Susa and Elam.

Archaeological, Philological, Historical

and Geographical Perspectives
Mémoires de la Délégation en Perse
Edited by
Michel Tanret
Hermann Gasche
Béatrice André-Salvini
Daryoush Akbarzadeh

VOLUME 58

The titles published in this series are listed at brill.com/mdp

 Susa and Elam.
Archaeological, Philological, Historical
and Geographical Perspectives
Proceedings of the International Congress Held at
Ghent University, December 14–17, 2009

Edited by
Katrien De Graef
Jan Tavernier

LEIDEN • BOSTON
2013
Library of Congress Cataloging-in-Publication Data

Susa and Elam. Archaeological, philological, historical and geographical perspectives : proceedings of the international congress held
at Ghent University, December 14-17, 2009 / edited by Katrien De Graef, Jan Tavernier.
pages cm. – (Mémoires de la Délégation en Perse, ISSN 1782-4168 ; volume 58)
Includes bibliographical references and index.
ISBN 978-90-04-20740-0 (hardback : acid-free paper) – ISBN 978-90-04-20741-7 (e-book) 1. Susa (Extinct
city)–Antiquities–Congresses. 2. Elam–Antiquities–Congresses. 3. Susa (Extinct city)–History–Congresses. 4.
Elam–History–Congresses. 5. Susa (Extinct city)–Languages–Congresses. 6. Elam–Languages–Congresses. 7. Susa (Extinct
city)–Geography–Congresses. 8. Elam–Geography–Congresses. I. Graef, Katrien de. II. Tavernier, J. (Jan)

DS262.S9S87 2012
935'.764–dc23
2012029256

ISSN 1782-4168
ISBN 978-90-04-20740-0 (hardback)
ISBN 978-90-04-20741-7 (e-book)

Copyright 2013 by Koninklijke Brill NV, Leiden, The Netherlands.

Koninklijke Brill NV incorporates the imprints Brill, Global Oriental, Hotei Publishing,
IDC Publishers and Martinus Nijhoff Publishers.

All rights reserved. No part of this publication may be reproduced, translated, stored in
a retrieval system, or transmitted in any form or by any means, electronic, mechanical,
photocopying, recording or otherwise, without prior written permission from the publisher.

Authorization to photocopy items for internal or personal use is granted by Koninklijke Brill NV
provided that the appropriate fees are paid directly to The Copyright Clearance Center,
222 Rosewood Drive, Suite 910, Danvers, MA 01923, USA.
Fees are subject to change.

This book is printed on acid-free paper.

 CONTENTS

List of Abbreviations. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ix
Preface . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . xiii

I
ARCHAEOLOGICAL AND ARCHAEOLOGICAL-HISTORICAL PERSPECTIVES

A New South-Eastern Iranian Glyptic Evidence . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3

Enrico Ascalone
New Evidences on Emergence of Complex Societies in the Central Iranian Plateau . . . . . . . 27
M. Hessari and R. Yousefi Zoshk
The Trans-Tigridian Corridor in the Early Third Millennium bc . . . . . . . . . . . . . . . . . . . . . . . . . . . 43
Steve Renette
Elam and Eshnunna: Historical and Archaeological Interrelations during the Old
Babylonian Period . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 51
Luca Peyronel
Transferts culturels de la Babylonie vers Suse au milieu du e millénaire av. n. ère . . . . . . . . . 71
Hermann Gasche
Elams Kulturkontakte mit seinen Nachbarn im Spiegel der Glyptik des . Jahrtausends v.
Chr. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 83
Georg Neumann
In the Shadow of Kurangun: Cultural Developments in the Highlands between
Khuzestan and Anšan . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 129
Daniel T. Potts
Essai d’application de la méthode de la psychologie environnementale à travers
l’exemple de la ville méso-élamite de Dûr-untaš (Tchoga-zanbil, Iran), site inscrit au
patrimoine mondial de l’Unesco . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 139
Zsolt Gábor Lantos
Some Chronological Aspects of the Building Structures at Haft Tappeh . . . . . . . . . . . . . . . . . . . . 161
Behzad Mofidi-Nasrabadi
Distribution, Materials and Functions of the “Wall Knobs” in the Near Eastern Late
Bronze Age: From South-Western Iran to the Middle Euphrates . . . . . . . . . . . . . . . . . . . . . . . 173
Françelin Tourtet
Re-assessing Elamite Highland Boundaries: New Evidence for the Middle and
Neo-Elamite Periods in the Mamasani Valleys, South-Western Iran . . . . . . . . . . . . . . . . . . . . 191
Bernadette McCall
Braids of Glory. Elamite Sculptural Reliefs from the Highlands: Kūl-e Farah IV . . . . . . . . . . . 207
Javier Álvarez-Mon
vi contents

From Susa to Persepolis: The Pseudo-Sealing of the Persepolis Bronze Plaque . . . . . . . . . . . . . 249
Gian Pietro Basello
Seal Impressions from Susa. Re-evaluating Some of the Findings in Susa Available in the
National Museum of Iran and Introducing Some Unpublished Samples . . . . . . . . . . . . . . . 265
Sedigheh Piran

II
PHILOLOGICAL, PHILOLOGICAL-HISTORICAL
AND HISTORICAL PERSPECTIVES

Ur-Nammâ(k)’s Conquest of Susa . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 285

Gianni Marchesi
Puzur-Inšušinak at Susa: A Pivotal Episode of Early Elamite History Reconsidered . . . . . . . 293
Piotr Steinkeller
Les premiers sukkalmah et les derniers rois de Simaški. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 319
Jean-Jacques Glassner
La «suprématie Élamite» sur les Amorrites. Réexamen, vingt ans après la XXXVIe RAI
() . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 329
Jean-Marie Durand
«Ainsi parle l’empereur» à propos de la correspondance des sukkal-mah . . . . . . . . . . . . . . . . . . 341
Dominique Charpin
Prosopographische Untersuchungen anhand der Rechtsurkunden aus Susa . . . . . . . . . . . . . . . 355
Sheyda Jalilvand Sadafi
Scribal Training in Old Babylonian Susa . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 365
Mehrnoush Malayeri
Abiešuh, Elam and Ashurbanipal: New Evidence from Old Babylonian Sippar . . . . . . . . . . . . 377
Frans van Koppen
Sugirs of Anšan . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 399
Matthew W. Stolper
Approche historique et philologique du titre royal ‘likame/we rišakki’ . . . . . . . . . . . . . . . . . . . . . 417
Stéphanie Anthonioz and Florence Malbran-Labat
Beziehungen zweier Großmächte – Elam und Babylonien in der . Hälfte des . Jt. v.
Chr. Ein Beitrag zur internen Chronologie . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 429
Susanne Paulus
Dynamics of the Fall: Ashurbanipal’s Conquest of Elam . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 451
Peter Dubovský
Elamite and Old Iranian Afterlife Concepts . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 471
Jan Tavernier
 contents vii

III
GEOGRAPHICAL PERSPECTIVES

Geoarchaeological Research in Lower Khuzestan: State of the Art . . . . . . . . . . . . . . . . . . . . . . . . . . 493

Vanessa M.A. Heyvaert, Peter Verkinderen and Jan Walstra

Index of Divine Names . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 535

Index of Personal Names . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 537
Index of Topographical Names. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 542
Index of Text Citations . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 548
 LIST OF ABBREVIATIONS

AA Arts Asiatiques
AB Assyriologische Bibliothek
ABAWPh Abhandlungen der Bayerischen Akademie der Wissenschaften. Philosophisch-histori-
sche Klasse
ABL R.F. Harper, Assyrian and Babylonian Latters belonging to the Kouyunjik collections of the
British Museum, London, –
AcSum Acta Sumerologica
ADFU Ausgrabungen der Deutschen Forschungsgemeinschaft in Uruk-Warka
ADOG Abhandlungen der Deutschen Orient-Gesellschaft
AfO Archiv für Orientforschung
AHw Akkadisches Handwörterbuch
AION Annali dell’Istituto Universitario Orientale
AJA American Journal of Archaeology
AJSLL American Journal of Semitic Languages and Literature
AMI Archäologische Mitteilungen aus Iran
AMIT Archäologische Mitteilungen aus Iran und Turan
AnOr Analecta Orientalia
AOAT Alter Orient und Altes testament
AOS American Oriental Series
ArAs Arts Asiatiques
ArOr Archiv Orientalní
ARM Archives royales de Mari
AS Assyriological Studies
AUCT Andrews University Cuneiform Texts
AUWE Ausgrabungen in Uruk-Warka. Endberichte
BA Beiträge zur Assyriologie und vergleichenden semitischen Sprachwissenschaft
BagM Deutsches Archäologisches Institut. Baghdader Mitteilungen
BaM Deutsches Archäologisches Institut. Baghdader Mitteilungen
BAR-IS British Archaeological Reports. International Series
BBV Berliner Beiträge zur Vor- und Frühgeschichte
BBVO Berliner Beiträge zum Vorderen Orient.
BBVOT Berliner Beiträge zum Vorderen Orient. Texte
BE The Babylonian Expedition of the University of Pennsylvania
BiblAr The Biblical Archaeologist
BiOr Bibliotheca Orientalis
BIWA R. Borger, Beiträge zur Inschriftenwerk Assurbanipals : die Prismenklassen A, B, C = K, D,
E, F, G, H, J und T sowie andere Inschriften, Wiesbaden, .
BM British Museum
BN Beiträge zur Namenforschung
BPOA Bibliotecé del Proximo Oriente Antiguo
BZAR Beihefte zur Zeitschrift für altorientalische und biblische Rechtsgeschichte
CAD Chicago Assyrian Dictionary
CAH Cambridge Ancient History
CANE J. Sasson (ed.), Civilisations of the Ancient Near East, New York, .
CBS Collection of the Babylonian Section (University of Pennsylvania, Philadelphia).
CDAFI Cahiers de délégation archéologique française en Iran
CDOG Colloquien der Deutschen Orient-Gesellschaft
CIRPL E. Sollberger, Corpus des Inscriptions ‘Royales’ Présargoniques de Lagaš, Genève, 
CM Cuneiform Monographs
CPOA Civilisations du Proche-Orient Ancien. Série : Archéologie et environnement
CPOP Civilisations du Proche-Orient Ancien. Série : Philologie
CRAIBL Comptes Rendus des séances de l’Académie des Inscriptions et Belles-Lettres
x list of abbreviations

CT Cuneiform Texts from Babylonian Tablets in the British Museum

CUAS Columbia University Oriental Studies
CUSAS Cornell University Studies in Assyriology and Sumerology
DA Documenta Asiana
DB Dictionnaire de la Bible
DMOA Documenta et Monumenta Orientis Antiqui
EKI F.W. König, Die elamischen Königsinschriften (AfO. Beih. ), Graz, .
ElWb W. Hinz & H. Koch, Elamisches Wörterbuch (AMI. Ergänzungsband ), Berlin, .
EncIr Encyclopedia Iranica
ETCSL Electronic Text Corpus of Sumerian Literature
FAOS Freiburger Altorientalische Studien
FM Florilegium Marianum
GS P. Amiet, Glyptique susienne des origines à l’époque des Perses Achéménides. Cachets,
sceaux-cylindres et empreintes antiques. Découvertes à Suse de  à  (MDP ),
Paris, .
HANES History of the Ancient Near East. Studies
HdO = HdOr Handbuch der Orientalistik
HEO Hautes études orientales. Publications de l’Ecole Pratique des Hautes Etudes. Ive Section.
Sciences historiques et philologiques
HKL R. Borger, Handbuch der Keilschriftliteratur,  vol., Heidelberg, –
HSAO Heidelberger Studien zum alten Orient
IrAnt Iranica Antiqua
ITT Inventaire des Tablettes de Tello
JA Journal Asiatique
JAOS Journal of the American Oriental Society
JBVO Jenaer Beiträge zum Vorderen Orient Bd. 
JCS Journal of Cuneiform Studies
JCS/S Journal of Cuneiform Studies. Supplement Series
JEOL Jaarbericht van het Vooraziatisch-Egyptisch Genootschap Ex Oriente Lux
JESHO Journal of the Economic and Social History of the Orient
JNES Journal of Near Eastern Studies
JRGS Journal of the Royal Geographic Society
LAPO Littératures anciennes du Proche-Orient
LKA E. Ebeling, Literarische Keilschrifttexte aus Assur, Berlin
MAD Materials for the Assyrian Dictionary
MARI Mari. Annales de Recherches Interdisicplinaires
MC Mesopotamian Civilisations
MDAI Mémoires de la délégation archéologique en Iran
MDP Mémoires de la délégation en Perse
MHEM Mesopotamian History and Environment Series. Memoirs
MHEOP Mesopotamian History and Environment Series. Occasional Publications
MHET Mesopotamian History and Environment Series. Texts
MMAI Mémoires de la mission archéologique en Iran
MSKH Materials and Study for Kassite History
MSL Materialen zum Sumerischen Lexikon
MVN Materiali per il vocabolario neosumerico
MVSt Münchener Vorderasiatische Studien
NABU Nouvelles Assyriologiques Brèves et Utilitaires
NAPR Northern Akkad Project Reports
NESt University of California Publications. Near Eastern Studies
OBO Orbis Biblicus et Orientalis
OECT Oxford Editions of Cuneiform Texts
OIC Oriental Institute Communications
OIP Oriental Institute Publications
OLA Orientalia Lovaniensia Analecta
OLZ Orientalistische Literaturzeitung
OPSKA Occasional publications of the Samuel Noah Kramer Fund
 list of abbreviations xi

Or Orientalia
OrA Orient-Archäologie
OrAn Oriens Antiquus
OrEx Orient-Express. Notes et nouvelles d’archéologie orientale, Paris.
PBS University of Pennsylvania. The University Museum. Publications of the Babylonian
Section
PDT Die Puzriš-Dagan Texte der Istanbuler archäologischen Museen
PF Persepolis Fortification Tablets (published by R.T. Hallock, Persepolis Fortification Tablets
(OIP ), Chicago.
PF-NN Unpublished Persepolis Fortification texts cited from draft editions by Richard T. Hallock
and collated by Wouter F.M. Henkelman
PIHANS Publications de l’institut historique et archéologique néerlandais de Stamboul
RA Revue d'Assyriologie et d'Archéologie Orientale
RAcc F. Thureau-Dangin, Rituels accadiens, Paris, 
RGTC Répertoire Géographique des Textes Cunéiformes
RIMA Royal Inscriptions of Mesopotamia. Assyria
RIMB Royal Inscriptions of Mesopotamia. Babylonia
RIME Royal Inscriptions of Mesopotamia. Early Periods
RlA Reallexikon der Assyriologie und vorderasiatischen Archäologie
RTC F. Thureau-Dangin, Recueil de tablettes chaldéennes, Paris, 
SAAS State Archives of Assyria. Studies
SANTAG SANTAG. Arbeiten und Untersuchungen zur Keilschriftkunde
SAOC Studies in Ancient Oriental Civilization
SAT Sumerian Archival Texts
SCCNH Studies on the Civilisation and the Culture of Nuzi and the Hurrians
SM Studia Mediterranea
StIr Studia Iranica
StOr Studia Orientalia
TAVO Tübinger Atlas des Vorderen Orients
TCL Musée du Louvre. Département des Antiquités Orientales. Textes cunéiformes
TCS Texts from Cuneiform Sources
TD Tell ed-Der
TTM I M.W Stolper, Texts from Tall-i Malyan I: Elamite Administrative Texts (–)
(Occasional Publications of the Babylonian Fund ), Philadelphia.
UAVA Untersuchungen zur Assyriologie und Vorderasiatischen Archäologie
UCP-NES University of California Publications. Near Eastern Studies
UE Ur Excavations
UET Ur Excavations. Texts
UF Ugarit-Forschungen
UMM University Museum Monograph
VB Vorderasiatische Bibliothek
VS Vorderasiatische Schriftdenkmaler der Königlichen Museen zu Berlin
WO Die Welt des Orients
WVDOG Wissenschaftliche Veröffentlichungen der Deutschen Orient-Gesellschaft
YOS Yale Oriental Studies
ZA Zeitschrift für Assyriologie und Vorderasiatische Archäologie
ZAR Zeitschrift für altorientalische und biblische Rechtsgeschichte
 PREFACE

This proceedings volume contains twenty-seven papers that were read at the international
congress “Susa and Elam. Archaeological, Philological, Historical and Geographical Perspec-
tives” held at Ghent University, December th–th .
The aim of this congress was to investigate—exactly  years after the th Rencontre Assyri-
ologique Internationale entitled “Mésopotamie et Elam” held at Ghent University—the present
state of affairs on our knowledge of the Elamite and Susian society from an archaeological,
philological, historical and geographical point of view.
The topic of this first “Susa and Elam Congress” was a multidisciplinary approach of the city
of Susa in the broader context of the Elamite region. Susa is one of the oldest cities of the Ancient
Near East and has known an almost uninterrupted occupation from the th millennium bc till
the Islamic period. Being one of the most important centres in Khuzestan, the South-Western
Iranian region east of Mesopotamia, Susa has been the capital of several Elamite kingdoms
and dynasties. Moreover, during the greatest part of its history, Susa has maintained variable
relations with the consecutive Mesopotamian empires and has eventually become one of the
seats of the great Persian Empire of the Achaemenids, who ruled the whole Ancient Near East.
The primary goal of this congress was to generate a multifaceted picture of the history
of Susa and Elam, based on archaeological, philological, historical and geographical research
results. The interdisciplinary character of the congress made a substantial contribution to the
present research into the fundamental landscape, socio-economic, historical and philological
developments in the Suso-Elamite region. Moreover, the congress wanted to give a new impulse
to these studies by assembling all specialists concerned, established and young scholars alike.
The idea is to organise a “Susa and Elam Congress” quinquennally.
This congress has been made possible thanks to funding by the IAP project “Greater Mesopo-
tamia. Reconstruction of its Environment and History” headed by Michel Tanret (Ghent
University), the Funds for Scientific Research Flanders and the Humanities Faculty of Ghent
University. I wish to thank each of these for their support.
Moreover, I wish to thank all participants and chairs for making this congress a great success
as well as the students, assistants and colleagues of the Department of Languages and Cultures
of the Ancient Near East of Ghent University for the organizational support.
Last but not least, I thank Jan Tavernier (Université Catholique de Louvain) for going along
with me in this adventure, for his support and for co-editing this volume.

Katrien De Graef
Ghent
 I

ARCHAEOLOGICAL AND ARCHAEOLOGICAL-HISTORICAL PERSPECTIVES

 A NEW SOUTH-EASTERN IRANIAN GLYPTIC EVIDENCE

Enrico Ascalone*

. Geographical and Chronological Notes

The excavations carried out during the second half of the Sixties and in the Seventies in
Iran have allowed a new historical and cultural approach to the south-eastern regions of
the Plateau. The explosion of fieldwork in the Kerman province, Bampur district and in
the Hilmand valley permitted us to know a more complex cultural horizon of the Iranian
eastern area. In the first studies on the new eastern Iranian excavations, the cultural com-
plex of Kerman and Lut province, as known from Tepe Yahya and Shahdad archaeologi-
cal evidences, was considered ‘intercultural’ by Philip Kohl to underline the civilizing inter-
ferences on the chlorite/steatite production (Kohl a; b; ; ; ; ).
According to Pierre Amiet’s evaluations on the cultural horizons of the region bordering
to the west with the historical Elam and to the east with the Lut desert and the Makran
coast, we believe more correctly to identify in the cultural production of Markhashi lands
(Steinkeller  contra Van Dijk ) a local, original and unbroken development (Amiet
; ).1 However the knowledge of a new eastern Iranian civilization, defined Trans-
Élamite by Amiet (: –), is not complete and it seems not to be so simple to recon-
struct its chronological limits and its geographical diffusion. If we could consider probable
to identify in the Kerman region, in the Halil valley, the cultural origin of a different artis-
tic thought (as attested in Shahdad, Tepe Yahya and in later Konar Sandal excavations), it is
not yet possible to comprehend the real geographical diffusion of south-eastern Iranian cul-
ture.
The new evidences collected in the Halil valley (Madjidzadeh a) and the excavations
carried out at Konar Sandal South and North (Madjidzadeh b; c; ; ) are
crucial to distinguish an autonomous regional context probably connected to the corpus of
seals known in Tepe Yahya and Shahdad excavations and from antiquary market (in particular
Porada ; ; ; Amiet ; ; ).
The archaeological and stylistic approach to the stratigraphical evidence of Tepe Yahya
have shown different and not exhaustive interpretations principally on IVB period; in the
first publications Yahya IVB period was attributed to the first half of the third millennium
(Lamberg-Karlovsky ; ; ; ; Lamberg-Karlovsky—Kohl ) and only sub-
sequently identified in later chronological framework (Beale ; Amiet ; Potts ).
The evidence for the continuation of the so-called ‘Intercultural Style’ in Akkadian times
(Ascalone : tab. ) in Mesopotamia and Iran (Miroschedji : ) and the evidence of
the glyptic art summarized by Amiet (), support a later date for IVB period.2 The glyptic

* University of Palermo.
1 Concerning the textual Mesopotamian evidences on Markhashi/Parakhshum see in particular Steinkeller ;
; Vallat : ; Heimpel : ; Horowitz : ; T.F. Potts : , n. .
2 D.T. Potts takes into consideration an Akkadian-type bull man representation, an alabaster unguent, square

based jar of a type common in Bactria, Iran and Gulf, a piece of incised grey ware (associated to Shahr-i Sokhta IV),
 enrico ascalone

Fig. a: South-eastern Iranian cylinder seal from Shahdad (after Hakemi : )

collected from Tepe Yahya shows that there is no continuity of occupation between period
IVC and IVB; the glyptic associated with period IVB can not to be earlier than Old Akkadian
period.
However on the base of last discoveries in the Halil Valley is possible to recognize three
different phases of production in the south-eastern Iranian corpus; the seals collected from
antiquary market and excavations projects (Shahdad, Tepe Yahya and Konar Sandal in the
regional context of production, Tôd, Gonur depe and Susa in the secondary contexts) allow
to identify at least three main stylistic phases of making:
1. Konar Sandal production (Trench III): ca. /–bc.
2. South-eastern Iranian production, classic group (= Konar Sandal Trench V, Yahya IVB;
Takab III): ca. –/bc.
3. South-eastern Iranian production, later group (= Yahya IVA; Takab III): ca. /–
/bc.
The earlier Konar Sandal sealings from Trench III seem to be very close to the ED III Mesopota-
mian ateliers (Madjidzadeh ), while the so-called ‘trans-elamite’ seals have to be considered
a different and very original production circumscribed to the last quarter of III millennium
bc with internal subdivisions. The identification of a later group in the south-eastern Iranian
corpus has been permitted by iconographical and stylistic changing in the figurative patrimony
of seals and for their significant archaeological contexts, in particular for seals coming from
Gonur  north (last century of III millennium bc), Tôd, in the Treasure of Amenemhat II
(ca. –bc), and Tepe Yahya IVA (ca. –bc).3

a Persian Gulf stamp seal and sherds of truncated pots similar to those from numerous sites in Bactria, Margiana and
Baluchistan for dating around – bc Yahya IVB period (Potts : –). In spite of Potts hypothesis,
P.C. Kohl placed Yahya IVB to the third quarter of Third millennium bc (around –bc) on the base of
chlorite vessels documentation and the new radiocarbon evidence (Kohl : ); in this way it seems possible to
conciliate the chlorite vessels Mesopotamian contexts, the later diffusion of ‘Intercultural Style’ in the Akkadian, Post-
Akkadian periods and the iconographical and stylistic cylinder and stamp seals production. Lamberg-Karlovsky’s
theory is again a lowering of dating of Yahya IVB period and represents a third different approach to the not clear
stratigraphical sequences of Yahya (Lamberg-Karlovsky : –).
3 For a more detailed analysis of south-eastern Iranian corpus see Ascalone : –.
 a new south-eastern iranian glyptic evidence 

Fig. b: South-eastern Iranian cylinder seal from Shahdad (after Hakemi : )

. South-Eastern Iranian Cylinder Seals

The south-eastern Iranian glyptic documentation collected from Tepe Yahya and Shahdad
shows a differential use of seals;4 stamp and cylinder morphologies are both well-attested in the
Kerman area (cf. with the specimens published in Amiet : –, fig. ; : , fig. ;
Hakemi : , figs. Ia. –; Lamberg-Karlovsky : , fig. C, E; : pl. XXVI: b;
Potts : , , fig. A; Pittman : nn. –). Although the stamp seals seem to be
a clear expression of the local culture, a dozen of cylinder seals were found during the Shahdad
and Yahya excavations. All excavated Shahdad specimens, except one from the surface,5 are
dated to the second half of the III millennium bc (using a periodical regional chronology
from Takab IV. to Takab III.). The five excavated cylinder seals from Shahdad were found
in the ‘Main Cemetery’, from the graveyards located in the area A (Hakemi ). A deeper
chronological evaluation, based on the iconographical comparisons, could allow us to date the
majority of Shahdad cylinder seals6 to the last quarter of III millennium bc (Fig. a–e).

4 On the south-eastern Iranian seals and their chronological, iconographical and more general discussion see

principally Porada  and Amiet : ; ; : –, fig. ; ; .
5 The alabaster seal from surface is height ,cm with a diameter of ,cm (Fig. d; Salvatori—Vidale : fig. :

; Hakemi : : Xg. a-b; Winkelmann : fig. ; Ascalone b: fig. c). The seal could be dated to a later
period.
6 The first marble cylinder seal (F./) from Shahdad was found in the ‘Main Cemetery’ (Area A), in Grave

 (,× , cm), from—, cm below the surface: its measures are , ×,cm (Fig. a; Hakemi : , :
Ib. ; Amiet : ; : –, fig. ; : , fig. : ; : fig. ; Dyson—Harris : n. ; Collon
: , n. ; Winkelmann : fig. d; : fig. : ; Ascalone a: fig. ; b: fug. a; : fig. c);
the second one specimen in chalky stone (F./, now placed in NMI ) measures ,×,cm and it comes
from Grave  (‘Main Cemetery’, Area A) (Fig. b; Hakemi : , : Ib. ; Amiet : ; : , fig. ;
: , fig. : ; Dyson—Harris : n. ; Winkelmann : fig. e; : fig. : ; Ascalone a: fig. ;
b: fig. b; : fig.  f.); an alabaster cylinder seal (F./ and NMI ) was found in funerary contexts
(‘Main Cemetery’, Grave ) and measures ,×,cm (Fig. e; Hakemi : , : Ib. ; Dyson—Harris :
n. ; Ascalone a: fig. ; b: fig. e; : fig. e); the fourth seal in limestone, now placed in NMI ,
was collected from Grave  (‘Main Cemetery’, Area A) and measures ,×,cm (Fig. c; Hakemi : , Obj.
n. ; Ascalone a: fig. ; b: fig. c; : fig. b).
 enrico ascalone

Fig. c: South-eastern Iranian cylinder seal from Shahdad (NMI )

Fig. d: South-eastern Iranian cylinder seal from

Shahdad (after Salvatore—Vidale : Fig. : )

Fig. e: South-eastern Iranian cylinder seal from Shahdad (NMI )
 a new south-eastern iranian glyptic evidence 

Fig. a: South-eastern Iranian cylinder seal from Tepe Yahya (NMI )

Fig. b: South-eastern Iranian cylinder seal from Tepe Yahya (NMI )

The Yahya evidence regarding the cylinder seals presence is linked to the IVB period where
six on seven ‘trans-elamite’ seals were found (Figs. a–f);7 all the specimens from Yahya are in
chlorite/steatite.8

7 TY  (from Area B-BW and now placed in NMI ) measures ,×,cm (Fig. d; Lamberg-Karlovsky

: fig. B; : pl. XXVI: c; Dyson—Harris : , n. ; Amiet : fig. : ; Winkelmann : fig. :
; Pittman : n. ; Ascalone b: fig. d); TY  was found in BW TT– in the north step trench (Second
Building Level) and its measures are ,× , cm (Fig. g; Lamberg-Karlovsky : , fig. ; : pl. XXXI: c;
Lamberg-Karlovsky—Tosi : fig. d; Amiet : fig. : ; Winkelmann : fig. : ; Pittman : n. ;
Ascalone b: fig. g; : fig. d); TY  is coming from the floor of the ‘Persian Gulf Room’ (Area B-BW...;
now in NMI ): its measures are , × , cm (Fig. b; Lamberg-Karlovsky : , pl. VI e fig. ; Lamberg-
Karlovsky—Tosi : fig. ; Amiet : ; : , fig. ; : fig. : ; : fig. ; Potts : –,
fig. ; Dyson—Harris : n. ; Winkelmann : f.; : fig. : ; Pittman : n. ; Ascalone b:
fig. b); TY  (NMI ) is , cm in height and , of diameter (Fig. a; Lamberg-Karlovsky : pl. XXVI: c;
Lamberg-Karlovsky—Tosi : , fig. c; Potts : , fig. ; Amiet : fig. : ; : fig. ; Collon :
, n. ; Winkelmann : fig. b; : fig. : ; Pittman : n. ; Ascalone b: fig. a); the seal from Baft
(near Tepe Yahya, now in NMI ) measures ,×,cm (Fig. f; Lamberg-Karlovsky : pl. ; Pittman :
n. ; Ascalone b: fig.  f.); TY , coming from BW.. (NMI ), is ,×,cm (Fig. c; Lamberg-Karlovsky
: pl. ; Amiet : fig. : ; : , fig. ; Winkelmann : fig. ; Pittman : n. ; Ascalone b:
fig. c; : fig. c); the last one cylinder seal from Yahya (Fig. e), TY , is coming from B.. (NMI )
(Lamberg-Karlovsky : fig. b; Pittman : n. ; Ascalone b: fig. e).
8 At the same chronological range could be associated the Yahya cylinder seal representing a human being in

front of a palm; strong affinities are known in a cylinder seal from private collection (cf. with Fig. a) where a deity is
sitting on his intertwined legs in front of a palm (in primis Amiet : fig. : ); the style and the iconographical
representation of the horned god have resemblance with the seals showing the twin horned deities without wings,
previously dated to the Akkadian period.
 enrico ascalone

Fig. c: South-eastern Iranian cylinder seal from Tepe Yahya (NMI )

Fig. d: South-eastern Iranian cylinder seal from Tepe Yahya (NMI )

Fig. e: South-eastern Iranian cylinder seal from Tepe Yahya (NMI )
 a new south-eastern iranian glyptic evidence 

Fig. f: South-eastern Iranian cylinder seal from Tepe Yahya (NMI )

The first attempts of identifying a development line of so-called ‘trans-elamite’ glyptic art were
born on the base of the primary glyptic documentation collected in Shahdad and Yahya and
on the analysis of the iconographical evidence placed on the surface of the engraved chlorite
vessels; some seals from private collection or without an archaeological context were related to
the south-eastern region overworking a new point of view on the south-eastern Iranian cul-
tural complex (Porada : –, figs. –; ; ; ; Amiet : ; ; :
–, fig. ; ; ; Winkelmann ; Ascalone a; ). The archaeological
evidences from Jiroft area have opened a new field of research and they have allowed a new inter-
pretation on the previously published seals;9 as far as we know, we can identify south-eastern
Iranian origin in the following seals coming from private collection or outside south-
eastern Iranian cultural complex:10
– from Louvre Collection (Amiet : fig. : ; Winkelmann : fig. : ; Ascalone
–: fig. c; b: fig. b; : fig. m) (Fig. a)
– from Louvre Collection (Amiet : fig. : ; Winkelmann : fig. : ; Ascalone
–: fig. b; b: fig. a) (Fig. b)
– from Susa (Legrain : n. ; Amiet : , fig. ; b: fig. ; Winkelmann :
fig. g; : fig. ; Ascalone –: fig. a; a: fig. a; b: fig. c; :
fig. g)11 (Fig. c)
– from Susa (Delaporte : , Pl. : –; Porada : , fig. ; Amiet : n. ;
: , , , fig. ; Winkelmann : tav. : h; : fig. ; Ascalone b:
fig. c) (Fig. n)
– from Susa (Ascalone –: fig. c; a: fig. c–d; b: fig. ) (Fig. l)

9 The unique south-eastern Iranian cylinder seals found outside Iranian sites are known at Gonur depe and

Tôd (Egypt): in Gonur seal the ‘vegetation deity’ is depicted on a snake moving from left to right (Sarianidi :
fig. : ) (Fig. k). A Susa seal, dating the end of Early Dynastic III period, seems to be strongly linked to the
iconographical and religious aspects of the later south-eastern Iranian experiences (Fig. n; cf. the attitude of the
deities with Shahdad cylinder seal published in Amiet : ; : , fig. ; : , fig. : ; Dyson—
Harris : n. ; Winkelmann : fig. e; : fig. : ; Hakemi : , : Ib. ).
10 For south-eastern Iranian stamp seals see the specimens coming from Tepe Yahya (Lamberg-Karlovsky :

fig. E), Shahdad (Hakemi a: , figs. Ia. , ), Tepe Giyan (Contenau—Ghirshman : tav. : ), and
Konar Sandal (Madjidzadeh : figs. a, g, h and j).
11 For P. Amiet the seal has to be dated to the Puzur-Inshushinak time (Amiet : ).
 enrico ascalone

– from Tôd (Bisson de la Roque : n. ; a; Landsberger : –; Porada
: , fig. ; Amiet : fig. : ; : fig. ; Winkelmann : fig. : ;
Ascalone –: fig. d; b: fig. d; : fig. e) (Fig. d)
– from Bailey Collection (Amiet : , , , fig. : ; : fig. ; : , fig. ;
Duchesne-Guillemin : , fig. ; Winkelmann : fig. a; : fig. : ; Ascalone
–: fig. e; b: fig. g; : fig. i; Pittman : fig. .) (Fig. e)
– from Foroughi Collection (Porada : , fig. ; : fig. ; : pl. IV; : pl. ;
Amiet : , fig. ; : , fig. ; : , fig. : ; : fig. ; : fig. ;
: fig. ; Harper—Aruz—Tallon : fig. ; Winkelmann : fig. j; : fig. :
; Ascalone b: fig. h; : fig. d) (Fig. f)
– from Rosen Collection (Porada : pls. I–III; : –,  e pl. ) (Fig. m)
– from Rosen Collection (Porada : ,  e pl. ; Amiet : fig. ; Winkelmann :
fig. c; : fig. : ; Ascalone –: fig.  f.; b: fig. e; : fig. h) (Fig. g)
– from Ligabue Collection (Winkelmann : fig. a–c; Ascalone b: fig.  f.; :
fig. d) (Fig. h)
– from Jalalabad (Ascalone –: figg. , h; b: fig. a–b; : fig. ) (Fig. j)
– from Gonur depe (Sarianidi : fig. : ; Salvatori : fig. : ; Amiet : , fig. ;
Ascalone –: fig. g; b: fig. ; : fig. g) (Fig. k)
– from Gonur depe (Sarianidi : , fig. ; Amiet : fig. ) (Fig. i)

Fig. a: South-eastern Iranian cylinder seal from

Louvre collection (after Amiet : Fig. : )

Fig. b: South-eastern Iranian cylinder seal from

Louvre collection (after Amiet : Fig. : )
 a new south-eastern iranian glyptic evidence 

Fig. c: South-eastern Iranian cylinder seal from Susa (after Legrain : n. )

Fig. d: South-eastern Iranian cylinder seal from Tôd (after Amiet : Fig. : )

Fig. e: South-eastern Iranian cylinder seal from

Bailey collection (after Amiet : Fig. : )
 enrico ascalone

Fig. f.: South-eastern Iranian cylinder seal from

Foroughi collection (after Amiet : Fig. : )

Fig. g: South-eastern Iranian cylinder seal from Rosen collection (after Porada : pl. )

Fig. h: South-eastern Iranian cylinder seal from

Ligabue collection (after Winkelmann : Fig. a)
 a new south-eastern iranian glyptic evidence 

Fig. i: South-eastern Iranian cylinder seal from Gonur depe (after Amiet : Fig. )

Fig. j: South-eastern Iranian cylinder seal from Jalalabad (NMI )

Fig. k: South-eastern Iranian cylinder seal from Gonur depe (after Salvatori : Fig. : )
 enrico ascalone

Fig. l: South-eastern Iranian cylinder seal from Susa (NMI /)

Fig. m: South-eastern Iranian cylinder seal from Rosen collection (Porada : pl. I)

Fig. n: South-eastern Iranian cylinder seal from

Susa (after Delaporte :  and Pl. : –)
 a new south-eastern iranian glyptic evidence 

Fig. : South-eastern Iranian cylinder seal from Jalalabad (NMI )

. Jalalabad Cylinder Seal

A new evidence of south-eastern Iranian glyptic art is a limestone seal now placed in the Bastan
Archaeological National Museum of Tehran (NMI ) (Fig. ).12 The measures of the cylinder
seal are ,cm in height and ,cm of diameter; the specimen was found at Jalalabad in the Fars
region but it has not a certain archaeological context; however the iconographical representa-
tion and the stylistic details seem to be linked to the south-eastern cultural traditions.13 The
seal represents, in the upper part, a figure with her head in profile and body facing frontward
with a pair of large, curving wings; breast are indicated by a pair of small circles on triangu-
larized torso; a long, rectangular lozenge, decorated with chevrons, is situated just below the
torso and may represent the clothed legs of the figure, as if it was sitting cross-legged. The hair
appears to be long and flowing without a crown as known in the other south-eastern Iranian
seals. The presumably female divinity is flanked by two figures knelled down represented in
profile; both seem to offer something to the winged personage. The left figure seems to have
a spherical element in the hands, while the right figure is depicted with his/her arm reached
to offer an incomprehensible vegetable element or a star placed on a vertical support (cf. with
Figs. a and h). Only one figure, on the left, has the typical hairstyle well-attested in other seals
from Tepe Yahya, Shahdad and private collections. In the lower part of seal is depicted a row
of quadrupeds, one of them with wings probably representing a fantastic creature or dragon as
known on the cylinder seal of Rosen collection (for bibliography see above).

12 I am especially grateful to Dr. Mohamad Reza Kargar and Mohammed Reza Mehrandish (Directors of the
National Museum of Iran) who entrusted the study of the III millennium cylinder seals of the Museum to me (now all
collected in my Ph.D. thesis). I would like express my thanks to Dr. Ciaici (vice-director) and Dr. Jaffar Mohammadi
(Director of the Treasure of the Bastan Museum) for their help and logistic assistance; I am very grateful to Mrs
Askari (Director of the Department of Seal and Coin) and his staff (in particular Mrs Akram Rezaee, Roghaiyeh
Chenary, Zainab Gavidol and Neda Amighi) for their kind availability and patience.
13 A seal with a south-eastern Iranian iconographical elements, pseudo-harappan inscription and Central Asia

morphology has been found on the surface of the same site of Jalalabad (site number ) in Paul Gotch’s survey
carried out in the Persepolis plain (Ascalone ).
 enrico ascalone

The closest comparisons of our seal are with the specimens discovered at Tepe Yahya; the
seal TY  (Fig. b) was found on the floor of the ‘Persian Gulf Room’ in B-BW area: the seal
represents two similar but distinctive figures with clothed legs; the deity with open wings, to the
left, has strong affinities with the personage represented in the Jalalabad cylinder seal. Although
the winged female deity of Yahya seal is depicted with a horned crown on the head (absent in
our seal), we suggest to see in both winged goddesses the same religious and/or mythological
origin.14
The female deity with long wings is attested also in another cylinder seals found in IVB
archaeological context of Tepe Yahya (TY ) (Fig. a), the phase to which the ‘Persian Gulf ’
room has been dated; the strongest affinities with our seal are connected to the iconographical
composition where the winged deity is represented, in the central part, and two figures knelled
down to the her flanks. Although the Yahya cylinder seal shows some iconographical differences
(like the presence of new figurative elements or the winged deity with horned crown sitting on
a throne), the central thematic motif is the same with the winged deity flanked by two female
personages.
A third cylinder seal from Tepe Yahya (IVB period) appears more difficult to connect to
Tehran Bastan Museum specimen (Fig. c); the winged figure is standing with horned crown
and without the bosoms. Vegetable elements seems to grow up from the ground and to be
connected to the winged personage. For this third cylinder seals we could only try to suggest a
possible link with our winged female deity.
The evidence of a seal from Shahdad could be very interesting because it could help us to
identify a new iconographical development of the winged deity as known in the previously
cylinder seals (Fig. e): the Shahdad seal (F./, excavation number ), coming from the
‘Main Cemetery’, Area A, Graveyard , was assigned to Takab III.-III. periods; although
the stylistic approach to the iconographical theme is changed, we could propose that the figure
with the large open wings flanked by two sitting female personages has strong affinities with
the previously analysed cylinder seals and with Jalalabad specimen. In the seal collected from
Shahdad we could see a new iconographic and stylistic version, dated to the same range, of the
same mythological and religious patrimony.
Summing up the iconographical comparisons with Yahya and Shahdad stratigraphied cylin-
der seals, their archaeological contexts and the stylistic evaluations could allow us to date the
Jalalabad seal to the XXIII century bc. The new evidence of a cylinder seal related to the so-
called ‘trans-elamite’ cultural complex allowed us to underline the chronological problems
connected to the south-eastern Iran. We suggest that the seal conserved in the Iran Bastan
Museum of Tehran should be dated, following a Mesopotamian terminology, to the Akkadian
period and linked to the iconographical patrimony known in the south-eastern Iranian cylin-
der seals corpus discovered in the Kerman region. The Mesopotamian evidence of a winged
divinity with horns (well-identified like Ishtar), attested for the first time, during the Akkadian
period could confirm our chronological proposal (Boehmer : figs. , , ; Amiet
: ; : ).15
It appears more difficult to comprehend the real significance and function of the winged
deity; the absence of a more deep knowledge concerning the Elamite or so-called ‘Trans-
elamite’ mythology doesn’t allow us to known the role played by the winged female divinity

14 Although the pair of god has been considered as the (trans-)Elamite equivalent of Mesopotamian Inanna and

Dumuzi or Elamite Pinikar and Khumban (Potts : –), according to P. Amiet (: ; : –;
: ) we believe in two distinct female deities or in different aspects of a single divine entity.
15 The (standing) winged goddess is attested in Bactria and Margiana as an unhorned woman surrounded by

animals; it seems probable that some peculiar iconographical aspects from south-eastern Iran are known in later
Bactria-Margiana glyptic art (Sarianidi : figs. .–, , –, .–; Winkelmann : –, fig. ).
a new south-eastern iranian glyptic evidence 

Fig. a: clay statue from Shahdad (after Hakemi : )
 enrico ascalone

Fig. b: clay statue from Shahdad (after Hakemi : )
 a new south-eastern iranian glyptic evidence 

Fig. c: standard of Shahdad (after Hakemi : )

in the Iranian pantheon. We could try to identify the female figure with one of the major
Iranian deity connecting her wings with the cosmic aspect of the divinity. The two personages
knelled down to the divinity flanks could be interpreted like praying. The winged figure and
the praying positions are not the same, probably to underline the specific significance of the
representation; the same attention used to depict the faithful in profile is due to necessity to
show the effective prayer act. The same position is attested in Yahya and Shahdad seals (Figs. e
and a), in Foroughi private collection (Fig. f), in Bailey (Fig. e) and Rosen seals (Fig. g).
This iconographical representation has strong links with the sculpture art of Shahdad where
some clay statuettes was found during  and  Hakemi excavations (Hakemi : Obj.
nn. , G.a, , G.b, , G., , G., , G., , G., ,
G., , G., , G., , G., , G.; see also Amiet : , fig. ; :
fig. ) (Figs. a–b).
On the contrary, figures with clothed legs, as known in our seal, appear to be a specific
representations of the eastern regions of Iran. Many evidences are known in the south-eastern
Iranian glyptic art during the III millennium bc and it represents a peculiar iconographical
evidence of the region. This iconography is well-known in Shahdad seals (Figs. a–c), in
Bailey collection (Fig. e), in the discussed seals from Yahya (Figs. b, g), in Louvre collection
(Figs. a–b), in Gonur specimen (Fig. k) and in the Tôd treasure (Fig. d).16 The evidence
seems to be also strongly linked to the so-called Shahdad standard (Hakemi : , n. ;
Amiet : , fig. ; : , , , fig. ) (Fig. c), to a silver Bactrian vase (Amiet
: , fig. ) (Fig. d) and to others relief in bronze and lapis from the Kerman province
(Amiet : , fig. ; : , figs. –) (Figs. e–f).

16 Although E. Porada suggested a possible iconographical links between south-eastern Iranian Tôd cylinder seal

and the protoelamite experiences (Porada : ), we believe possible to consider it as a later south-eastern
Iranian specimen related to the trans-elamite culture and probably dated to the end of III/beginning II millennium
bc (Amiet : ). For the archaeological contexts (Mont temple) see principally Kantor : –,  and for
strong iconographical comparisons see Untash Napirisha’s stele, published in P. Amiet (: , , figs. ,
), and Gudea’s basin (Sarzec—Heuzey –: pl. , fig. ). Finally P. Amiet has believed in a double role
carried out by the four standing figures connected to the water (fertility) and to the mountain aspects as known,
however, only from Mesopotamian Cassite period (Amiet : ).
 enrico ascalone

Fig. d: silver vessel from Bactria (after Amiet : Fig. )

Fig. e: lapis disk from Jiroft (after Madjidzadeh a: )
 a new south-eastern iranian glyptic evidence 

Fig. f.: bronze pin from Shahdad (after Amiet : Fig. )

. Conclusion

South-eastern Iranian complex could be used as a geographical and cultural terminology

without however certain chronological limits; our opinion is that the cultural assemblage, as
seen in the glyptic art, is attested in Yahya IVB–IVA, Shahdad III.– periods (as well-known
from Hakemi excavations) and Konar Sandal, Trench V. The ware comparisons on the Iranian
plateau permit us to link the Kerman region (from the third quarter to the end of the III
millennium bc) to Shahr-i Sokhta III–IV, early Kaftari period in the Marv Dasht valley and
Bampur IV–VI (corresponding to Susa IVB and V). In this chronological time-span a new
artistic experience was born in south-eastern Iran, developing an autonomous and local art, in
some case connected to the Elam entity and to the Bactrian region. The glyptic art, the earlier
steatite carved vessels (starting from Mesopotamian ED II period) and the metallurgical work
seem to be the firsts indicators to start every historical and cultural evaluations on the south-
eastern Iranian region. The beginning of ‘Intercultural Style’ appear to be earlier than the first
evidence of a specific glyptic art presumably linked to the second half of third millennium bc,
when the decorated chlorite vessels will be largely common in Yahya IVB period (Ascalone
; ). The glyptic evidence confirms a lower chronology of Yahya IVB allowing us to
believe in a stratigraphical gap between Yahya IVC (with Protoelamite seals and sealing) and
IVB periods (Potts ; contra Lamberg-Karlovsky ).
The glyptic evidence collected and associated to the south-eastern Iranian cultural complex
is not yet exhaustive and complete. The mythological corpus seems to have some important
difference with the Mesopotamian iconographies; the mythological fighters (well-known in
Mesopotamia) seem to be absent in the Iranian corpus that appears more interested in describ-
ing the cosmic organization within Iranian pantheon where are frequently depicted the major
female deities well-known in their cosmic, fertility (connected also probably to the snakes, see
 enrico ascalone

for example the specimen of Gonur depe, Fig. k)17 and stellar aspects (cf. with Figs. a, h
and ). The seals from Yahya and Shahdad were locally produced showing certain stylistic
features; the divine attitude is marked by one or more attributes, including horned headdress,
wings, grain sprouting from the body and stellar elements. The south-eastern Iranian glyptic art
seems to be circumscribed to the second half of the third millennium with only some sporadic
iconographical evidences directly influenced by the Akkadian engravers. Later evidences from
Bactrian glyptic art (bronze stamp seals) show a hypothetical iconographical links between
the south-eastern Iranian experience and Bactria-Margiana archaeological complex (while the
possible links with the previous protoelamite glyptic art are actually understandable; contra
Winkelmann ). This sporadic comparisons confirm an autonomous development of the
glyptic art of south-eastern Iran during the second half of the III millennium bc; an indigenous
glyptic art probably really mature only in the phases of Yahya IVB and Takab III.. Only with the
last quarter of III millennium bc it seems that the south-eastern complex developed an original
traits in their seal productions, now deeply far from the previously assemblage very close to
the ED III Mesopotamian ateliers. On the base of documentation collected from south-eastern
Iranian sites is possible to identify  macro-phases of cultural development:
1. Yahya IVC/Takab IV. (ca. –/ bc): protoelamite cultural diffusion (tablets
and seals) as known at Tepe Yahya and Shahdad; growth and diffusion of Halil valley sites;
relations with the western regions.
2. Takab IV./III. (ca. /–/ bc): crisis of Yahya settlement, first necropolis
of Shahdad and development of Halil valley centers. First early chlorite vessels production,
sealings and seals from Konar Sandal strongly linked to the ED III production (Trench III).
Wide relations with Mesopotamian reigns.
3. Yahya IVB/Takab III. (ca. /–/bc): new occupation of Tepe Yahya and
strong development of Halil civilizations (Barakhshum/Markhashi). New seal production
(Konar Sandal Trench V’) and still diffusion of early chlorite/steatite vessels. Relations
forced and unforced with Elam and Mesopotamia.
4. Yahya IVA/Takab III. (ca. /–/ bc): later chlorite vessels diffusion and
last south-eastern Iranian seals. New links with Bactrian and Persian Gulf entities and
diplomatic relations with Ur, Isin and Eshnunna.
The great diffusion of the cylinder morphology finished with the end of Yahya IVB–A civiliza-
tion, when a new strong cultural complex from north-eastern regions is attested in eastern Iran
(Lamberg-Karlovsky—Hiebert ); the development of an Anshanite glyptic art during the
first centuries of II millennium bc doesn’t seem to inspire the south-eastern Iranian engravers;
no Anshanite cylinder seals were found in Iran outside Fars (historical Elam) and Susiana plain.
The collapse of the Mesopotamian control system on the neighbouring territories, a definitive
new maritime market development and a strong and homogeneous presence of BMAC in east-
ern Iran carried out a decisive role in the change of the south-eastern Iranian cultural tradition.

17 A deity with snakes sprouting from the body is attested in the Rosen (Fig. g), Bailey (Fig. e) and Foroughi
(Fig. f) collections, while new more sporadic comparisons are possible with the other seal coming from Rosen
collection (Fig. h), from Shahdad (Figs. b, e) and Yahya (Fig. a) excavations. Although the snake-god is frequently
known in the Mesopotamian/Elamite glyptic corpora like Ninghishzida (Amiet : )/Inshushinak (Miroschedji
: –, n. ) and connected to the fertility aspects of the nature, our opinion is that the snake-divinity
known on the south-eastern Iranian seals had an autonomous and local development probably without ideological
or theological connections with the Elamo-Mesopotamian mythological world. For iconographical association
between god-snake and fertility see the Kurangun relief and an Old Elamite cylinder seal published in Trokay :
fig. .
 a new south-eastern iranian glyptic evidence 

Bibliography

Amiet, P. () Elam. Auvers-sur-Oise, Archée Editeur.

———. : Rois et dieux d’Elam. D’après les cachets et les sceaux-cylindres de Suse, Archéologia, ,
pp. –.
———. (): En Iran Central, la civilisation du désert de Lut, Archéologia, , pp. –.
———. (): Antiquités du désert de Lut, I. À propos d’objets de la collection Foroughi, Revue
d’Assyriologie et d’Archéologie Orientale, , pp. –.
———. (): Antiquités du désert de Lut, II, Revue d’Assyriologie et d’Archéologie Orientale, , pp. –.
———. (): La glyptique mésopotamienne archaïque. Paris, Editions du Centre National de la Recherche
scientifique , Quai Anatole-France.
———. (): L’âge des échanges inter-iraniens: – avant J.-C. Paris: Musée du Louvre. Départe-
ment des antiquités orientales. Editions de la Réunion des musées nationaux.
———. (): Elam and Bactria, in Ligabue, G.—Salvatori, S. (eds.), Bactria: An Ancient Oasis Civilisation
form the Sands of Afghanistan, Venezia, Casa Editrice Erizzo, pp. –.
———. (): Un sceau Trans-élamite a Suse, Revue d’Assyriologie et d’Archéologie Orientale, /, pp. –
.
———. (): La glyptique transélamite, in Coubet, A. (ed.), De Chypre à la Bactriane, les sceaux du
Proche-Orient ancien, Actes du colloque international organisé au musée du Louvre par la Service
culturel le  mars , Paris, Louvre Conférences et colloques, pp. –.
———. (): De l’Elam à la Margiane. Le renouvellement des problèmes archèologiques, in Winkel-
mann, S., Seals of the Oasis from the Ligabue Collection, Castello di Godego, pp. –.
———. (): Les sceaux de l’administration princière de Suse à l’époque d’Agadé, Revue d’Assyriologie
et d’Archéologie Orientale, , pp. –.
Ascalone, E. (–): L’indigeno ed il forestiero: elaborazione di nuovi percorsi culturali integrativi
tra seconda metà del III e inizio del II millennio a.C. in Susiana e sul plateau iraniano, Scienze
dell’Antichità, , pp. –.
———. (): Modalità, cause e dinamiche dei rapporti tra Siria e Turan durante la seconda metà del
III millennio a.C.: evidenze archeologiche, epigrafiche e storiche di un ‘indirect long-distance trade’,
Contributi e Materiali di Archeologia Orientale, , pp. –.
———. (a): Archeologia dell’Iran Antico. Integrazioni, interazioni e discontinuità nell’Iran del III
millennio a.C. (= Nisaba ), Di.Sc.A.M., Messina.
———. (b): Cultural Spheres on Iranian Plateau as Seen from Glyptic Evidence. Distribution and
Development of Seals in Iranian Regional Contexts during Mesopotamian Akkadian and Post-
Akkadian Periods (ca. –BC), in Baffi, F. et al. (eds.), Ina kibrat erbetti. Studies in Honor
of Paolo Matthiae, Presented by Colleagues and Friends on the Occasion of His th Birthday, Roma,
pp. –.
———. (): Archaeological Links between Syria and Iran. Stratigraphical Correlation and Regional
Connections, in Matthiae, P. et al. (eds.), From Relative Chronology to Absolute Chronology: The Second
Millennium bc in Syria-Palestine, Roma  novembre– dicembre , Roma , pp. –.
———. (): Cultural Interactions among Mesopotamia, Elam, Transelam and Indus Civilization.
The Evidence of a Cylinder-Stamp Seal from Jalalabad (Fars) and its Significance in the Historical
Dynamics of South-eastern Iran, in Kühne, H. et al. (eds.), Proceedings of the th International Congress
on the Archaeology of the Ancient Near East, Berlin .–. , Eisenbrauns, Winona Lake,
pp. –.
———. (): Glittica Elamita dalla metà del III alla metà del II millennio a.C. Sigilli a stampo, sigilli a
cilindro e impronte rinvenute in Iran e provenienti da collezioni private e museali, L’Erma di Bretschnei-
der, Roma.
Beale, T.W. (): Excavations at Tepe Yahya, Iran –, Cambridge, Peabody Museum of Archae-
ology and Ethnology.
Bisson de la Roque, F. (): Catalogue Général du Musée du Caire, Le Caire, Musée du Caire.
Boehmer, R.M. (): Die Entwicklung des Glyptik während der Akkad-Zeit, Berlin, Walter de Gruyter
and Co.
Collon, D. (): First Impression: Cylinder Seals in the Ancient Near East, London, British Museum
Press.
Delaporte, L. (): Musée du Louvre, Catalogue des cylindres orientaux. Catalogue des cylindres, cachets
et pierres gravées de style oriental I. Fouilles et missions, Paris, Librairie Hachette.
 enrico ascalone

Duchesne-Guillemin, M. (): Remarques d’organologie comparée, Mesopotamie-Iran, in De Meyer,

L.—Gasche, H. (eds), Mesopotamian History and Environment, Mesopotamie et Elam, Actes de la
XXXVIème Rencontre Assyriologique Internationale, Gand, – juillet , Ghent, University of
Ghent, pp. –.
Dyson, R.H.—Harris, S.M. (): The Archaeological Context of Cylinder Seals Excavated on the
Iranian Plateau, in M. Kelly-Buccellati (ed.), Insight Through Images: Studies in Honour of Edith Porada,
Malibù: Undena Publications, pp. –.
Hakemi, A. (): Shahdad, Archaeological Excavations of a Bronze Age Centre in Iran, Roma, Istituto
Italiano per il Medio ed Estremo Oriente.
Harper, P.O.—Aruz, J.—Tallon, F. (): The Royal City of Susa, Ancient Treasure in the Louvre, New
York, The Metropolitan Museum of Art.
Heimpel, W. (): Magan, Reallexikon der Assyriologie und vorderasiatische Archäologie, , pp. –
.
Horowitz, W. (): Two Abnu šilkinšu Fragments and Related Materials, Zeitschrift für Assyriologie
und Vorderasiatische Archäologie, , pp. –.
Kantor, H.J (): The Aegean and the Orient, American Journal of Archaeology, , pp. –.
Kohl, P.C. (a): Carved Chlorite Vessels: A Trade in Finished Commodities in the mid-third-
Millennium, Expedition, /, pp. –.
———. (b): The Archaeology of Trade, Dialectical Anthropology, , pp. –.
———. (): “Steatite” Carvings of the Early Third Millennium B.C., American Journal of Archaeology
, pp. –.
———. (): The Balance of Trade in Southwestern Asia in the Mid-Third Millennium B.C., Current
Anthropology, , pp. –.
———. (): The “World Economy” of Western Asia in the Third Millennium bc, in Taddei, M. (ed.),
South Asian Archaeology , Napoli, Herder, pp. –.
———. (): The First World Economy: External Relations and Trade in West and Central Asia in the
Third Millennium bc, in Nissen, H.-J.—Renger, J. (eds.), Mesopotamien und seine Nachbarn, Berlin,
Dietrich Reimer Verlag, pp. –.
———. (): Reflections on the Production of Chlorite at Tepe Yahya:  Years Later, in Potts, D.T.
(ed.), Excavations at Tepe Yahya, Iran, –: Periods IVC and IVB (–BC), Cambridge,
pp. –.
Lamberg-Karlovsky, C.C. (): Excavations at Tepe Yahya, Iran –, Progress Report I, Cam-
bridge, Massachusetts, Peabody Museum of Archaeology and Ethnology Harvard University.
———. (): The Protoelamite Settlement at Tepe Yahya, Iran, , pp. –.
———. (): Tepe Yahya . Mesopotamia and Indo-Iranian Borderlands, Iran, , pp. –.
———. (): Urban Interaction in the Iranian Plateau: Excavation at Tepe Yahya, –, Proceed-
ings of the British Academy, , pp. –.
———. (): Tepe Yahya Project, Iran, , p. .
———. (): Afterwards, in Lamberg-Karlovsky, C.C. (ed.), Excavations at Tepe Yahya, Iran –.
The Third Millennium, Cambridge, Massachusetts, Peabody Museum of Archaeology and Ethnology
Harvard University, pp. –.
Lamberg-Karlovsky, C.C.—Hiebert, F.T. (): Central Asia and the Indo-Iranian Borderlands, Iran,
, pp. –.
Lamberg-Karlovsky, C.C.—Kohl, P.C. (): The Early Bronze Age of Iran as Seen from Tepe Yahya,
Expedition, /–, pp. –.
Lamberg-Karlovsky, C.C.—Tosi, M. (): Shahr-i Sokhta and Tepe Yahya: Tracks on the Earliest
History on the Iranian Plateau, East and West, , pp. –.
Landsberger, B. (): Assyrische und “Dunkles Zeitalter”, Journal of Cuneiform Studies, , pp. –.
Legrain, L. (): Empreintes de cachets élamites. Mémoires de la Mission Archéologique de Perse ,
Paris, Editions Ernest Leroux.
Madjidzadeh, Y. (a): Jiroft. The Earliest Oriental Civilization, Tehran.
———. (b): La découverte de Jiroft, Dossiers d’Archéologie, , pp. –.
———. (c): La première campagne de fouilles à Jiroft dans le bassin du Halil Roud (janvier et février
), Dossiers d’Archéologie, , pp. –.
———. (): L’ancêstre des ziggourats à Jiroft, Archéologie, , pp. –.
———. (): Excavations at Konar Sandal in the Region of Jiroft in the Halil Basin: First Preliminary
Report (–), Iran, , pp. –.
 a new south-eastern iranian glyptic evidence 

Miroschedji de, P. (): Vases et objets en stéatite susiens du Musée du Louvre, Cahiers de la Délégation
Archéologique Française en Iran, , pp. –.
———. (): Le dieu élamite au serpent et eaux jaillissantes, Iranica Antiqua, , pp. –.
Pittman, H. (): Glyptic Art of Period IV, in Lamberg-Karlovsky (ed.), Excavations at Tepe Yahya,
Iran, –: Periods IVC and IVB (–BC), Cambridge, Massachusetts, Peabody Museum
of Archaeology and Ethnology Harvard University, pp. –.
———. (): La culture du Halil Roud, Dossiers d’Archéologie, , pp. –.
Porada, E. (): Antica Persia. Milano, Il Saggiatore.
———. (): Problems of Interpretation in a Cylinder Seal of the Akkad Period from Iran, in Kampman,
A.A.—Van der Ploeg, J.P.M. (eds.), Compte Rendu de l’Onziéme Rencontre d’Assyriologique Interna-
tionale, Leiden, Nederlands Instituut Voor Het Nabije Oosten, pp. –.
———. (): Remarks on the Tôd Treasures in Egypt, in Orthmann, W. (ed.), Societies and Languages
of the Ancient Near East. Studies in Honour of I.M. Diakonoff, Warminster, Arts and Phillips LTD,
pp. –.
———. (): Discussion of a Cylinder Seal, probably from Southeast Iran, Iranica Antiqua, , pp. –
.
———. (): Seals and Related Objects from Early Mesopotamia and Iran, in Curtis, J. (ed.), Early
Mesopotamia and Iran: Contact and Conflict –, London, British Museum Press, pp. –.
Potts, D.T. (): Tradition and Transformation: Tepe Yahya and the Iranian Plateau during the Third
Millennium B.C. Ph.D. dissertation, Department of Anthropology, Harvard University.
———. (): Echoes of Mesopotamian Divinity on a Cylinder Seal from South-eastern Iran, Revue
d’Assyriologie et d’Archéologie Orientale, , pp. –.
———. (): Excavations at Tepe Yahya, Iran, –: Periods IVC and IVB (–BC), Cam-
bridge, Massachusetts, Peabody Museum of Archaeology and Ethnology Harvard University.
Potts, T.F. (): Foreign Vessels of the Late Third Millennium bc from Southern Mesopotamia: Their
Origins and Mechanisms of Exchange, Iraq, , pp. –.
Salvatori, S. (): Bacria and Margiana Seals, A New Assessment of Their Chronological Position and
a Typological Survey, East and West, , pp. –.
Salvatori, S.—Vidale, M. (): A Brief Surface Survey of the Protohistoric site of Shahdad, Iran:
Preliminary Report, Rivista di Archeologia, , pp. –.
Sarianidi, V.I. (): The Bactrian Pantheon, UNESCO-Information Bulletin , pp. –.
———. (): Margiana and Protozoroastrism, Athens, Kapon Edition.
———. (): Gonurdepe. Türkmenistan. City of Kings and Gods, Asgabat, Miras.
Sarzec de, E. and Heuzey, L. (–): Découvertes en Chaldée, Paris, Editions Ernest Leroux.
Steinkeller, P. (): The Question of Marhaši: A Contribution to the Historical Geography of Iran in the
Third Millennium B.C., Zeitschrift für Assyriologie und vorderasiatische Archäologie, , pp. –.
———. (): Markhaši, Reallexikon der Assyriologie und vorderasiatische Archäologie, , pp. –.
Trokay, M. (): Les origines du dieu élamite au serpent, in De Meyer, L.—Gasche, H. (eds.), Mesopota-
mian History and Environment, Mésopotamie et Elam, Actes de la XXXVIème Rencontre Assyriologique
Internationale, Gand, – juillet , Ghent, University of Ghent, pp. –.
Vallat, F. (): Eléments de géographie élamite, Paléorient, /, pp. –.
Van Dijk, J. (): Išbi"Erra, Kindattu, l’homme d’Elam et la chute de la ville d’Ur, Journal of Cuneiform
Studies, , pp. –.
Winkelmann, S. (): Ein neues Trans-elamisches Siegel, Archäologische Mitteilungen aus Iran und
Turan, , pp. –.
———. (): Intercultural Relations between Iran, the Murghabo-Bactrian Archaeological Complex
(BMAC), Northwest India and Failaka on the Field of Seals, East and West, , pp. –.
 NEW EVIDENCES ON EMERGENCE OF
COMPLEX SOCIETIES IN THE CENTRAL IRANIAN PLATEAU

M. Hessari and R. Yousefi Zoshk*

. Introduction

The site of Sofalin (Fig. .) lies in the eastern Ray Plain1 of north-central Iranian plateau, at Lat.
00 0  N., Long. 00 0  E., at a general elevation above sea level of about  meter. This
location is situated some  kilometers east of the city of Varamin (Fig. .). The site takes its
name from the density of pottery sherds on its surface (Sofalin means pottery sherds). The site
itself is about  kilometers south of Khorassan Road, the major East-West trade route, between
southern Mesopotamia, the Iranian plateau and Central Asia.
The extensive remains of Tepe Sofalin, in an area about  meters long and  meters wide
with  meters high consists of material cultures whose sequence extends from the late-th
millennium to the Iron Age III. A very small portion of this extensive site, about less than .
percent of the total, was uncovered during  seasons of work (, ) by an expedition of
the Archaeological Service of Islamic Azad University of Varamin-Pishva, in cooperation with
the Iranian Center for Archaeological Research.2 Our excavations indicated that only about
some meters of this elevation consisted of occupational debris, the lower core being a vast
natural hill. The plan profile of the mound (Fig. .) is roughly ovoid and about  meters
in diameter; the main irregularity is a straight indention on the northern and eastern flank, due
apparently to some stage of human construction activities in recent years.
Tepe Sofalin provides a particularly clear illustration of cultural interaction in the th
and early rd millennia between the Iranian central plateau and the more densely populated
settlements on the alluvial plains of Khuzestan. Some of the trenches opened during the
excavation contained proto-Elamite tablets and tablet blanks, sealing impressions with strong
parallels to Susa, and polychrome proto Elamite / Jemdet Nasr ceramics. In this article, the
authors tend to discuss only the materials of the late th millennium.

* Isfahan University & Islamic Azad University, Varamin-Pishva Branch.

1 The Ray plain is located in the north-central Iranian Plateau at an elevation of between  to m. This semi-
arid fertile plain is limited by the Alborz mountain ranges in the north and the Dasht-e-Kavir desert in the south.
The plain is covered with water-transported alluvial sediment. The Rey plain is irrigated by couple of permanent
rivers that flow from the Alborz mountains. The main rivers are the Karaj, Shour and Jajrood. Irrigation is possible
in many locations through the manipulation of the many streams, rivers and springs. The Alborz Mountains create a
climatic border between the coastal plains of the Caspian region and the great central plateau of Iran by obstructing
precipitation from entering the interior of the country.
2 The authors wish to express their thanks to the following persons: Dr. Mohammad Reza Feizbakhsh, former

Head of Islamic Azad University of Varamin-Pishva for his kind aid in sponsoring the project; Dr. Hassan Fazeli,
Head of the Iranian Center for Archaeological Research for his kind permission to initiate the project; Dr. Malek
Shahmirzadi for suggesting useful modifications; Abdoreza Mohajerinejad, Representative of the Iranian Center
for Archaeological Research in Tepe Sofalin Expedition for his support and cooperation; Hassan Akbari, Faramarz
Ra"yati, Abulfazl Moheban for their cooperation in the fieldwork.
 m. hessari and r. yousefi zoshk

Fig. .. General View of Tepe Sofalin

Fig. .. Satellite Image of Rey Plain

the emergence of complex societies in the central iranian plateau 

Fig. .. D Reconstruction of Tepe Sofalin

Fig. .. Distribution of Late th Millennium Sites Across Iran

 m. hessari and r. yousefi zoshk

. Archaeological Evidences for

Late Fourth Millennium Societies in Iran (Fig. .)

Numerous archaeological investigations concentrated on the emergence of complex societies

have been initiated in Iran in the past decades. Under the direction of Perrot (, ),
French archaeologists at Susa have re-examined sections left from earlier work on the Acropolis,
conducted limited excavations, and produced a reliable, stratigraphically determined sequence
extending from late Susiana times (c.  bc) to the end of the Protoliterate period. Excava-
tions at Chogha Mish by Helen Kantor (Kantor , ; Alizadeh ) cover the complete
chronological span from the Neolithic up to the Proto-Literate period and provide vital new
information on cultural developments of late fourth millennium communities and early admin-
istrative systems. Both Sumner (Sumner , ) at Malyan, the ancient Elamite capital
of Anshan, and Young (Young , Weiss & Young ) at Godin Tepe have unearthed
large literate settlements on the Iranian plateau which date to the late th millennium. Proto-
Elamite tablets have been discovered in Tal-i Ghazir (Caldwell , Whitcomb ) and at
Tepe Yahya (Lamberg-Karlovsky ; Damerow & Englund ). Through a series of diag-
nostic chronological markers at Tepe Yahya, Lamberg-Karlovsky () has demonstrated the
contemporaneity of these developments in eastern Iran with the more spectacular changes in
Mesopotamia and Khuzistan. Excavations at Shahr-i Sokhta (Tosi ) have presented evi-
dence for craft specialization, and the work of Iranian archaeologists at Shahdad (Hakemi ,
) has shown the accumulation of wealth and, most likely, the beginnings of class stratifi-
cation on the eastern Iranian plateau in the early rd millennium. Bampur and Tal-i Iblis, two
sites originally investigated by Stein (), were re-excavated in the s by de Cardi ()
and Caldwell (Caldwell & Dougherty ; Caldwell ). The latter site yielded evidence for
the melting of copper in the early th millennium and contained rare but diagnostic ceramic
parallels with sites as far west as Mesopotamia.
The evidence from Sialk Period IV is similar to that from Godin V: tablets (which at Sialk are
indubitably Proto-Elamite economic texts), ceramics, and seal impressions were found in an
architectural complex situated on the highest part of the southern mound at the site. The oldest
occupation of the so-called Sialk IV period illustrates craft specialization and an administration
system in the central Iranian plateau (Chegini et alii ; Helwing ). The site of Arisman,
km southeast of Kashan, suggests settlement activities over the whole area from the late th
to the rd millennia bc.
Tepe Hissar, the largest known urban settlement in northeastern Iran provided the primary
archaeological record in the region, with its continuous habitation levels from the th to the nd
millennia bc. The importation of lapis and turquoise implies connections with the east, and at
the same time links with the west have been documented by blank clay tablets reminiscent of
Proto-Elamite tablets, and a cylinder seal.3 Farhardgerd is the far eastern settlement related to
the Proto-Elamite administrative organization near the modern city of Sabzevar in northeastern
Iran in which Gropp identified some beveled rim bowls (Gropp : –).
Unpublished polychrome pottery sherds from the Cheshme Ali collection in the British
Institute (Teheran) confirm the existence of late th millennium communities in the central
Iranian plateau.
Giyan (Dyson : ), Deshawer (Braidwood : ), and several sites in the Luristan
region—Mahidasht (Levine : , ), Hulilan (Mortensen : ; Id., ), Malayer
(Howell : ) and the Tarhan/Rumishan/Kuh-i Dasht Plains (Goff : –)—have

3 Unpublished report of the Hissar excavations under direction of E. Yagmaie.

 the emergence of complex societies in the central iranian plateau 

yielded occasional beveled-rim bowls or other Uruk-related ceramics. There is no published

mention of Jemdet Nasr- or Banesh-style polychrome pottery from these areas, and in fact much
of Luristan may have been abandoned between the mid-th and the early rd millennium bc.
(Mortensen : ).

. Description of Diagnostic of
Late th Millennium Wares of Tepe Sofalin

The ceramics discussed here (Tepe Sofalin) are dated to the late th millennium and the Early
Bronze Age and come from trenches excavated in . The late th and early rd millennium
occupation at Tepe Sofalin appears to have been relatively brief and this is reflected in overall
coherence of pottery assemblage. With few exceptions of local pottery assemblage, most types
occur throughout the Proto-Elamite sequence. The Proto-Elamite levels at Tepe Sofalin are
contemporary with Godin V (Young : –), Susa –B (LeBrun , Dittmann
a: –, –; Dittmann ), Ville Royalle – (Carter , Dittmann a:
–, ; b: ), Tepe Farukhabad (Wright ), Yahya IVc (Lamberg-Karlovsky
; Potts et alii ), the late middle Banesh building level  of Malyan (Banesh C–D)
(Sumner , Dittmann c: –; Nicholas ), Sialk IV (Ghirshman ) and
Ghabristan  (Majidzadeh ; Negahban ; Talaee ).
The Proto-Elamite pottery assemblages were produced at the site Sofalin. This is attested by
the presence of wasters and warped sherds on the surface and also by the discovery of one kiln
on the northern zone of the mound. Although Uruk occupation was not located during the
excavation, few Uruk standard pottery sherds appeared in disturbed layers and in a pit and on
the surface as well.
All the ceramics were analyzed during the excavation seasons. Body sherds were stored
according to the following categories and the numbers in each category were recorded: total
number, painted, combed, corrugated, with pot marks, scraped, washed, burnished, slag,
pierced sherds, sherds with bitumen coating, and neck-shoulder joins of jars. Diagnostics
of rims, bases, handles and other identifiable parts of vessels were classified into number
of types. Each type was named and a form produced with spaces for data on appropri-
ate variables for the type, then each sherd was recorded on the appropriate form for its
type, and data concerning variables within the type (diameter, core and surface color, tem-
per, etc.) were recorded. After all, the percentages were calculated using only examples for
which a given piece of information was recorded. Since the analysis of wares is currently
incomplete, we will limit the discussion of Tepe Sofalin Wares to a brief description of
those present.

.. Chaff-Tempered Wares

Generally speaking, formal classification of the chaff-tempered ceramics at Tepe Sofalin is
simple. All parts include bases, mid-sections and rims of low trays, drain pipes, beveled-rim
bowls and cups and are readily identifiable. Similarly, the bases and mid-sections of pedestal-
based goblets are easily recognizable.

... Beveled-Rim Bowls

Like beveled-rim bowls elsewhere, these vessels appear to have been made using prepared holes
in the ground as mold (Wright & Johnson ); the exterior side walls and the undersurface of
the base are crinkly and the interior of the base frequently shows a thumb impression where the
 m. hessari and r. yousefi zoshk

clay was pressed down into the mold. The interior of the vessel was smoothed, and the exterior
rim finished off by smoothing the upper exterior edge of the rim in such way as to create the
characteristic bevel. The walls are completely oxidized or have only a light grey tinge in their
center, even under low firing conditions. Except for one complete specimen, the Tepe Sofalin
beveled-rim bowls cannot be used as a further test of the hypothesis that in Mesopotamia such
bowls were used as ration containers.

... Pedestal-Based Goblets

These vessels are wheel-made, with string-cut bases. Their diameters vary from  cm to  cm
in the Sofalin pottery assemblages. The bases are not completely solid pedestals, but have hollow
centers spaces which have penetrated down to varying depths into the base from the body of
the vessel. The upper portions of the bases are distinctly narrower than the foot. The body of the
vessels is generally symmetrical rather than asymmetrical in appearance, with the upper walls
thinning significantly to end in one small variety of simple and pinched rims. The exterior and
interior of these vessels is smoothed.

... Unrestricted Chaff-Ware Forms (Fig. .–)

This vessel form frequently turned up as sherds of complete profile with a rim diameter in the
extra-large rang, although oval rather than round. This is handmade, has a flat base of about
cm and sides which flare smoothly down from the rim to the base.

.. Grit-Tempered Wares (Fig. .–)

By count, grit-tempered wares of Tepe Sofalin constitute approximately half of the total assem-
blage recovered from this site. Unfortunately, only a small number of complete profiles of vessels
with grit-tempered were discovered. For the reason explained earlier, it is clear that rims are
much more useful in classifying functional variation within a contemporaneous assemblage of
pottery than are bases and body sherds.

... Restricted Grit-Ware Forms

These have an inward orientation of upper body wall without a distinct neck.

... Hole-Mouth Wares (Fig. .–)

A hole-mouth rim is one that is inwardly rather than outwardly or vertically oriented, and that
arises directly from the wall of the vessel without the existence of a distinguishable neck area.

... Drooping Spout Wares (Fig. .)

These kinds of pottery assemblage do not appear in great number in Tepe Sofalin. The smoothed
sides of these vessels do not turn inward to create a waist, but rise more or less flaring from the
base. It is probable that the upper portion of these vessels ended in simple or pinched rims such
as those known to occur on the pedestal-based vessels. The spout is appliquéd erected with an
angle of  degrees.
the emergence of complex societies in the central iranian plateau 

Fig. . Unrestricted Chaff Forms: –) Banesh Trays: Grit-Tempered Wares;

Restriced Forms, –) Hole Mouth Wares; ) Drooping Sprout Ware
 m. hessari and r. yousefi zoshk

Fig. . Restricted Vessel with Uppermost Body Wall: –)

Everted Wares; –) Folded Wares; –) Ledged Rim Wares

... Restricted Vessels with Uppermost Body Wall (Necked Forms) (Fig. .–)
.... Everted Wares (Fig. .–)
This rim group consists of unexpanded forms which bend or flare outward generally at an angle
of  degrees from a point on the shoulder of the vessel; this bending point is thus an area of
construction.
 the emergence of complex societies in the central iranian plateau 

.... Folded Wares (Fig. .–)

A folded rim has the appearance of an everted or expanded rim that has been pushed first down
and back in approaching to or against the neck of the pot.
.... Ledge Rim Wares (Fig. .–)
A ledge rim bends out from the vessel wall at an approximate angle of  degrees, and is
generally thin and flattened in appearance, creating a shelf around the actual orifice of the vessel.

. Sofalin Tokens, Clay Balls, Tablets and Cylindrical Seal Impressions

.. Tokens (Fig. .–)

Sculpted and incised clay tokens, excavated from contexts as early as the th millennium bc,
have been cited as three-dimensional precursors for the first cuneiform signs (Schmandt-
Besserat : –). Simple baked tokens are ubiquitous artifacts found in administrative
contexts across excavations in Iran and Mesopotamia. These tokens are also referred to as
“counters” thus implying their function as numerical administrative tool.
During the excavation of Tepe Sofalin couples of clay tokens in different shape were dis-
covered. Similar in size to marbles, these tokens are found in a wide variety of shapes; simple
spheres, cones, rectangles, triangles, biconoids, and even jugs and animals. Complex tokens, in
addition to being plastically molded, have incised markings on the surface which presumably
added more detail to the information imparted by the plastic form alone.

.. Clay Balls (Fig. .)

Clay balls first appeared during the late Uruk period, a phase of development of administrative
technology in Mesopotamia and southwestern Iran. The first clay balls appear in the late Middle
Uruk period contexts with cylinder seal impression on the exterior. By the beginning of the late
Uruk period,4 these clay balls with cylinder seal impression are found with the addition of token
impression at Susa, Chogha Mish and Habuba Kabira. This combination is never found at Uruk
(Dittmann ). At Uruk the hollow clay ball is immediately followed by the appearance of
early tablets with numerical signs. Accordingly, the Late Uruk writing materials in Tepe Sofalin
can be divided into a bullae period, the earliest from the view point of stratigraphy, a phase of
development of administrative technology, in which the geometric clay counters were enclosed
in clay envelopes called bullae. The Tepe Sofalin’s bulla is bare of impressions5 from cylinder
seals or the impression of the tokens. It is . cm in diameter and is made from well-levigated
clay. Although it was broken when found, it still contained one token of unbaked clay, with
a size of cm. The Sofalin bulla was found at the lowest context on the sterile soil in which
almost all of the late Uruk administration materials emerged. The tokens are plain spheres; the
clay of the tokens is very fine and was obviously smoothed to remove all inclusions. The tokens
seem to have been modeled wet since the fingerprints of the maker is visible. The tokens are
completely identical with the tokens that have been found in the clay balls at other late Uruk
sites (Schmandt-Besserat , App. I, Type I, No. ). Most likely, the Tepe Sofalin’s bulla shows
great similarities with Southern Mesopotamia in the stage of Uruk expansion rather than with
Susa and southwestern Iran.

4 The chronology is based on Uruk IV, Susa , Habuba kabira, Jebel Aruda and Tel Brak.
5 The bulla’s surface is badly effaced and does not let us to determine if it was impressed by a seal or not! However,
since it does not have any token impression on its exterior surface, it cannot be later than Late Uruk bullae from
Chogha Mish, Susa and Habuba Kabira. Presumably, it covers an earlier time span of late Uruk period.
 m. hessari and r. yousefi zoshk

Fig. . Administrative and Bookkeeping Devices: ) Simple Sphere Tokens; )

Complex Geometric Tokens; ) Miniature Animal Tokens; ) Hollow Clay Ball

.. Early Numerical Tablets (Fig. .)

Flat and rounded, sealed and unsealed tablets appear for a short period of time towards the
end of the late Uruk period. These tablets bear marks made with a new administrative tool: the
stylus. It seems that the bolded clay balls from the previous stage had been flattened, since the
use of numerical impressions on the surface was sufficient enough to impart the information. A
more formalized version of the rounded tablets becomes common from about –bc.
During this  year period, the numerical notations become standardized. During the excava-
tion of , in trench Z, an early numerical tablet with early stylus impression was discovered
in the company of large storage jars. It is a completely preserved oblong, convex, unsealed tablet
with only three notations on its obverse, impressed with tokens or with stylus cut and shaped
 the emergence of complex societies in the central iranian plateau 

Fig. . ) Numerical Tablet; –) Ideo-Numerical Tablets

to imitate tokens, representing numerical values. In all of its general attributes and in most
particulars, the script is very similar to the early numerical scripts (Strommenger ). The
tablet is of a small size, measuring  × ×  mm, with the script running from right to left,
and consisting of three large deep circular numerical notations sunk . cm into the clay and
 m. hessari and r. yousefi zoshk

about .cm in diameter, on the same line on the obverse and the reverse and the edges. There is
not any effacement of the obverse. According to its format, it groups together with early scripts
from Susa Acropole , Levels , Godin V and the tablets in the Red temple of Uruk.
It is interesting that the tablet was left unfinished,6 which proves the local entity of the scripts.
Falkenstein (: ) identified the deep circular sign; ATU , as number,  but Vaiman
(: ) believes ATU  is measure of capacity equal to  units, whereas Friberg (:
; also Schmandt-Besserat : ) judges ATU  to indicate units of grain metrology,
equal to  bariga.

.. Numero-Ideographic Tablets (Fig. .–)

The first inscribed tablets with numero-idiographic tablets are dated to around bc. The
surface is marked with numerical notations, seal impressions and one or two inscribed images.
These inscribed images are referred to by scholars as signs because they are inscribed. Rather
than being impressed by the flat end of the stylus, these signs are drawn into clay with a
pointed end of the stylus. Akin to line drawings made in the clay, signs were shaped of some
type of commodity such as a jug of milk or a sheep; they only represent discrete objects, not
ideas or concepts. During the excavation of Tepe Sofalin, in , in trench  and , in the
southern edge of the mound,  numero-ideographic tablets were discovered. These tablets
can certainly be identified as Proto-Elamite texts.7 Proto-Elamite tablets of Tepe Sofalin are
the earliest complex written documents from the northern central Iranian plateau. The script
consists of both numerical and ideographic signs. The form and content of these eleven tablets is
entirely consistent with that of the standard and late Proto-Elamite tablets from Susa. Although
a majority of the inscribed objects from Tepe Sofalin are very fragmentary, they document
the existence of a developed administration system and of bookkeeping techniques. However,
only a limited number of tablet-content categories are present in the current sample of texts and
fragments. Some of the texts concern workers and rations, whereas some others concern animal
husbandry and others have an uncertain content. Many more tablets were found in subsequent
excavations, these will be published as soon as possible.
All the texts, except two of them date to the late (standard) period of the Proto-Elamite
writing system. A majority of the signs on the twelve tablets from Tepe are found in the Susa
repertoire as well. Tablets from subsequent seasons reveal more variation.
All of the numerical signs in the texts from Tepe Sofalin are known from the Susa texts (a
tablet found during season four, , has a new numerical system using the same signs found in
Susa and beyond). The seals on the tablets and sealings from Tepe Sofalin are strikingly similar
to seals found at Susa as well. Tepe Sofalin therefore provides a particularly good illustration
of cultural interaction in the late th and early rd millennia between the different parts of
the Iranian central plateau and the more densely populated settlements on the alluvial plains
of Khuzestan.

6A shallow circular sign near the numerical notation indicates that an owner wanted to incise another notation,
but perhaps because of a mistake he made, the script was discarded.
7 “Proto-Elamite” is the term for a writing system in use in the Susiana plain and the Iranian highlands

east of Mesopotamia between ca.  and bce, a period generally considered to correspond to the Jemdet
Nasr/Uruk III through Early Dynastic I periods in Mesopotamia.
 the emergence of complex societies in the central iranian plateau 

Fig. . Cylinder Seal Impressions from Tepe Sofalin

.. Cylinder Seal Impressions and Container Sealing (Fig. .–)

Another type of an image-bearing device in administrative systems in Tepe Sofalin, the cylinder
seal impression, functioned in the same way as the stamp seal, but was rolled across the surface
of the clay to create an impression that could cover a much greater surface area. These themes of
seal impressions of Tepe Sofalin were composed from a set number of figural elements, arranged
in a particular order. Most of the Tepe Sofalin cylinder seal imagery of the late th millennium is
similar to Proto-Elamite glyptic impressions of Susa. Human and animal figure are curved with
great care. Iconography of Sofalin seal impressions includes human figures, animals, geometric
 m. hessari and r. yousefi zoshk

designs, floral motives, etc. It appears that most Tepe Sofalin seal impressions are part of
the Proto-Elamite I tradition of southwestern Iran. There are few images that depict stylistic
traits of the central Asian late th millennium tradition. Approximately, more than   of the
sealing imagery fits stylistically and iconography into the Proto-Elamite  Period, as has been
mentioned before.

. Conclusions

The research presented in this article has produced a picture of a developed proto-urban
center in the northern central Iranian Plateau with a strong administrative system related to
the Proto-Elamite economic organization. The landscape features of Tepe Sofalin documented
during two seasons of expedition confirms the association of long-distance trade and late
th millennium urban centers and demonstrate that this association extends to the sites of
southwestern Iran and perhaps Central Asia. Probable evidence of long distance trade between
the Susiana plain and the central Iranian plateau and central Asia is surprisingly scarce. The
present evidence of Tepe Sofalin thus suggests an increasing dependence of the Proto-Elamite
centers in southwestern Iran on raw materials. It seems that in the Sofalin complex society an
elite group in mutual relation to Proto-Elamite communities in southwestern Iran controls both
the political and economic institutions and manipulates these institutions for their own benefits.
Within the sphere of economic activity, it seems that proto-Elamite elites in southwestern Iran,
perhaps Susa (?), attempt to control the supply and minimize procurement costs for the items
and materials that their societies require by establishment of some administrative offices on
their trade routes. As an example of this, Tepe Sofalin is a major socio-political organization on
the major East-West trade route, between southern Mesopotamia, southwestern Iran, Iranian
plateau and Central Asia and Afghanistan.
The impact of complex societies of southwestern Iran on a less complex society is dependent
on the specific attributes of both societies. The social, political and economic organizations
that emerge in the adjacent society will in some ways determine the type of community it
evolves into, in response to interaction. The level of sociopolitical organization and economic
specializations of Tepe Sofalin community indicates that strong mechanisms of finance with
southwestern Iranian Susiana plain based of trade, led to the emergence of a developed complex
society in the northern central Iranian plateau.

Bibliography

Alizadeh, A. (): Chogha Mish. The First Five Seasons of Excavations – (OIP ), Chicago.
Braidwood, R.J. (): “Seeking the world’s first farmers in Persian Kurdistan: A full scale investigation
of Prehistoric sites near Kermanshah”, Illustrated London News , –.
Caldwell, J.R. (): Investigations at Tal-i Iblis (Illinois State Museum Preliminary Reports ), Spring-
field.
———. (): “Ghazir, Tell-i”. RlA , Berlin, –.
Caldwell, J.R. & Dougherty, R.C. (): “Evidence of Early Pyrometallurgy in the Kerman Range in
Iran”, Science, N.S. , No. , –.
de Cardi, B. (): Excavations at Bampur, A Third Millennium Settlement in Persian Balutchistan, New
York.
Carter, E. (): “Excavations in the Ville Royale at Susa: The Third Millennium B.C. Occupation”,
CDAFI , –.
Chegini, N.N, Momenzadeh, M., Parzinger, H., Pernicka, E., Stöllner, T., Vatandost, R. & Weisgerber,
G. (): “Preliminary Report on Archaeometallurgical Investigation around the Prehistoric Site of
Arisman near Kashan, western Central Iran”, AMIT , –.
 the emergence of complex societies in the central iranian plateau 

Damerow, Peter & Englund, R. (): The Proto-Elamite Texts from Tepe Yahya (American School of
Prehistoric Research Bulletin ), Cambridge.
Dittmann, R. (a): “Susa in the Proto-Elamite Period and Annotations on the Painted Pottery of
Proto-Elamite Khuzestan”, U. Finkbeiner & W. Röllig (eds.), Ğamdat Nasr, Period or Regional Style?
Papers given at a symposium held in Tübingen, November  (Beihefte zum Tübinger Atlas des
vorderen Orients. Reihe B: Geisteswissenschaften ), Wiesbaden, –.
———. (b): “Seals and Sealings and Tablets”, U. Finkbeiner & W. Röllig (eds.), Ğamdat Nasr, Period or
Regional Style? Papers given at a symposium held in Tübingen, November  (Beihefte zum Tübinger
Atlas des vorderen Orients. Reihe B: Geisteswissenschaften ), Wiesbaden, –.
———. (c): Betrachtungen zur Frühzeit des Südwest-Iran. Teil . Regionale Entwicklungen vom . bis
zum Frühen . vorchristlichen Jahrtausend (Berliner Beiträge zum vorderen Orient ), Berlin.
———. (): “Bemerkungen zum Protoelamischen-Horizont”, AMI , –.
———. (): “Iran als Mittler zwischen Ost und West”, A. Hausleiter, S. Kerner & B. Müller-Neuhof,
Rezeption archäologischer Denkrichtungen in der Vorderasiatischen Altertumskunde. Internationales
Symposium Berlin .-. Juni  (im Druck).
Dyson, R.H. (): “Problems in the Relative Chronology of Iran, –BC”, R.W. Ehrich (ed.),
Chronologies in Old World Archaeology, Chicago, –.
Falkenstein, A. (): Archaische Texte aus Uruk (ADFU ), Leipzig.
Friberg, J. (): “Numbers and Measures in the Earliest Written Records”, Scientific American ,
–.
Ghirshman, R. (): Fouilles de Sialk, Vol. . Paris.
Goff, C. (): “Luristan Before the Iron Age”, Iran , –.
Gropp, G. (): Archäologische Forschungen in Khorasan, Iran (Beihefte zum Tübinger Atlas des
vorderen Orient, Reihe B: Geisteswissenschaften ), Wiesbaden.
Hakemi, A. (): Catalogue de l’exposition Lut-Xabis (Shahdad) (Premier symposium annuel de la
recherché archéologique en Iran), Tehran.
———. (): Shahdad. Archaeological Excavations of a Bronze Age Center in Iran (Reports and Memoirs
), Rome.
Helwing, B. (): “Long-Distance Relations of the Iranian Highland Sites during the Late Chalcolithic
Period: New Evidence from the Joint Iranian-German Excavations at Arisman, Province Isfahan,
Iran”, U. Franke-Vogt & H. Weisshaar (eds.), South Asian Archaeology . Proceedings of the 
International Conference of the European Association of South Asian Archaeologists (– July ,
Bonn), Aachen, –.
Howell, R. (): “Survey of the Malayer Plain”, Iran , –.
Kantor, H.J. (): “The Prehistoric Cultures of Chogha Mish and Boneh Fazili”, M.Y. Kiani (ed.),
The memorial volume of the VIth International Congress of Iranian Art and Archaeology, Oxford,
September –th, , Tehran, –.
———. (): “Chogha Mish and Chogha Bonut”, Iran , –.
Lamberg-Karlovsky, C.C (): Excavations at Tepe Yahya –. Progress Report  (Harvard
University. Peabody Museum. American School of Prehistoric Research. Bulletin  / The Asia
Institute of Pahlavi University. Monograph series ), Cambridge.
———. (): “The Proto-Elamite Settlement at Tepe Yahya”, Iran , –.
———. (): “Trade mechanisms in Indus-Mesopotamian interrelations”, JAOS , –.
Le Brun, A. (): “Recherches stratigraphiques á l’acropole de Suse (–)”, CDAFI , –.
Levine, L.D. (): “Survey in the Province of Kermanšāhān, : Māhidasht in the Prehistoric
and early Historic Periods”, F. Bagherzadeh (ed.), Proceedings of the IVth Annual Symposium on
Archaeological Resarch in Iran, Tehran, –.
Majidzadeh, Y. (): “Excavations in Tepe Ghabristan: The first two seasons, –” Marlik ,
–.
Mortensen, P. (): “Survey and soundings in the Holailan Valley ”, F. Bagherzadeh (ed.), Proceed-
ings of the IIIrd Annual Symposium on Archaeological Research in Iran, Tehran, –.
———. (): “Chalcolithic Settlements in the Holailan Valley”, F. Bagherzadeh (ed.), Proceedings of the
IVth Annual Symposium on Archaeological Resarch in Iran, Tehran, –.
———. (): “The Hulailan Survey: A Note on the Relationship between Aims and Method”, Akten des
VII. Internationalen Kongresses für iranische Kunst und Archäologie, München, .-. September 
(AMI. Ergänzungsband ), Berlin, –.
Negahban, E. (): “Preliminary report of Qazvin Expedition: Excavation of Zaghe, Qabrestan, Sagz-
abad, –”, Marlik , –.
 m. hessari and r. yousefi zoshk

Nicholas, I.M. (): The Proto-Elamite Settlement at TUV (University Museum Monograph ),
Philadelphia.
Perrot, J. (): “Mission de Susa”, Iran , –.
———. (): “Fouilles françaises à Susa”, Archéologia , –.
Potts, D.T, Lamberg-Karlovsky, C.C. & Pittman, H. (): Excavations at Tepe Yahya, Iran, –:
the third Millennium (Bulletin of the American School of Prehistoric Research ), Cambridge.
Schmandt-Besserat, D. (): “An Archaic Recording System in the Uruk-Jemdet Nasr Period”, AJA ,
–.
———. (): Before Writing, vol. , Austin.
Stein, A.S. (): Archaeological Reconnaissances in North-Western India and South-Eastern Iran, Lon-
don.
Strommenger, E. (): “The Chronological Division of the Archaic Levels of Uruk-Eanna VI to III/II:
Past and Present”, AJA , –.
Sumner, W.M. (): “Excavations at Tall-i Maliyan, –”, Iran , –.
———. (): “Proto-Elamite Civilization in Fars”, U. Finkbeiner & W. Röllig (eds.), Ğamdat Nasr,
Period or Regional Style? Papers given at a symposium held in Tübingen, November  (Beihefte zum
Tübinger Atlas des vorderen Orients. Reihe B: Geisteswissenschaften ), Wiesbaden, –.
Talaee, H. (): Qazvin Plain with  thousand years history, Tehran.
Tosi, M. (): “The Notion of Craft Specialization and its Representation in the Archaeological Record
of Early States in the Turanian Basin”, M. Spriggs (ed.), Marxist Perspectives in Archaeology, Cambridge,
–.
Vaiman, A.A. (): “Über die protosumerische Schrift”, Acta Antiqua Academiae Scientiarum Hungar-
icae , –.
Weiss, H. & Young, C.T. (): “The Merchants of Susa. GodinV and Plateau-Lowland. Relations in the
late Forth Millennium B.C.”, Iran .–.
Whitcomb, D.S. (): The Proto- Elamite Period at Tall-i Ghazir, Iran, Master Thesis, Athens-Georgia
(Unpublished).
Wright, H.T. (): An Early Town on the Deh Luran Plain: Excavations at Tepe Farukhabad (Memoirs
of the Museum of Anthropology ), Ann Arbor.
Wright, H.T. & Johnson, G.A. (): “Population, Exchange, and Early State Formation in Southwestern
Iran”, American Anthropologist , –.
Young, T.C. (): Excavations at Godin Tepe: First Progress Report (Royal Ontario Museum. Division
of Art and Archaeology. Occasional Paper ), Toronto.
———. (): “Godin Tepe Period VI/V and Central Western Iran at the End of the Fourth Millennium”,
U. Finkbeiner and W. Röllig (eds.), Ğamdat Nasr, Period or Regional Style? Papers given at a symposium
held in Tübingen, November  (Beihefte zum Tübinger Atlas des vorderen Orients. Reihe B:
Geisteswissenschaften ), Wiesbaden, –.
 THE TRANS-TIGRIDIAN CORRIDOR
IN THE EARLY THIRD MILLENNIUM BC

Steve Renette*

The Trans-Tigridian Corridor (the area between the Tigris and the Zagros Mountains1) is for
the most part still a blank spot on the archaeological map of the Ancient Near East. As a
result it is often treated as an empty, marginal zone merely serving communication between
the Mesopotamian lowlands and the Iranian highlands. The few pieces of information we have
(which will be discussed in this article) are so limited that they can be interpreted in a variety
of ways, but so far most interpretations have assumed domination by external powers and a
provincial or marginal character of the sites.
The aim of this article is to stir interest in this largely unexplored region by illustrating
its specific local dynamics and developments during the early third millennium bc. I argue
that the Trans-Tigridian Corridor was not necessarily a marginal, empty backwater under
complete cultural, economic, or political domination by Mesopotamian or South-Western
Iranian polities. The people who inhabited the Trans-Tigridian Corridor had unique forms of
social organization and cultural elements, particular to the region, while constant interregional
interaction, so typical for the Ancient Near Eastern world, resulted in mutual influences and
connected developments.
When archaeologists unexpectedly discovered large monumental constructions on small
sites in the Hamrin valley during salvage operations along the Middle Diyala River in the
late s (fig. ), these were almost immediately described as colonies of the famous Lower
Diyala cities (at Khafajah, Tell Asmar, and Tell Agrab). However, a reassessment of the evidence
has shown that alternative interpretations are possible which have wider implications on the
entire Trans-Tigridian Corridor. In a previous article, I have argued that the monumental
constructions (for examples see fig. ) that were found at various sites in the Hamrin valley
could not have been fortresses (as has mostly been argued), but that they were probably
communal storage facilities that were part of the intensive agricultural exploitation of the valley
by mobile communities which occupied a wider region (for the complete discussion, see Renette
 and the references there included). If this was the case, then the Hamrin sites could not
have been unique and isolated, and comparable settlements and material culture should be
found in many other valleys throughout the Trans-Tigridian Corridor.
In this article I will show that a similar material culture, as well as comparable burial customs,
occupational patterns, settlement history and exploitation of the available resources, evidencing
cultural ties or parallels have indeed been found at other locations within the Trans-Tigridian
Corridor: the Deh Luran valley close to Susa, the Pusht-i Kuh in Luristan (especially at the
necropoles of Bani Surmah (Haerinck/Overlaet ) and Kalleh Nissar (Haerinck/Overlaet
)), and several sites on the Adhaim and Lower Zab Rivers. Additionally, limited textual
sources from later in the third millennium can provide some complementing clues to the nature

* University of Pennsylvania, Dept. of Art and Archaeology of the Mediterranean World.

1 Due to geographical, environmental, and ultimately also cultural characteristics, the Lower Diyala region is
included with Mesopotamia proper. This means that the Trans-Tigridian Corridor denominates the hilly landscape
of the low range of the Zagros Piedmont east of the Jebel Hamrin.
 steve renette

of the societies that occupied the Trans-Tigridian Corridor, distinct from the Lower Diyala
region. These pieces of information make it possible to carefully posit a hypothesis about the
societies that inhabited this long-stretched region between the Jebel Hamrin and the Zagros
Mountains. As such, the origins of Zagros tribal confederations and dynasties such as Awan
and Shimashki, which played such an important role in the formation of the Elamite world in
the later third millennium, may be traced back to developments that started centuries earlier
in the Trans-Tigridian Corridor.

. The Trans-Tigridian Corridor

The interpretation of the Hamrin occupation in the early third millennium as intense agri-
cultural exploitation by dispersed mobile communities implies that this was not an isolated
phenomenon, but that the entire Trans-Tigridian Corridor was inhabited by the same or related
groups. The Hamrin valley was only one of the many valleys that were intensively exploited.
This interpretation also implies that there might have been a communal cultural tradition in
lifestyles, architecture and material culture. Unfortunately the Trans-Tigridian Corridor is one
of the least researched areas of the Ancient Near East, and as a result the available information
is extremely limited.
First of all, we would expect similar constructions elsewhere. So far, there are a few sites
north of the Hamrin valley, on the Adhaim River (McDonald/Simpson :  and )
and on the Lower Zab (McDonald/Simpson : , Shakir , Forest in press), where
identical buildings have been uncovered (for their location, see the boxes indicated in Fig. ).
Just like the settlements in the Hamrin valley, these northern sites were also discovered during
the construction of dams on rivers that cut through the Jebel Hamrin (which stretches all the
way from the Tigris, close to Assur, in the north to the edge of the Susiana plain in the south).
Regrettably, little to nothing has been published about these sites. Nonetheless, their discovery
is evidence that an architectural tradition of monumental circular constructions as well as the
associated social organization existed at least as far north as the Lower Zab. It is therefore likely
that this tradition also existed in valleys further east of the Jebel Hamrin, although this needs
to be confirmed by future research in the Trans-Tigridian Corridor.
Much further south, at the southern reaches of the Jebel Hamrin, the Deh Luran valley
provides additional information about the inhabitants of the Trans-Tigridian Corridor. During
the early third millennium this valley underwent a strong expansion of occupation, exactly like
the Hamrin valley. An extensive comparison of the material of the Deh Luran valley with the
material of the Hamrin valley has been made by E. Carter in the s. In sum, she noticed
that both valleys have many general similarities, while showing differences on a more detailed
level (Carter : –). They share similar geographical features in that they are both small
valleys which offer some agricultural potential within this hilly landscape and offer relatively
easy access to higher valleys in the Pusht-i Kuh in Luristan. They have striking parallels in
occupational history and dynamics, with the appearance or substantial growth of many small
villages with an emphasis on agricultural exploitation during the early third millennium and a
subsequent abandonment of most of these sites. In both areas there is a strong Scarlet Ware
tradition, which has distinct local characteristics.2 Finally, in both valleys necropoles were

2
These local characteristics are much better noticeable in the Deh Luran valley considering its greater distance
from the Lower Diyala region. The Hamrin valley borders directly on the Lower Diyala region and logically shows
many more similarities with the Scarlet Ware found there.
 the trans-tigridian corridor in the early third millennium bc 

found with mudbrick-built vaulted tombs and with burial customs and grave goods similar
to those in the Hamrin valley (although very little is known from the Deh Luran necropoles)
(Carter : , Eickhoff : –). However, despite intensive survey work and limited
excavations at Tepe Farukhabad (Hole, Flannery/Neely , Wright , Neely/Wright )
our knowledge of the communities living in this valley during this time is still very incomplete.
Nevertheless, the existing evidence does support my view that the Trans-Tigridian Corridor
was a distinct cultural region in the Ancient Near East of the early third millennium, with
mobile societies exploiting the agricultural potential of the valleys in the otherwise hilly
landscape.
Carter also pointed out the comparisons with the isolated necropole sites in the Pusht-i Kuh,
in Luristan (Carter :  and ). Just as in the Trans-Tigridian Corridor very little Uruk
occupation has been found, while a strong cultural tradition started developing around bc.
However, this cultural tradition has only been attested at grave sites, while no settlements have
been discovered so far. This might indicate that these necropoles were used by pastoral nomads
on their seasonal migratory routes (Haerinck/Overlaet : –). In his analysis of the
graveyards at Ahmad al-Hattu and Kheit Qasim in the Hamrin valley, T. Eickhoff emphasized
strong similarities (although not identical parallels) in burial customs, burial goods, and grave
types with the necropoles at Bani Surmah and Kalleh Nissar in the Pusht-i Kuh (Eickhoff :
–). Moreover, there was a long lasting Scarlet Ware tradition in the Pusht-i Kuh, albeit
most vessels were either imports or local imitations of a lower quality (Haerinck/Overlaet
: ). These similarities in material culture and customs indicate close interaction with
communities in the Trans-Tigridian Corridor. It is even possible that the necropoles in the
Pusht-i Kuh were used by families who lived in the lower valleys and who took their flocks
higher up during summer. In any case there was an interaction network that extended from the
Jebel Hamrin into the Pusht-i Kuh during the early third millennium. This network was most
likely sustained by the high degree of mobility that existed in the Trans-Tigridian Corridor, of
which the isolated nature of the Luristan necropoles are a clear indication.
In addition, several later textual sources offer us an indirect source of information for
the Trans-Tigridian Corridor in the early third millennium. The archives discovered at Tell
Shemshara far upstream on the Lower Zab constitute the most important source of information
(Eidem/Laessoe ). In later parts of the third millennium, according to the texts, several
“cities” (Der, Hamazi, Gassur …) and “kingdoms” (Turukkum, Lullubum, Gutium, Awan …)
existed in the Trans-Tigridian Corridor, although it seems more likely that they refer to central
towns of various sizes and tribal coalitions or chiefdoms. Because the textual sources mainly
deal with specific events and political contexts, it is hard to derive a general view on this region
based on them. They speak of nomadic threats, as well as cities and towns which are mentioned
in association with trade and military conquests. It seems that the Trans-Tigridian Corridor was
a politically fragmented landscape in the later third millennium with many local potentates and
various tribes, centred on cities and towns which were the focal points for communication and
trade (see also Steinkeller ).

. Tentative Interpretations

Without merely projecting the picture we derived from the textual sources back into the early
third millennium, it does seem obvious that the Trans-Tigridian Corridor has always been a
tribal landscape with a dimorphic structure (mobile elements, and towns and cities). However,
there is no need to invoke a dichotomy between sedentary villagers and city-dwellers on the
one hand, and pastoral nomads on the other hand, as scholars of the Ancient Near East so often
 steve renette

do. In the Trans-Tigridian Corridor, in all probability no difference was made between those
who moved around and those who stayed behind in the villages, while there was most likely a
significant degree of flexibility and mobility. The main economic activity was goat and sheep
herding for which the hilly landscape was perfectly suited. To make full use of the resources
available in the region, pastoralism was supplemented by intensive agricultural exploitation
of riverine valleys by members of the same society. This lifestyle probably existed throughout
the Trans-Tigridian Corridor and the continuous interaction, stimulated by the high degree of
mobility, resulted in shared customs and similar mobility and settlement patterns. This does
not necessarily mean however that the entire region was inhabited by the same homogeneous
groups, culturally or ethnically.
Interregional contact and trade networks caused several wide contact spheres to emerge
during the early third millennium bc. Central Mesopotamia, and especially the Lower Diyala
region, had close ties with the tribal societies of the Trans-Tigridian Corridor, at least partly
as a result of the trade that extended from the Iranian plateau all the way to western Syria (as
evidenced by the Piedmont style or glazed steatite style seals), and because of the geographical
proximity of the two regions. The exact nature of these ties and connections still eludes us and
many forms of interaction, cooperation or even integration of the Lower Diyala region and the
tribal landscape can be imagined through ethnic affiliations, trade connections, or social and
political ties. Whatever the case may have been, it is clear that Scarlet Ware is a common feature
both in Central Mesopotamia and in the Trans-Tigridian Corridor and that it might have
communicated identity and the participation within a far reaching network through its painted
designs and use in specific social situations, similar to the network in Northern Mesopotamia
with Nineveh V ceramics (see also Emberling  for the interpretation of Scarlet Ware as an
ethnic marker). It is of course this contact and trade network that was the source of the relative
wealth that we found in the Hamrin valley in the form of beads, shells, seals, and especially
metal artefacts.
A couple of centuries into the third millennium (during Early Dynastic II–III) things
started to change. After a period of relative isolation and focus on the Persian Gulf region,
South Mesopotamian city states shifted their attention back to the northern and eastern trade
routes. The first mentions of Elam not accidentally appear in this time and context. It seems
furthermore that this was accompanied by military endeavours and attempts to seize control
as evidenced by royal inscriptions talking of conquest and Sumerian mythological stories
(especially about the land of Aratta). The Trans-Tigridian Corridor obviously was in the midst
of these events and undoubtedly the region must have felt the consequences of military and
political pressure.
However, there are also internal developments that could have played a role. In the Deh Luran
valley we know that the site of Musiyan took on a dominant role and grew in size, probably
incorporating some of the population and becoming a true centre in the valley (Neely/Wright
: fig. V.). This process of contraction of the population into larger sites might also have
occurred throughout the entire region where larger agglomerations existed and tribal elites
may have created centres dominating a larger area. In the Hamrin valley it is possible, but not
known for certain, that the larger sites of Tell Yelkhi, Tell Abqa, and/or Tell Suleimeh were
inhabited throughout the entire third millennium and grew in size after Early Dynastic I–
II. They certainly were local centres during the Akkadian empire.
Another possibility is that some groups migrated to the Lower Diyala cities which underwent
significant expansion and must have exercised some attraction on a wider region. Perhaps the
tribal elites took residence in these cities, taking part in the local politics.
Finally, also the emergence of polities such as Awan and Shimashki might be the result
from continuous political developments and contractions of populations and tribal groups
 the trans-tigridian corridor in the early third millennium bc 

into stronger polities, competing on the interregional level, ultimately having their roots in the
cultural and social developments that took place in the Trans-Tigridian Corridor during the
early third millennium. Tribal confederations, chiefdoms and even states do not originate out
of nothing. They are the result of pre-existing social complexity and increasingly elaborate social
organization (Khoury/Kostiner ). External military and political pressure from South
Mesopotamian city states, as well as an increase in competition in interregional trade, could
have accelerated and amplified socio-political processes (for an overview of the discussion
about the origins of Awan and Shimashki, I refer to Potts : –).

. Prospects of Future Research

The main obstacle in understanding the Trans-Tigridian societies is the lack of research and
data. Basic fieldwork can answer many questions and reveal fascinating and perhaps unexpected
information. With the current state of knowledge it is impossible to provide solid explanations
and there is a lack of a comprehensive framework to understand and interpret correctly any
new information that may come to us. New research in this region, although momentarily
still extremely difficult or even impossible given the current political situation, can drastically
improve this situation, instantly prove or refute hypotheses, and change the way we think about
the role of this region and the people who lived there.
The first issue that needs to be resolved is whether the occupational pattern, monumental
constructions, and necropoles occur further east from the Jebel Hamrin as well. The specific
nature of the occupation in the Trans-Tigridian Corridor during the early third millennium
needs to be analyzed (agricultural exploitation, control of trade routes, military outposts?)
and we need to establish whether there are architectural and cultural traditions specific to
the region that inspired the monumental constructions in the Hamrin valley. In this article I
have presented a hypothesis based on the limited information that is available to us, but more
research is needed to provide solid evidence.
Secondly, an interesting field of study might focus on the distribution pattern, the diversity,
and the chronology of Scarlet Ware. Groundbreaking research has been undertaken on Nin-
eveh V ceramics and their distribution in Northern Mesopotamia. It would only make sense
to have similar research and debates on the Scarlet Ware network. Especially the contact zone
between these two ceramic styles could provide interesting information. It is sometimes still
assumed that Scarlet Ware and Nineveh V can be associated with distinct ethnic groups. It is
in the contact zone, presumably around the Lower Zab River, that this can be examined in
more detail. Most likely ethnicity is just one of the various factors, and not necessarily the most
important one.
Finally, a quick look at satellite images available on the internet already reveals how many
sites are spread over this largely unexplored region. The rescue excavations in the Ham-
rin valley alone have already delivered many surprises and important textual sources (for
the late third millennium and the Old Babylonian period), just as the important archives
at Tell Shemshara and Nuzi/Gassur. It can only be expected that much more can be
discovered.

Bibliography

Carter, E. (): “The Piedmont and the Pusht-i Kuh in the Early Third Millennium BC.” In: Huot, J.-L.
(Ed.) Préhistoire de la Mésopotamie. La Mésopotamie préhistorique et l’exploration récente du Djebel
Hamrin. Paris, -- décembre , Paris: –.
 steve renette

Eickhoff, T. (): Grab und Beigabe: Bestattungssitten der Nekropole von Tall Ahmad al-Hattu und
anderer frühdynastischer Begräbnisstätten im südlichen Mesopotamien und in Luristan, Münchener
Vorderasiatische Studien , Munich-Vienna.
Eidem, J. and Laessoe, J. (): The Shemshara Archives Vol.  The Letters, Copenhagen.
Emberling, G. (): “Ethnicity in Complex Societies: Archaeological Perspectives,” Journal of Archae-
ological Research /: –.
Forest, J.-D. (): “Some Thoughts about the Scarlet Ware Culture.” In: Miglus, P.A. and Mühl, S. (Eds.)
Between the Cultures. The Central Tigris Region from the rd to the st Millennium BC. Conference at
Heidelberg January nd–th, , Heidelberger Studien zum Alten Orient , Heidelberg: –.
Haerinck, E. and Overlaet, B. (): Bani Surmah. An Early Bronze Age Graveyard in Pusht-i Kuh,
Luristan, Luristan Excavations Documents Vol. VI (= Acta Iranica ), Leuven.
———. (): The Kalleh Nisar Bronze Age Graveyard in Pusht-i Kuh, Luristan, Luristan Excavations
Documents Vol. VII (= Acta Iranica ), Leuven.
Hole, F., Flannery, K.V. and Neely, J.A. (): Prehistory and Human Ecology of the Deh Luran Plain,
Ann Arbor.
McDonald, H. and Simpson, S.J. (): “Recent Excavations in Iraq,” Iraq : –.
Neely, J.A. and Wright, H.T. (): Early Settlement and Irrigation on the Deh Luran Plain, Ann Arbor.
Potts, D.T. (): The Archaeology of Elam, Cambridge.
Renette, S. (): “A Reassessment of the Round Buildings in the Hamrin Valley (Central Iraq) during
the Early Third Millennium BC,” Paléorient /: –.
Shakir, B. (): “Tell an-Nemel,” Sumer : –.
Steinkeller, P. (): “The Historical Background of Urkesh an the Hurrian Beginnings in Northern
Mesopotamia.” In: Buccellati, G. and Kelly-Buccellati, M. (Eds.) Urkesh and the Hurrians. Studies in
Honor of Lloyd Cotsen (= Urkesh/Mozan Studies ), Malibu: –.
Wright, H.T. (Ed.) (): An Early Town on the Deh Luran Plain. Excavations at Tepe Farukhabad,
Ann Arbor.
 the trans-tigridian corridor in the early third millennium bc 

Fig. . Geographical overview of Mesopotamia and the Trans-Tigridian Corridor with

boxes indicating the areas mentioned in the text. (Renette : Fig. ).
 steve renette

Fig. . Partially reconstructed plans of three separate circular buildings and their
immediate surrounding with indication of excavated parts (in grey). (Renette
: Fig. ) (a drawn after Fujii : fig.  and II, : fig. ; b drawn
after Gibson : plate  and plate ; c drawn after Roaf : fig. ).
 ELAM AND ESHNUNNA:
HISTORICAL AND ARCHAEOLOGICAL INTERRELATIONS
DURING THE OLD BABYLONIAN PERIOD

Luca Peyronel*

Analyzing archaeological evidences connected with trade relations and interregional policies
at the time of Eshnunna’s reign, focusing on exchange, in particular with the eastern regions
of the Iranian plateau and of Susiana, it is useful to follow the main “steps” of the reign’s
historical events, highlighting specific “junctions” in which the markedly diplomatic or military
element was found to have influenced the economy connected to trade exchanges, not always
strictly linked to the logics of politic control. The political, cultural and commercial interactions
manifest from the point of view purely of events in episodes of alliances, confrontations and
wars, with an alternation of abrupt front changes, characteristic of the so-called age of Mari.1
The unbalance derived from the Middle-Euphrates city’s archives, from which come the widest
and most detailed historical information of the time, imposes a privileged but at the same
time partial point of view, not always allowing a deep evaluation of the relations between
the other players of the near eastern board during the Early Old Babylonian period.2 This
is the case regarding the relations between Eshnunna and Elam that, with the exception of
the Mari sources, we are able to investigate only through limited and incomplete data.3 The
tentative integration of historical aspects with archaeological evidence thus results even more
complicated, but certainly unrenounceable. In this contribution attention is drawing on some
aspects that reveal historical data coinciding with archaeological record, focusing the interest
mainly on Eshnunna-Elam but also stressing the interaction dynamics between Syria, Central-
Eastern Mesopotamia and Western Iran through some significant archaeological indicators of
contact.4
Generally speaking the interaction processes between Eshnunna and the other political
entities can be better detected along the main directories of territorial expansion, towards North
and West, and along the main commercial routes that penetrated in the Iranian plateau through
the Zagros mountains. In the latter case, the prerogative of strategic control in the access of
goods and raw materials of different nature (tin and lapis lazuli, timber and semi-precious
stones) into the Mesopotamian alluvium, strongly influenced the historical events, peaceful or
not, between Eshnunna and Elam. At the same time, the direct overlapping of areas of influence
in the Northern regions between Tigris and the Zagros chains instead made Shubartu/Assyria
the natural antagonist of Eshnunna, especially at the time of Shamshi-Addu I, when also
Mari and the Middle Euphrates were under the political domain of the Northern Mesopo-
tamian kingdom.

* Università IULM di Milano (Italy).

1 See Liverani : –, Wu Yuhong , Charpin : –.
2 Lafont , Charpin/Ziegler .
3 See e.g. Saporetti : –, Charpin : –, in which diverging sequence of Eshnunna’s rulers are

present; the Elamite history between the Dynasty of Shimashki and the Sukkalmah period (Paléo-élamite II–III)
is outlined in Carter/Stolper : –, tabs. –; Vallat a (with a specific attention to the Mesopotamian
relations) and recently by Steve et al. , with some differences and open questions; see also Potts : tabs. .,
., Mofidi-Nasrabadi : –.
4 For a more detailed analysis of the pattern of archaeological interrelations between Elam, Eshnunna and

Subartu/Assyria during the first two centuries of the nd millennium bc see Peyronel .
 luca peyronel

The Eshnunna-Elam relations can be followed during the reign’s chronological lifespan, al-
though it is possible to detect two phases in which the political features are more emphasized
with the consequent cultural and socio-economic relapses. It is striking that these phases
roughly correspond with the beginning and the final years of the kingdom: the first can be traced
at the beginning of the territorial reorganization with the consequent independence of the
Diyala region, after the disruption of the Neosumerian power,5 and the second occurred after
Eshnunna’s peak of power under Ipiq-Adad II, Dadusha and the first part of the reign of Ibal-pi-
El II, when Eshnunna was conquered by Elam, corresponding to the general Elamite expansion
in Mesopotamia after the death of Shamshi-Addu, immediately before the Hammurabi’s seizure
of power (bc).6

At the time of Ibbi-Sîn of Ur, the Eshnunna governor, Shu-iliya, son of Ituriya, does not
refer to himself as ensi of Ur but “beloved of Tishpak” or “son of Tishpak”. We know his
seal from two impressions from the so-called Ilushu-iliya-Nurakhum palace at Tell Asmar.7
The seal (Fig. ) shows a representation of the governor in front of the god Tishpak: the god
holds in one hand the rod-and-ring and in the other a fenestrated axe laying on his shoulder,
standing upon two crouching figures of enemies held with a rope fastened to their nose;
Shu-iliya too holds a battle-axe and the meaning of the scene is clearly at the same time the
commemoration of royal power through the victory over the enemies and the transmission of
this power from the god to the ruler. This is a very interesting iconography, unusual for a seal,
but rooted in the previous periods and modified according to a new propaganda that can be
observed in some victory stela and monumental rock-reliefs. A meaningful link can be traced
looking at the Annubanini relief at Sar-i-Pol, where the king represented his victory adopting
an iconographic model replicated through the centuries on the same rock wall, but that had
been introduced for the first time by Naram-Sin of Akkad to celebrate his victory over the
Lullubites.8
Nurakhum, Shu-iliya’s successor, is again “beloved of Tishpak” and “ensi of Eshnunna” and
probably reigned during the years in which Ur was conquered and destroyed by Elam and Ishbi-
Erra of Isin tried to obtain the role of direct heir of Ur through his policy of military campaigns
against Elam to the East and against Martu to the West.9 Nurakhum was succeeded by Kirikiri,
probably his brother. He has a non-Semitic name, related to the Elamites and he called himself
“ensi of Eshnunna” on behalf of Tishpak.10 Kirikiri was then followed by his own son, Bilalama,
on the throne of Eshnunna, as indicated by the legend of the seal of Bilalama.11 This seal (Fig. )
recovered from a dealer in Baghdad during the Oriental Institute excavations at Tell Asmar

5 From bc until the end of the XX cent. bc, following the traditional Middle Chronology, used by Potts 

and Charpin , or from bc until the half of the XIX cent. bc, according to the low chronology proposed by
Gasche et al. , Gasche , and accepted e.g. by M.-J. Steve and F. Vallat (Steve et al. , Vallat ); The
problem of the absolute chronology during the nd millennium bc and the syncronization between Anatolia, Syria-
Mesopotamia and Egypt has been recently discussed by several scholars without reaching a consensus, revealing
the difficulties to choose between middle, low and ultra-low chronologies: see now Pruzsinszky , with updated
bibliography.
6 See Charpin , , Charpin/Durand , Durand .
7 Frankfort et al. :  fig. , Frankfort : –, pl.  n. , Frayne : .
8 See now Braun-Holzinger : –, –, AB – for Sar-i Pol and AB  for the relief of Iddin-Sin

of Simurrum.
9 Van Dijk , Vanstiphout –.
10 According to Saporetti, Nurakhum and Kirikiri could be considered related and sons of Shu-iliya and a

Shimashkian princess: Saporetti : –. However, beside the possible Elamite origin of their names, this
hypothesis is based only on the controverse interpretation of a Nurakhum dating formula (Saporetti : , C).
11 Frankfort et al. : , –, fig. , : fig. e, Frankfort : n. , Frayne : –.
 elam and eshnunna: historical and archaeological interrelations 

Fig. .

Fig. .

was probably stolen from a public building located over the Shu-Sin temple at the site. Three
sealings bearing multiple impressions of the same seal were also found in the Shu-Sin temple
and the palace of the rulers. However, the inscription was recut and the owner of the original
seal could have been Nurakhum, as recently argued by C. Reichel, who also doubts the family
relationship between Nurakhum and Kirikiri.12 The presentation scene shows the seated god
Tishpak with heads of snake-dragons (perhaps an addition at the time of the recutting of the
legend) emerging from the shoulder, stretching his hand towards a bald male figure with a
suppliant goddess at his back.

12 Reichel : –.

 luca peyronel

It is certain that Bilalama forms, through an inter-dynastic marriage, a stable alliance with
Elam, giving his daughter Mê-Kubi in marriage to Tan-Ruhuratir, eighth ruler of the Shimashki
dynasty.13
The queen’s activities at Susa are testified by some fragmentary inscribed bricks commemo-
rating the building of the Inanna temple at Susa by Mê-Kūbi or Tan-Ruhuratir.14 A cretula with
an impression of a seal of one of the queen’s servants also bears a chronological value since it
was found by Ghirshman in a stratified context of Ville Royale, Chantier B, Level B/VI.15 It is
interesting to note that the standard presentation scene of a male figure, introduced by a suppli-
ant goddess to a seated god holding the ring-and-staff, reveals a Mesopotamian origin, related
to a Diyala workshop or a manufacture by an Eshnunna seal-cutter at Susa. This is more evident
in the comparison with other seals of Elamite officials during the Shimashki period which show
‘anomalous’ presentation scenes, modified according the elamite ideology, as for example the
well-known seal of Kuk-Simut, scribe of Idaddu II, son of Tan-Ruhurater.16
These strong political and cultural interrelations probably contributed to the elaboration of
the iconography of Tishpak, the tutelary god of Eshnunna. This warrior god and his symbol,
the snake-dragon mušhuššu appeared in the Diyala only at the end of the rd millennium bc.17
Tishpak, therefore, is not a deity native of central Mesopotamia, and his son Nanšak bears a
non-Akkadian name. Even though its derivation from Ninazu seems to be probable, we must
recall that in Elamite religious milieu the representation of the god Inshushinak could have
been that of a deity (well-known from seals and sculptures) seated on a throne shaped as a
snake-dragon enveloped in spires, rooted in the earlier hybrid form of a snake-god attested in
the Akkadian glyptic.18

The glyptic material found in the Diyala/Hamrin region showing a direct influence from Iran
surely reveals a general cultural permeability of the Eshnunna kingdom. On the other hand
the strong Mesopotamian imprinting of the Susa sphragistic, self-evident in the more than one
hundred seals related to the Old Babylonian style, has often forced the attention on the general
reception and variation of Mesopotamian iconographic models in Susiana.19 However, only in
rare cases inscriptions or archaeological contexts allow a precise historical location, linking the
evidence to a specific phase of political relations between Mesopotamia and Elam. In fact the
exchange system and the wide web of cultural interactions overlap the historical developments
and are in part independent from the histoire evenementialle, following dynamics that are not
clear-cut, but chronologically elastic and not easily defined in the processual mechanisms.
It is equally clear that in specific ‘moments’ is possible to see the strong link between the
historical and cultural spheres as already brought forward for the emblematic case of the
daughter of Bilalama.

13 The policy of reconciliation of Susa with some Mesopotamian cities seems to be indicated also by an earlier

wedding between the daughter of Iddin-Dagan of Isin and Imazu, crown prince at the time of Kindattu and known
from a seal legend as sukkal of Anshan although not present in the Elamite royal list: Vallat b.
14 Scheil : , pl.  n. ; : –, see Frayne : – and Malbran-Labat : –. The latter

identifies two separate inscriptions attributed to Mê-Kūbi and Tan-Ruhuratir; see also Potts : – nn. –
.
15 Amiet : –,  n. , pls. :, :, Frayne : –.
16 Lambert , Seidl .
17 Wiggermann  and Lambert  on the dragon.
18 Miroschedji , Trokay .
19 A simplistic view of the Old Elamite glyptic as an adaptation of the canonic sphragistic production of

Mesopotamia was criticized by Amiet (: – and ) and recently the necessity to build up chronological
framework of the Susa seals, between the end of the Shimaski dinasty and the period of the Sukkalmakh has been
stressed by Ascalone ().
 elam and eshnunna: historical and archaeological interrelations 

The Hamrin region, a “frontier” region, last eastern stretch of the Diyala valley before
the Zagros, must be considered a critic area for cultural interactions, other than being a
natural commercial outpost of Eshnunna, and consequently a point of passage of the cara-
van route coming from the Iranian plateau. In this respect the glyptic evidence shed light
on these links, from the Akkadian period, when cylinder seals with Indus Valley icono-
graphic influences are attested (Tell Suleimeh)20 to the late Old Babylonian phase, as sug-
gested by seal impressions from Tell Yelkhi showing a clear Suso-Elamite iconographic deriva-
tion, such as evident in the sealing with a representation of a god with a high rounded tiara
with asymmetric horns is in front of an offerer with a peculiar old Elamite crested head-
dress.21
A local production of terracotta seals, known in two distinct groups, one characterized by
cultic or presentation scenes, related to stone productions of higher quality, and another one
with schematic figures, symbols and geometric motifs, is attested frequently in the Diyala sites
during the Old Babylonian period.22 The more schematic seals (Fig. ), characterized by figures
rendered with a rectangular body and arms square-bent or curved in semicircles, standing
or seated, associated with plant elements, symbols or schematic animals, dominant in the
Hamrin, show a strong link with Susiana, where a similar production is traceable (Fig. ).
The circulation of specimens westwards, most likely driven from the Diyala through Suhûm,
is demonstrated by sealings and by an original terracotta example found at Tell Bi"a/Tuttul
(Fig. ).23
Although identifying with certainty the place of origin and primary elaboration of this
class appears difficult, it could be related to post-Akkadian Iranian piedmont productions in
frit or clay, later developed by Eshnunna and Susa workshops sharing a common pattern of
iconographies.
On the contrary the seals with presentation scenes are documented only in the lower Diyala
valley and in some Southern Mesopotamian sites, lacking at Susa, where the common non-
Babylonian style is represented by the Anshanite seals in bituminous rock. However, the
diffusion of Ashanite cylinders doesn’t reach beyond the local Iranian sphere: it lacks in the
Diyala and is generally extremely rare in all Mesopotamia. From a historical-cultural point of
view, the bituminous rock seals thus appear surely referable to the unified culture of Anshan and
Susa during the final period of Shimashki and of the sukkalmahs, spread locally together with
specimens of a more refined Elamite style, of Old-Babylonian influence, often characterized by
a cuneiform inscription, while seal legends of Elamite officials in this glyptic class are extremely
rare.24
The use of a bituminous rock or, as initially proposed by Connan, of an artificial bitumen
compound25 characterizes a peculiar South-Western Iranian handicraft production for the
manufacture of small objects, seals and containers, usually shaped with a plastic zoomorphic
decoration, and attested especially at Susa with several items.26 The presence of this kind of
vessel is very rare in Mesopotamia and up-to-now the only sure evidence comes from Uruk
and Ishchali/Neribtum.

20 Peyronel : –, n. ..

21 Boehmer : –, n. , which stresses comparisons from Susa and Surkh Dum.
22 Al-Gailani Werr .
23 Otto : –, M –, taf. :–.
24 Amiet : –, nn. – (série élamite populaire); Connan/Deschesne : –, nn. –
.
25 Connan/Deschesne : –, and the changed view in Deschesne .
26 Kantor ; Connan/Deschesne : –.
 luca peyronel

Fig. .

Fig. . Susa.
 elam and eshnunna: historical and archaeological interrelations 

Fig. . Tell Bi"a—Tuttul.

A ram protome bowl fragment with only a part of the leg comes from the Sinkashid Palace at
Uruk (Fig. ):27 it has been found in locus a, a small square room opening to the distributive
central court  of sector C, in which several cuneiform documents suggest some kind of
administrative activities.28
A large portion of a vessel, the so-called ‘mouflon bowl’, was found in the Temple of Ishtar
kittitum at Neribtum (Fig. ).29 It has a plastic decoration with three recumbent rams, whose
heads and necks are protruding from the vessel body and turned at a right angle. It comes
from a room at the back of the main cella of the temple (room -Q.), a kind of ‘sacresty’ in
which precious objects and ritual offerings were stored, and it is related to the last architectural
phases (IV) of the temple dated to the th–th centuries bc.30 The vessel must considered
therefore an object imported from Susa and the cultic context point at a high ideological value
perhaps related to a system of reciprocal gift exchange between the highest social hierarchies
of the Eshnunna and Elam kingdoms. The chronological attribution cannot completely rule
out the alternative hypothesis of an object dedicated in the temple by an Elamite ruler himself,
at the time of the sukkalmahat Diyala political control.

27 Lindemeyer/Martin :  n. A, taf.  n. A.

28 Margueron : –, esp. , –, , .
29 Frankfort : , figs. –; Hill et al. : –, pls. –.
30 Hill et al. : , –.
 luca peyronel

Fig. .

