wisdom of Ancient Sumer. Bethesda, MD: CDL Press. Attinger, Pascal. 1993. Eléments de linguistique sumérienne: La construction de du11/e/di “dire.” OBO Sonderband. Göttingen: Vandenhoeck und Ruprecht. Attinger, Pascal and Manfred Krebernik. 2005. L’Hymne à Hendursaga (Hendursaga A). Pp. 21–104 in Von Sumer bis Homer: Festschrift für Manfred Schretter, ed. Robert Rollinger. AOAT 325. Münster: Ugarit Verlag. Behrens, Hermann. 1978. Enlil und Ninlil. Ein sumerischer Mythos aus Nippur. Studia Pohl, series maior, 8. Rome: Pontifical Institute. Black, Jeremy. 1985. A-še-er Gi6-ta, a Balag of Inana. ASJ 7: 11–87. Black, Jeremy. 2005. Songs of the Goddess Aruru. Pp. 39–62 in An Experienced Scribe who Neglects Nothing: Ancient Near Eastern Studies in Honor of Jacob Klein, ed. Yitschak Sefati Bethesda, MD: CDL Press. Brinkman, John A. 1976. Materials and Studies for Kassite History vol. I: A Catalogue of Cuneiform Sources Pertaining to Specific Monarchs of the Kassite Dynasty. Chicago: Oriental Institute. Bonechi, Marco and Alaura, Silvia. 2012. Il carro del dio del sole nei testi cuneiformi dell’Eta del Bronzo. SMEA 54: 5–115. Cavigneaux, Antoine. 1996. Miettes de l’Edubbâ. Pp. 11–26 in Tablettes et Images aux Pays de Sumer et d’Akkad: Mélanges Offerts a Monsieur H. Limet. Association pour la Promotion de l’Histoire et de l’Archéologie Orientales, ed. Önhan Tunca and Danielle Deheslle. Liège: Université de Liège. Cavigneaux, Antoine. 2014. Une version sumérienne de la légende d’Adapa (Textes de Tell Haddad X). ZA 104: 1–41. Cavigneaux, Antoine and Farouk Al-Rawi. 1993. New Sumerian Literary Texts from Tell Haddad: A First Survey. Iraq 55: 91–106.

49 See Frahm 2010: 169. 50 For a recent discussion of the meager but intriguing evidence of private libraries in the Old Babylonian period, see Wilcke 2012: 3–4 and Michalowski 2014. See Wagensonner 2011 for the probable existence of private libraries at MA Assur.

280

Jeremiah Peterson

Civil, Miguel. 1976. The Song of the Plowing Oxen. Pp. 83–96 in Cuneiform Studies in Honor of Samuel Noah Kramer, ed. Barry Eichler AOAT 25. Neukirchen-Vluyn: Neukirchener Verlag. Civil, Miguel. 1983. The Marriage of Enlil and Sud. JAOS 103: 43–66. Civil, Miguel. 1987. Sumerian Riddles: a Corpus. AuOr 5: 17–37. Civil, Miguel. 1987. An Edubba Text from Boghazköy. NABU 47: 25. Civil, Miguel. 1994. The Farmer’s Instructions: A Sumerian Agricultural Manual. AuOr Supplementa 5. Barcelona: Editorial AUSA. Civil, Miguel. 1995. Ancient Mesopotamian Lexicography. Pp. 2305–14 in Civilizations of the Ancient Near East 4, ed. Jack M. Sasson. New York: Scribner. Civil, Miguel. 1997. The Instructions of Ur-Ninurta: A New Fragment. AuOr 15: 43–53. Civil, Miguel. 2000. From the Epistolary of the Edubba. Pp. 105–18 in Wisdom, Gods, and Literature, Studies in Assyriology in Honor of W. G. Lambert, ed. Andrew R. George Winona Lake: Eisenbrauns. Civil, Miguel. 2002. The Forerunners of Marû and Ḫamṭu in Old Babylonian. Pp. 63–71 in Riches Hidden in Secret Places: Ancient Near Eastern Studies in Memory of Thorkild Jacobsen, ed. Tsvi Abusch. Winona Lake: Eisenbrauns. Civil, Miguel. 2008. The Early Dynastic Practical Vocabulary A (Archaic har-ra A). ARES 4. Rome: Missione Archeologica Italiana in Siria. Civil, Miguel. 2013. Remarks On ad-gi4 (A.K.A. “Archaic Word List C” or “Tribute”). JCS 65: 13–67. Civil, Miguel and Robert D. Biggs. 1966. Notes sur des textes sumériens archaïques. RA 60: 1–16. Cohen, Mark. 1975. The Incantation-Hymn: Incantation or Hymn? JAOS 95: 592–611. Cohen, Mark. 1981. Sumerian Hymnology: The Eršemma. HUCA Annual Supplement 2. Cincinnati, CT: Hebrew Union College. Cooper, Jerrold. 1978 The Return of Ninurta to Nippur. AnOr 52. Rome: Pontifical Institute. Cooper, Jerrold. 1983. The Curse of Agade. Baltimore: The Johns Hopkins University Press. Delnero, Paul. 2006. Variation in Sumerian Literary Compositions: A Case Study Based on the Decad. Ph.D. diss., University of Pennsylvania. Delnero, Paul. 2012. The Textual Criticism of Sumerian Literature. Journal of Cuneiform Studies Supplemental Series 3. Boston: American Schools of Oriental Research. van Dijk, Johannes J. A. 1987. Literarische Texte aus Babylon. VS 24. Berlin: Akademie Verlag. van Dijk, Johannes J. A. 1998. Inanna raubt den “grossen Himmel”: Ein Mythos. Pp. 9–38 in Festschrift für Rykle Borger zu seinem 65. Geburtstag am 24. Mai 1994: tikip santakki mala bašmu …, ed. Stefan M. Maul. CM 10. Groningen: Styx Publications. Ellis, Maria deJ. 1979. Akkadian Literary Texts and Fragments in the University Museum. JCS 31: 216–231. Ferrara, A. J. 1976. Texts and Fragments 91–96. JCS 28: 93–97. Fincke, Jeanette C. 2012. The School Curricula from Ḫattuša, Emar and Ugarit: a Comparison. Pp. 85–101 in Theory and Practice of Knowledge Transfer: Studies in School Education in the Ancient Near East and Beyond: Papers Read at a Symposium in Leiden, 17–19 December 2008, ed. Wolfert S. van Egmond and Wilfred H. van Soldt. PIHANS CXXI. Leiden: Nederlands Instituut voor het Nabije Oosten. Finkel, Irving K. 1986. On the Series of Sidu. ZA 76: 250–53. Frahm, Eckart. 2010. The Latest Sumerian Proverbs. Pp. 155–84 in Opening the Tablet Box: Near Eastern Studies in Honor of Benjamin R. Foster, ed. Sarah C. Melville and Alice L. Slotsky. Leiden: Brill.

A Middle Babylonian Sumerian Fragment of the Adapa Myth from Nippur

281

Gabbay, Uri. 2011. Lamentful Proverbs or Proverbial Laments? Intertextual Connections between Sumerian Proverbs and Emesal Laments. JCS 63: 51–52. Gabbay, Uri. 2014. A balaŋ to Enlil from the First Sealand Dynasty. ZA 104: 146–70. Geller, Markham J. 1985. Forerunners to Udughul: Sumerian Exorcistic Incantations. FAOS 12. Stuttgart: Franz Steiner Verlag. Geller, Markham J. 2007. Evil Demons: Canonical Utukkū Lemnūtu Incantations. SAACT 5. Helsinki: The Neo-Assyrian Text Corpus Project. George, Andrew R. 2007. The Civilizing of Ea-Enkidu: An Unusual Tablet of the Babylonian Gilgamesh Epic. RA 101: 59–77. George, Andrew R. 2012. Review of Peterson, J., Sumerian Literary Fragments in the University Museum, Philadelphia (BPOA 9). BSOAS 75: 368–70. George, Andrew R. 2013 Babylonian Divinatory Texts Chiefly in the Schøyen Collection. CUSAS 18. Bethesda, MD: CDL Press. Goodnick Westenholz, Joan. 2005. Sing a Song for Šulgi. Pp. 343–73 in An Experienced Scribe who Neglects Nothing: Ancient Near Eastern Studies in Honor of Jacob Klein, ed. Yitschak Sefati Bethesda, MD: CDL Press. Green, Margaret. 1982. Review of Enlil und Ninlil: Ein sumerischer Mythos aus Nippur by H. Behrens. BiOr 39: 339–44. Hallo, William W. 1963. On the Antiquity of Sumerian Literature. JAOS 83: 167–76. Hallo, William W. 1976. Toward a History of Sumerian Literature. Pp. 181–203 in Sumerological Studies in Honor of Thorkild Jacobsen on his Seventieth Birthday June 7, 1974, ed. Stephen J. Lieberman. AS 20. Chicago: Oriental Institute. Hallo, William W. 1977. Seals Lost and Found. Pp. 55–60 in Seals and Sealings in the Ancient Near East, ed. MacGuire Gibson and Robert D. Biggs. BM 6. Malibu: Undena Publications. Hallo, William W. 1989. Nippur Originals. Pp. 237–47 in dumu-e2-dub-ba-a: Studies in Honor of Åke W. Sjöberg, ed. Hermann Behrens, Darlene Loding, and Marths T. Roth. Occasional Publications of the Samuel Noah Kramer Fund 11. Philadelphia: University Museum. Izreel, Shlomo. 2001. Adapa and the South Wind: Language Has the Power over Life and Death. MC 10. Winona Lake: Eisenbrauns. Kilmer, Anne Draffkorn. 1992. Musical Practice in Nippur. Pp. 101–112 in Nippur at the Centennial: Papers Read at the 35e Rencontre Assyriologique Internationale, Philadelphia, 1988, ed. Maria deJ. Ellis. Occasional Publications of the Samuel Noah Kramer Fund 14. Philadelphia: University Museum. Klein, Jacob. 1976. Šulgi and Gilgamesh: Two Brother-Peers (Shulgi O). Pp. 271–92 Cuneiform Studies in Honor of Samuel Noah Kramer, ed. Barry Eichler AOAT 25. NeukirchenVluyn: Neukirchener Verlag. Klein, Jacob. 1981. The Royal Hymns of Shulgi King of Ur: Man’s Quest for Immortal Fame. Transactions of the American Philosophical Society vol. 71/7. Philadelphia: American Philosophical Society. Klein, Jacob. 2004. An Old Babylonian Edition of an Early Dynastic Collection of Insults (BT 9). Pp. 135–49 in Literatur, Politik und Recht in Mesopotamien: Festschrift für Claus Wilcke, ed. Walther Sallaberger, Konrad Volk, and Annette Zgoll. OBC 14. Wiesbaden: Harrassowitz Verlag. Kleinerman, Alexandra. 2011. Education in Early 2nd Millennium BC Babylonia: The Sumerian Epistolary Miscellany. CM 42. Leiden: Brill. Kraus, Fritz. 1985. Mittelbabylonische Opferschauprotokolle. JCS 37: 127–218.

282

Jeremiah Peterson

Krebernik, Manfred. 2001. Ein ki-dutu-Gebet aus der Hilprecht-Sammlung. ZA 91: 238–52. Krecher, Joachim. 1976–80. Interlinearbilinguen und sonstige Bilinguentypen. RlA 5: 124– 28. Krispjin, Theo J. H. 1990. Beiträge zur altorientalischen Musikforschung: 1. Šulgi und der Musik. Akkadica 70: 1–27. Lambert, Wilfred G. 1960. Babylonian Wisdom Literature. Oxford, Clarendon Press. Lambert, Wilfred G. 1962 A Catalogue of Texts and Authors. JCS 16: 59–77. Meltcalf, Christopher. 2015. The Gods Rich in Praise: Early Greek and Mesopotamian Religious Poetry. Oxford: Oxford University Press. Michalowski, Piotr. 1987. On the Early History of the Eršahunga Prayer. JCS 39: 37–48. Michalowski, Piotr. 1995. Review of Sumerian Literary Texts (Cuneiform Texts from Babylonian Tablets in the British Museum 58) by B. Alster and M. J. Geller. JNES 54: 49–51. Michalowski, Piotr. 1995 Sumerian Literature: An Overview. Pp. 2279–91 in Civilizations of the Ancient Near East, vol. 4, ed. Jack M. Sasson. New York: Scribner. Michalowski, Piotr. 1998. Literature as a Source of Lexical Inspiration: Some Notes on a Hymn to the Goddess Inana. Pp. 65–74 in Written on Clay and Stone: Ancient Near Eastern Studies Presented to Krystyna Szarzyń ska, ed. Jan Braun Warsaw: Agade. Michalowski, Piotr. 2011. The Royal Correspondence of the Kings of Ur: an Epistolary History of an Ancient Mesopotamian Kingdom. MC 15. Winona Lake: Eisenbrauns. Michalowski, Piotr. 2010. Learning Music: Schooling, Apprenticeship, and Gender in Early Mesopotamia. Pp. 199–239 in Musuker und Tradierung: Studien zur Rolle von Musikern bei der Verschriftlichung un Tradierung von literarischen Werken, ed. Regine Pruzinsky WOO 8. Vienna: LIT Verlag. Michalowski, Piotr. 2014. From the Collections of An Old Babylonian Literary Connoisseur. RA 107: 15–22. Michalowski, Piotr. 2016. Literary Journeys from Babylonia to Assyria: Second Millennium Copies of a Bilingual Poem Concerning Ninurta (pp. 205–30 in this volume). Milstein, Sara J. 2015. The Origins of Adapa. ZA 105: 30–41. Molina, Manuel and Marcos Such-Gutiérrez. 2004. On Cutting Plants and Noses in Ancient Sumer. JNES 63: 1–16. Peterson, Jeremiah. 2007a. A Study of Sumerian Faunal Conception with a Focus on the Terms Pertaining to the Order Testudines. Ph.D. diss., University of Pennsylvania. Peterson, Jeremiah. 2007b. The MB Nippur Combination Extract N 3783+N 5031 Obverse. NABU 34: 34–35. Peterson, Jeremiah. 2010 Sumerian Literary Fragments in the University Museum, Philadelphia III: Hymns to Deities. UF 42: 573–612. Peterson, Jeremiah. 2011a. Sumerian Literary Fragments in the University Museum, Philadelphia. BPOA 9. Madrid: CSIC. Peterson, Jeremiah. 2011b. The Personal Name Lists in the Scribal Curriculum of Old Babylonian Nippur: an Overview. ZA 101: 246–73. Peterson, Jeremiah. 2013b. A Catalog of Old Babylonian Sumerian Incantations and Rituals from Nippur in the University Museum, Philadelphia. NABU 1: 1–2. Peterson, Jeremiah. 2015. An Adab Composition of Nergal/Meslamtaea at Lagaš and Girsu for Šulgi. JCS 67: 45–63. Rutz, Matthew. 2006. Textual Transmission Between Babylonia and Susa: A New Solar Omen Compendium. JCS 58: 63–96. Rutz, Matthew. 2013. Bodies of Knowledge in Ancient Mesopotamia: The Diviners of Late Bronze Age Emar and their Tablet Collection. AMD 9. Leiden: E. J. Brill.

A Middle Babylonian Sumerian Fragment of the Adapa Myth from Nippur

283

Sassmanshausen, L. 2008. Babylonische Schriftkultur des 2. Jahrtausends v. Chr. in den Nachbarländern und im östlichen Mittelmeerraum. AuOr 26: 263–93. Schwemer, Daniel. 2015. Secret Knowledge of Lu-Nanna, the Sage of Ur: Six Astral Rituals for Gaining Power and Success (BM 38599). Pp. 211–28 in Saeculum: Gedenkschrift fur Heinrich Otten anlasslich seines 100. Geburtstags, ed. Andreas Muller-Karpe Wiesbaden: Harrassowitz Verlag. Shaffer, Aaron. 2000. A New Look at Some Old Catalogues. Pp. 429–36 in Wisdom, Gods, and Literature, Studies in Assyriology in Honor of W. G. Lambert, ed. Andrew R. George Winona Lake, IN: Eisenbrauns. Shaffer, Aaron and Marie-Christine Ludwig. 2006. Ur Excavations Texts VI: Literary and Religious Texts, Third Part. London: British Museum Press. Sjöberg, Åke. 1974. A Hymn to dLama-sa6-ga. JCS 26: 158–77. Sjöberg, Åke. 1975. In-nin šà-gur4-ra: A Hymn to the Goddess Inanna by the en-Priestess Enḫeduana. ZA 65: 161–253. Talon, Philippe. 1990. L’Mythe d’Adapa. SEL 7: 43–57. Taylor, Jon and Caroline Cartwright. 2011. The Making and Re-Making of Clay Tablets. Scienze dell’Antichità 17: 297–324. Vanstiphout, Herman L. J. 2003. The Old Babylonian Literary Canon: Structure, Function, and Intention. Pp. 1–28 in Cultural Repertoires: Structure, Function and Dynamics, ed. Gillis L. Dorleijn and Herman L. J. Vanstiphout. Leuven: Peeters. Veldhuis, Niek. 2000. Kassite Exercises: Literary and Lexical Extracts. JCS 52: 67–94. Veldhuis, Niek. 2008. Kurigalzu’s Statue Inscription. JCS 60: 25–51. Veldhuis, Niek. 2014. History of the Cuneiform Lexical Tradition. Guides to the Mesopotamian Textual Record 6. Münster: Ugarit Verlag. Viano, Maurizio. 2008–2009. La tradizione litteraria sumerica negli archive siro-anatolici durante il Tardo Bronzo. Ph.D. diss, Università degli studi di Trieste. Viano, Maurizio. 2012a. A Middle Babylonian Fragment of Angim in the University of Pennsylvania. NABU 31: 42–44. Viano, Maurizio. 2012b. A Sumerian Hymn from Boǧhazköy. WO 42: 231–37. Viano, Maurizio. 2015. Writing Sumerian in the West. Pp. 381–92 in Tradition and Innovation in the Ancient Near East: Proceedings of the 57th Rencontre Assyriologique Internationale at Rome 4–8 July 2011, ed. Alfonso Archi. Winona Lake, IN: Eisenbrauns. Wagensonner, Klaus. 2011. A Scribal Family and its Orthographic Peculiarities: On the Scientific Work of a Royal Scribe and his Sons. Pp. 645–701 in The Empirical Dimension of Ancient Near Eastern Studies, ed. Gebhard Selz. WOO 6. Wien: LIT Verlag. Wagensonner, Klaus. 2014. A Note on the Colophon of VAT 9487. CDLN 19. Wasserman, Nathan and Uri Gabbay. 2005. Literatures in Contact: The balaŋ uru2-am3-ma-irra-bi and its Akkadian Translation UET 6/2, 403. JCS 57: 69–84. Wilcke, Claus. 1976. Formale Gesichtspunkte in der sumerischen Literatur. Pp. 205–316 in Sumerological Studies in Honor of Thorkild Jacobsen on his Seventieth Birthday, June 7, 1974, ed. Stephen J. Lieberman. AS 20. Chicago: Oriental Institute. Wilcke, Claus. 2012. The Sumerian Poem Enmerkar and En-suhkeš-ana: Epic, Play, Or? Stage Craft at the Turn from the Third to the Second Millennium B.C. AOS Essay 12. New Haven: American Oriental Society. Wilson, Mark. 2008. Education in the Earliest Schools: Cuneiform Manuscripts in the Cotsen Collection, Los Angeles. Los Angeles: Cotsen Occasional Press.

Gonzalo Rubio

Sumerian Temples and Arabian Horses: On Sumerian e2-gal Para Miguel, amigo y maestro, “El que descubre con placer una etimología.” (Borges)

The Sumerian term e2-gal is commonly regarded as the ancestor of words meaning “temple” and “palace” in several Semitic languages: Ugaritic hkl, Ara), Clasmaic hkl/hykl, Hebrew hêkhāl (‫ל‬‫י‬), Syriac haykəlā, Arabic haykal ( sical Ethiopic haykal/ḫaykal, etc.1 The nexus between these Sumerian and Semitic lexical items is often assumed to be Akkadian ekallu, which would have been borrowed into Ugaritic in the Bronze Age and into Aramaic in the Iron Age. From Aramaic it would have spread to other Semitic languages: Sumerian > Akkadian > Aramaic > Syriac > Arabic > Ethiopic (Fraenkel 1886: 274–75; Zimmern 1917: 8; Leslau 1990: 79). This line of borrowing, however, poses a serious problem. Whereas Akkadian ekallu (clearly from e2-gal) has no initial consonant, the alleged loanwords in all the other Semitic languages exhibit an initial h, normally a voiceless glottal fricative.

1 E2 and the Ghostly Fricative In order to solve this conundrum, it has been proposed that the Sumerian word e2-gal was realized as /hekal/ or /haykal/. According to this scenario, West Semitic languages would have borrowed the term directly from an early (otherwise unattested) dialect of Akkadian, which would have preserved the initial

1 Del Olmo & Sanmartín 2015: I 330; Hoftijzer & Jongeling 1995: I 278; Koehler-Baumgartner 2001: 244–45; Brocklemann-Sokoloff 2009: 340–41; Dozy 1881: II 783; Leslau 1991: 221. The alternation h/ḫ in Classical Ethiopic is the result of the confusion between laryngeals that occurs in most Ethiopic manuscripts, a confusion that is reflected in the traditional, inherited pronunciation of Ge‘ez (Mittwoch 1926: 8). Thus, the form ḫaykal may be a hyper-corrective spelling triggered by the increasing and indiscriminate merger of ḫ with h and ḥ in the actual pronunciation (cp. Ullendorff 1955: 35–36; Leslau 1991: xix-xx). In Amharic, a language in which the characters for inherited /ḫ/, /ḥ/, and /h/ are all realized as [h], the word is simply haykäl, but spelled with the character that would etymologically correspond to /ḥ/ in Ge‘ez (Kane 1990: I 26) https://doi.org/10.1515/9781501503696-017

Sumerian Temples and Arabian Horses: On Sumerian e2-gal

285

segment (/h/), which all other dialects of Akkadian lost, as ekallu attests (Kaufman 1974: 27; Kogan 2015: 355). It has also been suggested, with far less likelihood, that West Semitic would have borrowed the word directly from Sumerian (Falkenstein 1959: 24; Lipiński 1975: 26). Others have searched for an explanation within the cuneiform writing interface and its use for early East Semitic. The sign E2 (LAK-732), read ’à in 3rd-millennium Semitic names and texts, almost certainly corresponded to /ha(y)/. Ebla and Mari texts exhibit forms with ’à corresponding to /ha/ and /ḥa/; e.g.:2 Ebla: ’à-a-gúm /halākum/ (VE 826 D; Conti 1990: 198) and i-ba-ti-’à-an /yippatiḥ-an/ “it was opened” (ARET 13: 1 obv. iv 8), as well as den-ki = ’à-u9 /ḥayyu(m)/ (VE 803; Conti 1990: 193). See Catagnoti 2012: 16. Mari: il-ga-’à /ilqaḥa/ “he took” and ’à-wa-tum (probably /hawātum/) “word” in liver models from the šakkanakku period. See Gelb 1981: 20–2.

The spelling ’à-wa-tum matches the Ugaritic term hwt (Gelb 1981: 20–21; Del Olmo & Sanmartín 2015: 344–45; Fronzaroli 2004: 44). In fact, in Sargonic Akkadian this word is spelled with E2: ’à-wa-a-ti /hawātī/ “my word,” ’à-waad-zu /hawātsu/ “his word,” etc. (Hasselbach 2005: 79). It is unclear, however, why this word would have an initial /h/, from an alleged root *hw(w) or *hw(y), which would occur with this meaning, “word, speech,” only in Akkadian and in Ugaritic (Cohen et al. 1994: V, 386). A connection with Arabic hawwata “he cried” seems tenuous (pace Ward 1969). In Akkadian, awātu is a feminine noun from the root of the verb awû “to argue,” atwû “to discuss” (CAD A/2: 29, 86). In Arabic, the /t/ is part of the root, hwt, a variant of hyt: hayyata, hawwata “he hailed, shouted to people.” This Arabic root would have originated in the common exclamation hayta, attested already in the Qur’ān, but whose exact meaning, origin, and grammatical nature remains elusive; e.g., Qur’ān 12:23: hayta laka “come here” or “here, I’m yours.”3 In Sargonic Akkadian, E2 can write /ḥ/: ’á-ru-uš /ḥaruθ/ “cultivate”; ’á-rašè /ḥarāθi/ “to cultivate”; da-la-’á-mu /talaḥḥamu/ “you will eat”; na-’á-ás-su /naḥassu/ “his life”; na-’á-ás /naḥas/ “life”; na-’á-si /naḥasi/ “life” (Hasselbach 2005: 81). Although our knowledge of Ur III Akkadian is limited mostly

2 See especially Tonietti 2003; Rubio 2006: 115–16 (with references); Catagnoti 2013: 52. Additionally, on the reading ’à of E2, see Lieberman 1977: 125, 216–17. Concerning texts from Fāra and Abū Ṣalābīḫ, see Krebernik 1998: 289. 3 See Lane 1863–93: 2905, 2909–10; Bosworth 1976: II, 341; Wehr 1976: 1042a; Corriente & Ferrando 2005: 1250a, 1254–55; Badawi & Haleem 2008: 998–99. On the Ugaritic form, see also Smith 1994: 160 n. 79.

286

Gonzalo Rubio

to names, one does find a-na na-’à-si-šu /ana naḥasišu/ “for his life” in an Akkadian inscription of Šu-Sîn (Gelb & Kienast 1990: 259a; Hilgert 2002: 477). Moreover, in Sargonic texts, E2 can also stand for ‘ayin: ’á-zum /‘azzum/ “fiery” (Hasselbach 2005: 83). Note, on the other hand, that neither in Ebla nor in the Sargonic corpus does E2 seem to be used for inherited alif (Hasselbach 2005: 79; Catagnoti 2012: 16). Thus, one can state that in third-millennium East Semitic texts, the cuneiform sign E2 can be used to write a sequence of fricative plus vowel, but not a stop: it can represent /h/ (voiceless glottal fricative), /ḥ/ (voiceless pharyngeal fricative), and even /ˁ/ (voiced pharyngeal fricative), but not /ˀ/ (voiceless glottal stop). In regard to the reading /ha(y)/ of E2, it has also been suggested that the existence of another Sumerian word for “house,” ga2, would have constituted an attempt to write the glottal fricative of /ha(y)/ (Gelb 1961: 26; Rubio 2006: 115). The latter piece of the puzzle does not really hold up, as the emesal equivalent of ga2 is ma, pointing thus to a nasal velar (ŋa2), which generally rules out a reading with an initial glottal fricative (Krecher 1978: 38; Schretter 1990: 189). Building on speculations first put forward by Falkenstein (1959: 24) and Gelb (1961: 25–26), Edzard (2003: 19) and Jagersma (2010: 48–49) argued that Sumerian would have had, early on, an /h/ segment. This segment would have been similar to the Semitic /h/, which Akkadian lost (as in *ahlu > ālu), but different from the other Sumerian fricative segment, which is written with the syllabograms normally used to write etymological /ḫ/ in Akkadian. This ghostly segment in early Sumerian would have occured in a small handful of lexical items: a2 “arm” (*/ha/); id2 “river” (*/hid/); ab2 “cow” (*/hab/); amar “calf” (*/’amahar/); la2 “to hang” (*/lah/); gada “linen” (*/gadah/); perhaps u2 “grass (*/hu/).4 Unfortunately, the presence of a segment in so few lexical items seems quite unlikely: this hypothetical segment would have been lost in all phonotactic contexts (initial, medial, final) and it would have been limited

4 The idea that id2 “river” could mask an earlier */hid/ is predicated on assuming a relation between the former and the Arabic toponym Hīt ( ), an Iraqi city on the Euphrates, as well as the first part of the Hebrew name of the Tigris, Ḥiddeqel (‫ל‬ < Akk. idiglat); see Gelb 1961: 26; Edzard 2003: 19; Jagersma 2010: 49. On the possibility that modern Hīt could correspond to Old Babylonian Ida/Ita, Middle Babylonian Idu, and 1st-millennium Itu/Ītu, see RGTC 3: 104– 15; 5: 135; 8: 184; and Postgate, RlA 5: 33. On the other hand, the theory that modern Hīt may correspond to a hypothetical “Tuttul on the Euphrates” (e.g., Astour 2002: 68–69) – as opposed to Tuttul on the Balīḫ, modern Tell Bī‘a – is probably to be discarded, since there is little or no basis to support the existence of such a Tuttul (Krebernik 2001: 3; Pappi, RlA 14: 233). Thus, if modern Hīt cannot be this hypothetical additional Tuttul, then the Ida/Ita, Idu/ Itu of Babylonian texts may well be the forerunner of this Arabic toponym.

Sumerian Temples and Arabian Horses: On Sumerian e2-gal

287

to a minute fraction of the lexicon.5 Moreover, Edzard (2003: 19–20) suggested that this phantom */h/ would have been present in instances in which a number of verbal and nominal stems ending in a vowel are followed by a suffix beginning with a vowel, without assimilation noted in the spelling: e.g., munus u4-da-e-ne (Ukg. 6 iii 23′), “women of today,” would have been /munus udah-ene/. In such instances, /h/ would be working mostly as a glottal stop (like an alif hamza), rather than a fricative, and it would be reflected in a graphic hiatus coinciding with a morphophonemic boundary.6 The correspondence between graphic hiatus and glottal stop is well attested in the writing interface of languages across the world (for instance, see Minkova 2004: 143– 44). Nevertheless, the kind of argument deployed by Edzard and Jagersma misunderstands the relation between language and writing interface, as it assumes that the latter is a careful (either phonological or phonetic) reflection of the realities of the former, a scenario rather alien to actual natural languages.

2 From Sumerian Temples to Arabian Horses Even if we were going to accept the existence of a long-lost /h/ segment in early (proto-historical) Sumerian, the problem would still remain unsolved: why does Akkadian not have /hekallu/ or /haykallu/? Since we only have ekallu, how would it be possible for that initial /h-/, allegedly attested in Sumerian but apparently lost in Akkadian, to eventually resurface in all Semitic languages over a millennium after the death of Sumerian as anyone’s mother

5 See also a critical review of the alleged evidence in Keetman 2006 (esp. 11–12). The proposal that Sumerian had an otherwise invisible /h/ is the sort of ad hoc explanation that tends to abound in some Sumerological literature, especially when it pertains to language: grammatical rules are fabricated in order to explain isolated attestations and simple idiolectic phenomena, with blissful disregard for the factual distinction between competence and performance, langue et parole. As a result, this widespread modern scholastic version of “Sumerian” does not look like a natural language, but resembles instead a construct that knows no actual hierarchical or functional rules: a language whose grammar seems governed by exceptions rather than by cross-linguistically attested rules. 6 The same idea can be found in Attinger (1993: 211–12). However, Attinger and Edzard used an archiphonemic capital /H/ to indicate a glottal, which is not clearly defined as either a fricative (/h/) or a stop (/ˀ/). On the other hand, note that Jagersma (2010: 38–41) advocates for a glottal stop as a fully phonemic segment, mostly in similar morphophonemic contexts and for the same reasons as Attinger and Edzard. On the use of Semitic readings of 3rd-mill. cuneiform to support such theories, see Meyer-Laurin 2011: 35 (for the Sargonic Akkadian evidence, see Hasselbach 2005: 64, 125–26).

288

Gonzalo Rubio

tongue? Of course, this problem would disappear if this were originally a Semitic loanword into Sumerian (from a hypothetical root *hykl) that had entered Sumerian with a spelling e2-gal, which had performed the double duty of writing the initial /h/ with E2 = ’à and providing a “learned folk” etymology for the term (“large house”).7 Such a word would have then been passed to Akkadian as ekallu, whereas the forms with initial /h-/ in all other Semitic languages would be mere cognates of Akkadian instead of loanwords. Unfortunately, there is very little basis to reconstruct a root *hykl or hkl.8 It is important to note that, whereas Sumerian e2-gal and Akk. ekallu can mean both “temple” and “palace,” their apparent heirs often show one meaning rather than the other. Hebrew hêkhāl does mean both “palace” and “temple,” and it can also be used to refer to the “middle area” or “holy place” (qōdeš) in Solomon’s temple (Ottosson 1997; Koehler-Baumgartner 2001: 244– 45). However, earlier on, in the Late Bronze Age, Ugaritic hkl meant only “palace” (Del Olmo & Sanmartín 2015: I 330). Conversely, the term hykl generally means “temple” in most ancient Aramaic dialects, with a few exceptions, such as the Story of Aḥīqar and Qumran, in which it means “palace.”9 Likewise, in Judean Aramaic (the dialect of non-literary texts from the Maccabean to the Tannaitic period), in Jewish Palestinian Aramaic (3rd cent. C.E. to the Arab Conquest), and in Jewish Babylonian Aramaic (Talmud Bavli and Gaonic literature), hykl/hêklā’ consistently means “temple,” with a few exceptions attested in incantation bowls.10 In Syriac, however, haykəlā exhibits the whole gamut of meanings from “palace” to “temple,” including “holy place (in a temple),” “shrine,” “nave,” “church,” and even the name of a star (Brockelmann-Sokoloff 2009: 340–41). Arabic haykal can designate any object perceived as enormous or colossal, from a celestial body to a building (Dozy 1881: II 783). In Classical Arabic,

7 Krebernik (1984a: 200, 321, 342) seems to hint at this possibility. 8 Cohen’s etymological dictionary (1994-) lists both a root *hykl (V, 401–402) and a root *’kl (I, 18b) –homophonous with the root of the verb “to eat” – although it states that the latter would ultimately be a loanword from Sumerian. However, it is most likely a loanword from Sumerian into Akkadian, not from Sumerian into Semitic at large, as a language family. The evidence concerning all subsequent borrowings (from Aramaic into Hebrew and Arabic, etc.) casts doubt on the historical basis to reconstruct a common Semitic root, whether originally borrowed from Sumerian by Semitic or not. It is clear, nonetheless, that there is no Semitic quadriliteral root *hykl, which would have been extraordinarily unproductive in any event. 9 See Hoftijzer & Jongeling 1995: I 278; Cook 2015: 64. For the occurrences in Aḥīqar, see Porten & Yardeni 1993: 26–31 (C.1.1 obv. i 9; ii 17, 23; iii 44). In Biblical Aramaic (Daniel and Ezra), it can mean both “palace” and “temple” (Koehler-Baumgartner 2001: II, 1859b). 10 See Sokoloff 2002a: 163; 2002b: 377; 2003: 43. For hykl meaning “palace” in Jewish Babylonian Aramaic bowls, see Shaked et al. 2013: 80.

Sumerian Temples and Arabian Horses: On Sumerian e2-gal

289

haykal had three basic meanings: a large horse; a large, tall building; and a Jewish temple or a Christian church (Boutz 2009: 127–29). In his famous mu‘allaqa, the pre-Islamic poet Imrū’ al-Qays (6th cent. C.E.) uses haykal to refer to a horse, in a context that makes clear this was a sturdy and sizable mount (53):11

wa-qad ’aġtadī wa-ṭ-ṭayru fī wukunāti-hā bi-munjaridin qaydi l-awābidi haykali I would often be off with the morning, the birds still in their nests, on a sleek steed (haykal) that overstrips wild beasts (awābid).

The expected meaning of haykal as a place of worship does occur in a Ġassānid ode by Ḥassān ibn Thābit of Yathrib/Madīna (7th cent.), one of the few poets whose œuvre bridges pre-Islamic and Islamic times (poem 13, 20):12

fa-la-qad yarānī mū‘idīyya ka-’annanī fī qaṣri dūma ’aw sawā’i al-haykali My ill-wishers saw me as though I were in the fortress (qaṣr) of Dūma or in the centre of the church (haykal).

The word haykal (pl. hayākil) acquired a vast array of meanings in later Arabic, all the way to modern times: from the earlier ones designating large or high structures (“temple,” “large building,” “altar,” “edifice”) to more specific denotations pertaining to objects with some sort of frame or well-defined contour (“skeleton,” “framework,” “frame,” “chassis,” “pagoda”), more often than not in keeping with the earlier general connotation of a gigantic or colossal object or being (Wehr & Cowan 1976: 1043b; Corriente & Ferrando 2005: 1256b). This was precisely Dozy’s (1881: 783b) early attempt at defining the term: “ [haykal] désigne dans l’origine tout ce qui est colossal, d’un grandeur ou d’une étendue extraordinaire … les corpes célestes … en parlant de montagnes, etc.”

11 See Zwettler 1978: 252–53 (with sources and variants); Pinckney Stetkevych 1993: 254, 315. A similar kind of semantic development based on an architectual metaphor was common at least in early and mid-20th century Spanish, when the word monumento (“monument”) was deployed as an objectifying term to designate a very attractive, usually tall and slight curvaceous, woman (cp. Seco et al. 2011: II, 3096b). In English, the word stacked, when applied to a woman, seems to also carry the connotations of a well-built structure, albeit in a more anatomically-targeted fashion (Green 2010: III, 1288). 12 The Ġassānids were a Byzantine client state in Syria and North Arabia in this period. For Thābit’s ode, see ‘Arafāt 1971: I 74 l. 20, II 78; Boutz 2009: 212–13. On the location of qaṣr dūma and the meaning of sawā’ al-haykal, see Shahîd 2002: 241–42, 283–87.

290

Gonzalo Rubio

3 Phonology and Spelling Aside from internal developments throughout the history of the Arabic lexicon, the etymological chain is marred by the presence of that initial /h/, which is absent in the entire Akkadian corpus. The Mesopotamian origin of all these otherwise obvously related Semitic words may be doubted for the reasons seen above: a) the /h/ is absent in Akkadian; b) even if one argues for a possible reading /ha(y)/ of E2, this is exceedingly unlikely to be a direct loan from Sumerian into Ugaritic, Aramaic, and so forth. Nonetheless, one should observe the occurrence of */haykalayn/ (hykryn) and */haykalaym/ (hykrym) in Egyptian texts from the New Kingdom, as Fremdwörter with the meanings “palace” and “temple” (Hoch 1994: 213). The term also appears as haikalli/ haigalli (ha-i-kal-li, ha-a-i-kal-li, ha-a-i-ga-al-li) in Hurrian texts from Boğazköy.13 In Hittite, a word meaning “temple,” which is translated as haigalli in Hurrian, exhibits a number of variants and inflectional oddities, as Puhvel (1991: 15–19) noted: halent(i)u-, hali(n)tu-, halant(i)u-. These irregularities have led to the suggestion that this could actually be a Hattic loanword in Hittite (Puhvel 1991: 19; Soyzal 2004: 143). Regardless, a connection between Hittite halent(i)u- and ekallu is quite far-fetched. Blau (1970: 49–50) argued that the /h-/ in Hebrew (hêkhāl) and other Semitic languages may have originated as a hypercorrection: speakers of early Aramaic, for instance, would have been aware that Akkadian lacked /h/ where Aramaic and other West Semitic languages did have it (e.g., alāku vs. hlk), and in this case they would have supplied it by false analogy. This would be similar to the phenomenon of h-insertion in English: speakers aware that they drop /h/ in words in which other (better educated, of higher social status) speakers do pronounce it, over-react by adding /h/ to words that should not have it, as Eliza Doolittle does in My Fair Lady, “in ’Ertford, ’Ereford, and ’Ampshire, ’urricanes ’ardly hever ’appen.” There is, however, no parallel for such a hypercorrection with initial /h-/ in Akkadian loanwords in West Semitic (Mankowski 2000: 50–51). A reader familiar with the history of Romance languages may think here of the unetymological initial h that precedes the spelling of some words in French and in Spanish:

13 See Neu 1993: 113; 1996: 229–230, 548; Catsanicos 1996: 225, 238; Richter 2012: 117–18. This must be distinguished from the Hurrian word that occurs in Nuzi (HSS 14: 105 l. 30: a-i-gal-luḫu) and in an Amarna letter from Tušratta (EA 25 iii 49: ḫa-i-gal-la-at-ḫé), which refers to a horned animal and to a rhyton shaped like it (CAD A/1: 231b; S: 78b).

Sumerian Temples and Arabian Horses: On Sumerian e2-gal

– – – –

291

French huile “oil” < Old French oile, oille, uelie, uelie, huille, etc. < Latin ŏleum (von Wartburg et al. 1922–2002: 7, 341–45); French huit “eight”< Old French oit, uit, huict, huit, etc. < Latin ŏctō (von Wartburg et al. 1922–2002: 7, 305–308); Spanish huevo “egg” < Medieval uevo < Latin ōuum (Corominas & Pascual 1980–84: 3, 422); Spanish húmedo “wet” < Latin ūmĭdus (Corominas & Pascual 1980–84: 3, 425–26).

These are, however, internal orthographic developments in these Romance languages – e.g, in Spanish, all words that start with the diphthong /ue/ are spelled with an initial h- (hueco “hole,” huérfano “orphan,” huelga “labor strike”) – so any resemblance is of no significance. The assumed borrowing from Akkadian into languages as diverse as Ugaritic and Hurrian would have to date to the Middle or early Late Bronze period. Still, Akkadian is well-attested in various dialects throughout that time span, and nowhere does the word ekallu occur with an initial /h/. Kaufman (1974: 27) states that there is no parallel to this phenomenon. Nonetheless, it has been argued that there is at least another word whose origin is ultimately Sumerian and which exhibits an intial /h-/ in several Semitic languages, but not in Akkadian. The word for “steward,” abarakku (< Sumerian abrig), has been regarded as the etymon of hbrk in the Phoenician inscription of Azatiwada at Karatepe (Lipiński 1974: 45–47; Arbeitman 1980, Krebernik 1984b). As Krebernik (1984b: 91) noted, in Ebla one finds (VE 706–707; Conti 1990: 183): ABRIG = ’à-ba-ru12-gú /habarukku(m)/ ABRIG.MUNUS = ’à-ba-ru12-[ga]-tum /habarukkatu(m)/

Note the variants ’à-ga-ra-gú-um and ’à-ga-ra-gú, as well as ’à-ga-ra-ga-tum. However, there are some complications in reading hbrk b‘l as “majordomo of Ba‘al” or “majordomo of the master” in the Karatepe inscription (Röllig apud Çambel 1990: 58). For one, the title abarakku had been replaced by masennu (NBab mašennu) in Neo-Assyrian documents, and perhaps even in the Middle Assyrian period already, as there seem to be no syllabic spellings of abarakku after the Old Babylonian period (Whiting apud Millard 1994: 7 n. 14). This parallel, thus, remains unsubstantiated.14 14 If it is speculative to connect hbrk to abarakku, it is even more so to add to this equation the Biblical Hebrew hapax ’abrek (), as Lipiński (2004: 124–27) does. See Gen. 41:43 (in front of Joseph’s chariot): ֑  ‫ ֖יו‬ ‫֥אוּ‬ “and they cried in front of him, ‘abrek.’” Cp. Koehler-Baumgartner 2001: 10a; Westermann 2002: 95. On the relation between abarakku and abriqqu, see CAD A/1: 35, 62–63.

292

Gonzalo Rubio

4 The Sumerian Word for Temple In discussing the date of a possible borrowing from Akkadian into other Semitic languages, it is important to note that e2-gal is well-attested in ED IIIa and IIIb texts, but normally, as far as one can tell, with the meaning “palace.” On the other hand, eš3 is the common word for “temple” in documents and lexical texts from ED IIIa, from both Fāra and Abū Ṣalābīḫ, as well as Uruk (lunga eš3, W 17258 [ UVB 10 pl 26 b, obv. v 6]).15 In this regard, eš3 was most likely the earlier word for “temple” and “temple household” in Sumerian (Gelb et al. 1991: 40).16 In fact, the sign AB, probably with the reading eš3, most likely corresponded to the term for “temple” already in the Uruk period (Selz 1997: 187 n. 26).17 The reading eš3 of AB may also be found in the archaic texts from Ur (Lecompte 2013: 98–100, 110, 115), in which e2-gal occurs as “palace” (UET 2: 112, 235, 349). Nevertheless, aside from specific readings, a general meaning “dwelling” (particularly of deities) matches most attestations of AB in archaic texts from the Uruk period, and it explains its occurrence in the compound signs used to write the names of a number of 3rd-millennium cities in Sumer (Michalowski 1993; Sallaberger, RlA 10: 200; RlA 13: 519–520). If e2-gal is probably not the earliest word for “temple” (nor probably “palace”) in Sumerian, one may entertain the idea that it reflects the etymologizing spelling of a word of unknown origin, a Kulturwort or Wanderwort that would have spread throughout a number of Ancient Near Eastern languages from the Early Bronze Age on (Watson 2007: 147).

5 Inconclusive Conclusion In order to explain the oddities involved in the assumed relation between Sumerian e2-gal, Akkadian ekallu, and the various Semitic words in question, the evidence mentioned above can lead to a number of different scenarios: 15 In Fāra, eš3 seems to refer mostly to the e2 dsud3 “the temple of Sud”; see Martin et al. 2001: 25–26. 16 This makes particular sense when compared to the general assumption that, in Mesopotamia, temples predated palaces as building structures of social and architectural significance (Trigger 2003: 202). However, it is not always easy to determine the original function of a specific structure, as in the case of the so-called “Four-hall Building” (“Bau mit den vier Sälen”), Building E in the Eanna precinct at Uruk (Bretschneider 2007). 17 Englund (1998: 70, 209) argues that AB simply designates a “large household,” as opposed to E2 referring to a regular “household,” in Uruk texts. However, he concedes that, in texts from Jemdet Naṣr, AB does seem to correspond to a “temple household” (Englund 1998: 81).

Sumerian Temples and Arabian Horses: On Sumerian e2-gal

293

a) Sumerian e2-gal represented /haygal/ or /haykal/, and this word was borrowed by Akkadian. Some Akkadian dialect or dialectal variant, which has not been preserved in writing, kept Semitic *h and used, therefore, a realization /hekallu/, /hēkallu/, /haykallu/, or the like. All other Semitic languages (beginning with Ugaritic and Aramaic) then borrowed this otherwise unattested Akkadian form, and so did both Hurrian (as a loanword) and Middle-Kingdom Egyptian (as a Fremdwort). b) Some Semitic language (perhaps Ugaritic) borrowed e2-gal directly from Sumerian, with an original Sumerian realization /haygal/ or /haykal/. Akkadian also borrowed it from Sumerian, but with the expected disappearance of the voiceless glottal fricative. The number of alleged direct borrowings from Sumerian into Ugaritic is quite small, and almost all these lexical items can be explained without resorting to Sumerian (Watson 2007: 145– 46). c) Sumerian e2-gal was borrowed into Akkadian as ekallu. Other Semitic languages, however, added an initial voiceless glottal fricative by some sort of hypercorrection, perhaps by analogy with other words that had initial /h-/ in Ugaritic or Aramaic but had lost it in Akkadian (as is regular). Hurrian and Egyptian then borrowed it from West Semitic. The lack of parallels makes this theory unlikely. d) There was a Semitic root *hykl or *hkl, which was borrowed into Sumerian as e2-gal, in a sort of etymologizing spelling that matched the meaning, which therefore represented /haygal/ or /haykal/. Akkadian subsequently borrowed it from Sumerian as ekallu, and this would not be a true cognate of the other Semitic words, but a re-borrowing through Sumerian (Watson 2005: 193). While possible, the internal Semitic evidence points to a chain of borrowing rather than to a common reconstructable root. e) Akkadian borrowed ekallu from Sumerian, but all the other Semitic words that are usually assumed to be related are not: all resemblance, in shape and meaning, is merely coincidental. Ugaritic and Aramaic borrowed quite a number of words (especially bureaucratic and administrative terminology) from Akkadian, so the idea that ekallu and hkl/hykl could be mere lookalikes is difficult to accept. f) The Akkadian spelling ekallu constituted a standardized, graphicallygrounded, and pseudo-morphophonemic spelling of an actual form

See also Szarzyńska 2011; Cohen 2005:132–33. Note that the sequence E2.GAL does not seem to occur in the archaic texts from Uruk (cp. ZATU: 196), although there are a couple of titillating, albeit ambiguous, sequences in a list of toponyms from Uruk (ATU 3: W 19948,24+) and in a document from Jemdet Naṣr (MSVO 1: 128).

294

Gonzalo Rubio

/haykallu/ or the like (not unlike the spelling of the name of the god Ea), which was borrowed from Sumerian e2-gal realized as /haygal/ or /haykal/. In the Middle Bronze Age, West Semitic speakers and scribes were exposed to the realization of the word behind this frozen spelling and borrowed it with its initial /h-/. This hypothesis must also assume the existence of a variety or varieties of Akkadian, in which the voicless glottal fricative /h/ was somehow preserved in the spoken dialect but not in any scribal orthographic tradition, since the syllabograms with {ḫ} were reserved mostly for etymological *ḫ. Albeit tempting, this remains impossible to prove.18 g) The term in question was neither Sumerian nor Semitic in origin, but a Wanderwort or Arealwort, which made it into Akkadian, Hurrian, and Semitic languages throughout the Bronze Age and continued to spread within Semitic in later periods. This is always possible, and it would explain the oddities surrounding the distribution of the initial /h-/. To various degrees of verisimilitude, all these possibilities have their merits, as well as their shortcomings. One can still put forward another setting that can perhaps explain the different treatment of the initial /h/ in Akkadian visà-vis the other Semitic languages. Sumerian e2-gal was borrowed directly both by Akkadian (as ekallu, with the expected loss of /h/) and by Hurrian.19 Hurrian haikalli/haigalli (see § 3 above) originated in Sumerian itself through a Semiticized understanding of the sign E2: scribes used the sign E2 because its Semitic readings (/h/, /ḥ/, /ˁ/) were retrofed into the Sumerian syllabary (see § 1 above).20 There is then the possibility that, in Sumerian itself, e2-gal was the etymologizing spelling of a Wanderwort or Arealwort. The latter would have contained an initial fricative different from Sumerian /h/, since this is a fricative segment whose exact point of articulation and phonation cannot be reconstructed. Sumerian /h/ is written with the syllabograms normally employed for inherited /ḫ/ in Akkadian, which corresponds to the Semitic voiceless velar fricative; but this does not necessarily define the exact nature of the Sumerian segment written with the same syllabograms.

18 There are instances in which Akkadian ḫ (or rather Akkadian syllabograms with {ḫ}) actually corresponds to Semitic *ġ; see Kogan 2001: 269–75. Likewise, there are other instances of irregular cognates, in which Akkadian ḫ corresponds to Semitic *ḥ, for instance; see Tropper 1995; Kogan 1995. 19 On the Hurrian linguistic presence in the third millennium, see Michalowski 1986; Steinkeller 1998; Wegner 2007: 21–24. 20 On the matter of Semiticization of Sumerian, see Civil & Rubio 1999.

Sumerian Temples and Arabian Horses: On Sumerian e2-gal

295

According to this hypothesis, it was from Hurrian that the term spread throughout Middle and Late Bronze Syria (Ugaritic, later on Aramaic). There are in fact quite a number of well-documented Hurrian loanwords in Ugaritic (Watson 2007: 124–135). This bifurcated line of borrowing from Sumerian into both Akkadian and Hurrian presents the advantage of doing away with the problem posed by the relation between ekallu and hkl/hykl, and it reflects the usage of the sign E2 in third-millennium texts.21 Nevertheless, in dealing with etymologies and loanwords, certainty often evades us.

Bibliography ‘Arafāt, Walīd N. 1971. Dīwān of Ḥassān ibn Thābit, I–II. Gibb Memorial, 25. London, Luzac. Arbeitman, Yoël L. 1980. E Luvia Lux. JANES 12: 9–11. Astour, Michael C. 2002. A Reconstruction of the History of Ebla (Part 2). Eblaitica 4: 57–195. Attinger, Pascal. 1993. Eléments de linguistique sumérienne: La construction de du11/e/di. OBO Sonderband. Göttingen: Vandenhoeck & Ruprecht. Badawi, El-Said M. and Muḥammad Abdel Haleem. 2008. Arabic-English Dictionary of Qur’anic Usage. Leiden: Brill. Blau, Joshua. 1970. On Pseudo-Corrections in Some Semitic Languages. Jerusalem: Israel Academy of Sciences and Humanities. Bosworth, Clifford E. 1976. The Mediaeval Islamic Underworld: The Banū Sāsān in Arabic Society and Literature, I–II. Leiden: Brill. Boutz, Jennifer Hill. 2009. Ḥassān ibn Thābit, a True mukhaḍram: A Study of the Ghassānid Odes of Ḥassān ibn Thābit. Ph.D. diss. Georgetown University. Bretschneider, Joachim. 2007. The “Reception Palace” of Uruk in its Architectural Origin. Pp. 11–22 in Power and Architecture: Monumental Public Architecture in the Bronze Age Near East and Aegean, ed. J. Bretschneider et al. OLA 156. Leuven: Peeters. Brockelmann, Carl and Michael Sokoloff. 2009. A Syriac Lexicon. Winona Lake, Ind.: Eisenbrauns. Catagnoti, Amalia. 2012. La grammatica della lingua di Ebla. QdS 29. Florence: Università di Firenze. Catagnoti, Amalia. 2013. La paleografia dei testi dell’amministrazione e della cancelleria di Ebla. QdS 30, Firenze: Università di Firenze. Catsanicos, Jean. 1996. L’apport de la bilingue de Ḫattuša à la lexicologie hourrite. Amurru 1: 197–296. Civil, Miguel and Gonzalo Rubio. 1999. An Ebla Incantation Against Insomnia and the Semiticization of Sumerian: Notes on ARET 5 8b and 9. Or. n.s. 68: 254–66.

21 In this scenario, Akkadian ekallu should be better normalized as ēkallu, with /ē/ as the result of the monophthongation of /ay/. The diphthong /ay/ contracts as /ē/ in Sargonic Akkadian as well as in Assyrian; Babylonian probably experienced a two-step development: /ay/ > /ē/ > /ī/. See Gelb 1981: 25; Hasselbach 2005: 41–42, 48, 91 n. 186.

296

Gonzalo Rubio

Cohen, Andrew C. 2004. Death Rituals, Ideology, and the Development of Early Mesopotamian Kingship: Toward a New Understanding of Iraq’s Royal Cemetery of Ur. AMD 7. Leiden: Brill/Styx. Cohen, David et al. 1994-. Dictionnaire des racines sémitiques ou attestées dans les langues sémitiques, 1-. Leuven: Peeters. Conti, Giovanni. 1990. Il sillabario della quarta fonte della lista lessicale bilingue Eblaita. MisEb 3 (QdS 17). Florence: Università di Firenze. Cook, Edward M. 2015. Dictionary of Qumran Aramaic. Winona Lake, Ind.: Eisenbrauns. Corominas, Joan and José A. Pascual. 1980–84. Diccionario crítico etimológico castellano e hispánico, I–VI. Madrid: Gredos. Corriente, Federico and Ignacio Ferrando. 2005. Diccionario avanzado árabe, I: Árabeespañol. Barcelona: Herder. Çambel, Halet, 1999. Corpus of Hieroglyphic Luwian Inscriptions, II: Karatepe-Aslantaş. Berlin: De Gruyter. Dozy, Reinhart. 1881. Supplément aux dictionnaires arabes. Leiden: Brill. Edzard, Dietz Otto. 2003. Sumerian Grammar. Leiden: Brill. Englund, Robert K. 1998. Texts from the Late Uruk Period. Pp. 13–233 in Mesopotamien, 1: Späturuk-Zeit und Frühdynastische Zeit. OBO 160/1. Göttingen: Vandenhoeck & Ruprecht. Falkenstein, Adam. 1959. Das Sumerische. Leiden: Brill. Fraenkel, Sigmund. 1886. Die aramäischen Fremdwörter im Arabischen. Leiden: Brill. Fronzaroli, Pelio. 2004. Le baiser cultuel à Ébla. N.A.B.U. 2004 no. 46, pp. 44–45. Gelb, Ignace J. 1961. Old Akkadian Writing and Grammar. 2nd ed. MAD 2. Chicago: University of Chicago Press. Gelb, Ignace J. 1981. Ebla and the Kish Civilization. Pp. 9–73 in La lingua di Ebla, ed. L. Cagni. Naples: Istituto Universitario Orientale. Gelb, Ignace J. and Burkhart Kienast. 1990. Die altakkadischen Königsinschriften des dritten Jahrtausends v. Chr. FAOS 7. Stuttgart: Franz Steiner. Gelb, Ignace, Piotr Steinkeller, Robert M. Whiting. 1991. Earliest Land Tenure Systems in the Near East: Ancient kudurrus, I–II. OIP 104. Chicago: Oriental Institute. Green, Jonathon. 2010. Green’s Dictionary of Slang, I–III. London: Chambers. Hilgert, Markus. 2002. Akkadisch in der Ur III-Zeit. Imgula 5. Münster: Rhema. Hoch, James E. 1994. Semitic Words in Egyptian Texts of the New Kingdom and Third Intermediate Period. Princeton, N.J.: Princeton University Press. Hoftijzer J. and K. Jongeling. 1995. Dictionary of the North-West Semitic Inscriptions, I–II. Leiden: Brill. Jagersma, Abraham H. 2010. A Descriptive Grammar of Sumerian. PhD. diss. University of Leiden. Kane, Thomas. 1990. Amharic-English Dictionary. Wiesbaden: Harrasowitz. Kaufman, Stephen A. 1974. The Akkadian Influences on Aramaic. AS 19. Chicago: University of Chicago Press. Keetman, Jan. 2006. Gab es ein h im Sumerischen? Babel und Bibel 3: 9–30. Koehler, Ludwig, Walter Baumgartner, et al. 2001. The Hebrew and Aramaic Lexicon of the Old Testament, I–II. Leiden: Brill. Kogan, Leonid. 1995. О нерегулярных рефлексах протосемитских ларингалов в аккадском языке. Вестник древней истории: 156–62. Kogan, Leonid. 2001. *ġ in Akkadian. UF 33: 263–98.

Sumerian Temples and Arabian Horses: On Sumerian e2-gal

297

Kogan, Leonid. 2015. Genealogical Classification of Semitic: The Lexical Isoglosses. Berlin: De Gruyter. Krebernik, Manfred. 1984a. Die Beschwörungen aus Fara und Ebla. TSO 2. Hildesheim: Olms. Krebernik, Manfred. 1984b. hbrk b‘l in den phön. Karatepe-Inschriften und ’à-ba-ra-gú in Ebla. WdO 15: 89–92. Krebernik, Manfred. 1998. Die Texte aus Fāra und Tell Abū Ṣalābīḫ. Pp. 235–427 in Mesopotamien, 1: Späturuk-Zeit und Frühdynastische Zeit. OBO 160/1. Göttingen: Vandenhoeck & Ruprecht. Krebernik, Manfred. 2001. Tall Bi‘a/Tuttul, II: Die altorientalischen Schriftfunde. WVDOG 100. Saarbrücken: SDV. Krecher, Joachim. 1978. Das sumerische Phonem /g̃/. Pp. 7–73 in Festschrift Lubor Matouš, ed. B. Hruška and G. Komoróczy. Assyriologia 5. Budapest: Eötvös Loránd Tudományegyetem, Ókori Történeti tanszékek. Lane, Edward W. 1863–93. Arabic-English Lexicon. London: Williams & Norgate (reprint, Cambridge: Islamic Texts Society, 1984). Lecompte, Camille. 2013. Archaic Tablets and Fragments from Ur. Ed. L. Verderame. Nisaba 28. Messina : DiCAM. Leslau, Wolf. 1990. Arabic Loanwords in Ethiopian Semitic. Wiesbaden: Harrassowitz. Leslau, Wolf. 1991. Comparative Dictionary of Ge‘ez. Wiesbaden: Harrassowitz. Lieberman, Stephen J. 1977. Sumerian Loanwords in Old Babylonian Akkadian, I. HSS 22. Missoula: Scholar Press. Lipiński, Edward. 1974. From Karatepe to Pyrgi: Middle Phoenician Miscellanea. Rivista di Studi Fenici 2: 45–61. Lipiński, Edward. 1975. Studies in Aramaic Inscriptions and Onomastics, I. OLA 1. Leuven: Leuven University Press. Lipiński, Edward. 2004. Itineraria Phoenicia. OLA 127. Peeters: Leuven. Mankowski, Paul V. 2000. Akkadian Loanwords in Biblical Hebrew. HSS 47. Winona Lake, Ind.: Eisenbrauns. Martin, Harriet P., Francesco Pomponio, Giuseppe Visicato, and Aage Westenholz. 2001. The Fara Tablets in the University of Pennsylvania Museum of Archaeology and Anthropology. Bethesda, Mad.: CDL Press. Meyer-Laurin, Vera. 2011. Die „Zeichenpaare“ im sargonischen Akkadisch aus sumerologischer Sicht. WdO 41: 27–68. Michalowski, Piotr. 1986. The Earliest Hurrian Toponymy: A New Sargonic Inscription. ZA 76: 4–11. Michalowski, Piotr. 1993. On the Early Toponymy of Sumer: A Contribution to the Study of Early Mesopotamian Writing. Pp. 119–33 in kinattūtu ša dārâti: Raphael Kutscher memorial volume, ed. A. F. Rainey. Tel Aviv: Tel Aviv University. Millard, Alan. 1993. The Eponyms of the Assyrian Empire, 910–612 BC. SAAS 2. Helsinki: The Neo-Assyrian Text Corpus Project. Minkova, Donka. 2004. Alliteration and Sound Change in Early English. Cambridge: Cambridge University Press. Mittwoch, Eugen. 1926. Die traditionelle Aussprache des Äthiopischen. Berlin: De Gruyter. Neu, Erich. 1993. Kešše-Epos und Epos der Freilassung. Studi micenei ed egeo-anatolici 31: 111–20. Neu, Erich. 1996. Das hurritische Epos der Freilassung, I. StBoT 32. Wiesbaden: Harrassowitz.

298

Gonzalo Rubio

del Olmo Lete, Gregorio and Joaquín Sanmartín. 2015. A Dictionary of the Ugaritic Language in the Alphabetic Tradition, I–II. 3rd ed. Leiden: Brill. Ottosson, M. 1997. ‫ל‬‫י‬ hêkhāl. Pp. 382–88 in Theological Dictionary of the Old Testament, III, ed. G. J. Botterweck & H. Ringgren (transl. J. T. Willis et al.). Grand Rapids, Mich.: Eerdmans. Pinckney Stetkevych, Suzanne. 1993. The Mute Immortals Speak: Pre-Islamic Poetry and the Poetics of Ritual. Ithaca, N.Y.: Cornell University Press. Porten, Bezalel and Ada Yardeni. 1993. Textbook of Aramaic Documents from Ancient Egypt, 3: Literature, Accounts, Lists. Jerusalem: The Hebrew University. Puhvel, Jan. 1991. Hittite Etymological Dictionary, 3: H. Berlin: Mouton de Gruyter. Richter, Thomas. 2012. Bibliographisches Glossar des Hurritischen. Wiesbaden: Harrassowitz. Rubio, Gonzalo. 2006. Eblaite, Akkadian, and East Semitic. Pp. 110–39 in The Akkadian Language in its Semitic Context, ed. N. J. C. Kouwenberg & G. Deutscher. Leiden: Nederlands Instituut voor het Nabije Oosten. Schretter, Manfred K. 1990. Emesal-Studien. Innsbrucker Beiträge zur Kulturwissenschaft 69. Innsbruck: Universität Innsbruck. Seco, Manuel, Olimpia Andrés, and Gabino Ramos. 2011. Diccionario del español actual, I–II. Madrid: Aguilar. Selz, Gebhard J. 1997. ‘The Holy Drum, the Spear, and the Harp’: Towards an Understanding of the Problems of Deification in Third Millennium Mesopotamia. Pp. 167–213 in Sumerian Gods and Their Representations, ed. I. J. Finkel & M. J. Geller. CM 7. Groningen: Styx. Shahîd, Irfan. 2002. Byzantium and the Arabs in the Sixth Century, 2/1: Toponyms, Monuments, Historical Geography and Frontier Studies. Washington, D.C.: Dumbarton Oaks. Shaked, Shaul, James Nathan Ford, and Siam Bhayro. 2013. Aramaic Bowl Spells: Jewish Babylonian Aramaic Bowls, 1. Leiden: Brill. Smith, Mark S. 1994. The Ugaritic Baal Cycle, I. Leiden: Brill. Sokoloff, Michael. 2002a. A Dictionary of Jewish Palestinian Aramaic. 2nd ed. Baltimore: Johns Hopkins University Press. Sokoloff, Michael. 2002b. A Dictionary of Jewish Babylonian Aramaic. Baltimore: Johns Hopkins University Press. Sokoloff, Michael. 2003. A Dictionary of Judean Aramaic. Ramat-Gan: Bar-Ilan University Press. Soysal, Oğuz. 2004. Hattischer Wortschatz in hethitischer Textüberlieferung. Leiden: Brill. Steinkeller, Piotr. 1998. The Historical Background of Urkesh and the Hurrian Beginnings in Northern Mesopotamia. Pp. 75–98 in Urkesh and the Hurrians: Studies in Honor of Lloyd Cotsen, ed. G. Buccellati & M. Kelly-Buccellati. BiMes 26. Malibu: Undena. Szarzyńska, Krystyna. 2011. Observations on the Temple Precinct EŠ3 in Archaic Uruk. JCS 63: 1–4. Tonietti, Maria Vittoria. 2003. É = bītum or = ’à? About Ea in Early Dynastic Sources. Pp. 666–79 in Semitic and Assyriological Studies Presented to Pelio Fronzaroli. Wiesbaden: Harrassowitz. Trigger, Bruce G. 2003. Understanding Early Civilizations: A Comparative Study. Cambridge: Cambridge University Press. Tropper, Josef. 1995. Akkadisch nuḫḫutu und die Repräsentation des Phonems /ḥ/ im Akkadischen. ZA 85: 58–66.

Sumerian Temples and Arabian Horses: On Sumerian e2-gal

299

Ullendorff, Edward. 1955. The Semitic Languages of Ethiopia: A Comparative Phonology. London: Taylor. Ward, William A. 1969. The Semitic Root hwy in Ugaritic and Derived Stems in Egyptian. JNES 28: 265–67. von Wartburg, Walther et al. 1922–2002. Französisches Etymologisches Wörterbuch, 1–25. Bonn: Klopp. Watson, Wilfred G. E. 2005. Loanwords in Semitic. AuOr 23: 191–98. Watson, Wilfred G. E. 2007. Lexical Studies in Ugaritic. AuOr Suppl. 19. Sabadell, Barcelona: AUSA. Wegner, Ilse. 2007. Hurritisch: Eine Einführung. 2nd ed. Wiesbaden: Harrassowitz. Wehr, Hans and J. W. Cowan. 1976. Arabic-English Dictionary. 3rd ed. Ithaca, N.Y.: Spoken Language Sevices. Westermann, Clauss. 2002. Genesis 37–50. Transl. J. J. Scullion. Minneapolis: Fortress. Zimmern, Heinrich. 1917. Akkadische Fremdwörter als Beweis für babylonischen Kultureinfluß. Leipzig: Hinrichs. Zwettler, Michael. 1978. The Oral Tradition of Classical Arabic Poetry. Columbus, Ohio: Ohio State University Press.

Gebhard J. Selz

To Carry Coals to Newcastle or Observations Concerning Sign Formation as an Early Concept in Mesopotamian “Commentaries” A house, based on foundations like the skies A house one has covered with a veil (like) a (secret) tablet box E2 uz-gin7 ki-gal-la gub-ba A house set on a base like a ‘goose’ Igi-nu-bad ba-an-ku4 One enters it blind, Igi-bad ba-an-ta-e3 Leaves it seeing. Ki-bur2-bi e2-dub-ba-a Answer: the School. Sumerian Riddle UET 6: 34 with parallels; Civil 1987: 19–20

E2-an-gin7 uru4 ğar-ra E2 dub-šen-gin7 gada mu-un-dul

It is with hesitation that I offer the following modest observations to such an outstanding scholar as Miguel Civil, without whose work, which is so rich in pertinent observations, Sumerology today would look much different. It is well known that Mesopotamian lexical lists are not only word lists (dictionaries) and sign lists.1 Recently, the diri (w)atru(m) lists have attracted much interest. Miguel Civil aptly defined diri signs as follows2: “[T]hey are signs with semantic, but not phonological contents, used to indicate to which semantic set a noun belongs. In some cases the classificators were in fact parts of a nominal, lexical compound and were pronounced, while a true classificator was never pronounced.”3

In other words, diri signs are compound graphemes representing Sumerian lexemes. Their phonetic shape is neither identical with nor similar to the combined syllabic values of the compound grapheme’s elements.4 However, the

1 For an excellent recent overview, see Veldhuis 2014, esp. 16–23. 2 See Civil 2004: 3. 3 The general notion that cuneiform classifiers were never pronounced will be challenged in a forthcoming article by Selz, Goldwasser, and Grinevald. 4 There are numerous cases in the diri lists where part of the compound sign most certainly functions as a phonetic indicator (mater lectionis). Another question is the pronunciation of the so-called “determinatives”, better “classifiers”; cp. preceeding fn. https://doi.org/10.1515/9781501503696-018

Sign Formation as an Early Concept in “Commentaries”

301

sign semantics – hereafter termed “sign etymologies” – provide a major source for reconstructing and understanding the Mesopotamian epistemic world.5 Such graphemes may even represent more than one Sumerian lexeme. It seems obvious that these writings date back to the invention of writing; numerous compound signs are attested already in the Uruk IV and Uruk III,6 some of them surviving until the late periods of cuneiform traditions. Rarely, such signs7 later decomposed into sign combinations.8 In a few cases, later-preserved compound / combined signs9 display diri-like features. One example is the sign ag2 = ninda2×ne, which perhaps occurred already in Uruk period texts (ZATU 21). The word /ninda/ (written ninda2) went as a loan into Akkadian as ittû, and the sign form depicts what the word designates: a “(seed-)funnel”. But the inscribed ne may convey the meaning attested with the reading ne = bir9 “to scatter, disperse” Akk. sapāḫu. If accepted, the sign etymology would be “the funnel for dispersing the seed (into the furrows)”. This fits very well with the usual meaning of ağ2 = madādu “to measure out (barley)”. The related sign ninda2ךe = šam2, better sa10, later with the Akkadian equation šâmu, is usually rendered “to pay for, buy; to be paid for, sell”. The basic meaning

5 Hilgert 2009; Johnson 2013. 6 Many of them consist of a frame sign with an inscribed second sign. Some of them survived until very late periods, especially those with ga2 and lagab as the frame sign. Many others simply disappeared. Whether a writing like dugךe [VESSEL]×[BARLEY] should be understood as “vessel for barley” or “vessel with barley” remains often uncertain. 7 Combined signs which consist of a restricted number of “frame signs” with other signs inscribed are well-attested among the Uruk period’s combined signs. Important frame signs include, according to the sign descriptions of Green and Nissen (1987): a (ZATU 12–15), dug (ZATU 89–124), ezem (ZATU 44, 151–157); ga2 (ZATU 163–183); gan (ZATU 191–194); lagab (ZATU 22, 76, 127, 200, 251, 309–322, 370, 430, 488, 616(?)); mah(?) (ZATU 242–351); ninda2 (ZATU 39, 402,-411, 510), and ZATU 737 (ZATU 185, 738–748). In later tradition the number of comparable signs is significantly smaller. 8 Cp., e.g. GANUN: the older form ga2×nun was latter written as ga2+ nun . Also pa5 has the older form e×pap, but was later written pap.e. Further dur, originally perhaps ZATU 126(?)), then gu2×kak, later gu2+kak; geš×tin (ZATU 202) fuses into a single sign and therefore could receive an additional classifier (or phonetic indicator) . Similarly, the sign šagan (ZATU 506) was originally a sign combination (according to Green and Nissen (1987) “a ligature dug+ku3”) that evolved into the sign to which a little later dug was added as an additional classifier: . It is suggestive, but due to the present stage of paleographic research rather difficult to substantiate, that not only compound signs but also diri writings appear in the earlier periods as combined signs. 9 As indicated, I tend to interpret a number of early compound signs as covert diri writing which is suggested by the fact that some of these very early signs are later written as sign combinations.

302

Gebhard J. Selz

seems to be something like “to let correspond to” which might be derived from the assumption that sa10 is a specific writing for the homophonic Sumerian verb sa2, also written sax(zag) or se3, “to equal, compare, compete, be equal to, rival”, in Akkadian kašādu; mašālu; šanānu. Therefore, the writing sa10 as ninda2ךe makes an implicit reference to barley as the most common medium of exchange. Another example, much more complicated example is the sign sug, also ambar, a combined sign of lagab×a, sug and ambar being the Sumerian words for “reed-beds” and “marshes”. The sign-etymology of this is something like [ENCIRCLED AREA]×[WATER]. Interestingly lagab×a also has a reading /buniĝ (bunin)/, a kind of trough, bowl, or bucket (= kuninu; pattû). In this case the sign etymology is slightly different: the writing must be understood as [ROUND CONTAINER]×[WATER]. The forms involving buniĝ, (bunin) provided by ePSD are:

[1]

ĝeš

buniĝ (gešbunin)

[2]

gi

[3]

buniĝ (bunin)

[4]

gi

[5]

ĝeš

buniĝ2 (gešbunin2)

[6]

ĝeš

buniĝ3 (gešbunin3)

[7]

kunin

buniĝx(|a.lagab×a|) (gibuninx(|a.lagab×a|))

buniĝ (gibunin)

Forms one [1] and [4] simply add the classifier for [WOOD] or [REED] to the basic word. Forms [5] and [6] use lagab×gar and lagab×hal, respectively, instead of lagab×a. Form [5] may have developed just by replacing [WATER] by [NOURISHMENT]. lagab×hal – form [6] –, on the other hand, is usually read engur, /engur/ being the Sumerian word for “underground water”. This is difficult to understand unless we take into account that lagab×hal is, in Early Dynastic texts, a writing for esir2 (RSP 264), in other texts written lagab×numun. /esir/ (=iṭṭû; kupru) designates “bitumen” This interpretation also has the reading /kunin/ and /gunin/ = is corroborated by the fact that kuninu, kuninnu; pattû, “a trough, bowl or bucket” (CAD K 539).10 The proposed sign etymologies therefore not only provide lexical information about the vessel type, they also indicate that buniĝ (bunin) and guniĝ

10 Thus, the sign certainly refers to a vessel made water-tight with bitumen!

Sign Formation as an Early Concept in “Commentaries”

303

(kunin) are the same word, displaying the well-known alteration between the consonants /b/ and /g/. So far such observations are still rare, especially due to our limited understanding of the evolution of the semasiographic writing system. However, they do shed some light on the origins and principles of diri writings. Judith Pfitzner recently studied a large number of OB Diri compounds. The major aim of her research was to explore the Old Babylonian scholarly interpretations of these writings. She focused on NOUN+NOUN combinations and demonstrated that in all interpretable cases the diri writings followed the rule of the normal Sumerian noun combinations, specifically the common left-headed NOUN+NOUN combinations, some of them even conforming to the normal structure of a Sumerian genitive compound. This suggests that in practice the diri writings could be read sign by sign by the scribes. This opens a new window on the practical significance of these diri writings, an issue to which I will return below. To the best of my knowledge, the formation and evolution of diri signs has never been studied systematically. A brief look at three complex writings will illustrate the issue: 1. The Sumerian word for “army” ugnim / uĝnim: {ugnim} “army; camp” = ki.kuš.lu.ub₂.gar11 Sign etymology = ki [PLACE] kuš [LEATHER] lu.ub {lub, bean-shaped bags} gar [TO PLACE] Hence “camp”, “place where the bean-shaped leather bags are placed” which in metonymic use also designates “army”. The etymology of the Sumerian word /uĝnim/ “soldiers, troops” is probably something like “tall men” (cp. Prussian „Lange Kerls“). 2. The Sumerian word for “nest” gud5: {gud5}[(gi.)u2.ki.se3.ga =gud5 “nest” Sign etymology = u2 [PLANT] ki [PLACE, EARTH] se3.ga [PLACED, SET ON GROUND]12 Hence: “nest”, “plants placed on ground” Note that the classifier gi [REED] is attested only in later texts, indicating the [OBJECT CLASS], which is the material from which nests were made.13

11 Compare Nigga 540. LEX/Old Babylonian/unknown [[ugnim]] = ki.su.lu.ub₂.gar = um-manu-um OB Diri “Oxford” 342. 12 Note that se3-ga has the grammatical form of a past perfective / passive participle. 13 Cp. the Debate between Bird and Fish l. 22: “Then Fish laid its eggs in the lagoons; Bird built its nest in a gap in the reedbeds.”

304 3.

Gebhard J. Selz

Also relevant to our topic are the written forms of a number of early place names. For example the writing of Akšak(ki): (ud/a/itu.kusu2) < a-kušu2 “Place of the Crab /Turtle-Waters”.14 15

Such examples suggest that diri writings possessed the inherent function of a commentary,16 providing definitions of a given Sumerian word in the form of an underlying question and answer.17 Compare, for example: Question: “What is uĝnim?” = “(a camp for) soldiers” (“army”, by metonymy)18 Answer: “A place where the bean-shaped leather bags are placed.” Question: “What is a nest?” Answer: “Plants placed on the ground.” Question: “What does the place name Akšak mean?” Answer: “Place of the Crab /Turtle-Waters.” The hypothesis that diri writings often provide definitions – or commentaries – of/on Sumerian lexemes is illustrated by the further, by no means exhaustive, examples: ) “to a) Q: “What is er2?” (“tear”; later texts also verbally šešx and šeš3 (= weep”) WATER+EYE

A.IGI

er2 (ir2, era2, ira2) eseš, esiš3, eš9, issš2, šešx (|A.IGI|)

A: “Water (flowing) from the eyes.”

14 Lambert 1990: 77–78. 15 The different ways of rendering place names − a) purely logographic; b) mixed, logographic with phonetic indicators (“syllables”); and c) purely syllabic –might reflect the different paleographic periods in which they originated. I discussed this system earlier (2013), and therefore it will suffice to refer here to the rather complex writings and the related connotations for the place name /Lagas/; see below Ill. 1. 16 I use “commentary” here in the very broad sense of elucidating the meaning of a given word and thus the specific culture in which it was used (also Selz in press). The term here does not refer to the more specific – and later − “commentaries on texts” so profoundly studied by Frahm 2011. However, my claim is that the writings discussed below display features of the same kind of “self-conscious reflection” as the later commentaries (Frahm 2011: 3 f.). Therefore these writings may be understood as “forerunners” of the Mesopotamian commentary tradition. 17 Question – answer reasoning is best known from the Sumerian Riddles (see Civil 1987). Concerning our discussion of DIRI writings the Question − Answer formula is, of course, reconstructed. 18 Note the divergence between the sign etymology and the etymology of uğnim: the first alludes to “camp”, the latter most probably to the soldiers.

Sign Formation as an Early Concept in “Commentaries”

305

b) Q: “What is /agrig/?” (= “a chief administrator”) or “What is izkim?” (= “(ominous) sign”) A: “(One with) the eye on the tablet” / “(What) the eye (sees) on a tablet.” EYE+TABLET

IGI.DUB

EYE+THROW(ING)

IGI.RU

agrig (aĝrig) ĝiskim ĝiškim (ĝiskim izkim giskim iskim giškim gizkim) c) Q: “What is /pad/?” (=“to find, discover; to name, nominate”) pad3 (pa3)

A: “To cast an eye toward / on something.” d) Q: “What is /tu/?” (=“to bathe” (in ED sources a composite verb a–tu5 “to bathe in water”; in later texts attested a.šu.naga = tu17) HAND+SOAP(WORTH)

ŠU.NAGA

tu5

A: “To apply soap with hands.” e) Q: “What is a /lukur/?” (= a type of priestess) WOMAN+(DIVINE) ORDINANCES f)

SAL.ME

lukur

A: “A woman concerned with (divine) ordinances.” Q: “What is a sipa(d)?” (= “shepherd”) OVERSEER+SHEEP/PASTURE

PA.UDU/ LU

sib2 sipad (sibad sipa)

AB2.KU/ TUŠ

udul unud (utul unu3 udul)

A: “An overseer of the sheep.” g) Q; “What is an /unud/?” (=“cowherd”19) COW/BOTTOM/SIT

A: “One who lets pasture / camp the cows.”

19 The original reading may have been /unu(d)/ and meant “(the one who) let rest (the animals in) the grass”, an interpretation suggested by numerous literary references; for the late ED period see Luzag. 1 2:17–28 and 3:22–23 (ABW = FAOS 5/1). Note that for utullu(m) ED sources have the syllabic rendering u2-du(-ul) (FAOS 6, 340). For the interpretation of the not entirely clear sign KU/TUŠ as “bottom” or “anus” cp. Sumerian Proverbs 2.100 (following ECTSL):165–166: gala-e bid3-da-ni ḫa-ba-an-da-ze2-er / ağ2 ga-ša-an-an-na ga-ša-an-ğu10 ba-ra-zi-zi-de3-en-e-še “The lamentation priest wipes his bottom: ‘One should not remove what belongs to my mistress Inana.’” Note that gala, a term for a cultic, especially mourning singer (a castrato?), is itself a DIRI-writing: NITA.KU/TUŠ [MALE]+[BOTTOM/SIT].

306

Gebhard J. Selz

h) Q: “What is /šeĝ/?” (wr. šeĝ3, also im.a.a: and im.a.an) (=„rain“) WATER+HEAVEN i)

A.AN

am3, šeĝ3

A: “Water from Heaven.” Q: “What is (kuš)/ummud/?” (= “waterskin”) (LEATHER+)WATER+STEPPE+HANGING

(kuš)

ummu3, ummud

A: “A hanging leather skin for water in the steppe.” The basic observation concerning these examples is, in principle, generally accepted. Sign combinations with one sign functioning as phonetic indicator are also well attested. However, things can be rather complicated, as our forthcoming article on “secondary iconization” (Selz and Pfitzner in press) will demonstrate. While the standard writing for the word /lirum/ “strength; quarrel, combat, fight; wrestling” and so on is the diri forms lirum HAND+STRONG, and lirum4 ARM+STRONG, the lexical tradition attests to signs indicating in various ways the notion of “duels.” Compare the writing , lirum6 , lirum7 , and lirum8 . lirum3 In conclusion, I have here attempted to show, in a preliminary and rather unsystematic way, that some of the semasiographic writings – sign combinations as well as combined signs – that is diri writings – contain implicit commentaries. That the reasoning of questions and answers was quite common among Sumerian intellectuals can be inferred from the so-called Pre-Sargonic Riddles from Lagas (BM 3, 26; Biggs 1973). Though not universally accepted,20 Biggs’ notion was that in the repeated sequence of these riddles “Its fish is XY, its snake is XY, its canal is XY, and its deity is XY” the deictic /bi/ refers to the underlying, but not explicitly expressed, question “What town is it: …”. The likewise unwritten answer then was the name of the specific town. Miguel Civil much improved not only our understanding of the Sumerian Riddles, but also that of the related Sumerian contest literature, the dialogues and diatribes. They all attest to vivid oral reasoning, especially in the e2-dubba-a. When, during the last decades, Sumerologists have become less blind, this was pre-eminently the result of Miguel Civil’s outstanding work.

20 Compare Marchesi 1999: 3; also Krecher 1992: 294.

***

“Buru-bird” – which is either a raven or a bird of prey (eagle?, Hence nu11.bur is classified as nu11.bur mušen vulture?) [all carnivorous!]; nu11 with the phonetic value /nu/ appears to function as a (pronounced) classifier(?); with mušen is added as additional classifier (pronounced?) classifier for purposes of clarification and disambiguation.

nu11.bur perhaps nu11bur = a bird containing the syllable /bur/ as in the attested /buru/

nu2mušen; also ğešnu2mušen Is the latter really indicative of the reading or rather motivated by the graphic ğeš ğeš relation to nu2 / na2, probably to be read /ĝešnu(d)/ “bed”?

nu11 = /sud/ also “a bird” (ED birds 124); cp. also nu11 mušen (ED birds 29); a possible reading /nu/(?) may be related to the bird nu2mušen (ED birds 40)? [ePSD lists /sud/ and /nu/ as separate lemmata!]

nu11:bur (šir (better nu11!).bur) “alabaster bowl” = “sacred storage room”(?) (or ancient writing for */na-bur/(?))

nu11 “various sorts of bright stones, including ‘alabaster’”; the reading /nu/ is uncertain when written with the classifier as na4nu11 ; in the light of later evidence sometimes transcribed as /ĝešnu/ (analysis (?))

bur-saĝ “main bowl; in metonymic use: “storage-room”

bur ; na4bur “(food) offering, sacrifice; meal(-time); (stone) bowl; a priest” = Akk. abru; naptanu; nīqu; pūru; perhaps pronounced originally as */nabur/ “stone bowl”

The place name Lagaš/s – a multi-layered form of writing

Sign Formation as an Early Concept in “Commentaries”

307

There is additional evidence corroborating this interpretation: 1. The place name Lagas appears in the early Fāra texts as nu11.bur.mušen.ki.la and bur.nu11.mušen.ki These slightly earlier writings clearly connect the place name with the /buru/-bird. 2. OB Diri Nippur Seg.9, 7 has [[lagaš/ski]] = = [šir.bur].la#-ki = na-kam-tum “storehouse,” providing evidence that the underlying intra-textual connection was still known at the beginning of the 2 nd millennium.

nu11.bur LA thus indicating that one needs to read /lagas/; for further clarification, one may even add the place “Lagasclassifier KI (as happens in texts after Ur-Nanše’s reign), resulting in nu11.bur.laki or nu11.bur.la.ki Place”

If one wants to write a toponym which is both connected to a “bird of prey” AND the salient function of the place as “storage house”(within the early household economy), you may add − for clarity − a new phonetic indicator and write:

308 Gebhard J. Selz

Sign Formation as an Early Concept in “Commentaries”

309

Bibliography Biggs, Robert D. 1973. Pre-Sargonic Riddles from Lagash. JNES 32: 26–33. Civil, Miguel. 1987. Sumerian Riddles: A Corpus. AuOr 5: 17–38. Civil, Miguel. 1995. Ancient Mesopotamian Lexicography. Pp. 2305–14 in Civilizations of the Ancient Near East vol. 4, ed. Jack M. Sasson. New York: Scribner. Civil, Miguel. 2004. The Series diri = (w)atru. MSL 15, Roma: Pontificium Institutum Biblicum. Civil, Miguel. 2008. The Early Dynastic Practical Vocabulary A (Archaic har-ra A). ARES IV. Roma: Missione archeologica italiana in Siria. Civil, Miguel. 2010: The Lexical Texts in the Schøyen Collection, CUSAS 12, Bethesda: CDL Press. ePSD: Electronic Pennsylvanian Sumerian Dictionary: http://psd.museum.upenn.edu/epsd/ nepsd-frame.html (last accessed June 2015) FAOS Freiburger Altorientalische Studien. Frahm, Eckart. 2011. Babylonian and Assyrian Text Commentaries. Origins of Interpretation. Guides to the Mesopotamian Textual Record 5. Münster: Ugarit Verlag. Green, Margaret W. and Hans J. Nissen. 1987. Zeichenliste der archaischen Texte aus Uruk. Archaische Texte aus Uruk 2. Berlin: Gebr. Mann Verlag. Hilgert, Markus. 2009. “Of ‘Listenwissenschaft’ and ‘Epistemic Things’. Conceptual Approaches to Ancient Mesopotamian Epistemic Practices. Journal for General Philosophy of Science 40 (2): 277–309. Johnson, J. Cale. 2013. Indexical iconicity in Sumerian belles lettres. Language & Communication 33/1: 26–49. Krecher, Joachim. 1992. ud.gal.nun versus ‘Normal’ Sumerian: Two Literatures or One? QdS 18: 295–303. Lambert, Wilfred George. 1990. The Names of Umma. JNES 49: 75–80. Maiocchi, Massimo. 2012. A Sign List and Syllabary for the Study of Third Millennium Cuneiform Inscriptions Based ePSD. Private ePublication. Marchesi, Gianni. 1999. Notes on Two Alleged Literary Texts from Al-Hiba/Lagaš. SEL 16: 3–17. Pfitzner, J. 2014. Die lexikalische Liste Diri − Watru und ihre Diri-Komposita. Masterarbeit, Universität Wien. Philologisch-Kulturwissenschaftliche Fakultät; Online Publication UB Wien e-theses (http://othes.univie.ac.at/34586/) Pfitzner, J. (forthcoming). How to Analyze Diri Compounds? Selz, G. J. 2002. ‘Babilismus’ und die Gottheit dnindagar. Pp. 647–84 in Ex Mesopotamia et Syria Lux. Festschrift für Manfried Dietrich, ed. Oswald Loretz AOAT 281. Münster: Ugarit Verlag. Selz, G. J. 2013. On Some Early Dynastic Sumerian Toponyms. Pp. 317–26. in Florilegium Aegyptiacum − Eine wissenschaftliche Blütenlese von Schülern und Freunden für Helmut Satzinger zum 75. Geburtstag am 21. Jänner 2013, ed. Julia Budka Göttinger Miszellen, Beihefte 14, Göttingen: Seminar für Ägyptologie und Koptologie der Georg-August-Univ. Selz, G. J. (in press). The World Behind the Words. Reflections on Concepts Crossing Linguistic Boundaries. Selz, Gebhard J. and J. Pfitzner. (in press). Whirl Winds and the “Bable of the Tongues.’ Remarks on Secondary Iconization. (Fs. O. Pedersén).

310

Gebhard J. Selz

Veldhuis, Niek. 2012. Cuneiform: Changes and Developments. Pp. 3–23 in The Shape of the Script. How and Why Writing Systems Change, ed. Stephen D. Houston. Santa Fe: School of Advanced Research Press. ZATU s. Green and Nissen 1987.

Marcel Sigrist, Uri Gabbay and Mark Avila

Cuneiform Tablets and Other Inscribed Objects from Collections in Jerusalem The following inscriptions on tablets and other objects come from various private collections in Jerusalem.1

I. Ur III 1 A tablet from Irisaĝrig listing arrears Provenance: Irisaĝrig. Date: ca. Šu-Sîn 4-Ibbi-Sîn 3. The tablet lists the arrears left from previous transactions (lá-ni), and the individuals to whom they are to be paid. According to the prosopography, the tablet should probably be dated to the end of the reign of Šu-Sîn or the beginning of the reign of Ibbi-Sîn (see details in the notes below). Obv. 1 2 3 4 5 6 7 8 9

lá-ni 3;1 gur nanna-ì-gi ˹kuš7˺ 5 gur d.aššir-gi4-al-˹su˺ 6;1 gur me?-á x 5;2 gur puzur4-ma-m[a] 2;4 gur tu-ra-a 2;1 gur ˹a?-pi˺-la-núm 2;3 gur ip-qú-ša 10;4 ˹gur˺ a-bu-ba-qar d

1 Some of the objects were brought to our attention by the Israel Antiquities Authority as part of its attempt to control the antiquities coming in and out the state of Israel. We thank JeanMichel de Tarragon for photographs of texts nos. 2–6. The photographs of no. 1 are by the owner of the tablet. Editorial note: The authors of this contribution did not follow the conventions employed to transliterate Sumerian in this volume. In order to avoid potential replacement errors, however, the editors decided to leave the Sumerian as submitted. https://doi.org/10.1515/9781501503696-019

312

Marcel Sigrist, Uri Gabbay and Mark Avila

Rev. 10 11

8;1 gur ba-ab-bu 11;1 gur šu-eš4-tár

12

˹57;3˺ ˹gur˺

13

lá- ni-àm

Notes 2. This individual, bearing the same title, is known from the years Šu-Sîn 4, 9, and Ibbi-Sîn 1 from Irisaĝrig; see references in Owen 2013, 1: 511. 3. This personal name is known from the years Amar-Sîn 7, Šu-Sîn 9, and IbbiSîn 3 from Irisaĝrig (the last two attestations probably correspond to the individual in our tablet); see references in Owen 2013, 1: 474. Note that the similar name Ašgi-tillat-su(2) is mentioned in texts dating to the same period, namely years Šu-Sîn 3 to Ibbi-Sîn 3 from Irisaĝrig; see references in Owen 2013, 1: 474. For a possible identification of these two names with each other, see Sigrist and Gabbay 2014: 297. 4. We are unable to read this name. Cf. the name sú-me-ed-dingir in Owen 2013, 2: 412 obv. 5 (Šu-Sîn 7), but this does not correspond to the traces following the sign á in our text, and in addition it seems that there is nothing between the sign gur and what seems like the sign me in our line. 5. This personal name is known from the years Amar-Sîn 8, 9, Šu-Sîn 4, 5, 9, and Ibbi-Sîn 2 from Irisaĝrig; see Owen 2013, 1: 518. 6. This personal name is known from the years Šu-Sîn 6 and Ibbi-Sîn 2 from Irisaĝrig; see, Owen 2013, 1: 534. 7. This personal name is known from the years Šu-Sîn 6 and Ibbi-Sîn 2 from Irisaĝrig, where it is written using the sign pi5 (ne) instead of pi in our line; see Owen 2013, 1: 470, s. v. a-pil-la-lum. This writing of the second element of the name, but probably with á as the first element instead of a in our line, i.e., ˹á˺-pi-la-núm, is attested also in Molina 2014: no. 337 obv. 9,2 where Adab is cautiously suggested as the provenance. Note that PuzurMama is mentioned in this text too (rev. 3), as in our line 5, so perhaps it also originates from Irisaĝrig. 8. This personal name is known from the years Amar-Sîn 8 to Ibbi-Sîn 3 from Irisaĝrig; see Owen 2013, 1: 498.

2 The original publication is not available to us, but a transliteration and photograph are available in BDTNS no. 157867 (http://bdtns.filol.csic.es/ficha_simple_ventana_lexema.php? miReferencia=0; accessed March, 2015).

Cuneiform Tablets and Other Inscribed Objects from Collections in Jerusalem

313

9. This personal name, written a-bu-um-ba-qar, is known from the years ŠuSîn 9 and Ibbi-Sîn 1 from Irisaĝrig; see Owen 2013, 1: 465. 10. This personal name is known from the years Ibbi-Sîn 1 and 2 from Irisaĝrig; see Owen 2013, 1: 476. 11. This personal name is known from the years Amar-Sîn 7, Šu-Sîn 2, 9, 12, and Ibbi-Sîn 1, 2 from Irisaĝrig; see Owen 2013, 1: 527–28.

2 A tablet listing rations to royal messengers from Irisaĝrig Provenance: Irisaĝrig. Date: Ibbi-Sîn 2, month 4. Obv. 1 2 3 4 5 6 7 8 9 10 11 12 13

5 sìla kaš 5 sìla ninda dingir-dan lú-kíĝ-gi4-a-lugal u4 ki-maški-ta ki lugal-šè ba-ĝen-na-a 5 sìla kaš 5 sìla ninda ḫu-la-al lú-kíĝ-gi4-a-lugal u4 kaskal anše-kúnga-zi-gu5-um-šè im-ĝen-na-a 2 sìla kaš 2 sìla ninda pu-lu-lu kuš7 u4 anše-kúnga-zi-gu5-um-šè im-ĝen-na-a 5 sìla kaš 5 sìla ninda šeš-kal-la àga-ús-lugal tu-ra u4 en-nun-ĝá še buru14-ka-šè im-ĝen-na-a / lú-sa-gaz-ke4 in-sìg-ga

Rev. 14 15 16 17 18 19 20 21 22

5 sìla kaš 5 sìla ninda a-ḫu-ba-qar lú-kíĝ-gi4-a-lugal u4 érin še ĝiš ra-ra zi-zi-dè / im-ĝen-na-a 3 sìla ˹kaš˺ 2 sìla ninda da[m]?-x-˹na?˺ lú-kíĝ-gi4-a-lugal / ur-niĝarxĝar ša13-dub-ba u4 ki énsi-ka-šè im-ĝen-na-a 2 sìla kaš 2 sìla ninda lugal-amar-kù lú-ùr-ra u4 mun-gazi-šè im-ĝen-na-a

314

Marcel Sigrist, Uri Gabbay and Mark Avila

23 24 Edge

zi-ga iti-ezem-dli9-si4 mu en dinanna unugki-ga / máš-e in-pà u4-10-kam

1–4

5 sila beer, five sila bread (for) Ilum-dan, the royal messenger, when he went from Kimaš to the king’s place. 5 sila beer, five sila bread (for) Ḫulal, the royal messenger, when he came to the sikum-mule road. 2 sila beer, two sila bread (for) Pululu the equerry, when he came for the sikum-mules. 5 sila beer, 5 sila bread (for) Šeškala, the royal gendarme (when) he was wounded, when he came for the guarding of the grain for the harvest (and) bandits struck him. 5 sila beer, 5 sila bread (for) Aḫu-waqar, the royal messenger, when he came to muster workers to thresh grain. 3 sila beer, 2 sila bread (for) Dam-..-na(?), the royal messenger, Urniĝar the archivist, when he came to the governor’s place. 2 sila beer, 2 sila bread (for) Lugalamarku the miller, when he came for spices.

5–7 8–10 11–13

14–16 17–19 20–22

23–24 Expenditures, month of the Festival of Lisi (= month 4), the year he chose the en-priestess of Inana in Uruk by means of extispicy. Edge Day 10. Notes For the subordinate temporal clauses in the tablet, see Owen 2013, 1: 165– 82. For most of the messengers mentioned in this tablet, see Owen 2013, 1: 155–63. 9. For the individual mentioned in this line, see references in Owen 2013, 1: 516. 12. For the individual mentioned in this line, designated as “wounded, sick,” in the context of bandits in the next line, see references in Owen 2013, 1: 366. 18. We are unable to read the name of the first individual in this line. For the second individual, see references in Owen 2013, 1: 540. 21. For the individual mentioned in this line, see references in Owen 2013, 1: 446.

Cuneiform Tablets and Other Inscribed Objects from Collections in Jerusalem

315

3 A tablet from Umma concerning the absence of a seal Provenance: Umma. Date: Ibbi-Sîn 3.3 Obv. 1 2 3 4

kišib lú-dnanna zà mu si-mu-ru-/umki ba-ḫul-šè šà-nin-ĝá kišib nu-un-da-ĝál-l[a]?

Rev. 5

mu lugal-bi [i]n-/pà

Seal [šà]-nin-ĝá [dub-sar] [dumu] lugal-[ùšur] The seal of Lu-Nanna (is used) until the “edge” of the year “Simurrum was destroyed.” That Šaninĝa will not have a seal with him he swore by the name of the king. Seal impression: [Ša]ninĝa, [the scribe, son of] Lugal-[ušur]. Notes The tablet, bearing the seal impression of Šaninĝa, documents that Šaninĝa swore, using the promissory formula for future events (Sallaberger 2009), that he will not be sealing tablets until the end of the year,4 and that Lu-Nanna will be sealing tablets for him.5

3 A transliteration of the obverse of the tablet can be found in BDTNS no. 193794 (http:// bdtns.filol.csic.es/ficha_simple_ventana_lexema.php?miReferencia=0; accessed March, 2015) according to a photograph in an auction publication. 4 We understand zà here as “edge, end,” independently from mu, i.e., not as zà-mu; cf. zà iti, “the ‘edge’ of the month” (see MVN 13: 311 obv. 3; MVN 21: no. 208 rev. 1; Yildiz and Ozaki 2000: no. 3232 rev. 4). Nevertheless, one cannot exclude the possibility that this refers to the zà-mu “high season” of the year (not necessarily the New Year); see Sallaberger 1993, 1: 142– 43 with n. 669. 5 Because the seal impression is present, it does not seem to be the case here that the oath is used instead of a seal impression, as is documented, e.g., in Owen 2013, 2: 156 rev. 2: mu kišib nu-tuku-šè mu lugal pà “because he did not have a seal, he swore by the name of the king.”

316

Marcel Sigrist, Uri Gabbay and Mark Avila

The two individuals mentioned in the tablet, Šaninĝa and Lu-Nanna, are known from tablets dating mainly to the reign of Šu-Sîn, but also to the end of the reign of Amar-Sîn, and the beginning of the reign of Ibbi-Sîn.6 These two individuals are mentioned together also in other tablets, of which one is dated to the year Šu-Sîn 4 and one to the year Ibbi-Sîn 3 (the date formula is not preserved in the other two).7 Tablets mentioning the seal of Šaninĝa are known from the year Amar-Sîn 9 to the year Ibbi-Sîn 2, only a few in the first and last years and a wealth of such documents dating to the year Šu-Sîn 7. Impressions of Šaninĝa’s seal are similarly found on tablets dating from the year Amar-Sîn 6 to the year Ibbi-Sîn 2, mostly corresponding to the tablets that mention his seal, and again only in scattered examples in the beginning and end of this period, the majority of tablets dated to the year Šu-Sîn 7.8 The impressions of Šaninĝa’s seal found on other tablets, containing an introductory scene, seem to be those of the seal found on our tablet.9 On the other hand, there are a few tablets mentioning the seal of Lu-Nanna from the second half of the year Ibbi-Sîn 3.10 Although this may be a chance find, this fits the declaration on our seal that Šaninĝa did not have a seal in the year Ibbi-Sîn 3.11

6 The tablets dealt with here and in the following paragraphs will not be all listed here, and can easily be consulted through a search in BDTNS (http://bdtns.filol.csic.es/). 7 Nisaba 23: 50 (no date; same Lu-Nanna?); Nisaba 24: 9 (no date; obv. ii 32′: kišib lú-dnanna; obv. ix 29′: kišib šà-n[in-ĝá]); Sauren 1977: 24 no. 21 ovb. i 13, 15 (Šu-Sîn 4); MVN 3: 312 obv. 5, rev. 1 (Ibbi-Sîn 3, month 7; kišib lú-dnanna but seal impression of lú-dšára; see below). 8 The latest tablet is Yildiz and Ozaki 2001: no. 3825, containing the notation kišib šà-nin-ĝá and his seal accordingly. 9 For Šaninĝa’s seal impression, see Mayr 1997, no. 599. 10 MVN 3: 312 (month 7), MVN 5: 87 (month 9), BPOA 7: 2147 (month 11); cf. also Pettinato 1997: no. 47 rev.16. 11 It is interesting that Lu-Nanna himself uses his own seal until Ibbi-Sîn 1 (lú-dnanna dubsar dumu la-ni-mu; see AUCT 3: 358, dated Ibbi-Sîn 1, month 8), but in Ibbi-Sîn 3 he uses his brother’s(?) seal (lú-dšára dub-sar dumu la-ni-mu), even though the notation kišib lúd nanna appears in the tablet; see MVN 3: 312 (mentioning Šaninĝa as well; month 7); MVN 5: 87 (month 9); BPOA 7: 2147 (month 11); Owen and Wasilewska 2006: 264, 286, no.15. Note that he uses his brother’s seal also in Šu-Sîn 7: Sigrist 2000: no. 1821 (kišib aš lú-dnanna).

Cuneiform Tablets and Other Inscribed Objects from Collections in Jerusalem

317

II Old Babylonian 4 An Eye-stone of Hammurapi, probably from Sippar (Ebabbar) Provenance: Sippar. Date: reign of Hammurapi. A chalcedony eye stone (diameter: 2.45 cm),12 flat in its white bottom and curved on its dark brown top, with a round hole drilled on its iris, contains a dedicatory inscription of Hammurapi to the goddess Aya, written in five lines, three above the hole, and two below it. 1 2 3 4 5

d

1 2 3 4 5

O Aya, Lady of the Ebabbar! Hammurapi, Your favorite prince, Keep alive!

a-a nin é-babbar-ra ḫa-am-mu-ra-pí nun še!-ga-zu(text: ba) ti-la-ì

Notes A very similar inscription dedicated to Utu is found in a different eye stone in the British Museum (BM 130829).13 This inscription is carved on a brown-white stone (diameter: 2.9 cm), containing no hole, and written on the flat side and not on the curved side. The inscription reads (RIME 4.3. 6. 2004): dutu / en gal an ki-ke4 / ḫa-am-mu-ra-pí nun še-ga-zu ˹ti˺-la-ì “Utu, great lord of heaven and earth! Hammurapi, your favorite prince, keep alive!”

12 For inscribed and uninscribed eye-stones, see Lambert 1969, and a detailed study by Clayden 2009. 13 For information, copy and photograph, see Clayden 2009: 56, no. 6, and pls. 1b, 6a-b. See also http://www.britishmuseum.org/research/collection_online/collection_object_details. aspx?objectId=1426782&partId=1&searchText=130829&page=1; accessed March, 2015).

318

Marcel Sigrist, Uri Gabbay and Mark Avila

5 A tablet regarding supplies for the temples of Nippur, probably from Dūr-Abiešuḫ Provenance: Dūr-Abiešuḫ(?). Date: Samsuditana 15, month 9. The following tablet can be assigned to the archives of Dūr-Abiešuḫ according to its content and according to some of the names mentioned in it. The tablet lists supplies for the Ekur temple of Enlil, the Ekiur temple of Ninlil, and the Ešumeša temple of Ninurta,14 as well as to the temple of Sîn, and to various officials.15 Obv. 1 2 3 4 5 6 7 8

2 še-gur é-kur ĝìr nu-èš-meš 1 gur é-ki-ùr ĝìr gudu4!-meš dnin-líl 1 gur é-šu-me-ša4 ĝìr gudu4!-meš dnin-urta 4 še gur ša a-na šuku é den.zu ˹m?˺ib-ni-dim ugula mar-tu érin ḫa-la-baki [i]k?-pu-tu

Edge 9 10

4 še-gur i-na qá-ti ti-ik-la-a-na-da[mar.utu(?)] / šà-tam érin lugal

Rev. 11 12 13

ša!? be-lí-a-ša-re-ed šà-tam érin lugal ˹m?˺na-bi-den.zu ša13-dub-ba il-qú-ú

14 Van Lerberghe and Voet (2009: 1) regard the mention of Ekur in this archive as a “new Ekur” in Dūr-Abiešuḫ. However, one cannot exclude the hypothesis that although the administrative system of the temples was conducted from Dūr-Abiešuḫ, the cult in the temples continued in Nippur itself, even though the city was mostly abandoned by then; cf. George 2009: 137–39. 15 For the economic and cultic activities at the temples taking place in Dūr-Abiešuḫ, see Van Lerberghe and Voet 2009: 2–4. For texts from this archive listing sheep offerings, see Van Lerberghe and Voet 2009: 23–38 (dated from years Amiṣaduqa 8 to Samsuditana 5). A similar text to ours is Van Lerberghe and Voet 2009: 54 (Amiṣaduqa 2), listing barley for the rations of Enlil, Ninlil, Ninurta and Nuska, received by nu-èš-priests and gudu4-priests.

Cuneiform Tablets and Other Inscribed Objects from Collections in Jerusalem

14 15 16 17 18

nam-ḫar-ti šà-tam érin lugal-lá ša13-dub-ba den-líl-lá ša13-dub-ba-meš šà-tam-gal [lugal?-l]a šà-tam é den.zu? iti-gan-gan-è u4-21-kam

Edge 19 20 21

mu sa-am-su-di-ta-na lugal-e alan igi ka-kešda ugnim ma-da

319

Seal 1 [qí]?-iš?-ti-dingir(–)[..] [dum]u? lu-ša-lim-be-[lí]? [ìr?] x an? […] Seal 2 d

nin-[…] ša den.zu? […] an kal x […] Seal 3 an pa […] ka tu? […] a? […] 1–4

2 kor of barley (for) the Ekur, supervised by the nešakku-priests; 1 kor (for) the Ekiur, supervised by the pašīšu-priests of Ninlil; 1 kor (for) the Ešumeša, supervised by the pašīšu-priests of Ninurta, 5–8 4 kor of barley that Ibni-Adad, the general of the troops of Ḫalaba took care of(?) for the rations of the House of Sîn. 9–13 4 kor of barley in the posession of Tikla-ana-[Marduk(?)], the šatammu of the troops of the king, that Bēlī-ašarēd, the administrator of the troops of the king, (and) Nabi-Sîn, the archivist, took. 14–17 Received by the administrator of the troops of the king, the archivist of Enlil, the archivists, Ibni-E[nli]l(?), the administrator of the House of Sin. 18–21 Month: 9. Year: Samsuditana, the king, (his) statue in front of the troops of the army.

320

Marcel Sigrist, Uri Gabbay and Mark Avila

Notes 7. The name Ibni-Adad occurs also in Van Lerberghe and Voet 2009: 267 no. 19 seal B’, dated to Ammiṣaduqa 17+, but it is doubtful whether this is the same person. According to Kathleen Abraham (personal communication), the “troops of Ḫalaba” (érin ḫa-la-baki) are also mentioned in a few unpublished texts from Dūr Abiešuh. 8. The restoration of the verb is based on the preserved end of the first sign and the small broken space left before it that would seem to correspond to the sign ig. The problem is that the form of this verb ending with t is not otherwise found in Old Babylonian texts, but only in Middle Assyrian texts (CAD K, 172–74). Nevertheless, the meaning of the verb (“to take care of,” especially in Old Babylonian sources; CAD K, 173–74) seems to fit the context of line. 10. For Tikla-ana-Marduk, see George 2009: 17, 75–76, dated Samsuditana 5 (there a “barber” šu-i, and perhaps not identical with the individual in our tablet; note especially the character of that text, see George 2009: 146–48). 11. For Bēlī-ašarēd šà-tam érin lugal-la, see Van Lerberghe and Voet 2009: 53, 15, rev. 4 (Samsuditana 13). 19–21. For the year formula of Samsuditana 15, see Pientka 1998: 139 (cf. Richardson 2010: nos. 35–36). Seal 1. For the name Lū-šalim-belī, see CAD Š/I, 257a.

III. Middle Babylonian 6 A brick bearing a royal inscription of Karaindaš Provenance: Uruk. Date: reign of Kara’indaš. The following inscription cut out of a brick is known from other exemplars excavated in Uruk.16

16 See Brinkman 1976: 170–71, N.2.2; Walker 1981: no. 58. For a discussion of this inscription, see Bartelmus 2010: 153 with n. 48, 160 n. 101, 163 with n. 115. The current brick inscription was previously in the collection of A. Dupont-Sommer, and was donated to the Ecole biblique by his family.

Cuneiform Tablets and Other Inscribed Objects from Collections in Jerusalem

321

1 [k]a-ra-in-da-a[š] 2 lugal kala-ga 3 lugal ká.dingir.raki 4 lugal ki-en-gi˹ki˺ 5 uriki-uriki-bi 6 lugal ˹ka-ru˺-du-ni-ia-aš 7 sipa še-ga-ni ˹é˺-an-na 8 ˹é˺ ki-áĝ-ĝá-ni 9 ˹mu-un˺-gibil-ba 10 Karaindaš, mighty king, king of Babylon, king of Sumer and Akkad, king of Babylonia, her (= Inana’s) favorable shepherd, renewed the Eana, her beloved house.

322

Marcel Sigrist, Uri Gabbay and Mark Avila

Photographs

Fig. 1: No. 1 obv.

Cuneiform Tablets and Other Inscribed Objects from Collections in Jerusalem

Fig. 2: No. 1 rev.

323

324

Marcel Sigrist, Uri Gabbay and Mark Avila

Fig. 3: No. 2 obv.

Cuneiform Tablets and Other Inscribed Objects from Collections in Jerusalem

Fig. 4: No. 2 rev.

Fig. 5: No. 2 edge.

325

326

Marcel Sigrist, Uri Gabbay and Mark Avila

Fig. 6: No. 3 obv.

Fig. 7: No. 3 rev.

Cuneiform Tablets and Other Inscribed Objects from Collections in Jerusalem

Fig. 8: No. 4(a).

Fig. 9: No. 4(b).

327

328

Marcel Sigrist, Uri Gabbay and Mark Avila

Fig. 10: No. 4(c).

Cuneiform Tablets and Other Inscribed Objects from Collections in Jerusalem

Fig. 11: No. 5 obv.

Fig. 12: No. 5 rev.

329

330

Marcel Sigrist, Uri Gabbay and Mark Avila

Fig. 13: No. 5 right edge.

Cuneiform Tablets and Other Inscribed Objects from Collections in Jerusalem

Fig. 14: No. 5 left edge.

331

332

Marcel Sigrist, Uri Gabbay and Mark Avila

Fig. 15: No. 5 bottom edge.

Fig. 16: No. 5 top edge.

Cuneiform Tablets and Other Inscribed Objects from Collections in Jerusalem

Fig. 17: No. 5 right edge.

333

334

Marcel Sigrist, Uri Gabbay and Mark Avila

Fig. 18: No. 6.