From the same floor level a large amount of materials was collected. In particular  cylinder
seals can be ascribed to a wide range of different periods, from Jemdet Nasr to the Isin-Larsa and
the heterogeneous group was surely kept in the temple for a symbolic and ritual reason maybe
related to earlier votive deposits. Another finding coming from the preceding floor level of the
room (th century bc) could be considered in the light of Elam-Eshnunna relations. It is an
Early Dilmun stamp seal showing four hatched squares alternated with crescents and schematic
birds, placed around a central sun disc. Other three stamp seals can be associated to the Dilmun
seal on the floor, and it seems probable that all the seals were collected together as a votive
offering.31 Since several items kept in the sacresty during both phases suggest an eastern link,
the custom of offering precious objects as ritual gifts to the goddess must have had at least in
such cases a foreign origin, specifically Elamite.32 The Dilmun seals, which could be considered
an ex-voto of a seafaring merchant of the Gulf, may in alternative suggest a commercial route
which linked Eshnunna, through Susa, with Failaka and Bahrain, taking also into account the
widely attested Susa-Dilmun relations.33

31 Hill et al. : pl. d, Peyronel , : –, n. ., tav. IV n. ..
32 For a list of the objects found in the room see Hill et al. : –.
33 Amiet , Peyronel –.
 elam and eshnunna: historical and archaeological interrelations 

Fig. .

White-filled Black/Gray Incised Ware surely represent another good archaeological marker of
Mesopotamian-Western Iranian interconnections at the beginning of the nd millennium bc.
Its distribution is wide, from Gasur to Uruk and from Malyan to Godin Tepe, but the higher
concentrations are in the Diyala-Hamrin region, at Girsu and in Susiana. The detailed and
seminal study of Börker-Klähn has shown that a group of middle-sized cylindrical necked jars
with pierced ledge-handles can be manufactured by few production centres if not by the same
workshop.34 All the vessels share the same kind of decoration, combined in a few ways: a series
of metopes alternating geometric motifs (dot-in-circles, dots, triangles), schematic animals
(water-birds, a bird catching a fish, fishes and caprids), vegetal (schematic plant and palms)
(Fig. –). Finally a very specific motif is a boat with multiple standards crowned by crescents
or discs, attested at Susa, Tell Suleimeh, Tello (Fig. ).

34 Börker-Klähn : –; see also Carter : –.

 luca peyronel

Fig. .

Fig. . Tello (not to scale).

elam and eshnunna: historical and archaeological interrelations 

Fig. . Susa.

Fig. .
 luca peyronel

Although a derivation from metallic prototypes has been suggested it seems more plausible
to look at this production as related to the earlier steatite vessels and their imitation in painted or
gray wares. Furthermore, the distributive map points to a circulation along the Zagros foothills
with entrance points in the Mesopotamian alluvium in the Hamrin and in the Lagash region.
Moreover, the presence at Susa of this kind of pottery in rich funerary assemblages, and the
co-occurrences in these graves of metal weapons and vessels, bitumen containers and jewels
suggest its south-western Iranian origin and the direct relation with the Susian élites. Again, it
is difficult to say if the diffusion in Mesopotamia might be explained through a gift-exchange
pattern or also as an element transferred in Mesopotamia during the time of the Elamite
dominion in the Diyala.

After Eshnunna’s peak (under Ipiq-Adad II, Dadusha and first years of Ibal-pi-El II) follows
a phase of political subordination of the Diyala-Hamrin region to Elam. The destruction of
the administrative centre of Tell Harmal and of the Hamrin settlements is probably dating at
that time and we cannot exclude that the conquest was followed by a period of residence at
Eshnunna of the sukkalmah himself.
For the commercial relations between Mesopotamia and Iran a quite radical reorganization
can be registered, showing a tendency to the adaptation of trade circuits to mutated political-
territorial conditions. The Mari archives have, in fact, provided a series of texts regarding tin
supply, obtained directly from Elam through diplomatic relations between Zimri-Lim (years –
) and Siwe-palar-huppak/Sheplarkak of Anshan and Kudu-zulush of Susa: these were two and
a half years of quite intense exchanges, although abruptly interrupted by the aggressive policy
of Elam itself in Upper Mesopotamia that determined the deterioration of the relationship with
Mari, that must have supported together with Babylon the Elamite conquer of the Diyala.35
Tin commerce thus emerges clearly as one of the most important economic activities man-
aged and controlled by Eshnunna, strongly entangled with political strategies and surely related
to the ambivalence of relations both with Assyria northwards and Elam south-westwards. The
interest to exert control of metal trade, intercepting or in any case in part diverting part of
the trade flux that had Assur as a terminal for the tin, must have likely caused at least in part
the belligerence between the two kingdoms. Since it is most likely that the material was collected
in Afghanistan a competition between a “northern” and a “southern” route is not to be excluded,
with an effective division of markets, in a balance that nevertheless always resulted at risk.
The crossing of the two tin routes seems to be confirmed by epigraphic evidence, not
necessarily contradictory, that attest tin traffic from Assur to Larsa and the redistribution of
tin from the Diyala towards Larsa. The overlapping between the lapis lazuli and tin routes,
theorized by several scholars for the rd millennium, in fact appears possible, while it seems
clear that a further southern route was the one through the Gulf, that connected the Indus
valley and Oman to the cities of lower Mesopotamia and South-Western Iran, thanks to the
fundamental mediation role of Dilmun.
The archaeological traces of these exchanges are obviously almost completely “invisible”,
while the typological relations between bronze objects can help in theorizing influences and
cultural contributions pushed by goods circulation.
The analysis of the Hamrin metalworking has shed light on strong typological interrelations
characterizing the metal weapons in the Diyala/Hamrin regions during the first centuries of

35 For a detailed study of this political relations see Durand : –, , Joannès , Michel :

–.
 elam and eshnunna: historical and archaeological interrelations 

the nd millennium bc.36 We cannot identify a predominant and general direct influence from
Iran in the bronze production, although we have to face the problem of the ‘Luristan’ bronzes
without reliable contexts which has confused the issue, as rightly pointed out by several scholars.
The available materials show the use at the same time of several different types of weapons,
and in particular of axes attested in the Syro-Palestinian, Northern Mesopotamian and Iranian
regions. This evidence could be explained with the peculiar role of the area at the crossing of
commercial routes linking the Levant with the East, through the corridor of the Diyala river.
As an appropriate example of this complex pattern of interaction it is useful to briefly discuss
the still unsolved question of the shaft-hole fenestrated axe, notwithstanding the wide and up-
to-dated literature on the topic.37
It is ascertained that two main types of shaft-hole fenestrated axes developed during the MB
period in the Levant with a partial chronological overlapping: the earlier so-called ‘broad type’
and the later duck-bill ‘long type’.38 The several specimens from rich funerary assemblages and
cultic deposits show the ceremonial and symbolic meanings of these weapons, and it is also
quite sure that the broad type derives from the hammered lunate and ‘epsilon’ axes, at the turn
of the rd and nd millennia bc.39 But the direction of the diffusion of the type from West
to East, as is usually accepted, does not explain in a satisfactory way all the archaeological
evidence. We probably have to face a complex pattern of mutual influences which involves
the general higher level of bronze technology in the Iranian area and the direction of raw
materials, tin in primis, from the East. Also in this case the Diyala/Hamrin region is a key-
area for the enquiry, having supplied examples of fenestrated axes bridging the gap between
Iran and the Levant. Two specimens from Tell Hassan and Tell Yelkhi attest the presence
of the broad fenestrated axe together with the duck-bill type, and other axes of the broad
type found in Diyala sites show the diffusion of the weapon in the region (Fig. ).40 Only
the broad type seems to be therefore acquired and re-elaborated in Iran through the Diyala
according to specific and particular modes, resulting also in the manufacture of hybrid and
anomalous items.41 According to Tallon, the Susa evidence allows to reconstruct a development
similar to the Levant, with the evolution of the type from the “epsilon” and “anchor” axes, and
an interconnection between the western and the eastern typological sequences seems to be
probable (Fig. ).42
The adoption of symbolic-ideological values related both to the divine and royal sphere
of this kind of ceremonial battle-axe is also suggested by the presence of the axe holding by
gods or kings on seals (as the seal of Shu-iliya mentioned above) and especially on monu-
mental rock reliefs in the Zagros regions. Thus the appearance of the epsilon and fenestrated
axes as key-elements in the representation of military victory is clearly intersected with the

36 Philip a.
37 See e.g. Philip : –, b, Miron : –, Tubb  with previous bibliography.
38 For the first distinction of the shapes see Maxwell-Hyslop : –, types B–, Hillen .
39 Tubb , Nigro b.
40 Dietre : –, nn. –, fig.  nn. –, fenestrated axes of the broad type are also attested at Tell

Suleimeh, Tell Harmal and Khafaja (Hauptmann/Pernicka : tafn. :, :–) and one earlier and
related type, without shaft-tube, comes from the late Akkadian grave PG  at Ur (Ibid: taf. :).
41 Few specimens can be ascribed to the western Iranian region: two shaft-hole fenestrated axes comes from Susa

(Tallon : nn. –), three from Luristan, one without provenance and one probably from Amlash (Calmeyer
: –). The latter is decorated with two lions resembliong the specimens from Tell Hassan. Some peculiar
“hybrids” and anomalous types suggest the re-elaboration of the shape in the Iranian cultural milieu, such as the
axes with three holes from the Barbier and Bach collections (Tallon : ), or the double-axe from the antiquity
market which combines eastern and the western shapes (Huot ).
42 Tallon : –.
 luca peyronel

Fig. .

Fig. .

development of the iconography of the royal triumph over the enemies in Mesopotamia and its
diffusion also in western Iran, and probably is also linked with the Syrian peculiar meaning of
this kind of weapon.43

43 The ceremonial meaning of the fenestrated axe during the Old Syrian period is suggested especially by the
presence of a figure holding the axe in the ivory talisman of Ebla (Matthiae et al. : n. ) and by the specimens
in precious metals found at Byblos. It has been also suggested that only the broad type had specific ritual and symbolic
values linked to the royal power, whereas the duck-bill type, more frequently found in the so-called “warrior tombs”,
could be considered a purely functional weapon (Philip : ).
 elam and eshnunna: historical and archaeological interrelations 

In this respect some archaeological and iconographic evidences from Ebla dating to the Old
Syrian period can be taken as examples of the difficulties which arise if we adopt a simplistic
view of linear diffusion. Fenestrated axes of broad type have been found in the so-called “Tomb
of the Lord of the Goats” under the Western Palace and bivalve moulds for the manufacture of
the same type of axe were discovered in a smith’s burial located in an area of the Acropolis which
can be considered a craft quarter possibly linked to the Royal Palace.44 The Eblaic specimens
show a ratio : between height and length, circular eyes and a blade with multiple ridges which
cannot be ascribed to the classic Levantine types but rather to eastern variants of the weapon,
where a similar shape and the same kind of blade decoration are attested (Fig. ).45 Moreover,
in the same tomb of the royal necropolis four bronze rattles recently analysed revealing an
unusual high percentage of tin, resemble specimens found at Giyan and at the Barbar temple
on the island of Bahrain (Fig. a).46 Finally, a lapis lazuli eagle pendant has comparisons at
Susa and in Bactria and could be considered therefore a precious object made in the royal
Eblaic workshop under a strong eastern influence (Fig. b).47 Although these evidences can
be considered only clues for a possible “receptivity” of the Old Syrian culture to Eastern
iconographies and typological influences, passing through the Middle Euphrates and central
Mesopotamia, another important piece of evidence from Ebla testifies for a direct link with
Eshnunna.
Two carnelian lion beads or amulets discovered in the favissae of the Ishtar’s sacred area
at Ebla have a counterpart in three other found in the Ishtar kittitum temple at Neribtum
(Fig. a–b):48 they can all be interpreted as ex-voto of Ishtar, and the strict similarity in
execution and material suggest a common origin from stone workshop at Eshnunna.

In the context of Elam-Eshnunna relations, the Eblaic evidence can be considered a “peripheral”
and indirect observatory. Nevertheless, it shed light on the complex and articulated pattern
of relations existing between Syria, central Mesopotamia and the East during the Middle
Bronze Age, which can be in a general way linked to the historical framework of the period
revealed by the Mari archives. What appears from the archaeological data is in fact a scattered
presence of elements showing cultural interactions with Elam, and the route along which these
“contacts” happened passed through the Middle Euphrates and then up the Diyala and Hamrin.
The strategic geo-political importance of the Euphrates “corridor” South of Mari, the Land of
Suhûm, is strongly documented by epigraphic data and it is not a chance that Eshnunna (and
also Elam) tried more than once to conquer that region, which ensured not only the exploitation
of the important bitumen sources but also the control of a key border zone between the Syrian
and Mesopotamian worlds.49

44 Matthiae , Matthiae et al. : nn. , , , Pinnock : –; see also Nigro a.
45 Matthiae : –, Pinnock : . Recently Tubb (: ) argues against the existence of an eastern
metalworking production of fenestrated axes, considering all the Iranian specimens imported from the Levantine
area.
46 Pinnock , : –.
47 Matthiae et al. : n. ; contra Ascalone : –, where a direct provenence of the item from Bactria

is suggested.
48 Ebla: Matthiae et al. : n. , Marchetti/Nigro : , figs. , –; Ishchali: Hill et al. : –,

pl. :a-c (Ish. : and Ish : from the main cella of the temple, Ish. : from a secondary context), where a
wrong function as balance weight is indicated; cf. Peyronel : –.
49 Lacambre . See also Kepinski , in which a peculiar role of the region in relation to the develepment

of the Dilmun culture is proposed.

 luca peyronel

Fig. .

Fig. a.
elam and eshnunna: historical and archaeological interrelations 

Fig. b.

Fig. a–b.
 luca peyronel

Bibliography

Al-Gailani Werr, L. (): “Cylinder Seals Made of Clay,” Iraq : –.
Amiet, P. (): Glyptique Susienne. Des origines à l’époque des Perses Achéménides (= MDP ), Paris.
———. (): “La glyptique du second millénaire en provenance des chantiers A et B de la ville royale
de Susa,” IrAnt : –.
———. (): “Susa and the Dilmun Culture.” In Al-Khalifa, H.A. and Rice, M. (Eds.), Bahrain Through
the Ages, London: –.
Ascalone, E. (): “Archaeological Links between Syria and Iran. Stratigraphical Correlation and
Regional Connections.” In: Matthiae, P. et al. (Eds.), Proceedings of the International Colloquium ‘From
Relative Chronology to Absolute Chronology: The Second Millennium bc in Syria-Palestine’ (Rome th
November—st December ), Rome: –.
———. (): “Sigilli paleoelamiti dei primi secoli del II Millennio a.C. I codici figurativi della glittica
paleoelamita e le nuove formule di espressione dinastica dei sovrani di Simashki.” In Matthiae, P. et al.
(Eds.), Proceedings of the tth International Congress on the Archaeology of the Ancient Near East, May,
th–th , “Sapienza”—Università di Roma. Volume , Harrassowitz Verlag, Wiesbaden: –.
Boehmer, R.M. () “Glyptik aus den Italienischen Ausgrabungen im Hamrin-Gebiet,” Mesopotamia
: –.
Börker-Klähn, J. (): Untersuchungen zur altelamischen Archäologie, Berlin.
Braun-Holzinger, E.A. (): Das Herrscherbild in Mesopotamien und Elam. Spätes . bis frühes . Jt. v.
Chr. (= AOAT ), Münster.
Calmeyer, P. (): Datierbare Bronzen aus Luristan und Kirmanshah, Berlin.
Carter, E. (): “Elamite Exports.” In: Vallat, F. et al. (Eds.) Contribution à l’histoire de l’Iran. Mélanges
offerts à Jean Perrot, Paris: –.
Carter, E. and Stolper, M.W. (): Elam: Surveys of Political History and Archaeology, Berkeley.
Charpin, D. (): “Les Élamites à Šubat-Enlil.” In De Meyer, L. et al. (Eds.), Fragmenta historiae
Aelamicae. Mélanges offerts à M.J. Stève, Paris: –.
———. (): “Šubat-Enlil et le pays d’Apum,” MARI : –.
———. (): “Histoire politique du Proche-Orient amorrite (–).” In Charpin, D. Edzard, D.O.
and Stol, M., Mesopotamien. Die altbabilonische Zeit (= OBO /), Fribourg: –.
Charpin, D. and Durand, J.-M. (): “La suzeraineté de l’empereur (Sukkalmah) d’Elam sur la
Mésopotamie et le ‘nationalisme’ amorrite.” In: De Meyer, L. and Gasche, H. (Eds.), Mésopotamie et
Elam. Actes de la XXXVIéme Rencontre Assyriologique Internationale, Gand, – juillet , Ghent:
–.
Charpin, D. and Ziegler, N. (): Mari et le Proche-Orient à l’époque amorrite: essai d’histoire politique
(= Florilegium Marianum ), Paris.
Connan, J. and Deschesne, O. (): Le bitume à Suse. Collection du Musée du Louvre, Paris.
Cros, G. (): Nouvelles fouilles de Tello, Paris.
———. (): Nouvelles fouilles de Tello, Paris.
Delougaz, P. (): Pottery from the Diyala Region (= OIP ), Chicago.
Deschesne, O. (): “Le mastic de bitume: Un matériau remis en question,” IrAnt : –.
Dietre, C. (): “L’area di Tell Yelkhi: i piccoli oggetti,” Mesopotamia : –.
Durand, J.-M. (): “Fragments rejoints pour une histoire élamite.” In: De Meyer, L. et al. (Eds.),
Fragmenta historiae Aelamicae. Mélanges offerts à M.J. Stève, Paris: –.
———. (): “Commerce de l’étain à Mari,” NABU : .
———. (): “L’empereur d’Elam et ses vassaux.” In: Gasche, H. et al. (Eds.), Cinquante-deux réflexions
sur le Proche-Orient ancien, offertes en hommage à Leon de Meyer, Ghent: –.
Frankfort, H. (): Progress of the Work of the Oriental Institute in Iraq, /. Fifth Preliminary
Report of the Iraq Expedition (= OIC ), Chicago.
———. (): Stratified Cylinder Seals from the Diyala Region (= OIP ), Chicago.
Frankfort, H. et al. (): Tell Asmar and Khafaje. The First Season’s Work in Ashnunna / (= OIC
), Chicago.
———. (): The Gimilsin Temple and the Palace of the Rulers at Tell Asmar (= OIP ), Chicago.
Frayne, D.R. (): The Royal Inscriptions of Mesopotamia Early Periods Volume . Old Babylonian Period
(–BC), Toronto.
———. (): The Royal Inscriptions of Mesopotamia. Early Periods Vol. /. Ur III Period (–bc),
Toronto.
Gasche, H. (): “La fin de la première dinastie de Babylon: une chute difficile,” Akkadica : –.
 elam and eshnunna: historical and archaeological interrelations 

Gasche, H., Armstrong, J.A., Cole, S.W. and Gurzadyan, V.G. (): Dating the Fall of Babylon (= MHEM
), Ghent-Chicago.
Genouillac, H. de (): Fouilles de Telloh. II: époques d’Ur III Dynasties et de Larsa, Paris.
Hauptmann, H. and Pernicka, E. (): Mesopotamian Metal Industry from the Beginnings to the nd
Millenium B.C.. A Catalogue of Analyzed Metal Objects from Iraq and Syria and Results of the X-ray
Fluorescence and Neutron Activation Analyses (= OrA ), Rahden.
Hill, H.D. et al. (): Old Babylonian Public Buildings in the Diyala Region (= OIP ), Chicago.
Hillen, C. (): “A Note on Two Shaft-Hole Axes,” BiOr : –.
Huot, J.-L. (): “Une double hace du Lūristān,” Iran : –.
Joannès, F. (): “Routes et voies de communication dans les archives de Mari,” Amurru : –.
Kantor, H.J. (): “The Elamite Cup from Choga Mish,” Iran : –.
Kepinski, Ch. (): “Mémoires d’Euphrate et d’Arabie, les tombes à tumulus, marqueurs territoriaux
de communautés en voie de sédentarisation.” In Kepinski, Ch., Lecomte, O. and Tenu, A. (Eds.), Studia
Euphratica. Le moyen Euphrate iraquien révélé par les fouilles préventives de Haditha, Paris: –.
Lacambre, D. (): “La région du Suhûm à l’époque des archives de Mari (XIXe–XVIIIe siècles av.J.C.).”
In Kepinski, Ch., Lecomte, O. and Tenu, A. (Eds.), Studia Euphratica. Le moyen Euphrate iraquien
révélé par les fouilles préventives de Haditha, Paris: –.
Lafont, B. (): “Relations internationales, alliances et diplomatie au temps des royaumes amorrites,”
Amurru : –.
Lambert, M. (): “Investitures de functionnaires en Elam,” JA : –.
Lambert, W.G. (): “The History of muš-huš in Ancient Mesopotamia.” In: Borgeaud, P. et al. (Eds.),
L’animal, l’homme, le dieu dans le Proche-Orient ancien. Actes du Colloque de Cartigny, Centre de
d’Étude du Proche-Orient Ancien (CEPOA), Univeristé de Geneve, , Leuven: –.
Lindemeyer, E. and Martin, L. (): Uruk Kleifunde III. Kleinfunde im Vorderasiatischen Museum zu
Berlin: Steingefäße und Asphalt, Farbreste, Fritte, Glas, Holz, Knochen/Elfenbein, Muschel/Permutt/
Schnecke (= AUWE ), Mainz.
Liverani, M. (): Antico Oriente. Storia, società, economia, Roma-Bari.
Malbran-Labat, F. (): Les inscriptios royales de Suse. Briques de l’époque paléo-élamite à l’Empire néo-
élamite, Paris.
Marchetti, N. and Nigro, L. (): “Cultic Activities in the Sacred Area of Ishtar at Ebla during the Old
Syrian Period: The Favissae F. and F.,” JCS : –.
Margueron, J.-C. (): Recherches sur les palais mésopotamiens à l’âge du Bronze, Paris.
Matthiae, P. (): “Sulle asce fenestrate del “Signore dei Capridi”,” Studi Eblaiti : –.
Matthiae, P. et al. (Eds.) (): Ebla. Alle origini della civiltà urbana. Trenta anni di scavi in Siria
dell’Università di Roma “La Sapienza”, Milano.
Maxwell-Hyslop, R. (): “Western Asiatic Shaft-Hole Axes,” Iraq : –.
Mecquenem, R. de (): “Fouilles de Suse (Campagnes –),” RA : –.
———. (): “Fouilles de Suse (–),” MDP : –.
Michel, C. (): “Le commerce dans les textes de Mari,” Amurru : –.
Miron E. (): Axes and Adzes from Canaan. Prahistorische Bronzefunde IX, ., Stuttgart.
Miroschedji, P. de (): “Le dieu élamite au serpent et aux eaux jaillissantes,” IrAnt : –.
Mofidi-Nasrabadi, B. (): Aspekte der Herrschaft und Herrscherdarstellungen in Elam im . Jt. v. Chr.
(= AOAT ), Münster.
Nigro, L. (a): “The Smith and the King of Ebla. Tell el Yahudiyeh Ware, Metallic Wares and the
Ceramic Chronology of Middle Bronze Syria.” In Bietak, M. (Ed.), The Synchronisation of Civilisations
in the Eastern Mediterranean in the Second Millennium bc, Wien: –.
———. (b): “L’ascia fenestrata e il pugnale venato: due tipologie di armi d’apparato dell’età del
Bronzo Medio in Palestina,” Bollettino dei Monumenti, Musei e Gallerie Pontificie : –.
Otto, A. (): Tall Bi #a/Tuttul—IV. Siegel und Siegelabrollungen (= WVDOG ), Saarbrücken.
Peyronel, L. (): “Nota su di un sigillo dilmunita dalla valle del Diyala,” OrEx : –.
———. (): “Sigilli harappani e dilmuniti dalla Mesopotamia e dalla Susiana. Note sul commercio nel
Golfo Arabo-Persico tra III e II Mill. a.C.,” Vicino Oriente : –.
———. (–): “Commercianti dilmuniti nella Susiana? Evidenze Archeologiche ed epigrafiche
sulle relazioni culturali tra Golfo Persico e Iran sud-occidentale agli inizi del II mill. a.C.,” Scienze
dell’Antichità : –.
———. (): “Politica e commercio interregionale in Mesopotamia. Uno sguardo dalla Diyala agli inizi
del II millennio.” In Dolce, R. (Ed.), Omaggio a un Maestro. Studi di Arte e Archeologia del Vicino
Oriente in memoria di Anton Moortgat a trenta anni dalla sua morte, Palermo: –.
 luca peyronel

Philip, G. (): Metal Weapons of the Early and Middle Bronze Ages in Syria-Palestine (= BAR-IS ),
Oxford.
———. (a): “New Light on North Mesopotamia in the Earlier Second Millennium B.C.: Metalwork
from the Hamrin,” Iraq : –.
———. (b): “Warrior Burials in the Ancient Near Eastern Bronze Age: The Evidence from Mesopota-
mia, Western Iran and Syria Palestine.” In Campbell, S. and Green, A. (Eds.), The Archaeology of Death
in the Ancient Near East, Oxford: –.
Pinnock, F. (): “Nota sui “sonagli” della “Tomba del Signore dei Capridi”,” Studi Eblaiti : –.
———. (): “The Relations Between North-Syria and Iran in the Early Second Millennium bc. A
Contribution from Ebla Metalworking.” In Dittmann, R. et al. (Eds.), Variatio Delectat. Iran und
Westen. Gedankschrift für Peter Calmeyer, Münster: –.
Potts, D.T. (): The Archaeology of Elam. Formation and Transformation of an Ancient Iranian State,
Cambridge.
———. (): “Elamite Temple-Building, in From the Foundations tot he Crenellations.” In Boda, M.J.
and Novotny, J. (Eds.), Essays on the Temple Building in the Ancient Near East and Hebrew Bible,
Münster: –.
Pruzsinszky, R. (): Mesopotamian Chronology of the nd Millennium bc. An Introduction to the
Textual Evidence and Related Chronological Issues, Wien.
Quarantelli, E. (Ed.) (): La terra tra i due Fiumi: Venti anni di archeologia italiana in Medio Oriente.
La Mesopotamia dei tesori, Torino.
Reichel, C.D. (): “A Modern Crime and an Ancient Mistery: The Seal of Bilalama.” In Selz, G.J.
(Ed.), Festschrift für Burkhart Kenast zu seinem . Geburstage dargebracht von Freunden, Schülern
und Kollegen, Münster: –.
Saporetti, C. (): Formule dalla Diyāla nel periodo paleobabilonese, . Trascrizioni e commenti, Pisa.
———. (): La rivale di Babilonia. Storia di Ešnunna, un potente regno che sfidò Hammurapi, Roma.
Scheil, V. (): Textes Élamites-Sémitiques (= MDP ), Paris.
Seidl, U. (): “Altelamitische Siegel.” In Vallat, F. et al. (Eds.), Contribution à l’histoire de l’Iran.
Mélanges offerts à Jean Perrot, Paris, –.
Steve, M.-J. (): “Des sceaux-cylindres de Simashki?,” RA : –.
Steve, M.-J. et al. (): “Suse,” Suppl DB : –.
Tallon, F. (): Métallurgie susienne I: De la fondation de Suse au XVIIIe siècle avant J.C., Paris.
Trokay, M. (): “Les origines du dieu élamite au serpent.” In De Meyer, L. and Gasche, H. (Eds.),
Mésopotamie et Elam. Actes de la XXXVIéme Rencontre Assyriologique Internationale, Gand, –
juillet , Ghent: –.
Tubb, J.N. (): “A Crescentic Axehead from Amarna (Syria) and an Examintation of Similar Axeheads
in the Near East,” Iraq : –.
———. (): “The Duckbill Axe as a Cultural and Chronological Indicator.” In Matthiae, P. et al. (Eds.),
Proceedings of the International Colloquium ‘From Relative Chronology to Absolute Chronology: The
Second Millennium bc in Syria-Palestine’ (Rome th November—st December ), Rome: –.
Vallat, F. (a): “L’Elam à l’époque paléo-babylonienne et ses relations avec la Mésopotamie,” Amurru
: –.
———. (b): “Šu-ilišu, Iddin-Dagan et Imazu, roi d’Anšan,” NABU : –.
———. (): “L’Elam du IIe millénaire et la chronologie courte,” Akkadica –: –.
Van Dijk, J. (): “Išbi-Erra, Kindattu, l’homme d’Îlam et la chute de la ville d’Ur,” JCS : –.
Vanstiphout, H.L.J. (–): “The Man from Elam: A Reconsideration of Ishbi-Erra ‘Hymn B’,” JEOL
: –.
Yaseen, G.T. (): Old Babylonian Pottery from the Hamrin, London.
Wiggermann, F.A.M. (): “Tišpak, His Seal, and the Dragon Mušhuššu.” In: Haex, M.C. et al. (Eds.),
To the Euphrates and Beyond. Archaeological Studies in Honour of Maurits N. van Loon, Rotterdam:
–.
Wu Yuhong (): A Political History of Eshnunna, Mari and Assyria during the Early Old Babylonian
Period, Changchun.
 TRANSFERTS CULTURELS DE LA BABYLONIE
VERS SUSE AU MILIEU DU 2E MILLÉNAIRE AV. N. ÈRE

Hermann Gasche*

Dans une étude parue voici  ans, nous avons pu établir que les centres urbains sud-babylo-
niens n’ont pas produits de textes après l’ an  de Samsuiluna ; une vingtaine d’ années plus
tard (Si ), le cœur de la plaine centré sur Nippur et Isin ne donnait plus d’ écrits non plus1.
Dans le dossier archéologique, on note l’ absence de la céramique paléo-babylonienne tardive
sur ces terres méridionales, répertoire maintenant bien connu grâce aux fouilles de Tell ed-Dēr,
Abū Habbah et Tell al-Deylam, ces trois villes se trouvant au nord-ouest de la plaine2. Vers la
fin du˙règne de Samsuiluna, Babylone ne contrôlaient ainsi plus que le territoire teinté en jaune
sur notre fig. ; mais là encore les villes seront progressivement désertées au cours des règnes
d’Ammisaduqa et de Samsuditana.
˙ que de petits sites – villages et hameaux – n’ ont pas été fouillés dans le sud, mais
Il est vrai
les grandes prospections de ces régions n’ ont pas livré le matériel paléo-babylonien tardif 3. Par
ailleurs, il existe des indices relatifs à des transferts de culte du sud vers le nord et à un exil de la
population du sud4. Un troisième constat n’ est pas sans intérêt non plus : comme Samsuiluna,
ses successeurs Abi-ešuh, Ammiditana et Ammisaduqa ont construit des forteresses dont le
nom indique qu’elles devaient protéger les territoires ˙ encore sous le contrôle de Babylone.
Abrité par ses murailles, on pouvait donc « vivre en paix » à Dūr-Abi-ešuh selon une formule
employée par Karel Van Lerberghe et Gabriella Voet (). ˘
On peut formuler l’hypothèse, mais le dossier est encore mince, qu’ une activité réduite
était maintenue dans les sanctuaires des grandes villes entre la fin de l’ occupation paléo-
babylonienne et le début des Cassites. La tombe  du Chantier F de Tell ed-Dēr soutiendrait
cette conjecture: de longues baguettes en ivoire entre les mains du dernier corps déposé et des
plaquettes du même matériau trouvées dans les débris d’ une (ou plusieurs) inhumation précé-
dente5 sont inhabituelles dans le contexte funéraire et suggèrent que cette tombe n’ est pas de
type domestique; elle pourrait donc être celle de personnes au service du temple d’ Annunı̄tum,

* Université de Gand. Cette recherche s’ inscrit dans le cadre du programme Pôles d’attraction interuniversitaires

nº / pour le compte de l’ Etat belge, Services fédéraux des affaires scientifiques, techniques et culturelles. Nous
remercions Steven Cole et Michel Tanret pour leurs précieuses informations, mais gardons pour nous les éventuelles
imprécisions.
1 Gasche , – et plan  ; à la p. , remplacer maintenant «Samsuiluna » par «Ammiditana »

pour Sippar-rābum (BM ).

2 Gasche , – ; Gasche, Pons  (ed-Dēr); Abdulillah Fadhil, Zuhair Rajab Abdallah –; id.

 (Abū Habbah) ; Armstrong  et  (al-Deylam). Le texte paléo-babylonien actuellement le plus récent
˙ Habbah) date de l’ an  de Samsuditana (Çig, Kizilyay, Kraus , ). Récemment, plus de 
de Sippar (Abū
˙
textes ont été trouvés dans les fouilles irakiennes; les dates connues à ce jour de ces documents se situent entre
Abi-ešuh et Ammisaduqa, an  (Abdulillah Fadhil, Zuhair Rajab Abdallah al-Samarraee , –; v. aussi
˘ Fadhil, Zuhair
Abdulillah ˙ Rajab Abdallah –, –). Cette archive provient d’une chambre (!) détruite
par un incendie au cours duquel l’ effondrement de la toiture aurait même tué l’un des occupants (id. , ). Cet
incendie évoque celui de la maison d’ Ur-Utu au Tell ed-Dēr en As , mais il faut peut-être attendre des résultats
plus complets avant de tirer un parallèle entre les deux événements.˙
3 Adams  et  ; Adams, Nissen ; Wright .
4 Voir maintenant Charpin , , –, après van Koppen ,  et Gasche , –.
5 Voir Gasche, Pons ,  et –.
 hermann gasche

Fig. . Image satellite de la plaine alluviale mésopotamienne avec les

villes et réseaux fluviatiles actuels. En jaunâtre, le territoire contrôlé par
Babylone après l’an  de Samsuiluna. D’après Gasche , plan .

divinité tutélaire de la ville. Par ailleurs, la poterie de ce caveau est plus ancienne que celle des
premières installations cassites de Tell ed-Dēr6 et plus récente que celle que nous connaissons
par les fouilles de la maison d’Ur-Utu détruite en As  = 7. Chronologiquement cette
tombe se situe donc dans l’intervalle durant lequel la ville˙ proprement dite n’ était pas occupée,
plus précisément vers le milieu du e siècle, soit quelque  ans après le début de la désurba-
nisation du site.
Une forme de survie des lieux sacrés expliquerait en partie au moins la raison pour laquelle
Kurigalzu, probablement le premier du nom, a trouvé notamment l’ emplacement du temple de
Gula à Isin plus de deux siècles après son abandon, ceci malgré l’ absence d’ une ziggurat qui ne
facilitait pas le repérage du quartier sacré8.

6 Minsaer .
7 Voir en dernier lieu Tanret . Les dates introduites pour le Paléo-babylonien se réfèrent à Gasche et al. a
et b.
8 For Gula, lady of Isin, his lady, Kurigalzu, governor for Enlil, built and restored to its place Egalmah, the ancient

temple, which for a long time had been in ruins (Walker, Wilcke , , traduction du texte IB ). ˘ Dans une
plaine à teneur en sel élevée, une construction de briques crues abandonnée pendant deux siècles pouvait, selon les
circonstances, ne pas laisser beaucoup de traces sur le terrain.
 transferts culturels de la babylonie vers suse au milieu 

Au cours des dernières décennies, un grand nombre de textes sont apparus sur le marché
des antiquités9. Un lot particulièrement intéressant des Cornell Collections est une sélection de
textes paléo-babyloniens tardifs provenant d’ une archive de Dūr-Abi-ešuh ; ils nous apprennent
que … the former inhabitants of Nippur, particularly the clergy of the old ˘Enlil temple, created a
new religious center and built a new Ekur dedicated to Enlil [à Dūr-Abi-ešuh]. These texts also
prove that Nippur itself was not totally abandoned and that some clergy lived˘there at least until
the end of the reign of King Ammisaduqa (– bce [chron. moyenne]). The most recent
text referring to Nippur dates from˙the year Ammisaduqa +d10.
˙
Ces informations, pourtant, ne donnent pas une image très claire de l’ évolution démographique
de la plaine mésopotamienne au cours des cinq derniers règnes de la première dynastie de
Babylone. Le sud s’est dépeuplé et des indices montrent qu’ il y a eu exil vers le nord de la
plaine11 ; mais les terres entre Dilbat et Sippar, et leur prolongement dans la vallée de l’ Euphrate,
pouvait difficilement accueillir une population plusieurs fois plus importante que celle de la
région encore sous le contrôle de Babylone.
C’est lors d’une étude de la poterie et des tombes de Suse que sont apparus quelques
nouveaux éléments; nous allons tenter de les associer à cette dynamique qui a tant modifié
les paysages démographique et politique de la Babylonie en ce milieu du e millénaire av. n. ère.

. Transfert du savoir faire dans le domaine de la poterie12

Au cours du Paléo-babylonien, une technique très particulière a été observée sur les récipients
les plus populaires de l’industrie de la poterie, industrie la plus importante de l’ Antiquité
orientale après la brique crue. Ces récipients, notamment les vases-gobelets (fig. ), étaient
produits rapidement et en très grandes quantités, mais il y avait une faiblesse à hauteur de la
base. En effet, si le potier montait directement la forme définitive, la base et la panse inférieure
auraient été trop épaisses et se seraient déjà fissurées lors du séchage. Afin de pallier cet
inconvénient, le bas du récipient est tourné très large (fig.  : ) pour être resserré ensuite à
la dimension voulue (fig. : ). Après le détachement (fig.  : ) et un pré-séchage, le récipient
est replacé à l’envers sur le tour afin d’ obturer l’ espace vide avec une pâte (plug) fortement
dégraissée avec du végétal, donc plus plastique et moins exposée à la fissuration (fig.  : ). En
même temps, la base est tournassée pour lui donner sa forme définitive.

9 S. Dalley () vient de publier  lettres et documents administratifs de l’époque de Pešgaldarameš et

Ayadaragalama, fils de Gulkišar, dynastes de la première Dynastie de la Mer. Comme souvent, la provenance de ces
textes n’ est pas connue, mais il n’ existe aucun argument décisif pour localiser les sites occupés par ces rois et leurs
successevis dans le sud de la plaine alluviale, entre les anciens réseaux du Tigre et de l’Euphrate; nous reviendrons
ailleurs sur cette épineuse question.
10 Van Lerberghe, Voet ,  ; à la p. , les auteurs donnent un autre texte intéressant qui nous apprend qu’en

Ammiditana , l’ Ekur de Nippur a été attaqué par  cavaliers et cinq jours plus tard par . Voir aussi Gibson
, –, en particulier p.  pour une hypothèse qui validerait une survie dans le quartier sacré.
11 Voir supra n. . A Tell ed-Dēr toujours, la grande demeure du Chantier F (première moitié du e siècle)

présente un plan inhabituel pour la région (Gasche, Pons , plan ); son plus proche parallèle est une maison
un peu plus ancienne de Tello (Parrot , – et fig. ). Voir aussi le sceau-cylindre agadéen avec une
représentation du mythe d’ Etana (Degraeve ) – motif ordinairement attesté dans le sud – trouvé sur le sol de
cette demeure avec de la poterie du e siècle ; Ann Degraeve s’est déjà posé la question si ce cylindre n’appartenait
pas à une de ces familles du sud qui émigra vers le nord avec son mode d’habitation et au moins un objet souvenir
de sa terre natale.
12 Les informations qui suivent ont été réunies dans le cadre d’un travail sur la poterie du e millénaire conduit

avec James A. Armstrong.

 hermann gasche

Fig. . Schéma montrant le tournage des vases-gobelets avec une base

obturée (plugged base). D’après Franken et Kalsbeek , , fig. .

L’attestation actuellement la plus ancienne a été relevée sur un vase-gobelet trouvé à Tello,
examiné au Louvre (fig. : )13 ; nous le datons de la première moitié du e siècle. Sur cet
exemplaire, le resserrement de la base, plus fort que sur celui de notre figure , a laissé une
membrane de terre qui sera percée lors de l’ obturation, détail bien visible à l’ intérieur du
récipient (fig. )14. Un peu plus tard (fig.  : ), la technique est reconnue dans le nord de la plaine
où elle perdure jusqu’au début de l’ époque cassite (fig.  : –), mais elle ne sera plus utilisée
après cette période, seule une autre méthode, tout aussi ancienne et plus rapide, survivra en
Babylonie (fig. : –, voir infra).
Vers le milieu du e siècle, enfin, la base obturée apparaît à Suse sur quatre récipients
trouvés dans deux tombes creusées à partir du niveau XIII du Chantier A de R. Ghirshman15
(v. fig. : ). Selon la documentation connue, la technique a été notée sur des récipients à
panse globulaire (fig. : ), alors qu’ en Babylonie seuls les vases-gobelets sont concernés à
cette époque (fig. : –). Ce n’est qu’ un peu plus tard – dans le niveau XII récent (fig.  : )16 –
que le vase-gobelet susien sera montés avec cette méthode qui sera ensuite utilisée durant tout
le Méso-élamite, aussi bien à Suse et en Susiane.

13 Nous remercions Madame Annie Caubet pour nous avoir permis d’examiner cet objet. D’autres vases-gobelets,
notamment deux exemplaires trouvés à Larsa par André Parrot (, figs.  et ), pourraient avoir le même type
de base, mais ces objets n’ ont pas pu être examiné.
14 La pâte plus dégraissée est bien visible sur la face extérieure de la base.
15 Deux exemplaires identiques du GS- (= fig. :  = Ghirshman, Steve , fig. ) et le GS-

(Ghirshman, Steve , fig. ) proviennent de la tombe  et le GS- (inédit) de la tombe ; sur les dessins
de ces récipients publiés il y a plus de  ans, la différence de pâte dans la base n’est pas indiquée, mais elle a bien été
notée. Pour la stratigraphie et la chronologie du Chantier A, voir fig.  et Steve et al. /, tableau b.
16 Deux exemplaires publiés (Gasche , Pl. : ) et six inédits.
transferts culturels de la babylonie vers suse au milieu 

Fig. . Schéma stratigraphique des bases remplies et obturées en

Babylonie et à Suse entre  et le début de l’epoque cassite.
 hermann gasche

Fig. . Schéma montrant le tournage des vases-gobelets avec une base remplie (filled-in
base). Technique non attestée à Suse. D’après van As et Jacobs , , Fig. .

Ces observations laissent peu de doute ; c’ est durant la période de dépeuplement de la

Babylonie qu’apparaît à Suse la technique de la base obturée qui n’y était pas connue auparavant.
Elle a donc très probablement été introduite par des Babyloniens qui ont choisi cette région
plus orientale et plus sûre contrôlée par les Elamites ; de plus, faut-il le rappeler, on y parlait
leur langue.

Une autre technique contemporaine pour pallier le même défaut de fissuration – la base dite
remplie (filled-in base) – est uniquement mentionnée ici pour compléter le dossier (figs.  et ).
Avant de monter l’objet, le potier introduit une pâte plus dégraissée dans une cavité pratiquée au
centre de la motte centrée sur le tour. Il tourne ensuite le récipient en prenant soin de maintenir
cette pâte dans le bas du futur récipient. Plus rapide et plus efficace, il est surprenant qu’ elle
n’ait pas été reconnue à Suse jusqu’à maintenant.

. Autre indice :
augmentation des tombes dans l’ habitat privé à partir du e siècle

Pour cette enquête, nous retiendrons les cinq niveaux les plus profonds fouillés par R. Ghirsh-
man dans le Chantier A au nord de la Ville Royale ; l’ ensemble date entre environ  et 
 transferts culturels de la babylonie vers suse au milieu 

avant notre ère17. Le sol naturel a été atteint dans deux secteurs du niveau XV, mais des sondages
réalisés en d’autres points indiquent que des installations un peu plus anciennes se cachent
toujours sous une partie de ce niveau18.
Le niveau XV (fig. a), le plus ancien fouillé par Ghirshman, comprend au moins 
demeures19 aux dimensions inégales. Dans le XIV, une grande résidence occupe le centre20 et
une autre le secteur est du Chantier; cette dernière sera reconstruite en XIII, XII et XI ancien.
En XIII, la parcelle de la grande résidence centrale sera occupée par des fours de poteries et
de modestes maisonnettes, mais vers la fin du XII, une nouvelle résidence plus petite – ou une
pinte selon Trümpelmann () – y sera bâtie en même temps qu’ un mur d’ enceinte dont
nous ne savons pas ce qu’il abritait (fig. b).
Dans les niveaux XIV à XI, l’espace entre les grandes résidences est réservé à des habitats
aux dimensions plus petites; c’est sous ces maisons que se trouvent les tombes. On soulignera
donc que les grandes résidences n’en abritaient pas … à une exception près : deux nouveau-nés
furent enterrés sous celle du niveau XIII, mais dans des locaux de service (fig. a)21.
Avertissement : les observations qui suivent ne concernent que les tombes reconnues dans
le Chantier A. Il serait audacieux de vouloir étendre les conclusions à l’ ensemble du site ou
la région, mais les quelque .m2 fouillés dans ce quartier urbanisé pour la première fois
à l’époque des sukkalmah présentent une superficie suffisamment grande pour formuler une
hypothèse. Par ailleurs, on n’entrera pas dans le détail des tombes, toutes les catégories étant
représentées; seule l’absence ou la présence est prise en compte ici.

La distribution des tombes dans ces niveaux donne le résultat suivant :

Niveau XV (ca. –):
 tombes (dont  nouveau-nés dans le même locus) dans  maisons sur un total d’au moins .
Niveau XIV (ca. –):
 tombes dans  maisons sur un total d’au moins  (les deux grandes résidences ne sont pas prises
en compte dans le total).
Niveau XIII (ca. –):
 tombes dans au moins  maisons et deux ruelles sur un total de  maisons souvent incomplètes.
On notera l’exception que forment les deux tombes de nouveau-nés dans des locaux de service de
la grande résidence à l’est du chantier. C’est dans une des tombes du XIII qu’apparaissent pour la
première fois des yeux en terre crue déposés vers le crâne (voir aussi A XII).
Niveau XII (ca. –):
 tombes dans au moins  maisons sur un total d’au moins , souvent incomplètes (les deux
grandes résidences ne sont pas prises en compte dans le total). Comme en XIII, une tombe contient
des yeux en terre crue; par ailleurs, deux tombes sont caractérisées par la présence de huit têtes
(portraits?) en terre crue.
Niveau XI ancien (ca. –):
le nombre de tombes diminue;  ont été reconnues, mais la reconstitution des maisons est délicate
dans ce niveau perturbé.

17 Les niveaux XV à XI ancien (fig. ). Le XI récent est caractérisé par une absence de constructions dans le secteur

jusqu’ au niveau X (vers ).

18 Malgré ce constat, R. Ghirshman avait décidé de transférer les activités dans le Chantier B situé au sud de la

Ville Royale où il mit au jour des constructions du début du e millénaire.

19 Le secteur est du Chantier a été très incomplètement fouillé.
20 Elle couvre sept maisons du niveau XV.
21 Cette absence de tombes suggère de voir dans ces grands bâtiments des résidences de notables proches du

pouvoir, soit des élamites qui enterraient leurs morts ailleurs.

 hermann gasche

Fig. a et b. Plans partiellement reconstitués des niveaux XI à XV du

Chantier A avec, en grisé, les habitations qui abritaient des tombes
sous leurs sols; les chiffres indiquent le nombre de tombes par locus.
transferts culturels de la babylonie vers suse au milieu 
 hermann gasche

Si l’on écarte le niveau XV avec  sépultures seulement, on observe une augmentation de plus
de  du nombre de tombes dans le même type d’ habitat entre le XIV (ca. –) et le
XII (ca. –); durant cette même période le nombre total de maisons n’ augmente que
de . On constate aussi que la période en question commence une bonne dizaine d’ années
après le début du règne de Samsuiluna (–) – époque où il perd le contrôle sur le sud
de la Babylonie – et qu’elle se prolonge bien au-delà de la fin de la ère dynastie. Dans le niveau
XI, le nombre de tombes diminue, mais les bâtiments sont moins bien conservés et la Susiane
est maintenant gouvernée par les Kidinuides dont l’ origine est totalement méconnue.
Ces statistiques ne prouvent pas que l’ augmentation de sépultures est due à une immigration
babylonienne; elles ne le contredisent pas non plus. Les Susiens enterraient déjà sous leurs
maisons avant le e siècle, mais il convient de rappeler que le quartier nord au moins de la
Ville Royale a été urbanisé au plus tôt au cours du e siècle22 et que peu de tombes sont encore
attestées dans le XV. On ne peut donc pas écarter l’ hypothèse d’ une population partiellement
nouvelle; il serait alors étrange qu’il ne s’ agisse pas de Babyloniens d’ autant plus qu’ il y a déjà
l’apport des potiers.

En XIII apparaît également une pratique funéraire qui n’ est pas connue auparavant : deux yeux
en terre crue sont déposés près du crâne de la tombe S.23 ; sur la même parcelle, mais en
XII, une autre tombe (S.)24 contenait également deux yeux25. Dans les deux cas, la calotte
crânienne portait des taches noires de formes irrégulières, peut-être dues à la présence d’ un
masque en matière périssable. Surprenant aussi est l’ apparition, dans deux tombes du XII, de
huit têtes (portraits?) d’hommes et de femmes, grandeur nature, en terre crue polychrome26 ;
une troisième tombe, plus modeste et probablement creusé à partir du début du XI récent
(ca. –) en contenait également une, mais ce témoin plus récent et plus fruste était
modelé autour d’un vase-gobelet qui lui servait d’ armature27. Toutes ces sculptures sont de la
même famille que les très belles têtes découvertes sur le site voisin de Haft Tepe28, probablement
dans un atelier royal de l’époque de Tepti-ahar (vers )29.
Nous ne connaissons pas de comparaisons ˘ pour ces œuvres de terre crue ; les yeux appa-
e
raissent au cours du  siècle et sont encore attestés, en gros, durant la première moitié du
e. Quant aux têtes polychromes, les plus anciennes sont du XII (ca. –) et la plus
récente (XI récent) appartient à la première moitié du e siècle30, date proche des exemples de
Haft Tepe.

22 Pour des installations antérieures au XV, voir le début de cette section.

23 Ghirshman , fig.  (à droite). R. Ghirshman donne toujours comme provenance le niveau dans lequel
la tombe a été découverte. Dans le présent texte, les tombes sont attribuées au niveau à partir duquel elles ont été
creusées ; l’ étude de ce corpus est en cours.
24 Ghirshman , figs. – et  à gauche; Spycket , nº  (ces yeux n’appartiennent pas à une tête en

terre crue comme indiqué à la p. ).

25 A la fig. , les tombes concernées sont indiquées avec un astérisque.
26 Tombe S. : GS-, GS- et  (Ghirshman , figs. –). Tombe S.: GS- (Ghirshman,

Steve , fig.  ; Ghirshman , fig.  ; Spycket , Pl. ), GS- (Ghirshman, Steve , fig. ; Spycket
, Pl.  ; id. , nº ), GS-, GS- et GS-.
27 Tombe S., GS-b.
28 Negahban , –, Pl.  :  et ; du même atelier provient également un masque (p.  et Pl. :

) dont l’ expression du visage est proche de celle de la tête .

29 Voir Cole, De Meyer  pour un très probable synchronisme entre Tepti-ahar et Kadašman-Harbe I.
30 Des têtes en terre crue ou cuite, beaucoup plus petites (h. conservées entre ˘ , et ,cm,˘soit deux fois
moins que nos exemples plus anciens), proviennent de contextes funéraires des anciennes fouilles de Suse; elles
sont attribuées par Amiet (, figs. –) aux e–e siècles av. n. ère. Il est difficile d’établir un lien avec les
têtes plus anciennes, mais il faut rappeler que nous manquons cruellement d’informations archéologiques sur la
période entre le e et le e siècle.
 transferts culturels de la babylonie vers suse au milieu 

Une nouvelle fois un changement est observé à l’ époque qui correspond à la fin de la première
dynastie de Babylone, mais ce changement pose problème. En effet, une quinzaine de ces têtes
seulement proviennent des anciennes fouilles de Suse et de celles plus récentes de Ghirshman ;
c’est peu comparé aux centaines de tombes mises au jour sur le site. Pourtant, le phénomène
est nouveau et il apparaît à une époque politiquement complexe ; il pourrait donc avoir été
introduit par des immigrés, mais la coutume n’ est ni babylonienne, ni attestée ailleurs dans
le dossier documentaire actuellement connu du Proche-Orient. Par ailleurs, les « portraits »
d’homme reproduisent généralement la coiffure projetée en avant sur le front31, coiffure que
d’aucuns considèrent comme caractéristique des Susiens. Concernant cette étrange tradition
funéraire, le débat n’est donc pas clos et c’ est sans doute bien ainsi.

Bibliographie

Abdulillah Fadhil, Zuhair Rajab Abdallah (–): «Sippar – Results of Prospecting. /[th
Season].» Sumer : – (en arabe).
Abdulillah Fadhil, Zuhair Rajab Abdallah al-Samarraee (): «Ausgrabungen in Sippar (Tell Abu
Habbah) – Vorbericht über die Grabungsergebnisse der . Kampagne .» BaM : –.
Adams, R.McC. (): «Settlement and Irrigation Patterns in Ancient Akkad.» In Gibson, McG. (Ed.),
The City and Area of Kish, Coconut Grove: –.
———. (): Heartland of Cities. Surveys of Ancient Settlement and Land Use on the Central Floodplain
of the Euphrates, Chicago-London.
Adams, R.McC., Nissen, H.J. (): The Uruk Country Side. The Natural Setting of Urban Societies,
Chicago-London.
Amiet, P. (): Elam, Auvers-sur-Oise.
Armstrong, J.A. (): «West of Edin: Tell al-Deylam and the Babylonian City of Dilbat.» BiblAr :
–.
———. (): «Late Old Babylonian Pottery from Area B at Tell ed-Deylam (Dilbat).» In Breniquet, C.,
Kepinski, C. (Eds.), Etudes mésopotamiennes. Recueil de textes offert à Jean-Louis Huot, Paris: –.
Charpin, D. (): «Histoire politique du Proche-Orient amorrite (–).» In Attinger, P.,
Sallaberger, W., Wäfler, M. (Eds.), Mesopotamien. Die altbabylonische Zeit. Annäherungen  (= OBO
/), Fribourg-Göttingen: –.
Çig, M., Kizilyay, H., Kraus, F.R. (): Altbabylonische Rechtsurkunden aus Nippur, Istanbul.
Cole, S.W., De Meyer, L. (): «Tepti-ahar, King of Susa, and Kadašman-dkur.gal.» Akkadica : –
. ˘
Dalley, S. (): Babylonian Tablets from the First Sealand Dynasty in the Schøyen Collection (= Cornell
University Studies in Assyriology and Sumerology ), Bethesda.
Degraeve, A. (): «Une représentation du mythe d’Etana sur un sceau-cylindre trouvé à Sippar-
Amnānum.» NAPR : –.
Franken, H.J., Kalsbeek, J. (): «Some Techniques used by the Potters of Tell ed-Dēr.» TD : –.
Gasche, H. (): La poterie élamite du deuxième millénaire a.C. (= VRS I = MDP ), Leiden-Paris.
———. (): La Babylonie au e siècle avant notre ère: approche archéologique, problèmes et perspectives
(= MHEM ), Gent.
Gasche et al. a = Gasche, H., Armstrong, J.A., Cole, S.W., Gurzadyan, V.G. (): Dating the Fall of
Babylon. A Reappraisal of Second-Millennium Chronology (= MHEM ), Ghent-Chicago.
Gasche et al. b = Gasche, H., Armstrong, J.A., Cole, S.W., Gurzadyan, V.G. (): «A Correction
to Dating the Fall of Babylon. A Reappraisal of Second-Millennium Chronology.» Akkadica : –.
Gasche, H., Pons, N. (): «Tell ed-Dēr . Un quartier du e siècle avant notre ère (rapport
préliminaire).» NAPR : –.
Ghirshman, R., Steve, M.-J. (): «Suse. Campagne de l’hiver –. Rapport préliminaire.» ArAs
: –.

31 Ghirshman, Steve , fig. . Amiet , fig. ; voir aussi les figs.  et  à  pour d’autres exemples

de la seconde moitié du e millénaire.

 hermann gasche

Ghirshman, R. (): «Suse au temps des sukkalmah. Campagne de fouilles –. Rapport
préliminaire.» ArAs : –.
———. (): «Suse au début du IIe millénaire avant notre ère et un sanctuaire iranien dans les monts
des Zagros.» CRAIB : –.
Gibson, McG. (): «Patterns of Occupation at Nippur.» In deJong Ellis, M. (Ed.), Nippur at the
Centennial. Papers read at the e Rencontre Assyriologique Internationale. Philadelphia  (=
OPSKA ), Philadelphia: –.
Minsaer, K. (): «II. La poterie du chantier E .» NAPR : –.
Negahban, E.O. (): Excavations at Haft Tepe, Iran (= UMM ), Philadelphia.
Parrot, A. (): Tello. Vingt campagnes de fouilles (–), Paris.
———. (): «Les fouilles de Larsa. Deuxième et troisième campagnes ().» Syria : –.
Spycket, A. (): La statuaire du Proche-Orient ancien (= HdOr, Abt. , Bd. , Abschn. , B – Lief. ),
Leiden-Köln.
———. (): «Funerary Heads.» In Harper, P.O., Aruz, J., Tallon, F. (Eds.), The Royal City of Susa.
Ancient Near Eastern Treasures in the Louvre, New York: –.
Steve et al. / = Steve, M.-J., Vallat, F., Gasche, H., Jullien, C. et F. (/): «Suse.» DB Suppl.
–: –.
Tanret, M. (): «As Years Went by in Sippar-Amnānum …» In Abusch, T., Beaulieu, P.-A., Huehner-
gard, J., Machinist, P., Steinkeller, P. (Eds.) Historiography in the Cuneiform World. Proceedings of the
XLVe Rencontre Assyriologique Internationale. Harvard University, Bethesda: –.
Trümpelmann, L. (): «Eine Kneipe in Susa.» IrAnt : –.
Van As, A., Jacobs, L. (): «Second Millenium b.c. Goblet Bases from Tell ed-Deir – The Relationship
Between Form and Technique.» In Van As, A. (Ed), A Knapsack full of Pottery, Archaeo-Ceramological
Miscellanea dedicated to H.J. Franken on the Occasion of his Seventieth Birthday. July ,  (=
Newsletter ), Leiden: –.
Van Lerberghe, K., Voet, G. (): «Living in Peace at Dur-Abiešuh in the Late Old Babylonian Period.»
communication faite à la e Rencontre Assyriologique Internationale (Münster, .–. Juli ).
———. (): A Late Old Babylonian Temple Archive from Dūr-Abiešuh (= Cornell University Studies in
Assyriology and Sumerology ), Bethesda. ˘
van Koppen, F. (): «Abum-waqar Overseer of the Merchants at Sippar.» NABU : .
Walker, C.B.F., Wilcke, C. (): «Preliminary Report on the Inscriptions, Autumn , Spring ,
Autumn .» In Hrouda, B. (Ed.), Isin-Išān Bahrı̄āt  (= ABAWPh NF ), München: –.
Wright, H.T. (): «The Southern Margins of Sumer. Archaeological Survey of the Area of Eridu and
Ur.» In Adams, R.McC., Heartland of Cities. Surveys of Ancient Settlement and Land Use on the Central
Floodplain of the Euphrates, Chicago-London: –.
 ELAMS KULTURKONTAKTE MIT SEINEN NACHBARN
IM SPIEGEL DER GLYPTIK DES 2. JAHRTAUSENDS V. CHR.

Georg Neumann*

Siegel bzw. Siegelabdrücke sind seit jeher für den Archäologen und Kunsthistoriker von enor-
mer Bedeutung. Man kann mit ihnen datieren, Motive analysieren und daran Sitten, Gebräu-
che, ja teilweise sogar ganze Mythen rekonstruieren und somit eine vergessene Welt zumindest
in Teilen wiederauferstehen lassen. Darüber hinaus bieten sphragistische Erzeugnisse die Mög-
lichkeit, auch komplexe administrative Vorgänge in gewissem Umfang nachzuvollziehen.
Neben diesen Bereichen der Auswertung und Interpretation bieten Siegel und deren Ab-
drücke auch die Chance, Handelsbeziehungen und Kulturkontakte aufzuzeigen. Im folgenden
soll anhand der im elamischen Kerngebiet – worunter hier die Susiana, also das Gebiet des
antiken Elam und die Fars, also der Bereich des alten Anšan, zu verstehen sind – gefundenen
Siegel und Abdrücke des ausgehenden . und . Jahrtausends v. Chr. untersucht werden,
inwieweit sich Beziehungen mit anderen Regionen der damals bekannten Welt nachweisen
lassen.
Das zu analysierende Material stammt vor allem aus den vier im größeren Rahmen ergra-
benen antiken Stätten Susa, Haft Tepe, Tchoga Zanbil und Tall-i Malyan. Für die altelamische
Periode spielen vor allem Susa und Tall-i Malyan eine entscheidende Rolle. In der mittelelami-
schen Zeit gewinnen dagegen Haft Tepe und Tchoga Zanbil zusätzlich stark an Bedeutung.
Aus der Šimaški- und Sukkalmah-Zeit stehen uns bislang mehr als  Siegel und Abrollun-
˘
gen zur Verfügung, wobei jedoch weniger als   davon aus stratifiziertem Kontext stammen.
Diese schicht- bzw. kontextbestimmten Funde seien der Vollständigkeit halber nachfolgend
noch einmal tabellarisch aufgeführt1:

Nr. Fundort
 Susa
 Susa
 Susa
 Susa
 Susa
Fundkontext2 /
Datierung Siegelbild und Stil Siglen und Abb.-Nr.
VR B VI; Tan-Ruhuratir I. stehender Gott, GS  – Abb. 
˘ davor Beter?;
Ur III-/Isin-Larsa-Stil
VR B VI; Mekubi, Frau Fürbittszene; GS  – Abb. 
des Tan-Ruhuratir I. Ur III-/Isin-Larsa-Stil
˘
VR A XIV, Locus ; Ur III-/Isin-Larsa-Stil GS  – Abb. 
Idadu II.
VR A XIV, Locus , Ur III-/Isin-Larsa-Stil, GS  – Abb. 
Spät-Šimaški-Zeit (vgl. Nr. linear
)
VR A XIV, Brandmauer Fürbittszene vor GS  – Abb. 
–; Frühe einem stehenden Gott;
Sukkalmah-Zeit altbabylonischer Stil
˘

* Eberhard Karls Universität Tübingen.

1 Es handelt sich um Siegel und Abrollungen aus Susa, Shemshara, Tepe Giyan, Djamšidi und Tepe Sialk.
2 VR = Ville Royale.
 georg neumann

Fundkontext/
Nr. Fundort Datierung Siegelbild und Stil Siglen und Abb.-Nr.
 Susa VR A XIII, Locus , Beter vor stehendem GS  – Abb. 
Grab ; etwa Mitte der Wettergott;
Sukkalmah-Zeit schematisierender
Stil und
Kugelbohrergebrauch
 Susa VR A XII; Spätsukkalmah Mittelelamisches GS  – Abb. 
bis Frühmittelelamisch ˘ Frittesiegel
 Susa VR A XII, Kuk-Našur Spät-Sukkalmah-Stil, Amiet : Nr.  –
III. (TS.XII.); ˘
Gott auf Schlangenthron Abb. 
Spät-Sukkalmah bis auf Podest mit Ecken
˘
Frühmittelelamisch
 Susa VR A XII, Kidinû; Opfertierträger vor Amiet a: , Nr.
Spät-Sukkalmah bis thronendem Gott?,  – Abb. 
˘
Frühmittelelamisch Frühmittelelamisch
(TS.XII.)
 Susa VR A VI (Parthisch) Tonsiegel, gebrannt; GS bis – Abb. 
linearer Stil, eventuell
altbabylonisch
 Susa VR A XII; Bitumen, geometrischer GS  – Abb. 
Spät-Sukkalmah bis Stil
˘
Frühmittelelamisch
 Susa VR I , Periode V3 Einführungsszene, Carter : Fig. , –
Ur III-/Isin-Larsa-Stil Abb. 
 Susa VR I , Periode V Siegelung, Kapride, Carter : Fig. , –
unklarer Stil Abb. 
 Susa VR I a, Periode V Personen und geflügelte Carter : Fig. , –
Tiere?, schematisierendes Abb. 
Tonsiegel, ungebrannt
 Susa VR I a, Periode V Siegelung, Gott auf Carter : Fig. , –
einem Fellschemel, Abb. 
Ur III-/Isin-Larsa-Stil
 Susa VR I , Periode V Tierkampf, Post-Akkad Carter : Fig. , –
A Abb. 
– Shemshara Šamši-Adad I.4 modellierender Eidem – Læssøe :
altbabylonischer Stil Seal – (Tf. –) –
Abb. –

3
Die Periode V in Susa entspricht der Ur III- bis Isin-Larsa-Zeit.
4
Es handelt sich um mehrere Siegelungen auf Brieffragmenten, die insgesamt sechs verschiedene Siegelbilder
repräsentieren und aus einem Šamši-Adad I. (ca. –) -zeitlichen Archiv stammen; vgl. Læssøe : –
; vgl auch Eidem & Læssøe : Tf. –. Da es sich hierbei um sehr kleine Fragmente handelt, konnten
weder über den Siegelinhaber noch über die Herkunft der Brieffragmente genauere Aussagen gemacht werden.
Ob es sich demnach um Siegel elamischen Ursprungs handelt, ist fraglich, da die Briefe ohne weiteres aus dem
mesopotamischen Raum stammen können.
 elams kulturkontakte mit seinen nachbarn im spiegel der glyptik 

Fundkontext/
Nr. Fundort Datierung Siegelbild und Stil Siglen und Abb.-Nr.
 Tepe Sialk Nekropole A, Grab 5 rein linearer, Ghirshman :
schematisierender Stil S.a – Abb. 
 Tepe Giyan „Construction II“6 kerbschnittartig?, Série Contenau & Ghirshman
Élamite Populaire : Tf. : – Abb. 
nahestehend
 Tepe Djamšidi Grab , Godin III27 linearer, Contenau & Ghirshman
schematisierender : Tf. : –
altbabylonischer Stil Abb. 

Die Feindatierung der Funde insgesamt muss also vor allem auf kunsthistorischen Betrachtun-
gen fußen und durch Vergleiche mit anderen Befunden untermauert werden.
An den Anfang der Untersuchung seien zunächst die altelamischen Siegelfunde aus Susa
gestellt8. Hier fallen vor allem vier Siegel (GS –; Abb. –) auf, die eindeutig als
Import anzusprechen sind. Es handelt sich dabei um Stempelsiegel, die aus grau-weißem Steatit
gefertigt und den sog. Dilmun-Siegeln9 zuzuordnen sind. Die Siegelbilder zeigen die typischen
Motive der zeitgleichen Glyptik des Persischen Golfes: langhornige Kapriden, teils neben
Menschen und architektonischen Elementen abgebildet. Sie datieren aus dem Übergang vom
. zum . Jt. v. Chr.10 Daran lassen sich sieben weitere Stempel anschließen, die in ihrer Form
eindeutig versuchen, die zeitgleichen Siegel aus der Region der Persischen Golfes zu imitieren,
jedoch lokal gefertigt wurden, was durch den Gebrauch von Bitumen als Rohmaterial zusätzlich
nahegelegt wird11. Die Oberseite ist bei diesen Siegeln durch einzelne oder doppelte Linien in
vier Kreissektoren unterteilt. Diese Bereiche weisen auf fünf Siegeln (GS –; Abb. –
) sternförmige Einschnitte auf und scheinen damit das für die Dilmun-Siegel typische „Dot-
in-Circle“-Motiv nachzuahmen. Ein weiteres Siegel (GS ; Abb. ) ist ebenfalls dieser
Gruppe zuzuweisen, trägt jedoch nicht die zuvor beschriebenen Einschnitte. Die Siegelbilder
selbst erinnern allerdings nur entfernt an die Funde aus Bahrain, dem alten Dilmun. Die
abgebildeten Figuren, seien es Menschen oder Tiere, sind in groben Schraffuren ausgeführt und
erinnern eher an die Siegel der sog. Série Élamite Populaire. Auch der Gebrauch des weichen

5 Vgl. Ghirshman : Tf. : S.a (Nekropole A, Grab ). Innerhalb dieses Grabes wurde auch sog. Sagzābād-

Keramik (vgl. ebd. Tf. : S.) gefunden, die jedoch zur Zeit lediglich der ersten Hälfte des . Jt. zugewiesen werden
kann und somit keine näheren Anhaltspunkte in Bezug auf die Feindatierung dieser beiden Siegel liefert; vgl. Piller
–: – und . Zu einem Parallelstück aus Susa vgl. GS .
6 .–. Jh.; die sog. „Construction II“ datiert etwa Godin III -zeitlich; vgl. Dittmann : –, Anm.
2/3
 und .
7 . Jh.; vgl. ebd.
8 Zu den altelamischen Siegeln aus Susa vgl. Amiet ; Börker-Klähn  sowie Delaporte .
9 Zu den Dilmun-Siegeln vgl. Potts a: ; zu Susa und Dilmun vgl. ebd. – sowie Potts : –.

Fundorte ähnlicher Siegel sind z. B. Sar, Failaka, Qalat al"Bahrain.

10 Zur Datierung vgl. auch Buchanan : . Die Tafel, auf der sich die Siegelung befindet, lässt sich auf

Gungunum von Larsa datieren. Eine Tontafel aus Susa ist hier von besonderem Interesse. Sie ist mit einem Dilmun-
Siegel gesiegelt. Bei der Tafel handelt es sich um einen Vertrag über  Minen Silber, die von den Geschwistern
Elamatum, A"abba und Milki-El, Söhne des Tem-Enzag, an einen Ekiba verliehen wurden; vgl. Lambert : –.
Der Gott I/Enzag/k ist der Hauptgott Dilmuns und tritt hier interessanterweise in Verbindung mit dem elamischen
Wortelement tem auf. Zu Susa und Dilmun vgl. Amiet : –. Zu Enzak und Susa vgl. Vallat . Ein
Inschriftenfragment, dass  in Bushire am Persischen Golf gefunden wurde und einen Simut-wartaš, Sohn des
Širuk-tuh, nennt, könnte eventuell den gleichnamigen sukkal meinen und würde damit eventuell auf eine spätere
Kontrolle˘ des Persischen Golf-Raums durch die sukkalmahhū hindeuten; vgl. Potts : .
11 Zur Verwendung von Bitumen vgl. Connan & Deschesne ˘ ˘ .
 georg neumann

Erdölderivats Bitumen als Werkstoff macht deutlich, dass die Siegel aus einer jener Werkstätten
stammen, die sonst Siegel im sog. populären Stil schnitten. Lediglich das zweimal auftretende
Thema des tierbezwingenden Helden hat Parallelen in der Dilmun-Glyptik.
Die vorliegende Siegelgruppe zeigt eindeutig, welche wichtige Rolle der direkte Handel mit
der Region des Persischen Golfes in Susa und damit auch in Elam gespielt hat. Ergänzt wird
dieser Befund durch eine in Susa gefundene Vertragstafel, die in die Zeit des Königs Gungunum
von Larsa datiert, also nach mittlerer Chronologie in das . Jh. v. Chr.12 An die in Rede
stehende Gruppe lässt sich ein weiteres Siegel (GS ; Abb. ) anschließen, das von Amiet
aufgrund seines geometrisierenden Schnitts den Siegeln der Série Élamite Populaire zugeordnet
wurde und eine thronende Person, zwei weitere Figuren, die beide den gleichen Speer in der
Hand halten, und einen Kapriden zeigt. Dieses Siegel entspricht jenen Stücken, die bei den
Ausgrabungen auf Failaka und Bahrain zutage kamen13, und ist mit großer Wahrscheinlichkeit
ein Import von einer dieser Inseln bzw. aus der Region des Persischen Golfes. Ein zweites
Siegel (GS ; Abb. ) aus Susa, das von Amiet als Siegel aus der Mitte des . Jt. v. Chr.
charakterisiert wurde, stammt ebenfalls aus besagter Region.
Ein weiteres aus Bitumen gefertigtes Siegel (GS ; Abb. ) verdient ebenso besondere
Aufmerksamkeit, scheint mit dem doppelten Stempel und dem Flechtband jedoch eher auf den
mittelasiatischen Bereich zu verweisen, wie ein rechteckiges Stempelsiegel aus der Margiana
zeigt14.
Zwei Siegel (GS –; Abb. –) und eine Siegelung (GS ; Abb. ) aus Susa
deuten in Richtung Kappadokien und Syrien und sind mit großer Wahrscheinlichkeit gleich-
falls Importstücke. Das erste der drei Siegelbilder zeigt einen syrischen Gott, eventuell Ba"al,
einen Mann mit konischer Kappe in einem Wulstsaummantel sowie zwei sich gegenüberste-
hende, einen Stab haltende Personen. Die gesamte Szene ist ober- und unterhalb durch ein sog.
Syrisches Flechtband begrenzt15. Das zweite Siegel der Gruppe verweist ebenfalls auf den Wes-
ten Vorderasiens bzw. auf Kappadokien, was an dem flächigen Schnitt, der Frontalgesichtigkeit
der abgebildeten Personen und dem horror-vacui zu erkennen ist. Vergleichbare Siegelbilder
fanden sich in der Schicht Kültepe Ib16. Die Abrollung eines weiteren Siegels auf einem Tür-
oder Gefäßverschluss ist auf Grund der zweiregistrigen Darstellung, der engen Staffelung der
Figuren, des Flechtbandes als Szenentrenner und der oftmals in der nordmesopotamischen
bzw. kappadokischen Glyptik vorkommenden standartentragenden Stiermenschen ebenfalls
in diese Region einzuordnen17.
Die einzige altelamische Siegelgruppe, die m.W. bis heute auch in Fundorten westlich des
Iran entdeckt wurde, ist die sog. Série Élamite Populaire, die mit über  Siegeln in Susa ver-
treten ist und somit die größte Glyptikgruppe stellt. Sie ist gekennzeichnet durch den häufigen
Gebrauch von Bitumen als Siegelmaterial, durch einen oftmals sehr groben und schraffurarti-
gen Schnitt sowie einen sehr einheitlichen Motivschatz. Beliebt sind Einführungsszenen und

12 Vgl. Buchanan .

13 Vgl. Potts a: –; vgl. auch Amiet : –. Diese Siegel wurden von Kjærum : , Nr.
, in die altbabylonische Zeit datiert.
14 Vgl. Sarianidi : . Doppelseitig mit Gravuren versehene Siegelsteine finden sich in Mittelasien sehr

häufig; vgl. z. B. ebd. , .

15 Vgl. Otto .
16 Dies würde eine Datierung auf Hammurapi oder Samsu-iluna wahrscheinlich machen; vgl. Özgüç : .
˘ vgl. auch Parrot : Tf. –,  (),  (). GS  (= Abb. ).
17 Vgl. z. B. Collon : , Nr. ;

Eventuell ist GS  (= Abb. ) auch anatolischen Ursprungs, jedoch ist dies auf Grund des schlechten Erhal-
tungszustandes nicht mehr sicher festzustellen. Amiet charakterisiert diese Siegel als ‚provinziell altbabylonisch‘;
vgl. Amiet : .
 elams kulturkontakte mit seinen nachbarn im spiegel der glyptik 

Tier- bzw. Pflanzendarstellungen. Desweiteren tragen sie oftmals Pseudoinschriften18. Drei Sie-
gel stammen aus Tell Asmar19 (Abb. –), allerdings aus unstratifiziertem Kontext, und zeigen
neben dem typischen schraffurartigen Stil den häufig in der populären Glyptik Elams vorkom-
menden Vogel als Ersatz für einen Opfertisch. Zwei weitere Siegel stammen aus dem südme-
sopotamischen Telloh20 (Abb. –). Während das eine Siegel aus dem Isin-Larsa-zeitlichen
Wohnviertel kommt, ist das zweite Stück21 unstratifiziert und wurde von Parrot (fragend) als
akkadzeitlich charakterisiert. Beide Siegel zeigen die für die Série Élamite Populaire so typi-
schen Pflanzendarstellungen. Ein weiteres Stück ist zwar nur mit Vorsicht als elamisch anzu-
sprechen, soll aber hier der Vollständigkeit halber mit aufgeführt werden. Es wurde im Palast
des Nūr-Adad in Larsa22 (Abb. ) gefunden.
Interessant ist, dass es sich in den vorliegenden Fällen jeweils um Orte handelt, deren weiteres
Fundmaterial aus dem . und . Jt. v. Chr. zum Teil ebenfalls eine Affinität zu Elam aufweist. So
stammen aus dem Diyala-Gebiet z.B. das Fragment eines Bitumengefäßes (aus Ishchali)23, wie
es vor allem zur Šimaški-Zeit in Gebrauch war, und sog. inkrustierte Keramik24. In Telloh, dem
alten Girsu, fanden sich inkrustierte Keramik25 und aus dem . Jt. v. Chr. Fragmente der sog.
Susa II-Keramik26. Auch in Larsa wurden Stücke der für die altelamische Zeit charakteristischen
inkrustierten Keramik ausgegraben27.
Den teilweise engen Kontakt mit Elam verdeutlichen hierbei natürlich auch die keilschriftli-
chen Quellen mit den entsprechenden Informationen. Es sei an dieser Stelle nur exemplarisch
die dynastische Verbindung zwischen Elam und Ešnunna durch die Ehe von Mekubi, der Toch-
ter des Bilalama von Ešnunna, und Tan-Ruhuratir I. genannt28. Interessant ist zugleich, dass ein
altelamisches, der späten Ausprägung der Série ˘ Élamite Populaire zugehöriges Siegel in Nuzi als
Siegel der Winnirke (Abb. ), der Frau des Tehip-Tilla, noch im . Jh. v. Chr. in Gebrauch war.
Das Siegel selbst ist m.E. in den Zeitraum vom ˘ . bis zum . Jh. v. Chr. zu datieren und mag
bereits zu jener Zeit nach Nuzi und in den Besitz der Familie von Tehip-Tilla oder Winnirke
gelangt sein29. ˘

18 Zu dieser Siegelgruppe vgl. Amiet : – sowie Neumann  mit weiterführender Literatur.
19 Vgl. Frankfort : Tf. : (= Abb. ); Tf. : (= Abb. ); Tf. : (= Abb. ). Die Siegel wurden
von Frankfort als Isin-Larsa-zeitlich angesehen.
20 Erstmals bemerkte Amiet den elamischen Charakter eines der beiden Siegel, welches in den er Jahren des

vorigen Jahrhunderts von Parrot im Tell de l’ Est in einem Isin-Larsa-zeitlichen Wohnviertel ausgegraben wurde (vgl.
Parrot : , Nr.  [= Abb. ]); vgl. Huh : ; vgl. auch Amiet : , Anm. .
21 Vgl. Parrot : , Nr.  (= Abb. ).
22 Vgl. ebd. , Nr.  (= Abb. ).
23 Vgl. Börker-Klähn :  ff., Nr. ; vgl. auch Hill, Jacobsen & Delougaz : Pl. –.
24 Vgl. Börker-Klähn :  ff. Nr. Bf–, Cf–, Cf, Cf–, Cf–, Cf, Cf–, Cf–, Cf,

Df–.
25 Vgl. ebd.  ff. Nr. Cf–, Cf–, Cf, Cf, Cf, Cf.
26 Vgl. Nagel : Tf. : und a/b.
27 Vgl. Börker-Klähn :  ff. Nr. Bf, C.
28 Vgl. Röllig –: , Nr. , sowie Edzard –.
29 An das hier besprochene Siegel lassen sich eine Reihe weiterer Stücke anschließen. Diese wurden erstmals von

Seidl :  unter der Bezeichnung Winnirke-Gruppe zusammengefasst. Die Datierung der Gruppe bereitet auch
heute noch große Probleme, da kaum ein Siegel eine Inschrift trägt. Den ersten Versuch einer zeitlichen Einordnung
unternahm Porada : , indem sie das Siegel der Winnirke in die . Hälfte des . Jh., also zeitgleich mit
besagter Person und dem Archiv der Familie von Winnirke datierte und es mit dem Siegel eines Dieners des
Temti-Agun verglich. Amiet vermutete, dass es sich bei dem Siegel der Winnirke um ein älteres Siegel gehandelt
haben muss, da sich auf einer Tontafel sowohl die Abrollung eines série-b-Siegels (zur sog. série b vgl. Amiet :
–; zur Gleichsetzung bzw. Umbenennung in die Winnirke-Gruppe vgl. Seidl : ) als auch die eines
spätaltbabylonischen Siegels befindet (vgl. Amiet : Tf. :.). Er schlug eine Datierung in das ausgehende .
bzw. in das . Jh. vor (vgl. ebd. –). Eine Anfrage von Seidl bei Vallat bezüglich der Datierung einer Tontafel
(GS ), auf der sich zwei Siegelungen befinden, die der Winnirke-Gruppe zuzuordnen sind (das Siegel des
 georg neumann

Will man die Kulturkontakte des Südwestiran mit dem Osten bzw. mit Mittelasien anhand
der Glyptik nachvollziehen, muss man die sog. Anšan-Siegel einer genaueren Betrachtung
unterziehen. Bevor man nun jedoch anhand dieser Siegel Beziehungen rekonstruiert, empfiehlt
es sich, die Gruppe selbst genauer zu untersuchen, da die Bezeichnung Anšan-Siegel bzw.
Anšan-Stil an sich bereits einen Bezug zum iranischen Hochland impliziert und oftmals
schwammig gebraucht wird.
Bereits 30 fielen Amiet zwei Siegel auf, die sich vor allem in einzelnen Bildelementen von
jenen Siegeln abgrenzten, die er als Série Élamite Populaire zusammengefasst hatte. Es handelte
sich um die Siegel GS  (Abb. A) und  (Abb. B). Während bei dem letztgenannten
die thronende Person förmlich mit dem Fellschemel verschmilzt und sich dazu ein schwer zu
definierendes Fabelwesen mit einer Mondstandarte auf dem Rücken gesellt, weist das andere
eine auf einem Podest sitzende Person auf, deren Füße nicht zu sehen sind, und in der Amiet
die Personifikation eines Berges oder eines Bauwerkes vermutete. Die diese Figur umgebenden
Weinreben spielten zu diesem Zeitpunkt für Amiet allerdings noch keine Rolle.
Dies änderte sich erst , als Amiet erstmals einen Anšan-Stil postulierte31, den er an
den in Tall-i Malyan, dem alten Anšan, gefundenen Siegeln aus einer kaftarizeitlichen32 Abfall-
niederlegung (Trash Deposit) festmachte. Es handelt sich hier um Siegel in der Machart der
Série Élamite Populaire, die allerdings eckiger und gröber als der Großteil der Siegel aus Susa
geschnitten sind33. Auffällig für Amiet war auch, dass auf einem der Siegel aus Tall-i Malyan
die menschenköpfige Schlange abgebildet ist. Desweiteren waren zwei Stempelsiegel gefunden
worden, von denen eines eine Kultszene zeigt, in der ein Beter vor einer blockartigen, auf einem
kleinen Podest sitzenden Person, umgeben von Weinreben, steht. Diesem Siegel schloss Amiet,
vor allem auf Grund der Figur und der Weinreben, das Susa-Siegel GS  und weitere aus
dem Kunsthandel sowie aus Susa stammende Siegel an34.
Nur wenige Siegel dieser Gruppe weisen Inschriften auf. Umso bedeutsamer ist es, dass zwei
Siegellegenden altelamische Herrscher erwähnen. Es handelt sich zum einen um ein Siegel

Šamaš-rabi [GS ] und eines Sohnes des Nūr-Šamaš), ergab, dass ein Šamaš-rabi in juristischen Texten aus Susa
belegt ist, die auf Kutir-Nahhunte I. und Kudu-Zuluš II. datieren, und ein Nūr-Šamaš in Texten, die auf Kuk-Našur
˘ ˘ Temti-Halki datieren. Seidl nahm auf Grund der damaligen Chronologiekenntnis –
II. (vgl. Seidl : ) und
˘
sie ging davon aus, dass beide Herrscher zueinander zeitnah regiert haben – eine Datierung in das . Jh. an.
Nach der gegenwärtig gültigen Herrscherabfolge der Sukkalmah-Zeit müsste man nun jedoch eine Laufzeit vom
.–. Jh. vermuten, da Kutir-Nahhunte I. teilweise zeitgleich ˘oder kurz nach Zimri-Lim von Mari regierte, da
er der Nachfolger von Kudu-Zuluš˘I.˘ war, bei dem ein Synchronismus mit dem zuvor genannten Mari-Herrscher
bezeugt ist (vgl. Durand : ). Dies ist für zwei Personen – Šamaš-rabi und den Sohn des Nūr-Šamaš –, deren
Siegel sich hier auf derselben Tontafel befinden, natürlich anzuzweifeln. Was das Aussehen der Siegelbilder der série
b anbelangt, so sind diese sowohl im modellierenden als auch im schematisierenden Stil gearbeitet. Desweiteren
weisen einige Siegel enge Bezüge zu der wahrscheinlich frühaltbabylonisch zu datierenden Série Élamite Populaire-
Gruppe auf (vgl. Neumann ). Demzufolge ist m.E. eine Datierung der sog. Winnirke-Gruppe in die Zeit vom
. bis zum . Jh. auf Grund der zuvor genannten stilistischen Merkmale als wahrscheinlich anzusehen.
30 Vgl. Amiet : .
31 Amiet : –. In den folgenden Jahre veränderte sich die Definition dieser Siegelgruppe, so dass die

Siegel mit Schlangendarstellungen, die nach Amiet auf Grund ihrer Ähnlichkeit mit den in Anšan gefundenen
Siegeln bezüglich ihres groben und schematischen Schnitts als selbständige Gruppe zusammengefasst worden waren,
später keine Erwähnung mehr in den Betrachtungen zum Anšan-Stil fanden. Die neu angeführte ‚Stildefinition‘
bezog sich vor allem auf Antiquaria, wie Personen im Reifrock und rebenartige Pflanzen.
32 Definiert anhand der Keramik von Tall-i Kaftari (Fars). Diese Epoche umfasst etwa den Zeitraum von –

; vgl. Vanden Berghe : .

33 Vgl. GS  (= Abb. A), – (= Abb. E, ,  und B),  (= Abb. B); Sumner : Fig.  (= Abb.

C, –).
34 Vgl. GS  (= Abb. A),  (= Abb. E),  (= Abb. D); Lambert : Pl. V, Nr.  (= Abb. H–I); Amiet

b: , Fig. b (= Abb. J); Sumner : Fig. a (= Abb. ), d (= Abb. ), i (= Abb. C), j (= Abb. );
vgl. dazu Amiet : – und Fig. –.
 elams kulturkontakte mit seinen nachbarn im spiegel der glyptik 

aus dem Gulbenkian Museum35, das einen Ebarat36 notiert, und zum anderen um ein Siegel aus
Susa37, dass Pala-Iššan38 nennt. Dies sind bislang die einzigen Siegel, die genauer datiert werden
können. In den Jahren  und  fügten Stève39 und Porada40 der von Amiet definierten
Gruppe weitere Siegel, vor allem aus dem Kunsthandel stammend, hinzu. Im folgenden seien
die den Anšan-Siegelkomplex charakterisierenden Merkmale noch einmal aufgeführt.
1. Blockartige Frauendarstellungen, die oft als Frauen im Reifrock (Krinoline) bezeichnet
werden; diese Personen stehen zumeist auf einem Podest.
2. Darstellung von rebenartigen Rankpflanzen, manchmal als Nuristanreben41 oder soma/
haoma42 bezeichnet.
3. Darstellung von Personen in Zottengewandung, die oft in den fellüberzogenen Schemel,
auf dem sie sitzen, übergehen.
Die Siegel der Anšan-Gruppe können in das .–. Jh. v. Chr. datiert werden, und zwar auf
Grund der Siegelinschriften und der Vergleiche mit der Série Élamite Populaire43. Die Frage, die
man sich jedoch stellen muss, ist, ob es sich bei allen Funden, die dieser Gruppe zugewiesen
wurden, auch wirklich um Siegel ein und desselben Stils handelt. Um diesen Punkt zu klären,
empfiehlt es sich, die Siegel aufzulisten und nach Material und Schnitttechnik zu ordnen.
Bislang gehören  Siegel und Siegelungen44 der Anšan-Gruppe an:

Paläographie/
Siegel (R, S, A)45 Fundort Datierung Keilform Material Stil
Abb. H–I (R) – Ebarat Strichartig Chalcedon Gravur, Kugelbohrer-
stilisierung im Bereich des
Gesichtes
Abb. L–M (R) – – Offene Amethyst Kugelbohrerstilisierung im
Keilköpfe Bereich des Gewandes, feine
lineare Gewandzeichnung,
Gravur
Abb. N–O (R) – – Strichartig weicher schematisch, flache
schwarzer Ausführung, kein Kugelbohrer
Stein und nur Linien

35 Vgl. Lambert : Pl. V, Nr.  (= Abb. H–I).

36 Zur Lesung der Legende vgl. weiter unten.
37 Vgl. Amiet b: Fig. b. Die Inschrift der Abrollung auf einer Tontafel aus Susa wurde erstmals von Dossin

: Nr. ,  publiziert; vgl. auch Mofidi-Nasrabadi :  Nr. .

38 Zur Legende vgl. Mofidi-Nasrabadi :  Nr. .
39 Stève : –.
40 Porada : –.
41 Eine Rebsorte, die in der ostafghanischen Provinz Nuristan noch heute beheimatet ist.
42 Es handelt sich um eine Rauschpflanze (Steppenraute), die vor allem in Indien und im Steppengebiet Mittela-

siens verbreitet war; vgl. Stève : .

43 Die einzige bisher vorgeschlagene Datierung stammt von Porada : . Sie datierte die Siegel in das .–

. Jh. Die dagegen hier vorgenommene Datierung beruht auf dem zeitlichen Ansatz der Série Élamite Populaire,
da dieser Gruppe zwei Siegel des Anšan-„Stils“ entstammen.
44 Zwei weitere Siegel lassen sich eventuell ebenfalls dieser Gruppe zuordnen: GS  und GS . Das erste

Siegel ist jedoch so stark abgerieben und zerstört, dass sich keine genaueren Aussagen mehr zum Siegelbild treffen
lassen. Das Siegel GS  könnte auf Grund des Zottengewandes und der Verschmelzung der Person mit dem
Thron durchaus dieser Gruppe zugeordnet werden.
45 R = Rollsiegel; S = Stempelsiegel; A = Abrollung.
 georg neumann

Paläographie/
Siegel (R, S, A) Fundort Datierung Keilform Material Stil
Abb. P–Q (R) – – – weicher, schematisch, flache
schwarzer Ausführung, kein Kugelbohrer
Stein und nur Linien
Abb. R–S (S) – – – keine Angabe plastischer Stil, Gravur, kein
Kugelbohrer
Abb. W (R) – – – Hämatit plastischer Stil, Gravur, kein
Kugelbohrer46
Abb. J (A) Susa Pala-Iššan Geschlossene – plastischer Stil, Gravur, kein
Keile Kugelbohrer47
Abb. G (R) – – – keine Angabe Gravur, Kugelbohrerstilisie-
rung im Bereich des Gesichtes
und der Reben
Abb. A (R) Susa – – Bitumen grober Schnitt, schematisch
aus Linien, Kerbschnitt
Abb. B (R) Susa – – Kalkstein Gravur, plastischer Stil
Abb. D (R) Susa – – Steatit rein linearer Schnitt, schlecht
erhalten
Abb. V (A) Anšan – – – schematisch, kein
Kugelbohrer?48
Abb. K Susa – Pseudo- brauner schematisch, kein
Inschrift Kalkstein49 Kugelbohrer, sehr flach
Abb. U (R) – – – dunkel-grüner schematisch, kein
Kalkstein Kugelbohrer, sehr flach
Abb. F (R) – – – – kantig, tief eingeschnitten,
plastisch, Gravur, Linear
Abb. T (A) Anšan – Offene – keine Angabe möglich50
Keilköpfe
Abb. C (S) Anšan – – keine Angabe strichartig, linear

Die vorliegende Tabelle zeigt, dass es sich weder um einen einheitlichen Stil51 noch um einheit-
liches Rohmaterial52 bei den Siegeln der Anšan-Gruppe handelt. So sind einige Siegel aus harten
Materialien, wie violettem Amethyst und blau-grünem Chalzedon, andere aus weichen Werk-
stoffen, wie weißem Steatit und schwarzem Bitumen, gearbeitet. Die Herstellungstechniken,
die den Stil der Siegel wesentlich bestimmen, variieren stark. So existieren linear-schematische
Stile neben plastisch-modellierenden. Auch gibt es neben Rollsiegeln mehrere Stempelsiegel.

46 Durch Amiet ist lediglich die Umzeichnung publiziert; vgl. Amiet b.
47 Durch Amiet ist lediglich die Umzeichnung publiziert; vgl. ebd.
48 Durch Amiet und Sumner ist lediglich die Umzeichnung publiziert; vgl. ebd.
49 Vgl. Amiet : . Dort wurde das angegebene Material mit einem Fragezeichen versehen. M.E. könnte es

sich bei dem braunen Material auch um mit Sand versetztes Bitumen handeln.
50 Durch Pittman ist lediglich die Umzeichnung publiziert. Desweiteren ist die Abrollung stark fragmentiert; vgl.

Pittman .
51 Anhand der Siegel lassen sich linearer Stil, Kugelbohrerstil und Gravurstil identifizieren.
52 Das Spektrum der Materialien reich von weichem Stein (Steatit, Härte = ) bis zu Quarziten, den härtesten in

der Siegelherstellung Verwendung findenden Materialen (Chalcedon, Amethyst, Härte = ).

 elams kulturkontakte mit seinen nachbarn im spiegel der glyptik 

Nicht einmal die Paläographie der Keilschrift ist einheitlich, wie man es bei einer Siegelgruppe,
die räumlich und zeitlich so stark – nämlich auf das Gebiet von Anšan und den Beginn des .
Jt. v. Chr. – beschränkt gewesen sein soll, erwarten würde.
Die Unterschiede im Duktus der Keilschrift deuten vielmehr auf eine lange Laufzeit der Sie-
gel hin, wenn man sie denn als mehr oder weniger geschlossene Gruppe weiterhin definieren
möchte. Während lineare und geschlossene Keile auf Siegeln der Šimaški- bis Sukkalmah-Zeit
˘ den
belegt sind, weisen offene Keilköpfe auf die Mitte des . Jt. und finden sich beispielsweise in
Siegellegenden des Common Style der Mittani-Glyptik. Fasst man dies zusammen, so ergibt
sich eine Laufzeit besagter „Gruppe“ vom ./. bis etwa in das ./. Jh. v. Chr. Da es sich
hier lediglich um eine über das Siegelbild tradierte Bildkomposition und eine eventuell damit
verbundene religiöse Vorstellung handelt und nicht um einen Stil, ist eine derartige Langlebig-
keit der Motivik durchaus nicht ungewöhnlich.
Was ist daran jedoch für die Kulturkontakte Elams mit anderen Regionen herauszulesen?
Hierzu soll exemplarisch das wohl bekannteste Siegel dieser Motivgruppe analysiert werden.
Es handelt sich um das Siegel (Abb. H–I) aus dem Gulbekian Museum in Durham, das
 von Lambert publiziert und bereits mehrfach diskutiert wurde53, zuletzt von Mofidi-
Nasrabadi 54. Beim dem Artefakt handelt es sich m. E. um ein Siegel, welches zwar auf
Ebarat Bezug nimmt, jedoch weder ihm noch seiner Frau zuzuordnen ist55. Dies ergibt sich aus
der nachfolgend vorgeschlagenen Neuinterpretation der Siegellegende56:
d
[E-ba-r]a-at lugal [Ebar]at, der König,
x […] [die] x,
dam Na- ram -[X-X] Ehefrau des Na rām -[x-x]

Das Siegel ist also m.E. eher einer Würdenträgerin bzw. Hofdame unter Ebarat I. oder II.
zuzuordnen und zeigt drei Personen, die in ihrer Ausarbeitung stark an die sog. baktrischen
Kompositfigurinen erinnern, die in das ausgehende . und beginnende . Jt. v. Chr. zu datieren
sind57. Frisur und Kopfbedeckung des Mannes sind im elamischen Raum m.W. unbekannt und
könnten ebenfalls in den mittelasiatischen bzw. östlichen Bereich verweisen, da die Hauben
der Frauen klar in dieser Region zu verorten sind58. Wirft man nun einen Blick auf die drei
schwer zu deutenden Objekte in der Hand der Zentralperson, so kommt einem unweigerlich
die bereits von Stève vorgeschlagene Deutung als Tulpen in den Sinn59, was abermals in die
Gebiete Mittelasiens führen würde. Die wenigen ausgegrabenen Stücke dieser Motivgruppe
wurden in Susa und Tall-i Malyan gefunden und zeigen deutlich, dass religiöse Vorstellungen
und Ikonographie aus den östlichen bzw. nordöstlichen Gebieten des Iran und Zentralasiens
bis in den elamischen Kernbereich übernommen wurden.
Für die erste Hälfte des . Jt. v. Chr. lässt sich anhand der Glyptik zeigen, dass Elam bis weit in
den Osten hinein Kulturkontakte pflegte. Man darf insbesondere auf Grund des umfangreichen

53 Vgl. Lambert : –, Nr. .

54 Vgl. Mofidi-Nasrabadi : –.
55 Zur Legitimation über die Siegellegende bzw. über die dort angeführten Personen/Herrscher, vgl. Winter .
56 An dieser Stelle wird weder dem Vorschlag von Stève  noch dem von Lambert :  sowie Lambert

 gefolgt. Beide Lesungen seien hier noch einmal wiedergegeben: Lambert: d[E-ba-r]a-at LUGAL / x […] / DAM
na- ram -t[a-ka/šu] „[Ebar]at, der König / die x / [deine / seine g]eliebte Ehefrau“; Stève: d[E-ba-r]a-at LUGAL / s[i-
maš-giki] / DAM na- ram -t[a-šu] „[Ebar]at, der König / von Š[imaški] / [seine g]eliebte Ehefrau“. Die neue Lesung
ergibt sich m. E. auf Grund eines in Susa gefundenen Siegels, das eine vergleichbare Inschrift trägt (GS ); zur
Lesung besagter Inschrift vgl. Mofidi-Nasrabadi : .
57 Vgl. Potts : –. Dort auch die Abbildung (Nr. ) einer in Gonur Depe gefundenen Kompositfigu-

rine.
58 Vgl. Aruz : –, Nr. b.
59 Vgl. Stève : –.
 georg neumann

Befundes aus Susa vielleicht vermuten, dass Händler aus Syrien, Mesopotamien, der Region des
Persischen Golfes und aus Mittelasien nach Susa kamen, um dort entsprechende wirtschaftliche
Beziehungen zu unterhalten. Dieser Befund unterstreicht die Bedeutung Susas in der ersten
Hälfte des . Jt. v. Chr. Die lokal gefertigten Imitate lassen auf eine Gruppe von Personen
schließen, die bereits seit mehreren Generationen hier ansässig war und die sich durch ein
Siegel in der Machart ihrer ursprünglichen Heimat, der Region des Persischen Golfes bzw.
Mittelasiens, ein Stück von dieser bewahren wollte.
Die zweite Hälfte des . Jt. v. Chr. versorgt uns leider mit weit weniger bzw. mit weniger
spezifischen Informationen. Das publizierte Siegelmaterial aus Susa, Tchoga Zanbil, Haft Tepe
und Tall-i Malyan weist enge Bezüge zur zeitgleichen Glyptik Babyloniens auf, was sich vor
allem in der sog. pseudo-kassitischen Glyptik widerspiegelt60. Siegel dieser Gruppe wurden vor
allem in Susa61 und Tchoga Zanbil62 gefunden. Die Siegelabrollungen aus Haft Tepe lassen enge
Bezüge zur spätaltbabylonischen Glyptik erkennen63. Dies zeigt sich vor allem im häufigen
Gebrauch des Kugelbohrers und des oftmals schematisierenden linearen Schnittstils64. Die
Siegel sind jedoch – soweit es sich aus der Publikation von Negahban ersehen lässt – lokal
gefertigt, wie die Gewandungen und die Hörnerkronen der Götter zeigen65.
Die Igi-Halkiden- bzw. Untaš-Napiriša-zeitlichen Siegel aus Tchoga Zanbil sind im sog.
˘
pseudo-kassitischen Stil geschnitten und verdeutlichen so den engen politischen Kontakt mit
Mesopotamien. Eine eigene Ausprägung erfährt die Glyptik im Stil der sog. Série Élamite
Élaboré-Siegel66, der vor allem durch die Darstellung von Volutenbäumen, Bogenschützen
und schraffierten Bändern gekennzeichnet ist67. Ein Siegel dieses Stils fand sich in Marlik im
Nordiran68, vergesellschaftet mit Mittani-Siegeln69 und einem Siegel kassitischen70 und einem
mittelassyrischen71 Stils. Die zeitgleichen Rollsiegel aus der Region des Persischen Golfes,
beispielsweise aus Failaka, sind teilweise ebenfalls im pseudokassitischen Stil geschnitten72.
Jedoch sind hier m.W. keine Siegel der Série Élamite Élaboré belegt.
In der Nachfolgezeit, also ab der zweiten Hälfte der Igi-Halkiden scheinen sich der Stil und
die Motivik der Siegel zu ändern und sich von Babylonien ˘ weg zu orientieren. So zeigt eine
ganze Reihe von Siegeln aus Tchoga Zanbil nicht mehr die gewohnten kassitischen Motive der
sog. . und . kassitischen Gruppe73. Die Darstellung von Adoranten wird sehr selten, und der

60 Zur pseudo-kassitischen Glyptik vgl. Porada : – und Matthews : – und –.
61 Vgl. GS , –, –.
62 Vgl. Porada : Nr. –.
63 Vgl. Negahban : – und Seidl : . Die dort definierte sog. schematisierende Gruppe, Amiet’s

série d (vgl. Amiet : –), zeigt enge Bezüge zu einigen Abrollungen aus Haft Tepe auf, vgl. z.B. H.T.S. /a
(vgl. Negahban : Nr. ) und  (vgl. Negahban : Nr. ).
64 Vgl. etwa Negahban : Pl. :, Pl. : und Pl. :.
65 Dies zeigt sich vor allem an den Gewandungen, die teils zottenrockartig (Negahban : Pl. :), teils mit

einem Fischgrätenmuster stilisiert sind (ebd. Pl. :). Typisch und bereits in der altelamischen Glyptik gut belegt
sind glockenförmige Röcke bei Adoranten (ebd. Pl. :). Auch die Hörnerkronen der Götter zeigen deutlich,
dass die Haft Tepe-Glyptik lokalen Ursprungs ist. So weichen diese oftmals stark von dem ab, was man zeitgleich in
Babylonien erwarten würde (ebd. Pl. :). Sie können als typisch elamisches Element nach außen geschwungene
Hörner haben (ebd. Pl. :).
66 Vgl. Porada : –, Nr. –.
67 Vgl. ebd., z. B. Nr.  und .
68 Negahban : –.
69 Ebd. Nr. –.
70 Ebd, Nr. .
71 Ebd. Nr. .
72 Vgl. Matthews : – und Kjærum : Nr. , , , , –. Wahrscheinlich ist, dass auch

die hier nicht angeführten Siegelsteine der pseudokassitischen Gruppe zuzuweisen sind. Auf Grund ihrer starken
Verwitterung wäre jedoch jede versuchte Zuordnung spekulativ.
73 Vgl. z. B. Porada : Nr. –.
 elams kulturkontakte mit seinen nachbarn im spiegel der glyptik 

ehemals modellierende Schnittstil weicht einem abstrahierenden, oftmals linearen Stil, wie die
Siegel aus Susa zeigen. Es wurde auch ein Motiv eingeführt, das sich dann in der neuelamischen
Zeit besonderer Beliebtheit erfreute: Es sind die s-förmig gebogenen Tiere bzw. Mischwesen,
die oftmals heraldisch an einen Lebensbaum gestellt sind74.
In der späten mittelelamischen Zeit, also am Ende des . Jt. v. Chr., wird – wenn man das
in die Schicht IV zu datierende Siegel aus Tall-i Malyan berücksichtigt – oft ein Leiterband
als Begrenzung des Siegelbildes, welches häufig stark geometrisierend geschnitten ist, hinzuge-
fügt75. Dieses Band findet sich auch auf Siegeln der neuelamischen Zeit, wie eines der wenigen
stratifizierten Siegel aus Susa beweist76. An dieser Stelle ist ein Siegel anzuführen, dass bei den
Ausgrabungen in Tell Abraq, einem Hügel in der Region des Persischen Golfes, aus einer frü-
heisenzeitlichen Schicht zutage kam77. M. E. zeigt es das spätmittelelamische bzw. neuelamische
Leiterband, stark linear ausgeführte Bäume und mit Stacheln versehene Kugeln. Vergleichbare
Siegel stammen aus Susa78, Tchoga Zanbil79 und Hasanlu80. Es handelt sich bei dem Fundstück
mit großer Wahrscheinlichkeit um ein Siegel aus Elam.
Fasst man nun die Ergebnisse zusammen, so wird deutlich, dass die Siegel und Siegelungen
aus dem elamischen Bereich bis zur ersten Hälfte der Igi-Halkiden-Zeit in ihrer Motivik durch-
weg stark südmesopotamisch beeinflusst sind. Vor allem in ˘ der ersten Hälfte des . Jt. v. Chr. tritt
Susa und damit Elam als wichtiges Handelszentrum hervor. Dies zeigen nicht zuletzt Importe
und Imitate von Siegeln sowie Übernahmen in der Motivik aus Syrien, Kappadokien, dem Per-
sischen Golf bis hin zum mittelasiatischen Raum. Ab der frühen mittelelamischen Zeit scheint
die Bedeutung Elams im Handel abzunehmen. Dies wird dadurch gestützt, dass sich nach dem
Zusammenbruch des altassyrisch-kappadokischen Handelssystems und im Zuge der gleichzei-
tigen Intensivierung des levantinischen Handels der Schwerpunkt der Austauschbeziehungen
Babyloniens und Assyriens in den Westen Vorderasiens – und zwar unter Einbindung Ägyp-
tens – verlagerte und sich darauf konzentrierte.

Abkürzungsverzeichnis

AMI NF Archäologische Mitteilungen aus Iran. Neue Folge (Berlin).

AMIT Archäologische Mitteilungen aus Iran und Turan (Berlin).
AOAT Alter Orient und Altes Testament ([Kevelaer – Neukirchen-Vluyn] – Münster).
AS Assyriological Studies (Chicago).
ArAs Arts Asiatiques (Paris).
BBV Berliner Beiträge zur Vor- und Frühgeschichte (Berlin).
BBVO Berliner Beiträge zum Vorderen Orient (Berlin).
CDAFI Cahier de la Délégation Archéologique Francaise en Iran (Paris).
CRAI Compte Rendu de la Rencontre Assyriologique Internationale.
DMOA Documenta et Monumenta Orientis Antiqui (Leiden).
GS P. Amiet, Glyptique susienne des origines à l’époque des Perses Achéménides. Cachets,
sceaux-cylindres et empreintes antiques. Découvertes à Suse de  à  (MDP ).
IrAnt Iranica Antiqua ([Leiden] – Ghent).
JNES Journal of Near Eastern Studies (Chicago).

74 Vgl. ebd. Nr. – (= Abb. –).

75 Vgl. Carter : Fig. : (= Abb. ). Die Schicht IV des mittelelamischen Baus (in EDD) datiert an das Ende
des . Jt. v. Chr., wie Ziegelinschriften Hutelutuš-Inšušinaks belegen; vgl. ebd. –.
76 Vgl. GS  (VR IX) (= Abb. ).
77 Potts b: –, Fig. – (= Abb. –).
78 Vgl. GS – (= Abb. , –), – (= Abb.  und ) und  (= Abb. ) sowie die von

Delaporte  publizierten Siegel S. (= Abb. ) und S. (= Abb. ).
79 Vgl. Porada : Nr.  (= Abb. ) und Nr.  (= Abb. ).
80 Marcus : , Nr.  (= Abb. ).
 georg neumann

MDP Mémoires de la Mission Archéologique de Perse (Paris).

OBO Orbis Biblicus et Orientalis (Freiburg [Schweiz] – Göttingen).
OIP Oriental Institute Publications (Chicago).
RA Revue d’Assyriologie et d’Archéologie Orientale (Paris).
RlA Reallexikon der Assyriologie und Vorderasiatischen Archäologie ([Leipzig] – Berlin – New
York).
SAOC Studies in Ancient Oriental Civilization (Chicago).
SM Studia Mediterranea (Pavia).
UAVA Untersuchungen zur Assyriologie und Vorderasiatischen Archäologie (Berlin – New York).
ZA Zeitschrift für Assyriologie und Vorderasiatische Archäologie (Berlin – New York).

Literaturverzeichnis

Amiet, P. (): Glyptique susienne des origines à l’époque des Perses Achéménides. Cachets, sceaux-
cylindres et empreintes antiques. Découvertes à Suse de  à  (MDP ), Paris.
———. (): „Glyptique Élamite, à propos de Documents Nouveaux“, ArAs , –.
———. (a): „La glyptique du second millénaire en provenance des chantiers A et B de la Ville Royale
de Suse“, IrAnt , –.
———. (b): „Antiquités de serpentine“, IrAnt , –.
———. (a): L’âge des échanges inter-iraniens – avant J.-C., Paris.
———. (b): „Susa and the Dilmun culture“, A. Shaikh A. & M.R. Al Khalifa (Hrsg.), Bahrain through
the Ages: The Archaeology, London – New York – Sydney – Henley, –.
Aruz, J. (): Art of the First Cities, New York – New Haven – London.
Börker-Klähn, J. (): Untersuchungen zur altelamischen Archäologie, Berlin.
Buchanan, B. (): „A Dated „Persian Gulf “ Seal and its Implications, in: Studies in Honor of Benno
Landsberger“, AS , –.
E. Carter, E. (): „Excavations in Ville Royale I at Susa: the third millennium BC occupation“,
CDAFI , –.
———. (): Excavations at Anshan (Tal-e Malyan): the Middle Elamite period, Philadelphia.
Collon, D. (): First Impressions: Cylinder Seals in the Ancient Near East, London.
Connan, J. & Deschesne, O. (): Le Bitume à Suse, Paris.
Contenau, G. & Ghirshman, R. (): Fouilles du Tépé Giyan près de Néhavend  et  (Musée du
Louvre. Département des antiquités orientales. Série archéologique ), Paris.
Delaporte, L. (): Catalogue des Cylindres cachets et pierres gravées de style Oriental, Paris.
Dittmann, R. (): „Eisenzeit I und II in West- und Nordwest-Iran zeitgleich zur Karum-Zeit Anato-
liens?“, AMI NF , –.
Dossin, G. (): Autre Textes Sumériens et Accadiens (MDP ), Paris.
Durand, J.-M. (): „Fragments rejoint pour une histoire élamite“, L. De Meyer, H. Gasche & F. Vallat
(Hrsg.), Fragmenta Historiae Elamicae. Mélanges offerts à M.J. Steve, Paris, –.
Edzard, D.O. (–): „Mekubi“, RlA , .
Eidem, J. & Læssøe, J. (): The Shemshara Archives . The Letters, Viborg.
Frankfort, H. (): Stratified Cylinder Seals from the Diyala Region (OIP ), Chicago.
Ghirshman, R. (): Fouilles de Sialk. Vol. I, Paris.
Hill, H.D., Jacobsen, Th. & Delougaz, P. (): Old Babylonian Public Buildings in the Diyala Region
(OIP ), Chicago.
Huh, S. (): Studien zur Region Lagaš: Von der Ubaid- bis zur Altbabylonischen Zeit (AOAT ),
Münster.
Kjærum, P. (): Failaka / Dilmun. The Second Millennium Settlements. Vol. :. The Stamp and Cylinder
Seals, Aarhus.
Læssøe, J. (): The Shemshāra Tablets. A Preliminary Report, Kopenhagen.
Lambert, M. (): „Tablet de Suse avec cachet du Golfe“, RA , –.
Lambert, W.G. (): „Near Eastern Seals in the Gulbenkian Museum of Oriental Art, University of
Durham“, Iraq , –.
———. (): „Further notes on a Seal of Ebarat“, RA , –.
Marcus, M.I. (): Emblems of identity and prestige: the seals and sealings from Hasanlu, Iran; commen-
tary and catalogue, Philadelphia.
 elams kulturkontakte mit seinen nachbarn im spiegel der glyptik 

Matthews, D.M. (): The Kassite Glyptic of Nippur (OIP ), Chicago.
Mofidi-Nasrabadi, B. (): Aspekte der Herrschaft und der Herrscherdarstellungen in Elam im . Jt. v.
Chr. (AOAT ), Münster.
Nagel, W. (): Djamdat Nasr-Kulturen und Frühdynastische Buntkeramiker (BBV ), Berlin.
Negahban, E.O. (): Excavations at Haft Tepe, Iran, Philadelphia.
———. (): Marlik, Philadelphia.
Neumann, G. (): “Siegel für jedermann. Neue Erkenntnisse zur sog. Série Élamite Populaire und
zur magischen Bedeutung von Siegelsteinen”, G. Wilhelm (Hrsg.), Organization, Representation,
and Symbols of Power in the Ancient Near East. Proceedings of the th Rencontre Assyriologique
Internationale at Würzburg – July , Winona Lake, –.
Otto, A. (): Die Entstehung und Entwicklung der Klassisch-Syrischen Glyptik (UAVA ), Berlin.
Özgüc, N. (): Seals and Seal Impressions of Level Ib from Karum Kanish, Ankara.
Parrot, A. (): Glyptique Mésopotamienne. Fouilles de Lagash (Tello) et de Larsa (Senkereh) (–
), Paris.
———. (): Mission Archéologique de Mari II (). Le Palais, Paris.
Pecorella, P.E. (): „Un gruppo di sigilli cilindrici del Vicino Oriente“, SM /, –.
Piller, Ch.K. (–): „Zur Mittelbronzezeit im nördlichen Zentraliran – Die Zentraliranische
Graue Ware (Central Grey Ware) als mögliche Verbindung zwischen Eastern und Western Grey Ware“,
AMIT –, –.
Pittman, H. (): „Reconsidering the Trouvaille de la Statuette d’Or“, N.F. Miller & K. Abdi (Hrsg.), Yeki
bud, yeki nabud. Essays on the Archaeology of Iran in Honor of W.M. Sumner, Los Angeles, –.
Porada, E. (): „The Origin of Winnirke’s Cylinder Seal“, JNES , –.
———. (): Tchoga Zanbil (Dur-Untash). La Glyptique. Vol. IV (MDP ), Paris.
———. (): „More Seals of the Time of the Sukkalmah“, RA , –.
Potts, D.T. (a): The Arabian Gulf in Antiquity. Vol. I,˘Oxford.
———. (b): A Prehistoric Mound in the Emirate of Umm al-Qaiwain, U.A.E., Munksgaard.
———. (): The Archaeology of Elam: Formation and transformation of an ancient Iranian State
(Cambridge World Archaeology), Cambridge.
———. (): „Puzur-Inšušinak and the Oxus Civilization (BMAC): Reflections on Šimaški and the geo-
political landscape of Iran and Central Asia in the Ur III period“, ZA , –.
Röllig, W. (–): „Heirat, politische“, RlA , –.
Sarianidi, V. (): Die Kunst des alten Afghanistan, Leipzig.
Seidl, U. (): „Altelamische Siegel“, F. Vallat (Hrsg.), Contribution à l’Histoire de l’Iran. Mélanges offerts
à Jean Perrot, Paris, –.
Stève, M.-J. (): „Des Sceaux-Cylindres de Simaški“, RA , –.
Sumner, W. (): „Excavations at Tall-i Malyan –“, Iran , –.
Vallat, F. (): „Le dieu Enzak: une divinité dilmunite venérées à Suse“, D.T. Potts (Hrsg.), Dilmun: New
Studies in the Archaeology and Early History of Bahrain (BBVO ), Berlin, –.
Vanden Berghe, L. (): Archéologie de l’Irān Ancien (DMOA ), Leiden.
Winter, I. (): „Legitimation of Authority through Image and Legend: Seals belonging to Officials in
the Administrative Bureaucracy of the Ur III State“, McG. Gibson & R.D. Biggs (eds.), The Organization
of Power: Aspects of Bureaucracy in the Ancient Near East (SAOC ), Chicago, –.
 georg neumann

Abbildungen

Abb. . Amiet : Tf. , Nr. .

Abb. . Amiet : Tf. , Nr. .

Abb. . Amiet : Tf. , Nr. . Abb. . Amiet :
Tf. , Nr. .
 elams kulturkontakte mit seinen nachbarn im spiegel der glyptik 

Abb. . Amiet : Tf. , Nr. . Abb. . Amiet : Tf. , Nr. .

Abb. . Amiet : Tf. , Nr. . Abb. . Amiet : Tf. X, Nr. .

Abb. . Amiet a: , Nr. . Abb. . Amiet : Tf. , Nr. bis.

Abb. . Amiet : Abb. . Carter : Fig. ,.
Tf. , Nr. .
 georg neumann

Abb. . Carter Abb. . Carter : Fig. ,.

: Fig. ,.

Abb. . Carter : Fig. ,.

Abb. . Carter : Fig. ,.

 elams kulturkontakte mit seinen nachbarn im spiegel der glyptik 

Abb. . Eidem & Læssøe : Tf.  (Seal ). Abb. . Eidem & Læssøe
: Tf.  (Seal ).

Abb. . Eidem & Læssøe : Tf.  (Seal ).

 georg neumann

Abb. . Eidem & Læssøe : Tf.  (Seal ).

Abb. . Eidem & Læssøe Abb. . Ghirshman : Tf. , Nr. S.a.
: Tf.  (Seal ).

Abb. . Contenau & Abb. . Contenau & Ghirshman : Tf. :.
Ghirshman : Tf. :.
elams kulturkontakte mit seinen nachbarn im spiegel der glyptik 

Abb. . Amiet : Tf. , Nr. .

Abb. . Amiet : Abb. . Amiet : Tf. , Nr. .
Tf. , Nr. .

Abb. . Amiet : Tf. , Nr. .

 georg neumann

Abb. . Amiet : Abb. . Amiet :

Tf. , Nr. . Tf. , Nr. .

Abb. . Amiet : Abb. . Amiet :

Tf. , Nr. . Tf. , Nr. .
elams kulturkontakte mit seinen nachbarn im spiegel der glyptik 

Abb. . Amiet : Tf. , Nr. .

Abb. . Amiet : Tf. , Nr. .

 georg neumann

Abb. . Amiet : Tf. , Nr. .

Abb. . Amiet : Tf. , Nr. .

 elams kulturkontakte mit seinen nachbarn im spiegel der glyptik 

Abb. . Amiet : Tf. , Nr. . Abb. . Amiet : Tf. , Nr. .

Abb. . Amiet : Tf. , Nr. . Abb. . Amiet : Tf. , Nr. .

Abb. . Frankfort Abb. . Frankfort : Pl. , Nr. .
: Pl. , Nr. .

Abb. . Frankfort : Pl. , Nr. . Abb. . Parrot
: Pl. XII, .
 georg neumann

Abb. . Parrot : Pl. XII, . Abb. . Parrot : Pl. IX, .

Abb. . Porada : Fig. .

Abb. . Amiet : Tf. , Nr. . Abb. . Amiet : Tf. , Nr. .

Abb. . Sumner : Fig. a. Abb. . Sumner : Fig. b.
 elams kulturkontakte mit seinen nachbarn im spiegel der glyptik 

Abb. . Sumner : Fig. c. Abb. . Sumner : Fig. d.

Abb. . Sumner : Fig. e. Abb. . Sumner : Fig. f.

Abb. . Sumner Abb. . Sumner : Fig. h.

: Fig. g.

Abb. . Sumner : Fig. j. Abb. . Sumner : Fig. l.
 georg neumann

Abb. . Porada : Pl. V,. Abb. . Porada : Pl. V,.

Abb. . Porada : Pl. V,. Abb. . Carter : Fig. ,.

Abb. . Amiet : Tf. , Nr. . Abb. . Potts b: , Fig. .
 elams kulturkontakte mit seinen nachbarn im spiegel der glyptik 

Abb. . Potts b: , Fig. .

Abb. . Amiet : Tf. , Nr. . Abb. . Amiet : Tf. , Nr. .

Abb. . Amiet : Abb. . Amiet : Tf. , Nr. .
Tf. , Nr. .
 georg neumann

Abb. . Amiet : Tf. , Nr. . Abb. . Delaporte
: Tf. , S..

Abb. . Delaporte : Tf. , S.. Abb. . Porada : Pl. III, .

Abb. . Porada : Pl. V,. Abb. . Marcus : , Fig.  (Nr. ).
elams kulturkontakte mit seinen nachbarn im spiegel der glyptik 

Abb. . Amiet : Tf. , Nr. .

Abb. . Amiet : Tf. , Nr. .

 georg neumann

Abb. A. Amiet : Tf. , Nr. . Abb. B. Amiet : Tf. , Nr. .

Abb. C. Sumner : Fig. i. Abb. D. Amiet :

Tf. , Nr. .

Abb. E. Amiet : Tf. , Nr. . Abb. F. Amiet : Fig. ,.

Abb. G. Amiet : Fig. ,.

 elams kulturkontakte mit seinen nachbarn im spiegel der glyptik 

Abb. H. Lambert : Tf. V, Nr. . Abb. I. Amiet : Fig. ,.

Abb. J. Amiet b: , Fig. b. Abb. K. Amiet : Fig. , .

Abb. L. Porada : Tf. I, Nr. a. Abb. M. Porada : Tf. II, Nr. b.

Abb. N. Porada : Tf. I, Nr. a. Abb. O. Porada : Tf. II, Nr. b.
 georg neumann

Abb. P. Porada : Tf. I, Nr. a. Abb. Q. Porada : Tf. II, Nr. b.

Abb. R. Porada Abb. S. Porada

: Tf. II, Nr. a. : Tf. II, Nr. b.

Abb. T. Pittmann : Nr. .. Abb. U. Pecorella : , Abb. .

Abb. V. Amiet b: , Fig. d. Abb. W. Amiet b: , Fig. c.
 elams kulturkontakte mit seinen nachbarn im spiegel der glyptik 

Appendix: Konkondanzen zu den altelamischen Siegeln

in Delaporte , Börker-Klähn  und Amiet 

. Sb – Louvre
Börker- Börker-
GS, ArAs. Klähn GS, ArAs. Klähn
Sb-Nr. 81 – Nr.  Sb-Nr.  – Nr. 

Sb   Sb  

Sb   Sb   
Sb   Sb   
Sb   Sb  
Sb   Sb  
Sb   Sb   
Sb   Sb  
Sb   Sb  
Sb   Sb  
Sb   Sb   
Sb   Sb  
Sb   Sb  
Sb   Sb  
Sb   Sb   
Sb    Sb   
Sb   Sb  
Sb   Sb   
Sb   Sb  
Sb   Sb   
Sb   Sb  
Sb   Sb  
Sb   Sb  
Sb   Sb  
Sb   Sb  
Sb   Sb  
Sb   Sb   
Sb   Sb  
Sb   Sb  
Sb   Sb   
Sb   Sb  
Sb   Sb  
Sb   Sb  
Sb   Sb  
Sb   Sb  
Sb   Sb  
Sb   Sb  
Sb   Sb  
Sb   Sb  
Sb   Sb  
Sb   Sb  
Sb   Sb  
Sb   Sb  
Sb   ,  Sb  
Sb   Sb   
Sb   Sb  

81 Amiet .
 georg neumann

Börker- Börker-
GS, ArAs. Klähn GS, ArAs. Klähn
Sb-Nr.  – Nr.  Sb-Nr.  – Nr. 

Sb   Sb  

Sb   Sb  
Sb   Sb  
Sb   Sb  
Sb   Sb  
Sb   Sb  
Sb   Sb  
Sb    Sb   
Sb    Sb   
Sb   Sb  
Sb   Sb  
Sb   Sb  
Sb    Sb  
Sb   Sb   
Sb    Sb   
Sb   Sb   
Sb   Sb  
Sb   Sb   
Sb   Sb  
Sb    Sb  
Sb    Sb  
Sb   Sb  
Sb   Sb  
Sb   Sb  
Sb   Sb  
Sb    Sb  
Sb    Sb  
Sb   Sb  
Sb   Sb  
Sb   Sb  
Sb   Sb  
Sb   Sb  
Sb   Sb  
Sb   Sb  
Sb   Sb  
Sb   Sb  
Sb   Sb  
Sb   Sb  
Sb    Sb  
Sb   Sb  
Sb    Sb  
Sb   Sb  
Sb    Sb   
Sb   Sb  
Sb   Sb  
Sb   Sb   
Sb   Sb   
Sb   Sb  
Sb   Sb  
Sb   Sb bis 
Sb   Sb , Sb  
Sb   Sb bis 
Sb   Sb bis 
 elams kulturkontakte mit seinen nachbarn im spiegel der glyptik 

Börker- Börker-
GS, ArAs. Klähn GS, ArAs. Klähn
Sb-Nr.  – Nr.  Sb-Nr.  – Nr. 

Sb bis  Sb  

Sb bis  Sb  
Sb , Sb ,  Sb  
Sb , Sb  Sb  
Sb   Sb  
Sb   Sb  
Sb   Sb  
Sb   Sb  
Sb   Sb  
Sb   Sb  bis
Sb   Sb  
Sb   Sb  
Sb   Sb  
Sb   Sb  
Sb   Sb  
Sb   Sb  
Sb   Sb  
Sb , Sb , Sb  
Sb , Sb   Sb  
Sb   Sb  
Sb   Sb  
Sb , Sb , Sb  
Sb , Sb   Sb , Sb ,
Sb   Sb , Sb  
Sb   Sb  
Sb   Sb  (E )  
Sb   Sb  (E ) 
Sb    Sb  (E )  
Sb   Sb  (E )  
Sb   Sb  (E ) 
Sb  (GS )  Sb  (E )  
Sb   Sb  
Sb , Sb   Sb  (E ) 
Sb    Sb  (E ) 
Sb   Sb  (E ) 
Sb   Sb  
Sb   Sb  
Sb   Sb  (F )  
Sb   Sb  (F)  
Sb   Sb  (F ) 
Sb   Sb  (F ) 
Sb   Sb  (E )  
Sb   Sb  (F )  
Sb   Sb   
Sb   Sb   
Sb   Sb  (G ) 
Sb   Sb  (G )  
Sb   Sb  (G ) 
Sb   Sb  (G )  
Sb   Sb  (H ) 
Sb   Sb  (H )  
Sb   Sb  (H ) 
Sb   Sb  (H )  
 georg neumann

Börker- Börker-
GS, ArAs. Klähn GS, ArAs. Klähn
Sb-Nr.  – Nr.  Sb-Nr.  – Nr. 

Sb  (H )   Sb  

Sb  (H )   Sb  
Sb    Sb  
Sb  (H )   Sb  
Sb  (H )   Sb  
Sb  (H )   Sb  
Sb  (H )   Sb  
Sb    Sb  
Sb  (K )   Sb  
Sb    Sb  
Sb    Sb  (TS B ) 
Sb  (K )   Sb  
Sb  (K )   Sb , Louvre 
Sb  (K )   Sb  ArAs ,
Sb  (K )   Nr. 
Sb  (I )   Sb , Château de
Sb  (K )   Suse, Nr. 
Sb  (K )   und Louvre 
Sb  (K )   Sb  ArAs ,
Sb  (K )   Nr. 
Sb  (K )   Sb , Sb ,
Sb  (I )  Louvre 
Sb  (K  bis)  Sb , Sb ,
Sb  (I )  Louvre 
Sb   Sb , Louvre 
Sb   Sb  ArAs ,
Sb   Nr. 
Sb    Sb  ArAs ,
Sb   Nr. 
Sb   Sb  ArAs ,
Sb   Nr. 

. MT – Museum Teheran
Börker- Börker-
GS, ArAs. Klähn GS, ArAs. Klähn
MT-Nr. 82 – Nr.  MT-Nr.  – Nr. 

MT  MT 
MT  MT 
MT  MT ArAs ,
MT  Nr. 
MT  MT (GS ) 
MT  MT (GS ) 
MT  MT (GS ) 
MT  MT  
MT  MT  
MT  MT  
MT  MT  
MT  MT  
MT   MT  

82 = Amiet .
 elams kulturkontakte mit seinen nachbarn im spiegel der glyptik 

Börker- Börker-
GS, ArAs. Klähn GS, ArAs. Klähn
MT-Nr.  – Nr.  MT-Nr.  – Nr. 

MT   MT  

MT   MT  
MT   MT  
MT  A ()  MT  
MT  A ()  MT  
MT  A ()  MT  
MT  B ()  MT  
MT  B ()  MT  
MT   MT   
MT   MT  
MT   MT  
MT   MT  
MT   MT  
MT   MT  
MT   MT   
MT   MT  
MT   MT  
MT   MT  
MT   MT  
MT    MT  
MT   MT  
MT   MT  
MT   MT  
MT   MT  
MT  ()  MT  
MT   MT  
MT   MT  
MT  ()  MT  
MT  ()  MT  
MT   MT  
MT   MT  
MT  ()  MT  
MT   MT  
MT   MT   
MT   MT ? ? 
MT   MT   
MT    MT  
MT   MT  (GS ) 
MT   MT  (GS ) 
MT  

. AO – Louvre
AO-Nr. Delaporte  Börker-Klähn 
AO  A. 
 georg neumann

. AS – Louvre
Börker- Börker-
Delaporte Klähn Delaporte Klähn
AS-Nr.   AS-Nr.  

AS  S.  AS  S. 

AS  S.  AS  S. 
AS  S.  AS  S. 
AS  S.  AS  S. 
AS  S.  AS  S. 
AS  S.  AS  S. 
AS  S.  AS  S. 
AS  S.  AS  S. 
AS  S.  AS  S. 
AS  S.  AS  S. 
AS  S.  AS  S. 
AS  S.  AS  (, ,
AS  S.  , , ) S. 
AS  S.  AS  S. 
AS  S.  AS  () S. 
AS  S.  AS  () S. 
AS  S.  AS  S. 
AS  S.  AS  S. 
AS  S.  AS  S. 
AS  S.  AS  S. 
AS  S.  AS  S. 
AS  S.  AS  S. 
AS  S.  AS  S. 
AS  S.  AS A  S. 
AS  S.  AS A  S. 

. Château de Suse
Château de Suse-Nr. ArAs , Nr., GS Börker-Klähn 
Château de Suse ArAs , Nr. 
Château de Suse ArAs , Nr. 
Château de Suse, Nr.  ArAs , Nr. 
Château de Suse, Nr. – 
Château de Suse, Nr.  
Château de Suse, Nr.  ArAs , Nr. 
Château de Suse, Nr.  ArAs , Nr. 
Château de Suse, Nr.  ArAs , Nr. 
Château de Suse, Nr. –, Sb   
Château de Suse, Nr. – 
Château de Suse, Nr.  
Château de Suse, Nr.  ArAs , Nr. 
Château de Suse, Nr.  ArAs , Nr. 
Château de Suse, Nr.  
Château de Suse, Nr.   

. Tafeln des Tehraner Museums

Tafel-Nr. GS
Museum Tehran, Tafel Nr.  
 elams kulturkontakte mit seinen nachbarn im spiegel der glyptik 

. MET – Metropolitan Museum of Art

MET-Nr. Börker-Klähn 
.. Porada : , fig.  

. Glyptique Susienne
Börker- Börker-
Fund-/ Klähn Fund-/ Klähn
GS-Nr. Museums-Nr.  GS-Nr. Museums-Nr. 

 Sb  (F)   Sb 

 Sb    MT  
 Sb   Sb 
 MT  MT  
 MT   ? MT ? 
 Sb   Sb 
 MT   Sb 
 MT  Sb 
 SB   Sb  
 Sb   Sb  
 Sb   MT
 Sb   Sb 
 MT   Sb 
 Sb   Sb 
 Sb   Sb 
 Sb   MT 
 MT  A ()  Sb bis
 Sb   Sb 
 Sb   Sb 
 Sb   Sb 
 MT   Sb 
 Sb   Sb 
 Sb   Sb 
 Sb   Sb  
 Sb    Sb  (H ) 
 Sb   Sb  (H ) 
 MT  ()  Sb  (H ) 
 Sb   Sb 
 Sb   Sb 
 Sb   Sb 
 Sb   Sb 
 Sb   MT 
 Sb   Sb 
 Sb  (TS B )  MT 
 Sb   MT 
 Sb , Sb ,  Sb 
Sb , Sb   Sb 
 Sb   MT 
 M. de Téhéran,  Sb 
Tafel Nr.   Sb 
 Sb   Sb 
 Sb   Sb 
 Sb    Sb 
 MT  Sb , Sb 
 Sb    Sb 
 Sb  (H )   MT 
 georg neumann

Börker- Börker-
Fund-/ Klähn Fund-/ Klähn
GS-Nr. Museums-Nr.  GS-Nr. Museums-Nr. 

 MT   Sb  

 Sb   Sb 
 Sb bis  Sb 
 MT   Sb 
 Sb  (K )   MT
 Sb  (I )  Sb 
 MT (GS )  MT
 Sb    Sb 
 MT   Louvre
 MT  Sb 
 Sb   MT 
 Sb   Sb  
 MT  Sb 
 Sb   MT 
 Sb   Sb 
 Sb   Sb 
 Sb   Sb  
 Sb   Sb  
 MT   –
 –  Sb  (K ) 
 Sb   MT 
 MT  Sb 
 Sb   Sb 
 Sb   Sb 
 –  MT  
 Sb  (H )   MT (GS )
 MT   Louvre
 Sb   –
 MT   Sb 
 MT   Sb 
 MT   MT 
 Sb    Sb 
 Sb  (K )   MT 
 Sb   MT  
 Sb  (K )   Sb 
 Sb  (GS )  Sb 
 Sb  (K )   Sb 
 Sb   MT 
bis Sb   Sb 
 Sb  (E )   Sb 
 Sb    Sb  (E )
 Sb   Sb 
 Sb  (K )   Sb 
 MT   Sb 
 MT   Sb 
 MT   Sb 
 MT  MT 
 Sb   Sb 
 Sb   MT 
 MT  Sb  (G ) 
 Sb  (K )   Sb 
 Sb   Sb 
 Sb   Sb 
 elams kulturkontakte mit seinen nachbarn im spiegel der glyptik 

Börker- Börker-
Fund-/ Klähn Fund-/ Klähn
GS-Nr. Museums-Nr.  GS-Nr. Museums-Nr. 

 Sb   Sb 

 Sb   Sb  (I ) 
 Sb   MT 
 Sb   MT 
 Sb   Sb 
 Sb   Sb  , 
 MT  B ()  Sb  (H ) 
 MT   MT 
 Sb   Sb  
 Sb   Sb  (K ) 
 –  Sb  
 –  Sb  
 Sb   Sb 
 Sb   Sb  (E ) 
 –  Sb  
 Sb   Sb  
 Sb   Sb 
 Sb   Sb 
 Sb   MT 
 Sb   MT 
 MT   MT 
 Sb   Sb  
 Sb   Sb bis
 Sb   Sb 
 MT   Sb 
 –  MT 
 Sb  (H )   Sb 
 Sb   Sb 
 MT   Sb 
 Sb    Sb  
 Sb   MT 
 Sb   Sb 
 Sb   Sb 
 Sb  (K )   Sb 
 Sb   Sb 
 Sb   MT 
 Sb  (K  bis)  Sb 
 Sb    MT 
 Sb    Sb 
 MT   –
 Sb   MT 
 Sb    Sb 
 MT   Sb 
 MT   Sb  (K ) 
 Sb   Sb 
 Sb   Sb 
 Sb   MT 
 Sb   Sb 
 Sb    MT
 Sb   Sb 
 Sb   Sb  
 Sb   –
 MT    MT 
 georg neumann

Börker- Börker-
Fund-/ Klähn Fund-/ Klähn
GS-Nr. Museums-Nr.  GS-Nr. Museums-Nr. 

 –  MT  A ()

 MT   Sb  
 Sb    Sb 
 Sb  (F )   Sb 
 Sb   MT 
 –  Sb 
 Sb   Sb  
 Sb   Sb  
 MT   Sb 
 Sb  (F )   Sb 
 Sb   Sb 
 Sb   MT 
 MT   Sb  (E ) 
 Sb   Sb 
 Sb   Sb 
 MT   Sb  
 Sb   Sb 
 Sb   MT  ()
 Sb   MT 
 Sb   Sb  
 Sb   Sb 
 Sb   MT 
 Sb   Sb  
 Sb   Château de Suse, 
 MT  Nr. 
 Sb    Sb 
 Sb   MT 
 Sb   Sb 
 MT   Sb  
 MT   Louvre 
 Sb   Louvre 
 MT   Louvre
 –  Sb 
 Sb   Sb bis
 MT   Sb 
 Sb    Sb 
 Sb   Sb 
 Sb   Sb 
 Sb   Sb 
 Sb    Sb 
 Sb    Sb bis
 Sb   Sb 
 Sb   Sb 
 MT   Sb 
 MT  ()  Sb 
 MT   MT
 MT  ()  Sb 
 Sb   Sb 
 Sb   Sb 
 Sb   Sb 
 Sb   Sb 
 Sb   Sb  (I )
 Sb   Sb  (E )
 elams kulturkontakte mit seinen nachbarn im spiegel der glyptik 

Börker- Börker-
Fund-/ Klähn Fund-/ Klähn
GS-Nr. Museums-Nr.  GS-Nr. Museums-Nr. 

 Sb   Sb 

 Sb    Sb 
 Sb    Sb 
 MT  Sb 
 MT  (GS )  Sb  (G ) 
 Sb   Sb 
 Sb   Sb 
 MT   Sb 
 MT  (GS )  Sb 
 Sb  (E )   Sb 
 Sb   Sb  (G )
 MT  B ()  MT (GS )
 Sb   Sb  (F )
 Sb   Sb 
 Sb  (G )  Sb  (F )
 Sb   –
 Sb   MT 
 Sb  (E )  Château de Suse,
 Sb  (E )  Nr. –
 Sb  (H )  Château de Suse,
 Sb   Nr. 
 Sb   Château de Suse,
 Sb  (E ) Nr. 
 Sb  (E )  Sb , Louvre
 Sb  (H )  Sb , Sb ,
 Sb  Louvre
 Sb   Château de Suse,
 Sb  Nr. –
 MT  A ()  Château de Suse,
 Sb  Nr. 
 Sb   Sb , Louvre
 MT   Sb , Château
 Sb  (H )  de Suse, Nr. 
 MT  und Louvre
 Sb   Château de Suse, Nr. 
 Sb  –, Sb 
 Sb 

. Delaporte 
Börker- Börker-
Delaporte Fund-/ Klähn Delaporte Fund-/ Klähn
 Museums-Nr.   Museums-Nr. 

A. AO   S. AS  

S. AS   S. AS  
S. AS   S. AS  
S. AS  (, , ,   S. AS  
) S. AS  
S. AS  ()  S. AS  
S. AS   S. AS  
S. AS  ()  S. AS  
S. AS   S. AS  
 georg neumann

Börker- Börker-
Delaporte Fund-/ Klähn Delaporte Fund-/ Klähn
 Museums-Nr.   Museums-Nr. 

S. AS A   S. AS  

S. AS   S. AS  
S. AS   S. AS  
S. AS   S. AS  
S. AS   S. AS  
S. AS   S. AS  
S. AS   S. AS  
S. AS   S. AS  
S. AS A   S. AS  
S. AS   S. AS  
S. AS   S. AS  
S. AS   S. AS  
S. AS   S. AS  
S. AS   S. AS  
S. AS   S. AS  
S. AS   S. AS  
S. AS  

. ArAs 83

ArAs  – Nr. Fund-/Museums-Nr.
ArAs , Nr.  Sb 
ArAs , Nr.  Château de Suse
ArAs , Nr.  Château de Suse
ArAs , Nr.  Château de Suse, Nr. 
ArAs , Nr.  Sb 
ArAs , Nr.  Sb 
ArAs , Nr.  Château de Suse, Nr. 
ArAs , Nr.  Château de Suse, Nr. 
ArAs , Nr.  Château de Suse, Nr. 
ArAs , Nr.  MT
ArAs , Nr.  Sb 
ArAs , Nr.  Teheran
ArAs , Nr.  Sb 
ArAs , Nr.  Château de Suse, Nr. 
ArAs , Nr.  Château de Suse, Nr. 

. Börker-Klähn 84

Börker- Börker-
Klähn Fund-/ GS, Dela- Klähn Fund-/ GS, Dela-
 Museums-Nr. porte   Museums-Nr. porte 

 AS  S.  MDP  (), –

 MT   , fig. :
 Sb  (H )   AS  S.
 Sb  (K )   AS  S.
 AS  S.  AS  S.
 Sb    Sb  

83
= Amiet .
84
Da sich bei einigen Angaben nicht nachvollziehen ließ, welche Fund-Nr. oder Delaporte - und GS-Nr. den
Siegeln bei J. Börker-Klähn entsprechen, werden in diesen stattdessen hier Literaturzitate verwendet.
 elams kulturkontakte mit seinen nachbarn im spiegel der glyptik 

Börker- Börker-
Klähn Fund-/ GS, Dela- Klähn Fund-/ GS, Dela-
 Museums-Nr. porte   Museums-Nr. porte 

 MDP  (), – ,  Sb  

, fig. :  MDP  (), –
 K  – , fig. :
 Sb    Sb  (I ) 
 Sb  (E )   AS  S.
 Sb  (K )   MT  
 Sb  (H )   Sb  
 AS  S.  Sb  (H ) 
 K  –  Sb  
 MDP  (), –  Sb  
, fig. :  Sb  (K ) 
 RA  (),  –  AS  S.
 AS  (, , , S.  Sb  (F ) 
, )  AS  S.
 AS  () S.  Sb  (K ) 
 RA  (),  –  Sb  (H ) 
 AS  () S.  AS  S.
 Sb  (H )   Sb  
 Sb    Sb  
 Sb    MT  
 AO  A.  Sb  
 Sb    MDP  (), –
 AS  S. , fig. :
 Sb    Sb  
 MT    AS A  S.
 MT ? ?  Sb  (E ) 
 Sb    Sb  
 Sb    Sb  
 Sb    Sb  
 AS  S.  Sb  
 AS  S.  MDP  (), –
 AS  S. , fig. :
 AS  S.  Sb  
 Sb  (G )   MDP  (), –
 Sb  (E )  , fig. :
 Sb  (H )   Sb  (E ) 
 AS  S.  AS  S.
 MT    Sb  
 Sb  (H )   AS  S.
 Sb  (H )   Sb  
 MT   AS  S.
 Sb    AS  S.
 Château de Suse,   Sb  
Nr.   Sb  (G ) 
 Louvre   Sb  (K ) 
 AS  S.  AS  S.
 Sb  (K )   Sb  (F ) 
 AS  S.  AS  S.
 Sb    Sb  
 Sb    AS  S.
 Sb    AS  S.
 AS  S.  Sb  
 Sb    Sb  
 georg neumann

Börker- Börker-
Klähn Fund-/ GS, Dela- Klähn Fund-/ GS, Dela-
 Museums-Nr. porte   Museums-Nr. porte 

 AS  S.  Sb  

 Sb    AS  S.
 Sb    AS  S.
 MDP  (), –  AS  S.
, fig. :  Sb  (F) 
 MT    Sb  (K ) 
 Sb    Sb  
 AS  S.  Sb  
 Sb    MET .. JNES , 
 AS  S. fig. 
 AS  S.  AS  S.
 Sb  (K )   AS  S.
 Château de Suse, Nr.   AS  S.
–, Sb   AS  S.
 Sb  (K )   Sb  (K ) 
 AS  S.  AS  S.
 AS  S.  Louvre 
 Sb    AS  S.
 AS  S.  AS  S.
 AS A  S.  Sb  (E ) 
 IN THE SHADOW OF KURANGUN:
CULTURAL DEVELOPMENTS IN THE HIGHLANDS
BETWEEN KHUZESTAN AND ANŠAN

Daniel T. Potts*

. Introduction

On April th, , Ernst Herzfeld entered the Mamasani region of western Fars, an area he
had first visited in .1 In his Reisebericht, published in , Herzfeld suggested that no
European had visited it in the interim, and that it was even then difficult to access because of
the wildness of its Kurdish inhabitants.2 He noted that the first time he had crossed the Fahliyan
plain, ‘ganz jung’, he had not even noticed what he called ‘eine ganze achämenidische Stadt, mit
Säulenbasen und anderen Architekturtrümmern auf dem Boden’—a site we now call Qaleh
Kali, Tepe Survan or Jinjun—as well as, ‘im Hügel, auf dem das Dorf Tulespid liegt, Mauerreste
aus elamischer Zeit mit Ziegeln in elamischer Keilschrift, wohl um – v. Chr.’, referring
to an inscribed brick recording the building of a temple to Kilah-Šupir by the Middle Elamite
king Šilhak-Inšušinak (Herzfeld : –).
One of the many insights that Herzfeld had during his second visit to the area stemmed from
a mistaken attribution. In describing, for the first time, the rock-cut tomb at Da-o Dukhtar,
he attributed it to an ancestor of Cyrus the Great’s whom he identified as a king of Anšan.
The discovery of Da-o Dukhtar was, he felt, as important historically and geographically as it
was from the standpoint of the history of architecture because it contributed to the solution of
the question of where the unknown land of Anšan was located.3 Later, in his  Schweich
Lectures, Herzfeld refrained from further historical speculation on Da-o Dukhtar, limiting
himself to a paragraph on its place in the history of rock-cut tombs in Iran. He was more precise,
though still wrong, about the date, placing the tomb between  and bc, but said nothing
about its significance for the question of Anšan’s location (Herzfeld : ). By , when
Herzfeld completed the text of The Persian Empire, speculations linking Da-o Dukhtar with the
Anšan question had vanished completely and in discussing the location of Anšan, he instead
drew attention to the Akkadian king Maništušu’s expedition in which he crossed the Lower
Sea. Reckoning that Maništušu must have embarked from Rešahr or Bušehr, where inscribed

* Institute for the Study of the Ancient World, New York University.
1 This is an only slightly revised version of the lecture delivered orally at the Susa and Elam conference.
2 ‘Von Shāpūr aus brach ich am . April ins Mamasēni-Gebiet auf, in dem ich  gewesen war, und das seither

nie von einem Europäer besucht war. Es ist auch jetzt noch ein wegen der Wildheit seiner kurdischen Bewohner
schwer zugängliches Gebiet’ (Herzfeld : ).
3 ‘Und endlich, weiter westlich [from Kurangun], in dem sehr unzugänglichen Gebiet von Khāk i Rustam,

ein vorachämenidisches Königsgrab, ein Zwischenglied zwischen den medischen Gräbern und denen von Naqsh
i Rustam und Persepolis, mit proto-ionischen Säulen. Es muß wohl einem der Vorgänger des Kyros, einem Könige
von Anshan gehören, und ist geschichtlich und topographisch von ebenso weittragender Bedeutung wie architek-
turgeschichtlich: denn es trägt bei zur Lösung der Frage nach dem unbekannten Lande “Anshan”, und es beweist,
zusammen mit dem Bilde des Tempels von Musasir aus dem Sargonspalaste, daß die gleichen Architekturfor-
men, die im Westen, an der Küste Kleinasiens, als ionischer Stil in Erscheinung treten, über ganz Kleinasien und
das westliche Iran, bis nach Fārs hin verbreitet waren, und daß die Beziehungen zwischen achämenidischer und
ionischer Architektur nur auf Grundlage dieser Anschauung verstanden werden dürfen, nicht wie es immer geschah,
als direkte Einwirkungen griechischer Architektur auf die iranische’ (Herzfeld : ).
 daniel t. potts

Elamite bricks had been found at Tol-e Peytul (ancient Liyan) he noted, ‘Thus Anšan, from the
time of Cyrus back to Maništusu of Kiš, was always modern Fārs’ (Herzfeld : ).
About  years later, of course, Maurice Lambert published a late Middle Elamite brick
inscription of Hutelutuš-Inšušinak’s, commemorating the construction of a temple to Napiriša,
Kiririša, Inšušinak and Simut, at Anšan, that was said to have come from somewhere between
Persepolis and Širaz (Lambert ), and the following year Erica Reiner published her well-
known paper on the location of Anšan, which had finally been confirmed by similar inscribed
bricks mentioning Anšan from Tal-e Malyan (Reiner ). All of this is well-known, and it is
not my intention to dwell any further on Anšanite historiography, except perhaps to point out
that, in defining Anšan’s geographical extent on the basis of the cuneiform inscriptions found
near Bušehr, Herzfeld was seemingly unaware that the same logic had been applied in 
and again in  by A.H. Sayce who invoked the bricks discovered in  near Bušehr by
the English East India Company troops in their war against Qajar Iran when discussing the
location of the toponym Anšan in the Cyrus Cylinder and, mistakenly, in the Kul-e Farah and
Eškaft-e Salman Elamite inscriptions (cf. Potts a).
There is, therefore, well over a century of scholarship, much of it written by people who never
visited the region, attempting to sketch the broad outlines of Anšan, and of course with the
eventual excavations at Tal-e Malyan, a great deal of generalizing speculation has been replaced
by actual data of many kinds. But prior to  it was still the case that, aside from studies of
rock reliefs like Kurangun, Da-o Dukhtar, and Sarab-e Bahram, or Sasanian sites further south
such as Bišapur and Firuzabad, very little was known about the intervening area between Susa
and Anšan. Sir Aurel Stein’s report of his travels through western Fars; studies by Louis Vanden
Berghe, Eric de Waele and others at Kul-e Farah and Eškaft-e Salman; soundings by Donald
McCown at Tal-e Ghazir; survey by Liz Carter and Henry Wright around Ram Hormuz; survey
around Behbehan and soundings at Tepe Sohz by Hans Nissen; and finally intensive survey and
soundings by Abbas Moghaddam and Negin Miri along the Ghaggar river east of Šuštar, are all
important but they still leave us with a large blank between Behbehan and Malyan. Trying to
understand what was going on in that intervening area was, in fact, one of the rationales behind
the decision in , by Kourosh Roustaei, Cameron Petrie, Lloyd Weeks, and me, to work in
the Mamasani region, between Pol-e Pirim in the west and Nurabad-e Mamasani in the east.
The results of excavations at Tol-e Spid and Tol-e Nurabad, each of which has a stratigraphic
and ceramic sequence from the Neolithic through to the Parthian or post-Achaemenid period,
with over  C dates, were reported on in a  page monograph published in , which
has since appeared in a revised edition (Potts et al. a). Environmental work, including the
coring of lakes in the area by Matthew Jones and Lloyd Weeks, is giving us an excellent climatic
history while excavations at Qaleh Kali, also known as Jinjun or Tepe Survan, have revealed a
major Achaemenid building with columned portico (Potts et al. , b). Here I do not
wish to review the entire project. Rather, in view of the theme of this conference, I wish to
concentrate only on a few issues that are fundamentally related to cultural identity in this part
of Iran and how that relates to the region’s position between and relations with the better known
centres of Elam and Anšan.

. Cultural Identity in the Mamasani Region

Herzfeld’s initial enthusiasm for the identification of the Mamasani region with Anšan was
based, firstly, as noted, on the presence of a tomb which he attributed to one of Cyrus’ ancestors,
known to have been ‘king of Anšan’; and secondly, one can probably assume, on the discovery
of the inscribed brick of the Middle Elamite king Šilhak-Inšušinak at Tol-e Spid, which clearly
 in the shadow of kurangun 

associated the area in his mind with both Anšan and Elam. It was not until our excavations at
Tol-e Spid and Tol-e Nurabad began, however, that we could actually investigate how closely
the material culture of the region did or did not match that known from Tepe Farukhabad,
Susa, Haft Tepe and Choga Zanbil in Khuzestan, and Tal-e Malyan, Darvazeh Tepe, Tal-e Šoga,
Tal-e Teimuran and Tal-e Qaleh in Fars, in other words, what are conventionally accepted as
Elamite and Anšanite assemblages. Before we examine this, however, I would like to briefly
say a few words about the pre-Elamite periods, by which I mean those prehistoric periods in
which we have no evidence that would allow us to determine whether or not the inhabitants of
southwestern spoke Elamite.
The study of the Neolithic material from Tol-e Nurabad (Weeks et al. ) showed that
relatively little pottery of Muški and considerably more Jari-type (Lashkari et al. ), as
known in Marv Dašt, appears in the Mamasani region and of course without analysis we cannot
say whether those sherds were locally made or imported. But the basic conclusion is that the Kur
River basin sequence, based on Tal-e Muški, Tal-e Jari and now Tol-e Baši (Pollock et al. ),
is simply inadequate to characterise the material from Nurabad where the range of diversity
in the Neolithic is much greater than can be encompassed by the Marv Dašt sequence. The
same goes for sequences further west. While a few parallels can be drawn with Hans Nissen’s
excavations at Qale Rostam in the Bakhtiari mountains, and Čoga Miš in Archaic Susiana
levels, the conclusion remains the same—there is a higher degree of regional variation than
commonality.
During the Chalcolithic one might question whether this continued to be the case. While
there are relatively few parallels with the west, there are very strong parallels with the Bakun/
Gap sequence (Weeks et al. ) while our stratigraphic sequence has allowed my collabo-
rator Cameron Petrie to differentiate earlier Lapui, later Lapui, transitional Lapui-Baneš and
Baneš assemblages at Tol-e Spid (Petrie et al. ). These parallels, however, should not be
interpreted as signs of either exchange or distribution from a common source by other means,
since unpublished analyses by Cameron Petrie have shown that all of the Bakun, Lapui and
Baneš-looking material from Tol-e Nurabad and Tol-e Spid that has been analysed thus far was
locally produced in the Mamasani region (C.A. Petrie, pers. comm.).
When we come to the rd and nd millennia, however, we continue to see regional distinc-
tions. What we found was a relatively small number of what might be termed generic Elamite
diagnostics of the sort that appear in both Khuzestan and Fars during the nd and early st mil-
lennium, and a large number of types that are clearly local, suggesting, if one wishes to define
cultural zones on the basis of ceramic types, that the Mamasani region is distinctly different than
its neighbours to the east and west. Amongst the painted material of early nd millennium date,
which might be considered more Anšanite than Elamite because it is found in Fars but not in
Susiana, there are certainly parallels with the Kaftari tradition, leading those members of our
team that have worked on the material from Tol-e Nurabad and Tol-e Spid, to describe it as
Kaftari-related in the absence of proof that it was made either locally or in Marv Dašt (Petrie
et al. ). Similarly, there is material of the same genre from the soundings at Tol-e Peytul
excavated by Maurice Pézard in  and even further afield on Bahrain and in eastern Arabia,
where the few examples are clearly exports and have nothing to do with the local ceramic tra-
ditions (Potts ; Laursen : ). But while immediately recognisable as Kaftari-related,
the material from Tol-e Spid and Tol-e Nurabad never looked quite like true Kaftari material
from Marv Dašt, and the reason became clear in , for in that year Kouroush Alamdari and
Kourosh Roustaei, both of whom have worked with us in Mamasani, and two colleagues, pub-
lished a short, preliminary account of the excavations at a cemetery near Lama, a site c.  kms.
north of Yasuj in Kuhgiluyeh-Boirahmed, to the north of Mamasani, or roughly  kms north
of Tol-e Nurabad (Rezvani et al. ). The parallels between Lama (Rezvani et al. ) and
 daniel t. potts

Tol-e Spid/Tol-e Nurabad appear much closer than between either of our sites and Tal-e Malyan
or Tol-e Peytul, particularly the unusual device of vertical brush strokes over the horizontal
bands on the shoulder of a vessel, and the triangles and modified Maltese cross or hour-glass
panel, suggesting the existence of a ceramic province within this region of western Fars and
Kuhgiluyeh-Boirahmed, that was clearly different than the classical Kaftari of Marv Dašt, or
the Kaftari-related of Tol-e Peytul, Fasa, Tal-e Nakhodi and other areas.
The later second millennium is characterised by even more regionalism, judging by the num-
bers of unparalleled painted wares encountered both in our area and in the Tal-e Teimuran/
Tal-e Šoga traditions. In  Bruno Overlaet published an account of Vanden Berghe’s sound-
ings at these sites (Overlaet ), and ten years later followed this up with the publication of
his soundings at Tal-e Kamin (Overlaet ). These painted pottery styles are clearly differ-
ent than the generic Middle Elamite material at Tal-e Malyan and its Qaleh successors. The
impression is certainly strong that in Mamasani, as in Marv Dašt, painted pottery traditions
were very much alive at this time, but I don’t think we can go so far as to agree with Nicol and
Sumner that the Šogha/Teimuran wares represent the arrival of the Persians. From my perspec-
tive these wares show too much continuity with the preceding Kaftari and Lama traditions to
support such a hypothesis, unless one wants to suggest that the Persians arrived not in the late
nd millennium but in the late rd millennium bc with the start of the Kaftari phase in the
Marv Dašt. That would present many complications, however, for our understanding of Anšan
in the sukkalmah period, unless one wanted to suggest this was a time when Persians were in
the ascendancy, only to be replaced by Elamites in the Middle Elamite/Qaleh phases later on.
But all of this is mere speculation. What this rapid review of the ceramic evidence does show,
however, is the originality of the Mamasani-Kuhgiluyeh-Boirahmed corridor vis-à-vis areas to
the west and to the east, at a time which sits squarely within the periods historically associated
with the Šimashki, sukkalmah and Middle Elamite phases of Elamite history. By originality I do
not for a moment mean to suggest isolation, simply that what we see in our region is different
than, although similar to, developments in neighbouring areas. This is not to be wondered at,
for even if Elamite or Anšanite political hegemony enveloped the region between Anšan and
Susa, there is no reason to think that local cultural differences were completely suppressed.

. Elamite or Anšanite Cultural Hegemony?

Two indications that this cultural corridor involved a real Elamite presence, and extended even
further to the north, are provided by an unpublished Middle Elamite inscribed stone object,
perhaps a large macehead, found near Yasuj (K. Alamdari, pers. comm.), and an inscribed brick
from Tol-e Afghani, near Lordegan, found some years ago by Mr. Norouzi of the Šahr-e Kord
office of the Iranian Cultural Heritage and Tourism Organization (Anonymous –).
Until these have been studied properly it would be unwise to say too much about them, but the
Tul-e Afghani brick, of Hutelutuš-Inšušinak, raises interesting questions about his activities
in the east, since we already know that he was active at Anšan, to which it was thought by
Maurice Lambert that he had retreated after being driven out of Susa by Nebuchadnezzar I (cf.
Potts : –). Whether a site like Tol-e Afghani, where Hutelutuš-Inšušinak presumably
engaged in temple building, was something like a summer residence for this late Middle Elamite
king we can only speculate. Certainly it would have made a mountain retreat well away from
the reach of any possible Babylonian enemy.
At this point it is appropriate to turn to the rock-relief at Kurangun, mentioned in the title
of this talk. From the perspective of height, this relief clearly dominates the Dašt-e Rustam-e
Yek where Tol-e Spid is located. What can it contribute to a discussion of cultural autonomy
 in the shadow of kurangun 

or assimilation in the Mamasani region during the nd millennium bc? First brought to the
attention of the scholarly world by Herzfeld as a result of his  visit to the region, Kurangun
was discussed briefly in Debevoise’s  paper on Iranian rock reliefs (Debevoise : –
) and was later the subject of studies by Louis Vanden Berghe in  (Vanden Berghe
) and Wolfram Kleiss in  (Kleiss ). The most detailed publication, however, is
Ursula Seidl’s  authoritative fascicle in the series Iranische Denkmäler (Seidl ). I do not
wish to dwell on the date of the relief, which has been discussed extensively by Pierre Amiet
and others, beyond noting that, based on parallels with glyptic representations from Susa, the
central deity on the serpent throne strongly suggests a date in the sukkalmah era, i.e. in the early
nd millennium bc. This corresponds ceramically, therefore, with the later Kaftari period we
have just discussed. Wolfram Kleiss was certainly rash, however, to suggest that the very visible
remains of stone walling just north of the relief were as old as the relief. Soundings conducted
by Ardashir Javanmard Zadeh and me in the main building in  revealed late material,
principally Sasanian and Islamic in date. With their massive stone walls, these buildings are
more likely to represent the remains of a heavily fortified lookout post over the main road
running through the valley from Istakhr or Širaz to Ahwaz and Susa, than a religious structure
of some sort. Some of the material recovered may be Elamite and Achaemenid, and it is obvious
that people have been climbing up to the relief at least since it was carved in the early nd
millennium, but nothing suggests the buildings are pre-Sasanian and they certainly do not
represent a temple or buildings with ritual purposes associated with the relief, tempting though
such a possibility seemed before we opened our sondage there.
The deities in the centre of the main panel of the Kurangun relief—interpreted as such on
the basis of their horned headgear—have been discussed by many scholars, including Walther
Hinz, Pierre Amiet, Pierre de Miroschedji, Madeleine Trokay, Françoise Grillot, François Vallat,
Louis Vanden Berghe, Ursula Seidl, Mark Garrison and myself. In spite of some diverging views,
the similarities are clear between the seated deities, their headgear, and the serpent thrones
at Kurangun and in seal impressions from Susa dating to the reigns of Atta-hušu, Šilhaha,
Kuk-našur II and Tan-Uli, and the badly destroyed rock relief at Naqš-e Rustam, which can
barely be seen because of the later carving of a relief honouring Bahram II from the late rd
century ad (Potts : – with earlier lit.). Even if the serpent throne appears later on
seals of Tepti-Ahar from Haft Tepe, it is clear that the motif was typical of sukkalmah-era glyptic.
Most scholars have agreed that the deities depicted are either Elamite or Anšanite, by which I
mean, in this context, deities of lowland Khuzestan or highland Fars. Hinz identified the deities
as Humban and Kiririša or Parti; Vanden Berghe and Amiet suggested they were Napiriša and
his consort (confirmed by EKI : ) Kiririša; and de Miroschedji identified the male deity as
Inšušinak, largely because of the fragmentary inscription on the upper portion of the Untash-
Napiriša stele from Susa, brought there from the siyan kuk at Čoga Zanbil, and dedicated, in all
likelihood, to Inšušinak.
In his review of R.T. Hallock’s Persepolis fortification texts, Walther Hinz suggested that
Kurangun or Tol-e Spid might be identical with a place called Zila-Humban in the Elamite
texts from Persepolis, interpreting zila as ‘relief ’, and identifying the central male deity in
the Kurangun relief with Humban (Hinz : –). As Wouter Henkelman has shown
recently, however, this interpretation of zila is far from certain (Henkelman : ) and
Hinz, apparently persuaded by Heidemarie Koch, who felt this location was too far east,
rescinded this identification before his death.
In the Festschrift for Peder Mortensen I referred to François Vallat’s suggestion that, in
MDP , no. , Ea and Enzag functioned as epithets of Inšušinak, and that as Ea was equated
with Napiriša in the incantation series Šurpu, perhaps de Miroschedji, Amiet and Vanden
Berghe were all correct in identifying the male deity at Kurangun as Inšušinak and Napiriša
 daniel t. potts

(Potts ). However, there are other texts, such as the brick from Malyan published by
Lambert, in which Inšušinak and Napiriša are clearly distinguished from one another, as indeed
they are at Choga Zanbil (de Miroschedji : ; Potts b: –, no. ). In the
end, even though many scholars have argued that Inšušinak and Napiriša are the most likely
candidates, Ursula Seidl suggested that none of the previously suggested identifications—by
Hinz, Amiet or de Miroschedji—stood up well enough to critical scrutiny to be acceptable and
in fact she questioned whether the relief necessarily represented the well-known Elamite deities.
This problem is not going to be resolved by me or anyone else unless some epigraphic or much
better iconographic data comes to light. It is important to stress that this is not simply a matter
of religion, however, but rather one of broader cultural hegemony.
Was the Mamasani region, in which Kurangun is located, subject to the cultural hegemony
of Susa or of Anšan during the sukkalmah period? Is such a distinction even tenable? Or are the
religious icons of Kurangun, which are clearly shared both to the west at Susa and to the east at
Naqš-e Rustam, either indigenous to the Mamasani area, or part of a broadly Elamite-Anšanite
cultural milieu that encompassed a zone extending from Deh Luran in the west to Marv Dašt
in the east, north to Tol-e Afghani and south to Liyan on the Persian Gulf?
The signs of this cultural hegemony are clear. Inšušinak, for example, whose name appears
as early as the mid-rd millennium bc (dnin-šušinak) in a list of divine names from Abu
Salabikh, was, as the etymology of the name suggests, the city-god of Susa, where there are
at least  different brick inscriptions in hundreds of copies extending in time from the post-
Akkadian and Ur III periods to the Neo-Elamite era, attesting to his popularity (Potts b:
–). Outside of Susa, shrines to Inšušinak are attested epigraphically at Čogha Zanbil,
Tappeh Deylam and Čogha Pahn West in Khuzestan, and at Tal-e Malyan in Fars, all of which
date to the Middle Elamite period (Potts b: Table ). Therefore, if the Kurangun relief does
depict Inšušinak, this could, in one sense, be interpreted as a sign of the spread of the cult of
Inšušinak into the highlands between Susa and Anšan, by whatever means. By the time we
reach the Middle Elamite period, though, when Inšušinak was honoured at Tal-e Malyan, the
god may have been worshipped for  years or more in the highlands to the east of Khuzestan
and should probably no longer be considered specifically Susian but rather a fully assimilated,
if originally foreign, deity from the lowlands of Susiana.
What of Napiriša and Kiririša, whom Amiet, Grillot and Vallat have seen as particularly
Anšanite? In Pierre Amiet’s words, Napiriša was ‘chef du panthéon, honoré particulièrement
au pays d’Anšan’ (Amiet : ). Grillot and Vallat considered Napiriša and Kiririša the
divine couple Anšanite par excellence (Grillot and Vallat : –). In her edition of
the inscribed bricks from Susa in the Louvre, Florence Malbran-Labat somewhat bizarrely
characterised Kurangun as a rock sanctuary ‘near Persepolis’ (‘près de Persépolis’) ‘où les
populations nomadisantes du haut pays d’ Anšan venaient honorer leur grand dieu Napiriša’,
depicted in the central panel with his consort Kiririša (Malbran-Labat : ). The case of
Napiriša is not clear. Two different shrines to Napiriša at Susa are attested by brick inscriptions,
while one is known to have existed at Liyan and at Tal-e Malyan (Potts b with refs.). As
noted already, Napiriša was worshipped along with several other deities in the siyan tarin.
Although Napiriša is mentioned in a contract from Susa believed to date to the early sukkalmah
period and in a small number of later texts (de Miroschedji : ), the shrines to him there
were both founded in the Middle Elamite period and Pierre de Miroschedji has suggested, as
indeed the name suggests, that Napiriša was the personal god of Untaš-Napiriša, the famous
builder of Čoga Zanbil, who introduced the god into Susiana from Anšan, where he was the
great god of the Anšanite pantheon (cf. Roche : , n. ), just as Inšušinak was of the
Susian pantheon. Čoga Zanbil, in which so many deities were honoured, was interpreted by
Malbran-Labat as a religious centre symbolising the unification of Anšan and Susa.
 in the shadow of kurangun 

As for Kiririša, this goddess had shrines at Susa and Čoga Zanbil, as well as one at Liyan
that was restored repeatedly by successive Middle Elamite kings beginning with Humban-
numena. Kiririša is also mentioned in a curse formula at Ayapir/Eškaft-e Salman (EKI : ;
cf. Henkelman : , n. ). Kiririša was most often referred to as Kiririša-of-Liyan, even
in texts from Susa (e.g. IRS ), and it is very possible that Liyan was her city. On analogy
with homonymous Mesopotamian deities, such as Ištar-of-Uruk, Ištar-of-Babylon, Ištar-of-
Akkad, Ištar-of-Arbela and Išhtar-of-Nineveh, who are often considered local manifestations
of a single deity (Beaulieu : –), it is possible but not absolutely certain that we
should distinguish between Kiririša, when mentioned without her geographical epithet, and
Kiririša-of-Liyan, both of which forms are attested at Susa. At Liyan itself, and at Choga Zanbil,
the goddess is simply called Kiririša without the addition of her city’s name. One could naturally
suggest that Kiririša was simply the tutelary goddess of Liyan. Archaeological sites—albeit
unexcavated ones—where stray inscribed bricks have been picked up, also provide references to
specific deities, such as Ruhurater at Tol-e Bormi; Manzat at Deh-e Now and Tappeh Horreeye
near Čoga Zanbil; Upurkupak at Tappeh Gotvand and Čoga Pahn West; Mašti and Tepti at
Deylam (Vallat : ); and Kilah-Šupir at Tol-e Spid and Čoga Zanbil (Potts b: ,
no. ), but this does not necessarily mean that those deities were particularly or exclusively
associated with those towns. Whether Kilah-Šupir was indigenous to the Mamasani area, we
do not know, but certainly there was a shrine to Kilah-Šupir as well as at Čoga Zanbil. If one
views Čoga Zanbil as a project undertaken by Untaš-Napiriša to create an artificial cult centre
honouring the deities of the entire Elamite-Anšanite world, then one certainly could suggest
that Kilah-Šupir was indigenous to the Mamasani region, perhaps even Tol-e Spid in particular,
and transplanted from there, along with many other non-native (i.e. not native to Khuzestan or
Susiana) deities to Čoga Zanbil.
Although we cannot identify the Fahliyan river with one of the hydronyms in the Elamite
texts from Susa or Persepolis (Vallat : –), five rivers—Ayanharišda, Hubutiš, Marriš,
Rannakarra and Šaušanuš—received allocations of commodities or had lan and other cer-
emonies performed for them, as discussed extensively by Wouter Henkelman (). This,
combined with the iconography of water and the flowing stream at the base of the Kurangun
relief, suggests the possibility of considerable continuity in religious praxis in Fars from the nd
through the st millennium bc, continuity that is also attested by the addition of the frieze of
worshippers descending steps, much like the steps that descend from the crest of the hill to the
relief of Kurangun itself, during the Neo-Elamite period.
In the end, even though we cannot positively identify the deities represented in the Kurangun
relief, their headger and snake-thrones, which can be found at Susa, Kurangun and Naqš-e
Rustam, do suggest a shared iconography extending from Anšan through the Mamasani region
to Susa, as do the pious activities of Humban-numena, Šutruk-Nahhunte, Kutir-Nahhunte and
Šilhak-Inšušinak at Liyan, Anšan and Tol-e Spid. Because of the later role of Bušehr, as the
gateway to Širaz and Fars Province, we often think of Liyan as the gateway to Anšan, but we
should also remember that the route up from the coast, via Kazerun, leads on to Mamasani and
westward towards Khuzestan as well as eastward towards Anšan. Kurangun and the Middle
Elamite Kilah-Šupir temple that must have existed at Tol-e Spid, are thus located squarely
between the two poles of Anšanite and Susian cultural and political power. Everything we know
today, which is admittedly not as much as one would like, confirms Herzfeld’s early suspicions
of an Elamite or Anšanite presence in the Mamasani region in the nd and st millennia bc
and of the intrinsic importance of this region for our understanding of pre-Achaemenid Iran.
 daniel t. potts

Abbreviations

AfO Archiv für Orientforschung

EKI König 
IrAnt Iranica Antiqua
RA Revue d’Assyriologie

Bibliography

Anonymous. (–): A guide of the historical sites and ancient monuments of Chaharmahal va
Bakhtiari, Shahr-e Kord.
Amiet, P. (): “Glyptique élamite: A propos de documents nouveaux”, Arts Asiatiques , –.
Beaulieu, P.-A. (): The pantheon of Uruk during the Neo-Babylonian period (Cuneiform Monographs
), Leiden—Boston.
Debevoise, N.C. (): “The rock reliefs of ancient Iran”, JNES , –.
Grillot, F. & Vallat, F. (): “Dédicace de Šilhak-Inšušinak à Kiririša”, IrAnt , –.
Henkelman, W.F.M. (): The other gods who are: Studies in Elamite-Iranian acculturation based on the
Persepolis fortification texts (Achaemenid History ), Leiden.
Herzfeld, E. (): “Reisebericht”, ZDMG , –.
Herzfeld, E. (): Archaeological history of Iran, London.
Hinz, W. (): “Die elamischen Buchungstäfelchen der Darius-Zeit”, Orientalia , –.
Kleiss, W. (): “Kurangun, die Burganlage am elamischen Felsrelief in Südwest-Iran”, M. Mellink,
E. Porada & T. Özguç (eds), Aspects of art and iconography: Anatolia and its neighbors, studies in honor
of Nimet Özguç. Ankara, –.
König, F.W. (): Die elamischen Königsinschriften (AfO. Beih. ), Graz.
Lambert, M. (): “Hutélutush-Insushnak et le pays d’Anzan”, RA , –.
Lashkari, A., Javanmard Zadeh, A., Potts, D.T., Weeks, L.R. & Taylor, H. (): Report on the 
excavations at Tol-e Nurabad, Mamasani, Iran. Tehran (Unpublished report submitted to the Iranian
Center for Archaeological Research).
Laursen, S.T. (): “The emergence of mound cemeteries in Early Dilmun: New evidence of a proto-
cemetery and its genesis c. –BC.”, L. Weeks (ed.), Death and burial in Arabia and beyond:
Multidisciplinary perspectives (Society for Arabian Studies Monographs ) Oxford, –.
Malbran-Labat, F. (): Les inscriptions royales de Suse. Paris.
Overlaet, B. (): “A report on the  and / soundings at Tall-i Taimuran (Fars), Iran”, IrAnt
, –.
———. (): “Soundings at Tall-i Kamin (Kur River basin), Fars, Iran”, IrAnt , –.
Petrie, C.A., Asgari, A. & Seyedin, M. (): “From Anshan to Dilmun and Magan: The spatial and
temporal distribution of Kaftari and Kaftari-related ceramic vessels”, Iran , –.
Petrie, C.A., Sardari Zarchi, A., Alamdari, K. & Javanmard Zadeh, A. (): “Transformations in fourth
millennium B.C. Fars: Further excavations at Tol-e Spid”, Iran , –.
Pollock, S., Bernbeck, R. & Abdi, K. (): The  excavations at Tol-e Baši, Iran: Social life in a
Neolithic village (Archäologie in Iran und Turan ), Mainz.
Potts, D.T. (): The archaeology of Elam: Formation and transformation of an ancient Iranian state,
Cambridge.
———. (): “Anshan, Liyan and Magan c. B.C.”, N. Miller & K. Abdi (eds.), Yeki bud, yeki nabud:
Essays on the archaeology of Iran in honor of William M. Sumner, Los Angeles, –.
———. (): “The numinous and the immanent: Some thoughts on Kurangun and the Rudkhaneh-e
Fahliyan”, K. von Folsach, H. Thrane & I. Thuesen (eds.), From handaxe to Khan: Essays presented to
Peder Mortensen on the occasion of his th birthday, Aarhus, –.
———. (a): “A note on the limits of Anšan”, J. Alvarez-Mon & M.B. Garrison, (eds.), Elam and Persia.
Winona Lake, –.
———. (b): “Elamite temple building”, M.J. Boda, M.J. and J.R. Novotny, (eds), From the foundations
to the crenellations: Essays on Temple building in the Ancient Near East and Hebrew Bible (AOAT )
Münster, – and –.
Potts, D.T., Roustaei, K., Petrie, C.A. & Weeks, L.R. (): The Mamasani Archaeological Project, Stage
One: A report on the first two seasons of the ICAR—University of Sydney Expedition to the Mamasani
District, Fars Province, Iran, nd ed., Oxford.
 in the shadow of kurangun 

Potts, D.T., Asgari Chaverdi, A., McRae, I.K., Alamdari, K., Dusting, A., Jaffari, J., Ellicott, T.M., Setoudeh,
A., Lashkari, A., Rad, S.A. & Yazdani, A. (): “Further excavations at Qaleh Kali (MS ) by the
Joint ICAR-University of Sydney Mamasani Expedition: Results of the  season”, IrAnt , –
.
Reiner, E. (): “The location of Anšan”, RA , –.
Rezvani, H., Roustaei, K., Azadi, A. & Alamdari, K. (): Lama ancient cemetery, Yasouj.
Rezvani, H., Roustaei, K., Azadi, A. & Ghezelbash, E. (): Final report of the archaeological excavations
at Lama Cemetery, Tehran.
Roche, C. & Overlaet, B. () (app. ): “Briques élamites du IIe millénaire av. J.-C. conservées
aux Musées royaux d’Art et d’Histoire de Bruxelles”, Bulletin des Musées Royaux d’Art et d’Histoire
/ Bulletin van de Koninklijke Musea voor Kunst en Geschiedenis , –.
Seidl, U. (): Die elamischen Felsreliefs von Kurangun und Naqš-e Rustam (Iranische Denkmäler
/II/H), Berlin.
Vallat, F. (): Les noms géographiques des sources suso-élamites (Répertoire Géographique des Textes
Cunéiformes ), Wiesbaden.
———. (): “Elam vi. Elamite religion”, EncIr , –.
Vanden Berghe, L. (): “Données nouvelles concernant le relief rupestre élamite de Kurangun”, L. de
Meyer, H. Gasche & F. Vallat (eds), Fragmenta historiae elamicae: Mélanges offerts à M.J. Steve, Paris,
–.
Weeks, L.R., Alizadeh, K., Niakan, L., Alamdari, K., Zeidi, M., Khosrowzadeh, A. and McCall, B. ():
“The Neolithic settlement of highland SW Iran: New evidence from the Mamasani District”, Iran ,
–.
Weeks, L.R., Petrie, C.A. & Potts, D.T. (): “#Ubaid-related-related? The “black-on-buff ” ceramic
traditions of highland southwest Iran” G. Philip & R. Carter (eds.), The Ubaid expansion? Cultural
meaning, identity and integration in the lead-up to Urbanism, –.
 ESSAI D’APPLICATION DE LA MÉTHODE
DE LA PSYCHOLOGIE ENVIRONNEMENTALE À TRAVERS L’ EXEMPLE DE
LA VILLE MÉSO-ÉLAMITE DE DÛR-UNTAŠ (TCHOGA-ZANBIL, IRAN),
SITE INSCRIT AU PATRIMOINE MONDIAL DE L’ UNESCO

Zsolt Gábor Lantos*

À la mémoire du Professeur Serge Cleuziou

. Définir le sujet, définir la méthode

.. Présentation du site

Au-delà des confins orientaux de la Mésopotamie, sur la rive gauche du Tigre, et s’ étendant à
travers la chaîne montagneuse du Zagros jusqu’ au plateau iranien (Hansman , Lambert
, Reiner a Carter/Stolper , Vallat ), fleurissait entre le XIVe et le XIIe siècle1
la Monarchie Confédérative de l’Élam (RLA  : –). L’ objet de notre étude est Dûr-Untaš,
une des villes nouvelles de l’époque de la Pax Mesopotamica2 fondée dans le cadre d’ un nouveau
programme de réformes – au plan politique, religieux3, social, et économique4 – par le souverain
élamite Untaš-Napiriša (–5), roi d’ Anšan et de Suse6. Il apparaît certainement comme

* Université Paris I –Panthéon –Sorbonne, Faculté Histoire de l’Art et Archéologie, Département d’Archéologie

Orientale et Université des Sciences Techniques et Économiques de Budapest, Faculté Architecture, Département de
l’ Histoire de l’ Architecture et des Monuments. Je profite de cette occasion pour remercier Behzad Mofidi-Nasrabadi
d’ avoir bien voulu attirer mon attention sur les récents résultats des recherches de son équipe effectuées à Tchoga-
Zanbil, qui sont présentés dans son ouvrage () intitulé Archäologische Ausgrabungen und Untersuchungen in
Čoga Zanbil.
1 Amiet  (–) appelle cette période «l’âge d’or de la civilisation élamite».
2 Je propose d’ employer l’ expression Pax Mesopotamica pour résumer la situation pacifique générale –grâce aux

bonnes relations internationales entre les grandes puissances indépendantes, qui étaient renforcées par des mariages
inter-dynastiques –caractérisant alors le Proche-Orient au XIVe siècle à l’époque des tablettes d’Amarna. À ce sujet
voir la correspondance diplomatique royale –presque toute écrite en une forme stylisée de babylonien, la langue
officielle internationale, soit la lingua franca de l’époque –dans Amarna Reports I–IV de Kemp et alii () et
la traduction française dans Les lettres d’ El-Amarna, correspondance diplomatique des Pharaons de Moran ().
Voir aussi les royaumes indépendants en correspondance avec l’Égypte dans les lettres d’Amarna sur la carte du
Proche-Orient dans l’Atlas de la Mésopotamie et du Proche-Orient Ancien de Roaf (/: –), l’article
sur « L’ Élam du IIe millénaire et la chronologie courte» de Vallat (: –) et le catalogue de l’exposition Babylone
( : – et ).
3 Ghirshman  : .
4 Kuhrt  : .
5 En dernier lieu voir l’ article sur « La chronologie méso-élamite et la lettre de Berlin» de Vallat (), qui

résume l’ essentiel à savoir à propos sur la problématique du règne; tout récemment une nouvelle interprétation de la
lettre de Berlin, à propos de la dynastie élamite des Igihalkides, a été proposée par Quintana dans sa communication
intitulée « Filiation et accès au pouvoir en Élam dans la e moitié du IIe millénaire» durant ce congrès.
6 Untaš-Napiriša introduit le dieu élamite Napiriša du plateau iranien (Anšan) dans le panthéon pour rapprocher

les deux composantes de l’ Élam (Anšan et Suse). En même temps, comme son père avant lui et comme son fils
ensuite, le roi épouse une princesse cassite, en l’occurrence la fille de Burnaburiaš II (–), pour conserver
avec le puissant voisin mésopotamien de bonnes relations politiques et économiques (Vallat : ); voir la
stèle d’ Untaš-Napiriša (Vallat  : –) où l’on aperçoit le roi entouré de son épouse Napirašu et probablement
de sa mère : cette stèle, avec les statues du roi et de la reine, sont les rares représentations des membres de la cour
 zsolt gábor lantos

le plus grand constructeur de l’histoire élamite, car il a construit ou restauré bons nombres
d’édifices religieux à travers toute la Susiane et l’ Élam (Steve/Vallat/Gasche / : –
). Le nom du fondateur de la ville fut identifié grâce à des inscriptions trouvées sur place (sur
des briques, des dalles de pierre, des plaques à pommeaux et des clous), à Suse mais aussi ailleurs
(briques7), rédigées majoritairement en élamite cunéiforme (de Mecquenem  : , Steve
, Ghirshman :  et : ) : « Je suis Untaš-Napiriša ». On a également rencontré son
nom sur d’autres objets, comme par exemple sa hache trouvée sur le site (Ghirshman : ,
Amiet : fig. ) et les fragments de statues8 trouvés à Suse. Il est tout à fait probable que
Dûr-Untaš soit la ville de Dūr-Undasi, mentionnée dans les annales assyriennes d’ Assurbanipal
(Streck : ,  et , Harmatta / : ). Le nom actuel de la ville méso-
élamite (–9) de Dûr-Untaš-Napiriša (« Fort du Grand Untaš » en akkadien) dans le
dialecte local est Tchoga-Zanbil, ce qui signifie « butte de corbeille » (de Mecquenem  : ,
Ghirshman : ). Cette dénomination rappelle l’ état du site avant le dégagement des restes
érodés de la ziggurat, quand la colline dominante s’ élevant au centre avait l’ aspect d’ un panier
renversé (Ghirshman , Pl. V.). On sait que le roi élamite Šutruk-Nahhunte Ier a rapporté de
Tchoga-Zanbil à Suse des stèles (König , nº  II), ainsi que des éléments architecturaux
(Vallat : ): en effet, une inscription de Šutruk-Nahhunte Ier (König , nº ) indique
que le roi «enlève les stèles que Untaš-Napiriša avait placées à siyan-kuk, ainsi que Inšušinak,
mon dieu, me le demandait …». Grillot et Vallat ( : , n. ) confirment l’ équation siyan-
kuk = Tchoga-Zanbil déjà suggérée par König ( : –, n. ). Les inscriptions portant
la mention siyan-kuk peuvent donc être attribuées à Tchoga-Zanbil (Ghirshman  : ,
Vallat :  et : ): on sait en effet que le mot siyan-kuk désignait spécifiquement
le temenos de Tchoga-Zanbil, c’est-à-dire le sanctuaire par excellence (de Miroschedji  :
). Le siyan-kuk est à Tchoga-Zanbil, ce que le kizzum sera plus tard à Suse, c’ est-à-dire
l’ensemble du complexe religieux (Steve/Vallat/Gasche / : ). Le site, découvert en
, a commencé à être fouillé par de Mecquenem. C’ est finalement Ghirshman (Kharkov,
 – Budapest, ) qui y entreprit d’ importants travaux de dégagement entre  et
. Dans les années , les techniques modernes utilisées par l’ équipe de Mofidi-Nasrabadi
ont profondément modifié notre connaissance de la ville. Le site se trouve dans la plaine de
Susiane, dans la province du Khousistan, qui correspond à la partie sud-ouest de l’ Iran. Ses
coordonnées géographiques sont: N °0,00 et E °0,00. La ville elle-même a été
érigée sur un vaste plateau bordé par de hautes collines au Nord et à l’ Est – qui atteignent
et dépassent en certains endroits une dizaine de mètres de hauteur – surplombant de plus
de  mètres la rive orientale de l’Âb-è Diz. Cet affluent du Karun, coulant actuellement à 
kilomètres au nord de la ville (de Mecquenem  : ), est connu sous le nom d’ Ididé dans les
annales assyriennes d’Assurbanipal (Streck  : , Harmatta / : , Ghirshman

méso-élamite parvenue jusqu’à nous. S’ il n’est pas possible de savoir laquelle des princesses cassites a été enterrée
dans la tombe IV du Palais-hypogée à Tchoga-Zanbil (même s’il est vraisemblable qu’il s’agit de la reine Napirašu),
au moins est-il possible, à la suite de la tablette publiée par Van Dijk en , de connaître sa nationalité (Ghirshman
 : – ; Vallat  : ).
7 Voir les villes où ont été retrouvées des briques au nom d’Untaš-Napiriša sur la carte dans l’Atlas de la

Mésopotamie et du Proche-Orient Ancien de Roaf (/: –).

8 Il est probable que ces statues ont été commandées par le roi et ont été fabriquées pour la décoration de la

ziggurat ou du kukunnum de Tchoga-Zanbil (Vallat : ). Elles ont été rapportées dans l’antiquité à Suse où
elles ont été retrouvées, comme un grand nombre d’objets destinés à Tchoga-Zanbil (Grillot/Vallat : , n. ).
9 Le plus important document est sans doute la lettre de Berlin (Van Dijk ), mentionnée à l’instant, qui

est décisive pour permettre à reconstituer une chronologie méso-élamite acceptable. Voir la chronologie révisée et
actualisée, ainsi que les différents points de vue des collègues, dans le Supplément au Dictionnaire de la Bible de
Steve/Vallat/Gasche (/ : – et –); en dernier lieu voir aussi la chronologie utilisée par Potts
( : ).
 essai d’application de la méthode de la psychologie environnementale 

 et ). L’importance du site justifie, qu’ il ait été inscrit parmi les premiers en Iran
sur la liste du patrimoine mondial de l’ Unesco en . Les recherches qui se poursuivent
à l’heure actuelle vont encore certainement apporter des nouveautés, et permettre de mieux
comprendre, l’occupation, l’évolution et l’ architecture urbaine de Dûr-Untaš. La méthode que
l’on se propose d’utiliser dans notre étude devrait elle aussi contribuer à résoudre le mystère de
la véritable nature du site …

.. Lire et comprendre la ville

L’objectif de cette étude est de présenter l’ intérêt de la psychologie environnementale (environ-
mental psychology10) développée par Lynch (a, b) pour l’ archéologie à travers l’ étude
de cas d’une ville antique, baptisée Dûr-Untaš par son fondateur : cette méthode principalement
utilisée par les architectes11 – mais également appréciée par des géographes, des anthropologues
et des sociologues (Lippai/Dúll : ) – aide à décrire et à mieux comprendre la nature
de l’environnement urbain. C’est à partir de la psychologie environnementale que Lynch a
posé les fondements de la synthèse esthético-topologique des approches urbanistiques et éta-
blit la sémiotique urbanistique (Vidor , Ankerl ). Les analyses de Lynch font intervenir
l’examen scientifique des images mentales produites par l’ environnement (Lippai/Dúll  :
). Ainsi, son travail a fourni à la fois un cadre théorique et une méthode pour l’ analyse scien-
tifique de la «cognition environnementale » (Holahan a/). L’ urbanistique, qui est la
science des villes, sera plus tard développée par Geddes (/) : cette méthode consiste
essentiellement à analyser les circonstances de la naissance, l’ histoire de l’ évolution, la struc-
ture spatiale des villes, ainsi que la façon dont la surface est répartie (Vidor ). Étant donné
que notre analyse porte avant tout sur la structure de la ville, on se contentera d’ appliquer la
méthode la plus adaptée à cette étude, celle de la psychologie environnementale, qui permet
de contribuer à la description et à une meilleure compréhension de la nature urbaine de Dûr-
Untaš. La lisibilité (legibility) de la ville reflète la structure de la physionomie urbaine (image of
the city). Pour avoir une vue d’ensemble lucide de la ville, il est nécessaire d’ analyser chacune de
ses composantes fondamentales. Les principaux éléments « doivent être aisément discernables
les uns des autres et doivent s’inscrire dans un schéma global » (Lynch b : , Lippai/Dúll
:  et ):
1. réseaux de larges avenues réalisées avec des matériaux de base de bonne qualité ;
2. limites (naturelles ou construites) nettes, morcelant la structure urbaine – mur(s) d’ en-
ceinte(s) comme moyen de délimitation et de défense, par exemple – et
3. qui encadrent de vastes espaces et secteurs à l’ intérieur de l’ enceinte de la ville ;
4. points stratégiques où se trouvent des
5. constructions remarquables (comme la ziggurat) aidant l’ orientation visuelle, et permet-
tant donc de se repérer.
D’après l’approche de Lynch (a, b, Vidor , Ankerl ), les éléments physiques
exceptionnels et immédiatement perceptibles de la physionomie urbaine de Tchoga-Zanbil
peuvent être répartis de la façon suivante (à ce propos, voir encore : Holahan b/, Séra
, Lawrence /, Bonnes-Secchiaroli  ; Dúll/Urbán , Dúll , , ,
):

10 Cette expression a été introduite par Ittelsonen  lors d’une conférence tenue à New-York, dans un exposé
intitulé « la psychologie environnementale et le projet de construction architectural».
11 Ce sont les psychologues qui ont d’ abord formulé cette problématique. Il faut cependant préciser que Lynch,

qui est à l’ origine du développement de la méthode de la psychologie environnementale, n’était pas un psychologue,
mais un urbaniste.
 zsolt gábor lantos

. Étude de la structure urbaine de Tchoga-Zanbil

.. Les voies (paths)

... Des voies dallées désaxées: une caractéristique élamite ?
.... Le cas des temples
Le sanctuaire élamite s’éloigne par certains détails de son « homologue » babylonien : son entrée
principale n’est jamais dans l’axe de la porte d’ entrée de l’ antecella, mais est décalée et en même
temps reliée à celle-ci par une voie dallée tracée de biais : depuis l’ entrée principale dans la cour
du temple de Pinikir du Complexe Est, un dallage de , mètres de largeur, fait de briques
cuites entières appareillées avec soin, venait buter obliquement contre la porte du bâtiment
central (Ghirshman : ). À partir de l’ entrée qui donnait sur la cour du temple IM et
Šala, un dallage très soigné, large de  mètres, traversait cette dernière de biais et venait buter
contre la porte de l’édifice central, identique à celui qui s’ élevait au même endroit dans la
cour du temple de la déesse Pinikir. Le long de ce dallage-trottoir, un embranchement dallé
menait vers une table d’offrandes en briques cuites cassées et l’ entourait (Ghirshman  :
). Une fois la porte dépassée, on se trouvait sur un dallage de , mètres qui traversait la
cour du temple de Šimut et NIN-ali du même complexe, toujours de biais, et butait contre
la porte d’entrée du bâtiment central qui était le sanctuaire proprement dit. Le long de ce
dallage-trottoir, un embranchement dallé menait ici aussi vers une table d’ offrandes en briques
cuites cassées et l’entourait (Ghirshman  : ). En revanche, dans le temple des Napratep –
le seul à avoir, parmi les quatre temples du Complexe Est, un type de plan élamite, selon
Ghirshman (: ) – le fidèle s’ engageait tout droit vers l’ entrée du temple, au lieu de
suivre un dallage oblique qui évitait d’ aborder la demeure divine de front : ce souci d’ intimité,
observable ailleurs à Dûr-Untaš, n’apparaît pas dans ce cas précis. La cour de la partie principale
du sanctuaire Hišmitik et Ruhuratir était elle aussi traversée, depuis la porte d’ entrée, par un
dallage oblique, large de trois briques. Au milieu de la cour, un embranchement du dallage, large
de , mètre, se détachait à gauche du passage oblique et atteignait la porte du petit bâtiment
central composé d’une antecella et d’ une cella. Le passage oblique traversait la cour et butait
contre la porte d’une autre antecella, élevée contre le mur extérieur nord-est du sanctuaire.
Ces deux branches distinctes du dallage menaient ainsi aux portes des deux sanctuaires, ce
qui est compréhensible puisque le temple était dédié à deux dieux (Ghirshman  : ).
La cour traversée, le dallage oblique battait contre la porte de l’ antecella du temple Kiririša
Ouest (Ghirshman : ). L’architecture religieuse élamite évitait, semble-t-il, les ouvertures
placées dans l’axe (Ghirshman  : ). Le temple de Dûr-Untaš, comme en Babylonie,
comprenait une antecella et une cella, mais, si en Mésopotamie (voir le temple de Tell-Harmal
ou celui de Ningal à Ur) les deux pièces étaient reliées par une porte percée dans l’ axe de l’ entrée,
en Élam cette porte se trouve être décalée (Baqir  : fig. , , , , Guardata/Dolce ,
fig.  et ). Ghirshman (: ) pense que le principe de l’ exposition des images cultuelles
qui devaient être vues de la cour, était commun aux deux pays. Cependant, si en Babylonie elles
étaient exposées dans la cella et vues à travers l’ antecella, à Dûr-Untaš les deux autels se trouvent
dans la première pièce que Ghirshman ( : ) désigne comme antecella. Ghirshman ( :
) présume que la seconde pièce était une cella où était placée une image cultuelle dans
une petite niche aménagée dans l’épaisseur du mur. Cette pièce pouvait peut-être servir aux
consultations de l’oracle auxquelles pensait Woolley ( : ), ou aux mystères pour lesquels
penche Lenzen (: ).
 essai d’application de la méthode de la psychologie environnementale 

.... Le cas des portes et des voies de circulation sur le parvis et le temenos
On a aussi constaté cette particularité devant la ziggurat où les portes de l’ enceinte des parvis
ne se trouvaient jamais dans l’axe des portes de la Tour (Ghirshman  : ). On peut
remarquer la même organisation avec les portes de l’ enceinte du temenos et de la ville. En
dehors de ces temples, des voies dallées – faites de dalles soit empierrées, soit en briques cuites
cassées ou entières soigneusement installées – tracées de biais sont observables dans les deux
secteurs religieux, qui pourraient indiquer une utilisation uniquement réservée au domaine
sacré, puisque le réseau de voirie est absent dans le secteur palatial, ainsi que dans le reste de
la ville elle-même (Ghirshman : ) : y en avait-il, ou est-ce le résultat de l’ érosion ? Il est
impossible de retrouver une hiérarchie urbaine par l’ intermédiaire des circulations (Margueron
: ). Regardons les exemples de dallages : un dallage, large de cinq briques, relie la porte
Nord-Est du parvis à celle de la ziggurat, mais cette voie ne se trouve pas strictement dans
l’axe et est légèrement déviée pour atteindre l’ angle Est de la triple rampe de la ziggurat. Selon
Ghirshman (: ), cette direction désaxée est certainement voulue. La porte Nord est
reliée par un dallage de quatre briques avec le « trottoir » Nord-Est de la ziggurat (Ghirshman
: ): cette voie n’est pas perpendiculaire par rapport au plan de la ziggurat et s’ oriente
de façon significative depuis la porte Nord vers le milieu du parvis Nord-Est, comme on l’ a
déjà constaté avec le dallage reliant la porte Nord-Est à celle de la tour, qui était aussi tracé en
biais. Dans les deux cas, cette direction désaxée est certainement voulue. La porte Est est reliée
à la «voie processionnelle» de circumambulation autour de la ziggurat par un dallage large de
quatre briques (Ghirshman : ). Un dallage légèrement oblique très large s’ étend depuis
l’escalier de la porte Sud-Est de la ziggurat jusqu’ au mur d’ enceinte du parvis (Ghirshman
: ). L’importance de la «porte des chars » se manifeste au fait qu’ elle était la seule de
toutes les portes du parvis à avoir un sol couvert de grosses dalles de pierre jointoyées au
bitume (Ghirshman : ). Une fois nettoyées, on a pu y reconnaître des traces de roues
de chars12, et c’est cette observation qui a conduit Ghirshman ( : ) à parler de « porte des
chars». Devant elle, sur le parvis, s’étendait un dallage rectangulaire soigneusement fait avec
des briques cuites entières, qui tranchait sur le reste du dallage, réalisé avec des briques cassées
(Ghirshman : ). Près de ce rectangle soigné, furent mises au jour les fondations d’ une
stèle qui avait été arrachée et emportée. La proximité de ces trois éléments (une porte dallée de
pierres, un carrelage rectangulaire et la trace au sol d’ une stèle), ne devait pas être accidentelle,
car l’on peut supposer que tout devait avoir une valeur religieuse, voire liturgique aux abords
immédiats de la Tour sacrée. Les ornières creusées par les roues des véhicules qui passaient par
cette porte posent un problème. Que venait-on chercher sur les parvis qui s’ étendaient devant
le temple «inférieur» dédié au dieu Inšušinak ? Les véhicules ne semblent pas avoir servi pour
le transport du monarque qui venait assister aux cérémonies religieuses. Le roi et la reine, suivis
de leur entourage, pénétraient sans doute sur le parvis par la monumentale « Porte Royale »,
dont les deux battants étaient probablement rehaussés d’ un revêtement particulièrement riche
et dont les murs intérieurs portaient, de chaque côté, des rangées de briques inscrites insérées
dans l’appareil en briques crues. L’hypothèse d’ un char transportant le couple royal jusque
sur le parvis paraît difficile à admettre. Le roi et la reine passaient sans doute à pied par la
plus belle porte et trouvaient immédiatement après, une jarre remplie d’ eau pour les ablutions,
trois tables d’offrandes et, selon Ghirshman, leurs sièges à côté. Les chars qui arrivaient sur le
parvis de la ziggurat pouvaient faire partie d’ un « transport sur roues » cérémoniel. On pourrait
supposer qu’en arrivant devant le temple « inférieur », réservé à l’ épiphanie du dieu Inšušinak,

12 La largeur entre les deux ornières était de , mètre et la largeur de chaque ornière variait entre , et ,

mètre (Ghirshman  : ).

 zsolt gábor lantos

ces chars étaient chargés de transporter l’ image du dieu vers le temple B d’ Inšušinak se trouvant
dans l’épaisseur du premier niveau de la ziggurat (Ghirshman  :  et  : ). Cette
proposition trouve un écho dans ces nombreux modèles de char en terre cuite trouvés en
Mésopotamie et datant de l’époque paléo-babylonienne (première moitié du IIe millénaire).
L’un d’eux porte une scène estampée à l’ intérieur du parapet qui représente un dieu coiffé de la
tiare à quatre rangs de cornes et vêtu de la robe à volants, assis sur un tabouret stylisé. Devant lui,
le disque du soleil semble indiquer qu’ il s’ agit de Šamaš, le dieu-Soleil et de la justice (AO ).
Sur le temenos, on peut compter au total six larges carrelages (Ghirshman , plan I ; Mofidi-
Nasrabadi , plan ) toujours implantés de façon désaxée. Une fois de plus, on observe
donc une volonté d’éviter systématiquement de placer deux portes sur le même alignement
et de tracer les allées de façon oblique, comme c’ est également le cas pour les temples. Ce
parti peut expliquer la disposition apparemment asymétrique des  portes des trois murs
d’enceinte, toutes décalées (Margueron  : ). Sept portes sur le temenos et dans la ville,
sept portes et deux fois sept tables de sacrifice sur le parvis Sud-Est de la ziggurat (Ghirshman
: , –): ce chiffre se révèle dans les installations religieuses et rituelles, et semble avoir
été révéré autant qu’en Mésopotamie où l’ un des exemples les plus significatifs est l’ oeuvre
de Lugalannemundu, roi d’Adab, qui a bâti pour la déesse Nintu un temple ayant sept portes
consacrées par des sacrifices de «sept fois sept » boeufs gras et brebis grasses (Kramer  : ).

.. Les limites (edges)

... Les trois murs d’enceinte concentrique
Sur les plans comme sur le terrain, la ziggurat de Tchoga-Zanbil apparaît indubitablement
comme le pôle central d’un vaste ensemble de constructions, dont les différents éléments se
répartissent à l’intérieur d’un système constitué par trois murs d’ enceinte approximativement
concentrique (Tourovets : ). Margueron ( : ) a noté que ces dernières n’ ont pas été
tracées suivant une configuration de type géométrique (Tourovets  : ) : « ces enceintes
ne paraissent pas avoir fait l’objet d’ un tracé théorique ou privilégié fondé sur un parti
géométrique». Cependant, Tourovets ( : ) suit difficilement l’ argument de Margueron,
quand celui-ci affirme ensuite que leur situation résulte d’ une adaptation à la topographie
des lieux («apparemment, elles – les enceintes – sont simplement adaptées aux conditions
topographiques»). Selon Tourovets ( : ), le terrain parfaitement plat sur lequel elles sont
construites ne présente aucune particularité ayant pu influencer de quelque manière que se soit
leur disposition; ainsi, trois murs concentriques s’ adaptent au monument le plus important et
le mieux conservé, qui est la ziggurat se trouvant au milieu de la ville (Margueron  : ).
Ces trois murs d’enceinte sont manifestement disposées d’ une manière concentrique autour
de cette énorme masse de , mètres de côté, même si la plus extérieure d’ entre-elles à
été légèrement prolongée vers l’est pour intégrer le secteur du Quartier Royal à l’ ensemble
(Tourovets : –). Dûr-Untaš possède une série de trois murs d’ enceinte concentriques
qui forment l’ossature principale de la ville. Selon Huot ( : ) « La ville … est enfermée
dans une triple enceinte». Margueron ( : ) précise que seule la ziggurat est dans cette
situation, puisque la ville proprement dite a seulement un mur et a été installée à l’ intérieur
de la plus grande enceinte. Ghirshman ( : ) fait la distinction entre le mur d’ enceinte du
temenos et celui de la ville. Bien qu’indirectement, le roi-bâtisseur Untaš-Napiriša a aussi son
propre mot à dire sur la question (selon le témoignage d’ une des briques inscrites en élamite en
provenance du site, cf. Steve : ) :
Moi Untash-GAL, fils de Humbannumena, roi d’Anzan et de Suse, l’emplacement de la ville (une
fois) enclos,
 essai d’application de la méthode de la psychologie environnementale 

j’ai construit là la ville d’Untash-GAL et le Lieu-saint; dans une enceinte extérieure et inférieure je
l’ai enfermée.
Les anciens rois n’avaient pas construit de kukunnum, moi je l’ai construit; aux dieux GAL et
Inshushinak (Seigneurs) du Lieu-saint
J’en ai fait don. Mes travaux et mes oeuvres qu’ils en acceptent l’offrande. L’exécution de l’ordre
des dieux GAL et Inshushinak
(Seigneurs) du Lieu-saint ici a été (réalisée). La ville que j’ai bâtie puissé-je … (Quant à) cette entrée
«la Grande Porte», (tel) est son nom.

... Tours, décorations et reconstitution de la hauteur des murs d’ enceintes

On ignore la hauteur du mur de la deuxième enceinte (celle du temenos). Cependant, on peut
tenter de l’évaluer. Derrière le temple des Napratep du Complexe Est qui s’ élevait près de l’ angle
Est du temenos, les eaux ont miné le mur, dont un pan s’ est effondré. Cette partie tombée était
constituée de  lits successifs de briques crues, ce qui correspondait à une hauteur d’ environ
, mètres. Ce pan de mur était tombé au pied de sa partie inférieure, encore attestée sur ,
mètres de hauteur. Ainsi, il serait possible d’ évaluer approximativement la hauteur primitive de
ce mur d’enceinte du temenos à au moins  mètres, soit le double de son épaisseur. L’ enceinte
extérieure, marquant la limite de la ville, est large de , mètres en moyenne (Ghirshman
: ), ce qui est assez surprenant vu l’ épaisseur du deuxième mur d’ enceinte qui atteint
, à  mètres (Ghirshman : ). Si l’ on suppose que la hauteur de celui-ci atteignait 
à  mètres, soit le double de son épaisseur (Ghirshman  : ), le mur extérieur de la ville
aurait été moins haut (Ghirshman  : ). Il se peut que les grands pilastres extérieurs du
troisième mur d’enceinte (le mur extérieur) formaient une suite de petites tours (Ghirshman
: ); il en va de même avec les saillants extérieurs qui dépassaient en hauteur le mur de la
deuxième enceinte et qui formaient peut-être de petites tours à intervalles réguliers (Ghirshman
: ). Sur toute la longueur du deuxième mur d’ enceinte, il n’ a été reconnu qu’ une seule
tour entièrement bâtie avec des briques cuites. Elle se situait, en saillie vers l’ extérieur, environ
au milieu du mur Sud-Est, à , mètres au Sud du « Passage Royal ». Large de , mètres et
formant une projection de , mètres, cette tour – qui semble avoir été postérieurement ajoutée
à la construction du mur d’enceinte – était pleine, sans trace d’ escalier, et conservait encore
près de  mètres de hauteur (apparemment on ne pouvait pas monter dessus) ! Trente-deux
briques inscrites alignées mentionnent un nur kiprat ou « lumière des régions (du monde) »,
une appellation qui s’applique peut-être à cette tour ou à toute l’ enceinte (de Mecquenem  :
). Cinquante-cinq autres briques d’un texte identique furent mises au jour près de l’ angle Est
de l’enceinte du temenos, derrière le Complexe Est où apparaît un décrochement dans le mur.
Il se peut qu’on ait projeté d’y élever une autre tour (Steve  : ) ! Celle qui flanquait le mur
Sud-Est, qui – vu ses dimensions – dépassait certainement le mur en hauteur, était richement
décorée sur sa face qui regardait la ziggurat. On a découvert à ses pieds les fragments de six
grandes plaques carrées émaillées, bleues, à pommeau, et au moins de douze clous dispersés
sur les restes du dallage de briques cassées qui reliait cette tour au « Passage Royal ». On ignore
si le mur du temenos était décoré de créneaux. On peut toutefois l’ admettre si l’ on se base sur
deux maquettes (ci-dessous), en terre cuite que les fouilleurs ont trouvées dans le Complexe Est
et dont le haut des murs est décoré d’une suite de « dents de scie » qui représentent peut-être
le dessin de merlons. Quant au décor de plaques à pommeaux, de clous émaillés, et de briques
émaillées, tous ces éléments d’embellissement architectural ne rehaussaient de leur éclat que
les façades des portes et n’ont jamais été identifiés sur la longueur du mur, sauf devant la tour
nur kiprat (Ghirshman : ).
 zsolt gábor lantos

... Des murs, comme fonction de défense ?

Selon Ghirshman (: ), aucun des trois murs d’ enceinte ne présente de particularité
propre à des fortifications élevées pour la défense : on n’y trouve ni portes destinées à une sortie
pour contre-attaquer des assaillants, ni poternes, ni contrescarpes, ni casemates, ni glacis, ni
douves, ni portes en chicane; les portes, qui n’ ont qu’ un seul escalier étroit pour accéder à la
crête du mur, le confirment largement. En ce qui concerne le deuxième mur d’ enceinte : un
escalier de , mètre de large et qui n’ a conservé que quatre marches, permettait de monter
sur les tours de la porte Nord-Est (Ghirshman  : ). Dans son mur Nord-Ouest, s’ ouvrait
un escalier de , mètre de large qui menait en haut des tours de la « Porte de Suse » et
qui conservait encore neuf marches, jadis couvertes de dalles de pierre aujourd’ hui disparues
(Ghirshman : ). Enfin, dans l’ angle Sud de la chambre-vestibule du « Passage Royal »,
face à la crapaudine, commençait un escalier qui menait en haut des tours et dont les trois
premières marches mesuraient , mètre et les trois suivantes un mètre de largeur. Ces six
marches formaient une volée qui, après un petit palier, tournait à ° pour continuer la montée.
Cette porte est la seule à faire connaître un escalier coudé, ce qui rendait la montée au sommet
des tours et du mur d’enceinte plus aisée. « L’ ascension n’ aurait-elle pas été prévue là pour un
visiteur princier» s’exclame Ghirshman ( : ). Chaque marche était recouverte de dalle de
pierre. Le nombre limité des portes (deux par rapport aux deux autres murs) de l’ enceinte de
la ville suggère néanmoins un certain souci défensif, un moyen physique de protection : c’ est
plutôt un obstacle pour pouvoir entrer dans une enceinte de  mètres de circonférence
(Ghirshman : ). Notons que selon l’ hypothèse de Ghirshman () l’ Entrée Royale
collée au mur du troisième enceinte était peut-être un casernement pour les soldats et les
chevaux. D’après Ghirshman ( : ), malgré ses murs, la ville de Dûr-Untaš devait rester
dans l’ esprit de ses créateurs, comme une ville ouverte conforme à la nature sacrée de ce grand
centre de pèlerinage. Selon Margueron ( : ), l’ existence des enceintes concentriques met
particulièrement en valeur la focalisation sur le phénomène religieux de la ville accentué par
la ziggurat se trouvant au centre du dispositif. Les nombreux temples existant à Dûr-Untaš,
dont chacun était dédié à une divinité particulière, firent aussi penser qu’ il s’ agissait d’ une
ville exclusivement religieuse. Mais aujourd’ hui cette hypothèse est remise en question car le
site, entouré par une puissante fortification qui délimite une zone d’ une centaine d’ hectares,
est loin d’avoir été entièrement fouillé. Il incluait très probablement des quartiers d’ habitation,
comme le laisse à penser le matériel ramassé en surface (Joannès  : ) et les résultats
géomagnétiques (Mofidi-Nasrabadi ). Untaš-Napiriša, lors de la réalisation de son vaste
plan de construction du centre impérial qui devait porter son nom, ne se doutait pas que
sept siècles plus tard, un de ses successeurs, aux destinées moins heureuses que les siennes,
en quittant «Madaktu, ville de sa résidence » (de Miroschedji ), viendrait se réfugier à
Dûr-Untaš: selon les anciennes annales assyriennes, Humban-Haltaš III, le dernier roi élamite
(–?), fuyant devant l’armée d’ Assurbanipal, arriva à Dûr-Untaš, mais les défenses de
cette cité lui parurent beaucoup moins sûres que le fleuve qui coulait à ses pieds, et, abandonnant
la ville d’Untaš-Napiriša, il traversa l’ Ididé (l’ Âb-è Diz) « et fit de ce fleuve sa ligne de défense » –
combien éphémère (Streck , vol. II : ). Il est possible que la ville ait déjà été abandonnée
depuis un certain temps13 et que les murs en ruines n’ étaient plus en état de la défendre, faute

13 Dûr-Untaš ait été déjà plus ou moins abandonnée un peu plus d’un siècle après la disparition de son fondateur
(Pons  : ). Il est possible que la ville ait été définitivement abandonnée au alentour de  à la suite des
campagnes militaires entrepris par le roi Nabuchodonosor Ier (–) contre l’Élam (Steve/Gasche/De Meyer
 : , Pons  : ) : le pillage de Dûr-Untaš en  ou  ne pouvait être l’oeuvre des troupes d’Assurbanipal,
mais le fait plus ancien du roi babylonien, qui se glorifie d’avoir «conquis l’Élam et pillé ses richesses». Quel roi de la
 essai d’application de la méthode de la psychologie environnementale 

de réparation et de restauration (Steve/Gasche/De Meyer  : , Tourovets  : ). Quoi
qu’il en soit, il semble que ni les murs, ni la barrière de collines se trouvant sur le plateau, ni le
fleuve ne constituait un véritable moyen de défense.

.. Les secteurs (districts)

... Les trois zones concentriques
Le modèle urbain de la zone concentrique défini par Burgess (/) est bien applicable
à la ville antique de Dûr-Untaš: selon lui, le centre d’ une ville est le noyau entouré de zones
de transition. L’unité de la structure urbaine de Dûr-Untaš justifie donc la théorie de Burgess
(/): trois zones concentrique encadrées par trois murs d’ enceinte de diamètre de
plus en plus petit entourent le noyau urbain de la ville. On reconnaît donc une organisation
concentrique avec la ziggurat dominant le centre de la première zone, avec les temples se situant
dans la seconde, ainsi qu’avec le quartier palatial se trouvant dans la troisième (Margueron
: ): ces différents secteurs paraissent donc remplir chacun un rôle bien précis ; Kohl
(Csanádi/Ladányi ) avance qu’en se dirigeant du centre vers l’ extérieur, « l’ importance »
de la population diminue (Lippai/Dúll  : ), mais dans notre cas, c’ est la « sacralité » qui
diminue. On peut constater une réflexion urbanistique, concernant le choix des emplacements
destinés à recevoir les différentes constructions (Tourovets  : ) : l’ observation quelque
peu attentive du plan topographique montre que l’ ensemble relève d’ un système planifié
obéissant à des règles prédéterminées (Tourovets  : ).

... La difficulté de définir la vraie nature de la ville

À Dûr-Untaš, on constate la présence d’ un foyer religieux dominant, ainsi que d’ un quartier
royal et de quelques bâtiments, bien qu’à fonction indéterminée : à première vue, on pourrait
voir dans cet assemblage une ville sacrée. Mais une étude plus profonde démontre – comme on
va le voir ci-dessous – que cette question n’ est pas aussi simple qu’ elle le paraît. Le caractère
sacré de la ville est par exemple confirmé par les briques inscrites (ou autres objets) d’ Untaš-
Napiriša, trouvées soit dans la ville, soit à Suse en provenance de Tchoga-Zanbil, qui définissent
cette agglomération comme Lieu-saint (Steve  :  ou ) ou siyan-kuk : ce terme, désignant
spécifiquement le temenos de Tchoga-Zanbil, souligne la fonction sacrée accentuée de la ville
(de Miroschedji : ). Dans le secteur entouré par le premier mur d’ enceinte où se trouve
les parvis, on reconnaît l’existence de plusieurs constructions cultuelles, parfois disposées l’ une
à côté de l’autre, et identifiées à des temples, chacun d’ entre-eux étant voué au culte de l’ une
ou l’autre divinité de la religion élamite : les raisons de tels regroupements restent cependant
inconnues. La surface de la fouille, depuis la face Sud-Ouest jusqu’à la face Nord-Est du mur
d’enceinte et depuis la façade des temples du Nord-Ouest jusqu’ au mur d’ enceinte du Sud-Est,
couvrait une superficie de . mètres carrés. Avec les temples, elle atteignait environ .
mètres carrés (Ghirshman : ). D’ après le témoignage des fouilles, Ghirshman a désigné
le secteur se trouvant entre le premier et le deuxième mur d’ enceinte, comme étant le temenos,
c’est-à-dire le sanctuaire par excellence (Tourovets  : ). Ce secteur intermédiaire est donc
apparu au fouilleur, comme un véritable temenos, puisqu’ en dehors de quelques bâtiments à
fonction difficilement identifiable (habitations provisoires pour les ouvriers selon Ghirshman

Mésopotamie ne se présente pas, dans ses propres écrits, comme le plus grand conquérant du monde? Les quelques
armes élamites et assyriens (pointes de flêche, poignards et haches) trouvés essentiellement du côté du parvis Nord-
Ouest de la ziggurat dateraient donc de cette époque (Ghirshman ).
 zsolt gábor lantos

: ) et sans grande importance (Margueron  : ), il renfermait principalement des
temples, voués aux divinités élamites : à côté du Complexe Est qui comprenait les temples de
Pinikir, de IM et Šala, de Šimut et NIN-ali et des Napratep, on doit mentionner le temple
de Hišmitik et Ruhuratir. Cet espace comprend une ziggurat de plus de , mètres de côté
et une douzaine de temples de  à  sur  mètres en moyenne (Margueron  : ) : les
temples A et B d’Inšušinak sont placés dans le premier niveau de la ziggurat, face Sud-Est ;
les temples de Kiririša Ouest et Est; les temples d’ Išnikarab, de GAL ; les deux temples carrés ;
plusieurs briques trouvées sur place témoignaient d’ autres noms de divinités, dont les temples
n’ont finalement jamais été réalisés: Adad, Nabu ou Nahhunte (Ghirshman  : ). Selon
Margueron (: ), il apparaît clairement que l’ organisation générale du site et la répartition
des différentes constructions ont été conçues en fonction de la ziggurat et de son emplacement
au centre du dispositif. L’importance de ces deux secteurs religieux (le troisième étant le secteur
royal) se situant à l’intérieur de la deuxième enceinte, apparaît curieusement surdimensionné
par rapport à l’ensemble de la ville. Le temenos à lui seul devait dépasser de près de trois fois
la superficie du quartier religieux de la grande capitale qu’ était Suse (Ghirshman  : ). En
dehors des édifices sacrés, quelques bâtiments non cultuels en bordure des murs Nord-Ouest
et Sud-Ouest du mur du temenos ont été également mis au jour par Ghirshman ( :  et
). En décidant d’élever les trois murs d’ enceinte, les « urbanistes » élamites prévoyaient de
larges espaces libres entre le mur du temenos et le mur de la ville (Ghirshman  : ), afin
d’accueillir les futurs bâtiments destinés à être construits. Par rapport à la superficie du site
atteignant une centaine d’hectares, le petit nombre des vestiges de bâtiments sacrés et profanes,
ainsi que d’habitations provisoires pourrait nous faire penser que la ville n’ a pas été achevée
à cause de la mort de Untaš-Napiriša, mais aussi parce que ses successeurs n’ ont pas souhaité
continuer l’oeuvre du roi: suivant les traditions fréquentes en Orient, un prince ne poursuivait
pas l’oeuvre de son prédécesseur, surtout quand il s’ agissait d’ un prince de dynastie différente
(Ghirshman ). En tout cas, c’est ce que les fouilles effectuées entre  et  par l’ équipe
dirigée par Ghirshman ( et ) ont démontré. C’ est en se basant sur ces résultats, que
Margueron (: ) a rappelé qu’ aucune activité commerciale ou artisanale n’ a pu être
décelée et que la ville elle-même, dans le troisième secteur compris entre le le deuxième et
le troisième mur d’enceinte, semble plus occupée par des bâtiments officiels (Quartier Royal),
à l’exception du sanctuaire de Nusku, que par des habitation populaires. A l’ époque, on ne
connaissait pas les résultats des prospections géomagnétiques.

... Activité de l’artisanat local sacré de Tchoga-Zanbil au temps d’ Untaš-Napiriša

D’après ce qui a été mis au jour à Tchoga-Zanbil, il n’y a nul doute sur l’ existence d’ une industrie
manufacturière sacrée: plusieurs centaines de plaques à pommeaux14 et des briques émaillées de
couleur bleue et verte – dont le Temple supérieur de la ziggurat de Tchoga-Zanbil étaient jadis
composé, donnant certainement au bâtiment exposé au soleil un éclat vif aux reflets argentés et
dorés15 – étaient très bien représentées au temps d’ Untaš-Napiriša. Les briques émaillées étaient
fabriquées en Élam à une très large échelle depuis au moins le XIVe siècle. L’ industrie du verre
qui se développe à cette époque, joue ici un rôle très important : on en ignorait tout avant la
découverte des monuments de Dûr-Untaš (Ghirshman  : ). En Élam, les premières

14 Je dois attirer l’ attention sur la communication de Francelin Tourtet présentée au même congrès, qui est

intitulée « Distribution, Material and Functions of the Wall Knobs in the nd Millennium bc : from Southwest Iran
to the Middle Euphrates » ; sur ce sujet voir également les travaux antérieurs de Annie Caubet.
15 Ghirshman  : .
 essai d’application de la méthode de la psychologie environnementale 

briques ou carreaux émaillés inscrits datent donc d’ Untaš-Napiriša16 (Sauvage b : ). À
cette époque on utilisait le mot upkumi pour qualifier les matières émaillées (Vallat  : ).
On peut également mentionner les tiges de verre noir et blanc, qui servaient aussi comme
éléments de décoration architecturaux sacrés sur les battants des portes en bois (Ghirshman
: ). Il est légitime de supposer que les briques émaillées sont une invention élamite et
que leur diffusion partait de l’Élam. Parmi les différents objets en métal retrouvés sur le site et
à Suse, on peut mentionner une remarquable hache en argent, dont la lame inscrite au nom du
roi est «crachée» par une tête de lion et dont le talon est orné d’ un sanglier couché en électrum
(alliage or et argent). Elle a été trouvée dans la cella  de la partie Est du temple Kiririša
(Ghirshman : , Amiet : fig. ). Mais c’ est surtout la statue en bronze massif de la
reine Napirašu (sans doute rapportée à Suse depuis Dûr-Untaš peut-être par Šutruk-Nahhunte
Ier), l’épouse du roi Untaš-Napiriša, qui permet de mesurer le haut degré de technologie que
les métallurgistes élamites ont alors atteint à cette époque (Amiet  : fig. ). Ils ont dû
pratiquer en effet deux coulées successives : une coulée pour la coque faite de cuivre et d’ étain
selon la technique de la cire perdue et une coulée en fonte pleine dans un alliage de bronze
et d’étain pour le noyau, qui est habituellement en terre réfractaire. Les deux parties ont été
maintenues par des broches et des barres. Un revêtement en or ou en argent devait être fixé sur
les côtés. Bien qu’acéphale, ce monument d’ une hauteur de , mètre, grandeur nature, pèse
encore  kilogrammes, soit presque deux tonnes ! (Steve/Vallat/Gasche / : ).
Les chambres  à  des annexes de la partie Ouest du temple de Kiririša étaient occupées par
les ateliers et les magasins qui dépendaient du sanctuaire et où étaient probablement fabriqués
des objets en fritte, peut-être aussi en argile : figurines de personnages, hommes et femmes,
animaux, copies d’objets en métal reproduits en fritte, des cylindres qui ont été mis au jour
tant dans les temples que dans les tombeaux de Dûr-Untaš. Le clergé les produisait sur place
où les obtenait des pèlerins qui déposaient ces objets par dévotion sur les tables d’ offrandes
ou les autels (Ghirshman : ). On peut compléter cette liste artisanale avec plus d’ une
centaine de sceaux-cylindres et cachets de type pseudo-cassite et élamite élaboré (classé par
Porada) trouvés essentiellement dans les chapelles II, III, IV de Tchoga-Zanbil, qui sont des
produits également locaux (Ghirshman  : –) : il va sans dire que Tchoga-Zanbil est un
site provincial comparé à Suse et que les cylindres sont pour la plupart de très médiocre qualité ;
cependant on voit se dessiner les grandes lignes d’ une évolution (Porada  : ).

... Dûr-Untaš est-elle une ville ordinaire, une cité-temple ou un cas spécial ?
Jusqu’à une époque récente on pensait, que les quelques bâtiments fragmentaires d’ origine
non cultuelle n’étaient pas regroupés en un quartier urbain, mais isolés. À présent, on sait
que des constructions se serraient les unes contre les autres. Autrefois, il semblait difficile de
parler de ville au sens strict du terme, à en juger seulement par les résultats des fouilles et le
témoignage des briques inscrites, d’autant que l’ expression siyan-kuk ne se référait qu’ au seul
temenos. C’est la raison pour laquelle le site a été défini comme une cité-religieuse (Ghirshman
: ), une cité-sacrée (Ghirshman  : ), une ville-sacrée (Ghirshman  : ), une ville-
sanctuaire (Steve/Gasche/De Meyer  : ), une cité-temple (Huot/Thalmann  :  et
Garcin : ) ou un complexe politico-religieux (Steve/Vallat/Gasche / : ).
Margueron (: ) propose de ne pas considérer Dûr-Untaš comme une ville – malgré
ses dimensions et ses caractères urbains – mais comme un cas particulier. Il y voit plutôt une

16 On a trouvé depuis des carreaux émaillés plus anciens à Haft-Tépé qui sont datés du XVe siècle (Harper et al.,

 : , Sauvage b : ).

 zsolt gábor lantos

fondation religieuse, un centre de pèlerinage qui devait pouvoir recevoir le roi et les fidèles lors
d’une manifestation religieuse. Dans ces conditions, toujours selon le chercheur, il paraissait
difficile de penser que le roi, à commencer par Untaš-Napiriša, ait pu envisager d’y résider en
permanence: pouvait-on alors dire que ses successeurs étaient retournés à Suse (Ghirshman
 et , Huot : )? Grâce aux résultats des récentes recherches géomagnétiques
effectuées par l’équipe archéologique dirigée par Mofidi-Nasrabadi (), on peut cependant
constater une occupation nettement plus dense de constructions certainement domestiques,
que l’on ignorait antérieurement. D’ après ces résultats, les habitations domestiques paraissent
regroupées. Les images géomagnétiques montrent des structures architecturales dans certaines
zones là où il n’y avait rien avant. Dans les zones où on ne voit rien, on ne peut pas dire qu’ il
n’y a absence de construction, car les briques crues apparaissent beaucoup moins nettement
sur ces images. Je pense que la présence de quartiers d’ habitations ne doit pas exclure la
réalité de l’aspect religieux accentué de Dûr-Untaš, que l’ on doit considérer comme acquis. De
plus, la présence de ces habitations dans la ville-même pourrait même renforcer le fait, qu’ il
s’agit de bâtiments prêts à accueillir justement les fidèles arrivant de différentes régions du
pays. C’est ainsi que les conditions de logement étant résolues, la ville était prête à accueillir
les pèlerins venant participer aux fêtes, qui se déroulaient aussi en dehors de la ville dans le
cadre de processions (Ghirshman  : –) : la porte du côté du fleuve devait être un
lieu de passage et de circulation pour la masse de pèlerins, car les grandes cérémonies, selon
Ghirshman (: ), incluaient aussi le fleuve, qui avait un rôle clé (Ghirshman  : ,
Tourovets ). C’est ici qu’on doit noter qu’ il est tout de même surprenant de constater que
l’enceinte la plus petite était paradoxalement la plus riche en portes (Margueron  : ) : en
effet, la largeur relativement limitée des entrées, ainsi que le peu de place disponible à l’ intérieur
du mur excluent l’accès d’un grand nombre de fidèles sur les parvis. Un nombre aussi exagéré
de portes devait être inutile, si l’on ne prenait pas en compte le facteur certainement complexe
du rite élamite, en grande partie inconnu. D’ après le contexte architectural, seul l’ élite – à savoir
sans doute le roi, la reine, la cour et le clergé, ce dernier étant responsable du bon déroulement
de la liturgie – aurait eu le privilège de passer les portes de l’ enceinte de la ziggurat lors de la
fête printanière de l’ akitu. Si l’on a parlé des portes, il est certainement utile de compléter cette
étude par une intéressante suggestion faite par Ghirshman : une des deux maquettes mises au
jour derrière le mur Sud-Est du temple de Pinikir a un plan ovale inconnu dans l’ architecture
élamite selon Ghirshman (: ) : le fouilleur songe plutôt à y reconnaître la maquette de
la ville de Dûr-Untaš (?) dont l’enceinte extérieure dessine un ovale irrégulier et était percée
de deux portes. Quant à cette maquette et une autre (Ghirshman  : ), je pense qu’ elles
étaient en relation avec le caractère religieux de la ville, tout comme les autres pièces votives
et objets d’offrandes confirment cet aspect (Ghirshman  : – et , Porada ). Or
la ville semble avoir compté elle aussi deux entrées, dont celle du côté du fleuve recevait les
pèlerins (Tourovets ): toutes deux jouaient sans doute un rôle important lors des grandes
cérémonies (Ghirshman : ). Pour un nombre élevé de fidèles arrivant à Dûr-Untaš (dans
l’hypothèse de Ghirshman), ou même pour la population locale, ces deux portes semblent bien
insuffisantes, et je suis convaincu que des portes supplémentaires nous échappent encore. Les
deux seules grandes portes connues, percées dans le mur d’ enceinte de la cité, l’ une du côté
Sud-Est et l’autre au Nord-Est, face au fleuve (Ghirshman  : ) ne me paraissent pas
représentatives, même si Lenzen (, pl. ) n’ attribue lui aussi que deux portes au mur
extérieur de la ville d’Uruk.
 essai d’application de la méthode de la psychologie environnementale 

.. Les points stratégiques (nodes)

... Des carrefours de circulation
En étudiant attentivement le plan du temenos (Ghirshman , plan I, Mofidi-Nasrabadi ,
plan ), on remarque que les grandes voies dallées se rencontrent en plusieurs endroits. Le plus
grand dallage, qui relie la porte Nord-Est et la Porte Royale, devait certainement avoir un rôle
privilégié (comme le souligne Tourovets à propos de la cérémonie qui se dirigeait peut-être vers
le Passage Nord-Est menant au fleuve), puisque c’ est là qu’ aboutissentent quatre voies sur cinq,
et parce que trois portes – dont deux monumentales – lui sont liées. Deux voies se dirigent vers
les portes Nord et Nord-Est de l’enceinte du parvis, les deux autres vers le Passage Royal où
finalement elles se rejoignent. Seule la voie isolée qui relie la Porte de Suse et la porte Ouest
est dépourvue de carrefour. Cette constellation de voies se croisant les unes avec les autres ne
devait pas être accidentelle: rien dans les approches de la Tour sacrée ne se réalisait sans un
sens voulu, une valeur religieuse, voire liturgique. Ainsi, ce réseau de circulation ne pouvait
avoir une destination «laïque». Ces dallages qui furent tracés en plusieurs directions sur le vaste
péribole de la «ville sainte» de Dûr-Untaš, apparaissent donc comme des voies de processions
religieuses. Toutes les six grandes avenues du temenos, reliant les temples et les portes, se dirigent
essentiellement vers le monument le plus important et le mieux conservé, la ziggurat qui se
trouve au centre de la ville.

.. Les points de repère (landmarks)

... Le reflet du pouvoir central sur les constructions à la verticale
Moi Untash-GAL, fils de Humbannumena, roi d’Anzan et de Suse,
j’ai obtenu une longue vie, ma santé est prospère, de terme … à ma lignée je n’aurai pas.
C’est pourquoi j’ai construit un temple aux voûtes de briques, sanctuaire sublime;
au dieu Inshushinak, (Seigneur) du Lieu-saint j’en ai fait don.
Moi, j’ai élevé une ziggurat vers le ciel.
Mes travaux et mes oeuvres que le dieu Inshushinak en accepte l’offrande! (Steve : )
Au point de vue architectural, une des principales caractéristiques de la physionomie urbaine
est la construction à la verticale. En raison de ses dimensions et de sa masse importantes, les plus
remarquables bâtiments des villes de la sphère mésopotamienne étaient les ziggurats. Derrière
le symbole politico-religieux se cachait le véritable rôle des ziggurats : une économie basée et
contrôlé sur la redistribution des biens produits et stockés aux noms des dieux. Le livre de la
Genèse : – nous décrit à travers l’ histoire de Joseph un bel exemple de l’ organisation du
pouvoir central au IIe millénaire: «Qu’ils rassemblent tout les produits de ces bonnes années à
venir, qu’ils amassent, sous l’autorité du pharaon, du blé et des vivres dans les villes et qu’ ils en
aient la garde. Ces provisions formeront une réserve pour le pays, pour les sept années de famine
qui frapperont l’Égypte, afin que le pays ne soit pas détruit par la famine ». Les tours à étages
exprimaient le pouvoir central, la puissance et la splendeur. Andrae ( : ) résume bien
l’essentiel: «Il existe une possibilité de rendre la puissance divine encore plus grande, c’ est de
monter plus haut et d’élever au dieu une voie ascendante par laquelle celui-ci pourrait descendre
sur terre. C’est le temple suprême, semblable à une montagne qui se dresse de la terre vers le
ciel, toute en marches destinées aussi bien aux dieux qu’ aux hommes. En Babylonie, ces temples
s’appelaient zikkurat …» (Ghirshman ). De même que Thureau-Dangin () : « (ce)
temple, comme une montagne dans le ciel et sur la terre, il s’ éleva vers le ciel » (Cyl. A, col. IX :
); «le temple est établi vers le ciel comme un grand mont » (Cyl. B. col. I :  et col. XXIV :  –
Steve , p. ). Cette montagne artificielle marquait l’ emplacement de l’ axis mundi, c’ est-à-
dire l’axe du monde (Vicari : ). On en a l’ écho biblique à travers l’ histoire de la Tour de
 zsolt gábor lantos

Babel: «Ils se dirent l’un à l’autre: Allons ! Faisons des briques et cuisons-les au feu ! La brique
leur servit de pierre, et le bitume de ciment. Il dirent encore : Allons ! Construisons-nous une
ville et une tour, dont le sommet touche le ciel et faisons-nous un nom, afin de ne pas être
dispersés sur toute la surface de la terre. L’ Éternel17 descendit pour voir la ville et la tour, que
construisaient les hommes» (Genèse  : –).

... Des «gratte-ciel» antiques: les ziggurats

Avec une hauteur totale18 variant de  à  mètres à l’ origine19, ces tours à étages étaient
incontestablement les premiers «gratte-ciel » du monde mésopotamien (Dethier  : ), les
premiers grands monuments de l’humanité entièrement construits par l’ homme dont l’ accès
au sommet était assuré par des escaliers : parmi les plus grandes, on peut citer l’ exemple d’ Aqar-
Quf (l’ancienne Dûr-Kurigalzu) construite vers  sous le règne du roi cassite Kurigalzu Ier
qui, encore conservée sur  mètres de haut, devait s’ élever à l’ origine à près de  mètres
(Gullini : , Joannès : , Montero-Fenollós  : , fig. ) ; la ziggurat de
Babylone devait à l’origine avoir une hauteur similaire : d’ après les récents calculs de l’ équipe
dirigé par Montero-Fenollós ( : –), la Tour de Babel était un monument plus
modeste avec ses  mètres de hauteur par rapport aux  mètres traditionnellement acceptés
(André-Salvini : –). Ce que l’ on trouve sur place – quelques bribes de briques
cuites dispersées sur le site – ne peut qu’ évoquer ce que fut autrefois la tour de Babylone,
puisque le monument est aujourd’hui en ruine : le fossé en forme de poêle carrée est l’ empreinte,
autrement dit le négatif imprimé dans le sol, du périmètre des fondations de la tour et de
l’escalier frontal. C’est ce dernier qui dessine le manche de la poêle. Le petit monticule fait de
briques crues, occupant le centre de cette dernière, s’ est érodé au fur et à mesure qu’ il perdait
son corset de briques cuites: en effet, les millions de briques cuites qui servaient de couche
protection (coffrage) à la Tour ont été démantelées au cours des siècles par les habitants du
lieu pour construire des villes et des villages aux alentours. Ces maigres vestiges sont tout ce
qui reste de ce qui fut la tour de Babel (Vicari  : –). D’ après les calculs effectués par
Ghirshman (: –) et Montero-Fenollós ( : , fig. ), la ziggurat de Tchoga-Zanbil
devait atteindre avec le temple supérieur entre  et  mètres à l’ origine, c’ est-à-dire à peu près
le double de la hauteur des vestiges actuels. Une telle hauteur correspondrait à la moitié de la
longueur de son côté (Ghirshman  : ). Avec ses dimensions imposantes, la ziggurat de
Tchoga-Zanbil était certainement un des exemples les plus grand de ce type.

... La «brique» de Dûr-Untaš

Contrairement à celle de Babylone, la ziggurat élamite de Tchoga-Zanbil est heureusement
restée une des tours les mieux conservées du Proche-Orient, atteignant quelque  mètres de
hauteur au socle du troisième étage (Vicari  : ) : aussi présente-t-elle encore un beau relief
aux yeux des voyageurs, fascinés par les ruines archéologiques, quelque trente-quatre siècles
après son érection sur un plateau, sans doute soigneusement choisi par le roi Untaš-Napiriša sur
des critères topographiques et religieux, peut-être conseillé par ses concepteurs et ses prêtres
(Ghirshman : , Tourovets ). Selon Perret « la bonne architecture est ce qui fait de

17 L’ hébreu utilise trois noms principaux pour désigner Dieu. Les traductions suivantes ont été retenues dans la
Bible () : Yhvh (généralement prononcé Yahvé) a été traduit l’Éternel; Adonaï a été traduit Seigneur; El, Elohim,
Eloha, ont été traduits Dieu.
18 Y compris le temple supérieur (supposé) sur chacune des ziggurats présentées prochainement.
19 Pour comparaison, la limite inférieue d’un gratte-ciel est aujourd’hui de  mètres.
 essai d’application de la méthode de la psychologie environnementale 

bonnes ruines»: avec cette affirmation, les bâtisseurs ont dû penser à l’ emploi de matériaux de
construction soigneusement choisis – en fonction de leur disponibilité, de leur résistance et de
leurs caractères esthétiques – pour la réalisation d’ un édifice destiné à immortaliser le souvenir
de l’activité du souverain. Car construire et maintenir les édifices de ses prédécesseurs était
une prérogative, une obligation envers les dieux, un devoir d’ État et de piété dynastique. Cela
répondait sans doute au désir universel des puissants, c’ est-à-dire de laisser une preuve de leur
existence et de leurs actions (Lackenbacher  : –). Il s’ agit là d’ un point fondamental,
puisqu’on a effectivement la possibilité de constater dans le matériau même de construction,
la marque concrète du pouvoir politique (Sauvage b, ) : tel est le cas de la ziggurat
de Tchoga-Zanbil. Ainsi, le pouvoir central marquait incontestablement son empreinte, sa
signature officielle en quelque sorte, dans le matériau même de construction. Au Proche-Orient,
c’est la brique qui a rempli ce rôle. Les « gratte-ciel » antique étaient principalement bâtis de
briques crues (c’est-à-dire séchées au soleil) et de briques cuites au four (Sauvage a : ).
Grâce à sa résistance à la pression et à sa remarquable durée de vie par rapport à la brique crue,
la brique cuite est devenue dans l’antiquité le symbole de l’ éternité. Comme elle représentait le
triomphe de l’homme sur le temps, ce n’ était pas un matériau banal aux yeux de nos ancêtres.
À vrai dire, c’était tellement un produit de luxe, que l’ on s’ en servait surtout pour édifier des
temples et des palais, c’est-à-dire les demeures des dieux et des rois (Cambell/Pryce  :
). Son importance dans l’architecture mésopotamienne était telle, que l’ on se servait du mot
«brique» pour désigner soit le temple ou soit l’ ensemble d’ une ville (Sauvage b : ). Le fait
que le même terme peut désigner à la fois le temple et la ville, amène à évoquer deux maquettes
en terre cuite provenant du quartier du Complexe Est. L’ une a été mise au jour derrière le mur
Sud-Est du temple de Pinikir. Une autre derrière le mur Nord-Est du temple de Šimut et NIN-
ali. Il s’agirait selon Ghirshman de pièces votives déposées dans ces temples par un urbaniste,
un architecte ou un maçon. On connaît bien ce genre de maquettes architecturales en terre
cuite (Parrot ; Margueron  et , Müller ). Le fait que les maquettes de Dûr-
Untaš ont leur toit bordé de merlons triangulaires invite à y reconnaître autre chose que de
simples maisons d’habitation. Du reste, dans une ville fondée pour devenir peut-être un grand
centre de pèlerinage, avec une ziggurat et de multiples temples, on hésiterait à penser qu’ une
maquette d’une simple demeure de laïc ait pu être déposée dans un sanctuaire. Ghirshman
(: ) proposait donc de voir dans l’ un de ces deux objets la maquette simplifiée d’ un
temple à une seule porte. Quant à l’autre, qui a un plan ovale inconnu dans l’ architecture élamite
selon Ghirshman (: ), il ne doit pas non plus correspondre à une habitation puisqu’ il a,
lui aussi, le toit décoré de merlons. Le fouilleur songeait à y voir plutôt la maquette de la ville de
Dûr-Untaš dont l’enceinte extérieure dessine un ovale irrégulier et était percée de deux portes.
Force est de conclure que Dûr-Untaš ne livre pas facilement ses secrets : en effet, on a
du mal à comprendre par exemple la fonction des murs. Le choix de voies désaxées est-
il une caractéristique de l’architecture urbaine élamite ? Il est certain, que rien ne devait
se réaliser sans un sens voulu, vu l’immense effort investi dans la construction de cette
ville. La fonction originelle de la ville-même n’ est pas claire, malgré plus d’ un demi-siècle
de recherches approfondies: est-ce une ville ordinaire comme on tendrait actuellement à le
suggérer ou bien une cité-sacrée, comme Ghirshman le pensait à la suite de ses recherches ?
On ne devrait pas s’étonner que ces soient principalement les bâtiments religieux ou royaux,
les plus soigneusement construits, qui aient attiré l’ attention du fouilleur. Notre connaissance
de la ville s’améliore cependant de façon significative, grâce aux techniques modernes pouvant
contribuer à la recherche: des quartiers d’ habitations ont été mis au jour dans des secteurs
que l’on croyait vides. La ville a effectivement été bâtie, même si tout le terrain n’ a pas été
occupé. La méthode d’analyse utilisée permet de mieux comprendre la nature urbaine du
site. Dûr-Untaš est le meilleur exemple d’ une ville élamite et sa ziggurat est restée une des
 zsolt gábor lantos

tours les mieux conservées du Proche-Orient. De ce fait, le statut du site, inscrit au patrimoine
mondial de l’Unesco justifie la base de données qui se développe et s’ enrichit d’ année en année.
Cette documentation est capitale pour la recherche, la valorisation touristique, la protection
des monuments et la conservation du site (Guillaud/Okada/Vatandoust ). L’ homme qui
détruit parfois plus que le temps les vestiges du passé, a une part de responsabilité dans le
cas de Tchoga-Zanbil: l’habitant des environs du site se contentait de sa hutte en roseaux ou
de sa tente en poil de chèvre; le matériau durable que représentaient les belles briques cuites,
étonnamment résistantes malgré leur vénérable âge de près de trois millénaires et demi et, qui
atteignaient un poids de  kilogrammes, ne l’ intéressait pas. Les temps changent pourtant et
la menace de destruction de ce qui fut mis au jour se précise. Aux responsables d’y veiller, nous
avertis Ghirshman.

Bibliographie

Amiet, P. (): Elam, Auvers-Sur-Oise: Archée Éditeur.

André-Salvini, B. (): Catalogue de l’exposition Babylone, Paris: Musée du Louvre.
Andrae, W. (): Alte Feststrassen im nahen Osten, Leipzig: Hinrichs Verlag.
Ankerl, G. (): Építészet és kommunikáció, Budapest: Műszaki Könyvkiadó.
Baqir, T. (): Aqar Quf, Bagdad.
Bible () Segond , Société Biblique de Genève.
Bonnes, M. et Secciaroli, G. () : Environmental psychology. A psycho-social introduction, London:
Sage.
Burgess, E.W. (): «A városfejlődés: hipotézisek egy kutatási javaslathoz.» In Szelényi I. (szerk.,
): Városszociológia, Budapest: Közgazdasági és Jogi Könyvkiadó: –.
Campbell, J. et Pryce, W. (): A tégla világtörténete, Budapest: Kossuth Kiadó.
Carter, E. et Stolper, M.W. (): «Middle Elamite Malyan,» Expedition /: –.
Csanádi, G. et Ladányi, J. (): Budapest térbeni-társadalmi szerkezetének változásai, Budapest: Aka-
démiai Kiadó.
de Mecquenem, R. et Michalon, J. () Recherches à Tchoga Zembil (= MDP ), Paris: Geuthner.
de Miroschedji, P. () «Le dieu élamite au serpent et aux eaux jaillissantes,» IrAnt : –.
———. () «La localisation de Madaktu et l’organisation politique de l’Élam à l’époque néo-élamite.»
In De Meyer, L., gasche, H. et Vallat, F. (Eds.), Fragmenta Historiae Elamicae, Paris: –.
Dethier, J. (): Des architectures de terre ou l’avenir d’une tradition millénaire, Paris: Centre Georges
Pompidou, Centre de Création Industrielle.
Dúll, A. (): «A helyidentitásról,» Magyar Pszichológiai Szemle –: –.
———. (): Hétköznapi otthoni rutinok környezetpszichológiai vizsgálata (Ph.D. értekezés, ELTE, Buda-
pest).
———. (): «A környezetpszichológia története,» Magyar Pszichológiai Szemle : –.
———. (): «Ember és környezet affektív kapcsolata: a helykötődés,» Alkalmazott Pszichológia IV/:
–.
Dúll, A. et Urbán, R. (): «Az épített környezet konnotatív jelentésének vizsgálata: módszertani
megfontolások,» Pszichológia /: –.
Garcin, J.-C. () «La ville pré-moderne,» JESHO :–.
Geddes, P. (): «Teremtő városfejlesztés.» In Vidor F. (szerk., ), Urbanisztika, Budapest: Gon-
dolat Kiadó: –.
Ghirshman, R. () Tchoga Zanbil (Dur-Untash), vol. I., La Ziggurat (= MDP ), Paris: Geuthner.
———. () Tchoga Zanbil (Dur-Untash), vol. II., Temenos, temples, palais, tombes (= MDP ), Paris:
Geuthner.
Grillot, F., et Vallat, F. (): «Le verbe élamite «pi(š)ši»,» Cah DAFI : –.
Guardata, F.B. et Dolce, R. (): Archeologia della Mesopotamia, l’età cassita e medio-assira, Roma:
Giorgio Bretschneider.
Guillaud, H., Okada, Y. et Vatandoust, A. (): Catalogue de Chogha Zanbil (Projet de conservation et
gestion), UNESCO.
Gullini, G. () : Catalogue de l’exposition La terra i due fiumi, Venti anni di archeologia italiana in
Medio Oriente, La Mesopotamia dei tesori, Torino: Museo di Antichità.
 essai d’application de la méthode de la psychologie environnementale 

Hansman, J. (): «Elamites, Achaemenians and Anshan,» Iran : –.

Harmatta, J. (szerk., /): Ókori Keleti Történeti Chrestomathia, Budapest: Nemzeti Tankönyv-
kiadó.
Harper, P.O. et al. (): The royal City of Susa. Ancient Near Eastern Treasures in the Louvre (Catalogue
de l’exposition du Metropolitan Museum of Art, New-York,  novembre – mars ),
New-York: Metropolitan Museum of Art.
Holahan, C.J. (a): «A környezet megismerése: Környezeti kogníció.» In Dúll A. et Kovács Z. (szerk.,
), Környezetpszichológiai Szöveggyűjtemény, Debrecen: Kossuth Egyetemi Kiadó: –.
———. (b): «A környezetpszichológia természete és története.» In Dúll A. et Kovács Z. (szerk., )
Környezetpszichológiai Szöveggyüjtemény, Debrecen: Kossuth Egyetemi Kiadó: –.
Huot, J-L. (): La ville neuve, une idée de l’Antiquité?, Paris: Éditions Errance.
Huot, J.-L., Thalmann, J.-P. et Valbelle D. (): Naissance des cités, Paris: Éditions Nathan.
Joannès, F. (): Dictionnaire de la civilisation mésopotamienne, Paris: Robert Laffont.
Kemp, B.-J. et al. (): Amarna Reports I.–IV., London.
König, F.W. (): Die elamischen Königsinschriften (= AfO Beih. ), Graz.
Kramer, S.N. (): The Sumerians. Their history and character, Chicago.
Kuhrt, A. (): Az ókori Közel-Kelet (The Ancient Near East, c. –BC I–II, London), Pázmány
Péter Katolikus Egyetem, Bölcsészettudományi Kar.
Lackenbacher, S. (): Le roi bâtisseur: les récits de construction assyriens des origines à Teglatphala-
sar III, Paris.
Lambert, M. (): «Hutélutush-Inshushinak et le pays d’Anzan,» RA : –.
Lawrence, J.C. (): «Földrajzi tér, társadalmi tér és a nagyáruház birodalma.» In Gyáni G. (szerk.,
) A modern város történeti dilemmái, Debrecen: Csokonai Kiadó: –.
Lenzen, H. (): «Mesopotamische Tempelanlagen von der Frühzeit bis zum zweiten Jahrtausend,»
ZA : –.
———. (): XXI. Vorläufiger Bericht über … Ausgrabungen im Uruk-Warka, Berlin.
Lippai, E. et Dúll, A. (): «Városutópiák környezetpszichológiai és szimbolikai elemzése», Magyar
Pszichológiai Szemle LVIII/: –.
Lynch, K. (a) The image of the city, Cambridge.
———. (b) «A város szemléletének struktúrája.» In Vidor F. (szerk., ), Urbanisztika, Budapest:
Gondolat Kiadó: –.
Margueron, J.-C. (): «Maquettes architecturales de Meskéné-Emâr,» Syria, : –.
———. (): «Le matériel, les maquettes.» In Beyer, D., Meskéné-Emâr, Paris: –.
———. (): «Fondations et refondations à l’âge du Bronze Récent au Proche Orient.» In Mazzoni,
S. (Ed.), Nuove Fondazioni Nel Vicino Oriente Antico: realtà e Ideologia, Seminari di Orientalistica ,
Dipartimento di Scienze Storiche del Mondo Antico, Sezione di Egittologia e Scienze Storiche del Vicino
Oriente, Università degli studi di Pisa, Atti del colloquio – dicembre , Pisa: –.
Moran, W.L. (): Les lettres d’El-Amarna, correspondance diplomatique des Pharaons, [traduction
française par D. Collon et H. Cazelles] (= LAPO ), Paris: Cerf.
Mofidi-Nasrabadi, B. (): Archäologische Ausgrabungen und Untersuchungen in Čoga Zanbil, Müns-
ter.
Montero-Fenollós, J.L. et al. (): «Etemenanki: nuova ipotesi di ricostruzione dello ziggurati di
Nabucodonosor II nella cittá di Babilonia,» Revista sobre Oriente Próximo y Egipto en la antigüedad
: –.
Müller, B. (): Les maquettes architecturales du Proche-Orient ancien: usage et signification. Les cases
de l’Amunia. Maquettes arqui tectoniques de l’Antiguitat (–BC) (thèse inédite).
———. (): «Le thème du «temple» vide dans l’iconographie orientale.» In Müller, B. (ed.), Maquettes
architecturales de l’Antiquité, Regards croisés, Actes du Colloques de Strasbourg – décembre ,
Strasbourg: –.
Parrot, A. (): «Les fouilles de Mari, dixieme campagne (automne ),» Syria : –.
Porada, E. (): Tchoga Zanbil (Dur Untash), vol. IV. La Glyptique (= MDP ), Paris: Geuthner.
Pons, N. (): «Tchoga-Zanbil après Untaš-Napiriša.» In Gasche, H., Tanret, M., Janssen, C. et
Degraeve, A. (eds.), Cinquante-deux réfléxions sur le Proche-Orient ancien offertes en hommage à Léon
De Meyer (= MHEOP ), Leuven: –.
Potts, D.T. (): «Elamite Temple Building.» In Bodaand, M.J. and Nobotny, J. (eds.), From The
Foundations to the Crenellations, Essays on Temple Building in the Ancient Near East and Hebrew Bible
(= AOAT ), Münster: Ugarit-Verlag: – et –.
Reiner, E. (): «The location of Anshan,» RA : –.
 zsolt gábor lantos

Roaf, M. (/): Atlas de la Mésopotamie et du Proche-Orient Ancien, Brepols.

Sauvage, M. (a): «La construction des ziggurats sous la troisième dynastie d’Ur,» Iraq : –.
———. (b): La brique et sa mise en oeuvre en Mésopotamie des origines à l’époque Achéménide, Paris.
Séra, L. (): «A pszicológia és a környezet,» Pszichológia /: –.
Steve, M.-J. (): «Textes élamites de Tchoga-Zanbil,» IrAnt : –.
———. (): Tchoga Zanbil (Dur Untash), volume III., Textes élamites et accadiens de Tchoga Zanbil (=
MDP ), Paris: Geuthner, Paris.
Steve, M.-J., Gasche, H. et De Meyer, L. (): «La Susiane au deuxième millénaire: à propos d’une
interprétation des fouilles de Suse», IrAnt : –.
Steve, M.-J., Vallat, F. et Gasche, H. (–): «Suse,» Suppl DB –: –.
Streck, M. (): Assurbanipal und die letzten assyrischen Könige bis zum Untergang Niniveh’s (= VB ),
Leipzig: Hinrichs Verlag.
Tourovets, A. (): «Observations concernant l’existence d’une ancienne voie cérémonielle au Nord-
Est du site de Tchoga Zanbil,» IrAnt : –.
Thureau-Dangin, F. (): Les cylindres de Gudéa, Paris.
Vallat, F. (): Suse et l’Elam, Paris.
———. (): «L’inscription de la stèle d’Untash-Napirisha,» IrAnt : –.
———. () «Légendes élamites de fragments de statues d’Untaš-Napiriša et Tchoga-Zanbil,» IrAnt :
–.
———. (): «L’hommage de l’élamite Untash-Napirisha au Cassite Burnaburiash,» Akkadica –
: –.
———. (): «L’Élam du IIe millénaire et la chronologie courte,» Akkadica –: –.
———. (): «La chronologie méso-élamite et la lettre de Berlin,» Akkadica : –.
Van Dijk, J.J. (): «Die dynastischen Heiraten zwischen Kassiten und Elamern: eine verhängnisvolle
Politik», Or NS : –.
Vicari, J. (): La Tour de Babel, Que sais-je?, Paris: Puf.
Vidor, F. (): «Az urbanisztika világáról.». In Vidor F. (szerk.), Urbanisztika, Budapest: Gondolat
Kiadó: –.
Woolley, C.L. () Ur Excavations, vol. IX. The Neo-babylonian and Persian Periods, London.
essai d’application de la méthode de la psychologie environnementale 

Fig. . Dûr-Untaš (Tchoga-Zanbil), plan général de la ville (d’après

Ghirshman, : plan I; Mofidi-Nasrabadi, ; plan ).
 zsolt gábor lantos

Fig. . Plan du temenos incluant les images géomagnétiques (d’après

Ghirshman, : plan I; Mofidi-Nasrabadi, ; plan ).
essai d’application de la méthode de la psychologie environnementale 

Fig. . Plan du temenos avec les portes et les voies désaxées (d’après
Ghirshman, : plan I; Mofidi-Nasrabadi, ; plan ).
 SOME CHRONOLOGICAL ASPECTS
OF THE BUILDING STRUCTURES AT HAFT TAPPEH

Behzad Mofidi-Nasrabadi*

Haft Tappeh is situated about km South-East of the ancient city Susa in the region Khuzestan
in the South-West of Iran (Fig. ). The geophysical prospection indicated that the ancient site
at Haft Tappeh should be about .×. km large (Fig. ).1 A wide area in the Northern part
of the site was excavated from  to  by E. Negahban and included a tomb building
and parts of three complexes (Fig. ).2 He distinguished two terraces made of mud bricks, near
which some halls and rooms were situated.3 Negahban named them as terrace complex I and
II. On the South-East side of the first terrace there was an artist’s workshop where some finished
and semi finished products made of bone, clay and bronze were found.4 An oven was built in
front of the workshop in its courtyard. The tomb building was situated on the North-Western
side of the terrace complexes. An important fact that was not mentioned by Negahban is the
difference between the levels of the tomb building and the terrace complexes. The pavement
of the courtyard of the tomb building lay approximately  m higher than the pavement of the
premises close to the terrace complexes.5
With the use of geophysical prospection it was possible to obtain an overview of the building
structures in the city area (Fig. ).6 In the Northern part of the city at least five monumental
complexes can be distinguished on the geomagnetic map. Some of the walls are about  m
long and more than m large. Furthermore there are other structures in the Southern part
of the city. We decided to carry out excavations in one of the constructions situated in the
Southern part of the site (area I) and also in the complex C in the Northern part (area II and
III) (Fig. ).7
Area I is situated about m South-West of the terrace complex II. In  we were
able to uncover a large part of a building made of mud brick and clay in an area of 
square meters (Fig. ). Its walls were between  and  cm large and usually covered with
a plaster of gypsum. The premises of the building were filled with rubble. In some parts we
observed a layer of ashes and rests of burnt roof beams over the floor. The large part of
the building was extremely damaged. Room , in which we found many cuneiform tablets,
seems to be the West corner of the building. Its walls were painted in red colour and the
floor was made of mud bricks which became hard due to the fire. A thick layer of ashes
with many coal pieces of tamarisk covered the floor. Within the ashes lay several cuneiform
tablets near the North-Eastern, North-Western and South-Western walls. It seems that the
tablets stood in a rack of tamarisk wood close to the walls. A narrow canal crossed the

* Johannes Gutenberg-Universität, Mainz.

1 Mofidi-Nasrabadi : –.
2 Negahban ; . Negahban (: Pl. ) distinguished two terrace complexes. The premises around the

first terrace are divided by a large wall between hall  and hall . Therefore this part was composed of two complexes.
See also Mofidi-Nasrabadi –: –.
3 Negahban : –.
4 Negahban .
5 Mofidi-Nasrabadi (in press ).
6 Mofidi-Nasrabadi –.
7 Mofidi-Nasrabadi : –.
 behzad mofidi-nasrabadi

Fig. . Map of Northern Khuzestan.

Southern part of the room to the South-West wall which could have led the water to the outside.
Hence water was used in this room to make clay tablets for cuneiform inscription. Beside tablets
we found a big mortar and a lamp. The mortar could have been used to crush the soil in order to
make fine clay for the tablets. Near the tablets lay several blades of a roughly triangular shape.
Some of them were made of good quality. Most probably they were used for engraving the
cuneiform inscription on the clay tablets. The door of the room was situated in the South-East
wall and connected it with a large square courtyard of about  ×  m. Room  must have been
a workroom for the scribes. The narrow canal in the room as well as all objects found there
indicate that the room served as a workshop for the production of clay tablets and inscribing
them. The tablets were archived in the racks near the walls.8
Besides room  there were also other rooms around the courtyard . Through a badly eroded
long room () at the South-West side of the courtyard, it was possible to reach three rooms
which were arranged one behind the other. On the North-East side of each one of these rooms
there was a small vestibule (rooms , , ). Each of them led to a long room (, , ). The
vestibules were heavily eroded and we could determine their walls only because of the traces of
their gypsum plaster.

8 See Mofidi-Nasrabadi (in press ); : –.

 some chronological aspects of the building structures at haft tappeh 

Fig. . Overview of the site extension of Haft Tappeh.

The long rooms , , and  ran parallel to each other in the South-West wing of the building.
The structure of the building didn’t allow reaching them easily from the courtyard and one had
to pass first other rooms in order to enter one of them. Their complicated accessibility points
out that the long rooms were used as storages. We hardly found any objects in the rooms, except
some rests of valuable metals, like some gold and bronze pieces in the ashes covering the floor
of room . Furthermore, several cuneiform tablets lay in this room. More tablets were found
in room  which was the anteroom of long room . The tablets are generally lists of different
objects like bows, arrows, and horse-gear which most likely were stored in the long rooms. In
view of this fact the construction could have been an administrative building in which different
objects were registered and stored.
The section with three parallel long rooms and their anterooms was separated through
a corridor (–) from the North-East part of the building, where we uncovered another
long room () made symmetrically to room . This was also not directly accessible from the
courtyard  and could be reached only through the anteroom . Its walls had a gypsum plaster
with red painting. We found only a knob of gypsum on the floor. This long room must also have
been a store.
 behzad mofidi-nasrabadi

Fig. . General plan of area excavated by Negahban

(after Negahban : Pl. , with alterations).
some chronological aspects of the building structures at haft tappeh 

Fig. . Geomagnetic map of the northern part of the city area (left); schematic plan of the
complexes based on geomagnetic plan (after Mofidi-Nasrabadi –: Abb. ).

Fig. . The areas excavated at Haft Tappeh by Mofidi-Nasrabadi.

 behzad mofidi-nasrabadi

Fig. . Plan of the administrative building in Area I in the southern part of the site.
 some chronological aspects of the building structures at haft tappeh 

Fig. . Excavated areas and geomagnetic map of the northern part of complex C.

The layer of ashes and burnt roof beams in some of the rooms shows that the building was
destroyed most probably due to an invasion The stores must have been either cleaned out by
the city inhabitants or plundered by the enemy. In room  a part of a skeleton lay on the floor.
We found a bronze arrowhead close to it which doesn’t belong to the common arrows from
Haft Tappeh. Similar arrows from the Middle Babylonian period were discovered in Susa, Uruk
and Dur-Kurigalzu.9 Probably the devastation of the building took place during a Babylonian
invasion. Thereafter the building was abandoned and used as a funerary place. We found in
room  close to room  several jar graves made of two ceramic jars situated vertically one
upon the other.10 Also there was a sarcophagus near them. In the Southern part of the building
there were simple pit graves.11 The skeletons lay in flexed position and the grave goods included
ceramic goblets and bowls. In the jar graves we found several bronze bracelets and earrings. All
graves were embedded in the rubble on the floor of the rooms or upon the floor. It means that the
burial was carried out in the ruins of the building. Since the grave goods, like pottery belong
to the first phase of the Middle Elamite period, the interments must have happened shortly
after destruction of the building. It seems that the construction was not restored in the Middle
Elamite period, because there was no evidence for younger Elamite layers. In the Sassanian
period poor housing was built in the ruins of the Middle Elamite administrative building.

9 See Tallon : , Nr. , ; Van Ess—Pedde :  and Taf. , Nr. –; Baqir : Fig. .
10 Mofidi-Nasrabadi : Taf. , .
11 Ibid.: Taf. , –.
 behzad mofidi-nasrabadi

Fig. . The wall of the later building level in complex C.