Cuneiform Tablets and Other Inscribed Objects from Collections in Jerusalem

335

Bibliography Al-Rawi, Farouk N. H. and Lorenzo Verderame. 2009. Neo-Sumerian Administrative Texts from Umma Kept in the British Museum, Part Three (NATU III). Nisaba 23. Messina: Di.Sc.A.M. Al-Rawi, Farouk N. H., Franco D’Agostino, and Jon Taylor. 2009. Neo-Sumerian Administrative Texts from Umma Kept in the British Museum, Part Four (NATU IV). Nisaba 24. Messina: Di.Sc.A.M. Bartelmus, Alexa. 2010. Restoring the Past: A Historical Analysis of the Royal Temple Building Inscriptions from the Kassite Period. Kaskal 7: 143–71. Brinkman, John A. 1976: Materials and Studies for Kassite History, vol. 1. Chicago: The Oriental Institute. Clayden, Tim. 2009. Eye-Stones. Zeitschrift für Orient-Archäologie 2: 36–86 George, Andrew R. 2009: Babylonian Literary Texts in the Schøyen Collection. CUSAS 10, Bethesda, MD: CDL Press. Koslova, Natalia. 2000. Neusumerische Verwaltungstexte aus Umma aus der Sammlung der Ermitage zu St. Petersburg, Rußland. Materiali per il Vocabolario Neosumerico 21. Rome: Multigafica Editrice. Lambert, W. G. 1969. An Eye-Stone of Esarhaddon’s Queen and Other Similar Gems. RA 63: 65–71. Mayr, Rudolf H. 1997. The Seal Impressions of Ur III Umma. Ph.D. diss., Leiden University. Molina, Manuel. 2014. Sargonic Cuneiform Tablets in the Real Academia de la Historia: The Carl L. Lippmann Collection. Madrid: Real Academia de la Historia. Owen, David I. 1975. The John Frederick Lewis Collection. Materiali per il Vocabolario Neosumerico 3, Rome: Multigafica Editrice. Owen, David I. 2013. Cuneiform Texts Primarily from Iri-Saĝ rig/Āl-Šarrākī and the History of the Ur III Period, Nisaba 15/1–2. Bethesda, MD: CDL Press. Owen, David I. and Ewa Wasilewska. 2006. Cuneiform Texts in Utah Collections. Pp. 259–96 in If a Man Builds a Joyful House: Assyriological Studies in Honor of Erle Verdun Leichty, ed. Ann K. Guinan Cuneiform Monographs 31. Leiden: Brill. Pettinato, Giovanni. 1997. L’uomo cominciò a scrivere. Iscrizioni cuneiformi della Collezione Michail, Milan: Electa. Pientka, Rosel. 1998. Die Spätaltbabylonische Zeit: Abiešuḫ bis Samsuditana: Quellen, Jahresdaten, Geschichte. Imgula 2/1. Münster: Rhema-Verlag. Richardson, Seth F. C. 2010. Texts from the Late Old Babylonian Period. JCS Supplemental Series 2. Boston: ASOR. Sallaberger, Walther. 2008. Der Eid im Gerichtsverfahren im neusumerischen Umma. Pp. 159–76 in On the Third Dynasty of Ur: Studies in Honor of Marcel Sigrist, ed. Piotr Michalowski. JCS Supplemental Series 1. Boston: ASOR. Sallaberger, Walther. 1993. Der Kultische Kalender der Ur III-Zeit. Untersuchungen zur Assyriologie und Vorderasiatischen Archäologie 7/1–2. Berlin and New York: De Gruyter. Sauren, Herbert. 1977. Keilschrifturkunden in den Sammlungen zweier deutschen Museen. Orientalia Lovaniensia Periodica 8: 5–31. Sigrist, Marcel. 2000. Texts from the Yale Babylonian Collection II. Sumerian Archival Texts 3. Bethesda, MD: CDL Press. Sigrist, Marcel, and Uri Gabbay. 2014. Eighteen Cuneiform Inscriptions from the Ur III and Old-Babylonian Periods. Pp. 283–315 in Studies in Economic and Social History of the Ancient Near East in Memory of Péter Vargyas, ed. Zoltán. Csabai. Ancient Near Eastern

336

Marcel Sigrist, Uri Gabbay and Mark Avila

and Mediterranean Studies 2. Budapest: Department of Ancient History, University of Pécs, L’Harmattan. Sigrist, Marcel, Carney. E. S. Gavin, Diana Stein, and Constance Menard. 1988. Neo-Sumerian Account Texts in the Horn Archaeological Museum. Andrews University Cuneiform Texts 3. Berrien Springs, MI: Andrews University Press. Sigrist, Marcel, David I. Owen, and Gordon D. Young. 1984. The John Frederick Lewis Collection, Part II. Materiali per il Vocabolario Neosumerico 13. Rome: Multigafica Editrice. Sigrist, Marcel and Tohru Ozaki. 2009. Neo-Sumerian Administrative Tablets from the Yale Babylonian Collection, Part Two. BPOA 7. Madrid: CSIC. Sollberger, Edmond. 1978. The Pinches Manuscript. Materiali per il Vocabolario Neosumerico 5. Rome: Multigafica Editrice. Van Lerberghe, Karel and Gabriella Voet. 2009. A Late Old Babylonian Temple Archive from Dūr-Abiešuḫ. CUSAS 8. Bethesda, MD: CDL Press. Walker, Christopher B. F. 1981. Cuneiform Brick Inscriptions in the British Museum, the Ashmoleon Museum, Oxford, the City Museum of Birmingham Museums and Art Gallery, the City of Bristol Museum and Art Gallery. London: British Museum. Yildiz, Fatma and Tohru Ozaki. 2000. Die Umma-Texte aus den Archäologischen Museen zu Istanbul, Band V (Nr. 3001–3500). Bethesda, MD: CDL Press. Yildiz, Fatma and Tohru Ozaki. 2001. Die Umma-Texte aus den Archäologischen Museen zu Istanbul, Band VI (Nr. 3501–3834). Bethesda, MD: CDL Press.

Claudia E. Suter

On Images, Visibility, and Agency of Early Mesopotamian Royal Women 1 Different Interpretations Early Mesopotamian elite women are well attested in the visual record. From the beginning of urban life in the late 4th millennium BCE and throughout the 3rd millennium they are represented in statuary, dedicatory reliefs, luxury goods, and glyptic. By contrast, hardly any such images from the 2nd millennium BCE have come down to us. Does this change reflect a decline in women’s social standing, as some scholars argue, or were there other reasons for their earlier depiction? How visible were their images really and what did they represent? In 3rd millennium Mesopotamia, elite women were not only carved in stone; they actively participated in cult festivals, owned estates in their name, headed economic enterprises, and corresponded and exchanged gifts with elite women from other states.1 Some scholars take this as evidence for women’s independence and agency at this time. Aage Westenholz (1999: 70–71), for one, contends that Old Akkadian women – even if only few of them – were “emancipated:” they “did not have to lead secluded lives. From the queen down to ordinary citizenry, women, whether married or unmarried, Sumerian or Akkadian, were free to participate in public life on a par with men and mingling freely with men.” Rita Wright (2008: 272) holds that the organization of the state in the Ur III period “was the product of a strategy implemented by men and women in which women at the highest levels of society were on a par with their male counterparts.” Other scholars are more cautious. After stating that women played a very active role in the economic life of Sumer, Marc Van De Mieroop (1989: 66) long ago warned: “We do not know how far these indications about women in the economy reflect the social attitudes toward them. The economic system in which the women function may be entirely controlled by the men, and the

1 See Van De Mieroop 1989; Weihershäuser 2008; Marchesi 2011. Acknowledgements: I am grateful to Eva von Dassow and Ligia Ravé for their perceptive reading of an earlier manuscript, to Gary Beckman for sharing a manuscript of his forthcoming contribution with me, to Karen Wilson for identifying the museum number of no. 13 in table 1, and to George Contomichalos for improving my English. https://doi.org/10.1515/9781501503696-020

338

Claudia E. Suter

women mentioned in administrative positions in this study, may have been very unusual. It is clear that the major economic powers lay in the hands of men, and that only a small percentage of the texts preserved reflect the women’s role.” In the same vein, Tonia Sharlach (2007) questions whether the archive of Šulgi-simti was a woman’s archive when men wrote the tablets and men ran the livestock foundation. Gender specialist Julia Asher-Greve (2012: 374), who takes a more positive outlook, claims that conditions for women declined with the onset of Amorite rule in Mesopotamia in the early 2nd millennium. As evidence for her allegation, she primarily enlists elite women’s visibility in images: “Although we cannot trace this development, apart from laws restricting women’s agency, there is other evidence of change, in particular much reduced visibility of royal wives, termination of the office of en-priestess, women including princesses living in cloisters (nadītu), and the disappearance of women from seal imagery with the exception of a generally small-sized nude female figure. Images of women are also signs of their agency and potential models for identification; with the disappearance of women in imagery, they became symbolically invisible.” Four years earlier, Frauke Weiershäuser (2008: 278) drew a different picture: “In Mesopotamien ist auch in den auf die Ur III-Zeit folgenden Epochen der Isin-Larsa- und der altbabylonischen Zeit zu beobachten, daß die königlichen Frauen insbesondere im wirtschaftlichen Bereich ihre starke Position halten konnten. Bekannt ist die leitende Funktion, welche die Königin Šîbtu in der Administration des Palastes von Mari einnahm. Ob diese Frauen jedoch auch im kultischen Bereich eine Stellung bekleiden konnten, die jener der Königinnen des ausgehenden dritten Jahrtausends vergleichbar ist, oder ob sich das Konzept des Königtums in der ersten Hälfte des zweiten Jahrtausends soweit gewandelt hatte, daß die Königinnen nicht mehr als integraler Bestandteil des Herrscherpaares, insbesondere auch bei der Vertretung des Landes vor den Göttern, angesehen wurden, sollte Gegenstand weiterer Untersuchungen sein.” When taking into consideration the nature of 3rd millennium images of elite women, where they were set up, and what ultimate purpose they had, it is unlikely that they were “signs of their agency and potential models for identification,” as assumed by Asher-Greve (2012: 374). As Weiershäuser observes, royal women continued to be active in the 2nd millennium.2 Changes in that millennium seem to have affected less their agency than their role in the representation of the crown.

2 See, for example, Colbow 2002; Batto 1974; Dalley 1984: 97–111.

On Images, Visibility, and Agency of Early Mesopotamian Royal Women

339

In his significant contribution “Les limitations de l’information textuelle,” Miquel has demonstrated that ancient texts require not only decipherment and translation, but also a careful analysis and contextualization before they can be used toward the reconstruction of that society (Civil 1980). The same applies to images. “The idea that representations of natural things convey information as simply and directly and truthfully as nature itself is very seductive” (Molyneaux 1997: 2). Images, however, never depict real things. Ceci n’est pas une pipe. The historian Peter Burke (2001: 183) observes that “images are neither a reflection of social reality nor a system of signs without relation to social reality, but occupy a variety of positions in between these extremes. They testify to the stereotyped yet gradually changing ways in which individuals or groups view the social world, including the world of their imagination.”3 To give one instructive comparative example for the topic in question: Roman empresses appear as personification of Roman virtues, as authoritarian mothers of Roman legions or the patria on coins, but they can certainly never compete with male roles (Scheer 2006). Their images reflect the Roman discourse on gender and the ideology of the male establishment. Burke acknowledges the value of the visual record as supplementing as well as supporting the evidence of written documents. He emphasizes that the testimony of images needs to be placed in a number of contexts, that series of images offer more reliable testimony than single images, and that the historian needs to read between the lines. In the following, I will scrutinize the ostensible disappearance of elite women from 2nd millennium images and query their agency in early Mesopotamian society before discussing the extant relief sculpture that features them in narrative contexts. I warmly dedicate my musings to Miquel, who loves women without fear, in the hope that he will find some pleasure in them.

2 Images of Historical Individuals in the 2nd millennium BCE Before drawing far-reaching conclusions about a segment of society, it is prudent to take the larger picture into consideration. In fact, it is not only women that disappear from images, but also men, with the exception of the king. The surviving sculptural record of the 2nd millennium is extremely poor. In part

3 For a theoretical discussion relating to Mesopotamia, see Bahrani 2001.

340

Claudia E. Suter

this is due to increased use of reusable or perishable materials (Orthmann 1975: 288). Marie-Thérèse Barrelet’s (1974: 30) comparison of actual royal images with textually attested ones illustrates especially well the vagaries of discovery for the Old Babylonian period. An additional problem resides in the dating of insufficiently identified sculpture of late Ur III to Old Babylonian times, since the royal figure hardly changed from the later reign of Šulgi to Hammurabi.4 The second half of the 2nd millennium is even darker. The Kassite Dynasty has left us practically no images representing historical individuals. Exceptions are fragments of a monumental statue of Kurigalzu I (Spycket 1981: 294–296; Veldhuis 2008); a few hereditary entitlements carved on stone boulders that depict the involved parties rather than the typical host of deities represented in symbolic form (Slanski 2003); and fragmentary wall paintings of Mardukappla-iddina I at Dur-Kurigalzu apparently depicting rows of officials (Nunn 1988: 98–101). In view of this dearth, it is misleading to declare one-sidedly the disappearance of women in images. In fact, some are extant as, for example, a statue of Enanatuma, daughter of Išme-Dagan of Isin (Suter 2007: 355 fig. 9), or the relief of Hunnubat-Nanaya, daughter of Melišipak of Babylon (Orthmann 1975: pl. 191). More high priestesses of the early 2nd millennium may hide in anonymous statuettes often interpreted as goddesses (Suter 2007: 333–38). By comparison, there is only one identified image extant of an elite man: the bronze statuette that Lu-Nanna dedicated to Martu for Hammurabi’s life if indeed it represented the dedicator (Braun-Holzinger 1991: no. St. 172). Traditional dedicatory objects continue into the Isin-Larsa period, some lingering into the Old Babylonian period, and women still appear as donors.5 While there is some continuity form the 3rd to the 2nd millennium with regard to pictorial conventions and dedicatory gifts, a major difference and perhaps another reason for the scarcity of images of mortals, is the tendency toward increasing religiosity, what Thorkild Jacobsen (1976: 145–64) interpreted as the rise of personal religion. In 2nd millennium glyptic, deities and supernatural beings predominate (Boehmer in Orthmann 1975: 336–44). Some of these supernatural figures, such as the “figure with mace,” are actually 3rd millennium royal figures transformed into protective spirits (Suter 2010: 342– 46). Another protective spirit is the nude female figure (Wiggermann 1998),

4 The tendency to attribute such sculpture to the Isin-Larsa and Old Babylonian periods rather than to the late Ur III period, no doubt an attempt to fill the gap, is not warranted (Suter 2010). 5 Braun-Holzinger 1991: nos. G 423, S 12, P 25. Since Sumerian names rarely indicate gender, more women may hide in dedications for the life of a king (Braun-Holzinger 1991: nos. G 419, 421, 422, 424–26, S 16), especially of vessels, a typically female gift (Braun-Holzinger 1991: 96).

On Images, Visibility, and Agency of Early Mesopotamian Royal Women

341

which Asher-Greve inaccurately mentions in the context of mortal women. In parallel to the predominance of divinities in imagery, many Old Babylonian seal inscriptions identify the seal owner in terms of his/her personal or family deity rather than his profession or rank in state administration, or simply invoke deities without naming the seal owner (Braun-Holzinger 1996: 263, with further literature), while Kassite seal inscriptions can consist of entire prayers leaving no or very little room for visual imagery (Limet 1971). Dietz Edzard (2004: 615) explained Old Babylonian religiosity with both the disappearance of divine kingship and the reversion of the political landscape into a series of small kingdoms. Divine kingship disappeared in favor of a new image of kingship that Nicole Brisch (2007, 2011) sees transpiring in the hymns praising Larsa and Old Babylonian kings: they abandoned several themes typical of the hymns praising Ur III and Isin kings, including the “sacred marriage.” Kathryn Slanski (2003) interprets the fact that the mostly non-royal recipients of Kassite entitlement boulders appropriated a traditionally royal form of monument as a weakening of monarchic power at this time. Along these lines, the scarcity of images of mortal men and women may not exclusively be due to the vagaries of discovery, but also to prevailing ideologies and weak monarchies. Mesopotamian sculptural images were issued by the ruling establishment and conveyed ideology of the power.

3 Women’s Agency in the 3rd Millennium BCE Throughout its ancient history Mesopotamia was a monarchic society where kings were men. It was patriarchal in the sense of a system in which men held power and women were largely excluded from it, at least officially.6 Strongcharacter elite women could certainly become influential on the highest level, but their official role in politics remained symbolic (Melville 2005: 228). Stereotypical gender roles are already attested in a mid 3rd millennium birth incantation: the great midwife brings the girl to the world with a spindle and needle and the boy with two types of weapons (Krebernik 1984: 36–47). These symbols of femininity and masculinity related women to the domestic textile industry and men to military or hunting expeditions. Their recurrence in Ur III and Isin

6 Asher-Greve (1997: 232) contests the appropriateness of the term patriarchy for early Mesopotamia, holding that fatherhood was not fundamental to the definition of gender. However, individuals largely identify themselves by patronyms, and stereotypical gender roles did exist.

342

Claudia E. Suter

royal hymns confirms their validity in traditional early Mesopotamian royal ideology.7 In inscriptions on dedicatory gifts and seals, women are largely identified through their husband or father. Occasionally they are given a title and/or profession. But unlike kings, neither queens nor high priestesses are given epithets, nor are they praised in hymns. I am aware of only three 3rd millennium royal women mentioned by name in Sumerian literature. A Love Song for ŠuSuen names his mother and wife, both of whom are relegated to female gender roles: they are likened to different parts of the loom (Jacobsen 1987: 95–96). By contrast, the high priestess Enheduana appears as 1st-person narrator and compiler of hymns. While many scholars celebrated and continue to celebrate her as the first author, several recent voices led by Civil (1980: 229) are skeptical.8 I would not categorically exclude women poets or female voices. Andromache Karanika (2014) demonstrates how songs that women sang at work inspired ancient Greek poetry, and Jerry Cooper (1997) argues that the Dumuzi-Inana Songs betray a female approach to sexuality. Yet, Ninmešarra, the only hymn that includes “autobiographical” references, reflects ideology of the male regime (Glassner 2009). Following Niek Veldhuis’ lead, Alhena Gadotti (2011: 196–99) understands Enheduana’s appearance in literature, like the similar case of Sin-Kašid’s daughter Ninšatapada, as part of the common Sumerian heritage that was transmitted in Old Babylonian scribal schools. With regard to women’s role in the economy, it must be recalled that agriculture was the prime economic resource in ancient Mesopotamia. The rationale behind estates nominally headed by royal women and princes alike, was doubtlessly the expansion of the crown’s control over the economy. If women took charge of business, especially during absences of their husbands due to military or mercantile ventures, they acted on behalf of their husbands (Melville 2005). The marriage policies of the Akkad and Ur III dynasties underscore the subservient role of royal women. Princesses were married off to high functionaries or rulers of peripheral regions in the service of the establishment. Name changes of some princesses betrothed to foreign rulers turned these women into signs of power and prestige (Michalowski 2010). Other royal daughters were installed as high priestesses and symbolically married to major gods of the realm. This served both practical and ideological ends: to bring the respective temple estates under the crown’s control, and to link the king directly to the divine world by making him the father-in-law of the respective gods. The

7 Ur-Namma A: 88–131; Šu-Suen A: 5–6; Išme-Dagan K: 24. 8 For example, Glassner 2009; Rubio 2009: 27–28; Lion 2011: 96–97.

On Images, Visibility, and Agency of Early Mesopotamian Royal Women

343

commemoration in year names of both marriages to foreign rulers and elections and installations of high priestesses emphasizes the political and ideological importance of these events for the king. Another arena in which royal women represented the crown was state festivals. Weiershäuser (2008: 276) observes that queens’ participation in the official state cult made them an important part of kingship: together with their husbands they provided for the gods, which, in turn, guaranteed the wellbeing of the country. Although there is only indirect evidence, it seems that the king and his wife were the protagonists in annual festivals that celebrated divine weddings (Sallaberger 1999: 155–56). A similar role fell to high priestesses whose installation in office was sumptuously celebrated along the lines of a wedding to the god (Sallabeger & Huber Vuillet 2005: 622–23). Piotr Michalowski (2013: 175) suggests that performance arts had much more impact in the propagation of power than the stone images that have come down to us. It must have been in state festivals that queens and high priestesses were on public view. While performance arts are lost to us, stone images provide us with a glimpse.

4 Relief Sculpture Featuring Royal Women in Narrative Contexts In contrast to statuary, relief sculpture featuring early Mesopotamian elite women in narrative contexts is scarce. I count only twenty-seven items over a period of 1,200 years (see Table), the majority dating to the long Early Dynastic period. With one possible exception (no. 2), these items belong to the type of objects that a donor dedicated to a deity in a temple. The identified donors include Ur-Nanše of Lagaš (nos. 22–23), two elite men (nos. 14, 25), and Enheduana (no. 24). The most likely patrons of the remaining reliefs were kings or elite men: a Late Uruk king for the monumental vessel (no. 1); an elite man who may have been a local ruler, namely Ušumgal, pap-šeš of the god Šara, for the Early Dynastic I stela (no. 2); a king for the stela fragment that can be attributed to Gudea of Lagaš (no. 26); and kings or elite men for the anonymous door plaques if we extrapolate from the donors of inscribed door plaques (Braun-Holzinger 1991: 304–12). This would leave us with Enheduana as the only female donor of a relief image featuring herself. With the exception of the monumental stela fragment of Gudea, all reliefs are small to very small in scale: the largest is the 105 cm high Uruk Vase, followed by Ur-Nanše’s 91 cm high stela and his door plaque measuring 40 × 47 cm; Enheduana’s unusual, two-sided object found damaged has been

344

Claudia E. Suter

restored to a disk of 27 cm in diameter, and Ušumgal’s stela is only 22 cm high. The remaining objects are door plaques, all smaller than that of Ur-Nanše. No relief is made of the dark Gulf stone that kings used for their most prestigious monuments from the late Early Dynastic period on.9 Rather they are of easily available light-colored soft stones; only the Ur III door plaque (no. 27) is of steatite, probably from Syria.10 Although none of the reliefs have been found in their original context of use, some inferences about their visibility can be made. The target audience for dedicatory gifts is difficult to establish not only because most of them were found in secondary contexts, but also because their inscriptions only address the gods and future generations, and we have no knowledge of whether the population at large, the elite or only a small circle of elite members had access to temples (Michalowski 2013: 174–75). Inscriptions on a few royal monuments specifically mention their installation in courtyards (Braun-Holzinger 1991: 238). This rare bit of information may be mentioned only because it was exceptional (Civil 1980: 228), applying only to life-size or over life-size statues and large-size stelae, all of which were exclusive to kings. Such royal monuments conceivably also targeted a contemporary audience and may have stood in more accessible outdoor areas of temple compounds. The only candidate for such a setting among the reliefs under consideration is the stela of Gudea. By contrast, the small scale of the other reliefs, just like that of the extant female statuettes, speaks against an outdoor placement. Function corroborates this suggestion. The Uruk Vase was a cult vessel. Moreover, it includes a representation of identically shaped vessels, together with other cult paraphernalia and food supplies, in what appears to be the temple’s storeroom. Door plaques had a central peg that served to secure a door with a rope or hook, which could then be sealed (Braun-Holzinger 1991: 306–307 with literature). They must have been installed at the entrance of such storerooms. Thus, if access to indoor areas of temples was restricted to a small elite circle, these sculptural images of royal women were not on public display. The imagery carved on the reliefs under review pertains largely to cult or court ceremonies which, at the same time, conveyed social hierarchy and authority. The predominant scene is the banquet of a royal couple (nos. 3–17), to which I will return below. Most other scenes also depict the elite woman together with the (known or presumed) male patron of the relief (nos. 1–2, 18, 19, 22–23, 26), while high priestesses either preside over a libation (nos. 20, 21?, 24) or banquet with their divine husband (nos. 25, 27?). 9 Only few Lagaš II to Ur III statuettes of royal women are made of Gulf stone: Suter 2007: fig. 9; 2008: St. 16, 20, 26–28. 10 On materials, see Moorey 1994: 24–30.

On Images, Visibility, and Agency of Early Mesopotamian Royal Women

345

The Uruk Vase (no. 1) visualized the new order of the first urban society: one single man brings about prosperity due to his special relationship with the goddess and provides for his people, who depend on him (Suter 2014). The female figure who meets him in front of Inana’s temple/storehouse, was smaller in size than the now damaged king, wears a less elaborate garment, and lacks an entourage in contrast to the king. She must be a mortal inferior to the king, who represented the goddess. Although the Ušumgal Stela commemorates a transfer of property (Gelb et al. 1991: no. 12), it may have been the record of a local ruler (no. 2): Ušumgal’s hairstyle and beard are somewhat reminiscent of the Late Uruk royal figure, in contrast to the other typical Early Dynastic figures. Ušumgal appears on the main side of the four-sided stela about to perform the key act of the ritual that sealed the transfer, while his daughter Šara-igizi-Abzu(?), who is equal in size and whose vessel signals involvement in the ritual, approaches him on the adjacent narrow side. Ušumgal is followed by three male officials, Šara-igiziAbzu(?) by IGI.RU?-NUN, also daughter of a pap4-šeš and bearing the same title; the secondary figures are smaller than the protagonists. The male figure carrying a lamb, followed by a female figure, on a fragmentary door plaque from Girsu probably represent a royal couple taking offerings to a deity (no. 18). The peculiar image carved on a door plaque from Tell Asmar depicts, according to Licia Romano (2010: 947–48), a couple making butter in a churn (no. 19). Both reliefs of Ur-Nanše commemorate his temple building and his import of timber form far-away Dilmun (nos. 22–23). The main scene on the four-sided stela shows the king approaching an enthroned goddess, together with an entourage of sons and male officials, while a self-contained sub-scene below the king and his cupbearer, depicts his wife and daughter facing each other in banquet (Fig. 1); the women share with the goddess seated position, cup, and vegetal attribute. The door plaque depicts two superimposed scenes presided over by the king as builder and banqueter, respectively. He receives his entourage in audience and is considerably larger than all other figures. Only his daughter, who immediately faces him on the upper register, sticks out not only by her prominent place and larger size compared to the remaining, all male entourage, but also by wearing a garment of the same elaborate material as the king. She may well represent a high priestess destined to be married to the god of one of the temples the king claims to have built.11

11 The identity of this figure is disputed: Glassner 2003; Tunca 2004; Braun-Holzinger 2007: 60; Selz 2010; Romano 2014: 190–91.

346

Claudia E. Suter

Fig. 1: Stela of Ur-Nanše from al-Hiba, 91 cm high (drawing by author).

Gudea commemorated his temple building in much more detail. Various episodes of construction and inauguration were depicted in superimposed rectangular registers, while arched top registers showed the king before a deity, similar to Ur-Nanše, but without an entourage or banqueting women. The fragment depicting two women clapping their hands probably formed part of a musical performance during the inauguration (no. 26). Although the women’s attire and hairstyle is not distinguished from that of royal women, they may represent court musicians. Female subordinates of royal women are rarely distinguished in attire and hairstyle from their superior. The clappers’ multiplication recalls later “harems” which included large amounts of female musicians (Ziegler 2007). The libation scene presided over by a high priestess conveyed her role as that of an intermediary between the human and divine worlds, perhaps on the occasion of her installation in office. The two damaged figures following behind Enheduana may represent the governor of Ur and his wife who took part in the celebration (no. 24), analogous to the royal couple on the Early Dynastic door plaque that depicts a similar scene and was found in the same Gipar (no. 20 = Fig. 2). High priestesses who were daughters of hegemonic kings represented the crown on the local level.

On Images, Visibility, and Agency of Early Mesopotamian Royal Women

347

Fig. 2: Door Plaque from Ur, 22 cm high (courtesy of the Trustees of the British Museum).

The plaque that Nidupae, an archivist from Urusagrig/Šarrakum, dedicated to Ningublaga for the life of his lord Šaratigubisin, his own, and those of his wife and children depicts a woman in banquet with a god, presumably Ningublaga (no. 25 = Fig. 3). Šaratigubisin controlled a kingdom along the middle Tigris in Guti times and may have been either Šarkališari’s son (Sallaberger & Schrakamp 2015: 127) or the son of a Guti king (Steinkeller 2015: 284–85). I suspect that the image emulated now lost images of daughters of the house of Akkad who were depicted in banquet with their divine husband, not unlike peripheral rulers stepping into the vacuum left by the fall of Ur emulated images of Ur III kings (Suter 2010: Figs. 14–15, 24–25). The presumed prototype would have evoked the king becoming father-in-law of a god. The fragmentary Ur III parallel of such a scene preserves only the figure of a high priestess in her by then standard attire: flounced robe and circlet on long loose hair (no. 27).12 The woman on Nidupae’s plaque wears the flounced robe combined 12 On these two plaques, see also Suter 2007: 326–27, 337–38.

348

Claudia E. Suter

Fig. 3: Door Plaque of Nidupae, 25 cm high (courtesy of the Louvre, RMN/Christian Larrieu).

with a shawl over her head as worn by late Early Dynastic royal women from Lagaš (no. 23), Mari, and Ebla (Marchesi & Marchetti 2011: 208, pl. 63:1, 6).13 She may represent Nidupae’s wife, whose name appears next to her, or a daughter of Šaratigubisin, either in her role as a high priestess. Fifteen door plaques depict the banquet of a human couple (nos. 3–17). A hallmark of the Early Dynastic period, this scene occurs also in glyptic, on luxuriously inlaid objects, in sculptural groups of seated male and female statuettes with banqueter’s attributes, and possibly also on the lost top of the Bedre Stela.14 Gudrun Selz (1983: 441–62) convincingly identifies the banqueters facing each other on equal footing as royals. Recent suggestions to reduce this scene to funerary cult (Selz 2004; Cohen 2005), or identify it as “female 13 Compare also the statues Asher-Greve 1985: nos. 400, 401, 445. 14 Selz 1983, and for the stela, Börker-Klähn 1982: no. 12.

On Images, Visibility, and Agency of Early Mesopotamian Royal Women

349

Fig. 4: Seal, 5.1 cm high (courtesy of the Yale Babylonian Collection).

iconography” (Evans 2012: 188–91) do not convince me.15 Banquets were celebrated not only on recurring cult festivals (Asher-Greve 1985: 97–108), but also on special occasions, such as a temple inauguration (nos. 22–23) or a military victory as on the Standard of Ur. No matter what the occasion was, communal drinking cemented the recognition of authority and hierarchy and was mirrored in the world of the gods (Michalowski 1994). Mirroring of the earthly banquet in the divine sphere is not limited to poetry, but also evoked in visual imagery. While the depiction of divine banquets is rare and confined to glyptic (Braun-Holzinger 2013: 85–88), other scenes featuring deities in sculpture and glyptic can strikingly parallel royal banquets in composition and particulars. A good illustration is the two-registered seal combining such a scene with a royal banquet (Fig. 4).16 A divine

15 Contra the former, see also Baadsgaard et al. 2012: 151–52. 16 Buchanan 1981: no. 463.

350

Claudia E. Suter

couple or single deity is enthroned, often attended by servants, and can hold the cup and/or vegetal attribute.17 If the woman on the Uruk Vase (no. 1) indeed depicted a queen (Suter 2014), then the propagation of the royal couple as representing the gods on earth would go back to the beginnings of royal ideology in Mesopotamia. This would be in line with the emergence of a repertoire of royal images at this time which remained formative for millennia. The only identified donor of a door plaque depicting the royal banquet is the master stonecutter (GAL.ZADIM) Lumma (no. 14 = Fig. 5). Before this plaque was excavated, Henri Frankfort (1939: 47), who argued for a recurring cult festival – possibly the New Year’s festival – as subject of such plaques, suggested that they may have been dedicated by rich members of the community who contributed to the costs of the celebrations. The typically three-registered plaque of Lumma depicts the standard banquet of a royal couple on the top and another, unfortunately damaged banquet on the bottom. Since the depiction of two banquets is untypical, Lumma may have sponsored the second banquet, while the patrons of other door plaques may also have included kings. Seals depicting royal banquets may have been royal gifts. The extended banquet on the Standard of Ur conveyed, as Michalowski (2013: 176) convincingly argues, the king’s claim of command over military, economic, and ideological power networks. What was the queen’s role in a scene celebrating male power networks? Frankfort (1939: 46) already suggested that the vegetal attribute of banqueters pointed to procreation. Female figures – both queens and goddesses – usually hold what looks like a date cluster, while male figures’ vegetal attribute is usually different in shape and can, in some cases, be identified as a male date spathe (Hansen 2003: 31).18 Since fertilization of the date palm requires artificial pollination, such gendering leaves little doubt about the evocation of procreation. The link between the vegetal attribute – whether pertaining to the date palm or other crop – and procreation is corroborated by its presence in female banquets (e.g., no. 22), and its absence in banquets that do not revolve around a royal couple and celebrated ceremonies other than recurrent cult festivals as, for example, on the Standard of Ur, Ur-Nanše’s door plaque (no. 23), and the plaques depicting high priestesses with their divine husband (nos. 25, 27). Thus it was procreation rather than the banquet that queens were associated with. Procreation not only evoked agricultural prosperity, but also guaranteed succession to the throne. An unusual door plaque combines a banquet with a scene generally interpreted as depicting sexual intercourse, but perhaps more probably representing childbirth (no. 11 = Fig. 6). The damaged surface does 17 Braun-Holzinger 2013: Relief 2, 4–9, 12, Siegel 6–8. 18 For drawings, see Asher-Greve 1985: table 2.

On Images, Visibility, and Agency of Early Mesopotamian Royal Women

351

Fig. 5: Door Plaque of Lumma from Nippur, 31 cm high (3N133-134, courtesy of Richard L. Zettler).

not allow for a clear recognition of details beyond the identification of the left banqueter and the figure lying on the bed as women (Asher-Greve 1985: 101– 2). Westenholz (1999: 72) sees a midwife at work in the similar bed scene on a cylinder seal that also includes a figure standing at the bottom of the bed and touching the recumbent woman. The comparatively small, shapeless figure on top of the recumbent woman on the door plaque may then be a newborn child rather than a man penetrating her. Alternatively, there may be no figure on her belly at all, but simply the recumbent mother with a still pregnant belly holding her baby (which has previously been interpreted as the head of the penetrating man) in her outstretched arms.

352

Claudia E. Suter

Fig. 6: Fragmentary Door Plaque from Tell Asmar, 19 cm high (As. 32:930 & 1178, courtesy of the Oriental Institute of the University of Chicago).

It was not only in Mesopotamia that the queen’s official role centered on procreation. In Egypt, “on the ideological level, the queen represented the female principle of the universe through which the male king could renew himself. On the practical level, the royal women provided potential heirs to the throne. Both the king’s mother and the king’s principal wife had important ritual roles to play, and it is possible that these offered potential power to the holder” (Robins 1993: 55). In Hittite Anatolia, religious ideology was resistent to the patriarchal nature of the society owing to the equally important role that the Sun-goddess and the Storm-god at the head of the pantheon played in procreation. This assured for Hittite femmes sages essential spheres of activity in cultic life, and let queen Puduhepa invoke “I am a woman of the birth stool” in her plea with the gods on behalf of her ailing husband (Beckman 2000, and forthcoming). During the last quarter of the 3rd millennium, the banquet scene was gradually superseded by the presentation scene, which conveyed a similar ideological message with regard to authority and hierarchy (Zajdowski 2013). In contrast to banquets, however, presentations normally have a single human protagonist. Selz (1983: 578) explains the disappearance of the royal banquet scene with the emergence of deities in images by late Early Dynastic times,

On Images, Visibility, and Agency of Early Mesopotamian Royal Women

353

which made their representation by surrogate of the royal couple dispensable. In Lagaš II and Ur III times, the theme of procreation was visualized in scenes that depicted the king (or queen?) before an enthroned divine couple with the goddess sitting on the lap of her husband while he embraces her (Ornan 2010).19 With the dismissal of the banquet scene and the growing number of monumental victory stelae, in which elite women play no part, or monumental temple building stelae, in which neither a queen nor a high priestess seem to appear, narrative images of elite women become almost exclusively confined to glyptic. On seal images, queens and high priestesses are seen receiving subordinates in audience, like kings; queens can pour libations to goddesses and supervise women’s cult festivals; high priestesses are depicted with similar insignia and divine attributes as kings (Suter forthcoming). Yet queens are no longer depicted on an equal foot with their husbands as representatives of the gods. By Ur III times, seals that the king gifted to his concubines, sisters, and the wives of high functionaries depict them as subordinates, just like elite men. This may be related to a growing polygamy and foretokens queens identifying themselves in letters to their husbands as being their servants.

5 Conclusions Early Mesopotamian images of royal women were neither potential models for identification nor signs of their agency. They were not on public display, and the patrons of sculptural reliefs were generally men. Narrative scenes that feature royal women rather provide us with a glimpse into live performances of celebrations in which they, together with their husbands or on behalf of their fathers, represented the power before a contemporary audience. The gradual disappearance of such images in the 2nd millennium reflects a change in royal ideology rather than a decline in women’s social standing. Images of mortals in general are scarce in the 2nd millennium due to diverse factors including the vagaries of archaeological discovery, the abolition of divine kingship, and weak monarchies. Images pertaining to princesses married to gods naturally disappeared together with divine kingship in Old Babylonian times. Images featuring

19 Whereas the Ur-Namma Stela depicts the king before the divine couple, the fragmentary door plaque from Tello dedicated to Baba for Gudea’s life (Braun-Holzinger 1991: no. W 24) may have depicted the queen if the now lost dedicator was Gudea’s wife.

354

Claudia E. Suter

Fig. 7: Terracotta Plaque from Tello, 11 cm high (courtesy of the Louvre, RMN/Franck Raux).

queens together with their husbands alluding to procreation had already been abandoned by Ur III times, although the royal couple still seems to have represented the gods in state cult festivals at this time. Early 2nd millennium terracottas depicting embraced Ur III-style royal couples may hark back to such celebrations (Fig. 7).20 The same applies to Ur III-style figures of royal women apparently transformed into protective spirits on Old Babylonian seals (Suter 2008: 11). Whether royal women also disappeared from public view in performance arts at that time is difficult to answer. Due to the dearth of stone images that included them and texts recording the administration of state festivals, we lose our window into such ceremonies.

20 Barrelet 1968: no. 523 (pp. 290–91).

On Images, Visibility, and Agency of Early Mesopotamian Royal Women

355

The unofficial role of royal women is hard to assess because neither sculptural images nor literary texts took any interest in individual careers. Enheduana, the only known female patron of a relief image of herself, may have been a case of a woman of influence even if subservient to the regime. As the first en-priestess of Nanna in Ur, she may have made this office, which already existed in some form in Early Dynastic times and continued to be held by a royal daughter until the last Larsa king, prestigious. Perhaps it was not only Sumerian heritage inculcated in scribal schools, but also the memory of an extraordinary woman that caused Old Babylonian scribes to include her prominently in hymns? Perhaps her fame was related to some role she had in her nephew’s becoming the first self-deified king?

Period

Late Uruk

ED I

ED II-IIIa

ED II-IIIa

ED II-IIIa ED II-IIIa ED II-IIIa

ED II-IIIa ED II-IIIa

ED II-IIIa

ED II-IIIa

ED II-IIIa ED II-IIIa

ED II-IIIa ED II-IIIa ED II-IIIa ED II-IIIa

No.

 1

 2

 3

 4

 5  6  7

 8  9

10

11

12 13

14 15 16 17

Khafaje: Houses II? Tell Asmar: Abu Square II Tell Asmar: Abu Single Shrine I Tell Asmar: Abu Single Shrine I Nippur: test pit Nippur: Inana V? debris ED/Ur III Nippur: Inana VIIB – – Nippur: Inana VIII

Tell Agrab: Šara Main Tell Agrab: Šara Main Khafaje: Oval I Khafaje: Sin IX Khafaje: Sin IX

–

Uruk: Eanna “Sammelfund”

Provenance

Patron

DP 31 × 28.3 Lumma DP 26 × 24 DP 30 × 30 DP ?

DP 32 × 30 DP 27 × 26.5

DP 28 × 30

DP 22 × 22

DP *6 × 6 DP 17 × 14

DP 32 × 29.5 DP 20 × 20 DP 27 × 28

DP 25 × 22

Stela Ušumgal 22 × 14 × 9 DP 18 × 17.5

Vessel 105 × 36

Object

Tab. 1: Women in Stone Relief from the Late Uruk to the Ur III period.

Banquet Banquet Banquet Banquet

of of of of

royal royal royal royal

couple couple couple couple

Banquet of royal couple? + childbirth? Banquet of royal couple Banquet of royal couple

Banquet of royal couple

Banquet of royal couple Banquet of royal couple

Banquet of royal couple Banquet of royal couple Banquet of royal couple

Banquet of royal couple

King at head of procession facing woman in front of goddess’ temple/storehouse Men and women in rite sealing transfer of property Banquet of royal couple

Image

IM 66157 unknown IM 47225 IM 66154

MMA 59.41.10 IM 60950

IM 15547

IM 19794

IM 14661 OI A.12417 AO 14053 +OI A.12392 IM (Kh III 583) OI A.12305

OI A.18073

IM 27869

MMA 58.29

IM 19606

Museum

Braun-Holzinger 1991: W 13 Boese 1971: K 2 Boese 1971: K 8 Asher-Greve 1985: 565

Boese 1971: N 4 Boese 1971: N 5

Boese 1971: AS 4

Boese 1971: AS 3

Boese 1971: CH 2 Boese 1971: AS 2

Boese 1971: CT 2 Boese 1971: CS 7 + K 7 Boese 1971: K 1 + CS 4

Boese 1971: AG 2

Boese 1971: AG 1

Braun-Holzinger 2007: FD 27

Braun-Holzinger 2007: FS 1

Reference

356 Claudia E. Suter

ED III

ED IIIb ED IIIb

ED IIIb

Akkad

Akkad

Lagaš II

Ur III

20

21 22

23

24

25

26

27

–

–

–

Ur: Gipar

Girsu: Tell K

Ur: Gipar al Hiba: Ibgal

Stela *16 × 21 × 8 DP 14 × 15

Disk? 27 × 7.7 DP 30 × 26

DP 40 × 47

DP *12 × 14 Stela 91 × 47 × 17

Girsu: Tell V DP 35 × 30 Tell Asmar: DP 14 × 14 Abu Single Shrine III Ur: Gipar DP 22 × 22

[…]

(Gudea?)

Nidupae

Banquet of high priestess? & god Female clappers in musical performance? Banquet of high priestess & god?

Libation of high priestess + women before god […] Fragment of dito? Ur-Nanše King & male entourage before goddess + banquet of queen & princess Ur-Nanše King as builder & banqueter receiving entourage Enheduana Libation of high priestess

Royal couple offering Couple making butter?

AO 2761

AO 10235

AO 4799

CBS 16665

AO 2344

CBS 16682 IM 61404

BM 118561

AO 3290 OI A.11410

Braun-Holzinger 1991: W 30

Braun-Holzinger 1991: W 1; RIME 1.9.1.2 Braun-Holzinger 1991: Varia 5; RIME 2.1.1.16 Braun-Holzinger 1991: W 23; RIME 2.8. 1. 2001 Suter 2000: ST.15

Braun-Holzinger 1991: W 14 Braun-Holzinger 1991: Stele 5; RIME 1.9.1.6a

Boese 1971: U 4

Boese 1971: T 9 Boese 1971: AS 5

ED = Early Dynastic, DP = door plaque. Measurements are in cm; for door plaques they concern their image field, complete or estimated; an asteriks flags items too fragmentary for estimating their original size.

ED II-IIIa ED III

18 19

On Images, Visibility, and Agency of Early Mesopotamian Royal Women

357

358

Claudia E. Suter

Bibliography Asher-Greve, Julia M. 1985. Frauen in altsumerischer Zeit. BiMes 18. Malibu, CA: Undena Publications. Asher-Greve, Julia M. 1997. Feminist Research and Ancient Mesopotamia: Problems and Prospects. Pp. 218–37 in A Feminist Companion to Reading the Bible: Approaches, Methods and Strategies, ed. Athalya Brenner and Carole Fontaine. Sheffield: Sheffield Academic Press. Asher-Greve, Julia M. 2012. Women and Agency: A Survey from Late Uruk to the End of Ur III. Pp. 359–77 in The Sumerian World, ed. Harriet Crawford. London: Routledge. Baadsgaard, Aubrey, Janet Monge, and Richard L. Zettler. 2012. Bludgeoned, Burned, and Beautified: Reevaluating Mortuary Practices in the Royal Cemetery of Ur. Pp. 125–58 in Sacred Killing: The Archaeology of Sacrifice in the Ancient Near East, ed. Anne M. Porter and Glenn M. Schwartz. Winona Lake, IN: Eisenbrauns. Bahrani, Zainab. 2001. Women of Babylon: Gender and Representation in Mesopotamia. London: Routledge. Barrelet, Marie-Thérèse. 1968. Figurines et reliefs en terre cuite de la Mésopotamie antique. Paris: Geuthner. Barrelet, Marie-Thérèse, with a contribution of Jean-Marie Durand. 1974. La ‘figure du roi’ dans l’iconographie et dans les textes depuis Ur-Nanše jusqu’à la fin de la Ire dynastie de Babylone. Pp. 27–138 in Le palais et la royauté: Compte rendu de la XIXe Rencontre Assyriologique Internationale, ed. Paul Garelli. Paris: Geuthner. Batto, Bernard Frank. 1974. Studies on Women at Mari. Baltimore, MD: The Johns Hopkins University Press. Beckman, Gary M. 2000. Goddess Worship: Ancient and Modern. Pp. 11–23 in A Wise and Discerning Mind: Essays in Honor of Burke O. Long, ed. Saul M. Olyan and Robert C. Culley. BJS 325. Providence, RI: Brown Judaic Studies. Beckman, Gary M. forthcoming. Females as Sources of Authority in Hittite Government and Religion. In Structures of Power: Law and Gender across the Ancient Near East and Beyond, ed. Ilan Peled. OIS 11. Chicago: The Oriental Institute. Boese, Johannes. 1971. Altmesopotamische Weihplatten. Eine sumerische Denkmalsgattung des 3. Jahrtausends v. Chr. UAVA 6. Berlin: De Gruyter. Börker-Klähn, Jutta. 1982. Altvorderasiatische Bildstelen und vergleichende Felsreliefs. BaF 4. Mainz: Philipp von Zabern. Braun-Holzinger, Eva Andrea. 1991. Mesopotamische Weihgaben der frühdynastischen bis altbabylonischen Zeit. HSAO 3. Heidelberg: Heidelberger Orientverlag. Braun-Holzinger, Eva Andrea. 1996. Altbabylonische Götter und ihre Symbole. Benennung mit Hilfe der Siegellegenden. BaM 27: 235–359. Braun-Holzinger, Eva Andrea. 2007. Das Herrscherbild in Mesopotamien und Elam. Spätes 4. bis frühes 2. Jt. V. Chr. AOAT 342. Münster: Ugarit-Verlag. Braun-Holzinger, Eva Andrea. 2013. Frühe Götterdarstellungen in Mesopotamien. OBO 261. Fribourg: Academic Press. Brisch, Nicole Maria. 2007. Tradition and the Poetics of Innovation: Sumerian Court Literature of the Larsa Dynasty (c. 2003–1763 BCE). AOAT 339. Münster: Ugarit-Verlag. Brisch, Nicole Maria. 2011. Changing Images of Kingship in Sumerian Literature. Pp. 706–24 in The Oxford Handbook of Cuneiform Culture, ed. Karen Radner and Eleanor Robson. Oxford: Oxford University Press.

On Images, Visibility, and Agency of Early Mesopotamian Royal Women

359

Buchanan, Briggs. 1981. Early Near Eastern Seals in the Yale Babylonian Collection. New Haven: Yale University Press. Burke, Peter. 2001. Eyewitnessing: The Uses of Images as Historical Evidence. London: Reaktion Books. Civil, Miguel. 1980. Les limites de l’information textuelle. Pp. 225–232 in L’archéologie de l’Iraq du début de l’époque néolithique à 333 avant notre ère: Perspectives et limites de l’interprétation anthropologique des documents, ed. Marie-Thérèse Barrelet. Paris: Centre National de la Recherche Scientifique. Cohen, Andrew C. 2005. Death Rituals, Ideology, and the Development of Early Mesopotamian Kingship: Toward a New Understanding of Iraq’s Royal Cemetery of Ur. Leiden: Brill. Colbow, Gudrun. 2002. Priestesses, either Married or Unmarried, and Spouses without Title: Their Seal Use and their Seals in Sippar at the Beginning of the Second Millennium BC. Pp. 85–90 in Sex and Gender in the Ancient Near East: Proceedings of the 47 th Rencontre Assyriologique Internationale, Helsinki, ed. Simo Parpola and Robert M. Whiting. Helsinki: The Neo-Assyrian Text Corpus Project. Cooper, Jerrold S. 1997. Gendered Sexuality in Sumerian Love Poetry. Pp. 85–97 in Sumerian Gods and Their Representations, ed. Irving L. Finkel and Markham J. Geller. Groningen: Styx. Dalley, Stephanie. 1984. Mari and Karana: Two Old Babylonian Cities. London: Longman. Edzard, Dietz O. 2004. Altbabylonische Literatur und Religion. Pp. 485–640 in Mesopotamien Annäherungen 4: Die altbabylonische Zeit, ed. Pascal Attinger, Walther Sallaberger, and Markus Wäfler. Fribourg: Academic Press. Evans, Jean M. 2012. The Lives of Sumerian Sculpture: an Archaeology of the Early Dynastic Temple. Cambridge: Cambridge University Press. Frankfort, Henri. 1939. Sculpture of the Third Millennium B.C. from Tell Asmar and Khafajah. OIP 44. Chicago: University of Chicago Press. Gadotti, Alhena. 2011. Portraits of the Feminine in Sumerian Literature. JAOS 131: 195–206. Glassner, Jean-Jaques. 2003. Áb-šu-tur, devin d’Ur-Nanshe? NABU no. 106. Glassner, Jean-Jaques. 2009. En-hedu-ana, une femme auteure en pays de Sumer, au IIIe millenaire? Pp. 219–231 in Femmes, cultures et sociétés dans les civilisations méditerranéennes et proche-orientales de l’Antiquité, ed. Françoise Briquel-Chatonnet, Saba Farès, Brigitte Lion, and Cécile Michel. Paris: De Boccard. Jacobsen, Thorkild. 1976. The Treasures of Darkness: A History of Mesopotamian Religion. New Haven: Yale University Press. Jacobsen, Thorkild. 1987. The Harps that Once …: Sumerian Poetry in Translation. New Haven: Yale University Press. Karanika, Andromache. 2014. Voices at Work: Women, Performance, and Labor in Ancient Greece. Baltimore: Johns Hopkins University Press. Krebernik, Manfred. 1984. Die Beschwörungen aus Fara und Ebla: Untersuchungen zur ältesten keilschriftlichen Beschwörungsliteratur, Texte und Studien zur Orientalistik. Hildesheim: Georg Olms. Limet, Henri. 1971. Les légendes des sceaux cassites. Bruxelles: Palais des Académies. Lion, Brigitte. 2011. Literacy and Gender. Pp. 90–112 in The Oxford Handbook of Cuneiform Culture, ed. Karen Radner and Eleanor Robson. Oxford: Oxford University Press. Marchesi, Gianni. 2011. Goods from the Queen of Tilmun. Pp. 189–199 in Akka de is King: A Collection of Papers by Friends and Colleagues Presented to Aage Westenholz, ed.

360

Claudia E. Suter

Gojko Barjamovic, Jacob L. Dahl, Ulla S. Koch, Walter Sommerfeld, and Joan G. Westenholz. Leiden: NINO. Marchesi, Gianni and Nicolò Marchetti. 2011. Royal Statuary of Early Dynastic Mesopotamia. MC 14. Winona Lake, IN: Eisenbrauns. Melville, Sarah C. 2005. Royal Women and the Exercise of Power in the Ancient Near East. Pp. 219–28 in A Companion to the Ancient Near East, ed. Daniel C. Snell. Oxford: Blackwell. Michalowski, Piotr. 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, ed. Lucio Milano. Padova: Sargon. Michalowski, Piotr. 2010. On the Names of Some Early Ancient Near Eastern Royal Women and on a Seal Impression from Karum Kanesh. Pp. 211–16 in DUB.SAR É.DUB.BA.A: Studies presented in honour of Veysel Donbaz, ed. Şevket Dönmez. İstanbul: Ege Yayınları. Michalowski, Piotr. 2013. Networks of Authority and Power in Ur III Times. Pp. 169–205 in From the 21 st Century BC to the 21 st Century AD, ed. Steven Garfinkle and Manuel Molina. Winona Lake, IN: Eisenbrauns. Molyneaux, Brian Leigh (ed.). 1997. The Cultural Life of Images: Visual Representation in Archaeology. London: Routledge. Moorey, Peter R. S. 1994. Ancient Mesopotamian Materials and Industries: the Archaeological Evidence. Oxford: Clarendon Press. Nunn, Astrid. 1988. Die Wandmalerei und der glasierte Wandschmuck im Alten Orient. Handbuch der Orientalistik. Leiden: Brill. Ornan, Tallay. 2010. Divine Love: Nanna, Ningal and their Entourage on a Clay Plaque. Pp. 119–43 in A Woman of Valor: Jerusalem Ancient Near Eastern Studies in Honor of Joan Goodnick Westenholz, ed. Wayne Horowitz, Uri Gabbay, and Filip Vukosavović. Madrid: Consejo Superior de Investigaciones Científicas. Orthmann, Winfried (ed.). 1975. Der Alte Orient. Berlin: Propyläen Verlag. Robins, Gay. 1993. Women in Ancient Egypt. London: British Museum Press. Romano, Licia. 2010. Who was Worshipped in the Abu Temple in Tell Asmar? KASKAL 7: 51–65. Romano, Licia. 2014. Urnanshe’s Family and the Evolution of Its Inside Relationships as Shown by Images. Pp. 183–92 in La famille dans le Proche-Orient ancien: réalités, symbolismes et images: Proceedings of the 55 th Rencontre Assyriologique Internationale, Paris, ed. Lionel Marti. Winona Lake: Eisenbrauns. Rubio, Gonzalo. 2009. Sumerian Literature. Pp. 11–75 in From an Antique Land: An Introduction to Ancient Near Eastern Literature, ed. Carl S. Ehrlich. Lanham: Rowman & Littlefield. Sallaberger, Walther. 1999. Ur III-Zeit. Pp. 121–390 in Mesopotamien Annäherungen 3: Akkade-Zeit und Ur III-Zeit, ed. Pascal Attinger and Markus Wäfler. Fribourg: Universitätsverlag. Sallaberger, Walther and Fabienne Huber Vulliet. 2005. Priester. A. I. Mesopotamien. RlA 10: 617–40. Sallaberger, Walther and Ingo Schrakamp. 2015. The Gutean Period: A Problem of 3rd Millennium Chronology. Pp. 113–130 in History & Philology, ed. Walther Sallaberger and Ingo Schrakamp. ARCANE 3. Turnhout: Brepols. Scheer, Tanja S. 2006. Bilder der Macht? Repräsentationen römischer Kaiserinnen. Pp. 295– 321 in Images and Gender: Contributions to the Hermeneutics of Reading Ancient Art, ed. Silvia Schroer. Göttingen: Vandenhoeck & Ruprecht.

On Images, Visibility, and Agency of Early Mesopotamian Royal Women

361

Selz, Gebhard J. 2004. Feste in Stein. Der frühmesopotamische Kult der Bilder: Identität und Differenz. Archiv für Religionsgeschichte 6: 19–38. Selz, Gebhard J. 2010. Immer nur Söhne und keine Töchter? Zu einem Familierelief des UrNanše. Pp. 187–96 in Von Göttern und Menschen: Beiträge zu Literatur und Geschichte des alten Orients: Festschrift für Brigitte Groneberg, ed. Dahlia Shehata, Frauke Weiershäuser, and Kamran V. Zand. Leiden: Brill. Selz, Gudrun. 1983. Die Bankettszene. Entwicklung eines überzeitlichen Bildmotivs in Mesopotamien, von der frühdynastischen bis zur Akkad-Zeit. FAOS 11. Wiesbaden: Franz Steiner. Sharlach, Tonia M. 2007. Shulgi-simti and the Representation of Women in Historical Sources. Pp. 363–68 in Ancient Near Eastern Art in Context: Studies in Honor of Irene J. Winter, ed. Marian Feldman and Jack Cheng. Boston: Brill. Slanski, Kathryn E. 2003. The Babylonian Entitlement ‘narûs’ (kudurrus): A Study in their Form and Function. Boston: American Schools of Oriental Research. Spycket, Agnès. 1981. La statuaire du Proche-orient ancien. Leiden: Brill. Steinkeller, Piotr. 2015. The Gutian Period in Chronological Perspective. Pp. 281–88 in History and Philology, ed. Walther Sallaberger and Ingo Schrakamp. ARCANE 3. Turnhout: Brepols. Suter, Claudia E. 2000. Gudea’s Temple Building: the Representation of an Early Mesopotamian Ruler in Text and Image. CM 17. Groningen: Styx Publications. Suter, Claudia E. 2007. Between Human and Divine: High Priestesses in Images from the Akkad to the Isin-Larsa Period. Pp. 315–59 in Ancient Near Eastern Art in Context: Studies in Honor of Irene J. Winter, ed. Marian Feldman and Jack Cheng. Boston: Brill. Suter, Claudia E. 2008. Who are the Women in Mesopotamian Art from ca. 2334–1763 BCE? KASKAL 5: 1–55. Suter, Claudia E. 2010. Ur III Kings in Images: A Reappraisal. Pp. 319–49 in Your Praise is Sweet: A Memorial Volume for Jeremy Black from Students, Colleagues, and Friends, ed. Heather D. Baker, Eleanor Robson, and Gábor Zólyomi. London: British Institute for the Study of Iraq. Suter, Claudia E. 2014. Human, Divine or Both? The Uruk Vase and the Problem of Ambiguity in Early Mesopotamian Visual Arts. Pp. 545–68 in Critical Approaches to Ancient Near Eastern Art, ed. Brian A. Brown and Marian H. Feldman. Boston: De Gruyter. Suter, Claudia E. forthcoming. Images of Queens, High Priestesses, and Other Elite Women in 3 rd Millennium Mesopotamia. In Women in Antiquity: Real Women Across the Ancient World, ed. Stephanie L. Budin and Jean M. Turfa. London: Routledge. Tunca, Öhnan. 2004. A propos d’une figure de la plaque perforée d’Ur-Nanše: fille/femme, fils ou devin? NABU no. 22. Van De Mieroop, Marc. 1989. Women in the Economy of Sumer. Pp. 53–69 in Women’s Earliest Records: From Ancient Egypt and Western Asia, ed. Barbara Switalski Lesko. Atlanta, GA: Scholars Press. Veldhuis, Niek. 2008. Kurigalzu’s Statue Inscription. JCS 60: 25–51. Weiershäuser, Frauke. 2008. Die königlichen Frauen der III. Dynastie von Ur. GBAO 1. Göttingen: Universitätsverlag. Westenholz, Aage. 1999. The Old Akkadian Period: History and Culture. Pp. 17–117 in Mesopotamien Annäherungen 3: Akkade-Zeit und Ur III-Zeit, ed. Pascal Attinger and Markus Wäfler. Fribourg: Universitätsverlag. Wiggermann, Frans A. M. 1998. Nackte Göttin A. Philologisch. RlA 9: 46–53.

362

Claudia E. Suter

Wright, Rita P. 2008. Gendered Relations and the Ur III Dynasty: Kinship, Property, and Labor. Pp. 247–79 in Gender through Time in the Ancient Near East, ed. Diane Bolger. Lanham, MD: Altamira Press. Zajdowski, Karol. 2013. Transformation of the Mesopotamian Banquet Scene into the Presentation Scene in the Early Dynastic, Akkadian and Ur III Periods. Akkadica 134: 1–16. Ziegler, Nele. 2007. Les musiciens et la musique d’après les archives de Mari, Mémoires de NABU. Antony: SEPOA.

Niek Veldhuis

Words and Grammar: Two Old Babylonian Lists The present contribution offers discussions of two lexical lists: a bilingual Old Babylonian exemplar of the thematic list har-ra (section 1) and a Sealand dynasty bilingual paradigm of the verb gub = izuzzu (section 2). Section 1 is concerned with the semantics of a number of words, both Akkadian and Sumerian, for which the new text provides additional evidence. Section 2 discusses the organization of the gub = izuzzu paradigm and includes an edition of the text.1 Throughout the article tablets and composite (lexical) texts are identified by a six-digit P or Q number, respectively. These numbers are recognized by CDLI (http://cdli.ucla.edu) and by all ORACC projects (http://oracc.org), giving access to editions, photos, hand copies and metadata, where available. The six digit P and Q numbers may be entered in the search box, or added directly to the URL (e.g., http://oracc.org/dcclt/P247857 or http://cdli.ucla.edu/P247857).

1 The Old Babylonian Bilingual HAR-ra Exemplar BM 85983 1.1 Introduction BM 85983 is a large tablet of unknown provenance written in an unmistakable Old Babylonian hand. It contains the sections leather (kuš), copper (urudu), and bronze (zabar) from the thematic series har-ra2 in a bilingual format with occasional glosses to the Sumerian. Many of the more than 250 preserved lines are well-known from other versions of har-ra and require no comment. A good number of entries, however, are attested here for the first time, or throw additional light on previously known words and expressions. The goal of the present section is to discuss those entries and their contribution to Akkadian and Sumerian lexicography. A full edition with translation of the tablet is

1 Jay Crisostomo read the article and the gub = izuzzu paradigm text and offered many insightful suggestions and corrections, for which my sincere thanks. 2 For the conventional label har-ra, see Civil 2008a: 5 n. 13. https://doi.org/10.1515/9781501503696-021

364

Niek Veldhuis

found on DCCLT at the address http://oracc.org/dcclt/P247857. Occasional references to the tablet may be found in Civil 2008a: 91 (where BM 85893 should be corrected to 85983), Rubio 2000: 214, and Veldhuis 2014: 151–53 (with line drawings), 247. Among the thousands of Old Babylonian exemplars of the thematic series har-ra from Nippur, Ur, Uruk, Isin, Sippar, Larsa, Kisurra, Terqa, and other places, BM 85983 is highly unusual because of its bilingual format. The only other instances of bilingual Old Babylonian har-ra are a couple of translation glosses in a few Nippur exemplars3 and a group of lenticular exercises, mostly from Susa, that provide two or three lines in Sumerian on one side and the same text in syllabic Sumerian and Akkadian translation on the other side.4 In addition to being bilingual, BM 85983 has a number of other characteristics that are expected in Kassite or Middle Babylonian peripheral exemplars, but not in the Old Babylonian period. Among these characteristics are: – repetition of an Akkadian translation by numbers (well-known from Middle Assyrian and Middle Babylonian lexical texts; see Finkel 1982: 5–6). – Akkadian column frequently left empty (because the translation is obvious?). Compare, for instance, SLT 44 (P227759), a Kassite copy of har-ra 13 (domestic animals). – omission of determinatives (zabar section on reverse ii-iv). Well known from Kassite sources and contemporary texts from Emar (Syro-Hittite tradition)5 and Ugarit, but not known in Old Babylonian exemplars. In its paleography,6 however, BM 85983 has all the characteristics of an Old Babylonian tablet and the text as preserved is rather far removed from the established version of har-ra 11 and 12 as known from first millennium sources. Both features argue against a Kassite dating. BM 85983 may thus serve to emphasize how artificial labels such as “Old Babylonian” or “Middle Babylonian” are, how complex the history of lexical texts in the second millennium really is and how much of what we know or think to know depends on textual finds from short periods and from a small number of sites – with large blanks in between.

3 The texts are: SLT 37 + SLT 46 + (with several additional fragments) = P227892; HS 1659+ = P229794; HS 1799 = P229962; and perhaps SLT 250 (P228002). 4 See Malayeri 2012; Civil 1975. 5 The Syrian tradition from Emar is slightly older and largely conforms to Old Babylonian standards (Cohen 2009). 6 For instance, har (rev. i 11) and ah (obv. iii 14) are written hi×aš2 and hi×nun (rather than hi.aš2 and hi.nun), respectively.

Words and Grammar: Two Old Babylonian Lists

365

In terms of Akkadian orthography, the text usually omits mimation. The sign pi is used for both pi/e and for wa/e/i/u. An unusual aspect is the use of the sign ga₂ (rather than ma) to represent original /ĝ/ of a Sumerian loanword. There are two examples: urud

obv. v 24 rev. ii 18

a₂-aš-ĝar šu-uš-ĝar

a₂-aš-ga₂-ru šu-uš-ga₂-ru

a battle ax (bronze) basin

The second of these words is well known and is listed as šušmarû in the dictionaries. The first will require more discussion – see below 1.2.2.5. BM 85983 begins with hides and leather; only the final section of this listing is preserved in what probably was column 3 of the obverse. The text continues with a short section of earths (im; beginning of obv. iv), followed by copper and copper objects (obv. iv–rev. i). Two lines inscribed on the right edge have been added to the end of rev. column 1 in the DCCLT edition, at the very end of the copper section. Thematically these items belong in the section that lists qualities of copper and copper fragments (obv. iv 8–20). It seems likely that these entries were accidentally skipped and added afterwards. Many years ago, during a visit to the Oriental Institute, I read parts of this tablet with Miguel Civil, who generously offered his time and his insights. This study is dedicated to him as a token of appreciation for his scholarship and generosity.

1.2 Leather Objects The first column of BM 85983 is entirely broken away; of the second column only a few traces of Akkadian translations remain. The third column preserves the final 50 entries of the section kuš (leather).

1.2.1. zim(b)uḫaru “waterskin” This rare word so far appears exclusively in the lexical tradition. BM 85983 had no less than eight entries for this word, of which five are preserved: obv. iii 1. 2. 3.

kuš

x niĝ₂-umbin-na kuš niĝ₂-hul-a kuš

= [3 (zimuḫaru)] = [4 (zimuḫaru)] = 5 (zimuḫaru)

366

4. 5.

Niek Veldhuis

kuš

hi-ru zi-mu-ha-ru-um

kuš

= 6 (zimuḫaru) = 7 (zimuḫaru)

More than 40 lines below, at the very end of the kuš section, the text returns to the zimuḫaru theme: kuš niĝ₂-ĝal₂-la 49. kuš 50. a-ĝa₂-la₂ i₃-ĝeš (end of column)

= zi-mu-ḫa-ru = a-ia-ṣum

water skin leather container for oil

Of these 5 preserved Sumerian equivalencies for zimuḫaru, three are known elsewhere in the lexical tradition in slightly different forms. kušhi-ru (obv. iii 4) is related to the entry hirin₂(ku₇) = zinbuḫaru in Ea 4, 194 (MSL 14, p. 363; see Civil 2007: 25). The entries kušniĝ₂-umbin-na and kušzi-mu-ḫa-ru-um are known from first millennium har-ra 11 (see MSL 9, p. 197), unilingual Emar exemplars and Old Babylonian copies. The Sumerian equivalences kušniĝ₂hul-a and kušniĝ₂-ĝal₂-la are new. All these Sumerian entries may now be associated with a leather bag or water skin on the strength of Malku II 245 hiri-in-nu = na-a-a-du (Hruša 2010: 70).

1.2.2 kuš niĝ₂ nu-si-is = ša lā pīdu “leather strap” or “lash” (obv. iii 43) Sumerian si-is (or si-iz) is attested exclusively in the lexical tradition, where it is usually translated by Akkadian pedû. A brief section on si-is is found in the enigmatic lexical text CUSAS 12: 7.1 section 2 i 4–8 (Civil 2010: 247 and 254; P253232; P253256). In the present text the word is used to designate an otherwise unknown leather object, presumably a strap or something similar used to restrain or perhaps to lash a captive. kuš niĝ₂ nu-si-is = ša la pi-du

the merciless one

It is likely that the same Akkadian word may be restored in obv. iii 31: 31. 32.

kuš

lu2×gan2tenû-pap lu2×gan2tenû-pap

kuš

= la ˹pi?˺-[du] = šem-x-bu

The Sumerian items in obv. iii 31–32 refer to leather items for a captive (heš₅ or šaĝa); they are known from Old Babylonian and Middle Babylonian versions of har-ra in the following forms:

Words and Grammar: Two Old Babylonian Lists

367

OB Nippur har-ra 2 (Q000040): 447 lu2×gan2tenû-du

kuš

MB Emar Msk 74103b obv. iii 12–13′ (Emar 6/1: 257; P271466) ˹kušlu2×gan2tenû˺ = […] kuš lu2×gan2tenû […] = […] In both versions the entry is found among words for whips and straps.

1.2.3

kuš

lu-ub₂sar = parūtu “leather pouch” (obv. iii 46)

The word parūtu appears in Old Babylonian texts from Mari and Isin. It is always found in the context of leather objects and was interpreted as a quiver by Von Soden in AHw (followed by Van de Mieroop 1987: 149; Schrakamp 2010b: 169), but as a tarpaulin by Arkhipov 2012a: 10–11 and as “a kind of leather” in CAD P. The present equation with Sumerian kušlu-ub₂sar (“bag,” or “pouch”; Civil 2013: 45–46) favors the meaning “leather container,” which may well include the meaning “quiver.” Arkhipov argued for “tarpaulin” because the word often qualifies a chariot. However, the association with chariots is consistent with the meaning “quiver.” In the post-OB lexical tradition the word is found as an Akkadian loan in Sumerian in the form kušbar-ru-tum (Emar, Syrian tradition) or kušpa-ru-tum (Emar, Syro-Hittite tradition, and first millennium sources).

1.3 Copper and Copper Objects 1.3.1 urud uš₂ = uššu “nugget” (obv. iv 10) This entry appears in the context of different forms and qualities of copper:  8.  9. 10. 11. 12. 13. etc.

urud urud urud urud urud urud

niĝ-kalag-ga uš₂ [iš]-me-na kalag-ga luh-ha

= = = = = =

we-ru-u₂ min uš-šu uš-me-na-ku uš-šu dan-nu we-ru-u₂

copper (high quality) copper copper nugget malachite high quality copper nugget refined copper = copper

Akkadian uššu is tentatively defined in the dictionaries as a copper vessel, based primarily on the entry in Diri 6 B 73 (see below), but a comparison of its

368

Niek Veldhuis

various attestations and contexts will show that the meaning “nugget” is more likely. The most revealing contextual use of Akkadian uššu is in the Old Babylonian administrative text HMA 9–1811, edited by Reiter 1997: 104*, where 1.5 ma.na urud uš-šu and 2/3 ma.na 2 gin2 an-na-ku are added to an alloy.7 A further indication for the meaning of uššu may be derived from a comparison between the passage in Diri 6 B and some entries in Aa. Diri 6B 70–73 (MSL 15, p. 192) has three readings of the sign combination urud.si.a, as follows:8 (nindar) (du₂₆) (uš₁₅)

nin-da-ar₂ var: nin-du-ub du uš

urud.si.a

kirinnu

lump

urud.si.a urud.si.a

kamāru uššu

to heap up nugget

Akkadian kirinnu (70) is translated as “lump” in CAD, based on the form kirinnû in Aa 1/2 30 (MSL 14, p. 209), where it stands in parallel with kirṣu: gi-ri-ak

lagab

kirṣu ša pahāru kirinnû min

lump (of clay) of a potter lump-shaped (piece of clay), of a potter

The word kirinnu turns up once more in Aa 2/3 in the section that deals with the sign bad in the reading uš₂ (Aa 2/3 B 4′–6′). Among the Akkadian translations are the following (MSL 14, pp. 276–77; K 7751 = P382622): (uš₂) (uš₂) (uš₂)

kug.sig7 kug.babbar kirinnu

(nugget of) gold (nugget of) silver lump; nugget

We may thus conclude that the Akkadian words kirinnu (Sumerian nindar or uš₂) and uššu (uš₂ or uš₁₅) are near-synonyms, meaning “lump” (of clay) or “nugget” (of metal). In summary, the entry uš₂(bad) = kirinnu in Aa 2/3 reflects the same Sumerian word as urud uš₂ = uššu in BM 85983 which is found in an alternative writing as uš₁₅(urud.si.a) = uššu in Diri 6B.

7 This attestation was not used in CAD U/W. 8 Line 71 is only represented in PTS 1 and is omitted here.

Words and Grammar: Two Old Babylonian Lists

369

1.3.2 urud za-ri₂-in = zarinnu piṣṣum “broken pieces of copper” (obv. iv 17) Sumerian za-ri₂-in and its Akkadian equivalent zarinnu are well-known qualifications for metals and textiles, meaning “coarse” or “low quality.” Akkadian piṣṣu was interpreted by Arkhipov 2012b: 41 n. 130 as a word for scraps or broken pieces, a meaning that is well supported by the present entry. The word reappears in the section bronze in the entry [za]bar za-ri₂-in = pi-iṣ-ṣu₂ hušu-u = za-ri₂-in-nu (rev. ii 4; for hušû see below 1.2.2.7).

1.3.3 urud giŋ₄ ti-a-ma = “Tēma copper” (obv. iv 19) This entry is found among various qualities of copper and related terms and is left without Akkadian translation. Sumerian ti-a-ma no doubt refers to the same geographical name encountered in administrative texts from the SinKašid palace in Old Babylonian Uruk (urud ti-a-ma; see Sanati-Müller 1990: no. 140; Reiter and Waetzoldt 1996: 408) and from Mari (in the form urud tema-yu; see Arkhipov 2012b: 15–16 with previous literature). The word may well refer to the oasis Tēma in Arabia, but our entry does not yield any further arguments pro or contra that identification.9

1.3.4

urud

šen (obv. iv 21–26)

The section urudšen provides various words known from copper texts from Mari. 21. 22. 23. 24. 25. 26.

urud

šen šen urud šen im-ma urud šen im-ma urud šen-dili₂ urud dilim₂ im-ma urud

šu-un-nu ru-uq-qu sa-an-ku-tu ša-pil₂-tum min (blank)

cauldron cauldron/sheet metal remainder remainder lower part

The English translations follow the Akkadian, but are far from assured. Akkadian šunnu (21) is a by-form of šennu (cauldron) and is unproblematic. The alternative rendering ruqqu (22) may mean cauldron, but is also well attested 9 Tēma appears in Neo-Assyrian and Neo-Babylonian sources and a stele of Nabonidus has been excavated at the site (Eichmann, Schaudig, and Hausleiter 2006; Schaudig 2011). Achaeological evidence suggests that the oasis already played an important role in overland trade at a much earlier date (Hausleiter 2011).

370

Niek Veldhuis

in the sense of “sheet metal.” The latter meaning may be more appropriate here because of the following item urudšen im-ma = sankuttu (23). The discussion of sankuttu in Arkhipov 2012b: 19–20 arrives at various different meanings, including “sample,” or “remainder.” The translation “remainder” is chosen here because of the association with Sumerian im-ma “last year.” Akkadian šapiltum (24–25) may be used literally in the sense of “lower part” or “base,” or may be another term for “remainder.” It is proposed here that both meanings are used in sequence, but the passage is far from clear.

1.3.5 a₂-aš-ĝar “battle ax” (obv. v 24–25) BM 85983 provides two Akkadian renderings for uruda₂-aš-ĝar (obv. v 24–25): urud

a₂-aš-ĝar a₂-aš-ĝar

urud

a₂-aš-ga₂-ru nap-ta-qu

The word a₂-aš-ĝar is known from Gilgameš and Huwawa 55 (ETCSL 1.8.1.5) and from a number of Ur III sources. The Ur III references that I am aware of include: Tab. 1: a₂-aš-ĝar in Ur III texts. AnOr 7: 315 rev. 8′ AnOr 7: 315 rev. 10′ MCS 1: 37 (BM 105452) MVN 16: 1142 SA 061 UTI 4: 2336 UET 3: 0721 Limet Métal: 286 14 UTI 6: 3529? Nisaba 15: 462

Umma Umma Umma Umma Umma Umma Ur Girsu Umma Iri-sagrig

zabar zabar zabar zabar zabar zabar uruda zabar

2 mana 2 mana 2 mana each 1 ¹⁄₃ mana 1 ⁵⁄₆ mana 2 mana

2 mana zabar

It appears that a₂-aš-ĝar is a heavy weapon or tool (2 mana or about 1 kilogram is the most common weight), mostly made of bronze, but also of copper. In preparing for the journey to the cedar forest Gilgameš had a copper a₂-aš-ĝar as well as a battle ax (urudaga-silig) cast for him. The contribution of our text is, first of all, to confirm the reading a₂-aš-ĝar (rather than a₂-sur; see already Civil 2003: 81). Akkadian a₂-aš-ga₂-ru is unlikely to be related to the word azmarû (also asmarû, zamirītum, etc.), a

Words and Grammar: Two Old Babylonian Lists

371

spear or lance (Schrakamp 2010a). This word is written in Sumerian as za-amru-tum (with many different spellings; see Michalowski 2011: 421–22, who prefers the translation “sword”). The Umma references associate a₂-aš-ĝar with common agricultural tools such as ha-bu₃-da and ha-zi-(in), rather than with weapons, which makes it likely that a₂-aš-ĝar is an agricultural tool that could double as a weapon. The second equation a₂-aš-ĝar = naptaqu (line 25) was already known from Nabnitu 7 210 f. (MSL 16, p. 111; VAT 8755 rev. i 49–50 = P368989): urud

a₂-aš-ĝar = nap-ta-q[u] tun₃ šu min₃ = min

urud

a’ašĝar = cast object double-handed ax = cast object

Akkadian naptaqu, which derives from patāqu, to cast, is only known in this Nabnitu entry and now in our text. Like Nabnitu, the Gilgameš and Huwawa story associates a₂-aš-ĝar with casting and with axes. One may entertain the possibility that the first millennium word ešmarû (ašmarû) is distantly related to a₂-aš-ga₂-ru in our text. Akkadian ešmarû refers to a quality of silver that is used in casting ornaments or in plating. If the two are related, their common aspect is the technique of casting.

1.3.6

urud

he₂-me-gi-na = maškanu “fetters” (rev. i 9)

Sumerian he₂-me-gi-na is known exclusively from lexical lists, where it is always associated with šer₃-šer₃ “chains” (see OB har-ra 2 = Q000040, 257a; unilingual Emar texts have the variant urudhe₂-niĝ₂-gi-na: Msk 731084z = P271379 rev. 3′; Msk 74171c = P250373 rev. i 1).10 The association with maškanu “fetters” is new; the Akkadian word was recently discussed by Arkhipov 2012b: 169–70.

1.3.7

urud

niĝ₂-dim₂ gul = hušû “copper scraps” (rev. i 27)

Sumerian urudniĝ₂-dim₂ gul means “broken (copper) artifacts,” which makes good sense with Akkadian hušû, “metal scraps.”11 The entry [niĝ₂-d]im₂ gul = re-šu-u₂ (Nigga Bilingual B 91; MSL 13, p. 117; YBC 13524) is a misreading for

10 The copy of Msk 74171c shows urudsiki-niĝ₂-gi-na. This fragment joins ASJ 9: 289–91; see Gantzert 2008: 141. 11 For a recent discussion of the word see Reiter 1997: 196.

372

Niek Veldhuis

hu-šu-u₂ (the sign is somewhat damaged) and the hapax rešû adj. “reckless” in CAD is to be deleted.12

1.4 Bronze and Bronze Objects 1.4.1 zag-šu-la₂ = merdītu “cultic vessel” (rev. ii 21) Akkadian merdītu is associated with a bewildering variety of Sumerian words that are not always easily compatible with the well-established meaning “an offering,” or “a ritual” that is primarily found in late texts. BM 85983 adds another entry to this mix: zag-šu-la₂ = me-er-di-tum

Since the entry is found among bronze vessels the Sumerian must be identical to zag-še₃-la₂, a well-known name for a vessel made of clay or metal that is attested in Sargonic, Ur III, and Old Babylonian administrative texts as well as in lexical sources from Old Babylonian to Hellenistic times. The word was discussed by Sallaberger 1996: 109 (and 69) and his data collection shows that the vessel had ritual and royal purposes. The present entry is to be understood as “a ritual vessel for the merdītu ceremony.” This same vessel, now with the ĝeš determinative for wooden objects is found in Neo-Assyrian Murgud, with a slightly more specific translation: ĝeš

niĝ₂-zag-še₃-la₂ = ša₂ mer-di-it hur-ri = bur-[…] (K 242 + K 4574 obv. i 9′ = P393806)13

In this entry ḫurru does not mean “hole” but rather “day-time” (urru) as the parallel from Late Babylonian Uruk makes clear: ĝeš niĝ₂-zag-še₃-la₂ = ša₂ mer-di-it ur-[ri = …] (SpTU 2: 53 rev. ii 39 = P348656)

The entry thus means (wooden) ritual vessel for the day-time merdītu ceremony.

12 I wish to thank Enrique Jiménez (Yale University) for collating the line for me and for sending me excellent photographs of the passage. 13 MSL 6, p. 78; see har-ra 6, 147. The third column word was restored burtum (well) in MSL 6, probably on the assumption that ḫurru is “hole.” Several words for “offering” begin with bur- (burgû; bursaggû, etc.).

Words and Grammar: Two Old Babylonian Lists

373

The secondary meaning of merdītu, listed in the dictionaries as “an implement for lifting or climbing” (CAD M/2), or a “Trittleiter” (AHw), does not exist. This rendering was based on the spurious Murgud entry ĝešgu₂-si-ki-ir = merdi-tum = sim-m[il-tu]m (MSL 6, p. 79). An inspection of the source (K 242+ = P393806) shows that the word in the third sub-column does indeed begin with a nam sign, but that the other two signs are faint traces. The association of merdītum or ĝešgu₂-si-ki-ir with some kind of ladder, therefore, has no basis in the lexical tradition or anywhere else. To follow this trail further we may note that in the lexical tradition the Sumerian word ĝešgu₂-si-ki-ir is also known with /b/ instead of /k/: ĝešgu₂-zibi-ir-ra = merdētum (har-ra 6, 215). Variants of this form are attested in two Ur III texts. Civil 2007: 27 drew attention to ĝešgud ze₂-be₂-er in TPTS 1: 248 (P126937); still another variant is found in the Garšana text CUSAS 3: 1269 (P324074): urudgud zi-bi-ir, weighing 1/2 pound (mana) and one shekel. These Ur III references indicate that the form with middle /b/ or /p/ (rather than /k/) is the earlier one, but a clear (Akkadian) etymology remains difficult. The meaning “step ladder,” suggested in CUSAS 4, p. 69, and no doubt based on the same Murgud entry discussed above, may now be rejected. The name of the merdītu ritual (Linssen 2003: 162) is derived from the verb redû (to lead animals); the ritual consists of an offering of sheep or oxen and may include a libation. The Sumerian words discussed above may all refer to various objects in wood, copper, or bronze, that were used during the ceremony, either for leading or restraining the animals or for libation, and are thus associated with merdītu in the lexical tradition. This certainly does not imply that the Ur III words urudgud zi-bi-ir and ĝešgud ze₂-be₂-er necessarily refer to such a ritual since the objects may well have had a more general use and the relevant lexical entries are many centuries later than the administrative texts.

1.4.2 bulug-kin-gur₄ = marṣadu = lancet (rev. iii 1) Sumerian bulug-kin-gur₄, “lancet,” is known from literary sources (IddinDagan D 8 = ETCSL 2.5.3.4; and Nin-Isina A 10 = ETCSL 4.22.1), in both cases associated with ĝiri₂-zal “scalpel.” The present entry clarifies the word marṣadu, which remained undefined in CAD. In the first millennium lexical tradition Akkadian marṣadu is equated with kin-gurum-mezabar (SpTU 3: 111 = P274484 obv. ii 21), or [urudkin-gurum]-ma in W 23830d (BaM 10, 129). The word may well be identical with masdaru, most recently discussed by Stol 2007: 238 as a kind of knife used in medical contexts and associated with a scalpel (karzillu).

374

Niek Veldhuis

2 gub = izuzzu Paradigm Among a group of scholarly tablets (divinatory, literary, and lexical) from the Sealand dynasty14 and dated to the early sixteenth century, is a new bilingual paradigm of the verb gub = izuzzu “to stand.” The tablet is available in excellent black-and-white photographs at http://cdli.ucla.edu/P431315. Its present whereabouts are unknown to me and I have thus not been able to collate the original. The tablet exhibits some of the typical sign forms of the Sealand dynasty texts, in particular uz (see e.g., rev. 10) and gi (rev. 23–30).15 It is likely that the tablet originally had four columns (two on each side), of which the columns 1 and 4 are largely preserved. Parts of the text are damaged and in a few places the photographs are difficult to read. Thanks to the systematic organization of the paradigm, most of the text in columns 1 and 4 can be reconstructed. The edition of the text is presented in 2.4 and may also be consulted at http://oracc.org/dcclt/P431315.

2.1 Overview of the Paradigm The paradigm is divided into sections of 7 or 8 lines each, indicated by horizontal rulings on the tablet. Within each section subject, mode, and main stem are held constant, but the Akkadian forms vary in tense (present, preterit), derived stem (main stem, /t/ infix, /tan/ infix), and presence or absence of a first person pronominal suffix. Some sections add an extra modal or imperative form that does not correspond to forms in other sections. The new paradigm only has two Akkadian forms in common with the other gub = izuzzu paradigm, OBGT X (= A 24186; MSL 4, pp. 111–13), and those forms are not associated with the same Sumerian forms: OBGT X 27. bi₂-in-gub 28. bi₂-ib₂-gub 29. mi-ni-ib₂-gub 69.

uš-zi-iz uš-zi-iz uš-zi-iz

bi₂-ib₂-gub-be₂ uš-za-a-az

P43131516 o 7 [in-ne]-en-gub

uz-zi-iz

o5

uz-za-az

[…]-gub-bu-nam

14 George 2013: 129–31; see also Gabbay 2014. 15 See Gabbay 2014: 148; George 2013: 131–32. 16 For the assimilation of the prefixed /š/ to /z/ in uzzīz < ušzīz etc., see 2.2.1.

Words and Grammar: Two Old Babylonian Lists

375

This lack of overlap is not surprising, because OBGT X follows a very different pattern and does not deal with plural forms, whereas the preserved part of the new paradigm has singular forms only in obv. 1–8.17 Table 2 illustrates the preserved scope of the paradigm by collecting one form from each section, namely the main stem in the present tense without first person pronominal suffix. Tab. 2: Sections of the gub = izuzzu Paradigm. §

Sumerian

Akkadian

 1

[…-g]ub-bu-nam

uz-za-az

 2

mi-ni-ib₂-gub-bu-un-ne uz-za-az-zu

 3

nam-mi-ni-ib₂-gub-buun-ne nu-mu-ni-i[b₂-gub-buun-ne] u₃-mu-[…]-gub-bu-unne

 4  5

[la uz-za-az-zu] [u₂-ul uz-za-az-zu] [uz-za-a]z-zu-ma

he makes it stand Š 3rd sing. indicative18 they make it Š 3rd pl. stand indicative they shall not Š 3rd pl. vetitive make it stand they do not make Š 3rd pl. negative it stand they make it Š 3rd pl. stand and … indicative +-ma

approximately 5 sections missing 11 12 13

mi-[…] nam-[…] [nu-…]

[…] […] […]

… indicative … vetitive [… negative]

1 or 2 more sections missing 15 16 17 18

mi-ni-ib₂-gub-bu-unze₂-en nam-mi-ni-ib₂-gub-buun-ze₂-en nu-mi-ni-ib₂-gub-buun-ze₂-en nu-uš ma?-sug₂-ge-enze₂-en

tu-uz-za-az-za la tu-uz-za-az-za u₂-ul tu-uz-za-az-za lu-ma-an ta-az-za-azza

you (pl.) make it stand you (pl.) shall not make it stand you (pl.) do not make it stand if only you (pl.) would stand

Š 2nd pl. indicative Š 2nd pl. vetitive Š 2nd pl. negative N 2nd pl. irrealis

17 Note that uš-zi-iz-zu in OBGT X 37 is third person singular with pronominal suffix and is thus not comparable with uz-zi-zu in obv. 15 (third person plural). 18 Or, possibly, 1st singular, although a 1st sing. causative with 1st accusative suffix (such as o 2 uštanazzazzanni “I keep making myself stand”) seems rather far-fetched. The Sumerian half of this first section is badly damaged and what remains is irregular in any interpretation.

376

Niek Veldhuis

The overview shows that most of the preserved sections are in the plural (2nd or 3rd person), with the exception of § 1 (3rd singular). §§ 1–5 and §§ 15–17 list transitive forms (Š stem in Akkadian); § 18 has intransitive forms (N stem). The sections 2–4 (3rd plural), 11–13 (almost entirely broken), and 15–17 (2nd plural) list basically the same forms in indicative, vetitive and negative, and thus allow for restorations of broken sections. Section 1, which is heavily damaged on the Sumerian side, may not so easily be restored, and some of the forms (or what is left of them) remain rather mysterious (in the form cited above, the role of -nam, presumably /n/ plus the copula, remains unclear). The last section on the obverse (§ 5) is equally difficult to restore in its entirety. All Akkadian forms have the -ma suffix, corresponding in some entries with the Sumerian prefix of anteriority u₃- (Civil 2008b). Section 18, finally, has the irrealis particle lūman in all Akkadian forms, corresponding to Sumerian nu-uš. The structure of each section is very straightforward and regular on the Akkadian side (obv. 9–16 = section 2: “they made me/it stand;” all forms 3rd plural Š): Tab. 3: Section 2 of the gub = izuzzu paradigm. tense/mode he₂-de₃-ib₂-gub-bu-un-ne mu-ni-ib₂-gub-bu-un-ne mu-un-ne-da-gub-be₂-en ba-an-ne-da-gub-be₂-en mi-ni-ib₂-gub-bu-un-ne mi-ni-ib₂-gub-gub-bu-un-ne in-ne-en-gub-be₂-e-eš ba-an-ne-en-gub-be₂-e-eš

lizzīzūninni uzzazzūninni uzzīzūninni uštazizzūninni uzzazzū uštanazzazzū uzzīzū uštazizzū

precative present preterit preterit present present preterit preterit

infix

t tan t

suffix 1st 1st 1st 1st ø ø ø ø

acc acc acc acc

Although the sections vary slightly in the forms that they admit, they all create a basic contrast between forms with and without a first person pronominal suffix (accusative in §§ 1–5 and 15–17, and dative in § 18). In the Sumerian this contrast corresponds with the distinction between a comitative and a /ni/ /nne/ infix. In a more abstract form the structure of each section may be summarized as follows:

Words and Grammar: Two Old Babylonian Lists

377

Tab. 4: Morphological analysis of a section. Conj. Pref

Infix

base

tense

infix

suffix

mumubamumui₃ba-

comitative comitative comitative ni ni nne nne

marû ḫamṭu ḫamṭu marû redupl. marû ḫamṭu ḫamṭu

present preterit preterit present present preterit preterit

ø ø t ø tan ø t

1st person 1st person 1st person ø ø ø ø

The preserved traces in reverse column 1 follow essentially the same pattern.

2.2 Orthographic and Morphological Peculiarities The Akkadian of our paradigm is very systematic and mostly conforms to what is known about Akkadian grammar and the verb izuzzu (Huehnergard 2002). A few remarkable aspects are discussed in 2.2.1. The Sumerian, on the other hand, contains many rare and irregular forms, that may throw some light on how Sumerian verbal forms were analyzed (2.2.2). Many aspects of the Sumerian in this paradigm will require further study.

2.2.1 Akkadian Orthography Most preserved Akkadian forms derive from the transitive Š stem, translated here, perhaps awkwardly, “to make stand.” Preformative /š/ assimilates to a following /z/, as in: uzzīzū (< *ušzīzū) Such forms, which superficially look like D stems, have been noted before and were discussed by Kouwenberg (2010: 490). There is one exception: rev. 16 has u₂-ul tu-uš-za-az-za-ni-in-ni, without assimilation. The final root consonant of izuzzu is usually reduplicated before a vowel (see Huehnergard 2002). Our paradigm follows that rule, except in the preterit of the main stem (obv. 13–16): uzzazzū uštanazzazzū

they make it stand they keep making it stand

Š present Štn present

378

uzzīzū uštazizzū

Niek Veldhuis

they made it stand they made it stand up

Š preterit Št preterit

Why the preterit of the main stem does not reduplicate the final /z/ remains unclear to me but the pattern is followed throughout the paradigm (uzzīzanni; uzzīzūninni; uzzīzūma; tuzzīzā; tazzīzānim; etc.). Otherwise, the only irregularity in the Akkadian is the combination of modal lā with a preterit in obv. 19 (lā uzzīzūninni) and elsewhere. These forms are produced by the systematicity (on the Akkadian side) of the paradigm, where every form in the sections 2 and 15 receives vetitive and negative forms in the sections 3–4 and 16–17, respectively.

2.2.2 Sumerian Orthography and Morphology Sumerian morphology in the paradigm usually corresponds in a systematic way with Akkadian morphology. Akkadian present tense, for instance, always corresponds to Sumerian marû and the presence of an accusative suffix in the Akkadian always corresponds to a comitative infix in the Sumerian. We may thus study the Sumerian morphology by investigating the contrasts that the paradigm creates between the Akkadian forms of a single section, or between corresponding Akkadian forms of different sections. Table 2, above, shows that the preserved sections of the paradigm produce permutations of the following (Akkadian) forms: Obverse column 1: §  1 Š 3rd singular §  2 Š 3rd plural §  3 Š 3rd plural vetitive §  4 Š 3rd plural negative §  5 Š 3rd plural + -ma Reverse column 2: § 15 Š 2nd plural § 16 Š 2nd plural vetitive § 17 Š 2nd plural negative § 18 N 2nd plural irrealis The most immediate contrast that comes out of this arrangement is between 3rd plural (§§ 2–5) and 2nd plural forms (§§ 15–18; 2.2.2.1). The vetitive and nega-

Words and Grammar: Two Old Babylonian Lists

379

tive forms are mostly regular and require little comment (2.2.2.2). The -ma forms (§ 5) and the irrealis (§ 18) are briefly discussed in 2.2.2.3. How the contrast between intransitive N (§ 18) and transitive Š (all other sections) is represented in Sumerian is addressed in 2.2.2.4. Akkadian tenses are regularly represented in Sumerian by a combination of bases (marû vs. ḫamṭu) and conjugation prefixes (2.2.2.5). The presence or absence of an accusative suffix in the Akkadian is represented by comitative vs. /ni/ /nne/ infix in Sumerian (2.2.2.6).

2.2.2.1 Agreement on the Verb: 2nd and 3rd plural The agreement markers on the verb are written: V-en .n-V-en-ze₂-en V-un-ze₂-en .n-V-e-eš V-un-ne

1st singular 2nd plural (ḫamṭu) 2nd plural (marû) 3rd plural (ḫamṭu) 3rd plural (marû)

The writing -un-ne with reduplicated /n/ is unusual, but is also found before the comitative in this paradigm (see below). Third person plural forms in the ḫamṭu with a first person object (absolutive) add the first person -en suffix and displace the third person plural marker to the comitative. Contrast the forms that correspond to Akkadian preterit without and with the first person accusative suffix (o 15 and 11): in-ne-en-gub-be₂-e-eš mu-un-ne-da-gub-be₂-en

uzzīzū uzzīzūninni

they made it stand they made me stand

The first form has the regular third person plural marker (ḫamṭu) .n-V-e-eš. In the second form the first person absolutive suffix occupies the post-verbal slot and the third person plural marker is moved to the prefix chain where it is represented by the .nne- before the comitative -da- (for the comitative see 2.2.2.6). Second person plural agreement markers work somewhat differently. First, unlike 3rd plural, these markers are essentially the same for ḫamṭu and marû.19

19 According to Jagersma 2010: 350 the suffix -enzen represents the second person plural in both absolutive and ergative in imperfect as well as perfect in third millennium Sumerian.

380

Niek Veldhuis

It is likely that ḫamṭu forms take an .n- prefix before the verbal base, as in inne-en-gub-be₂-en-ze₂-en = tuzzīzā (rev. 7), presumably in analogy with the third person plural forms.20 Second person plural forms (ḫamṭu) always keep the -enzen suffix, whether or not there is a first person object. The second person plural forms corresponding to those quoted above are (rev. 7, 3): in-ne-en-gub-be₂-en-ze₂-en mu-un-da-gub-be₂-en-ze₂-en

tuzzīzā tuzzīzāninni

you (pl.) made it stand you (pl.) made me stand

The analogous third person plural form lost both the preverbal .n- and the plural suffix -e-eš in order to make place for the first person absolutive suffix. The second person plural forms, however, only loose the preverbal .n- and do not seem to represent the first person absolutive at all.

2.2.2.2 Negative, Vetitive The negative nu- in the sections 4 and 17 is prefixed to the forms in the sections 2 and 15, respectively, with the regular transformation of nu-ba- to la-ba-. In a few cases the form is slightly altered, but the significance of those changes remains difficult to gauge, for instance (obv. 11, 27): mu-un-ne-da-gub-be₂-en nu-un-ne-da-[gub-be₂-en]

uzzīzūninni they made me stand [ul uzzīzūninni] they did not make me stand

Obv. 27 appears in a sequence of nu-un- forms where one would expect numu-un-. Similarly, the vetitive entries regularly add nam- as a prefix to the indicative Sumerian form; for instance (obv. 13, 21): mi-ni-ib₂-gub-bu-un-ne nam-mi-ni-ib₂-˹gub-bu-un-ne˺

uzzazzū [lā uzzazzū]

they make it stand they shall not make it stand

In a few cases this procedure results in unusual forms, for instance (rev. 7, 14): in-ne-˹en˺-gub-be₂-en-ze₂-en nam-in-ne-en-gub-be-en-ze₂-en

tuzzīzā lā tuzzīzā

you (pl.) made it stand you (pl.) should not make it stand

20 Also ba-an-ne-en-gub-be₂-en-ze₂-en = tuštazizzā (rev. 8) and the corresponding vetitive and negative forms.

Words and Grammar: Two Old Babylonian Lists

381

An irregular entry is obverse 24, which is the only form preserved that has the negative modal ba-ra- instead of nam-. The corresponding forms in sections 2, 3, and 4 are (obv. 16, 24, and 32): ba-an-ne-en-gub-be₂-e-eš ba-ra-an-ne-en-gub-[be₂-e-eš]

uštazizzū [lā uštazizzū]

la-ba-an-[ne-en-gub-be₂-e-eš]

[ul uštazizzū]

they made it stand up they should not make it stand up they did not make it stand up

2.2.2.3 Anteriority and Irrealis Third person and second person plural are both represented by four sections: indicative, vetitive, negative, plus an additional section. The additional section for third person plural (§ 5) has -ma forms in the Akkadian column. The additional section for the second person plural has irrealis forms (lūman) of the N stem (§ 18; see also 2.2.2.4). Section 5 is, unfortunately, rather damaged. It seems to follow the established pattern rather closely, probably beginning with a modal form, now lost. As in all other sections, the first half is concerned with forms that have a first person pronominal suffix in the Akkadian, followed by forms that do not have such a suffix. Those latter forms correspond to u₃- forms in Sumerian, representing the prefix of anteriority (Civil 2008b). The initial four forms of § 5 do not begin with u₃-; how they represent Akkadian -ma remains unclear because of damage to the text. Section 18 is also damaged, but enough is preserved to suggest that Akkadian lūman (“if only“), systematically corresponds to nu-uš in Sumerian (Civil 2005: 39–40).

2.2.2.4 Transitive vs. Intransitive; gub vs. sug₂ Almost all Akkadian forms in the paradigm are in the (transitive) Š stem, except for § 18, which has (intransitive) N forms of the 2nd person plural with the lūman particle (2.2.2.3). Accordingly, instead of first person accusative suffixes (as in §§ 1–5 and 15–17), the first four forms of § 18 have first person dative suffixes, represented by the (second person!) -ra- infix in the Sumerian.21 The second person plural intransitive subject (absolutive) triggers the plural form of the verb gub, which is sug₂. This is in accordance with the current

21 The Sumerian in this section is badly broken, but what is preserved suggests that the infix -ra- is understood as a generalized dative of any person.

382

Niek Veldhuis

understanding of Sumerian suppletive plural stems (Steinkeller 1979; Michalowski 2004), which agree with the plurality of the absolutive (in the transitive forms in the sections 2–5 and 15–17 the plural subject of the Akkadian corresponds to a plural of the ergative in Sumerian and thus does not trigger the plural stem).

2.2.2.5 Bases and Conjugation Prefixes Akkadian tenses (preterit, present) and the derived stems with -t- and -taninfix are regularly represented by a combination of conjugation prefixes (mu-, ba-, and i₃-) and bases (hamṭu, marû, and reduplicated marû), as follows: mu-, marû mu-, ḫamṭu i₃-, ḫamṭu ba-, ḫamṭu mu-, reduplicated marû

present preterit preterit preterit + t infix present + tan infix

The table shows that there are two ways to represent the Akkadian preterit in Sumerian: with mu-, ḫamṭu or with i₃-, ḫamṭu. The mu- forms are used when a first person accusative suffix is present (corresponding to a comitative in the Sumerian; see 2.2.2.6); the i₃- forms when the accusative suffix is absent. This works well with a view of mu- emphasizing prototypical transitivity, agency, and animacy (see Woods 2008; Delnero 2010). Marû forms in this paradigm are always combined with the mu- prefix.22 The combination mu- + -ni- is usually written mi-ni-, but note nu-mu-ni-ib₂[gub-bu-un-ne], etc. in obv. 29–30.23 The prefix i₃- appears once in each section, always in the form in-ne- and with the ḫamṭu base. This corresponds to a preterit form in Akkadian without infixes or suffixes, phonologically the lightest form available in each section. The correspondence between the conjugation prefix ba- and Akkadian -tinfix is well known from other Old Babylonian paradigm texts.24 There is one exception; the form tattazizzā[nim] (r ii 26) does not correspond to a ba- form in the Sumerian column.

22 See also Delnero 2010. 23 The form mu-ni-ib₂-gub-bu-un-ne in obv. 10 is exceptional (and probably erroneous) in several regards; see 2.2.2.6. For mu-ni- and mi-ni-, see Delnero 2012. 24 See Black 1991: 27–28; Huber 2007: 8.

Words and Grammar: Two Old Babylonian Lists

383

The distinction between the ḫamṭu and marû bases of the verb is expressed by gub (perfect/ḫamṭu) vs. gub-bu (imperfect/marû). This results in forms such as (rev. 2–3) mu-un-de₃-ib₂-gub-bu-un-ze₂-en tuzzazzāninni mu-un-da-gub-be₂-en-ze₂-en tuzzīzāninni

you (pl) make me stand you (pl) made me stand

Reduplicated forms of gub are always marû (gub-gub-bu), have the conjugation prefix mu- and correspond to a present tense form with -tan- infix in the Akkadian (obv. 14):25 mi-ni-ib₂-gub-gub-bu-un-ne

uštanazzazzū

they keep making it stand

2.2.2.6 Comitative and /ni/-/nne/ infix The main contrast within each section is between forms with first person pronominal suffix (the first 3 or 4 entries) vs. forms without such a suffix. The suffixed entries always have a comitative -da- (or -de₃-) in the Sumerian, the non-suffixed ones have the infix -ni- (or .nne-). This is illustrated by § 16 (rev. 9–15): Tab. 5: Section 16. nam-mu-un-de₃-gub-bu-un-ze₂-en

lā tuzzazzāninni

nam-mu-un-da-gub-be₂-un-ze₂-en26

lā tuzzīzāninni

nam-ma-e-da-gub-be₂-un-ze₂-en

lā tuštazizzāninni

nam-mi-ni-ib₂-gub-bu-un-ze₂-en nam-mi-ni-ib₂-gub-gub-bu-un-ze₂-en

lā tuzzazzā-ø lā tuštanazzazzā-ø

nam-in-ne-en-gub-be-en-ze₂-en

lā tuzzīzā-ø

nam-[ba]-an-ne-en-gub-be₂-en-ze₂-en

lā tuštaziz[zā-ø]

you (pl.) shall not make me stand you (pl.) should not make me stand you (pl.) should not make me stand up you (pl.) shall not make it stand you (pl.) shall not keep making it stand you (pl.) should not make it stand you (pl.) should not make it stand up

25 The exception may be obv. 2, where a form ending in […-e]n-nam (presumably ḫamṭu) corresponds to present tense uštanazzazzanni (“he keeps making me stand”). The expected Sumerian form here is …-gub-gub-bu-nam, but too much is broken in this section to be confident about any reconstruction – let alone emendation. 26 In this line and the next -un- is a mistake for -en- (*nam-mu-un-da-gub-be₂-en-ze₂-en).

384

Niek Veldhuis

Because of the consistency of this pattern throughout the paradigm, obv. 10: mu-ni-ib₂-gub-bu-un-ne = uzzazzūninni is exposed as an error (Sumerian should be mu-un-de₃-ib₂-gub-bu-un-ne or perhaps mu-un-de₃-gub-bu-unne). The comitative has the forms -da- and -de₃-. The writing -de₃- is found almost exclusively with Akkadian present tense forms with first person accusative suffix and we may thus conclude that -de₃- somehow represents the comitative infix plus the first person accusative (marû). This works well with the form nam-mu-un-de₃-gub-bu-un-ze₂-en, quoted above, where the first person accusative marker (marû), corresponds to the inanimate accusative (marû) pre-verbal /b/ in forms like nam-mi-ni-ib₂-gub-bu-un-ze₂-en, associated with Akkadian present tense forms without first person accusative. Unfortunately, there are quite a few forms which have the sequence -de₃-ib₂-Vmarû, as in rev. 2: mu-un-de₃-ib₂-gub-bu-un-ze₂-en tuzzazzāninni

you (pl.) make me stand

It is not clear what the pre-verbal /b/ in these forms could refer to or how comitative -da- could have changed to -de₃-. The following such forms are found: he₂-de₃-ib₂-gub-bu-un-ne nam-de₃-ib₂-gub-bu-un-ne nam-de₃-ib₂-gub-gub-bu-un-ne nu-un-de₃-ib₂-˹gub˺-[bu-un-ne] nu-un-de₃-ib₂-gub-[gub-bu-un-ne]

˹he₂?-en˺-de₃-ib₂-gub-bu-un-ze₂-en mu-un-de₃-ib₂-gub-bu-un-ze₂-en nam-de₃-ib₂-gub-gub-bu-un-ne nam-mu-un-de₃-gub-bu-un-ze₂-en nu-mu-un-de₃-gub-bu-un-ze₂-en

The usage of the comitative to reflect an Akkadian accusative suffix was briefly discussed by Civil 2004: 209 with reference to the paradigm of the verb ĝar in OBGT VI (A 24187; MSL 4, pp. 79–87). This same usage of the comitative is also attested in the gub = izuzzu paradigm in OBGT X (A 24186; MSL 4, pp. 111–13)27 and in our present text. The infix -ni- or .nne- is written ni in mu-ni- or mi-ni- and as ne in inne- and ba-an-ne-. The forms in-ne- and ba-an-ne- are superficially similar to in-ne- and ba-ne- in OBGT VII (ĝen = alāku; A 24189), where they are also found together (MSL 4, p. 97: 268–273). In OBGT VII, however, those forms must be interpreted as third person plural datives (corresponding to Akkadian šunūši), an interpretation that is not available here. The infix -ni- is quite frequently used to represent the causative Š stem in verbal paradigms (see

27 See Black 1991: 20, 38; Huber 2007: 12.

Words and Grammar: Two Old Babylonian Lists

385

Black 1991: 30–32); it is analyzed as the “Oblique Object of the third person singular human” prefix in Jagersma 2010: 421–23.

2.3 Conclusions The new gub = izuzzu paradigm is in many ways unusual, with an organization that is unlike any other such text known so far. At the same time, it exhibits many of the same systematic correspondences that are attested elsewhere. The Sumerian conjugation prefix ba-, for instance, is regularly associated with the Akkadian -t- infix and the usage of the comitative, discussed above (2.2.2.6) corresponds to well-known patterns in other such texts. Old Babylonian scholars shared the knowledge of these Sumerian – Akkadian morphological associations – even if they occasionally seem rather odd to us. This shared procedural knowledge did not lead to the redaction of standardized texts: each verbal paradigm is essentially a new manifestation of this common knowledge. This same pattern is also found in mathematical problem texts, divinatory compendia, and Akkadian literary texts of the Old Babylonian period. One aspect of the text, not discussed so far, relates to the interface between morphology and writing. Throughout the text an important role is played by the sign ne, read ne or de₃, and this sign is responsible for quite a few of the unusual forms discussed above. Taken in isolation, forms like he₂-de₃-ib₂gub-bu-un-ne (obv. 9), nam-de₃-ib₂-gub-gub-bu-un-ne (obv. 18), or nu-muun-de₃-gub-bu-un-ze₂-en (rev. 16) might better be read he₂-bi₂-ib₂-gub-buun-ne, nam-bi₂-ib₂-gub-gub-bu-un-ne, and nu-mu-un-ne-gub-bu-un-ze₂en, respectively. It is only because of the strong correlation in this text between accusative suffix in Akkadian and comitative infix in Sumerian that we are forced to read de₃. It is possible that the ambivalence of the writing system was used here with some perverse intentionality – or at least, I hope so.

2.4 P431315: Edition The translation below follows the same somewhat rigid patterns that the Akkadian text uses. Akkadian present tense is translated by English present tense, preterit by past tense. The Š of izuzzu is translated “to make stand” the ingressive28 Št “to make stand up” and the iterative Štn “to keep making stand.” All this is, of course, rather artificial – and so is the text itself.

28 For the ingressive meaning of the /t/ infix, see Kouwenberg 2005, 2010.

386

Niek Veldhuis

Obverse column 1 § 1: 1 […] 2 […-gub-gub-be₂]-˹en˺nam 3 […]-˹gub˺-be₂ 4 […]-gub-be₂-en-nam 5 […]-˹gub˺-bu-x-nam 6 […]-˹gub˺-gub-bu-nam 7 [in-ne]-en-gub 8 [ba-an-ne]-en-gub-am₃ § 2: 9 10 11 12 13 14 15 16 § 3:17 18 19 20 21 22 23 24

§ 4:25 26 27

[uz-za]-az-za-an-ni uš-ta-na-az-za-azza-an-ni uz-zi-za-an-ni uš-ta-zi-iz-za-an-ni uz-za-az uš-ta-na-az-za-az uz-zi-iz uš-ta-zi-iz

he₂-de₃-ib₂-gub-bu-un-ne mu-ni-ib₂-gub-bu-unne29 mu-un-ne-da-gub-be₂-en ba-an-ne-da-gub-be₂-en

li-iz-zi-zu-ni-in-[ni] uz-za-az-zu-ni-in[ni] uz-zi-zu-ni-in-[ni] uš-ta-zi-iz-zu-ni-in[ni] mi-ni-ib₂-gub-bu-un-ne uz-za-az-˹zu˺ mi-ni-ib₂-gub-gub-bu-un- uš-ta-na-az-za-az˹zu˺ ne in-ne-en-gub-be₂-e-eš uz-zi-˹zu˺ ba-an-ne-en-gub-be₂-e-eš uš-ta-zi-iz-˹zu˺ nam-de₃-ib₂-gub-bu-unne nam-de₃-ib₂-gub-gub-buun-ne nam-mu-un-ne-da-gubbe₂-en nam-ba-an-ne-da-gubbe₂-en nam-mi-ni-ib₂-˹gub-buun-ne˺ nam-mi-ni-ib₂-gub-[gubbu-un-ne] nam-in-ne-en-gub-[be₂-eeš] ba-ra-an-ne-en-gub-[be₂e-eš]

la uz-za-az-zu-ni˹in˺-[ni] la uš-ta-na-az-za˹az˺-zu-[ni-in-ni] la ˹uz-zi-zu-ni˺-[inni] [la uš-ta-zi]-˹iz-zuni-in˺-[ni] […] […]

he makes me stand he keeps making me stand he made me stand he made me stand up he makes it stand he keeps making it stand he made it stand he made it stand up may they make me stand they make me stand they made me stand they made me stand up they make it stand they keep making it stand they made it stand they made it stand up they shall not make me stand they shall not keep making me stand they should not make me stand they should not make me stand up they shall not make it stand

[…]

they shall not keep making it stand they should not make it stand

[…]

they should not make it stand up

nu-un-de₃-ib₂-˹gub˺-[bu- […] un-ne] nu-un-de₃-ib₂-gub-[gub- […] bu-un-ne] nu-un-ne-da-[gub-be₂-en] […]

29 Likely an error for mu-un-de3-ib2-gub-bu-un-ne.

they do not make me stand they do not keep making me stand they did not make me stand

Words and Grammar: Two Old Babylonian Lists

28 29 30 31 32

§ 5:33 34

la-ba-an-ne-˹da˺-[gub50zz1be₂-en] nu-mu-ni-˹ib₂˺-[gub-buun-ne] nu-mu-ni-˹ib₂˺-[gub-gubbu-un-ne] nu-un-ne-[en-gub-be₂-eeš] la-ba-an-[ne-en-gub-be₂e-eš] e-mu-un-[…] mu-˹un?-de₃˺-[…]

35

mu-un-[…]-˹gub-be₂˺-[en]

36

ba-an-[…]-gub-be₂-˹en˺

37

u₃-mu-[…]-˹gub-bu˺-un˹ne˺ u₃-[mu-…]-˹gub˺-gub-unne u₃-[…-gub]-be₂-e-eš u₃-x-[…]-gub-be₂-eš

38 39 40

387

[…]

they did not make me stand up

[…]

they do not make it stand

[…]

they do not keep making it stand

[…]

they did not make it stand

[…]

they did not made it stand up

[…]30 [uz-za-az-z]u-ni-inni-ma [uz-zi-z]u-ni-in-nima [uš-ta-zi]-˹iz-zu˺-niin-ni-˹ma˺ [uz-za]-˹az?˺-zu-ma uš-ta-na-az-za-azzu-ma uz-zi-zu-ma uš-ta-zi-iz-zu-ma

they make me stand and … they made me stand and … they made me stand up and … they make it stand and … they keep making it stand and … they made it stand and … they made it stand up and …

Obverse column 2 §§ 6–10 broken Reverse column 1 § 11: 1 x-[…]  2 mu-[…]  3 ma?-[…]  4 ba-[…]  5 mi-[…]  6 mi-[…]  7 in?-[…]  8 ba-[…] § 12: 9 10 11 12

nam-[…] nam-[…] nam-[…] nam-[…]

30 Most likely to be restored as lizzīzūninnima “may they make me stand and …”. The Sumerian is unclear.

388

Niek Veldhuis

13 nam-[…] §§ 13–14 broken Reverse column 2 ˹he₂?-en˺-de₃-ib₂-gub-bu§ 15: 1 un-ze₂-en  2 mu-un-de₃-ib₂-gub-buun-ze₂-en  3 mu-un-da-gub-be₂-enze₂-en  4 ba-e-da-gub-be₂-en-ze₂en  5 mi-ni-ib₂-gub-bu-un-ze₂en  6 mi-ni-ib₂-˹gub˺-gub-buun-ze₂-en  7 in-ne-˹en˺-gub-be₂-enze₂-en  8 ba-an-ne-˹en˺-gub-be₂en-ze₂-en § 16: 9 10 11 12 13 14 15

§ 17:16 17 18

nam-mu-un-de₃-gub-buun-ze₂-en nam-mu-un-da-gub-be₂un-ze₂-en31 nam-ma-e-da-gub-be₂un-ze₂-en32 nam-mi-ni-ib₂-gub-buun-ze₂-en nam-mi-ni-ib₂-gub-gubbu-un-ze₂-en nam-in-ne-en-gub-be-enze₂-en nam-[ba]-an-ne-en-gubbe₂-en-ze₂-en nu-mu-un-de₃-gub!-buun-ze₂?-en nu-mu-un-da-gub-be₂en-ze₂-en la-ba-e-da-gub-be₂-enze₂-en

šu-zi-iz-za-ni-in-ni

make me stand (pl.)!

tu-uz-za-az-za-niin-ni tu-uz-zi-za-ni-in-ni

you (pl.) make me stand

tu-uš-ta-zi-iz-za-niin-ni tu-uz-za-az-za

you (pl.) made me stand you (pl.) made me stand up you (pl.) make it stand you (pl.) keep making it stand

tu-uš-ta-na-az-zaaz-za tu-uz-zi-za

you (pl.) made it stand

tu-uš-ta-zi-iz-za

you (pl.) made it stand up

la tu-uz-za-az-zani-in-ni la tu-uz-zi-za-ni-inni la tu-uš-ta-zi-iz-zani-in-ni la tu-uz-za-az-za

you (pl.) shall not make me stand you (pl.) should not make me stand you (pl.) should not make me stand up you (pl.) shall not make it stand

la tu-uš-ta-na-azza-az-za la tu-uz-zi-za

you (pl.) shall not keep making it stand you (pl.) should not make it stand

la tu-uš-ta-zi-iz33

you (pl.) should not make it stand up

u₂-ul tu-uš-za-azyou (pl.) do not make me stand za-ni-in-ni u₂-ul tu-uz-zi-za-ni- you (pl.) did not make me stand in-ni u₂-ul tu-uš-ta-zi-iz- you (pl.) did not make me stand za-ni-in-ni up

31 Error for nam-mu-un-da-gub-be₂-en-ze₂-en. 32 Error for nam-ma-e-da-gub-be₂-en-ze₂-en. 33 There seems to be no space for the final -za.