In contrast to the situation of the administrative building in the Southern excavated area, we
found in complex C in the Northern part of the site different construction layers. This complex
was determined due to the geomagnetic prospection and we were able to distinguish several
rooms in its Northern section on the geomagnetic map (Fig. ). We uncovered large walls of
mud brick of the premises in this area about – cm below the surface (Fig. ). According to
the pottery found near them, the construction belongs to the first phase of the Middle Elamite
period. This construction was the latest Elamite building in this area. It lay above a deep layer of
rubble and ashes. Under this layer we found other walls of mud brick (Fig. ). Their upper part
was situated approximately m below the walls of the later construction. So there was another
earlier building below the layer of rubble. Also the pottery close to the walls of this earlier
building seems generally to be from the first phase of the Middle Elamite period. Near the walls
of the later structure we found in some areas simple constructions made of mud and baked brick
fragments. They must belong to poor housing which was built near the later structure after it
was destroyed.
The excavations in complex C showed that there were different building levels in this area.12
They were all probably from the first phase of the Middle Elamite period, because there is no
significant difference between their potteries. The presence of various building levels in the
Northern part of the site clarifies why the tomb building excavated by Negahban lay about

12 For here used stratigraphic terminology see Echt  and Eichmann . For English equivalent see

Hachmann /: , footnote .

 some chronological aspects of the building structures at haft tappeh 

Fig. . Different building levels in complex C.

m higher than the terrace complexes. The structures which Negahban excavated must belong
to different building levels. It means that the tomb building was built later than the terrace
complexes and most probably is contemporary with the later building level in complex C
(schema ).

Schema : Relative chronology between excavated structures by Negahban and different building levels
in complex C
Tomb building ↔ Later building level in complex C
Layer of rubble and ashes
Terrace complexes ↔ Earlier building level in complex C

The absolute dating of different buildings is rendered difficult, because inscriptions from the
reign of Tepti-ahar were found near the artist’s workshop close to terrace complex I as well as in
the courtyard of the tomb building.13 Several cuneiform tablets from terrace complex I mention
Tepti-ahar and other officials from his period like Athibu.14 Therefore terrace complex I must

13 Negahban : , Negahban : –. For the inscriptions from Haft Tappeh see Herrero ,
Herrero—Glassner , , , , Glassner , Beckman .
14 HT  (Herrero/Glassner : Text ), HT  (Herrero/Glassner : Text ), HT  (Herrero/Glassner
 behzad mofidi-nasrabadi

Schema : Relative chronology between different parts of the site.

belong to Tepti-ahar’s time. On the other hand an inscribed fragment of a stele discovered in
the courtyard of the tomb building refers to sacrifices before the chariot of Tepti-ahar.15 Most
probably this stele fragment did not lie originally in the courtyard of the tomb building and
was brought there later, because another fragment which maybe belonged to the same stele
was found in the courtyard in front of the artist’s workshop in terrace complex I.16 So the
location of the first mentioned stele fragment in the courtyard of the tomb building can not
be a convincing reason for dating this building in the reign of Tepti-ahar. As mentioned above,
the tomb building must have been made in a later period. Otherwise one has to assume that
terrace complex I as well as the tomb building with a difference of about  m in their levels both
belong to Tepti-ahar’s reign.
Unfortunately, the exact location of all cuneiform tablets found by Negahban is not clear. We
only know that most of the inscriptions were found in terrace complex I. Some of them belong
to the period of Tepti-ahar while others belong to the time of Inšušinak-šar-ilani.17 Therefore,
most probably terrace complex I was used during the reign of both rulers. There are a few
personalities who seem to have lived during the reign of both of them. An individual named as
Ina is cited in the inscription H.T.  together with Athibu18 as well as in H.T.  with Adad-
ereš.19 Athibu and Adad-ereš named themselves on their seals one as servant of Tepti-ahar20 and
the other as servant of Inšušinak-šar-ilani.21 It seems that both rulers reigned successively in the
period when terrace complex I as well as the structure of the lower building level in complex C
were in use.

: Text ), HT  (Herrero/Glassner : Text ), HT  (Herrero/Glassner : Text ), HT  (Her-
rero/Glassner : Text ), HT  (Herrero/Glassner : Text ), HT  (Herrero/Glassner : Text ).
15 Reiner : .
16 Negahban : , , Herrero/Glassner : Text .
17 Glassner : –.
18 Herrero/Glassner : Text , line 0 and reverse 0.
19 Herrero/Glassner : Text , line  and .
20 Herrero : Nr. .
21 Herrero/Glassner : Text .
 some chronological aspects of the building structures at haft tappeh 

Another problem regarding the chronology of structures in Haft Tappeh is related to the
dating of the administrative building in the Southern area of the site and its relation to
complexes in the Northern part. For an absolute dating of the administrative building the
study of cuneiform tablets could give some information. Until now only a small number of
the tablets was studied.22 As mentioned above, the administrative building was destroyed and
after that abandoned. Therefore it had only one building level which belongs to the first phase
of the Middle Elamite period. Its chronological relation to different layers in complex C is not
exactly clear. According to archaeological evidence it was earlier than the layer of ashes and
rubble which was caused by an invasion. Therefore the administrative building was most likely
contemporary with the earlier building level in complex C (schema ). As argued above, the
structure below the layer of ashes in complex C was on the same level as the terrace complexes
excavated by Negahban. They must have been in use in the period of Tepti-ahar as well as
Inšušinak-šar-ilani, because Negahban found inscriptions of their time in terrace complex
I. Subsequently a large building was built there, above the layer of ashes. It is on the same level
with the tomb building and most probably was contemporary with it. In a later period after the
destruction of complex C some simple constructions of mud and brick fragments were built in
this area close to the large walls of the later building level.

Bibliography

Baqir, T. (): Excavations at #Aqar Qūf – (Iraq Supplement).

———. (): “Iraq Government Excavations at #Aqar Qūf. Third Interim Report, –,” Iraq :
–.
———. (): Aqar Quf, Baghdad.
Beckman, G. (): “A stray tablet from Haft Tépé,” IrAnt : –.
Echt, R. (): Stratigraphie (Kamid el-Loz ).
Eichmann, R. (): Uruk. Die Stratigraphie. Grabungen – in den Bereichen “Eanna” und “Anu-
Ziqqurrat” (= AUWE ), Mainz.
Glassner, J.-J. (): “Les textes de Haft Tépé, la Susiane et l’Élam au II millénaire.” In De Meyer, L. and
Gasche, H. (Eds.), Mesopotamie et Elam (= MHEOP ), Ghent: –.
Hachmann, R. (/): Review of “R. Eichmann, Uruk. Die Stratigraphie. Grabungen – in
den Bereichen “Eanna” und “Anu-Ziqqurrat”, AUWE , Mainz ”, AfO /: –.
Herrero, P. (): “Tablettes administratives de Haft—Tépé,” DAFI : –.
Herrero, P. and Glassner, J.J. (): “Haft-Tépé: Choix de textes I,” IrAnt : –.
———. (): “Haft-Tépé: Choix de textes II,” IrAnt : –.
———. (): “Haft-Tépé: Choix de textes III,” IrAnt : –.
———. (): “Haft-Tépé: Choix de textes IV,” IrAnt : –.
Mofidi-Nasrabadi, B. (–): “Archäologische Untersuchungen in Haft Tape (Iran),” AMIT –:
–.
———. (): Vorbericht der archäologischen Ausgrabungen der Kampagnen – in Haft Tappeh
(Iran), Münster.
———. (in press ): “Arbeitszimmer eines Schreibers aus der mittelelamischen Zeit,” e Rencontre
Assyriologique Internationale.
———. (in press ): “The Spatial Order in the Tomb Buildings of the Middle Elamite Period.” In Niehr,
H. and Pfälzner, P. (Eds.), (Re-)Constructing Funerary Rituals in the Ancient Near East, Akten eines
internationalen Symposiums in Tübingen im Mai , Qatna Studien .
Negahban, E. (): “Haft Tepe,” Iran : –.
———. (): Excavations at Haft Tepe, Iran, Philadelphia.

22 Prechel in Mofidi-Nasrabadi : –.

 behzad mofidi-nasrabadi

———. (): “The Artist’s Workshop of Haft Tepe.” In Gasche, H., Tanret, M., Janssen, C. and Degraeve,
A. (Eds.), Cinquante-deux reflexions sur le proche-orient ancien. Offertes en hommage à Léon De Meyer
(= MHEOP ), Gent: –.
Reiner, E. (): “Inscription from a Royal Elamite Tomb,” AfO : –.
Tallon, F. (): Métallurgie Susienne I. De la fondation de Suse au XVIII e avant J.-C., Paris.
Van Ess, M. and Pedde, F. (): Uruk. Kleinfunde II, Metall und Asphalt, Farbreste, Fritte/Fayence, Glas,
Holz, Knochen/Elfenbein, Leder, Muschel/Perlmutt/Schnecke, Schilf, Textilien (= AUWE ), Mainz.
 DISTRIBUTION, MATERIALS AND FUNCTIONS OF
THE “WALL KNOBS” IN THE NEAR EASTERN LATE BRONZE AGE:
FROM SOUTH-WESTERN IRAN TO THE MIDDLE EUPHRATES

Françelin Tourtet*

. Introduction

In the Shorter Oxford English Dictionary,1 a “knob” is defined as following: “A small rounded
lump or mass, esp. at the end or on the surface of something; a rounded protuberance, boss,
stud […]”. Thus, this word can be related to any kind of support. The expression “wall knob” has
mainly been used in Near Eastern archaeology to design a group of artefacts having the shape
of a “knob” and assumed to have been fixed to a wall. However, it is not the only expression
which has been used to designate such artefacts. One can find through the English literature
the following expressions, used more or less indifferently: “wall tiles, wall nails, wall pegs,
wall cones”. In some cases, several of these expressions have been used in one and the same
publication.2 However, these expressions do not all express completely the same idea: while a
cone barely describes a shape, the use of the words “nail” or “peg” implies that these artefacts
were aimed at being inserted into a support, here a wall. Considering the word “tile”, the artefact
designed by such a name is associated to the wall nails by its fixing method: it is either mere
a pierced plaque fixed at the wall by a nail (Fig. b), or it is associated to a hollow head which
does not fix it at the wall like a nail but requires a third (and probably wooden) element, to fix
all parts of this kind of tiles at a wall (Fig. c). Considering this, it seems logical to study all
these three associated shapes (nail, tile + fixing nail, tile + hollow head) together. In this paper,
the word “nail” will design every nail-shaped artefact which can be inserted directly into a wall
(fig. a), the word “tile” will refer to the artefacts resulting of the combination of a tile and a nail
(fig. b) or of a tile and a hollow head fixed by a third element (fig.  c), while the expression
“wall knob” describes the whole category, including the nails as well as the tiles. All of these
terms agree on the original localisation of such artefacts: on a wall.
Changing the reference language does not help further, expressions with similar differences
being also used in German3 and in French.4 The common reference to the word sikkatu will be
avoided here, since it would limit the investigation field to the Akkadian speaking sphere. Such
artefacts however have been found outside this sphere, at least in Elam, where they could have
been named huphuppu.5

* Deutsches Archäologisches Institut, Berlin.

1 rd edition, Oxford, , , s. v. “knob”.
2 See for example an early publication of the material from Rimah, where “wall nails” is used in the text (Carter

: ) and “wall pegs” in the caption of the corresponding illustration (Carter : ).
3 Knauf, Nagel and Knaufplatte are the terms commonly used in German to describe the category of artefacts

considered here (e.g. Nunn ).

4 One can find in the literature the following appellations: pommeau (e.g. Ghirshman : –, Caubet/

Pierrat-Bonnefois : ), clou (e.g. Margueron : ) or carreau d’antéfixe (Caubet/Pierrat-Bonnefois :
).
5 See Hinz/Koch : , s.v. hu-up-hu-pu-um. The authors collected the different translations which have been

proposed, expressing some doubts on the association between hu-up-hu-pu-um and decorative knob suggested by
Steve (Steve : ). Even if Akkadian was well known in the Late Bronze Age Elam, the Akkadian word sikkatu
is not yet attested there for this period.
 françelin tourtet

Fig. . Typology of wall knobs: (a) nail, (b) nail and

tile, (c) tile fixated by a wooden element. F. Tourtet.

The Late Bronze Age offers an ideal occasion to study these artefacts, the cultural intercon-
nections through the whole Near East reaching a peak in this period. At this time, wall knobs are
also widespread throughout the main part of the Near East, at least from Tchoga Zanbil (south-
western Iran, ancient Dūr-Untaš) to Meskene (Syrian Middle Euphrates, ancient Emar). Thus,
the following question can be raised: which was, or were, the function(s) of the wall knobs
throughout such a large region marked as well by a cultural koine as by local practices?

. Andrae’s Functional and Evolutionist Hypothesis

This question has often been considered as answered by W. Andrae. He assumed that there was
a direct evolutionary link between the clay cones of the Obeid and Uruk periods and the sikkatū
of the Neo-Assyrian period (fig. ).6 On the other hand, critical reviews of this hypothesis are
only scarce.
His main critic has been F. Kraus in his study of inscribed clay nails.7 He thought that the
original function could not be the one assumed by W. Andrae, because this hypothesis could
be supported by no archaeological evidence. Furthermore, he highlighted that if any one wished
to answer the question of the origin and function of the wall knobs, he should remain aware
that any proposal would remain highly hypothetical.8 Although F. Kraus dealt mainly with third
millennium or early second millennium bc material, his critique is noteworthy: if W. Andrae’s
hypothesis does not work for the third millennium bc, why should it work for the late second
millennium bc?

6 In Farbige Keramik aus Assur, W. Andrae considered them as marking the end of the building of an edifice but

first of all as decorative elements (Andrae : ). In Das Gotteshaus und die Urformen des Bauens im Alten Orient,
he tried to show that the wall knobs had their roots in mud architecture, where reeds were fixed by such nails on the
wall. Over time, they would have lost their practical function, becoming purely decorative elements (Andrae :
 ff.). R. Starr (Starr : ) and C. Hemker (Hemker : –) also proposed similar theories, assuming
that these objects belonged to the sphere of architectural decoration. E. Unger considered also an evolution from the
Sumerian foundation documents to the Assyrian specimens: “Sie [the clay nails] kommen während der ganzen Zeit
der mesop. Kultur vor und werden von den Assyrern in technischer Verbesserung pilzförmig und hohl ausgeführt
und als zigâti zu Gründungsurkunden für Stadtmauern verwendet” (Unger :  §).
7 Kraus .
8 “Die Unsicherheit dieser Vermutungen wird durch den schon erwähnten Umstand erhöht, dass wir die

Anwendungsweise des Tonnagels nur ungenügend und in ihren frühen Phasen gar nicht kennen. Wollen wir die
Frage nach der Herkunft des Tonnagelfrieses der Zeit Gudeas und der dritten Dynastie von Ur nicht mit einem
non liquet beantworten, so können wir ihre Lösung nur mit der—vielleicht nie erfüllten—Hoffnung auf neue
Ausgrabungsfunde und -beobachtungen der Zukunft überlassen” (Kraus : ).
distribution, materials and functions of the “wall knobs” 

Fig. . Evolution from the Obeid wall nail to the Assyrian

sikkatu following W. Andrae. After Andrae , p. .
 françelin tourtet

Furthermore, the method adopted by W. Andrae must be criticised. He actually did compare
material belonging to more or less three millennia, without delivering any argument sustaining
the study of such a large chronological frame. Moreover, he recognized himself that some
links were missing to his evolution scheme but was certain they did exist.9 Furthermore, he
based his assumption on the only similarity of shapes. However, he neither did take in account
the size or detailed shape of these objects (the Obeid-period examples are plain and only a
few centimetres long while the Neo-Assyrian examples are mainly hollow, sometimes fixed
to a tile, and can reach /cm in length), nor did he analyse the contexts in which they
had been found. Furthermore, even if he took into account the presence of inscriptions when
considering the inscribed wall knobs as a pendant to foundation documents—they should mark
the accomplishment of the construction10—he did not analyse the content of these inscriptions.
Even if wall knobs are attested for other periods, this paper will exclusively focus on the
Late Bronze Age, quoting the findings belonging to other periods only as possible—but not
secured—comparisons. By restricting the chronological frame in this way, it should be possible
to establish an overview of such artefacts and their function(s) for a limited period and thus to
check whether or not Andrae’s hypothesis can be relevant for this period. Aiming at pinpointing
the possible function(s) of the wall knobs, this paper will take into account as many criteria
as possible: archaeological context, material, inscription (if present), geographical distribution
and historical background.

. Geographical Distribution

First, the extent of the presence of the wall knobs through the Near East (fig. ) should be
considered taking into account different aspects of the archaeological finds: the architectural
remains in relation with which they have been found, their contexts within such architectural
remains as well as the presence or not of an inscription.

.. Buildings in Which Wall Knobs Have Been Found

Considering the function of the places where wall knobs have been found, four groups can be
distinguished, relying both on archaeological and textual data. Indeed, the inscriptions of some
of the knobs found out of context (or at least for which the context has not been published)
mention the place for which they had been manufactured.
The first category of buildings in which wall knobs have been found, recognized pretty soon
by different scholars, consists of temples.11 This is the case at Dūr-Untaš, Isin, Assur and Emar.
The situation at Nuzi requires particular attention. For R.F.S. Starr, wall knobs were associated
exclusively with sacred spaces.12 On this basis, he interpreted a room of the palace in which wall
knobs had been discovered in situ on the walls (fig. ) as a chapel.13

9 Andrae , .

10 W. Andrae once considered the inscribed wall knobs as being documents set in the masonry when finishing the
erection of a building, naming them “Schlußstein” in opposition and comparison to foundation documents (Andrae
: ).
11 Starr :  and , Carter : .
12 Starr : .
13 Starr : . This interpretation has been criticised by Heinrich (Heinrich : –). In his view, the

equation sikkatu—temple had not been confirmed. Moreover, the layout of this room in the Nuzi palace did not
correspond to the layout one would expect for a temple.
 distribution, materials and functions of the “wall knobs” 

Fig. . Attestation of Late Bronze Age wall knobs. F. Tourtet.

However, the evidence of wall knobs belonging exclusively to palaces—the second category
of buildings to which wall knobs are related to—increased over the years. Once again, this
evidence does not rely only on archaeological finds but also on textual sources. Such a relation
could be retraced in Assur,14 Ţābētu (Tell Ţābān)15 and Kār Tukulti Ninurta.16 Thus, temples
were not the only place adorned with wall knobs. Furthermore, it invalidates any interpretation
of a room as a cultic place on the only basis of the presence of knobs within the remains of this
room.
A further association between knobs and another kind of architecture, i.e. enclosure walls,
relies mainly on the textual evidence from Assur and was already highlighted by E. Unger as
being an Assyrian development of the earlier Sumerian foundation clay nails.17 As illustrated
by an example of the Middle Assyrian king Aššur-bēl-nišēšu,18 this category seems to include

14 e.g. Ass  + , text published in Andrae : –, Ebeling et al. : –, Donbaz/Grayson
: , Grayson : – (= RIMA A....).
15 e.g. T II-– = Text Nr.  in Maul , from the palace of Adad-bēl-gabbe (Maul : ,  and ).
16 T , T , T , published in Eickhoff , Pl. .–.
17 Unger :  § , Unger : .
18 The text XIII- in Ebeling et al.  refers to the city wall and is attested on several wall knobs fragments

(Ebeling et al. , XX).

 françelin tourtet

Fig. . Wall knobs found in situ in a wall at Nuzi. After Starr , Fig. .

all kinds of enclosure walls, even city walls. Also the numerous wall knobs found in close rela-
tion to the temenos wall of Dūr-Untaš19 support the existence of this third group.
A fourth and last category regroups all unclear contexts. It includes surface finds (e.g. Tell
Ţābān) as well as the material for which no context has been registered (e.g. Susa) or could be
identified (e.g. Tell Fecheriyeh), even with the help of the textual data (e.g. some of the inscribed
fragments from Tchoga Zanbil or Assur).

.. Contextualization
Once, Enrica Fiandra raised the hypothesis that wall knobs were to be found mainly in relation
with gates or doors as parts of a locking system.20 In order to verify this hypothesis, the location
of the knobs within the buildings should be taken in account.
As already mentioned, some wall knobs have been found still in situ on some walls of the
rooms L and L of the Nuzi Palace.21 Starr did not mention any relation with doors. In the Nuzi
Temple A, wall knobs were found mainly scattered over the courtyard, assumed to have been
removed from their original places in the walls of the cella.22 Here also, there is no clue that they
were originally in connection with gates or doors. In Isin or Larsa, wall knobs cannot be related
to doorways. In Isin, only one knob was found, near to the dead end of a room.23 In Larsa, some
tiles have been found but cannot be related to doorways. Even if they are not illustrated, their
description allows a comparison with well known tiles belonging to the category of the wall
knobs: “Some of these plaques are concave trapezoïds, perforated in the middle. […] Certain

19 Ghirshman : .
20 Fiandra .
21 Starr : , fig. .
22 Starr : , –.
23 IB , found in Room XXIV of the Gula temple at Isin (Hrouda : ,  and Pl. ).
 distribution, materials and functions of the “wall knobs” 

Fig. . “Bobine” and wall knob: (a) Larsa, (b) Nuzi. (a)
after Huot , Fig. ; (b) after Starr , Pl. c.

fragments, with engraved decoration, are petal-shaped”.24 This description enables a compari-
son with the Late Bronze Age plaques from Isin and Susa. Some further knobs (“bobines” in
the French description25) have been related to doors, but they have a shape which distinguish
them from the artefacts defined here as wall knobs: they are shorter, without shaft and the
head is smaller (fig. a). J.-L. Huot first considered them as possibly belonging to the category
of the wall knobs.26 Later, reviewing different possible interpretations, he interpreted them as
being part of door-locking system.27 At Tell Brak, the wall knobs cannot be specifically related
to doorways, most of them having been found in upper room fills.
At Tell al-Hamı̄dı̄ya (Syria) and Tell ar-Rimah (Iraq), some wall knobs were related to
doorways, but this cannot be considered as a systematic fact. At Tell al-Hamı̄dı̄ya, wall knobs
have been mainly found in the room R  of the so-called “Maittani Palace”, but only  of a
total of  specimens can be associated with a doorway.28
At Emar, some of the wall knobs were also related to a doorway, more precisely to the entrance
of the temple M (fig. ), but most of them are not directly related to architectural remains,29 as
it is also the case at Tell Bazi, where they were mainly found scattered in later layers.30

24 Huot et al. : .

25 Huot et al. : .
26 Huot et al. : .
27 See Huot .
28 Their findspots have been published in Wäfler , Plan .
29 Personal communication F. Sakal.
30 Personal communication A. Otto.
 françelin tourtet

Fig. . Context of the wall knobs at Emar, temple M. Margueron , p.  Fig. .
 distribution, materials and functions of the “wall knobs” 

The situation is different at Dūr-Untaš. If the specimens found inside the storage rooms at
the base of the zigurrat are not taken into account, since they were not in use but simply stored
there, then it can be said that the majority of the knobs from Dūr-Untaš had a relation with the
gates of the temenos.31 However, it must be stressed that many exemplars have been found in
the ruins of the ziggurat itself,32 thus not in relation with doors.
To sum up, even if an association between wall knobs and doorways could be verified
in some cases, it does not seem possible to consider it as being the rule. The only pattern
which seems to be identifiable is an exclusive association between wall knobs and monumental
architecture, but more detailed conclusions cannot be drawn from the published contexts of
the wall knobs.

.. Distribution of the Inscribed Knobs

Considering now the distribution of the inscribed wall knobs,33 it should be noted that they
are not attested everywhere. One can find them at Dūr-Untaš, Susa, Assur and Ţābētu only. All
these cities have a common characteristic: they all are capitals, either of a kingdom or of an
empire.34 But it is not possible to say that inscribed wall knobs have been found in every capital
of the Late Bronze Age: at Kār Tukulti Ninurta, no inscribed wall knob has ever been attested.
One could say that this situation is the result of the good or bad fortune of any excavation.
However, places like Emar, Nuzi or Tell Bazi have been excavated on a very large scale, so that
it is possible to assume that no inscribed knob had ever been used at these sites.

. Materials of the Wall-Knobs

The late nd millennium bc is not only a period of intense commercial and cultural contacts but
it is also the period in which innovations in the use of vitreous materials reach an apex. Vitreous
materials35 are not involved any more only in the manufacturing of small artefacts, e.g. beads
or small vessels, but also in the production of larger objects, for example façade adornments
and statues.36 This change applies also to faïence, which in this period is used in similar ways
to glazed terracotta, although the dimensions of faïence artefacts remains first constrained by
its lack of malleability in comparison to a clay-based paste. It seems thus that the colourful and
shiny aspect of vitreous materials reached a particular significance for wall knobs in this period,
even if clay specimens are not rare.

.. Clay Knobs

The material most commonly used for the production of wall knobs is simple terracotta.
Such items are attested at numerous sites. At Meskene, Tell Ţābān, Tell Fecheriyeh and Kār
Tukulti Ninurta, they even represent the majority of the wall knobs. It seems that specimens
manufactured in clay belong only to the category of the nails. Until now, no terracotta nail

31 Ghirshman : .
32 “Les fragments [des plaques émaillées à pommeau] étaient dispersés sur toutes les pentes et les parvis”
(Ghirshman : ).
33 Nails and tiles could both be inscribed with similar texts.
34 Following Maul’s interpretation, Ţābētu was the capital of the kingdom of Māri, a vassal of the Assyrian

overlord, and maybe also its secondo genitur (Maul : –, partic. –).
35 For a definition of the different vitreous materials (glass, faïence, glazed terracotta, frit), see Moorey : .
36 Caubet : .
 françelin tourtet

could be associated to a tile. However, within the category of the nails, the shapes could vary
greatly. In some case, the head of the knob has been painted, as attested by an example from
Tell Fecheriyeh.37

.. Glazed Knobs

Terracotta knobs could also have been decorated by the use of glaze, a vitreous cover. However,
because glazed terracotta is not restricted to the category of the nails but also includes tiles
and represents an intermediary technique between terracotta and faïence wall knobs, it seems
logical to consider it independently from both of these specific materials. Furthermore, the use
of glazed terracotta is an innovation of the Late Bronze Age, supporting the assumption that it
is an intermediate step between terracotta and faïence. Glazed terracotta wall knobs are attested
at several sites: Susa and Dūr-Untaš for Elam, Isin for Babylonia,38 Assur39 for Assyria and Nuzi
for Mitanni. Their absence in other Late Bronze Age sites does not seem to be related to the lack
of the technological know-how needed to manufacture them, other artefacts made of vitreous
materials (as well glass as faïence or glazed terracotta) being attested by many items at sites
like Tell Ţābān or Emar. Therefore, it seems that during the Late Bronze Age, the use of glazed
terracotta for wall knobs was restricted to some sites only, but the reasons for this phenomenon
remain, at present, unclear.

.. Faïence Knobs

Less frequent than the glazed knobs are those made of faïence. They are attested with certitude—
i.e. confirmed by scientific analyses—only at Dūr-Untaš and Susa. The situation at Assur and
Nuzi remains unclear, since the fabric of the wall knobs recovered at these two sites have not
yet been subjected to specific analyses.40 At Dūr-Untaš, it seems that faïence has been used only
for specimens smaller than those made of glazed terracotta, as noticed by Annie Caubet.41 This
fact could be related to the technology itself, faïence not being as malleable as clay.
At Susa, it is often difficult to date the wall knobs, even for a few inscribed items. For example,
a plaque inscribed with the name of Šutruk-Nahhunte could be attributed as well to the first as
to the second king of the name. A similar polychrome tile, assumed to be made of faïence,42 has
been found at Anšan. However, this item does not help to resolve the chronological problem,
since it is dated to around  bc.43 Being uninscribed, its inclusion in the studied corpus

37 McEwan et al. , Pl. :–. Up to now, it is the only known painted example from the Late Bronze Age.
38 The glass exemplars from Larsa (Huot et al. ) are unfortunately not illustrated. Should these exam-
ples be actually of glass, they would be the first and only known specimens of this type. Another possibility
would be that they have been made of glazed clay or faïence, as it is the case for the specimen from Isin. This
comparison would also be supported by the already mentioned description of the decoration of some of the
tiles found at Larsa (see note ). The glaze cover (and not the core) alone having been analysed (Huot et al.
: ), it is not possible to make more than assumptions whether they were made of glazed terracotta or
faïence.
39 It seems that A. Nunn did not consider the wall knobs found in the “jüngeren Ischtar Tempel” from Assur,

probably assuming they were made of faïence (“Quarzkeramik”), a material she did not take into account in her
study (Nunn :  n. ). However, the size of these artefacts and the way the glaze has been applied (see for
example Ass  in Andrae , Pl. ) seems to indicate that they are actually made of glazed terracotta and not
of faïence, the glaze not covering the whole knob but only its head.
40 R.J. Gettens analysed the composition of the green glazes on some objects from Nuzi (Gettens ), but he

made only a few remarks concerning the support of the glaze. The analysis of the faïence items from Assur carried
out by U. Löw has not been published at this day (personal communication J. Renger).
41 Caubet : .
42 No chemical analysis have been carried out on the Anšan material.
43 This dating of the Anšan evidence is the reason why the Anšan material have been excluded from this study,
 distribution, materials and functions of the “wall knobs” 

would not help further to identify the king named on the Susa exemplar. Šutruk-Nahhunte,
even the first of the name, having reigned later than Untaš-Napirriša, the Susa exemplar is in
any case younger than the knobs from Dūr-Untaš. One can conclude that Dūr-Untaš was an
innovative, maybe even experimental centre for the development of the technology pertaining
vitreous materials in general and faïence in particular. In a slightly earlier period, Nuzi played
a similar role in Northern Mesopotamia.44 Even if a direct link between both sites cannot be
established by technological studies, it should, however, not be completely excluded. This does
not mean that the technologies were transferred directly from Nuzi to Dūr-Untaš. A more likely
scenario would be that under an external impulse, maybe the contact with Nuzi craftsmen, the
Elamite arts and crafts developed further the new technologies, leading at least to the invention
of glazed bricks with a sintered quartz body,45 a production technique for which no parallel is
known either in Babylonia or Assyria.
To sum up, it seems that Nuzi and Dūr-Untaš were both centres of innovations for the
technology of vitreous materials. In other cities where wall knobs have been found, even if
the knowledge of glass-working techniques is attested by archaeological as well as by textual
sources, no relation can be made with the wall knobs. Clear relations between peculiar shapes
and materials cannot be retraced, except maybe for the complete absence of clay tiles (all are
made either of glazed terracotta or of faïence).

. The Function(s) of the Wall-Knobs in the Late nd Millennium bc

The question of the function of wall knobs has been addressed by many scholars since the early
th century. However, most of these studies focus on the material of a single site and assume
that these objects had only one function. Using a different approach and after introducing
briefly the main hypotheses about the function of the knobs, I will consider the meaning of
the presence or absence of an inscription on them, an approach which requires to consider
the possibility of a multi-functionality of the wall knobs. Hypotheses dealing with the material
from other periods (e.g. with the Early Dynastic or Neo-Sumerian foundation cones) will only
be summoned for possible—but not secured—comparisons.46

.. Main Hypotheses

Often, wall-knobs were interpreted as supports for garments.47 However, this assumption could
not be verified, neither by archaeological nor by textual evidences. On the one hand, as Astrid
Nunn emphasizes it in her study on the Assur knobs, one can doubt whether the idea of

although many knobs are known. For a discussion of the dating of the building EDD, in which wall knobs have been
found, see Stolper : –.
44 For the innovative aspect of Nuzi, see Moorey : –.
45 This material is named akti- in Elamite and Šutruk-Nahhunte claimed to be its inventor (Steve ). On the

mention of faïence in Elamite texts, see also Caubet/Pierrat-Bonnefois : .

46 Since this paper uses a different method, i.e. analyses the archaeological contexts of wall knobs and the texts

mentioning them in a well delimited chronological frame and does not consider these artefacts first as the evolution
of an older category of artefacts, any study relaying on material from other periods (e.g. Kraus , Hansen ,
Ellis , Leichty , Braun-Holzinger : ff.) cannot be used as a solid argument but can only be quoted
as possible comparisons.
47 All following examples are related to periods other than the Late Bronze Age: for the th millennium, Hemker

: . For the Neo-Assyrian period, W. Andrae considered once the wall knobs as symbolizing the fastening
of a garment, represented by wall paintings, as the continuation of a tradition from the earliest periods (Andrae
: ). A. Hausleiter considers them as possible elements on which garments were fixed, without excluding other
possible functions like decorative or even symbolic ones (Hausleiter : ).
 françelin tourtet

hanging garments on walls would not be too modern.48 On the other hand, the attestation
of wall knobs on the ziggurat of Dūr-Untaš or on the city-wall of Assur raises some doubts
on the likelihood that garments would have adorned them. If an association between wall
paintings and wall knobs is testified beyond any doubt at least by the examples from Khorsabad,
where the tiles of the wall knobs exist only as a painted motif, and if it also seems possible
to recognize such knobs with a tile in some painted friezes from Til Barsip, these depicted
specimens do not seem to be connected with garments. The example described by W. Andrae
as a Middle Assyrian carpet representation49 could also be interpreted as the representation
of a knob with its plate, since all the depicted elements (central point surrounded by floral
elements in a square, the corners of which are decorated with palmettes and whose borders
consist of a strip of alternating light and dark coloured squares) are known also from wall knobs
themselves.
In her own interpretation, Enrica Fiandra50 considered the knobs as parts of a system for
closing doors or gates (fig. ). On the one hand, the concentration of wall knobs by gates,
as shown earlier, should not be underestimated. But, on the other hand, several elements
challenge this hypothesis. First, it has been shown that there was no systematic relation
between wall knobs and doors / gates. Furthermore, if part of a closure system, some traces
of use made by the rope which would have been wrapped around them to close the door
and removed to open it would be expected if the knobs had been used regularly. But such
traces have never been documented. Moreover, in the examples Fiandra is quoting, artefacts
of other materials (like bronze or stone), being more resistant and thus probably better suitable
for such functions, have been found in contexts undoubtedly related to door closing, likely
having the function she assumed for the clay knobs. The most striking example of this is
probably a doorway from Dūr-Untaš, where the whole system has been found, completely
made of stone,51 and no reference to wall knobs of clay or faïence in its vicinity has been
published.
A third hypothesis, presented first by Starr on the basis of the Nuzi evidence and mentioned
later by Annie Caubet on the basis of parallels with the use of wall paintings, considers these
knobs as antefixes, underlying the architectonic structure of an edifice or marking the end of
beams. As in the case of the first hypothesis, this is neither supported by archaeological nor by
textual evidences. The link attested at Mari between wall paintings and architectonic structure52
cannot be transferred to the wall knobs with certainty. Considering the sites where knobs have
been found still inserted in the walls, no architectonic characteristic to which they could be
related has ever been mentioned, neither for the nd (Nuzi palace) nor for the st millennium
bc (Alter Palast, Assur).53

48 Nunn : .

49 Andrae :  fig. .
50 Fiandra .
51 Ghirshman : –.
52 Pierre : –, Pierre : –.
53 See the description of the so-called “Fliesenzimmer” (Room ) in Pedde/Lundström : –.
 distribution, materials and functions of the “wall knobs” 

Fig. . Reconstruction proposal for the gate. After Fiandra , Fig. .

.. Some Considerations on the Meaning of the Inscriptions

In all of these hypotheses, the inscriptions have never been considered as decisive for deter-
mining the function of the wall knobs and if they have sometimes been mentioned, they do
not belong to the core of the argumentation.54 If one considers these inscriptions, it has to be
noted that they are neither restricted to a particular shape or to a special place on the knobs.
Three types of inscriptions can be distinguished (fig. ). The first one is attested at Assur and
Susa: the inscriptions are pretty long and similar to those of the “foundation deposits”, record-
ing the name of the ruler responsible for the erection or repairing of the monument, naming
this monument and blessing the one who, in the future, will remember the name of his pre-
decessor when restoring it. The second group consists of shorter inscriptions, mentioning the
name of the ruler responsible for the construction, its name and the name of the god to whom
the temple is dedicated. The third group is only attested by the inscriptions of Untaš-Napirriša,
which mention neither his title nor the name of the building, but record only his own name “I,
Untaš Napirriša”.

54 For example, even if S. Heim mentioned the presence of inscriptions on some wall knobs, relating them to a
“commemorative function”, she did not mentioned the content of these inscriptions in her review of the function(s)
of the wall knobs (Heim : –).
 françelin tourtet

Fig. . The three different types of inscriptions attested on wall knobs (translation by the author).
Type (reference) Translation
 (Ass  = EŞ  = Aššur-rêm-nišēšu / vice-regent of (the god) Aššur / son of Aššur-nārāri
RIMA A...) / vice-regent of (the god) Aššur, son of Aššur-rabi / vice-regent. The
wall, which Kikia, Ikūnum, Sargon, Puzur-Aššur, Aššur-nārāri son of
Išme-Dagan, my forefathers, built, had become dilapidated. For my life
and the well-being of my city, I rebuilt (it) from its bottom to its top and
its knobs, to their place / I placed them back. The later vice-regent, /
when the wall will be dilapidated, / who will rebuild it, he will pray
Aššur and Adad and they will listen to him. May he place back the
knobs at their place.
 (T III-– = Maul  [Mannu-lū-j]ā`u, king of the land of Māri, [re]stored the temple [of the
Text ) god GN (of Ţābētu)].
 (GTZ 55) I, Untaš Napirriša.

A question has to be raised concerning the inscriptions: were they meant to be read or
not? Some being placed on the head of the knobs, one could assume they were intended to
be read. However, if the original position of the inscribed knob is taken into account, this
assumption becomes doubtful. If the in situ exemplars from Nuzi and Assur are representa-
tive of the location of this class of objects, a reconstructed height of ./. m above the
floor could be considered being normal for wall knobs. In that way, and even at this rela-
tively low height, it would have been difficult to read them when standing on the floor. This
hypothesis is also supported by at least some of the Tchoga Zanbil wall knobs, for which the
inscription had been set on an edge of the tile.56 The issue of the accessibility of the foun-
dation inscription is not restricted to those placed on wall knobs, but is more general. At
Dur-Untaš for example, such inscriptions are attested also on bricks, which were used in the
construction of walls. That these inscriptions were legible is doubtful: on the one hand, the
façades of the walls were, most probably, covered with plaster, on the other hand, one of
the inscribed faces of the brick could have been placed towards the inner of the wall and not
towards its façade.57 Thus, one can wonder whether such inscriptions were really intended to
be read by a visitor. Here, a comparison with foundation documents could let assume that
the inscriptions were not there to be read but merely to keep alive the name of the builder
and/or of the owner of the building, as it has been suggested by Kraus for earlier examples,
relating the phraseology and the vocabulary of the third-millennium clay nails to sale docu-
ments.58
What about the non-inscribed examples? One possibility could be that the meaning of these
knobs was well known to everyone and that an inscription was not necessary to establish a
parallel between inscribed and uninscribed knobs. The simultaneous use of both uninscribed
and inscribed specimens at the same sites (e.g. Assur, Ţābētu, Susa or Dūr-Untaš) prevents to
consider the possibility that the inscription was an element which importance decreased with
the time.

55Illustrated in Ghirshman , Pl. –.

56See for example Sb  (Paris, Louvre).
57 An inscription of Šilhak-Inšušinak has been written on  of the  sides of a brick, see Malbran-Labat , Text

 / takkime nº: –. See also the discussion on so-called “Display texts” by Grayson (Grayson : –).
58 Kraus : –. See also Radner , , .
 distribution, materials and functions of the “wall knobs” 

.. A Possible Multi-Functionality

If a parallel with foundation documents is possible for the inscribed knobs, it does not work for
uninscribed specimens and does not exclude the possibility that these artefacts, either inscribed
or not, could have had, simultaneously, other functions. If they were visible, they also had,
undoubtedly, a decorative function. This assumption seems to be supported by the existence
of glazed and faïence examples, animating the brick architecture with relief and colours. The
painted knob from Tell Fecheriyeh points also towards a decorative function. The decorative
function of the wall knobs is furthermore supported, at least for the Neo-Assyrian period,
by the evidence from Dūr-Šarrukin, where nails were inserted in painted motives, the whole
composition being much similar to the wall knobs including a tile.
Considering the antefixe hypothesis, the archaeological evidence neither supports nor con-
tradicts it. None of those knobs have been found still in situ, nor have been noticed traces
compatible with this hypothesis. Should they be considered as antefixe for roof beams, the low
height of the in situ exemplars from Nuzi and Assur does contradict such a hypothesis. Should
have they underlined the structure of the wall, as assumed for some non-figurative wall paint-
ings,59 such structures have never been recorded at the sites where wall knobs have been found.
The only site for which the antefixe hypothesis could be envisaged would be Emar, where wall
knobs have been found at the foot of the entrance to the temple M (fig. ), so that they could
have been used as antefixes, decorating the main façade. However, how could have they been
fastened to beam ends? From the two kinds of nails known at Emar,60 only some of the cylin-
drical shaped ones were open at both ends, indicating the likely use of a stick to fix them. This
means that such wall knobs could have been set theoretically everywhere, in a beam as well as in
a wall. The other attested shape of the Emar nails, with a sharp (and closed) end, could not have
been set in by this way, neither could have they been hammered down into beams or bricks. It
is more likely that they were integrated in the mud-brick architecture, either during the erec-
tion of the walls,61 or that a hole had been managed in the wall after its erection, necessitating
probably to seal them in place with some plaster.
The use of wall knobs to reflect architectonic structures would not be completely unparal-
leled. At the “Haute Terrasse” of Susa, wall knobs have been found, similar in shape to those of
the Late Bronze Age. Considering their findspots, the excavators proposed a reconstruction of
the façade, with rows of wall knobs set at regular intervals, most probably reflecting in a way
or in another the architectonic structure of the façade.62 But this comparison has weaknesses.
Firstly, nothing is known about the actual architectonic structure of this façade: the wall knobs
could also have been purely decorative, nothing in the published evidence allows to consider
the one hypothesis more likely than the other one. Secondly, the façade of the “Haute Terrasse”
of Susa dates back to the th millennium bc, thus this comparison cannot be considered as
more accurate than that of W. Andrae. Even the very similar shapes, with a tile fixed by a nail,
should not be considered as being a sufficient argument as long as the evolution of the fourth-
millennium clay nails down to the wall knobs of the Late Bronze Age could not be proven.

59 See above and note .

60 See Margueron , fig. .
61 This hypothesis could be assumed for the “Fliesenraum” of the “Alter Palast” at Assur, where the wall plaster was

found leaning on the tile edge (Preusser : ). However, it could also be possible that the tiles were set in place
after the erection of the wall and before its plastering, but no published element allows to prefer the one possibility
to the other.
62 Perrot/Ladiray : .
 françelin tourtet

Considering the inscribed specimens, next to the decorative function which is assumed if the
wall knobs were protruding from the wall, they had simultaneously a commemorative function:
they recorded the name of the ruler who erected the building in the walls of which the knobs
had been set. This last point could find a parallel in earlier legal texts from Susa, where the action
of the sikkatam mahaşum, the “driving in of a nail”, was performed to made public a change of
property, giving the name of the new owner of a property.63 In a similar way, F. Kraus suggested
that the use of clay cones in earlier times was used to stress the change of owner for a building:
before the setting of the cone, the building is only a “house” (E2), and only the setting of the
cone transfers its property to a god, transforming the “house” in a “temple” [E2.GN(-AK)].64
If this can be said of some inscribed wall knobs related to temples (i.e. naming gods) or
palaces (i.e. naming rulers), it is not a systematic rule. The specimens from Dūr-Untaš do not
mention every time the god who owns the temple, they sometimes bear the only name of the
ruler who built the temple. Thus, they do not stress the name of the owner of the building,
but of the one who dedicated it. Inscribed knobs are thus either a commemoration65 of the
owner and/or of the ruler who let erect the building, or they are dedicative if the inscription
explicitly gives the purpose of the action, e.g. for someone’s sake. In any case, as long as the
knobs were not hidden but set on walls, these functions do not exclude the decorative one but
all are simultaneous.

. Conclusion

To sum up, one can say that the understanding of the use of wall knobs in the Late Bronze
Age is not as clear as some papers would lead to assume. It is however possible to point out
some elements which could be relevant for the determination of the function of these artefacts.
First, wall knobs cannot be said to have been used exclusively for cultic places, but they are
more generally connected to monumental architecture, including palaces and city-walls. If
a concentration of wall knobs around gates can be observed at Dūr-Untaš, such a relation
knob / door is far from systematic. As far as the inscribed knobs are concerned, they seem
to be restricted to a few sites, all of them being capitals. Considering the material used for the
manufacture of the knobs, it seems that glazed pottery and faïence were restricted to the Elamite
capitals (Susa and Dūr-Untaš), Nuzi and Assur.
The combination of the discussed criteria (related architecture, findspot, inscription, mate-
rial) does not draw any specific cluster, which would identify a shared use of wall knobs over a
particular region. On the opposite, this rather supports the hypothesis of a highly local varia-
tion of their use. It however does not prevent the presence of similarities between different sites,
but these similarities are always restricted to a few particular aspects of the use of wall knobs.
Dealing with the function of these knobs, it is also possible to retrace parallels between
different areas. All the inscriptions on knobs can be compared to foundation inscriptions,
making durable the ownership over a building and/or the name of the instigator of the building,

63 For further details, see Malul : esp. –.

64 In his study of rd millennium and early nd millennium inscribed nails, Kraus considered that inscribed pieces
were reserved for deities, the ownership of a temple being eternal. On the contrary, transactions between people,
being recurrent and known to any person implicated in the transaction, they did neither need to be inscribed nor to
be made of non perishable materials (Kraus : ). Even if such an hypothesis cannot be used for the Late Bronze
Age (uninscribed knobs have been found in temples), it illustrates the fact that for similar objects, here clay nails,
different uses are attested, or at least should be considered.
65 Ellis compared the commemorative function of the inscribed Assyrian wall knobs to the pegs deposits of the

rd millennium bc and considered them all as foundation deposits (Ellis : ).
 distribution, materials and functions of the “wall knobs” 

bringing in a similarity with dedicatory and historical inscriptions. Moreover, the inscribed
knobs shared the decorative function of the non-inscribed knobs. Their use as antefixes or for
hanging draperies is a possibility which cannot be excluded, like their use as part of a door
closing system, but for which the lack of systematic archaeological or textual evidence should
be stressed. Thus, it seems that this category of objects is more heterogeneous than their shape
would lead to assume at a first glance. Furthermore, considering the multiple possibilities of
use, which could be highlighted only through a detailed analysis of the archaeological as well
as of the textual evidences, no general assessment on the function of the wall knobs can be
supported for the Late Bronze Age. As a consequence, it appears necessary to analyse in details
all available data related to this category of artefacts before assuming that artefacts similar in
shape attested over several millennia are directly related to each other.

Bibliography

Andrae, W. (): Die Festungswerke von Assur (= WVDOG ), Leipzig.

———. (): Farbige Keramik aus Assur und ihre Vorstufe in altassyrischen Wandmalereien, Berlin.
———. (): Das Gotteshaus und die Urformen des Bauens im Alten Orient, Berlin.
———. (): Die jüngeren Ischtar-Tempel (= WVDOG ), Leipzig.
Braun-Holzinger, E. (): Mesopotamische Weihgaben der frühdynastischen bis altbabylonischen Zeit,
Heidelberg.
Carter, E. (): “Early Assyrians in the Sinjar,” Expedition /: –.
Caubet, A. (Ed.) (): Faïences et matières vitreuses de l’Orient ancien. Étude physico-chimique et
catalogue des œuvres du département des Antiquités orientales, Paris.
Caubet, A. and Pierrat-Bonnefois, G. (Eds.) (): Faïences de l’Antiquité. De l’Égypte à l’Iran, Paris.
Donbaz, V. and Grayson, K.A. (): Royal Inscriptions on Clay Cones from Ashur now in Istanbul (=
RIM Suppl. ), Toronto-Buffalo-London.
Ebeling, E., Meissner, B. and Weidner, E.F. (): Die Inschriften der altassyrischen Könige, Leipzig.
Eickhoff, T. (): Kar Tukulti Ninurta, eine mittelassyrische Kult- und Residenzstadt (= ADOG ),
Berlin.
Einwag, B. and Otto, A. (): “Tall Bazi.” In del Olmo-Lete, G. (Ed.): Archaeology of the Upper Syrian
Euphrates. The Tishrin Dam Area. Proceedings of the International Symposium Held at Barcelona,
January th–th , Sabadell: –.
Ellis, R.S. (): Foundation Deposits in Ancient Mesopotamia, New Haven-London.
Fiandra, E. (): “Porte e chiusure di sicurezza nell’Antico Oriente,” Bollettino d’Arte : –.
Gettens, R.J. (): “Chemical and Microscopic Examination of the Green Glaze on Objects Found at
Nuzi.” In: Starr, R.F.S. (Ed.): Nuzi. Report on the Excavations at Yorgan Tepa near Kirkuk, Iraq conducted
by Harvard University in Conjunction with the American School of Oriental Research and the University
Museum of Philadelphia –, vol. I, Cambridge: –.
Ghirshman, R. (): “Tchoga-Zanbil près Suse. Rapport préliminaire de la ème campagne,” ArAs /:
–.
———. (): Tchoga Zanbil (Dur-Untash). La ziggurat (= MDP ), Paris.
———. (): Tchoga Zanbil (Dur-Untash) II: Temenos, Temples, Palais, Tombes (= MDP ), Paris.
Grayson, K.A. (): “Histories and Historians of the Ancient Near East: Assyria and Babylonia,” Or :
–.
———. (): Assyrian Rulers of the Third and Second Millennia bc (to BC) (= RIMA ), Toronto-
Buffalo-London.
Hansen, G.P. (): “New votive plaques from Nippur,” JNES : –.
Hausleiter, A. (): “Neuassyrische Kunstperiode. VI. Keramik.” In Edzard, D.O. (Ed.), Reallexikon der
Assyriologie und Vorderasiatischen Archäologie , Berlin-New-York: –.
Heim, S. (): Glazed Architectural Elements in Elam and Related Material from Luristan (unpublished
PhD thesis, New York University).
Heinrich, E. (): Die Paläste im Alten Mesopotamien, Berlin.
Hemker, C. (): “Wandnägel im Alten Orient,” MDOG : –.
Hinz, W. and Koch, H. (): Elamisches Wörterbuch, Berlin.
Hrouda, B. (): Isin—Ishan Bahriyat II. Die Ergebnisse der Ausgrabungen –, München.
 françelin tourtet

Huot, J.-L. (): “Fermetures de porte?” In: Gasche, H. and Hrouda, B. (Eds.), Collectanea Orientalia.
Histoire, arts de l’espace et industrie de la terre. Études offertes en hommage à Agnès Spycket, Neuchâtel-
Paris: –.
Huot, J.-L., Calvet, Y., Charpin, D., Cleuziou, S. and Forest, J.-D. (): “French Archeaological Mission
in Iraq: Larsa. Preliminary Report on the Sixth Campaign,” Sumer : – (the same article was
first published in French in  in Syria : –).
Huot, J.-L., Bachelot, L., Braun, J.-P., Calvet, Y. and Forest, J.-D. (): “Rapport préliminaire sur la
huitième campagne (),” Syria : –.
Kraus, F.R. (): “Altmesopotamische Tonnaegel mit Keilinschriften.” In: Halil Edhem Hâtira Kitabı
Cilt: I / In Memoriam Halil Edhem Vol. , Ankara: –.
Leichty, E. (): “Omens from doorknobs,” JCS /: –.
Malbran-Labat, F. (): Les inscriptions royales de Suse. Briques de l’époque paléo-élamite à l’Empire
néo-élamite, Paris.
Malul, M. (): “GAG.RÚ: sikkatam mahāşum/retûm ‘To Drive in the Nail’—An Act of Posting a
Public Notice,” OrAn : –.
Margueron, J.-C. (): “Architecture et urbanisme.” In Beyer, D. (Ed.), Meskéné—Emar. Dix ans de
travaux –, Paris: –.
Maul, S. (): Die Inschriften von Tall Ţābān (Grabungskampagnen –). Die Könige von Ţābētu
und das Land Māri in mittelassyrischer Zeit (= AcSum Suppl. Series ), Tokyo.
McEwan, C.W., Braidwood, L.S., Frankfort, H., Güterbock, H.G., Haines, R.C., Kantor, H.J. and Kraeling,
C.H. (): Soundings at Tell Fakhariyah (= OIP ), Chicago.
Moorey, P.R.S. (): Ancient Mesopotamian Materials and Industries. The Archaeological Evidence,
Winona Lake.
Nunn, A. (): Knaufplatten und Knäufe aus Assur (= WVDOG ), Saarwellingen.
Pedde, F. and Lundström, S. (): Der Alte Palast in Assur. Architektur und Baugeschichte (= WVDOG
), Wiesbaden.
Pedersén, O. (): Katalog der beschrifteten Objekte aus Assur. Die Schriftträger mit Ausnahme der
Tontafeln und ähnlicher Archivtexte (= ADOG ), Saarbrücken.
Perrot, J. and Ladiray, D. (): “La découverte de la Haute Terrasse,” Dossiers Histoire et Archéologie
: –.
Pierre, B. (): “Décor peint à Mari et au Proche-Orient,” MARI : –.
———. (): “Décor peint à Mari et au Proche-Orient (II),” MARI : –.
Preusser, C. (): Die Paläste in Assur (= WVDOG ), Berlin.
Radner, K. (): Die Macht des Namens. Altorientalischen Strategien zur Selbsterhaltung (= SANTAG
), Wiesbaden.
Starr, R.F.S. (): Nuzi. Report on the Excavations at Yorgan Tepa near Kirkuk, Iraq conducted by
Harvard University in Conjunction with the American School of Oriental Research and the University
Museum of Philadelphia –, vol. II, Cambridge.
———. (): Nuzi. Report on the Excavations at Yorgan Tepa near Kirkuk, Iraq conducted by Harvard
University in Conjunction with the American School of Oriental Research and the University Museum
of Philadelphia –, vol. I, Cambridge.
Steve, M.-J. (): “Fragmenta Elamica,” Or : –.
Stolper, M.W. (): Texts from Tall-i Malyan I. Elamite Administrative texts (–), Philadelphia.
Unger, E. (): “Gründungsurkunde.” In: Ebert, M. (Ed.), Reallexikon der Vorgeschichte , Berlin: –
.
———. (): “Nagelurkunde.” In: Ebert, M. (Ed.), Reallexikon der Vorgeschichte , Berlin: .
Wäfler, M. (): Tall al-Hamı̄dı̄ya . Vorbericht – (= OBO Series Archaeologica ), Fribourg-
Göttingen.
 RE-ASSESSING ELAMITE HIGHLAND BOUNDARIES:
NEW EVIDENCE FOR THE MIDDLE AND NEO-ELAMITE PERIODS
IN THE MAMASANI VALLEYS, SOUTH-WESTERN IRAN

Bernadette McCall*

. Introduction

This paper is not so much concerned with re-drawing the boundaries of an Elamite presence in
the highlands of Fars but aims instead to re-assess the extent and duration of Elamite settlements
in this region in light of recent archaeological investigations carried out by the joint Iranian-
Australian Mamasani Archaeological Project (MAP).1 Since  a programme of excavation
and survey has centred on the Mamasani district of western Fars, concentrating on several
small intermontane valleys north of the modern city of Nurabad-e Mamasani (Fig. ). Prior to
this project much of our understanding of the archaeology of Fars was limited geographically
to the high altitude plains of the Kur River basin to the south-east of Mamasani. There the
results of regional surveys and excavations have established an archaeological record covering
the Epipalaeolithic to post-Achaemenid periods, with much of this work focused on the large
multi-period site of Tal-e Malyan, ancient Anshan (Sumner , ).2 Less attention has
been paid to the smaller highland regions outside of the Kur River basin but the archaeological
potential of the Mamasani district had not escaped notice. In , on a visit to the valleys
that make up the current survey area, Ernst Herzfeld recorded Elamite period rock reliefs high
above the plains at a site known as Kurangun, the rock-cut tomb of Da-o Dokhtar, the remains
of an Achaemenid period building near the village of Jinjun,3 the mound site of Tol-e Spid
containing an inscribed Middle Elamite brick, and the ruins of a large stone wall which cross
the central pass within the valleys (Herzfeld , ). In the following decade Sir Aurel Stein
revisited these sites and undertook reconnaissance of several other mounds also visiting Tol-e
Spid and the multi-period mound known as Tappeh Sorna (Stein ).4 It was this clear
potential to add significant new information to the archaeology of Fars that led the current
project directors to focus their investigation in the Mamasani area (Potts et al. a: xiii).

* University of Sydney.
1 The combined excavation and survey results of the first stage of the Mamasani Project were initially published
in a single monograph in  with a second more widely available version published in  (Potts et al. ).
Throughout this paper references to the Mamasani Project publication refer to the second edition.
2 The chronological framework was for the most part developed out of the surveys and soundings by Vanden

Berghe in the s and Sumner in the late s (Sumner ). These regional surveys built on the results of
excavations at several key prehistoric sites, Tal-e Bakun, Tal-e Jari and Tal-e Mushki prior to the excavation of the
multi-period site of Tal-e Malyan (Sumner ). A more complete summary of archaeological investigation in the
Kur River basin with full bibliographic references can be found in Potts et al. (a: –).
3 Jinjun was briefly investigated by a Japanese team in  but has recently been the subject of ongoing

excavations during Stage Two of the Mamasani Project under the name of Qaleh Kali (Potts et al. , b for
full references).
4 For a summary of archaeological investigation conducted in the Mamasani district consult Potts et al. (a:

–) or McCall (: –).

 bernadette mccall

Fig. . Location of the Mamasani survey area.

The initial stage of the Mamasani Archaeological Project comprised three areas of investi-
gation: the first two were the excavation of deep soundings at the multi-period sites of Tol-e
Nurabad and Tol-e Spid which established a regional chronology from absolute and relative
dates spanning the Neolithic to post-Achaemenid periods (Weeks et al. ; Petrie et al.
a), and the third part of the project consisted of an extensive regional survey of the two
valleys immediately north of Nurabad-e Mamasani which mark the modern geographic bound-
ary between Fars and Khuzestan.5 Evidence of Epipalaeolithic to Islamic period occupation was
identified from surface collections, highlighting a regional settlement system with broad cul-
tural links to other areas of south-western Iran, in particular with the archaeological cultures of
the Kur River basin in Fars and the Susiana region in lowland Khuzestan (McCall ; Zeidi
et al. ).

5 The field survey component was carried out by the author in collaboration with core ICAR representatives
Alireza Khosrowzadeh and Mohsen Zeidi with additional assistance from other Mamasani project team members.
The preliminary assessment of the survey data is found in Zeidi, McCall & Khosrowzadeh (). More comprehen-
sive analysis of the regional survey data from the Epipalaeolithic to Elamite periods formed the basis of the author’s
doctoral research (McCall ) with a fuller publication on the later historical periods in preparation.
 re-assessing elamite highland boundaries 

. Geographic Location

The Mamasani survey area is composed of two valleys: the southern valley is the Dasht-e
Rostam-e Yek (Dasht-e Rostam-e One, DR), and the Dasht-e Rostam-e Do (Dasht-e Rostam-e
Two, DR) is in the northern part of the survey area.6 The valleys are situated c.  km east of
the head of the Persian Gulf within the mid-altitudes of the Zagros Mountains. The centre of
the survey area is c. km north-west of Shiraz, or c.  km north-west of Tal-e Malyan in the
Kur River basin and c. km south-east of Susa.7 The valley floors sit at an altitude of  m
above sea level on average, and the surrounding mountains, which essentially form a discrete
geographically bounded unit for the survey, rise to a maximum height of  m. The valleys
generally follow the main orientation of the Zagros Mountains on a north-west to south-east
axis. The main plain of DR is sub-rectangular measuring approximately  km wide by  to
km long. There are two smaller connected plains extending to the west and south-east. DR
is a longer narrow alluvial valley of about  km long by  to . km wide. Together they form
an area that covers approximately  square km.
Within the survey area are several environmental zones: the well-watered central alluvial
plains; a zone of drier colluvial deposits along the base of the mountains bordering the valleys;
a large alluvial fan in DR (where the site of Tol-e Spid is located); and an area of marshland
and ephemeral lake in DR (Fig. ). The valleys are also in an area referred to alternatively
as the Fahliyan region so named from one of two large braided rivers, the Rud-e Fahliyan,
that flows through DR, which itself is also known as the Dasht-e Fahliyan. There are several
concentrations of permanent springs that issue from the base of the mountains to form a
network of perennial streams and small rivers that feed the marshy area of DR. This network is
augmented by seasonal rainfall runoff forming a catchment that drains into the Rud-e Fahliyan
and the other major river, the Tang-e Shib that flows though the DR. These rivers converge
in an adjacent parallel valley and eventually join the Zuhreh River to flow into the Persian
Gulf. The climate of Mamasani is typically hot and dry in summer, cool and wet in winter
and compared to other climatic zones in south-western Iran, is less susceptible to summer and
winter temperature extremes. From available weather observations the average annual rainfall
is between about –mm and at least two cropping seasons are possible per year.8 This
relatively mild climate makes the area attractive on a year round basis; the range of exploitable
microenvironments within the valleys has supported a variety of subsistence strategies reflected
by changes in site distribution over the course of human occupation (McCall :  ff.).
Perhaps one of the more important features of the Mamasani valleys is their precise highland
location. Situated in the southern Zagros Mountains they form an integral part of a major
communication and transport corridor that links Fars and Khuzestan. Movement through the
valleys and surrounding area is made possible by several passes which connect the Mamasani
district to regional and smaller local networks within the Zagros. The most important route to
Khuzestan is accessed via the north-western pass out of DR and travels via the Behbehan, Ram
Hormuz and Mianab plains. Several passes to the south and south-east of DR form alternate

6 Various toponyms have been used for each of these valleys in the historic and recent past. The region is often
referred to as the Fahliyan district in much of the contemporary literature and the term is essentially interchangeable
(McCall : –).
7 The approximate centre of the survey zone, the Yagheh Sangar pass, is located at ° 0 00 E and ° 0 00

N.
8 A complete discussion of the environment of the survey area and the Mamasani district is found in McCall

(: –) and Potts et al. (a: –).

 bernadette mccall

Fig. . The two Mamasani valleys DR and DR showing the
location of the major environmental zones identified during fieldwork.

routes out of Mamasani all of which lead to the Kur River basin. The main southern pass through
the Dasht-e Nurabad also branches south-west towards the Persian Gulf. Other local routes
through the mountains are reached by northern passes out of each valley which connect with
Yasuj, the Bakhtiari Mountains and beyond. As noted above, in terms of size the individual
valleys of the Mamasani district are small but their combined size means they form one of only
a few open areas of fertile and habitable land between Khuzestan and upper Fars (Potts et al.
a: ).

. The Elamite Presence in the Highlands:

Historical and Archaeological Background

Although considerably smaller in size and population numbers than either of the larger regional
centres in the Kur River basin or the Susiana hinterland, Mamasani settlement trends generally
follow a similar trajectory to these larger regions. But of particular interest here, is the period of
Middle Elamite occupation. From the mid-nd millennium bc to the final Neo-Elamite period
in the first half of the st millennium bc local settlement patterns diverge from those observed
in the Kur River basin. The epigraphically attested close association of the Elamite kingdoms of
Susa and Anshan and the primacy of the mountains as the defining element of Elamite territory
during the Middle Elamite period is not clearly reflected in the archaeology of the Kur River
basin (Stolper : ; Potts : ). Middle Elamite ceramics have been documented at
the site of Tal-e Malyan (Anshan) in stratified archaeological contexts from the EDD building
 re-assessing elamite highland boundaries 

which date from c. –bc, but were not found in surface collections elsewhere in the
region (Carter : ; Sumner : ). The precise nature of Middle Elamite occupation
at Malyan remains unclear but it has been described as “an outpost of empire on the far eastern
edge of the Elamite world” seemingly with little involvement with other settlements in the
surrounding area (Carter : ). The ceramic repertoire at Malyan contains typical lowland
Elamite vessel forms which initially appeared in the archaeological record in conjunction with
highland Qaleh wares. By the Late Middle Elamite period however lowland forms dominated
the assemblage. On the basis of stylistic parallels and radiocarbon determinations occupation
in this area of Malyan is dated from c. –c. bc when the building was severely damaged
by fire (Carter : ; : –, ). The area was briefly re-used but by about bc
this part of the site was abandoned (Carter : –; Carter : –).
Only sporadic evidence of any occupation after the Middle Elamite phase has been observed
at Malyan. No substantial structures have been identified which date from the Neo-Elamite
period, apart from the few burials assigned to Building level II in the EDD operation dated to
the early st millennium bc (Carter : –, Fig. ; Carter : , Fig. ). This absence
of Neo-Elamite archaeological evidence may result from the limited extent of excavation at
Tal-e Malyan but the lack of evidence from the Kur River basin surveys for either Middle
or Neo-Elamite ceramics at sites outside of Malyan, and the uncertain dating of other local
st millennium bc ceramic traditions, makes it extremely difficult to characterise settlement
patterns after the Middle Elamite period (Sumner : –). Sumner’s proposal that the
Elamite population of Anshan left the highlands and retreated to Susa may explain the lack of
settlement in the immediate region but does not fit the evidence elsewhere (Sumner : ).
Continuity of settlement between the Middle Elamite, Neo-Elamite and Achaemenid periods
has been identified outside of Susa in the Ram Hormuz plain in Khuzestan already (Carter
), and the archaeological evidence from the Mamasani Project area thus far suggests a
more enduring period of Elamite settlement extending much further into the highlands. As
such the results discussed below may provide additional archaeological correlates for a number
of Elamite-based toponyms from the Fahliyan area that are mentioned later in the Persepolis
Fortification texts (Henkelman : ). It is against this brief archaeological and historical
background regarding Elamite settlement in the highlands that the results from Mamasani take
on greater significance.

. The Mamasani Survey Results

As noted above the Elamite potential of the Mamasani area was known to archaeologists well
before the current project. It is clear from previous investigations that the valleys were included
within the Elamite world, due primarily to the presence of the carved rock reliefs of Kurangun
in the central part of the survey area, and from the inscribed Middle Elamite brick at Tol-e Spid,
which records a dedication by Shilhak-Inshushinak (c. –bc) to the deity Kilahshupir
(Potts : ; see Potts et al. a:  for further references). What was missing until
now was more substantial evidence for Elamite settlement within the area. Based on parallels
with stratified nd to st millennium bc ceramics from the soundings at Tol-e Spid and Tol-e
Nurabad, and with comparative data from stratified Elamite sequences in Khuzestan and the
Kur River basin, the excavation and survey collections have provided confirmation of more
widespread Elamite occupation in Mamasani.9

9 All illustrations and parallels for the Elamite corpus and references to comparative data used to identify the

survey assemblage can be found in McCall (: Figs. .–. and Tables .–.). Collections used to identify
 bernadette mccall

Fig. . Frequency of Kaftari-Elamite ceramics.

Of the  sites that were recorded during the survey at least  contained evidence of Elamite
period occupation (Fig. ).10 Sites numbered up to and including MS are located in the
northern valley of DR whereas sites numbered above this are located in the southern DR
plains. Given the small sample base and limited surface collections the data is presented as raw
sherd counts and the relative proportions of Kaftari-related ceramics (plain and painted wares),
Qaleh-type wares, Middle Elamite wares and Neo-Elamite sherds are shown at each site. The
principal difficulty in dating the surface finds was differentiating between ceramics of the later
Middle Elamite and Neo-Elamite phases. Some diagnostic sherds displayed characteristics that
could be assigned to either of these phases at other sites and in these instances relative dating
relied upon the combination of local fabric and form parallels and the dates of the majority of
published parallels. This problem was more apparent with ceramics from MS, MS, MS,
MS and MS and each of these sites would benefit from further investigation. On the basis
of surface distribution alone this discussion of settlement focuses on the main survey sites
identified for each period: MS, MS and MS for the Middle Elamite period and MS,
MS, MS, MS and MS for the Neo-Elamite period.

the surface finds come from Susa, Haft Tepe, Tepe Sharafabad, Chogha Zanbil and Tepe Farukhabad in Khuzestan
and Deh Luran, from surveys conducted in the Susiana plain, the Mianab plain, the Eastern Corridor, the Izeh and
Ram Hormuz plains and Bushehr surveys. To the south-east of Mamasani, assemblages from the Kur River basin
survey and excavations at Tal-e Malyan and Darvazeh Tepe were consulted. However the major advantage of the
current survey was the ability to match wares with stratified ceramics derived from the local mid nd millennium
Kaftari to Elamite period strata at Tol-e Spid and Tol-e Nurabad.
10 Sites are here referred to by their survey number, prefaced with MS rather than by local names.
 re-assessing elamite highland boundaries 

Fig. . Distribution of Middle Elamite sites in the survey area.

. Middle Elamite Settlement Patterns

Fourteen sites from the survey area were dated from the middle to late nd millennium bc based
on the presence of Middle Elamite or Qaleh related ceramics (MS, MS, MS, MS, MS,
MS, MS, MS, MS, MS, MS, MS, MS and MS—Fig. ). With just over 
per cent of the total surface collection assigned to the Middle Elamite period, MS (Tappeh
Dozak) stands apart as the major site of this period. Middle Elamite ceramics were also found
in significant quantities at MS and MS, in conjunction with local painted and plain ware
sherds typical of the Qaleh ware tradition of the Kur River basin. The relative proportion of their
surface components for this period indicates that these sites were also important population
centres. Each of the remaining sites contained fewer diagnostics sherds.
Middle Elamite sites are located predominantly in DR or in the northern part of DR.
MS and MS represent a continued occupation presence from the Kaftari period and each
is located on the alluvial plain of DR towards the southern side of the valley next to a small
perennial river. MS on the northern side of the plain in DR is one of the few sites in the
area which contained Kaftari period painted wares but its smaller Qaleh/Middle Elamite period
surface component indicates it experienced a decline in population at this later stage. By far the
most noteworthy site of the period is MS. It was established sometime in the middle of the
nd millennium bc and is also found on the southern side of the DR. But unlike earlier sites
it was set against the foot of the mountains rather than on the more fertile central plain, and is
much closer to the central pass which connects DR and DR. MS occupies an area that has
no known settlements from any other period and which is still only sparsely populated today.
MS (Tol-e Spid) had only a limited Middle Elamite/Qaleh period surface component making
 bernadette mccall

Fig. . Middle Elamite ceramics from MS and MS.

it difficult to discuss the nature of occupation during this period.11 Much like the sites of MS
and MS the prominent position of Tol-e Spid on the edges of the alluvial fan in DR, close
to the two main passes at the northern end of that valley, represents continuity from previous
settlement patterns. Two sites that are also worthy of note are MS,12 which marks the position
of the Kurangun rock carvings high above the valley floor and the small site of MS, situated
at the foot of the mountains just below the reliefs.

. Middle Elamite Ceramics

Mamasani survey ceramics of this period include various typical Middle Elamite vessel forms
which were found to have clear lowland parallels, as well as the painted and plain wares more
typically associated with Kur River basin Qaleh wares (Fig. ). Stratified lowland parallels
were found with ceramics from Susa Ville Royale, Haft Tepe and Tepe Sharafabad (McCall
: Tables .–.). Highland Qaleh ware and Middle Elamite parallels were found at Tal-
e Malyan in the Middle Elamite building complex in Operation EDD that dates from c.  to
bc (Carter : ). Local parallels were matched by form or fabric to ceramics from
Phases  and  at Tol-e Spid, which include Late Kaftari/Qaleh type wares and Middle
Elamite/Qaleh wares respectively, and to Phases A and A at Tol-e Nurabad, covering only

11 The sequence of Kaftari period levels (Phases –) in the Tol-e Spid sounding indicates that the site was a

major settlement of the period. The levels containing Middle Elamite and Qaleh ceramics (Phase ) are more mixed
in character with ceramics of earlier periods and the nature of occupation during this phase is uncertain (Petrie et
al. a: –, –).
12 The remains of a small building were found just behind the platform which houses the Kurangun reliefs, but

they are probably of a later date as surface finds were mostly Sasanian or later.
 re-assessing elamite highland boundaries 

the Late Kaftari/Qaleh phases (Petrie et al. a; Weeks et al. ).13 The painted wares at
MS and MS (MS–) show affiliations with later Kur River basin painted ceramic
assemblages, particularly Qaleh wares from Malyan and to a lesser degree with Shogha-
Teimuran pottery types but they also display a regional Mamasani character (D.T. Potts,
pers.comm.). Further reinforcing these highland links, especially at MS and MS, is the
presence of a class of handmade coarse ware jars (MS–) with a distinctive dark grey core
and signs of external blackening which are comparable to so-called ‘cooking pots’ from Tal-e
Malyan (Carter :–). Painted wares (MS–) were also paralleled with materials
found in survey collections in the Ram Hormuz, Izeh and Bushehr regions showing that the
distribution of these pottery types covered a broad region, which included Mamasani.
Unlike the other Mamasani sites there were no clearly highland-related plain or painted
wares in the surface collection from MS. The assemblage was dominated by lowland Elamite
ceramic forms of which the majority were plain wares (Fig. ). Two bases, both with traces of
a red slip or wash (MS– and MS–) are among the earliest dated finds and hint at
an establishment date for MS sometime in the middle of the nd millennium bc, perhaps
contemporary with the Late Kaftari period of Fars or early Middle Elamite I in Khuzestan
and was probably part of the same move into Fars which led to an Elamite occupancy at Tal-
e Malyan.14 Parallels with ceramics from Malyan were found in the MS corpus but only
with forms which are typical of lowland Elamite assemblages representing an imported style
in both Mamasani and the Kur River basin (MS–, , —Carter :). A point
of difference between MS and other sites in Fars to date is the relative lack of painted wares.
These ceramics were not common at MS and only three diagnostic sherds were collected
(e.g. MS–). They display design elements which can be paralleled with Qaleh wares from
the Middle Elamite building at Tal-e Malyan (level IV) but similar motifs are also found on
contemporary ceramics from Izeh, Bushehr, and Ram Hormuz making it difficult to determine
which area or areas were more influential on the painted wares found at MS. Unlike at MS
and MS there were no highland-style cooking pots at MS. This absence implies further
cultural differences in the makeup of their respective populations at a domestic level.
This point, in conjunction with the overall nature of the surface assemblage, could also be
interpreted as evidence of functional differences between MS and other Mamasani sites, an
argument that is bolstered by the proportion of diagnostic sherds from large storage vessels at
MS (MS–, ). Along with the smaller typical Elamite ceramic items such as bowls,
goblets and small jars which could be used for serving or consumption, there was a high
proportion (c. ) of larger capacity storage vessels, with rim diameters falling between  cm
to cm. This is considerably different to the corpus at Malyan where large vessels accounted
for only  of the assemblage and while the high proportion of such vessel types at MS
may reflect a bias in the surface collection, the possibility that these vessels types provide
further evidence of a specific functional nature cannot be ruled out. Overall this site has the
greatest variation of vessel forms compared to other Mamasani survey sites and displays all the
classic elements of the lowland Middle Elamite ceramic repertoire and virtually no highland
types. MS represents a departure from other highland areas of Fars in the mid to late nd
millennium as does the wide distribution of Middle Elamite ceramic styles in association with
local highland ceramic wares.

13 Unfortunately neither of the local soundings displays a clear stratigraphic progression between Qaleh and

Elamite levels so the use of external parallels is important but also not without problems.
14 The solid base is paralleled with Susa Ville Royale A, Group b elongated vases from levels AXIV–AXII

(Gasche : Pl. , ; Carter : Fig. .; .) and the other base with forms from Levels AXIII-AXII. If
an early date can be established it would be tempting to propose that MS may have been settled not long after the
Kurangun rock reliefs were carved or represent contemporaneous expansion into the region.
 bernadette mccall

Fig. . Middle Elamite ceramics from MS.

. Neo-Elamite Settlement

Evidence for Neo-Elamite settlement was tentatively identified at  sites in the survey (MS,
MS, MS, MS, MS, MS, MS, MS, MS, MS and MS) but at most of these the
surface collections were small. Given the inherent lack of chronological certainty in dating sites
particularly from limited surface assemblages, the possibility that some represent later Middle
Elamite occupation cannot be completely ruled out. Several of the proposed Neo-Elamite sites
 re-assessing elamite highland boundaries 

contain pottery that combine characteristics of both Middle and Neo-Elamite ceramics, and
no doubt the inclusion of some of these sites in this phase may need to be revised as local
sequences are further refined. However from the results to date it appears that settlement
continued into the Neo-Elamite period at MS. The relatively high frequency of surface finds
of this phase compared with other survey sites attests to its longstanding regional importance.
Similarly the proposed Neo-Elamite occupation phase at MS suggests further continuity
in settlement patterns from the Middle Elamite period. The limited chronological precision
provided by the surface collections also presents difficulties in determining the duration of Neo-
Elamite occupancy, but at least six sites, all located in DR (MS, MS, MS, MS, MS, and
MS) contain compelling evidence for later Neo-Elamite II (c. late th to late th centuries bc)
occupation. Few sherds were collected from the surface at Tol-e Spid and until further intensive
survey can be carried out there is little evidence of other sites between DR and Tol-e Nurabad
to the south.

. Neo-Elamite Ceramics

The identification of Neo-Elamite period wares relied mainly on parallels with ceramics from
Neo-Elamite levels at Susa Ville Royale II, Susa Ville Royale-Apadana and Chogha Zanbil
(Miroschedji a; Miroschedji b; Mofidi Nasrabadi ) and with local ceramics from
the Tol-e Nurabad sequence in phases B to B (Weeks et al. ). As for the Middle Elamite
sites there was still a variety of small and large vessel types among Neo-Elamite forms but
generally the profiles tended to be less elaborate (Fig. ). Fabrics were similar to those of the
previous phase, mostly light brown or light reddish-brown pastes with fine mineral, vegetal,
occasional limestone inclusions, and crushed sherd temper in some instances. Surfaces were
generally slipped and smoothed but were not as well finished as earlier Elamite wares, which in
many respects matches the published descriptions of fabrics of Neo-Elamite wares at Susa Ville
Royale II (Miroschedji a: –).15 The later Neo-Elamite period ceramics were identified
on the basis of parallels found in Neo-Elamite II levels, dated from c. /bc to c.  B.C
at Susa Ville Royale II, levels – and Ville Royale-Apadana levels A–B; with the slightly
earlier Neo-Elamite I–II levels at Chogha Zanbil, Phase , c. th–th centuries bc and from
the proposed Neo-Elamite II phase B at Tol-e Nurabad (Miroschedji a:; Tab. ; Mofidi
Nasrabadi : –; –; Potts et al. a: Fig. .). 16

. Conclusions and Future Directions

In conclusion there are several points of note arising from the survey results. Settlement and
excavation evidence from Mamasani shows that this part of the southern Zagros highlands
not only lay on the route to Anshan but was itself incorporated into the Elamite sphere by
the middle of the nd millennium bc and remained so well into the st millennium bc. This

15 Local fabrics of this period have not as yet been characterised and this general overview of survey wares is
based on visible features only.
16 The potential Neo-Elamite levels identified for Tol-e Nurabad (Phases Bb and a) still await confirmation from

absolute dating methods and have been proposed with some caution (Petrie et al. b: ). The identification of
surface finds is based therefore on these cautionary dates and on the limited sequences from Susa and Chogha Zanbil
and may also require some revision as the chronology of this period is further refined. Further details of these finds
and parallels are presented in the author’s doctoral dissertation (McCall ).
 bernadette mccall

Fig. . Neo-Elamite ceramics from MS, MS, MS and MS.

contrasts with the archaeological data from highland Fars where there is no evidence yet of Neo-
Elamite occupation and population trends in the nd millennium followed a slightly different
trajectory. Rather than witnessing a decline in settlement from peak levels in the Early to Middle
Kaftari period as was the case in the Kur River basin (Sumner : ; Tab. ), the population
of the Mamasani area continued to grow from the Late Kaftari to Middle Elamite periods.
A further point of difference between Mamasani and the Kur River basin is the combination
of highland Qaleh-type ceramics and Elamite wares that are found at several Mamasani sites.
In the Kur River Basin a clear geographic division was observed between settlements with
Kaftari, Qaleh, Middle Elamite and Shogha-Teimuran ceramics (Sumner : –, ).
A mixing of highland and lowland regional ceramics was observed in the earliest Middle
Elamite levels at Tal-e Malyan but not at any other sites (Carter : –). In the Mamasani
communities there is more evidence of intermingling of ceramic traditions in survey and
excavation assemblages. Although the collection from MS sets the site apart culturally
and presumably ethnically from contemporary Mamasani sites, the combination of Qaleh
and Middle Elamite wares in the Mamasani region is widespread. This indicates that local
populations may have interacted to a greater extent with the occupants of MS and can be
interpreted as evidence that the region was assimilated to some degree within the Elamite
sphere.
Perhaps the most important site for understanding the nature of Elamite settlement in the
region is MS. It was a single period site located in an area of DR that had not been settled
previously and in terms of size only the multi-period mound sites of Tol-e Spid and Tol-e
Nurabad cover a larger surface area. Field observations suggested that the low mound of MS
covered only ha, but later analysis of aerial photographs suggests the surrounding area may
contain other remains related to the settlement which warrant further intensive survey. The
 re-assessing elamite highland boundaries 

appearance of the distinctive lowland Elamite ceramic assemblage at this new site is in part
similar to the sudden appearance of these lowland wares at Tal-e Malyan during the Middle
Elamite period, but unlike Malyan, MS represents a completely new settlement and provides
evidence of a singular Middle Elamite presence in highland Fars. The surface collection con-
tained a small number of Middle Elamite diagnostic sherds that may belong to an earlier phase
than the EDD building at Malyan but this relative chronological position could change with
further investigation of Kaftari to Qaleh levels at Malyan. For the present it would seem that
the Elamite settlement of MS and the Mamasani area pre-date the Elamite presence at Tal-e
Malyan but the character of the surface assemblage points to a complementary role between
the sites.
The variety and types of vessels identified in the MS surface collection provide some
evidence which hints at a specific functional role for the site and one that differs to its
neighbours. The concentration of large capacity vessels indicates that commodity storage was
important and implies a scale of organisation that is not seen at the other sites. A later
Neo-Elamite text underlines the economic importance of grain to the economy of Elam and
describes the centrally organised systems which were in place to collect and redistribute grain
(Henkelman :  for further references). The storage vessels at MS are found alongside
a variety of smaller vessels typically associated with the serving or consumption of food which
adds weight to the hypothesis that the site played a role in distributing supplies or rations to a
local Elamite population and to those who frequented the site on the way to or from Anshan.
A large quantity of administrative texts found at Malyan attest to exchanges of precious metals
and, less frequently food and utilitarian items which no doubt were moved along the main route
to Khuzestan through Mamasani (Stolper : ; Potts : , ). Just as the Kurangun
rock-reliefs signal the inclusion of the Mamasani district in the Elamite realm from the nd
millennium bc onwards, so too does the establishment of MS. The investment of resources
necessary to create the Kurangun monument and to maintain commercial interests at Anshan
could both have been met from a settlement such as MS.
If the trade in commodities from Tal-e Malyan declined in importance during the final
stages of the Middle Elamite period and eventually ceased, there appears to have been no
accompanying retreat of Elamite interests from the Mamasani district. The discovery of a
Middle Elamite III brick at Tol-e Spid attributed to Shilhak-Inshushinak, –bc attests
to a substantial building phase late in the period which was not matched at Malyan only a few
decades later when the EDD building at Tal-e Malyan was destroyed by fire and not rebuilt
(Carter : ; : –, ). The mounting evidence for Neo-Elamite occupation at
several locations in the Mamasani district further reinforces the view that a more settled Elamite
population remained in Fars even if the survey data suggests a smaller overall population
occupying a reduced number of sites. It is impossible to say if all sites were contemporary during
the final Elamite phase, but if so the choice of site locations did not alter to any great degree
representing a level of continuity from the Middle Elamite period.
Further evidence that an Elamite population remained in Fars throughout the early st
millennium bc can be found in the Achaemenid period Persepolis Fortification archive. The
continued use of Elamite as an official record-keeping language for the majority of texts
demonstrates that people who used and understood Elamite were still present in Fars. Scrutiny
of personal names and toponyms in the texts further supports this view as ten per cent of these
names are identified as Elamite and many of the toponyms are concentrated in the Fahliyan
area (Henkelman : , n. ). It would be hard to argue that the continued use of the
language and the existence of these place names would have been possible without an Elamite
population reservoir within the highlands. With no evidence to date of Neo-Elamite settlement
elsewhere in highland Fars, the results of the survey and excavations illustrate how important
 bernadette mccall

the Mamasani area will be for understanding the final stages of the Elamite kingdom and the
eventual rise of an Achaemenid state. The Mamasani survey and excavation results have now
identified an area where there appears to have been no prolonged gap in occupation between
the Middle Elamite, Neo-Elamite and Achaemenid periods. While the situation remains unclear
further south-east in the highlands, it is apparent that the limits of Elamite settlement in the st
millennium bc can now be firmly extended into the highland valleys of western Fars.

Bibliography

Carter, E. (): “Archaeology”, E. Carter & M.W. Stolper, Elam: Surveys of Political History and
Archaeology, Berkeley, –.
———. (): “Bridging the gap between the Elamites and the Persians in southeastern Khuzistan”,
H. Sancisi-Weerdenburg, A. Kuhrt & M.C. Root (eds), Achaemenid History VIII: Continuity and
Change, Leiden, –.
———. (): Excavations at Anshan (Tal-e Malyan): The Middle Elamite Period (University Museum
Monographs ; Malyan Excavation Reports ), Philadelphia.
Gasche, H. (): La poterie élamite du deuxième millénaire a.C. (MDP ), Leiden.
Henkelman, W.F.M. (): The Other Gods Who Are: Studies in Elamite-Iranian Acculturation based on
the Persepolis Fortification Texts (Achaemenid History XIV), Leiden.
Herzfeld, E. (): Archaeological History of Iran, London.
———. (): The Persian Empire. Studies in Geography and Ethnography of the Ancient Near East,
Wiesbaden.
McCall, B. (): The Mamasani Archaeological Survey: Epipalaeolithic to Elamite settlement patterns in
the Mamasani district of Zagros Mountains, Fars province, Iran, Ph.D. diss., University of Sydney.
Miroschedji, P. de. (a): “Fouilles du Chantier Ville Royale II à Suse (–) I. Les niveaux
élamites”, Cahiers de la Délégation archéologique française en Iran , –.
———. (b): “Observations dans les couches néo-élamites au nord-ouest du tell de la Ville Royale à
Suse”, Cahiers de la Délégation archéologique française en Iran , –.
Mofidi Nasrabadi, B. (): Archäologische Ausgrabungen und Untersuchungen in Čoğā Zanbı̄l, Münster.
Petrie, C.A., Asgari Chaverdi, A. & Seyedin, M. (a): “Excavations at Tol-e Spid”, D.T. Potts, K. Rous-
taei, C.A. Petrie & L.R. Weeks (eds), The Mamasani Archaeological Project Stage One: A Report on the
First Two Seasons of the ICAR-University of Sydney Expedition to the Mamasani District, Fars Province,
Iran (BAR S), Oxford, –.
Petrie, C.A., Weeks, L.R., Potts, D.T. & Roustaei, K. (b): “Perspectives on the Cultural Sequence of
Mamasani”. D.T. Potts, K. Roustaei, C.A. Petrie & L.R. Weeks (eds), The Mamasani Archaeological
Project Stage One: A Report on the First Two Seasons of the ICAR-University of Sydney Expedition to
the Mamasani District, Fars Province, Iran (BAR S), Oxford, –.
Potts, D.T., Roustaei, K., Petrie, C.A. & Weeks, L.R. (eds). . The Mamasani Archaeological Project
Stage One: A Report on the First Two Seasons of the ICAR-University of Sydney Expedition to the
Mamasani District, Fars Province (BAR S), Oxford.
Potts, D.T., Roustaei, K., Weeks, L.R. & Petrie, C.A. (a): “The Mamasani District and the Archae-
ology of Southwestern Iran”. D.T. Potts, K. Roustaei, C.A. Petrie & L.R. Weeks (eds), The Mamasani
Archaeological Project Stage One: A Report on the First Two Seasons of the ICAR-University of Sydney
Expedition to the Mamasani District, Fars Province, Iran (BAR S), Oxford, –.
Potts, D.T., Asgari Chaverdi, A., McRae, I.K., Alamdari, K., Dusting, K., Jaffari, J., Ellicott, T.M., Setoudeh,
A., Lashkari, A., Rad, S.A. & Yazdani, A. (b): “Further excavations at Qaleh Kali (MS ) by the
Joint ICAR-University of Sydney Mamasani Expedition: Results of the  season”, IrAnt , –
.
Potts, D.T., Asgari Chaverdi, A., Petrie, C.A., Dusting, A., Farhadi, F., McRae, I.K., Shikhi, S., Wong, E.H.,
Lashkari, A. & Zadeh, A.J. (): “The Mamasani Archaeological Project, Stage Two: Excavations at
Qaleh Kali (Tappeh Servan/Jinjun [MS])”, Iran , –.
Potts, D.T. (): The Archaeology of Elam: Formation and Transformation of an Ancient Iranian State
(Cambridge World Archaeology), Cambridge.
Stein, A. (): Old Routes of Western Iran, London.
Stolper, M.W. (): “Political History”. E. Carter & M.W. Stolper, Elam: Surveys of Political History and
Archaeology, Berkeley, –.
 re-assessing elamite highland boundaries 

Sumner, W.M. (): Cultural Development in the Kur River Basin, Iran: An Archaeological Analysis of
Settlement Patterns, Ph.D. diss., University of Pennsylvania.
———. . “Anshan in the Kaftari Phase: Patterns of Settlement and Land Use”, L. de Meyer &
E. Haerinck (eds), Archaeologia Iranica Et Orientalis: Miscellanea in Honorem Louis Vanden Berghe.
Gent, –.
———. . “Archaeological Measures of Cultural Continuity and the Arrival of the Persians in Fars”,
Sancisi-Weerdenburg, H., Kuhrt, A. and Cool Root, M. (eds), Achaemenid History VIII: Continuity
and Change. Leiden, –.
———. (): Early Urban Life in the Land of Anshan: Excavations at Tal-e Malyan in the Highlands of
Iran (Malyan Excavation Reports / University Museum Monographs ), Philadelphia.
Weeks, L.R., Alizadeh, K.S., Niakan, L. & Alamdari, K. (): “Excavations at Tol-e Nurabad”. D.T. Potts,
K. Roustaei, C.A. Petrie & L.R. Weeks (eds), The Mamasani Archaeological Project Stage One: A Report
on the First Two Seasons of the ICAR-University of Sydney Expedition to the Mamasani District, Fars
Province, Iran (BAR S), Oxford, –.
Zeidi, M., McCall, B. and Khowsrowzadeh, A. (): “Survey of Dasht-e Rostam-e Yek and Dasht-e
Rostam-e Do”. D.T. Potts, K. Roustaei, C.A. Petrie & L.R. Weeks (eds), The Mamasani Archaeological
Project Stage One: A Report on the First Two Seasons of the ICAR-University of Sydney Expedition to
the Mamasani District, Fars Province, Iran (BAR S), Oxford, –.
 BRAIDS OF GLORY.
ELAMITE SCULPTURAL RELIEFS FROM THE HIGHLANDS:
KŪL-E FARAH IV*

Javier Álvarez-Mon**

Denagh was kneeling, sitting back on her heels, with her back to Mani, who,
with an accustomed hand, was rebraiding her hair, which had come undone.
Malchos was speechless. Normally, he thought, it is young girls who plait
the hair of warriors; so who is this descendant of Parthian warriors who is
taking such pains to braid a woman’s plait! (…). when Mani’s companion
was peaceful, serene, she instinctively wore the plait in front over her right
shoulder; when she felt joyful, but her joy was mixed with expectation,
impatience, she wore it over her left shoulder; finally, when she was anxious,
distressed, unhappy, her plait hung down her back. During the coming
period, Denagh’s plait would not remain for long in the same place.
Amin Maalouf, The Gardens of Light ().

. Introduction: Sculptural Reliefs from the Elamite Highlands

The Zagros highland region of Īzeh/Mālamı̄r is nested in a mountain valley located about 
meters over sea level and extending approximately  km2 [Fig. ]. High rainfall and snow
melting runoffs generate streams and springs contributing to the creation of two large seasonal
lakes occupying the central part of the valley. Excellent grazing and mountain slopes including
oak trees add to the stunning scenery. In the words of A.H. Layard (: ): Mál Amír is
perhaps the most remarkable place in the whole of the Bakhtiyárí Mountains. On all sides the
most precipitous mountains rise almost perpendicularly from the plain. Carved over the sides of
cliffs and boulders in four different locations are found a total of twelve Elamite bas-reliefs
without parallel in the artistic historical record of the ancient Near East: four in Shekaft-e
Salmān (henceforth SS), six in Kūl-e Farah (henceforth KF), one in Shāh-Savār, and one in
Xong-e Azdar (Hung-i Naurūzı̄-i) [Figs. –].
By any chronological estimation, the reliefs are considered to be more than  years old.
Those from Shāh-Savār and Xong-e Azdar are generally dated within the Old Elamite period.
Most scholars presently date the reliefs from Shekaft-e Salmān to the end of the Middle Elamite
period, or sometime between  and  bc.1 A specific time for five of the six reliefs carved
at Kūl-e Farah, however, is far from being asserted. A late Neo-Elamite date for KF I is accepted
in accord to the contents of a large Elamite cuneiform inscription engraved over its surface but

* This study began in April  after a memorable week spent in Īzeh/Mālamı̄r accompanied by Dr. Djāffar

Mehrkian and members of the Ayapir Research Center (ARC). This article is dedicated to D. Mehrkian and the ARC
in hope to foster and further promote the candidacy of The Natural Landscape of Īzeh for inclusion in the inventory
of UNESCO World Heritage.
** University of Sydney.
1 For details on chronology see below: § . Chronological Considerations.
 javier álvarez-mon

there is lack of consensus regarding KF II, III, IV, V, and VI.2 These five reliefs used to be con-
sidered together as a single group sculpted sometime during the Neo-Elamite period or, more
precisely: the “époque Elamite récente” (Jéquier : ; Vanden Berghe : ); the period
of Elamite-Persian cohabitation (Calmeyer : ); a broader Neo-Elamite sequence ranging
from the th to the th centuries bc;3 and a period comprising the th–th centuries bc for all
reliefs, including those of Shekaft-e Salmān (Vanden Berghe : –). In  Carter
(: ) proposed a pre-bc date for KF IV. This was followed by Amiet (: –)
who, making exception for KF I and V, suggested the reliefs manifested the expression of a local
monarchy that developed in eastern Elam after the invasion of the Babylonian king Nebuchad-
nezzar I (ca. bc), or perhaps slightly later, at the beginning of the th century bc.
The earliest reference to the existence of the reliefs was made in  by H.C. Rawlinson
() followed by A.H. Layard () whose work included a preliminary account of the
cuneiform inscriptions.4 A study of choice reliefs was attempted by members of the French
archaeological mission in Persia; in particular: G. Jéquier (), together with the ensuing
epigraphic work of father V. Scheil, and sketches made by J. de Morgan. More than half a century
later, a Belgian archeological mission under the direction of L. Vanden Berghe () began
a project of documentation followed by an in-depth study undertaken by Eric De Waele as
the main focus of his doctoral dissertation.5 Work by De Waele lasted between  and 
(totalizing close to two months of actual in-field work). The end result came in the form of an
unpublished doctoral dissertation and a string of seven articles including choice illustrations
and line-drawings.6 Despite the considerable value of these investigations, however, a systematic
and inclusive analysis supported by detailed photographic documentation and line-drawing
has been missing.7 This absence has prevented from establishing constructive chronological
and interpretative parameters and from articulating the originality and significance of these
cultural manifestations of the Elamite highlands.8
In April  I had the opportunity to visit Īzeh/Mālamı̄r and make a digital photographic
record of the reliefs. I have examined these photographs alongside previous published photo-
graphic records and used the combined result to propose descriptions, and create composite
line-drawings. In addition to its prospective academic value, this work seeks to establish a
blueprint of documentation by which to encourage the protection and eventual restoration of
this important manifestation of the rich cultural heritage of ancient Iran.9

2 The reliefs are cited according to the classification proposed by De Waele in a and b.
3 De Waele (a: ; : ). This sequence stipulated the first half of the th century bc for KF II; the
th–th centuries for KF III; the th century for KF IV; the mid th century for KF V; and the th century for KF VI.
4 Commenting on an overall impression of the artistic quality of Kul Fara" ún (I) Layard (: ) indicated: the

design is bold and the execution good. For a complete bibliography see De Waele (a: –).
5 For the interlude – see historiographic record in Vanden Berghe ().
6 De Waele (a, b, , a, b, , , and ).
7 The doctoral dissertation by De Waele (a) was a worthy attempt at addressing some of these shortcomings.

Unfortunately, this work was never published having little or no repercussions in the field of ancient Near Eastern art.
The Ph.D. manuscript reveals a number of insights not included in related published work. It also reveals important
limitations such as the absence of supporting line-drawings, quality photography, and art historical analysis.
8 It ought to be said that this state of affairs was pre-determined by challenges of practical nature. Earlier scholars

traveling to a secluded region of the Zagros Mountains faced the task of documenting reliefs positioned high up above
ground or carved inside small crevices seldom exposed to natural light. These challenges vary depending on location
but is particularly observable in cliff carved reliefs (versus boulder carved reliefs) such as in the uppermost section
of KF I, and crevice sections of KF IV. In fact, some details of KF IV are only discerned when sunlight sheds on the
narrow crevices. As noted by De Waele (: ): “Il est impossible par example de photographier en detail Kūl-e
Farah III et IV”.
9 Judging by line-drawings made during the early th century, photographic records taken in the s’, and
 elamite sculptural reliefs from the highlands 

This present study is limited to Kūl-e Farah number IV, a relief that has never been studied
in its entirety or discussed in detail, and has played little or no part in the corpus of Elamite
art and culture, not to mention in the broad cultural heritage of the ancient Near East. As I will
attempt to argue here, KF IV provides a distinctive range of artistic characteristics which lead to
a redefinition of the sculptural arts of the Elamite highlands and, by the same token, introduces
new parameters of understanding for the society, political and religious history of Elam during
the th–th centuries bc.

. Kūl-e Farah IV

Three scholars were drawn to examine the peculiar characteristics of KF IV, namely: Layard
(), Jéquier (), and De Waele (a: –; ; ). I will build upon their com-
bined contributions to demarcate documentary criteria through a detailed systematic analysis
of the relief and to discuss related artistic and technical characteristics [Fig. ]. Before doing so,
a note of clarification on the accompanying photographs and line-drawings is necessary. KF IV
has never been photographed or drawn in its entirety. The published photographic record is lim-
ited to general overviews,10 choice photographs of panel A,11 and miscellaneous photographs.12
The earliest extant drawing-sketch was made by de Morgan after a heliogravure of panel A
(Jécquier : Fig. , a). To my knowledge, no other drawings or sketches of KF IV were
made until those provided by De Waele in  (figs. , ).
When considering the new photographs, accompanying line-drawings, and the final com-
posite sketch presented here two features of technical character ought to be kept in mind: ()
The photographs were taken from surface level implying that a certain degree of distortion
increasing proportionally with height is embedded in the documentation. I attempted to cor-
rect this distortion by digitally manipulating the angles of the photographs; clearly this solution
is not ideal and therefore all line-drawings remain not-to-scale renderings. () The function and
value of the line-drawings is of practical character, namely: they seek to offer an overall view of
the composition and a visual framework into which to plot a course of analysis.
Kūl-e Farah IV is located in the left bank of what is presently a seasonal creek. The relief
was carved along the vertical surface of the rock cliff extending over a space of c. . m long
by m high (De Waele : ). As I will clarify below, KF IV was conceived as a single unit
exhibiting a communal banquet centred on the figure of a high ranking individual sitting on a
throne. The ending result must have materialized under a quite different light than the disjoint
and weathered masses of rock sculpture encountered at present.
Jéquier and De Waele attempted a systematic approach to the study of KF IV by dividing the
imagery into groups and panels (Jéquier’s tableaux). Jéquier (: ) differentiated between
two groups of seven panels (Jéquier : , fig.  parts a, b, c, a, b, c, and d). De
Waele (: , ) distinguished ten main panels but they are nowhere articulated in the
published record.13 I have adopted seven formal divisions into panels originally proposed by

digital photographs taken in , the reliefs have deteriorated at an alarming speed suggesting that a campaign of
restoration and preservation is a most urgent task at hand.
10 Hüsing (: Fig. ); Vanden Berghe (: Pl. XV); De Waele (: Pl. IVa).
11 Herzfeld in Vanden Berghe (: Pl. xvii); Vanden Berghe (: Pls. xvi, xix); De Waele (a: Fig. ).
12 Musicians: De Waele (: Pls. Va, Vb); Miscellaneous: De Waele (: ).
13 De Waele (: ) also suggested that KF IV is intimately linked to a fire altar located in the vicinity of the

relief, next to Panel DIII: –. The association of this “fire altar” with the relief remains highly hypothetical (its
style is does not seem to have parallels on the altars exhibited in KF I, V, and SS II).
 javier álvarez-mon

G. Jéquier but with slight variations in nomenclature [Fig. ]. Accordingly, these partitions take
as reference the scenery represented in panel A and progress to its right side (-B and -C) and to
its left side (B, C, D, and E) [see Fig. ]. These divisions are of practical nature. They are meant to
individualize by number all visible features—so that, for instance, figure AI: a corresponds
to panel A, register I, individual figure a or, in other words, the enthroned individual or
king—and do not characterize the unity of the composition which is articulated by registers
crossing over panels.

.. Panel A (Main Scene)

Carved along the surface of a large protruding area of the cliff. It comprises five registers. The
register on top (AI) has a trapezoidal shape; the four registers below (AII–V) are arranged
horizontally. From the viewer’s vantage viewpoint all five registers continue around the corner
of the boulder into panel -B [see Fig. ].
Register AI (figures –). This register exhibits a banquet ceremony depicting a high-status
individual sitting on a chair (hereafter a king in a throne) accompanied by servants and elite
entourage arranged in rows [Fig. a–c].
– [AI: a] King sitting on a throne; his hair appears to be gathered into a braid; the beard
is short; he wears a sleeved long garment with a hem at the lower end; the waist is very
narrow and includes a belt. The extended left arm reposes on the knee; the right arm is bent
upwards with the hand positioned at mouth level.14 This gesture is identified throughout
this study the “right-hand holding food” (abbreviated RHF; for explanation see panel CI:
 below; see individual CI: ).
– [AI: b] Armless throne with straight legs characterized by a long backrest with a
broad terminal head bending backwards. Under the throne there is an object of difficult
identification: it has a large head, a narrow segmented body, and a pedestal made of
two small legs. This feature could be identified as an ornamented part of the throne;
alternatively it could also be a perfume holder, an incense or a lamp stand.
– [AI: ] Low square table-stand situated behind the monarch. It is holding three elongated
pear-shaped vessels (for a description see below: AIII: ) [Fig. d–e].
– [AI: ] Individual standing next to the table wearing a knee-long garment; one hand seems
to be held at chest level. The remains of a short beard are still visible.
– [AI: ] Horizontal line connects the shoulders of individuals AI:  with AI: . A few inches
above stands a small individual wearing a short skirt and with withboth hands raised, one
closer to the mouth. The interpretation of this individual is open to speculation.15
– [AI: ] Individual wearing an ankle-long garment; the right hand is raised at the level of
the mouth (most likely making the RHF gesture). The waist is narrow and the lower back
modelled.
– [AI: ] Individual similar to AI: .
– [AI: ] Badly preserved individual. Possibly wearing short (?) garment.

14 De Waele (a: ) suggests the right hand of the king is holding a vessel.
15 It should be noted that reliefs KF III and KF VI centre on a ceremony of a king or a divinity being carried atop a
platform. I do not think we have a similar feature here. Difference of hand gestures between KF IV (RHF) and KF III
and KF VI (hands held together at waist level for KFIII, and hands held upright with fingers pointing for KF VI)
suggest that at KF IV we have the representation of another participant to the communal meal. Hence, horizontal
lines at KF IV appear to be used to mark the separation between registers and spatial plans (see also figures AI: , 
and figure Panel -B: ).
 elamite sculptural reliefs from the highlands 

– [AI: ] Square table in front of the king. The horizontal top surface holds articles of
difficult identification. The table is characterized by a broad middle rail with a convex
underside. The terminals of the legs recall the shape of animal hoofs or paws.
– [AI: ] Individual oriented to the left wearing a short knee-long skirt. He is slightly
perched over the table with arms and hands reaching forward.
– [AI: ] Individual oriented to the left wearing a knee-long skirt; he has both hands
holding an object of difficult identification.
– [AI: ] Individual kneeling oriented to the left, both hands are stretched forward handling
an indefinite bulky mass (perhaps an animal carcase?).
– [AI: , ] Pair of individuals oriented to the left side standing over a horizontal line. They
wear short kilts and make the RHF gesture. The individual in front has the left arm and
hand extended at waist level. The second individual could be holding a staff.
– [AI: ] Individual wearing an ankle-long garment belonging to the row depicted in panel
-B: register I (see below).
Register AII (figures –). Register situated below AI and occupied by two rows of individuals
oriented to the right and left of a vacant space [see Fig. ]. The register continues around the
boulder with the inclusion of two more individuals (-BII: , ).
– [AII: ] Individual with broad shoulders and narrow waist wearing a knee-long kilt. He
carries a large quiver with arrowheads, holds a bow with the left hand and a weapon of
imprecise shape with the right hand. I will refer to this individual as a weapon bearer and
chief archer.
– [AII: , ] Two individuals wearing short kilts and holding a bow with the left hand; the
right hand is raised at mouth level probably making the RHF gesture.
– [AII: –] In between these two bow-holders stand three individuals making the RHF
gesture and with the left arm held stretched in a  degree angle.16
– [AII: –] Five individuals oriented to the left wearing short knee-length garments and
espousing the RHF gestures. The left arm is positioned in about  degree angle.
Register AIII (figures –). Horizontal register below register AII. It includes two rows of
individuals oriented to the left and to the right of a space occupied by two tall vases (). The
register continues around the boulder (-BIII: –).
– [AIII: , ] Two individuals wearing long garments with a fringe at the bottom end;
right hand making the RHF gesture; open left hand open with thumb upwards.
– [AIII: –] Three individuals wearing short kilts making the RHF gesture and left hand
open upwards; the front leg of number  and the kilt of number  preserve some
evidence of plaster (see §.. Technical Characteristics).
– [AIII: –] Row of four individuals oriented to the left wearing short kilts and making
the RHF gesture; the left hand open upwards.
– [AIII: ] Pair of tall vessels with an elongated pear-shaped body; the mouth is flat and
wide; the neck is narrow and broadens at shoulder level; the foot is narrow [Fig. f]. The
bottom end is not clearly discerned. Similar vessels may be present on the table stand
located behind the king (AI: ).
Register AIV (figures –). This horizontal register includes four individuals oriented to the
right side (–) and five individuals to the left side (–) of an empty space. The register

16 A.H. Layard (: ) thought these individuals had their hands bound.
 javier álvarez-mon

continues on the other side of the boulder (-BIV: –). They wear short kilts, and are making
the RHF gesture. The group on the left has their left free arm held at a near  degree angle or
slightly higher; the group on the right has the left arm positioned in a lower position.
Register AV (figures –). Horizontal register with row of six individuals oriented to the
right wearing a short kilt and making the RHF gesture with the left hand open upwards. As
many as three additional individuals may have been missing, as it is estimated that this corner
section has broken off.17 The register continues around the boulder (-B: ).

.. Panel -B [-B]

Panel facing the creek bed and situated to the right side of Panel A [Fig. ]. The surface has
significantly eroded and, judging by the sketch-drawing made by J. de Morgan at the beginning
of the th century, a section atop the panel (-B: ) is missing. The panel exhibits five registers
continuing those exhibited on Panel A. All sixteen individuals (or seventeen counting AI: )
stand oriented in rows towards the left. They share identical characteristic gestures: the RHF
gesture and the left arm stretched at waist level with the palm of the hand opened upwards. The
hair is collected into a short braid. Visible differences regard to the type of garment.
– [-BI: –] Six individuals (counting AI: ) wearing ankle long garments.
– [-BII: –] A break in the rock cuts through the waist and skirt of the two individuals
represented.
– [-BIII: –] Row of six individuals sharing gestures, garment, and hairstyle.18
– [-BIV: –] Three individuals with short kilt.
– [-BV: ] Remains of a single individual.

.. Panel -C [-C]

This panel is poorly preserved. It includes three registers continuing those depicted in panels A
and -B [Fig. ]. There are remains of one register occupied by a row of seven individuals (IV: –
); two additional registers (III and V) can be said to have been present, as indicated by random
remains of plaster (III and V).
– [-CIV: –] Two rows of individuals separated by a small space sharing identical charac-
teristics with registers AIV (–) and -BIV (–).

.. Panel B [Composed of BI and BII]

Panels located to the left side of panel A and separated by a large crevice in the rock [Fig. ].
Inside this gap and facing panel A, there are two registers with three rows of individuals each
(BI: –). A third register below seems to have been chiselled away.19 Also the surface of panel
BI appears to have been defaced with a sharp tool. All individuals are oriented to the left, make
the RHF gesture, and have their left hand extended.
– [BI: –] Row of three individuals wearing a short kilt, short sleeved upper garment, and
have a narrow waist marked by a belt. Their legs are muscular. Individual  has preserved
evidence of facial hair, a perpendicular garment fringe, and the thumb of the right hand.

17 Note: The sketch made by J. de Morgan fails to accurately represent the long diagonal break on the surface of
the rock (for instance the break goes through individual IV:  and not IV: ).
18 The smaller scale of individual  can be explained by the lack of actual space left to represent him.
19 De Waele (a: , n. ) estimates this block may have be about .m long and believes it may have had a

third register with enough space to include four additional individuals.

 elamite sculptural reliefs from the highlands 

– [BI: –] Row of three similar individuals to the left of BI: –. Number  has
preserved a narrow belt, the perpendicular hems of the garment, the left arm, the outline
of the shirt, the contour of the strong muscular legs, and some facial features such as the
ear and the tip of the beard.
– [BII: –] Three superimposed horizontal registers occupied by rows of fifteen identical
individuals oriented to the right sharing gestures, garment, and hairstyles. They make the
RHF gesture; the left arm is extended with the hand open upwards. Individual BII:  has
retained detail indicating braided hairstyle, the eye, the ear, a narrow belt, and the band
decorating the hem of the short sleeved shirt.

.. Panel C [CI, CII]

The vertical surface of the rock to the left side of panel BII can be divided in two areas: the lower
section below a rock ledge exhibits a group of seven individuals at first view randomly organized
(group panel CI) [Fig. ]; the section above the ledge includes large groups of individuals
organized in rows (group panel CII) [Fig. ].
– Panel CI (figures –) exhibits the best preserved sections of Kūl-e Farah IV. The panel
depicts a group of seven individual whose distribution sharply contrasts with the mar-
tial sequence of those observed in the previous registers.20 Two groups may be defined:
a trio on top (CI: –) and a quartet below (CI: –). Of these, individual CI:  is most
exceptional.
– [Panel CI: ] Individual oriented to the right [Fig. ]. The uniqueness of this figure was
noted by L. Vanden Berghe (:  n. ) who indicated his physical features epitomize
the highland Elamite identity type already discerned in the celebrated relief of Kūrangūn:
“on distingue nettement chez ces personages, dont nous donnons ici une photo de detail
(Pl. XIX), les caractéristiques du type élamite: une tête plutôt arrondie, un œil très grande,
presque en forme de huit, un nez busqué aux narines larges et développées, des lèvres fort
marquées, des cheveux réunis en longues nattes tressées à boucle terminale, retombant sur le
dos, et une barbe très courte.” It is important to further underline some of these physical
features.

... Facial Hair and Hairstyle

The surface of the head preserves the remains of a white coloured plaster-like material skilfully
incised with series of parallel stripes collected into U-terminals to depict a short neatly trimmed
beard and a hair style characterized by a long braid reaching below the shoulder. The braid can
be divided in an upper and middle section segmented by parallel, slanted, striated outlines
beginning around the crown and by a terminal section marked by long perpendicular, parallel,
lines.

... Face
The face is illustrated in profile. It is characterized by an almond-shaped eye with a marked
eyeball, a large eyebrow arching from the level of the temple to the bridge of the nose. The nose
is prominent, straight, and tip slightly rounded; nostrils are visible, as well as the lips. The shape
of the ear is defined by incisions similar to the U-terminal sections of the hair but forming a
continuous band.

20 There is no apparent reason to justify this arrangement other than it may have something to do with a possible

identity as archers.
 javier álvarez-mon

... Body and Garments

The body also conserves a layer of plaster modelled over the sculpted surface of the rock
and extending to the surrounding area of the body (see § .. Technical Characteristics). This
individual has broad shoulders, a narrow waist, fleshy buttocks, muscular thighs and narrow
knees. He is wearing a short shirt and kilt separated by a wide waist band. The left shoulder
and arm is shown to the right side of the right arm. The short sleeves of the shirt have bands
made of incised lines: one at the edge, the other further up; modelled biceps are discernable.
The left arm extends downwards with the left hand closed into a fist (holding a bow). The right
arm is flexed in a right angle; all five fingers of the hand are clearly represented. The remains of
nails are also visible as well as some folds on the fingers. The thumb and index fingers hold a
triangular piece of what can best be interpreted as a meatball; here the more neutral right-hand
holding food gesture (RHF) has been adopted.
– [Panel CI: –, –] Individuals similar to CI: . The left hand appears to be closed into a
fist possibly clasping a bow similar to those depicted in AII: , .
– [Panel CI: ] Individual similar to the previous ones except than his left arm is held in
about a  degrees angle.
Panel CII (–). This section is located high above ground; hence the photographic record
available to me, and presented here, is sharply distorted [Fig. ]. Four main rows of individuals
can be distinguished according to garment style, position, and type of activity: the individuals
in group CIIa wear a long robe (–); those of group CIIb wear a short kilt (–); group
CIIc is made by a small orchestra of six musicians faced by a “musical director” (–); and a
group of seven individuals below wear a short kilt (CIId: –). Except for individual CIIc: 
the general orientation is towards the right.
– [CIIa: –] Panel composed of three rows of individuals wearing long robes arranged in
a diagonal divided into a group of eight (–), a pair (–), and a single individual ().
– [CIIa: –] Individuals located directly over panel BII. Except for their long garment,
they share similar gestures and braided hairstyle.
– [CIIb: –] Isolated trio wearing short kilts.21
– [CIIc: –] Harp orchestra composed of three musicians each: the group at the back is
arranged in a horizontal line; the one at the front in a diagonal line. An individual labelled
here the “musical director” faces them with both arms held slightly above the waist; the
right hand is visibly open.22 Two types of harps are represented: two small horizontal
open harps lacking a forepillar whose frame is composed of a horizontal resonance-box
and a vertical neck, and four large vertical open harps whose frame is composed of a
vertical resonance-box and a horizontal neck. The resonance-box lays positioned in the
background to the left side of the face. Some strings are visible as well as the index and
little fingers of the right hand.
– [CIId: –] Three groups of seven individuals wearing short kilts arranged in three rows:
a pair (–), a trio (–), and another pair (–). They are oriented to the right.
They espouse the RHF gesture, the left arm is stretched downwards.

.. Panel D [D]

Area situated in the lower left periphery of the main relief; it comprises three panels: DI (–),
DII (–), and DIII (–). All individuals are oriented towards the right [Fig. ].

21 According to De Waele (a: ) these three individuals may wear their hair on a bun.
22 This individual is the only one represented in the entire relief to have his back turned against the central scene
in panel A. He is also not participating in the consumption of food.
 elamite sculptural reliefs from the highlands 

– [DI: –] Long rectangular register descending in a diagonal accommodating two rows of
nine individuals: five on a horizontal surface (–) and four on a diagonal surface (–).
They make the RHF gesture and have the left arm held high.
– [DII: –] Small panel with a pair of participants roughly preserved. The garment is
knee long. The gestures are similar to those represented in panel DI.
– [DIII: –] L shaped rectangular panel situated furthest away from the main composi-
tion including a pair (DIII: –, not shown in the photographs) and trio (DIII: –).
All five wear long garments, the right hand is placed in the direction of the mouth and the
left hand outstretched closely behind. They appear to have a short hairstyle with absence
of the characteristic long braided hair.23

. Notes on Artistic and Technical Characteristics of Kūl-e Farah IV

The relief of Kūl-e Farah IV exhibits a long-backed throne, a low table-stand holding three
vessels, a square table, two long-necked vessels, and counts with the presence of a minimum of
one hundred and forty one individuals.24 Social hierarchy seems to be determined by registers,
placement, and type of garment. Amongst the individuals present we can distinguish specific
individuals and groups: a king sitting on a throne [AI: ], various high status individuals
wearing long-garments [AI: , AI: ?; AI: ; AIII: , AIII: ; -BI –; CIIa: –; DIII: –],
a weapons-bearer/chief archer [AII: ], two archers [AII: , ], harp players [CIIc: –],
possibly, but not certainly, a group of seven archers [CI], attendants to the king [AI: ?, , ,
], and more than a hundred extra participants wearing short kilts. This last group is by far the
most numerous and is personified by an individual wearing the hair on a braid and making the
“right-hand with food” gesture [CI: ]. Visible stylistic variations are minimal and pertain to
the position of the arms and hands. Most participants stand oriented in the direction of panel A.

.. Technical Characteristics: Dimensions and Manufacture

Estimated Dimensions of Kūl-e Farah IV (in meters)25

Panel DIII DII DI CII CI BII BI A –B –C

Length 1.2 0.4 1.9 3.5 1.0 0.9 0.85 2.3 1.2 1.8
Height c.1.0 0.55 0.9 c.2.6 1.2 1.35 0.92 2.8 2.65 0.65
Height 0.48 0.5 0.47 note26 0.4 0.42 0.44 note27 0.48 0.51
of 0.53 0.52 0.51 0.45 0.46 0.55 0.55
People 0.42

23 De Waele (: ) suggested this group of five individual depicted priests distinguished by buns on their
heads and long garments. He further related their presence in this particular location to a rock with two circular
cavities on top which he interpreted as a fire altar (Waele : ).
24 As mentioned, additional participants to the communal ritual may have been depicted in registers AIII, -

BIII, AV, and -BV may have extended further to the right (possibly in panel -C where there is attestation of seven
individuals). There are also broken sections over panel BII.
25 All dimensions adapted after E. De Waele (a: –). The correspondences between De Waele’s divisions

and the panel divisions espoused here are as follows: DIII (Waele ), DII (Waele ), DI (Waele ), CII (Waele ), CI
(Waele ), BII (Waele ), BI (Waele ), A (Waele ), -B (Waele ), -C (Waele ).
26 CIIa: – (./.); CIIa: – (.); CIIb: – (./.); CIIc: –, Orchestra, vertical harp players

(./.); “musical director” and horizontal harp players (./.); CIId: – (./.).
27 Panel A register I (height . m); height of seating king: .; height of standing individuals: ./.; height

of kneeling individual : .; height of the table in front of the king: .; height of table behind the king: .;
 javier álvarez-mon

Technically speaking, within the categorical boundaries of art making, KF IV cannot be

considered just rock-sculpture. L. Vanden Berghe (: –) was first to notice the
exceptional technical characteristics employed in the manufacture of this relief. While in need
of support by scientific analysis, my own observations corroborate these remarks. While risking
to be premature, I will suggest the manufacture of this and related reliefs may have involved the
following general steps: () a flat panel was cut over the rock;28 () the imagery was carved
in low-relief; () the surface was plastered and modeled with a solid bitumen-based, plaster-
like material;29 () detail was added by incision with the help of a sharp object;30 and ()
the reliefs were painted.31 The possibility that the manufacture of KF IV involved sculpting,
plastering, modeling, engraving, and painting, presents a multifaceted understanding of the
field of sculptural crafts; if confirmed, these attributes would prove to be distinctive attributes
of Elamite highland art.32

.. Elements of Style and Composition

The preceding analysis suggests a highly experienced artistic enterprise characterized by a dis-
tinctive stylistic grammar and iconography. () At the individual level, the survival of partial
motifs together with the remarkable details exhibited by participant CI:  provide the oppor-
tunity to further articulate notions of style. () At the general level, the wide measurements of
the composition (. ×m) and its distribution, arrangement and structure hint at a com-
plex planning and an artistic enterprise of genuine originality without parallel in the arts of the
ancient Near East.

... Individual Style

Key criteria of body representation entails depicting the lower half of the body and the head
in profile with chest and shoulders showed frontally. This arrangement can be said to present
trivial challenges when no parts of the body interact with each other but test the ability of the
artist when the arms interact with chest and shoulders. As we have seen, the iconography of
KF IV stipulated that the largest majority of individuals partaking on the ritual meal were

height of individuals , , and : ./.; height of individuals in panels AII, AIII, AIV, and AV: ./.; height
of the two vessels in register AIII: ./..
28 Evidence of this stage is apparent in those parts of the relief divided into registers (for instance the perpendicular

line atop AI). There are empty cut-out panels on a boulder next to KF III and in the vicinity of KF V.
29 For instance, this layer is observed over the surface of figure BI: , some of the individuals along BII: –,

and DI: , and, most particularly, over the body and head of individual CI:  Additional supporting evidence of the
existence of this plaster occurs in less damaged areas of reliefs from Kūl-e Farah, Shekaft-e Salman, Kūrangūn. At
KF III, in the southern face, the fourth individual in the th line from the left reveals the incised details of his hair,
beard, and tunic (Vanden Berghe : , plate XIX; : ) [Fig. b]. At Shekaft-e Salmān the better preserved
relief of SS II (depicting the royal family) includes substantial evidence of this plaster over the long garment worn by
the queen [Fig. a, a]. At Kūrangūn, this layer can be observed in the eye, the belt and the bottom of the long
dress of individual  (Vanden Berghe : ) [Fig. c].
30 Evidence can be clearly observed in individual CI:  (note, for instance, the hair, beard, eyes and ear and segment

of the back).
31 This last stage is not clearly recognized at KF IV since most of the plaster did not survive; although there is some

evidence of paint left over the surface of CI:  and perhaps BI: . Evidence suggesting that KF IV and other reliefs
from Īzeh/Mālamı̄r were painted is provided by Shekaft-e Salmān II. This relief depicts the royal family (the king, a
child, and the queen) in worshipping gesture directed to a cave and related stream. The headdress and garment of
queen preserves remains of a white layer of plaster and red paint.
32 The outcome of this research could be greatly enhanced by ongoing technological improvements on pigmenta-

tion analysis, including recent break-troughs on the C analysis of pigmentation remains (Rowe & Steelman ;
Steelman & al. ).
 elamite sculptural reliefs from the highlands 

depicted with the right hand directly in front of the mouth holding a piece of foodstuff while the
“free” left arm extended forward at various angles. In accord to their orientation two inventive
representational challenges and, correspondingly, solutions had to be applied: (a) with body
oriented to the right, the right arm was required to be depicted in front of the chest; in this
case, the left shoulder and arm are shown as appendices to it (see for instance CI: ; Fig. );
(b) when the body is oriented to the left, the left arm crosses over the chest and the right arm
remains in place in front of the body (see for instance AIII: –; Fig. ).33
The depiction of physical characteristics was enacted with extraordinary precision at the
service of a specific ideal of corporeal representation. This is demonstrated by a realistic attempt
at depicting individual features most particularly documented by individual CI: , namely:
the treatment of hair style (long hair pulled back and collected into a braid), the treatment
of hands (re. presence of nails and finger folds); and precise physical features (large eyebrow,
a substantial nose, proportionate ear, full lips, beard with well differentiate hair locks, strong
neck, arched backline, narrow waistline, muscular legs and arms but not exaggerated). These
elements of style were conceived as part of an ambitious artistic program encompassing at least
 individuals. This shared commonness implies the existence of a master plan and a canon
of artistic proportions expressed through the hand of expertly trained artisans. As evidenced
by the facial and hair details observed in CI: , one is tempted to raise the possibility that the
original artwork was intended to be a portrayal of individuals defined by different treatments of
facial features; if such was the case, the words communal portrait may indeed accurately describe
Kūl-e Farah IV.

... Structure of the Composition

The apparent complexity of Kūl-e Farah IV can be presently dissected into two well thought-out
compositional structures integrating: (a) horizontal registers occupied by rows of participants
to the communal banquet; and (b) a large open area including panels CI and CII [see Fig. ]. (a)
The great majority of the partakers to the banquet are found inside registers martially organized
on rows oriented on both sides of an imaginary focal axis point. This axis is located in panel A
and runs down in a vertical band from the upper register (I), separating registers II, III, and IV
in two. This vertical axis serves as the back bone of the composition. (b) A second key structural
reference is indicated by the arrangements of individuals in panels CI and CII, outside the
boundaries imposed by registers. These are the seven bowmen freely dispersed in panel CI, the
harp musical ensemble with their leader (CII: –), rows of high status individuals wearing
long robes (CII: ), and pairs and trios of individual participants (CII: –; CII: –).
Group CII is arranged in a diagonal oriented in the direction of main panel A. As mentioned,
the freer distribution of the “archers” (panel CI) is intriguing and perhaps suggests an attempt
at representing a mountainous background environment.
The focus of attention is patently established by register AI: the king is surrounded by
attendants, and possibly community leaders. Framed on both sides at the same hierarchic level
panels -BI and CII include two groups of individuals wearing long garments. They may be
identified as religious leaders, but this cannot be confirmed. The register below AI includes
the presence of a weapons bearer/chief archer (AII: ) and two individuals holding bows (AII:

33 There is no apparent rationale in my mind to include the left shoulder and arm to the side of the body when this
is oriented to the right. The “common” alternative, to have a body fully represented in profile, can be observed in the
treatment of the braided individual worshippers from KF III and Kūrangūn II. Yet, this approach was not espoused
at KF IV perhaps because there was a theoretical and ideological choice at maximizing the wholeness of the body
(which gives precedence to a frontal chest and strong shoulders and arms).
 javier álvarez-mon

, AII: ).34 To the left, organized along descending perpendicular lines, stand the orchestra
composed of two groups of harp musicians (CII: –) whose presence further underlies the
significance of the ceremonial act.

. Chronological Considerations

The manufacturing date of KF IV has been stipulated in art historical grounds, with both
identification and interpretation of the evidence contriving chronological arguments. In the
ensuing analysis I will review key art historical arguments so far offered and present additional
diagnosis arguing in favor of dating KF IV to the th–th centuries bc.
Two main reasons were originally given for a presumed Neo-Elamite dated for KF IV:
the presence of a late Neo-Elamite inscription engraved at Kūl-e Farah I (and closely related
inscriptions from Shekaft-e Salmān) and compositional iconographic parallels exhibited at Kūl-
e Farah, Kūrangūn, and monumental Achaemenid Persian sculpture from Persepolis (Jéquier
: ; Vanden Berghe : , : –; Calmeyer : ; De Waele : ).
The trend for an earlier date began when the inscriptions from Shekaft-e Salmān were found
to be secondary additions made by Hanni son of Tahhi (the author of KF I).35 Thereafter,
art historical comparisons stressed analogies between the reliefs from Shekaft-e Salmān and
imagery developed at the time of the Shutrukid Dynasty (th century bc). This led to the
view that the House of Šutruk-Nahhunte had been responsible for some of the reliefs carved
at Kūl-e Farah.36 The latest argument was introduced by Amiet (: –) who argued that
KF II, III, IV, and VI were manufactured around the end of the second millennium bc.37 The
significance of these dates does extend beyond KF IV, as it is generally recognized that the rows
of participants depicted at KF III and the rows of worshipers added to Kūrangūn (hereafter
Kūrangūn II) share the presence of similar braided worshippers and must, therefore, be closely
related if not contemporary.38
Two art historical arguments have been made in support of dating KF IV to dates ranging
from the th to the th centuries bc: the presence of stylistic similarities with tall Elamite
ritual goblets represented in the archaeological records of Susiana and Tal-i Malyan (§ .); and
iconographic similarities with banquet scenes exhibited in seals from Tchoga Zanbil (§ .).

.. The Elamite Goblet

Panel figure AI:  exhibits a table-stand supporting three vessels whose shape is not clearly made
distinct by the photographs. Hence, the following observations are made with reservation: ()
contrary to the interpretation suggested by the sketch made by de Morgan [Fig. a], the vessels
seem to be hanging from the table and not resting on their foot; () the neck of the vessel is
elongated and narrow; () the shape of the body is pear-like; () the outline of the foot is not
clearly discerned. These elements suggest similarities with the pair of vessels shown in register
AIII: ; the shape of the foot, however, remains elusive.

34 Besides the attendants to the king, the musicians and their “musical director”, AII:  seems to be the only
participant who is not holding a piece of food with two fingers.
35 De Waele (a: ; : –). For concordance amongst the inscriptions see De Waele (b). For a

review of the inscriptions see Stolper ().

36 Carter (: ), Seidl (: ) and Calmeyer (: ); Amiet (; : –).
37 This reclassification does not include KF V which is dated together with KF I to the th and th centuries bc

(this is intriguing to me because KF II and V are closely associated).

38 To clarify my position, there are stylistic differences such as the depiction of the body’s profile and the length

of the braid suggesting they are not contemporary.

 elamite sculptural reliefs from the highlands 

Carter fittingly suggested the shape of the vessels depicted at KF IV recalled the distinctive
ceremonial Elamite goblet whose attestation in the archaeological record concentrates in the
Middle Elamite period and is significantly reduced beyond c. bc. The goblet is characterized
by a long cylindrical neck broadening to the mouth, smooth lips, an elongated pear-like
shoulder that narrows at the base, and a round broadening foot [Fig. g,h,i]. It is well
attested in Middle-Elamite levels of Tchoga Zanbil, Tal-i Malyan, and Susa.39 At first view,
these stylistic features suggest a close match for the vessels depicted at KF IV. Nonetheless,
three concerns must be raised. () The popular Elamite goblet did not disappear completely
from the archaeological record after  bc;40 () the shape of the foot in the vessels from
KF IV remains elusive. Within an Elamite context this difference bears significant chronological
weight as amphorae style vessels follow in time the Elamite goblet [Fig. g];41 finally () little
is known regarding the archaeological record (and ceramic sequences) of Īzeh/Mālamı̄r (or for
the same token, the associated urban and political entities of Huhnur and Ayapir, wherever the
latter is to be located). Consequently, it is perhaps hazardous to assume that KF IV exhibits
ceramic sequences matching those from Susa, Anšan or Tchoga Zanbil. These concerns do
not outright reject a correspondence between the mainly Middle-Elamite ceremonial goblet
and Kūl-e Farah IV but raise uncertainties regarding the actual shape of the vessels, their
chronological parameters, and their archaeological distribution.

.. Elamite “Banquet” Seals

Carter pointed to banquet scenes depicted in th century bc seals from Tchoga Zanbil to
suggest possible parallels with the banquet scene depicted at KF IV.42 The well represented
corpus of “banquet seals” from Tchoga Zanbil (classified by Porada as Group VII; see Fig. a)
forms an iconographic and stylistic unity characterized by: () an individual (identified as a
king) for the most part sitting on a () square taboret without backing holding (and most likely
in the act of drinking from) a () small vessel. Many samples include the presence in front of
the king of a () small table with vessels, a fish or a small caprid (Porada : –). These
four elements (the king, the tabouret, the vessels, and the table) will be evaluated against their
counterparts from KF IV.
– () In the seals the sitting king wears a fringed garment and has the right hand extended
at about shoulder level holding (or drinking from) a vessel. He may be wearing a band or
a circular cap with hair reaching shoulder level. The king from KF IV is of a larger scale
than the other participants, the left hand rests on the lap at the level of the knee. The right
hand may be making the RHF gesture; his hair is held together in a braid.