Words and Grammar: Two Old Babylonian Lists

19 20 21 22

§ 18:23 24

25 26 27 28

29 30

nu-mi-ni-ib₂-gub-bu-unze₂-en nu-mi-ni-ib₂-gub-gub-buun-ze₂-en nu-[in]-ne-en-gub-be₂-enze₂-en [la-ba]-˹an˺-ne-en-gubbe₂-en-ze₂-en [nu-uš x-r]a-ab-sug₂-geen-ze₂-en [nu-uš x]-ra-ab-sug₂sug₂-ge-en-ze₂-en [nu-uš x]-da-sug₂-ge-enze₂-en ˹nu-uš˺ mu-da-sug₂-geen-ze₂-en nu-uš ma?-sug₂-ge-enze₂-en nu-uš ma?-sug₂-sug₂-geen-ze₂-en ˹nu-uš in˺-sug₂-ge-enze₂-en [nu-uš b]a-sug₂-˹ge˺-[enze₂-en]

389

u₂-ul tu-uz-za-az-za you (pl.) do not make it stand u₂-ul tu-uš-ta-naaz-za-az-za u₂-ul tu-uz-zi-za

you (pl.) do not keep making it stand you (pl.) did not make it stand

u₂-ul tu-uš-ta-zi-izza

you (pl.) did not make it stand up

lu-ma-an ta-az-zaaz-za-˹nim˺ lu-ma-an ta-at-tana-az-za-az-za[nim] lu-ma-an ta-az-ziza-[nim] lu-ma-an ta-at-tazi-˹iz-za˺-[nim] lu-ma-an ta-˹az˺za-˹az-za˺ lu-ma-an ta-at-tana-˹az-za-az˺[za] lu-ma-an ta-az˹zi?˺-[za] lu-ma-an ta-at-ta˹zi˺-[iz-za]

if only you (pl.) stand with me if only you (pl.) keep standing with me if only you (pl.) stood with me if only you (pl.) stood up with me if only you (pl.) stand if only you (pl.) keep standing

if only you (pl.) would stand if only you (pl.) would stand up

390

Niek Veldhuis

Plates

Fig. 1: BM 85983 (P247857) obverse. Published by permission of the Trustees of the British Museum.

Words and Grammar: Two Old Babylonian Lists

391

Fig. 2: BM 85983 (P247857) reverse. Published by permission of the Trustees of the British Museum.

392

Niek Veldhuis

Fig. 3: BM 85983 (P247857) obverse. © British Museum.

Words and Grammar: Two Old Babylonian Lists

Fig. 4: BM 85983 (P247857) reverse. © British Museum.

Fig. 5: BM 85983 (P247857) lower edge. © British Museum.

393

394

Niek Veldhuis

Bibliography Arkhipov, Ilya. 2012a. Les véhicules terrestres dans les textes de Mari, II: le harnachement et l’équipement de chars en cuir et en tissu. Babel und Bibel 6: 5–14. Arkhipov, Ilya. 2012b. Le vocabulaire de la métallurgie et la nomenclature des objets en métal dans les textes de Mari. Matériaux pour le dictionnaire de babylonien de Paris 3. Leuven; Walpole, Mass: Peeters. Black, Jeremy A. 1991. Sumerian Grammar in Babylonian Theory. StPohl SM 12. Rome: Editrice Pontificio Istituto Biblico. Civil, Miguel. 1975. A School Exercise from Tell Asmar. StOr 46: 39–41. Civil, Miguel. 2003. Reading Gilgameš II: Gilgameš and Huwawa. Pp. 77–86 in Literatur, Politik und Recht in Mesopotamien: Festschrift für Claus Wilcke, ed. Walther Sallaberger, Konrad Volk, and Annette Zgoll. Orientalia Biblica et Christiana 14. Wiesbaden: Harrassowitz Verlag. Civil, Miguel. 2004. Review of Ur III Period (2112–2004 BC) by Douglas R. Frayne. JNES 63: 207–10. Civil, Miguel. 2005. Modal Prefixes. ASJ 22: 29–42. Civil, Miguel. 2007. Early Semitic Loanwords in Sumerian. Pp. 11–33 in From the Workshop of the Chicago Assyrian Dictionary Volume 2: Studies Presented to Robert D. Biggs, ed. Martha T. Roth AS 27. Chicago: The Oriental Institute of the University of Chicago. Civil, Miguel. 2008a. The Early Dynastic Practical Vocabulary A (Archaic har-ra A). ARES 4. Rome: Missione Archeologica Italiana in Siria. Civil, Miguel. 2008b. A Sumerian Connective Particle and Its Possible Semitic Counterparts. AuOr 26: 7–15. Civil, Miguel. 2010. The Lexical Texts in the Schøyen Collection. CUSAS 12. Bethesda, Maryland: CDL Press. Civil, Miguel. 2013. Remarks on ad-gi₄ (A.K.A. ‘Archaic Word List C’ or ‘Tribute’). JCS 65: 13–67. Cohen, Yoram. 2009. The Scribes and Scholars of the City of Emar in the Late Bronze Age. HSS 59. Winona Lake, Indiana: Eisenbrauns. Delnero, Paul. 2010. The Sumerian Verbal Prefixes im-ma- and im-mi-. Pp. 535–61 in Language in the Ancient Near East: Proceedings of the 53e Rencontre Assyriologique Internationale Vol. 1/2, ed. Leonid Kogan Babel und Bibel 4. Winona Lake, Indiana: Eisenbrauns. Delnero, Paul. 2012. The Sumerian Verbal Prefixes mu-ni- and mi-ni-. Pp. 139–64 in Altorientalische Studien zu Ehren von Pascal Attinger mu-ni u₄ ul-li₂-a-aš ĝa₂-ĝa₂-de₃, ed. Catherine Mittermayer and Sabine Ecklin. OBO 256. Fribourg: Academic Press. Eichmann, Ricardo, Hanspeter Schaudig, and Arnulf Hausleiter. 2006. Archaeology and epigraphy at Tayma (Saudi Arabia). Arabian Archaeology and Epigraphy 17: 163–76. Finkel, Irving L. 1982. The Series sig7.alan = Nabnītu. MSL 16. Rome: Pontificium Institutum Biblicum. Gabbay, Uri. 2014. A Balaĝ to Enlil from the First Sealand Dynasty. ZA 104: 146–70. Gantzert, Merijn. 2008. The Emar Lexical Texts. Part 1: Text Edition. Maastricht: Uitgeverij Boekenplan. George, Andrew R. 2013. Babylonian divinatory texts chiefly in the Schøyen Collection: with an appendix of material from the papers of W. G. Lambert. CUSAS 18. Bethesda, MD: CDL Press. Hausleiter, Arnulf. 2011. Tēmā. B. Archäologisch. RlA 13: 515–19.

Words and Grammar: Two Old Babylonian Lists

395

Hruša, Ivan. 2010. Die akkadische Synonymenliste malku = šarru. Eine Textedition mit Übersetzung und Kommentar. AOAT 50. Münster: Ugarit-Verlag. Huber, Peter J. 2007. On the Old Babylonian Understanding of Grammar: A Reexamination of OBGT VI–X. JCS 59: 1–17. Huehnergard, John. 2002. Izuzzum and itūlum. Pp. 161–85 in Riches Hidden in Secret Places. Ancient Near Eastern Studies in Memory of Thorkild Jacobsen, ed. T. Abusch. Winona Lake, Indiana: Eisenbrauns. Jagersma, Abraham Hendrik. 2010. A descriptive grammar of Sumerian. Doctoral dissertation, Leiden: Leiden University. https://openaccess.leidenuniv.nl/handle/1887/16107. Kouwenberg, Norbert J. C. 2005. Reflections on the Gt-Stem in Akkadian. ZA 95: 77–103. Kouwenberg, Norbert J. C. 2010. The Akkadian Verb and its Semitic Background. LANE 2. Winona Lake, Ind: Eisenbrauns. Linssen, Marc J. H. 2003. The Cults of Uruk and Babylon: The Temple Ritual Texts As Evidence for Hellenistic Cult Practice. CM 25. Leiden: Brill/Styx. Malayeri, Mehrnoush. 2012. Scribal Training in Old Babylonian Susa. Pp. 365–76 in Susa and Elam. Archaeological, Philological, Historical and Geographical Perspectives: Proceedings of the International Congress Held at Ghent University, December 14–17, 2009, ed. Katrien D. Graef and Jan Tavernier. MDP 58. Leiden: Brill. Michalowski, Piotr. 2004. Sumerian. Pp. 19–59 in The Cambridge Encyclopedia of Ancient Languages, ed. R. Woodard. Cambridge: Cambridge University Press. Michalowski, Piotr. 2011. The Correspondence of the Kings of Ur: an Epistolary History of an Ancient Mesopotamian Kingdom. MC 15. Winona Lake, Indiana: Eisenbrauns. Van de Mieroop, Marc. 1987. Crafts in the Early Isin Period. A Study of the Isin Craft Archive from the Reigns of Išbi-Erra and Šu-Ilišu. OLA 24. Louvain: Departement Oriëntalistiek. Reiter, Karin. 1997. Die Metalle im Alten Orient. AOAT 249. Münster: Ugarit-Verlag. Reiter, Karin and Hartmut Waetzoldt. 1996. Neue Lesevorschläge und Kollationen zu den altbabylonischen Texten aus Uruk. BagM 27: 401–409. Rubio, Gonzalo. 2000. On the Orthography of the Sumerian Literary Texts from the Ur III Period. ASJ 22: 203–25. Sallaberger, Walther. 1996. Der babylonische Töpfer und seine Gefässe nach Urkunden altsumerischer bis altbabylonischer Zeit sowie lexikalischen und literarischen Zeugnissen. MHEM 3. Ghent: University of Ghent. Sanati-Müller, Shirin. 1990. Texte aus dem Sînkašid-Palast. Dritter Teil. Metaltexte. BagM 21: 131–213. Schaudig, Hanspeter. 2011. Tēmā. A. Philologisch. RlA 13: 513–15. Schrakamp, Ingo. 2010a. Speer und Lanze. RlA 12: 630–33. Schrakamp, Ingo. 2010b. Krieger und Waffen im frühen Mesopotamien. Organisation und Bewaffnung des Militärs in frühdynastischer und sargonischer Zeit. Doctoral dissertation, Marburg: Philipps Universität Marburg. http://archiv.ub.uni-marburg.de/diss/z2010/ 0486/pdf/dis.pdf. Steinkeller, Piotr. 1979. Notes on Sumerian Plural Verbs. OrNS 48: 54–67. Stol, Marten. 2007. Remarks on some Sumerograms and Akkadian Words. Pp. 233–42 in From the Workshop of the Chicago Assyrian Dictionary Volume 2: Studies Presented to Robert D. Biggs, eds. Martha T. Roth AS 27. Chicago: Oriental Institute of the University of Chicago. Veldhuis, Niek C. 2014. History of the Cuneiform Lexical Tradition. GMTR 6. Münster: UgaritVerlag. Woods, Christopher. 2008. The Grammar of Perspective: The Sumerian Conjugation Prefixes as a System of Voice. CM 32. Leiden: Brill.

Lorenzo Verderame

The Seven Attendants of Hendursaĝa: A study of animal symbolism in Mesopotamian cultures The almost complete Sumerian hymn1 dedicated to the god Hendursaĝa contains a long section in which the god is described as a guardian of the night and a psychopomp.2 In these functions the god is assisted by three groups of seven beings. The first heptad (ll. 78–84), whose components are called “warriors” (ursaĝ) and may well be related to the later seven demons (Sebettu),3 are described as sharing features and acting like animals. In this article I will analyse each of the seven animals and their features mentioned in the description of the Seven (the fox, the dog, the raven, the vulture, the wolf, the owl, and the shark), discussing parallels from Sumerian and Akkadian literature in order to highlight how animal symbolism works in ancient Mesopotamia.

1 The fox The first of the Seven is “(like) a fox that drags/shuffles its tail” (1 ka5-a-am3 kun im-ur3-ur3-re, l. 78). The fame and fortune of the fox in Mesopotamian literature and in world folklore is well known.4 In Mesopotamia, the fox is often associated with the god Enlil (ŠI V 94) and, in its astral form, with Erra.5 1 The hymn, ca. 271 lines, has recently been edited by Attinger and Krebernik (2005). An updated translation has been made available by Attinger (2012) on his webpage. In quoting and numbering the lines I will refer to Attinger’s reconstruction of the fragments. 2 For Hendursaĝa and his identification with Išum, see Attinger and Krebernik 2005: 21–33, George 2015, Verderame 2015. 3 Verderame 2015. 4 Suffice it here to mention Aesop’s fables or the medieval collection Roman de Renart; for examples from Asia, see Walravens 2002. For the Akkadian Tale of the Fox see Kienast 2003; Vanstiphout 1988; further references to the fox in CAD Š2 sub šēlebu 268–70. 5 The Fox, identified with one of the stars of Ursa Major, is also a learned name for Mars and it is associated with the infernal gods Erra and Nergal; cf. mul.apin I i 17 “the Fox (is) Erra, Acknowledgements: It is a pleasure and honor to dedicate this article to Miguel Civil. The paper applies one of his favorite approaches, combining lexicography and philology with material culture and interdisciplinary perspectives. https://doi.org/10.1515/9781501503696-022

The Seven Attendants of Hendursagˆa

397

Together with its womanly eyes, the fox’s tail appears frequently both as a pars pro toto of the animal and as a symbol of its attitudes and qualities.6 In a fragmentary letter to the Assyrian king, an anonymous sender, who had previously had to justify himself in front of the king,7 denounces his colleague Binunî and emphasizes his accusation by quoting the following proverb, possibly referring to the length and resistance of the fox’s tail:8 A man who caught the tail of a lion sank in the river, but the one who caught the tail of a fox was rescued (SAA 13: 45 rev. 3–6)9

An entry in the fifth tablet of the teratological series Šumma izbu also mentions the fox’s tail: [If a ewe] gives birth to a [li]on and it has the tail of a fox: request for […] (ŠI V 89)10

Unfortunately, the rest of the line is broken and we do not know how the reference to the fox tail may have related to the apodosis. The following omens suggest a relation between the animal(s) mentioned in the protases and the “value” attributed to it (or them) in the Mesopotamian culture.

If a ewe gives birth to a wolf (ur.bar.ra / barbaru) a badger (ur.ki / kalab urṣi)

ŠI V there will be plague in the land; madness; destruction of the land; disease of the herd the land of the prince will prosper; the king will behave badly toward his land 11

90

91

powerful among the gods” (mul.ka5.a der-ra gaš-ri dingir.meš). See Gössmann 1950: 81 no. 205, Kurtik 2007: 239–41 k06. 6 See the Latin proverb Cauda de vulpe testatur and Erasmus of Rotterdam’s explanation in his Adagia no. 835. The tail of the fox is often associated with fire, as in the history of Samson attaching torches to fox’s tails (Judges 15:4), the Roman festival of Cerialia, and Aesop’s fable of the farmer and the fox. The Japanese kitsune can light fire with its tail. The heaviness and length of the tail, the feature that the passage in Hendursaĝa hymn seems to stress (see below), appears in the depiction of the kitsune, as well as other fox spirits in the oriental tradition (Chinese Huli jing, Korean kumiho), as a nine-tailed fox. Note that in these traditions the fox is associated with cereal crops and their deity. 7 “I confirmed the king’s order, and gave (what was due) to the king. Now then, Nergal-bēluuṣur, the chief cook, can report on me. The king’s order is now fixed in my mouth, and I keep the king’s treaty” (SAA 13: 45 obv. 2–8). 8 Alster 1989. 9 guruš šá si-bat / né-ši iṣ-bat-u-ni ina id2 / iṭ-ṭu-bu šá si-bat še-li-bi / iṣ-bat-u-ni ú-se-zib. 10 [be u8 ur].mah u2.tu-ma kun ka5.a gar e-reš-ti li-di-[iš …]. 11 Cf. the parallel apodosis “If a miscarriage has fur like a badger, (var.:) a dog or a pig: [there will be] famine [in the land]” (ŠI XVII 61′). Cf. the passage “(if) a badger is continually seen in

398

Lorenzo Verderame

If a ewe gives birth to a domesticated dog (ur.gir15 / kalbu) an otter (ur.gir15.a / kalab mê)

ŠI V there will be carnage in the land; the king will weaken his land the king will provide his land with food

92

93

In the first omen, wolves (l. 90) are associated with devastation and periods of anarchy, as shown in the very common apodoses “raging of lions and wolves” (see below). The following omens forecast the future of the land and the behaviour of the king towards it. In the first case (l. 91) these two features contrast with each other: prosperity of the land (+) vs. king’s bad behaviour (–); while in the second case the protasis and apodosis coincide in their negative sign. The omen referring to the otter is apparently positive.12 In all these omens the attributes associated with the animal, as perceived by the Mesopotamians, are reflected in the king’s behaviour. The tail of the fox appears again in a Šumma izbu section at the beginning of the seventh tablet. Here a series of omens describe the appearance of a miscarriage with the head of a lion.

If a miscarriage has the head of a lion (ur.mah / nēšu) the prince will seize universal kingship a lion and the tail of a lion the prince will rule the four quarters a lion and the tail of a fox the prince will serve in a work gang in his own city

ŠI VII 1 2 3

The head, as a symbol of election, pre-eminence and supremacy, determines the subject of the apodoses, the prince. The lion, symbol of power and kingship, conditions the nature of the apodoses. In l. 2, the head and the tail, that it is to say the beginning and the end, express the idea of totality and so forecast

houses” (ur.ki ina e2.meš it-tan-mar) in the collection of omens portending the fall of Akkade, Rm. 155 CT 29: 48–49 obv. 12. See the catchline of the namburbi “Against the evil (portended) by a badger that is (seen) in the field of a man” (ana hul ur.ki šá ina a.gar3 lu2 gal2-[ú], KAR 64: rev. 22); Maul 1994: VIII.5. 12 Note however that the Šumma izbu commentary (ll. 207–209) provides a negative interpretation, for the verbal form ip-pi-ir from epēru “to feed” is assumed to be a substantive, ippi/īru “struggle”, and equated to bubūtu “starvation” and mānaḫtu “fatigue”; see Leichty (1970: 81 and 218).

399

The Seven Attendants of Hendursagˆa

universal leadership for the prince. In l. 3, the apodosis is reversed due to the presence of the fox’s tail. The fox, whose submissive but at the same time deceitful status is in stark contrast to the leadership of the lion in proverbs and tales, determines the nature of the apodosis. The sequence continues with a miscarriage which has a lion’s head with two bodies and two tails (ŠI VII 4)13 or a vulva (ŠI VII 5), and then again with animal parts:

If a miscarriage has the head of a lion and the hair of an ass (anše / imēru) a lion and the body of an ass a wolf a domesticated dog a domesticated dog and the body of a lion

ŠI VII the crown prince will seize the throne of his father

6

– ditto

7

[…] will devour (= pestilence) the prince will weaken his land; […] will become sick the prince will become strong; […] will become sick

8 9

10

The omens in ll. 6–7, which bear similar apodoses, show the workings of the pars pro toto principle (hair = body). A comparison between the two groups of omens confirms the general interpretation of the animal “nature”: the wolf forecasts pestilence and famine (ŠI V 90 / VII 8); the dog, “disloyal” behaviour (ŠI V 92 / VII 9),14 etc. This analysis could be extended to the other animals in the series,15 but this is not the aim of this article. Returning to the line in the Hendursaĝa hymn, the tail is said to be “dragged” by the fox, a standard description which finds parallel in other texts. The idea of the heaviness and strength of the fox’s tail is expressed by the verb “to drag” (ur3)16 as the animal’s main feature, which recurs in other literary passages.

13 This omen is Narām-Sîn’s and it is related, like the others, to the idea of duality. 14 The two omens have the same apodosis (“the prince will weaken his land”). For the dog, see below and fn. 21. 15 The following line (ŠI VII 11) records the birth of a miscarriage with the head of a fox. Unfortunately the apodosis is lost and therefore does not help our analysis. 16 For ur3, see Civil 2003: 80 f.; cf. Proverbs collection 5.123, where the beaten dog “dragged its tail and sat on the street” (kun-a-ni ba-an-ur3 sila ba-ni-in-tuš). Cf. fn. 6.

400

Lorenzo Verderame

The fox keeps dragging its tail (4R 11: 45 f. // SBH 62 No. 33: 21 f.)17 May the fox that flattens the ruin mounds drag her tail on your uzga precinct, established for purification rites (The Curse of Agade 256 f.)18

In commenting on the passage, Cooper (1983: 256) suggests that a contrast between the holiness of the uzga precinct and the unholy fox is meant here. A further hypothesis might also be advanced: in fact, the fox flattens the mounds19 dragging its heavy tail, and together with its habit of making its lair in wasted and destroyed lands, it may suggest the idea of abandonment and destruction of the cultic places20 rather than (or rather than only) an opposition between holy/unholy.

2 The sniffing dog The second being “sniffs like a domestic dog” (2-kam-ma ur-gir15-gin7 si-imsi-im i3-ak-e, l. 79). The dog, domesticated or not, conveys contrasting values. It may be considered under its protective and helpful aspect and thus conceived as psychopomp and symbol of healing goddesses; or it might be perceived as a deceitful being and a rival to humans, and from this viewpoint it is included among the dangerous beings that menace humanity.21 On that day when there will be no snake, no scorpion, no hyena, no lion, neither dog nor wolf, neither fear nor trembling, man will have no rival! (Enmerkar and the lord of Aratta 136–140)

17 ka5-a kun-bi mi-ni-ib2-ur4-ur4-re (var.: mi-ni-ib2-ur3-ur3-ra) / še-le-bu zib-bat-su im-tana-aš-šar. For more duplicates, see HKL I 402, II 228; Heimpel 1968: 349 f. 18 ki uz-ga šu-luh-ha ĝar-ra-zu / ka5 du6 gul-gul-la-ke4 kun he2-ni-ib-ur3-ur3-re; a manuscript presents the variant ka5-zu ki du6-ul-du6-ul-la-ke4, see Cooper 1983: 223 l. 257 L1. 19 Note the use of ur3 with the destructive storm or flood, e.g., “the reaping storm wiped out the country” (ud šu ur4-ur4-re kalam i3-ur3-ur3-re, Lament for Ur 197). 20 Perhaps a parallel may be drawn here with the words of Tobiah the Ammonite about the rebuilding of Jerusalem: “What they are building – if even a fox climbed up on it, he would break down their wall of stones!” (Nehemiah 4:3). 21 For the ur-gir15, see Heimpel 1968: 354–59. The complex figure of the dog, with its dual, contradictory value, as well as its relation with healing goddesses, deserves a more detailed analysis; see Sibbing Plantholt 2015.

The Seven Attendants of Hendursagˆa

401

In both these aspects the dog is a common character in Mesopotamian literature.22 Among its different and symbolical habits,23 the passage of the Hendursaĝa hymn focuses on the act of sniffing, which is paralleled in many other compositions. In Ur, the dog no longer sniffs at the base of the wall (Lament for Sumer and Ur 350)24 Like a dog sniffing around may you not succeed in escaping, face towards me! (The home of the fish 42)25 A dog sniffing with the raised head, (you are) a liar! (Dialogue between a scribe and his son 158)26 A sniffing dog entering all the houses (Proverbs collection 2.109)27

This basic act of seeking through sniffing the air is shared with two other antagonistic animals, the mongoose and the serpent.28 The three of them appear together in the passage describing evil udugs roaming, lurking, and sniffing the air in search of their victims. It is in their nature to slither like a snake, like a mongoose, they are the ones who sniff the latrine, like a dog, prowling they roam around in all directions. (UH VI 174′–176′ ≈ UHF 607–609)29

22 See above. 23 A passage from of the eršemma composition The beaten bull lives no more (am mu-ra nuun-ti) associates the dog with death: “The dog lies upon his spread garment” (tug2am2-bara3ga-na ur ba-e-na2, l. 38); see Cohen 1981: 89–93 E.60, and fn. 33 below. 24 urim5ki-ma ur-bi ur2 bad3-da si-im-si-im nu-mu-un-ak-e. 25 ur-gin7 ki si-im-si-im ak-zu-še3 na-an-ga-ba-e-da-du-e igi-zu ki-ĝu10-uš-še3. 26 ur-gir15 saĝ us2-sa si-im-si-im al-ak-e lu2-IM. 27 ur si-im-si-im e2-e2-a ku4-ku4. 28 “May the quick snake of the mountain sniff it!” (muš ul4 kur-ra-ke4 si-im he2-em-ši-akde3, Lugalbanda in the mountain cave 360//370); cf. “the serpent smelled the fragrance of the plant” (muš i-te-ṣi-in ni-piš šam-mu, The epic of Gilgameš XI 305). For further Akkadian references to the moongose, see CAD Š2 sub šikkû 433–35. 29 UHF 607 ni2-ba ˹x˺ muš-gin7 mu-˹un˺-sur-sur-˹e-de3˺ UH 174′ ni2-bi-a muš-gin7 mu-un-sur-sur-e-[de3] ina ra-ma-ni-šú-nu ki-ma ṣe-ri it-ta-na-áš-lá-lu UHF 608 UH 175′

nin-˹kilim-gin7˺ ur2-e2-gar5-ra-˹ka˺ si-˹im˺-si-im-bi [in-na-a]k-de3 nin-kilim-gin7 ur2-e2-gar8-ra-ke4 ir-si-im in-na-ak-e-de3 ki-ma šik-ke-e a-sur-ra-a uṣ-ṣa-nu šu-nu d d

UHF 609 ˹ur-gi4˺-gi4-da niĝ2 igi m[u-un-ši-i]n-˹bar˺-[e-de3] UH 176′ ur-gir15-gin7 niĝin-e igi mu-un-ši-in-bar-re-e-de3 ki-ma kal-bi ṣa-i-du it-ta-nap-ra-ár-ru šú-nu Here, another feature common to predators and wild animals as well as demons is introduced, that of roaming around: see below, in particular fn. 73.

402

Lorenzo Verderame

3 The raven The third being “like a raven, (its) bite pecks larvae” (3-kam-ma ugamušen-gin7 za-na gug im-kul-e, l. 80).30 The raven is the main character of the first part of the Inanna and Šukaletuda myth. Here, the bird is a creature of Enki and, through the instruction and the magic of his master, introduces date palm cultivation into Sumer. In The šumunda grass, Inanna seized the “fire carrier” šumunda, put it in fetters, and “seized a raven there and set it on top of it” (ba-e-dab5 ugu-ni-še3 ba-gub, l. 32), probably as a guardian. In Enlil and Namzitarra, the god appears under the guise of a raven (ll. 13 ff.). Like the previous animals, the raven is one of the main characters in proverbs and tales:31 for instance, in the following one, which stresses its belonging to the predator group: In the sky there is the raven; on the earth there is the mongoose; in the desert there is the lion …; my husband! Where shall I go? (Proverbs collection 1.128 and parallels)32

In proverbs the raven appears as a competitor to man, constantly menacing the crop; while in lamentations it is a symbol of destruction and abandonment,33 being as it is a messenger and instrument of Enlil.34 As a star, the Raven (Corvus) is associated with both Enlil and Adad, but it is also a learned name for Mars.35

30 For uga (Akk. āribu), possibly “raven,” see Veldhuis 2004: 299–301 and Weszeli 2007 with the previous bibliography. 31 The fragmentary proverb beginning “A raven … a porous jar” (ugamušen dugsahar an-da[…], Proverbs collection 8 D 2) may recall the folk motif of the animal trying to fill a river or the sea, paralleled in the Sumerian literature by the fox urinating in the Tigris or the sea (Proverbs collection 2.67, and UET 6/2 216). 32 [an] ˹ugamušen-am3˺ / [ki] dnin-kilim-e / ˹edin˺-na ur-˹mah˺-[am3] / [X X] KAŠ4 SA […] / [mu]-ud-na-ĝu10 me-[še3 ga-ĝen]. 33 “In my man’s fold dwells the raven” (mu-lu-ĝa2 rig7-ga-na ugamušen ba-e-dur2, The beaten bull lives no more 39), see Cohen 1981: 89–93 E.60, and fn. 23 above. Thompson (1903: L–LII) describes some popular beliefs regarding evil birds, among them the owl and the raven, in Near Eastern folklore. 34 “The raven, the messenger bird of the gods” (uga musen niĝir2 diĝir-re-e-ne-ke4 / a-riba iṣ-ṣu-ra na-gi-ir dingir.meš , UH VIII 33). For the relation of the raven with the Enlil circle, see the line “Raven, elder brother of Su’en” (ugamušen š[eš-s]aĝ dsu’en-na, YOS 11: 61, 3) and the incantation incipit “ Raven … (a container?) of Enlil” (en2 ugamušen mud.la2 den.lil2.la2. ke4 […]; see Maul 1994: 198 l. 14, cf. 200 l. 14″ and fn. 376); cf. “the raven containers of Papsukkal and Guškinkubanda” (nam-ḫa-ri.meš a-ri-bi šá dpap.sukkal ù dguškin.ku3.ban3.da, RAcc 89: 10). 35 See the references in Gössmann 1950: 47–50 no. 132–133 and Kurtik 2007: 557–61 u11.

The Seven Attendants of Hendursagˆa

403

As for the line in the Hendursaĝa hymn, its general meaning appears clearer than the meaning of its single parts. The term za-na has been taken here as “larva” (Akk. mūnu), on the basis of evidence from lexical lists, while it remains a hapax in the literary texts.36 The basic meaning of gug is assumed to be “tooth” on the basis of the lexical equivalence with the Akk. šinnu.37 In literary passages it has the meaning of something that pierces and leaves a mark, like a bite, which possibly results in the extended meaning of “mole, mark on skin” (Akk. ḫalû and pendu). While the bite is always described as that of a dog, the general expression is used as a metaphor for sharpening or beating weapons and, in one case, for the bite of an animal with no teeth, but with a beak (the turtle), like the raven in our passage.38 The verbal root kul expresses the idea of a vertical movement both upwards (to pick up, to collect, to remove, etc.) and downwards (to pick, to beat, to pierce, etc). The idea of a repetitive vertical movement is reflected in compound nouns like the wooden peg (ĝišsaĝ-kul / sikkūru) which, moving up and down, blocks the door’s locking system. The meaning of an upward movement is rendered in Akkadian by the equivalent verb laqātu, whose basic meaning of picking up, particularly of animals gathering food from the ground with the mouth or the beak,39 would also fit the sense of the Hendursaĝa line, for the raven picking up (kul) bugs (za-na) with its “bite” (gug).40

36 Other homophones for za-na (or za-na(-ru)?) are known from Akkadian equivalents: “doll” (Akk. passu); “upper part” (elītu) or “breach” (pilšu). The latter is frequently referred to in relation to the ear, in the expression za-na-ĝeštu2-ĝu10 in the monolingual Ugu-mu 142, corresponding to za-na-ĝeštu-ĝu10 : pí-li-iš uz-ni-ia, e-li-it uz-ni-ia “earhole, upper part of the ear” in a bilingual version of the series, see Couto Ferreira 2009: 38, 70 f., and 192; cf. za-na-ĝeštu, za-na-ru-ĝeštu = e-lit uz-nim in SIG7.ALAN XXV (= L) 176 f. (MSL 16: 228). See also nin-ĝu10 suh10 za-gin3 za-na gun3!-a “My lady, (adorned) with a lapis-lazuli diadem and a multicolored z.” (Ur III catalogue at Yale (Y1) 31 // Old-Babylonian catalogue in the Louvre (L) 67). 37 Attinger and Krebernik 2005: 66. 38 “My battle-axe, like a dog with a bitter tooth, gnashes continually (lit. beats bites) … my m.-weapon at the head of the land “(mitum-ĝu10-u3 ur zu2 sis-a-gin7 gug hu-mu-u3-ra-rara, ma-sa2-tum-ĝu10-u3 saĝ kalam-ma-ka gug sar ha-mu-u3-ak-ke4, Šulgi D 158 and 189; cf. Iddin-Dagan D 12, […](–)x-ba gug sar ak); “my bite is that of a small dog” (gug-(a-)ni gug ur-tur-ra-ke4, The heron and the turtle 54, passim). 39 This basic meaning is paralleled by the Hebrew ‫ לקט‬and the Arabic ; for the idea of picking up with the tip of something: cf. the Neo-Assyrian passages referring to the act of picking up seeds with the tip of the tongue as a punishment (SAA 14: 166 e. 9′ and parallels quoted in CAD L: 100). An extended meaning of laqātu is that of uprooting and exterminating the progeny in curses. 40 The analysis and discussion of the Sum. kul and its compounds deserves further attention, which cannot be given here; see also Peterson 2007: 373. Note the comparison with the IndoEuropean root PIC-/BIC-, from which the act of picking, beating, or piercing (it. picchiare, sp.

404

Lorenzo Verderame

4 The huge “vulture” The fourth being “overwhelms like a huge carrion devouring vulture” (4-kamma te8mušen maḫ ad6 gu7-a-gin7 ka i3-ša-an-ša-ša, l. 81). The te8-bird has been identified with various birds of prey, mainly the eagle, the buzzard, or the vulture;41 the latter will be used here for the sake of convenience. The main reference in Sumerian literary sources for the te8-bird appears in the composition Nanše and the birds. Here the name of the te8-bird is repeatedly followed by the apposition or epithet uš-bar and a closing line states “its [wings] together, its feet together.” 42 The term uš-bar means weaver and a preSumerian origin has been proposed (Rubio 1999: 4). This might be related to a Semitic root sbr, whose Akkadian cognate ṣabāru means “to move quickly” (to twinkle, to blink, to twitter, etc.) and “to bend, twist.”43 Both verbs may be used to describe the act of spinning the spindle, which provides a good basis for the interpretation of uš-bar as “weaver” as well as for its use as a descriptive reference to the vulture.44 In fact, the basic idea of a circular movement may recall the flight of carrion birds.45 Vultures and other large birds of prey such as eagles are represented on reliefs as devourers of enemies’ corpses together with other animals like dogs.46 They are symbols of war and particularly of victory through the annihilation of the enemy.47 No wonder the te8 star (Aquila), often translated as Eagle

picar, fr. piquer, etc.), the name of a weapon (sp. pica, it. picca, fr. pique etc.) as well as the term for the beak (it. becco, fr. bec, sp. pico, etc.) and derived bird names. 41 See the discussion in Veldhuis 2004: 286 f. The Akk. equivalent erû is generally translated as “eagle”. 42 ˹mušen˺-e an-na te8mušen uš-bar te8mušen uš-bar / ˹te8˺-uzmušen-e an-na te8mušen ušbar te8mušen uš-bar / [a2]-bi teš2-am3 ĝiri3-bi teš2-am3, Nanše and the birds Seg. A 54–56; the rest of the lines are fragmentary. 43 The two verbs have two different stem vowels (u and a/u) and are listed separately in the dictionaries as ṣabāru I and II, see CAD Ṣ s. v. 2–5; AHw 1065–66. Note, however, that CAD and AHw present different distributions of the passages in which the two verbs have a similar form, showing a close relation in meaning between the two terms. 44 Note the use of bal for ṣabāru in the bilingual UH V 6, which provides a further link with the idea of a circular movement and the spindle (ĝišbal). 45 See Black 1996: 35 f. 46 The obscure Sumerian proverb “A vulture enters a man’s mouth” (te8mušen ka ˹lu2˺-kam ba-an-kur9, Proverbs collection 28.6) may be related to the birds pecking the lips of severed heads, as in the Daduša stele; cf. the fragmentary passage of Nanše and the birds (Segment B 10), [x] te8mušen […] x-na ib?-ši-ku4-ku4. 47 For the ideological involvement of the corpse devoured by animals, see Cassin 1982.

The Seven Attendants of Hendursagˆa

405

star, is associated with warrior gods like Zababa, Ninurta, and Nergal, as well as being a learned name for Mars.48 However, this idea is also based on the spread wings as a symbol of enveloping and overwhelming.49 This finds a further connection in the verbal construction in the line of the Hendursaĝa hymn, ka(-ša-)an-ša-ša, a well-known expression for overwhelming.50 It is often found in bilingual texts with its Akkadian equivalent kašāšu to describe an overpowering attack by evil beings:51 Fainting and daze have covered him like a garment, have overwhelmed him incessantly (Šurpu V–VI 15–16) (Demons) are an overwhelming storm which leaves nothing behind (UH V 29 // UHF VII 703)

5 The wolf The fifth being “although not a wolf, swoops on black lambs” (5-kam-ma urbar-ra nu-me-a sila4 ĝi6-ga i3-šub, l. 82). The wolf seizing lambs is a wellknown image in Sumerian literature.52 The wolf, a predator and symbol of dis48 “The star of Zababa: the Vulture and the Corpse” (diš mul dZa-ba4-ba4 mul.te8 u mul.ad6, MUL.APIN I ii 12); see Gössmann 1950: 1–2 no. 2 and Kurtik 2007: 22–26 a04. 49 In this regard, compare the similarity between large birds of prey and the spread battle net falling over and enveloping the enemy; an association that may be found in the Anzû-handle on the Ninĝirsu battle net in Eannatum’s Stele of Vultures. Note the passage “circling around him like an eagle” (gim a-re-˹e˺ i-sa-a[r ugu-šú], Epic of Gilgameš VIII 60), where the verb sâru is generally used to describe the action of winds and demons (CAD S s. v. 190), cf. “At that time, the Seven, evil gods, were whirling about at the base of the heavens” (u4-bi-a imin-bi diĝir hul-a-meš ul-he2-a du7-du7-meš / i-nu-šu si-bit-ti-šú-nu dingir.meš lem-nu-ti ina šu-puk an-e i-sur-ru, UH XVI 29). 50 For overpowering the enemy, see Ninurta C 69–70 (erim2-ma ka-ša-an-ša-ša); Sîn-iddinam to Utu 19 (erim2 ga-an-ša-ša). See Heimpel 1968: 478, cf. 425; Hallo 1973: 168 f. and fn. 37, Sjöberg 1973–74: 173. 51 niĝ2-me-ĝar-ra tug2-gin7 ba-an-dul ka-ša-an-ša-ša / qu-lu ku-ú-ru gim ṣu-ba-ti ik-tùm-šúma ik-ta-na-áš-šá-áš-šú (Šurpu V–VI 15–16); u4 ka-ša-an-ša-ša niĝ2-nam-ma nu-un-tak4tak4 / ud mu-uk-taš-šá-áš-šú šá mim-ma šum-šú la iz-zi-bu (UH V 29 // UHF VII 703). See CAD K s. v. 286 and CAD A sub ašāšu B 424 f.; note that like the other verbs describing the demon’s actions, these appear in the iterative form (-tan-) as well. 52 “The wolf was not seizing lambs” (ur-bar-ra-ke4 sila4 nu-ub-kar-re, Enki and Ninhursaĝa 14); “like a wolf which has seized a lamb, he runs quickly … the wolf seized the ewe and its lamb” (ur-bar-ra sila4 šu ti-a-gin7 ul4-ul4-e im-ĝen … ur-bar-ra u8 sila4!-˹bi˺ in-kar, Enmerkar and En-suhgir-anna 49 and 235). Sheep and lambs are wolves’ favourite victims, but goats are not such easy prey (Proverbs collection 8, 12 f.).

406

Lorenzo Verderame

order,53 is a frequent character in Mesopotamian tales, together with the fox, the lion, and the dog, and is often compared and contrasted with the latter. The proverbs stress its voracity, ferocity, and treacherousness.54 Furthermore, there are often references to its contradictory social behaviour, that is, as a solitary animal that hunts in a group. The story of the shepherd turned into a wolf by Ištar in the narrative of the goddess’s lover in the Epic of Gilgameš is otherwise unknown in the Mesopotamian sources, but has several parallels in other cultures. You loved the shepherd, the grazier, the herdsman, who regularly piled up for you (bread baked in) embers, slaughtering kids for you every day. You struck him and turned him into a wolf, so his own shepherd boys drive him away, and his dogs take bites at his thighs. Epic of Gilgameš (VI 58–63)55

The literature on human beings transformed into wolves is too vast to be discussed here, but it is worth mentioning the story of Lycaon, king of Arcadia, who serves human flesh to test the divine nature of his host (Zeus) and is therefore transformed by the god into a wolf. Several versions of the story exist,56 but all deal with the subject of transgression (the killing of a human being for sacrifice or consumption) and the transformation of the evildoer into a wolf, banished from the human community and obliged to eat human flesh. The brief narrative in the Epic of Gilgameš offers few details of the story of the shepherd, but we may hypothesize that his daily habit of slaughtering kids for Ištar may represent the link with his successive transformation into a wolf. While this element of continuity creates a connection between the shepherd and the wolf, the transformation itself implies the crossing of a boundary (man/animal) and an inversion of roles, from the shepherd who drives back

53 See above. The Wolf star (α Trianguli) is a learned name for Mars; see Gössmann 1950: 65 no. 161 and Kurtik 2007: 567–69 u18. 54 “You, Wolf, are an image of slander, an evildoer, one who cuts his friend’s throat” (at-ta bar-bar ṣa-lam [ṭa]-píl-ti / ˹e-pí-iš lem˺-né-ti na-ki-su na-piš-ti tap-pí-šu, Tale of the Fox I iv 20 f., see Kienast 2003: 46 f.). 55 Quoted after George 2003: I 622 f. 56 The identity of victim (a child, a serf, a prisoner, etc.) and the reason for the killing (sacrifice or consumption) change as well as the protagonists. In the latter case, the author of the misdeed may be Lykaon alone or his numerous sons, transformed into a solitary wolf or into a pack of wolves, thus recalling the animal’s dual social nature. For a study of the story of Lykaon and similar mythemes, see Piccaluga 1968.

The Seven Attendants of Hendursagˆa

407

the wolves to the wolf chased away by his own shepherd boys and bitten by his own dogs. Other lovers of Ištar recalled by Gilgameš experience similar fates, but it is worth mentioning a transformation in the opposite direction mentioned in the Epic of Gilgameš: Enkidu’s conversion from animal to man. In the Enkidu story too we find an element of continuity and inversion in the transformation from animal to human (or from uncivilised to civilised). The “animal” Enkidu avoids the shepherd, runs with the gazelles and frees animals from the hunter’s snares; but once humanised, he is avoided by the animals, becomes a shepherd, and chases off wild animals (lions and wolves). The shepherd transformed into a wolf is banished from the human community, while the “civilised” Enkidu joins it.

6 The nin-immax bird The sixth being “shouts like a n.-bird, which resides in …” (6-kam-ma dninimmaxmušen-gin7 x x tuš-a-gin7 gu3 mi-ni-ib-ra, l. 83).57 The term dninimmax(lagab×sig7) and its variants (dnin-ninna(2)/dnin-niĝin2) refer to a bird of prey that attacks small birds and is even capable of snatching a “lamb from the sheep house.”58 The literary passages depict this bird as a predator and as a symbol and messenger of death. The Akkadian equivalents seem to be associated mainly with this latter feature, since all are qualified in lexical entries as “evil portending bird” (iṣṣur lemutti).59 The eššebu is a night bird, 60 often interpreted as a messenger, possi57 Edzard and Wilcke (1976: 148 f.): “Der Sechste schreit(?), wenn er … sitzt(?), wie ein Falke(?)”; Attinger and Krebernik (2005: 55): “le sixième, tel un oiseau …, hurle dans la ville qui s’est endormie.” 58 dnin-ninna2mušen-ta-e ĝa2 udu-ka sila4 šu ba-ni-ib-ti (Dumuzi’s dream 35 // 60); buru5mušen dnin-ninna2mušen-ta šu-ta šub-ba-gin7 (Nungal A 50); cf. “Then I arose like a n.-bird (var.: a falcon)” (dnin-ninna2/immaxmušen (var.: sur2-du3mušen)-gin7 ha-ba-zig3ge-en, Šulgi A 60). Veldhuis (2004: 272–75) excludes the identification with the owl; see also Heimpel 1968: 419; Focke 1998, 1999–2000. 59 Other night birds are considered birds of ill omen, such as the “owl” (qadû) whose Sumerian name, uru-ḫul-a, refers both to the town and the forecasting of evil; see CAD Q s. v. 51. 60 For eššebu, see CAD E: 370–71, in particular the final comment. The lexical entries for dninninna2 provide equivalences with two other unidentified birds, anpātu and kilīlu; the former, defined as the bird of Gula (STT 400: 34) is related to the Hebrew ‫ﬣ‬, an unclean animal identified with the heron (Leviticus 11:19; Deuteronomy 14:8). For kilīlu or kilili, possibly the owl, see the references and commentary in CAD K s. v. 357; as a demon, see also Wiggermann 2011: 315.

408

Lorenzo Verderame

bly to be identified with the hoopoe.61 A bilingual passage from the Utukkū lemnūtu relates the demons to the dnin-ninna / eššebu bird: They are the owl which screeches in the city (UH V 9)62

7 The shark (kušu2) The seventh being “is (like) a shark (that) surfs the waves” (7-kam-ma kušu2ku -am3 a-ĝi6 im-bu-bu-bu, l. 84). The kušu2 and its Akkadian cognate kušû have been equated with several aquatic animals: the seal, the shark, the crocodile, the crab, the turtle.63 The tentative identifications are complicated by the later confluence with the kud-da (another unidentified animal) in the lexical lists and by the confusion with the sign nagar identifying the Cancer constellation in later texts.64 The reference to the use of the skin (kuš) of this animal in an Old Akkadian and a Lagaš II economic text from Ĝirsu seems to exclude the identification with the crab or a turtle and to favour that with other animals.65 Sharks, for their part, are not uncommon in Near Eastern rivers66 6

61 CAD E: 370–71; cf. “Concerning the hoopoe about which you said: “It has been sent as a message” (ina ugu eš-še-bu-u.mušen šá taq-bu-u-ni / ma-a šip-ru šú-u šap-ir, SAA 10: 183, 1– 2). 62 dnin-ninnamušen uru-a šeg10 gi4-gi4-a-meš / eš-še-bu šá i-na a-li i-šag-gu-mu šú-nu. 63 See Cohen 1973: 203 ff., Boehmer 2002: 30 f., Attinger and Krebernik 2005: 66. Against the distinction of different animals (shark, seal, crab) proposed by Landsberger (1962: 89–94), Cohen (1973) relates all the references to the Akkadian kušû and the Sumerian kušu2/kud-da to a single animal, the turtle, while Boehmer (2002) in his study on sharks discusses only the references that may be related to these animals. In view of what follows, I prefer to maintain a distinction between at least two animals, the shark and a crab/turtle(?), corresponding to the Sumerian kušu2. 64 See the detailed discussion of most of the references by Landsberger (1962: 89–94) and Cohen (1973); for the constellation of Cancer see Gössmann 1950: 114; Kurtik 2007: 269–270 k31. 65 DPA 30, 4: 3 (30 kuš kušu2); RTC 222: rev. ii 6 (5 kuš kušu2); cf. the reference to the “fleece” (bar) of the kušu2 in Diatribe C 2, see fn. 70. For the uses of sharkskin see fn. 67. 66 Boehmer (2002). Before the construction of dams and the drainage of the Marshlands, sharks and shark attacks were very common in this area as well as in the upper stream of the Tigris and the Euphrates (personal communication of A. al-Hamdani); see also Coad (2010: 72–75) and the article The Riverine Bull Sharks of Iran and Iraq in the blog Corner of the Cabinet (http://cornerofthecabinet.com/2014/10/08/the-riverine-bull-sharks-of-iran-and-iraq-a-briefmodern-history/). The following passage from the Pre-Sargonic riddles of Lagaš (Biggs 1973) may well refer to the shark: “Its canal is “(the canal) going to Nina,” its deity is Nanše, the mighty lady, its fish is “the man-eater” (ux-lu2 gu7-[g]u7), its snake is […]” (2H-T25 ll. i 1–3). It

The Seven Attendants of Hendursagˆa

409

and their skin was used for various purposes in several cultures.67 Finally, the similarity of kušu2/kušû with the modern Iraqi term for shark ( ) may support a fascinating hypothesis of a local survival of the ancient term in modern times.68 References to the kušu2 in literature are scarce. Its tooth is mentioned in Lugal-e69 and possibly referred to in the obscure passage in Lugalbanda and

is possible to identify the river monster conjured in the short Akkadian incantation from Susa published by Cavigneaux (2003: 61 f.) with the shark. 67 The skin of the shark, like that of the stingray, has a rough surface that makes it suitable for a variety of uses. Note the adjective shagreened (it. zigrinato) from shagreen (fr. chagrin, it. zigrino; cf. also en. chagrin) “sharkskin,” a term that originally meant the rough skin of the back of a horse or donkey from Turkic sāğrī “horse back”; it. zigrino is also a folk name for the kitefin shark (Dalatias licha), whose skin was used as an abrasive. The rough surface of the sharkskin offers a good grip and is commonly used as the handle of bows and swords (in China, Japan, and elsewhere). It is a fine abrasive used as a skin scraper for beauty treatments (in Polynesia) or for polishing wood and ivory; the latter (it. smerigliare “to sander, polish”) is the origin of the Italian folk name smeriglio for the porbeagle (Lamna nasus). Due to its resilient texture, shagreen is a leather used in luxury goods such as shoes, belts, luggage, boxes, book covers, and so on. The use of shark and stingray skin is widespread, and was very common in Japan from the eighth century CE. Introduced into Europe as an exotic leather from the Orient in the eighteenth century – the term shagreen is of Turkic origin and was not documented before – it became popular thanks to Louis XV of France’s master leather artisan, JeanClaude Galuchat, from whose name galuchat, an alternative term for shagreen, derives. For the use of fish and water mammal bones and skins in ancient Israel and the ancient Near East, see Neufeld 1973; for other by-products of sharks in Iraq, see Coad 2010: 75. 68 The term (kōsaj), which we find alongside the classical Arabic , is spelled in various ways and it is commonly believed to be a Persian loanword. Mario Casari, who at my request kindly checked the Arabic and Persian sources, informed me that the Persian word (gūše) means both shark and a person with a goatee, or wisps of hair on his chin. was the nickname of the former Iranian president Rafsanjani, due both to his physical feature (his goatee) and to his political behaviour (a shark). Arabic dictionaries have both meanings under the term (e.g., Lane, Arabic-English Lexicon, London 1863, pp. 2809– 10): “goatee”, is recorded in the earliest dictionaries (ninth century CE) and labelled as a loanword from the Persian “shark” appears later (possibly after the tenth century CE) in lexical lists, related to the same root as “goatee,” but without information on its origin. In contrast, the oldest Persian dictionaries record only the meaning “goatee” for ; the meaning “shark” appears only in modern dictionaries and without any comment on its origin. A plausible explanation is that even in Antiquity the Arabic term for “shark” was related on the base of homophony to the Persian term for “goatee” and then fused together. Thus, the Arabic “shark” may well be related to the old Sumerian kušu2 and Akkadian kušû. 69 “For them a k. tooth has grown up (var.: came forth) in the kur, dragging away the trees” (kur-ra zu2 kušu2-a mu-ne-mu2 (var.: mi-ni-in-e3) ĝiš mi-ni-ib-ur3-ur3, Lugal-e 39). An interpretation of ĝiš … ur3 “harrowing” recalling the idea of a penetrating tooth may be suitable, but it is made less plausible by the parallel line 176 (ĝiššinig mi-ni-ib2-ur3). The Akkadian

410

Lorenzo Verderame

the Anzu bird (l. 47) as well, where the reference to the claw of Anzû like that of the kušu2 should be amended.70 The interpretation of the Hendursaĝa hymn passage is made more difficult by the triplicated verbal form im-bu-bu-bu, which the previous editors of the text have not interpreted.71 The verbal root could be related to bu “to flit” or to gid2 “to drag,” often used in relation to water (“to drag or tow a boat”). Both verbs may fit the sense of the movement of the kušu2 in the water, darting across the waves or producing waves with its movement.72 The former verb (bu) suggests further associations. In fact, bu describes the roaming around of beings in the steppe, such as animals, winds, and ghosts.73 The verb is particularly frequent with lil2 (gale, ghost, or lillu-demon), which is said to roam around (…-bu-bu-…) in the plain.74 The strict relation of this root with the features of beings populating the steppe is confirmed by the Akkadian equivalents of Sumerian bu. These, often in the iterative form -tan-, refer to the act of roaming around in order to chase the victim (ḫalālu, našarbuṣ/ṭu, saḫāru, sanāqu) or to the cause or aim of roaming around (u3-bu-bu-ul / bubūtu “famine”; ḫabātu “to plunder”). As a result of the continuous movement of an animal or of one of its parts, Sumerian bu may be equated to a verb referring to the sound produced by insects, the hissing of snakes, or to murmuring (ḫabābu, ḫalālu). Verbs referring to movement or the production of sounds have been

translation reads “A k. tooth came forth in the mountain and she began to tremble” (ina kuri šin-ni ku-ši-i a-ṣa-at i-ta-nar-ra-ar). 70 See Wilcke 1969: 151. In the text of the Diatribe C, among the insults proffered against the opponent there is that of “fox with a k. skin” (ka5-a bar kušu2ku6, l. 2). For a wind vane possibly in the shape of a, shark see Neumann and Parpola (1983). Note that in the Underworld vision of an Assyrian prince (SAA 3: 32), the evil Alad tramples on a kušû with its left foot (gir3.2 kab-šú ku-še-e ka-bi-is, l. rev. 4). 71 Edzard and Wilcke (1976: 148 f.); Attinger and Krebernik (2005: 56). See Peterson (2007: 215 fn. 935) for further parallels and an interpretation of the verb bu as “to howl”. 72 This may point to large animal able to dive fast, like the shark; cf. the destructive flood of Gilgameš and Huwawa A 42 / B 43, discussed below. 73 “The wild four-legged animals did not run about” (maš2-anše niĝ2-ur2-4 niĝ2-zi-ĝal2 numu-un-bu-e, The Lament for Sumer and Ur 131); “no flies buzz around” (nim nu-mu-un-bubu-bu, The home of the fish 8); “like a hurricane whirling in the midst of heaven” (tum9maruru5 an-ša3-ga-še3 bu4-bu4-gin7, The debate between Bird and Fish 112). See the references in CAD N2 sub našarbuṭu 60 (lex.sec.). For the idea of roaming around of beings that populate the steppe, see Verderame 2011: 114; 2012: 120; see above fn. 29. 74 “Chased by lillu in the arid plain” (edin bar-rim4 lil2 bu-bu-da, Išbi-Erra B 6); “roams about like a (male) lillu” (le-e-la-a i-bu-bu / nita-lil2-la2-am3 i3-bu-b[u], in Geller 1989: 194 l. 2); “the lillu, dweller of the plain, roams about” (lil2-la2-am3 tuš-edin-na i3-bu-bu-de3-eš, UHF III 18, see Geller 1985: 20 f. l. 18, 87; cf. [e-ne-n]e-ne bu-bu-meš, UHF V 406, [… mu?un?-d]a-˹bu?-bu?˺-de3-eš, UHF III 271); cf. lil2-bu-bu in Proto-Lu 829 (MSL 12: 36).

The Seven Attendants of Hendursagˆa

411

entered separately in the Akkadian dictionaries, but in some cases, for instance ḫalālu, they are clearly cognates of the same root. A third interpretation would be to relate the verb to bu(r) “to tear out, to uproot, eradicate” (Akk. nasāḫu), often reduplicated bu-bu or even triplicated, possibly to be read bur12 or bu7 according to ePSD.75 A key passage for this association is Lugal-e 254, where the reduplicated verbal root bu may be related to the act of tearing out or biting, and paralleled to a line of the same composition mentioning the tooth of the kušu2 (l. 39).76 The passage in the Hendursaĝa hymn may therefore describe the shark chasing through the water with open jaws and “biting” the waves.

Conclusions The seven beings are described by comparison with a feature or an attitude of an animal. All of them are predators, with the exception of the raven, and all are wild except for the dog. Three are land animals (the fox, the dog, the wolf), three are birds (the raven, the vulture, the owl), one is a fish (the shark). Most of them are considered dangerous and competitors to humans. They are all directly or indirectly related with death, destruction, abandonment; as a result, the astral forms of all of them are associated with the planet Mars.77

75 Cf. “… uproot …” (… ur2-ba …-bu(3)-bu(3)-…, Lugalbanda in the mountain cave 295, 310, 322; Šulgi D 223 and 341, Šulgi C seg. B 31); “its date clusters … were torn out” (zu2-lum-mabi … ba-ra-an-bu-bu-de3-eš, The lament for Sumer and Ur 417); “It uprooted […]” ([…] im-miin-bu-bu, Inanna and Šukaletuda 65); “With her fingers she pulled out the boxwood stake” (šu-si-ni bulug ĝištaškarin-na mu-na-an-bu-bu-re, Nanna-Su’en’s journey to Nippur 237A // 248A); “the house and the smitten city, all have been uprooted, like a shepherd’s sheepfold it has been uprooted” (e2 uru2 sig9-ga mu-un-bu-bu-bu, The lament for Ur 130); “(the turtle) uprooted the heron’s (nest) made of reeds” (igira2 gi du3-a-ni mu-na-an-bu-bu, The heron and the turtle A 61 ≈ 101); see in general Volk (1995: 154 and fn. 680). The term im-bu-bu-a of the Proverbs collection 2.54 (šidim pe-el-la2 im-bu-bu-kam, l. 95) may be related to this root and translated “one who extracts clay,” rather than a “clay dragger.” 76 gugx-ka-gin7 i3-bu-bu-de3 / [… -s]u i-šal-li-šu; see the discussion in Klein (1981: 97–98) and Seminara (2001: 282–83). Cf. fn. 69. 77 Gössmann 1950; Reynolds 1998; Kurtik 2007.