39 The samples from Susa (Ville Royale II, level ) and Malyan measure more than cm in height (Miroschedji

, figs. , ; Carter , fig. ). The measures of the samples found in room  of the “Palais-hypogée” at Tchoga
Zanbil are unclear (the scale accompanying fig.  in Ghirshman : , pl. :  and pl. :, are mistaken).
40 At Susa, Ville Royale II, levels  and  of Susa Ville Royale II (c. – bc) include a small percentage of the

Elamite goblet ( ) (Miroschedji : , fig. : ). It should be also noted that Level I of the so-called Village
Perse-Achéménide includes samples of a related form (the th–th centuries bc date for Level I proposed by Ghirsh-
man [: ] was revised by Stronach [: ] to ca.  to bc). Incidentally, the Elamite goblets may have
been transported on a tray shaped like a high walled hexagon partitioned into seven circular chambers [see Fig. i].
41 The most significant group of wares characterizing the Neo-Elamite II period (c. – bc) are elongated

amphorae style jars without handles measuring between  and cm in height; Little jars (measuring between 
and  cm high; de Miroschedji : . fig. : –); medium-size jars (between  and cm high; Miroschedji
: , fig. : –); and large jars (between  and cm high, fig. : –).
42 With reference to Amiet MDP  () nos. ,  and Porada MDP  (), –. The th

century bc date for the Tchoga Zanbil seals assumes these were deposited at around the time of Untaš-Napiriša
(ca. – bc); Thus F. Vallat ().
 javier álvarez-mon

– () The tabouret depicted in the seals is generally square, has no backing, and includes
an internal X-shaped frame. The throne from KF IV has a high back-rest ending on some
ornamental protuberance. The internal frame of the chair includes a low horizontal rail
and perhaps an ornamented rail or, as I have noted above, an incense stand or a lamp.
– () The tables depicted in the seals present little diversity: a simple square box with an
internal X-shaped frame and a base (Porada : fig. ). At KF IV we have two sets
of tables represented (AI:  and AI: ). The one behind the king probably included a
perforated top surface into which to slide the vessels. The square table in the front has a
broad middle rail with a convex underside. The terminals of the legs may be ornamented
in the shape of animal hoofs or paws.
– () The vessels represented in the seals are generally small globe-shaped flasks (sometimes
known as aryballos). The ones from KF IV are large, elongated and pear-shaped.
It is noticeable, that neither the “king”, nor the table, nor the tabouret, nor the vessels, nor
the general iconography (drinking from a vessel) depicted in the banquet seals from th
century bc Tchoga Zanbil have correspondence with imagery exhibited at KF IV. If similarities
do exist these occur at the ideological and ritual level: the performance of a sacred banquet
ceremony before the king that may have taken place at the royal court (Porada : ) or/and
was associated with ritual performed in the main religious installations of the Ziggurat.43
Perceptively, Porada pointed to series of Neo-Assyrian/Babylonian banquet scenes of linear
style suggesting they owed main characteristics to the influence of Elamite glyptic art (:
–, –; see Figs. c–d–e).44 The Neo-Assyrian banquet seals housed at the British
Museum have been classified into three main groups according to the presence of a pot-stand, a
table with offerings, and an incense burner. Most seals in the Assyrian pot-stand group are dated
to the th century bc but the theme may have continued throughout the th century.45 One of
the main differences between the th century bc Elamite banquet seals and their latter th–th
century bc Mesopotamian cousins is the presence of high-backed thrones, according to Collon
(: ) a feature appearing during the second-half of the th century bc. In these examples
the backrest of the throne has a curved terminal, in one clear sample seemingly depicting an
animal head, perhaps to be identified as a duck or an ostrich neck and head (Fig. c; Keel-Leu
& Teissier : seal z).
More pertinent to this discussion is the representation of a royal banquet scene depicted in
an ivory strip found in Fort Šalmaneser at Nimrud and presumably dated to the time of Ashur-
nasirpal or his son Šalmaneser III, that is the th century bc [f, f].46 The core scene
depicts the king seated on a high-backed throne holding a vessel with the right hand and with
the left arm resting on the lap. The table stand located behind the king holds three small spher-

43 It is of interest to note that all banquet seals from Tchoga Zanbil were found inside Chapels III and IV. These

two Chapels framed the central niche located facing the main staircase providing access to the superior stages
of the ziggurat. R. Ghirshman (: ) wondered wether there was an association between the banquet scene
representation exhibited on the seals and their particular presence in Chapels linked to the main access to the
ziggurat.
44 Porada has traced the origin of the banquet representation exhibited at Tchoga Zanbil to Old Elamite imagery

(: ; Collon : ). This insight can be further refined in light of the more recent publications of late nd and
st millennia Mesopotamian glyptic (Matthews , ; Collon ; See Fig. b).
45 One of the most ubiquitous presences in the Assyrian banquet scenes including a pot-stand is the presence of

an individual holding a square fan (Collon : ). The fan scene is popular in Elam during the th century bc,
but Matthews (: –) suggests a Middle-Assyrian prototype of possible Mitannian ancestry appearing first
in the mid th century bc.
46 For the seals see Keel-Leu & Teissier (: , ; Seals  and , but see also seal ; with references).

For the ivory plaque found in SE Quadrant, Room SE (British Museum ND) see Mallowan & Davies (:
, pl. V, fig. ; Hermann, Coffey & Laidlaw : ).
 elamite sculptural reliefs from the highlands 

ical aryballos with narrow neck and pointy foot; it has terminals ending in lion paws perhaps
resting on pine cones. The two tables in front of the king are of a standard Assyrian type: they
have a nearly square frame, two horizontal middle rails, a vertical rail placed under the convex
underside of the table; and terminals ending in horizontal lion paws supported by pine cones.47
To recapitulate, Kūl-e Farah IV combines elements of a banquet scene: the king sitting on
a throne, the presence of a rectangular stand supporting vessels, a square table perhaps with
legs ending in lion’s paws, and probably (under the throne) an ornamented object perhaps to
be identified with an incense burner, lamp, or candelabrum. On all five accounts (the king,
the vessels, the two tables, and the “incense burner”) KF IV shows great complexity of style
and elaborate composition that the abridged banquet scenes observed in the seals from Tchoga
Zanbil. Instead, greater similarities appear to be shared with Neo-Assyrian glyptic imagery and
most particularly with the ivory strip representing an Assyrian royal banquet, all dated to the
th and th centuries bc.

.. Braided Highlanders and Royal Elamite Imagery

A noteworthy trait of cultural identity appearing in the relief of Kūl-e Farah IV is the presence of
individuals wearing long hair collected into a medium-length braid. As mentioned, this feature
directly links KF IV with worshippers represented in the reliefs of KF III and Kūrangūn (II).
These peculiar hairstyles are not restricted to Īzeh/Mālamı̄r and Kūrangūn. In fact, there is
ample and consistent evidence in the Elamite artistic records to suggest that long braided hair
is a significant cultural accoutrement associated with the peoples residing in the Elamite Zagros
highlands.48 To illustrate this point and further examine the chronological boundaries of KF IV,
I will review four main sources of artistic evidence ranging from the ca. th century bc to the
ca. th century bc which exhibit the distinctive braided hairstyle presently associated with the
Elamite highlands.

... The Šutrukid Dynasty (–)

Appreciation of the artistic program of the Šutrukid dynasty (named after the king Šutruk-
Nahhunte, ca. –) is often obscured by the infamous actions of the House of Šutruk
in Mesopotamia.49 The resources accumulated during this period of Elamite imperial expan-
sionism produced unprecedented building activity throughout Elamite territory.50 It is under

47 Evidence of continuity and longevity of the banquet scene in Elam is exhibited on a cylinder seal from the

P. Morgan Library brought into discussion by Porada (Fig. i; : , Pl. xv, fig. ). Porada suggests the
date of this seal is clearly indicated by the pointed visor hair-style worn by the enthroned figure. This hairdo
bears close parallels with that spoused by Hanni of Ayapir, Atta-hamitti-Inšušinak, and the “unknown king” from
the Elamite relief at Naqs-e Rustam. This may be an indication of a late Neo-Elamite date for this seal unless,
as indicated by Porada, analogies are established with the hairstyle-helmet worn by the seven Elamite warriors
represented on a bronze stele dated by Amiet to the th century bc (but see comments on the dating of this stele
below).
48 The inhabitants of the Zagros may have adopted this peculiar hairstyle since the latter part of the third

millennium bc. As attested by the portrayal of the Lullubi people during the Akkadian period (see below: §.
Postscript: Braids of Glory; See also Potts ).
49 Šutruk-Nahhunte and his two sons Kutir-Nahhunte (–) and Šilhak-Inšušinak (–) continued

a foreign policy of vindication that—since the marriage of prince Pahir-iššan with the eldest daughter of the Kassite
king Kurigalzu (in bc) and a long succession of marriages between Elamite kings and Kassite princesses—
asserted the claim of the Elamite kings over the Babylonian throne. This claim incurred numerous raids on
Mesopotamian cities and eventually ended with the collapse of the Kassite dynasty in  bc and the presence
at Susa of large amounts of “trophy”.
50 From Anšan to the shores of the Persian Gulf and the Susiana plain spread the construction of new temples,

reconstruction and ornamentation of old temples, and installation of new sanctuaries (Stolper : ; Potts ).
 javier álvarez-mon

this cultural golden age that an evaluation of the active artistic program of the Šutrukids ought
to take place, including the likelihood that they were responsible for the art manifested at
Kūl-e Farah IV. To examine this possibility four key artistic references will be also summar-
ily reviewed. A fifth artistic reference of imprecise date will be discussed, namely a large bronze
plaque representing seven highland “divine” warriors.
.... Royal Couple from the Acropole Mound [Fig. e]
Twenty five glazed bricks belonging to a monumental façade located in the Susa Acropole and
dated, after the inscription, to the time of the brothers Kutir-Nahhunte (–) and Šilhak-
Inšušinak (–). Amiet (: –) was able to reconstruct parts of this relief showing
the royal Elamite couple with head and lower body in profile and frontal chest. Of the king’s
head only the long beard and pair of long tressed sideburns are visible; the chest is broad and
the waist narrow; he is depicted in a worship gesture holding hands (with the right hand over
the left), and appears to have a bracelet in the right wrist. Associated glazed bricks depict a
large shoe with folds and the curved rim of a long garment. Of the queen’s head only the face
is preserved; the garment over the left arm seems to include a seam and a clasp; the right hand
has extended fingers and crosses over the wrist of the left hand; the left hand is closed on a fist
holding a dropping object, possibly a towel; two additional bricks showing the ending corners
of a long garment have been associated with this queen.
Because of the presence of an inscription and the secure dating this relief is of pivotal
significance to establish comparisons with the royal couple (and child) depicted at Shekaft-e
Salmān I and II. All individuals represented in the four reliefs from Shekaft-e Salmān (including
reliefs III, and IV) wear a mid-chest level long braid ending in a characteristic upwards loop or
knob.51
.... Moulded Monumental Brick Façade from the Apadana Mound
Šilhak-Inšušinak was responsible for restoring a symbolic monumental sacred garden enclosure
composed of moulded bricks for the “exterior chapel” probably located in the Apadana at Susa.52
Of note, is the presence of long braided sideburns ending on large loops framing the face of the
divine bull-man.53
.... The Jasper Bead of Princess Bar-Uli [Fig. b–b]
Small blue chalcedony bead engraved with a royal image representing the enthroned Elamite
king Šilhak-Inšušinak (c. –bc) making an offering to his daughter Bar-Uli.54 The

51 This long style of braid is not shared by the individuals depicted at Kūl-e Farah IV. Neither, I think, there is

clear evidence suggesting the participants at Kūl-e Farah III share the same type of braid (but this relief has much
suffered and only a detailed analysis of the photographic record may clarify this point).
52 R. de Mecquenem (: ; : ) reported to have found in the vicinity of the eastern necropolis and

at a depth of  to  meters large quantities of modeled terracotta bricks presenting the remains of monumental
wall decoration including a sequence of at least twenty bull-man belonging to a temple of Inšušinak built by Kutir-
Nahhunte (–) and Šilhak-Inšušinak (–). Their presence and significance in the Apadana mound
remains controversial (Steve & Gasche :  n. ; Malbran-Labat : , plan of Susa; Caubet ; Álvarez-
Mon ).
53 Compare the representation of these divine bull-men with a likely close reference from the facade of the Inanna

Temple at Uruk built around bc by the Kassite ruler Karaindaš. Since Šilhak-Inšušinak says to be the restorer
(and not the original builder) of this façade we have to reckon with the possibility that the Elamite temple was built
much earlier.
54 Measurements: .   cm; housed at the British Museum ; purchased in  (Calmeyer : ).
× ×
This precious bead includes a pierced hole along its longest axis for suspension. The translation by Sollberger (:
) reads: I, Šilhak-Inšušinak, enlarger of the kingdom, this jasper (of/in/to) Puralsiš I took. My completed work I
placed there, and to Bar-Uli, my beloved daughter, I gave Bar-Uli was a child of the union between Šilhak-Inšušinak
and Nahhunte-utu. For the special role played by incest between a king and his sister (and, exceptionally, a king and
his daughter) see Steve & al. (: –).
 elamite sculptural reliefs from the highlands 

carefully engraved imagery sharply contrasts with the banquet scenes exhibited at Tchoga
Zanbil and public royal imagery of Šilhak-Inšušinak. The king has a short beard and a typical
Elamite visor hair-style; he wears a long a sort-sleeved garment ending in a fringe and a pair
of bracelets on each wrist; the right arm and hand rest over his lap with extended fingers; the
left hand is raised and holds a small round-shaped object (maybe a self-reference to the jasper
pebble?). The high-backed chair is characterized by an upper terminal ending in duck-head and
legs ending on bull hoofs. The legs appear to be connected by a middle horizontal rail. Under
the throne stands an object shaped like a cone with circular head which could be identified with
an incense stand. The presence of a high-backed throne, the gesture of the hand lying on the
lap, and the intriguing presence under the throne of an “incense stand” take us a step closer
to the iconography exhibited in KF IV. Main differences, however, remain: this is not a public
banquet ceremony but a private scene, the king has not a braid, the throne’s backrest is shorter
and less elaborated.
.... Babylonian-Elamite Stele
Upper section of a Babylonian stele found at Susa exhibiting an enthroned divinity offering the
rod and ring to an Elamite king (Louvre Museum Sb ; for description see Seidl : –
). The king stands holding hands next to an incense burner with triangular-head. He has a
visor hair-style complemented with a pair of long sideburn braids ending on a loop and a small
braid hardly visible emerging under the line of the hair.55 The date of this Elamite addition to the
Babylonian stele used to be contentious. Based on historical and iconographic grounds, Amiet
(: , fig. ; followed by E. Carter : ; and Calmeyer : ) vouched for a
representation of a th century Šutrukid ruler. But Harper (: –), following Seidl
(: ), maintains the th century bc for the Kassite original and the th century bc for
the Elamite recut addition.56
.... Highland Elamite Warriors [Fig. a–a]
Fragmentary bronze bas-relief found during the early excavations of the Acropole mound at
Susa.57 The relief included at least two registers separated by a horizontal line. Engraved on
the lower register there is a garden setting with trees and two large birds. The upper register
includes two scenes: the remains of an animal next to the lower body of a person oriented to
the left and wearing a long garment. Underneath, there is a row of seven similarly represented
male warriors oriented towards the right.
The right arm of the warriors is held behind the head at about a  degree angle. The hand is
holding a sickle-looking weapon. The left arm extends straight downwards, the hand is holding
a small bow with bended extremities. They carry a large quiver over the right shoulder held
perpendicularly with a band string going across the chest. The upper end of the quiver is filled
with arrows; the lower end can be seen at the level of the left elbow. The left arm (and perhaps
the hand?) includes a series of bands attached to the elbow, circling around, and crossing at wrist
level. The garments worn include an under skirt (the neck line is visible) and a single piece with
short sleeves and short skirt with the right side overlapping the left and held together with a
broad waist belt. Next to the corners of the skirt, on the lower edges, hang two intriguing hook-
shaped extensions. The peculiar shoes have pointy upwards toes. They are wearing a peculiar
helmet whose visor ends on a pointy edge which, as Amiet (: , fig. ) suggests, may

55 Seidl (: ) hesitated between a braid and the end of a headband.
56 As should be recalled, Seidl (: , and nt. ) proposed this Neo-Elamite date based on close analogies
with the reliefs from Shekaft-e Salmān at the time they were thought to have been made by Hanni of Ayapir (at the
time KF I was dated to the th century bc).
57 de Morgan (: –, pl. ); Börker-Klähn (: ). Housed at the Louvre Museum Sb .
 javier álvarez-mon

represent horns. To the side and back of the helmet the headdress combines a pair of long side
braids together with a single long braid at the back. The beard is divided into small stripes of
long hair matching the length of the side braids.
The seven warriors are integrated into a background covered by an Elamite inscription
(engraved vertically inside registers). V. Scheil (: ) offered a summary translation and
attributed its composition to the time of the Šutrukids.58 Amiet (: ) has indicated the
possibility that this bronze relief may have originated from an Elamite highland environment
(possibly Anshan) and was brought to Susa by a member of the Šutrukid dynasty. He has also
interpreted the individuals depicted as “a procession of ancient divinised warrior kings”.
The artistic references just reviewed present a sample of the vigorous artistic agenda and
originality of the Šutrukid Dynasty (c. –). As noted, some specific features such as
the visor hairstyle or helmet, the braided sideburns, or the long braided hair with upward loop
find close parallels in the reliefs from Shekaft-e Salmān; accordingly it is fitting to suggest that
this dynasty was directly involved in sponsoring some of these highland reliefs. The possibility
however that they were involved in the manufacture of Kūl-e Farah IV is, in my view, unlikely.
KF IV includes stylistic and iconographic features that distance its conception from those
espoused by the th century bc art of the Šutrukid House; to name the most pertinent: the
conventions used for the representation of human profile is significantly different, the hair is
collected into one single and much shorter braid (which does not end on a loop), there is lack
of sideburns, and the banquet scene is highly developed. At this regard, the iconography of the
Bar-Uli bead with the king sitting on a high-backed chair with duck-head terminal and the
“incense stand” could be said to announce that of KF IV.59 Equally important, the treatment of
the bronze plaque with rows of braided-hair archers, narrow waist line marked by a broad band
does also share some important elements with KF IV. In this latter case, though, as suggested
by Amiet, the plaque may postdate the Šutrukid period.
The end of the Shutrukid dynasty is often associated with an attack by the Babylonian
king Nebuchadnezzar I (– bc) and the demise of Hutelutuš-Inšušinak, son of Šilhak-
Inšušinak.60 In this context, Amiet (: –) suggests that KF II, III, IV, and VI (but not
KF V, which is dated together with KF I to the th–th centuries bc) manifest the expression
of a local monarchy that developed in eastern Elam after the invasion of Nebuchadnezzar I or
perhaps slightly later. The Elamite period between ca.  and  bc is considered a “dark
age” of Elamite history which, in fact, is a reflection of a gap in the archaeological record. Some
scholars have suggested that absence of evidence reflects the collapse of urban Elam which
came hand by hand with territorial fragmentation and adaptation of mainly pastoral-nomadic
based existence (de Miroschedji : ). Inspired by a similar model, Amiet has placed the
creative genius responsible for the highland reliefs from Kūl-e Farah (and Kurangūn II, which is
characterized by the presence of large numbers of participants), into a phenomenon of ethno-
genesis emerging after a process of settlement by nomadic populations that took place after
c. bc (Amiet :  n. ).61

58 The fragmentary text describes ram offerings by an unknown ruler to various Elamite divinities: Manzat
(inscription behind the second individual), Nahhunte and Laqamar (behind the th individual), Lagamar, Pinigir,
and Kiririša (behind the th individual), and Nahhunte and Kiririša (behind the th individual).
59 Incidentally, it is interesting to note that on the public domain Šilhak-Inšušinak shows continuity with

traditional figurative models depicting royal and divine imagery with long side-burns and braided hair but in the
private domain (the bead) these distinctive elements of Elamite royalty are missing.
60 For two additional kings of the dynasty (Šilhina-hamru-Lagamar and perhaps a Humban-numena) that may

have reigned well into ca. bc see Potts (: ) and Steve, Vallat & Gasche (–: ).
61 According to Amiet, the large participation of individuals and the presence of royal images at Kūl-e Farah III

and VI, “doivent être évocateurs de la nation entourant son roi; ils pourraient donc illustrer une prise de conscience
très nouvelle alors”.
 elamite sculptural reliefs from the highlands 

... The Rock-Cut Sanctuary of Kūrangūn (II)

Kūrangūn is situated on the ancient highway linking the Elamite capitals of Susa and Anšan
(Tal-e Malyan). The sanctuary was carved on a rock-cliff ca.  m high atop an outcrop
of the Kūh-e Pātāwēh which overlooks the Fahliyān River flowing through the panoramic
Mamasani region, in the southwestern province of Fārs. It was discovered in  by Herzfeld.
His description and line-drawings were rectified by Vanden Berghe ( and ) during
fieldwork achieved in  and  and, again, by a detailed study of the relief by Seidl ().
Debevoise () was the first to propose two distinct time periods for the carving of the
relief. Accordingly, it is believed that the three flights of stairs, the vertical central panel and
the horizontal ledge (Kūrangūn I) were made during the Old Elamite period.62 The three rows
of braided worshippers sculpted along the stairs and their associated leaders arranged in profile
(Vanden Berghe : ; , numbers –) were thought to be Neo-Elamite additions
(Kūrangūn II) but this date has been under scrutiny with some commentators suggesting a
possible end of the nd millennium bc.63 The worshipers are represented in profile with a long
braid at the back ending on a looped knob; their arms are extended with hands held together
at the waist level. They wear a short tunic held at the waist by a belt (Vanden Berghe ,
fig.  nos. , , , ). Their braid style and hand gesture contrasts with those exhibited by
the worshippers represented at KF IV.

... Lorestān Bronze Beakers (th–th Century bc)

The culture of Iron Age Lorestān is perhaps best known through a distinctive class of bronze
artifacts, especially horse paraphernalia, which today are classified under the generic title of
“Lorestān bronzes.”64 In  there were more that ninety-five decorated bronze beakers known
and it is suspected that many more have made their way into the antiquities market since.
Despite problems with forgeries and irregularities regarding origin and ownership Muscarella
(: ) indicates: “there can be no doubt that the art represented on many of the beakers
(‘situla art’) is of high quality and is a major source of cultural and artistic information”. They
most likely originated from an artistic environment sharing close analogies with Kassite and
Elamite artistic traditions. One of the most popular groups of series of beakers includes a
decoration with banquet scenes [Fig. h].65 The presence of high-backed thrones with duck-
head terminals suggests elements of comparison with KF IV.66

62 King Šilhak-Inšušinak dedicated a temple to Kilah-šupir at Tul-e Spid, in the vicinity of Kūrangūn (König :

, A); a fact suggesting close correspondences between the Šutrukid Dynasty (–) and Kurangun II.
63 In addition, Seidl (: –) proposes that four additional figures carved on the south-eastern face may be

a later addition (Vanden Berghe , numbers –). Seidl (: , nt. ) notes parallels between the four added
figures and ones depicted at KF II; a relief she places in the late Elamite period, following Vanden Berghe (), De
Waele () and Calmeyer ().
64 Notwithstanding the more than  publications on the subject of the Lorestān bronzes, the Iron Age cultures

of Lorestān remain, for the most part, an enigma (Muscarella : ; Overlaet : ).
65 Key recurring elements characterizing these scenes are: () a high backed throne with duck’s head end; () the

presence of musicians (mainly lute players); () the presence of a servant holding a fan or fly chaser; () heavily
ceremonial ornamented garments; () a cross-legged table with legs in animal’s hooves; () the men tend to have
prominent noses, large eyes, long hair resting on the shoulders and visible ears; () the seating elite individual holds
a beaker with the right hand, the left arm is generally extended and resting on the lap, the palm of the hand on the
knee.
66 Within this corpus, a beaker of doubtful authenticity includes a harp player (Muscarella : ). It is said

to have been purchased in the Kermanshah area (Maléki : , n. ) and has been dated to the th century bc, in
accord to assumed connections with Assyrian art of the time of Ashurbanipal (Calmeyer : –, C; Akurgal
: –). If this beaker is authentic—something still to be demonstrated—it would represent an evolving, more
sophisticated, version of the Neo-Assyrian and Lorestān banquet scenes from the th and th centuries bc.
 javier álvarez-mon

... Neo-Elamite Artistic Evidence: Kūl-e Farah I (Late th Century bc)
KF I is the most recent of all highland Elamite reliefs and consequently it bears the marks of
a multifaceted artwork assimilating aspects of two corpora of reliefs.67 A main “innovative”
aspect of KF I within the context of Elamite highland reliefs pertains to its manufacture. Even
if significant features such as the head of Hanni seems to have been defaced, the visible remains
suggest that it was finely carved with detailed engraving and had its surface highly polished (and
painted?; although no evidence of pigmentation has been observed by this author). An Elamite
cuneiform inscription occupying the upper half of the relief was carved over the imagery.68
A significant parallel between KF I and KF IV is the presence of small orchestras [Figs. c,
e]. At KF I the trio of musicians has been badly damaged, still we can observe two harp
players followed by a percussionist playing a square drum.69 The musicians are characterized
by hair gathered at the back into a bun, long ceremonial robes and a belt. Stylistically, the two
orchestras have few things in common.70 With regard to the chronology of the reliefs it seems
that the harps from KF I are quite similar to those played by the Elamite orchestras exhibited
at Assyrian palace slab reliefs and also the harp depicted in register II of the Arjan bowl.71
KF I is the only carving from this part of the valley whose late Neo-Elamite date is undisputed,
although a precise date is wanted.72 From the artistic view-point it is of interest to underline that
Hanni adopted a dome-shaped head-dress similar to the bulbous hat supported by the Elamite
king Humban-haltaš III (– bc). Of equal importance are the noted correspondences with
Elamite orchestras exhibited at the time of Ashurbanipal in c. bc and in the Arjan bowl

67 To note some of the most obvious: from Shekaft-e Salmān Hanni integrated the royal iconography of the

Šutrukids (the long braid with knobbed end, and the hieratic frontal position of the body and gesture of worship).
More significantly, he co-opted the ancestral royal imagery by adding his own inscriptions to reliefs carved in Shekaft-
e Salmān (EKI –). From Kūl-e Farah, reference is made to a number of features: () the body orientation and
gesture of worship from the sizable figure represented on the north side of KF III (ultimately with reference to
Shekaft-e Salman); () the trio of musicians find close analogies with those represented in KF III and IV (even if
the musicians and instruments themselves are treated differently); () the usage of scale to define status; thus the
presence behind him of the religious and military leaders, one on top of the other (reference to KF III, V, and VI);
() the sacrifice of animals associated with a fire altar (reference to KF V; but ultimately to Shekaft-e Salmān II).
Noticeably, these references suggest that the makers of KF I sought to integrate its visual and textual characteristics
into an artistic (socio-political and religious) tradition already well established.
68 This carving style contrasts with the sculptural “conventions” noted at KF IV. As mentioned, these insights need

corroboration by scientific analysis.

69 A fragmentary copper vase found by E. Herzfeld in Nahāvand (to the south of Hamadan) and dated by

Duchesne-Guillemin to c. bc, depicts a row of musicians including a harp player, a square drum player, and
two individuals clapping their hands (Duchesne-Guillemin : –).
70 The muscular bodies, narrow waist, and precise fingering gesture of the hand observed in the musicians from

KF IV is contrasted by the elegant but rather bulky shapes of the musicians from KF I. At KF I the type of portable
harps represented is also open (no presence of pillar) but the resonance-box of the vertical harp is represented on
a diagonal, forming an about  degree angle with the horizontal neck; as a result, the resonance box is not placed
directly to the left side of the face (as in KF IV or KF III) but slightly in a forward position. These small differences
may perhaps bear significance on the history of evolution of portable harps.
71 The royal Elamite orchestra depicted in the Southwest Palace of Sennacherib, Room , marks the arrival to

the royal city of Madaktu of the Elamite king Ummannigaš/Huban-nikaš II, newly appointed by Ashurbanipal just
after the Ûla(-)ya/Ulai river (Tell Tuba) battle and the death of Te"umman in bc (Barnett, Bleibtreu & Turner
, Room , Pl. ). The Arjan bowl has been dated by this author to the end of the th, beginning of the th
century bc (Álvarez-Mon ).
72 The inscriptions were first published and studied by Weissbach (); Scheil (: –); Hinz ();

and König (: –). For commentary see De Waele (b) and Stolper (: –). Within the text
there is an obscure passage mentioning king Šutur-Nahhunte, son of Indada. Based on epigraphic similarities with
other late Neo-Elamite inscriptions, Stolper (: ) proposes a date anywhere between the end of the th and
early th century bc with Vallat () supporting a roughly th century bc and, most recently, Tavernier (: ,
) proposing the last quarter of the th century bc.
 elamite sculptural reliefs from the highlands 

(c. bc).73 The relief represents the last link of a continuing tradition of Elamite sculptural
reliefs from the highlands and, consequently, provides the lowest chronological boundary
for KF IV. Its characteristics reveal continuity with past artistic, socio-political, and religious
references, including the existence of ceremony counting with a sacrificial feasting accompanied
by music.74

.. Chronology: Summary and Discussion

The goal of the previous discussion has been to provide artistic references against which to eval-
uate the likely manufacturing time of Kūl-e Farah IV. To recapitulate: () It remains uncertain
whether the ceremonial vessels represented in the relief are Middle Elamite goblets or Neo-
Elamite amphorae; () None of the banquet scenes represented in the glyptic from Tchoga
Zanbil or contemporary Mesopotamia depict correspondences or show a parallel level of com-
plexity; () High backed thrones are rarely attested during the Middle Elamite period (with
the notable exception of the th century bc throne of Šilhak-Inšušinak represented in the
jasper bead of Bar-Uli). They are characteristic of Neo-Assyrian glyptic of the th and th
centuries bc; () High backed thrones with protruding curved terminals representing duck-
heads are at home in Lorestān style bronze beakers made during the th and th centuries
bc; () The dimension and structure of the composition from KF IV recalls larger sculptural
reliefs panels and wall paintings so far lacking from the Elamite artistic record; () instead,
it is a th century bc royal banquet represented on the ivory strip found in Fort Šalmaneser
(in particular, the presence of a throne with curved back rest terminal, and two tables prob-
ably with lion feet) that seems to reveal some of the closest parallels; and finally () The
relief of Kūl-e Farah I, dated here to the end of the th century-beginning of the th cen-
tury bc, marks the last link on the artistic sculptural artistic production of the Elamite high-
lands.
In short, I believe the date of KF IV is framed by two opposite artistic poles: the rich
artistic heritage of the art of the th century bc Šutrukid dynasty and the relief of Kūl-e
Farah I (th–th century bc). In between, the unique style and mature iconography of the
relief, coupled with a sophisticated structural organization, and with the noted parallels with
Neo-Assyrian court-based banquet scenes (and possibly with Lorestān beakers) suggest the
possibility that KF IV was manufactured in the interval embracing the th and th centuries bc.
In the final analysis, this date reflects the dearth of direct artistic and archaeological references.
This absence underscores the exceptional attributes of KF IV and fosters the prospect that
we may be in presence of an original—to date not wholly understood—artistic production
representing the culture of the Elamite highlands. Hence, according to this conjecture, the
cultural epicentre from which this art originated remains elusive and cannot be located in
Susiana nor (as far as is it presently known) in the Anšanite highlands but most likely it ought to
be sought after in the highlands and piedmont associated with the territories of Īzeh/Mālamı̄r
and Kurangūn.

73 These correspondences therefore add further support to the date advocated by Tavernier for KF I, that is: the

last quarter of the th century bc or possibly shortly thereafter. Interestingly, Hanni lacks the side pair of side-burn
braids characterizing Elamite royalty and displayed by both Humban-haltaš III and the Elamite kings represented at
Shekaft-e Salmān. This absence may perhaps underlie his less-than-royal status.
74 Various aspects of the inscription do also reveal significant insights into the religious and royal ideology of the

late Neo-Elamite period (see Henkelman ).

 javier álvarez-mon

. Significance of Kūl-e Farah IV

The archaeological history of the valley of Īzeh/Mālamı̄r remains unknown. The reliefs men-
tioned in these pages, together with archaeological surveys, the existence of an ancient mound
(Layard : –) and of a carved limestone stele found in the village of Qal’ eh-ye Tol (about
 miles south of Izeh) suggests this area was inhabited since the late th millennium bc and,
in particular, belonged to a political and cultural koine during the nd and the st millennium
bc.75 The possibility exists that the nearby Rām Hormuz plain which counts with the attested
presence of large, but still unexcavated, Elamite urban centers may have played a pivotal posi-
tion on fostering a network of highland-lowland political and cultural communities.76 Within
this limited but promising archaeological context, it is becoming increasingly apparent that the
reliefs from Īzeh/Mālamı̄r provide a critical insight into various aspects of Elamite history. In
the ensuing comments, and keeping within the concentration on Kūl-e Farah IV, I will advance
a few remarks on the socio-political and ritual significance of this relief.
The study of KF IV has disclosed a distinctive artistic enterprise illustrating genuine orig-
inality without parallel in the arts of the ancient Near East. It is apparent to me that such
an accomplished and ambitious production materialized out of a mature and well established
artistic tradition. This tradition was intimately attached to a notion of place (Kūl-e Farah and
Īzeh/Mālamı̄r), the notion of “ethnic” self-identity (a social-group characterized by distinctive
physical features, most particularly braided long-hair), and the notion of custom and ritual
(defined by a communal shared meal). All together: place, community, and tradition provide a
nexus of identity markers defining a population characterized by a specific socio-political and
ritual ideology.
As a single relief, KF IV captures a “frozen-in-time” communal banquet whose participants
partake on a ritual entailing the consumption of a morsel of food, most likely a piece of
meat [see Fig. ]. This ceremony takes place within the musical background provided by
an orchestra. The creative genius of this artistic production is based on a keen approach to
the characterization of a ritual whose gravity is expressed through a highly organized scheme
of relative formulaic simplicity but profound ideological significance. The aesthetic choices
and distribution of the various registers and groups of individuals illustrate hierarchical order
planned along the lines of social status. This organization provides a well thought-out structural
representation of a social order whose hierarchical zenith is determined by the presence of a
king surrounded by community leaders.
Following sociological categories established for the analysis of the Greco-Roman sympo-
sium, five attributes of this social organization can be underlined: () the meal represents an
idealized model of the community; that is, an image of the community defining their view

75 According to the report by Stein (: : fig. ), the stele measured approximately / feet by  feet and
exhibits two rulers in long bell-shaped robes and their retinue facing each other. A survey of the Īzeh/Mālamı̄r and
the Qalah Tol plain provided evidence of settlement reaching back to the late th millennium and representing the
th and rd millennia bc (Stein : ; Wright & Redding ; Wright ).
76 The sites of Tepe Bormı̄ ( ha.) and Tall-e Ghazı̄r (. ha.), were both occupied during the second and first mil-

lennium bc (de Miroschedji : ; Carter ; Wright & Carter ). The ancient city of Huhnur/Hunar/Unar,
already mentioned in third millennium bc Mesopotamian texts, can presently be identified with a mound located in
this area (Luckenbill , No. ; Nasrabadi ). Funerary architectural evidence and pottery from Tepe Bormi
and Tall-e Ghazir can be compared to those found at Arjān and Susa Neo-Elamite II phase (/– bc; Álvarez-
Mon ). The recent fortuitous discovery in  of a burial containing two “bathtub” coffins and related precious
goods further underscores the cultural links between Susa, Rām Hormuz, and Arjān and the significance of the Rām
Hormuz region (and the city of Huhnur) in the transmission of Neo-Elamite urban traditions (de Miroschedji :
–; Carter : ; Henkelman : ).
 elamite sculptural reliefs from the highlands 

of themselves; () the meal includes social boundaries defined along age and gender lines. It
remains unclear whether or not female members of the community are present at KF IV (it
is possible that a group of individuals characterized by buns could be females; DIII: –).
The presence of community elders cannot be determined (although individual A may be a
possible candidate). Otherwise, most individuals represented are males of a similar “ideal”
body type and height; () sharing the same meal indicates a customary form of bonding and
some levels of equality amongst the participants; () careful ranking is established in relation
to close proximity to the ruler. Thus the noble and priestly class make up the close royal
entourage; () the presence of musicians mark a celebratory occasion; the culmination perhaps
of a set of rituals leading to consumption of the “first portion” of the meal. In all accounts, the
idealization of a community represented by a group of chosen individuals sharing a meal within
a highly sophisticated social structure demonstrates emphasis on the participatory nature of the
performance (the sharing of a piece of meat). This does not necessarily reject the exaltation of
the royal figure but, I think, reflects on an equalizing aspect of a society.
Despite the possible presence of a priestly class defined by long ankle-long ceremonial
garments, the relief does not appear to include religious symbols or make reference to a
supernatural order. This idiosyncratic absence of divinities (or divine symbols) could lead to
interpret the scene as “secular ritual”. In my view, however, this opinion needs to be tempered
by the unique majestic properties of the natural setting and by the fact that other reliefs
from Izeh-Malamir are clearly dedicated to divinities (KF I), may include divine presence
(represented perhaps by a divine sculpture being carried atop a platform; KF III and VI),
and religious liturgical attributes (symbolized by a fire altar; KF I).77 Focus on the communal
partaking of food implies that we are assisting at a ceremony where the sharing of the meal
takes ritual significance.78 The location of the Kūl-e Farah IV next to a natural spring and to
reliefs illustrating ritual processions, music, and banquets suggests that this place must have
enjoyed a particular religious significance; perhaps of transcendent nature. This possibility is
further supported by the evidence from Kurangūn and from Shekaft-e Salmān (where the royal
worshipers are oriented towards the grotto and its related natural spring), suggesting that a
distinctive characteristic of Elamite highland religious practices included ritual processions and
the enactment of ceremonies in natural open-air sanctuaries (Potts ). Indeed, this may
advocate a belief system where the notion and experience of the supernatural developed out of
an association with landscapes of extraordinary natural properties.

77 A study of religious symbolism from the relief of Izeh-Malamir remains to be written.

78 The ramifications of this subject cannot be treated here adequately. I address the reader to the noteworthy
publication of Henkelman (: –) where the author investigates parallels between Elamite and Persian
Achaemenid lan banquet ritual ceremonies. In the context of the Persepolis Tablets context, the lan ceremony was a
monthly (and sometimes daily) animal sacrifice, a regular way of making offerings, Accordingly, the author proposes
the existence of a “particular type of offering that had an Elamite origin and was reshaped in the process of Elamo-
Iranian acculturation to become a Persian offering practiced in the Achaemenid heartland”. More patently, and in
reference to Kūl-e Farah I, Henkelman states (: ): “The grand sacrificial feast regularly celebrated at Kūl-e
Farah, presided over by the local ruler and organized in a way that expressed the unity and social build-up of an entire
society, is a logical antecedent for a very similar feast (šip) documented in the Persepolis texts.” For further discussion
on the possible Elamite heritage of ritual community meals in ancient Iranian religion see Hultgård ().
 javier álvarez-mon

. Postscript: Braids of Glory

The topic of long hair and identity in the history of the ancient Near East remains dominated
by by popular accounts of individual heroes. Few characters have captured artistic and literary
imagination of the West more that the single figure of the tragic hero Samson and the tales
associated with his hair. Samson wears his long hair braided into seven locks (Judges ).79
His hair is a mark of identity implying a culture of hair representing warrior status, charisma,
manliness, and divine selection (Niditch : –). As a folk tragic hero, he is a kind of
Israelite Gilgameš, a heroic figure of unmatched physical force. Within the biblical chronology,
the Samson narrative is set in pre-Monarchic times (before c. bc).
Gilgameš and his twin companion Enkidu provide a broader Near Eastern cultural back-
ground to the role of long hair. Both heroes wear their hair long but in the context of the epic,
long hair is qualified by both gender and urban-centred boundaries. Enkidu’s wears his hair
loosed in tresses more “like a woman”, while Gilgameš wears his carefully braided (and, one
has to assume, properly maintained).80 Both hair styles echo a recurrent dichotomy running
through the epic of Gilgameš, namely: the contrast between the primarily uncivilized, untamed,
and coarse status of Enkidu and the urban-based civilized, status of the king of Uruk, Gilgameš.
In parallel to this stereotypical aspect of the tale, there is also a social element associated with
long hair that is introduced by the sudden death of Enkidu. To mark a distinctive mourning
and grief status Gilgameš returns to a “primitive condition” symbolized by leaving his long hair
bedraggled.
A similar bias and dichotomy between the civilized and uncivilized worlds can be observed in
a more historical context. The victory celebration stele of Naram-Sîn and the closely related, but
probably not contemporary, carved relief of Darband-i Gawr (in the Zagros piedmonts, south of
Sulaimaniya) portray a victorious Mesopotamian ruler crushing the bodies of defeated Zagros
highlanders characterized by long hair whose outline is in the shape of an elongated cone.81 The
defeated enemies are supposed to be the highland Lullubi whose homeland was located to the
east of present day Irbil and were most-likely closely related to their southern Elamite speaking
neighbors.82 In this context the long hair of the Lullubi has been interpreted as a symbol of the
emasculated enemy, a visual mechanism to gender the enemy as “female” and thus stripping off
their pride (Winter : ).
The textual and visual references associated with Samson, Gilgameš, Enkidu, and the Lullubi
highlanders, provide legendary, cultural, and historical backgrounds to the evidence discussed
in these pages. In the case of Kūl-e Farah IV, however, we are assisting to a cultural display

79 Then Delilah said to Samson, “Up to now you have mocked me, and told me lies; tell me how you may be bound.”

And he said to her, “If you weave the seven locks of my head with the web and make it tight with a pin, then I shall
become weak and be like any other man.”
80 Gilgameš Standard Version, see both šārtu and pērtu, in particular I:  and VIII:  (George ).
81 The stele of Naram-Sîn was brought to the western Elamite capital by Šutruk-Nahhunte who “at the command

of Inšušinak struck down Sippar”. The Elamite king co-opted its original significance by consecrating the work to
Inšušinak (König :  n. ). The similarities between both reliefs were first noted by Sydney Smith (Edmonds
: ). He also noted the difference in headdress style (in the first case representing a divine horned helmet, in
the second a plain rolled cap akin to that worn by Gudea). Boese (: –) argued for a date in the Ur III period
because of this headdress difference; Roaf (in Postgate & Roaf : ) suggests that the Darband relief was copied
from the Naram-Sîn stele.
82 For the territory of the Lullu (bum) see Zadok (). Ashurnasirpal II confronted the Lullu in the area around

mount Lara (Grayson ,  and ), and in the same area mentions tribute by inhabitants wearing their hair like
women (see also comments by Winter : , nt. ). Note also the exceptional representation of a Kassite king
on a Kudurru housed at the Walters Art Gallery, Baltimore with long (braided?) hair ending on a knob (Calmeyer
: , fig. ). See also Potts ().
 elamite sculptural reliefs from the highlands 

celebrating the properties of hair as an element of personal and communal identity. Lack of
specific additional evidence prevents from asserting whether the carefully braided long hair of
the Elamite highlanders can be construed in light of a military victory, urban-centered biases,
or divine granted power. Yet, at the end, the skillful visual display of Kūl-e Farah IV discloses
a collective communing with itself and, no doubt, a community gazing at itself. It is tempting
to be rhetorical about a composition where aesthetic and ideological choices merge to manifest
an elegant display of order through unity; in sum, a highland culture in all its glory.

Abbreviations

KF Kūl-e Farah
SS Shekaft-e Salmān
RHF Right-Hand holding Food

AA Arts Asiatiques
AJA American Journal of Archaeology
AMI Archäologische Mitteilungen aus Iran
IrAnt Iranica Antiqua
JCS Journal of Cuneiform Studies
JNES Journal of Near Eastern Studies
JRGS Journal of the Royal Geographical Society
MDAI Mémoires de la Délégation Archéologique en Iran
MDP Mémoires de la Délégation en Perse
MMAI Mémoires de la Mission Archéologique en Iran
RA Revue d’Assyriologie
RlA Reallexikon der Assyriologie und vorderasiatische Archaologie
SIr Studia Iranica
ZA Zeitschrift für Assyriologie

Bibliography

Akurgal, E., (): Urartäische und altiranische Kunstzentren, Ankara.

Álvarez-Mon, J. (): “Imago Mundi: Cosmological and Ideological Aspects of the Arjan Bowl”, IrAnt
, –.
———. (): The Arjan Tomb: at the Crossroads of the Elamite and the Persian Empires, Leuven.
Amiet, P. (): Glyptique Susienne (MDAI ), Paris.
———. (): “Disjecta Membra Aelamica, le décor architectural en briques émaillées a Suse”, AA ,
–.
———. (): “La glyptique du second millenaire en provenance des chantiers A et B de la Ville Royale de
Suse, Annexe ”, M.-J. Steve, H. Gasche & L. de Meyer, “La Susiane au deuxieme millenaire: A propos
d’une interprétation des fouilles de Suse”, IrAnt , –.
———. (). Suse,  ans d’histoire, Paris.
———. (). “Victory Stele of Naram-Sin”, P.O. Harper, J. Aruz & F. Tallon (eds.), The Royal City of Susa:
Ancient Near Eastern Treasures in the Louvre Museum, New York, –.
———. (). “Glanes élamites”, R. Dittmann et al. (eds.), Variatio Delectat Iran und der Westen Gedenk-
schrift für Peter Clameyer (AOAT ), Münster, –.
Barnett, R.D., Bleibtreu, E. & Turner, G. (): Sculptures from the Southwest Palace of Sennacherib at
Nineveh, London.
Boese, J. (): “Zur stilistischen und historischen Einordnung des Felsreliefs von Darband-i Gaur”, StIr
, –.
Börker-Klähn, J. (): Altvorderasiatische Bildstelen und vergleichbare Felsreliefs (DAI. Baghdader
Forschungen ), Mainz am Rhein.
Calmeyer, P. (): “Eine Westiranische Bronzewerkstatt des / Jahrhunderts v. Chr. Zwischen Zalu
Ab und dem Gebiet der Kakavand-I”, Berliner Jahrbuch für Vor- und Frühgeschichte , –.
 javier álvarez-mon

———. (): “Zur Genese altiranischer Motive”, AMI , – (Tafeln –).
———. (): “Mālamı̄r. C. Archäologisch”, RlA , –.
———. (): “Middle Babylonian Art and Contemporary Iran”, J. Curtis (ed.), Later Mesopotamia and
Iran: Tribes and Empires –BC, London: –.
Carter, E. (): “Archaeology”, E. Carter & M.W. Stolper, Elam, Surveys of Political history and Archae-
ology (University of California Publications. Near Eastern Studies ), Berkeley, –.
———. (): “The Middle Elamite Period ca. –BC”, P.O. Harper, J. Aruz & F. Tallon (eds.),
The Royal City of Susa: Ancient Near Eastern Treasures in the Louvre Museum, New York, –.
———. (): “Bridging the Gap Between the Elamites and the Persians in Southeastern Khuzestan”,
H. Sancisi-Weerdenburg, A. Kuhrt & M.C. Root, Continuity and Change: Proceedings of the Last
Achaemenid History Workshop, April –, Ann Arbor (Achaemenid History ), Leiden, –.
———. (): Excavations at Anshan (Tal-e Malyan): the Middle Elamite Period, Philadelphia.
Caubet, A., (): “Le temple d’Inshushinak de Suse et l’architecture monumentale en “faience””,
D.T. Potts, M. Roaf, & D. Stein (eds.), Culture through Objects: Ancient Near Eastern Studies in Honour
of P.R.S. Moorey, Oxford: –.
Collon, D. (): Catalogue of the Western Asiatic Seals in the British Museum, Cylinder Seals V, Neo-
Assyrian and Neo-Babylonian Periods, London.
Debevoise, N.C., (): “The Rock reliefs of Ancient Iran”, JNES , –.
De Waele, E. (a): Shutruk-Nahunte II et les Reliefs Rupestres dits Néo-Elamites d’Iseh/Mālamı̄r,
Revue des Archéologues et Historiens d’Art de Louvain : –.
———. (b): Quelques aspects de la religion d’Élam à travers l’art rupestre d’époque Neo-Élamite
D’ Īzeh/Mālamı̄r, Iran Bastan Museum Publication, Tehran.
———. (): Une page d’art Iranien: les reliefs rupestres d’ Īzeh Mālamı̄r, Archeologia : –.
———. (a): Les Reliefs Rupestres Élamites de Shekāf-e Salmān et Kūl-e Farah près d’Izeh (Mālamir).
Doctoral Dissertation, Université Catholique de Louvain, Belgium.
———. (b): Remarques sur les Inscriptions Élamites de Šekākt-e Salmān et Kūl-e Farah près Izeh,
I. Leur Corrélation avec les Bas-Reliefs, Le Muséon , fasc. –: –.
———. (): Les Processions avec Statues Divines sur les Reliefs Ruprestres Elamites, Kūl-e Farah III et
Kūl-e Farah VI (Īzeh), Akten de VII. Internationalen Kongresses für Iranische Kunst und Archäologie,
Berlin, –.
———. (): Travaux Archéologiques à Shekaft-e Salmān et Kūl-e Farah prés d’ Īzeh (Mālamı̄r), IrAnt
: –.
———. (): Musicians and musical instruments on the rock reliefs in the Elamite sanctuary of Kūl-e
Farah (Īzeh), Iran : –.
Duchesne-Guillemin, M. (): “Note sur la provenance Asiatique d’un tambour Égyptien”, AMI ,
–.
Edmonds, C.J. (): “Two Ancient Monuments in Southern Kurdistan”, The Geographical Journal /,
–.
George, A. (): The Epic of Gilgamesh, a New Translation, New York.
Ghirshman, R. (): Village perse-achéménide (MDP ), Paris.
———. (): Tchoga Zanbil (Dur-Untash): Vol. I, La Ziggurat (MDP ), Paris.
———. (): Tchogha Zanbil (Dur-Untash): Vol. II, Temenos, temples, palais, tombes (MDP ), Paris.
Grayson, A.K. (): Assyrian Rulers of the Early First Millennium bc.  (–bc) (RIMA ),
Toronto.
Hansman, J., (): “Elamites, Achaemenians and Anshan”, Iran , –.
Harper, P.O. (): “Stele with an Elamite Ruler Approaching a Seated God”, P.O. Harper, J. Aruz &
F. Tallon (eds.), The Royal City of Susa: Ancient Near Eastern Treasures in the Louvre Museum, New
York, –.
Henkelman, W.F.M. (): “Persians, Medes and Elamites: Acculturation in the Neo-Elamite Period”,
G.B. Lanfranchi, M. Roaf, & R. Rollinger (eds.), Continuity of Empire (?) Assyria, Media, Persia
(History of the Ancient Near East Monographs ), Padova, –.
———. (): The Other Gods Who Are, Studies in Elamite Iranian Acculturation based on the Persepolis
Fortification Texts (Achaemenid History ), Leiden.
Hermann, G., Coffey, H. & Laidlaw, S. (): The Published Ivories from Fort Šalmaneser, a scanned
archive of photographs, London.
Hinz, W. (): “Die Elamischen Inschriften des Hanne”, W.B. Hennig & E. Yarshater (eds.), A Locust’s
Leg. Studies in Honour of S.H. Taqizadeh, London, –.
 elamite sculptural reliefs from the highlands 

Hultgård, A. (): “Ritual Community Meals in Ancient Iranian Religion”, M. Stausberg (ed.), Zoroas-
trian Rituals in Context, London, –.
Hüsing, G., (): Der Zagros und seine Völker. Eine archäologisch-ethnographischte Skizze, Leipzig.
Jéquier, G. (): “Appendice: Description du site de Mālamı̄r”, V. Scheil, Textes élamites-anzanites.
Première série (MDP ), Paris, –.
Keel-Leu, H., & Teissier, B. (): Die vorderasiatischen Rollsiegel der Sammlungen “Bibel+Orient” der
Universität Freiburg Schweiz, Fribourg.
König, F.W. (): Die elamischen Königsinschriften (AfO. Beih. ), Graz.
Layard, A.H. (): “A Description of the Province of Khuzistan”, JRGS , –.
Luckenbill, D.D. (): Ancient Records of Assyrian and Babylonia vols. I & II, London.
Malbran-Labat, F. (): Les Inscriptions Royales de Suse, Briques de l’époque paléo-élamite à l’Empire
néo-élamite, Paris.
Maléki, Y. (): “Situle à scène de Banquet”, IrAnt , –.
Mallowan, M. & Davies, L. (): Ivories in Assyrian Style. Ivories from Nimrud (–), Fascicule II,
London.
Matthews, D.M. (): Principles of Composition in Near Eastern Glyptic in Near Eastern, Glyptic of the
Later Second Millennium bc, Freiburg.
———. (): The Kassite Glyptic of Nippur, Freiburg.
Mecquenem, R. de. (): “Fouilles de Suse: Campagnes des Annees --”, RA , –.
———. (): “Fouilles de Suse –”, R. de Mecquenem, G. Contenau & R. Pfister (eds.),
Archéologie susienne (MMAI ), Paris, –.
Miroschedji, P. de. (): “Stratigraphie de la periode neo-elamite a Suse (c. –c. )”, Paléorient ,
–.
———. (): “Notes sur la glyptique de la fin de l’Elam”, RA , –.
———. (): “La fin du royaume d’Ansan et de Suse et la naissance de l’Empire perse”, ZA , –.
———. (): “La fin de l’Elam: essai d’analyse et d’interpretation”, IrAnt , –.
———. (): “Susa and the Highlands. Major Trends in the History of Elamite Civilization”, N.F. Miller &
K. Abdi (eds.), Yeki Bud, Yeki Nabud: Essays on the Archaeology of Iran in Honor of William M. Sumner,
Los Angeles, –.
Morgan, J. de (): Recherches Archéologiques (MDP ), Paris.
Muscarella, O.W. (): “Decorated Bronze Beakers from Iran”, AJA /, –.
———. (): Bronze and Iron, Ancient near Eastern Artifacts in the Metropolitan Museum of Art, New
York.
Nasrabadi, B.M. (): “Eine Steininschrift der Amar-Suena aus Tappe Bormi (Iran)”, ZA , –.
Niditch, S. (): “My Brother Esau Is a Hairy Man”, Hair and Identity in Ancient Israel, Oxford.
Overlaet, B. (): The Early Iron Age in the Posht-e Kuh, Lorestan, LED IV (Acta Iranica ), Leuven.
Porada, E. (): “More seals of the time of the Sukkalmah”, RA , –.
Postgate, J.N. & Roaf, M.D. (): “The Shaikhan Relief ”, Al-Rāfidān , –.
Potts, D.T. (): The Archaeology of Elam. Formation and Transformation of an Ancient Iranian State,
Cambridge.
———. (): “The Numinous and the Immanent. Some Thoughts on Kūrangūn and the Rudkhaneh-e
Fahliyān”, K.von Folsach, H. Thrane & I. Thuesen (eds.), From Handaxe to Khan. Essays Presented to
Peder Mortensen on the Occasion of his th Birthday, Aarhus, –.
———. (): “Elamite Temple Building”, M. Boda and J.R. Novotny (eds.), From the Foundations to the
Crenellations: Essays on Temple Building in the Ancient Near East and Hebrew Bible, Münster, –.
———. (): “The ABBUTTU and the Alleged Elamite ‘Slave Hairstyle’”, L. Vacín (ed.), U DU-
GA-NI SÁ MU-NI-IB-DU, Ancient Near Eastern Studies in Memory of Blahoslav Hruška, Dresden,
–.
Rawlinson, H.C. (): “Notes on a march from Zohab, at the foot of Zagros along the mountains to
Khuzistan (Susiana), and thence through the province of Lorestān to Kirmanshah in the Year ”,
JRGS .
Rowe, M.W. & Steelman, K.L. (): “Radiocarbon Dating of Rock Painting Using Plasma-Chemical
Extraction”, American Laboratory , –.
Scheil, V. (): Textes élamites-anzanites, première série (MDP ), Paris.
———. (): Textes élamites-anzanites, quatrième série (MDP ), Paris.
Seidl, U. (): “Zur Umarbeitung zweier Stelenbekronungen aus Susa und anderer altorientalischen
Reliefs”, Berliner Jahrbuch für Vor- und Frühgeschichte , –.
———. (): Die Elamischen Felsreliefs von Kurangūn und Naqsh-e Rustam, Berlin.
 javier álvarez-mon

Sollberger, E. (): “A New inscription of Šilhak-Inšušinak”, JCS , –.

Steelman, K.L. et. al. (): “Direct Radiocarbon Dating of Megalithic Paints from North-west Iberia”,
Antiquity , –.
Stein, A., (): Old Routes of Western Iran, London.
Steve, M.-J. & Gasche, H. (): “L’acces a l’au-dela, a Suse”, H. Gasche & B. Hrouda (eds.), Collectanea
Orientalia Histoire, Arts de l’Espace et Industrie de la Terre (CPOA ), Paris, –.
Steve, M.-J., Vallat, F. & Gasche, H. (–): “Suse. F. Suse dans l’histoire”, DBS , –.
Stolper, M.W. (): “Political History”, E. Carter & M.W. Stolper (eds.), Elam: Surveys of Political History
and Archaeology, Berkeley, –.
———. (): “Mālamı̄r. B. Philologisch”, RlA , –.
Stronach, D. (): “Achaemenid Village I at Susa and the Persian Migration to Fars”, Iraq , –.
Tavernier, J. (): “Some Thoughts on Neo-Elamite Chronology”, ARTA ., –.
Vallat, F. (): “Nouvelle analyse des inscriptions neo-elamites”, H. Gasche & B. Hrouda (eds.),
Collectanea Orientalia. Histoire, arts de l’espace et industrie de la terre (CPOA ), Neuchatel—Paris,
–.
———. (): “Atta-hamiti-Insusinak, Shutur-Nahhunte et la chronology neo-elamite”, Akkadica ,
–.
Vanden Berghe, L. (): “Les reliefs élamites de Mālamir”, IrAnt , –.
———. (): Reliefs Rupestres de L’Iran Ancien, Bruxelles.
———. (): “Données Nouvelles Concernant le Relief Rupestre Elamite de Kūrangūn”, L. de Meyer,
H. Gasche & F. Vallat (eds.), Fragmenta Historiae Elamicae, Paris: –.
Weissbach, F.H. (): Neue Beiträge zur Kunde der susischen Inschriften (Abhandlungen der Konigl.
Sachsischen Gesellschaft der Wissenschaften XXIV/Phil.-hist. Klasse XIV/), Leipzig.
Winter, I. (): “Tree(s) on the Mountain: Landscape and Territory on the Victoria Stele of Naram-Sîn
of Agade”, L. Milano et al. (eds.), Landscapes, Territories, Frontiers and Horizonts in the Ancient Near
East, Part I, Padova, –.
Wright, H.T. (): “The Susiana Hinterlands during the Era of Primary State Formation”, F. Hole
(ed), The Archaeology of Western Iran. Settlement and Society from Prehistory to the Islamic Conquest,
Washington, –.
Wright, H.T. & Carter, E. (): “Archaeological Survey on the Western Ram Hormuz Plain”, N.F. Miller
& K. Abdi (eds.), Yeki Bud, Yeki Nabud: Essays on the Archaeology of Iran in Honor of William M. Sum-
ner, Los Angeles, –.
Wright, H.T. & Redding, R. (): “The Paleolithic periods on the Izeh Plain”, H.T. Wright (ed.),
Archaeological investigations in Northeastern Xuzestan,  (University of Michigan. Museum of
Anthropology. Technical Reports . Research Reports in Archaeology ), Ann Arbor, –.
Zadok, R. (): “Lulubi”, EncIr (http://www.iranica.com/articles/lulubi; accessed --).
 elamite sculptural reliefs from the highlands 

Fig. : Location of the Valley of Īzeh/Mālamı̄r; [a] Landsat map

of Iran; [b] Landsat D rendering after Cameron Petrie (http://www
.archatlas.dept.shef.ac.uk/IndexAAP.htm); [c] Map of Southwest Iran
by Hansman (: , fig. ); [d] Map of the region after De Waele (a).
 javier álvarez-mon

Fig. : Reliefs from Kūl-e Farah (photographs by the author).

 elamite sculptural reliefs from the highlands 

Fig. : Relief from Shekaft-e Salmān (photographs by the author).

 javier álvarez-mon

Fig. : View of Kūl-e Farah IV; [a] after Vanden Berghe (); [b, c]
photographs by the author; [d] line-drawing after De Waele ().

Fig. : Views of Kūl-e Farah IV (after Vanden Berghe )

 elamite sculptural reliefs from the highlands 

Fig. : View of Kūl-e Farah IV: A (sketch by de Morgan ; photograph by the author).

Fig. : [a-e] Views of panel IV: AI (a. sketch by de Morgan; b-d. photograph by
the author; c-e. photograph after Vanden Berghe ); [f] View of the vessels in
register AIII (photograph by the author); [g, h, i] line-drawings and photographs of
Middle-Elamite goblets (drawings after de Miroschedji ; photographs by the
author) [g] neo-Elamite amphorae vessels (after de Miroschedji ).
 javier álvarez-mon

Fig. : Panels -B (top), and -C (bottom); Photographs by the author.

elamite sculptural reliefs from the highlands 

Fig. : Panels B and C (Photographs by the author).

 javier álvarez-mon

Fig. : Individual C:  [a,b,d] top photographs and line-drawing

by the author; [c] photograph after Vanden Berghe .
 elamite sculptural reliefs from the highlands 

Fig. : Panel CII; [b] photograph after Vanden Berghe (); [c, d, e] Elamite
orchestras from Kūl-e Farah IV (c), III (d), and I (e); Photographs and line-drawings
by the author (after De Waele ); [f] Elamite Royal orchestra represented in
the Southwest Palace of Sennacherib, Room , marking the arrival to the royal
city of Madaktu; ca. bc (Barnett, Bleibtreu & Turner , Room , pl. ).
 javier álvarez-mon

Fig. : Panel [photographs and line-drawings by the author

except main photograph by Vanden Berghe ()].
 elamite sculptural reliefs from the highlands 

Fig. : Banquet Representations from:

[a] Tchoga Zanbil seal, th century bc (photograph by the author).

[b] Kassite-style III c. –bc; BM  (line drawing by Matthews : fig. ).
[c] Neo-Assyrian seal, th–th century bc (Keel-Leu & Teissier , seal z).
[d] Neo-Assyrian seal, th–th century bc (Keel-Leu & Teissier , seal ).
[e] Neo-Assyrian seal, th–th century bc; BM  (Collon , Pl. XI: ).
[f1,f2] Ivory strip, Fort Šalmaneser, Nimrud, th century bc
(Hermann, Coffey & Laidlaw : ).
[g] Bitumen plaque from Susa; th century bc (photograph by the author).
[h] Line-drawing of scene exhibited in a bronze beaker allegedly found
in the Luristan region, th–th century bc (after Muscarella ).
[i] Elamite seal? (Porada : , Pl. xv, fig. ).
 javier álvarez-mon

Fig. :

[a1, a2] Shekaft-e Salmān II with detail of the queen’s robe (photographs by the author).
[b] Detail from Kūl-e Farah III (photograph by the author).
[c] Detail from Kūrangūn II: individual  (photograph by
the author; line-drawing by Vanden Berghe : ).
[d] Hanni from Kūl-e Farah I (heliogravure by Dujardin
published by Scheil ; line-drawing by the author).
[e] Line drawing of glazed bricks from the Acropole
representing the Elamite royal couple (Amiet : –).
 elamite sculptural reliefs from the highlands 

Fig. : [a] Bronze bas-relief

from Susa; Louvre Museum Sb ;

[a1] photograph by the author;

[a2–a3] line drawings by J. de Morgan (: –, pl. );
[a4] the author.
[15b] Jasper pendant of princess Bar-Uli; British Museum 
(photo after Calmeyer : ; line-drawing by the author).
 javier álvarez-mon

Fig. : Line-drawing and structural divisions of Kūl-e Farah IV (by the author).
 FROM SUSA TO PERSEPOLIS:
THE PSEUDO-SEALING OF THE PERSEPOLIS BRONZE PLAQUE

Gian Pietro Basello*

Like other bronze objects from Elam (the statue of queen Napir-asu,1 the “bronze aux guerriers”,2
the “barrière de bronze”3 and the model commonly called Sit šamši4), the bronze plaque found
in the so-called Treasury at Persepolis is a unique discovery. According to Erich F. Schmidt, a
single iron spearhead, three onyx eye stones, a faceted stone bead, a corrugated bronze strip, a
slightly curved bronze disk pierced by three iron nails, a curved spouted bronze object and
a green chert pestle were found in the same room.5 Nineteen uninscribed but sealed clay
tablets of peculiar oblong shape were also found there.6 If the sacking of the building could
explain such arrangement of things, the association of the Persepolis bronze plaque (henceforth
PBP) with such objects rises several questions, especially on when and why it entered the
Treasury.
PBP is inscribed on both sides with a long Elamite text. Some scholars cite PBP as “Ururu”,7
after the personal name, written u-ru-ru, that appears at least eight times in the text.8 After
the discoveries of the Persepolis Treasury and Fortification Wall administrative tablets, the
documentation of Elamite language in Persepolis is no more surprising. Yet, PBP stands out

* ‘L’ Orientale’ University, Naples (Italy).

1 Louvre Museum number Sb , . m tall, kg, fragmentary cast bronze statue (the head and the left
arm are missing) discovered in , from Susa, temple of Ninhursag. Published in MDP VIII: – (‘Statue
de la reine Napir-Asou’ by G. Lampre) and pls. XV–XVI; inscription published by Scheil in MDP V : – =
EKI . Dated to ca. – bc. See also Tallon in Harper & al. : –, no. , and Potts : –
.
2 Louvre Museum number Sb , . .m, fragmentary bronze relief with seven extant warrior gods,
×
from Susa, pit . Published in MDP I: – (‘Bas-relief de bronze’ by de Morgan) and pl. XIII (heliogravure).
Inscription summarized in MDP XI :  and fig.  (autograph), published in EKI . Dated to th–th cent.
bc. See also Potts : , pl. ., and the brief discussion in Steve & al. –: col.  (‘Ce monument n’est
certainement pas susien’).
3 Two hollow bronze cylinders, one . m long and .m in diameter, the other shorter, probably with the same

inscription of Šilhak-Inšušinak I (ca. – bc), discovered in , from the Acropolis of Susa. Described in
MDP VII: . Inscription published in MDP V : – and pls. – (heliogravure), and in EKI .
4 Louvre Museum number Sb , base ×cm, three-dimensional bronze model reproducing a cultual

scene discovered in the – campaign, from Susa. Published in Gautier  with transliteration and
translation of the inscription of Šilhak-Inšušinak I (ca. –bc) by Scheil; text published also in MDP XI
: – (‘Texte de Šilhak In Šušinak. Plateau votif de bronze’ by Scheil) and pl. , no. , and in EKI . See also
˘
Tallon in Harper & al. : –, no. , and Basello .
5 Schmidt : .
6 Schmidt : – (‘clay labels’ in Schmidt’s terminology) and fig. ; also Schmidt : . The impressed

seals are Schmidt , no.  (on PT ), no.  (on PT , PT , PT ), nos.  and  (on PT ,
PT , PT , PT , PT –, PT , PT , PT –, PT ). On the uninscribed tablets
from the Treasury, see Garrison & Root : ; on those from the Fortification Wall, see Garrison . Other
uninscribed tablets are published in Henkelman & al. .
7 ‘Oruru’ in Hinz & Koch ; ‘Ururu’ in Waters ; ‘Ururu Bronze Tablet’ in Tavernier . Other

conventional name: ‘Elamite bronze plaque’ in Schmidt ; ‘Hub Pe’ (abbreviation for Huban-Šuturuk, Persepolis)
in Vallat .
8 According to Cameron apud Schmidt : : Ururu son of Dununu in PBP:, , rev., rev. and rev..

According to the unpublished transliteration by Steve: u-ru-ru in PBP:, , , , –, ; u-ru-ru DUMU
šá-du-nu-nu in PBP:–, . According to Hinz & Koch : , s.v. ‘u-ru-ru’: PBP:, , , , rev. (=
PBP:S), rev. (= PBP:S), rev.– (= PBP:–S), rev.  (= PBP:S).
 gian pietro basello

as an exceptional piece in Persepolis: the fact that it is an inscribed bronze plaque makes it clear
that it is not a standard administrative document; nor is it comparable to the gold and silver
plaques (DPh) found below the so-called Apadana,9 being far from themes and lexicon of the
royal Achaemenid inscriptions. At its discovery, the text was labelled as ‘neo-Elamite’ or ‘late
Elamite’, i.e. ‘pre-Achaemenian’.10

. Find-Spot and Physical Description

PBP was found during the excavations of the Persepolis Treasury led by Erich F. Schmidt in
. It lay covered with charred matter on the floor near the South-East corner of room
.11 Room  leads to room  to the North and to room  to the South, both being
dead ends. No windows seem to be opened on the external wall.12 The walls of the rooms
are scorched by fire.13 No traces of doors were found except for the one between hall 
and room , swinging into the room; Schmidt remarked: ‘this doorway had frequent use,
as shown by the patched strip of floor’.14 In room  four Elamite tablet fragments were
found;15 one of these fragments has possibly been published as PT ,16 a fragmentary ‘list
of names of individuals belonging to the army’ without parallel among texts from the Trea-
sury.17
PBP measures ca. . ×.cm and is . cm thick.18 The length of the long side of the plaque
is approximately comparable to the lengths of the sides of square metal plaques and stone tablets
bearing Achaemenid foundation inscriptions, e.g., the gold and silver plaques from the so-
called Apadana of Persepolis (DPh) measuring . × . × . cm,19 and the stone tablets from
the Apadana mound of Susa (DSz and DSaa) measuring .× . × . cm.20 It is noteworthy
that ca. cm was a standard Persepolitan unit of length, probably derived from Mesopotamia.21
From the ends of the upper edge of PBP project ‘two tabs, pierced by remnants of bronze wire,
which terminate on the obverse of the object in pronounced lobes’.22 Remnants of the bronze
wire were found in the vicinity, in at least five curved pieces.23 If a ‘looped wire’ really hung the

9 Except partly for the size (see below).

10 ‘Neo-Elamite’: Schmidt : ; ‘late Elamite, pre-Achaemenian’: Schmidt : .
11 Schmidt : , and photograph on p. , fig.  E. Field number PT .
12 Schmidt : .
13 Schmidt : .
14 Schmidt : , also p. .
15 Schmidt : ; Schmidt :  and fig. .
16 The ‘Index of tablet numbers and locations’ in Cameron : –, lists only four tablets whose field

number begins with PT (corresponding to the  campaign): two from hall  (PT , see Schmidt : ,
table ‘Tablets with Impressions of Seal No. 0; PT , see Schmidt : , table ‘Tablets with Impressions of Seal
No. 0), one from room  (PT , see Schmidt : , fn. ); the field number of the fourth tablet, PT  (=
PT ), reminds of PT  (Schmidt , pl. , no. ), PT  (idem, pl. , no. ), PT  (idem, , no. ),
PT e (idem, , no. ), all from room . Against my hypothesis, see Schmidt : : ‘The fragments are
not well enough preserved to show whether the tablets differed in text or seal from those uncovered in the archive
room () and in other units of the Treasury’. Unfortunately, no field number is available for the tablets published in
Cameron  and  (as remarked in Cameron : , and Cameron : ).
17 Cameron : , no. .
18 Schmidt :  provided only the thickness.
19 Size according to the silver exemplar found under the wall at the North-East corner of the main hall as given

by Curtis & Razmjou in Curtis & Tallis : ; ×cm is the measure in Schmidt : .
20 Vallat : .
21 Roaf : –. See also the discussion in Bivar : –, with further references; for some evidence

from the Apadana of Persepolis, see Schmidt : .

22 Schmidt : .
23 The number of pieces is deduced from the photographs in Schmidt , pl.  A–C.
 the pseudo-sealing of the persepolis bronze plaque 

plaque on a wall,24 the visible face should have been the obverse, i.e. the face where the lobes
protruded. In the photograph of the find-spot,25 cracks are visible at the junctures of the tabs;
in subsequent photographs, the tabs are missing.26
PBP lay on the floor with the obverse face up, so the reverse, protected by the floor, suffered
less damage. The beginning of the text on the obverse, at least a dozen lines, is wholly lost under
a patina of oxidation;27 the text of the following  lines is partly visible among several encrusted
areas. The reverse is nearly entirely preserved with  lines of text, followed by  blank lines,
then a final line of text; several stained areas hamper the reading.28 After the last isolated line, the
image of a sealing, turned upside down, follows. Each line of text is framed by a horizontal rule.
In some way PBP recalls an administrative tablet from Persepolis: we have the holes on the
short side edge, the wire in place of the string and the sealing parallel to the short side. The text
direction is different, being parallel to the short side and not to the long side as in the small
tablets, but similar to that of the so-called account tablets.29

. Brief Overview on Previous Treatments

In March , PBP was sent to the Oriental Institute of Chicago as a loan from the Iranian
Ministry of Education for ‘cleaning, study and publication purpose’.30 There the plaque was
chemically cleaned.31 In May , PBP was returned to the Imperial Legation of Iran in the
USA. PBP is now in the Inscription Department of the National Museum of Iran in Tehran.32
The first passing reference to the discovery is given by Schmidt in Persepolis I.33 A first
study on PBP was published by Schmidt in Persepolis II, embedding a summary of the text
by Cameron;34 according to Cameron ‘from the signs upon it, the bronze plaque may be dated
approximately anywhere between the early years of the th century bc and the third quarter
of the th’. The text is still unpublished; a synopsis can be found in the Survey of Neo-Elamite
History published by Waters in .35
Stolper, Steve, Vallat and Tavernier all placed the text in the historical framework of the
late Neo-Elamite period. According to Stolper, PBP ‘is, as its paleography suggests, approx-
imately contemporary with the late Elamite tablets from Susa’,36 i.e. the so-called Acropole
tablets, administrative documents mainly dealing with clothing and weapons.37 Steve, whose

24 ‘A looped wire for suspension’ is mentioned in Schmidt : .

25 Schmidt : , fig.  E. The two protruding lobes are not visible in the photograph and must have been
removed.
26 E.g. Schmidt , pls.  D and .
27 Cameron (apud Schmidt : ) believed that here there was ‘a frieze of seal patterns (or of a seal pattern)—

indeterminate but for the probability that they were different from the seal shown on the reverse’.
28 See the photographs in Schmidt , pl. . The original lines were  on the obverse and  on the reverse

according to Hinz & Koch : , s.v. ‘Oruru’.

29 See for example PF  (photograph in Henkelman : , fig. .).
30 Loan-in folders LI-. I owe the information on the loan to the kindness of Matthew W. Stolper (The Oriental

Institute of the University of Chicago) and Helen McDonald (The Oriental Institute Museum). See also Schmidt
: , fn. .
31 Schmidt : .
32 Museum number B.K. . I am thankful to Daryoush Akbarzadeh (Head of the Inscription Department)

and to Simin Piran (National Museum of Iran) for providing information and new photographs.
33 Schmidt : .
34 Schmidt : – and pls. –.
35 Waters : –.
36 Stolper in Carter & Stolper : .
37 Published by Scheil in MDP IX and re-edited in Jusifov . See the general presentation in Waters :

–.
 gian pietro basello

unpublished autograph and transliteration circulate among scholars,38 ascribed PBP to N III A
(ca. –bc), i.e. slightly before the last Neo-Elamite phase (N III B, ca. –) to which
Steve dated the Acropole tablets.39 Vallat, followed by Potts, supported Stolper’s chronology,
dating both the Acropole tablets and PBP to N III B (ca. –ca. ), afterwards subsumed
by Steve, Vallat & Gasche in an undivided N III period.40 Tavernier, on the ground of broken
writings in PBP, reached similar results, even if he slightly shifted back the dating of the Acrop-
ole tablets, suggesting ca. –ca. .41 Henkelman, in his detailed study on the Elamite and
Iranian acculturation, used repeatedly the textual evidence of PBP in connection with the Perse-
polis Fortification tablets, suggesting a range ‘– or even – bc’.42 A higher date (th
or th century bc) was advanced by Hinz & Koch.43
As to the function of PBP, according to Cameron ‘it was a memorial plaque recounting
the founding of a temple, the fixing of a temple offering or of temple offerings’;44 also Waters
referred to PBP as ‘Ururu’s dedicatory inscription’.45 According to Steve, Vallat & Gasche, the
text is a royal charter which assigned some lands, estates and vineyards left by a man named
Ururu, without filiation, to another Ururu, son of Šadanunu, as a consequence of the dis-
appearance of Addaten, son of Huban-ahpi; the properties are located in two distinct areas,
Hamun and Gisat; the text is closed by the mention of seven gods: Šašum, Napiriša, Dilbat,
Laliya, Nahhunte, Šati and Nanna.46 Henkelman linked PBP to the existence of a ‘sanctuary’
in Gisat where offerings are attested with continuity from Neo-Elamite to Achaemenid peri-
ods.47

. The Persepolis Bronze Plaque and the Persepolis Fortification Tablets

The main connection between PBP and the Persepolis Fortification tablets is provided by
PF .48 In this tablet we find a priest (dšá-tin), Ururu (HALu-ru-ru) by name, who received
 sheep/goats instead of  quarts of barley ( ŠE BARMEŠ) having made (hu-ut-táš-tá)
something related to a divine service (dna-ap-na, apparently ‘of god’,49 perhaps ‘divine’, i.e. an
offering). The text ends with the location and the date: Gisat (AŠgi-sa-at) in the nd year (of
Darius I). The spellings of the names Ururu and Gisat are the same attested in PBP.

38 As indicated in Vallat a: , fn. ; see also Steve : , fn. : ‘nous avons collationnée [la plaque]

à deux reprises ( et ) au Musée de Téhéran’. Henkelman : , fn. , attributed the unpublished
transliteration to Steve & Reiner.
39 Steve : , sub ‘N III A’, no. , and ‘N III B’, no. . This opinion was already expressed in Steve : ; cf.

Miroschedji : , fn. : ‘Sans donner ses raisons, Steve :  et , date la plaque entre “ et ” et la
juge “antérieure de quelques décennies” aux tablettes de l’Acropole’.
40 Vallat a:  (N III B), Vallat b: ; Potts :  and ; Steve & al. –, cols. –

(N III). See also Vallat .

41 Compare Tavernier : , and the final overview on p. .
42 Henkelman : .
43 Hinz & Koch : , s.v. ‘Oruru’.
44 Cameron apud Schmidt : .
45 Waters : .
46 Steve & al. –, col. : ‘Ce document … se prèsente comme une charte royale qui redistribue des

biens fonciers (territoires, immeubles et vignobles) d’un nommé Ururu (sans filiation) à un autre Ururu, fils de
Šadanunu, à la suite de la disparition d’ un certain Addaten, fils de Huban-ahpi, dans deux régions distinctes … le
pays de Hamun … et … le pays de Gisat … Cette charte se termine par des imprécations envers les contrevenants et
des invocations à différentes divinités: Šašum, Napiriša, Dilbat, Laliya, Nahhunte, Šati et Nanna’. On Dilbat and Šati,
see also Vallat b: .
47 Henkelman : : ‘[PBP] mentions offerings at a sanctuary in Gisat’.
48 See an abstract of the text in Henkelman : .
49 Probably to be interpreted as a plural form, i.e. ‘for the gods’ (Henkelman : ).
 the pseudo-sealing of the persepolis bronze plaque 

A priest named Ururu (spelled as in PBP and PF ) is attested also in PF , dated to the
th year (of Darius I); unfortunately, no place is mentioned in the text and the sealings are not
identifiable.
Gisat is mentioned in seven other published tablets and in one tablet now available online
thanks to the Persepolis Fortification Archive Project;50 the spelling is the same of PF  except
for AŠgi-sa!-ut in PF :. Gisat is connected to Hidali in PF  and PF ; on this basis, it
is placed on the boundary between Elam (i.e., roughly, Khuzestân) and Persia (Fârs) by Koch.51
Some similarities can be found between the Fortification tablets and PBP in lexicon52 and
onomastics. The verbal form tumanra (du-man-ra) is attested only in PBP and the Persepolis
tablets.53 Šati, mentioned as god in PBP, is attested as theophoric element in the onomastics
of Persepolis.54 In particular, the addressee of one of the texts mentioning Gisat (PF :),
Šati-Dudu, is a man whose theophoric name is composed with Šati. Another theophoric name
with the name of the god Šati, Šati-hupiti, father of Huban-šuturuk, is known from PBP. Also
Huban-ahpi is attested in other sources. It occurs in the inscription of seal PFS * impressed on
 Persepolis Fortification tablets for a total of  impressions.55 In the inscription, Huban-ahpi
is son of Šati-hupan,56 again a theophoric name with the god Šati. The spelling, hu-pan-a-ah-pi,
is the same attested on PBP. In seal PFS *, impressed on  Persepolis Fortification tablets,57
Huban-ahpi is the father of a woman Šeraš.58 Since the name Huban-ahpi is not otherwise
attested at Persepolis, both Hinz and Koch suggested that seal PFS * was an heirloom from
the Neo-Elamite period.59

. The Persepolis Bronze Plaque and the Acropole Tablets

Actually, Huban-ahpi is well attested in the tablets from the Acropole of Susa. According to
Steve, Vallat & Gasche, several elements link PBP to the Acropole tablets from Susa.60 First of
all, the onomastics: tablet MDP IX  mentions Ururu and Huban-ahpi. Tablet MDP IX 
mentions Addaten, the woman Ammaten and Huban-ahpi [DUM]U sunki ‘son of the king’;
all these names are attested also on PBP where Addaten is the son of Huban-ahpi.61 In sum,
Huban-ahpi is attested  times on the Acropole tablets.62

50 PF  (rd year),  (nd),  (st?), :– (no date), :– (nd), : (th), : (th);

PF-NN :– (th). Gisat is rendered as ke-sa-at (ke for ké) in the simplified transliteration by Hallock. Gisat is
often associated with Mamannuwiš (in PF , , ,  and ) and with seal PFS  (on the left edge of
PF , , , ) and  (on the reverse of the same tablets).
51 Koch : –. See Vallat : , s.v. ‘Gisat’, with further references. See also Tavernier : , and

Álvarez-Mon : –.

52 Waters : , fn. : ‘There are numerous vocabulary parallels with the Persepolis texts’.
53 PBP:rev.H and rev.H = PBP:S and S. Persepolis Fortification tablets: occurrences in Hinz & Koch

: , s.v. du-man-ra. The spelling du-man-ri is attested in the Persepolis Treasury tablets; see Hinz & Koch
: , s.v. du-man-ri.
54 E.g. in PFS * = Garrison & Root , no.  (see also Vallat : ). See Zadok : –, no. ,

and Mayrhofer : , nos. .– (except no. .).

55 Garrison & Root , no. . The tablets are PF –, , , –, –, , ,  and

 (see Garrison & Root : –).

56 |1 [DIŠ] d hu-pan- a - |2 ah- |3 pi |4 DUMU d |5 ša- ti -hu-pan (Jones apud Garrison & Root : ).
57 PF – and – (see Hallock : ).
58 Jones apud Garrison & Root : .
59 Garrison & Root : , with further references.
60 Steve & al. –, cols. –.
61 Addaten: PBP:S (= PBP:H) and S (= H). Ammaten: PBP:S (= PBP:H). Huban-ahpi: PBP:S (=

PBP:H) and S.

62 MDP IX  (two times), , , , , , , , , , .
 gian pietro basello

Among the seven gods evoked in PBP, two are known also from the Acropole tablets. The god
Šati is attested only in PBP and in a dozen of Acropole tablets; Šati is attested also as theophoric
element in the onomastics of the Acropole tablets (as it is in that of PBP) (Table ). The goddess
Laliya is mentioned only in one Acropole tablet, whereas she is attested as theophoric element
in a frequently recurring anthroponym, Lalintaš (Table ).63 Another goddess mentioned on
PBP, Dilbat,64 is attested in some Neo-Elamite texts: an inscription of Šilhak-Inšušinak II who
dedicated to her a temple in Susa,65 an inscription of Hanni at Izeh66 and on the vessels of
Samati.67 Finally, the title GAL.E.GALMEŠ is attested at least  times in the Acropole tablets;68
in PBP, GAL.E.GALMEŠ seems to be written on the last isolated line of text of the reverse.69

Table . Occurrences of the gods Lali and Šati as divine names and theophoric elements in personal
names on the Acropole tablets.
transliteration and occurrences in
god(dess) name type context MDP IX total occ.
d
Lali Laliya DN f la-li-ya  
BE
Lalintaš PN la-li-in-taš , ,  ( occs.), , 
, , ,  (?)
BE
Lali-sunki PN la-li-EŠŠANA , ,  
d
Šati Šati DN šá-ti BE.dšá-ti- , , :– 
du-du-pè-ra
d
šá-ti , :rev., , 
, , ,  (?),
 (?), , :
Šati … DN (?) dšá-[ti …]  
BE.d
Šati PN šá-ti ,  
Šati PN (?) [d]šá-ti SAL.d[…]  
BE.d
Šati (?) PN šá-ti al-[…]  
BE.d
Šati … PN šá-ti-[…] , ,  
BE.d
Šati-dudu PN šá-ti-du-du , , :– 
(BE).d
Šati-duš PN šá-ti-du-iš  (?), : 
BE.d
Šati-hupiti PN šá-ti-hu-pi-ti  
BE.d
Šati-kitin PN šá-ti-ki-tin  
BE.d
Šati-šilhak PN šá-ti-šil-ha-ak  
BE.d
Šati-te … PN šá-ti-te(?)-[…]  
BE
Šati / Humban Šatin-humban PN šá-tin-dh[u]-ban  
BE.d
Šati / Humban Šati-humban PN šá-ti-hu-ban , :, ,  
BE.d
Šati / Napiriša Šati-napiriša (?) PN šá-ti-DINGIR.GAL  ( occs.) 

An additional connection is provided by the dating formula on PBP:rev., ITI ra-hal UD-
ma, which follows the usual pattern of the Acropole tablets; this pattern is well differentiated
from the formula in the Persepolis tablets where nan ‘day’ is used in place of the logogram UD;
moreover, nan is always followed by the day number while in the Acropole texts the day number

63 See also the occurrence of Lalintaš on the rhyton published in Vallat b.
64 On Dilbat, see Vallat a: –, Vallat : –, and Henkelman b: , fn. .
65 EKI . See also Waters : .
66 EKI ; text and translation also in Hinz .
67 Henkelman b: –. See also Steve & al. –, cols. –, with further references (especially

Vallat b and ).

68 MDP IX , , , , , , ; it has to be restored in MDP IX  according to Hinz & Koch : ,

s.v. hw.ráb.E.GAL.lg; see also Hinz & Koch : , s.v. ráb.lg. Another occurrence is in Nin . See also Steve
: –, and Henkelman : , fn. .
69 Steve : , and Waters : .
 the pseudo-sealing of the persepolis bronze plaque 

is nearly always lacking, as is lacking in PBP. The usual translation is ‘a day in the month MN’;
Grazia Giovinazzo suggested that this could be not ‘a day’ but ‘the day’, i.e. the day established
for offerings or the market day. The month-name is also interesting, since rahal is attested both
in the Acropole and in the Persepolis Fortification tablets; in both corpora this month-name
seems not to be part of the usual sets of month-names.70
A general connection between PBP and the Acropole tablets is provided by the palaeography.
Already Cameron wrote that there is little difference between the PBP and the tablets, though
hinting at a closer link of PBP to the inscriptions of Šutruru from Susa and Hanni from Izeh.71

. The Use of the Sign GAM as Personal Determinative

Another peculiarity of PBP is the use of the sign GAM as personal determinative. Leaving
aside the occurrences in the Elamite administrative tablets from Tall-e Malyân, where Matthew
Stolper pointed out the usage of GAM at the juncture of an anthroponym divided in two lines,72
the rare occurrences of this sign as a true personal determinative in Elamite are chronologically
closer to the Achaemenid period.73 In a royal inscription of Hallutaš-Inšušinak (ca.  bc),
GAM precedes the name of the king.74 In one of the so-called Nineveh letters, before the two
occurrences of the logogram EŠŠANA ‘king’, a GAM and then a debated BE is to be found.75
GAM occurs several times in another isolated Elamite text, the omen tablet found in Susa.76
In PBP, GAM occurs several times, not only before anthroponyms,77 but also before a social
group designation (šá-al-hu-ip ‘nobles’, as the Achaemenids are styled in DB/AE),78 before a
denomination of temple personnel (pu-hu dzí-ya-nu-ip, the young apprentices of the temple)79
and before KÙ.BABBAR ‘silver’.80
An apparently anomalous occurrence of GAM can be dated without doubt to the Achaeme-
nid period, in the Elamite text of the short inscription CMc in Palace P at Pasargadae. The
inscription is engraved along a fold of the garment of the king in two bas-reliefs facing one
another in a doorway.81 GAM is attested before the logogram for ‘king’ and before the word
[h]a-ak-ka4-man-nu-ši-ya-ra ‘Achaemenid’.
So the use of GAM is not limited to Neo-Elamite documents and one should emphasize that
it cannot be used as a distinctive feature for dating purposes. Its presence both in Susa and in
Pasargadae is puzzling and could be related to the origin of the Old Persian word-divider, since
the personal determinative probably had functions very similar to those of the word-divider;82
not by chance determinatives are not attested in scripts using word-dividers systematically.

70 Basello , especially pp.  and .

71 Cameron apud Schmidt : . The inscriptions of Šutruru and Hanni are published as EKI  and EKI –
.
72 As a hyphen, but at the beginning of the new line. Stolper a: . See also Steve  and Steve : ,

nos. –. In Elamite script, also in monumental inscriptions like the royal Achaemenid ones, it is usual to break
words at the end of a line.
73 See especially Steve  and Vallat : –.
74 EKI :; see Steve . The indexing no. : in König , pl. , and the reference to EKI : in Steve

: , no. , are misprints.

75 Nin :–.
76 Published in Scheil .
77 E.g. PBP:S and S (mentioned in Steve : , no. ), S and S (Steve ).
78 PBP:S (= PBP:H). DIŠšá-lu-ú-ut in DB/AE col. I:.
79 PBP:S (= PBP:rev.H). Giovinazzo  distinguished three categories of puhu ‘young’: the apprentices, the

servants and the assistants (especially of travellers).

80 PBP:S (= PBP:rev.H).
81 Photograph in Stronach , pl. b. Steve  and : , no. ; also Vallat : –.
82 Basello, in print.
 gian pietro basello

. The Pseudo-Sealing on the PBP

I mentioned above the image of a sealing on the reverse of PBP. It is a reproduction of a

sealing, engraved on the bronze plaque.83 The pseudo-sealing is upside down, i.e. oriented
in the same manner of the text on the obverse. It was drawn using the full width of the
plaque so to have four instances of the figuration alternating with three instances of a five-
line panelled inscription. According to the transliteration by Steve, the inscription is to be read
as follows:
|1 EŠŠANA DIŠ.d|2hu- ban -šu-|3tur- uk? DUMU |4 dšá-ti-hu-|5pi-ti-na84
i.e. ‘king Huban-šuturuk son of Šati-hupiti’. I added in the transliteration the half square
brackets and the interrogation mark, since the second sign in line , commonly read uk, is
of difficult interpretation: only a nu or man (i.e. two converging horizontal wedges) in one of
the forms attested on the Acropole tablets (but not in PBP itself) is visible.
The figurative part represents two symmetrical rampant animals, one facing the other. One
foreleg of each animal is raised, touching the raised foreleg of the facing animal; the other
foreleg is flexed in front of the chest. The animals are labelled as ‘perhaps lions’ by Schmidt.85
However, the slimness of their bodies, necks and heads points rather to horses, as the styling of
the mane and the descending tail (the lions are usually styled with an upright curved tail86) do.
The pseudo-sealing, being on the reverse (i.e. not on the face where the lobes protruded if the
plaque was hung on a wall), could not be an ornamentation and should therefore have had the
function of a real sealing.87
Starting from the assumption that we are facing a reproduction of a real sealing, I looked for
comparisons among seals and sealings. I found one in a sealing attested on a tablet from the
Acropole of Susa and published by Amiet in .88 The seal has been applied three times: in
the two impressions below the last line of text, as long as the width of the tablet and one above
the other, the seal has been rolled for at least two complete turns; one impression of the animals
is on the left edge. Only the latter impression is shown in the photograph published by Amiet;
the drawing of Amiet is based mainly on the impressions on the face of the tablet, probably the
upper one, left part, for the animals and the lower one, right part, for the inscribed panel.

83 Cameron apud Schmidt : : ‘representations of the impression of a cylinder seal’; Miroschedji : ,

fn. : ‘… pas à proprement parler un sceau-cylindre mais—cas unique dans l’histoire de la glyptique élamite—sa
reproduction gravée sur une plaque de bronze découverte à Persépolis’; Steve : : ‘… gravé au burin sur une
imitation de sceau-cylindre’; Waters : : ‘imprint of a cylinder seal with inscription’. Also Henkelman : ,
fn. .
84 Also in Waters : . Cameron (apud Schmidt : ), who offered only a translation, omitted the

logogram ‘king’. The sign GAM before the name of the king, restored by Hinz & Kock : , s.v. [GAM].hu-
ban.šu-tur-uk, is probably to be substituted with DIŠ, since it would be unusual to use GAM in a sealing; see
Giovinazzo  for a peculiar use of DIŠ in administrative texts. Vallat’s translation is given in Miroschedji :
, fn. .
85 Schmidt : .
86 E.g. in Amiet , nos.  and .
87 Miroschedji : , fn. : ‘Le texte de cette plaque est apparemment une charte dont les clauses sont

garanties par le roi, qui devait donc y apposer son sceau’.

88 Amiet : – and , pl. III, no. . The similarity was already noted (without a particular emphasis) in

Miroschedji : , fn. : ‘Le décor de ce dernier est typique de la glyptique élamite du VIe siècle (cf. Amiet …
no. – et plusieurs ex. parmi les empreintes inédites des tablettes de Persépolis)’. In the discussion following
my lecture at the conference in Ghent, Javier Álvarez-Mon supported this comparison, which has been already
emphasized in his forthcoming work on the Arjân tomb; see now Álvarez-Mon : – and pl. e–f. See
Garrison  for a reassessment of the stylistic evidence of the sealings from Susa.
 the pseudo-sealing of the persepolis bronze plaque 

The figurative section is nearly the same of the PBP. The forelegs, both the touching and the
flexed ones, are clearly visible, even if the ends were not drawn by Amiet; the tail is thick and
falls vertically at the bottom of the back. Note also the hoof of the rear leg nearly touching the
end of the tail. The head is stylized in the same way, and the snout has the same orientation.89
The proof that this very seal was imitated in the PBP pseudo-sealing should come from the
comparison of the inscribed panels. Unfortunately, the seal was lightly pressed and the inscribed
panel can be barely seen. Amiet could see only a few wedges, parts of three or four signs. In the
first line, I can see the lower half of a possible logogram EŠŠANA ‘king’; in the second line,
it seems possible to see two nu signs; traces of a logogram DUMU are cut by a slit, probably
the impression of a nail; however, it is likely that I am influenced by the mention of a king
Ummanunu in the text of the tablet.
The tablet bearing this seal, MDP IX ,90 is a special tablet. First of all, is one of the few
tablets from the Acropole mentioning a king.91 So it is possible that the impressed seal was
a royal one. The name of the king mentioned in the text is Ummanunu (BEum-ma-nu-nu),92
attested only here as a king, just as Huban-šuturuk, the king named in the pseudo-sealing of
PBP, is otherwise unknown. A theophoric compound name implying the god Šati (BE.dšá-ti-
[x-x-x(-x)]) is also attested in the tablet.93 Moreover, this tablet is exceptional in the Acropole
corpus, as it is the only tablet mentioning a measure of capacity while the other tablets deal
with clothing, weapons and other tools.94 Even the administrative formulae are different from
the other Acropole tablets.
If it were possible to prove the identity of the sealings on PBP and MDP IX , it would
be likely that PBP also originated in Susa as a royal chart or grant regarding some estates
in the East, and perhaps kept there. Probably the emphasis on Gisat may be reduced, on
account of the other toponyms attested on PBP.95 Maybe PBP should not be a proof of a
fragmentation of power in the last Neo-Elamite phase anymore.96 On the contrary, PBP could be
a witness of the existence of a regional system were Susa was still the focal point, as attested also
from the mention of people from other places (e.g. Samati and Ayapir) in the Acropole tablets,
and notwithstanding the localization of Gisat, probably closer to Persepolis than to Susa.

. Why Was PBP in the Persepolis Treasury?

As mentioned above, PBP was found in a subsidiary room of the Treasury together with
various other objects. Assuming that no-one would bring objects into the building while looting
it, one should assume that these findings were already in the room or, at most, in the two
interconnected rooms. According to Erich F. Schmidt, the uninscribed tablets found in the

89 Miroschedji : , provided the following general description of the sealings on the Acropole tablets: ‘Les

animaux y sont représentés suivant des conventions particulières …: ils ont une silhouette efflanquée, avec des
membre grèles et un ventre excessivement allongé, mais avec une cuisse puissante et en fort relief; la ligne de la
tête, du cou et du poitrail cambré forme un “S” ’.
90 Louvre Museum number Sb .
91 See Waters : –, and Tavernier : –, for a discussion on the references to kings in the Acropole

tablets.
92 MDP IX :; see also the transliteration in Jusifov : , no. . The sign ma in Ummanunu was read

ba by Scheil; actually it is not ba but it appears to be slightly different from the ma on line .
93 MDP IX :.
94 Basello : –.
95 See Vallat , s.v. Asampi, Babili (TIN?.TAR), Bahar, Hamun, Irkume?, Kummama, Šumurtan-Duri, Tartin,

Tartin-Ammak?, Udazammin, Udman, Zanu and Zippa.

96 On late Neo-Elamite political fragmentation, cf. Henkelman a, cols. –, with further references.
 gian pietro basello

same room might be attached to valuable objects deposited in the Treasury.97 Starting from this
interpretive context, that of the scene of a secure storehouse violated by looting, Schmidt tried
to explain the presence of PBP as ‘apparently a trophy of war’;98 George G. Cameron reinforced
this view:
it is unlikely that this plaque was inscribed at Persepolis, a new city founded by Darius. Presumably
it was written some generations earlier in some other (Elamite) center and brought to the capital
as a war trophy or the like.99
This judgment was issued in the context of the other findings from the Treasury such as the
inscribed Neo-Assyrian and Neo-Babylonian eyestones, seals and beads from Mesopotamia100
which, as Potts remarked, ‘certainly … must have arrived long after they have originally been
manufactured’.101 Recently Razmjou gave a slightly different perspective, defining the Treasury
as ‘an early kind of museum’.102
Henkelman suggested another possibility:
the late Neo-Elamite Persepolis Bronze Plaque, found in the Treasury at Persepolis, was in fact kept
there as retroact and had a relevance for the current rights and obligations of the Gisat sanctuary
and its administrators.103
In the light of the connection between PBP and the tablet MDP IX , it seems possible to
hypothesize that PBP was written in Susa, as a grant related to a far provincial centre in the
East, maybe Hamun or Gisat. The polarization created by the establishing of Persepolis as the
chief administrative centre in the area once called Anshan, may account for the “migration” of
the text from Susa to Persepolis. The administrative link between Gisat and Persepolis seems to
be proven by PF . As remarked by Henkelman, PBP was kept not for its material value but
because it was a still valid document. The discovery of PBP in Persepolis may attest the shift of
administrative control from Susa to Persepolis. It is also possible that PBP had been transferred
to the previously pre-eminent centre of the area, maybe Matezziš.104
While this interpretation remains a tentative attempt, it is hoped that a thorough study of
the “exceptional” PBP and the publication of its text could shift our comprehension of the th
century bc in the South-Western Iran a little bit further.

Acknowledgements

This research benefited from discussions held at the meetings of the Italo-Iranian Project
DARIOSH (Digital Achaemenid Royal Inscription Open Schema Hypertext, partly financed
by the Italian Ministry of Education, under PRIN contracts  and ZKPPSM),
directed by Adriano V. Rossi (‘L’Orientale’ University, Naples). Ela Filippone (University of
Tuscia, Viterbo) and Grazia Giovinazzo (‘L’ Orientale’) are members of the Project.

97 Schmidt : : ‘When the building was despoiled by Alexander, these labels were torn from the objects to
which they were attached, or the labeled objects were left with things found useless, to be burned during subsequent
conflagration’. See also the recent observations on the uninscribed tablets in Garrison : –.
98 Schmidt : .
99 Cameron apud Schmidt : .
100 See the section ‘Votive objects from Mesopotamia’ in Schmidt : –.
101 Potts : .
102 Razmjou : –, referring to PBP as ‘a Neo-Elamite inscribed plaque’ (p. ).
103 Henkelman : , fn. .
104 On Matezziš, see Stolper b: –; Vallat : –, s.v. Matezziš, and Tavernier :  (§..),

 (§ ..) and – (§§ ..–), with further references.

 the pseudo-sealing of the persepolis bronze plaque 

I would like to thank the following scholars for the following reasons, but especially for
making the research work on an ancient item a lively experience: Daryoush Akbarzadeh (head
of the Inscription Department, National Museum of Iran, Tehran; now director of the Museum)
and Simin Piran (keeper, National Museum of Iran) for looking after PBP; Javier Álvarez-Mon
(University of Sydney) for sharing his ideas and forthcoming publications; Mark B. Garrison
(Trinity University, San Antonio, Texas) for putting at my disposal his experience with seals and
sealings; Helen McDonald (The Oriental Institute of Chicago) for providing information on
the loan of PBP to the Oriental Institute; Grazia Giovinazzo and Adriano V. Rossi (‘L’ Orientale’
University, Naples) for reading attentively this text in all its phases; Matthew W. Stolper (The
Oriental Institute of Chicago) for reading and commenting the text; Katrien De Graef (Ghent
University) and Jan Tavernier (Université catholique de Louvain) for reuniting so many Elamite
scholars in Ghent and for bringing to publication this contribution.

Abbreviations

DB/AE Grillot-Susini & al. .

DPh see Schweiger , vol. , pp. –, and Schmitt , pp.  and –, with further
bibliography and the Old Persian text.
DSaa Vallat ; see also Vallat .
DSz Vallat ; see also Vallat .
EKI König .
MDP I Morgan & al. .
MDP V Scheil .
MDP VII Morgan & al. .
MDP VIII Jéquier & al. .
MDP IX Scheil .
MDP XI Scheil .
Nin Weissbach  (autograph) and Hinz  (transliteration). See also Vallat b and
Waters , pp. –, with further references.
PBP Persepolis bronze plaque. The text is unpublished. See Schmidt , pp. –, Vallat
 for the toponyms (under the siglum ‘Hub Pe’ for Huban-šuturuk, Persepolis), Waters
, pp. –. Line numbers, differing among scholars, are marked with the following
sigla: C: Cameron apud Schmidt , pp. –; H: references in Hinz & Koch 
(under the siglum ‘Oruru’); S: unpublished transliteration by Steve (followed also in Vallat
).
PF Hallock .
PF-NN unpublished Persepolis Fortification tablets edited by R.T. Hallock. Online access: <http://
ochre.lib.uchicago.edu/PFA_Online>.
PFS seals on the Persepolis Fortification tablets, partially published in Garrison & Root . See
the list in Hallock , pp. –.
PT Cameron .

Bibliography

Álvarez-Mon, J. (): The Arjān Tomb. At the Crossroads of the Elamite and the Persian Empire (Acta
Iranica, ), Leuven: Peeters.
Amiet, P. (): “La glyptique de la fin de l’Elam”, Arts Asiatiques , –.
Basello, G.P. (): “Elam and Babylonia: The Evidence of the Calendars”, in A.C.D. Panaino & G. Pet-
tinato (eds.), Ideologies as Intercultural Phenomena (Melammu Symposia ), Milano, –.
———. (): “Finding a name for an archaeological finding: the sit-šamši from Šuš”, <www.elamit
.net>.
———. (): “Elamite as Administrative Language: from Susa to Persepolis”, in J. Álvarez-Mon &
M.B. Garrison (eds.), Elam and Persia, Winona Lake, –.
 gian pietro basello

———. (in print): ‘The Sign GAM in Elamite and the Old Persian Word-divider’, in R. Schmitt, A.V. Rossi,
A.C.D. Panaino, E. Filippone & V. Sadovski (eds.), Achaimenidika, I, Wien: Verlag der Österreichische
Akademie der Wissenschaft.
Bivar, A.D.H. (): “Achaemenid Coins, Weights and Measures”, in I. Gershevitch (ed.), The Cambridge
History of Iran, vol. : The Median and Achaemenian Periods, Cambridge, –.
Cameron, G.G. (): Persepolis Treasury Tablets (Oriental Institute Publications, ), Chicago.
———. (): “Persepolis Treasury Tablets Old and New”, Journal of Near Eastern Studies , –.
———. (): “New Tablets from the Persepolis Treasury”, Journal of Near Eastern Studies , –.
Carter, E. & Stolper, M.W. (): Elam. Surveys of Political History and Archaeology, Berkeley.
Curtis, J.E. & Tallis, N. (): Forgotten Empire. The World of Ancient Persia, Berkeley & Los Angeles.
Garrison, M.B. (): “The ‘Late Neo-Elamite Glyptic Style’: A Persepective from Fars”, Bulletin of the
Asia Institute N.S. , –.
———. (): “The uninscribed tablets from the Fortification archive: a preliminary analysis”, in P. Briant,
W. Henkelman, M.W. Stolper (eds.), L’archive des Fortifications de Persépolis. État des questions et
perspectives de recherches (Persika, ), Paris, –.
Garrison, M.B. & Root, M.C. (): Seals on the Persepolis Fortification Tablets. Volume I: Images of
Heroic Encounter (Oriental Institute Publications, ), Chicago.
Gautier, J.E. (): “Le “sit šamši” de Šilhak In Šušinak”, Recueil de travaux relatifs à la philologie et à
l’archéologie égyptiennes et assyriennes ,˘ –.
Giovinazzo, G. (): ““NP hiše” dans les textes achéménides”, Annali dell’Istituto Universitario Orien-
tale , –.
———. (): “I ‘puhu’ nei testi di Persepoli. Nuove interpretazioni”, Annali dell’Istituto Universitario
Orientale , –.
Grillot-Susini, F., Herrenschmidt, C. & Malbran-Labat, F. (): “La version élamite de la trilingue de
Behistun: une nouvelle lecture”, Journal Asiatique , –.
Hallock, R.T. (): Persepolis Fortification Tablets (Oriental Institute Publications ), Chicago.
Harper, P.O., Aruz, J. & Tallon, F. (): The Royal City of Susa. Ancient Near Eastern Treasures in the
Louvre, New York.
Henkelman, Wouter (a): “Defining Neo-Elamite History”, Bibliotheca Orientalis , –.
———. (b): “Persians, Medes and Elamites: acculturation in the Neo-Elamite period”, in G.B. Lan-
franchi, M. Roaf & R. Rollinger (eds.), Continuity of Empire (?). Assyria, Media, Persia (History of the
Ancient Near East. Monographs, ), Padova, –.
———. (): The Other Gods Who Are. Studies in Elamite-Iranian acculturation based on the Persepolis
Fortification Texts (Achaemenid History, ), Leiden.
Henkelman, W., Jones, C.E., Stolper, M.W. (): “Clay Tags with Achaemenid Seal Impressions in the
Dutch Institute of the Near East (NINO) and Elsewhere”, ARTA, ..
Hinz, W. (): “Die elamischen Inschriften des Hanne”, in W.B. Henning & E. Yarshater (eds.), A
Locust’s Leg. Studies in honour of S.H. Taqizadeh, London, –.
———. (): “Zu den elamischen Briefen aus Ninive”, in L. de Meyer, H. Gasche & F. Vallat (eds.),
Fragmenta Historiae Elamicae. Mélanges offerts à M.-J. Steve, Paris, –.
Hinz, W. & Koch, H. (): Elamisches Wörterbuch (Archäologische Mitteilungen aus Iran. Ergänzungs-
band ), Berlin.
Jéquier, G., de Morgan, J., Gautier, J.E., Lampre, G., Jouannin, A., de la Fuye, A., de Morgan, H. ():
Recherches archéologiques. Troisième série (MDP ), Paris.
Jusifov, Ju.B. (): “Ėlamskie chozjajstvennye dokumenty iz Suz”, Vestnik Drevnej Istorii /, –
and /, –, Moskva (in Russian).
Koch, H. (): Verwaltung und Wirtschaft im persischen Kernland zur Zeit der Achämeniden (Beihefte
zum Tübinger Atlas des vorderen Orients, Reihe B ), Wiesbaden.
König, F.W. (): Die elamischen Königsinschriften (AfO. Beih. ), Berlin & Graz.
Mayrhofer, M. (): Onomastica Persepolitana. Das altiranische Namengut der Persepolis-Täfelchen
(SÖAW ), Wien.
Miroschedji, P. de (): “Notes sur la glyptique de la fin de l’Élam”, Revue d’Assyriologie et d’Archéologie
Orientale , –.
———. (): “La fin du royaume d’Anšan et de Suse et la naissance de l’Empire perse”, Zeitschrift für
Assyriologie und vorderasiatische Archäologie , –.
———. (): “La fin de l’Élam: essai d’analyse et d’interprétation”, Iranica Antiqua , –.
Morgan, J. de, Jéquier, G. & Lampre, G. (): Recherches archéologiques. Première série. Fouilles à Suse
en – et – (MDP ), Paris.
 the pseudo-sealing of the persepolis bronze plaque 

Morgan, J. de, Jéquier, G., de Mecquenem, R., Haussoullier, B. & Graadt Van Roggen, D.L. ():
Recherches archéologiques. Deuxième série (MDP ), Paris.
Potts, D.T. (): The Archaeology of Elam. Formation and Transformation of an Ancient Iranian State
(Cambridge World Archaeology), New York.
Razmjou, Sh. (): “Persepolis: A Reinterpretation of Palaces and Their Functions”, in J. Curtis &
S.J. Simpson (eds.), The World of Achaemenid Persia. History, Art and Society in Iran and the Ancient
Near East. Proceedings of a conference at the British Museum th September—st October ,
London & New York, –.
Roaf, M. (): “Persepolitan Metrology”, Iran , –.
Scheil, V. (): Textes élamites-anzanites. Deuxième série (MDP ), Paris.
———. (): Textes élamites-anzanites. Troisième série (MDP ), Paris.
———. (): Textes élamites-anzanites. Quatrième série (MDP ), Paris.
———. (): “Déchiffrement d’un document anzanite relatif aux présages”, Revue d’Assyriologie et
d’Archéologie Orientale , –.
Schmidt, E.F. (): Persepolis I. Structures, Reliefs, Inscriptions (Oriental Institute Publications, ),
Chicago.
———. (): Persepolis II. Contents of the Traesury and Other Discoveries (Oriental Institute Publica-
tions, ), Chicago.
Schmitt, R. (): Die altpersischen Inschriften der Achaimeniden. Editio minor mit deutscher Überset-
zung, Wiesbaden.
Schweiger, G. (): Kritische Neuedition der achaemenidischen Keilinschriften (in zwei Bänden); vol. ,
Textband. Transkribierter Text und Übersetzung; vol. , Katalog. Archäologische Gegebenheiten und
kritische Lesungen in Transliteration, Taimering.
Steve, M.-J. (): “La fin de l’Élam: a propos d’une empreinte de sceau-cylindre”, Studia Iranica ,
–.
———. (): “Le déterminatif masculin BE en néo-élamite et élamite-achéménide”, Nouvelles Assyri-
ologiques brèves et utilitaires, /, pp. –.
———. (): Syllabaire Elamite. Histoire et Paleographie (Civilisations du Proche-Orient: Serie II,
Philologie, ), Neuchâtel & Paris.
Steve, M.-J., Vallat, F. & Gasche, H. (–): “Suse A–F”, in J. Briend & M. Quesnel (eds.), Supplément
au dictionnaire de la Bible –, Paris, –.
Stolper, M.W. (a): Texts from Tall-i Malyan I. Elamite Administrative Texts (–) (Occasional
Publications of the Babylonian Fund, ), Philadelphia.
———. (b): “The Neo-Babylonian Text from the Persepolis Fortification”, Journal of Near Eastern
Studies , –, Chicago.
Stronach, D. (): Pasargadae. A report on the excavations conducted by the British Institute of Persian
Studies from  to , Oxford.
Tavernier, J. (): “Some Thoughts on Neo-Elamite Chronology”, ARTA, ..
———. (): Iranica in the Achaemenid Period (ca. –B.C.). Lexicon of Old Iranian Proper Names
and Loanwords, Attested in Non-Iranian Texts (OLA ), Leuven.
Vallat, F. (): “Deux nouvelles “chartes de fondation” d’un palais de Darius Ier à Suse”, Syria /–,
–.
———. (): “Deux inscriptions élamites de Darius Ier (DSf et DSz)”, Studia Iranica , –.
———. (): “Kidin-Hutran et l’époque néo-élamite”, Akkadica , –.
———. (): “Table accadienne de Darius Ier (DSaa)”, in L. de Meyer, H. Gasche & F. Vallat (eds.),
Fragmenta Historiae Elamicae. Mélanges offerts à M.-J. Steve, Paris, –.
———. (): Les noms géographiques des sources suso-élamites (Répertoire Géographique des Textes
Cunéiformes, ), Wiesbaden.
———. (): “Šutruk-Nahunte, Šutur-Nahunte et l’imbroglio néo-élamite”, Nouvelles Assyriologiques
brèves et utilitaires, /, Paris.
———. (a): “Nouvelle analyse des inscriptions néo-élamites”, in H. Gasche & B. Hrouda (eds.),
Collectanea Orientalia. Histoire, arts de l’espace et industrie de la terre. Etudes offertes en hommage
à Agnès Spycket, Neuchâtel & Paris, –.
———. (b): “Le royaume élamite de SAMATI”, Nouvelles Assyriologiques brèves et utilitaires, /,
–.
———. (a): “Elam, i. The History of Elam”, in E. Yarshater (ed.), Encyclopædia Iranica , –.
———. (b): “Le royaume elamite de Zamin et les “Lettres de Ninive””, Iranica Antiqua , –
.
 gian pietro basello

———. (a): “La ‘clergé’ élamite”, in S. Graziani (ed.), with the collaboration of M.C. Casaburi &
G. Lacerenza, Studi sul Vicino Oriente antico dedicati alla memoria di Luigi Cagni, , –,
Napoli.
———. (b): “Une inscription élamite sur un rhyton en argent à tête de bélier”, Akkadica , –.
———. (): “La Dame faite prisonnière à Babylone”, Akkadica /, –.
———. (): “L’inscription néo-élamite de Manaka[…]-untaš et l’emploi des déterminatifs à basse
époque”, in M. Bernardini & N.L. Tornesello (eds.), Scritti in onore di Giovanni M. D’Erme, Napoli,
–.
———. (): “Légendes élamites sur des cylindres de Persépolis”, Akkadica , –.
Waters, M.W. (): A Survey of Neo-Elamite History (State Archives of Assyria Studies, ), Helsinki.
Weissbach, F.H. (): “Susische Thontäfelchen”, Beiträge zur Assyriologie und semitischen Sprachwis-
senschaft , –.
Zadok, R. (): The Elamite Onomasticon (Supplemento agli Annali dell’Istituto Universitario Orien-
tale, ), Napoli.
the pseudo-sealing of the persepolis bronze plaque 

Figures

Fig. . The pseudo-sealing at the bottom of the

reverse of PBP (National Museum of Iran, Tehran).

Fig. . PBP, detail of the third figuration of the

pseudo-sealing (National Museum of Iran, Tehran).
 gian pietro basello

Fig. . MDP IX , second sealing below the text (courtesy by G. Giovinazzo).

Fig. . MDP IX , left part of the first sealing below the text (courtesy by G. Giovinazzo).

Fig. . MDP IX , left side with sealing (courtesy by G. Giovinazzo).

 SEAL IMPRESSIONS FROM SUSA.
RE-EVALUATING SOME OF THE FINDINGS IN SUSA
AVAILABLE IN THE NATIONAL MUSEUM OF IRAN
AND INTRODUCING SOME UNPUBLISHED SAMPLES

Sedigheh Piran*

. Introduction

The documentation project of objects in the National Museum of Iran began last year. The
registration of all the tablets and inscriptions present in the Inscriptions Department of the
museum, including nearly , tablets and inscriptions was prepared. During the course of
this work, issues such as the archaeological context and the dating of the objects were also
researched when possible, with quite some success.
Among the tablets found in the Inscriptions Department of National Museum of Iran, a
majority of the tablets belong to the Susa excavations. In the years after their discovery, a number
of clay lumps bearing seal impressions as well as sealed cuneiform tablets were put into tin boxes
in a confused and mixed manner without reference to the dates and archaeological context,
and were transferred to the storage area of the National Museum of Iran. When the boxes
were opened in recent years,  seal impressions of various historical periods and layers were
retrieved, still wrapped in newspapers dating to the ’s. None of these pieces is registered
by Field Number, so their documentation required a review of excavation reports and other
publications related to them.
Seal impressions can be classified in terms of different historical periods. Furthermore, they
can be categorized by shape, usage and iconography. In this study, the seal impressions were
initially classified according to their iconography. Then, using the scientific literature, they were
attributed to historical periods.
It is noteworthy that a number of designs were repeated, and in some cases it is probable that
some were broken pieces of the same seal impression, sometimes already cited elsewhere.
As a result, all of these seal impressions could be attributed to six chronological stages: Susa I,
Susa II, Susa III, Early Dynastic I/ II, Old Elamite, Old Babylonian. Only a few pieces could not
be identified due to damage.
The stratigraphy of Acropole I at Susa was established by Perrot, with layers – belonging
to Susa I, layers – to Susa II and layers – to Susa III (Perrot : –).
According to the final report of Le Brun on the oldest layers of Acropole I, stamp seals occur,
in layer  cylinder seals and finally, in layer , the first accounting documents (round clay
balls or bullae). In layer  numerical signs appear on tablets and round clay balls. Layer  saw
the first tablets and finally in layer  the inscriptions have been found (Le Brun : ). Seal
impressions of Old Babylonian and Old Elamite types have also been reported from the Ville
Royale (Amiet ).

* National Museum of Iran, Tehran. I cordially thank professors Yousef Majidzadeh, Elizabeth Carter and Holly

Pittman for a final review of these seal impressions and for their kind suggestions. The comments of Prof. Carter
have been added in footnote at the adequate places. My gratitude also goes to Mrs. Farideh Shirkhodaii, head of the
Library of the National Museum of Iran for her unfailing help.
 sedigheh piran

. Seal Impressions of the Susa I Period

“The earliest C determination from the Susa I levels on the Acropole falls between  and 
cal. bc while the latest date from the period can be placed between  and  cal. BC” (Potts
: ). In southwestern Iran, stamp seals appear in the Middle Village Period and then in
the Late Village Period they are used in the whole of western Iran (Hole : ).
Seals were impressed on lumps of clay sealing jar caps or attached like large lentils to knots
closing small packages, guaranteeing their security (Amiet : ). In fact “seals and seal
impressions were present in some numbers, implying specialized closure of containers for
shipment and/or storage” (Johnson : ). Some Susa I seals have also been used to seal
locked doors. Fiandra has studied this practice of sealing doors (Fiandra ).
Also in this period figures were engraved on convex buttons used as stamp seals. The ones
from the Susa excavations belonging to the Susa I period include about  stamp seals and
their impressions (Potts : ). Among the  seal impressions mentioned,  belonging to
this type were identified and can be classified in two groups: one with geometric and one with
human engravings.
According to Amiet, seals and seal impressions from Susa excavations symbolize political
power, and those who used them are part of the governing elite of the land. In his opinion,
these individuals probably played religious roles, participating in rituals as a sovereign together
with his partner sharing his rank (Amiet : ).
There are  broken and incomplete seal impressions (B.K. , B.K. , B.K. ,
B.K. , B.K. , Fig. –, : B.K. ) in this collection all belonging to the same seal
impression. Its scene is comparable to that of a complete one published by Amiet (Amiet ,
pl. : ).
Amiet interprets this scene as a ritual performed in presence of an important figure with a
spiritual role, maybe a god. In his opinion, the long garment of this person is similar to the
dress of an animal god with a dome-shaped crown; another person present at the scene behind
him is his servant and wears the same clothes. The third person carries gifts and is moving in
front of him. According to Amiet, this scene could be an example of a ceremonial observed at
the beginning of the Proto-Urban period in the Sumerian city of Uruk (Amiet : , ,
fig. )1 and the emergence of this role in the early warehouses of equipment and goods could
have a special meaning (Amiet : ). Other elements in this scene are the moon and the
sun above the figures.
There is another example of a seal impression with two people facing each other (de Mec-
quenem , fig. : ) that depicts a religious ceremony. The one figure is a half naked man
with a zigzag striped skirt and a belt of hanging ropes who holds something in his right hand,
above the head of the one in front of him. The latter also is half naked but wears pants and holds
something in his raised right hand. Other elements fill the scene, such as an oven with a conical
top placed between the two figures and probably related to the ceremony depicted. On the left
side of the scene there is a head of a goat. There are  impressions of this seal in this collection
(B.K. , B.K. , Fig. –, : B.K. ).
Another seal impression (Fig. –, : B.K. ) shows two people sitting with round hats,
they seem to be holding two pieces of wood2 over their head; on top of this scene part of the
bottom of a long skirt with its zigzag design and a pair of other legs in front of it is visible (de

1 According to E. Carter, the net skirt is like those seen on representations of the Uruk Period ruler.
2 Nowadays nomads in Iran, especially in the east, west and south, play wooden instruments in ceremonies and
therefore, the life style of nomads continued from past through modern time certainly with some changes we can
suggest that the main material of instruments presented on seals probably are made of wood.
 seal impressions from susa 

Mecquenem , fig. : ). Comparison of the remaining part at the top of this scene with
the remnants of the bottom part of the seal impression previously described (Fig. , , B.K.
), shows that these two pieces are related and complement each other. This connection was
not made up to now because of the fragmentary state of the impressions and the approximate
drawings made from them. Until now the impressions were thought to have been made with
two separate seals. In all three pieces of this seal impression humans have been depicted in a
specific way. They are all portrayed moving their hands or feet no doubt in order to emphasize
the actual performance of the ritual. The objects in the hands of the figures probably are musical
instruments for the performance of music accompanying the rituals. The importance of joining
the two pieces into two scenes of the same seal is that this may indicate that both ritual scenes
were performed within the same musical context.
Some scholars recognize Susa as a religious center and residence of priests on the basis of
these important ceremonial-ritual seal impressions (Hole : , Johnson : ).
Stamp seal impressions with geometric designs in this collection do not show much variety.
On the whole,  broken and imperfect pieces of this type were identified  of which bore the
repeated cross motif (B.K. , B.K. , B.K. , B.K. , B.K. , B.K. , B.K.
, B.K. , Fig. –, : B.K. ); the other two pieces had starlike designs (Fig. –, :
B.K. , Fig. –, : B.K. ). Another seal impression (Fig. –, : B.K. ) of this type
with an ambiguous design has been published before (Amiet , pl. : ).

. Seal Impressions of the Susa II Period

Wright and Johnson have divided the period between  and  B.C into Early, Middle and
Late Uruk. Potts on the other hand, prefers the term Susa II because of the indigenous nature
of the Khuzestan culture. In this period, Khuzestan was colonized by Mesopotamia, Susa was
almost a second Sumer (Potts : –).
The use of engraved cylinder seals becomes prevalent now. By rolling the cylindrical seals
over the clay lumps or surfaces of tablets, the full scene was reproduced. According to Amiet,
the engraved cylindrical seals represent the beginning of the real Old World civilization (Amiet
: ).
The diversity of seals in this period is considerable as compared to the previous one. They
were used as a registration system of economic, administrative and business documents; Le
Brun emphasizes the capacity of seal impressions for guaranteeing data and information (Le
Brun : ). This type of seal impression has only one number in this collection (Fig. -, :
B.K. ) that has already been published (de Mecquenem , fig. : ), its scene consists
of unclear depictions of nude people moving in a row.
Artists in Susa have also been inspired by daily human activities. Thirteen samples of this type
of seal impression were found. One of these was impressed on a jar or bottle stopper (Delougaz
and Kantor , part : , part : pl. ) (Fig. -, : B.K. ) it represents a scene of daily
activities (de Mecquenem , fig. : ); another example (Fig. –, : B.K. ) shows part
of a boating scene with people carrying objects on their heads (de Mecquenem , fig. : ).
Seven seal impressions previously published include a variety of scenes showing fruit picking by
women with a hair lock or pony tail, wearing long skirts (B.K. , B.K. , B.K. , B.K.
, Fig. -, : B.K. , Fig. -, : B.K. , Fig. -, : B.K. ) that has been already
published (Amiet , vol. II, pl. : –). Finally four pieces with designs including a
container and animal motif (Fig. -, : B.K. , Fig. -, : B.K. , Fig. -, : B.K. ,
Fig. -, : B.K. ) show characteristics of the art of seal engraving of the Susa II period
(Amiet , vol. II, pl. : ). A type of engraving rendering a boar with zigzags representing
 sedigheh piran

the animal’s back can also be found on contemporary seals in Mesopotamia (Amiet : pl. :
). An example of this impression was identified in National Museum collection (fig. -, :
B.K. ).

. Seal Impressions of the Susa III Period

Scheil, called this period “Proto-Elamite” (Scheil : ). In Voight and Dayson’s chronology
it is “Susa III” (Voight/Dyson : , fig. ). Several studies have provided substantial
insights on the writing and management system of this period dated approximately between
 and  bc (Potts : –).
In this period, “after the invention of writing, the art of drawing on the seals has also changed,
and seal impressions give useful information about the way of life and the environment of Susa”
(Majidzadeh : ). Among the seal impressions of the National Museum of Iran,  are
comparable to pieces of early to late Susa III periods. Their scenes are very diverse and include
depictions of activities of craftsmen and workers (three pieces: Fig. , : B.K. ,3 Fig. , :
B.K. , Fig. , : B.K. ), realistic images of animals such as a lion? a cow and a bird,
standing or on the move (one piece from each: Fig. , : B.K. , Fig. , : B.K.  and
four pieces from the last: B.K. , , , Fig. , : B.K. ), scenes of rams grazing in
prairies (two pieces: Fig. , : B.K. , Fig. , : B.K. ) and finally part of a cow (Fig. , :
B.K. ) that have already been published (Legrain : Pl. VI: ).
Another type of illustration in the late Susa III period presents intertwined animals (Amiet
, pl. : ). Four related broken pieces of such seals were identified in the collection
(Fig. , , , , : B.K. , B.K. , B.K.  and B.K. ).
In the late part of this period, i.e. the Post Proto-Elamite period, the scenes feature remarkable
changes. Animals mimic humans, in other words, they are shown in human positions. This kind
of images is unprecedented in carvings of the first cylinder seals (Amiet : , Figs: –).
In the National Museum collection two pieces of seal impression were identified showing scenes
of animals mimicking activities of humans (Fig. , : B.K. , Fig. , : B.K. ).
Geometric and rosette motifs are among other varieties of designs used in this period. Two
examples of the geometric type (Fig. , : B.K. , Fig. , : B.K. ) and one seal impression
with a rosette (Fig. , : B.K. ) were identified. The rosette or flower design in this seal
impression is comparable to a similar design from the early Elamite period (Legrain :
Pl. II, ).

. Seal Impressions of the Early Dynastic Period

The growth of city states in Mesopotamia was simultaneous with the Post Proto-Elamite civi-
lization in Susa. This period, known as “Early Dynastic” in Mesopotamia, is to be approximately
dated between  and  or bc. Susa was partly under the influence of Sumerian
power in this period (Amiet : –). Comparative stratigraphy (Carter ) shows
that in Susa the late period III (or late Proto-Elamite) and Period IVA correspond to the Early
Dynastic Period in Mesopotamia. The only find related to the recording of administrative affairs
in early Susa IV period is a cache of sealings found in Susa, “which can be dated from ED I–III
period on the basis of Mesopotamian parallels” (Carter , vol. /: ).

3 E. Carter remarks that this sealing looks earlier than I think (might be Uruk style cf. fig. -, , B.K. ).
 seal impressions from susa 

This Early Dynastic period is divided into three parts according to changes in artistic styles,
especially in seals and architecture in the Diyala region of Mesopotamia (Majidzadeh :
). Most designs of the first period include simplified combined forms of animals, plants and
linear designs. This type of style has been called “Brocade Style” by Frankfort (Porada ,
vol. : ). In the collection of the National Museum there are two pieces of seal impressions
with similar motifs that could be part of one and the same (Fig. , : B.K. , Fig. , : B.K.
). The motif of these two seal impressions is comparable to a published one from the first
Early Dynastic period (Porada , vol. : pl. VIII. ; Moortgart : Tafel . ).
In the Second Early Dynastic period, the use of simple lines to create the first period motifs
has largely been lost, and instead flat surface images with no visual value appear (Majidzadeh
: ). The main themes of this period include legendary themes, heroes, animals and
monsters and for the first time the banquet scene (Porada , vol. : ). Two pieces of seal
impressions from this period were identified (Fig. , , : B.K. , B.K. ), both are already
published (Amiet , pl. : , ).
In the Third Early Dynastic period a popular style is the “Fara style” (Porada , vol. I: ,
Moortgart : ). In this time, the cylinder seal impressions found in Susa are not different
from those of the Sumerian civilization in Mesopotamia. Amiet believes that in this period
Elam had lost the authenticity of its art (Amiet : , fig. –). Since we do not have
exact parallels with Fara motifs in Susa, we cannot decide whether there was a real local style
in this city or a borrowed one.
In this collection, only one of the seal impressions shows any similarity with the Fara style.
It has a scene with the “master of the animals” (Fig. : B.K. ). The seal impression is largely
damaged and details of the subdued animals are not detectable. On the whole, from what
remains, it seems that the way human and animal are rendered is similar to “Fara style” motifs.

. Seal Impressions of the Early Elamite and Early Babylon Periods

The division of Elamite history into three periods, old, middle, and new was first proposed by
H.H. Paper for Elamite texts (Paper, : ). This period begins with the Šimaški Dynasty
from  to c.  bc. At the beginning of this period, the Ur III dynasty in Mesopotamia
dominated Susa, and in the early second millennium bc, the city expanded and became
a regional and even international capital (Carter : –). Around – bc, the
Sukkalmah period was contemporary with the dynasty of Babylon in Mesopotamia (Schacht
: ). According to Potts the position and influence of the Sukkalmah dynasty in the early
second millennium bc represents the peak of Elamite political influence in Western Asia (Potts
: –).
These seals in the early second millennium bc are heterogeneous. On the one hand, the style
of some of them is similar to the “Old Babylonian” showing the characteristics of Mesopotamian
iconography. On the other hand, in this period we see an increasing number of seals in a style
called “Popular Elamite” (Potts : ).
Old Babylonian seal motifs are a compilation of the traditional theme of entry into the
presence of gods with single motifs added to differentiate the seal. Scenes related to religious
ceremonies continue in imitation of the past—the Akkad period—but there is greater emphasis
on the seal legends4 (Majidzadeh , –). “During the revival the scene of the presenta-
tion to a deified king was favoured more and more” (Moortgart : –, Pl. G, G–).