412

1 2 3 4 5 6 7

Lorenzo Verderame

English

Sumerian

Element

Predators

Domain

fox dog raven vulture wolf owl shark

ka5-a ur-gir15 ugamušen te8mušen ur-bar-ra d nin-immaxmušen kušu2

tail nose/sniffing pecking mouth? preying voice roaming/biting

x x ? x x x x

Earth Earth Sky Sky Earth Sky Water

The analysis of the passages from the Hendursaĝa hymn shows the workings of figurative language and symbolism. In their constant interaction with human society, animals are a source of observation, abstraction, and symbolism. In accordance with a well-chosen anthropological definition, animals are “good to think”.78 The animal is attributed a “moral” essence or “nature” (Aelian’s natura animalium) according to the relation it establishes with human society; thus it may be considered as positive, negative, or ambiguous. The animal aptitude, “nature”, and qualities are reduced to an idiosyncratic action it performs or to its relevant and functional body parts (claws, jaws, wings, horns, etc.). These features (habits or body parts), separated from the single species and shared with animal with similar aptitudes, are thus a common denominator of that class of beings − its mark and marker. They become evocative carriers of meanings (semantemes). The claws and the jaws are features, attributes, and symbols of the bird of prey or other predators, as are wings in the case of the birds. Transcending from the single animal or class, these features have an independent life as symbols conveying concepts (claws and jaws for strength, fierceness, etc.; wings for freedom of action) used as allegories and metaphors in figurative language. The seven beings in the Ḫendursaĝa hymn have no name. Their bodies are left undefined. Instead, they are described through analogy with the actions

78 Sperber 1996; for an overview of human/animal relations, see Ingold 1994 and, for ancient Near East, Collins 2002. The anthropological debate on animal symbolism has focused mainly on taxonomies related to the problem of totemism and taboo (C. Lévi-Strauss, M. Douglas, V. W. Turner, S. J. Tambiah, etc.), leaving aside the analysis of the main concept, as well as other aspects of the topic (Tullio-Altan 1991: 225), which have been taken up in recent decades by other disciplines such as semiotics and cognitive sciences; see Sperber 1975, 1996. For an overview of the debate and new perspectives on symbolism in anthropology, see Tullio-Altan 1991. A more productive debate on symbolism has developed in historical schools influenced by anthropological theories or sensitive to interdisciplinary research; for Classical studies, see the overview by Tullio-Altan (1991: 226), for Mediaeval studies, see for example the works of M. Pastoureau.

The Seven Attendants of Hendursagˆa

413

characteristic of different animals which suggest their functions and qualities as attendants of the god. The reference to the feature of a single animal suggests that the origin of the Seven may lie in the class of the magic animals, rather than that of the monsters or demons whose body is the agglutinative result of the symbolic body parts of other animals (e.g., Lamaštu, Pazuzu, etc.).79

Bibliography Alster, Bendt. 1989. An Akkadian Animal Proverb and the Assyrian Letter ABL 555. JCS 41: 187–93. Attinger, Pascal. 2012. Ḫendursaĝa A (4.06.1). Unpublished manuscript: http://www.iaw. unibe.ch/unibe/philhist/ifaw/content/e246526/e255000/e274658/e274665/e379923/ e379995/4_6_1.pdf Attinger, Pascal and Manfred Krebernik. 2005. L’hymne à Ḫendursaĝa (Ḫendursaĝa A). Pp. 21–104 in Von Sumer Bis Homer. Festschrift Für Manfred Schretter Zum 60. Geburtstag Am 25. Februar 2004, ed. R. Rollinger. AOAT 325. Münster: Ugarit Verlag. Biggs, Robert D. 1973. Pre-Sargonic Riddles from Lagash. JNES 32: 26–33. Black, Jeremy A. 1996. The Imagery of Birds in Sumerian Poetry. Pp. 23–46 in Mesopotamian Poetic Language: Sumerian and Akkadian, eds. M. E. Vogelzang and H. L. J. Vanstiphout. CM 6. Groningen: Styx. Boehmer, Rainer M. 2002. Vom Hassek Höyük bis zum Buch Tobias. Von Sägefischen und Haien im Altertum. BaM 33: 7–43. Cassin, Elena. 1982. Le mort: Valeur et représentation en Mésopotamie ancienne. Pp. 355–72 in La mort, les morts dans les sociétés anciennes, ed. Gherardo Gnoli and Jean-Pierre Vernant. Cambridge / Paris: Cambridge University Press / Editions de la Maison des Sciences de l’Homme. Cavigneaux, Antoine. 2003. Fragments littéraires susiens. Pp. 53–62 in Literatur, Politik und Recht in Mesopotamien. Festschrift für Claus Wilcke, ed. Walther Sallaberger et al. Wiesbaden: Harrassowitz. Civil, Miguel. 1987. Feeding Dumuzi’s Sheep. The lexicon as a source of literary inspiration. Pp. 37–55 in Language, Literature, and History. Philological and historical studies presented to Erica Reiner, ed. F. Rochberg-Halton. AOS 67. New Haven: American Oriental Society. Civil, Miguel. 1998. “Adamdun,” the Hippopotamus, and the Crocodile. JCS 50: 11–14. Civil, Miguel. 2003. Reading Gilgameš II: Gilgameš and Huwawa. Pp. 77–86 in Literatur, Politik und Recht in Mesopotamien. Festschrift für Claus Wilcke, ed. Walther Sallaberger et al. Wiesbaden: Harrassowitz. Coad, Brian W. 2010. Freshwater fishes of Iraq. Sofia: Pensoft.

79 I have discussed the question of hybrids elsewhere: see Verderame 2012, 2015 with previous bibliography. The iconographic sources for the Sebettu are limited to the First Millennium BC; they show seven human bodies with animal heads.

414

Lorenzo Verderame

Cohen, Mark E. 1973. The identification of the kušû. JCS 25: 203–10. Cohen, Mark E. 1981. Sumerian Hymnology: The Eršemma. HUCA Suppl. 2. Cincinnati: Hebrew Union College. Cooper, Jerrold S. 1983. The Curse of Agade. Baltimore − London: The John Hopkins University Press. Collins, Billie J. (ed.). 2002. A History of the Animal World in the Ancient Near East. HdO I/64. Leiden − Boston − Köln: Brill. Couto Ferreira, M. Erica. 2009. Etnoanatomía y partonomía del cuerpo humano en Sumerio y Acadio. El léxico ugu-mu. PhD. Dissertation, Universitat Pompeu Fabra. Edzard, Dietz O. and Claus Wilcke. 1976. Die Ḫendursanga-Hymne. Pp. 139–76 in Kramer Anniversary Volume. Cuneiform Studies in Honor of Samuel Noah Kramer, ed. Barry L. Eichler. AOAT 25. Kevelaer / Neukirchen-Vluyn: Butzon & Bercker / Neukirchener Verlag. Focke, Karen. 1998. Die Göttin Nin-imma. ZA 88: 196–224. Focke, Karen. 1999/2000. Die Göttin Ninimma. II. Ninimma in der keilschriftlichen Überlieferung des 3. Bis 1. Jts. V. Chr. AfO 46–47: 92–110. Geller, Markham J. 1989. 1985. Forerunners to Udug-hul : Sumerian exorcistic incantations. FAOS 12. Stuttgart: Franz Steiner. Geller, Markham J. 1989. A New Piece of Witchcraft. Pp. 193–205 in DUMU-E2-DUB-BA-A: Studies in honor of Åke W. Sjöberg, eds. Hermann Behrens et al. Occasional Publications of the Samuel Noah Kramer Fund 11. Philadelphia: The University Museum. George, Andrew R. 2003. The Babylonian Gilgamesh Epic, I–II. Oxford − New York: Oxford University Press. George, Andrew R. 2015. The Gods Išum and Ḫendursanga: Night watchmen and streetlighting in Babylonia. JNES 74: 1–8. Gössmann, Felix. 1950. Planetarium Babylonicum oder die sumerisch-babylonischen SternNamen. ŠL IV/2. Rome: PIB. Hallo, William W. 1973. Choice in Sumerian. JANES 5: 165–72. Heimpel, Wolfgang. 1968. Tierbilder in der sumerischen Literatur. StPohl 2. Roma: Pontificium Institutum Biblicum. Ingold, Tim. 1994. Introduction. Pp. 1–16 in What is an Animal?, ed. Tim Ingold. London − New York: Routledge. Kienast, Burkhart. 2003. Iškar šēlebi : Die Serie vom Fuchs. FAOS 22. Stuttgart: Franz Steiner. Klein, Jacob. 1981. Three Šulgi Hymns. Ramat-Gan: Bar-Ilan University. Kurtik, Gennadij E. 2007. Звездное небо Древней Месопотамии [The Star Heaven of Ancient Mesopotamia]. St. Petersburg: Aletheia. Landsberger, Benno. 1962. The Fauna of Ancient Mesopotamia, II. MSL VIII/2. Rome: Pontificium Institutum Biblicum. Leichty, Erle. 1970. The Omen Series Šumma Izbu. TCS IV. Locust Valley: J. J. Augustin. Maul, Stefan M. 1994. Zukunftsbewältigung : Eine Untersuchung altorientalischen Denkens anhand der babylonisch-assyrischen Löserituale (Namburbi). BaF 18. Mainz am Rhein: Philipp von Zabern. Neufeld, Edward. 1973. Fabrication of Objects from Fish and Sea Animals in Ancient Israel. JANES 5: 309–24. Neumann, Jehuda and Simo Parpola. 1983. Wind Vanes in Ancient Mesopotamia, About 2000–1500 B.C. Bullettin American Meteorological Society 64: 1141–43. Pastoureau, Michel. 2004. Une histoire symbolique du Moyen Âge occidental. Paris: Éditions du Seuil.

The Seven Attendants of Hendursagˆa

415

Peterson, Jeremiah. 2007. A Study of Sumerian Faunal Conception With a Focus on the Terms Pertaining to the Order Testudines. PhD. Dissertation, University of Pennsylvania. Piccaluga, Giulia. 1968. Lykaon, un tema mitico. Roma: Edizioni dell’Ateneo. Reynolds, Frances. 1998. Unpropitious Titles of Mars in Mesopotamian Scholarly Tradition. Pp. 347–58 in Intellectual Life of the Ancient Near East: Papers presented at the 43 rd Rencontre Assyriologique Internationale Prague, July 1–5, 1996, ed. Jiri Prosecký. Prague: Academy of Sciences of the Czech Republic. Rubio, Gonzalo. 1999. On the Alleged “Pre-Sumerian Substratum.” JCS 51: 1–16. Seminara, Stefano. 2001. La versione accadica del Lugal-e. La tecnica babilonese della traduzione dal sumerico e le sue “regole”. MVS 8. Roma: Università degli Studi di Roma “La Sapienza.” Sibbing Plantholt, Irene. 2015. Black Dogs in Mesopotamia and Beyond. In From the Four Corners of the Earth. Studies in the Iconography of the Ancient Eastern Mediterranean and Near East in Honour of F. A. M. Wiggermann, eds. David Kertai et al. PIHANS. Leiden: Nederlands Instituut Nabije Oosten (in press). Sjöberg, Åke W. 1973/74. Miscellaneous Sumerian Texts, I. OrSuec 23/24: 159–81. Sperber, Daniel. 1975. Rethinking Symbolism. Cambridge: Cambridge University Press. Sperber, Daniel. 1996. Why Are Perfect Animals, Hybrids, and Monsters Food for Symbolic Thought? Method & Theory in the Study of Religion 8: 143–69. Thompson, Reginald C. 1903. The Devils and Evil Spirits of Babylonia, I. “Evil Spirits.” London: Luzac. Tullio-Altan, Carlo. 1991. L’interpretazione dei simboli in antropologia. Pp. 225–58 in Tra uomo e animale, ed. Ernesta Cerulli. Bari: Dedalo. Vanstiphout, Herman L. J. 1988. The Importance of 'The Tale of the Fox'. ASJ 10: 191–227. Veldhuis, Niek. 2004. ḪI-(še3) la2. CDLB 2004.4. http://cdli.ucla.edu/pubs/cdlb/2004/ cdlb2004_004.html. Verderame, Lorenzo. 2011. L’immagine della città nella letteratura sumerica. Pp. 99–126 in Città nel Vicino Oriente e nel Mediterraneo. Linee di storie e di simboli dall’antichità ad oggi, ed. Rita Dolce and A. Pellitteri. Palermo: Flaccovio. Verderame, Lorenzo. 2012. “Their Divinity is Different, Their Nature is Distinct!” Nature, origin, and features of demons in Akkadian literature. Archiv für Religionsgeschichte 14: 117–27. Verderame, Lorenzo. 2015. On the Early History of the Seven Demons (Sebettu). In From the Four Corners of the Earth. Studies in the Iconography of the Ancient Eastern Mediterranean and Near East in Honour of F. A. M. Wiggermann, eds. D. Kertai et al. PIHANS. Leiden: Nederlands Instituut Nabije Oosten (in press). Volk, Konrad. 1995. Inanna und Šukaletuda. Zur historisch-politischen Literaturwerkes. SANTAG 3. Wiesbaden: Harrassowitz. Walravens, Hartmut (ed.). 2002. Der Fuchs in Kultur, Religion und Folklore Zentral- und Ostasiens. Wiesbaden: Harrassowitz. Weszeli, Michaela. 2007. Rabe(nvögel). RlA 9: 210–13. Wiggermann, Frans A. M. 2011. The Mesopotamian Pandemonium. SMSR 77: 298–322. Wilcke, Claus. 1969. Das Lugalbandaepos. Wiesbaden: Harrassowitz.

Christopher Woods

The Abacus in Mesopotamia: Considerations from a Comparative Perspective It is generally assumed, but rarely discussed, that calculations in early Mesopotamia were carried out not with cuneiform numerals, but – following a seemingly universal practice – with the use of an abacus of some kind. That is, the numbers so abundantly present in cuneiform texts do not directly represent calculations themselves, but are the records of computations carried out in a different, operational, medium. The only recent discussions that consider this issue in any detail have been by Høyrup (2002a) and Proust (2000), who have investigated what calculation errors reveal about mathematical procedures, and the implications that these have for the design of calculation devices in Mesopotamia. Although any conclusions remain necessarily provisional in the absence of an archaeologically attested instrument, or textual descriptions for its use, this article will explore what more can be said about early counting devices in Mesopotamia by drawing upon the scant but suggestive evidence presented by textual sources and numerical notation itself. The goal of this paper is primarily to remind the reader of the central role these devices likely played in Mesopotamian numerical life, and to explore their potential parameters by considering the evidence in light of cross-cultural and historical parallels as well as the cognitive advantages and limitations posed by such counting aids. As such, this study is very much a first approximation rather than a definitive investigation of the abacus in Mesopotamia. The central assumption

Acknowledgements: I would like to thank Profs. David Barner, Michael C. Frank, and James W. Stigler, who availed their time to answer numerous questions about the cognitive aspects of abacus practice. I also acknowledge the assistance of Totton Heffelfinger and, particularly, Stephen Stephenson, who shared their knowledge of abacus design. A particular debt of gratitude is owed to Master Jeonghee Lee, one of the world’s foremost abacus masters, and her associates, Kevin D. Kim, Pema Kongpo, and Woo Song, who responded to numerous questions by phone and email about contemporary abacus methods. Certainly this paper has benefited greatly from their assistance. Finally, I thank Jennie Myers, Piotr Steinkeller, Andy Wilent, and especially Jens Høyrup and Christine Proust, for their comments, suggestions, and corrections on an earlier draft. Stephen Chrisomalis and Wayne Horowitz kindly provided me with references and drafts of their unpublished work. It goes without saying that all errors are mine alone. https://doi.org/10.1515/9781501503696-023

The Abacus in Mesopotamia: Considerations from a Comparative Perspective

417

behind the arguments presented here, reasonable if unproven, is that the computational devices used in Mesopotamia operated on the same general principles as those broadly attested across a range of cultures, from the Greco-Roman world, to Russia, China, and Japan. Needless to say, it is an honor to devote this study to Miguel Civil, the great master of all things Sumerian. Beyond the groundbreaking studies that cover every aspect of Sumerology, a perpetual source of amazement for me personally has been Miguel’s uncanny ability to drop the most penetrating of observations into casual conversation or to bury one of these gems of insight in a footnote. Perhaps it relates to a topic to which you have devoted much time and energy, convinced that your efforts would be rewarded with some meaningful contribution to the field – only to learn that Miguel had figured it all out long ago, having distilled the problem to it essence and reported his results almost as an afterthought. As will become apparent, this article draws its inspiration in no small part from one such encounter, namely, a note on early calculation devices in his article “From the Epistolary of the Edubba” in honor of W. G. Lambert (Civil 2000: 115–116).

1 The Abacus – A Description The abacus has a venerable history in Europe and Asia, prior to the modern era, as the primary instrument for effecting calculations that strained the limits of mental reckoning.1 In Europe it is attested from Greco-Roman times, and remained in common use through the 18th century. In China the abacus, the suanpan, was certainly well established by the 16th century, while 14th-century illustrations likely push the date of the first evidence back by at least two centuries. Paleography, however, may suggest much earlier origins for an ancestor of the device. At the root of suanpan (literarally, ‘counting board’), is the verb suan ‘to calculate’, which is represented by a graph that may depict a counting board or, perhaps more likely, counting rods (Fig. 1),2 as its central element 1 For a general history and description of the abacus, see Adkins 1956; Barnard 1916; Burnett and Ryan 1988; Jami 1998; Kojima 1954 and 1963; Kwa 1922; Menninger 1969; Moon 1971; Pullan 1969; Reynolds 2008; Tani 1964. Schärlig 2001 represents the most comprehensive treatment. 2 Counting rods (suan or chou) constituted a base-10 positional system developed for representing numbers in computations by a series of short sticks oriented horizontally and vertically and manipulated on a flat surface (in general see Chrisomalis 2010: 264–70; Martzloff 2006: 210–11; Needham 1959: 70–72). Literary references and the appearance of counting rod notations on coins demonstrate that this system was in existence from at least the Warring States

418

Christopher Woods

(算) – the graph is attested already in the 4th century BC.3 It was probably not until the 15th century that the suanpan was transmitted to Korea and Japan (Jami 1998: 3–4; Martzloff 1997: 215). The device remains an essential calculating instrument in Asia, having only lost ground in recent decades owing to the advent of electronic calculators. In the Andes the abacus took an entirely different form as attested by the Peruvian yupana used by the Inka, on which calculations were carried out that were recorded with the knotted-string khipu (Fig. 2a, Fig. 3).4 Although there is no consensus on the functional principles of the yupana, these devices may have represented storehouses in miniature – their striking geometric grid structure possibly reproducing the architecture of grain storage facilities (Brokaw 2010: 84). In Ecuador, stone counting boards known as taptana (Fig. 2b) – boasting a similar antiquity as their Peruvian counterparts but differing in organization and design – remain in use to this day.5 The abacus likely has a far greater antiquity than the earliest textual references and archaeological evidence currently admit (see § 1.3). The term abacus technically refers to two devices: the counting board, otherwise known as the western abacus, which consists of a flat surface on which counters were moved, and the wooden-framed, fixed-bead eastern abacus, the instrument most closely associated with the term in common parlance. To add to the confusion in nomenclature, the Roman term abacus, refers specifically to the counting board, being derived from the Greek ἄβαξ denoting broadly a flat surface.6 In this article, I will use the term abacus to refer to both devices, but will specify either counting board (western abacus) or fixed-bead, or fixedcounter device (eastern abacus) when a distinction is warranted. The counting

period (475 − 221 BC). As observed by Martzloff (2006: 211–121), similarities between the counting rod notation and that of the suanpan – and, in particular, the special role of 5 as a subbase (§ 1.2) – are striking and may speak to the origins of the abacus in China. 3 I thank my colleague, Prof. Edward Shaughnessy, for pointing out this early attestation and clarifying the Chinese etymology (see also Martzloff 2006: 210; Menninger 1969: 310; Needham 1959: 4). 4 Salomon 2002: 310, with previous literature. 5 Quilter 2002: 215; Tun and Montaluisa Chasiquiza forthcoming. 6 The Greek term is often claimed (e.g., Nesselmann 1842: 107 n. 5) to stem from Semitic abq meaning ‘dust’, a derivation that is not without complications. While Classical and Medieval sources make reference to boards covered with dust, these were used to draw geometrical figures and later numerals as a precursor to pen reckoning; in fact, when Hindu-Arabic numerals were first introduced to Europe in the 10th century, they were referred to as “dust numerals” (Burnett and Ryan 1998: 5). The counting board in its basic, and likely original, form is a flat, even surface that facilitates the sliding of counters – dust likely played no role, and, indeed, would have been a hindrance (see Menninger 1969: 301; Pullan 1969: 17–18).

The Abacus in Mesopotamia: Considerations from a Comparative Perspective

419

board is presumably the older instrument, having given rise to the fixed-bead device. The two instruments operate on the same basic principles of spatial reckoning in which numerical value is relative and positional within the device. The two differ primarily in the movement of the counters, with the fixed-bead device representing, essentially, a moving counting board. Abaci of both kinds are used to perform the basic operations of arithmetic – most commonly addition and subtraction – but also multiplication and division. Skilled users are able to perform more complicated operations such as extracting square- and cube-roots.

1.1 The Counting Boards of the West The western abacus, or counting board, consists of a series of parallel lines upon or between which counters are placed.7 Value is assigned to these lines or the spaces between them according to the rules of replacement – the bundling steps by which lower order units are replaced by higher order units at regular intervals – for the underlying numerical system. For instance, the typical counting board used in the West from Classical to early modern times has decimal columns representing I, X, C, and M (or their Greek acrophonic counterparts), with intermediate quinary columns or lines, corresponding to V, L, D values (Figs. 4–9). Consequently, there is a complete one-to-one relationship between the positions on the board and the replacement values of Greek and Roman numerals: five I’s are replaced by one V, two V’s are replaced by one X, etc. In other words, the numerals and the abacus are both base-10 systems that employ a sub-base of 5, allowing numbers to be expressed in terms of alternating replacements of 5 and 2 (see Fig. 10). Each counter on the board corresponds to a sign of the numerical phrase; for instance, the number CXXIII would be represented by six counters: one of value 100, two of value 10, and 3 of value 1. The counters themselves have no intrinsic numerical value, but rather acquire the value of units, or groups of units, based on their placement on the board. There is no direct representation of zero on these devices; rather, zero is indicated by the absence of a counter in a given position. Extant examples from the Classical world occasionally attest markings for numerical values, but generally these are not indicated. This allows the same board to be easily

7 The most recent and comprehensive treatment of the abacus in antiquity lists 30 known objects from the Classical world, mainly from the Aegean, although the identification of all of these objects as counting boards is not certain (Schärlig 2001: 61–104).

420

Christopher Woods

adapted to accommodate various numerical bases and metrological systems by applying to neighboring positions the necessary replacements, since value is inherently positional and relative. The western abacus is a simple device. It may be as simple as a few crude lines scratched on the bare earth with pebbles used for counters – and, indeed, the device may have such humble origins (Netz 2002: 327, 332–33; Pullan 1969: 18). The improvised nature of such devices, coupled with the use of perishable materials to fashion counting boards in other instances, as well as the eventual erosion of distinctive grid lines for those etched on stone, likely account for the scant attestation of counting boards in the archaeological record. The operation of the counting board is based solely on the replacement values of the numerical system being represented. There are no limits to the number of counters that may be placed on the board; the only rule of operation is that once the number of counters in a column reaches the corresponding replacement value, it is replaced by one counter in the adjacent, higher-order column. In this way, basic arithmetic, particularly addition, is reduced to a mechanical process that requires nothing more than an ability to count to ten. Addition, for instance, is effected by representing individual addends as counters, which are placed in the appropriate positions on the board (Fig. 4, Fig. 7). As the counters accrue, they are simply replaced when there are enough to represent a counter in the next larger unit. On the typical counting board with alternating decimal and half decimal positions, addition of even very large numbers is reduced to a series of sums that are limited by five and two respectively. The counting boards of Europe, essential instruments of commerce and science between the Middle Ages and early modern times, operated on the same principles and utilized the same numerical system as their Roman ancestors (Fig. 4, Figs. 11–18). Differences were primarily cosmetic – whereas the Classical abacus had columns oriented vertically to the user, the counting boards of Europe often exhibited a horizontal arrangement and varied in the representation of decimal and quinary units on lines or the spaces between them.8 The pebbles and calculi of glass, bone, or ivory of the Greco-Roman world were replaced by brass counters, called jetons (< Fr. jeter ‘to cast, throw’; see Fig. 18). The counting board was used in Europe throughout the Middle Ages until the 16th century when it began to give way to pen and paper reckoning, the ascendency of the new method owing in part to the increasing availability of paper. The legacy of the device lives on in English words such as

8 Lang 1957: 275–76; Pullan 1968: 58–60; Schärlig 2001: 144.

The Abacus in Mesopotamia: Considerations from a Comparative Perspective

421

‘counter’ (as in countertop), ‘check’ or ‘cheque’ (‘exchequer’), ‘calculate’, ‘calculus’, and the idiom ‘to cast accounts’ (cf. jeton < Fr. jeter).9

1.2 The Fixed-Bead Devices of the East The eastern abacus is a more complex device in terms of both construction and operation. These same characteristics, however, make the eastern abacus the more agile and portable instrument, and one that is more capable of rapid calculation and better equipped to handle large numbers. Unlike the counting board, on which an unlimited number of counters may be placed and moved from column to column in accord with the corresponding replacement values, the eastern abacus is a fixed-counter device that does not permit the movement of counters between columns. Operation, therefore, requires procedures that involve the carrying-over of counters from one column to the next, borrowing, replacement, and equivalences between columns, all of which account for the greater complexity of the device (Stigler 1984: 149–150). The abacus typical of east and southeast Asia is a rectangular wooden-framed instrument containing columns of rods, grooves, or strings, which are oriented vertically and bear beads; running parallel to the length of the frame is the traverse bar, which divides the abacus into upper and lower registers. Like its western counterpart, the eastern abacus relies on place value and the base-10 system of numeration, whereby each column is assigned to an incrementally increasing power of 10 from right to left, corresponding to a column of a base-10 number. And, like the counting board, the abaci of Asia employ intermediate quinary positions and so make use of an intra-exponential sub-base of 5 within a base-10 framework. The distinction between integers and fractions – the “decimal place” – on the abacus, may be assigned arbitrarily to the space between any two columns since columns have no fixed value, only a value relative to one another. The Chinese abacus, the suanpan, in current use is by design a 2 : 5 device in that it consists of two beads in the upper register, each with a weight of 5 ×, and five beads, each representing 1 ×, in the lower register (Fig. 19). Consequently, each rod has a maximum value of fifteen for a given multiple of ten. As with the counting board, there is no direct representation of zero. Counters of the upper and lower registers only attain value when they are engaged and brought towards the traverse bar; moving counters away from the traverse bar clears the column to zero. The Japanese soroban differs in being a 1 : 4 device in which the single upper register bead, known as the “heavenly” bead, has a

9 Pullan 1969: xi–xii, 72–73.

422

Christopher Woods

weight of 5 ×, while the four beads on the lower register, the “earthly” beads, have a value of 1 × (Fig. 20).10 The soroban represents a minimal abacus design since the total potential value of each column is the smallest possible number, 9. By necessity, the number of beads per column on a minimal abacus, for a numerical system of a given base, b, is b – 1 (Moon 1971: 76). On the soroban, each column has a maximum value of 9 × 10n (where n = the power of ten assigned to the corresponding column of beads); the number 10 is represented by engaging the first lower register bead of the adjacent column. A minimal abacus involves the manipulation and apprehension of fewer beads and any given number has a unique specification as there is a one-toone relationship between numbers and their representation on the device. However, these same qualities require that procedures used for addition must likewise be unique. This often results in complex movements of beads that may involve as many as three parts of the abacus to effect the simple addition of two numbers (Moon 1971: 76, 94).11 A non-minimal design, conversely, requires

10 Martzloff (1997: 212) cautions that the minimal Japanese design should not be taken as an evolutionary improvement; indeed, abaci other than the 2 : 5 design were known in China in earlier periods. Notably, in Japan a 1 : 5 soroban was in use until 1930 (Moon 1971: 84). 11 The bead movements and procedures that typify an abacus of minimal design may be exemplified by several simple additions on the soroban (modeled after Stigler 1984: 149–150).⁠ For instance, adding 3 + 1 is trivial as it requires adding one bead to the three already moved to the traverse beam so that all four beads of the lower register are engaged. However, adding 4 + 1 is somewhat more complicated as there are not enough beads in the lower register. The solution is to engage the upper register bead and subtract the four lower beads by returning them to the zero position. The procedure for such inter-column additions when the addend cannot be directly summed is to subtract the complement of that addend to 5 (in this case the complement of 1 is 4), and to add 5 by engaging the bead in the upper register. More generally, if x+y cannot be added directly within a column, the addition is rendered as x ˗ (5 ˗ y) + 5, which in this example would be 4 ˗ (5 ˗ 1) + 5 = 5. Additions of sums greater than 9 require carryingover to the adjacent column and therefore the next place value. The rule in cases in which x + y cannot be added in a single column is to subtract the complement of y to 10, and then add one bead from the adjacent column, which has the value of 10, or, x ˗ (10 ˗ y) + 10. For instance, in adding 8 + 4, one first sets the 8 in the device by engaging the 5 bead and three unit beads. Adding the 4 encounters the problem that only one unit bead is now available in lower register. The solution is to subtract the complement of 4 to 10, which is 6, and to add 10, which is represented by one bead from the adjacent lower register: 8 ˗ (10 ˗ 4) + 10 = 12. Frequently, sums require applying both rules and necessitate carrying-over within the column and to the adjacent column. For example, adding 6 + 7 runs into the difficulty of attempting to subtract 3, the complement of 7, since only one unit bead is engaged from the setting of 6. To subtract 3, 5 has to be first subtracted (i.e., the upper register bead set to the zero position), and then two beads (the complement of 3 to 5) added from the lower register. The combined rule for sums of this kind is: x + (5 ˗ (10 ˗ y)) ˗ 5. However, theoretical descriptions and mathematical formulae obscure the inherent simplicity of abacus operations. The seemingly com-

The Abacus in Mesopotamia: Considerations from a Comparative Perspective

423

the manipulation and reading of more beads, but also allows for greater flexibility in the representation of numbers and permits certain short-cuts in the procedures for addition that avoid triple-bead movements (Moon 1971: 76). In the case of the Chinese suanpan, the one extra bead in the upper and lower registers has the additional advantage of accommodating hexadecimal numerals that were used in weight metrology (Martzloff 1997: 211). While the abacus could be theoretically constructed with any number of columns, tradition dictates certain conventions. The Chinese suanpan typically has 13 rods while the soroban has 21 or 23 (Jami 1998: 3; Stigler 1984: 147). Despite the distinction commonly made between eastern and western abaci, remarkably, a fixed-bead device was known in the West at least since Roman times.12 The four known Roman specimens are portable, or pocket, abaci that consist of small bronze plates, roughly 10–14 × 7–10 cm in size, with decimally-assigned slots that hold attached sliding beads (Fig. 21). An example housed in the Museo Nazionale Romano, dates to the first century AD, and, like the other three, represents a minimal 1 : 4 device, with one bead in the upper register with a value of 5 ×, and four beads in the lower register each representing 1 ×. The device, then, is in design remarkably similar to the Japanese soroban. In Russia, a bead-framed abacus, schety (from schitat ‘to count’; cf. Sum. NIG2.ŠID devices discussed in § 3), operating on quite different principles than its Roman, Chinese, or Japanese counterparts, is attested from the early 17th century onwards, though there are indications that the device is of much older origin (Burnett and Ryan 1990: 6). Each horizontal wire on the Russian schety represents a power of 10 and bears 10 beads (Fig. 22). As such, the schety is purely a base-10 instrument without an intra-exponential base-5 substructure. While it is clear that the Japanese soroban derives from the Chinese suanpan both in design and name, these instruments have origins that are almost certainly distinct from both the Russian schety and the Roman pocket abacus (Burnett and Ryan 1998: 7; Hénin 2012: 206). The independent invention of at least three fixed-bead abaci speaks to the universality of such devices, and the instinctive human tendency to enlist beads in the service of counting.

plex example of 6 + 7 is accomplished with three simple flicks of the fingers. With practice and increasing expertise, calculations on the abacus become completely automated and subconscious. Chen (2013 and forthcoming) details the history of abacus education as it pertains to the evolution of bead manipulation procedures in China and Japan between the 15th and 20th centuries. 12 See Burnett and Ryan 1998: 5; Fellmann 1983; Hénin 2012: 205–206; Pullan 1969: 89; Schärlig 2001: 123–26; Schärlig 2003b.

424

Christopher Woods

1.3 Counting and Cognition Although archeological and textual evidence is relatively late, the abacus – at least in its most basic form of crude lines scratched on the earth – is probably as old as arithmetic itself. Counting, it is generally held, began with fingers or with fingers and toes, hence decimal and vigesimal number systems, and proceeded to the use of pebbles or other readily available discrete objects to represent higher numerosities.13 The abacus expands the envelope for efficient counting farther still, while extending the limits of what is possible by means of mental reckoning alone. Ultimately, the device is based on the inter-related concepts of replacement, number base, and place value.14 For instance, we can envision a system of pebble counting whereby 10 small pebbles are replaced by 1 medium-sized pebble, 10 medium sized pebbles are replaced by 1 large pebble, and so on. In this way, the limit that can be counted efficiently is extended (e.g., the number 257 would be represented by 14 pebbles – 2 large, 5 medium, and 7 small – rather than 257 small pebbles). The system of sequential replacements at a set interval has the consequence of establishing an additive number base equal to the replacement number, in this case 10. In such a system, where units are distinguished by size, it is natural to separate units, maintaining distinct containers for small, medium, and large pebbles. A system that distinguishes size could provide the basis for a system that is positional – that is, place value replacing size value. The first step would be establishing a consistent order of these containers. Since value would then be defined by both size and position, redundancy could be eliminated by only using pebbles of one size, in which case value would be determined solely by the container in which it is placed. In this way, this system becomes a purely positional, or place-value, one. In effect, we have invented an abacus and achieved a level of abstraction in terms of the representation of number. The ordered containers define numerical value, while the pebbles of uniform size serve as counters with no intrinsic value. The abacus, then, emerged from counting, and its development cross-culturally may have been closely intertwined with the origins of various early systems of numerical notation, specifically, those that are cumulative-additive in structure, such as the Roman, Egyptian and Cretan hieroglyphic, Aztec, and proto-cuneiform numerals,15 among others. As described by Chrisomalis (2010: 11–12), cumulative-additive systems are cumulative in that intra-exponentially 13 E.g., Flegg 1983; Gullberg 1997; Ifrah 2000; Menninger 1969; Rudman 2007. 14 The following description is based on Rudman (2007: 31–33). 15 That is, prior to the advent of sexagesimal place value notation at the end of the third millennium, which rendered the system cumulative-positional.

The Abacus in Mesopotamia: Considerations from a Comparative Perspective

425

(i.e., within each base), graphs are iconically repeated and their values are added. They are additive in that inter-exponentially, each power base is represented by a distinct sign, and the total of the entire numerical expression represents the sum of the intra-exponential values. For example, the Roman numeral CCCLXXVIII, consists of three 100s (102), one 50 (5 × 101), two 10s (101), one 5 (5 × 100), and three 1s (100), which together expresses 378. These numeri, cal systems do not reflect spoken language – the iconic repetition of for instance, has nothing to do with the Sumerian word /eš/ ‘three’ – and, as has often been commented in connection with Roman numerals, are cumbersome and ill-suited to the purposes of arithmetic.16 Rather, these numerical systems are best accounted for within the context of tallying and the abacus.17 Taisbak, for instance, makes the bold claim that “the notation of Roman numerals originates from the abacus reckoning” (1965: 158 [emphasis in the original]). Similarly, Netz (2002: 329), in comparing the transitory use of counters to the permanence of numerical notation, observes that counters “of course cannot function as a medium for permanent record, but their centrality as a medium for manipulation was such that, in the classical Greek world (as in the Roman world, well into early modern times) their use in the abacus shaped the form of the medium of record itself.” The fate of Roman numerals would remain bound to the abacus until the end of the Middle Ages. When Hindu-Arabic numerals were introduced in the West at the end of the first millennium, they did not instantaneously replace the cumbersome archaic Roman system (Fig. 16). On the contrary, five hundred years would pass before the new system became commonplace and overtook the old one (Netz 2002: 323; Pullan 1969: 30, 36, 42). One of the driving forces behind the perseverance of Roman numerals was the counting board, which reduced basic arithmetic to a mechanistic process that could be mastered without the requirement of literacy or formal education. In fact, when Hindu-Arabic numerals were first used in Europe they could be manipulated in much the same way as counters on a counting board.18 Whether it was the numerical notation that emerged from the abacus

16 See Chrisomalis 2010: 116–18, with previous literature. 17 Rudman (2007: 129–30) describes cumulative-additive numerical systems, such as Roman, Egyptian hieroglyphic, and proto-cuneiform, as “abacus notations.” 18 Gerbert d’Aurillac, who would become Pope Sylvester in 1000 AD, is credited with devising a hybrid counting board, the “abacus with apices,” which made use of counters that were individually marked (the apex) for each of the Hindu-Arabic numerals from 1 to 9. The device paved the way for pen-reckoning first on parchment (Burnett and Ryan 1998: 5; Chrisomalis 2010: 220; Schärlig 2001: 135–40; more generally, see Brown 2010) and eventually on paper. In the centuries before the ascendency of Hindu-Arabic numerals, the use of counters was taught along-side pen-reckoning (see Fig. 16).

426

Christopher Woods

or the abacus that gave rise to the numerical notation may be debated, but the two systems are no doubt related to one another and are rooted in concrete counting.19 Both share the same decimal and half-decimal structure, while the inter-exponential iconicity of the numerical system finds an exact correspondence in the tangible counters of the abacus for a given exponentially assigned column – that is, there is a one-to-one relationship between each counter on the abacus and a graph of the numerical notation for a given power (see § 1.1). In short, cumulative-additive systems are the clearest means of representing the notation of the abacus. The structural similarity between various cumulative-additive notations may, more generally, suggest a functional similarity that binds these numerical systems to abacus practice. The power of the abacus – attested by it prevalence and perseverance even in the face of more technologically sophisticated rivals – lies in representing the abstract number concept in concrete, spatial terms that are non-linguistic and tailored to human visual-cognitive capacities (see Frank and Barner 2011; Netz 2002: 344). It is a deceptively simple device that leverages the primacy of visual perception to extend the natural limits of number processing. The inputs and outputs of abacus calculations may be linguistic, but the calculations themselves are carried out in a non-linguistic format that relies primarily on visual-motor resources. Indeed, experiments with abacus users suggest that the structure of the fixed-bead abacus itself – “rectilinear, horizontally extended, and segregated into columns” – may be optimized to the design and limitations of the human visual system (Frank and Barner 2011: 12–15). Once the basic procedures for manipulating counters are mastered – trivial with the counting board, but requiring greater practice with the eastern abacus – arithmetic becomes an almost automated process. As described above, calculations involving even large numbers are reduced to series of sub-computations involving numbers less than 10 for decimal-based devices. The abacus excels at addition, in particular, rendering it an entirely mechanical, subconscious operation that obviates the need for addition tables (Moon 1971: 41, 72, 127). Multiplication requires greater mental investment and the use of multiplication tables. The basic process is to reduce the problem to the products of all possible pairs of digits between multiplicands and to sum these partial products. With use of a multiplication table, the multiplication of even large numbers is reduced to addition on the abacus. Subtraction involves the same basic counter

19 For the origins of the Roman numerals, for which these issues have received the most attention, see the important article by Keyser (1988; see p. 540 for the relationship to the abacus).

The Abacus in Mesopotamia: Considerations from a Comparative Perspective

427

manipulations as addition; division is more complex, but may be treated as a series of subtractions. A common feature of all of these devices, both eastern and western, is that their designs reflect the limits of visual cognition, and, consequently, they all operate on general principles of subitization and chunking. Subitization refers, generally, to the fast enumeration of small quantities – specifically, the cognitive ability to instantly and accurately perceive a number of similar, proximate objects without resorting to manual counting (e.g., Dehaene 1993: 12–14). Early experiments by Kaufman et al. (1949), who coined the term, found this limit to be seven, based on the number of dots test subjects could accurately perceive when patterns were flashed on a screen for one-fifth of a second. This finding was reinforced in the seminal paper of Miller (1956), “The Magical Number Seven, Plus or Minus Two,” often taken as the departure point for modern cognitive science, which maintained seven items, within the broader range of five through nine, as a capacity limit of human cognition. More recent studies, however, have tended to reduce this threshold to four or five, though it is subject to various factors, such as subject practice and the characteristics and arrangement of the objects being quantified, which may increase this limit (see Dehaene 1993: 14–18; Mander and Shebo 1982). The critical point is that beyond the subitization limit, the number of items can only be discerned through the mediation of explicit counting or estimating. Chunking, a term that originates with Miller’s 1956 paper, is closely related to subitizing; chunking operates across larger quantities, organizing them into easily subitized groups to facilitate cognitive processing.20 Chunking, for instance, accounts for the use of spaces and dashes in telephone, credit card, and social security numbers – breaking long 9–12 digit numbers into manageable, subitized chunks of two to four digits that are more readily committed to memory than long unbroken sequences. Indeed, maintaining reasonable chunking and subitization limits accounts for why cuneiform numerals are written with graphic arrays that rarely extend ‘5’, ‘90’), a graphic convention already attested beyond 3–5 signs (e.g., for proto-cuneiform. Chunking, in effect, extends the limits of subitization. Subitization explains the inclusion of an intra-exponential sub-base of 5 for all of the abaci discussed above that represent an underlying decimal number system – from the counting boards of the Classical world and medieval Europe, to the Chinese suanpan and the Japanese soroban. The interspersing of half decimal 5 × 10n and decimal 10n positions meant that abacus users never had to contend with clusters of more than four counters in a given position;

20 For recent literature, see, for instance, Feigenson 2008; Halberda et al. 2008.

428

Christopher Woods

for the suanpan and soroban the maximal number of counters is respectively five and four in the lower register – numbers that all fall within subitization limits, which naturally avoids manual counting and facilitates rapid computations – the raison d’être of these devices. Even the apparent exception, the Russian schety, proves the rule (Fig. 22). Each row represents a power of 10 and bears 10 beads; however, the use of color-coding divides these beads into easily subtilized “chunks” of 4-2-4 (Frank and Barner 2011: 15). As Netz (2002: 327) notes in connection with the counting board – a comment that applies equally to the eastern abacus – “One can say that the operations of the western abacus are nothing but concrete subitization: clusters of counters are seen as unities – and are replaced by unities. The abacus is effective because it breaks calculation down into a sequence of operations, mental (subitization) and bodily (opposition) that, each, require no effort at all.”

2 Counter Culture in Mesopotamia The major role played by counters in Mesopotamian numeracy and administration is widely acknowledged, particularly for the proto-literate period. Wellknown, of course, are the claims of Schmandt-Besserat (1992), who sees not only early Mesopotamian numerical notation but writing itself as evolving out of a prehistoric counter culture that was as extensive in geographical scope as it was in diachronic depth.21 We need not subscribe to the details of SchmandtBesserat’s theory of the origins of cuneiform to agree with the general proposition that counters were in many ways the driving force behind numeracy in Mesopotamia. Although it is quite unlikely that small clay artifacts assuming a variety of simple geometric shapes – excavated from sites throughout the Near East and attested from the 9th millennium through historic periods – represent a single class of objects, many were no doubt used for counting. In particular, the tokens appearing within proto-literate clay envelopes – which represent the best evidence for these objects belonging to a single, contextually-defined system – likely represent a variety of numerical and metrological systems used to count distinct classes of commodities, analogous, or perhaps even identical, to those systems attested in proto-cuneiform texts.22 Indeed, there is some evidence to connect tokens as counters to the numeration of

21 The term “counter culture” draws from Netz’s (2002) stimulating study of the role of counters in Greek numeracy. 22 Woods (forthcoming).

The Abacus in Mesopotamia: Considerations from a Comparative Perspective

429

later, historic periods. The names for 602 and 603, which are respectively, šar2 and šar2-gal, literally ‘everything’ and ‘big everything’ respectively, may trace their ancestries back to spherical tokens of two different sizes, that is, “ball” (cf. the circular impression to write 602 in early cuneiform) and “big ball,” which commonly occur in proto-literate token assemblages.23 This may suggest that the numerical notation of the counter system was based in part on sizevalue, in which physical size is iconic with value (see § 1.3).24 That these counters were ordered sequentially and hierarchically is shown by those few clay envelopes, particularly from Susa, which bear the impressions of tokens (or corresponding finger or stick marks) on their exteriors that match in number and profile the tokens contained within. In these cases, identical units are grouped together, and higher value units are iconically larger than lower value ones for a given shape and are symbolically placed above lower value units.25 This is the same numerical syntax that is attested in the numerical tablets and which would persist through the third millennium (see also Englund 1998: 112). Tokens may be convincing forerunners of the numerical systems of cuneiform, but it does not necessarily hold that these counters were used in computations any more than their successors were, or than Roman numerals were for that matter. In fact, clay tokens are unlikely candidates for use on an abacus of any kind despite claims to the contrary.26 Many of the most common shapes – spheres, ovoids, cones, cylinders, biconoids – are hardly appropriate for a counting board where stable and precise positioning are of paramount importance.27 For this reason, counting boards from antiquity through the nineteenth century have used disk-shaped counters, or jetons, which can be easily slid from one column to another but do not risk rolling across the board (Fig. 9, Fig. 18). Moreover, as described above, abaci of all types, eastern and western, work on the principle that the counters are identical and value is dictated by position, whether on a board or on a given rod or string, which allows for efficient unit replacements. Consequently, clay tokens that appear in a wide repertory of shapes and sizes (consisting of 16 types and 500 subtypes according to Schmandt-Besserat’s classification [1992: 17]) – whether they be

23 Powell 1973: 78; Powell 1987: 480–81. 24 See Lieberman 1980: 342–43; cf. Chrisomalis 2010: 242. 25 Note, for example, Susa envelope Sb 1927 (Schmandt-Besserat 1992: 126 n. 73), which bears markings for 1 large cone, 3 small cones, and 3 lenticular disks. 26 Schmandt-Besserat 1979a: 22; see also 1979b: 26. But see 1979b: 31 where it is stated that the token system simplified after the invention of writing and was reduced to “a single shape as in the modern abacus.” 27 See also Lieberman 1980: 344; cf. Powell 1995, who explores, on a hypothetical basis, various ways in which tokens could have been used for calculations.

430

Christopher Woods

solid or perforated – are inappropriate as counters on a counting device akin to an eastern or western abacus. Rather, tokens, like their numerical counterparts of writing, are best understood as archival devices, which recorded abacus notations, rather than as operational devices themselves. Clay tokens, which can be stored in various receptacles – envelopes being one known means – would provide a permanent, yet flexible, means of maintaining records of quantities. Since position is irrelevant in such a system, with units distinguished by size and shape, quantities represented in this way could be stored loosely in pouches, or similar, without danger of ambiguity, provided that shapes had contextually distinct values. As there is reason to believe that the proto-literate token systems operated on the same principles as the succeeding numerical and metrological systems of the proto-cuneiform texts, the tokens may have had the additional utility of providing some restricted information about the commodities being quantified. The invention of writing did not render tokens of hoary antiquity obsolete. Recent finds at Ziyaret Tepe, the Neo-Assyrian provincial capital Tušhan, demonstrate that tokens were used side-by-side with writing down to the end of cuneiform culture (MacGinnis et al. 2014). Tokens persevered as a parallel semiotic system because they addressed information storage needs beyond the remit of writing – offering access for illiterates, facilitating the keeping of dynamic accounts and tallies that required continual updating, and serving as an intermediary between fleeting calculations of abaci and the immutable records of writing (MacGinnis 2014: 289–290; Oppenheim 1959: 126). As noted at the outset, the archaeological record offers no certain examples of Mesopotamian abaci, while various probable factors – from the use of perishable materials to that of improvised, impermanent marks – are not encouraging for future prospects. Liverani (1983) tentatively proposed that fragments of clay bearing circular impressions, excavated from late-Uruk period Malatya in an administrative context, constituted an abacus designed for sexagesimal computations. This proposal, however, is unconvincing; the fragments (none of which can be joined) may simply represent tallies of some kind rather than a computation device. Furthermore, counters, which may have consisted of nothing more than pebbles, are easily overlooked or discarded in excavations. But here, fortunately, we are on firmer ground with regard to the evidence. The primary sources come from Nuzi, where the famous “egg-shaped” envelope and short notes in administrative texts make clear that pebbles (Akkadian abnu ‘stone’, always plural and written logographically, NA4.MEŠ) were used as counters, primarily for livestock, which could be deposited (nadû), removed (šūlû), and transferred (šubalkutu) from different receptacles (Oppenheim 1959: 125–26). Lieberman (1980: 346), argued that the corresponding Sumerian term

The Abacus in Mesopotamia: Considerations from a Comparative Perspective

431

was /imna/, literally, ‘clay stones’ on the basis of corresponding attestations in various recensions of the Old Babylonian Lu-list, which provide the title for the official responsible for manipulating counters: lu2-im-na4 na = ša abnē ‘counter man’, literally, ‘man of the clay-stones’.28 In terms of both material and shape, the Akkadian and Sumerian terms may only have been conventional. The figurative use of the term “stones” for counters made of other materials, as Lieberman (1980: 342) points out, finds a parallel in Roman calculi (“pebbles”), which were often made of glass, bone, ivory, and occasionally clay; depictions of these counters reveal them to be coin-like disks much like their successors, the jetons of the Middle Ages and beyond (Fig. 9). Excavations have unearthed a variety of artifacts that may possibly represent counters that would be appropriate for a counting board or fixed-bead device. Notable, for instance, is the presence at Abū Ṣalābīkh of more than one hundred reused potsherds that have been filed down to form crude disks averaging 3 cm in diameter (Fig. 23). The administrative context of these sherds – having been excavated from buildings that have been connected with commercial activities – led Postgate to conclude that they were used as counters.29 Unsurprisingly, rounded sherds of pottery found at Roman sites have been similarly connected with calculi (Pullan 1969: 20). More tentatively, evidence for counters used on a fixed device may be suggested, for instance, by perforated stone disks such as those found at Chogha Mish (Fig. 24). These artifacts are uniform in size, ca. 4–5 cm in diameter, and appear in the general vicinity of administrative debris such as tokens, clay envelopes, and tallying slabs; the high polish within the holes may be indicative of regular use on strings or rods (Delougaz-Kantor 1996: 277; plate 251: F–K).

3 Lexical, Literary, and Mathematical Attestations Lexical sources provide evidence for the use of a wide range of instruments for calculation purposes in Mesopotamia, although these remain for the most part of uncertain identification. Cross-cultural parallels from the ancient and premodern worlds lead us to expect tallies, counters, and certainly the ubiquitous

28 Lieberman 1980: 346, 349. Other calculating devices (and offices connected with them) include dubdimmu and šumekkû; the latter Lieberman (1980: 347, 350) tentatively connected with the abacus, but without elaboration or argument. 29 Postgate 1980: 92; see also Green 1993: 125 and figs. 5 : 1 and 5 : 2.

432

Christopher Woods

abacus to be among the realia behind these lexemes. The presence of multiple instruments, fulfilling different niches and used in distinct functional and social spheres, should come as no surprise. In China, for instance, the suanpan existed alongside counting rods between the 13th and 16th centuries before the abacus eclipsed the latter. During the period of their coexistence, as Jami (1998: 4) observes, counting rods were used by scholars in mathematics and the sciences, while the abacus was the device of popular arithmetic. In the Classical world, the counting board was the primary device for calculations, but the discovery of several fixed-bead abaci from Roman times (see § 1.2) clearly demonstrates that this instrument was also in regular, if more limited and specialized, use. For ancient Greece, Lang (1957: 282) argues for the use of two different types of counting boards; no doubt, tallies of various kinds also had their place. Finally, in the Middle Ages two versions of the counting board were employed alongside reckoning cloths and tallies. As described by Schärlig (2003) and emphasized by Proust (2003), the “abaques à lignes,” which assigned decimal values to lines and half-decimal values to the spaces between them (Fig. 4), were true computing devices. As purely base-10 instruments, lacking metrological divisions, theses devices were appropriate, not only for addition and subtraction, but also for multiplication and division, operations that were generally the purview of scholars and specialists. The “abaques à bandes,” on the other hand, typically assigned value only to the spaces, or bands, between the lines. The divisions were based on various monetary and metrological systems, which may exhibit a decimal organization for multiple units represented by the upper bands, but differ for unit subdivisions expressed by the lower bands (cf. Fig. 5, Figs. 21–22). These were primarily instruments of accounting and commerce, which, owing to their heterogeneous bases, were only suitable for addition and subtraction.30 The devices that are our concern are written as compounds of the graph ŠID, specifically, as derivatives of NIG2.ŠID (cf. nig2-ka9(ŠID) = nikkassu ‘account, calculation’). The compound, which likely refers to several related counting instruments, occurs with both the giš and tug2 classifiers and is glossed u2-tuk-ku and u2-tu-uk respectively; the word also appears syllabically as u2-tu-gu-um in the one attestation in connected texts, Dialogue 1: 16, discussed below (ex. 6). Importantly, gišNIG2.ŠID (written NIG2.GIŠ.ŠID) occurs in an Early Dynastic mathematical text (MS 3047 iii 7 [Friberg 2007: 150–53]), possibly from Fara, which provides a table with the length and width dimensions, along with the calculated areas, for a series of six similar rectangles. The

30 I thank Christine Proust for pointing out the distinction between these two Medieval abaci, and for providing me with the reference to her review (Proust 2003) of Schärlig 2003a.

The Abacus in Mesopotamia: Considerations from a Comparative Perspective

433

table concludes with a summation of the areas and the notation gišNIG2.ŠID, which strongly suggests that the calculations were carried out on this device.31 The Akkadian equivalents of NIG2.ŠID with both determinatives are the loanword uttuku and māḫiṣātu; gišNIG2.ŠID is also equated with Akkadian iṣṣi nikkassi. A further device NIG2.ŠID(-ma) also bears the translation māḫiṣātu as well as iṣṣi minûti. (1) Hh. IV 16–20 (MSL 5, p. 151–152): GIŠ.ŠID-ma iṣṣi minûti MIN nikkassi GIŠ.NIG2.ŠID uttuku GIŠ.NIG2.ŠID MIN GIŠ.NIG2u2-tuk-kuŠID māḫiṣātu GIŠ.NIG2.ŠID (2) Hh. XIX 147–148 (MSL 10, p. 132): utukku TUG2.NIG2u2-tu-ukŠID māḫiṣātu TUG2.NIG2MINŠID (3) Diri III 20–23 (MSL 15, p. 136), following entries for another administrative device, gišrinnu ‘balance’: uttuku gi[š- ] GIŠ.NIG2.ŠID māḫiṣātu (4) DIRI V 137–138 (MSL 15, p. 172): ut-tu-ku TUG2.NIG2.ŠID uttuku māḫiṣātu (5) Malku VI 153: NIG2.ŠID

māḫiṣātu

The Akkadian equivalents māḫiṣātu and uttuku belong to a common semantic field, both terms evoking connotations of weaving. māḫiṣāṭu, a plurale tantum noun, is a derivative of maḫāṣu, which has the well-known specialized meaning ‘to weave’ (CAD maḫāṣu mng. 3; cf. māḫiṣu ‘weavers’). Sumerian uttuku is likely to be derived from Uttu (TAG×TUG2), the spider goddess of weaving, presumably as an unbound regens-less genitive, (d)Uttu.ak ‘that of the goddess Uttu’, i.e., ‘belonging to the goddess Uttu’, and which was likewise the basis of the Akkadian equivalent, uttuku. Nothing certain can be claimed about how these counting devices, which relate in some way to textiles, differed from one

31 Reference courtesy of our honoree, Miguel Civil. The writing was mistakenly read as x.giš.sanga and understood as a signature by Friberg (2007: 151).