4 According to E. Carter the key difference between the popular style seals and the more Mesopotamian style
 sedigheh piran

Porada also refers to similarities between seal scenes of the Elamite period with the general
representation of worship scenes on Old Babylonian seals (Porada : –). There are three
pieces of broken seal impressions among the seal impressions of the National Museum of Iran
with Old Elamite characteristics. Only one has part of the cuneiform inscription (Fig. , : B.K.
). Another seal impression shows part of a worship scene in which two persons are standing
facing each other; there is a framed inscription on two sides of the scene. Part of the impression
has been lost due to wear and its details are not discernable (Fig. , : B.K. ). The third seal
impression of this period is a scene showing two individuals standing in front of each other
wearing long Elamite robes with the profile of a bird looking leftward between them. On the
bird’s back there is the design of a crescent connected to a long rod (Fig. , : B.K. ). The
crescent with a long rod is comparable to a seal impression from the Old Babylonian period
carried by a god (Von Der Osten : , fig. : ). Elamite robes of individuals represented
in the scene are comparable to Old Elamite seal impressions published (Porada : –,
figs. –). Bird’s design and crescent are also present in this art (Rutten , No. : pl. V:
), (de Mecquenem , p: , fig. ).
Two other seal impressions (Fig. , : B.K. , Fig. , : B.K. ) on clay have been
impressed several times—maybe used to seal doors. These two impressions have a similar scene
and it could well be that they were made with the same seal on the same object, broken in
two. The seal impression bears the scene of bringing offerings to a god or king comparable
to Old Babylonian examples (Porada , vol. I: ). In this scene three persons in a row
face an enthroned deity. The first of them is a minor goddess introducing the other ones. The
other two bear the offerings. They wear different clothes. There is an indistinguishable person,
perhaps a kneeling male figure between the god and the minor goddess. Between the last two
persons there are two similar small jars of the same size, one under and the other above the
scene. These jars are similar to the design of a jar on a seal of the Old Babylon period (Porada
, vol. II: pl. LVI: ). The god wears a long mantle, the skirt decorated with a fringe. His
conical headdress seems to be decorated with horns. This kind of garment is similar to examples
from the Old Babylon period (Von Der Osten : , Fig. : ). The garment of one of the
offering bearers is similar to that of the god: a long mantle decorated with a fringe and a high
headdress with a rounded top. This garment is also comparable to examples of garments of the
Old Babylonian period (Von Der Osten : ).
Unidentifiable seal impressions in this collection are B.K. , B.K. , B.K. , B.K.
, B.K. , B.K. , B.K.  and B.K. .

. Conclusion

Artefacts from archaeological excavations stored in museum storerooms often need to be

reviewed, studied and republished. The excavations of Susa are important for archaeologists
because of the large number of interesting artefacts they uncovered. These objects have been
stored in bulk often in less than ideal conditions in museums around the world. Completing
their study or reviewing it on the basis of modern information often leads to new discoveries.
Reviewing the rich collections of the National Museum of Iran with the cooperation of
archaeologists in the recent years has contributed a lot to publishing updated information in this
regard. Re-opening boxes with tablets and seal impressions found in Susa, allowed to complete
our knowledge and understanding of the art of the early periods.

pieces is linked to the material. Local style seals use bitumen and thus are deeply cut. Some have “fake” inscriptions
in imitation of Mesopotamian style seals.
 seal impressions from susa 

In this project, out of  seal impression,  were identified, classified by style and attributed
to historical periods. Some of them are published here for the first time. The remaining  pieces
were unidentifiable.

Bibliography

Amiet, P. (): Élam, Auvers-sur-Oise.

———. (): Glyptique susienne (= MDP ), Paris.
———. (): La Glyptique Mésopotamienne Archaïque, Paris.
———. (): “A propos de l’usage et de l’iconographie des sceax à Suse,” Paléorient /: –.
Carter, E. (): “Note on Archaeology and the Social and Economic History of Susiana,” Paléorient
/: –.
Delougaz, P. and Kantor, H.J. (): Chogha Mish vol. I/II,e First Five Seasons of Excavations –,
Alizadeh, A. (Ed.) (= OIP ), Chicago.
de Mecquenem, R. (): “Fouilles de Suse, –,” MDP : –.
Fiandra, E. (): “Porte e chiusure di sicurezza nell’Antico Oriente,” Bollettino d’Arte : –.
Frankfort, H. (): Cylinder Seals. A Documentary Essay on the Art and Religion of the Ancient Near
East, London.
Hole, F. (): “Settlement and Society in the Village Period.” In: Hole, F. (Ed.), The Archaeology of
Western Iran, Settlement and Society from Prehistory to the Islamic Conquest, Washington: –.
Johnson, G. (): “Nine Thousand Years of Social Change in Western Iran.” In: Hole, F. (Ed.), The
Archaeology of Western Iran, Settlement and Society from Prehistory to the Islamic Conquest, Washing-
ton: –.
Le Brun, A. (): “Le niveau  de l’Acropole de Suse. Mémoire d’argile, memoire du temps” Paléorient
/: –.
Legrain, L. (): Empreintes de cachets élamites (= MDP ), Paris.
Majidzadeh, Y. (): Elam History and Civilisation, Tehran.
———. (): Ancient Mesopotamia: History and Civilization, Vol. III: Art and Architecture, Tehran.
Moortgart, A. (): Vorderasiatische Rollsiegel, Ein Beitrag zur Geschichte der Steinschneidekunst,
Berlin.
———. (): The Art of Ancient Mesopotamia, London.
Paper, H.H. (): “Elamite Texts from Tchogha-zambil,” JNES : –.
Perrot, J. (): “La séquence archéologique de Suse et du Sud-Ouest de l’Iran antérieurement à la
période Achéménide,” Paléorient : –.
Porada, E. (): Corpus of Ancient Near Eastern Seals in North American Collections, Vol. I/II: The
Collection of the Pierpoint Morgan Library, Washington.
———. (): The Art of Ancient Iran Pre-Islamic Cultures, New York.
Potts, D.T. (): The Archaeology of Elam, Formation and Transformation of the ancient Iranian State,
Cambridge.
Rutten, M. (): “Glyptique susienne-empreintes et cylindres,” RA : –.
Schacht, R. (): “Early Historic Cultures.” In: Hole, F. (Ed.), The Archaeology of Western Iran, Settle-
ment and Society from Prehistory to the Islamic Conquest, Washington: –.
Scheil, V. (): Texts élamites-sémitiques, troisième série (= MDP ), Paris.
Voight, M. and Dyson, R.H. (): “Chronologies in the Near East-Iran.” In: Erich, R.W. (Ed.), Chronol-
ogy in Old World Archaeology, vol. II, Oxford: –.
Von der Osten, H.H. (): Ancient Oriental Seals in the Collection of Mrs. Agnes Baldwin Brett (= OIP
), Chicago.
 sedigheh piran

Fig. . Map of topography of  grand hills of Susa (de Mecquenem , fig. ).
seal impressions from susa 

Fig. -—Susa I seal impressions.

 sedigheh piran

Fig. -—Susa I seal impressions.

seal impressions from susa 

Fig. -—Susa II seal impressions.

 sedigheh piran

Fig. -—Susa II seal impressions.

seal impressions from susa 

Fig. —Susa III seal impressions.

 sedigheh piran

Fig. —Proto Elamite – late Susa III seal impressions.

Fig. —Post Proto Elamite – late Susa III seal impressions.

 seal impressions from susa 

Fig. —Early Dynastic I, II seal impressions.

 sedigheh piran

Fig. —Early Dynastic III (Fara Style?) seal impressions.

Fig. —Old Elamite seal impressions.

 seal impressions from susa 

Fig. —Old Babylonian seal impressions.

 II

PHILOLOGICAL, PHILOLOGICAL-HISTORICAL
AND HISTORICAL PERSPECTIVES
 UR-NAMMÂ(K)’S CONQUEST OF SUSA

Gianni Marchesi*

During my stay in the tablet room of the University of Pennsylvania Museum of Archaeol-
ogy and Anthropology in , I discovered two “forgotten” fragments of inscribed vessels
from Woolley’s excavations at Ur, whose inscriptions mention the conquest of Susa by Ur-
Nammâ(k),1 the founder of the so-called Third Dynasty of Ur. They are published here by the
kind permission of Steve Tinney, co-curator of the Babylonian Section of the Penn Museum.2
To tell the truth, the inscriptions in question have long been known: they were first edited by
Braun-Holzinger (, , G  and G ); and, more recently, by Frayne (, –,
nos.  and ). However, neither Braun-Holzinger nor Frayne understood their historical
significance, not realizing that the dedicant of the inscriptions was, in both cases, king Ur-
Nammâ(k) of Ur.
In previous scholarship, the capture of Susa was generally counted among the deeds of
Sulgi(r) (= “Šulgi”),3 whom the majority of scholars considered to be the true builder of the
Ur III empire.4 Beyond rendering unto Ur-Nammâ(k) that which is Ur-Nammâ(k)’s, the texts
published here document a key episode in the history of the Ur III empire and of its eastwards
expansion.5

. CBS  (= Fig. )

This fragment of a white calcite vase, which measures × . × . cm,6 was found in the fill of
the É.NUN-mah.7 Its inscription reads as follows:
˘
* University of Bologna.
1 Used to be read also as “Ur-Namma” or “Ur-Nammu.”
2 I would like to thank Steve Tinney for allowing me to publish these two objects (plus the related piece CBS ,

from Nippur). The photos reproduced in this article were provided by Jeremiah Peterson and adapted for publication
by Massimo Bozzoli. To both of them go my heartfelt thanks. I am also grateful to Glenn Magid, who checked
and corrected my English. The abbreviations used in this article are those of the Reallexikon der Assyriologie und
Vorderasiatischen Archäologie, vol. . This study was made possible by a research grant from the Department of
Archaeology of the University of Bologna.
3 See, among the others, Cameron : –; Hinz : –; Lambert : –; Stolper : ; Amiet

: –; André/Salvini : ; Steve : ; Carter in Harper et al. : ; Brentjes : ; Malbran-
Labat : ; Pittman : ; Quintana : –; Vallat : ; Huot : –. On the other hand,
Michalowski (: ) thought it possible that the conquest of Susa by Ur “took place earlier during the reign of
Ur-Namma” (also cf. T. Potts : – and D.T. Potts : ). For the reading of the name DUN-gi as sul-gi
(= /sulgi(r)/) rather than šul-gi, see Marchesi : – n. .
4 Note, however, the contrary view of Michalowski (: ): “Later Mesopotamian traditions celebrated

Shulgi, the second king of the new dynasty, above all other members of his family […]. Modern scholars have
followed suit, […] mesmerized by the opinions of others; it is time, however, to rehabilitate Ur-Namma, the man
who actually created the Ur III kingdom and who set the foundations for generations to come. […] The borders of
the core of the empire were established under his rule and were not to change until the collapse of his creation two
generations later.”
5 In view of the significance of this event, it is likely that one of the lost year names of Ur-Nammâ(k) was *mu

susinxki ba-hulu, “Year: Susa was smitten;” or *mu ur-dnamma lugal-e susinxki mu-hulu, “Year: Ur-Nammâ(k), the
king, smote ˘Susa” (cf. Frayne : –). ˘
6 According to Braun-Holzinger : , ad G . Cf. Woolley : , ad *U. (with slightly different

dimensions:  × ×. cm).

7 Woolley : , ad *U..
 gianni marchesi

0. [ur-] d [namma]

0. [lug]al uri5[ki-m]a-ke4
0. [… M]ÙŠ.ERENki
0. [m]u-hulu-a
0. nam-ra˘ -[…]

Frayne (: ) read the first line as “ d […]” and commented: “The fact that the royal
titulary ends with the epithet ‘[k]ing of Ur’ would suggest an attribution of the fragment to
Šulgi, since Amar-Suena, Šū-Sîn and Ibbi-Sîn all used the title ‘king of the four quarters’ after
‘king of Ur’ in their inscriptions. If line 0 is a royal name, then it cannot be Ur-Nammu; his
name never appears with the prefixed DINGIR sign in contemporary royal inscriptions.”
However, the DINGIR sign is not at the beginning of the case, but in the middle. Moreover,
Sulgi(r) always makes the epithet ninta kalag-ga, “strong male,” or diĝir kalam-ma-na, “(patron)
god of his country,” precede lugal uri5ki-ma(-ke4), “king of Ur.” Therefore, [ur-] d [namma] is the
only restoration possible in line 0.
On the basis of the partial duplicates UET , , CBS  (= Fig. )8 and CBS  (see
below, sub II), we may reconstruct the entire text as follows:

. [d …]
(one or two lines broken)
0. [ur-] d [namma]
0. [lug]al uri5[ki-m]a-ke4
0. [u4 s]usinx(MÙŠ.EREN)ki
0. [m]u-hulu-a
0. nam-ra ˘ -[aš]
0
 . [mu-na-AK-a]
0. [nam-ti-la-né-šè]
0. [a mu-na-ru]

“[To DN, (…,) his lord/lady, Ur-Nammâ(k), ki]ng of Ur, [wh]en he smote [S]usa and [turned it
into] booty [for him/her, presented (this vase) for his own life].”

.. Philological Remarks

0: Lafont (: ) argued for a reading /šušum/ of the sign complex MÙŠ.ERENki in Ur III
texts. However, the syllabic spellings su-sí[nki]-na (CIRPL , Ean.  rev. vii 0; locative) and
su-sín-naki (SAT ,  rev. ; genitive) definitively prove that the Sumerian reading of the
logogram for Susa was /susin/ in the third millennium bc.9
As for “kaskal šu-šu-umki,” quoted by Lafont (: ), cf. KASKAL urušu-šiki (TCL ,
A: [envelope]10) and ha-ra-an šu-ši-im (Leemans : , AO :11). Clearly, šu-šu-
˘
umki represents the undeclined (i.e., in the nominative) Akkadian name of Susa.12 This is

8 Cf. Frayne : –, nos.  and , respectively.

9 Cf. MÙŠ.ERENki-na (genitive) in Lambert : –, no. : and MVN , :. For additional occurrences
of the spelling su-sín(ki) in Sumerian texts, see RGTC ,  s.v. “Šušina/Śuśin;” and RGTC , – s.v. “Śuśin.”
Sumerian /susin/ is obviously related to Elamite /šušen/ (cf. Krebernik : –). Even assuming that Sumerian
s was pronounced as [š] in third-millennium Sumerian in view of the spelling su-sín(ki) we should for the sake of
consistency transliterate MÙŠ.ERENki in Sumerian context as susinx rather than as “šušin” or “šušina” (as scholars
typically do). In other words, the sibilant that occurs twice in the word for “Susa” was, in Sumerian, the same as the
s of su, “flesh,” and different from the š of šu, “hand,” regardless of how they were actually pronounced.
10 The tablet (= B), line , has KASKAL MÙŠ.ERENki, instead.
11 The envelope (ibid., p. ), line , has h[a-ar]- ra-an MÙŠ.EREN [ki], instead.
12 Cf. Krebernik : –. ˘
 ur-nammâ(k)’s conquest of susa 

confirmed by, among other things, Proto-Diri Oxford : MÚŠ.ERENki = šu-ú-šu-um (MSL
, ).13
0: The sign HUL is generally transliterated hul, in spite of the fact that, when it is followed by
the suffix {"a}, it˘ is never written **HUL-la or ˘**HUL-lá in third-millennium texts,14 but rather
as HUL-a. The latter spelling suggests ˘ that HUL ended
˘ with a vowel. In fact, the Old Babylonian
˘
syllabary ˘
Proto-Ea gives hu-lu ( sources), beside hu-ul ( source), as the pronunciation of HUL
(MSL , : Proto-Ea ). ˘ In transliterating texts˘ from the third millennium bc, the former ˘
15
reading, hulu, is therefore to be preferred.
˘ the meaning of the word is concerned, it should be noted that the usual translation of
As far as
hulu as “to destroy” is questionable. The verb hulu is frequently found in year names consisting
˘ clauses/sentences such as mu GN (al-)hulu-a
of ˘ / ba-hulu or mu PN (TITLE) GN mu-hulu(-a).16
˘
One of these (year Sulgi(r) ) is called: mu si-mu-ru-um ˘ ki
ù lu-lu-bu-umki a-řá  lá˘-kam-aš
ba-hulu, “Year: Simurrum and Lullubum were …ed for the ninth time.”17 As Sollberger (:
˘
–) already observed, “it is hardly likely […] that Simurrum could have been ‘destroyed’
 times [in the period] from Šul-gi [year]  to [year] , and twice in two consecutive years
(–).”18
The usual translation “to destroy” is based on Akkadian renderings in first millennium
bilingual texts, in which Sumerian hul(u) is translated into Akkadian as either ubbutu or
šulputu.19 However, the rendering of ˘hul(u) as ubbutu is a “hapax” translation that may stem
erroneously from a confusion between ˘ hul(u) and gul—the latter being the usual equivalent
of Akkadian abātu/ubbutu, “to destroy.”20˘ As regards šulputu, in the dictionaries, in addition to
“to destroy,” one also finds the meanings “ruinieren, brandschatzen,”21 “to overthrow, defeat”
and “to desecrate, defile (a temple, a palace, etc.).”22 Be that as it may, neither abātu/ubbutu nor
šulputu was employed in third-millennium texts for translating hulu. In their place, na" ārum,
“to smite, strike (mortally),”23 or hulluqum, “to annihilate,” was used. ˘ 24
˘

13 Cf. Krebernik :  with n. . Needless to say, the newly introduced Sumerian Lautwert “šušun” for

MÙŠ.EREN (Mittermayer/Attinger :  and ) has no basis, and its use should be avoided.
14 Pace Wilcke :  n. . Wilcke quoted two alleged occurrences of the spelling “hul-la,” but neither of them,

in fact, exists: “in-hul-la” is a misreading of NG : by Falkenstein (: ; cf.˘photo of the tablet: http://
˘
cdli.ucla.edu/P); and “é-da-hul-la” in PDT , : is in all likelihood a misprinting of the well-known PN
é-da-húl-la, “He who rejoices over the ˘ house/temple” (see Nik. , :; OIP ,  rev. ; UDT  rev. ii ; etc.).
˘ also Krecher :  with n. .
15 Cf.
16 See Gelb/Kienast : –: D-, D-, D-, D-, D-, D-, D-; ibid., : D-; Sigrist/Gomi : –

, s.v. “hul.”

˘
17 See Frayne : .
18 Consequently, Sollberger (loc. cit.) suggested that in year names, hulu should rather be translated as “to raid, to

carry out a punitive expedition, to sack.” See also Michalowski :  ˘ and Owen :  with n. : the former
observed that “the verb ‘to destroy’ (Sumerian hul), in such contexts, must not be taken literally, and should really
˘ proposed the meaning “to raid” or “to conquer” for hulu.
be rendered as ‘defeat’, or [the] like,” while the latter
19 See CAD A/, , s.v. abātu A, lex.; L/, –, s.v. lapātu, lex. ˘
20 See CAD A/, –, s.v. abātu A, lex.
21 See AHw, , s.v. lapātu(m), Š.
22 See CAD L/, –, s.v. lapātu, b–c.
23 Cf. Gelb : , sub N’ R. Note that the logogram for this verb is SAĜ.ĜIŠ.RA, which means “to club/cudgel
3
the head,” i.e., “to deal a mortal blow.”
24 For hulu corresponding to na" ārum in the bilingual inscriptions of Sargon of Akkad, see Kienast/Sommerfeld

: ,˘ s.v. nêrum, b. For the rendering of hulu as hulluqum, cf. Sulgi(r) year  (mu dsul-gi … ki-maški hu-ur5-ti
ù ma-da-bé u4 -a mu-hulu) with Frayne : ˘ ,˘“Šulgi” :– (ì-nu / ma-at ki-maški / ù hu-ur-timki˘ / ù-ha-
li-qú-na); and ibid., ,˘ “Šū-Sîn” :0–0 (u4 ma-da za-a[b]-ša-liki! / ù ma-d[a-m]a-da / simaškiki˘{-ka} / mu-hulu-a)
˘
0 ki ˘
ki
with ibid., –, “Šū-Sîn”  ex. :—rev.  ( ma -ta-at si-[m]aš-ki-im / [ù]-ha-li-iq / ma-at za-ab-ša-li / … /
ù-ha-l[i-i]q-s[u?-nu?-ti?]). ˘
˘
 gianni marchesi

0–0: Cf. “Puzur-Sulgi to Ibbi-Sîn”  (Michalowski :  and ): ha-ma-ziki nam-ra-
aš im-mi-in-AK (var. im-ma-an-lah5), “he (= Išbi-Erra) turned Hamazi into ˘ booty” (var. “…
˘
led away (the people of) Hamazi as booty”). 25 ˘
˘

. CBS  (Figs. a–b)

This alabaster vessel fragment ( × ×.~. cm) is also from Ur. No information on its
findspot is available. The inscription reads:

0. [enim mah(?)] d

nanna [lugal-na(?)]-ta
0. u4 susinxki ˘
0. mu-hulu-a
0. nam-˘ ra -a[š]
0. [mu-na-AK-a]
(rest broken)

“[…], according to [the great word of] Nanna, [his master], when he (= Ur-Nammâ(k)) smote
Susa and [turned it in]to booty [for him, …].”

.. Philological Remarks

This inscription is obviously related to the previous one. Although the two inscriptions are not
exact duplicates, the attribution of CBS  too to Ur-Nammâ(k) is almost certain (note that
lines 0–0 are identical to lines 0–0 of CBS ).
The proposed restoration for the difficult line 0 is based partly on Wilcke :  ff.,
“Kodex Urnamma” – = – = – (á dnanna / lugal-ĝá-ta, “by the might of
Nanna, my master”), partly on Nasrabadi : , i – (i-na / a-wa-at dELLIL / ra-bí-tim,
“according to the great word of Ellil;” inscription of Amar-Zu"ena(k)). Note, however, that the
syntax of the sentence is odd. One would rather expect: u4 susinxki / mu-hulu-a / nam-ra-aš /
mu-na-AK-a / enim mah dnanna lugal-na-ta, “(… Ur-Nammâ(k), …,) when ˘ he smote Susa and
˘
turned it into booty for him, according to the great word of Nanna, his master, (did this and
that).”26
As an alternative, one might consider restoring line 0 as [u4 á] d nanna [lugal-na]-ta, and
emend line 0, accordingly, as {u4} susinxki. In this case, we would have: “[when], by [the might]
of Nanna, [his master], he smote Susa …”

Bibliography

Amiet, P. (): Suse:  ans d’histoire, Paris.

André, B. and Salvini, M. (): “Réflexions sur Puzur-Inšušinak,” IrAnt : –.
Behrens, H. and Steible, H. (): Glossar zu den altsumerischen Bau- und Weihinschriften (= FAOS ),
Wiesbaden.
Braun-Holzinger, E.A. (): Mesopotamische Weihgaben der frühdinastischen bis altbabylonischen Zeit
(= HSAO ), Heidelberg.
Brentjes, B. (): “The History of Elam and Achaemenid Persia: An Overview,” CANE : –.
Cameron, G.G. (): Histoire de L’Iran antique, Paris.
Falkenstein, A. (): Die neusumerischen Gerichtsurkunden, II. Umschrift, Übersetzung und Kommen-
tar (= AbhMünchen NF ), München.

25 Cf. Gelb : –.

26 Cf., e.g., Frayne : –, “Ur-Nammu”  i —ii .
 ur-nammâ(k)’s conquest of susa 

Frayne, D.R. (): Ur III Period (– bc) (= RIME /), Toronto, Buffalo and London.
Gelb, I.J. (): Glossary of Old Akkadian (= MAD ), Chicago.
———. (): “Prisoners of War in Early Mesopotamia,” JNES : –.
Gelb, I.J. and Kienast, B. (): Die altakkadischen Königsinschriften des Dritten Jahrtausends v.Chr. (=
FAOS ), Stuttgart.
Harper, P.O., Aruz, J. and Tallon, F. (Eds.) (): The Royal City of Susa. Ancient Near Eastern Treasures
in the Louvre, New York.
Hinz, W. (): Persia c. – B.C., Cambridge (republished as Chap. XXIII of CAH, rd ed.,
vol. I/, –).
Huot, J.-L. (): Une archéologie des peuples du Proche Orient, I. Des premiers villageois aux peuples des
cités-États (X e–III e millénaire av. J.-C.), Paris.
Kienast, B. and Sommerfeld, W. (): Glossar zu den altakkadischen Königsinschriften (= FAOS ),
Stuttgart.
Krebernik, M. (): “Philologische Aspekte elamisch-mesopotamischer Beziehungen im Überblick.”
In Kogan, L., Koslova, N., Loesov, S. and Tishchenko, S. (Eds.), Babel und Bibel : –.
Krecher, J. (): “Die marû-Formen des sumerischen Verbums.” In Dietrich, M. and Loretz, O.
(Eds.), Vom Alten Orient Zum Alten Testament. Festschrift für Wolfram Freiherrn von Soden zum .
Geburtstag am . Juni  (= AOAT ), Kevelaer and Neukirchen-Vluyn: –.
Lafont, B. (): “A propos de la ville de Suse et d’un fragment d’enveloppe,” RA : –.
Lambert, M. (): “Objets inscrits du Musée du Louvre,” RA : –.
———. (): “Le prince de Suse Ilish-mani et l’Elam de Naramsin à Ibîsin,” JA : –.
Leemans, W.F. (): Foreign Trade in the Old Babylonian Period as Revealed by Texts from Southern
Mesopotamia, Leiden.
Malbran-Labat, F. (): Les inscriptions royales de Suse. Briques de l’époque paléo-élamite à l’Empire
néo-élamite, Paris.
Marchesi, G. (): “Who Was Buried in the Royal Tombs of Ur? The Epigraphic and Textual Data,”
OrNS : –.
Michalowski, P. (): The Royal Correspondence of Ur (PhD diss., Yale University), Ann Arbor.
———. (): The Lamentation over the Destruction of Sumer and Ur (= MC ), Winona Lake.
———. (): “Memory and Deed: The Historiography of the Political Expansion of the Akkad State.” In
Liverani, M. (Ed.), Akkad, the First World Empire. Structure, Ideology, Traditions (= HANES ), Padua:
–.
———. (): “The Ideological Foundations of the Ur III State.” In Meyer, J.-W. and Sommerfeld, W.
(Eds.),  v. Chr—Politische, wirtschaftliche und kulturelle Entwicklung im Zeichen einer Jahrtausend-
wende (= CDOG ), Saarbrücken: –.
Mittermayer, C. and Attinger, P. (): Altbabylonische Zeichenliste der sumerisch-literarischen Texte (=
OBO Sonderband), Fribourg and Göttingen.
Nasrabadi, B.M. (): “Ein Steininschrift des Amar-Suena aus Tappeh Bormi (Iran),” ZA : –
.
Owen, D.I. (): “The Royal Gift Seal of Silluš-Dagan, Governor of Simurrum.” In Graziani, S. (Ed.),
Studi sul Vicino Oriente antico dedicati alla ˙ memoria di Luigi Cagni, Napoli: –.
Pittman, H. (): “Susa.” In Meyers, E.M. (Ed.), The Oxford Encyclopedia of Archaeology in the Near
East, vol. , New York and Oxford.
Potts, D.T. (): The Archaeology of Elam. Formation and Transformation of an Ancient Iranian State,
Cambridge.
Potts, T. (): Mesopotamia and the East. An Archaeological and Historical Study of Foreign Relations
ca. –BC, Oxford.
Quintana, E. (): Historia de Elam, el vicino mesopotàmico, Murcia.
Sigrist, M. and Gomi, T. (): The Comprehensive Catalogue of Published Ur III Tablets, Bethesda.
Sollberger, E. (): The Business and Administrative Correspondence under the Kings of Ur (= TCS ),
Locust Valley.
Steve, M.-J. (): “Elam: histoire continue ou discontinue?” In De Meyer, L. and Gasche, H. (Eds.),
Mésopotamie et Elam. Actes de la XXXVIème Rencontre Assyriologique Internationale, Gand –
juillet  (= MHEOP ), Ghent.
Stolper, M.W. (): “Political History.” In Carter, E. and Stolper, M.W., Elam. Surveys of Political History
and Archaeology (= NESt. ): –.
Vallat, F. (): “La date du règne de Gudea,” NABU /.
Wilcke, C. (): “Der Kodex Urnamma (CU): Versuch einer Rekonstruktion.” In Abusch, T. (Ed.),
 gianni marchesi

Riches Hidden in Secret Places. Ancient Near Eastern Studies in Memory of Thorkild Jacobsen, Winona
Lake: –.
Wilcke, C. (): “Sumerian: What We Know and What We Want to Know.” In Kogan, L., Koslova,
N., Loesov, S. and Tishchenko, S. (Eds.), Language in the Ancient Near East. Proceedings of the e
Rencontre Assyriologique Internationale, vol. /I (= Babel und Bibel /), Winona Lake: –.
Woolley, C.L. (): The Buildings of the Third Dynasty (= UE ), London and Philadelphia.
 ur-nammâ(k)’s conquest of susa 

Fig. : CBS  Fig. : CBS 

Fig. a: CBS  Fig. b: CBS , detail

 PUZUR-INŠUŠINAK AT SUSA:
A PIVOTAL EPISODE OF EARLY ELAMITE HISTORY RECONSIDERED*

Piotr Steinkeller**

In memory of Edith Porada, whose eye was matched

only by her sense of class

. Introduction

This paper has a long history, since it goes back to a presentation I gave in the fall of  at
the “Columbia University Seminar for the Archaeology of the Eastern Mediterranean, Europe,
and the Near East,”1 which was run by Edith Porada for nearly thirty years. As in that Columbia
presentation, which was entitled “The Date of Gudea and Others,” and which never saw the
printer’s press, the task I face today is to make some sense of the events that took place at the
very beginning of the Ur III period, that is, the time of Utu-hegal, Gudea, and Ur-Namma.
There is yet another individual who played an exceedingly important role in that chapter of
ancient Near Eastern history, and that person was Puzur-Inšušinak of Susa.2

. Puzur-Inšušinak’s Beginnings

According to the Old Babylonian historical tradition, as it is reflected in the king-list of the
rulers of Awan and Šimaški,3 Puzur-Inšušinak4 was the last, i.e., th, king of the Awan dynasty.
All of the inscriptions that Puzur-Inšušinak has left to us come from Susa. In those sources, his
standard (and apparently earlier) titles are those of the “governor of Susa (and) the general of
the land of Elam, son of Šimpi-išhuk.”5
The origins of Puzur-Inšušinak are still obscure. Given the fact that his father’s name is
Elamite (Šimpi-išhuk), he must have been of Elamite origin. However, it is unlikely that Puzur-
Inšušinak was a descendent of the Awan dynasty, and that, therefore, he stemmed from Awan.
I will talk more about his relationship to Awan later on.

* I wish to offer my warm thanks to Jason Ur, who was kind enough to design the two maps accompanying this

paper (figs.  and ).

** Harvard University.
1 “The Date of Gudea and Others,” November , .
2 Among the earlier discussions of Puzur-Inšušinak, see especially André/Salvini , Potts : – and

André-Salvini –: –. It is interesting to note that, at the very discovery of Puzur-Inšušinak’s existence,
V. Scheil (MDP : ), described him as “une sort de Goudéa susien.”
3 For the most recent discussion, see Steinkeller  and in press.
4 Since the so-called “linear Elamite” writing remains undeciphered, the practice of reading Puzur-Inšušinak’s

name as “Kutik-Inšušinak,” which has been adopted by some scholars following W. Hinz’s suggestion, is unwarranted.
Here note that, in the Old Babylonian Susa, the name was understood as Puzur-Inšušinak. This is demonstrated by
the writing Puzur(MAN)-dInšušinak in the list of the kings of Awan (RA  []  line ). For MAN = puzru,
see pu-zur MAN = pu-zu-ru (var.: pu-uz-[ru] (EA II  = MSL , ); pu-zur MAN = pu-uz-rum (A II/= line
 = MSL , ); and the uses of MAN for puzru in extispicy sources from Susa (CAD P, b).
5 ÉNSI Šušinki ŠAGINA ma-ti Elamki DUMU Šim-bi-iš-hu-uk (Gelb/Kienast : –, Elam , , , , , ,

 [only énsi of Susa],  [only énsi of Susa]).

 piotr steinkeller

The fact that Statue B of Gudea and the prologue to the “Code of Ur-Namma,” which allude
to the conflict with Puzur-Inšušinak,6 both mention Anšan in this connection, raises a distinct
possibility that he actually came from Anšan. If Puzur-inšušinak indeed was an Anšanite,7 we
would have to assume that he adopted his personal name only after he had come to Susa, in
recognition of the local cult of Inšušinak.
The only certain fact we know about Puzur-Inšušinak’s early career is that he had somehow
been able to establish himself as a ruler of Susa.8 Some time after that event, he launched a
massive campaign in the Zagros, from as far as Huhnuri in the east9 to as far as the Hamadan
plain in the northwest. The high point of this campaign, which is described in his victory
inscription,10 was the capture of the lands of Kimaš and Hurti.11 As Puzur-Inšušinak emphasizes
in that source, Kimaš and Hurti were his main opponents—and apparently also his main
target—during these military operations. As we shall see in the following, it was undoubtedly
the possession of these two localities that enabled him subsequently to move into and conquer
the Diyala Region and northern Babylonia.
But, before I discuss that next phase of Puzur-Inšušinak’s conquests, I need to digress
on the location of Kimaš and Hurti, since this problem is of crucial importance for the
proper understanding of the historical geography of the central Zagros at the end of the third
millennium. As I demonstrate in detail in the Appendix to this paper, these two places can quite
confidently be located along the Great Khurasan Road, in the general vicinity of the modern
towns of Islamabad-e Gharb and Khermanshah.
The question of Kimaš and Hurti must be considered within the broader issue of the Ur III
foreign policy,12 which had largely been formulated during the reign of Šulgi. As is well known,
Šulgi campaigned in the northeastern periphery of Babylonia for more than thirty years. It is
clear that the primary objective of this incessant military activity was to obtain for Babylonia
an access to and the control of the Great Khurasan Road. The main obstacle in meeting this
objective were the lands of Šimurrum and Lullubum, which were situated in the upper valley of
the Sirwan River and the region of modern Sulaymaniyah respectively. These two lands blocked
Babylonia’s access to the Great Khurasan Road, thus frustrating Šulgi’s plans. More than ten

6 See below pp. –.

7 Such a possibility may find support in the use of the “linear Elamite” writing in some of Puzur-Inšušinak’s
inscriptions. This writing is otherwise attested only on the pieces from the Iranian plateau, and thus can assuredly
be considered a native Elamite phenomenon. See most recently Potts : – and Dahl : –.
Unfortunately, too little is known about this writing to make any historical inferences from its use by Puzur-
Inšušinak. There is certainly no evidence at this time to think that this writing was “invented” by Puzur-Inšušinak
(as it has been speculated by some authors).
8 Frayne : , attempted to restore the name of Puzur-Inšušinak in an inscription of Puzur-Mama, a late

Sargonic ruler of Lagaš: [PÙ.Š]A-[dMÙŠ.ER]EN / [MÙŠ].ERENki (i0 0–0). While it is certain that i0 0 names Susa,
a restoration of Puzur-Inšušinak’s name in the preceding line is impossible. First, the sign read by Frayne as ŠA
definitely is not ŠA; cf. ŠA ibid. iii0 0. Second, the last sign in i0 0 cannot be EREN, since it is different from EREN
in the following line.
9 Thanks to a recently published inscription of Šu-Suen from Iran, which describes the conquest of Huhnuri

in year Amar-Suen , Huhnuri can now confidently identified as the site of Tappeh Bormi near Ramhormoz. See
Mofidi-Nasrabadi : –.
10 MDP , – = Gelb/Kienast : –, Elam . The lower section of a seated statue of Puzur-Inšušinak.

The total height of the surviving fragment is cm; the height of the socle is cm. The same inscription is also
recorded on two fragments of a stele, which were stolen during the transportation from Susa to the Persian Gulf. See
MDP , – and MDP , –.
11 ì-nu-me Ki-maški ù ma-at Hu-úr-timki i-ge-ru-uš (MDP , – = Gelb/Kienast : –, Elam :–).

Huhnuri (written Hu-hu-un-rí ki) is mentioned in line .

12 See, in detail, Steinkeller, The Grand Strategy of the Ur III Empire: Babylonia’s Foreign Policy and Territorial

Expansion at the End of the Third Millennium bc (a monograph; in preparation). This monograph builds on my
paper “The Grand Strategy of the Ur III Empire: Exquisite Design, Perfect Failure,” which was presented at the th
RAI, Würzburg, July , .
 puzur-inšušinak at susa 

separate campaigns had to be launched against these two lands before they were finally subdued.
Once this had been achieved—in year Šulgi —Šulgi was then free to accomplish his main and
ultimate strategic objective, which was the possession of Kimaš and Hurti, the two critical points
in charge of the Great Khurasan Road along its passage through the Western Zagros onto the
Hamadan plain.
The conquest of Kimaš and Hurti, which took three years to complete, was the crowning
event of Šulgi’s reign. Significantly, it is the only foreign expedition mentioned in Šulgi’s
historical inscriptions.13 This fact underscores the importance that this campaign had to Šulgi
and his policies. In this connection, it may not be by accident that the wording of this inscription
is almost identical to that of Puzur-Inšušinak’s victory stela, which celebrates that ruler’s victory
over Kimaš and Hurti. Thus, Šulgi’s victory over Kimaš and Hurti was a reversal of Puzur-
Inšušinak’s deed, in that it opened up the Elamite highlands to Babylonia, as well as the final act
of settling the accounts with the Susian, which had been began by his father Ur-Namma. We
may be absolutely certain, therefore, that it was the conquest of Kimaš and Hurti that made it
possible for Puzur-Inšušinak to move into the Diyala Region and, subsequently, into northern
Babylonia as well. According to one of Ur-Namma’s inscriptions, which describes his conflict
with Puzur-Inšušinak, the latter occupied the cities of Awal, Kismar, and Maškan-šarrum, and
the lands of Ešnuna, Tutub, Zimudar, and Akkade.14 The prolog to Ur-Namma’s code adds to
list the northern Babylonian cities of Marda, GIRkal, Kazalu, and probably Akšak?, plus their
rural settlements.15 See fig. , which illustrates the sequence of Puzur-Inšušinak’s conquests.
Of those places, Awal, Maškan-šarrum, and Kismar are of particular interest, since these
three localities had exceptional strategic importance. Awal, modern Tell es-Suleimeh, was
located at the point where the Diyala River cuts through the Jebel Hamrin, ca.  km downstream
from Sa"adiya.16 Maškan-šarrum, on the other hand, lay on the Tigris, at the mouth of the Lower
Zab, representing the southernmost point of Assyria.17 As for Kismar, which was a relative
neighbor of Maškan-šarrum, it probably was situated on the opposite bank of the Tigris.18 In this
way, these three places controlled much of the traffic between the lower section of the Diyala
Region and the points north of the Jebel Hamrin and Makhoul ranges. The importance of these
localities is reflected in the fact that, during the reign of Šu-Suen, Maškan-šarrum and Awal
were both governed by Babati, the brother of queen Abi-simiti, and one of the most important
men of the realm.19

13 See below p. .

14 Frayne : –, Ur-Namma  v0 0–0.
15 Frayne : –, Ur-Namma :–.
16 See Steinkeller : –.
17 See ibid.: . See also Steinkeller a: . To the data cited there add the year-formula Samsu-iluna  (A-

ma-alki Ar-ku-umki Maš-gán-LUGALki bàd-da níg-gul-la etc.), which indicates the proximity of Maškan-šarrum to
Awal. Further, note the following passage from an inscription of Tukulti-Ninurta I, where Maškan-šarrum is linked
with Šišil, a well-known Diyala locality: be-ri-it URU Šá-ši- la ù URU Maš-hat-MAN e -be-er-ti Za-be šu-pa-li,
“(the region) between the towns Šašila (and) Mašhat-šarri (= Maškan-šarri) on˘ the
˙ opposite bank of the Lower Zab”
(Grayson : –, Tukulti-Ninurta  iv –). ˘ ˙ For this passage, see Wilcke :  and n. . Finally, note
an unpublished Ur III tablet from Urusagrig (courtesy of D.I. Owen), which lists barley allotments for the “soldiers
of Awal, Kismar, Maškan-šarrum, and Šurbu” (še-ba éren ki en-nu-gá A-wa-alki Ki-is-marki Maš-gán-LUGALki ù
Šu-úr-buki). Šurbu, which was part of the gún ma-da belt (see AUCT   ii –:  udu ú  máš gal ú àga-ús lú
Šu-úr-buki-me<-éš> ugula Da-hi-iš-a-tal; YOS   ii –:  lá  udu  máš éren Šu-úr-buki; PDT  :–
: the tax of [Šu-úr]-buki, followed by that of Der), was situated in the Diyala Region, in an area bordering on the
territory of Der. See “Sargon Geography” line : ultu Šur-bu adi Ib-rat KUR Derki, “from Šurbu to Ibrat (is) the land
of Der” (Grayson –: ). For the location of Ibrat, see most recently K. Volk in George : –.
18 See Steinkeller : .
19 His various titles were as follows: pisan-dub-ba, šà-tam lugal, šagina Maš-gán-LUGALki, énsi A-wa-alki ù A-pi-

akki, sabra nin -a-bi, sanga dBe-la-at-šuh-nir ù dBe-la-at-ter-ra-ba-an (Frayne : –, Šu-Sin  and ).
 piotr steinkeller

An independent proof of Puzur-Inšušnak’s possession of the Diyala region is provided by

the fact that one of his inscriptions is dedicated to the goddess Belat-Teraban.20 That goddess,
together with her companion Belat-Šuhnir, had her home in the Diyala Region, probably in the
city of Ešnuna itself.21 Here it may be noted that the aforementioned Babati also bore the title
of the temple administrator (sanga) of Belat-teraban and Belat-šuhnir.22
We can be quite confident that it was as a consequence of all these conquests combined,
in particular, the capture of northern Babylonia and the city of Akkade, that Puzur-Inšušinak
assumed the titles of the “powerful one” (da-núm) and “the king of Awan” (LUGAL A-wa-anki).23
These titles are attested only in the inscriptions describing the construction of a monumental
stone stairway by Puzur-Inšušinak; in fact, these texts are inscribed on the individual steps of
that staircase.24 According to those particular inscriptions, Inšušinak looked at Puzur-Inšušinak
with favor, and then “gave him four quarters to rule.”25
As is well known, the rule over the four quarters of the world had earlier been exercised by
Naram-Suen, who, in fact, was the first Near Eastern monarch to make such a claim. It was
also Naram-Suen who for the first time used the title of the “powerful one.” We will be justified
in assuming, therefore, that it was as a result of his capture of Akkade that Puzur-Inšušinak
felt entitled to adopt these two designations for himself. To those he added the ancient and
prestigious title of the king of Awan, the highest distinction an Elamite could aspire to.
The Awan dynasty had likely been terminated during the Sargonic period, probably by
Naram-Suen. This would make Hita, the penultimate king of Awan (no. ), a contemporary
of Naram-Suen. Accordingly, one should probably conclude that, subsequent to Hita’s reign,
the title of the king of Awan had not been used for several decades, perhaps even a century
(depending on how one reconstructs the chronology of the post-Sargonic period), until it was
revived by Puzur-Inšušinak.26
The location of Awan is still unknown. This problem has recently been studied by P. Micha-
lowski (: ). It is characteristic that, with one notable exception, Awan is never named
in Ur III documentation. This shows that, by the Ur III period, Awan no longer was a viable,
living toponym. The only exception here are the historical sources of Ibbi-Suen dealing with his
campaign against Susa and AdamDUN in his “fourteenth” regnal year, where Awan is named
together with AdamDUN:27 A-dam-DUNki ma-da A-wa-anki-ka.28 Michalowski has made a very
important observation that the phrase in question is to be translated “AdamDUN of the land of
Awan,” and not “AdamDUN and the land of Awan,” as it had been interpreted previously. This
led him to suggest that AdamDUN was the main city of Awan.

20 Gelb/Kienast : –, Elam :.

21 See Sallaberger : –. Further, note the seal of Šu-iliya, an early OB ruler of Ešnuna (here referred to as
ma-at Wa-ri-im), who calls himself a “beloved of Tišpak, Belat-Teraban, and Belat-šuhnir” (Frayne : –,
).
22 See above n. .
23 Note that in the victory inscription (Elam ) and the dedication to Belat-Teraban (Elam ), Puzur-Inšušinak

still uses the (more modest) titles of the governor of Susa and the general of Elam.
24 a-na (dInšušinak) be- lí -[su] Puzur -dInšušinak da-núm LUGAL A!-wa-anki DUMU Šim-bi-iš-hu-uk (Gelb/
4
Kienast : –, Elam :–, Elam :– and André/Salvini : , Sb  // Sb  lines –).
25 in  MU dInšušinak ib-ba-al-su-šum ki -ib-ra-tim ar -ba-im a-na ša?-x-[x] i-ti-nu GIŠ.KUN (NA ) i-bu-uš
5 4
(Gelb/Kienast : –, Elam :–; André/Salvini : , Sb  // Sb  lines –). The reading and
meaning of line  ( a-na ša?- x -[x]) is obscure; the signs should be collated.
26 For this conclusion, see already André/Salvini : .
27 Civil (: –), has suggested for this toponym a possible reading A-dam-šáh. However, as being based

on circumstatial evidence, this reading is far from conclusive. Cf. also M. Krebernik : –.
28 Frayne : –, Ibbi-Sin :–, – Ibbi-Suen :–; year-formula Ibbi-Suen “” (UET  

rev. ii –, :–).

 puzur-inšušinak at susa 

As for the location of AdamDUN, various textual data indicate that this place is to be sought
to the east of the Susiana. A strong argument in favor of this localization is the fact that the
inscription of Gudea describing his building activity in AdamDUN appears to have been found
at the site of Tépé Surkhegan, which is situated in the vicinity of Šūštar.29
Another datum supporting this localization is the fact that AdamDUN could be reached
by boats from southern Babylonia. Thus, a tablet from Ur refers to a shipment of barley by
boat to AdamDUN.30 Similar conclusion is reached by the comparison of two Lagaš tablets
from the reign of Gudea, in which the same official (Lu-NINA) once receives a shipment of
timber from AdamDUN, and, another time, a shipment of timber that arrived from Guabba
through “long-distance trade” respectively.31 The obvious inference must be that both of these
shipments originated in AdamDUN, from where the timber was transported, apparently over a
waterway, to Guabba, which was Lagaš’ main seaport on the Persian Gulf. From Guabba, it was
then carried on boats to Lagaš itself.32 If AdamDUN is in fact Tépé Surkhegan, the waterway in
question must have been the Karun, which flows right by Šūštar.33
Assuming that, as argued by Michalowski, AdamDUN was situated in the land of Awan, we
may tentatively conclude that Awan denoted the southeastern portion of the modern province
of Khuzistan. This conclusion finds support in the fact that Awan was not too far from Susa,
since, in the inscriptions of Rimuš, one of Rimuš’ battles with Marhaši is said to have taken
place “between Awan and Susa, on the ‘Middle River’ (= the Karun?).”34

. Puzur-Inšušinak and the Gutians

It is quite certain that the foreign conquests of Puzur-Inšušinak coincided broadly with the reign
of Utu-hegal of Uruk. We do not know when precisely did Puzur-Inšušinak occupy northern
Babylonia, of course. But it is a good guess that, throughout Utu-hegal’s reign (of eight years),
northern Babylonia remained under Puzur-Inšušinak’s control.35 It is equally clear that the
occupation of northern Babylonia by Puzur-Inšušinak must have weakened the position of the
Gutians, who at that time resided at Adab, and who may actually have ruled over northern

29 See below p. .

30  dug mangaga má še A-dam-DUNki-šè ki dNam-ha-ni ugula má-lah4 šu ba-ti (UET  :–; Ibbi-Suen
/vi).
31  pieces of timber gìr Eren-[da(-ni)] sagi Lú-NINA[ki] šu ba-ti A-dam-DUNki-ta (RTC :–; Gudea “” =

mu é dGá-tùm-dùg ba-dù-a); , pieces of timber giš nam-ga-eš8 Gú-ab-baki-ta Lú-NINAki šu ba-ti (RTC :–;
undated).
32 Here note that timber was delivered from AdamDUN to Lagaš also in Ur III times. See ITT   and Nesbit

: .
33 Identical conclusions about the reverain traffic between southern Babylonia and the Susiana have indepen-

dently been reached by Bertrand Lafont : – (which became available to me only after this article had
been completed), Lafont demonstrates, using a completely different set of data, that, in Ur III times, boats were navi-
gated from Girsu/Lagaš to Susa and AdamDUN. See, especially, the following statement: “Ainsi, à partir de Gu"abba,
au débouché du grand canal reliant Girsu, Lagaš, Kinunir/Nigin et Gu"abba, il faut imaginer que les bateaux circu-
laient dans les eaux du Golfe ou dans les marais jusqu’à l’embouchure du Karun, puis remontaient ce fleuve jusque
vers l’ actuelle Ahwaz, pour emprunter la Karkheh jusqu’à Suse.” Lafont’s findings had been anticipated by Gasche
:  n. , and, even earlier, by Leemans : .
As a result of the recent researches by S. Cole and H. Gasche, the ancient coast line of the Persian Gulf and the
courses of the Karun and Karkheh rivers can now be reconstructed with great precision. See Gasche : – (esp.
pp. – fig. ). In a personal communication, Hermann Gasche tells me that the Karun is fully navigable all the
way to Šūštar.
34 in ba-rí-ti A-wa-anki ù Su-si-imki in ÍD gab-lí-tim (Frayne : –, Rimuš :– and –, Rimuš :–

).
35 Of special importance here is the fact that there is no evidence of Utu-hegal’s having controlled any territories

in northern Babylonia.
 piotr steinkeller

Babylonia before its conquest by Puzur-Inšušinak.36 We must consider, therefore, that the
involvement of Puzur-Inšušinak in Babylonian affairs was one of the reasons why Utu-hegal
defeated Tirigan, the last Gutian ruler, so easily and with such an incredible swiftness.
First of all, Puzur-Inšušinak’s conquests in the Zagros would have weakened the position of
the Gutians in that region. It is even conceivable that Puzur-Inšušinak had taken possession of
the Gutian homeland. Furthermore, if the Gutian presence in northern Babylonia had, thanks
to Puzur-Inšušinak’s invasion, been eliminated, then what was left to the last Gutian kings was
the state of Adab and some of the adjoining territories. This would have made their position
very precarious indeed.

. Puzur-Inšušinak and Ur-Namma

Following Utu-hegal’s brief reign, the crown of Uruk was assumed by Ur-Namma of Ur, who
probably was Utu-hegal’s younger brother. The chronology of Ur-Namma’s reign, which lasted
eighteen years, is known very poorly. Even more to the truth, we know nothing certain about
it. However, since the majority of his inscriptions name him a king of Sumer and Akkad, it is
clear that Ur-Namma became the master of northern Babylonia relatively early in his reign.
Accordingly, his conflict with Puzur-Inšušinak should probably be dated to the very beginning
of his reign. As one can gather from the so-called “Cadaster of Ur-Namma,”37 Ur-Namma
recovered from Puzur-Inšušinak entire northern Babylonia. It is likely that he also regained
the control of the Diyala Region, but the evidence here is lacking as yet.
We also now know that Ur-Namma conquered Susa.38 Since the year-formulae of Šulgi and
his inscriptions make no references to any military actions in the Susiana = Khuzistan, chances
are that this entire region, with such cities as Urua, Sabum, and Pašime, had already been
conquered and incorporated into the Ur III state by Ur-Namma.

. Gudea’s Participation in the War on Puzur-Inšušinak

Various data indicate very persuasively that Gudea of Lagaš too was part of the war on Puzur-
Inšušinak.39 The most likely assumption is that Gudea and Ur-Namma had formed a military
alliance against Puzur-Inšušinak. However, we have no direct confirmation of this fact so far.
We also do not know the exact chronology of the events. One might envision, for example, that
the expulsion of Puzur-Inšušinak from northern Babylonia came first, to be then followed by an
expedition to the Susiana and the territories farther east, in which both Ur-Namma and Gudea
took part. But, of course, it is equally possible that the course of events was just the opposite.
And there are still other scenarios to be considered.
As for the evidence on Gudea’s participation in the operations against Puzur-Inšušinak, there
is, first of all, the laconic passage from Statue B vi , according to which Gudea “smote the city

36 For a recent revaluation of the Gutian period, see Steinkeller, “The Gutian Period in a Chronological Perspec-
tive,” to appear in a collection of papers on third millennium chronology, edited by W. Sallaberger and I. Schrakamp
(forthcoming).
37 See Frayne : –, Ur-Namma . Two new OB mss. of this document have recently been published by

this author in George : –.

38 See Marchesi in this volume, “Ur-Namma’s Conquest of Susa.”
39 For the earlier attempts to link Gudea with Puzur-Inšušinak, see Steinkeller a: – and Vallat .

That Gudea and Ur-Namma were allies has recently been argued also by C. Wilcke. See his edition of a new bilingual
inscription of Gudea (MS ) in George : –.
 puzur-inšušinak at susa 

of Anšan and (the land of) Elam” (or “cities of Anšan and Elam”).40 This information is probably
corroborated by a partly preserved year-formula, which appears on a tablet belonging to the
Gudea dynasty.41
That Gudea did in fact campaign in the east is confirmed by one of his inscriptions.42
According to that source, which allegedly was found at the site of Tépé Surkhegan,43 situated
ca. km southwest from Šūštar, Gudea built, evidently in AdamDUN itself, a structure of some
sort for the “mistress of AdamDUN.” Unfortunately, due to the poor quality of the hand-copy,
neither the name of the goddess nor the name of her building can be read with confidence.44
The obvious implication of this building activity is that Gudea had conquered AdamDUN
at one point during his reign. Gudea’s control of AdamDUN is confirmed by a tablet from
his reign, which records a delivery of timber from AdamDUN.45 Still another indication of
this control is the fact that a Lagaš tablet from the reign of Ur-Ningirsu II, Gudea’s son and
immediate successor, records an expenditure of garments to an unnamed ensi of AdamDUN.46
Another proof of Gudea’s conquests in the east are the Lagaš tablets MVN   and ,
which, based on the prosopographic data contained in them,47 may be dated to the reign of
Gudea. These two documents record barley rations, which were distributed among Elamite
women and children from the locality called Elanir or Elašir (Elam, lú E-la-NIRki-me).48
Without any doubt, these individuals had been brought to Lagaš as prisoners-of-war, almost
certainly in connection with Gudea’s forays into Elam.
However, the most important evidence for Gudea’s participation in the war against Puzur-
Inšušinak is provided by the texts RTC  and MVN  . These two tablets, which duplicate
each other, record expenditures of barley to the gangs of workers or soldiers. Summarized in
the total (col. iv) as the “Lullubeans,” these men include various foreigners, among them the
men of Šimurrum, a Šimurrean smith (simug), the men of Huhunuri, and the Lullubeans. Of
special importance here is the presence in this group of four “sons of Šimbi-išhu” (ii –). One
recognizes, of course, in Šimbi-išhu the name of Puzur-Inšušinak’s father. Accordingly, the four
men in question were either Puzur-Inšušinak’s brothers or his kinsmen.49

RTC  + MVN   (distribution of lines follows RTC )

40 gištukul uru An-ša-an Elamki mu-sìg.

41 [mu An-ša-anki?] gištukul ba-sìg-a (NFT AO  left edge). See Steinkeller a:  n. .
42 Steve : –.
43 Ibid.: –. Potts (: –) has recently questioned the veracity of this information. However, Hermann

Gasche, who accompanied Steve when the tablet in question was reported to them, assures me that the facts were
exactly as described by Steve, op. cit., –. In Gasche’s words: “I was with Steve in  (Hubert mentioned in Steve’s
article was the photographer). The son of the owner of the tablet, a local teacher who had found it on the tell (which
is located about  m north of the village of Surkhegan or Qal’eh Surkheh) gave us the information at that time.
We then went to the tell, and recognized its precise location on the /, maps of the Iranian Oil Operating
Companies (dated ). There is no doubt about the location; it corresponded precisely to what the son of the owner
had said. You can see the tell on Google, but the village is bigger than in , and part of the site, as elsewhere in
Khusistan, has been partially leveled for agricultural purposes. Houses are now very close to the ‘Acropolis’; they are
on top of the remains Steve interpreted as a possible ‘ville basse’” (e-mail letter, June , ).
44 d[x(-x)]- AB? / [n]in! A-dam-DUNk[i] / [n]in-a-ni / x-x -ka-ni / mu- na -dù (lines –). Contrary to Steve’s

claim, ibid., , it is by no means certain that line  names Nanše.

45 See RTC :–, cited above p.  and n. .
46  garments énsi A-dam-DUNki zi-ga dŠára-ì-šag (RTC :–; mu Ur-dNin-gír-su énsi mu ús-a-bi = Ur-
5
Ningirsu ).
47 Both of them are designated as zi-ga Šu-na, “expenditures of Šu-na,” and therefore they belong to Gudea’s time.

Here note that two of the tablets with the “zi-ga Šu-na” rubric (MVN   and ) bear a Gudea date (mu mi-ì-tum
sag-ninnu ba-dím = Gudea “”).
48 In MVN  : the toponym is apparently spelled É-la-NIRki (transliteration only).
49 See Steinkeller a:  n. .
 piotr steinkeller

i ) […]
) [x]-ga-ar
) [x] guruš .. gur-ta
) x dumu ..-ta
) še-bi .. gur
) .. giš-kin-ti
) .. I-sar-GIM
) še-bi .. gur
) lú Si-mu-ru-um-me
) .. dumu A-ba-bí-um
) .. Lú-dNin-gír-su
) .. Á-bí-lí
) še-bi .. gur
) ki ha-za-núm
)  guruš .. gur-ta
) [] ama ..
) še-bi .. gur
) [lú L]u?-lu-a-me
ii ) […]
) libir-me
)  guruš .. [gur-ta]
)  guruš .. gur
) še-bi ..
) dumu Šim-bí-iš-hu (MVN  ) / Šim-bí-iš-me (RTC )
) .. dumu Ú-ki-iš
) .. DINIGIR-su-dan
) .. ama Bu-zu-zu
) .. La-gi-bu-um
) šeš A-ba-an-hu-me
) .. simug Si-mu-ru-um
) ugula Da-da-a
) .. Šu-e-li lú-didli
) .. NE-zi-a
) ugula Dul4-ga-an
) .. Na-dar-ri
) Zi-bí-rí(-lí in MVN  )
) lú Hu-hu-ri-me
iii ) .. lá .. šeš Maš
) .. Za-an-ga-dar
)  guruš ..-ta
)  dumu ..-ta
)  ugula .. gur
) .. Ar-ši-ah
) še-bi .. lá .. gur
) ugula Ù-rí-gi4-ar
)  lá  guruš ..-ta
)  dumu ..-ta
)  dumu .  sìla-ta
)  ugula ..-gur
) še-bi .. gur
) ugula LAK-(= GADA.TAG4.SI)-zé-a
) Lu-lu-bu-um-me
)  guruš .. gur
 puzur-inšušinak at susa 

) ugula Ù-áš-ti-a

) .. Zé-ga
) .. La-ši-in-a
iv ) ugula dNanše-ul4-gal
) .. Ú-uh-NI lú Pù-pù
) ugula Ur -ab-ba
) .. Bur-ra
) .. šeš Da (or: Á )-an-har-ra(-ri in MVN  )
) .. NI-gu-gu
(space)
) šu-nígin-nígin .. še gur A-ga-dèki
) Lu-lu-bu-um-me
) [ez]en-munu4-kú
(space)
) […]- x

Based on various prosopographic data, these two texts can securely be dated to Gudea’s reign.
Thus, the man by the (foreign) name of Zangadar (iii ) is mentioned in two other texts, one of
which can be assigned to Gudea’s reign through prosopographic evidence.50
It is striking that the Lagaš documents dating to the time of Gudea’s dynasty mention surpris-
ingly many foreigners, especially easterners. Among those we find various military personnel
and officials from Susa, AdamDUN, Huhnuri, Marhaši, Anšan, Kimaš, Gutium, Šimurrum,
Lullubum, and Zimudar.51 Included among them there is also an ensi of Gasur,52 and even
men from the far-away Syrian Tutul.53 The presence of all these foreigners in Lagaš at that
particular moment in time is puzzling. An assumption that they came (or were brought)
there in connection with the war on Puzur-Inšušinak would provide a plausible explana-
tion.
These facts have also important implications for chronology, namely, the question of the date
of Gudea vis-à-vis the reign of Ur-Namma. This is not the place to delve into this exceedingly
complicated problem.54 Nevertheless, one fact appears to be absolutely certain: unless Puzur-
Inšušinak’s occupation of northern Babylonia had been terminated, it would not have been
possible for Gudea to obtain for the Eninnu project all the foreign materials he claims to
have brought from Syria (Amanum, Jebel Bishri, Ebla, Uršu, Madga = modern Hit on the

50 In MVN  :– (undated), Za-an-ga-dar receives two garments. In MVN  :– (undated), he receives

.. of barley, with Lu-NINA acting as a requsitioner, and Šu-na being the disbursing party (Lú-dNINAki maškim
ki Lú-NINAki-ta zi-ga Šu-na). Two of the tablets bearing the “zi-ga Šu-na” rubric (MVN   and ) use a year-
formula of Gudea (mu mi-ì-tum sag-ninnu ba-dím(-ma) = Gudea “”). The Lu-NINA appearing in MVN   is
plausibly identical with his namesake who is known from the documents dating to Gudea’s reign. See above n. .
Further, note MVN  , in which both Lu-NINA and Šu-na are listed.
Another prosopographic link is likely provided by Bur-ra (iv ). See NN šeš Bur-ra, who appears in a tablet from
Gudea’s reign (MVN  :; mu šita sag-ninnu ba-dím-ma = Gudea “”). Cf. also Ù-rí-gi4-ar (iii ), who is also
mentioned in RTC : ( gišbanšur Ù-a-rí-gi4-ar).
51 lú Šušinki (MVN  :); rá-gaba A-dam-DUNki (RTC :); Si-im-hu-zi- a? énsi Hu-hu-nu-ri (RTC :);

NN énsi Hu-hu-nu-riki … lú-kin-gi4-a Mar-ha-šiki … àga-ús Elam Mar-ha-šiki-ta (RTC :–); eme-bala … lú
Mar-ha-šiki … lú Hu-hu-nu-riki … eme-bala-me (MVN  :–); lú An-ša-na (RTC  iii 0; MVN  :,
:); ugula eme-bala-me … lú Si-mu-ru-umki (RTC :–); Lu-lu-bu-um (MVN   i ); lú Ki-maški (MVN
 :; RTC :0, :0, :); Gu-ti-um (RTC :; MVN  :); dumu Zi-mu-dar (MVN  :), lú-tur
Zi-mu-dar (MVN  :).
52 NN énsi Ga-súrki (RTC :).
53 lú Du-du-ulki (RTC :), lú Du -du -ulki-me (MVN  :).
11 11
54 This issue has recently been re-examined by C. Wilcke, with a conclusion that Gudea’s tenure coincided with

the reign of Ur-Namma, perhaps even extending into Šulgi’s reign. See his contribution to George : –.
 piotr steinkeller

Euphrates),55 and the Zagros mountains (Kimaš). For the same reasons, it also follows that those
foreign acquisitions must have been later than the removal of Tirigan by Utu-hegal. Here it is
highly significant that both Ur-Namma and Gudea claimed, using virtually the same language,
to have opened up trade routes from the south to the north.56
Similar chronological implications has the fact that, during Gudea’s reign, Lagaš was involved
in commercial exchanges with Makkan (Oman). Thus, Lagaš shipped textiles to Makkan,57
while Makkan, as demonstrated by Gudea’s dedicatory inscriptions, provided Lagaš with dior-
ite, from which statues of its rulers were carved. As in the case of Lagaš’s commercial contacts
with Upper Mesopotamia and the Zagros region, it is unlikely that such exchanges could have
taken place before the reign of Ur-Namma, since it is Ur-Namma who claims to have re-opened
the sea connections with Makkan.58

. Puzur-Inšušinak and the Šimaškians

Lastly, we need to consider what may be termed the “Šimaškian connection” of Puzur-Inšuši-
nak’s saga.59
Toward the very end of Puzur-Inšušinak’s victory inscription, there is a passage describing
how, following Puzur-Inšušinak’s victories in the Zagros, a king of Šimaški, being apparently
impressed by that event, came to Puzur-Inšušinak to pay obeisance to him.60 Significantly, this
is not only the earliest attestation of a Šimaškian ruler on record, but also of Šimaški’s name
itself. The fact that Puzur-Inšušinak attaches so much importance to his encounter with the
unnamed ruler of Šimaški, and that he recognizes him as a “king,” must be interpreted that this
individual was a political figure in his own right, whose power, while inferior to that of Puzur-
Inšušinak, was something to be reckoned with. At the same time, one has the impression that
this mysterious figure appeared on the scene somewhat unexpectedly and as if from nowhere—
the tell-tale signs of a newcomer.
It will not be unreasonable to consider, therefore, that the unnamed Šimaškian partner of
Puzur-Inšušinak was none other than Kirname, the first ruler of the Šimaškian dynasty. Such a
hypothesis would certainly fit chronologically. Since Puzur-Inšušinak belonged to Ur-Namma’s

55 Note the records of workers sent to Madga: x (guruš) Ma-ad-ga (RTC :, ; mu íd dBa-ú-hé-gál-SUD ba-ba-

al-la = Ur-Bau “”). Further, see .. še gur A-ga-dèki má Ma-ad-gaki Ur-dÙZ ì-gub gìr Ur-LI sabra zi-ga Šag4-ge
(RTC :–). For the identification of Madga with Hit, see now Heimpel a: –.
56 mu Ur-dNamma lugal-e sig-ta igi-nim-šè gìr bí-sá-a, “year when king Ur-Namma put in order the routes from

the Lower Country to the Upper Country” (year-name Ur-Namma “”; RTC , , ; ITT  ); ud é dNin-
gír-su-ka mu-dù-a dNin-gír-su lugal ki-ág-gá-ni-e a-ab-ba igi-nim-ta a-ab-ba sig-ga-šè gìr-bi gál mu-na-taka4, “when
he was about to build Ningirsu’s temple, Ningirsu, master who loves him, opened for him the routes from the Upper
Sea to the Lower Sea” (Gudea Statue B v —vi ); kur gišeren-na lú nu-ku4-ku4-da Gù-dé-a en dNin-gír-su-ke4 gír
mu-na-ni-gar, “the lord Ningirsu established for Gudea a route to the land of cedar, (the place) which no one can
enter” (Gudea Cylinder A xv –).
57 Garments Má-ganki-šè (MVN  :–; mu é-ba-gára ba-dù-a = Gudea “”).
58 gaba a-ab-ba-ka-ka ki-sar-a nam-ga-eš bí-sá má Má-gan šu-na mu-ni-gi , “on the coast of the sea he made
8 4
long-distance trade to flourish; he brought the ships of Makkan back to him (i.e., Nanna)” (Frayne : –, Ur-
Namma :–); ki -sur-ra má Má-g an-na … šu-na mu -ni-gi4, “the flourishing of the ships of Makkan … he
brought back to him” (ibid.: – Ur-Namma  i —ii ); ki- sar -ra má Má-ganki-na dNanna á dNanna lugal- gá -
ta hé-mi-gi4, “by the power of Nanna, his master, he restored for Nanna the flourishing of the ships of Makkan” (ibid.:
– Ur-Namma :–). I tentatively assume that ki-sar means something like “growth, cultivation, flourishing.”
Cf. Akkadian mušaru. If so, ki-sur-ra in Ur-Namma  is a phonetic writing.
59 The following is a summary of my discussion of this problem in Stolper Studies.
60 in UD  / ù-ga-ti- id / a-na DU-su / ù / ki-ma LUGAL / Si-maš-giki / il-li-kam-ma / DU Puzur -dInšušinak
4
/ is-ba-at / dInšušinak / ik-rí-bi-su / iš-me-má, “in a single day he made (those lands) fall prostate at his feet; and,
˙ the king of Šimaški came up (on learning about it), he seized the feet of Puzur-Inšušinak (in submission)”
when
(Gelb/Kienast : –, Elam :–).
 puzur-inšušinak at susa 

generation, Kirname would have been a contemporary of Ur-Namma. In turn, this would make
him two generations removed from Ebarat I, the third ruler of Šimaški, whose rule seems to
have begun in or shortly before the year Šulgi .61
Such a hypothesis would also help to understand Šimaški’s subsequent ascent to power.
Since Puzur-Inšušinak was eventually defeated by Ur-Namma, who, very likely with Gudea’s
active cooperation, expelled Puzur-Inšušinak both from northern Babylonia and the Susiana,
Kirname (if indeed it is he of whom Puzur-Inšušinak talks in his victory inscription) would
be expected directly to have profited from Puzur-Inšušinak’s demise. Indeed, it is even con-
ceivable that Kirname aided Ur-Namma in the latter’s war on Puzur-Inšušinak. According to
this scenario, Puzur-Inšušinak’s territorial possessions might very well have been divided up
between Ur-Namma and Kirname, with the former taking the hold of the Susiana plain (with
the cities of Susa, Urua, Sabum, (AdamDUN), and Pašime, to name only the most important
settlements there), and with the latter seizing the Elamite highlands. Very likely, it was this
dramatic transfer of rule over (at least some territories of) Elam from Puzur-Inšušinak to Kir-
name that is reflected in the Šimaškian King List, which closes the Awan chapter of Elam’s
history with Puzur-Inšušinak, beginning the next, Šimaškian chapter of that history with Kir-
name.

. Conclusions

Although there is evidence that various Elamite rulers made incursions into Babylonia during
Pre-Sargonic times, it appears quite certain that Puzur-Inšušinak was the first Elamite ruler who
was able to conquer, and then to control for a period of time, a major chunk of Babylonia. That
chunk was not insignificant, since included in it were northern Babylonia and the adjoining
Diyala Region, therefore more than half of the traditional Babylonian territories.
Puzur-Inšušinak’s conquests in the east were equally (if not even more) impressive, since,
apart from the Susiana and the state of Awan, he put under his rule the Zagros territories as
far as the Hamadan plain (Kimaš and Hurti). If he also controlled Anšan, Puzur-Inšusinak
probably was the first Elamite ever both to establish hegemony over the entire western section
of the Iranian plateau and to integrate the Susiana with Elam.
In view of the huge geographical scale of Puzur-Inšušinak’s conquests, it will not be unjusti-
fied to call his state an “empire.” Although this empire was short-lived, its historical importance
cannot be overstated, since the act of putting of much of the Iranian plateau under a single rule,
and of incorporating the Susiana into Elam, was a watershed event of the early Elamite history.
Clearly, it was this achievement that gave impetus, only a few decades later, to the formation of
the Šimaškian state.
The next Elamite ruler to conquer Babylonia following Puzur-Inšušinak’s feat was Kindattu,
the sixth ruler of Šimaški. By sacking Ur, and by controlling it for some twenty years, this
political heir of Puzur-Inšušinak’s made Babylonia pay for the annihilation of the latter’s realm
by Ur-Namma and Gudea. This pattern of invasions countered by retaliations, which had began
with the Sargonic conquests in Elam, was to characterize the Elamite-Babylonian relations
down to the very end of these two great civilizations.

61 See Steinkeller :  and .

 piotr steinkeller

Appendix: The Location of Kimaš and Hurti

The localization of Kimaš and Hurti62 is of critical importance for the proper understanding of
the geography of central Zagros ranges, and, consequently, of the Ur III military and political
engagement in that region. Another place that must also be considered in this connection is
Harši,63 which, together with Kimaš and Hurti, was the target of Šulgi’s campaign during his
forty-eigth regnal year, and which, therefore, must have been situated in the same general area.
The most important data bearing on these three places may be identified as follows:

() Kimaš and Hurti were situated near each other. This is demonstrated, first of all, by the fact
that these two lands were a joint target of the campaigns of years Šulgi  and . According to
the respective year-formulae,64 Kimaš, Hurti, and their lands were all conquered “in one day”
or “at once” (ud -a). This is not an exaggeration, since, as the archival data show (see below),
the conquest of these two localities was accomplished within less than a week.
That Kimaš and Hurti were neighbors is further indicated by the fact that these two places are
identified as the main target of Puzur-Inšušinak’s military operations in the Zagros. See above
p. ; below p.  n. .
From year  of Šulgi there survives a remarkable group of documents, which document
the progress of that year’s campaign against Kimaš and Hurti. The sequence begins with a
reference, dated to the second month, to the drinking party that celebrated the conquest of
Kimaš, during which twenty-nine oxen were consumed.65 A similar party, for which fifty-one
oxen were expended, took place during the following (third) month, in celebration of the
capture of Hurti.66 Importantly, the two texts referring to these festivities use the ús-sa form
of year Šulgi , thus indicating that the formula of Šulgi  had not yet been introduced.
But the campaign had not been over yet. Additional military operations against both Kimaš
and Hurti were necessary, since, on the twenty-third, twenty-fourth, and on either the preceding
or the following day of the fourth month, a banquet was celebrated at Nippur, in the temples of
Enlil and Ninlil, when Hurti was captured for the “second time.”67 The “second time” must refer
here to the operation against Hurti in the third month (see above n. ).
The final victory over Kimaš took place some days later, after the ruler of Kimaš had
been captured. This crowning event was celebrated at Nippur, on the first three days of the
fifth month, with a three-day-feast, during which fifty oxen and fifty suckling lambs were
consumed.68 This later group of records is dated with the formula of Šulgi , showing that
the victory over Kimaš and Hurti had now been made official.

62 Kimaš and Hurti are documented only in post-Sargonic and Ur III periods. The only later mention of Kimaš

comes from Goetze : , UIOM : (lú Ki-ma-aški), which dates to the reign of Sin-iddinam of Larsa. For
the alleged occurrences of Kimaš in the early Old Babylonian year-formulae from Ishchali, see below n. .
63 Harši is attested only in Ur III texts.
64 mu dŠul-gi … Ki-maški Hu-ur -tiki ma-da-bi ud -a mu-hul, “year Šulgi … destroyed Kimaš, Hurti, and their
5
lands in one day” (Šulgi ); mu Ha-ar-šiki Ki-maški Hu-ur5-tiki ù ma-da-bi ud -bi ba-hul, “year Harši, Kimaš, Hurti,
and their lands were destroyed in one day” (Šulgi ).
65  gud niga  gud kaš-dé-a ud Ki-maški ba-hul (YOS  :–; Š /ii).
66  gud-niga  gud kaš-dé-a ud Hu-ur -tiki ba-hul-a (AUCT  :–; Š /iii).
5
67 Sheep for Ninlil, Suen, Nintinuga, Dumuzi, and Nisaba, šag é dNin-líl-lá; and  sheep for Du -kug, gizbun é
4 6
dEn-líl-lá ud Hu-ur -tiki a-rá -kam-aš ba-hul (SAT  :–; Š /iv/);  sheep for dNin-líl ninda du -a gizbun
5 8
šag4 é dEn-líl ud! (Sigrist: mu) Hu-ur5-tiki a-rá -kam-ma-aš ba-hul (Sigrist , :–; Š /iv/);  oxen and 
lambs gizbun šag4 Nibruki ud Hu-ur5-tiki a-rá -kam-ma-aš ba-hul (BPOA  :–; Š []/iv/ [ or ]).
68  oxen and  suckling lambs for gišbun é dEn-líl dNin-líl ud énsi Ki-maški in-ma-dab - ba -a (OIP  :–
5
; Š /v/–).
 puzur-inšušinak at susa 

The scale of these victory celebrations is absolutely extraordinary,69 thus attesting to the great
importance that Šulgi attached to the victory over Kimaš and Hurti. Here it is also significant
that these two lands—which had to be retaken once again (this time together with Harši) in year
Šulgi —were the last foreign territories to be conquered by Šulgi. Thus, these two lands likely
were the main and the ultimate objective of all his conquests in the northwestern periphery.

() The campaign against Kimaš and Hurti is the only foreign expedition mentioned in Šulgi’s
historical inscriptions.70 This fact again underscores the importance that this campaign had to
Šulgi and his policies. In this connection, it may not be by accident that the wording of this
inscription is almost identical to that of Puzur-Inšušinak’s statue (see below under ()), which
celebrates that ruler’s victory over Kimaš and Hurti. In fact, it is even conceivable that Šulgi (or
his officials) had actually seen Puzur-Inšušinak’s display inscriptions, as such likely were present
in northern Babylonia and the Diyala Region as a result of the occupation of those territories
by Puzur-Inšušinak at the very beginning of the Ur III period.

() A close neighbor of Kimaš and Hurti was the land of Harši. Harši’s proximity to Kimaš and
Hurti is demonstrated by the fact that these three localities were taken together “in one day” or
“at once” during the campaign of year Šulgi . Since Harši was attacked already in year Šulgi
, apparently as part of the first (and failed) attempt to take Kimaš and Hurti,71 it evidently was
situated closer to Babylonia than Kimaš and Hurti, probably to the southwest of them.
That Harši was a neighbor of Kimaš and Hurti is also proved by the Puzriš-Dagan sources
from year Šulgi , in which the booty of Harši is listed jointly with that of Kimaš.72

() After their final pacification in year Šulgi , Kimaš, Hurti, and Harši were incorporated
into the ma-da system of defensive settlements, as reflected in the fact that all three of them
are documented as payers of the gún ma-da tax and related types of duties.73 This places them

69 Although this could be due to the incompleteness of the surviving record, no other Ur III military victory is

known to have been celebrated in a similarly grand and sumptuous way.

70 Frayne : –, Šulgi :–: ì-nu ma-at Ki-maški ù Hu-ur-timki ù-ha-li-gu-na. A brick of unknown
˘
provenance, published by von Schuler (: – and pl. ). Von Schuler’s attribution of this piece to Susa (“als
Herkunftsort des … Ziegels wird Susa genannt,” ibid.: ) is but a guess. The same attribution is given by Frayne,
:  (“… which is thought to have come from Susa”), again without any justification.
71 See Steinkeller, “Grand Strategy”.
72 nam-ra-ak Ki-maški Ha -ar-šiki (Sigrist , :–; Š /vii); nam-ra-ak Ha-ar-ši[ki] ù Ki-mašk[i] (AUCT 

:0; date not preserved, but almost certainly Šulgi ; mentions also gún Šušin[ki] in lines 0–0). For the booty of
Harši alone, see the following examples, all from Šulgi :  gud lá-NI šag4 nam-ra-ak Ha-ar-šiki gìr Ur-nìgin-gar
ugula Eš4-dar-al-su (TCL  :–; Š /vii);  gud lá-NI šag4 nam-ra-ak Ha-ar-šiki gìr Ur-dNin-ti ugula Hu-ba-a
(SAT  :–; Š /vii), for which cf.  gud šu-gíd šag4 maš-da-ri-a kaskal-la ir-ra gìr Ur-mes àga-ús lú Hu-ba-
a mu-DU (JCS :  lines –; Š /viii); animals šag4 nam-ra-ak Ha-ar-šiki (unpublished, Science Museum of
Minnesota, Cuneiform Collection, SMM : = BDTNS no. ; Š /vi/);  guruš se12(SIG7)-a, “blind” lú
Ha-ar-šiki (envelope: se12-a xxxx Ha-ar-šiki-me) (PDT :–; Š /vii). For se12-a, “blind,” see Heimpel b.
73  amar az I-sar-a-lí-iš-sú lú Ki-maški (OIP  :–; Š /ii/);  gud  máš Hu-un-hi-li  gud  máš Ra-

ši-ši lú Ki-maški-me (Torino  :–; AS /vii/);  gud  áb éren Hu-ur5-tiki (PDT :–; AS /i/);  sheep
from mu-DU lú Gu- ma -ra-šiki ù lú Ha-ar-šiki-me (AUCT  :–; Š /ix/[x]);  goats šag4 mu-DU éren Ha-
ar-šiki (BPOA  :–; /iv/);  goats and  bears Ià-ši-wi-ir énsi Ha-ar-šiki ugula I-ti-dDa-gan (MVN 
:–; ŠS /ix/);  sheep and goats engar-ne  goats ab-ba-uru-me-éš lú Ha-ar-šiki-me-éš (MVN  :–;
ŠS /ii/).
Somewhat problematic are the records of the cattle delivered by the “soldiers of Kimaš” in years Šulgi , , and
, which precede the conquest of Kimaš in years Š –:  gud éren!(UD) Ki-maški ud -kam (BPOA  :;
Š /ii-min/);  gud éren!(VD) Ki-maški (TRU :–; Š /iv-min/;  gud éren Ki-maški (Hirose :–;
Š /iii/—mu ús-sa Ur-bí-lumki ba-hul). The same problem is presented by the deliveries of Harši in years Šulgi 
 piotr steinkeller

comparatively close to Babylonia, excluding any far-away localizations, such as, e.g., the central
or northern sections of the Iranian plateau.
In this connection, it should be noted that Kimaš remained loyal to Babylonia as late as the
ninth month of year Ibbi-Suen . This is shown by the record of an oath of allegiance, which
was taken at that time by some officials from Kimaš at Nippur, in Ninurta’s temple there.74

() Apart from being payers of the gún mada tax, Kimaš, Hurti, and Harši also supplied Elamite
soldiers (Elam) to the Ur III state.75 This was true especially of Kimaš, which was, after Šimaški
and DuhduhNI, the third largest provider of this type of military personnel.76 This fact places
all three of them—Kimaš, Hurti, and Harši—safely within the Elamite horizon, excluding any
possibility of these places being located to the west of the Zagros mountains.77

() In the unpublished Ur III texts from Urusagrig,78 which record the movement of messengers
and military personnel between the city of Urusagrig, situated ca.  km northeast of Nippur,79
and the Iranian plateau, the most commonly named foreign place is Der (modern Badrah80),
which is attested over  times. The second most frequent destination in these sources is
Kimaš ( attestations), followed by Šimaški ( attestations), Diniktum ( attestations), and
Hurti ( attestations). Harši is mentioned three times. This particular distribution of toponyms
strongly suggests that the traffic proceeded from Urusagrig to Der, from where a route then led
onto the Iranian plateau, toward Kimaš, Hurti, Harši, and Šimaški. When one considers the
geographic position of Der, it becomes obvious that the route in question must have ran via the
modern towns of Mehran, Amirabad, and Ilam, connecting with the Great Khurasan Road in
the area of Islamabad-e Gharb (See fig. ).81 Clearly, this was the main access route to the Great
Khurasan Road when traveling from Nippur and southern Babylonia more generally. This is
confirmed by the fact that the Urusagrig texts do not mention any toponyms situated in the
Diyala Region,82 which offers the (more usual in the later periods) access to the Great Khurasan
Road when traveling from the area of Baghdad. Consequently, a conclusion is unavoidable that
Kimaš, Hurti, and Harši were situated along that particular stretch of the Great Khurasan Road.

and  (AUCT   and BPOA  , cited above). Should we assume, therefore, that the conquest and colonization
of Kimaš, Hurti, and Harši had actually began a few years earlier? Here note that Babylonia had had contacts with
Kimaš already in year Šulgi , as shown by the following record of an equid delivered from Kimaš to Umma in that
year:  dùr amar-ga dÌr-ra-dan mu-DU Ki-maški (BPOA  :–; Š /xi; Umma).
74  udu  máš é dNin-urta mu lú Ki-maški-ke -ne-šè (MVN  :; IS /x/).
4
75 Kimaš: RGTC , ; NISABA  texts , , , , , ; Sigrist, Messenger Texts, p. ; and passim. Hurti:

RGTC , . Harši: ibid.: . See now extensively Notizia : –, – and Tabela L.
76 See Notizia, op. cit.: – and Tabela L.
77 Another indication of this is the fact that, like Der (see Steinkeller, “Grand Strategy”), Kimaš and Harši were

suppliers of bear cubs. See MVN :– and OIP  :– cited above in n. .
78 Information courtesy of D.I. Owen, June , . I am deeply grateful to him for sharing—in his customarily

generous manner—this important information with me. A full publication of these texts by Owen is forthcoming.
79 For the approximate location of Urusagrig, see Steinkeller :  Map  and –.
80 More precisely, Tell al-#Aqar near Badrah. See Hrouda : – and pls. –.
81 From Mehran, this route branched out southeast toward the Susiana, via Dehloran and Musiyan. In this way,

Der oversaw traffic directed toward both of these two main destination points in western Iran.
82 The only possible exception here is Diniktum ( attestations), which has been located by some scholars in the

Diyala Region (see RGTC , ). However, as is indicated by the Mari letter ARM  :– (courtesy of D. Charpin),
which describes the retreat of an Elamite army from Ešnuna to Elam via Diniktum, Diniktum was located to the east
of the Diyala Region, along the route to Elam: [LÚ ELA]M.MEŠ a-lam Eš-nun-naki im-šu-uh [a-na D]i-ni-ik-timki
úr-ta-am-mi [iš-tu] Di-ni-ik-timki a-na ma-ti-šu [ú-da-a]p-pa-ar, “the Elamites looted the city˘of Ešnuna, (and then)
they retreated to Diniktum. (Now) they have left Dinikitum for their land.” The fact that Diniktum is mentioned so
often in the Urusagrig texts points to a location in the vicinity of the road leading from Urusagrig to Der.
 puzur-inšušinak at susa 

This alternative access-route to the Great Khurasan Road: Badrah (Der) > Mehran > Amira-
bad > Ilam > Islamabad-e Gharb, was in fact used both in ancient and in modern times. See, in
particular, the description of this route offered by J.V. Harrison (: –):
Although Kabir Kuh appears to be an unbroken barrier a few paths cross right over it, some passable
only by men on foot, and some suitable for donkey traffic. North-west of this range other tracks
existed which had a steady movement over them whilst the Wali of Pusht-i-Kuh ruled the region
from his fort in #Ilam, and for years afterwards smugglers used them to bring in the secondhand
clothing (…) It is the line of one of these old tracks which has been followed approximately by
the motor road from #Ilam to the frontier. The motor road to Kermanshah keeps well to the west
of the tribal track which climbed out of the plain, passed under the crags of flat-topped Kuh-i-
Manisht, and dropped down on the east of it to travel over rounded hills and through open valleys
to the north, where good going made long stages possible. It joined the main road at Shahabad [=
Islamabad-e Gharb] where the #Ilam motor road still branches off.
Kabir Kuh, which runs for  miles from Pul-i-Zal to #Ilam, has no main road across it to Pusht-
i-Kuh; but a mule train not far from the present motor road used to come across the outer ranges
from Bedra on the plains to #Ilam.83
The same route was also used by Freya Stark during her travels in the Pusht-i-Kuh.84 Having
crossed into Iran from Badrah, and having spent some time in Ilam (which she calls Husain-
abad), Stark took the motor road from Ilam to its junction with the main Karind-Kermanshah
road as it passes by Manisht Kuh.85
It is important to note here that the region of Ilam had a significant population during the
Bronze Age. Some km to the northwest of Ilam a number of Bronze Age cemeteries were
excavated by L. Vanden Berghe and his team in the ies. The most important of those are
Bani Surmah (near the village of Chavar on the Islamabad-e Gharb road) and Kalleh Nissar.86

() As we have seen earlier, Kimaš and Hurti were the main opponents of Puzur-Inšušinak
during his military operations in the Zagros.87 It was undoubtedly the possession of these two
localities that enabled him subsequently to move into and conquer the Diyala Region and
northern Babylonia.88 If so, Kimaš and Hurti must have provided access to the Great Khurasan
Road when traveling west from the Iranian plateau.

() An Ur III cylinder seal invokes a ruler of Kimaš named Hunnili: Hu-un-ni-li / ÉNSI Ki-
maški / ŠAGINA ma-at Elamki.89 Hunnili’s designation, “the governor of Kimaš (and) the general
of the land of Elam,” immediately brings to mind the titulary of Puzur-Inšušinak: ÉNSI Šušinki
ŠAGINA ma-ti Elamki.90 Given the fact that Puzur-Inšušinak had controlled Kimaš at one point,
it is highly likely that Hunnili’s titulary derives directly from that of Puzur-Inšušinak’s.

83 Harrison : . This route from Badrah via Ilam to Islamabad-e Gharb is shown on the map at the end of
the article. See also ibid., fourth photograph following p. , with a caption: “Mule road to #Ilam (Deh Bala) crossing
high pass under Kuh-i-Manisht.”
84 Stark : –. A more popular account of this trip was offered in her essay “The Hidden Treasure ”

in Stark .
85 Stark : . Another “early” traveler using this route was F.R. Maunsell. See Maunsell : –.
86 See Haerinck/Overlaet  and .
87 See above p. .
88 See above p. .
89 Frayne :  .
90 Gelb/Kienast : –, Elam :–, – Elam :–, – Elam :–, – Elam :–,

– Elam :–, – Elam :–. The title ŠAGINA ma-ti Elamki was introduced in the Sargonic period.
See Kienast/Sommerfeld : .
 piotr steinkeller

Hunnili is the only ruler of Kimaš whose name is known so far. Was he the same person as
the énsi who was captured by Šulgi in year Šulgi ? Since Hunnili’s grandiose titulary makes it
unlikely that he could have ruled over Kimaš after its conquest by Šulgi,91 this is quite probable,
in my view.92 Be that as it may, his (highly exaggerated) claim to have ruled over Elam, and his
familiarity with Puzur-Inšušinak’s titulary, locate Kimaš firmly in Iran.

() A messenger tablet (unfortunately not dated by year) records an expenditure of flour for
thirty-five prisoners-of-war from Kimaš, who came to Girsu/Lagaš from Urua.93 Since Urua can
confidently be located in northwestern Khuzistan, in a strategic point controlling the passage
from Southern Babylonia onto the Susiana plain,94 the prisoners in question very likely had
come to Urua from the north, over the route that ran from the region of Kimaš and Hurti
southward toward the Susiana, via modern Mehran, Dehloran, and Musiyan.95

() According to the inscriptions of Gudea of Lagaš, Kimaš—or, more precisely, its mountain
range Abullat—was, during Gudea’s reign, a source of copper:
Abullat(KÁ.GAL)-atki hur-sag Ki-maš-ka urudu mu-ni-ba-al, “in Abullat, the mountain range of
Kimaš, he (i.e., Gudea) mined copper” (Gudea Statue B vi –);
hur-sag urudu-ke4 Ki-maški-ta ní-bi mu-na-ab-pàd urudu-bi gi-si-a-ba mu-ni-ba-al, “the mountain
range of copper made itself known to him (lit.: found itself for him) from Kimaš; from there he (i.e.,
Gudea) excavated its copper in the baskets” (Gudea Cylinder A xvi –).
This information about Kimaš is tantalizingly corroborated by two Ur III tablets from Puzriš-
Dagan,96 both dating to year Amar-Suen , which list ritual “baskets” made of the “copper of
Kimaš”:
 ma-sá-ab zabar urudu Ki-maški úr dNin-KAS4mušen urudu kug-babbar šub-ba si-ga dInana Unugki-
ga a-ru-a lugal … šag4 Unugki-ga, “one bronze ‘basket’ of Kimaš copper (or: one ‘basket’ of bronze
and copper from Kimaš), on (its) bottom there is a Nin-KAS4 bird of copper, which is overlaid and
inlaid with silver, for Inana of Unug, the king’s ex-voto gift … in Unug”;97
 ma-sá-ab za[bar] urudu Ki-maški úr la -ha-ma kug-babbar igi ka kug-sig17 gar-ra! si-g[a] dEn-ki
Eriduki a-ru-a lugal … šag4 Eriduki, “… on (its) bottom there is a Lahmu figure of silver, (his) eyes
and mouth are overlaid and inlaid with gold, for Enki of Eridu, the king’s ex-voto gift … in Eridu.”98
It appears quite likely these two objects had been fashioned in Kimaš (or somewhere else in
Iran), from where they were brought to Babylonia following the sack of Kimaš in year Šulgi .
There is a good chance, in my view, that these copper “baskets” are to be connected with
the characteristic type of a decorated basin, examples of which have come from Shahdad, Jiroft

91 Unless he ruled in the post-Ur III period, which appears less likely, however.
92 It is tempting to think that Hunnini is the same person as the man of Kimaš named Hu-un-hi-li, who delivered
animals to Puzriš-Dagan in year Amar-Suen :  gud  máš Hu-un-hi-li  gud  máš Ra-ši-ši lú Ki-maški-me (Torino
 :–; AS /vii/). Should this be the case, we would then need to conclude that Hunnini was allowed eventually
to return to Kimaš, to occupy there some subordinate administrative position.
93  sìla ninda Na-sá lú-kas  guruš  sìla zíd-ta ne-ra!-aš-ak Ki-maški-me gìr Na-sá lú-kas Ú.URUxAki-ta
4 4
du-ni (Lau :–; -/xii). Cf.  sìla ninda A-hu-a sukkal .. Elam ne-ra-aš-ak Ki-maški-me gìr A-hu-a sukkal
Ki-maški-ta du-ni (CUSAS  :; Girsu/Lagaš; -/iii).
94 Steinkeller b: .
95 See above p.  and n. .
96 These two tablets were brought to my attention by P. Paoletti and W. Sallaberger during my stay at the Universität

München in –. I am grateful to them for this contribution, as well as for the opportunity of discussing these
data with them.
97 AUCT  :–; AS /xiii.
98 AUCT  :–; AS /xii; collation courtesy of P. Paoletti.
 puzur-inšušinak at susa 

(ancient Marhaši), and Tepe Hissar, as well as from other, unidentified sites on the Iranian
plateau.99 As described by P.R.S. Moorey, these objects are “shallow circular bowls with flat
broad bottoms and low, slightly flaring, vertical sides. On the bottoms, designs were raised
by hammering the metal up from below so that the motifs appear in various degrees of relief
on the inside of the bowl.”100 See figs. , , and  for the examples of such basins. Needless
to say, this identification, if correct, would provide us with an exceedingly important dating
indicator.101
While these data show that copper and copper objects were available in Kimaš in Ur III times,
they do not prove the existence of copper mines there.102 Although it is possible that the region of
Kimaš indeed had some copper deposits, a more likely solution appears to be that Kimaš simply
was a place where copper could easily be obtained.103 And a location along the Great Khurasan
Road would fit such a possibility very well. This is probably all that one can realistically expect
from the author of Statue B and the Cylinders to have known about the origin of Gudea’s copper.
Given the great distance between Lagaš and Kimaš, and, even more important, the unsettled
nature of Gudea’s times, we can be certain that none of his men ever ventured to Kimaš, and
that the copper in question was acquired through intermediary trade.104
For the purposes of historical geography more informative is the toponym Abullat. This
place-name is attested in yet another Ur III tablet. Significantly, the source in question, which
comes from Susa, links Abullat with a Zagros principality called Zidanum. A Susa tablet dating
to year Amar-Suen  names a Ra-si, who almost certainly is the same person as the ruler of
Zidanum of that name:105

99 For the examples of such basins from Shahdad, see Hakemi : , Obj. No.  (.cm diameter, .cm

height; a large knob in the middle);  Gs.  (ca. cm diameter; a crab and four fish);  Gs.  (a ring with two
rows of nine fish);  Gs.  (ca.  cm diameter; two antelopes);  Gs.  (ca. cm diameter; a coiled snake); Jiroft:
Madjidzadeh :  and  ( cm diameter, cm height; decorated with a sitting bird). Another, unpublished
basin, which is decorated with a reclining lion, is on display in the Kerman Museum (photograph taken by this
author, May ); Tepe Hissar: Schmidt : – (.cm diameter; a lion attacking a bull); Unprovenanced:
Amiet :  no. , pl. ; also color plate following p.  (Louvre, AO ; cm diameter, .cm height; a
reclining bull) and :  fig.  (Louvre, AO ; cm diameter; a reclining bison in low relief); Pittman
: –,  fig.  (Metropolitan Museum, L..; .cm diameter; a reclining humped bull in low relief).
In addition, Amiet :  n. , lists three unprovenanced basins sold at auctions in  and , which are
decorated with a “bison,” a “lionne,” and a “ronde d’oiseaux-pêcheurs” respectively.
100 Moorey (sd): .
101 The only problem with accepting this identification is that, unlike the Ur III ma-sá-ab, none of the surving

basins have silver or gold overlay. However, such decoration could easily have been added in Babylonia, to increase
the attractiveness and value of these pieces.
102 The geographic information given in Gudea’s inscriptions is far from precise, as their author(s) must have had

but a vague idea about the sources of the materials used on the Eninnu project. The parade example here is the
passage uru Ur-suki hur-sag Eb-la-ta “from the city of Uršu, (which) is (at) the mountain range of Ebla” (Statue B iv
–), on the basis of which late M. Astour (: –), “proved” that Tell Mardikh cannot be Ebla, and that the
latter was situated near Gaziantep in Turkey!
103 So Moorey : : “… it [Kimaš] may have been just an etrepôt for copper from mines deep in Iran.” In this

way, Kimaš’s position would have paralleled that of Tilmun (Bahrain), which, in Old Babylonian times and earlier,
served as an entrepôt for the copper mined in Makkan (Oman). Cf. here also the case of Ilušuma of Assur, who
“washed” copper for the Babylonians. See Grayson : –, Ilušuma :–, discussed by Steinkeller (:
–).
104 The most likely source of copper on the Iranian Plateau in Ur III times was the mining complex at Veshnoveh

south of Qom, where the evidence of very extensive ancient workings has been discovered. See Piggot : .
Given the fact that the distance between the Hamadan plain and the region of Veshnoveh is only ca. km, Kimaš
could easily have obtained its copper from Veshnoveh.
105 For Ra-ši, see Steinkeller, Studies Stolper. He is conceivably identical with Ra-ši-ši, whose is identified (together

with Hun-hili = Hunnili) as a man of Kimaš:  gud  máš Hu-un-hi-li  gud  máš Ra-ši-ši lú Ki-maški-me (Torino
 :–; AS /vii/).
 piotr steinkeller

?(gur) ì-giš gur / mu Ra-si-šè / ki Za-ri-iq-ta / A-da-làl lú-kin-gi4 lugal / šu ba-ti / šag4 A-bu-
la-atki / in Zé-ti-anki / iti l[a]- lu-bu -um / mu en Unu6-gal dInana ba-hun, “ ? bushel of sesame
oil was received, on behalf of Raši, from Zariq by the royal envoy Adallal, in Abullat, (which is) in
Ziti"an (= Zidanum); date”.106
Abullat undoubtedly is an Akkadian word, abullāt(um), “gates,” a plural form of abullum. This
meaning immediately brings to mind “Zagros Gates,” the critical mountain pass (the gorge Pai-
i Taq) through which the Great Khurasan Road ascends to the Iranian plateau. The “Zagros
Gates” are situated ca. km above the village of Sar-i-Pul-i-Zohab.107
The great strategic—as well as symbolic—importance of this pass was recognized in antiquity,
as proved by the reliefs and inscriptions left there by the rulers of Šimurrum and Lullubum.108 To
hold the “Zagros Gates” meant the control over the Great Khurasan Road. Here it is significant
that it was precisely Šimurrum and Lullubum that were the main opponents of Šulgi in his
Zagros wars.
Clearly, the proposition that Abullat is an ancient designation of the “Zagros Gates” is a
very enticing one.109 However, one needs to be cautious here, since Abullat could equally well
have denoted some other, more easterly pass along the Great Khurasan Road. One thinks here
especially of the Asadabad pass west of Hamadan, where the Great Khurasan Road crosses Kuh-
i-Alwand, the most easterly of the western Zagros ranges, and the “first significant barrier to
east-west movement after the road ascends the plateau at the Zagros Gates.”110 “Beyond Alwand
lies the Hamadan plain, the largest flat open area in this part of the Zagros. This plain separates
the eastern and western ranges of the Zagros.”111 Therefore, we must leave this question open
for now.

The data presented above argue strongly (in my view, irrefutably) that Kimaš and Hurti were
situated along the Great Khurasan Road as it passes through the western Zagros ranges. I
offered this suggestion already in : “Kimaš and Hurti … can quite confidently be located in
the western section of the modern province Kermanshah, around the towns of Shahabad and
Kermanshah.”112 My proposal that Kimaš and Hurti may correspond to modern Kermanshah
and Shahabad-e Gharb (since  known as Islamabad-e Gharb)113 respectively, both of which

106 MDP  :–. For this interpretation, see already De Graef : –. Cf. the year-formula mu ús-sa

Zi-da-na ba- hul in MDP  :, a tablet dating to the Šimaški period, for which see De Graef : . For the
various spellings of Zidanum in Ur III sources, see Steinkeller, Studies Stolper.
107 “About  km. further on [coming from the east], the road climbs the Pai-i Taq pass (the ‘Gates of Zagros’). A

stone ivan is built into a curve of this road and is known as Taq-i Girreh, possibly being a Sassanian relay station
on the royal road linking Ctesiphon and Hamadan … About km. further along the main road is the Kurdish
village of Sar-i Pul-i-Zuhab” (Matheson : ); “A little east of Sar-i-Pul the road crosses the floor of a very fine
natural amphitheatre and climbs over its edge at Tak-i-Girreh, the ‘Zagrian Gates’” (Persia, B.R. , Geographical
Handbook Series [London, Sept. , Naval Intelligence Division],  + photos  and  next to p. , which show
the Taq-i-Girreh pass). Cf. also Levine : –.
108 See Frayne :  = Edzard : –, an inscription of the unknown king of Šimurrum; Frayne :

–, Annubanini . Although no Ur III victory reliefs have been identified in that region so far, it has been
suggested that the uninscribed relief from Darband-i Gawr, some km south of Sulaymaniyah, may be of an Ur III
date. See Boese : –, Postgate/Roaf : , Eppihimer : –. For the location of Darband-i
Gawr, see Postgate/Roaf : , fig. .
109 It is tempting to consider that this key Zagros pass is described as ig gal Elam-a, “the great door of Elam,” in

one of Ur-Namma’s inscriptions: ig gal Elam-a-ta zag Šušinki-na-šè, “from the great door of Elam to the border of
Susa” (Frayne : –, Ur-Namma  ii0 0–0). Alternatively, the place in question could be Urua/Arawa, which
is described as the “lock of Elam” in a hymn to Išbi-Erra. See Steinkeller b: , –.
110 Levine : .
111 Ibid.: .
112 Steinkeller b: , n. .
113 Shahabad-e Gharb had formely bore the name of Harunabad.
 puzur-inšušinak at susa 

are ancient settlements, is dictated by the fact that these are the largest modern settlements in
the region in question. Even more important is the fact that both of them occupy strategically
important positions.
An Iranian archaeological team has recently conduced excavations in the middle of Islama-
bad-e Gharb, at a site called Chogha Gavaneh.114 Although it presently measures only  ha in
size, Chogha Gavaneh may have originally occupied an area as large as  ha.115 Importantly,
this site produced a lot of fifty-six Old Babylonian tablets, dealing with agriculture and herding.
It is striking that the onomasticon of these texts is overwhelmingly Akkadian, with a small
admixture of Amorite names. There are virtually no Elamite personal names within it. This fact
suggests that this particular settlement was a Babylonian (commercial?) enclave.
It goes without saying that this discovery in Islamabad-e Gharb proves the importance of
this region in ancient times.116 However, the question where precisely Kimaš and Hurti were
located must remain open for now. My hypothesis could be verified only through a systematic
archaeological investigation of this general area, especially the region of Kermanshah.117
As for Harši, I tentatively suggest that it is identical with the area of modern Ilam. I base this
hypothesis on the fact that, as indicated by the texts from Urusagrig (see above, p.  n. ),
in Ur III times the Great Khurasan Road appears to have been reached (primarily) through a
connecting route that ran via Der and Ilam (earlier known as Deh Bala and Husainabad), and
which hooked-up with the Great Khurasan Road near modern Islamabad-e Gharb.
Such a conclusion finds further support in the fact that Harši was the target of a Babylonian
military expedition already in year Šulgi .118 This suggests that Harši was situated closer to
Babylonia than Kimaš and Hurti, which too would fit the region of Ilam.

114 Abdi/Beckman : –.

115 Ibid.: –.
116 For Islamabad-e Gharb, see also Matheson : : “Shahabad-Gharb … is a small town about km. from

Kirmanshah, built right up against a mound surveyed by Stein, who found bronze ornaments and dishes of Luristan
type in the upper layers, and chalcolithic sherds of the third or fourth millennium bc in the lower sections.”
117 Two alternative localizations of Kimaš have been suggested since my b article. The first of them was by

Lafont (: –), who tried to connect Kimaš with the famous copper mines at Anarak (ca. km northeast
of Isfahan), which appear to have been exploited already in antiquity. Lafont’s hypothesis was enthusiastically
endorsed by Steve (: –), who even identified Abullat with Kūh-I Damergil, which towers over Anārak
at  m. However, any connection between Kimaš and Anarak is excluded by the fact that the latter place is
situated at a prohibitively huge distance (over km) from Babylonia. In contrast, Kimaš and its neighbor Hurti lay
comparatively close to Babylonia, since, as I showed earlier, both of them belonged to the belt of the ma-da defensive
settlements.
The other suggestion was made by Frayne (: –), who, while agreeing that there was a Kimaš in Iran
(which he did not seek to locate, however), argued that there was yet another place called Kimaš, which, according
to him, lay near modern Tawuq, not far from Nuzi. It was that other Kimaš, which, in Frayne’s view, was the target of
Šulgi’s campaigns. Frayne based this hypothesis on a single datum, which is the alleged appearance of Kimaš in the
year-names from Ishchali: “Unequivocal evidence [emphasis mine] that a city named Kimaš was located in the greater
vicinity of the middle Tigris is found in one (or more) year names of Old Babylonian date from Išcālı̄ that mention
the two cities of Kimaš and Ekallātum being attacked by an apparent king of Ešnunna” (ibid.: ). However, an
examination of the formulae in question (MU ÉREN SU.BI]R4 BA RA GIŠ KU BAR URU Ki-maš É-kal-l[a-tim]
[Greengus ,  year , variant]; MU GUD.APIN KUG.GI URU Ki-maš É-kal-la-tim [ibid.,  year  variant])
reveals that none of this is true. To begin with, there is no mention of a king of Ešnuna in any of these formulae
(though one of them possibly mentions a Subarian army). More important, URU Ki-maš É-kal-la-tim is most likely
to be explained as URU.KI.BAR É-kal-la-tim, “the suburb of Ekallatum” (note the genitive ending!). Here note that
another variant of year  has instead URU.KI Ša-nu-hu-um (ibid., ), where Šunuhum evidently is the name of
that suburb. ˘ ˘
The question of the location of Kimaš has been recently reviewed also by Potts (: –). However, this
aricle does not add anything new or substantial to the discussion.
118 As I considered earlier (see above, p. ), this campaign may have been Šulgi’s first attempt to get control of

Kimaš and Hurti. To all appearances, this attempt was unsuccessful, due probably to Šimurrum’s and Lullubum’s
interference.
 piotr steinkeller

Other ancient localities that are to be sought within this general area are Zidanum and
Zidahri, and probably also the Šimaškian land Šigriš.119 One may locate all three of them pro-
visionally between Kermanshah and Hamadan, or perhaps even farther east on the Hamadan
Plain.120

Bibliography

Abdi, K. and Beckman, G. (): “An Early Second-Millennium Cuneiform Archive from Chogha
Gavaneh, Western Iran”, JCS : –.
Amiet, P. (): Les antiquités du Luristan, Paris.
———. (): L’âge des échanges inter-iraniens – avant J.-C., Paris.
André, B. and Salvini, M. (): “Réflexions sur Puzur-Inšušinak,” IrAnt : –.
André-Salvini, B. (–): “Puzur-Inšušinak,” RlA : –.
ARM  = Durand, J.-M. and Charpin, D. (): Archives épistolaires de Mari, Paris.
Astour, M. (): “Tell Mardikh and Ebla,” UF : –.
AUCT  = Sigrist, M. (): Neo-Sumerian Account Texts in the Horn Archaeological Museum, Andrews
University Cuneiform Texts , Berrien Springs.
AUTC  = Sigrist, M. (): Neo-Sumerian Account Texts in the Horn Archaeological Museum, Andrews
University Cuneiform Texts , Berrien Springs.
Boese, J. (): “Zur stilistischen und historischen Einordnung des Felsreliefs von Darband-i-Gaur,” StIr
: –.
BPOA  = Sigrist M. and Ozaki, T. (): Neo-Sumerian Administrative Tablets from the Yale Babylonian
Collection, Part One, Madrid.
BPOA  = Sigrist, M. and Ozaki, T. (): Neo-Sumerian Administrative Tablets from the Yale Babylonian
Collection, Part Two, Madrid.
Civil, M. (): ““Adamdun,” the Hippopotamus, and the Crocodile,” JCS : –.
CUSAS  = Garfinkle, S., Sauren, H. and Van De Mieroop, M. (): Ur III Tablets from the Columbia
University Libraries, Maryland.
Dahl, J.L. (): “Early Writing in Iran, a Reappraisal,” Iran (): –.
De Graef, K. (): “Annus Simaškensis. L’usage des noms d’année pendant la période simaškéenne
(ca. – av. notre ère) à Suse,” IrAnt : –.
Edzard, D.O. (): “Zwei Inschriften am Felsen von Sar-i-Pūli-Zohāb Anubanini  und,” AfO : –
.
Eppihimer, M.A. (): The Visual Legacy of Akkadian Kingship (unpublished Harvard PhD disserta-
tion).
Frayne, D. (): Old Babylonian Period (–BC) (= RIME ), Toronto.
———. (): Sargonic and Gutian Periods, –bc (= RIME ), Toronto.
———. (): Ur III Period (= RIME /), Toronto.
———. (): “The Zagros Campaigns of Šulgi and Amar-Suen,” SCCNH : –.
Gasche, H. (Ed.) (): “The Persian Gulf Shorelines and the Karkheh, Karun and Jarrahi Rivers: A
Geo-Archaeological Approach,” Akkadica : –.
Gelb, I.J. and Kienast, B. (): Die altakkadischen Konigsinschriften des dritten Jahrtausends v. Chr. (=
FAOS ), Stuttgart.
George, A.R. (Ed.) (): Cuneiform Royal Inscriptions and Related Texts in the Schøyen Collection (=
CUSAS ), Bethesda.
Goetze, A. (): “Sin-Iddinam of Larsa. New Tablets from His Reign,” JCS : –.
Grayson, A.K. (–): “The Empire of Sargon of Akkad,” AfO : –.
———. (): Assyrian Rulers of the Third and Second Millennia bc (to BC) (= RIMA ), Toronto.
Greengus, S. (): Old Babylonian Tablets from Ishchali and Vicinity (= PIHANS ), Istanbul.

119 See Steinkeller, Studies Stolper. Here note that Zidahrum and Šigriš are documented as messengers’ destinations
in the Ur III texts from Urusagrig (see above p. ). Each of them is attested four times (information courtesy of
D.I. Owen).
120 The distance between Kermanshah and Hamadan is ca. km. Kermanshah lies in the fertile valley of the

Qareh Su, which is one of the richest agriculatural areas of southern Kurdistan.
 puzur-inšušinak at susa 

Haerinck, E. and Overlaet, B. (): Bani Surmah: An Early Bronze Age Graveyard in Pusht-i Kuh,
Luristan (= Luristan Excavation Documents ): Leuven.
———. (): The Kalleh Nissar Bronze Age Graveyard in Pusht-i Kuh, Luristan (= Luristan Excavation
Documents ): Leuven.
Hakemi, A. (): Shahdad: Archaeological Excavations of a Bronze Age Center in Iran, Rome.
Harrison, J.V. (): “Some Routes in Southern Iran,” The Geographical Journal : –.
Heimpel, W. (a): “The Location of Magda,” JCS : –.
———. (b): “Blind Workers in Ur III Texts,” KASKAL : –.
Hirose = Gomi, T., Hirose, Y. and Hirose, K. (): Neo-Sumerian Account Texts of the Hirose Collection,
Potomac.
Hrouda, B. (): “Ergebnisse einer Ruinenbesichtigung im südöstlichen Iraq,” Bagh. Mitt. : –.
ITT  = de Genouillac, H. (): Textes de l’Éopque d’Ur, Paris.
ITT  = Delaporte, L. (): Tablettes de l’époque d’Ur (fouilles d’Ernest de Sarzec en  et ), Paris.
JCS  = NN. (): “Texts and Fragments (–),” JCS : –.
Kienast, B. and Sommerfeld, W. (): Glossar zu den ALtakkadischen Königsinschriften (= FAOS ),
Stuttgart.
Krebernik, M. (): “Philologische Aspekte elamisch-mesopotamischer Beziehungen im Überblick.”
In: Kogan, L. et al. (Eds.), Babel und Bibel / Memoriae Igor M. Doakonoff, Winona Lake: –.
Lafont, B. (): “L’extraction du minerai du cuivre en Iran à la fin du IIIe millénaire.” In: Tunca, Ö. and
Deheselle, D. (Eds.), Tablettes et images aux pays de Sumer et d’Akkad: Mélanges offerts à Monsieur
H. Limet, Liège: –.
———. (): “Sur quelques dossiers des archives de Girsu.” In: Kleinerman, A. and Sasson, J.M. (Eds.),
Why Should Someone Who Knows Something Conceal It? Cuneiform Studies in Honor of David I. Owen
on his th Birthday, Bethesda: –.
Leemans, W.F. (): Foreign Trade in the Old Babylonian Period, Leiden.
Levine, L.D. : Geographical Studies in the Neo-Assyrian Zagros, Toronto.
Madjidzadeh, Y. (): Jiroft: The Earliest Oriental Civilization, Tehran.
Matheson, S.A. (): Persia: An Archaeological Guide, Park Ridge.
Maunsell, F.R. (): “The Land of Elam,” The Geographical Journal : –.
MDP  = Scheil, V. (): Textes Élamites-sémitiques—ème série, Paris.
MDP  = Scheil, V. (): Textes Élamites-sémitiques—ème série, Paris.
MDP  = Scheil, V. (): Textes Élamites-sémitiques—ème série, Paris.
MDP  = Dossin, G. (): Autres texts sumériens et accadiens, Paris.
Michalowski, P. (): “Observations on ‘Elamites’ and ‘Elam’ in Ur III times.” In: Michalowski, P. (Ed.),
On the Third Dynasty of Ur: Studies in Honor of Marcel Sigrist (= JCS Suppl. Series ), Boston: –
.
Mofidi-Nasrabadi, B. (): “Eine Steinschrift des Amar-Suena aus Tappeh Bormi (Iran),” ZA : –
.
Moorey, P.R.S. (): Ancient Mesopotamian Materials and Industries: the Archaeological Evidence,
Oxford.
———. (sd): “High Relief Decoration on Ancient Iranian Metal Vessels: Development and Influence,” The
Circle of Ancient Iranian Studies – (www.cais-soas.com).
MVN  = Owen, D.I. (): The John Frederick Lewis Collection, Rome.
MVN  = Sollberger, E. (): The Pinches Manuscript, Rome.
Nesbit, W.M. (): Sumerian Records from Drehem (= CUOS ), New York.
NFT = Cros, G., Heuzey, L. and Thureau-Fangin, F. (–): Nouvelles fouilles de Tello, Paris.
NISABA  = Capitani, M. (): Girsu Messenger Texts in the British Museum, Messina.
Notizia, P. (): I testi dei messaggeri da Girsu-Lagaš della Terza Dinastia di Ur (= NISABA ), Messina.
OIP  = Hilgert, M. (): Cuneiform Texts from the Ur III Period in the Oriental Institute, vol. :
Drehem Administrative Documents from the Reign of Šulgi, Chicago.
PDT  = Çig, M., Kizilyay, H. and Salonen, A. (): Die Puzriš-Dagan-Texte der Istanbuler Archäolo-
gischen Museen Teil I: Nrr. –, Helsinki.
Piggot, V.C. (): “The Development of Metal Production on the Iranian Plateau.” In: Piggot, V.C. (Ed.),
The Archaeometallurgy of the Asian Old World (= UMM ), Philadelphia: –.
Pittman, H. (): Art of the Bronze Age: Southeastern Iran, Western Central Asia, and the Indus Valley,
New York.
Postgate, J.N. and Roaf, M.D. (): “The Shaikhan Relief,” Al-Rāfidān : –.
 piotr steinkeller

Potts, D.T. (): The Archaeology of Elam: Formation and Transformation of an Ancient Iranian State,
Cambridge.
———. (): “Adamšah, Kimaš and the Miners of Lagaš.” In: Baker, H.D. et al. (Eds.), Your Praise is
Sweet: A Memorial Volume for Jeremy Black from Students, Colleagues and Friends, London.
RGTC  = Edzard, D.O. and Farber, G. (): Die Otrs- une Gewässernamen der Zeit der . Dynastie von
Ur, Wiesbaden.
RGTC  = Groneberg, B. (): Die Orts- une Gewässernamen der altbabylonischen Zeit, Wiesbaden.
RTC = Thureau-Dangin, F. (): Recueil de tablettes chandéennes, Paris.
Sallaberger, W. (): Die kultische Kalender der Ur III-Zeit, Berlin.
SAT  = Sigrist, M. (): Texts from the Yale Babylonian Collections. I: Sumerian Archival Texts ,
Bethesda.
Schmidt, E.F. (): Excavations at Tepe Hissar Damghan, Philadelphia.
Sigrist, M. (): Tablettes du Princeton Theological Seminary: époque d’Ur III, Philadelphia.
———. (): Neo-Sumerian Texts from the Royal Ontario Museum. : The Administration at Drehem,
Bethesda.
Stark, F. (): “The Pusht-i-Kuh,” The Geographical Journal : –.
———. (): The Valleys of the Assassins and Other Persian Travels, London.
Steinkeller, P. (): “Early History of the Hamrin Basin in the Light of Textual Evidence.” In: Gibson,
McG. (Ed.), Uch Tepe I, Chicago-Copenhagen: –.
———. (a): “The Mesopotamian God Kakka,” JNES : –.
———. (b): “The Question of Marhaši: A Contribution to the Historical Geography of Iran in the
˘
Third Millennium B.C.,” ZA : –.
———. (a): “The Date of Gudea and His Dynasty,” JCS : –.
———. (b): “On the Identity of the Toponym LÚ.SU(.A),” JAOS : –.
———. (): “New Light on the Hydrology and Topography of Southern Babylonia in the Third
Millennium,” ZA : –.
———. (): “New Light on Šimaški and Its Rulers,” ZA : –.
———. (in press): “On the Dynasty of Šimaški: Twenty (or so) Years After.” In Woods, C. (Ed.), Extraction
and Control: Studies in Honor of Matthew W. Stolper, Chicago.
———. (in press): “The Gutian Period in a Chronological Perspective.” In: Sallaberger, W. and Schrakamp,
I. (Eds.), Handbook on Third Millennium Chronology (ARCANE project).
———. (forthcoming): The Grand Strategy of the Ur III Empire: Babylonia’s Foreign Policy and Territorial
Expansion at the End of the Third Millennium bc.
Steve, M.-J. (): “La tablette sumérienne de Šūštar (T. MK ),” Akkadica : –.
TCL  = de Genouillac, H. (): Tablettes de Drehem publiées avec inventaire et tables, Paris.
Torino  = Archi, A. and Pomponio, F. (): Testi cuneiformi neo-sumerici da Drehem, Milan.
TRU = Legrain, L. (): Le temps des rois d’Ur, recherches sur la société antique, d’après des texts
nouveaux, Paris.
von Schuler, E. (): “Eine neue Inscgrift König Šulgis,” Berliner Jahrbuch für Vor- und Frühgeschichte
: –.
Wilcke, C. (): “Die Inschrift ‘Tukultı̄-Ninurta I ’. Tukultı̄-Ninurtas I. von Assyrien Feldzug gegen
Gutäer und andere, nordöstliche und nordwestliche Feinde und der erste Bericht über den Bau seines
neuen Palastes” In Fincke, J.C. (Ed.), Festschrift für Gernot Wilhelm, Dresden: –.
YOS  = Keiser, C.E. (): Selected Temple Documents of the Ur Dynasty, New Haven.
YOS  = Goetze, A. (): Sumerian and Akkadian Texts, New Haven-London.
UET  = Legrain, L. (): Business Documents of the Third Dynasty of Ur, London.
Vallat, F. (): “La date du règne de Gudea,” NABU /.
 puzur-inšušinak at susa 

Fig. . A map illustrating the Conquests of Puzur-Inšušinak.

 piotr steinkeller

Fig. . A reconstruction of the route leading from Urusagrig to Harši, Kimaš, and Hurti.
 puzur-inšušinak at susa 

Fig. . A copper basin decorated with a sitting bird. cm

diameter, cm height. After Madjidzadeh :  and .

Fig. . A copper basin Fig. . A copper basin decorated with a

decorated with two antelopes. reclining bull. cm diameter, .cm height.
Ca. cm diameter. After Louvre, AO . After Amiet : 
Hakemi :  Gs. . no. , pl. ; also color plate following p. .
 LES PREMIERS SUKKALMAH ET LES DERNIERS ROIS DE SIMAŠKI

Jean-Jacques Glassner*

Dans un article paru en 1, je m’intéressais à l’ expression élamite ruhušak que les Mésopo-
tamiens traduisirent invariablement, à travers les siècles, par mār ahati(m), « fils de la soeur ».
J’ai montré qu’en Élam, le fils né d’un couple parental formé d’ un frère et de sa soeur n’ était
pas désigné par le mot ruhušak, mais par celui de šak, « fils », et que la filiation, dans ce cas, était
de type patrilinéaire. ruhušak désignant le neveu utérin, la traduction akkadienne était pleine-
ment justifiée. En un mot, à une période donnée de leur histoire, les familles royales élamites
adoptèrent un mode de succession matrilinéaire, un neveu utérin succédant à un oncle mater-
nel sur le trône royal. Cette vision d’une élite sociale ayant fait le choix de la matrilinéarité me
paraît aujourd’hui définitivement acquise.
Toujours en , nullement préoccupé par des problèmes de chronologie, alors insolubles,
je proposais le schéma suivant: les premiers sukkalmah se seraient présentés comme les fils
biologiques de la soeur de Šilhaha. Leurs successeurs, par contre, se seraient désignés comme les
fils de la soeur de leurs prédécesseurs immédiats. Plus tard, à dater de Kuk-Našur, le successeur
de Temti-Agun, les futurs souverains auraient fait référence, dans leurs titulatures, à la soeur
de leurs prédécesseurs lorsqu’ils n’étaient encore que sukkal de Suse, n’ invoquant la soeur de
Šilhaha qu’une fois devenus sukkalmah. Il se serait alors agi, pour eux, en se rapprochant des
figures fondatrices de la dynastie, de légitimer leur pouvoir avec plus de force. Cette seconde
hypothèse doit être abandonnée, à la lumière d’ une lecture plus assurée des sources et grâce à
l’apport de documents nouveaux.
Semblablement, avec d’autres auteurs, j’ accordais trop de crédit au témoignage d’ un sceau
susien2 qui énonce à la suite les uns des autres les noms de Šilhaha, Siruk-duh, Siwe-palar-
huhpak et Kudu-zuluš. Je pensais alors qu’ il était fait mention, dans l’ ordre de leur ˙succession,
des noms des quatre premiers sukkalmah, le lien entre Šilhaha et Ebarat demeurant imprécis.
Cette position est aujourd’hui intenable. Si Siruk-duh, Siwe-palar-huhpak et Kudu-zuluš sont
trois sukkalmah qui se succédèrent directement dans le˙ temps, il ne peut en être de même de
Šilhaha et de Siruk-duh.

Je prends pour point de départ de la présente étude le « cylindroïde » d’ Atta-hušu3. Ce texte

est habituellement traduit comme suit4 : « Ebarat étant roi d’ Anšan et de Suse, Šilhaha étant
sukkalmah et roi-père (AD.DA LU[GA]L) d’ Anšan et de Suse, Atta-hušu, sukkal et teppir de
Suse5, fils de la soeur de Šilhaha, a bâti le temple de Nanna ». F. Vallat ( : ) en propose,
récemment, une nouvelle interprétation : « (Pour) Ebarat, le roi d’ Anšan et de Suse, (pour)
Šilhaha, sukkalmah, l’ancêtre des rois (AD.DA LU[GA]L) d’ Anšan et de Suse, Atta-hušu, sukkal

* Arscan, CNRS-Paris-Paris-Ouest. Mes remerciements vont à Françoise Grillot et à Piotr Steinkeller avec

lesquels j’ ai discuté plusieurs points importants traités dans cet article. Il va de soi que les erreurs sont miennes.
1 Glassner, J.-J.  : –.
2 MDP ,  et . Voir Grillot, F., Glassner, J.-J. : .
3 MDP  : .
4 Sollberger, E., Kupper, J.-R.  : , IVOa.
5 On lit souvent ERÍN Šušim, « du peuple de Suse», mais les graphies de Suse varient et, à l’initiale, les signes UD

et DINGIR peuvent alterner.

 jean-jacques glassner

et eppir de Suse, fils de la soeur de Šilhaha, etc. ». Aucune de ces traductions n’ est véritablement
de nature à emporter la conviction. Comment concilier, en effet, la double titulature d’ Ebarat,
roi d’Anšan et de Suse, avec celle de Šilhaha, sukkalmah, roi-père d’ Anšan et de Suse. Il est
tout aussi impossible d’admettre que Šilhaha soit père des futurs rois d’ Anšan et de Suse, alors
qu’Ebarat, qui est mentionné avant lui, porte déjà ce titre ; pareille hypothèse suppose, en outre,
une projection des plus incertaines dans un avenir lointain et dont Atta-hušu aurait eu la géniale
intuition.
En acceptant à titre provisoire la lecture AD.DA LU[GA]L, après tout, le troisième signe est
en partie perdu dans une cassure, il est peut-être possible d’ offrir une autre traduction : « Pour
Ebarat, roi d’Anšan et de Suse, Šilhaha étant sukkalmah, il est le père du roi d’ Anšan et de Suse,
Atta-hušu, le sukkal et le teppir de Suse, le fils de la soeur de Šilhaha, etc. »
Šilhaha serait alors le père d’Ebarat et l’ oncle maternel d’ Atta-hušu. Ce dernier, qui est
l’auteur de l’inscription, doit montrer les liens de parenté qui l’ unissent à son souverain, Ebarat
d’Anšan. Se présentant comme le fils de la soeur de Šilhaha, il lui est nécessaire d’ introduire ce
dernier. Le parallèle avec une inscription de Kitten-rakittapi, sukkalmah et serviteur d’ Idattu
Ier de Simaški6, fait entrevoir qu’il est possiblement, en sa qualité de sukkalmah, un serviteur
d’un roi de Simaški. Son fils, s’émancipant de la suzeraineté simaškéenne, s’ emparerait du
pouvoir à Anšan et à Suse, au détriment des représentants de la dynastie de Simaški. On
comprendrait mieux, ainsi, que Šilhaha soit considéré comme le véritable fondateur de la
dynastie des sukkalmah dont il est la figure de référence.
Il est bien connu que l’affichage d’ une relation d’ ascendance ou de parenté est un mode
avéré de reconnaissance sociale. Comme le souligne A. Duplouy ( : ), une généalogie
constitue un instrument de prétention sociale ; or, il n’ est « rien de plus facile que de forger
une généalogie», l’unique problème étant « non pas de la rendre vraisemblable, mais de la faire
accepter par l’opinion». Si la référence à l’ ancestralité est, certes, un marqueur identitaire, il
n’empêche, la construction de la généalogie peut relever entièrement de l’ imagination. « Une
généalogie antique n’a rien d’un arbre généalogique », écrit encore A. Duplouy, « elle résulte
d’une volonté d’accumuler le renom du plus grand nombre possible d’ ancêtres et d’ endosser
le prestige» dont ils sont crédités (Duplouy  : ). Certes, il existe le témoignage tardif
de Šilhak-Inšušinak faisant de Šilhaha non pas le père, mais le « fils bien-aimé », šak hanik7,
d’Ebarat, celui-ci n’est cependant corroboré par aucune source incontestable.
L’incohérence du style semble plaider en défaveur de cette dernière hypothèse, mais elle n’ est
peut-être qu’apparente, elle résulte du mode d’ expression compliqué et contourné adopté par
l’auteur de l’ inscription qui est contraint de préciser, au moyen d’ une incise, sa place parmi les
membres de l’élite sociale qui accapare le pouvoir politique. Šilhaha n’ apparaît ici que parce
qu’il est l’oncle utérin de l’auteur de l’ inscription.
Quoique séduisante, l’hypothèse, toutefois, n’ est pas recevable, et cela pour deux raisons
essentielles. La première raison tient au fait que le nom de Šilhaha figure deux fois dans les
archives d’un certain Ašiši, des archives identifiées par K. De Graef. La première fois, son nom
apparaît sans titre (MDP , ), la seconde fois, il est qualifié de LUGAL8. À quelle fonction
ce dernier titre peut-il faire allusion ? On le découvre dans plusieurs formulaires de serments
susiens où se révèle l’ordre hiérarchique suivant : Kuter-Šilhaha sukkalmah et Teptiraptaš « roi »,
šar, de Suse; Kuter-Šilhaha sukkalmah et Sir-duh « roi », šar, de Suse ; Kuk-Našur sukkal d’ Élam

6Steinkeller, P.  : .

7König, F.W. ,  §  ; a §  ; b §. Sur cette expression, voir MDP : .
8 MDP , . Ce document est à rapprocher de MDP ,  qui traite d’une affaire comparable impliquant un

même acteur.
 les premiers sukkalmah et les derniers rois de simaški 

et Kudu-zuluš «roi», LUGAL, de Suse9. D’ où il ressort que le titre royal désigne le roi de Suse10
et que Šilhaha porta ce titre avant de poursuivre sa carrière politique et de revêtir la fonction de
sukkalmah. Enfin, dans une ultime formule de serment, l’ invocation du nom d’ Ebarat précède
celle du nom de Šilhaha11. Même si aucun titre ne leur est attribué, et si l’ on ne sait à quel
moment de leurs cursus politiques respectifs le document se situe, on voit qu’ ils mènent leurs
carrières de conserve. La seconde raison réside dans le fait que la lecture LU[GA]L dans le titre
énigmatique AD.DA LUGAL ne peut être conservée. Ce titre est totalement inconnu et, de ce
fait même, il fait difficulté, étant suspect aux yeux de l’ historien12. À la suite d’ une proposition
d’I.J. Gelb, M. Stolper13 suggère de restituer le signe KA[LA]M( !) en lieu de L[UGA]L, et de
lire «père du pays d’Anšan et de Suse». La collation de l’ objet, au Musée du Louvre14, montre
que le signe LUGAL est à exclure de façon définitive. Le début du signe ressemble fort à celui
de KALAM, les traits horizontaux de la première moitié du signe étant attendus au début de la
cassure. Quant à la fin du signe, V. Scheil copiait autrefois deux traits formant un angle obtus,
ce qui représente la fin du même KALAM, deux traits que la restauration a malheureusement
fait disparaître; partant, il faut à la vérité de dire que, en l’ état, la restitution d’ un signe LÚ n’ est
pas totalement à écarter. Quoi qu’il en soit, la proposition d’ I.J. Gelb reprise par M.W. Stolper
présente le triple mérite d’être acceptable sur le plan de la paléographie, de rendre le titre plus
admissible15, enfin de s’harmoniser avec celui d’ Ebarat au lieu de le dupliquer. C’ est donc elle
que l’on retiendra.
Le texte est donc à comprendre comme suit :
e-ba-ra-at
lugal an-ša-an ù MÙŠ.ERIN.KI
ší-il-ha-ha
SUKKAL.[M]AH
AD.DA *KA[LAM]!
an-ša-an ù MÙ[Š].ERIN-ÀM
at-tá-hu-šu
SUKKAL ù <te>-ep-pí-ir16 UD.MÙŠ.ERIN.KI
DUMU NIN9 ší-il-ha-ha
É dNANNA
BA.DÙ
Désormais, seules deux hypothèses demeurent plausibles :
a) Ebarat est prédécédé. Šilhaha est présenté comme son successeur avec les titres de sukkal-
mah et de adda, «père», du pays d’ Anšan et de Suse. Une question se pose à propos de
cette titulature: à quel pays s’applique le premier titre, celui de sukkalmah ? Deux possi-
bilités se présentent. Peu de temps auparavant, sous le règne d’ Idattu Ier, qui cumulait
les titres de «roi d’Anšan», ainsi que de « roi de Simaški et d’ Élam », Kitten-rakittapi
fut sukkalmah d’Élam17. Peu de temps plus tard, comme on va le voir, Temti-Agun sera

9 MDP ,  : – ;  : – ;  rev. – ; : –.

10 Sur ce titre, voir l’ hypothèse émise par L. De Meyer (: ).
11 De Meyer, L.  : .
12 Voir Seux, M.J.  : passim.
13 Carter, E., Stolper, M.W.  :  et , n. , où ils renvoient à I.J. Gelb; Stolper, M.W. : . Une leçon

admise par B. André-Salvini ( : , no ; on trouvera une excellente photographie de l’objet à la page ).
14 Je remercie vivement B. André-Salvini et A. Benoit pour m’avoir autorisé à collationner ce texte au Musée du

Louvre.
15 Sur abu/ad.da : Seux, M.-J. : passim.
16 Les titres sukkal et teppir se retrouvent en Mahboudian, H. : . Il faut donc restituer le signe TE à l’initiale

du titre. Sur ippir : MDP  :  n. .

17 Steinkeller, P.  : .
 jean-jacques glassner

sukkalmah d’Élam et de Simaški. Il semble qu’ Ebarat ne régna que sur Anšan et Suse, son
successeur Šilhaha, investi des titres de sukkalmah soit d’ Élam, soit d’ Élam et de Simaški,
et de père du pays d’Anšan et de Suse, aurait-il tenté de reconstituer l’ ancien groupement
de royaumes sur lequel régnaient les membres défunts de la dynastie de Simaški ? On ne
saurait le dire.
b) Ebarat est vivant, il règne sur Anšan et Suse ; on se souvient que sur le sceau de son épouse,
son nom est précédé du déterminatif divin18. Šilhaha est son sukkalmah qui exerce en son
nom l’autorité soit sur l’Élam seul, soit sur Simaški et l’ Élam.
Jusqu’à plus ample informé, je me range à la seconde hypothèse, Ebarat est vivant et Šilhaha
est son sukkalmah, comme le donne à penser l’ inscription inédite de Temti-Agun publiée ci-
dessous et qui fait certainement allusion à la mort de ce roi ou à la commémoration de celle-ci.
L’inscription d’Atta-hušu est donc à traduire : « Pour Ebarat, roi d’ Anšan et de Suse, Šilhaha
étant sukkalmah, père du pays d’Anšan et de Suse, Atta-hušu, sukkal et teppir de Suse, fils de
la soeur de Šilhaha, a bâti le temple de Nanna. »
Ce faisant, je rejoins, dans ce cas particulier, l’ hypothèse formulée voici quelques années par
P. Steinkeller (: , n. ) selon laquelle le sukkalmah serait au service du roi d’ Anšan, ce
que vient confirmer l’inscription d’un sceau-cylindre (MDP , no ) :
e-ba-ra-at LUGAL
ku-uk-dKAL.LA19
DUMU ku-uk-ša-rum
ÌR ší-il-ha-ha
«Ebarat roi. Kuk-Tanra, fils de Kuk-šarum, serviteur de Šilhaha»
Dans les deux cas de figure envisagés, enfin, la suffixation -ÀM ne s’ applique qu’à l’ élément
de phrase concernant Šilhaha. On possède, du reste, et comme on le verra, un second texte
construit sur le même modèle20.

K. De Graef (MDP : ss et MDP  : passim) met fort opportunément en lumière le fait que
certains textes autrefois édités dans MDP  appartiennent aux archives d’ un certain Kûyâ. Elle
invite également à prendre en considération les vestiges des archives d’ un certain Ašiši. On y
découvre la présence conjointe des noms d’ Ebarat, de Šilhaha et d’ Atta-hušu. J’ ai déjà évoqué
le cas de Šilhaha et d’Ašiši.
– Ebarat: son nom figure dans les archives de Kûyâ, sur l’ empreinte du sceau de son serviteur
Šū-Baba21.
– Atta-hušu: son nom figure également dans les archives de Kûyâ, sur l’ empreinte du sceau
de son serviteur Adad-rabi, fils de Rı̄m-Adad22. Il figure aussi dans une formule de serment
(MDP , ) où K. De Graef, non sans hésitations, propose de lire a-ta-hu-šu-ù ma-ar
ki-da-tum, faisant d’Atta-hušu un fils de Kidattu, un roi de Simaški. Mais, selon la copie,
le signe lu MA est, en réalité, à lire BA23, et le signe lu DA est un signe ŠA24 ; il convient
donc de lire, outre le nom d’Atta-hušu, celui de ba/pá-ar-ki-ša-tim !.

18 Lambert, W.G.  : , no  et ss.

19 Selon la photographie, le signe final est LA et non RA; le logogramme rend donc la graphie du nom divin Tanra.
20 Le gobelet dit d’ Ukal : Mahboudian, H. : . Voir ci-après.
21 MDP ,  et  (corriger MDP  : ).
22 MDP , ,  et .
23 Avec Vallat, F. .
24 Le signe DA figure à la ligne  de la face. Vallat, F.  lit IŠ, ce qui est proprement impossible.
 les premiers sukkalmah et les derniers rois de simaški 

En l’état des sources, Atta-hušu porte les titres et les épithètes de sukkal et teppir de
Suse, SIPA d’Inšušinak, «pasteur d’Inšušinak », et DUMU NIN9 Šilhaha, « fils de la soeur de
Šilhaha»25, mais il n’est jamais revêtu de celui de sukkalmah. L’ association des deux titres
sukkal et teppir est intéressante en elle-même. Avant Atta-hušu, Kitten-rakittapi avait été suk-
kalmah et teppir sous le règne d’Idattu Ier ; après Atta-hušu, le futur sukkalmah Temti-Agun
sera sukkal et teppir26. Le sukkal prenant la place du sukkalmah dans ce couple, une modifi-
cation se serait-elle opérée dans la hiérarchie des fonctions, le sukkalmah étant promu à un
rang royal auquel il n’avait pas eu droit antérieurement ? Un autre indice de changement est
révélé par la présence du toponyme Simaški dans la titulature de Temti-Agun. Les sources
manquent, toutefois, qui pourraient nous permettre de quitter le terrain mouvant des hypo-
thèses.
Selon F. Vallat27, Atta-hušu aurait été placé sur le trône de Suse par Gungunum de Larsa, vers
, et il aurait toujours détenu les rênes du pouvoir en , alors que Sumu-abum était roi
de Babylone. Son règne, cumulé avec celui de son successeur Tetep-mada28, aurait couvert une
période de quatrevingts ans. Nous savons, aujourd’ hui, que ce roman est totalement dénué de
fondement.
Il est indispensable, à ce sujet, de faire une mise au point. L’ histoire n’ est pas une connaissance
objective, c’est une construction savante fondée sur une érudition relative. Le document est
le matériau de l’historien, le temps, cet espace sur lequel s’ égrènent les événements toujours
singuliers que l’historien promeut au rang de faits historiques, est sa matière première. Pour
leur donner sens, l’historien est celui qui construit une intrigue, laquelle ne se veut en aucun
cas la vérité scientifique. Car il n’y a pas de vérité historique.

La nouvelle histoire attribue aux décisions personnelles de l’ historien une place prépondé-
rante, au détriment du document. Elle a tendance à ne laisser subsister que les règles du
genre littéraire. À distance des documents, la démonstration de l’ historien post-moderne tient
dans le déploiement de sa propre narration. Partant, dans son discours, il est menacé de ne
plus rendre compte des événements en exerçant sa propre liberté de penser, comme il est
attendu, mais, se tenant éloigné de cette démarche, il est conduit à se projeter lui-même,
prêtant à ceux auxquels il est supposé donner la parole ses propres idées et ses propres fan-
tasmes. Il n’y a aucune différence, dans ce cas, entre la fiction du romancier et la fiction de
l’historien.
Pour ma part, je considère que le fait historique est le produit que font exister, de conserve,
le document que l’historien établit et le questionnement que ce même historien met en oeuvre
et auquel il soumet le document. Car l’ historien élabore le document à partir des traces dont
il dispose, traces qu’il soumet au crible d’ une critique toujours plus érudite, dans le respect
des méthodes philologique, archéologique, ethnographique et historique qui sont les siennes,
et à des interrogations toujours plus nombreuses, plus diversifiées à mesure que le temps passe,
que les sociétés changent et que les générations se succèdent. Non, le document ne disparaît

25 Pour toutes références, voir Sollberger, E. : –, et Malbran-Labat, F. , nos  et . Les traducteurs

distinguent habituellement entre sipa dmùš.erin, «pasteur d’Inšušinak», et sipa érin mùš.erin.ki, « pasteur du peuple
de Suse ». En réalité, les signes UD et ÉRIN se confondant à cette époque, et UD pouvant remplacer DINGIR,
il faut vraisemblablement rapprocher les diverses graphies existantes et ne reconnaître qu’un seul titre: (a) sipa
šà u[dm]ùš.erin, (b) sipa udmùš.erin <<ki>>, KI étant ajouté, possiblement par erreur, sous la ligne, et (c) sipa
dmùš.erin.
26 Voir ci-après.
27 Dans Steve, M.-J., Vallat, F., Gasche, H., Jullien, C. et F. –: –.
28 Ce personnage n’ est connu que par l’ inscription de son sceau: te-te-ep-ma-da / SIPA MÙŠ.ERIN ! / DUMU

NIN9 ! ši-il-ha-ha (MDP , no ).

 jean-jacques glassner

pas progressivement. André Leroi-Gourhan parlait de bribes de documents ou de sources

fragmentaires, qu’il faut rendre philosophiquement utilisables. Ce sont elles qui constituent
le matériau auquel l’historien ne peut pas ne pas se référer. Ainsi s’ établit la différence entre la
fiction du romancier et la fiction de l’ historien.
À l’appui de ses allégations, F. Vallat tente de montrer que les membres de trois générations
d’une même lignée de scribes sont au service d’ Atta-hušu, Ibni-Adad, Rı̄m-Adad fils d’ Ibni-
Adad et Adad-rabi fils de Rı̄m-Adad29. Si le doute n’ est pas permis pour les deux derniers, il
s’instille, par contre, s’agissant de l’ identification d’ Ibni-Adad, père de Rı̄m-Adad avec l’ egir
teppir du même nom30, également serviteur d’ Atta-hušu.
Bref, on relève qu’Ebarat, Šilhaha et Atta-hušu sont contemporains, la carrière du dernier,
qui est en même le cadet du trio, se déroulant principalement au service et à l’ ombre de celles
des deux autres.

Le nom d’un autre personnage figure dans les archives découvertes à Suse, celui d’ Idattu, fils
de Tan-Ruhurater. Tout comme celui d’ Ebarat31, son nom est précédé du déterminatif divin sur
le sceau de l’un de ses serviteurs32. Dans ses propres inscriptions, il porte le titre d’ ensi de Suse,
à l’instar de son père. Mais il est paré, sur le sceau évoqué à l’ instant, de l’ épithète nita.kala.ga,
«mâle puissant», une épithète qui caractérise un personnage revendiquant un rang royal. On
comprend donc qu’il désire se comporter comme un roi, même s’ il se contente modestement
du titre d’ensi. Tout ce qu’il est possible de dire est qu’ il règne à Suse. Ces faits sont connus
depuis trop longtemps pour qu’il soit utile de s’y attarder.
L’un de ses subordonnés est un scribe du nom de Sir-ahu-pitir33 ; or, si sur deux sceaux
différents, celui-ci se dit un serviteur d’ Idattu (MDP , no  et MDP  : ), sur un
troisième il se présente comme un serviteur d’ Atta-hušu (MDP , no , où il est doté
d’un patronyme). Idattu et Atta-hušu sont donc au moins partiellement contemporains l’ un de
l’autre. Semblablement, on rencontre le nom d’ un certain Atta-puni sur un autre sceau où, en
compagnie d’un dénommé Turunkuz, il est qualifié de serviteur d’ Idattu (Amiet , no ).
Sur un second sceau, le même Atta-puni se dit serviteur de Kuk-sanit34 ; on peut lire, en effet,
sur ce dernier sceau:
at-ta-pu-ni
[EGI]R SUKKAL35
[ÌR ku]-ku-sa-ni-it
«Atta-puni, egir du sukkal, serviteur de Kuk-sanit»

29 MDP , ,  et , les sceaux ; ces textes appartiennent aux archives de Šū-Baba, serviteur d’Ebarat; MDP ,

nos  et .

30 Sollberger, E.  : –. Voir, déjà, les réserves de K. De Graef, MDP : ss.
31 Steinkeller, P.  : , n. , réserve l’inscription du déterminatif divin au seul Ebarat Ier. Toutefois, en

Mahboudian, H.  nos  et , ligne , le nom d’Ebarat II auquel Šilhaha présente des offrandes, est précédé
du déterminatif divin ; sans doute, ce monarque est-il possiblement prédécédé, mais l’attribution du déterminatif à
un monarque décédé n’ est documentée que pour des individus dont les noms avaient été agrémentés de ce même
déterminatif de leur vivant.
32 Le signe dingir est très nettement visible sur la photographie (MDP , no ); la première colonne du texte

peut être lue comme suit : d i -da-du / sipa dutu / [ki.á]g? mùs.erinki / [nita ? k]ala.ga, «Idattu, pasteur de Šamaš, qui
aime( ?) Suse, mâle( ?) puissant ».
33 Sur ce point, voir, déjà, Glassner, J.-J. a.
34 Amiet, P. , no . P. Amiet lisait : at-ta-pu-ni / [Ì]R SUKKAL / [DUMU] ku-sa-ni-it ; Vallat, F. , de son

côté, proposait une autre lecture : at-ta-pu-ni / [Ì]R SUKKAL / [ku]-ku-sa-ni-it. Or, le signe en début de ligne deux
est un signe long dont le clou vertical final survit après la cassure; à la ligne trois il y a place pour deux signes dans
la cassure.
35 Ce titre est connu, voir MDP ,  : ; comparer egir teppir: Solberger, E. : : .
 les premiers sukkalmah et les derniers rois de simaški 

Kuk-sanit n’est pas un inconnu. Son nom est invoqué après celui de Pala-iššan dans une
formule de serment (MDP , ). Il figure également dans une inscription sur un gobelet
récemment publié36 :
Te-em-ti-a-[gu-un]
SUKKAL.M[AH]
ku-uk-sa-ni-i[t]
DUMU KI.ÁG.A.NI
BÍ.IN.NA.DÍM
«Temti-Agun, sukkalmah. Kuk-sanit, son fils bien-aimé a fait pour lui»
Il est présent, enfin, dans une inscription inédite, sur un gobelet kunaggu apparu sur le marché
des antiquités, lors d’une vente chez Christies, en 37 :
– Premier cartouche:
e-ba-ra-at
LUGAL an-ša-an
te-em-ti-a-gu-un
SUKKAL.MAH NIM.[MA.KI] / ù si-maš-ki
DUMU NIN9 ší-il-ha-ha
MÙŠ.ERIN ù NIM.MA
NÍG.ZI ! IN.DÍM
gú-na-gi4 KÙ.BABBAR
IN.NA.DÍM
NAM.TI.LA.NI.ŠÈ
a-na dna-pi-ri-ša
IN.NA.AN.SUM
«Ebarat, roi d’Anšan. Temti-Agun, sukkalmah d’Élam et de Simaški, fils de la soeur de Šilhaha,
a établi la justice à Suse et en Élam; il a fait (ce) gobelet kunaggu en argent pour lui (et) pour sa
propre vie il l’a offert à Napiriša»
– Second cartouche:
ku-ku-sa-ni-it
te-ep-pí-ir
MÙŠ.ERIN.KI
DUMU KI!.<ÁG> SUKKAL.MAH
te-em-ti-a-gu-un
«Kuk-sanit, teppir de Suse, fils bien-aimé du sukkalmah Temti-Agun»
Kuk-sanit est donc un fils du sukkalmah Temti-Agun ; il exerce la fonction de teppir de Suse. Il
est aussi contemporain d’Idattu II et d’Atta-hušu. Il ressort de tout cela qu’ il existe un premier
sukkalmah du nom de Temti-Agun et qui règne au tout début de la dynastie homonyme, comme
l’avait découvert, naguère, F. Vallat ( : –). Mais une question demeure en suspens :
Ebarat, roi d’Anšan, est-il décédé? Tout semble le laisser entendre. Le texte de l’ inscription
précise que le vase kunaggu est offert à Napiriša. Or, un texte élamite rapporte que Šilhaha
présente des offrandes au défunt Ebarat pour que celui-ci dirige la défunte Amma-tedak,
possiblement sa fille, sur le chemin qui la conduit, précisément, vers Napiriša38. L’ offrande d’ un
gobelet kunaggu se ferait-elle à l’occasion d’ un décès ou de la commémoration de celui-ci ?

36 Glassner, J.-J. b.

37 Je remercie vivement Mme Anne Kevorkian qui a bien voulu procurer la photographie de l’objet et a autorisé
la présente publication de l’ inscription.
38 Mahboudian, H. ,  et . Comparer avec König, F.W.   §X où l’on retrouve la même phraséologie.
 jean-jacques glassner

L’inscription du second cartouche semble confirmer cette hypothèse. En effet, elle présente une
particularité, le titre de sukkalmah y précède le nom de son détenteur au lieu de le suivre. Or,
dans les inscriptions royales élamites postérieures, lorsqu’ il arrive que le titre précède le nom du
porteur, il s’agit toujours de personnes prédécédées39. Kuk-sanit aurait donc réutilisé le gobelet
kunaggu au moment du décès de son propre père.
En un mot, Idattu, Atta-hušu, Temti-Agun et Kuk-sanit sont proches dans le temps. Au sein
de cet agrégat d’anthroponymes, Ebarat fait figure d’ aîné, Kuk-sanit de cadet.
Tan-Ruhurater, le père d’Idattu, avait épousé Me-kūbi, une fille de Bilalama d’ Ešnunna, le
neveu de Nūr-ahum, le fondateur de la dynastie. R. Whiting ( : –) propose de dater le
règne du roi d’Ešnunna de la première moitié du XXe siècle. Celui d’ Idattu serait donc à dater du
milieu ou de la seconde moitié de ce même siècle. Šilhaha et Ebarat vécurent approximativement
à la meme époque. Atta-hušu fut leur contemporain.

Temti-Agun est associé, à son tour, à Pala-iššan dans une inscription récemment publiée40 :
pá-la-iš-ša-an
DUMU NIN9 ší-il-ha-ha
ŠEŠ KI.ÁG
te-em-ti-a-gu-un
SUKKAL ù te-ep-pi-ir
UD.MÙŠ.ERIN.KI-ÀM
----------
ú-kà-al
GÌR.NITA UD.MÙŠ.ERIN.KI
gu-na-gi4 KÙ.BABBAR
IN.NA.DÍM
«Pala-iššan, fils de la soeur de Šilhaha, frère bien-aimé de Temti-Agun, sukkal et teppir de Suse.
Ukal, gouverneur de Suse, a fabriqué (ce) gobelet kunaggu en argent pour lui»
Pala-iššan n’est pas un inconnu. Une nouvelle fois, on ne peut exclure que l’ offrande du gobelet
kunaggu soit présentée à l’occasion du décès de la personne honorée ou de la commémoration
de celui-ci. Pala-iššan serait donc décédé. Son nom est invoqué dans quelques serments susiens
avec, à sa suite, soit Kuk-Kirwaš41, soit Kuk-sanit (MDP , ). Dans le second cas, après
avoir évoqué la prestation du serment, ni-iš pá-la-iš-ša-an ni-iš ku-ku-sa-ni-it u ni-iš na-pi-ri-ša
(…) it-ma, «il prêta serment par le nom de Pala-iššan, le nom de Kuk-sanit et par le nom de
Napiriša», le texte résume la formule, au début du revers, en ces termes : a-na LUGAL u ni-
iš DINGIR, «par le roi et le nom de la divinité »42. Sachant que Kuk-sanit est teppir de Suse, il
ressort que c’est Pala-iššan qui est investi du titre royal, à l’ évidence celui de la ville de Suse. Son
nom figure enfin sur les sceaux de plusieurs de ses subordonnés. Parmi ceux-ci, on découvre
un scribe dont le nom est partiellement perdu dans des lacunes43 :

39 König, F.W.   §II ;  §V ; etc.

40 Mahboudian, H. ,  ; Vallat, F. : .
41 MDP ,  et  ; corriger Stolper, M.W. :  dans ce sens. En MDP ,  figure une date qui associe

les deux noms à l’ établissement de la justice et de l’équité.

42 Formule similaire en MDP , , où sont invoqués Kuk-Našur, sukkal de Suse, et Kuduzuluš, lugal de Suse;

MDP , , où sont invoqués Kuter-Nahhunte et Temti-Agun; MDP , , où sont invoqués Siruk-duh et sa
soeur Amma-haštuk.
43 Scheil, V.  : . MDP  : , n. , signale que l’empreinte du sceau a été perdue. En , F. Vallat annonce

qu’ il a trouvé une seconde empreinte complète à Suse. Il est à craindre que cette seconde empreinte ne soit autre que
celle qui fut égarée.
 les premiers sukkalmah et les derniers rois de simaški 

[…]-ni-d[…]
DUB.SAR
DUMU Ha-aš-tù-u[k]
ÌR pá-la-iš-ša / -an
«… ni …, scribe, fils de Haštuk, serviteur de Pala-iššan»
F. Vallat suggère de lire le nom d’Ibni-Adad, qu’ il rapproche de cet autre Ibni-Adad, père du
scribe et serviteur d’Atta-hušu Rı̄m-Adad et de l’ egir teppir, toujours du même nom, également
serviteur d’Atta-hušu44. Partant, il voit dans Pala-iššan un sukkalmah qui aurait régné entre
Šilhaha et Atta-hušu45. Or, en l’état de la documentation, aucune source historique digne de foi
ne fait de Pala-iššan un sukkalmah. Seul Šutruk-Nahhunte46, beaucoup plus tard, le considèrera
comme un monarque et situera son règne entre ceux de Siwe-palar-huhpak et de Pahir-iššan.
Toutefois, si dans notre inscription Temti-Agun porte les titres ˙ de sukkal et teppir de Suse, Pala-
iššan, dans le cas où il serait son frère aîné, a quelques chances d’ être lui-même le sukkal d’ Élam,
voire le sukkalmah en exercice dont Temti-Agun serait alors le successeur. Car si l’ on ne peut
affirmer que Pala-iššan fut sukkalmah, son frère, par contre, exerça sans l’ ombre d’ un doute
cette magistrature suprême, les textes présentés précédemment le proclamant sans l’ ombre
d’une ambiguïté.
En ce qui concerne les autres anthroponymes que les sources associent à Pala-iššan, à savoir
Kuk-Kirwaš, Kuk-Nahhunte, Kuk-Našur et Tem-sanit, seul Kuk-Kirwaš, le premier d’ entre eux,
est connu pour avoir revêtu le titre de sukkalmah47. Quant aux autres, les sources, en l’ état, nous
laissent dans une complète ignorance.
Kuk-sanit faisant, faut-il le répéter, figure de cadet dans l’ agrégat des anthroponymes pré-
cédents, on admet que Temti-Agun et Kuk-Kirwaš, auxquels son nom est associé, régnèrent à
l’extrême fin du XXe siècle ou au tout début du XIXe.

Si l’on tente de comprendre la situation politique de Suse à cette époque, tout semble se passer
comme si un membre de la dynastie de Simaški en la personne d’ Idattu, avec ses subordonnés
à Suse, avait tenté de s’opposer à la prise de possession de la ville par Ebarat et ses successeurs.
Ebarat est un souverain sans patronyme connu et que rien ne permet de rattacher à la dynastie
de Simaški48. Tel n’est pas le cas d’Idattu qui tente de reprendre pied à Suse et semble y parvenir,
dépossédant, certes pour une courte durée, Ebarat lui-même ou son successeur de leur autorité
sur la ville. Toutefois, les défections des fonctionnaires et des dignitaires susiens se multipliant
à son endroit49, il apparaît que le parti d’ Anšan, progressivement, l’ emporte.
Les documents nouveaux mettent donc en évidence des luttes pour le pouvoir qui opposent
des factions rivales au sein des élites politiques élamites pour la détention du pouvoir suprême
et, plus modestement, pour la domination de la ville de Suse laquelle est perdue par le parti
simaškéen, avant d’être reconquise par lui, pour être mieux perdue par la suite.
Au sein de ces élites, Šilhaha semble être un sukkalmah au service du roi d’ Anšan, dans
la continuité de l’ère précédente. Cependant, dans l’ hypothèse où il aurait porté le titre de

44 Sur ces personnages, voir supra.

45 Vallat, F.  :  ; Steve, M.-J., Vallat, F., Gasche, H., Jullien, C. et F. –: . Une hypothèse admise
par M.W. Stolper ( : ), et rejetée par K. De Graef (MDP : ss).
46 König, F.W. , A §XVIII.
47 Sur Kuk-Kirwaš : voir, à présent, Malbran-Labat, F. , no . Sur Kuk-Našur, voir Vallat, F. : .
48 Je maintiens que cette liste ne reproduit pas une réalité historique, voir Glassner, J.-J. a; l’opinion inverse:

Steinkeller, P.  : , n. . Il est vrai que si rien ne permet de rattacher Ebarat à la dynastie de Simaški, il n’existe
aucun argument pour l’ en dissocier. Dans ce cas, deux groupes, members de la même famille, s’affronteraient.
49 On pourrait penser à un mouvement inverse, les défections se faisant au détriment du parti d’Ebarat; Kuk-sanit

étant le plus jeune des notables mentionnés dans ce dossier, cette hypothèse a peu de chances d’être recevable.
 jean-jacques glassner

sukkalmah des pays d’Élam et de Simaški, on pourrait admettre que, en s’ émancipant, il prend
son indépendance par rapport à son souverain, un geste qui le mettrait en position d’ être le
véritable fondateur de la dynastie des sukkalmah.
Le règne de Pala-iššan demeurant dans une grande incertitude, en l’ état des sources, on est
tenté de formuler l’hypothèse que Temti-Agun Ier est le successeur immédiat de Šilhaha. On
observe, dans son cas, que Simaški fait partie de ses états, le titre de sukkalmah aurait-il alors
été promu, porté par un souverain indépendant et non plus par un ministre Le débat est ouvert.

Bibliographie

Amiet, P. (): Glyptique susienne des origines à l’époque des Perses achéménides. Cahets, sceaux-
cylindres et empreintes antiques découverts à Suse de  à  (= MDP ), Paris.
———. (): «Glyptique élamite. À propos de nouveaux documents.» ArAs : –.
André-Salvini, B. (): «Foundation Document Commemorating the Construction of the Nanna
Temple by Attahushu.» In Harper, P.O., Aruz, J., Tallon, F. (Eds), The Royal City of Susa, New York:
–.
Carter, E., Stolper, M.W. (): Elam: Surveys of Political History and Archaeology, Berkeley-Los
Angeles-London.
De Graef, K. (): Les archives d’Igibuni. Les documents Ur III du chantier B à Suse (= MDP ), Gand.
———. (): De la dynastie Simashki au sukkalmahat. Les documents fin PE II–début PE III du chantier
B à Suse (= MDP ), Gand. ˘
De Meyer, L. (): «Epart sukkalmah?» In Beek, M.A., de Liagre Böhl, F.M.T. (Eds.), Symbolae Biblicae
et Mesopotamicae Francisco Mario Theodoro de Liagre Böhl Dedicatae (= Studia Francisci Scholten
Memoriae Dicata ) Leiden: –.
———. (): «Les structures politiques en Susiane à l’époque des sukkalmah.» In Finet, A. (Ed.), Les
Pouvoirs locaux en Mésopotamie et dans les régions adjacentes, Bruxelles: –.
Duplouy, A. (): Le Prestige des élites, Paris.
Glassner, J.-J. (): «ruhušak – mār ahatim: la transmission du pouvoir en Élam.» JA : –.
———. (a): «Les dynasties d’Awan et de Simaški.» NABU : .
———. (b): «Kuk-Kirwaš, sukkalmah.» NABU : .
Grillot, F., Glassner, J.-J. (): «Problèmes de succession et cumuls de pouvoirs: une querelle de famille
chez les premiers sukkalmah?» IrAnt : –.
König, F.W. (): Die elamischen Königsinschriften (= AfO Beih. ), Graz.
Lambert, W.G. (). «Near Eastern Seals in the Gulbekian Museum of Oriental Art.» Iraq : –.
Mahboudian, H. (). Elam, Art and Civilization of Ancient Iran, – B.C., Salisbury.
Malbran-Labat, F. (): Les Inscriptions royales de Suse: Briques del’époque paléo-élamite à l’Empire
néo-élamite, Paris.
Scheil, V. (): Textes élamites-sémitiques, quatrième série (= MDP ), Paris.
———. (): Actes juridiques susiens (= MDP ), Paris.
———. (). Actes juridiques susiens (suite: nº  à nº ) (= MDP ), Paris.
———. (). Actes juridiques susiens (suite: nº  à nº ) (= MDP ), Paris.
———. (): «Raptim.» RA : .
———. (): Mélanges épigraphiques (= MDP ), Paris.
Seux, M.J. (): Épithètes royales akkadiennes et sumériennes, Paris.
Sollberger, E. (): «A tankard for Atta-hušu.» JCS : –.
Sollberger, E., Kupper, J.-R. (): Inscriptions royales sumériennes et akkadiennes, Paris.
Steinkeller, P. (): «New Light on Simaški and Its Rulers.» ZA : –.
Steve, M.-J., Vallat, F., Gasche, H., Jullien, C. et F. (–): «Suse.» DB Suppl. –: –.
Stolper, M.W. (): «On the Dynasty of Šimaški and the Early Sukkalmahs.» ZA : –.
———. (): «Pala-iššan.» RlA : .
Vallat, F. (): «Les trois Kuk-Našur.» NABU : .
———. (): «Temti-Agun I. Un nouveau sukkalmah.» Akkadica : –.
———. (): «Du règne de Kindadu à celui d’Atta-hušu.» NABU : .
Vallat, Le scribe Ibni-Adad et les premiers sukkalmah, NABU /.
Whiting, R.M. (): Old Babylonian Letters from Tell Asmar (= AS ), Chicago.
 LA «SUPRÉMATIE ÉLAMITE » SUR LES AMORRITES.
RÉEXAMEN, VINGT ANS APRÈS LA XXXVIE RAI (1989)

Jean-Marie Durand*

A Gand, en , lorsque nous avons présenté, Dominique Charpin et moi-même, notre
communication «La suzeraineté de l’empereur d’ Élam sur la Mésopotamie » (Charpin/Durand
), notre auditoire n’était certainement pas bien préparé à une telle vision des choses. Dans
leur excellent manuel, E. Carter et M. Stolper (Carter/Stolper ) considéraient plutôt la
période de Mari comme celle où le pouvoir de l’ Élam était limité à l’ Est, non comme s’ étendant
à toute la Plaine, comme semblaient pourtant déjà le montrer de plus en plus les archives de Mari
dont nous avions connaissance.

– Quasiment tous les renseignements qu’ ils avaient sur l’ Élam pour l’ époque paléobabylo-
nienne dans leur ouvrage (Carter/Stolper  : ), provenaient pourtant de Mari car la plupart
des références à l’Élam de RGTC : – en venaient, sans compter, depuis , l’ excellent
index des publications des textes de Mari qu’ offrait ARM XVI/.
Ce qu’ils savaient tenait en quelques faits massifs et importants :
(a) L’Élam avait su prendre le contrôle d’ Ešnunna ; des sceaux de documents originaires
de Malgi"um parlaient de «serviteurs de Kuduzulush ». Ils en concluaient logiquement :
«These are signs that rulers of Elam held influence, if not suzerainty, over eastern Meso-
potamian centers.»
Nous proposions donc une vision beaucoup plus large de cette influence, voire suzeraineté.
(b) Ils notaient la présence de messagers élamites au Proche-Orient et étaient au courant de
la route de l’étain (cf. aussi Joannès  : –).
Sur ce dernier point Georges Dossin ( :  sq.) avait tout à fait raison contre Leemans,
à ceci près que cette route de l’étain n’ a duré que trois ans, ce qui est peu par rapport à
l’autre route de l’étain, celle qui conduisait d’ Aššur en Cappadoce.
(c) Cependant beaucoup de leur présentation synthétique est aujourd’ hui dépassé car leurs
références étaient présentées un peu « en vrac » : depuis , ce qui était une entreprise
encore inconnue en , voire toujours en , il y avait eu, de fait, un grand effort des
mariologues pour collationner les documents et mettre en place la chronologie relative
et la géographie, tout particulièrement pour le règne de Zimrî-Lîm et la chronologie des
rois d’Ešnunna1. Les premiers résultats n’ en pouvaient être bien perçus qu’à partir de la
publication des ARM XXVI/ et XXVI/ ().
Parmi les gains les plus importants qui ont été faits, on compte :
(a) le réajustement concernant l’affirmation : « Shortly after Shamshi-Adad’s death, his suc-
cessor, Ishme-Dagan, assured the Assyrian regent at Mari that Elam and Eshnunna were
held in check.»

* Collège de France. Merci à Dominique Charpin pour sa relecture et les références ajoutées ici.
1 Toutes les références ne seront pas données dans cet article qui se réfère implicitement à l’exposé fondamental
sur l’ histoire événementielle de Mari publié par Dominique Charpin et Nele Ziegler, FM V, .
 jean-marie durand

Il s’agit là de ARM IV , un document dont la mé-compréhension a été très dommageable

pour l’histoire de la période; on le considère désormais (et sûrement) comme appartenant
à la correspondance entre deux petits rois de Haute-Djéziré, sous Zimrî-Lîm, avec dès lors
une autre interprétation du passage crucial (LAPO  ) ;
(b) la chronologie nouvelle ne permet plus une présentation comme, « combined armies of
Elam and Eshnunna invaded Idamaraz on the Balikh or Khabur » (Carter/Stolper  :
), même si l’on a de bonnes raisons de penser qu’ Atamrum, installé à Ešnunna, a
conduit les forces élamites contre les vassaux mariotes du Nord.
Vingt ans après notre communication, il importait donc de refaire le point. Les informations
ponctuelles se sont assurément multipliées depuis , au gré des publications et des déchif-
frements (éventuellement toujours aujourd’ hui inédits). Il n’y a rien jusqu’ ici dans l’ apport de
la documentation nouvelle qui permette une rupture fondamentale avec la globalité de la vision
que nous présentions D. Charpin et moi-même. La fourchette chronologique s’ est considéra-
blement précisée, mais la toute puissance de l’ Élam, voire sa suzeraineté, au moins entre la e
année de Zimrî-Lîm et sa e année, selon le nouveau comput de  ans de règne établi défini-
tivement établi pour ce roi par D. Charpin, est désormais un fait avéré.
Il faut ajouter ajouter néanmoins que ce n’ est cependant qu’ un moment de l’ histoire amorrite
et il faut désormais essayer d’expliquer le pourquoi lui-même de cette suzeraineté.

. La multiplication des renseignements sur l’ Élam

Les références à l’Élam ne concernent pas tant l’ histoire de Mari que celle en général de tout le
Proche-Orient que l’on connaît désormais de plus en plus dans le détail :
On peut ainsi énumérer:
(a) la prise d’Ešnunna par l’Élam, dont la date exacte reste néanmoins toujours sujet à débat ;
(b) les problèmes relationnels entre Babyloniens et Élamites à Babylone avec en corollaire la
montée de l’antagonisme entre Babylone et Larsa, la pro-élamite ;
(c) l’invasion de la Haute-Djéziré par l’ Élam, qui est de mieux en mieux connue, et la division
entre pro-élamites et pro-mariotes des rois de la Haute-Djéziré, au sein desquels émerge
désormais Arriyuk, prince de Kalhu, qui semble avoir joué double jeu (Durand ) ;
(d) la lutte des princes amorrites de la plaine mésopotamienne, au premier rang desquels il
y a Babylone et Mari, dans son sillage, aidés par Alep, contre les puissances du Plateau
Oriental, lutte qui culmine à la bataille de Hirîtum qui entraîna le repli élamite ;
Dans un autre domaine, nous sommes bien au courant des relations économiques entre Est et
Ouest:
(a) le commerce de l’étain via Ešnunna (Joannès ) ;
(b) le commerce des pierres précieuses entre Larsa et Suse (Guichard ) ;
(c) la diffusion vers le grand Est des formes artistiques (vaisselle et motifs) de l’ Ouest, Crète
et centres syriens, grâce à ARM XXXI, établi par M. Guichard.
Il y a donc bien là une question de sources : le silence total ou relatif d’ Aššur, de Babylone,
de Sippar, d’Ur et de Larsa – dont les textes nourrissent les rubriques « paléo-assyrien » et
«paléo-babylonien» de nos dictionnaires – doit nous faire réfléchir sur l’ utilisation toujours
très aventureuse de l’argumentum a silentio dans nos sources cunéiformes. Seules des archives
non locales et internationales, fournissant une très abondante documentation d’ ordre politique
sur des sujets internationaux, peuvent nous renseigner sur la réalité des relations entre États au
Proche-Orient au XVIIIe siècle. Pour diversifier cette documentation en fonction des différents
horizons nationaux, il faut attendre que soient trouvées ailleurs d’ autres archives analogues.
 la «suprématie élamite » sur les amorrites 

La documentation de Mari suffit néanmoins aujourd’ hui à nous montrer que l’ Élam est omni-
présent au Proche-Orient dans la seconde moitié du règne de Zimrî-Lîm et qu’ il faut le consi-
dérer comme l’un des acteurs majeurs de la scène internationale.

. Confirmation du leadership de l’ Élam

Vingt ans après notre communication de Gand, nous n’ avons donc pas l’ impression, à la lecture
accrue des textes de Mari, que le poids politique de l’ Élam au Proche-Orient à l’ époque des
textes de Mari ait diminué, mais même qu’ il s’ est accru.

Voici les faits qui montrent au mieux son premier rang.

(a) On apprend que c’est l’Élam qui a réparti les territoires de Samsî-Addu après la disparition
de ce dernier et qui a attribué Hît au royaume de Mari. C’ est là une information que l’ on
tire d’une allusion anecdotique (ARM XXVI ), pour laquelle nous manquons toujours
de détails, alors que les documents de Mari ne nous parlent pour la région concernée que
de la rivalité d’influences entre Ešnunna et de Mari, au détriment de Babylone qui semble
avoir ainsi payé son alliance avec le royaume de Haute-Mésopotamie.
(b) On voit la reconnaissance du pouvoir de l’ Élam s’ étendre jusqu’à l’ extrême-Occident
puisque la lointaine Qatna dit que « le pays lui est donné »2. La Bible doit sans doute
garder un lointain souvenir˙ du fait lui-même qui avait fort impressionné les contempo-
rains (à en juger par les échanges épistolaires entre Alep et Mari), lorsqu’ a été rédigée à
une époque bien plus récente3 l’histoire d’ Abraham, c’ est-à-dire un des souvenirs histo-
riques les plus anciens de la Genèse, portant sur les Patriarches historiques, ceux dont
la vie est en rapport avec les pays bien connus du Proche-Orient, car il est bien plus
vraisemblable de retrouver le souvenir d’ un roi de Qatna, Amudpa"el, derrière cet énig-
matique «Amraphel» (Durand  : ), sectateur des˙Élamites, que celui du plus illustre
(au moins pour nous!) Hammu-rabi de Babylone, certainement totalement oublié par
l’Occident récent.
(c) Quand l’Élam engage les opérations contre Ešnunna, il requiert l’ aide des princes amor-
rites, tout particulièrement de Babylone et de Mari, lesquelles obtempèrent apparemment
sans discuter. Le sukkal-mah a, en outre, l’ appui inconditionnel de Larsa, laquelle devait
payer au prix fort, par la suite, sa fidélité à l’ alliance élamite tout en opposant, semble-t-il,
Babylone et Larsa.
Pourtant Ešnunna nous apparaîtrait comme un pays frère des princes amorrites, tout par-
ticulièrement de Babylone, constituant effectivement la partie orientale du pays d’ Akkad.
Quand on parle des Ešnunnéens en Haute-Djéziré, on dit naturellement en parlant d’ eux :
«les Akkadiens» et il semble y avoir eu des liens si forts entre Ešnunna et Babylone
que Hammu-rabi aurait pu s’installer sur le trône d’ Ešnunna, restaurant l’ unité du pays
d’Akkad4.

2 Cf. ici-même la reprise du texte FM VII  avec un nouveau commentaire par D. Charpin, p. xx.
3 Le caractère récent de la rédaction pourrait être indiqué par le rattachement de cet Amudpa"el/Amraphel à
la ville de Babylone, ce qu’ indique la désignation de son pays par Shinéar, certainement dans la Bible récente une
désignation de la Babylonie. Quelle que soit l’ origine de cette appellation, pour un rédacteur récent du Ier millénaire
qui ne connaît plus à l’ Est que l’ empire babylonien comme grand pouvoir politique (l’ancienne Qatna, l’actuelle
˙
Mishrifé, n’ étant plus qu’ une ville déchue politiquement), dire que le pouvoir d’Amraphel était à Babylone devait
aller de soi.
4 Voir pour cela le commentaire à LAPO  .
 jean-marie durand

(d) Quand l’Élam envahit la Haute-Djéziré, après la chute d’ Ešnunna, on ne comprend pas
bien pourquoi le roi de Mari semble manifestement le laisser d’ abord faire à sa guise, alors
que des alliés de toujours, comme le roi de Razamâ, sont attaqués, et que l’ on voit ses
vassaux se ranger peu à peu du côté de l’ Élamite.
L’explication que nous en avons proposée est que l’ on sentait l’ Empereur d’ Élam chez lui,
non pas comme un envahisseur. Arriyuk semble lui livrer passage sans discussion au gué
de Kalhu (Durand : –). Ce n’ est que par la suite qu’ il sentira la nécessité d’ une
légitimation.
(e) On note, d’autre part, le ton incroyable d’ arrogance dont usait le sukkal-mah envers les
princes amorrites, ce qui fait l’objet de la communication de D. Charpin ici-même.
(f) Il y a plus: c’est Babylone qui engage les hostilités contre l’ Élam, dans une lutte à mort, sur
son territoire, alors qu’elle a obtempéré l’ année d’ avant à l’ ordre d’ envoyer des troupes
contre Ešnunna; Mari donne l’ impression de s’ être laissée entraîner à la guerre contre
l’Élam, même si elle suscite l’arrivée, par la suite, des forces de l’ extrême Occident, tout
particulièrement d’Alep. On a l’ impression d’ une prise de conscience tardive du danger
que représente l’Élam et que cette résistance revêt, au moins à Mari, les apparences d’ une
révolte contre un suzerain naturel.
(g) Lorsque Zimrî-Lîm s’affronte à l’ Élam, le roi de Mari donne l’ impression d’ entreprendre
une guerre impie, du genre de celle qu’ un vassal ne pouvait engager contre son suzerain
au Moyen-Âge. Un problème crucial à l’ époque est effectivement celui de la pyramide des
engagements: on voit par de nombreux exemples, dans diverses situations, que ce n’ est pas
là un cas d’école: ceux qui sont en bas de la pyramide des engagements doivent obéissance
au sommet. Il n’y a pas de fidélité qui s’ arrête à un niveau intermédiaire.
– Un des faits majeurs de la vie politique de Mari, et dont il faut désormais tenir compte, tient
aux grands serments jurés par la population du royaume de Mari, tout particulièrement les
femmes du royaume, en ZL  (= 0) au moment où le roi de Mari engage la guerre avec l’ Élam
aux côtés de Babylone (cf. Charpin/Durand  : ).
– De grandes prophéties, en outre, se multiplient dans le royaume de Mari, dont la plus
spectaculaire reste ARM X  (cf. LAPO  ). Il apparaît désormais de plus en plus que
les prophéties ne sont pas des messages divins gratuits : ils servent à légitimer l’ action royale
à des moments difficiles: elles se répartissent ainsi entre révolte des Benjaminites, paix avec
Ešnunna, guerre contre l’Élam, puis contre Babylone.
Les gens qui avaient juré obéissance au roi de Mari ont dû se poser la question, lorsque ce
dernier rompit son engagement envers l’ Élamite, de savoir quelle devait être leur attitude envers
leur suzerain immédiat.
Le cas s’est en tout cas posé pour ce qui est des vassaux de l’ Ida-Maras, comme le roi d’ Ilân-
sûrâ que le chef de la garnison mariote voit avec inquiétude céder peu à˙ peu aux sollicitations
˙du chef des forces élamites dans l’est de la Haute-Djéziré.
Cela est encore plus net en ce qui concerne la population de Tuttul, tenue à verser l’ impôt
au roi de Mari pour la guerre contre l’ Élam, mais qui choisit d’ ensorceler le tribut versé,
apparemment pour que le roi de Mari, en l’ utilisant, perde la guerre. Cela est un bel exemple
de «respect» de deux engagements contradictoires5.
Tous ces faits se combinent pour renforcer l’ idée que la lutte du roi de Mari contre l’ Élam est
bien celle, impie, d’un vassal contre son suzerain.

5 Pour ce texte remarquable, on se reportera à la citation préliminaire et au commentaire qu’en fait M. Guichard

(Guichard b : –).

 la «suprématie élamite » sur les amorrites 

. Le nationalisme amorrite

Il faut essayer maintenant de comprendre ce phénomène, en somme peu attendu, et, en tout cas,
en dissonance avec tout ce que l’on connaît pour d’ autres époques des relations de l’ Élam avec la
Plaine: pourquoi le Sukkal-mah d’Élam est-il à l’ époque amorrite accepté comme leur suzerain
par tous ces Occidentaux, alors qu’ils semblent tout prêts à se rebeller contre sa suprématie, et
d’ailleurs passeront très vite à l’acte?
Or, alors que nous commencions à parler de la prédominance de l’ Élam, nous avions
senti, Charpin et moi-même, qu’il nous fallait aborder complémentairement le problème du
nationalisme amorrite.
(a) La lettre à Zimrî-Lîm d’un chef benjaminite uprapéen, Hammî-ištamar, que j’ ai publiée
dans les Mélanges J. Perrot sous le titre de « fourmis blanches et fourmis noires » (cf.
désormais LAPO  ), n’a pas d’ autre sens que de montrer la parenté des deux branches
fondamentales des Bédouins qui constitue le « Hana », le groupe des gens qui ne vivent
pas que dans des villes, les Bensim’ alites et les Benjaminites, et l’ altérité des Élamites par
rapport à eux.
Quel que soit, en effet, le problème que semble poser au point de vue philologique le terme
de rimmatum – qui doit désigner un insecte ailé et dans aucun contexte sûr ne signifie une
«perle», comme d’aucuns l’ont répété –, le sens général du document est assez net comme
l’a dit Charpin dans FM V: l’altérité des Bédouins avec les Élamites est comprise comme
«raciale», et, cela, au niveau le plus primaire, celui de la couleur de la peau6.
(b) La lettre sur les prises de grands serment de l’ année ZL , publiée dans les Mélanges en
l’honneur de Cl. Wilcke (Charpin/Durand  : –), montre la haine de l’ Élam dans
le royaume de Mari: devant la montée des Élamites une femme parle de se réfugier dans
une forteresse, ce qui se fait effectivement lors d’ une invasion, mais son fils lui répond que
ces places ne sont pas aussi sûres qu’ un affrontement direct et que les « hommes » sont faits
pour aller à la guerre.

. L’Élam et la culture mésopotamienne

Il apparaît désormais de plus en plus qu’ il y avait une altérité profonde de mentalité entre
les gens de la Plaine et les gens d’Élam, même si la culture matérielle pourrait sembler les
rapprocher.
(a) On croirait à une communauté culturelle lorsque l’ on voit des gens envoyés d’ Élam jusqu’à
Hît pour se soumettre à l’ordalie, comme le dit ARM XXVI  : « Le Sukkal d’ Élam a
envoyé  Élamites pour plonger dans le fleuve. Alors le Sukkal d’ Élam a dit aux gens qui
nous avaient escortés: «Faites plonger les deux hommes ; c’ est quand vous retournerez à
Mari, qu’il faudra les emmener (avec vous) ». »7

6 Il est, pour l’ heure, impossible de savoir si l’opposition est de pure rhétorique, comme celle fondamentale de
« blanc » et « noir », ou s’ il faut la comprendre comme «naturaliste», en y voyant le fait que les Élamites utilisaient pour
leur opérations militaires des corps mercenaires venus de régions mélanodermes, du genre de celles qu’ils devaient
cotoyer à leur Orient, le pays de Meluhha, comme l’appelaient les sources sumériennes. Suse était, en effet, à l’époque
amorrite une ville dont la civilisation ne devait pas différer beaucoup de celle des grandes villes de la plaine. Il est
impossible, en revanche, de savoir comment se présentait la population d’Anshan où résidait apparemment le vrai
pouvoir élamite.
7 šukkal Elamtim šina elamâyî-ma ana Id šalê" im ittardam; u šukkal Elamtim ana awîlî âlik idi-ni ki" âm iqbi

ummâmi : šina awîlî šunûti Id šušlê-ma attûnu ana Ma[ri˙˙ alâkim], u šunûti terrânim.
 jean-marie durand

Le passage est néanmoins unique ; c’ est d’ autre part un moment privilégié, celui qui date
du début de l’affrontement avec Ešnunna (l. –) et de la jonction des troupes de Mari
avec les forces élamites et babyloniennes (l. –). On ne sait pas quels étaient ces deux
Élamites que l’on envoie à Hit, mais il est possible qu’ il s’ agisse de gens à qui l’ on fait
subir l’ordalie parce qu’il serait difficile de les condamner autrement ; on pouvait dès lors
s’en remettre au subterfuge de la justice divine, comme c’ en est le cas pour la reine du
Zalmaqum, cette femme représentant également un cas unique, qui ne pouvait être résolu
qu’en le commettant aux Dieux (ARM XXVI ).
(b) Des textes inédits montrent que des Élamites gardaient le sanctuaire international de
Terqa, au moment de la rébellion des Benjaminites ; ils semblent avoir été les garants de
la liberté du culte. Cette sorte de « corps expéditionnaire international » devait être tenu
pour «non engagé» et au dessus des problèmes locaux.
Cependant:
(c) S’il y a abondance de renseignements touchant à la politique événementielle concernant
les rapports de l’Élam avec Mari, les gens des Bords-de-l’ Euphrate semblent très mal au fait
de sa civilisation et on ne peut rien dire de cette dernière à partir des documents retrouvés
dans le palais de Mari.
– Le nom même de l’Élam (Haltammatum) que l’ on a cru un moment attesté par un texte
de Mari en a désormais disparu suite à une collation et n’ est plus que le NP d’ un particulier :
Haldu-Muluk (Durand : ).
– On hésite à Mari entre le titre spécifiquement élamite « sukkal-mah » et celui de lugal =
šarrum, «roi» (Šeplarpak, «roi d’Anšan » [ARM XXIII  ; XXV , , , ]) qui est la façon
banale d’appeler un souverain.
– L’Élam, c’est Anšan et Suse, l’une avec l’ autre, sans plus. La spécificité du système royal
élamite, tel qu’il est reconstitué aujourd’ hui par certains chercheurs (Grillot/Glassner ),
n’apparaît nullement. Les gens des Bords-de-l’ Euphrate ont dû plutôt avoir l’ impression qu’ il
y avait plusieurs rois en Élam, l’un à Anšan, l’ autre à Suse, sans qu’ on les situe bien l’ un par
rapport à l’autre8.
– Le nom même des dynastes n’est pas très bien connu9 : si certains textes parlent bien de
« Siweparlarhuhpak» d’Anšan, ce nom est aussi raccourci en « Šeplarpak » ; pour ce qui concerne
˙
Kuduzulluš de Suse, on se reportera aux deux notes de M. Guichard (Guichard  et ) ;
il y a désormais  mentions inversées du NP du souverain de Suse et une seule conforme à
ce que l’on attend. On pourrait se demander si ce « ŠulŠI-Kudur » que l’ on constate au lieu
de «Kuduzulluš», n’est pas une réinterprétation du NP non compris, plutôt qu’ une faute.
L’onomastique élamite est, en tout cas, réduite à celle de messagers ou chargés d’ affaires10 ainsi
que de très hauts personnages11 pour lesquels des variantes révèlent le sentiment d’ étrangeté
(ou de difficulté phonétique) qu’elle suscitait chez les Mariotes.

8 Cf. ARM XXIII  :  , šûbultum à Šeplarpak lugal Anšan, et :, šûbultum à Kudušuluš lugal Šušim (réf.
D. Charpin).
9 Le phénomène peut paraître semblable pour de petits rois de l’Ida-Maras; on constate ainsi que l’on appelle

éventuellement Išme-Addu Išîm-Addu (cf. Charpin :  n.  et ). Faut-il ˙ y voir néanmoins autre chose
qu’ une simple faute ? Écrire « Addu a exaucé» ou «Addu a décidé du sort» peut n’être pas innocent. Pour la
déformation de Simah-ilânê (« Joie de mes dieux») en Simma-Ila «Fléau de dieu», cf. le commentaire LAPO :
.
10 Ainsi Kayyayya se présente-t-il aussi comme Kuyyayya.
11 Ainsi trouve-t-on pour le vice-roi de Šubat-Enlil à la fois Kunnam et Kunnaman (ARM XXVI/).
 la «suprématie élamite » sur les amorrites 

On peut mentionner aussi le cas du «zibir Anšan », pris par les Mariotes pour un nom propre,
alors qu’il s’agirait du titre du «ministre », selon Fr. Vallat12.
– Il n’y a pas de mots élamites dans la culture mariote, à part éventuellement l’ habit mazzum
(cf. l’index de ARM XXX).
– Il n’y a pas d’alliance matrimoniale connue entre princes et princesses du Plateau et de la
Plaine.
(d) Ce qui apparaît désormais clairement, c’ est l’ idée de la force terrible et implacable de
l’Élamite:
– «L’Élam dévore aussi bien son ennemi que son allié. »13
– Dans ARM XXVIII  Arriyuk se plaint : « auparavant, l’ Élamite « avait fait sortir » mes
messagers pour les tuer.» (Durand  : )
– Le ton du sukkal-mah envers les dynastes de la plaine est d’ une arrogance analogue à celle
dont le grand roi (ho megas Basileus) parle aux Grecs.

. L’ombre des Benjaminites

Or, désormais, il est facile de voir que ce que l’ on dit de l’ Élam on le disait aussi d’ Ešnunna et
que les deux États ont même réputation auprès des contemporains.
– Dans une lettre éditée par Nele Ziegler, FM IX , on souligne à propos d’ Ešnunna le
caractère impitoyable de ces derniers: « Les messagers, gens d’ Ešnunna, d’ Élam et de Babylone,
se sont levés et ont affirmé: «Nous voulons nous en aller, pour l’ amour de Dagan et d’ Itûr-Mêr !
Ešnunna n’a (jamais) distingué pour (le) dévorer14 entre ce qui est à son adversaire et ce qui est
à sa marque15. Nous voulons (nous en) aller ! » »
– L’indifférence portée au caractère d’ inviolabilité des ambassadeurs par le roi d’ Ešnunna
est bien soulignée à propos de l’envoyé de Qatna par Zimrî-Lîm selon ARM XXVI  et ARM
XXVIII : «Je l’ai retenu chez moi. Ils ont mis˙ à mort ton messager précédent ; alors à présent,
allons-nous de la même façon jeter au feu celui-ci ? » (cf. ARM XXVIII : –)
On gagne ainsi l’impression qu’Ešnunna et l’ Élam font partie du même groupe de barbares
sanguinaires qui adoptent des conduites qui semblent innaceptables aux gens de la Plaine.

On peut donc se demander dans quelle mesure Ešnunna n’ était pas pour les Amorrites de la
Plaine comme un pays complètement inféodé aux pratiques de l’ Élam, voire même comme un
avant-poste élamite et s’il ne faut pas commencer, en conséquence, à rechercher la trace de
l’Elam à l’ombre de la documention sur Ešnunna.

L’Élam semble, de fait, très absent des préoccupations mésopotamiennes à l’ époque du royaume
de Haute-Mésopotamie, alors qu’au même moment Ešnunna tient une place extrêmement
importante. On a jusqu’ici pensé, au moins parmi les mariologues, que c’ était en fonction de
la puissance d’Ešnunna que l’Élam apparaît ou non et que la première faisait éventuellement
écran à la seconde. De fait Ešnunna nous apparaît comme une puissance colossale et, de ce fait,

12 Vallat  :  (réf. D. Charpin).

13 Cf. ARM XXVI  (réf. D. Charpin).
14 En m. à m. « a (toujours) réuni pour son ventre ce qui est à son adversaire …».
15 Nele Ziegler a compris « ce qui est à sa frontière»; j’interprête, plutôt, «ce qui porte son signe», à partir de

ittum.
 jean-marie durand

indépendante. Ce fait pose d’emblée le problème des relations entre Ešnunna et Élam : en fait,
Ešnunna pourrait s’interpréter dans une certaine façon comme une expression de la politique
élamite envers la Plaine mésopotamienne.

– On comprend le poids dont pèse le plateau iranien sur ceux qui se trouvent à l’ Est du Proche-
Orient, mais, il est plus difficile d’expliquer que l’ influence de l’ Élam aille si avant vers l’ Ouest
qu’elle atteigne Qatna.
˙
Ce qui a été compris comme une « trahison de Qatna » – ce qui a tant choqué les princes de
la Plaine au moment où ils avaient engagé leur lutte contre˙ les puissances du plateau iranien –
intervient assurément dans un climat d’ hostilité permanente entre les deux grands centres
politiques de l’Ouest, Alep et Qatna. On retrouverait là le jeu normal de la géopolitique de
l’époque: le simple fait qu’Alep soit ˙ opposée à l’ Élam ferait que Qatna, son éternelle rivale,
éprouverait de la sympathie pour l’Élam. ˙

Mais cela n’expliquerait toujours pas la réaction pro-élamite de Tuttul, dont la soumission
à Mari (il y a un hazzanum de Mari à Tuttul) est constitutive de l’ équilibre géopolitique de
l’époque, elle qui est si loin de l’Élam, au confluent du Balih et de l’ Euphrate.
C’est peut-être néanmoins le particularisme de Tuttul qui pourrait nous faire comprendre
mieux ce que nous prenons pour la suzeraineté de l’ Élam sur la Plaine.

Poser la question des sentiments pro-élamites de Tuttul revient en fait à entrer dans l’ examen
de la complexité des alliances tribales de l’ époque. La région du Balih où se trouve Tuttul est
de peuplement et d’obédiences benjaminites. Or, il s’ agit là de gens qui ont plus que de la
sympathie pour Ešnunna; cette dernière cependant a été vaincue par l’ Élam et, de ce fait, on
s’attendrait à ce que Tuttul veuille venger Ešnunna et soit anti-élamite.
(a) L’affaire est beaucoup plus complexe qu’ il n’y paraît.
– Nous savons, aujourd’hui, qu’Ešnunna a soutenu contre Zimrî-Lîm les princes benjaminites
et, cela, de façon humiliante pour le roi de Mari.
– Ešnunna est entrée en guerre contre Mari pour soutenir des « frères de race ».
– C’est un prince benjaminite Yagihhaddu qui commande les armées d’ Ešnunna qui s’ en
viennent attaquer les alliés de Mari au Nord (ARM XXVIII  :  et ARM XVIII  : ).
– La paix ne se fit entre Mari et Ešnunna qu’ en redonnant leur place aux Benjaminites, alors
que cela entraîna de grands bouleversements dans l’ organisation interne du royaume de Mari.
– Des Benjaminites se réfugièrent, après la révolte, à Ešnunna plutôt que de revenir dans leurs
possessions euphratiques (ARM XXVII ).
(b) Un autre fait essentiel est que l’ Élam a bien triomphé d’ Ešnunna mais que cette dernière
passe dès lors du côté de l’Élam et aide ce dernier à pousser son avantage vers le Nord.
Cela a été le rôle remarquable de quelqu’ un comme Atamrum qui se comporte en vice-roi
élamite d’Ešnunna et de Haute-Mésopotamie orientale, avant de devenir roi d’ Andarig.
C’est un exemple de ces carrières très embrouillées pour nous comme on en trouve tant à
l’époque amorrite.
Une fois vainqueur, et semble-t-il assez vite, d’ Ešnunna (la ville ne devait être détruite que
lorsque l’Élam s’est replié sur le plateau, devant l’ échec de ses entreprises), les Élamites
mettent alors leurs pas dans ce qui était jusque là une politique toute ešnunnéenne :
pousser vers le Nord-Ouest. Il est, dès lors, possible que ce que nous prenons pour une
«conquête élamite» n’ait été qu’ une remise au pas d’ un État vassal et l’ éviction du parti
nationaliste qui devait être le fait du roi local, Ibâl-pî-El II. On comprendrait dès lors
 la «suprématie élamite » sur les amorrites 

pourquoi Babylone et Mari dans un premier temps ont soutenu ce qu’ elles pouvaient
considérer comme une décision légitime de suzerain et comme la chance d’ affaiblir un
rival gênant, alors que, dans un second temps, lorsqu’ elles se sont rendu compte de la
lattitude que cela donnait au puissant Élam d’ intervenir dans la plaine, elles ont décidé de
résister à ce dernier.
(c) Nous ne savons pas comment la composante benjaminite de l’ Orient du pays d’ Akkad
se situait par rapport à l’Élam. Manifestement elle ne soutenait pas le parti nationaliste
et devait incarner l’Ešnunna pro-élamite. Mon idée est donc qu’ Ešnunna a dû faire sa
soumission à l’Élam de telle façon que les autres Benjaminites qui lui sont alliés de par le
Proche-Orient en fonction de liens tribaux très puissants se sont sentis ipso facto du côté
de l’Élam au moment où Ešnunna est désormais rentrée dans son alliance.
(d) En tout cas, Zimrî-Lîm eut bien du mal à mobiliser les Benjaminites pour aller faire la
guerre contre l’Élam16.
– En ce qui concerne son attentisme au moment de l’ invasion élamite des régions du Nord,
il est possible qu’il se soit rendu compte qu’ entrer en lutte frontale avec l’ Élam susciterait
une nouvelle rébellion d’une partie de ses sujets benjaminites. Ce que nous avons compris
comme l’acceptation de sa part que l’Élam se conduise en maître dans des territoires qui étaient
inféodés à Mari n’a dû être en fait que de la temporisation. Il est désormais intéressant de
voir que, parallèlement aux prophéties et aux serments populaires, c’ est auprès de certains rois
benjaminites du royaume que Zimrî-Lîm a trouvé l’ appui le plus net de fidélité. Ces derniers,
c’est lui d’ailleurs qui les avait nommés et, de ce fait, il avait fait arriver au pouvoir des gens qui
pensaient différemment de ce que d’autres Benjaminites pouvaient estimer avant eux.
– Lors de l’affrontement avec l’Élam, les Benjaminites devaient donc être divisés entre eux :
si une partie n’a pas voulu (et avec détermination !) de la guerre contre l’ Élam, une autre s’ est
rangée résolument derrière le roi de Mari. Le fait qu’ Alep entre dans la coalition anti-élamite a
dû peser d’un grand poids auprès de certains, comme l’ ethnie des Rabbéens, par exemple.

Conclusion

En conclusion, l’examen de la suprématie (momentanée) de l’ Élam devrait nous amener à une

réappréciation des relations entre Ešnunna (tenue pour un des États amorrites majeurs), sous
Zimrî-Lîm, avec son grand voisin élamite.
– Ešnunna a dû, à une époque mal repérable pour nous, passer sous la domination élamite
et servir à Anšan et Suse comme d’un tremplin pour leur politique occidentale. En cela, les
alliances très étroites qu’entretenait, mais on ne sait depuis quand, Ešnunna avec une série de
tribus benjaminites, dont une partie importante étaient établies au royaume de Mari, devaient
grandement favoriser de telles entreprises.
Considérer Ešnunna comme un prolongement de la puissance du Plateau vers la Plaine
ne doit pas paraître comme une chose aberrante. Il faudrait effectivement inclure dans cette
problématique d’autres réalités géopolitiques comme le royaume de Larsa et, éventuellement,
ceux de Diniktum, de Dêr et du Malgi"um. Dans la mesure où nous ne voyons que les relations
de tous ces États (à la puissance non négligeable) envers les autres États amorrites et que
nous constatons pour certains leur complexe fonctionnement administratif, nous les tenons,
en constatant leur autonomie, pour des réalités indépendantes et souveraines. Ils pouvaient ne

16 Cf. ARM XXIII // et FM V :  n. , ainsi que «Vie nomade», qui date selon D. Charpin de ce

moment là.
 jean-marie durand

l’être pas plus par rapport à l’Élam que les États de la Haute-Djéziré ne l’ étaient réellement
par rapport à Mari: la différence est que ces derniers ne sont connus que par leurs relations
internationales d’États subordonnés, pas par leur fonctionnement interne ni dans leurs liens
avec d’autres réalités politiques.
Faire de Larsa le prolongement au sud de la puissance élamite, ce qui serait l’ objet d’ un autre
exposé, permet en tout cas d’expliquer la rivalité entre elle et Babylone : la conquête de Larsa
par Babylone et Mari n’a pas dû être autre chose que la continuation de la politique consistant à
bouter l’Élam hors la Mésopotamie. C’ est sans doute de la même façon qu’ il faudrait expliquer
l’étonnante sommation du roi d’Alep à la lointaine Dêr17.

La suprématie élamite sur les princes de la Mésopotamie doit donc être revue : ces derniers
ont dû avant tout tenir compte de deux faits : la puissance énorme du Plateau iranien avec
ses ressources en hommes et en matières premières ; la réalité de sa suzeraineté sur des États
amorrites considérables (comme au moins Ešnunna et Larsa) laquelle devait, à cette époque,
leur paraître un fait légitime. Par là-même, l’ Élam devait profiter de leurs réseaux d’ alliances
tribales, avant tout benjaminites18.
«Comparaison n’est pas raison», dit-on, mais il existe néanmoins des « permanences his-
toriques». La situation se présenterait pour les relations entre Ešnunna et les Amorrites de la
même façon que pour une «province perse » aux forts accents hellènes comme Sardes avec
les Grecs qui avaient directement affaire avec elle et dont les relations obéissaient à la poli-
tique du Grand Roi, très lointain, jusqu’à tant que ce dernier intervienne directement et que
l’affrontement armé direct se produise entre lui et les États qui maintenaient leur indépendance
à ses frontières occidentales.
Dans une telle perspective, la bataille de Hirîtum a dû sonner le glas de grands projets occi-
dentaux du Plateau iranien et se présente mutatis mutandis comme les prodromes d’ autres
batailles stoppant, à des époques ultérieures et plus à l’ ouest, des entreprises analogues.
L’entreprise avortée du sukkal-mah se présente dès lors comme la première tentative (connue)
de ce qui sera le grand projet de l’empire perse à la fin du Ier millénaire av. JC.
Pour en rester à un niveau plus strictement mésopotamien, il semble qu’ apparaît ici en fait
une grande constante de la trame historique : il a semblé nécessaire à beaucoup d’ impérialismes
proche-orientaux d’avoir le contrôle de la grande transversale Sud-Est / Nord-Ouest qui passe
par la Djéziré et cherche à atteindre le domaine anatolien et, par delà la grande mer ouverte sur
l’extrême ouest. C’est la grande route du Nord dont une partie importante suit les piémonts,
nord ou sud, du Taurus arrosés de pluies pérennes et dont à l’ époque qui nous concerne une
partie avait été longtemps sous contrôle commercial, mais non encore militaire, de la cité
d’Aššur.

Bibliographie

ARM IV = Dossin, G. (): Correspondance de Šamši-Addu et de ses fils, Paris.

ARM XVI/ = Birot, M., Kupper, J.-R. and Rouault, O. (): Répertoire analytique – Partie I: Noms
propres, Paris.
ARM XVIII = Rouault, O. (): Mukannišum: l’administration et l’économie palatiales à Mari, Paris.
ARM XXIII = Bardet, G. et al. (): Archives administratives de Mari, Paris.
ARM XXV = Limet, H. (): Textes administratifs relatifs aux métaux, Paris.

17 Cf. la réédition de ce texte dans LAPO  .

18 Pour ce qui est de Larsa, on constate l’arrivée inopinée d’Atamrum après la prise de Larsa (ARM XXVII )
lequel, en tant que chef du Yamutbal du Nord, devait venir voir ce que devenaient ses contribules du Sud.
 la «suprématie élamite » sur les amorrites 

ARM XXVI/ = Durand, J.-M. (): Archives épistolaires de Mari I/, Paris.
ARM XXVI/ = Charpin, D. (): Archives épistolaires de Mari I/, Paris.
ARM XXVII = Birot, M. (): Correspondance des gouverneurs de Qattunân, Paris.
ARM XXVIII = Kupper, J.-R. (): Lettres royales du temps de Zimri-Lim, ˙˙ Paris.
ARM XXX = Durand, J.-M. (): La nomenclature des habits et des textiles dans les textes de Mari (=
Matériaux pour le dictionnaire de Babylonien de Paris – Tome I), Paris.
ARM XXXI = Guichard, M. (): La vaisselle de luxe des rois de Mari (= Matériaux pour le dictionnaire
de Babylonien de Paris – Tome II), Paris.
Carter, E and Stolper, M.W. (): Elam. Survey of Political History and Archaeology (= Near Eastern
Studies ), Berkely-LA-London.
Charpin, D. (): «Un souverain éphémère en Ida-Maras: Išme-Addu d’Ašnakkum,» MARI : –
. ˙
Charpin, D. and Durand, J.-M. (): «La suzeraineté de l’empereur (Sukkalmah) d’Elam sur la
Mésopotamie et le ‘nationalisme’ amorrite.» In De Meyer, L. and H. Gasche (Eds.), Mésopotamie et
Elam. Actes de la XXXVIe Rencontre Assyriologique Internationale, Gand, – juillet  (= MHEOP
I), Ghent: –.
———. (): «Des volontaires contre l’Élam.» In Sallaberger, W., Konrad, V. and Zgoll, A. (Eds.),
Literatur, Politik und Recht in Mesopotamien: Festschrift für Claus Wilcke (= OBC ), Wiesbaden:
–.
Dossin, G. (): «La route de l’étain,» RA : –.
Durand, J.-M. (): «Notes brèves: [.–] Le nom de l’Elam dans les archives de Mari,» MARI : –
.
———. (): «Fourmis blanches et fourmis noires.» In Vallat, F. (Ed.), Contribution à l’histoire de l’Iran:
mélanges offerts à Jean Perrot, Paris: –.
———. (): «De l’époque amorrite à la Bible: le cas d’Arriyuk.» In Kogan, L. et al. (Eds.), Memoriae
Igor M. Diakonoff (= Babel und Bibel ): –.
FM V = Charpin, D. and Ziegler, N. (): Mari et le proche-orient à l’époque amorrite: essai d’histoire
politique, Paris.
FM IX = Ziegler, N. (): Les musiciens et la musique d’après les archives de Mari, Paris.
Grillot, R. and Glassner, J.-J. (): «Problèmes de succession et cumuls de pouvoirs: une querelle de
famille chez les premiers sukkalmah?,» IrAn : –.
Guichard, M. (): «Résurrection d’un souverain élamite fantôme?,» NABU /.
———. (): «Une nouvelle mention de Šulši-kudur à Mari,» NABU /.
———. (a): «À la recherche de la pierre bleue,» NABU /.
———. (b): «Violation du serment et casuistique à Mari.» In Joannès, F. and Lafont, S. (Eds.),
Jurer et maudire: pratiques politiques et usages juridiques du serment dans le proche-orient ancien (=
Méditerranées –), Paris: –.
Joannès, F. (): «L’étain de l’Élam à Mari.» In De Meyer, L. and H. Gasche (Eds.), Mésopotamie et
Elam. Actes de la XXXVIe Rencontre Assyriologique Internationale, Gand, – juillet  (= MHEOP
I), Ghent: –.
LAPO  = Durand, J.-M. (): Les Documents épistolaires du palais de Mari, tome I, Paris.
LAPO  = Durand, J.-M. (): Les Documents épistolaires du palais de Mari, tome II, Paris.
LAPO  = Durand, J.-M. (): Les Documents épistolaires du palais de Mari, tome III, Paris.
RGTC  = Groneberg, B. (): Die Orts- und Gewässernamen der altbabylonischen Zeit, Wiesbaden.
Vallat, F. (): «L’Élam à l’époque paléo-babylonienne et ses relations avec la Mésopotamie.» In
Durand, J.-M. (Ed.), Mari, Ebla et les Hourrites, dix ans de travaux (= Amurru I), Paris: –.
 «AINSI PARLE L’ EMPEREUR »
À PROPOS DE LA CORRESPONDANCE DES SUKKAL-MAH

Dominique Charpin*

Pourquoi un mariologue, pour parler de Suse et de l’ Elam ? Il y a, bien entendu, des raisons
que je qualifierai de mi-historiographiques et mi-personnelles. On pourrait évoquer le fait que
G. Dossin, avant même d’être appelé par F. Thureau-Dangin en  à le seconder dans le
déchiffrement des tablettes de Mari, avait publié en  à l’ invitation du Père Scheil un volume
de «Textes sumériens et accadiens» de Suse, le tome  des MDP. On pourrait ajouter que
cette association entre Mari et Suse persista, puisque G. Dossin, alors directeur des Archives
Royales de Mari à côté d’André Parrot, donna au regretté Léon De Meyer comme sujet de thèse
«L’accadien des contrats de Suse», qui aboutit à la publication d’ un livre bien connu (De Meyer
). On le voit, ce sont là les contingences de l’ histoire de la recherche. Il y a toutefois une
réalité historique incontestable: les archives de Mari ont permis de voir à quel point Suse et
l’Elam ont tenu une place centrale dans l’ Asie occidentale du début du deuxième millénaire.
Nous avions tenté de le montrer lors de la RAI de Gand en  (Charpin/Durand ) et
J.-M. Durand fait le point sur ce sujet dans ce volume.
Se pose alors une autre question: pourquoi une deuxième contribution à partir des archives
de Mari? En dehors de l’amitié qui me lie depuis longtemps à Michel Tanret et plus récemment à
Katrien De Graef, il existe une raison scientifique. J’ ai entrepris une recherche de longue haleine
sur la correspondance à l’époque paléo-babylonienne, étudiée non pas quant à son contenu,
mais du point de vue du phénomène socio-politique et du genre littéraire qu’ elle constitue
(Charpin en prép.). Ce projet a pris corps à l’ occasion d’ une communication faite à Gand le 
mars , lors de la célébration des vingt ans de la Fondation assyriologique Georges Dossin,
organisée par Denise Homès-Fredericq et Léon De Meyer. C’ est un aspect de ce travail que je
voudrais aujourd’hui développer, qui concerne la correspondance des sukkal-mah.
Je voudrais d’abord présenter le corpus disponible, avant de mettre en relief les usages
particuliers que montre cette correspondance et d’ exposer pour finir une nouvelle hypothèse
concernant un message de l’empereur élamite cité dans une lettre de Mari.

. Un corpus limité

Le corpus de la correspondance des sukkal-mah est limité, pour plusieurs raisons.

.. Les corpus de Suse et de Mari

Il faut d’ abord rappeler que la fouille de Tall-i Maliyān, l’ antique Anšan, n’ a pas livré de tablettes
paléo-babyloniennes. Ajoutons que, même si tel était le cas, cela ne changerait sans doute pas
la situation: dans toutes les archives palatiales connues jusqu’à présent, on n’ a retrouvé que la

* École Pratique des Hautes Études (Sorbonne, Paris). Cette contribution a été rédigée dans le cadre du projet

« ARCHIBAB (Archives babyloniennes, xxe-xviie siècles)», financé pour – par l’ANR (Agence Nationale
de la Recherche) au titre de l’ appel d’ offres « Corpus et outils de la recherche en sciences humaines et sociales». Voir
le site www.archibab.fr.
 dominique charpin

correspondance passive des rois. Même à supposer qu’ on dispose un jour des archives d’ Anšan,
il est fort probable qu’on n’y retrouvera qu’ un nombre limité de lettres écrites par les sukkal-
mah: celles qu’ils écrivirent à des membres de leur famille ou à leurs serviteurs pendant leurs
absences, ou encore des brouillons ou des lettres non envoyées pour une raison ou pour une
autre. À titre de comparaison, dans le palais de Mari, le rapport entre les correspondances
passive et active des rois tient en quelques chiffres révélateurs :

correspondance passive: correspondance active:

roi lettres reçues par – lettres expédiées par – % du total
Yasmah-Addu 483 41 8,5%
Zimrî-Lîm 1625 86 4,7%

N.B. Ces chiffres ne prennent pas en compte les lettres encore inédites.
Rendant compte de la documentation écrite susienne, L. De Meyer avait noté à propos des
lettres: «Cette catégorie de textes est quasi absente dans les MDP » (De Meyer  : ). Il avait
également signalé que plusieurs dizaines de lettres ont été retrouvées dans les couches XII à XV
du chantier A de la ville royale de Suse, lors des fouilles Ghirshman de  à 1. Quarante
d’entre elles ont conservé leur adresse ; une dizaine ont le sukkal-mah comme expéditeur2. La
publication de toute cette correspondance avait été confiée à J. Bottéro ; F. Malbran-Labat et
S. Lackenbacher devraient la mener à son terme3. Il faut actuellement se contenter d’ indications
livrées ici ou là, en particulier dans le CAD, puisque J. Bottéro avait envoyé à Chicago le
manuscrit provisoire de son édition4.
On doit enfin remarquer qu’on n’a retrouvé dans le palais de Mari aucune lettre d’ un sukkal-
mah à un roi de Mari. Cela n’est en soi pas très étonnant : on a également très peu de lettres
de grands rois plus proches, comme Hammu-rabi de Babylone ou Ibâl-pî-El II d’ Ešnunna. Le
problème est de savoir si cette situation résulte du tri effectué par les Babyloniens, ou si elle
reflète la réalité5. Malgré tout, nous pouvons reconstituer un corpus de  lettres : il s’ agit soit de
courrier intercepté ( cas), soit de citations ( cas), soit de copie intégrale ( cas).

.. Des lettres interceptées

Le phénomène de l’interception du courrier en temps de guerre est de mieux en mieux connu.
Un exemple relatif au sukkal de Suse nous est donné par Meptûm6 :

1 Pour ces fouilles et la datation de la couche XII à cheval sur la fin de l’époque paléo-babylonienne et le début

de l’ époque médio-babylonienne, voir Steve, Vallat & Gasche /: –.

2 L. De Meyer parle de « plus de dix lettres» (De Meyer : ). M.J. Steve signale «que trois de ces lettres

(TS XII : , , ), ont été adressées à Attaru-Uktuh par un sukkalmah anonyme» (Steve : ). Il ajoute que
« TS XII:  est adressée par un sukkalmah anonyme à un personnage que l’état du texte ne permet pas d’identifier,
mais la tablette porte l’ empreinte d’ un sukkalmah: Kuk-Našur » (ibid.).
3 Rien n’ est paru jusqu’à présent, à part une note brève qui mentionne seulement «sept missives envoyées par le

sukkal.mah» (Malbran/Lackenbacher ).

4 Les citations sont le plus souvent accompagnées de la remarque «(Susa let., courtesy J. Bottéro)». Une recherche

dans la version électronique du CAD que l’ Oriental Institute met généreusement à la disposition des chercheurs m’a
permis de comptabiliser  citations dans  volumes, parus à partir de : M/ (p. b, b, b); M/ (a);
N/ (b) ; N/ (a, a, a, a) ; P (a, a, a, b, a, a, b); Q (a, b, a, a, b);
R (b, b, a, b) ; S (a, a, a); Š/ (a, b, b, a et b, a, a–b, b, a, a, b,
a, b, b, b, a) ; Š/ (b, a, b, a, b, b); Š/ (a, b); T (a, b, b, a).
5 Pour cette problématique, voir en dernier lieu Charpin a: –.
6 ARM   : ()  tup-pa-tum () ša SUKKAL šu-ši-im ša ELAM.MA ki () ša DUMU ! <É.>DUB !.BA -šu na-šu-
˙ il-qú-nim () [tup-pa-t]im ši-na-<ti> ep-te-e-[ma] () [um-ma a-na-ku-m]a* pí-qa-at ()
ú () ša i-na ge5-er-ri-im
˙
 «ainsi parle l’empereur» à propos de la correspondance des sukkal-mah 

4–8
Quatre tablettes du sukkal de Suse d’Elam, dont son scribe était porteur et dont on s’est emparé
lors de l’expédition, j’ai ouvert ces ! tablettes 9–11en me disant: «Sans doute y a-t-il écrit [à l’intérieur
des table]ttes des nouvelles concernant l’étranger ou le pays lui-même. 12–14[Il me faut absolument
les ouvrir], les écouter et écrire à mon seigneur; en outre, je dois prendre, moi-même, mes
dispositions.» 15–17(En fait) il n’y avait pas de nouvelles dans ces tablettes, concernant l’étranger ou
le pays lui-même. 18–19À présent, j’ai fait porter ces tablettes chez mon seigneur: que mon seigneur
les écoute.
Cette lettre nous apprend d’abord que les lettres du sukkal circulaient, selon la coutume de
l’époque, sous enveloppe. S’il fallait les ouvrir pour accéder au contenu7, c’ est que, selon
l’usage général à l’époque, l’enveloppe ne comportait que le nom du destinataire, ainsi que
le déroulement du sceau de l’expéditeur (Charpin  : –).
De fait, on a retrouvé dans le palais de Mari une lettre du sukkal-mah qui fut, soit interceptée,
soit transmise à Zimrî-Lîm par un roi qui lui resta fidèle. La tablette est malheureusement très
mal conservée, mais l’adresse est intéressante à plus d’ un titre8 :
1
Ainsi (parle) le Sukkal-mah: 2à [tous] les rois du Šubartum et de [ma] frontière, 3dis ceci. 4Aupara-
vant, [vous aviez donné] votre parole à Hammu-rabi. 5–6[À présent], quant à vous, que nos troupes
et les troupes de […] fassent ensemble leur jonction (tout le reste a disparu).
On se contentera à présent de relever que l’ empereur ne jugea pas nécessaire de personnaliser
ses envois. Dans les autres cas de «circulaires » connus, même si le texte est identique, le nom
du destinataire est chaque fois indiqué9. On a ici l’ impression qu’ au moment où la chancellerie
élamite envoya ces lettres, on ne savait pas exactement quels rois en seraient les destinataires : le
soin de les remettre aux souverains intéressés était apparemment laissé aux messagers porteurs
de ces lettres. Cela s’explique peut-être par les troubles de cette période, au cours de laquelle de
nombreux rois trouvèrent la mort10.

[te4-em A.ŠÀ ù ma-tim]-ma () [i-na li-ib-bi tup-pa-t]im* ša-te4-er () [šum-ma la e-pe-et-tu]-ma i-še-mu-ú-ma
˙ a-[n]a be-lí-ia a-ša-ap-p[a-r]u () ù a-na-ku
() ˙ te -mi a-s[a-ba]-tu
˙ () mi-im-[m]a te4-mu-um i-na li-ib-bi-ši-na
4
˙
() ú-ul i-ba-aš-ši aš-šum te4-em () A.ŠÀ ù ma-tim-ma ˙ () a-nu-um-ma tup-pa-tim˙ ši-na-ti a-na be-lí-ia ()
˙
uš-ta-bi-lam be-lí li-iš-me-ši-na-ti. ˙
Je propose ici une nouvelle interprétation de cette lettre, qui s’écarte de la traduction de J.-M. Durand, LAPO
  [J.-M. Durand me signale qu’ il n’ est pas nécessaire de suppléer <É> dans l’expression DUMU É.DUB.BA
car il a retrouvé l’ expression DUMU tup-pí-im dans l’inédit M.: ]. La difficulté concerne l’identification de
˙
l’ expéditeur. En principe, SUKKAL šu-ši-im devrait désigner Kudušuluš (cf. ARM  , qui distingue še-ep-la-ar-
pa-ak SUKKAL e-lam-<<ti>>-timki et ku-du-šu-lu-úš SUKKAL šu-ši-imki); l’ajout de ša ELAM.MA ki pourrait être
dû à la volonté d’ écarter une possible confusion avec le ministre (SUKKAL) du roi de Susâ en Haute Mésopotamie
(dont le nom est parfois écrit šu-ša-aki). Quoi qu’il en soit, il me semble exclu que ce texte annonce la mort de
Kudušuluš, comme le croit W. Heimpel (Heimpel : ).
7 « Ouvrir une tablette » (tuppam petûm) est une façon abrégée habituelle pour dire «ouvrir une enveloppe pour

accéder à la tablette intérieure ˙ ».

8 ARM   : () [um-m]a SUKKAL.MAH-[ma] () a-na šar-ra-ni ša šu-bar-tim ù pa-ti(DI)-[ia ka-li-šu-nu]
˙
() [q]í-bí-[ma] () [i-na p]a-ni-tim pí-i-ku-nu a-na ha-am-mu-ra-b[i …] () [i-na-an-na a]t-tu-nu-m[a] sa-bu-ni
ù sa-b[u]-u[m ša […] () [a-hu-um qa-d]u-um a-hi-i[m] li-in-[n]e-em-d[u-nim] () [x x x x -b]i il-la-am ša ˙ x […]
(la˙ suite n’ est pas conservée).
9 On peut mentionner celle que Zimrî-Lîm adressa à plusieurs rois au début de son règne: on en connaît un

exemplaire destiné à Tiš-Ulme (TH ., publié dans Birot : – = LAPO  ) et un autre adressé
conjointement à Abi-Samar et Ikšud-lâ-šêmêšu (ARM  ; cf. ARM /:  et Guichard :  n. ). Les
deux lettres ne sont pas exactement duplicat l’ une de l’autre, mais sont plus que de simples parallèles. Un bel exemple
est connu indirectement par une lettre d’ Ašmad (inédit A., cité dans Guichard : –). C’est aussi le
cas des lettres de Samiya (voir à ce sujet Charpin/Ziegler : ).
10 Charpin/Ziegler  : –.
 dominique charpin

.. Des citations

J’ai pu relever les citations de  lettres du sukkal dans l’ ensemble de la correspondance
actuellement publiée. Les différentes situations rencontrées sont les suivantes :
– citation d’une lettre interceptée ;
– citation d’une copie reçue;
– citation de lettres que leur destinataire a fait suivre ;
– résumé d’une lettre dont on a entendu lecture.

... Citation d’une lettre du sukkal d’ Elam à Atamrum interceptée

Une lettre de Nûr-Sîn au roi Zimrî-Lîm rapporte au roi de Mari qu’ un message de l’ empereur
élamite destiné à Atamrum a été intercepté11 :
1–4
Dis à mon Seigneur: ainsi (parle) Nûr-Sîn, ton [serviteur]. 5–7Avant le départ d’Isi-Qatar, [servi-
˙
teur] de mon seigneur, [j’ai entendu dire] dans mon entourage: « 7–10On a capturé quatorze ˙ âhizum
11–12
de [l’empereur] d’Elam [qui] portaient [un message] et on les a emprisonnés». [Voi]ci ce qui
[était é]crit à Atamrum:
« 13Fais la paix avec le sire de Razamâ; 14donne-lui [de l’argent et de] l’or et 15libère [ses serviteurs].
16
Je veux te donner un [éd]it 17et te l’envoyer par écrit, pour qu’ainsi, moi-même, 18je puisse rentrer
17
dans mon pays!»
18–19
Voilà ce que j’ai entendu dire. [Présentement, j’ai écrit] mon [rapport à mon seigneur].
L’affaire date de la période où les Élamites avaient envahi la Mésopotamie. L’ empereur se
trouvait dans la région d’Ešnunna, mais souhaitait rentrer. Atamrum faisait alors le siège de
Razamâ, où le roi Šarraya résistait12. Manifestement, une fois les porteurs de la tablette arrêtés13,
Nûr-Sîn prit sur lui d’ouvrir l’enveloppe pour prendre connaissance du contenu de la lettre,
qu’il recopia à l’intention du roi. Contrairement à Meptûm, il ne fit pas suivre la tablette à
Zimrî-Lîm.

... Citation d’une copie reçue: l’ultimatum du sukkal d’ Elam à Hammu-rabi de Babylone
Lorque le général Yasîm-Dagan se trouvait à Ešnunna à la tête de troupes mariotes, il écrivit
à Zimrî-Lîm au moment où les Élamites, après avoir conquis Ešnunna, assiégeaient Manki-
sum14 :

11 FM   : () [a-na] be-lí-[ia] () [qí]-bí-ma () [um]-ma nu-úr-dEN.[ZU] () [ÌR]-ka-a-ma () [i-na] pa-

an wa-sé-e i-sí-qa-tar () [ÌR] be-lí-i[a] i-na a-hi-ti-ia () [ke-em eš-me u]m-m[a]-a-[mi]   LÚ a-hi-zi () [ša
LÚ.SUKKAL˙ ˙L]Ú ˙[E]LAM.MAki () [ša ši-pí-ir-tam] ub-ba-lu is-ba-tu-/nim-ma (T.) [ú-še-r]i-bu-šu-nu-ti ()
[ke-e-e]m a-na a-tam-ri-im (R.) [iš-ša]-pí-ir um-ma-a-mi ()˙ [LÚ ra]-za-ma-aki sú-ul-li-im () [KÙ.BABBAR
KÙ].GI i-di-in-šum-ma () [ÌR.MEŠ-šu p]u-ut-ra-am () [ši-ip]-tà-am lu-ud-di-na-kum-ma () [lu-uš-p]u-ur ù
˙
a-na-ku a-na ma-ti-ia () [lu-tu]-ra-[a]m an-ni-tam eš-me () ˙[i-na-an-na a-nu-um-ma te4-m]i () [a-na be-
lí-ia aš-pu-ra-am]. Je ne suis pas sûr de la restitution du début de la l. , que je conserve ˙faute d’une meilleure
idée.
12 Pour le contexte historique, voir Charpin/Ziegler : .
13 Pour le sens du mot âhizum, voir Durand : –, qui traduit «colporteur».
14 Lettre inédite de Yasîm-Dagan A.: (R. 0) … LÚ.SUKKAL a-na ha-mu-ra-bi (0) ki-a-am iš-pu-ur um-

ma-a-mi a-tam-<rum> na-si-iq a-di wa-aš-ba-ku (0) a-wa-ti-ka gu-mu-ur a-la-nu ša ÈŠ.NUN.NAki ša tu-ka-al-lu
(0) ú-ul ú-yu-ut-tu-un wa-aš-ši-ir-šu-nu-ti ù ki-ša-ad-ka (0) a-na ni-ri-ia šu-ri-ib ú-la-šu-ma ma-a-at-ka (0)
ah-ta-na-bi-it iš-tu ma-an-ki-siki sa-bu-um i-sa-ba-tam (0) aš-ra-nu-um-ma i-ib-bi-ra-am ù a-na pa-an um-[ma-
˙ ma-ti-ka ˙ah-ha-ba-at (0) an-ni-tam LÚ.SUKKAL a-na ha-am-mu-ra-bi iš-pu-
na-ti-ia] (T.0) e-eb-bi-ir-ma a-na
r[a-am] (0) me-he-er tup-pí-im ú-ša-bi-lu-šum a-na a-tam-ri-im (0) LÚ.MEŠ ša-pí-ru-ut sa-bi-im ù a-na se-ri-ia
˙
(0) il-li-kam a-na nu-ku-ur-ti L[Ú] KÁ.DINGIR.RAki (0) [ma]-di-iš sa-ri-i[m …] … Une˙partie de cette citation˙
se trouve déjà dans Charpin/Durand  :  n. . Voir depuis Charpin ˙ : .
 «ainsi parle l’empereur» à propos de la correspondance des sukkal-mah 

190–200
L’empereur d’Elam a écrit à Hammu-rabi en ces termes: «C’est Atamrum qui a été choisi.
200–230
Achève ce que tu dois faire tant que je réside ici. Les villes d’Ešnunna que tu détiens ne sont-
elles pas miennes? Évacue-les et soumets-toi à mon joug!15 230–260Sinon, je pillerai ton pays de fond
en comble. L’armée fera route depuis Mankisum, elle franchira le fleuve à cet endroit. À la tête
de mes armées, je franchirai le fleuve et j’envahirai ton pays.» 270Voilà ce que l’empereur a écrit à
Hammu-rabi. 280–310Le double de la tablette qu’il lui a fait porter est parvenu à Atamrum, aux chefs
de la troupe et à moi-même; il brûle d’ardeur à faire la guerre au sire de Babylone.
L’original de cette lettre de l’empereur d’ Elam doit se trouver dans les archives du palais de
Hammu-rabi; on en connaît le contenu grâce à une lettre envoyée à Mari, où un général mariote
a reproduit le texte de la copie qui lui a été envoyée … On observera que Yasîm-Dagan n’ indique
pas qui lui fit porter le double de la lettre de l’ empereur.

... Citation de lettres que leur destinataire a fait suivre : les ultimatums du sukkal à Hammu-
rabi de Babylone et Rîm-Sîn de Larsa après la prise de Larsa
Yarîm-Addu était le représentant de Zimrî-Lîm auprès de Hammu-rabi. Il cita des extraits de
lettres du sukkal à Hammu-rabi de Babylone et Rîm-Sîn de Larsa16 :
1
Dis à mon seigneur: ainsi (parle) ton serviteur Yarîm-Addu.
3–4
Le sukkal d’Elam a écrit ainsi à Hammu-rabi: « 5–7Je me dispose à partir contre Larsa. Mobilise
ta troupe d’élite, la troupe du génie et tes sujets que j’ai vus à Ešnunna, afin qu’ils soient prêts pour
mon arrivée. 9–11Si un seul homme de la troupe que j’ai vue n’est pas employé, c’est à toi que je m’en
prendrai.» Voilà ce que le sukkal d’Elam a écrit à Hammu-rabi.
12–15
Celui-ci lui a répondu: «Comme tu me l’as écrit, ma troupe est prête et disponible pour ton
attaque. Le jour où tu attaqueras, ma troupe partira te rejoindre.» Voilà ce qu’il lui a répondu.
15–17
Et de la même façon que le sukkal d’Elam avait écrit à Hammu-rabi, il a écrit à Rîm-Sîn en
ces termes: « 18–23Je me dispose à partir contre Babylone. Mobilise ta troupe d’élite, la troupe du
génie et tes sujets de confiance, afin qu’ils soient prêts. Si un seul homme de la troupe dont je ne
cesse d’entendre parler n’est pas employé, c’est à toi que je m’en prendrai.» 24Voilà ce que le sukkal
d’Elam a écrit à Rîm-Sîn.
25–27
La tablette que le sukkal d’Elam a fait porter à Rîm-Sîn, cette tablette, Rîm-Sîn l’a fait porter à
Hammu-rabi; 28–30et Hammu-rabi de même a fait porter à Rîm-Sîn la tablette que le sukkal d’Elam
lui avait fait porter.
Le dernier point est sans doute précisé par Yarîm-Addu pour expliquer comment, étant à
Babylone, il pouvait connaître, non seulement la lettre de l’ empereur à Hammu-rabi, mais aussi
celle que l’empereur envoya au roi de Larsa.

15 Pour cette expression, voir Anbar .

16 ARM /  : () a-na be-lí-ia qí-bí-ma () um-ma ia-ri-im-dIM ÌR-ka -a-ma () SUKKAL ELAM.MA-
tim a-na se-er h[a-am-mu-ra-b]i () ki-a-am iš-pu-ra-am [um-ma šu-ú-ma] () a-na la-ar-saki pa-nam ša-ak-na-ku
˙ bé-e]h-ra-am () sa-ab tu-up-ši-ik-ka-nim ù ÌRdi.MEŠ-ka () ša i-na ÈŠ.NUN.NAki a-mu-ru ki-i-in-ma
sa -[ba-ka
˙a-na pa-ni-ia () re-ša-am li-ki-il-lu
˙ i-na sa-bi-im ša a-mu-ru ()  LÚ ip-pa-ra-ku-ma a-na se-ri-ka-ma () pa-
nam ú-sa-ah-<ha>-ra-am an-n[i-t]am SUKKAL ˙ ELAM.MA-tim () a-na se-er ha-[am-m]u-ra-bi ˙ iš-pu-ra-am-ma
() ki-a-am i-pu-u[l-š]u um-ma šu-ú-ma ki-ma ša ta-aš-pu-ra-am () sa-bi ˙ sa-mi-i[d-m]a re-eš te-bé-e-ka ú-ka-al
() u4-um te-te-eb-b[i-a]m sa-bi a-na pa-ni-ka it-ta-al-la-kam () an-ni-tam˙ ˙ [i]-p[u]-ul-šu ù ki-ma ša SUKKAL
ELAM.MA-tim () a-na s[e-er] ˙ ha-a[m-m]u-ra-bi iš-pu-ra-am () a-n[a se]-er ri-im-dEN.ZU ki-a-am iš-pu-ur
˙
um-ma šu-ma () a-na KÁ.DINGIR.RA ˙
ki p[a]-nam ša-ak-na-ku (T.) sa-ba-ka bé-eh-ra-am sa-ab tu-up-ši-ik-
ka-nim () ù ÌRdi.MEŠ-ka ták-lu-tim ki-i-in-ma (R.) re-ša-am li-ki-il-[lu] ˙ () i-na sa-bi-ka ša˙ eš-te-ne-mu-[ú 
LÚ ip]-pa-ra-ku-ma () a-na se-ri-ka-ma pa-nam ú- sa-ah-ha -ra-am () an-ni-tam ˙SUKKAL ELAM.MA-tim a-
˙
na se-er ri-im-dEN.ZU () iš-pu-ur-ma tup-pa-am ša SUKKAL ELAM.MA-tim () a-na ri-im-dEN.ZU ú-ša-bi-lu
˙ d ˙
tup-pa-am ša-a-tu ri-im- EN.ZU () a-na se-er ha-am-mu-ra-bi ú-ša-bi-la-aš-šu () ù ha-am-mu-ra-bi qa-tam-ma
˙tup-pa-am ša SUKKAL ELAM.MA-tim () ˙ ú-ša-bi-la-aš-šum a-na se-er ri-im-dEN.ZU () uš-ta-bi-il-šu.
˙ ˙
 dominique charpin

Comment prendre en compte de telles citations ? Chaque fois qu’ un contrôle est possible (et
le cas se présente plus d’une fois dans les archives de Mari), on s’ aperçoit que les citations ne
sont pas faites mot à mot, mais restent très près de l’ original. Mettre en doute radicalement
l’authenticité de ces citations relèverait, à mes yeux, de l’ hyper-critique17.

.. Une copie intégrale

Aqba-ahum était le représentant de Zimrî-Lîm auprès du roi de Kurdâ Hammu-rabi. Il recopia
l’intégralité d’une lettre du sukkal à ce roi18 :
1–3
Dis à mon seigneur: ainsi (parle) Aqba-Ahum, ton serviteur.
4–7
Voilà que j’ai fait porter à mon seigneur dans ma présente tablette le double d’une tablette de
l’empereur des Élamites, qu’il a envoyée à Hammu-rabi.

« 8Ainsi (parle) l’empereur 9à Hammu-rabi. 10–11Atamrum, un serviteur à moi, t’a pris en vassalité.
12–14
Or, je ne cesse d’entendre dire que tu ne cesses de faire porter des tablettes de toi à Babylone
et à Mari. 15–17Ne recommence pas à faire porter des tablettes de toi à Babylone et à Mari! 18–21Si à
nouveau tu fais porter des tablettes de toi à Babylone et à Mari, je soufflerai en tempête sur toi.»
22–25
Tel est le message que l’empereur des Élamites à envoyé à Hammu-rabi. J’ai moi-même entendu
cette tablette.
Du point de vue formel, cette lettre a des parallèles : la copie est séparée de l’ introduction de la
lettre par une ligne de séparation19. Il ne s’ agit donc pas d’ une simple citation, qui donnerait un
extrait de la lettre; néanmoins la copie est faite de mémoire.

. Des usages particuliers

Les lettres envoyées par les sukkal-mah offrent deux caractéristiques exceptionnelles au sein du
corpus épistolaire paléo-babylonien: elles placent l’ expéditeur en tête et donnent son titre sans
le nommer.

17 Pour plus de détails sur cette question, voir Charpin en préparation. Les archives de Shamsara donnent un

exemple supplémentaire de citation d’ une lettre d’un empereur élamite, dans une missive que Šepratu écrivit à
Kuwari, le chef turukkéen de Šušarra. ShA  : () … š[a-ni]-tam () Išu-ru-uh-tu-uh LUGAL ša NIM.MA-
tim () a-na Ita-bi-lu iš-pu-ra-am () um-ma šu-ú-ma a-na mì-nim ma-at i-ta-pá-al-hi-im () ma-ru ši-ip-ri-im
a-na se-ri-ia () la i-ša-pa-ra-am « Autre chose. Šuruhtuh, le roi d’Elam, a écrit à Tabilu en ces termes: “Pourquoi
le pays˙ d’ Itapalhum ne m’ envoie-t-il pas de messagers?”» La citation n’est manifestement pas complète, de sorte
que la formule d’ adresse n’ est pas reproduite. Mais de manière significative, l’extrait cité ne consiste qu’en un
reproche, qu’ on peut ainsi paraphraser : pourquoi l’empereur élamite est-il contraint de prendre l’initiative dans
l’ établissement de relations diplomatiques avec le pays d’Itapalhum? Autrement dit, pour le souverain élamite, sa
supériorité est telle que ce devrait être le roi d’Itapalhum qui lui envoie des messagers le premier. Le contexte permet
de comprendre ce reproche : traditionnellement, une alliance liait les Elamites et les Gutis. Or, cette fois, il est question
d’ un rapprochement entre Elamites et Turukkéens, ennemis jurés des Gutis.
18 A., publié dans Durand  : – (= LAPO  ): () a-na be-lí-ia () qí-bí-ma () um-ma aq-ba-a-hu-

um ÌR-ka-a-ma () a-nu-um-ma me-hi-ir tup-pí-im () ša LÚ.SUKKAL ELAM.MA.MEŠ X () ša a-na ha-mu-ra-bi
iš-pu-ru () i-na tup-pí-ia an-ni-im a-na˙se-er be-lí-ia/ ú-ša-bi-lam (ligne) () um-ma LÚ.SUKKAL-ma () a-na
ha-mu-ra-bi ()˙Ia-tam-{AŠ}-rum ÌR-di () ˙ a-na ma-ru-tim il-qí-ka () i-na-an-na eš15-te-né-me-em () tup-
pa-ti-ka a-na KÁ.DINGIR.RAki () ù a-na ma-riki tu-uš-ta-na-ba-al () la ta-ta-ar-ma tup-pa-ti-ka () a-na ˙
ki ki ˙
KÁ.DINGIR.RA ù a-na ma-ri () la tu-ša-ba-al () ta-ta-ar tup-pa-ti-ka () a-na KÁ.DINGIR.RA ù a-na ki
˙
ma-riki () tu-ša-ba-al-ma () a-ša-i-ra-kum () ša-pa-ra-am an-né-em () LÚ.SUKKAL ELAM.MA.MEŠ ()
a-na ha-mu-ra-bi iš-pu-ra-am () [t]up-pa-am še-tu () a-na-ku eš-me.
19 Voir de même Charpin  :  ˙ fig. ..
 «ainsi parle l’empereur» à propos de la correspondance des sukkal-mah 

.. L’ expéditeur en tête

La norme paléo-babylonienne consistait à placer le nom du destinataire avant celui de l’ expé-
diteur, quel que soit leur rang respectif. En effet, la formule initiale d’ une lettre conserve
le souvenir de l’origine orale de la transmission du message. Le début n’ est pas adressé au
destinataire, mais à celui qui doit lui communiquer les nouvelles20 :
À X, parle.
C’est ensuite seulement que le messager prend la parole :
Ainsi (parle) Y, (ton …).
Cela explique le recours à la deuxième personne, dans les lettres adressées à des rois, pour
qualifier l’expéditeur («ton serviteur» waradka-ma) : c’ est le messager, non l’ expéditeur, qui
l’emploie21. C’est après seulement que débute le discours de l’ expéditeur au destinataire.
L’ordre d’énonciation: destinataire d’ abord, expéditeur ensuite, ne connaît dans les archives
de Mari qu’une seule exception: celle de l’ empereur élamite. On possède en effet deux exemples
où le sukkal-mah écrivit en plaçant son nom en tête. Il s’ agit de la circulaire aux rois du
Šubartum22 :
Ainsi (parle) le sukkal-mah: dis ceci à [tous] les rois du Šubartum et de ma frontière.
Le deuxième exemple figure dans la copie d’ une lettre du sukkal à Hammu-rabi de Kurdâ incluse
dans une lettre de Aqba-ahum23 :
Ainsi (parle) le sukkal à Hammu-rabi.
Cette manière de faire correspond à l’ usage normal de la chancellerie élamite ; c’ est ce que
confirment les inédits découverts à Suse dans la couche XII du chantier A de la « ville royale ».
Comme l’a indiqué L. De Meyer ( : ) : « A une exception près, le régent suprême reste
anonyme: son titre seul figure toujours à la première ligne24. La disposition suivante a pu être
observée dans plus de dix lettres: um-ma sukkal-mah-ma / a-na NP wardi(di)ia qí-bí-ma /
˘ ». »
«ainsi parle le sukkalmah. A NP, mon serviteur, dis (ceci)
˘
Ce serait toutefois une erreur de croire que cet usage visait exclusivement à souligner
l’omnipotence des empereurs. Il semble en effet que les règles épistolaires du monde élamite
étaient analogues à celles en vigueur dans le monde des marchands paléo-assyriens : on y avait
coutume de nommer la personne la plus importante en tête25. En effet, les lettres publiées par
G. Dossin dans MDP  montrent les deux formules :

20 Ana X qibi-ma umma Y (…)-ma.

21 Normalement, on s’ adresse toujours à un souverain à la troisième personne. Les exceptions se trouvent dans
des lettres expédiées par des proches, comme des membres de sa famille.
22 ARM   : () [um-m]a SUKKAL.MAH-[ma] () a-na šar-ra-ni ša šu-bar-tim ù pa-ti(DI)-[ia ka-li-šu-nu]

() [q]í-bí-[ma]. ˙
23 A. : () um-ma LÚ.SUKKAL-ma () a-na ha-mu-ra-bi (Durand : ).
24 Deux exemples ont été cités par le CAD :

– umma SUKKAL.MAH-ma A XII/ :  (CAD S: a);

– umma SUKKAL.MAH-ma ˘ ana PN u šı̄būti ša PN2 qí-bí-ma A XI :  (CAD Q: b).
˘
25 On corrigera donc l’ appréciation de C. Michel à propos de la correspondance paléo-assyrienne: «Les en-têtes

de lettres correspondent aux formules épistolaires classiques paléo-babyloniennes: «Dis à NP1 : ainsi (parle) NP2 »
ou encore « Ainsi (parle) NP1 : dis à NP2 » : le personnage le plus important apparaît systématiquement en premier»
(Michel  : –). Dans les « formules épistolaires classiques paléo-babyloniennes», le destinataire est en tête,
quel que soit son rang par rapport à l’ expéditeur. On peut trouver de rares exceptions où l’expéditeur figure en tête,
comme AS  . Il ne s’ agit pas d’ un problème hiérarchique, mais du fait que la tablette comporte deux lettres
du même expéditeur (ce que l’ éditeur n’ a pas compris): la première est adressée à sa mère (l. –), la seconde au
SUKKAL.MAH (l. –) (il s’ agit ici du ministre du roi d’Ešnunna).
 dominique charpin

– type  (inférieur à supérieur, ou entre égaux) : ana NP1 qibîma umma NP2-ma : MDP
 , , ,  et . Dans ces deux dernières lettres, la position inférieure de
l’expéditeur est explicitée par l’ ajout de « ton fils » (ma-ru-ka-a-ma). On note également
l’égalité («ton frère») dans la lettre TS.B./nº, en lisant26 : () a-na […]-hu-tu () qí-[bí]-
ma () [u]m-ma ia ?- x -a a-hu-ka-[a]-ma. On ajoutera Vallat  :  (S)27.
– type  (supérieur à inférieur): umma NP1-ma ana NP2 qibîma : MDP  28 et .
Cet usage se retrouve aussi dans les lettres du chantier A XII29.
Il semble cependant que le sukkal de Suse suivait l’ usage babylonien, en ne se mettant pas
en tête. L. De Meyer a signalé qu’une dizaine de lettres « montre que le maire de Suse s’ adresse
d’abord au citoyen, qu’il s’individualise ensuite par son nom et qu’ il introduit très souvent, en
troisième lieu, la formule: é.gal ša-lim « le palais va bien » avant de passer au vrai contenu de sa
lettre; p.ex. TS.XV.: a-na Ku-uk zu-uh-zu wardu(di)-ia qí-[bí-ma] / um-ma Te-em-ti hal-ki /
é.gal ša-lim» (De Meyer : ). ˘ ˘

.. Pas de nom, mais un titre

Pour désigner l’expéditeur d’une lettre, les usages des chancelleries amorrites étaient variés. Il
faut distinguer selon que la lettre était adressée à un « fonctionnaire » ou à un autre roi.
Dans le cas du courrier «interne», on trouve trois formules30 :
a) umma NR-ma : c’est le cas des rois de Babylone, parfois de ceux de Larsa ;
b) umma NR bêlka-ma: la formule est attestée pour Rîm-Sîn de Larsa31 ;
c) umma bêlka-ma : la formule se rencontre à Ešnunna, à Ekallâtum/Šubat-Enlil (Samsî-
Addu), à Mari (sous Yahdun-Lîm, Yasmah-Addu et Zimrî-Lîm).
Pour la correspondance «internationale », on trouve toujours : umma NR, suivi simplement de
l’enclitique -ma ou d’un qualificatif: abuka-ma, ahuka-ma, waradka-ma32. De ce point de vue,
les usages élamites sont doublement atypiques. En effet, ils ne varient pas selon que l’ empereur
s’adresse à un de ses sujets ou à un autre roi : dans tous les cas, le sukkal-mah ne se nommait
pas, mais se désignait simplement par son titre33. Je rappelle les trois exemples cités plus haut
concernant les lettres adressées à d’autres rois :
– um-ma LÚ.SUKKAL-ma (A., citation dans une lettre de Aqba-ahum à Zimrî-Lîm) ;
– um-ma SUKKAL e-la-am-tim-ma (FM  , citation dans une lettre de Halu-rabi à Zimrî-
Lîm)34 ;
– umma SUKKAL.MAH-ma (ARM  , document original).

26 Avec la copie de De Graef  : . La lecture a-bu ?-ka-ma (De Graef : ) ne permettrait pas de

comprendre l’ ordre destinataire / expéditeur.

27 Ce petit fragment de lettre commence par: () a-na -p[u ?-… qí-bí-ma] () [um]-ma d[…-ma].
28 Ce texte comporte deux expéditeurs.
29 C’ est ce qui ressort implicitement de la note de F. Malbran et S. Lackenbacher: «Sur les autres lettres qui ne

comportent pas de bénédiction, neuf sont d’ un inférieur à un supérieur [i.e. du type  ci-dessus], trois d’un supérieur
à un inférieur [= type ] ; sur les vingt-et-une avec bénédiction, seize sont d’un inférieur à un supérieur et cinq d’un
supérieur à un inférieur » (Malbran/Lackenbacher ).
30 Pour plus de détails, voir Charpin . Les formules b) et c) peuvent comporter un autre possessif que la

deuxième personne masculin singulier (comme -ki ou -kunu).

31 On note seulement deux exceptions, où l’expéditeur est indiqué par um-ma šar-ru-um-ma (Veldhuis :

).
32 Pour l’ emploi de abum, ahum ou wardum dans ce contexte, voir Lafont : –, à compléter par

Guichard  : –.

33 Une exception a été signalée par L. De Meyer: «TS.XV.: um-ma Ku-te-er dNa-ah-hu-un-di-ma / a-na (I)Li-

li-ri-ša wardi(di)-ia / qí-bí-ma» (De Meyer :  n. ). ˘ ˘

34 Voir la discussion de ce texte plus bas au §.
 «ainsi parle l’empereur» à propos de la correspondance des sukkal-mah 

Il y a toutes chances pour que la formule authentique soit la dernière, la seule d’ ailleurs qui
soit attestée sur un original: c’est elle en effet qui correspond à ce qu’ on trouve sur les lettres
du sukkal-mah découvertes à Suse. Les indications de M.-J. Steve permettent de montrer que
l’anonymat de la formule d’adresse était compensée par l’ empreinte du sceau au nom du sukkal-
mah, qui figurait sur l’enveloppe35.

.. Portée politique de ces usages

On pourrait s’interroger sur la pertinence de cette étude et surtout sur sa portée. Cependant,
certains textes montrent explicitement l’ importance que la rédaction d’ une adresse de lettre
revêtait dans le contexte politique de l’époque. Je me bornerai à citer ce texte36 :
40–50
Et lorsque mon seigneur leur écrira, qu’il ne leur écrive pas en tant que père ou que frère, 60–70mais
qu’il leur écrive (seulement) cela: «ainsi (parle) Zimrî-Lîm.» 80–90De sorte que par la suite, lorsqu’ils
se seront réconciliés avec mon seigneur, 100–110d’eux-mêmes ils diront: «Oui, mon seigneur», en face
de mon seigneur.
Les découvertes susiennes montrent des usages qui varient dans les lettres rédigées en akkadien
dans le monde élamite; sans doute faudra-t-il prendre en compte la chronologie pour les
expliquer. Du moins une chose apparaît-elle clairement : les lettres du sukkal-mah, à l’ époque
de Zimrî-Lîm, se distinguaient absolument du reste du courrier échangé entre les rois de
cette époque. L’empereur élamite ne se nommait pas et ne plaçait le nom du destinataire
qu’en deuxième position. Les rois auxquels il s’ adressaient ne pouvaient manquer de sentir
la signification politique de cette façon de faire.

. Une message de l’ empereur élamite à un roi syrien

La dernière partie de cette communication sera dévolue à un message du sukkal reproduit dans
FM  37. Il s’agit d’une lettre écrite par Hâlû-rabi, qui, passée l’ introduction, ne contient que
la citation d’une lettre de l’empereur d’ Elam38.
1–2
Dis à mon seigneur: ainsi (parle) Hâlû-rabi, ton serviteur.
3–6
Des messagers élamites sont arrivés ici. Lorsqu’ils ont délivré leur message, nous étions présents
à leur exposé. Voici en quels termes ils ont délivré l’essentiel de leur message:
« 7–8Ainsi (parle) l’empereur d’Elam. Voici ce que tu m’as écrit: ‘8–13Tout ce que … de la bouche de
mes messagers mêmes … tes serviteurs … un autre …» (importante lacune)
10–30
… afin qu’il soit placé par devant Addu d’Alep, je ne lui ai rien donné. 40–50A présent, si tu me
l’écris, je veux bien lui donner un arc. 60–80Tout ce qu’un roi m’écrira pour te dénoncer, je te l’écrirai

35 Steve  : . Il en allait de même pour les rois d’Ešnunna, d’Ekallâtum ou de Mari (ci-dessus, cas c).
36 Inédit A. (Ašmad à Zimrî-Lîm) : (0) ù i-nu-ma be-lí a-na se-ri-š[u-n]u i-ša-pa-ru (0) a-bu-tam ù a-hu-
tam la-a i-ša-ap-pa-ar-šu-nu-ši-im (0) um-ma zi-im-ri-li-im-ma (0) ˙an-ni-tam be-lí li-iš-pu-ur-šu-nu-ši-im-ma (0)
wa-ar-ka-nu i-nu-ú-ma (0) it-ti be-lí-ia it-°i-bu (0) šu-ú-nu-ú-ma an-na be-lí (0) a-na pa-ni be-lí-ia i-ta-su-ú.
37 FM   : () a-na be-lí-ia [qí-bí-ma] ˙ () um-ma ha-lu-ra-bi [ÌR-ka-a-m]a () DUMU.MEŠ ši-ip-ri LÚ.EL-
[AM.MA ik-š]u-du-nim-ma () i-nu-ma te4-em-šu-nu id-di-nu () i-na te4-mi-šu-nu ni-iz-zi-iz () re-eš15 te4-mi-šu-
nu ki-a-am id-di-nu () um-ma SUKKAL ˙ e-la-am-tim-ma ki-a-am () ˙[ta]-aš-pu-ra-am um-ma at-ta-ma˙ mi-im -
?
ma () [ o o o] a -[n]a pi4 DUMU.MEŠ ši-ip-ri-ia-/ma () [ o o o o] x-ma ÌR.MEŠ-ka () [ o o o o o]-uB-Bu-tu ()
[ o o o o o-t]i-ik ?-šu ša-nu-um () [ o o o o o o ] x-x-ka (lacune de ca.  lignes) (R.0) […………… ma-h]a- ar (0)
[d]IM ša ha-la-abki li-iš-ša-ki-in (0) [mi]-im-ma ú-ul ad-di-in-šum (0) [i-na-an]-na šum-ma ta-ša-ap-pa-ra-am gišqa-
aš-tam (0) [lu]- ud -di-in-šum (0) mi-im-ma LUGAL ša ka-ar-sí-ka (0) a -na se-ri-ia i-ša-ap-pa-ra-am (0) ma-
ah-ri-tam-ma a-ša-ap-pa-ra-ak-kum (0) ak-ki-ma ge-er-ra-am la˙ pé-te-em (0) te-ep-te-ma
˙ i-na pa-ni DUMU.MEŠ
ši-ip-ri (0) ša LUGAL.MEŠ ka-li-šu-nu (T.0) [DUMU].MEŠ ši-ip-ri-ka a-na se-ri-ia (0) [t]a-aš-pu-ra-am.
˙
38 Une fois achevée sa citation, Halû-rabi aurait eu la place, sur la tranche inférieure et sur la tranche latérale, pour
ajouter un commentaire ; il ne le jugea pas nécessaire.
 dominique charpin

aussitôt; 90–130car tu as ouvert une route qui ne l’était pas, et tu as envoyé chez moi tes messagers,
avant les messagers (envoyés) à tous les (autres) rois!»
Le début de la lettre décrit une situation classique : les messagers élamites sont reçus lors
d’une audience, à laquelle assistent les envoyés du roi de Mari, présents dans cette capitale en
même temps qu’eux39. On remarque que Hâlû-rabi ne prétend pas faire une citation complète
du message donné par les messagers élamites40 : il reproduit le rêš têmim, non pas « le début
du message», mais plutôt (selon la traduction de J.-M. Durand) ˙« l’ essentiel du message ».
Selon l’usage de l’époque, lorsque des messagers lisaient une lettre à un roi chez qui ils
avaient été envoyés, ils ne nommaient pas le destinataire, mais donnaient seulement le nom
de l’expéditeur41.
On est frappé, pour une fois, par le ton aimable de ce message de l’ empereur élamite. Il
propose, si son interlocuteur le lui demande, d’ envoyer un arc (qaštum) au temple du dieu Addu
d’Alep. Il préviendra également ce roi de toute dénonciation contre lui qui lui parviendrait de
la part d’un autre roi. Il justifie son attitude bienveillante par le fait que le souverain auquel
il s’adresse a rétabli des relations diplomatiques jusqu’ alors interrompues, avant de prendre
contact avec toute autre puissance42.
Quel peut donc être le destinataire de cette lettre peu commune ? J.-M. Durand a proposé,
à juste titre, qu’elle date de l’an  de Zimrî-Lîm (ZL 0), au moment de l’ invasion élamite.
Le roi auquel s’adresse l’empereur élamite lui avait envoyé des messagers après une période de
rupture des relations diplomatiques. Comme J.-M. Durand avait identifié Halu-rapi au chef de
la localité de Sapîratum, dans le Suhûm, il avait naturellement pensé à « un prince du sud-est
de la Haute-Mésopotamie» (Durand  : ). Mais la mention de l’ offrande d’ un arc au dieu
Addu d’Alep serait bien étonnante dans ce contexte43. Aussi avait-il ajouté en note44 : « Il n’ est
cependant pas impossible qu’il s’agisse d’ une homonymie [i.e. d’ un autre Hâlû-rabi45] et que
le destinataire de la lettre de l’Élamite ne fût Yarîm-Lîm lui-même ! » On observe cependant,
dans le discours du sukkal, l’opposition entre une situation antérieure et celle d’ aujourd’ hui :
précédemment, l’empereur avait refusé d’ envoyer un arc en cadeau au dieu Addu d’ Alep, alors
que désormais, si son correspondant lui en fait la demande, il acceptera de le faire. Si l’ on admet
que l’affaire se passe à Alep, il y a ici clairement une allusion à la conduite de Yarîm-Lîm avant
son décès, qui fut notoirement anti-élamite. Dès lors, le roi à qui s’ adresse le sukkal serait le
nouveau roi d’Alep, Hammu-rabi. La date de son avènement peut être fixée assez précisément,
puisque la mort de Yarîm-Lîm eut lieu peu avant le /viii/ZL  [= 0], moment où une arme
précieuse fut envoyée de Terqa à Alep pour son tombeau46. La lettre FM   nous apprend donc

39 Charpin  : .

40 Il n’ est pas question ici de tablette, de sorte qu’on ne peut savoir si ces messagers lisent un texte écrit (comme
c’ est parfois le cas) ou récitent un texte appris par cœur (cf. ARM / , commenté dans Charpin : ).
41 Voir par exemple ARM /  : () … ki-a-am iq-bi () um-ma šar-ra-a-ia, ou ARM  : () um-ma

šu-nu-ma um-m[a] a-tam-rum ÌR-k[a-a]-m[a].

42 Il faut comprendre DUMU.MEŠ ši-ip-ri (0) ša LUGAL.MEŠ ka-li-šu-nu, non pas comme «les messagers

envoyés par tous les autres rois », mais « les messagers envoyés à tous les autres rois» C’est au cours d’une discussion
sur FM   avec J.-M. Durand que le sens exact des l. 0–0 a pu être établi; qu’il en soit ici remercié.
43 Toutefois, l’âpilum de Šamaš à Andarig se préoccupe de présents à faire à Addu d’Alep: ARM /  et

commentaire dans Charpin  : .

44 Durand  :  n. .
45 On connaît plusieurs Halu-rabi. L’ un d’eux est messager du roi d’Ilan-sura Haya-Sumu (ARM  : ). La

plupart des autres références concernent un responsable de Sapîratum du Suhûm, ˙ qui se décrit comme «fils» de
Meptum. Il a écrit les lettres A. (Charpin :  nº); M. (Charpin :  nº); A. (Charpin
 :  nº) ; A. (Charpin  :  nº). Mais il existe aussi un messager de Zimrî-Lîm également nommé
Halu-rabi.
46 D’ après ARM   : – ; voir Charpin & Ziegler :  n.  et désormais Charpin b.
 «ainsi parle l’empereur» à propos de la correspondance des sukkal-mah 

qu’à peine monté sur le trône, le jeune Hammu-rabi envoya des messagers à l’ empereur élamite.
On comprend mieux dès lors ce qui est raconté dans une lettre de Hammi-šagiš. Alors qu’ il avait
été envoyé à Alep par Zimrî-Lîm pour négocier une alliance avec le nouveau roi, il fut témoin
d’événements qu’il rapporta au roi de Mari en ces termes47 :
4–7
Le messager élamite, lorsqu’il est passé en direction d’Alep, a envoyé depuis Imâr trois48 ser-
viteurs à lui [à Qatna]. 8–9Hammu-rabi (d’Alep) ayant appris [cela], il a envoyé [des gardes] à sa
frontière. 10–11Ils se˙sont emparé de ces gens à leur retour. 12–13On les a questionnés et voici ce qu’ils
ont dit:
« 13–14Le sire de Qatna nous a envoyés avec le message suivant: ‘15–17Le pays t’est donné, monte à
moi! Si tu montes,˙tu ne seras pas surpris’49.»
18
Ces gens sont au secret dans un village. 19–22Or actuellement, le sire de Qatna vient d’envoyer
[à l’empereur50] deux messagers à lui, des […] ayant pris leur tête. 22–25Il faut˙ que mon seigneur
donne des ordres stricts et qu’il écrive au sire de Babylone, afin que ces hommes ne puissent pas
sortir.
Comment situer chronologiquement cette affaire ?51 Son contexte est manifestement l’ avène-
ment de Hammu-rabi d’Alep. On a surtout retenu de ce texte la déclaration de soumission
du roi de Qatna envers le souverain élamite. Mais il faut aussi souligner que le sukkal d’ Elam
˙ ambassadeur à Alep, non à Qatna : ce ne sont que trois serviteurs qui partent à
avait envoyé son
˙
Qatna et leur mission est manifestement secondaire. L’ empereur d’ Elam a donc tenté de profiter
˙
de la succession sur le trône d’Alep pour faire en sorte que l’ alliance anti-élamite conclue par
Yarîm-Lîm avec Mari et Babylone ne soit pas renouvelée par son fils52. Le message cité par Halu-
rabi dans FM   aurait donc été envoyé par le sukkal à Hammu-rabi juste au moment où ce
dernier monta sur le trône d’Alep53. L’empereur élamite aurait alors tenté d’ attirer le nouveau

47 A. : () DUMU ši-ip-ri-im LÚ e-la-mu-ú-umki () i-nu-ma a-[n]a h[a-l]a-abki i-te -qú () i[š-t]u i-[ma]-ar ki
9
* LÚ.TUR.MEŠ-šu () [a-na qa-tá-nim] ki iš-pu-ur () [an-ni-tam] ha-am-mu-ra-b[i] ìš-me-ma () [ o o o ] a-na
pa-tì-šu i[t-ru-u]d () [L]Ú.MEŠ˙ šu-nu-ti i-na ta-ri-šu-n[u] () is-ba-tu () ù [t]e4-ma-am i-ša-lu-šu-nu-ti-ma
()˙ ki-a-am
˙ iq-bu-ú um-ma-a-mi LÚ qa-tá-na-yu-um () ki-a-am ˙ iš-pu-ra-an-né-ti
˙ um-ma šu-ma () ma-a-
˙
tum a-na qa-te9-ka na-ad-na-at () e-le-e-em šum-ma [t]e-[é]l-le-em () ú-ul ka-á[š-d]a-ta LÚ.MEŠ šu-nu ()
i-na  kap-ri-im sú-um-mu-šu () ù a-nu-um-ma LÚ qa-tá-na-[yu-um] ()  DUMU.MEŠ ši-ip-ri-[š]u LÚ.[MEŠ
………] () pa-né-šu-nu sa-a[b-tu] a-na [LÚ.SUKKAL] () it-tà-ra-ad be-lí dan-na-t[im] () li-iš-ku-un ù a-na
LÚ KÁ.DINGIR.[RAki] () ˙ li-iš-pu-ur-ma LÚ.MEŠ [šu-nu] ()
˙ ˙ [l]a us-sú-ú. Texte publié dans Durand : 
n.  ; nouvelle traduction dans LAPO  . ˙˙
48 La photo (Durand  : ) montre clairement * (et non ).
49 J.-M. Durand a changé d’ avis sur la valeur de kašâdum (cf. LAPO   n. a). Le meilleur sens me semble être

celui de « surprendre » (CAD K a).

50 L. , il n’ est pas possible de restituer le nom de l’Elam, car l’empereur élamite résidait alors à Ešnunna. Je

préfère restituer a-na [LÚ.SUKKAL].

51 J.-M. Durand avait déjà indiqué une fourchette chronologique assez précise dans son édition du texte: «Cette

lettre fait allusion à une situation historique non encore complètement éclaircie mais qui ne peut s’interpréter que par
un refroidissement des rapports entre Qatna et Alep. Le fait que Hammu-rabi soit roi et qu’une ambassade élamite
puisse encore prendre la route d’ Alep fait˙choisir une date comprise entre la fin de ZL 0 et le début de ZL 0, avant
qu’ Alep ne s’ engage aux côtés de Mari et de Babylone contre l’Élam.» (Durand :  n. ; note reproduite dans
LAPO  p. ). Les progrès accomplis dans la reconstitution de l’histoire politique permettent d’être aujourd’hui
encore plus précis (voir déjà Charpin/Ziegler :  et n. ).
52 Pour le caractère personnel des alliances à l’époque amorrite, voir Charpin sous presse. On se rappelle par

exemple l’ insistance avec laquelle le roi d’ Ešnunna, Ibâl-pî-El II, envoya à trois reprises des messagers à Zimrî-Lîm
juste après son avènement ; mais le nouveau roi de Mari préféra la fidélité au roi Yarîm-Lîm d’Alep.
53 Le seul argument contre cette hypothèse pourrait être le fait que dans sa lettre au sukkal, le roi dont l’identité

est à déterminer emploie la deuxième personne (FM  :  ÌR.MEŠ-ka). Du coup, on pourrait se demander si le roi
mystérieux n’ est pas celui de Qatna. En effet, Amud-pî-El était alors sur le trône depuis une dizaine d’années, il était
˙
pro-élamite et il employait la deuxième personne en s’adressant au sukkal (A.: –). Mais on ne comprendrait
pas pourquoi l’ empereur élamite lui proposerait d’envoyer un présent au dieu Addu d’Alep (FM  : 0–0), ni quel
changement serait intervenu à Qatna pour expliquer le revirement du sukkal envers le roi de cette ville. Par ailleurs,
˙
 dominique charpin

roi du Yamhad dans son camp, en utilisant différents arguments, dont le fait qu’ il a reçu des
messagers envoyés par Hammu-rabi d’ Alep. On sait que finalement ce dernier se rallia aux rois
de Mari et de Babylone.

Conclusion

Cette enquête révèle donc une fois de plus l’ extrême sophistication de la circulation de l’ infor-
mation à l’époque amorrite. La supériorité de l’ empereur élamite, reconnue par tous les rois de
l’époque, se marquait doublement dans la correspondance. Dans le contenu des messages : à une
exception près, toutes les lettres que nous connaissons se caractérisent par un ton péremptoire.
Mais également dans le formulaire même des adresses de lettre : nous avons affaire à un monde
où il fallait faire sentir sa puissance, dans la forme comme dans le fond. Enfin, ce dossier montre
à quel point les relations diplomatiques étaient géographiquement étendues54 et comment elles
se traduisaient éventuellement en termes religieux : on a retrouvé à Terqa une offrande envoyée
par Kudur-Mabuk depuis Larsa55, le sukkal d’ Elam pouvait sérieusement envisager d’ en faire
porter une au dieu Addu à Alep.