434

Christopher Woods

another. With the tug2 classifier, uttuku/uttuku plausibly denotes a reckoning cloth – that is, a textile with suitable lines that mimicked the design of a counting board and served as a portable substitute for the latter. Parallels for portable reckoning cloths are well-known from Europe, and were in use from at least the Middle Ages through the French Revolution. Barnard (1916: 232–33), for instance, describes the design and accounting use of the Munich reckoning cloths (16th–18th century), which are among the few extant examples of such objects (Fig. 25). During the reign of Henry II, accounts were calculated on a table covered with a black-lined reckoning cloth that resembled a chess-board (Fr. échiquier), from which the nouns “exchequer” and “cheque, check” derive (Burnett and Ryan 1988: 5). The one literary attestation of one of these lexemes (uttuku) supports the supposition that uttuku/uttuku represented a woven textile:32 (6) šah2 tug2u2-tu-gu-um al-tuku5-tuku5 ‘(You are) a pig weaving a reckoning cloth!’ (Dialogue 1: 16). This line appears among a string of insults that play on weaving motifs. The Sumerian verb tuku5 (TAG) regularly means ‘to weave’ in connection with textiles, and is equated with maḫāṣu in lexical texts.33 By extension, uttuku, when appearing with giš, may possibly represent a wooden counting board, or wooden-framed reckoning cloth.34 The noun māḫiṣāṭu, on the other hand, which naturally conveys plurality coupled with the aforementioned connection with weaving, could refer to a wooden-framed fixed-bead abacus. An abacus – whose salient feature may have been vertical, loom-like parallel threads bearing beads for counters (cf. the plural use of abnu for counters at Nuzi [§ 2]) – could conceivably account for the connotations of weaving and plurality suggested by māḫiṣāṭu. Consisting of strings or threads and set in a wood frame,

32 Johnson and Geller (2015: 44, 116) provide a different understanding of this line: ‘Striking the batten (against the loom) for a pig’. 33 E.g., Attinger 1993: 721 n. 2105; Waetzoldt, Textilindustrie, 140–44. 34 The Hh IV 16–17 entries GIŠ.ŠID-ma = iṣṣi minûti and GIŠ.NIG2.ŠID = MIN nikkassi may similarly refer to wooden counting boards. Analogous to the distinction between the Medieval “abaques à lignes” and the “abaques à bandes” described above, these two lexemes could conceivably represent instruments dedicated to computations involving the basic sexagesimal counting system (iṣṣi minûti) and those structurally based on the various metrological systems of administration (iṣṣi nikkassi). It cannot be excluded, however, that the terms refer to wooden tallies of various kinds or perhaps even to computational devices similar to the counting rods known from China (see n. 2).

The Abacus in Mesopotamia: Considerations from a Comparative Perspective

435

such an abacus might take either the tug2 or giš classifiers, if indeed the two designations refer to the same device.35 Evidence for a fixed-bead abacus – or at least for the use of strung beads as a counting device – is found in a particular expression occurring in literary texts. In the Lambert Festschrift, Miguel Civil (2000: 115–16) drew attention to the idiom gu-dili-a(–e3), which occurs in Let. 3.3.17 (Letter of Lugal-ibila to Lugal-nisag) in a broken context, translating it as ‘to string together on a single thread’. The idiom occurs in at least three other literary texts where it has been typically translated along the lines of ‘to put in a row’. But as Civil shows, the trope is literally based on a method of counting that involved beads on a string. In all cases, the idiom refers to a head count in terms of the tallying of workers or a population census. The imagery may liken individuals, ordered in single file for a count, to the columnar beads of the counting device itself (in particular, see ex. 9). A critical piece of evidence cited by Civil for establishing the meaning of the idiom is the gloss e3 = šakāku ‘to thread, to string’ appearing in the In-nin-ša3-gur4-ra passage cited below.36 The gloss demonstrates that the counting device in question made use of beads that were threaded on strings – in other words, it was a type of abacus. Beyond the obvious parallels presented by the eastern abaci, we may point to other instruments that make use of counters on strings which represent orders of magnitude, and so constitute abaci in function if not prototypical shape. For instance, Menninger (1969: 255) describes the chimpu of Bolivia and Peru, which consists of decimally-assigned strings that bear perforated seeds for counters (Fig. 26). Similarly, Ifrah (2000: 125) notes the African use of pierced stones on sticks representing orders of magnitude, as well as threaded pearls or shells among the indigenous peoples of the Pacific and the Americas.37 Naturally, there are also those devices that make use of knots, such as the famous Inka khipu, and similar instruments mentioned in ancient Chinese literature (Martzloff 2006: 179; Needham 1959: 69). However, the semantics of the verb e3 ‘to thread, to string’ would appear to exclude this possibility in our case.

35 Civil speculates that the device represented by GIŠ.NIG2.ŠID in lexical lists may be an abacus frame (Civil 2000: 116). 36 The verb e3 is glossed ša-ak-[…] in CBS 15203. For e3 with the meaning ‘to thread, to string’, Civil cites: na4kišib a2-šuba šu u3-me-ti he-me-da u3-me-ni-e3 ‘You take hold of a seal of jasper, you string it on red yarn’ (STVC 16 rev. 1 f. [incantation]). 37 Compare the Madagascar tallying practice described by Ifrah (2000: 125), whereby a makeshift abacus, consisting of pebbles and decimally-assigned furrows that have been cut into the earth for the purpose, was used to count individuals as they pass in single file through a narrow passage.

436

Christopher Woods

(7) mas-su an ki-a en dNu-nam-nir-ra sag zid sag kal-la mu-ni-ib-sa4-sa4 sag-bi gu-dili-a mu-un-e3-de3 (vars. mu-un-ne-ed3-de3-a//mu-un-ni-ibdu8) dingir-re-e-ne-ra šukur2-še3 mu-un-dab5-be2 dEn-ki-ke4 gišal-ani zag-mi2 ba-an-dug4 ki-sikil dNisaba eš-bar-ra ba-an-gub ‘The leader of heaven and earth, the lord Nunamnir, names the worthy and important people. He recruits these people, whom he strings together on a single thread, to provide for the gods. Enki then praised (Enlil’s) hoe, and the maiden Nisaba was made responsible for (keeping records of) the decisions’ (Song of the Hoe 28–33). (8) gi-dub-ba guškin šu ba-ši-in-ti dNanše-er sag-e gu-dili-a si mu-na-absa2-e ‘(Nisaba) takes the golden stylus in hand and arranges for Nanshe the people on a single thread’ (Nanše Hymn 100–101). (9) nita munus-bi gu-dili-˹a˺ mi-ri-e3-de3 nig2-u4-da-ka-ne-ne-a ˹igi˺-zuše3 bi2-ib2-ta-la2 [u]n šar2-ra-zu kur7-ak-da-bi dUtu-gin7 igi-zu-še3 i3dib-be2-ne ‘You string together men and women on a single thread – their daily status hangs down before your eyes. Your numerous people parade before your eyes, as (before) the Sun, for their inspection’ (In-nin-ša3-gur4-ra 211– 212).

4 The Sign and Office of the Sanga It may very well prove to be the case that the closest we may come to a physical portrayal of a Mesopotamian abacus will be furnished not by a textual description or an excavated artifact, but by paleography. The proto-cuneiform SANGA sign (Fig. 27), which represents, notably, the verb šid ‘to count’ in later texts, in addition to the professions sanga ‘accountant’ and umbisang ‘expert, scribe’, is in all likelihood a pictograph of an object that symbolizes the basis of this semantic field, namely, the abacus.38 These are notions that are closely intertwined in early Mesopotamia. It is probable that the office of the sanga, the abacus, and the related numerical system all pre-dated writing. And it is plausible that writing sprung from the office of the sanga, possibly at Uruk,

38 See already Englund 1998: 112 n. 245; Nissen, Damerow, and Englund 1993: 134. Compare the Chinese graph 算 for suan ‘to calculate’, which is at the root of suanpan ‘counting board’. As discussed above (§ 1) the central portion of this graph may depict a counting board.

The Abacus in Mesopotamia: Considerations from a Comparative Perspective

437

whose duties, as the chief temple administrator and accountant, were bound up with counting, and whose primary tool of the trade, the abacus, came to represent, pictographically, the office itself. Indeed, there is some suggestion that scribal activity in the archaic period was originally subsumed by the sanga, or there was at least an overlap in function (e.g., Nissen, Damerow, and Englund 1993: 105; Visicato 2000: 3–4). Lexical and administrative texts from Uruk attest, not infrequently, a title sanga-dub, while evidence from the third millennium convincingly connects the sanga with scribal activity.39 At Fara, and more rarely at Abū Ṣalābīkh, holders of the title sanga appear in colophons; in fact, several of the personal names paired with this designation in Fara school texts appear in administrative texts with the title dub-sar, a title that makes its first appearance in the first half of the Early Dynastic period (Steinkeller 2013: 141). Caution, of course, must be exercised when basing the physical description of objects on their pictographic representations in the writing system, which even in the earliest epigraphic evidence are highly conventionalized. Yet the Gestalt described by the sum of Uruk III–IV attestations of the graph suggests a vertically oriented abacus like its later, Greco-Roman and Asian counterparts, and in contrast to the horizontal devices of the Middle Ages. A counting board would appear to be the most likely candidate, although a fixed-bead abacus cannot be ruled out. Indeed, the vertical orientation recalls the much later passage from In-nin-ša3-gur4-ra (ex. 9), which describes strings with beads as hanging down, and the weaving and loom imagery discussed in connection with māḫiṣāṭu (§ 3). The grid within the frame would represent positions – either strings or spaces – that correspond to the alternating 10 and 6 replacement values (e.g., 1s, 10s, 60s, 600s, 3600s, etc.), which describe the basic sexagesimal numerical system (Figs. 28–30; discussed further in § 5). In other words, just as there is a complete isomorphism between the abaci of the Europe and the Greek acrophonic and Roman numerals (base-10 with a sub-base of 5), we expect the Mesopotamian abacus to reflect the sexagesimal numerical system (base-60 with a sub-base of 10). Since positions are relative, different assignments could accommodate other counting sub-systems and metrologies, such as those attested for proto-cuneiform. The distinctive horizontal line, or lines, of the graph suggest that the device consisted of upper and lower registers, while the protruding box to the lower left conceivably could represent a

39 Biggs 1967: 59; Biggs 1974: 33 n. 29. Visicato (2000: 4) points out the parallels between the specialized types of sanga attested in Uruk III texts with the subtypes of dub-sar attested between the Fara and Sargonic periods.

438

Christopher Woods

space for recording partial sums, other intermediate calculations, or fractional units.40

5 Calculation Errors The direct evidence of our sources at present can take us no farther in our attempt to describe the Mesopotamian abacus. Details must now be sought in the indirect evidence provided by the computations that were effected on these devices. For these purposes, erroneous calculations are typically more revealing of the processes by which they were arrived at than correct ones. Given the data for a correct calculation, errors allow for the identification of the step in the process where the error was made, which is otherwise impossible with correct results. The method is exemplified by Lang (1957; see also 1964, 1965, 1968), in her seminal study of the abacus in the Greco-Roman world. Lang argued that the calculations attested in Classical texts, Herodotus among others, could be explained by assuming that they were performed on an abacus. Moreover, errors in these calculations coupled with extant examples of these devices and their depictions in art allowed her to reconstruct certain computational procedures and to propose aspects of the device’s design. This method has been taken up recently by Høyrup (2002a) and Proust (2000), who have considered what errors in mathematical texts reveal about the devices on which they were calculated. As will be clear from the foregoing discussion (§ 1), the conclusions that can be drawn from these errors are entirely consistent with abacus use. Specifically, Høyrup shows that the erroneous repetition of certain entries in additive Old Babylonian calculations indicates that the device represented a transitory medium in which a number, once entered in the computation, was no longer recognizable, in contrast with more permanent calculation means that may have maintained individual entries analogous to pen and paper reckoning.41 Furthermore, the errors exhibited in 40 Englund (1998: 112 n. 245) proposes that this box may represent a storage space for counters, understanding the graph to represent a tallying board with three compartments in the lower and upper registers respectively. However, as counters can be stored by any number of means (for instance, in pouches) and the prominence of this element in the shape of the graph suggests that it is a salient aspect of the design, the box more likely represents a functional feature of the instrument. 41 Høyrup 2002a: 195, on the basis of BM 13901 problem no. 12. Ouyang and Proust (forthcoming) discuss “scratch pad” notations in Ur III administrative texts, which appear to reflect the interface and transfer between abacus and written numerical notations. For example, blank spaces in the texts appear to reflect the state of a calculation device in which a counter is absent in a given position.

The Abacus in Mesopotamia: Considerations from a Comparative Perspective

439

MDP 34: 19 rev. 1–12 ([Problème D]; see p. 103 and pl. 29), an Old Babylonian mathematical text that includes several exercises, shed some light on how numbers were represented in the instrument (Høyrup 2002a: 196). In one problem, the square of 14.48.53.2042 is given as 3.39.[28].44.26.40 instead of the correct 3.39.28.43.27.24.26.40,43 a result that represents the combination of two sequential errors: a unit that belongs to the 603 position ended up in the 604 position, thus rendering 43.27 as 44.26; and a second haplographic error in which 44.26.24.26.40 was truncated to 44.26.40, owing to the repetition of 26, which in both instances is preceded by 4.44 Importantly, the fact that this erroneous result is propagated further in the calculation shows that the first error was made not by the scribe transcribing the results of the device but by the original calculator;45 the subsequent error is due to a mistaken transcription – the results of the device were miscopied before being transferred to a second device or stage of the procedure (Høyrup 2002a: 196; see § 6). Any conclusive statement on what mathematical errors say about the devices on which they were computed will have to await a systematic study, which may point to multiple devices in use in different periods and based on different organizational principles. However, the preliminary evidence of the first error in MDP 34: 19 rev. 1–12, which is supported by further evidence presented below, suggests, as Høyrup argues, a device that represented numbers as collections of identical counters, rather than as written distinct graphs, and one that distinguished 10s from 1s – in other words, an abacus. Studies of abacus users have shown that certain frequent errors are unique to the abacus and can be directly attributed to the particular characteristics of the device.46 Soroban users, for instance, are much more likely to make errors of ± 5, owing to the presence of the “heavenly” 5 bead on that device (Stigler et al. 1986: 459). Likewise, abacus users are prone to making mistakes in inter- and intra-column carry-overs (see § 1.2), and to confusing numbers with similar bead config-

42 Following standard convention, periods are used to distinguish powers of 60 in a sexagesimal place-value notation. For instance 14.48.53.20 represents: 14 × 603 + 48 × 602 + 53 × 601 + 20 × 600. 43 See Høyrup 2002a: 196 n. 7 for corrections to copy and transliteration in MDP 34. 44 Høyrup 2002a: 196. A second error in this same problem (ll. 6–7) is too complex to disentangle with certainty, but may likewise involve the misplacement of both unit and tens counters in adjacent positions representing neighboring orders of magnitude. 45 See also Proust’s (2012: 16) discussion of the distinction between calculation and copying errors. 46 See Stigler 1984: 169–70, 173. While these particular errors were detected in the course of a study of the mental abacus (see § 6), they may equally occur with the physical device, being characteristic of abacus calculations generally.

440

Christopher Woods

urations (e.g., 2 and 7 which differ only with regard to the engagement of the “heavenly” 5 bead). Cross-cultural parallels, from the counting boards of the Classical world to abaci of Russia, China, and Japan, suggest that the observed distinction between 10s and 1s in MDP 34: 19 rev. 1–12 was not made with the counters themselves, but spatially, in terms of assigning values either to the lines or spaces of a counting board, or to the columns of a fixed-bead abacus.47 Naturally, a device that has a superordinate organization based on multiples of 60 with a subordinate distinction between 10s and 1s would correspond to the sexagesimal number system, which attests a staggered sequence of replacements of ten 1s and six 10s, resulting in a base of 60 with a sub-base of 10 (Figs. 28–30). This is precisely what we would expect, namely, for the base of the abacus to mirror the replacement values of the numerical system (see § 1). Furthermore, that the error of MDP 34: 19 rev. 1–12 involves neighboring positions (603 vs. 604) may be telling of the physical design. It suggests that positions of 1s, and by extension of 10s, were represented adjacent to one another for respective orders of magnitude, as this type of error is best explained in terms of misplacing a counter in a neighboring column (Høyrup 2002a: 196). Although this need not necessarily be the case, an arrangement in which 10s columns were above 1s columns, analogous to the Roman pocket abacus, the suanpan, and soroban, would be iconic with the writing of numerals in text, where larger units are written first, either appearing above or preceding lower units. Further insights can be gleaned from the errors considered by Proust (2000). These errors occur in Old Babylonian and Seleucid mathematical texts and are characteristic of large – literally, wide – numbers, which exceed five sexagesimal positions. A convergence of evidence suggests that multiplication involving these large numbers was accomplished by splitting the numbers into two pieces, multiplying the two halves separately, and then arriving at a correct total by recombining the halves by addition. Proust hypothesizes that the limit of five positions reflects the design of the calculation device used, which, in turn, may ultimately have been based on finger counting.48 Although Proust 47 Høyrup (2002a: 196), however, assumes that counters for units and tens were different. 48 An abacus covering five sexagesimal positions, with columns for 600–604, would be capable of expressing numbers from 1 to 777,599,599 (i.e., 59 × 604 + 59 × 603 + 59 × 602 + 59 × 601 + 59 × 600, not accounting for columns potentially devoted to fractions), a range wide enough to cover most practical applications. Interestingly, certain early traditions suggest that 604 was considered a limit of natural counting, a type of infinity horizon. Ebla text MEE 3, 73 labels 604 (6 × 10 × 603) as nu-da-šid ‘that which cannot be counted’, while the Old Babylonian metrological text BE 20/1, 29, gives 604 as šar2-gal šu nu-tag ‘big everything that the hand cannot touch’. Similarly, parallel texts TSŠ 190 (pl. 74) (Šuruppak) and MEE 3, 73 (Ebla) respectively conclude with šar₂ an [ki-bi-da] and šu nu-gi₄, designations that may refer to cosmic or

The Abacus in Mesopotamia: Considerations from a Comparative Perspective

441

did not speculate on how such a device functioned or was designed, she astutely observed the paradox that this conclusion posed: the evidence suggests an instrument of limited capacity; however, any such device that is likely to have been used could, despite its presumed origins, be extended indefinitely by adding more rods, columns, etc., to accommodate larger numbers (Proust 2000: 302). Critical to the argument that the multiplication of large numbers was handled through splitting is Old Babylonian Nippur text N 3958 (ex. 10), a mathematical text for which 36 of the 40 lines of numerical entries can be restored (Proust 2000: 299–301; Sachs 1947: 228–29). Utilized as part of a favorite reciprocal algorithm,49 the numbers represent a geometric progression with a common ratio of two, in which each successive entry doubles the preceding one. Although the beginning of the text is broken, the extant portions demonstrate that the series, as is common in reciprocal tables, begins with the sexagesimal number 2.5. The remarkable feature of this sequence is the use of a dividing sign consisting of a vertical wedge with an appended diagonal, , a graph that Friberg (1990: 536) suggested may be a variant of kud = parāsu ‘to cut’. This dividing sign is represented in the transcription below by ‘+’, following the convention of Proust (2000: 300). The sign first appears in rev. 1, separating the fifth sexagesimal position from the first four, so that the number 10.6.48.53.20 is written in two parts as 10+6.48.53.20. The two halves are then doubled as if they were two distinct numbers; the dividing sign itself appears only in the first ten entries of the reverse, but is clearly implied thereafter given that the two halves of the number continue to be doubled separately. Again, following the conventions of Proust, an implied dividing sign is indicated by ‘(+)’, whereas the numbers in bold and underscored mark the position where the two halves must be recombined.

unreachable quantities as noted by Friberg (1990: 538); for these notations in general, see also Friberg 1986: 8–15; Powell 1971: 73, 76–78. These attestations are consistent with other evidence, which suggests that the Mesopotamian abacus may have conventionally or traditionally consisted of five columns, a design that may ultimately have its roots in finger counting (see § 6 and n. 56). However, the linguistic evidence for the conception of large numbers has little bearing on the theoretical limits of the device or the shape it may have taken in various contexts – the abacus represents number by non-linguistic means and is not constrained by the language of number (compare, for instance, the Japanese soroban which often includes up to 23 columns, i.e., 1023). 49 See Sachs 1947. The algorithm began with the relation 2.5−1 = 28.48 and generated new pairs on the basis of the continuous doubling of 2.5 and halving of 28.48.

442

Christopher Woods

(10) N 3958 Line

Obverse

Reverse

Calculation

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20

2.5 4.10 8.20 16.40 33.20 1.6.40 2.13.20 4.26.40 8.53.20 17.46.40 35.33.20 1.11.6.40 2.22.13.20 4.44.26.40 9.28.53.20 18.57.46.40 37.55.33.20 1.15.51.6.40 2.31.42.13.20 5.3.24.26.40

10 + 6.48.53.20 20 + 13.37.46.40 40 + 27.15.33.20 1.20 + 54.31.6.40 2.40 + 1.49.2.13.20 5.20 + 3.38.4.26.40 10.40 + 7.16.8.53.20 21.20 + 14 .32.17.46.40 42.40 + 29.4.35.33.20 1.25.20 + 58 .9.11.6.40 2.50.40 (+) 1.56.18.22.13.20 5.41.20 (+) 3.52.36.44.26.40 11.22.40 (+) 7.45.13.28.53.20 22.45.20 (+) 15.30.26.57.46.40 45.30.40 (+) 31.0.53.55.33.20 1.31.1.20 (+) 1.2.1.47.51.6.40 … … … …

10.6.48.53.20 20.13.37.46.40 40.27.15.33.20 1.20.54.31.6.40 2.41.49.2.13.20 5.23.38.4.26.40 10.47.16.8.53.20 21.34.32.17.46.40 43.9.4.35.33.20 1.26.18.9.11.6.40 2.52.36.18.22.13.20 5.45.12.36.44.26.40 11.30.25.13.28.53.20 23.0.50.26.57.46.40 46.1.40.53.55.33.20 1.32.3.21.47.51.6.40

The sequential placement of the dividing sign is revealing of the computational device and procedures used. Consistent with the conclusions drawn from the errors in MDP 34: 19 rev. 1–12 discussed above, the position of the divider shows that the device was positional and distinguished 10s and 1s, which were represented by collections of counters. The divider first appears when the fifth sexagesimal position reaches ten; for numbers less than ten in this position the graph is not used (obv. 18–20).50 At this point, calculations are carried out separately, ostensibly on two distinct devices, which for the present may be envisioned as set next to one another representing the left and right halves of each number (Fig. 31). In rev. 8–10, the dividing sign does not fall between 10s and 1s in the fifth position, but rather the ‘right’ and ‘left’ numbers include both 10s and 1s in this position, which must be combined. In other words, the right half number extends beyond the 1s and into the 10s for the fifth sexagesi-

50 Computing 10s separately from 1s, which once again suggests that they were handled in different parts of the device, is explainable in abacus terms, as it simplifies the multiplication procedures and the manipulation of counters (I thank Pema Kongpo [personal communication] for this observation).

The Abacus in Mesopotamia: Considerations from a Comparative Perspective

443

mal position. In rev. 11–13, the implied divider shows that the right number reaches the 1s of the sixth sexagesimal position; and in rev. 14–15 it occupies the tens position of the sixth position. In the final number that can be reconstructed with confidence, rev. 16, the right number extends into the 1s of the seventh sexagesimal position. Assuming the dividing sign represents the distinction between two calculating instruments, then, as the right numbers show, this instrument is not limited to five positions, but is capable of accommodating sexagesimal numbers as wide as at least seven positions. The initial recombination point remains the fifth position, but broadens to encompass the sixth and seventh positions as the numbers grow. The leftward drift of the dividing sign relative to the numbers calculated demonstrates that calculations proceeded from right to left, from lower orders of magnitude to higher, as on a Japanese soroban.51 And, as with the soroban, the expansion of both the left and right halves of the number to increasingly higher powers of 60 – which propels the dividing sign leftward – is explainable in abacus terms as the carrying-over (§ 1.2) of counters to the next power once the replacement value for a given column has been reached. Divisions of this kind involving big numbers are not limited to this text, but occur elsewhere with both explicit and implicit indication of the point of separation. The related Old Babylonian reciprocal text CBS 29.13.2152 writes 1.26.18.9.11.6.40 as 1.25.20.58.9.11.6.40, which indicates an underlying 1.25.20(+)58.9.11.6.40. In lines 8 and 9 of the same column there is 18.32.21.56.30/6.5.[37].30, which is to be interpreted as 30+6, where the separation is signaled by the division of the number across two lines. Other examples exhibiting this same type of split include the following, all encountered in Old Babylonian multiplication and reciprocal texts (see MCT, p. 13 n. 69): 5.55.30/ 3.20 and 37.2.10/3.20 (YBC 4716); 32.55.18.30/1.6.40 (CBS 8309); 32.5[5.18.30]/ 1.[6.40] (CBS 29.15.497); 4.25.23.10.50/6.47.24.26.40 (YBC 4704). Even in instances where there is no implicit or explicit indication of a split, several types of errors, characteristic of big numbers and appearing in medial position, suggest nonetheless the use of two calculation devices. On the basis primarily of the Seleucid reciprocal text AO 6456, Proust (2000: 296–299) shows that these are gluing or joining errors which were likely generated in the course of attempting to recombine the results from multiple instruments. These errors include: the erroneous addition of numbers that should be written consecutively, e.g., 46.49.19.54.58.53.20 for correct 46.49.19.40.14.48.53.20; the

51 Compare Chinese abacus techniques, in which computations are worked from left to right. 52 MCT, p. 13 n. 68; discussed by Proust (2000: 301) and Sachs (1947: 227–29).

444

Christopher Woods

faulty insertion of a zero notation, e.g., 2.16.41.0.15 for correct 2.16.41.15; the omission of medial digits in transcription (including zero), e.g., 1.58.36.15 for 1.58.39.8.26.15 (cf. the second, haplographic error in MDP 34: 19 rev. 1–12 discussed above). In one instance, the presence of two errors of the last type indicates that a particularly large number represented the joining of three pieces (2.0.25.38.4.52.25.29.46.*.29.37.46.40 for 2.0.25.38.14.52.25.29.46.0.29.37.46.40). In each case, the error can be attributed to a misreading or misidentification of the medial numerical positions of a counting device where the pieces of a big – that is, wide – number were to be joined. The evidence, taken as a whole, suggests a common procedure – in practice from at least Old Babylonian times through the end of the cuneiform record – by which numbers that span multiple sexagesimal positions were computed piecemeal. And it is safe to assume that the method was more widespread than these examples suggest, for it is only discernible when a revealing error is made or the split is graphically indicated in the transcription. It is also apparent that, while the fifth sexagesimal position was a common separation point, numbers may be split in the fourth, third, or even second, sexagesimal positions. And, as discussed above in connection with N 3958 (ex. 10), while the initial split was made in the fifth position, for the widest numbers in the progression the dividing sign “carries over” to the sixth and seventh positions. Assuming all of these attestations share a common explanation, and that the numerical split marks the use of different instruments, then this device was not necessarily fixed to the representation of five sexagesimal positions, but varied in its capacity. This brings us back to the paradox observed above that begs an explanation for why calculators resorted to the error-prone process of splitting numbers into parts to be computed on different devices, when the use of a single larger instrument would obviate these difficulties.

6 In the Mind’s Eye: Mental Abacus Calculations A solution to this enigma presents itself if we consider that the splitting of wide numbers into parts is not necessarily indicative of the use of multiple physical devices, but rather speaks to mental calculations and techniques that facilitate the cognitive computation of large numbers. These mental calculations would be carried out on what is known as a “mental abacus” – an abacus construed in the mind’s eye. As Netz points out, whether the abacus consists of an actual physical device or an imagined facsimile may itself be a distinction with little merit given its relatively simple design – “Ultimately, indeed, the very notion of the abacus as a clearly defined artifact is misleading. While scratches are

The Abacus in Mesopotamia: Considerations from a Comparative Perspective

445

useful, the lines can very well be imagined, perhaps referring to whatever irregularity the surface at hand may have. Thus any surface will do. The abacus is not an artifact: it is a state of mind” (2002: 327). While Netz’s comments refer specifically to the counting boards of the Classical world, they apply equally to the fixed-bead devices in contemporary use. Abacus use, with extensive training is a typical example of what is referred to in the psychological literature as an overlearned cognitive skill – one that has been learned beyond proficiency, so that the operation of the device becomes automatic, interiorized, and subconscious. In Japan, for instance, expertise of this kind is achieved by children who have typically practiced at least a few hours daily for several years (Hatano et al. 1977: 48). In fact, abacus experts typically opt to abandon the physical device altogether, and are able to construct a mental image of an abacus and perform calculations by manipulating the visualized beads in precisely the same manner, following the same procedures and intermediate steps as if they were using an actual, physical device. Many practitioners, with the exception of grand masters at the highest levels of expertise, even imitate the same finger movements used with the actual device to aid in moving the imagined beads on the mental abacus, a process that has been likened to the mouthing of words while reading.53 The process becomes so internalized and habitual with such users that they are even able to carry on conversations while computing complex calculations (Hatano et al. 1977). Like its physical counterpart, the mental abacus provides a means of visually representing number that is tailored to visuo-cognitive capabilities. The power and appeal of this method, as Frank and Barner (2011: 13) explain, lie in allowing users to make extremely fast and accurate calculations that remain beyond the reach of other mental arithmetic techniques. Experts of this method are capable of bewildering feats of mental calculation, such as finding the product of two 5-digit numbers in under 10 seconds, and adding five 3-digit numbers in under three seconds (Hatano et al. 1977; Stigler 1984). Typically, these mental abacus practitioners can perform calculations faster than their counterparts using physical devices because they are not slowed by the physical limitations of manipulating beads. Indeed, experts can often calculate more quickly even than users of electronic calculators, who are limited by the physical operation of the instrument. The mental abacus method is practiced throughout Asia, particularly in China, Japan, India, and South Korea, where it is frequently employed in abacus competitions of speed and accuracy.54 53 Stigler 1984, Hatano et al. 1977: 48; Frank and Barner 2011: 4. 54 For an overview of previous literature concerning the mental abacus, see Frank and Barner 2011.

446

Christopher Woods

Problems that involve particularly big, or wide, numbers that occupy multiple columns – pose a particular difficulty for mental abacus calculations. Since numbers are rendered as abacus images, users need to commit to visual memory the bead arrays for each column, tracking the precise position of each bead of the mental abacus. Naturally, this strains normal capacities for storing and refreshing detailed images (Stigler 1984: 154). A common method to deal with the memorization burdens imposed by big numbers is to split the problem in half, solve for each half separately, and then combine these halves for a final result. As explained by Stigler (1984: 174): “mental abacus experts, when faced with a problem which exceeds the width of two or three columns, may tend to break the problem in half vertically and solve the two parts separately. Presenting the problems visually made it possible for subjects to use such a method. The time required to remember the first half result while calculating the second, and then combining the two halves could account for the extra time required for wider problems.” Lee Jeohghee, one of the foremost abacus masters in the world, confirms this technique, reporting that when adding, for example, ten 10-digit numbers, she will break each number into two columns of 4- and 6-digits each, and add them separately. The process requires remembering the 7-digit answer to the 6-digit column (the discrepancy is due to carrying-over counters to the next column; cf. the leftward drift of the dividing sign in ex. 10) while she adds the 4-digit numbers, then combining the two sums. According to Master Lee, all abacus masters employ this strategy and she teaches it to her students as well.55 Of course, this mental abacus technique corresponds closely to the evidence of split calculations attested in our mathematical texts (§ 5). Not surprisingly, the width of the splits, in terms of the number of positions each part occupies, is governed by the subitization and chunking limits imposed by human cognitive capacities (§ 1.3). The splits exhibited in mathematical texts range from second to seventh sexagesimal positions, but most commonly occur in the fifth position, which may reflect the traditional organization of the physical abacus as a device based on five fingers.56 The splits reported by Stigler

55 Personal communication. With regard to numerical recall, Master Lee recounted that she was asked on a Korean TV show to memorize a 100-digit number. She did this by breaking the number into a sequence of twenty 5-digit numbers and proceeded to add them mentally, visualizing each of the 20 steps on her mental abacus. 56 See Proust 2000: 302; Høyrup (2000b and personal communication) suggests that šu-nigin ‘total’ and šu-tag₄ ‘remainder’ (TSŠ 50 ii 2), among other OB mathematical expressions involving the hand, may indicate that the Mesopotamian abacus was restricted to five columns (cf. n. 48 above, where the significance of five sexagesimal positions in Mesopotamian numeracy is discussed).

The Abacus in Mesopotamia: Considerations from a Comparative Perspective

447

and Master Lee likewise fall within this range. And this evidence is in agreement with recent experimental data as well. An experiment with children expert in the mental abacus, for instance, found that the crucial factor limiting mental abacus additions was not so much the number of addends or the number of beads in play, but rather their width in terms of the number of columns the individual addends occupied on the abacus.57 Whereas most children were able to consistently add 3-digit addends, there was a sharp drop off thereafter, with only the most highly skilled children adding 4-digit addends at an accuracy rate of roughly 50 %.58 In other words, the most salient limit to mental abacus calculations is the width of the addends, which translates to an image of individual abacus columns that must be committed to memory (Frank and Barner 2011: 7). The types of errors exhibited in the mathematical texts considered in § 5 are also consistent with mental abacus calculations. Errors such as the omission of columns or counters, or the confusion of neighboring columns, are particularly common in mental abacus calculations. For example, experiments comparing the types of errors Chinese mental abacus users make, in contrast to their American peers using familiar mental calculation techniques, show that the former are three times more likely to omit digits, particularly when they occur in medial position, which are envisioned in the mind’s eye as abacus columns (Stigler 1984: 154–55, 166–73). Other studies of mental abacus practice demonstrate that errors involving the confusion of columns occur most frequently when medial beads in adjacent columns exhibit similar bead configurations (Frank and Barner 2011: 12). More generally, as previously observed, errors involving the faulty insertion or omission of zero in transcription are consistent with abacus use. Since zero is not directly represented on the abacus, but rather is indicated by the convention of the absence of a counter in a given space, users are burdened with accurately recalling the number and position of the columns not engaged. As discussed above, these are precisely the types of errors that are encountered in the mathematical texts studied by Høyrup and Proust (see the discussion of MDP 34: 19 rev. 1–12 and AO 6456 in § 5). Making their appearance particularly at the medial join positions where the partial sums are to be recombined, they may represent not visual

57 Frank and Barner 2011: 4–7. As the researchers point out, this range is consistent with recent experimental studies (e.g., Feigenson 2008; Halberda, et al. 2006), which suggests that subjects can accurately count and track up to three or four sets of parallel, multiple objects in visual working memory. 58 Master Lee reports being able to multiple 8-digit by 7-digit numbers before having to resort to the splitting procedure (personal communication).

448

Christopher Woods

errors in reading a physical device, but memorization errors where the calculator had omitted or lost track of the relative positions of the columns or confused the similar configurations of adjacent columns.

7 Towards a Description of the Mesopotamian Abacus Although no one piece of the evidence presented here can be regarded as definitive, taken collectively, a compelling case can be made for the use of the abacus in ancient Mesopotamia, a conclusion that is entirely in line with the expectations of cross-cultural parallels. And, as attested elsewhere, we may expect a variety of computational devices, which took their place alongside other administrative counting instruments such as tallies and tokens. Certainly, the lexical attestations point in this direction. Although definitive identification will in all probability remain elusive, the writings of the lexemes giš NIG2.ŠID and tug2NIG2.ŠID show that wood and textiles were salient materials in the construction of such counting aids. The Akkadian equivalents māḫiṣātu and uttuku (and Sumerian uttuku), both of which have connotations of weaving, as well as the one attestation of these devices appearing in connected text (ex. 6), suggest the use of textiles or perhaps even a device with physical characteristics resembling a loom (cf. the SANGA sign, Fig. 27). These lexemes may refer to several related instruments (cf., for instance, the use of the term abacus, which refers to both counting boards and fixed-bead devices, § 1), but among these calculating tools the counting board and the reckoning cloth, which is merely a more portable version of the former, must be considered as particularly likely candidates. A variety of artifacts discovered in the course of excavations, such as rounded potsherds, may have served as counters on such devices, though the humble pebble could serve this purpose as well. Such unassuming objects, and not the geometric tokens or cuneiform numerical graphs of writing – both of which are better suited to recording and archiving the results of calculations – were likely the basic instruments of computation in Mesopotamia.59 The counting board is the abacus par excellence, the original calculator of apparent universal attestation. In Mesopotamia, the counting board almost

59 Compare Roman Numerals, which, in the view of most authorities, were employed throughout their two-thousand year history not for calculating, but for numerical recording, purposes (see Chrisomalis 2010: 115).

The Abacus in Mesopotamia: Considerations from a Comparative Perspective

449

certainly predates writing. The invention of writing brings to light for us the mathematical heights that had been achieved by the close of the pre-literate era, preserving computations of remarkable complexity that must have been carried out with the aid of a calculation device. Indeed, the counting board may have influenced the development of writing. In particular, the incipient place value system of proto-cuneiform numerical notation by which units are hierarchically ordered, the development of replacement values, and the oneto-one correspondence between the enumerated set and the numerical graphs of writing, may all be related to the use of counters on a counting board.60 For comparison, we may point to Roman numerals, whose structure and even eventual fate in medieval Europe were bound up with the counting board. Abacus notation may have also shaped the representation of numerals as subitized chunks, the omission of zero paleographically, and, perhaps, even tablet format. The division of text into cases, attested from the first writing, recalls the spatial representation – indeed the spatial syntax – of the counting board. The counting board thus may have provided a physical template for the organization of text into cells or cases, the relationships between which are expressed not linguistically but as a function of relative spatial position. The SANGA sign likely depicts an abacus – whether it is a primary counting board, or, at even this early date, a fixed-counter device, cannot be determined. In either case, the sign points to a device of a specific design and not to ad hoc rulings that might constitute a counting board in its most primitive form. Naturally, the attestation of gišNIG2.ŠID in an Early Dynastic mathematical text (§ 3) confirms the supposition that calculation devices were in use at an early date at least in specialized, scholarly contexts. The idiom gu-dili-a(–e3) ‘to string together on a single thread’ implies a counting device that makes use of threaded beads, which are oriented vertically. The literary attestations alone are not sufficient to establish beyond doubt the existence of a fixed-bead abacus analogous to the Chinese suanpan, Japanese soroban, or Russian schety. However, the types of calculations discussed in § 5, which extend into the decimal equivalent of hundreds of trillions, would be remarkably cumbersome on a counting board to the point of

60 Interestingly, the use of the abacus as a spatial template for other domains is suggested by the possible abacus origins of certain board games, which require the calculation of positions and the movement of counters across a numerically defined board (Netz 2002: 326, 342; Pullman 1969: 8, 20–21, 90; see also Tun and Montaluisa Chasiquiza [forthcoming] for the association of Andean taptana [§1] with indigenous board games). Jens Høyrup (personal communication) points out that the Medieval board game rithmomachia, used to drill Boethian ratios, utilized the board of the “Gerbert abacus” (see n. 18).

450

Christopher Woods

being unfeasible, all the more so if we assume that Mesopotamian devices operated on the basis of alternating replacements of 10 and 6 in accord with the structure of the numerical system. For instance, the number in N 3958 rev. 14 – 22.45.20 (+)15.30.26.57.46.40 – would require an imposing 62 counters on a counting board of this design, and 124 counters for the doubling procedure. And still the split of the number into two halves would be unaccounted for – indeed, the split increases the number of counters that need to be manipulated by 50 %, and requires the additional, evidently troublesome, step of recombination. Calculations involving big numbers of these kinds are much better explained as being effected on a fixed-counter device, which allows for much more rapid and efficient computations. The counting board, particularly those with alternating replacement values (Fig. 28a), is essentially a tallying device, most suitable for basic addition and subtraction (Lang 1957: 278–82). The fixedcounter abacus, on the other hand, for which replacement values occupy different parts of the device as on the suanpan and soroban, is a much more powerful instrument – it is a true calculating machine. Just as the system of Mesopotamian numeration grew out of the very act of counting itself (§ 1.3), the evidence provided by errors demonstrates that calculations were ultimately rooted in the manipulation of physical counters. The error in MDP 34: 19 rev. 1–12, for instance, in which a counter was misplaced by an order of magnitude indicates that numbers were represented in computations as collections of discrete objects, and suggests the use of an abacus in which unit positions were adjacent to one another; tens positions would presumably be arranged similarly, perhaps placed above corresponding units. Likewise, the progressive drift of the numerical divider in N 3958 (ex. 10) demonstrates that calculations proceeded from right to left, with counters carryingover to neighboring positions as they do on an abacus (§ 1.2). Errors relating to the placement of zero, and more generally the omission of a paleographic notation for zero until Seleucid times,61 are also, as discussed above, consistent with the concrete notation of the abacus.62 Also consistent with the use of the abacus is the plethora of multiplication tables but the absence of addition tables from the textual record. Multiplication on the abacus, which is effected

61 The exception being the use of a separation graph functioning as zero in two OB mathematical texts from Susa (MDP 34: 12 and 24; see Friberg 1990: 536 and Høyrup 2002c: 15–16 n. 19). 62 See Nissen, Damerow, and Englund 1993: 147–49 for early errors involving zero and the possibility that these were calculated on a counting board or fixed-bead device. Note also that the errors involving zeroes in medial position in Seleucid reciprocal text AO 6456, as discussed by Proust (2000), are likewise consistent with abacus use.

The Abacus in Mesopotamia: Considerations from a Comparative Perspective

451

by the addition of partial sums, relies, as does pen-reckoning, on the knowledge of the multiplication table.63 This necessity is all the more pressing in the case of a sexagesimal system for which the multiplication table is vastly more extensive than its decimal counterpart.64 On the other hand, addition on the abacus, which is entirely mechanical, completely eliminates the need for addition tables. As a working hypothesis, we might propose that the Mesopotamian sexagesimal abacus, in at least one form, represented a minimal 5 : 9 design (§ 1.2) with an upper register limited to five counters, each with a weight of 10 ×, and a lower register limited to nine counters, each with a weight of 1 ×, for each column assigned to a power of 60 (Fig. 28b, Fig. 30). Such a design would correspond to the structure of the Mesopotamian numerical system, which exhibits a base of 60 and a sub-base of 10, and finds support in the provisional data of numerical errors, which point to the use of counters in computations with values of 1 × and 10 × (§ 5). A systematic study of calculation errors could confirm or deny this proposal, and may very well suggest a device or devices based on different divisions of 60.65 These, of course, are not the only factors that must be taken into account when considering possible designs. There is, for instance, the issue of subitization. A sexagesimal counting board of 5 : 9 design would be particularly cumbersome since the presence of up to nine counters in the units columns exceeds reasonable subitizing limits and requires manual counting, which naturally impedes rapid and efficient calculations.66 Counters could theoretically be arranged in suitable subitized chunks, 63 See Proust (2012: 8–9) for multiplication algorithms in Mesopotamia involving the addition of partial sums, which is consonant with abacus practice. 64 A base-10 system has a multiplication table with 36 entries, while a base 60-table has 1711 entries. However, the existence of a sub-base reduces the size of the multiplication table; the sexagesimal system with a sub-base of 10 has a table of 91 entries (see Moon 1971: 44, 48). 65 Also worthy of further study, for the light it may possibly shed on abacus practice, are the spatial arrangements of numbers on certain mathematical texts, such as those considered by Proust (2012), in which text layout may reflect calculation practices. 66 While it is reasonable to expect the functional design of the abacus to be identical with the numerical systems it represents in terms of sharing common replacement values, this need not necessarily be the case (see Lang 1957: 282 on this point regarding the abacus in ancient Greece). Other abacus designs are possible, for instance, which would eliminate the problem of subitization. Rather than a two-tiered division of 60, conceivably there could be a device that relies on a further factorization of 60. One might envision, for instance, a non-minimal abacus that is four-tiered and represents replacements of 5:2:3:2, i.e., with columns for 1s (replacement = 5), 5s (= 2), 10s (= 3), 30s (= 2). More sophisticated devices that rely on 3 or 4 factors of a given base, or that employ negative counters (cf. the subtractive notation used in cuneiform as a type of administrative short-hand, e.g., 59 written as 60 la2 1), are often described in the literature as theoretical possibilities, but were unknown in the ancient world

452

Christopher Woods

as they are in writing, but this comes with its own hindrances as counters cannot be easily moved across columns. A fixed-counter 5 : 9 device (Fig. 30), however, could easily overcome this hurdle by marking the counters in such a way that breaks them into subitized groups, as is done, for example, with the Russian schety (§ 1.2).67 The cross-cultural evidence for the use of counting instruments (§ 3) suggests that the various devices hinted at in the lexical sources occupied distinct functional and social niches, being used for various purposes and by specialists of different kinds – tallies for simple counting purposes, tokens for dynamic accounts of specific commodities, and counting boards and reckoning cloths for the calculations of basic administration. More complex devices, such as perhaps the fixed-counter abacus, would be reserved for demanding administrative tasks and, particularly, for the production of mathematical and scientific texts, for which the basic counting board would be insufficient. And it is not unreasonable to assume that scholars engaged in the generation of these texts – such as our multiplication and reciprocal tables – would have been so intimately familiar with the instrument of their trade that they could conjure the device in the mind’s eye and perform the calculations mentally, manipulating imagined counters and using methods analogous to those employed by contemporary mental abacus users. Understanding the numerical splits and associated errors in terms of the natural limits of memory and visual-cognitive capacity provides a plausible explanation for these otherwise inexplicable phenomena. Memorization was a virtue with few rivals in the Mesopotamian scribal ethos, whether it involved literary, lexical, or mathematical texts. Within this context, internalizing computations may very well have been an esteemed skill among scribes of the mathematical milieu.

to my knowledge (see Moon 1971: 86). While such a device would typically minimize the number of counters employed, the operation is more complex and more prone to user error in terms of the placement of counters. Importantly, one would expect such a device not only to generate errors in 1s and 10s, but in the other factors as well, in this particular example in 5s and 30s. 67 As an aside, I would note my own experiences experimenting with these devices, which are necessarily impressionistic and likely say more about my own visuo-cognitive limitations than they do about the nature of the Mesopotamian abacus. Working through the reciprocal sequence of ex. 10 (N 3958) on two counting boards of the aforementioned 5 : 9 design (Fig. 28b) I found to be a time-consuming and error-prone process precisely because of the bewildering number of counters involved. On the other hand, performing these same computations on two fixed-bead abaci of the type illustrated in Figs. 30–31, was a rapid, efficient, and relatively error-free endeavor once the basic bead manipulations of the abacus had been mastered with practice (see § 1.2).

The Abacus in Mesopotamia: Considerations from a Comparative Perspective

453

Figures

Fig. 1a-b: Counting rods were in use in China from at least the Warring States period (475–221 BC), and were introduced to Japan, perhaps via Korea, by c. 600 AD. Commonly made of bamboo – but also wood, bone, ivory, iron, and jade – counting rods varied in length (c. 8–15 cm) and shape (attesting cylindrical, triangular, and rectangular crosssections) depending on the period and place of use (Martzloff 2006: 210; Smith and Mikami 1914: 20–29). Left: The number 38,057 expressed with Japanese rectangular counting rods. Right: A Japanese mathematician manipulating counting rods. From the Shojutsu sangaku zuye (1795) (cited from Martzloff 2006: 211, Fig 13.1 and Smith and Mikami 1914: 29).

Fig. 2a-b: Left: A wooden Inka yupana from the site of Chan Chan, Peru (16.5 × 13.5 cm) – a counting board used for calculations recorded by the khipu (reproduced after Radicati di Primeglio 1990: 222, Fig. 2). The rules governing the organization and use of the device remain debated, with individual artifacts exhibiting considerable geometric variation. Presumably, these are positional devices in which value is assigned to the cells, the organization of which may mimic the architecture of grain storage facilities. Pebbles, beans, seeds, or kernels of corn may have served as counters (Brokaw 2010: 18, 84). Right: An Ecuadorian taptana (reproduced after Quilter 2002: 218, Fig. 9.3).

454

Christopher Woods

Fig. 3: An Inka accountant known as a khipukamayuq (knot maker or keeper), holding a khipu. The grid depicted on the bottom left is probably a type of yupana (Browkaw 2010: 69) or, perhaps, a taptana (Quilter 2002: 215); reproduced after Quilter 2002: fig. 9.4 (Guaman Poma 1980: 332).

Fig. 4: The addition of 8,749 + 2,463 = 11,212 on a horizontal counting board typical of the 15th century. Although similar methods were likely used in Classical antiquity, the earliest detailed descriptions of abacus procedures date to this time (Schärlig 2001: 143–44). The counting board makes use of both lines for decimal values, and the spaces between them for quinary values. Addition is carried out by rendering the first addend as counters and placing them on the board to the left; the second addend is then placed to the right. A vertical line separates the two numbers, and a cross marks the thousands position. The counters are then pushed together and replaced according to the alternating 5 and 2 replacement values of the Roman numerical system. The sum of 11,212 is given on the far right.

The Abacus in Mesopotamia: Considerations from a Comparative Perspective

455

Fig. 5: The Salamis counting board (5th–4th century BC, National Museum of Epigraphy, Athens) – one of the earliest and best-preserved abaci from the ancient world. Discovered near Athens, the white marble slab measures 149 cm × 75 cm. Greek acrophonic numerals appear on three sides, representing monetary values from one talent (T) to 1/8 obol (X) (after Schärlig 2001: 66).

Fig. 6: The number 28,478 represented on the Salamis board.

456

Christopher Woods

Fig. 7: The addition of 8478 + 2287 = 10,765 on the Salamis board.

The Abacus in Mesopotamia: Considerations from a Comparative Perspective

457

Fig. 8: Detail from the Darius Vase (c. 300 BC, Museo Nazionale di Napoli), portrays Darius’s treasurer receiving tribute. He holds a writing tablet in his left hand and manipulates white pebbles with his right (Pullan 1969: 30).

458

Christopher Woods

Fig. 9: Roman bas-relief (1st century BC, Museo Capitolino, Rome), with detail, depicting a dying man dictating his will to his wife, as a calculator keeps track of the bequests with a counting board (Pullan 1969: 25).

The Abacus in Mesopotamia: Considerations from a Comparative Perspective

459

Fig. 10: The replacement values for two cumulative-additive systems: Roman numerals and the basic sexagesimal system (S) of the archaic texts (after Englund 1998: 118). The base-10 Roman system employs a sub-base of 5 with alternating replacements of 5 and 2. The base60 Mesopotamian system has a sub-base of 10 with alternating replacements of 10 and 6. Such numerical systems may be broadly described as “abacus notations,” as each graph represents the value of a counter on an abacus.

Fig. 11: German woodcut, c. 1500 (Barnard 1916, pl. 47b).

460

Christopher Woods

Fig. 12: Woodcut from Johann von Schwarzenberg’s Teutsch Cicero (1533) (Barnard 1916, pl. 59).

The Abacus in Mesopotamia: Considerations from a Comparative Perspective

461

Fig. 13: Woodcut from Cicero’s Officia (1531), presumably depicting the introduction of a new student (Barnard 1916, pl. 49).

462

Christopher Woods

Fig. 14: Woodcut from Cicero’s Officia (1531), portraying a merchant engaged in the keeping of accounts (Barnard 1916, pl. 50).

The Abacus in Mesopotamia: Considerations from a Comparative Perspective

463

Fig. 15: Woodcut (1543) depicting three methods of reckoning: casting counters on a counting board; pen and paper calculations using Hindu-Arabic numerals; and tallying with marks (Barnard 1916, pl. 55a).

464

Christopher Woods

Fig. 16: Woodcut from Gregor Reisch’s Margarita Philosophica (1503) (Barnard 1916, pl. 46), which speaks to the change of the numerical guard at the beginning of the 16th century. A trepidatious Pythagoras, who was credited with having invented the abacus, represents the old counting board method. He is pitted in a competition of mathematical wits against a confident Boethius, who was considered to have introduced Hindu-Arabic numerals to Europe. Dame Arithmetic looks with approval upon the new, ostensibly superior, method of pen and paper reckoning.

The Abacus in Mesopotamia: Considerations from a Comparative Perspective

Fig. 17: Strasbourg reckoning table (16th century, Musées de la Ville de Strasbourg) (Pullan 1969: 53).

Fig. 18: German brass jetons depicting counting boards, 16th century (Barnard 1916, pl. 42: 1–5).

465

466

Christopher Woods

Fig. 19: Chinese suanpan representing the number 3749.

Fig. 20: Japanese soroban representing the number 3749, with the decimal point arbitrarily assigned to the 9th column.

The Abacus in Mesopotamia: Considerations from a Comparative Perspective

467

Fig. 21: Roman pocket abacus (Bibliothéque National, Cabinet des medailles, Paris). The decimal organization is identical to that of the soroban. The rightmost slots represent fractional monetary values (unica). The number represented is 6,290,153, plus 5 unica, and fractions of 1 unica (after Schärlig 2001: Fig. 6 and 125).

468

Christopher Woods

Fig. 22: Russian schety representing the number 3975. The row with four beads serves as a decimal marker distinguishing rubles, above, and kopeks, below.

The Abacus in Mesopotamia: Considerations from a Comparative Perspective

469

Fig. 23: Rounded potsherds, ca. 3 cm in diameter (drawings scaled 1 : 2), from Abū Ṣalābīkh (Green 1993, figs. 5 : 1 and 5 : 2).

Fig. 24: Perforated stone-disks, ca. 4–5 cm in diameter, from Chogha Mish (Delougaz-Kantor 1996: 277; plate 251: F–I).

470

Christopher Woods

Fig. 25: Reckoning cloth (16th century, Bayerisches Nationalmuseum) from Munich, measuring 72 cm x 38 cm (Barnard 1916, pl. 40).

Fig. 26: The chimpu – a decimal abacus consisting of strung seeds, here representing the number 4456. The device is used by the indigenous peoples of Peru and Bolivia, and is likely a descendent of the khipu. Powers of ten are indicated by the number of strands of each string: 4 strands for thousands, 3 for hundreds, 2 for tens, 1 for units. Seeds serving as counters are then threaded on the appropriate strings; the recording of the number is completed when the four strings are tied together at the bottom (indicated by u-----u') (Menninger 1969, fig. 89).

The Abacus in Mesopotamia: Considerations from a Comparative Perspective

471

Fig. 27: Uruk III–IV attestations of the sign SANGA (after ZATU, no. 444 [Green and Nissen 1987).

Fig. 28a-b: Two possible organizations of a sexagesimal counting board that would correspond to the replacement values of Mesopotamian numerical notation. The number represented in both is 1.11.12 (= decimal 4272). The error recorded in MDP 34: 19 rev. 1–12, among other evidence, suggests the organization of the board on the right (Fig. 28b), which may possibly be reflected in the SANGA sign (Fig. 27).

472

Christopher Woods

Fig. 29: An example of addition on a sexagesimal counting board of the type given in Fig. 28b, in which upper and lower registers distinguish 10 × and 1 × counters for each column: (a) the addend of 39.56 is the first set on the board; (b) the second addend of 21.45 is set below the first; (c) applying the alternating replacements of 10 (60n lower register) and 6 (10 × 60n upper register) yields the answer of 1.1.41.

Fig. 30a-b: A hypothetical, schematic representation of a Mesopotamian five-column, fixedbead abacus, in which value is assigned to beads when they are moved towards the traverse bar as they are on a soroban. The abacus represents a minimal 5 : 9 design. The use of white beads divides the cluster of 5 upper-register beads into subitized chunks of 2-1-2 and the lower-register beads into chunks of 4-1-4, similar to a Russian schety. The abacus on the left (Fig. 30a) is set to zero; the abacus on the right (Fig. 30b) represents 37.0.42.26.40, which is the equivalent of decimal 479,672,800. It must be stressed that the purpose of this schema is to exemplify the aforementioned organizational principles, rather than to advocate for specific, physical aspects of the design, which may have been quite different from those depicted.

The Abacus in Mesopotamia: Considerations from a Comparative Perspective

473

Fig. 31: Calculations that involve splitting big, i.e., wide, numbers into two pieces are envisioned as taking place on two open-ended fixed-bead abaci. The number represented is the last confirmed entry for ex. 10 (N 3958 rev. 16), which combines 7- and 4-position numbers. The use of gray represents columns that may be utilized in subsequent entries of the sequence.

474

Christopher Woods

Bibliography Adkins, Julia Elizabeth. 1956. An Historical and Analytical Study of the Tally, the Knotted Cord, the Fingers, and the Abacus. Ph.D. dissertation, The Ohio State University. Attinger, Pascal. 1993. Eléments de linguistique sumérienne: La construction de du11/e/di “dire”. Orbis Biblicus et Orientalis. Fribourg, Suisse/Göttingen: Editions universitaires: Vandenhoeck & Ruprecht. Barnard, Francis Pierrepont. 1916. The Casting-counter and the Counting-board: A Chapter in the History of Numismatics and Early Arithmetic. Oxford: Clarendon Press. Baxter, William T. 1989. Early Accounting: The Tally and Checkerboard. The Accounting Historians Journal 16: 43–83. Biggs, Robert D. 1967. Semitic Names in the Fara Period. Orientalia NS 36: 55–66. Biggs, Robert D. 1974. Inscriptions From Tell Abū Ṣalābīkh. OIP 99. Chicago & London: University of Chicago Press. Brokaw, Galen. 2010. A History of the Khipu. Cambridge Latin American Studies 94. New York: Cambridge University Press. Brown, Nancy Marie. 2010. The Abacus and the Cross: The Story of the Pope Who Brought the Light of Science to the Dark Ages. New York: Basic Books. Burnett, Charles and W. F. Ryan. 1998. Abacus (western). Pp. 5–7 in Instruments of Science: An Historical Encyclopedia, ed. Robert Bud and Deborah Jean Warner. Garland Encyclopedias in the History of Science. The Science Museum, London and The National Museum of American History, Smithsonian Institution. Burns Glynn, William. 2002. Decodificación de Quipus. Lima, Peru: Banco Central de Reserva del Perú/Universidad Alas Peruanas. Chen, Yifu. 2013. L’étude des Différents Modes de Déplacement des Boules du boulier et de l’invention de la Méthode de Multiplication Kongpan Qianchengfa et son Lien avec le Calcul Mental. Ph.D. dissertation, Université de Paris Diderot (Paris 7). Chen, Yifu. (forthcoming). On the Abacus Addition Education Before the Early 20th Century in China and Japan. In Mathematics Education in Digital Era. Dordrecht, The Netherlands: Springer. Chrisomalis, Stephen. 2010. Numerical Notation: A Comparative History. New York: Cambridge University Press. Chrisomalis, Stephen. (forthcoming). Hooke’s Abaci: European Interpretations of East Asian Arithmetical Instruments, 1600–1750. Civil, Miguel. 2000. From the Epistolary of the Edubba. Pp. 105–18 in Wisdom, Gods and Literature: Studies in Assyriology in Honour of W. G. Lambert, ed. Irving I. Finkel and Andrew R. George. Winona Lake, Indiana: Eisenbrauns. Damerow, Peter. 1996. Abstraction and Representation: Essays on the Cultural Evolution of Thinking. Dordrecht: Kluwer. Darlington, Oscar G. 1947. Gerbert, the Teacher. The American Historical Review 52: 456–76. Dehaene, Stanislas. 1993. Numerical Cognition. Cambridge, MA: Blackwell. Delougaz, Pinhas and Helene K. Kantor. 1996. Chogha Mish, Volume I. The First Five Seasons of Excavations, 1961–1971. Part 1: Text; Part 2: Plates, ed. Abbas Alizadeh. OIP 101. Chicago: Oriental Institute Publications, Chicago. Englund, Robert K. 1998. Texts from the Late Uruk Period. Pp. 15–233 in Mesopotamien: Späturuk-Zeit und Frühdynastische Zeit. Orbis Biblicus et Orientalis 160/1. Freiburg, Switerzerland/Göttingen: Univ.-Verl./Vandenhoeck & Ruprecht.

The Abacus in Mesopotamia: Considerations from a Comparative Perspective

475

Feigenson, Lisa. 2008. Parallel Non-verbal Enumeration is Constrained by a Set-based Limit. Cognition 107: 1–18. Fellmann, Rudolf. 1983. Römische Rechnentafeln aus Bronze. Antike Welt 14: 36–40. Flegg, Graham. 1983. Numbers: Their History and Meaning. New York: Schocken Books. Frank, Michael C. and David Barner. 2012. Representing Exact Number Visually Using Mental Abacus. Journal of Experimental Psychology: General 141: 134–49. Friberg, Jöran. 1986. The Early Roots of Babylonian Mathematics III. Three Remarkable Texts from Ancient Ebla. Vicino Oriente 6: 3–25. Friberg, Jöran. 1990. Mathematik. Reallexikon der Assyriologie und Vorderasiatischen Archäologie 7: 531–85. Friberg, Jöran. 2007. A Remarkable Collection of Babylonian Mathematical Texts. Manuscripts in the Schøyen Collection: Cuneiform Texts I. Sources and Studies in the History of Mathematics and Physical Sciences. New York: Springer. Guaman Poma de Ayala, Felipe. 1980 [1612]. El primer nueva corónica y buen gobierno, ed. John V. Murra and Rolena Adorno. Mexico City: Siglo Veintiuno. Green, Anthony (ed.). 1993. The 6G Ash-Tip and Its Contents: Cultic and Administrative Discard from the Temple? Abu Salabikh Excavations 4. London: British School of Archaeology in Iraq. Green, M. W. and Hans J. Nissen. 1987. Zeichenliste Der Archaischen Texte Aus Uruk. Archaische Texte aus Uruk 2. Berlin: Gebr. Mann Verlag. Gullberg, Jan. 1997. Mathematics From the Birth of Numbers. New York: W. W. Norton. Halberda, Justin, Michèle M. M. Mazzocco, and Lisa Feigenson. 2008. Individual Differences in Non-verbal Number Acuity Correlate with Maths Achievement. Nature 455: 665–68. Hatano, Giyoo, Yoshio Miyake and Martin G Binks. 1977. Performance of Expert Abacus Operators. Cognition 5: 47–55. Hatano, Giyoo and Keiko Osawa. 1983. Digit Memory of Grand Experts in Abacus-derived Mental Calculation. Cognition 15: 95–110. Hénin, Silvio. 2012. Early Italian Computing Machines and Their Inventors. Pp. 204–30 in Reflections on the History of Computing: Preserving Memories and Sharing Stories, ed. Arthur Tatnall. IFIP Advances in Information and Communication Technology 387. Heidelberg: Springer. Henkelman, Wouter F. M. and Margaretha L. Folmer. (forthcoming). Your Tally is Full! On Wooden Credit Records in and after the Achaemenid Empire. Pp. 129–226 in Silver, Money and Credit: Festschrift for Robartus J. Van Der Spek on Occasion of His 65 th Birthday on 18 September 2014, ed. Kristin Kleber and Reinhard Pirngruber. Leiden: Nederlands Instituut voor het Nabije Oosten. Høyrup, Jens. 2002a. A Note on Old Babylonian Computational Techniques. Historia Mathematica 29: 193–98. Høyrup, Jens. 2002b. Review of Christine Proust (2000), La multiplication babylonienne: La part non écrite du calcul. MathSciNet American Mathematical Society Mathematical Reviews MR1858598 (2002i:01004). Høyrup, Jens. 2002c. Lengths, Widths, Surfaces: A Portrait of Old Babylonian Algebra and Its Kin. Sources and Studies in the History of Mathematics and Physical Sciences. New York: Springer. Ifrah, Georges. 2000. The Universal History of Numbers: From Prehistory to the Invention of the Computer. Trans. David Bellos. New York: John Wiley & Sons, Inc. Jami, Catherine. 1998. Abacus (eastern). Pp. 3–4 in Instruments of Science: An Historical Encyclopedia, ed. Robert Bud and Deborah Jean Warner. Garland Encyclopedias in the

476

Christopher Woods

History of Science. London/New York: The Science Museum, London and The National Museum of American History, Smithsonian Institution. Johnson, Justin Cale and Markham J. Geller. 2015. The Class Reunion: An Annotated Translation and Commentary on the Sumerian Dialogue, Two Scribes. Cuneiform Monographs 47. Leiden/Boston: Brill. Kaufman, Edna L., M. W. Lord, T. W. Reese, and John Volkmann. 1949. The Discrimination of Visual Number. The American Journal of Psychology 62: 498–525. Keyser, Paul. 1988. The Origin of the Latin Numerals 1 to 1000. American Journal of Archaeology 92: 529–46. Kojima, Takashi. 1954. The Japanese Abacus, Its Use and Theory. Tokyo: Charles E. Tuttle Company. Kojima, Takashi. 1963. Advanced Abacus: Japanese Theory and Practice. Tokyo: Charles E. Tuttle Company. Kwa, Tak Ming. 1922. The Fundamental Operations in Bead Arithmetic: How to Use the Chinese Abacus. San Francisco, CA: Service Supply Company. Lang, Mabel. 1957. Herodotus and the Abacus. Hesperia: The Journal of the American School of Classical Studies at Athens 26: 271–88. Lang, Mabel. 1964. The Abacus and the Calendar. Hesperia: The Journal of the American School of Classical Studies at Athens 33: 146–67. Lang, Mabel. 1965. The Abacus and the Calendar, II. Hesperia: The Journal of the American School of Classical Studies at Athens 34: 224–47. Lang, Mabel. 1968. Abaci from the Athenian Agora. Hesperia: The Journal of the American School of Classical Studies at Athens 37: 241–43. Lay-Yong, Lam. 1987. Linkages: Exploring the Similarities Between the Chinese Rod Numeral System and Our Numeral System. Archive for History of Exact Sciences 37: 356–92. Lieberman, Stephen J. 1980. Of Clay Pebbles, Hollow Clay Balls, and Writing: A Sumerian View. American Journal of Archaeology 84: 339–58. Liverani, Mario. 1983. Fragments of Possible Counting and Recording Devices. Origini 12: 511–21. John MacGinnis, M. Willis Monroe, Dirk Wicke, and Timothy Matney. 2014. Artifacts of Cognition: The Use of Clay Tokens in a Neo-Assyrian Provincial Administration. Cambridge Archaeological Journal 24: 289–306. Mackey, Carol, et al. (ed). 1990. Quipu y yupana: Colección de escritos. Lima, Peru: Consejo Nacional de Cienca y Techologia. Mandler, George and Billie J. Shebo. 1982. Subitizing: An Analysis of Its Component Processes. Journal of Experimental Psychology: General 111: 1–22. Martzloff, Jean-Claude. 2006. A History of Chinese Mathematics. New York: Springer. Menninger, Karl. 1969. Number Words and Number Symbols: A Cultural History of Numbers. Revised ed. Cambridge, MA: M. I. T. Press. Miller, George A. 1956. The Magical Number Seven, Plus or Minus Two: Some Limits on Our Capacity for Processing Information. Psychological Review 63: 81. Moon, Parry Hiram. 1971. The Abacus: Its History, Its Design, Its Possibilities in the Modern World. New York: Gordon and Breach Science Publishers. Murray, Harold J. R. 1951. A History of Board-games Other Than Chess. Oxford: Clarendon Press. Needham, Joseph. 1959. Science and Civilisation in China, vol. 3: Mathematics and the Sciences of the Heavens and Earth. Cambridge: Cambridge University Press.

The Abacus in Mesopotamia: Considerations from a Comparative Perspective

477

Nesselmann, Georg Heinrich Ferdinand. 1842. Versuch einer kritischen Geschichte der Algebra. Erster Theil: Die Algebra Der Griechen. Berlin: G. Reimer. Netz, Reviel. 2002. Counter Culture: Towards a History of Greek Numeracy. History of Science 40: 321–52. Nissen, Hans J., Peter Damerow, and Robert K. Englund. 1993. Archaic Bookkeeping: Early Writing and Techniques of Economic Administration in the Ancient Near East. Trans. Paul Larsen. Chicago, London: The University of Chicago Press. Oppenheim, A. Leo. 1959. On an Operational Device in Mesopotamian Bureaucracy. JNES 18: 121–28. Ouyang, Xiaoli and Christine Proust. (forthcoming). Place Value Notations in the Ur III Period: Marginal Numbers in Administrative Texts. In Cultures of Computation and Quantification, ed. Karine Chemla, Agathe Keller, and Christine Proust. Dordrecht, The Netherlands: Springer. Postgate, J. Nicholas. 1980. Excavations at Abu Salabikh, 1987–79. Iraq 42: 87–104. Powell, Marvin A. 1971. Sumerian Numeration and Metrology. Ph.D. dissertation, University of Minnesota. Powell, Marvin A. 1987. Maße und Gewichte. Reallexikon Der Assyriologie 7: 457–530. Powell, Marvin A. 1995. Metrology and Mathematics in Ancient Mesopotamia. Pp. 1941–1957 in Civilizations of the Ancient Near East, ed. Jack M. Sasson. Peabody, MA: Hendrickson Publishers, Inc. Proust, Christine. 2000. La multiplication babylonienne: La part non écrite du calcul. Revue D’histoire Des Mathématiques 6: 293–303. Proust, Christine. Review of Alain Schärlig (2003), Compter avec des jetons: Tables à calculer et tables de compte du Moyen Age à la Révolution. Pour La Science 313: 102. Proust, Christine. 2012. Interpretation of Reverse Algorithms in Several Mesopotamian Texts. Pp. 384–422 in The History of Mathematical Proof in Ancient Traditions, ed. Karine Chemla. Cambridge: Cambridge University Press. Pullan, J. M. 1969. The History of the Abacus. New York: F. A. Praeger. Quilter, Jeffrey. 2002. Yncap Cimin Quipococ’s Knots. Pp. 197–224 in Narrative Threads: Accounting and Recording in Andean Khipu, ed. Jeffrey Quilter and Gary Urton. Austin, TX: University of Texas Press. Radicati di Primeglio, Carlos. 1976. El sistema contable de los Incas : Yupana y quipu. Lima, Peru: Librería Studium. Radicati di Primeglio, Carlos. 1990. Tableros de escaques en el antiguo Peru. Pp. 219–34 in Quipu y yupana: Colección de escritos, ed. Carol Mackey Lima, Peru: Consejo Nacional de Cienca y Techologia. Reynolds, Barbara E. 2008. Pp. 1–3 in The Encyclopedia of the History of Science, Technology, and Medicine of Non-Western Cultures (2nd ed.), ed. Helaine Selin. Berlin/ Heidelberg/New York: Springer-Verlag. Robson, Eleanor. 2008. Mathematics in Ancient Iraq: A Social History. Princeton, N.J.: Princeton University Press. Rudman, Peter S. 2007. How Mathematics Happened: The First 50,000 Years. Amherst, New York: Prometheus Books. Sachs, Abraham J. 1947. Babylonian Mathematical Texts I. Reciprocals of Regular Sexagesimal Numbers. Journal of Cuneiform Studies 3: 219–40. Salomon, Frank. 2002. Patrimonial Khipu in a Modern Peruvian Village. Pp. 293–319 in Narrative Threads: Accounting and Recording in Andean Khipu, ed. Jeffrey Quilter and Gary Urton. Austin, TX: University of Texas Press.

478

Christopher Woods

Schärlig, Alain. 2001. Compter avec des cailloux: Le calcul élémentaire sur L’abaque chez les anciens Grecs. Lausanne: Presses polytechniques et universitaires romandes. Schärlig, Alain. 2003a. Compter avec des jetons: Tables à calculer et tables de compte du Moyen Age à la Révolution. Lausanne: Presses polytechniques et universitaires romandes. Schärlig, Alain. 2003b. Rechnen ganz einfach: Der römische abacus – sein griechischer Vorfahre und seine Nachkommenschaft bis nach Japan. Antike Welt 34: 517–20. Schmandt-Besserat, Denise. 1979a. An Archaic Recording System in the Uruk-Jemdet Nasr Period. American Journal of Archaeology 83: 19–48. Schmandt-Besserat, Denise. 1979b. Reckoning before writing. Archaeology 32: 22–31. Schmandt-Besserat, Denise. 1992. Before Writing, Volume I: From Counting to Cuneiform. Austin: University of Texas Press. Smith, David Eugene and Yoshio Mikami. 1914. A History of Japanese Mathematics. Chicago: The Open court publishing company. Steinkeller, Piotr. 2013. An Archaic “Prisoner Plaque” from Kiš. RA 107: 131–57. Stigler, James W. 1982. Abacus Skill in Chinese Children: Imagery in Mental Calculation. Ph.D. dissertation, University of Michigan. Stigler, James W. “Mental Abacus”: The Affect of Abacus Training on Chinese Children’s Mental Calculation. Cognitive Psychology 16: 145–76. Stigler, James W., Laurence Chalip, and Kevin F. Miller. 1986. Consequences of Skill: The Case of Abacus Training in Taiwan. American Journal of Education 94: 447–79. Taisbak, Christian M. 1965. Roman Numerals and the Abacus. Classica Et Mediaevialia 26: 147–60. Tani, Yukio. 1964. The Magic Calculator: The Way of Abacus. Rutland, VT/Tokyo: Japan Publications Trading Co. Tun, Molly. (forthcoming). Yupana. In The Encyclopedia of the History of Science, Technology, and Medicine of Non-Western Cultures (3rd ed.), ed. Helaine Selin. Berlin/Heidelberg/ New York: Springer-Verlag. Tun, Molly and Luis Octavio Montaluisa Chasiquiza. (forthcoming). Taptana. In The Encyclopedia of the History of Science, Technology, and Medicine of Non-Western Cultures (3rd ed.), ed. Helaine Selin. Berlin/Heidelberg/New York: Springer-Verlag. Visicato, Giuseppe. 2000. The Power and the Writing: The Early Scribes of Mesopotamia. Bethesda, MD: CDL Press. Volkov, Alexeï. 1994. Large numbers and counting rods. Extrême-Orient Extrême-Occident 16: 71–92. Volkov, Alexeï. 2002. Le bacchette. Pp. 125–33 in Enciclopedia Italiana, vol. II: La Scienza in Cine, India, Americhe, section I: Storia della Scienza, ed. Karine Chemla. Rome: Instituto della Enciclopedia Italiana. Woods, Christopher. (forthcoming). New Light on Administrative Device from the Dawn of Writing in the Ancient Near East.

Index of texts and compositions discussed 2H-T25 408 n. 66 3N-T 914 (proverb) 265 n. 8 4R 11 400 4R2 53 (lament catalogue) 86 a-ab-ba hu-luh-ha (balag) 90 Aa 1/2 30 (MSL 14 p. 209) 368 AAICAB 1/3: Bod. A 37 241 AAICAB 1/3: Bod. S 160 233 AAS 81 198 n. 43 AbB 3: 8 18 AbB 7: 116 28 AbB 10: 4 20 AbB 10: 15 19 AbB 10: 41 19 AbB 11: 11 13 AbB 11: 40 19 AbB 11: 108 21 AbB 14: 164 19 Adapa 262–279 AfO 3: 154 (Assur-dān II) 21 Aḥīqar 288 Amar-Sîn to Šulgi 220 An = Anum 268 an-gim 179 n. 32, 210, 211, 213–14, 224, 271, 275 n. 32 AnOr 1: 134 251 n. 6 AnOr 7: 315 370 AnOr 7: 316 198 n. 43 AnOr 7: 374 251 n. 6 An-ta-gal2 A 112 (MSL 17) 15 Anzu 26 AO 6456 443, 447, 450 n. 61 AoF 24: 39 (Iddin-Nūnum, Dēr) 21–22 ARET 13: 1 285 ARM 6: 51 // 52 24 ARM 26/1: 142 24 ARM 27: 1 18 ARM 28: 95 23 n. 22, 27 AS 22: 24 (Tell Asmar) 23 Ashm 1913–0318 2–10 (see BE 31: 7 and OECT 5:31) Ashm. 1924–0667 251 n. 5, 254 n. 19 ASJ 8: 113 no. 30 242 ASJ 13: 224 no. 70 241 https://doi.org/10.1515/9781501503696-024

ASJ 13: 225 no. 71 241 ASJ 14: 241 no. 8 241 Astrolabe B 201, 212, 214, 223 a-še-er gi6-ta (balag) 89, 96, 97 AUCT 1: 793 243 AUCT 2: 200 201 n. 51 AUCT 3: 358 316 n. 11 AUWE 23: 127 See W 17259 m AWL 124 163 Babylonian Gilgameš 17, 401 n. 28, 403 n. 49, 406 BaM 10, 129 (W 23830d) 373 BAM 235 // BAM 236 60 BE 20/1: 29 440 n. 48 BE 30/1: 11 See CBS 10085 BE 31: 7 2–10 (see OECT 5: 31) Bible – Deut 25:4 14 n. 4 – Ruth 2:17 26 – Isa 27:12 26 – Luke 3:17 21 BIN 8: 23 162 BIN 8: 347 164 Bird & Fish 303 n. 13, 410 n. 73 Bīt Rimki 277 BL 95+ (tigi) 277 BM 12917 242 BM 12945 242 BM 13002A 241 BM 15296 241 BM 27084 241 BM 85983 363–73, 390–93 BM 94502 242 BM 96691 (balag) 97 BM 81014 + BM 132267 86, 87–88, 89–90, 98, 99 BM 103436 243 BM 130829 317 BM 110126 250 n. 1 BM 111750 253 n. 18 BM 113032 236 BM 113089 235 BM 132095 90 BPOA 1: 207 242

480

Index of texts and compositions discussed

BPOA 1: 652 254 n. 25 BPOA 1: 852 233–34 BPOA 1: 918 235 n. 6 BPOA 2: 2545 235 BPOA 2: 2681 193 n. 21 BPOA 2: 2685 246 n. 14 BPOA 6: 4 254 n. 25 BPOA 7: 1721 235 BPOA 7: 1831 237 BPOA 7: 1893 235 BPOA 7: 2072 192 n. 16 BPOA 7: 2147 316 n. 10, n. 11 BPOA 7: 2668 243 BPOA 7: 2721 168 Būr-Sîn Hymn A 6 BWL: 253 (K 8216) 30 CBS 29-15-497 443 CBS 3558+ (Lipit-Eštar A) 270, 271 CBS 3832 (Inanna & An) 268, 271 CBS 6890 (lament) 96 CBS 7946 (in-nin ša3 gur-ra) 271 CBS 8309 443 CBS 8547 + N 3465 + N 6898 269 n. 19 CBS 9899 (Two Women B) 269 n. 19 CBS 10070 271 CBS 10085 (Enlil lament) 84–85 CBS 10295 269 n. 19 CBS 10417 (Enlil lament) 83–101 CBS 10433 271 CBS 10475 (En-lil2 su3-ra2-še3) 263 n. 6, 271 CBS 10900 (Šulgi O?) 271 CBS 10903 (En-lil2 su3-ra2-še3) 271 CBS 10986 (Lamašaga hymn) 270 CBS 11153 (an-gim) 271 CBS 11945 (Counsels of Wisdom) 271 CBS 12674 (OB lament fragment) 89, 97 CBS 13509 (Šulgi B) 271 CBS 13777 (Early Rulers) 271 CBS 13860 (in-nin ša3 gur-ra) 272 CBS 13905 (Udughul) 268, 272 CBS 13990 269 n. 19 CBS 14075 272 CBS 15080 (incantation) 272 CBS 15143 272 CBS 15203 (in-nin ša3 gur-ra) 272, 435 n. 36

CDLJ 2003/1: 1 235 CLAM 97–98 174 n. 13 CM 31: 286 no. 15 316 n. 11 CT 5: pl. 36 242 CT 29: 48–49 398 n. 11 CT 36: pl. 42–43 See BM 96691 CT 42: 1 See BM 132095 CT 58: 39A See BM 81014 + BM 132267 CT 58: 64 276 n. 40 CTMMA 1: 17 168 CTNMC 5 201 n. 51 Curse of Akkade 174 n. 13, 400 CUSAS 3: 43 234 CUSAS 3: 185 234 CUSAS 3: 189 234 CUSAS 3: 205 234 CUSAS 3: 235 245 n. 13 CUSAS 3: 236 245 n. 13 CUSAS 3: 237 244 n. 13 CUSAS 3: 246 244 n. 13 CUSAS 3: 258 244 n. 13 CUSAS 3: 261 234 CUSAS 3: 287 234 CUSAS 3: 288 234 CUSAS 3: 289 234 CUSAS 3: 290 234 CUSAS 3: 414 234 CUSAS 3: 462 233 CUSAS 3: 553 233, 234, 235, 237–39 CUSAS 3: 561 238 n. 9 CUSAS 3: 884 244–45 n. 13 CUSAS 3: 1269 373 CUSAS 8: 53 320 CUSAS 8: 54 318 CUSAS 8: 267 320 CUSAS 12: 7.1 366 CUSAS 17: 101 197 Death of Gilgameš 223 Dialogue 1 434 Dialogue 3 276 Dialogue between a scribe and his son Diatribe C 410 n. 70 Diatribe C 2 408 n. 65 Diri 433 Diri 6B (MSL 15 p. 192) 368 Diri Nippur OB Seg.9, 7 308 Diri OB “Oxford” 342 303 n. 11

401

Index of texts and compositions discussed

481

DoCuEPHE: 24 255 n. 31 DP 128 163, 165 DP 129 163 DP 132 163 DP 133 163, 164 n. 14 DP 157 162 DP 160 161–62 DP 218 161 n. 8 DP 226 163 DPA 30 408 n. 65 Dumuzi-Inanna F 179 n. 31 Dumuzi-Inanna P 55–56, 61, 63, 65 Dumuzi’s Dream 407 n. 58

HAR-ra 6 372 n. 13, 373 HAR-ra 11 366 HAR-ra 2 OB 371 HAR-ra OB Nippur 367 Hendursaga Hymn 267, 396–413 Heron & Turtle 403 n. 38, 411 n. 75 Hoe & Plough 66 Home of the Fish 401, 410 n. 73 HMA 9–1811 368 HS 1512 (prayer) 273 HS 1659+ 364 n. 3 HS 1799 364 n. 3 HSM 7165 236

Ea 4: 194 366 edin-na u2-sag-ga2-ke4 (balag) 96 Edubba A 276 Edubba C 276 Edubba E 276 Ekur Hymn 172–186 e-lum gu4-sun2 (balag) 90 Emar 6/1: 257 367 Enki & Ninhursag 57–58, 61, 405 n. 52 Enki & the World Order 177 n. 23 Enlil hymn 174–75 n. 14 Enlil, the Merchant 82–83 Enlil & Ninlil 57, 181 Enlil & Namzitarra 37–52, 402 Enlil & Sud 220 En-lil2 su3-ra2-še3 263 n. 6, 277 Enmerkar & Ensuhkešdanna 405 n. 52 Enmerkar & the Lord of Aratta 400 Erim-ḫuš (MSL 17) 16 Eršemma 60 401 n. 23, 402 n. 33 Ešnunna CodeA ii 40 23 Examenstexte 276 n. 40

Iddin-Dagan D 373, 403 n. 38 Inanna H (Balbale to Inanna) 55, 63, 181 Inanna & An 268, 276 Inanna & Dumuzi A 62–63 Inanna & Šukaletuda 411 n. 75 Inanna’s Descent 263, 268, 276 Ininšagurra 219, 436–7 ISET I: 221–22 See L.1492 Išbi-Erra B 410 n. 74 Išme-Dagan I (Išme-Dagan & Enlil’s Chariot) 60 n. 17 Išme-Dagan J 181 Išme-Dagan K 342 n. 7 Išme-Dagan M 180 n. 38 Išme-Dagan Q 6 Išme-Dagan W 177 n. 21 Ištar Louvre 15 Ištar’s Descent 263 ITT 2: 907 196 n. 34, 197 ITT 2: 3772 197 n. 43

FAOS 6, 340 305 n. 19 Farmer’s Instructions 14, 242 n. 11 Gilgamesh & Agga 81–82 Gilgameš & Huwawa 370, 410 n. 72 Green 2013–3152 231–33 Gudea Cyl. B 177 n. 21, 178 n. 27 Gudea hymn A (tigi to Baba) 173 n. 7 Hammurabi’s Code 13 HAR-ra 266 n. 9, 267 n. 13, 363–393, 433, 434 n. 34

JCS 24: 163 no. 68 242 JCS 54: 6 no. 41 242 K 242 + K 4574 obv. i 9′ = P393806 372, 373 K 7751 (MSL 14, pp. 276–77) 368 KAR 64 398 KAR 97 211, 212 KAR 119 207–225, 277 KAR 128 (+) 129 212 KAR 158 (song catalogue) 276 nn. 38–39, 277, 278 KAV 218 214

482

Index of texts and compositions discussed

Keš Temple Hymn 172 n. 1, 177 n. 21, 181 KM 89100 243 Kt 90/k, 178 (OAss incantation) 59–60 K. 263+10934 (NBab) 60 L.1492 (eršemma) 96 Laments (OB Nippur) 87 n. 7 Lament over Ur 400 n. 19, 411 n. 75 Lament over Sumer & Ur 178 n. 29, 401, 410 n. 73, 411 n. 75 Limet Métal: 286 14 370 Lipit-Eštar A 277 Lipit-Eštar B 44 n. 23 Lipit-Ištar & the Plow 263 n. 6, 276 Lisina Tale 38 LKA 102` See ša2.zi.ga Lugalbanda & the Anzu Bird 410 Lugalbanda in the Mountain Cave 401 n. 28, 411 n. 75 Lugal-e 179 n. 32, 210–11, 213–14, 224, 275 n. 32, 409, 411, 409 n. 69, 411 Lugal-ibila to Lugal-nesag 276, 435 L’oumo 47 316 n. 10 Luzag. 1 2:17–28 305 n. 19 Luzag. 3:22–23 305 n. 19 Malku 2: 245 366 Malku 4: 204 29 Malku 6: 153 433 Man & His God 10 27 Marriage of Martu 223 MBGT see MSL SS 1 MCS 1: 37 (BM 105452) 370 MCT, p. 13 n. 69 443 MDP 18: 250 26 MDP 34: 12 450 n. 61 MDP 34: 19 439–40, 447, 450, 471 MDP 34: 24 450 n. 61 MEE 3: 73 440 n. 48 MS 2291 269 n. 19 MS 3047 432 MS 3370 viii Msk 731084z =P271379 371 Msk 74171c = P250373 371 MSL 6, p. 79 373 MSL SS 1 (MBGT) 16 MTBM 327 241 MTBM 329 242

mul.apin 1 396 n. 5, 405 n. 48 mu-tin nu-nuz dim2-ma (balag) 91 n. 12, 98, 99 MVN 2: 8 242 MVN 3: 312 316 n. 7, 316 n. 10, n. 11 MVN 5: 57 193 n. 20 MVN 5: 87 316 n. 10, n. 11 MVN 6: 257 241 MVN 6: 539 242 MVN 7: : 175 241 MVN 9: 6 241 MVN 10: 105 234–35 MVN 10: 128 240, 241 MVN 11: 90 240 MVN 12: 26 104, 106 MVN 12: 84 241 MVN 12: 100 241 MVN 12: 102 241 MVN 12: 179 104–5 MVN 12: 245 241 MVN 12: 464 233 MVN 12: 466 233 MVN 13: 312 233 MVN 13: 222 198 MVN 13: 311 315 n. 4 MVN 14: 312 235 MVN 16: 1142 370 MVN 16: 1593 236 MVN 18: 404 235, 236 MVN 18: 644 237 MVN 18: 689 235 MVN 21; 208 315 n. 4 N 1634 273 N 1747 (Enlil & Ninlil) 273 N 1751 + N 4523 276 n. 38 N 1789 268 n. 14, 276 N 2203 (Enlil & Sud) 273 N 2243 (Enlil & Ninlil) 273 N 2306 273 N 2875 + N 4113 273 N 3395 (bilingual proverbs) 269, 270 N 3455 (Enlil & Ninlil) 273 N 3462 207–225, 273, 277 N 3465 + CBS 8547 + N 6898 269 n. 19 N 3495 (Lipit-Ištar & the Plow) 273 N 3498 (Lipit-Ištar & the Plow) 263 n. 6 N 3514 273

Index of texts and compositions discussed

N 3662 (Adapa) 262–279 N 3958 444, 450, 473 N 4188 + UM 29–13–596 (lament) 270 N 6064 276 N 6126 (Enlil & Ninlil) 273 N 6286 (an-gim) 271 N 6898 + N 3465 + CBS 8547 269 n. 19 Nabnītu 7: 210 f. (MSL 16 p. 111) 371 Nabnītu 25 403 n. 36 Nanna E 178 n. 25 & n. 29 Nanna L 173 n. 7 Nanna-manšum to Rīm-Sîn 1–10 Nanna’s Journey to Nippur 177 n. 23, 178 n. 29, 411 n. 75 Nanše & the Birds 404 n. 42, 46 Nanše Hymn 436 NATN 35 242, 243 NATN 59 241, 243 NATN 104 242 NATN 444 242 NATN 447 233, 242 NATN 451 233, 242 NATN 455 243 NATN 461 242 NATN 462 198 NATN 568 242 NATN 727 242 NATN 739 242, 243 NATN 748 240 NATN 757 243 NergalD 275 n. 32 Ni 2676+ (Udughul) 273 Ni 9602 See Dumuzi-Inanna P Ni 9734 + UM 29–13–419A 273 Ni 13227 (Šulgi O?) 273 Nigga 540 303 n. 11 Nigga Bilingual B 91 371 nig2-nam nu-kal 48 Nik 1: 209 161 n. 8 Nik 1: 227 178 n. 25 Nik. 2: 114 235 Nik. 2; 141 235 n. 6 Ningišzida Hymn 8–9 Ningišzida D 176 n. 20 Nin-Isina A 373 nin me šar2-ra 277 Ninšatapada to Rīm-Sîn 7 Ninurta A 224

483

Ninurta C 179 n. 31, 405 n. 50 Ninurta’s Journey to Eridu 179 n. 31, 224 Ninurta prayers 180 n. 38 nir-gal2 lu2 e3-ne (bala) 211 n. 13 Nisaba 3: 123 198 n. 45 Nisaba 5: 231 199 Nisaba 7: 12 240, 241 Nisaba 10: 61–62 241 Nisaba 10: 74 241 Nisaba 10: 87 241 Nisaba 15: 56 199 Nisaba 15/2: 146 196 Nisaba 15/2: 156 315 n. 5 Nisaba 15/2: 412 312 Nisaba 15/2: 462 370 Nisaba 15/2: 668 110–130, 140 Nisaba 15/2: 1031 110–130 Nisaba 15/2: 1032 110–130 Nisaba 17: 81 241 Nisaba 23: 7 235 n. 5 Nisaba 23: 10 237 Nisaba 23: 33 235 n. 5 Nisaba 23: 50 316 n. 7 Nisaba 24: 9 316 n. 7 Nisaba 24: 10 235, 237 Nisaba 24: 37 199 Nisaba 30: 50 192 n. 16 Niṣirti bārūti (AOAT 326: no. 65) 25 Nothing is of Value 38 Nungal A 407 n. 58 Nuska B 179 n. 31, 224 NYPL 335 252 n. 8 OB catalogue in the Louvre 403 n. 36 OBGT VI 384 OBGT VII 384 OBGT X 374, 384 OECT 5: 31 2–10 (see BE 31:7) OIP 14: 65 164 OLA 8: 24 no. 21 316 n, 7 Ontario 1: 81 201 n. 51 OrSP 18, pl. 14 no. 39 198 n. 45 OrSP 47–49: 382 246 n. 14 OrSP 47–49: 465 200 n. 50 P431315 374–389 Padānu (Koch 27:18) 23 Pān tākalti (Koch 59.64) 24

484

Index of texts and compositions discussed

Pān tākalti (Koch 59.66) 23 Pān tākalti Comm. (Koch 89 v 22) 24 PBS 1/1: 11 (bilingual lit.) 269 PBS 10/2: 17 See CBS 6890 PBS 11/1: 7 (Inanna-teš2) 278 PBS 12: 29+ 265 n. 8 PDT 1: 461 201 n. 51 PDT 1: 541 192 n. 16 PDT 2: 921+ 242, 243 PDT 2: 932 240 PDT 2: 933 240 PIHANS 100 (Michel, Fs. Larsen) 22 Plowing Oxen 276 PPAC 5: 605 242 PPAC 5: 619 240, 241 PPAC 5: 964 241 PPAC 5: 984 241 PPAC 5: 1117 241 PRAK B 389 (Enlil lament) 84 Princeton 1: 509 236 Proto-Lu2 410 n. 74 Proverb Collections 1: 128 402 Proverb Collections 2 278 Proverb Collections 2: 54 411 n. 75 Proverb Collections 2: 67 402 n. 31 Proverb Collections 2: 100 305 n. 19 Proverb Collections 2: 109 401 Proverb Collections 5: 123 399 n. 16 Proverb Collections 8 405 n. 52 Proverb Collections 8 D 2 402 n. 31 Proverb Collections 11.70 265 n. 8 Proverb Collections 28: 6 404 n. 46 RA 7: 153 (Šamši-Adad) 24 RA 27: 142 23 n. 22 RA 40: 91 23 n. 22 RA 53: 35 (Kiš) 25 RA 71: 102 164 n. 12 RA 79: 30 no. 23 240 RAcc 89 402 n. 34 RIMB 2.3.1 (Simbar-Šipak) 15 RIME 1/9.5.17 (Enmetena) 178 n. 27 RIME 1/9.9.1 178 n. 25 RIME 1/13.5.2 (Mesannepada) 161 n. 7 RIME 3/2.1.1.10 (Ur-Namma) 76 n. 19 RIME 3/2.1.1.31 (Ur-Namma) 76 n. 19 RIME 3/2.1.1.32 (Ur-Namma) 76 n. 19 RIME 3/2.1 2.2012 241

RIME 3/2.1.3.5 (Amar-Sîn) 77 n. 23 RIME 3/2.1.3.6 (Amar-Sîn) 77 n. 23 RIME 3/2.1.3.15 (Amar-Sîn) 76–78 RIME 4.3. 6.2004 (Hammurabi) 317 RIME 4.3. 10. 2001 (Ammiṣaduqa) 6 RIME 4.3.19 (Warad-Sîn) 9 Rīm-Sîn B 9, 173 n. 7 Rīm-Sîn C 6, 7 Rīm-Sîn D 7 Rīm-Sîn E 7 Rīm-Sîn G 7 RTC 12 165 n. 16 RTC 53 162 RTC 222 408 n. 65 Rulers of Lagaš 66 n. 38 SA 061 370 SAA 3: 32 410 n. 70 SAA 13: 45 397 SAA 14: 166 403 n. 39 SACT 2: 26 235 SANTAG 6: 119 255 n. 27 SANTAG 7: 108 178 n. 25 SAT 1: 420 233 SAT 2: 292 235 n. 5 SAT 2: 323 235 SAT 2: 1051 235 n. 5 SAT 2: 1053 235 n. 5 SAT 2: 1142 192 n. 16 SAT 3: 1189 198 n. 45 SAT 3: 1197 200 n. 48 SAT 3: 1654 235 SAT 3: 1657 236 SAT 3: 1733 237 SAT 3: 1821 316 n. 11 SAT 3: 2157 168 SBH 62 no. 33 400 Shepherd & Farmer 63–64 Si-du3 275, 276, 277, 279 Sîn-iddinam to Utu 405 n. 50 Sîn-iqīšam A 176 n. 20 SLT 37+SLT 46+ 364 n. 3 SLT 44 364 SLT 250 364 n. 3 SNAT 511 242 Song of the Hoe 436 SpTU 2: 53 372 SpTU 3: 111 373

Index of texts and compositions discussed

Sumer 43 (Tell Ḥaddād) 16 Sumerian King List 190 Sumerian Temple Hymns 43, 172 n. 1, 181 – 5 178 n. 26 – 51 6 – 66 177 n. 21 – 83 176 n. 20 – 148 179 n. 29 – 309 176 n. 20, 178–79 n. 29 Syria 19: 122 23 n. 22 ša2.zi.ga 62, 67 Šemšara Archives I: 36 24 Šemšara Archives I: 44 27 Šemšara Archives I: 411 23 n. 22 Šulgi A 276 n. 38, 407 n. 58 Šulgi C 411 n. 75 Šulgi D 178 n. 29, 179 n. 31, 403 n. 38 Šulgi F 178 n. 27 Šulgi G 173 n. 7, 178 n. 29 Šulgi N 276 n. 38 Šulgi O 180 n. 34 Šulgi Q 173 n. 7 Šumma izbu 5 397, 399 Šumma izbu 16 397 n. 11 Šurpu 5–6 405 The šumunda grass 402 Šu-Suen A 342 n. 7 Tale of the Fox 406 n. 54 TCL 5: 5680 252 n. 7 TCL 16: 86 See Šulgi F TCS 1: 237 243 TCS 1: 253 168 TCTI 1: 742 241 TCTI 1: 907 198 TCTI 2: 2816 253 n. 16 Tell Rimah 267 28 TH 02-T192 (Tell Ḥaddād) 28 TH see Sumerian Temple Hymns TIM 9: 88 14 n. 3 TMH 3: 53+4 277 n. 45 TMH 4: 49+88 224 TMH nF 1/2: 82+ 243 TMH nF 1/2: 88 242 TMH nF 1/2: 126 242 TMH nF 1/2: 131 242 TMH nF 1/2: 148 243 TMH nF 1/2: 171 241, 243

485

TMH nF 1/2: 225 199 TMH nF 1/2: 247 240 TMH nF 1/2: 249 240 TMH nF 1/2: 250 240 TMH nF 1/2: 253 240 TMH nF 1/2: 253 240 TMH nF 1/2: 293 242 TMH nF 1/2: 316 241, 242 TPTS 1: 248 373 TSA 5 163, 164 n. 14 TSŠ 50 446 n. 56 TSŠ 190 440 n. 48 TUT 160 233 TUT 177 241 UCP 9/2/1: 44 236 Udughul (Utukkū lemnūtu) 268, 272, 273, 277, 401, 402 n. 34, 405, 405 nn. 49 & 51, 408, 410 n. 74 UET 3: 242 200 n. 50 UET 3: 721 370 UET 6/1: 34 (riddle) 300 UET 6/2: 196 269 n. 20 UET 6/2: 216 402 n. 31 UET 6/3: 522 269 n. 20 Ugu-mu 403 n. 36 UM 29-13-419A + Ni 9734 273 UM 29-13-513 (nin me šar2-ra) 273 UM 29-13-560+ (Plowing Oxen) 274 UM 29-13-596 + N 4188 (lament) 270 UM 29-15-174 (proverbs) 270 UM 29-15-399 + Ni 9734 274 UM 29-16-35 (Inanna’s Descent) 268 UM 29-16-51 see Ekur Hymn Ur III catalogue at Yale 403 n. 36 Ur-Namma A 342 n. 7 Ur-Namma B (tigi) 173, 174 n. 13, 179 n. 33 Ur-Namma I 7 Ur-Namma Code 160–61, 163 Ur-Ninurta E 276 n. 39 uru2 am3-ma-i-ra-bi (balag) 82, 88, 97 uru2 hul-a-ke4 (balag) 90 Urukagina 6 287 UTI 3: 1845 193 n. 18 UTI 3: 1848 199 UTI 3: 1910 236 UTI 4: 2336 370 UTI 4: 2372 236

486

Index of texts and compositions discussed

UTI 4: 2401 233 UTI 4: 2585 237 UTI 4: 2823 252 n. 8 UTI 4: 2836 235 n. 6 UTI 4: 2881 236 UTI 5: 3232 315 n. 4 UTI 5: 3499 236 UTI 6: 3529 370 UTI 6: 3821 254 n. 25 UTI 6: 3825 316 n. 8 d utu-gin7 e3-ta (balag) 86 n. 5, 89, 90, 97 Utukkū lemnūtu See Udughul Utu šir3-gid2-da 224–25

VAS 27: 33 165 VAT 607 (lament) 95 VAT 611 (lament) 95 VAT 617 (eršemma) 95 VAT 1367 (Enlil lament) 84–85 VAT 1421 (Enlil lament) 84 VAT 1557 (Enlil lament) 84–85 VAT 8755 371 VAT 10610 see KAR 119 VE (Ebla) 285

VAS VAS VAS VAS VAS VAS VAS VAS VAS VAS VAS VAS VAS VAS

YBC 4704 443 YBC 4716 443 YBC 13524 371 Yondorf 24 103–109 YOS 4: 313 197 n. 43 YOS 10: 25 24 YOS 10: 33 27 YOS 10: 44 24 n. 24 YOS 11: 24 29 YOS 11: 61 402 n. 34

2: 2 See VAT 617 2: 11 See VAT 607 2: 26 See VAT 611 2: 71 See VAT 1557 2: 95 See VAT 1367 10: 179 See VAT 1421 14: 106 163, 164 14: 173 163 24: 43 276 n. 38 24: 120 214 25: 14 162 25: 37 162 25: 70 164 25: 71 164 n. 12

W 17259 m (Enlil lament) 84 WF 74 165 n. 16, 166 n. 18

ZA 49: 164 (Hammurabi) 29 za3-mi3 hymns 172 n. 1, 174 n. 14, 177 n. 21 zi-bu-u3-um zi-bu-u3-um (balag) 86 n. 5

Index of words and terms discussed Sumerian a – dug4 58 a-da-ab 265–66 a-da-ba 265–66 a-kušu2 304 a-ru-a 116 a-ša3 60–61 a-ša-an-gar3-ra 219 a-ša3 uzmušen 65 a2-aš-gar 370–71 a2 – dar 220 a2-la2 223 abbunnuna 269 n. 22 abrig 291 ad2 254 n. 21 ag2 301 urud aga-silig 370 agrig 305 an-za-gar3 233 ab-sin2 64–65 aga3-us2 113, 197 ambar 302 ar3-ra 112 ar3-tu munus 159, 162 aš(2)-dah 220–21 bal 404 n. 44 ba-al 56 n. 6 bala 213 giš bala-di 223 bar (mišlu) 266 kuš bar-ru-tum 367 bir9 301 bu 410 bu-bu (bur7, bu7) 411 bur 411 bulug-kin-gur4 373–74 bulugx (šim×uh3) 256 n. 39–40 bunig 302–303 (na4) bur 307 bur-sag 307 dan3-dan3 62 n. 24 de6 244 n. 13 https://doi.org/10.1515/9781501503696-025

du3 (for ak) 219 du3-a-tar 111, 116 n. 37 du6 65 dub-sar 437 dug4 175–76 dumu-ka 178 n. 27 e3 435 e2-duru5 114 n. 21 e2-gal 284–295 eme-sal 147–56 engur 302 er2 304 eren 253 n. 17 esir2 302 esir2-e2-a 254 n. 24 esir2-had2 255 n. 26 eš3 292 ga-du8 en 166 ga2 286 ga2-dub-ba 107 gada-mah 62 gal.zadim 350 gal4(-la) 56 gala 147–56 gam-gam-ma 257 n. 47 gamun2 252 n. 7 ganun 254 n. 25 gan2-ki-duru5 63 n. 28 garadin 221 gar3-du 197, 200–201 gašam 113 gaz 14 n. 4 geme2-kar-kid3 166 n. 18 gešpu2 223 gi (for gi-i3) 253 n. 15 gi-zi 177 n. 43 gid2 410 giri3 − sil 15 giri3-zal 373 giš-gi4-gal2 172–73 giš-i3 sur-sur 112 gu 62 gu-dili-a – e3 435, 449

488

Index of words and terms discussed

gu2 – gal2 9 geš gu2-si-ki-ir 373 gub 10, 374–85 gud5 303 gudu4 42–43 gug 403 gur zabar 106 ha-bu-da 371 ha-zi-(in) 371 he2-me-gi-na 371 heš5 366 hirin2 366 kuš hi-ru 366 hul2 177 n. 23 id2 56–60, 286 n. 4 igi – la2 223 im-babbar2 255 n. 35 im-ma 370 im-ri-a 219 im-na4 na 430–31 izkim 305 ka − de6 10 ka5-a 396–400 kab2-kud 237 ki-duru5 63–64 ki-ru-gu2 86, 172 kin-gurum-mazabar 373 kislah 61–63 kud-da 408 kul 403 kun-zi-da 235 kunin 302 kuš 363, 365 kušu2 408–410 la-ni 311 lagaš/s 307–308 lil2 410 lirum 223, 306 lirum4 306 kuš lu-ub2sar 367 lu2-igi-nigin2 164, 168 lu2-gi-zi 111, 117 lu2-im-na4 na 431

lugud4 221 lukur 305 ma2-an-na 65 mašda2 219 mi-ri2-za 252 n. 9 murub2 269 n. 22 murub4 269 n. 22 naga-si-e3 256 n. 36 nam 42 nam-dumu 178 n. 25, 180 n. 35 nam-ra-ak 116 narx 166 n. 18 urud nig2-dim2 gul 371–72 nig2-gig 209 kuš nig2-gal2-la 366 kuš nig2-hul-a 366 nig2-ka9 432 nig2-kas7-aka 250 giš nig2.šid 432–33, 448 kuš nig2-umbin-na 366 nin-immaxmušen 407–408 ninda 301 nindar 368 nu- (prefix) 380 nu-gig 159–168 nu-nu 62 n. 24 nu-uš 376 nu2 62 nu2mušen 307 nu11 307 nu11.bur 307–308 nu11:bur 307 pad 305 pa-ru-tum 367 pisan-dub-ba See ga2-dub-ba

kuš

ri (ra “beat”) 62 n. 24 ri (zâqu) 267 sa-har 166 n. 18 sa9 266 sa10 301–302 giš sag-kul 403 sanga 436–37

Akkadian

sanga-dub 437 si-is 366–67 si12-a 111, 116 n. 37 sig4-tu-tu 162–63 n. 10 sipa(d) 305 su-hi-rin/rum 221 su-gan 255 n. 28 sug 302 sug2 381–82 sukkal-mah 192 n. 15 šabra 241, 243 ša3-zu 165–66, 168 ša13-dub-ba See ga2-dub-ba šaga 366 šar2 429 šar2-gal 429 še – bad 61 n. 23 še-eb 94 še ur5-ra 193 n. 21, 196 še – (giš) ra 61 n. 23 še – us2 61 n. 23 še-li 253 n. 14, 254 n. 23 urudu šen 369–70 šeg3 306 šer3-šer3 371 šeš3 304 šešx 304 šid 436 šim 252 n. 13, 254 n. 22 šim-gan2 252 n. 11 šim-hi 257 n, 43 šim-im 256 n. 42 šir3-gid2-da 224 šu-i 320 šu – tu-tu 9 šub 14 n. 3 šub-lugal 164 šuku 240 šu-ur2-me 254 n. 20 tab 62 n. 24 tag-tag 62 n. 24 te8mušen 404 ti-a-ma 369 tu 305 tuku5 434

u2-tu-gu-um 432 u2-tu-uk 432 u2-tuk-ku 432 u3-suh5 252 n. 9 u4-sakar-gibil 65–66 ugamušen 40, 44 n. 20, 402–403 ugnim 303–304 um-me 159–168 um-me-da 159–168 umbisag 436 (kuš) ummud 306 unud 305 ur-bar-ra 405–406 ur-gir15 400 ur-sag 396 ur3 399, 400 n. 19 urudu uš2 367–68 urudu 363 uš2 368 uš-bar 112, 404 ušur3 163 n. 11 utukku 433–34, 448 uzmušen 65 za-am-ru-tum 371 za-gul 255 n. 33 za-ri2-in 369 za3 315 n. 4 za3-mu 315 n. 4 zabar 363 zabar-dab5 192 n. 15, 233 zag-šu-la2 372 zah3 111, 113 za-na 403 zar-(maš) 221 ze2-ze2 62 n. 24 zid2-ka-še3 105–6 zu2 (of plow) 57 zu2-lum 115

Akkadian abarakku 291 abnu 430 abšinnu 64 agarinnu 64–65 alāku 290

489

490

Index of words and terms discussed

allaḫaru 255 n. 30 alû 223 anpātu 407 n. 60 ardadillu 220 ardadu 220 arratum 221 artadillu 220 arurtu 220 a2-aš-ga2-ru 370 atwû 285 awātu 285 awû 285 balālu 16 bantu 64–65 barbaru 397 bubūtu 398 n. 12, 410 daṣû 220 diāšu 15 dubdimmu 431 n. 28 ekallu 284–295 epēru 398 n. 12 eriqqu 65 n. 36 errēšu 67 n. 39 erû 404 n. 41 ešmarû 371 eššebu 407–408 gukuru 257 n. 48 ḫabātu 12–33, 59 n. 12, 410 ḫabbātu 28–29, 31–32 ḫābitānu 31 n. 41 ḫābitānūtu 31 n. 41 ḫalālu 410 ḫalû 403 ḫubtu 13–14 ḫubuttātu 18, 20, 21, 32 ḫurru 372 ḫušû 371–72 imēru 399 ippi/īru 398 n. 12 izuzzu 374–85

kalab mê 398 kalab urṣi 397 kalbu 398 karzillu 373 kilīlu 407 n. 60 kirinnu 368 kušû 409 labānum 16 laqātu 403 maḫāṣu 62, 434 māḫiṣātu 433–34, 448 makāru 58 n. 11 mānaḫtu 398 n. 12 marru 59 marṣadu 373–74 masennu 291 n. 14 maškanu 61 n. 21, 371 merdītu 372–73 musukkatu 210 nadû 430 naptaqu 371 nārāru 29 nēšu 398–99 nidûtu 61 n. 21 niḫbussū 17 nikkassu 432 parūgu 221 n. 19 parūtu 367 pedû 366 pendu 403 qadû 407 n. 59 raḫiṣu 221 rakābu 62 rēṣūtum 221 ruṭibtu 63 sankuttu 370 sâru 405 n. 49 sarmāšu 221 sarru 220 sekēru 60 sikkūru 403

Turkic

ṣabāru 404 šaḫāṭu 16 n. 9, 25, 28 n. 30 šakāku 435 šamû 31 n. 40 šandabakku 107 šapiltum 370 šaqû 58 n. 11 šarbiš 223 šasurru 64–65 šer’u 64 šinnu 403 šubalkutu 430 šūlû 430 šūlûtu 219 šumekkû 431 n. 28

Aramaic

tarammu 221 tašgirtu-219 temmen(n)u 25 n. 25 terīqtu 61 n. 21 timbuttu 223 turbalû 61 n. 21 tupšikku 59

dūš 15 n. 6

ummu 64–65 urru 372 ūsu 65 n. 35 uššu 367–68 uttuku 433–34, 448 utullum 305 n. 19 zarinnu 369 zim(b)uḫaru 365–66

Arabic hawwata 285 haykal 284–294 hayta 285 hayyata 285 ḫbṭ 17, 20–21 kawsaj 409 n. 58 kōsaj (Iraq) 409 n. 68

haykəlā 284, 288 hbrk 291 hykl 288

Ethiopic haykal/ḫaykal 284

Hebrew ’abrek 291 n. 14 ’ănāphāh 407 n. 60

hêkhāl 284, 288, 290 ḥbṭ 17, 26, 32 lqṭ 403 n. 39

Hittite halent(i)u- 290

Hurrian a-i-gal-lu-ḫu 290 n. 13 ḫa-i-gal-la-at-ḫe 290 n. 13 haikalli 290, 294

Ugaritic hkl 284, 288 hwt 285

Persian gūše / kūse 409 n. 58

lqṭ 403 n. 39

Turkic

qaraš 409 n. 68

sāǧrī (Turkish çāğrī) 409 n. 67

491

Index of personal names discussed This index does not generally include the names of major and well-known rulers who are repeatedly mentioned throughout the volume; e.g., Gudea is not listed, but Namhani is.

A’akala 153 n. 18 Abī-simtī 190, 199 n. 47 A-bu-ba-qar 313 A-da-lal3 119 Aḫ-dam-ì-lí 119 Aḫušuni 111 Al-tuš 161–64 A-lu-lu 119 Ama-ša6 163 A-pi-la-núm 312 Aradmu 243–44 Arad-Nanna 194, 200 Ašgi-tillat-su(2) 312 Ayakalla 193 Babati 195 Ba-ab-bu 313 Baranamtarra 165 Bazi 241–242, 244 n. 13 Bēl-aḫḫā-iddina 213 Bēlī-ašarēd 320 Dukra 195 E2-dag-ga 235 n. 5 E2-mete 164–65 En-Amar-Suenara-ki’ag-Anna 193 Engar-zi 111, 121 Enheduanna 342–43, 346 En-me-šar2-ra 43 Erra-ḫābiš 30 n. 37 Erra-ḫābit 30 Gan-ezem 161–63, 165 Geme2-e2-dam 162–63 Geme2-id2-edin-na 161–63 Geme2-dNanše 161–62 Geme2-šu 162–62 Gisadu 166 Ḫabit-Sîn 30 https://doi.org/10.1515/9781501503696-026

IGI.RU?-NUN 345 Iḫbit-alsin 30 Iḫbit-elšunu 30 Ibni-Adad 320 Ilalum 195 Ilum-ḫābil 30 n. 37 Ilšu-muballiṭ 8 Ip-qú-ša 312 i-re-eb 119 Itti-ilīya-aḫbut 30 Karaindaš 321 Ki-na-at-e2-a 168 Ku8-ba-tum 167–68 Kubātum 190 Kudur-mabuk 8 Kuli 106 La-ma-ša 253 n. 18 Lu-Enlila 255 n. 29 Lu-Nanna 269 n. 21, 316 Lū-šalim-belī 320 Lugal-amar-ku3 314 Lugal-ebansa 255 nn. 26–27 Lugalkugzu 240, 242 Lugal-niglagare 251–54 Lu-kala 253 n. 18, 256 n. 37 Lumma 350–51 Lu-Nin-Šubur 251 n. 5–6 Marduk-balāssu-ereš 213, 214 Munus-sa6-ga 161–62 Namhani 194, 243 Nanna-kam 195 Nanna-manšum 1–10 Nanna-zišagal 194 Namzitara 37–52 d nanna-i3-gi 312 Naplanum 195

Index of personal names discussed

Nidupae 347–48 Nin-guru7 161–63 Nin-hedu 234 Nin-munuszida 241, 243 Ninšatapada 2, 5, 6, 342 Nin-uru-da-kuš2 164 Puduhepa 352 Pu-lu-lu 314 puzur4-ma-ma 312 Rīm-Sîn 1–10 Si-du3 275 n. 32 Simat-Ištarān 190 Sîn-šuma-iddina 213 sú-me-ed-dingir 312 Šaninga 315–16 Šara-kam 194, 255 n. 34 Šaratigubisin 347 Šat-Suen 243 Šaza-igizi-Abzu 345 Šeš-kal-la 314 d.aš šir-gi4-al-su 312 Šu-eš4-tár 313

d

Šu-Sîn-walid-dŠulgi 199 n. 47 Šuziana 43 Taddin-Eštar 168 Takunai 167 Tarām-Uram 190 Tikla-ana-Marduk 320 Tu-ra-a 312

U3-dag-ga 235 n. 5 Uqnitum 167 Ur-gar 105 Ur-Lama 194 Ur-li 106 Ur-Lisi 183 Ur-meme 45, 196, 200, 234, 240–423 Ur-mes 196 Ur-nigar 119, 121, 314 Ur-Nanše (dumu ur-li) 105, 106, 107 Ur-Nungal 252 n. 12 Ur-dšu-dsuen-an-dul3 199 Ušumgal 343–45 Zamena 167 Zu-zu 121, 275 n. 32

493