Bibliographie

Anbar, M. (): «To Put One’s Neck under the Yoke.» In Amit, Y. et al. (Eds.), Essays on Ancient Israel
in Its Near Eastern Context. A Tribute to Nadav Na" aman, Winona Lake: –.
ARM  = Birot, M. (): Lettres de Yaqqim-Addu, gouverneur de Sagarârum, Paris.
ARM  = Bardet, G. et al. (): Archives administratives de Mari , Paris.
ARM  = Limet, H. (): Textes administratifs relatifs aux métaux, Paris.
ARM / = Durand, J.-M. (): Archives épistolaires de Mari I/, Paris.
ARM / = Charpin, D. (): Archives épistolaires de Mari I/, Paris.
ARM  = Birot, M. (): Correspondance des gouverneurs de Qattunân, Paris.
ARM  = Kupper, J.-R. (): Lettres royales du temps de Zimri-Lim,˙˙ Paris.
AS  = Whiting, R.M. (): Old Babylonian Letters from Tell Asmar, Chicago.
Birot, M. (): «La lettre de Zimri-Lim à Tiš-Ulme.» In Lebeau, M. and P. Talon (Eds.), Reflets des
deux fleuves, volume de mélanges offerts à André Finet (= Akkadica Suppl. ), Louvain: –.
Charpin, D. (): «Usages épistolaires des chancelleries d’Ešnunna, d’Ekallâtum et de Mari,» NABU
/.
———. (): «Sapîratum, ville du Suhûm,» MARI : –.
———. (): «Hammu-rabi de Babylone et Mari: nouvelles sources, nouvelles perspectives.» In Renger,
J. (Ed.), Babylon: Focus mesopotamischer Geschichte, Wiege früher Gelehrsamkeit, Mythos in der
Moderne (= Colloquien der Deutschen Orient-Gesellschaft ), Saarbruck: –.
———. () Review of: Cohen, R. and R. Westbrook (Eds.), Amarna Diplomacy. The Beginnings of
International Relations, Baltimore-London, , RA : –.
———. () «Prophètes et rois dans le Proche-Orient amorrite: nouvelles données, nouvelles pers-
pectives.» In Charpin, D. and J.-M. Durand (Eds.), Florilegium marianum VI. Recueil d’études à la
mémoire d’André Parrot (= Mémoires de NABU ), Paris: –.
———. () «The Writing, Sending and Reading of Letters in the Amorite World.» In G. Leick (Ed.),
The Babylonian World, New York-London: –.
———. (a) «Tell Hariri / Mari: textes. II. Les archives de l’époque amorrite,» Suppl. DB : –.

la citation que fait Hâlû-rabi des propos des messagers élamites n’est sans doute pas mot à mot. Les intermédiaires
sont en effet nombreux : il s’ agit d’ une lettre de Hâlû-rabi, citant les propos de messagers élamites, qui transmettent
un message du sukkal, qui cite une lettre qu’il a reçue …
54 Pour l’ idée que l’ époque d’ El Amarna n’ait pas tant vu un élargissement des relations diplomatiques que leur

déplacement vers l’ Ouest, voir Charpin .

55 Wilcke .
 «ainsi parle l’empereur» à propos de la correspondance des sukkal-mah 

———. (b) « «Le roi est mort, vive le roi!» Les funérailles des souverains amorrites et l’avènement
de leur successeur.» In R. van der Spek (Ed.), Studies in Ancient Near Eastern World View and Society
Presented to Marten Stol on the occasion of his th birthday, Bethesda: –.
———. (sous presse): «Guerre et paix dans le monde amorrite et post-amorrite.» In Neumann, H.,
Dittmann, R., Paulus S., Neumann G. and Schuster Brandis, A. (Eds.), Krieg und Frieden im Alten
Vorderasien, AOAT , Münster.
———. (en prép.): La Correspondance à l’époque amorrite. Écriture, acheminement et lecture des lettres
d’après les archives royales de Mari.
Charpin, D. and Durand, J.-M. (): «La suzeraineté de l’empereur (Sukkalmah) d’Elam sur la
Mésopotamie et le ‘nationalisme’ amorrite.» In De Meyer, L. and H. Gasche (Eds.), Mésopotamie et
Elam. Actes de la XXXVIe Rencontre Assyriologique Internationale, Gand, – juillet  (= MHEOP
I), Ghent: –.
Charpin, D. and Ziegler, N. (): Florilegium marianum V. Mari et le Proche-Orient à l’époque amorrite:
essai d’histoire politique (= Mémoires de NABU ), Paris.
De Graef, K. (): «Les textes de V récent du chantier B à Suse (fin Sukkalmahat – ca. – av.
notre ère),» IrAn : –. ˘
De Meyer, L. (): L’accadien des contrats de Suse (= Ir.Ant. Suppl. ), Leiden.
———. () «Les structures politiques en Susiane à l’époque des sukkalmah.» In A. Finet (Ed.), Les
˘
pouvoirs locaux en Mésopotamie et dans les régions adjacentes, Brussels: –.
Durand, J.-M. (): «La Cité-État d’Imâr à l’époque des rois de Mari,» MARI : –.
———. (): «L’empereur d’Élam et ses vassaux.» In Gasche H., Tanret M., Janssen C. and Degraeve
A. (Eds.), Cinquante-deux réflexions sur le Proche-Orient ancien offertes en hommage à Léon De Meyer
(= MHEOP ), Ghent: –.
———. (): Les Documents épistolaires du palais de Mari, tome III (= LAPO ), Paris.
FM  = Durand, J.-M. (): Florilegium marianum VII. Le Culte d’Addu d’Alep et l’affaire d’Alahtum
(= Mémoires de NABU ), Paris.
Guichard, M. (): «Au pays de la Dame de Nagar.» In Charpin, D. and Durand, J.-M. (Eds.),
Florilegium marianum II. Recueil d’études à la mémoire de Maurice Birot ( = Mémoires de NABU ),
Paris: –.
———. (): «Lecture des Archives royales de Mari, tome XXVIII: lettres royales du temps de Zimrî-
Lîm,» Syria : –.
Heimpel, W. (): Letters to the King of Mari. A New Translation, with Historical Introduction, Notes,
and Commentary (= Mesopotamian Civilization ), Winona Lake.
Lafont, B. (): «Relations internationales, alliances et diplomatie au temps des royaumes amorrites.»
In Durand, J.-M. and Charpin, D. (Eds.), Mari, Ébla et les Hourrites: dix ans de travaux. Actes du
colloque international (Paris, mai ). Deuxième partie (= Amurru ), Paris: –.
LAPO  = Durand, J.-M. (): Les Documents épistolaires du palais de Mari, tome I, Paris.
LAPO  = Durand, J.-M. (): Les Documents épistolaires du palais de Mari, tome II, Paris.
Malbran, F. and Lackenbacher, S. (): «Les bénédictions dans les lettres de Suse,» NABU /.
MDP  = Dossin, G. (): Autres textes sumériens et accadiens, Paris.
Michel, C. (): Correspondance des marchands de Kanish (= LAPO ), Paris.
ShA  = Eidem, J. and Læssøe, J. (): The Shemshara Archives Vol.  The Letters (= Historisk-filosofiske
Skrifter ), Copenhague.
Steve, M.-J. (): «Suse: la couche XII du chantier ‘A’ de la ‘ville royale’ et la fin de l’époque des
Sukkalmah.» In Gasche H., Tanret M., Janssen C. and Degraeve A. (Eds.), Cinquante-deux réflexions
sur le Proche-Orient ancien offertes en hommage à Léon De Meyer (= MHEOP ), Ghent: –.
Steve, M.-J., Vallat, F. and Gasche, H. (/): «Suse,» Suppl. DB /: –.
Vallat, F. (): «Fragments de textes cunéiformes de Suse,» RA : –.
Veldhuis, N. (): «Old Babylonian Documents in the Hearst Museum of Anthropology, Berkeley,»
RA : –.
Wilcke, C. (): «Kudurmabuk in Terqa. » In Tunca, Ö. (Ed.), De la Babylonie à la Syrie, en passant par
Mari. Mélanges offerts à Monsieur J.-R. Kupper à l’occasion de son ° anniversaire, Liège: –.
 PROSOPOGRAPHISCHE UNTERSUCHUNGEN
ANHAND DER RECHTSURKUNDEN AUS SUSA

Sheyda Jalilvand Sadafi*

Bei den französischen Ausgrabungen in Susa wurden tausende von Tontafeln gefunden. Der
größere Teil von ihnen sind Rechtsurkunden. Der Rechtsurkundenkorpus aus Susa, der aus
Kaufurkunden, Pachturkunden, Darlehen, Erbteilungen, Schenkungs-, Heirats-, Geschäfts-
und Prozessurkunden besteht, ist in MDP , ,  und  von Vincent Scheil veröffentlicht
worden1. Der überwiegende Teil dieser veröffentlichten Urkunden wird teilweise im „National
Museum of Iran“ und im „Musée du Louvre“ aufbewahrt. Im Rahmen meiner laufenden
Dissertation über die Rechtsurkunden aus Susa in den vier genannten MDP-Bänden habe ich
diese Tafeln, soweit möglich, bearbeitet und kollationiert.
Die hier vorgestellte prosopographische Untersuchung, die ein Teil meiner Doktorarbeit ist,
ermöglicht uns einerseits, mehrere Familienarchive aus Susa systematisch zu rekonstruieren
und das Beziehungsgeflecht von lokalen Familien zu beschreiben2 und andererseits durch
die Analyse der Eidesformeln neue Einblicke in die chronologische Reihenfolge der lokalen
Herrscher zu gewinnen.
Hier wird exemplarisch der Stammbaum einer der Großfamilien aus Susa, Šušinak-šēmi (Teil
) und ihre lokale Herrscherabfolge (Teil ) rekonstruiert:

. Rekonstruktion des Stammbaumes der Familie Šušinak-šēmi

Von besonderem Interesse für die Rekonstruktion dieses Stammbaumes sind die Erbschafts-
urkunden, weil sie detaillierte Informationen zur Eltern-Kind-Abfolge3 und zu den Geschwis-
terbeziehungen liefern. Wir können die einzelnen Erben auf dem oben gezeigten Stammbaum
(Abb. ) in der folgenden Erbteilungsurkunde identifizieren:

Text A (MDP , :–)

I
Vs.  Gi-mil-li- Gimilli-Sîn,
I.d
 MÙŠ.EREN-ga-mil Šušinak-gāmil,
I.d
 UTU-še-me-e Šamaš-šemmê
 ù Ta-at-ta-a und Tattâ,
 DUMU.MEŠ dMÙŠ.EREN-še-mi die Erbsöhne des Šušinak-šēmi
 i-na tú-ba-ti-šu-nu haben in ihrer Güte,
 i-na ˙na-ar-a-ma-ti-šu-nu in ihrer Freiwilligkeit
 zi-i-zu me-e-sú-ú du-up-pu-ru (das Erbe) geteilt, sie sind rein, zufrieden,
 ta-mu-ú vereidigt.

Zur besseren Übersicht über die Mitglieder und Angehörigen der Familie des Šušinak-šēmi
wird hier der Stammbaum dieser Familie angegeben:

* Universität Tübingen. Ich bedanke mich bei Andreas Fuchs für seine Unterstützung und zahlreiche Anregun-
gen. Ebenso gilt mein Dank auch Catherine Mittermayer für ihre Anregung und Hinweise.
1 Scheil, V. , ,  und .
2 S. De Meyer, L. .
3 S. hierzu Klíma, J. .
 sheyda jalilvand sadafi

Abb. 4

Im Folgenden wurden für die genealogische Reihenfolge der Familie Šušinak-šemi Informa-
tionen aus den gesamten Rechtsurkunden aus Susa, die sich zu einzelnen Personen der Familie
gewinnen lassen, tabellarisch dargestellt:

Tabelle 5
Nr. Gen. Person Text Typ des Textes Beruf Funktion Herrscher
 I Šušinak- MDP ,  Erbteilung – Vater der Erbsöhne: Tan-Uli (Sukkalmah)
šēmi Gimilli-Sîn, und Temti-halki ˘
Šušinak-gāmil, ˘
Šamaš-šemmê und
Tattâ
MDP ,  (?) Prozessurkunde – eine der Parteien Kuk-Našur (Sukkal von
Elam) und Kuduzuluš
(Lugal von Susa)
MDP ,  (?) Erbteilung – Zeuge Temti-Agun
(Sukkalmah) und
Kuk-Našur˘
MDP ,  (?) Kaufurkunde – Verkäufer Kutir-Nahhunte und
˘˘
Temti-Agun
MDP ,  (?) Schenkungs- – Schenkende, Vater der –
urkunde fNarubtu

4Die römischen Ziffern I–V bezeichnen die Generationenabfolge. Die durchgehende Linie wird hier für die
Personen, die der Familie angehören, verwendet und die gepunktete Line für die Personen, die möglicherweise der
Familie angehören.
5 Belege, die nicht ganz sicher sind, wurden mit einem Fragezeichen „(?)“ gekennzeichnet. Nr. = Nummer der

Familienmitglieder im Familien-Stammbaum. Gen. = Generation.

 prosopographische untersuchungen anhand der rechtsurkunden aus susa 

Nr. Gen. Person Text Typ des Textes Beruf Funktion Herrscher
 II Gimilli- MDP ,  Erbteilung – Erbsohn des Tan-Uli (Sukkalmah)
Sîn Šušinak-šēmi und Temti-halki ˘
˘
MDP ,  Erbteilung – Vater des Gamāl-ili –
(Zeuge)
MDP ,  Erbteilung – Vater des Gamāl-ili –
(Zeuge)
 II Šušinak- MDP ,  Erbteilung – Erbsohn des Tan-Uli (Sukkalmah)
gāmil Šušinak-šēmi und Temti-halki ˘
˘
MDP ,  Erbteilung – Vater des Ātanah-ilu, –
Itti-ilı̄-balitu und˘
Warad-Ištar ˙
MDP ,  Erbteilung – Vater von Ātanah-ilu, –
Itti-ilı̄-balitu und˘
Warad-Ištar ˙
MDP ,  (?) Darlehen – Zeuge –
MDP ,  (?) Kaufurkunde – Verkäufer –
MDP ,  Pacht – Zeuge –
MDP ,  (?) Schenkungs – Schenkende, der –
urkunde Beschenkte ist unklar
MDP ,  Prozessurkunde – Bruder des Tattâ –
 II Šamaš- MDP ,  Erbteilung – Erbsohn des Tan-Uli (Sukkalmah)
šemmê Šušinak-šēmi und Temti-halki ˘
˘
MDP ,  Pacht – Zeuge –
MDP ,  (?) Kaufurkunde – Vater des Išmeanni Tan-Uli und
(Verkäufer) Kuk-Našur
MDP ,  (?) Kaufurkunde – Vater des Išmeanni –
(Verkäufer)
 II Tattâ MDP ,  Erbteilung – Erbsohn des Tan-Uli (Sukkalmah)
Šušinak-šēmi und Temti-halki ˘
MDP ,  Kaufurkunde – Vater des – ˘
Šušinak-kı̄nam-ı̄di
(Zeuge)
MDP ,  Schenkungs- – Vater des Šātu –
urkunde

MDP ,  Prozessurkunde – Bruder des –

Šušinak-gāmil
MDP ,  Erbteilung – Zeuge Tan-Uli (Sukkal) und
Temti-halki
MDP ,  (?) Prozessurkunde ba" iru Zeuge ˘
Temti-raptaš und
(Fän- Kuduzuluš
ger)

 II Narubtu MDP ,  Schenkungs – Beschenkte, Tochter –

urkunde des Šušinak-šēmi
 III Gamāl-ili MDP ,  Erbteilung – Zeuge, Sohn des –
Gimi[lli-Sîn]
MDP ,  Erbteilung – Zeuge, Sohn des –
Gimi[lli-Sîn]
MDP ,  Prozessurkunde – eine der Parteien –
 III Ātanah- MDP ,  Erbteilung – Erbsohn des –
ilu ˘ Šušinak-gāmil

MDP ,  Erbteilung – Erbsohn des –

Šušinak-gāmil
MDP ,  Erbteilung – Vater der Erbsöhnen, Temti-raptaš und
Kuk-Adar und Abı̄-ilı̄ Kuduzuluš
 sheyda jalilvand sadafi

Nr. Gen. Person Text Typ des Textes Beruf Funktion Herrscher
 III Itti-ilı̄- MDP ,  Erbteilung – Erbsohn des –
balitu Šušinak-gāmil
˙
MDP ,  Erbteilung – Erbsohn des –
Šušinak-gāmil
MDP ,  Prozessurkunde – eine der Parteien –
 III Warad- MDP ,  Erbteilung – Erbsohn des –
Ištar Šušinak-gāmil

MDP ,  Erbteilung – Erbsohn des –

Šušinak-gāmil
 III Išmeanni MDP ,  (?) Kaufurkunde – Verkäufer, Sohn des Tan-Uli und
Šamaš-šēmi Kuk-Našur
MDP ,  (?) Kaufurkunde – Verkäufer, Sohn des –
Šamaš-šēmi
 III Šātu MDP ,  Schenkungs- – Sohn des Tattâ –
urkunde

 III Šušinak- MDP ,  Kaufurkunde – Sohn des Tattâ –

kînam-ı̄di

 IV Kuk-Adar MDP ,  Erbteilung – Erbsohn des Temti-raptaš und

Ātanah-ilu Kuduzuluš
MDP ,  Darlehen – Zeuge˘ Temti-raptaš und
Kuduzuluš
MDP ,  Prozessurkunde – Bruder des Abı̄-ilı̄ Temti-raptaš und
Kuduzuluš (rabiānu)
MDP ,  Prozessurkunde – Vater des Temti-raptaā und
Puzur-Teppuna Kuduzuluā (rabiānu)
 IV Abı̄-ilı̄ MDP ,  Erbteilung – Erbsohn des Temti-raptaš und
Ātanah-ilu Kuduzuluš
MDP ,  Prozessurkunde – Bruder˘ des Kuk-Adar Temti-raptaš und
Kuduzuluš (rabiānu)
MDP ,  Prozessurkunde – Vater des Iqı̄šūni Temti-raptaā und
Kuduzuluā (rabiānu)
 V Puzur- MDP ,  Prozessurkunde – Sohn des Kuk-Adar –
Teppuna

 V Iqı̄šūni MDP ,  Prozessurkunde – Sohn des Abı̄-ilı̄ –

.. Erste Generation der Familie Šušinak-šēmi

In dem von mir ausgewählten Text (Text A) steht Šušinak-šēmi somit als Erblasser für die
erste Generation. Gimilli-Sîn, Šušinak-gāmil, Šamaš-šemmê und Tattâ sind die Erbsöhne des
Šušinak-šēmi, die das Erbe ihres verstorbenen Vaters rechtmäßig unter sich teilen.
Der Name Šušinak-šēmi wird außerdem in einer Schenkungsurkunde als Vater der Narubtu
genannt6. In dieser Urkunde schenkt Šušinak-šēmi seiner Tochter sein Vermögen, das er mit
zwei Frauen, Tepirti und Šušinak-nādā, nutznießt7. Ob diese Person mit dem Vater der Familie
identisch ist, ist nicht nachweisbar.

6 S. MDP , .

7 Diese Frauen sind möglicherweise die Schwestern des Šušinak-šēmi, die die Anteile ihres verstorbenen Vaters
mit ihrem Bruder im Besitz haben; s. MDP , .
 prosopographische untersuchungen anhand der rechtsurkunden aus susa 

Der Name Šušinak-šēmi tritt weiterhin in einer Prozess-8, einer Erbrechts-9 und einer Kau-
furkunde10 auf. Es ist aber auch hier nicht nachweisbar, ob die mit dem Personennamen
Šušinak-šēmi bezeichnete Personen in diesen Urkunden mit dem Vater der Familie identisch
sind.

.. Zweite Generation

Gimilli-Sîn tritt außer in der oben genannten Erbrechtsurkunde (Text A) in zwei anderen
Erbrechtsurkunden als Vater des Gamāl-ili (des Zeugen) auf 11.
Šušinak-gāmil kommt nicht nur in der oben genannten Erbrechtsurkunde (Text A), sondern
auch in zwei weiteren Erbrechtsurkunden vor12. In diesen Erbrechtsurkunden teilen Ātanah-ilu,
Itti-ilı̄-ballitu und Warad-Ištar, die Erbsöhne des Šušinak-gāmil, die Anteile ihres Vaters˘ auf.
Die beiden˙Urkunden sind inhaltlich ähnlich. Nur in MDP ,  wird Warad-Ištar als erster
Erbe und in MDP ,  wird Ātanah-ilu als erster Erbe genannt. Es ist anzunehmen, dass jeder
Erbe ein Exemplar der Urkunde bei˘sich behielt.
In einer Prozessurkunde steht Šušinak-gāmil als Vertreter seines Bruders, Tattâ, vor Ge-
richt13. Wir finden in dieser Urkunde die Namen weiterer Mitglieder der Familie, nämlich
Itti-ilı̄-balitu und Gamāl-ili14.
˙
Der Personenname Šamaš-šemmê kommt in dieser Namensvariante nur in von mir ausge-
wähltem Text (Text A) vor15. In zwei Kaufurkunden wird ein Šamaš-šēmi als Vater des Verkäu-
fers, Išmeanni, genannt16. Es ist nicht klar, ob die Person, die in diesen Kaufurkunden mit der
Namensvariante Šamaš-šēmi bezeichnet wird mit Šamaš-šemmê, dem Sohn des Šušinak-šēmi,
identisch ist. Der Name Šamaš-šemmê begegnet uns auch in einer Pachturkunde, dort als ein
Zeuge17.
Tattâ wird außer in der oben genannten Erbrechtsurkunde (Text A) in einer Prozessurkunde
als Bruder des Šušinak-gāmil erwähnt18.
In einer Erbrechtsurkunde wird Tattâ zusammen mit seinem Zeitgenossen Warad-Kūbi als
Zeuge genannt19. In einer weiteren Urkunde wird eine Person mit dem Namen Tattâ erwähnt
und hier möglicherweise in Bezug zur Tätigkeit als Fänger (bā" iru) gebracht20. Ob diese Person
mit Tattâ, dem Sohn des Šušinak-šēmi, identisch ist, ist nicht klar.

8 S. MDP , :–.

9 S. MDP , :.
10 S. MDP , :.
11 S. MDP , : Rs. ’ und : Rs. ’. Leider ist ein Teil des Namens abgebrochen. Scheil hat diesen Namen so

ergänzt: Gim[il]-Sîn. Da es nach der von mir vorgenommenen Kollation genug Platz für li an der abgebrochenen
Stelle gibt, werde ich den Namen des Vaters von Gamāl-ilı̄ als Gim[illi]-Sîn ergänzen. Desweiteren sind die Erben
in diesen Urkunden die Söhne des Šušinak-gāmil, des Bruders des Gimilli-Sîn; s. auch Anm. .
12 S. MDP , : und :. Ein Šušinak-gāmil, der Sohn des Šelebum, wird auch in diesen Erbrechtsurkunden

genannt.
13 S. MDP , ; s. auch Anm. . Tattâ war möglicherweise bei der Herstellung der betreffenden Urkunde

nicht anwesend oder nicht mehr am Leben.

14 S. MDP , :–: I.dMÙŠ.EREN-ga-mil IIt-[ti-DINGIR-ba-li-tu] ù IGa-ma-al-ì-[lí] a-na ÚRDU-Ku-bi id-

bu-[bu] um-ma Šu-nu-ma ITa-at-ta a-hi-[ni …] … Šušinak-gāmil, ˙It[ti-ilı̄-balitu] und Gamāl-i[li] haben dem
˘
Warad-Kūbu folgendermassen gesagt: „Tattâ, mein Bruder […] …“. ˙
15 S. Text A (MDP , :): dUTU-še-me-e. Zu šemmê „immer erhörend“ (Gewohnheits- oder Steigerungsadjek-

tive), s. von Soden, W. :  §m; Stamm, J.J. : ; CAD Š/II sub šemmû.
16 S. MDP , : und , : Vs. : dUTU-še-mi.
17 S. MDP , :: dUTU-še-em-me-e.
18 S. MDP , :–.
19 S. MDP , : Rs. ’-’; s. auch Anm. .
20 S. MDP , :. In dem Text ist nur das Zeichen ŠU von einem möglichen Sumerogramm ŠU.HA übrig

geblieben; vgl. MDP , :. ˘

 sheyda jalilvand sadafi

.. Dritte Generation

Gamāl-ili wird als Zeuge und Sohn des Gimilli-Sîn in MDP , : und : genannt21. Gemäß
den oben genannten Erbrechtsurkunden MDP ,  und  sind Ātanah-ilı̄, Itti-ilı̄-ballitu und
Warad-Ištar die Erbsöhne des Šušinak-gāmil22. ˘ ˙
Šušinak-kı̄nam-ı̄di, der Sohn des Tattâ, wird mit Šušinak-abi-enši, Sohn des Warad-Kūbi, als
Zeuge in einer Kaufurkunde genannt23. Šātu, der Sohn des Tattâ, wird als Zeuge seines Onkels
Šušinak-gāmil in einer Schenkungsurkunde erwähnt24.

.. Vierte Generation

Abı̄-ilı̄ und Kuk-Adar sind die Erbsöhne des Ātanah-ilu. In einer Erbrechtsurkunde teilen diese
zwei Brüder den Anteil des verstorbenen Vaters auf˘ 25.

.. Fünfte Generation

In einer Prozessurkunde steht Puzur-Teppuna, der Sohn des Abı̄-ilı̄, mit seinen Söhnen gegen
Iqı̄šūni, dem Erbsohn des Kuk-Adar, vor Gericht:

Text B (MDP , :–)

Vs.  Aš-šum-É šà i-na Te-em-
[ti-ra-ap-ta-aš]
 ù Ku-du-zu-lu-uš ra-bi-[a-nu]
I
 A-bi-i-li a-na Ku-uk-a-[da-ar]
 a-na ší-mi ga-am-ri id-di-n[u]
 IPúzur-dTe-ep-pu-na DUMU A-bi-i-li
 ù ap-lu-šu
 a-na I-qí-šu-ni a-na di-nim
it-te-[bu]
 um-ma šu-nu-ma É a-bu-ni
É a-bu-ni
 a-na a-bi-ka u-ul id-di-im-mi
Betreff des Hauses, das in der Zeit
des Temp[ti-raptaš]
und Kuduzuluš, den Ortsvorstehern,
Abı̄-ilı̄ dem Kuk-[Adar]
zum vollen Preis gegeben hatte.
Puzur-Teppuna, der Sohn des Abı̄-ilı̄
und seine Erbsöhne
sind gegen Iqı̄šūni für den Prozess
aufgestanden
(und) haben folgendes ausgesagt:
„Das Haus hat unser Vater
deinem Vater nicht gegeben (verkauft)“.

Aus dieser Urkunde geht hervor, dass Abı̄-ilı̄ dem Kuk-Adar ein Haus zum vollen Preis verkauft
hat. Zu einem späteren Zeitpunkt26, sagen Puzur-Teppuna und seine Erbsöhne aus, dass sein
Vater bzw. ihr Großvater, Abı̄-ilı̄, dem Vater des Iqı̄šūni, Kuk-Adar, das Haus nicht verkauft
habe.

. Präzisierung der Herrscherabfolge

Die prosopographische Untersuchung in den Rechtsurkunden aus Susa ermöglicht auch neue
Einblicke in die Rekonstruktion der Herrscherabfolge, denn die in den Eidesformeln genannten
Herrscher lassen sich mit der rekonstruierten Generationenabfolge verknüpfen.

21 S. auch MDP , :.

22 S. auch MDP , :–.
23 S. MDP , : und ; s. auch Anm.  und .
24 S. MDP , : Rs. ’.
25 S. MDP , :–.
26 Abū-ilı̄ und Kuk-Adar werden hier als Zeugen nicht aufgerufen, weil sie vielleicht bereits verstorben sind. Sie

haben in der Zeit der Herrscher Temti-raptaš und Kuduzuluš gelebt; s. Text B, Z. –.
 prosopographische untersuchungen anhand der rechtsurkunden aus susa 

Die Herrscher werden in der Eidesformel entweder mit einer abgekürzten Form des Titels
Sukkalmah bzw. Sukkal oder Lugal oder manchmal auch ohne einen Titel erwähnt.
˘
.. Generation I–V in Bezug zur Herrscherabfolge
Verfolgen wir die Familie des Šušinak-šēmi über fünf Generationen aus der Zeit der jeweiligen
Herrscher, so ergibt sich folgendes Bild27:

Tabelle 28
Gen. Person Funktion Herrscherperiode Text
I ) Šušinak-šēmi Erblasser Tan-Uli (Sukkalmah) / Temti-halki (?)29 MDP , 
˘ ˘
II ) Tattâ Erbsohn Tan-Uli (Sukkalmah) / Temti-halki MDP , 
Zeuge ˘
Tan-Uli (Sukkal) / Temti- halki˘ MDP , 
˘
III ) Ātanah-ilu Erblasser Temti-raptaš / Kuduzuluš MDP , 
˘
IV ) Kuk-Adar Erbsohn Temti-raptaš / Kuduzuluš MDP , 
Bruder des Abı̄-ilı̄ Temti-raptaš / Kuduzuluš (rabiānu) MDP , 
V ) Puzur-Teppuna Sohn des Kuk-Adar Temti-raptaš / Kuduzuluš (rabiānu) oder MDP , 
ihrer Nachfolger30

.. Erste Generation

Ob Šušinak-šēmi, der Erblasser aus der ersten Generation, wirklich während der Regierungszeit
der Herrscher Tan-Uli und Temti-halki gelebt hat, ist nicht sicher, da sein Name in dieser
Herrscherperiode selbst nicht bezeugt ˘ ist. Zu dem Zeitpunkt, als in dem von mir ausgewählten
Text (Text A) beim Leben der Herrscher Tan-Uli und Temti-halki geschworen wurde und die
˘
Erbkinder des Šušinak-šēmi die Anteile ihres verstorbenen Vater aufteilten, war dieser schon
31
nicht mehr am Leben .

.. Zweite Generation

Die zweite Generation der Familie Šušinak-šēmi, d. h. Šušinak-gāmil und seine Brüder, die
Erbsöhne des Šušinak-šēmi, waren eindeutig Zeitgenossen der Herrscher Tan-Uli und Temti-
halki, deren Namen beim Eid in der oben genannten Erbrechtsurkunde (Text A) angerufen
˘
werden 32
. Alle Erbsöhne waren bei der Herstellung der betreffenden Urkunde noch am Leben
und sie haben sich auf der Urkunde mit ihren Nagelabdrücken verewigt.

27 Für die Tabelle der Herrscherabfolge wurden nur die Tafeln ausgewählt, auf denen die Mitglieder der Familie,

die sicher zur Familie gehören, genannt werden; s. Tabelle .

28 In dieser Tabelle sind nur einige der Familienmitglieder mit ihrer identifizierten Nummer im Familien-

Stammbaum ausgewählt.
29 S. unter „Erste Generation“.
30 S. dazu unter „Fünfte Generation“.
31 S. MDP , : Rs. ’-’.
32 S. MDP , : Rs. ’-’.
 sheyda jalilvand sadafi

.. Dritte Generation

Ātanah-ilu, Itti-ilı̄-ballitu und Warad-Ištar, die Erbsöhne des Šušinak-gāmil, die die dritte
˘
Generation der Familie ˙repräsentieren, haben vor oder während der Regierungszeit der Herr-
scher Temti-raptaš und Kuduzuluš gelebt, denn in einer Erbrechtsurkunde, in der Kuk-Adar
und Abı̄-ilı̄, die vierte Generation der Familie, als die Erbsöhne des Ātanah-ilu genannt wer-
den, wird beim Leben der Herrscher Temti-raptaš und Kuduzuluš geschworen ˘ 33.

.. Vierte Generation

Gemäß der Erbrechtsurkunde in MDP ,  lebten Abı̄-ilı̄ und Kuk-Adar, die Erbsöhne des
Ātanah-ilu, in der Zeit der Herrscher Temti-raptaš und Kuduzuluš34.
˘
.. Fünfte Generation
Gemäß Text B (MDP , :–) hat die fünfte Generation des Familienzweiges, der auf
Šušinak-gāmil zurückgeht, nämlich Iqı̄šūni und Puzur-Teppuna, gleichzeitig oder direkt nach
der Regierungszeit der beiden Herrscher Temti-raptaš und Kuduzuluš gelebt.

Auf dieser Basis gelangen wir zu einer Herrscherabfolge, deren chronologische Reihenfolge
gesichert ist. Wie Tabelle  zeigt, kommt der Name des Herrschers Tan-Uli ab der dritten
Generation nicht mehr vor. Es zeigt sich, dass fünf Generationen der Familie Šušinak-šēmi
mit folgenden Herrscherpaaren in Relation zu bringen sind35:

Tabelle 

Generation Herrscherabfolge
I Tan-Uli (Sukkalmah) / Temti-halki (?)36
II ˘
Tan-Uli (Sukkal) / Temti- halki˘
˘ halki
Tan-Uli (Sukkalmah) / Temti-
III ˘
Temti-raptaš / Kuduzuluš ˘
IV Temti-raptaš / Kuduzuluš (rabiānu)
V Temti-raptaš / Kuduzuluš (rabiānu) oder ihrer Nachfolger

Das Ergebnis lässt sich erweitern, wenn wir den Befund von Itti-ilı̄-balit37, der aus einer ande-
ren Familie stammt, hinzuziehen. Itti-ilı̄-balit kommt in zahlreichen ˙Kaufurkunden vor. Er
war tätig in der Regierungszeit des Tan-Uli˙ und des Temti-halki38 bis in die Regierungs-
˘ wird er mit drei weiteren
zeit der Herrscher Kutir-Šilhaha bzw. Temti-raptaš39. Dazwischen
˘ ˘
Herrscherpaaren, Tan-Uli bzw. Kuk-Našur40, Temti-halki bzw. Kuk-Našur41, Kutir-Šilhaha bzw.
Kuk-Našur42, genannt: ˘ ˘ ˘

33 S. MDP , :– und –.

34 S. MDP , :–; s. auch Text B (MDP , :–).
35 S. auch Tabelle .
36 S. unter „Erste Generation“.
37 Nicht mit Itti-ilı̄-balitu, dem Erbsohn des Šušinak-gāmil zu verwechseln. Itti-ilı̄-balit ist der Erbsohn des Apil-

Kūbi; s. MDP ,  und˙. ˙

38 S. MDP ,  und .
39 S. MDP ,  und .
40 S. MDP , :.
41 S. MDP , ,  und .
42 S. MDP , .
 prosopographische untersuchungen anhand der rechtsurkunden aus susa 

Tabelle 

Person Herrscherabfolge Text

Itti-ilı̄-balit Tan-Uli / Temti-halki MDP , , 
(Sohn des ˙Apil-Kūbi) Tan-Uli (Sukkal)˘/ Kuk-Našur MDP , 
Temti-halki (Sukkal) / Kuk-Našur MDP , , , 
˘ haha (Sukkal) / Kuk-Našur
Kutir-Šil MDP , 
˘ ˘ 43
Šimut-wartaš MDP , 
Kutir-Šilhaha / Temti-raptaš MDP , , 
˘ ˘
Temti-raptaš MDP , , 

Durch die Kombination beider Tabellen ( und ) ergibt sich folgende mögliche Herrscherab-
folge:

Tabelle 

Tabelle  Tabelle 
(Familie Šušinak-Šēmi) (Itti-ilı̄-balit)
˙
Tan-Uli (Sukkal) / Kuk-Našur
Tan-Uli (Sukkal) / Temti-halki
˘
Tan-Uli (Sukkalmah) / Temti-halki
˘ ˘
Temti-halki / Kuk-Našur
˘
Temti-halki (Sukkal) / Kuk-Našur
˘ haha (Sukkal) / Kuk-Našur
Kutir-Šil
˘ ha
Kutir-Šilha ˘ / Temti-raptaš
˘ ˘
Temti-raptaš / Kuduzuluš
Temti-raptaš / Kuduzuluš (rabiānu)
Šimut-wartaš

Entsprechend dieser Rekonstruktion nimmt Tan-Uli den ersten Platz dieser Herrscherabfolge
ein. Tan-Uli, der zusammen mit Temti-halki regierte, ist auch in späteren Urkunden mit dem
Titel Sukkalmah bezeugt44. Der Sukkalma˘h, Kutir-Šilhaha regierte zusammen mit Temti-raptaš,
˘ und
Lugal von Susa˘45. In der Folge regierten Temti-raptaš
˘ ˘ Kuduzuluš46. Besonders zu beachten
ist die Platzierung von Temti-raptaš, da er sehr wahrscheinlich erst nach der Regierungszeit
von Tan-Uli regiert hat47.
Bisher liegen verschiedene Ansätze zur Darstellung der Herrscherabfolge in Elam für diese
Zeit vor48. Diese basieren auf archäologischen Befunden. François Vallat z. B. sah sich außer-
stande eine genaue Datierung für Tan-Uli und seine Nachfolger vorzunehmen49. Vallat setzte
ihre Herrschaft nach Kutir-Šilhaha und Temti-raptaš an. Meine Arbeit versucht einen Beitrag
zu einer möglichen neuen, d.h.˘ erweiterten,
˘ Betrachtung dieses Problems zu leisten.

43 Die Platzierung dieses Herrschers ist nicht sicher.

44 S. MDP , ; MDP , , , .
45 S. MDP ,  und ; MDP , .
46 S. MDP ,  und , .
47 S. auch Scheil, V. : –.
48 S. M.-J. Steve, H. Gasche und L. De Meyer (: ) notieren Nachweise zu Tan-Uli bzw. Temti-halki in der

Schicht XIII und in der Schicht XII von Chantier A; s. auch Scheil, V. : f.; s. auch Cameron, G.˘ : ;
Vallat, F. :  ff.; ders., : ; ders., : –, mit Tab. .
49 S. Schaubild, Vallat, F. : .
 sheyda jalilvand sadafi

Prospographische Untersuchungen dieser Art können also nicht nur Beziehungen innerhalb
der Familien, auf welche sich die Urkunden direkt beziehen, aufklären helfen, sondern darüber
hinausgehend bei Datierungsproblemen von Herrschergenealogien der Sukkalmah-Zeit hilf-
reich sein. Validere Ergebnisse wird man allerdings erst nach der Auswertung des ˘kompletten
Urkundenkorpus und weiterer prosopographischen Untersuchungen erhalten können.

Bibliographie

Cameron, G. (): History of early Iran, Chicago-London ().

De Meyer, L. (): „Une famille susienne du temps des Sukkalmahhu.“ IrAnt : –.
˘˘
Klíma, J. (): Untersuchungen zum Altbabylonischen Erbrecht, Prag.
Rutten, M. (): „Archéologie Susienne: deux fragments de tablette provenant de Suse au nom de
Siwepalarhuhpak.“ MDP : –.
˙ V. ():
Scheil, ˘ ˘ „La division du pouvoir à Suse.“ RA : –.
———. (): Actes juridiques susiens (suite: nº  à nº ) (= MDP ), Paris.
———. (): Actes juridiques susiens (suite: nº  à nº ) (= MDP ), Paris.
———. (): Actes juridiques susiens (suite: nº  à nº ) (= MDP ), Paris.
———. (): Mélanges épigraphiques (suite: nº  à nº ) (= MDP ), Paris.
Stamm, J.J. (): Die Akkadische Namengebung (= MVAeG ), Leipzig.
von Soden, W. (): Grundriss der Akkadischen Grammatik (= AnOr ) Roma.
Steve, M.-J., Gasche, H., De Meyer, L. (): „La Susiane au deuxième millénaire: à propos d’une
interprétation des Fouilles des Suse.“ IrAn. : –.
Vallet, F. (): „Réflexions sur l’époque des Sukkalmah.“ In Vallat, F. (Ed.), Contribution à l’histoire de
l’Iran: Mélanges offerts à Jean Perrot, Paris: –. ˘
———. (): „L’Elam du Ile Millénaire et la chronologie courte .“ Akkadica –: –.
———. (): „Temti-Agun I. Un nouveau sukkalmah.“ Akkadica : –.
 SCRIBAL TRAINING IN OLD BABYLONIAN SUSA

Mehrnoush Malayeri*

. Introduction

In recent years, the topic of scribal training in the Ancient Near East has become a focal
point of scientific discussion. Studies of school tablets have brought new insights into scribal
education. These findings, though, are mainly limited to the Mesopotamian educational centers
such as Nippur,1 Ur,2 Sippar-Amnanum,3 Meturan4 and Uruk5 since only the school tablets
found in these cities were taken into consideration, whereas “scribal education” in Susa, the
most important eastern neighbor of Mesopotamia since third millennium bc, remains yet
unknown.6
The following discussion tries to give an overview of the curricular approach of scribal
education as attested in the school tablets found in Susa. This paper is a summary of my ongoing
thesis at the Eberhard-Karls-University, Tübingen, Germany.

. Archaeology

The French excavations conducted over several decades at the site of Susa, yielded not only
the most spectacular pieces of the ancient Near Eastern art, but also thousands of clay tablets
inscribed with cuneiform writing. Among these tablets were a great number of school tablets.
The find location and context of the Susa school tablets is, unfortunately, only rarely given. The
documentation of these early excavations is inadequate based on modern standards. Only the
excavations by Roman Ghirshman in Chantier A and Chantier B in the Ville Royale, during
– are extensively documented (fig. ). He established a new method of studying
the stratigraphy, wherein all objects were documented precisely by archaeological context and
find numbers.7

* Institute of Ancient Near Eastern Studies (IANES), Eberhard-Karls-University, Tübingen. I would like to thank
my supervisor and teacher K. Volk for his guidance, helpful comments, and criticisms. Special thanks are due to my
dear friend C. Mittermayer, for her encouragement and discussions regarding this paper. Lastly, I should also thank
my dear friend J. Baldwin for proofreading my English.
1 See e.g. Veldhuis, N. , Robson, E.  and ibid. .
2 See e.g. Tinney, S. .
3 See Tanret, M. .
4 See e.g. Cavigneaux, A./Al-Rawi, F. .
5 See e.g. Cavigneaux, A., , ibid., .
6 Apart from a study on a limited number of exercise tablets from Chantier B, see Tanret, M./De Graef, K. .
7 See Steve, M.-J./Gasche, H./De Meyer, L. : –.
 mehrnoush malayeri

Fig. . The excavations by Ghirshman during –

(Steve, M.-J., Gasche, H. and L. De Meyer, , Fig. ).

. Dating

Since the archaeological reports deliver little usable information, it is not possible to date the
school tablets from Susa precisely. Again only school tablets from two spots, Chantiers A and B
in the Ville Royale, found during the excavations of Ghirshman, can be dated exactly based on
the archaeological context.8 According to the chronological table of the Chantiers A and B from
Steve, Gasche and de Meyer (fig. ), these tablets can be dated from the Old Elamite II to Middle
Elamite III, Early Old Babylonian to Middle Babylonian period. However, only the tablets from
the layers A XII to A XV (Old Elamite III) and all the tablets from the layers of Chantier B (Old
Elamite II to Old Elamite III) are included in this study, since they are chronologically relevant.
The paleography remains the only criterion, for dating the other school tablets, although it is
not very trustworthy. A preliminary paleographical analysis allows us to date them to the (early)
Old Babylonian period.

8 See Steve, M.-J./Gasche, H./De Meyer, L. : –, Annexe ; for stratigraphic Chantier B see De Graef,

: –; for school tablets from Chantier B see Tanret, M./De Graef, K. : –.
 scribal training in old babylonian susa 

Fig. . Date of the school tablets from Chantier A and B in the Ville Royale,
(derived from table  in Steve, M.-J./Gasche, H./De Meyer, L., : );
the number of the school tablets found in each layer is given between brackets.
 mehrnoush malayeri

Fig. .  large literary tablets in situ from the excavations by Ghirshman

in the winter of – (Ghirshman, R., : , Fig. ).
Fig. .  large literary tablets in situ (Steve, M.-J./
Gasche, H./De Meyer, L. : , Fig. )

. Text Corpus

This study is based on more than  published and unpublished school tablets from Susa,
which are located mostly in the “National Museum of Iran” and in the “Musée du Louvre”.
Two texts are kept in the “Museum of Susa”. The main corpus of published texts appeared in 
volumes of the series “Mémoires de la Délégation en Perse” (abbreviated as MDP): volumes 
(),  (),  (), which is dedicated to the mathematic texts from Susa, and
 (), which is dedicated to  large literary tablets found during the excavations by
Ghirshman in the winter of – (fig.  and ). Several other school tablets from Susa
were published individually or in small groups in various articles or monographs.9 Apart from
the published tablets, many unpublished ones were also identified in the “National Museum of
Iran” and the “Musée du Louvre”. They are included in this study.
Although all the school tablets from Susa are considered together as one corpus, it is
important to emphasize that they came from different archaeological contexts or even different
times within a period. Apart from the tablets from the Chantiers A and B, there is no certainty
about their exact dating or exact find spots and so it is almost impossible to speak about the
probable curricular changes from time to time or from place to place.

. Typology of Tablets

The physical typology of the school tablets from Susa follows, to some extent, the classification
established by Civil in MSL : ,10 however, the typological distribution differs strikingly from

9 E.g. Alster, B.  (proverbs), Cohen, M.E.  (The curse of Agade), Flückiger-Hawker, E.  (Urnamma
A), Klein, J.  (Šulgi A), Lambert, M.  (five lentils from Susa), Tanret, M.  and recently Tanret, M./De
Graef, K.  (the school tablets from Chantier B).
10 This classification is used to describe the typology of school tablets of the Old Babylonian period: (I) Multi-

columned large tablets, cylinders or prisms, (II) Large teacher-student copies; the obverse contains – columns
 scribal training in old babylonian susa 

the Mesopotamian ones. Large tablets and cylinders, referred to as type I, are well attested
among the school tablets from Susa. In contrast, no prism has so far been identified, which
definitely belongs to Old Babylonian period.
Remarkably type II tablets are almost unknown at Susa. This type of tablet, a well-known
type from the Mesopotamian sites, particularly from Nippur, carries usually different exercises
on the obverse than on the reverse.11 As these exercises generally belong to different phases
of the scribal education they can give us an opportunity to find out about the sequence of the
teaching materials. Such tablets can play a decisive role in the reconstruction of the curriculum.
At Susa type II is very rare. However, there are many tablets, mostly of type IV, which have the
same feature, i.e. the inscribed exercise on the obverse differs from the exercise on the reverse.
It seems that in Susa there was no specific type of tablet assigned to this feature.
Type III is also well attested, but the most common type among the school tablets from
Susa is type IV, i.e. lentil-shaped tablets, mostly with teacher-student exercises, which can be
divided into several subgroups.12 A very well known type of this kind is a tablet with two lines
of Sumerian terms on the obverse and a syllabic repetition of the same terms followed by their
Akkadian translations on the reverse13 (fig. ). This type, I shall call IVs (“s” for Susa), seems
to be characteristic of school tablets from Susa and except for one example from the Diyala
region14 it is not attested at the Mesopotamian sites.

Fig. . An example of a type IVs tablet (MDP , ).

of calligraphic exercises; the left column is the instructor’s model, the right, rarely preserved the student’s copy,
the reverse contains a long extract of another composition, or an earlier section of the same one, (III) Small one-
columned tablets and (IV) Small lentil-shaped tablets.
11 See Veldhuis, N. : –.
12 The subgroups can be defined due to the number of the lines on the obverse and/or on the reverse, e.g., type

IV1 = obverse – lines written by the schoolboy, reverse blank or type IV4 = obverse lines – written by the teacher,
lines – by the schoolboy, reverse blank.
13 Passim in MDP ,  and also among the unpublished tablets.
14 See Civil, M. : –.
 mehrnoush malayeri

. The Curriculum

In the scribal education, the apprentice scribe followed a specific curriculum, from the first
exercise of wedges up to complex texts. The reconstruction of the curriculum is mainly based
on the textual analysis of the tablets which carry different exercises on the same tablet, under the
assumption that the progress of the acquisition of skills and knowledge, from simple to complex,
from short to long texts, proceeded. This assumption assumed as correct  main phases can be
distinguished in curriculum of Susa: () Sign Exercises, () Word Exercises, () Mathematics,
() Sentence Exercises and () Literary Compositions.
The following figure shows the typological distribution of the school tablets at Susa (fig. ).

Fig. . The typological distribution of the school tablets at Susa.

.. Sign Exercises

The very first step in the scribal training of a student was to produce a clay tablet as a writing
medium and to handle a stylus as a writing instrument. After mastering this, the apprentice
scribe begun inscribing the wedges in the clay and practicing the three basic elements of
cuneiform signs, i.e. vertical, horizontal, or oblique wedges. Among the school tablets from
Susa, there is not one single tablet with these very basic exercises. It has not been so far
confirmed, whether they have been practiced at all or they are lost to us. The sign lists are the
most elementary exercises, which are known to me.

... Syllable Alphabet A

The first sign list is the Syllable Alphabet A (abbreviated as SA), which is well attested in
almost all the Old Babylonian schools except for in Nippur.15 It is a list of  lines of one
to four characters and contains a total of  different signs.16 The basic function of SA was

15 See Tanret, M. : .

16 Ibid.
 scribal training in old babylonian susa 

mastering the uncomplicated signs correctly. This sign list is well documented among the school
tablets from Susa. Except for two oblong tablets, all the other ones are typologically type IV
tablets.

... Susa Sign List

The other list, which the pupils had to learn, is a one-column list of more than  signs without
any columns for reading or meaning. Each entry is preceded by a vertical wedge.17 Within this
sign list no thematically composed sections can be detected, rather it follows the acrographic
principles. Apparently it is a sign list specific to the educational materials of Susa. No relation
between Susa Sign list and other sign lists such as Syllabary A or Proto-Ea has so far been
attested.
Unlike the tablets of “SA”, the typology of the tablets inscribed with the “Susa Sign List” varies
from type I, including cylinders, to type IV.
There are two tablets, which contain both lines of “SA” with the lines of “Susa Sign List”.18
But as they show no consistent sequence of learning between these two sign lists, it cannot be
concluded, which of them was first learned. Fig.  shows one of these tablets with four lines of
“SA” on the obverse and eight lines of “SA” with an excerpt from “Susa Sign List” on the reverse.

Fig. . A type IV tablet with exercises of SA and the Susa Sign List (MDP , ).

.. Word Exercises

After completing the first steps of the elementary phase, i.e. learning the cuneiform signs, the
pupils began to learn the words, through copying many lists of the words.

... List of Personal Names

Among the school tablets from Susa there are many exercises of personal names. Three different
groups can be identified: Sumerian personal names, Akkadian personal names, and Elamite
personal names. The exercises of the Sumerian personal names differ in structure and typology

17 It conforms to the format of “–”; see Civil, M. : .

18 MDP ,  and .
 mehrnoush malayeri

from the exercises of Akkadian and Elamite personal names. The Sumerian personal names
were mostly practiced in groups of three, on type IV tablet; whereas the Akkadian and Elamite
personal names appear mostly in one-column lists and the typology of the tablets varies from
type I to type IV, mostly type III.
Moreover, the curricular setting of the Akkadian and Elamite personal names seems not
to be the same as of the Sumerian personal names. Practicing the Sumerian personal names
together with the exercises of the previous phase19 indicates an earlier curricular setting, i.e.
at the beginning of the phase of the word exercises. In contrast to that, the co-occurrence of
especially Akkadian personal names with lexical lists20 on one tablet shows a much later level
in this phase.

... List of Names of Deities

The great numbers of tablets inscribed with names of deities indicates the important role of
theological issues in the scribal education. These tablets can be divided in two groups: ()
The tablets, which are all of type IV and inscribed with small groups of two to four lines of
divine names. They show structural and typological similarities to the exercises of the Sumerian
personal names. The majority of tablets with names of deities belong to this group. () The
tablets, which contain names of deities as a form of list. No type IV tablet was found among
them. As no significant correlation has so far been attested between these two groups, their
relation remains unclear.
Except for two names of Elamite gods, dMUŠ.EREN and dtu-zi, and three names of Akkadian
gods, da-a, dla-qi2-pu-um and dEN.ZU, all of the divine names practiced are Sumerian. A striking
fact is the preponderance of divine names that explicitly refer to the Lagaš-ĝirsu and Ur
pantheon, such as BaU, Ninĝirsu, Ninĝišzida, Hendursaĝa and dŠul-pa-e3.
˘
... Thematic Lists of Objects
The most important group of word exercises were the thematic lists of objects. Numerous tablets
were found on which one or more excerpts from lexical lists were practiced. Lists of wood,
doors, reeds, pots, hides, metals, domestic animals as well as wild animals, birds, stones, plants,
clothing and geographical names and terms have so far been identified. They are all unilingual
Sumerian, apart from the tablets of type IVs. As above mentioned, this type of tablet carries on
the reverse the Akkadian translations of the Sumerian words practiced on the obverse.21
As most tablets containing excerpts from lexical lists of objects are fragmentary, no complete
list of objects can be reconstructed. The curricular sequence of different topics or of different
sections within a list with same topic can also be reconstructed only in rare cases. The so far
attested sequences within or between the lists of objects differ considerably from either the
series of Uh or even each other.22
˘

19 Two tablets can be mentioned here: MDP , , which contains “SA” on the obverse and a Sumerian personal
name on the reverse and less probable MDP , , which is inscribed with the “Susa Sign list” on the obverse and
a list of Sumerian names on the reverse. As I have no access to the original tablet, it is not possible for me to collate
it and determine how much is missing. The presence of the genitive extension “la2”, if intended in this connection,
at the end of lines  and  on the reverse, could be an argument for a hypocoristic personal name with a theophoric
element, such as “<ur>-den-lil2-la2”, however, a list of names of deities cannot be excluded too.
20 As example MDP , ,  and .
21 Passim, e.g. tablets MDP ,  and  or MDP , .
22 See e.g. Tanret, M./De Graef, K. :  for the sequence of domestic animals.
 scribal training in old babylonian susa 

... Other Thematic Lists

In addition to different lists of objects, other thematic lists were also introduced. The so-called
lu2-list, a thematic list of titles and professions, and ugu-mu, a list of body parts, are attested
among the school tables from Susa so far.
The typology of the tablets inscribed with thematic lists varies from type I to type IV.

.. Mathematics
Besides learning the signs and the words, mathematics was also a very important element in
scribal education. On the basis of the complexity of the exercises, the mathematical tablets from
Susa can be divided in two groups: () The tablets, which contain exercises of elementary level
of mathematic, such as metrological lists, multiplication tables and tables of reciprocals. The
co-occurrence of mathematical exercises of this level and excerpts of lexical lists on the same
tablet,23 though very few cases, indicates that mathematic of elementary level was introduced
to the pupils along with the lexical lists. () The tablets, which contain more advanced level
of mathematic. These tablets seem not to be schoolboy exercises, but rather the work of an
advanced student or a master, in which mathematical problems and their solutions are listed.
All the tablets introduced in MDP , except for only one (MDP , ), belong to this group.

.. Sentence Exercises

After learning how to read and write Sumerian signs and words, a scribal apprentice completed
his knowledge of Sumerian with learning sentences. In order to learn this, short texts such as
proverbs or excerpts from long literary texts are usually practiced. Such exercises have three
features in common: the level of difficulty of the text, the length of the text and the typology
of the tablet, which is, except for two cases, type IV. Although these features allow us to place
these kinds of exercises at this stage of the curriculum, the curricular sequence of them remains
unclear. Here is a table of the sentence exercises with regard to their content, numbers of
excerpts found, and tablet typology:

Number of Type
Text excerpts found IV Small Oblong Tablet
Proverbs   –
Instruction of Šuruppag   
Gilgameš and Huwawa A   
˘
Royal Correspondence Šulgi to Aradmu   –
Puzur-Šulgi to Ibbi-Sin   –
A father and his perverse son   –
Nisaba A   –
Lipit-Ištar B   –

As we can see, the proverbs are the ones attested most. The presence of literary texts, such as
Nisaba A and Lipit-Ištar B, which are part of the Tetrad, and Gilgameš and Huwawa A, part of
the Decad,24 is remarkable. ˘

23 E.g. MDP ,  and , for the latter see Robson, E. : .
24 The terms “Tetrad” and “Decad” were used for the first time by Tinney, S. : .
 mehrnoush malayeri

.. Literary Compositions

During this phase of scribal education, Sumerian compositions with different levels of com-
plexity in vocabulary or syntax were copied. There are some examples of such texts among the
tablets from Susa. Typologically the tablets are either type I or type III as expected. The follow-
ing table summarizes the Sumerian literary texts found at Susa:

Category25 Composition25 Number

Narratives, Myths and Epics Dumuzi’s Dream 
Enlil and Sud 
Cursing of Agade 
Inana and Ebih 
˘
Hymns Inanna Hymn C 
Urnamma A 
Šulgi A 
The Keš Temple Hymn 
Lamentations balaĝ-composition 
Royal Correspondences Šulgi to Puzur-Šulgi 
Šulgi to Išbi-Erra 
Debate Poems Debate between bird and fish 
Debate between sheep and grain 
Debate between date palm and tamarisk 
Compositions with historical background Sumerian king list 
King list of Awan and Šimaški 

There are also a few tablets from Susa, which contain Akkadian compositions. It is not clear
whether they were really an element of the scribal education in Susa. The following table shows
these Akkadian compositions:

Text Number
The epic of Etana 
The myth of Anzu 
Hammurapi’s code 
Magic text 

. Conclusion

In conclusion, the school tablets from Susa are a case study for an education in a multicultural
social context. At a time (Old Babylonian Susa) when the mother tongue of the students was
probably Elamite and the language of administration was Babylonian, the most common lan-
guage of learning, i.e. language of scholars was Sumerian. The influence of the Mesopotamian
school system on the scribal education in Susa seems to have been so significant that Elamite
played almost no role in the scholastic context investigated here. As we have already seen, the
curriculum of Susa shows a similar structure to the Mesopotamian curriculum. The sequence of
the stages of scribal education is clearly traceable, but the curricular material itself shows strik-
ing differences. It reveals that the Mesopotamian curriculum was not just blindly adopted but

25 The titles of the categories follow the definitions used in ETCSL.

26 The titles of the compositions follow the definitions used in ETCSL.
 scribal training in old babylonian susa 

that it underwent a significant adjustment to fit the linguistic environment of Susa. The
characteristics of the school texts from Susa, such as the reproduction of Sumerian syllabic
terms27 or frequent use of non-standard orthographies,28 confirm this fact. The study of the
educational corpus from Susa uncovered many, but not all features of the curriculum and
process of scribal education at Susa. There are still many questions that need to be answered by
further investigations.

Bibliography

Alster, B. (): Proverbs of Ancient Sumer I–II, Bethesda.

Bruins, E.M. and Rutten, M. (): Textes Mathématiques de Suse (= MDP ), Paris.
Cavigneaux, A. (): “Schultexte aus Warka,” BagM : –.
Cavigneaux, A. and Al-Rawi, F. (): “New Sumerian Literary Texts from Tell Haddad (Ancient
Meturan): A First Survey,” Iraq : –.
Civil, M. (): “A School Exercise from Tell Asmar,” StOr : –.
———. (): Materials for the Sumerian Dictionary (= MSL XIV), Rome.
Cooper, J.S. (): The Curse of Agade, London.
De Graef, K. () Les Archives d’Igibuni. Les Documents Ur III du Chantier B à Suse (= MDP ), Ghent.
Dossin, G. (): Autres Textes Sumériens et Accadiens (= MDP ), Paris.
Edzard, D.O. (): Textes littéraires de Suse (= MDP ), Paris.
ETCSL = http://etcsl.orinst.ox.ac.uk/
Flückiger-Hawker, E. () Urnamma of Ur in Sumerian Literary Tradition (= OBO ), Fribourg.
Ghrishman, R. (): “Suse. Campagne de Fouilles –,” ArAs : –.
Lambert, M. (): “Cinq Textes Scolaires de Suse,” JA : –.
Robson, E. (): “The Tablet House: A Scribsl School in Old Babylonian Nippur,” RA : –.
———. (): “More than Metrology: Mathematics Education in an Old Babylonian Scribal School.” In
Imhausen A. and Steel, J. (Eds.), Under One Sky: Mathematics and Astronomy in Ancient Egypt and
Mesoptamia (= AOAT ), Münster: –.
Steve, M.-J., Gasche, H. and De Meyer, L. (): “La Susiane au deuxième millénaire: à propos d’une
interpretation des fouilles de Suse,” IrAnt : –.
Tanret, M. (): “Fragments de Tablettes Pour des Fragments d’Histoire.” In De Meyer, L., Gasche, H.
and Vallat, F. (Eds.), Fragmenta Historiae Aelamicae. Mélanges Offerts à M-J. Steve, Paris: –.
———. (): Per aspera as astra: l’apprentissage du cunéiforme à Sippar-Amnānum pendant la période
paléobabylonienne tardive (= MHET I/), Ghent.
Tanret, M. and De Graef, K. (): “The Exercise Tablets from Chantier B in Susa Revisited,” IrAnt :
–.
Tinney, S. (): “Texts, Tablets and Teaching, Scribal Education in Nippur and Ur,” Expedition /:
–.
———. (): “On the Curricular Setting of Sumerian Literature,” Iraq : –.
Van de Meer, P.E. (): Textes scolaires de Suse (= MDP ), Paris.
Veldhuis, N. (): Elementary Education at Nippur, Groningen.

27 This feature can be seen mostly on the type IVs. This type of tablet, as above mentioned, carries two Sumerian
terms on the obverse and syllabic repetitions of the same terms followed by their Akkadian translations on the
reverse. See “Typology of tablets”.
28 See e.g. Flückiger-Hawker (:  ff.), where this topic is discussed elaborately for the Susa version of

“Urnamma A”.
 ABIEŠUH, ELAM AND ASHURBANIPAL: NEW
EVIDENCE FROM OLD BABYLONIAN SIPPAR

Frans van Koppen*

Only one reference to Elam is currently known from the corpus of year names of the First
Dynasty of Babylon: in naming his th regnal year, Hammurabi of Babylon commemorated
how in the previous year he had thwarted an invasion by a host of nations under Elamite
high command. The purpose of the present communication is to draw attention to a second
attestation of Elam in this corpus, in a year name of Hammurabi’s grandson Abiešuh which
appears in abbreviated form in the dates of four hitherto unpublished tablets from Sippar in
the collection of the British Museum. Because a reconstruction of the full wording and order
of Abiešuh’s year names is still lacking,1 we must first address the chronological position of
this new date within his -year reign and its relation to other date formulas from his reign,
before we can turn to its historical significance, and its bearing on other traditions about the
interactions between Elam and Babylon under Hammurabi’s successors.

. The New Evidence

The full version of the th year name of Hammurabi is known from a date-list2 but is not
attested in actual tablet dates. In its place the abbreviation mu ugnim elam-ma(ki), “year: the
Elamite army,” was commonly used for the purpose of dating.3 Two tablets with the date mu
éren elam-maki were initially identified as also belonging to Hammurabi’s th year4 but on
inspection were found to bear the seal impressions of a servant of Samsuiluna, the son and
successor of Hammurabi. Therefore, they must be dated after Hammurabi’s reign. These two
tablets can be linked with two further tablets with partly damaged but identical dates, one of
which5 mentions the name of the king in question, Hammurabi’s grandson Abiešuh: [mu a-bi]-
e-šu-uh lugal-e [éren el]am-maki, “The year: Abiešuh the king <..> the army of Elam,” a typical
˘
short form of a year name that includes the ruler’s name and the object of his action while
omitting the main verb of the clause or any adjuncts. These four tablets, and one parallel text
carrying another year name of Abiešuh, are edited in the appendix to this contribution.

* Birkbeck, University of London. I would like to thank the Trustees of the British Museum for permission to
study tablets in the museum’s collections, the staff of the Department of the Middle East for their hospitality and
support, and Karen Radner for comments on a draft version of this article.
1 Important work on the year names of Abiešuh was done by Goetze () and Stol (), as well as Pientka

() and Horsnell (). A fresh attempt at a reconstruction on the basis of much new material is in preparation
by the author.
2 OECT  Pl. –: I –, edited by Hornell as Date-List O (: I –).
3 Passim, with variants mu éren elam-maki (CT   tablet), mu éren ugnim elam-ma (CT  c), mu éren-meš

elam-maki (Friedrich BA /:  no. ) and mu ugnim lú elam-ma, “year: the army of the ruler of Elam” (VS  ,
MHET / ), see Horsnell : II  note .
4 Sigrist/Figulla/Walker :  sub  and ibid.:  sub .
5 BM , read mu a-bi-e-šu-uh x kù-babbar in Sigrist/Figulla/Walker :  sub , and listed in

Horsnell : I  sub (). ˘

 frans van koppen

Turning to the question where this new date belongs within Abiešuh’s -year reign, and
whether and how it can be connected with any of his known year names, we must start with
the texts that are dated with the new formula. They belong to a single file and their contents
and prosopography (discussed in the appendix) allow us to narrow down their date to the first
decade of the reign of Abiešuh. Sin-iddinam, the owner of the seal impressed on four of the
five tablets, presumably had died by Abiešuh year “q”, which thus gives us a terminus ante quem
for the year under consideration. The chronological position of year name “q” has not yet been
determined with certainty but there are persuasive reasons to consider it the name of an early
year of Abiešuh, rather than one in the second half of his reign (as was maintained by Hornsell
using arguments developed by Goetze6). Firstly, a text from year “q” contains the most recent
evidence that one Ikūn-pi-Sin son of Sin-tayyār was still alive.7 This well-attested Sippar resident
was already old enough to be the father of a nadı̄tum-devotee by the th year of Hammurabi.8
In view of this individual’s life expectancy, it seems hardly possible to place year “q” later in
Abiešuh’s reign.9 Secondly, that same year “q” also saw an inquiry into the slave status of a young
girl who had lost her freedom after a Babylonian army had been defeated by the Kassites, forcing
her family to take refuge in Babylon.10 If we accept Wilcke’s suggestion and situate this military
encounter in the context of the war with the Kassites mentioned in the name of Abiešuh’s third
(or perhaps fourth) year (Abiešuh year name “d”),11 then again an early date for year name “q”
seems more plausible than one much later in his reign. It would seem opportune, therefore, to
restore the damaged entry for the seventh (or perhaps eighth) year of Abiešuh in Date-List B12
as mu alan [gal-gal-la], that is Abiešuh year name “q”.13
Date-List B contains recognizable signs of four year names:14

III  mu giša[š-te ..] Abiešuh year “e”

III  mu gìr x [..] Abiešuh year “f ”
III  mu šíta ma[h ..] Abiešuh year “g”
III  ˘
mu alan [gal-gal-la] Abiešuh year “q”

According to Horsnell’s careful reconstruction of this fragmentary tablet, there is space for
exactly three year names between the end of the section dealing with the reign of Samsuiluna
and this passage, meaning that the above year names designate Abiešuh’s fourth to seventh
year.15 This is however not easy to harmonize with the evidence from a number of economic

6 Horsnell : I –, Goetze : .

7 BM  (mu a-bi-e-šu-uh lugal-kam / alan gal-gal-la), where i-ku-un-ka-dsuen dumu dsuen-ta-ia-ar gives
out a loan. ˘
8 See CT  a and MHET /  with Harris : –.
9 Similar, but less extreme, is the case of our seal owner, Sin-iddinam son of Nūr-ilišu, who appears for the first

time (in the capacity of witness) in Samsuiluna year  (BE / :  with Wilcke : ).
10 BBVOT  , edited by Wilcke . I assume that šalāmum in lines  and  was used euphemistically.
11 Wilcke : .
12 BM  +  + , edited by Horsnell : I –.
13 Horsnell (: I – and I  note ) prefers to restore the damaged entry in III  as Abiešuh year

name “w” (mu alan [níg si-sá]), but this is excluded because the father of sons who divide their estate in Abiešuh
year “h” (OLA  ) appears in an active role in a text (BM : ) dated with a year name of Abiešuh now
provisionally labelled “ee” (mu aš-me n[a4 za-gìn]; this formula has not yet been recognized as a separate year name
and was inaccurately lumped together with year name “q” in Pientka : , and with year name “t” in Horsnell
: I –). Abiešuh year “h” immediately followed year “w” (see CT  c and OLA  , with the comments
by Goetze :  and Horsnell : I  on the first text) and restoring the damaged entry of the date-list as
Abiešuh year name “w” would leave no space to accommodate year name “ee” before the seventh (or perhaps eighth)
year of his reign.
14 Goetze : –, Horsnell : I – and I –.
15 Horsnell : I  and Figure .
 abiešuh, elam and ashurbanipal: new evidence from old babylonian sippar 

texts recording transactions early in the reign of Abiešuh and mentioning what seems like four
discrete year names (year names “a”-“d”).16 Therefore, the question whether three or four entries
for Abiešuh are lost in the break is not yet settled.
Be that as it may, all year names of Abiešuh prior to his year name “q” are now recognized and
their subject matter known in sufficient detail for a comparison with the new dating formula
involving the “Elamite army”, in order to see whether this succinct phrase can be linked with any
of them. With year “a” given over to the traditional accession theme,17 year name “b” celebrating
the performance of a redress act, year name “c” being a repetition of year name “b”,18 year name
“d” commemorating a victory over the Kassites, and finally year names “e” and “g” recording
the dedication of temple furnishings, only year name “f ” seems to qualify for consideration in
conjunction with our new formula.
Up to now year name “f ” was known only from Date-List B, where its two surviving signs
have not yet been convincingly deciphered,19 but not from any actual tablet dates—the only one
of Abiešuh’s year names. This implausible situation is remedied if we assign the dates invoking
the “Elamite army” on our tablets to this year name. More importantly, a date in year “f ” would
fit these texts particularly well because a related transaction took place in Abiešuh year “g”, that
is the very next year (appendix text ).
The discrepancy between the formula used in the date-list for year name “f ” and that
occurring in our tablets can easily be explained by assuming that the year name was abbreviated
in the date-list by its opening statement (probably referring to the king’s divine support),
whereas the target of royal action was the preferred short form for tablet dates. There are
many parallels for this. For example, Abiešuh year “d” is often abbreviated mu inim mah an
d
en-líl, “year: by the exalted command of An and Enlil”, but also mu éren ka-aš-šu-ú, “year: ˘
20
the Kassite army”. Or, to take an example from a date-list, Samsuiluna’s rd year appears in
the abbreviated form mu usu gìr-ra, “year: by the fierce power,” in a date-list21 and tablets,22
but in other tablets as mu ša-ah-na-a(ki), “year: (the city of) Šehna.”23 The opening statement
of year name “f ” cannot yet be˘ reconstructed on the basis of the ˘ poorly preserved entry in
Date-List B but relevant evidence may still turn up among yet unclassified tablet dates of this
period.24

. Another Conflict between Babylon and Elam

Abiešuh’s year name “f ” contains the first contemporary evidence for a second episode of war-
fare between the First Dynasty of Babylon and the Elamite state. Whereas the first confronta-
tion, between the forces commanded by the sukkalmah Siwepalarhuhpak and the coalition
of Mesopotamian states headed by Hammurabi of Babylon, ˘ is documented in rich detail by

16 See Goetze : – and the discussion by Horsnell : I –.
17 See Pientka : – and Horsnell : I –.
18 As well as year name “a” according to Horsnell : I –.
19 Goetze : , Pientka : , Horsnell : I  note .
20 References can be found in Pientka : .
21 VS  : rev. , edited by Hornell as Date-List G (: I –).
22 Passim, e.g. PBS /  and YOS  .
23 CT  a and YOS  : .
24 A good candidate is the date of BM , a record of Abum-waqar son of Liwwirum and Overseer of the

Merchants (van Koppen ). This individual is otherwise attested in Abiešuh year  (see footnote  below), year
“g” (BM ), year “q” (BM ), year “ee” (BM  and BM ; for Abiešuh year name “ee” see footnote
 above), and year “w” (BM  and BM ). I have however not been able to decipher the poorly executed
date formula of BM , but include a copy of the upper edge of this tablet containing the date as figure .
 frans van koppen

contemporary sources, until now just one, much later, allusion was available for what some
scholars have interpreted to be another clash between the two states. The reliability of this
later account can be called into question, and with no contemporary evidence at hand,25 other
scholars have rejected it for historical purposes. As a result this alleged second confrontation
does not feature in most surveys of Mesopotamian history of this period.
Reference to this second clash is made in inscriptions of Ashurbanipal of Assyria (–
c. bc), in connection with the return of a cult statue of Nanaya from Susa to Uruk26 following
this king’s second campaign against Humban-haltaš III (Ummanaldasi) of Elam and the sack
of Susa in bc.27 The prism annals in their editions A, F, and T28 describe how  (variants:
 and ) years before the goddess had become angry and left her abode, since then
staying in Elam in unbefitting circumstances. The return of the goddess is here represented
as the execution of an ancient divine command, with Ashurbanipal taking the role of the
predestined champion of the gods who brings Nanaya back to her sanctuary Ehilianna at
the Eanna temple in Uruk.29 The return of Nanaya, often together with two other Urukean
goddesses,30 is furthermore discussed in a number of texts known from Kuyunjik tablets.31
Particularly important for our purpose is Ashurbanipal’s building inscription for the temple
of Nergal in Kutha, again available in a tablet copy,32 where the king justifies his attack on Elam
as divinely ordered retribution for wrongs committed against his ancestors, more specifically a
destruction wrought  years ago:
“Kudur-Nanhundi, the Elamite (ku-dúr-na-an-hu-un-di lú e-la-mu-ú), who did not respect
˘
the oath by the great gods, who in his madness [trusted] in his own strength, brought his hand
against the sanctuaries of the land of Akkad and ruined the land [of Akkad]” (lines –).
Later in the text, after a description of his campaigns against Humban-haltaš, the king tells
how he escorted the goddesses Nanaya, Usur-amassa and Arkayı̄tu back to the Eanna temple
(lines –). This detail, in combination ˙with the mention before of a period of  years
(line ), shows that for Ashurbanipal the name of Kudur-Nanhundi was linked with the
abduction of Nanaya’s cult statue as described in his prism inscriptions.33

25 Note that Scheil’s interpretation (: –) of a fragmentary inscription in Elamite as referring to a

campaign by Kutir-Nahhunte and Temti-agun in northern Babylonia has been rejected by Vallat  and Glassner
; see Vallat a for the date of this document.
26 For the history of the cult of Nanaya of Uruk see most recently Beaulieu : –.
27 For this date see Frame : –.
28 Text Borger : –, with translation ibid.: ; see also ibid.:  (DT  || BM ).
29 For the theology of this passage see Nissinen : –.
30 Nanaya, Usur-amassa and Arkayı̄tu appear together in a number of these texts, and the name of the first is
˙
on occasion replaced by Ishtar (qarittu Ištar, K : obv  published in Bauer : ). This triad is known from
other first millennium contexts (Beaulieu  Chapter ) and has been compared with the Old Babylonian triad An-
dInanna, Nanaya and Kanisurra (Charpin : –). Ashurbanipal’s prism annals focus exclusively on Nanaya

(but note that a very similar account in the tablet K : III – published by Bauer :  mentions the three
goddesses), presumably because, as Nabû’s spouse (Nanaya etellet Eanna hı̄rat Nabû, --, :  published by
Bauer : ), her favours were particularly meaningful for the Assyrian ˘ king.
31 Fragments of what has been classified an “Ashurbanipal Epic” about his warfare in Elam prominently deal

with the revenge and return of Nanaya/Ishtar, Usur-amassa and Arkayı̄tu to Uruk-Eanna, where they will bless
Ashurbanipal and intercede on his behalf to Nabû ˙(references in HKL  p. ). The return of Nanaya is promoted to
casus belli in K  (CT  ) // --,  (Bauer : –), where we read how Ashurbanipal had written
in vain to Humban-haltaš before the attack, asking him to release the goddess to Uruk. A variant of this motive can
be found in the fragmentary source K  (Bauer : ), where his predecessor Humban-nikaš II (Ummanigaš)
apparently fails to return the deity.
32 Text and bibliographical references in Borger : –.
33 See also K  (Bauer : –), mentioning Nanaya’s anger and “the arrogant Elamite” (no name is

preserved), who has “ruined the land of Akkad”. Now, after an exile of [] years, the goddess conveys her desire
to return by means of dreams and prophetic messages for Ashurbanipal.
 abiešuh, elam and ashurbanipal: new evidence from old babylonian sippar 

Scheil () was the first to suggest an identification of Ashurbanipal’s Kudur-Nanhundi

with the sukkalmah Kutir-Nahhunte, the second successor of Hammurabi’s contemporary
Siwepalarhuhpak. A ˘ number of scholars have accepted this interpretation although others
have preferred to identify the nemesis of Ashurbanipal’s ancestors with Kutir-Nahhunte of
the Šutrukid Dynasty, who notoriously held power over Babylonia after the fall of the Kassite
Dynasty in the mid-th century bc.34 While the second alternative holds the attraction
of bringing all traditions about the hostility of Kudur/Kutir-Nahhunte together in a single
individual, it is nevertheless difficult to harmonize with the Distanzangabe of  years. This
is no random, or greatly exaggerated, number but a figure which Ashurbanipal’s scholars—
working with sources comparable to the Babylonian King List A—could easily have reached
when calculating the amount of time between the reign of a ruler of the First Dynasty of Babylon
and their own days.35 On the contrary, the reliability of the same chronographical tradition for
late second and early first millennium bc history makes it quite unlikely that counting from
the end of the Kassite Dynasty could have resulted in this number. It is always possible that
by the time of Ashurbanipal the name of the Šutrukid ruler was erroneously associated with a
tradition about earlier events, but the presence of another ruler of the same name at what seems
the right time does encourage the more straightforward explanation: that the sukkalmah Kutir-
Nahhunte attacked Babylonian territory at some time during the reign of one of Hammurabi’s ˘
successors.
The question when and where this encounter should have taken place has so far been
answered with the help of the few details that can be learned from Ashurbanipal’s inscriptions.
The abduction of the cult statue of Nanaya and her return to Uruk implied for Leemans an
attack on Uruk and southern Babylonia, which he understood to have been a factor in “the
disintegration of Hammurapi’s kingdom in the second part of Samsuiluna’s reign”.36 In this he
was followed by Gasche and Charpin, who both proposed to link the Elamite attack with the
abandonment of the cities in the southern alluvium, early in the second decade of the reign
of Samsuiluna.37 Ashurbanipal returns the deities to Uruk but it is important to realize that
the relevant passages nowhere state that the gods had also been taken from that city. Beaulieu
was therefore able to suggest instead that a raid may have been carried out on Kish,38 in the
Babylonian heartland, where the gods of Uruk had found shelter when their home town was
deserted early in Samsuiluna’s reign.39 This suggestion becomes all the more convincing if one is
prepared to link this raid with Abiešuh year name “f ”, as we propose in this article, because one
may safely assume that by then the breakdown of regular worship at Eanna had run its course.

34 Stolper : – note  and Vallat  (both with other references). For the Šutrukid Kutir-Nahhunte

and Babylonia see Brinkman : – and –, Stolper : –, and Lambert .
35 Mesopotamian scholars assumed that all past dynasties on the record had been consecutive, as can be shown

with the help of a Distanzangabe from the time of Enlil-nadin-apli (Brinkman : –). We are not familiar with
the sources for Babylonian history used by Ashurbanipal’s scholars, but may assume that their figures resembled
those of the Babylonian King List A, with  years for the First Sealand Dynasty and  years and  months for
the Kassite Dynasty. There is no complete Babylonian king list for the post-Kassite period: Brinkman’s chronology
(apud Oppenheim : –) recognizes  years between the last year of Enlil-nadin-ahi () and the
return of Nanaya’s statue in bc, and we may assume that Ashurbanipal’s scholars reckoned with a similar, or
perhaps somewhat higher, number. Subtracting all this from the figure of  (variants:  and ) years, we
are left with  (variants:  and ) years, which should be allocated to late Old Babylonian reigns. Since we don’t
know which specific figures were used in calculating this time-span, it would be futile to speculate which reign may
have been its starting point, but it seems clear enough that Ashurbanipal’s Distanzangabe began at some point in
time during the First Dynasty of Babylon.
36 Leemans : .
37 Gasche :  note , Charpin :  and note .
38 Beaulieu : .
39 For the transfer of cults from Uruk to Kish see Charpin : – and Pientka : – and –.
 frans van koppen

In addition, matters of military geography also favour a northern Babylonian theatre for this
assault. Any army invading Mesopotamia from the east inevitably entered the alluvium via the
Diyala corridor, either coming down the Zagros highlands (like Cyrus’ invasion of  bc), or
setting out from Susa and marching north via Der.40 This means that Uruk was effectively out
of reach for a direct attack from Elam.
Above we have seen that Abiešuh’s year name “f ” refers to a military encounter between
Babylonian and Elamite armies, more specifically one in which the Babylonians were successful
enough to merit celebration in a year name. A source from Sippar dating to the ninth month
of the preceding year (Abiešuh “e”) reveals that not long before that time some enemy force
had posed a threat to that city, because “the city gate of the land was opened” once again.41
This obviously refers to the relaxation of city defence measures, and it would appear that these
had been implemented over a wide area.42 We would therefore like to propose that Abiešuh’s
adversary was no other than the sukkalmah Kutir-Nahhunte, and that it is to this campaign
˘ this view because the earlier proposal of taking
that Ashurbanipal’s later accounts refer. We take
Kutir-Nahhunte as the opponent of Samsuiluna was not based on any factual evidence,43 and
an attack on Kish, as it can be deduced from the Ashurbanipal inscriptions, ties in well with the
reported state of emergency throughout the northern part of the Babylonian state. We therefore
need to consider whether the contemporary evidence from Kish reveals any trace of an assault
on that site.
Old Babylonian Kish was an urban conglomerate, with its two main settlements at Uhaimir
and Ingharra (ancient Kiš and Hursagkalama) occupied simultaneously. To my knowledge
the known Old Babylonian records ˘ from Kish contain no evidence which would confirm
the postulated Elamite assault early in the reign of Abiešuh, but relatively few sources dating
to this period are available. Their paucity however could be taken as an argument that no
major damage was inflicted. In that case one might reasonably expect to see an increase in
the number of records dating to the years leading up to the catastrophe, but no such pattern
is apparent.44 There is, perhaps, one feature in the settlement history of Kish which might tie
in with Ashurbanipal’s account: excavation of an area of houses west of the Zababa temple at
Uhaimir have produced very few, if any, post-Samsuiluna dates.45 This seems remarkable, given

40 For the routes between the Susiana and the Mesopotamian alluvium see Postgate : .
41 Van Lerberghe/Voet :  Di : : i-nu-ma ká.gal ma-tim ip-pé- tu -[ú] (transcription only). The
interpretation of the preceding line is unfortunately not entirely clear: wa-ar-ka i-ta-a-sí- a-am (transcription only)
can perhaps be translated “after (the enemy) had departed”, but the implicit subject and ˙ the unusual orthography for
a preterite of wasûm Gt require some explanation.
˙ of the land” presumably means the gate of every fortified settlement in the land, a meaning also
42 The “city gate

applicable to a passage in a Yahdun-Lim inscription (cited CAD A/ p. a).

43 Note also that a literary fragment from Kuyunjik mentioning the word sukkalmah and, perhaps, Samsuiluna

(published by Lambert : –) is too broken to be helpful for historical purposes and ˘ is better left aside for the
purpose of dating Kutir-Nahhunte’s campaign (pace Glassner ).
44 A file of loans contracts given out by Gimillum and dating to the later reign of Samsuiluna and the first three

(or four?) years of Abiešuh (dates collected in Goetze : –) is the only text group from Kish thus far known
that ends early in the reign of Abiešuh (for the Kish provenance see the unpublished tablet MLC ). Lacking
more dossiers with the same chronological profile, no archive-closing catastrophe can be postulated on the basis of
this file alone. Its homogeneous character suggests instead that we are looking at a group of tablets which had been
set aside early in the reign of Abiešuh for reasons unknown to us (its end date does not coincide with any known
debt cancellation act).
45 The vast majority of tablets from de Genouillac’s mission at Kish were found in this area (de Genouillac :

–), and his report contains just a handful of texts dating to later reigns (Pientka : –, items  and –
). No individual find spots are given, so that the possibility cannot be excluded that these tablets were actually
found elsewhere at the site; the Oxford-Chicago expedition in any case recorded no tablets from Uhaimir with
dates later than the reign of Samsuiluna. De Genouillac describes the area west of the ziggurat as pitted by fresh
excavations (ibid. p. ), and unprovenanced Kish material with the same date range and acquired in the years
 abiešuh, elam and ashurbanipal: new evidence from old babylonian sippar 

their close proximity to an important temple which is known to have been in use under later
kings, and one might consider linking this with an Elamite attack early in Abiešuh’s reign. But
the sources from this area do not display the typical catastrophe profile—there are ample records
dating to Sin-muballit, Hammurabi, and the first decades of Samsuiluna, but their number tails
off as one moves through this last reign. The decline of Uhaimir thus seems to be a gradual
process, one in which enemy havoc might have featured in a subsidiary role, but which cannot
be used to confirm an attack independently. Old Babylonian sources thus neither corroborate an
Elamite raid on Nanaya’s sanctuary (at Kish or elsewhere), nor rule out this hypothesis, and the
issue will merit reconsideration once more work has been done on the archival reconstruction
and the urban history of Old Babylonian Kish.

. Conclusion

While currently not attested, the name of Abiešuh’s year “f ” (which is the fifth, or perhaps sixth,
year of his reign) in its full version no doubt celebrated a Babylonian victory over the Elamites
in the preceding year,46 but the situation on the ground seems to have been far more precarious
than whatever view the royal rhetoric may have presented: the enemy had crossed the Tigris
and Irnina, watercourses that under normal circumstances served as secure boundaries at the
kingdom’s northern and eastern flanks,47 and was threatening the cities of northern Babylonia
long enough for “famine and hardship” to take hold of their besieged inhabitants.48 By entering
into Babylonian territory the Elamites were this time far more successful than they had been
under Siwepalarhuhpak in Hammurabi’s th year, when their assault on Babylon was brought
to a halt already at the Irnina frontier.49 This is confirmed also by the fact that this crisis was still
remembered almost a millennium later, whereas Siwepalarhuhpak’s invasion seems to have left
no mark in later tradition. We have proposed to link this event with a story about ‘crimes’ of the
distant past in Ashurbanipal’s inscriptions, from which we learn that they had been committed
 years ago, that the Elamites had carried off the statues of Urukean deities, and that their
ruler was called Kudur-Nanhundi. Chronologically, this statement matches our Old Babylonian
date, and while it remains possible that the other two details might result from contamination
by memories of events that had taken place in Babylonia in the th century bc, this need not
necessarily have been so: we have seen that an attack on Kish could indeed have been part of
the crisis of Abiešuh year “e”, and agree with an earlier suggestion that Ashurbanipal’s Kudur-
Nanhundi is identical with the sukkalmah Kutir-Nahhunte.
This proposition has some implications ˘ for the political history of the sukkalmah period.
Kutir-Nahhunte belonged to the first generation after Siwepalarhuhpak,50 and was his ˘ second
successor (after Kuduzuluš) in the supreme office of Elamite leadership.51 If he indeed was

before his expedition () can be confidently attributed to the same locus. This working hypothesis may not apply
to tablets of later date (reigns of Ammiditana to Samsuditana), which in view of their distinct subject matter (archives
of the Urukean clergy) perhaps were found elsewhere at the site (Ingharra?).
46 Or perhaps in the current year: since we have no evidence yet that this year name was in use before the tenth

month (see appendix), it cannot be excluded that year name “f ” was introduced in the course of the year, in which
case the celebrated event would have taken place earlier in the same year.
47 For Babylon’s northern border see van Koppen and Lacambre –: –.
48 Van Lerberghe/Voet :  Di : –.
49 Lacambre .
50 For Kutir-Nahhunte, “fils de la sœur de Siwepalarhuppak” see Steve/Gasche/De Meyer : –. The

reliability of this statement is challenged by Vallat˙ a: , but I assume it is based on the unpublished land grants
of Kutir-Nahhunte mentioned by Steve/Gasche/De Meyer : .
51 Most elegantly shown by the letter MDP  , for which see Grillot/Glassner :  and . For Kutir-

Nahhunte as sukkalmah see the comments of Vallat  and the response by Grillot/Glassner .
˘
 frans van koppen

Abiešuh’s opponent, then we have to take into account that the invasions of Hammurabi year 
and Abiešuh year “e” which took place  (or perhaps ) years apart were instigated by rulers
of two successive generations. This would then probably imply that Siwepalarhuhpak was still
relatively young by the time of the first, and Kutir-Nahhunte at a more advanced age at the time
of the second invasion. Accepting that Kutir-Nahhunte reigned as late as the fifth (or sixth) year
of Abiešuh moreover means that the statement of filiation given by his successor Temti-agun,52
who calls himself “sister’s son of Siruktuh”,53 cannot refer to a biological reality but must have
been used as a legitimizing identity.54
In contracts written before Temti-agun was recognized as sukkalmah he is frequently men-
tioned in a junior position of leadership alongside Kutir-Nahhunte.55 ˘From other documents
we also know of a number of obscure rulers of higher rank mentioned beside Temti-agun, and
these texts too date to the time before the latter had attained the highest office.56 The evidence
has been interpreted as reflecting a power struggle in the Elamite royal house,57 but much of
this episode remains unclear, including the question whether Kutir-Nahhunte was still alive
and in control of the supreme office throughout this time,58 and whether his war with Babylon
may somehow be connected.59 Our suggestion that Kutir-Nahhunte was well advanced in years
when he waged war on Babylon may be relevant to this issue. It allows us to assume that the
expedition took place towards the end of his reign, rather than at any time before, and to see this
act of aggression as a sign of instability within the ruling elite at a time when Kutir-Nahhunte’s
leadership waned. Whether Temti-agun’s rivals had secured their positions of authority before
that time, or whether some first came to power after the old king had died will remain unclear
for the moment, but more light on this period can be expected from the ongoing study of the
contemporary archival sources from Susa.
Temti-agun’s period as sukkalmah in all likelihood corresponds to the middle part of the
˘
reign of Abiešuh, and might have lasted longer. His second successor in the supreme office
(after Kutir-Silhaha ) was Kuk-Našur, the “sister’s son” of Temti-agun62 who first comes into
60 61

view during the latter’s term of sukkalmah as the junior ruler at his side.63 A land grant by the
sukkalmah Kuk-Našur dating to the first year˘ of Ammisaduqa of Babylon is a well-known point
of reference for the chronology of the Elamite rulers of˙ this period,64 but this date is not easily
˘
harmonized with the other chronological indicators at hand. We have proposed that Kutir-
Nahhunte was advanced in years by Abiešuh year “e”, and may assume that Temti-agun had
been his junior partner for quite some time before. It is then possible that Temti-agun’s “sister’s
son” Kuk-Našur was still in power in the first year of Ammisaduqa,  (or ) years later? Or,
setting aside the propositions made above, can we accept that ˙ three generations of the Elamite
royal family lasted for  years (from Hammurabi year  to the first year of Ammisaduqa)?65
˙

52 The second sukkalmah of that name according to Vallat .

53 Malbran-Labat no. :˘ , MDP  : , and MDP  : –.
54 Pace Grillot/Glassner : – and ; see already Vallat b: .
55 References in Börker-Klähn :  and Grillot-Glassner :  note .
56 References in Börker-Klähn :  and Grillot-Glassner : .
57 Grillot-Glassner .
58 This is the position maintained by Grillot-Glassner : –.
59 As was suggested by Börker-Klähn : .
60 Vallat .
61 The second sukkalmah of that name according to Vallat b.
62 According to MDP ˘ :  and the legend of his seal impressed on that tablet (Glassner ).
63 References in Vallat b.
64 VS  , see Steve/Gasche/De Meyer : , Stolper : , and Vallat : .
65 Siwepalarhuhpak was sukkalmah in Hammurabi year  (Vallat : –), and Kutir-Nahhunte, Temti-
˘
agun and Kuk-Našur are almost certainly three successive generations (see footnotes , – and  above).
 abiešuh, elam and ashurbanipal: new evidence from old babylonian sippar 

Although neither situation can be categorically excluded, it seems nevertheless improbable

in view of what can be expected with generations of average length. We therefore prefer to
take the Ammisaduqa synchronism as an terminus ante quem for Kuk-Našur, instead of an
terminus ad quem ˙ as has been done so far. A closer look at the tablet VS   mentioning Kuk-
Našur and Ammisaduqa supports this view as well, because the presence of a Babylonian year
name on an Elamite ˙ royal deed does not mean that the beneficiary of the grant recognized
the authority of both kings, but implies that we are dealing with a dated copy of an older
document. The fact that the tablet is not sealed but carries a drawing are further indications
for this view.66 How old the original was when it was copied, and why a Babylonian scribe
would copy an Elamite land grant, are questions that the tablet unfortunately does not help
answering.

Appendix: Five Tablets from Sippar

Four tablets dating to months  and  of Abiešuh year name “f ” record various transfers of
silver with the same subscript: namharti awı̄lim libbu  / mana ( šikil)67 kaspim ša qāt Marduk-
˘
nāsir “(silver) received by the gentleman, part of (a fund of)  / mina (and  shekel) of silver
˙
under the control of Marduk-nāsir.” The name of the anonymous recipient is found in the legend
of the seal impressed on three of ˙ the four tablets:68 “Sin-iddinam, son of Nūr-ilišu, servant of
Samsuiluna” (figure ).
This Sin-iddinam is a well-known resident of Sippir-Amnānum who inherited the office of
Overseer of the Merchants from his father Nūr-ilišu69 in the th year of Samsuiluna and held
it until the th (and last) year of this king.70 The Overseer of the Merchants was the high-
est post in the municipal administration that was reserved for members of the town’s leading
families. While up to that time control of the office alternated among different families, Nūr-
ilišu and his descendants retained it for three generations, with Sin-iddinam handing over
the position to his eldest son Marduk-nāsir by the first year of Abiešuh.71 The tablets pub-
lished here suggest that Sin-iddinam may ˙have stepped down because of new responsibilities
under the new king. His death seems to have occurred in, or shortly before, Abiešuh year

66 The original has not been checked but no sealings are recorded in the publication.
67 Text  omits the three shekels.
68 No seal impression can be seen on text .
69 The name of Nūr-ilušu, the son of Šarrum-Šamaš and Overseer of the Merchants from late in the reign of

Hammurabi (cf. MHET /  [collated], tablet rev 0: [igi s]ig-ì-lí-šu [..], case rev 0: [igi sig]-ì-lí-šu ugula dam-
gàr, oath by Hammurabi) until Samsuiluna year  (BE / ), is usually written sig-ì-lí-šu and has so far been read
Ipiq-ilišu (Harris :  and , Veenhof –, and Charpin ), but appears as nu-úr-ì-lí-šu in MHET
/ :  (Samsuiluna oath, date illegible), MHET / : rev 0 (not dated), and BM : 0 (date broken).
This shows that the sign sig could apparently be used as a graphic variant for zalag2, a sign with also consists of
four wedges in an arrangment rather similar to sig.
70 For this person see Harris : , Veenhof –: , and Charpin . Sin-iddinam holds the title

of Overseer of the Merchants for the first time in Samsuiluna year  (BM A: 0) and for the last time in
Samsuiluna year  (BM : ). Note that the Overseer of the Merchants Sin-iddinam son of Nūr-ilišu should
not be confused with Sin-iddinam son of Šērum-bāni, Overseer of the Merchants of Sippir-Yahrurum during much
of the reign of Abiešuh (e.g. CT  b, cited by Charpin ; see also footnote  below). ˘
71 See especially MHET /  (collated), dated to the eleventh month, day  [+x], of the first year of Abiešuh:
220 ki dmarduk -na-si-ir ugula dam-gàr Ie-tel-ka-dna-bi-um 230 Id suen -e-ri-ba-am Idmarduk-mu-ša-lim 240 ù x [x
de]n.zu-i-din-nam dumu˙ sig-ì-lí-šu 250 [ImunusPN lukur]-dutu dumu-munus a-bu-um-wa-qar 260 [in-ši-in-sa10]. The
second and third sign of line 0 are perhaps a[h-hi] or, less likely, du[mu-meš]. Sin-iddinam’s nonattendance is
remarkable, and might be due to his long-term absence ˘ ˘ from Sippar (see below). It seems quite plausible that the
buyer’s father was Abum-waqar son of Liwwirum and Overseer of the Merchants (see below).
 frans van koppen

name “q”, when we find a record of receipt for a quantity of barley, “part of the outstanding
delivery of Sin-iddinam son of Nūr-ilišu,” supplied by Sin-erı̄bam, known to be another of his
sons.72
Marduk-nāsir, the son of Sin-iddinam, was Overseer of the Merchants in Abiešuh’s first year
˙ this office for long.73 Sometime later Abum-waqar son of Liwwirum, who up to
but did not keep
that time may have held the position of Overseer of the Merchants for the local community of
Larsa residents,74 assumed this responsibility for the whole of Sippir-Amnānum. This transfer
of responsibilities was apparently brought about by royal appointment,75 but does not seem
to have dented the prestige and influence of Sin-iddinam’s family: Marduk-nāsir and his eldest
brother Etel-pi-Nabium continue to appear very prominently in contract witness ˙ lists and other
documents throughout the reign of Abiešuh.
The four tablets do not tell us the patronymic of Marduk-nāsir, but since Etel-pi-Nabium
˙
is mentioned as his brother in text , we can deduce that the documents concern the sons
76
of Sin-iddinam. They record expenditures from an account “under the control” (ša qāt) of
Marduk-nāsir but authorized by his father Sin-iddinam. Silver is spent on visitors to Babylon
(texts  and˙ ), for Šērum-ilı̄, whose status is unknown (text ), and is handed over to Sin-
iddinam (first post of text ) as well as Etel-pi-Nabium (second post of text ). This last post
also reveals that Etel-pi-Nabium had a similar account at his disposal, as we read that Sin-
iddinam transferred silver between his sons in order to “make them even” (mahārum Št). The
records summarize all silver spent as having been received by Sin-iddinam (nam ˘ harti awı̄lim),
˘
even though only one post involves a payment to him personally, the others disbursements
going out to third parties by his authorization. Sin-iddinam made these payments in Babylon,
and had the tablets brought to Marduk-nāsir in Sippar, where they would have been filed, in
case the owner of the capital would ever summon˙ Marduk-nāsir for a formal settlement of
accounts. Who this owner might have been is nowhere made explicit ˙ but the first post of text
 holds a clue. Here we see that Marduk-nāsir’s silver fund in reality consisted of agricultural
produce, in this case three-year-old bulls, the˙ proceeds of which were paid out to Sin-iddinam.
Three-year-old bulls are one of the staples in the well-documented “Palastgeschäfte” at the

72 BM , dated to the fifth month, day  [+x], of year Abiešuh “q” (mu alan gal-gal-la): .. še 2 gišbán
d[utu] 3 [š]à .. še-gu[r] 4 lá-u [mu-d]u 5 [I]dsuen-i-din-nam 6 dumu sig-ì-[lí]- šu 7 [nam]-ha-ar-t[i] 8 [x x] x
˘
gàr ša [o] x (LoE broken, probably uninscribed) rev 1 i-na qá-[ti] rev 2 Idsuen-e-ri-ba-[am] rev 3 ù tap-pé-šu. For Sin-
erı̄bam see MHET / :  . 0
73 Note that one Marduk-nāsir (patronymic unknown) who occurs the title of Overseer of the Merchants

(probably of Sippir-Amnānum, being ˙ the successor of Abum-waqar in that office) later in the reign of Abiešuh (e.g.
TCL  ) is quite likely a different person than the son of Sin-iddinam with the same name.
74 Assuming that Abum-waqar, Overseer of the Merchants, who appears alongside the judges of Larsa in

BM  (Jursa ) is the same individual as the son of Liwwirum (for whom see footnotes  and  above).
Since BM  dates to the seventh month of Abiešuh’s first year and MHET / , where Marduk-nāsir son of
Sin-iddinam carries the title of Overseer of the Merchants (footnote  above), to the eleventh month of ˙that year,
we have to assume that the responsibilities of Abum-waqar were initially different from those of Marduk-nāsir.
75 Abum-waqar’s promotion would appear to have taken place in the context of changes in the municipal ˙
administration at Sippar, as is suggested by the following observations. In the first place, from this time we find
again two Overseers of the Merchants in Sippar, one for each part of the double town. This meant a return to an
earlier arrangement, attested during the late reign of Hammurabi and the early years of Samsuiluna (Charpin ),
after the two posts had been merged into a single office held by Nūr-ilišu and Sin-iddinam during the second half of
Samsuiluna’s reign. Secondly, Abum-waqar and his equivalent at Sippir-Yahrurum (Sin-iddinam son of Šērum-bāni)
˘ in their seal inscriptions (van Koppen
are the first office holders to include their title of Overseer of the Merchants
 and RIME ...); this would be the norm henceforth, suggesting that the office had become subject to
royal appointment.
76 The text describes both as mār bı̄tim, “son of the house”. For Marduk-nāsir and Etel-pi-Nabium being sons of

Sin-iddinam, see e.g. MHET /  (footnote  above). ˙

 abiešuh, elam and ashurbanipal: new evidence from old babylonian sippar 

time of Ammiditana and Ammisaduqa, managed and commercialized by entrepreneurs on

˙
behalf of herd-owning palace dignitaries, 77
and we can therefore count Sin-iddinam among
their ranks. Sin-iddinam’s resignation from the post of Overseer of the Merchants hence seems
to have coincided with his appointment to one of the great offices of state. We do not know
the title of his new position, but it would seem to have given him access to state resources, the
day-to-day administration of which he delegated to his sons and other managers (see text ).

Text : Abiešuh Year Name “f ”, Month , Day  (Figure )

BM  (--, ) (.×.×.cm)
[..]  gín kù-babbar na4 d[utu]
 [ šám (x)] diš gu4 mu 
[ša ki dmarduk]-na-si-ir dumu é
 [iš-š]a-mu-ú-ma ˙
[a]-na a-wi-lim id-di-nu

 / ma-na kù-babbar na4 dutu
<ša> a-na e-tel-ka-dna-bi-um
 dumu é
a-na ki-ma dmarduk-na-si-ir
 šu-ta-am-hu-ri-im ˙
˘
in-na-ad-nu-šum
LoE (uninscribed)
Rev šu-ti-a a-wi-lim
šà  / ma-na kù-babbar
 nig-šu dmarduk-na-si-ir
(ruling) ˙
(uninscribed space)
[iti ab-u]d.du u4--kam
 [mu a-b]i-e-šu-uh lugal-e
[éren] elam-maki˘
UE [broken]

No sealing visible

[x +]  shekels of silver (according to) the stone of Šamaš, [the price] of x three-year-old
bulls that were bought from Marduk-nāsir, son of the house, and which he/they paid to the
˙
gentleman. One-third mina of silver (according to) the stone of Šamaš, which was given
to Etel-pi-Nabium, son of the house, in order to make him even with Marduk-nāsir. (All
silver) received by the gentleman. Part of (the total sum of)  / mina of silver under˙control
of Marduk-nāsir. Day  of the tenth month of the year: “Abiešuh the king: The army of
Elam”. ˙

Notes
: Either read [šám x+]  gu4 mu- or, alternatively, [šà šám]  gu4 mu-.

77 See Charpin :  and Stol : , , and .
 frans van koppen

Text : Abiešuh Year Name “f ”, Month , Day  (Figure )

BM  (--, ) (.×. ×.cm)
[ig]i--gál kù-babbar a-na uzu
 a?-na? re-eš x-gal
[i]gi--gál kù-babbar a-n[a] uzu
 a-na si-lá dumu-meš ši-ip x x tu?
i-nu-ma a-na x [..]
LoE a-na ša-da-d[i ..]
[o o o] x x [..]
Rev [il-l]i-ku-[nim]
[šu]-ti-[a] a-wi-lim
 [š]à  / ma-na [] gín kù-[babbar]
[ní]g-šu dmarduk-na-si-ir
(ruling) ˙
(uninscribed space)
 [iti ab]-è-a u4--kam
UE [m]u éren elam-maki

Seal of Sin-iddinam son of Nūr-ilišu

One-sixth shekel of silver for meat at the disposal of the palace (?); one-sixth shekel of silver for
meat provisions of the messengers from Amurrum (?), on the occasion when [..] came [(to ..)]
in order to … (All silver) received by the gentleman. Part of (the total sum of)  / mina and
 shekels of silver under control of Marduk-nāsir. Day  of the tenth month of the year: “The
army of Elam”. ˙

Notes
: ana rēš (if read correctly) is perhaps elliptic for ana rēš .. kullim?
: The exact connotation of the verb šadādum in this fragmentary context is unclear. Since the
passage concerns supplies of food, should one consider the usage of the verb for provisioning
(cf. CAD Š/ pp. –)?

Text : Abiešuh Year Name “f ”, Month , Day  (Figure )

BM  (--, ) (.×.×.cm)
 gín kù-babbar na4 [dutu]
 lál dše-rum-ì-l[í (o)]
ša a-na šám a?-šà? [(o)]
 in-na-ad-nu-šum
šu-ti-a a-wi-lim
 šà  / ma-na +[ gín kù-babbar]
LoE (uninscribed)
Rev níg-šu dmarduk-na-s[i-ir]
(ruling) ˙
(uninscribed space)
 iti ab-è u4--[kam]
mu éren elam-maki
UE (uninscribed)

Seal of Sin-iddinam son of Nūr-ilišu

 shekels of silver (according to) the stone of Šamaš, (the sum) lacking for Šērum-ilı̄ which was
given to him for the purchase of a field. (Silver) received by the gentleman. Part of (the total
 abiešuh, elam and ashurbanipal: new evidence from old babylonian sippar 

sum of)  / mina and  shekels of silver under control of Marduk-nāsir. Day  of the tenth
month of the year: “The army of Elam”. ˙

Notes
: or perhaps a-na šám  gín [kù-gi]. For payments of gold see Stol : .

Text : Abiešuh Year Name “f ”, Month  (?), Day  (Figure )

BM  (Bu --, ) (.×. ×.cm)
x / gín kù-babbar na4 dutu
(line left uninscribed)
 [s]i-lá sú-ha-re-e ša-pir aga-uš-meš
˙ ˘
ù b[e-l]í-qar-ra-ad
 ša k[a-ni-kam] ub-lu-nim
i-nu-ma a-wi-lum
 [o-n]a ká-dingir-raki
[o o o o] x x dsuen
LoE [..]
šà? [o] / ma-na? kù?-babbar?
 [..]
Rev x x x šum? x [..]
 a-na dmarduk-na-si-ir in?-[na]-a[d]?-nu-ma
a-na sú-ha-re-e ša˙ša-pir aga-uš-meš
˙ ˘
 ù be-lí-qar-ra-ad
i-na ká-dingir-raki ú-še-sú-ú
 šu-ti-a a-wi-lim ˙

šà  / ma-na  gín kù-babbar
 níg-šu dmarduk-na-si-ir
(ruling) ˙
UE [iti zíz]-a u4--kam
 [mu éren] elam-maki

Seal of Sin-iddinam son of Nūr-ilišu

x + / shekels of silver (according to) the stone of Šamaš <for …>, provisions of the servants
of the troops commander and Bēlı̄-qarrād, who brought a sealed tablet when the gentleman in
Babylon […] … […]. Out of x + / mina of silver [which …] was given to Marduk-nāsir and
˙
which in Babylon he/they paid out to the servants of the troops commander and Bēlı̄-qarrād.
(All silver) received by the gentleman. Part of (the total sum of)  / mina and  shekels of
silver under control of Marduk-nāsir. Day  of the eleventh (?) month of the year: “The army
of Elam”. ˙

Notes
: One line was left blank after line  for recording details of the type of foodstuffs which had
been bought for the individuals of lines – (similar to ana šı̄rim in text :  and ). This
information was never inserted.
0: The available space is insufficient for [iti ab-è]-a.

The following document does not concern Marduk-nāsir, nor is it dated to year Abiešuh “f ”. It
needs to be considered alongside the previous records˙ because it includes a similar subscript
and features the same seal impression. Sin-iddinam is here seen withdrawing from a fund (of
 frans van koppen

unspecified value) “under the control of Išme-Sin”. This name does not appear elsewhere in the
fragmentary file of archival records belonging to Sin-iddinam and his son Marduk-nāsir, and it
is therefore not possible to say more about his connection to the family. Whatever his˙ relation
might have been, in the light of the above discussion of texts – we can identify him as another
manager of Sin-iddinam’s ex officio assets. The text dates to Abiešuh year name “g”,78 that is one
year later than the other four tablets.

Text : Abiešuh Year Name “g”, Month , Day  (Figure )

BM  (--, ) (.× .×.cm)
xxxxxx
 a-na x x x x
[o o o] gi? mu? x x (x)
 [o o dna]-b[i-u]m-ma-lik
[o o o o] me?-še?-qí-im
LoE [o] x [o] x
pa-qá-di-im
Rev šu-ti-a a-wi-lim
šà kù-babbar
 níg-šu iš-me-dsuen
(ruling)
(blank line)
iti ne-ne-gar u4--k[am]
UE mu šíta mah kù-babbar kù-g[i]
˘
Seal of Sin-iddinam son of Nūr-ilišu
(Obverse and lower edge have not been deciphered). (Silver) received by the gentleman. Part
of (the fund of) silver under control of Išme-Sin. Day  of the fifth month of the year: “The
exalted weapon of silver and gold”.

Notes
The obverse is badly worn and almost illegible.
: This is to my knowledge the first instance where “silver” appears alongside “gold” as the
material of the “weapon” in Abiešuh’s year name “g” (see Pientka :  for attestations).

The Seal of Sin-iddinam (Figure )

Sin-iddinam is known to have used two different seals. The first bears a legend giving his father’s
name and an expression of devotion to the goddess dIštar-šar-ba-at (unpublished). The second
seal expresses loyalty to king Samsuiluna in its legend: dsuen-i-din-[nam] / dumu sig-ì-lí-šu /
[ì]r sa-am-su-i-lu-na-k[e4]. This seal is published here. It is impressed on tablets – above, as
well as on the tablets BM A (= case of MHET / , dated Samsuiluna year ) and
BM  (unpublished, date broken). Sin-iddinam used this seal during the last phase of his
career as Overseer of the Merchants at Sippar, and continued to use it in his new role in Babylon
under Abiešuh.
The seal is c. .cm high and was set in metal caps which are visible in the impressions; they
obscure details along the base-line of the scene. The composite drawing is based on all known
impressions listed above. Depicted is a variation on the three-participant presentation scene

78 The date was identified as the name of Hammurabi’s rd year in Sigrist/Figulla/Walker :  sub .
 abiešuh, elam and ashurbanipal: new evidence from old babylonian sippar 

involving a seated deity or king. On the left is the adorant with clasped hands, facing the seated
figure in a flounced robe who extends his right hand, presumably holding a cup (detail missing).
The figure is seated on a throne, his feet probably resting on a platform (detail missing). The
nature of his head-dress is not entirely clear but seems to be a turban, hence his identification as
a deified king. There is no supplicant goddess behind the worshipper, as is common in scenes of
this type. Instead a full-face goddess in a flounced robe stands behind the king, with her right
hand raised. Filling motives include a monkey (between the adorant and the king) and sun disc
and crescent (between the king and the goddess).
The composition of the scene reflects Sin-iddinam’s individual taste. The inclusion of a three-
line inscription box left insufficient space for a four-participant presentation scene that would
have included the supplicant goddess.79 By giving preference to the depiction of a full-face
goddess while dropping the usual supplicant goddess, Sin-iddinam opted to depict the two
subjects of his adoration on his seal: the king and dIštar-šar-ba-at, his family deity.

BM  Upper Edge (Figure )

The contents of this tablet was discussed by van Koppen (). It was suggested above
(footnote ) that its date may be the opening statement of Abiešuh’s year name “f ”, otherwise
only known from Date-List B, but the text has not yet been successfully deciphered (mu x x x /
an na? den-líl).

Bibliography

Bauer, T. (): Das Inschriftenwerk Assurbanipals vervollständigt und neu bearbeitet. II. Teil: Bearbeitung
(= AB NF ), Leipzig.
BBVOT  = Arnaud, D. (): Altbabylonische Rechts- und Verwaltungsurkunden aus dem Musée du
Louvre, Berlin.
BE / = Ranke, H. (): Babylonian Legal and Business Documents from the Time of the First Dynasty
of Babylon, Chiefly from Sippar, Philadelphia.
BE / = Poebel, A. (): Babylonian Legal and Business Documents from the Time of the First Dynasty
of Babylon, Chiefly from Nippur, Philadelphia.
Beaulieu, P.-A. (): The Pantheon of Uruk during the Neo-Babylonian Period (= CM ), Leiden-
Boston.
Borger, R. (): Beiträge zum Inschriftenwerk Assurbanipals Assurbanipals: die Prismenklassen A, B, C =
K, D, E, F, G, H, J und T sowie andere Inschriften (mit einem Beitrag von Andreas Fuchs), Wiesbaden.
Börker-Klähn, J. (): Untersuchungen zur altelamischen Archäologie, Berlin.
Brinkman, J.A. (): A Political History of Post-Kassite Babylonia – B.C. (= AnOr ), Rome.
Charpin, D. (): “Marchands du palais et marchands du temple à la fin de la Ire dynastie de Babylone,”
JA : –.
———. (): Le clergé d’Ur au siècle d’Hammurabi (= HEO ), Genève-Paris.
———. (): “Notices prosopographiques, : les “prévôts des marchands” de Sippar-Amnânum,” NABU
/.
———. (): “Immigrés, réfugiés et déportés en Babylonie sous Hammu-rabi et ses successeurs.” In:
Charpin, D. and Joannès, F. (Eds.), La circulation des biens, des personnes et des idées dans le Proche-
Orient ancien. Actes de la XXXVIII e RAI, Paris: –.
Collon, D. (): Catalogue of the Western Asiatic Seals in the British Museum: Cylinder Seals III,
Isin/Larsa and Old Babylonian Periods, London.
CT  = Pinches, T.G. (): Cuneiform Texts from Babylonian Tablets in the British Museum, part VI,
London.
CT  = Pinches, T.G. (): Cuneiform Texts from Babylonian Tablets in the British Museum, part VIII,
London.

79 For an unabbreviated four-participant scene of this type see e.g. Collon  no. .
 frans van koppen

CT  = Leeper, A.W.A. (): Cuneiform Texts from Babylonian Tablets in the British Museum,
part XXXV, London.
CT  = Finkelstein, J.J. (): Old Babylonian Legal Documents, London.
De Genouillac, H. (): Premières recherches archéologiques à Kich: mission d’Henri de Genouillac,
–: rapport sur les travaux et inventaires, fac-similés, dessins, photographies et plans. Fouilles
françaises d’el-" Akhymer I, Paris.
Frame, G. (): Babylonia –B.C.: A Political History, Istanbul.
Friedrich BA / = Friedrich, T. (): Altbabylonische Urkunden aus Sippara, Leipzig.
Gasche, H. (): La Babylonie au e siècle avant notre ère: approche archéologique, problèmes et
perspectives (= MHEM ), Ghent.
Glassner, J.J. (): “Chronologie élamite et chroniques mésopotamiennes,” NABU /.
———. (): “Collations susiennes,” NABU /.
Goetze, A. (): “The Year Names of Abı̄-ešuh,” JCS : –.
Grillot, F. and Glassner, J.-J (): “Problèmes˘ de succession et cumuls de pouvoirs: une querelle de
famille chez les premiers sukkalmah?” IrAnt : –.
———. (): “À propos des premiers sukkalmah. Addenda et corrigenda,” NABU /.
Harris, R. (): “Notes on the Babylonian Cloister and Hearth: A Review Article,” Or : –.
———. (): Ancient Sippar: A Demographic Study of an Old-Babylonian City (–B.C.), Istan-
bul.
HKL  = Borger, R. (): Handbuch der Keilschriftliteratur : Inhaltliche Ordnung der sumerischen und
akkadischen Texte. Anhang: Sekundärliteratur in Auswahl, Berlin-New York.
Horsnell, M.J.A. (): The Year-Names of the First Dynasty of Babylon. Vol. : Chronological matters,
the year-name system and the date-lists. Vol. : The year-names reconstructed and critically annotated
in the light of their exemplars, Hamilton.
Jursa, M. (): ““Als König Abi-ešuh gerechte Ordnung hergestellt hat”: eine bemerkenswerte altbaby-
lonische Prozessurkunde,” RA : –.
Koppen, F. van (): “Abum-waqar Overseer of the Merchants at Sippar,” NABU /.
Koppen, F. van, and Lacambre, D. (–): “Sippar and the Frontier between Ešnunna and Babylon,”
JEOL : –.
Lacambre, D. (): “La bataille de Hirîtum,” MARI : –.
Lambert. W.G. (): “Samsu-iluna in Later Tradition.” In: Tunca, Ö. (Ed.), De la Babylonie à la Syrie, en
passant par Mari: Mélanges offerts à Monsieur J.-R. Kupper à l’occasion de son e anniversaire, Liège:
–.
———. (): “The Fall of the Cassite Dynasty to the Elamites: An Historical Epic.” In: Gasche, H., Tanret,
M., Janssen, C. and Degraeve, A. (Eds.), Cinquante-deux réflexions sur le Proche-Orient ancien offertes
en hommage à Léon de Meyer (= MHEOP ), Leuven: –.
Leemans, W.F. (): “Old Babylonian Letters and Economic History: A Review Article with a Digres-
sion on Foreign Trade,” JESHO : –.
Malbran-Labat, F. (): Les inscriptions royales de Suse: Briques de l’époque paléo-élamite à l’Empire
néo-élamite, Paris.
MHET  = Dekiere, L.: Old Babylonian Real Estate Documents from Sippar in the British Museum, ():
Part : Documents from from the Reign of Hammurabi, (): Part : Documents from the Reign of
Samsu-iluna, (): Part : Post-Samsu-iluna Documents, (): Part : Documents from the Series
-- (from Zabium to Ammi-saduqa), Ghent.
˙
MDP  = Scheil, V. (): Actes juridiques susiens (suite: nº  à nº ), Paris.
MDP  = Scheil, V. (): Mélanges épigraphiques, Paris.
MDP  = Steve, M.-J. (): Nouveaux mélanges épigraphiques. Inscriptions royales de Suse et de la
Susiane, Nice.
Nissinen, M. (): References to Prophecy in Neo-Assyrian Sources (= SAAS ), Helsinki.
OETC  = Langdon, S. (): The Weld-Blundell Collection : Historical Inscriptions, Containing Princi-
pally the Chronological Prism W-B , London.
OLA  = Van Lerberghe, K. (): Old Babylonian Legal and Administrative Texts from Philadelphia,
Leuven.
Oppenheim, A.L. (): Ancient Mesopotamia: Portrait of a Dead Civilization (Revised Edition), Chi-
cago.
PBS / = Chiera, E. (): Old Babylonian Contracts, Philadelphia.
Pientka, R. (): Die spätaltbabylonische Zeit: Abiešuh bis Samsuditana: Quellen, Jahresdaten, Ge-
schichte (= Imgula ), Münster. ˘
 abiešuh, elam and ashurbanipal: new evidence from old babylonian sippar 

Postgate, J.N. (): Early Mesopotamia: Society and Economy at the Dawn of History (Paperback edition
with revisions), London-New York.
RIME  = Frayne, D.R. (): Old Babylonian Period (–BC), Toronto.
Scheil, V. (): “Kutir Nahhunte I,” RA : –.
˙ ˙ Walker, C.B.F. (): Catalogue of the Babylonian tablets in the British
Sigrist, M., Figulla, H.H., and
Museum II, London.
Steve, M.-J., Gasche, H., and De Meyer, L. (): “La Susiane au deuxième millénaire: à propos d’une
interprétation des fouilles de Suse,” IrAnt : –.
Stol, M. (): Studies in Old Babylonian History, Istanbul.
———. (): “Wirtschaft und Gesellschaft in altbabylonischer Zeit.” In: Attinger, P., Sallaberger, W. and
Wäfler, M. (Eds.), Mesopotamien: Die altbabylonische Zeit. (= OBO /), Fribourg-Göttingen: –
.
Stolper, M.W. (): “Political History.” In: Carter, E. and Stolper, M.W.: Elam: Surveys of Political History
and Archaeology (= UCP-NES ), Berkeley: –.
TCL  = Thureau-Dangin, F. (): Lettres et contracts de l’époque de la première dynastie babylonienne,
Paris.
Vallat, F. (a): “Deux inscriptions royales en élamite de l’époque des Epartides (sukkalmah),” NABU
/.
———. (b): “Réflexions sur l’époque des sukkalmah.” In: Vallat, F. (Ed.), Contribution à l’histoire de
l’Iran: Mélanges offerts à Jean Perrot, Paris: –.
———. (): “Les sukkalmah: errata,” NABU /.
———. (): “Le Kutir-Nahhunte d’Assurbanipal,” NABU /.
———. (): “L’Élam à l’époque paléo-babylonienne et ses relations avec la Mésopotamie.” In: Durand,
J.-M. (Ed.) Mari, Ébla et les Hourrites: Dix ans de travaux: Première partie: Actes du colloque interna-
tional (Paris, mai ) (= Amurru ), Paris: –.
———. (a): “Nouveaux problèmes de succession en Élam,” IrAnt : –.
———. (b): “Les trois Kuk-Našur,” NABU /.
———. (): “Temti-Agun I. Un nouveau sukkalmah,” Akkadica : –.
Van Lerberghe, K. and Voet, G. (): “A Poor Man of Sippar,” AoF : –.
Veenhof, K.R. (–): “The Sequence of the “Overseers of the Merchants” at Sippar and the Date
of the Year-Eponymy of Habil-kēnum,” JEOL : –.
VS  = Ungnad, A. (): Altbabylonische Urkunden, Leipzig.
VS  = Ungnad, A. (): Altbabylonische Urkunden, Leipzig.
VS  = Figulla, H.H. (): Altbabylonische Verträge, Leipzig.
Wilcke, C. (): “Nachlese zu A. Poebels Babylonian Legal and Business Documents From the Time
of the First Dynasty of Babylon Chiefly From Nippur (BE /), Teil I,” ZA : –.
———. (): “Nanāja-šamhats Rechtsstreit um ihre Freiheit.” In: Pongratz-Leisten, B., Kühne, H. and
Xella, P. (Eds.), Ana Šadî Labnani lū allik: Beiträge zu altorientalischen und mittelmeerischen Kulturen:
Festschrift für Wolfgang Röllig (= AOAT ), Kevelaer-Neukirchen-Vluyn: –.
YOS  = Feigin, S.I. (): Legal and Administrative Texts of the Reign of Samsu-iluna, New Haven-
London.
YOS  = Finkelstein, J.J. (): Late Old Babylonian Letters and Documents, New Haven-London.
 frans van koppen

Fig. . BM 

Fig. . BM 
abiešuh, elam and ashurbanipal: new evidence from old babylonian sippar 

Fig. . BM 

Fig. . BM 
 frans van koppen

Fig. . BM 

Fig. . Seal of Sin-iddinam

abiešuh, elam and ashurbanipal: new evidence from old babylonian sippar 

Fig. . BM  UE (Photograph by the author

reproduced with permission of the British Museum).
 SUGIRS OF ANŠAN

Matthew W. Stolper*

. Introduction

In the first volume of Texts from Tall-i Malyan (TTM I), I mentioned two Elamite administrative
documents of historical interest excavated at Malyan in  and . I cited them in
connection with the problem of dating the tablets found in and around the EDD building. I said
that passages in these texts mention at least two and perhaps three personal names following
the Elamite word su-gìr, and I commented that these passages might refer to at least one and
perhaps three otherwise unknown Elamite kings. I cited one of them (M-) a second time
in connection with the problem of the names and sequence of months in the Malyan Elamite
texts (Stolper :  and ).
Since then, these documents have been cited occasionally in connection with these and other
topics in Elamite language and history: for example, by Steve and others in connection with
Elamite political history at and after the end of the Šutrukid dynasty (Steve/Vallat/Gasche :
, Steve : , etc.); by Basello in connection with the development of Elamite calendars
(Basello : , www.elamit.net), and by Henkelman in connection with ceremonial feasting
in Achaemenid and pre-Achaemenid Iran (Henkelman : ).
I cited these texts in the expectation that more volumes of Texts from Tall-i Malyan would
follow soon to be a venue for publishing them, and in the hope that more Elamite administrative
texts would be found at Malyan to clarify their uncertainties. This expectation and this hope
were already moot by the time TTM I appeared. To allay my reservations about the value of
evidence that is cited without being displayed, I adopted interim expedients: I characterized
these and other unpublished Elamite administrative texts from the EDD excavations at Malyan
in reports to the annual Symposium on Archaeological Research in Iran (Stolper , Stolper
n.d.); and I made the MS and illustrations of those reports, along with copies and transliterations
of other Malyan Elamite texts, available to the editors of the Elamisches Wörterbuch (ElWb),
where these texts are almost completely excerpted.
The retrospective character of the present volume makes it an appropriate venue for present-
ing editions of M- and M-, general assessments of their contents in lieu of translations,
and a review of the relevance of these texts for suppositions about late and/or post-Middle
Elamite history. To call this presentation “preliminary” after more than thirty years seems ludi-
crous but appropriate, both in the sense that the absence of closely comparable material makes
confident interpretation of the particulars elusive; and in the sense that full presentation of the
remaining Elamite administrative documents from the EDD building and its vicinity is desir-
able for sharpening comment on readings in these texts. As comments on the texts will show, I
still do not understand them.1

*Oriental Institute, University of Chicago.

1I am indebted to Dennis Campbell and Wouter Henkelman for help with references, comments and corrections.
Responsibility for errors of substance and judgment remains with me.
 matthew w. stolper

. M-

.. Context and Chronology

In a stratigraphic sense, M- is the latest Elamite cuneiform tablet found in the EDD area. It
came from the fill of a bin (feature ) in the corner of an exterior courtyard (feature ) of
Level IIIA (Carter :  and  s.v. EE  Lot ). Since the walls of the building associated
with the courtyard were built to abut standing ruins of the monumental building that was
destroyed by fire at the end of Level IVA, it appears that the interval that separated IIIA, where
M- was found, from IVA, where all the other Elamite administrative tablets were found, was
short—perhaps two generations, probably not as much as a century.
The interval between the production of the Level IVA administrative tablets and the produc-
tion of M-, however, was probably shorter still. The bin where it was found and the building
associated with it are not plausible sites for storing records. M- is undoubtedly separated
from a larger original group of records. The find-spot is the result of secondary deposition, so
M- may not be much younger than the Level IVA administrative tablets. The archaeological
context provides a terminus ante quem, of course, but not a terminus ad quem. On the question
of absolute dating, see below, sub .

.. Contents
M- appears to record outlays during at least two months and part of another. It consists
of six sections: a heading (line ); a first set of four entries and a second set of three (perhaps
four) entries, each concluding with a subtotal (lines – and –); a grand total in two parts
(lines –); a date (line ); and a conclusion (lines –).
The two sets of entries (lines – and –) are similar in structure: three or four entries
and subtotal. They differ in substance: the entries of the first set are characterized by month
names, but at least two of the entries of the second set are characterized by the word sugir and
personal name. The subtotal of the first is characterized by an administrative title, teppir, and
proper name, but the subtotal of the second is characterized by a unique substantive, perhaps
indicating things “taken away” (corresponding to ZI.GA, “issued,” in the heading).
The numbers in the first set of entries are either erroneous or incomplete. The grand total
(lines –) is divided into two elements that do not correspond to the subtotals, but whose
sum equals the sum of the subtotals; that is:  (line ) +  (line ) =  =  (line ) +
 (line ) (see comments to lines  and ).
The commodity issued is not evident (see comments to lines  and ).
Interpretation of the name of the second sugir named here, Akšir-“ditto,” of course depends
on interpretation of the name of first, Šutruk(-)ša-x. All known Elamite personal names begin-
ning with Šutruk- continue with the divine name Nahhunte (or dUTU), but this restoration is
excluded by collation (see note to line ).

. M-

.. Context and Chronology

The second text cited in Stolper  as mentioning a sugir, M-, was found in the open
alley that ran along the southwestern wall of the EDD building, in a lot attributed to Level
IVA. Preserved rooms of a structure on the other side of the alley may be somewhat later than
the monumental EDD building (Carter :  and  CC  Lot ). The same may be true of
the surface of the alley, but the tablet was clearly discarded and the text may be somewhat older.
 sugirs of anšan 

.. Contents
Features suggesting that this text is a letter or letter-order (similarly ElWb  s.v. la-h-ti-
iš) include the two personal names in the first two lines, plausibly understood as naming an
addressee and sender (and contrasting with the amounts, measures and commodities that
commonly appear at the beginnings of administrative texts); a probable second-person verbal
form line  (ElWb  s.v. du-un-ti); a possible second-person verbal form in line  (ElWb 
s.v. hu-ma-ti); a possible second-person singular pronoun in line  (ElWb  s.v. ki-ni-ma-
ha); and a probable first-person possessive pronoun in line .
If this is a letter, then the apparent third-person forms lahtiš, šarraš may rather be the
imperatives expected in a letter-order, similar in form to Achaemenid Elamite halpiš ‘attack!,
kill!’, or turuš ‘speak, tell!’ (at the beginning of letters). The repeated word naš, four times
following a verbal predicate (lines , , , ) may also be an imperative, introducing
successive statements, echoing the introductory verb expected at the end of the first line, PN
turuš, ‘tell PN.’ But since it once precedes a conjunction ak, ‘and’ (line ), it seems more likely
that this naš is instead the third-person verb, ‘he says/said,’ and that it concludes successive
statements by resuming the introductory phrase expected in the second line, PN nan turuš,
‘[the sender] spoke, saying.’
In either case, the putative letter consists of a series of sentences, presented as quoted
utterances of the sender. If that is so, this is a syntactic departure from the usual marking of
quoted speech with initial nanri (plural nanpe) and final ma(n)ra (plural ma(n)ba). It is also
stylistic feature that is not found in any other known Elamite letter.
The addressee may be connected with a storehouse (lu-du, line , cf. Stolper :  and
). The text is concerned with grain (ŠE.BARMEŠ, lines  and ) and with disbursing it
(šar(r)aš, lines  and ). Of immediate interest here, the text concludes with the mention
of a sugir named Akšir-x, in hopelessly damaged context.

. M-

For the sake of completeness, I append M-, a fragmentary administrative text from Level
IVA of the EDD building. It was found with  other administrative tablets and fragments on
the floor of a room in the northeastern sector of the excavated part of the EDD building (room
, Carter :  and  FF  Lot ). Other texts found with it deal with livestock, hides,
grain, and perhaps grain products (e.g., M- = Stolper :  fig. ). An inscribed sherd
from a large storage jar was in the same room (M-, Stolper :  fig. ). The text lists
up to seven outlays to named individuals of some commodity measured by volume. Here, sugir
appears not to be a title, but the first component of a personal name.

. What Are These sugirs?

In the absence of objections, I take it that the statements in Stolper :  still stand: that su-gìr
is a spelling of an Elamite word for ‘king,’2 that the word-order title+name is unobjectionable,
and that the Malyan tablets therefore mention otherwise unattested Elamite rulers.3 The full

2 And consequently, that the reading of the logogram : (EŠŠANA) in other Malyan texts is su(n)gir.
3 I would emend the statement, however, to say that the Malyan texts name at least one and perhaps four
 matthew w. stolper

contexts of these two documents, M- and M-, offer some lexical, grammatical and
stylistic interest, but they do not help with the simplest questions of political history, namely,
when these rulers lived and what they ruled over.
Suggested responses to these questions depend in part on the dates assigned to the occupa-
tion and destruction of the EDD building. Broadly speaking, published opinions fall into two
camps—an “Anšan view,” that the building and the texts date to about ,bc and/or a century
or more earlier (e.g. Stolper : , Carter :, Potts : ); and a “Susa view,” that
the building and texts date to about ,bc and/or a century or more later (e.g. Steve :
, Steve , Steve/Vallat/Gasche :  f.).
These judgments differ little in chronological outcome and associated degrees of uncertainty.
Thus, the assessment of Pons :  ff., that the building and its contents belong to the
eleventh century bc, comes by a different route to approximately the same assessment offered
in Stolper : , that the tablets “were composed c. – bc, and most probably in the
last third of that interval.” The two views differ more in the matter of historical interpretation,
that is, on the question of how the evidence of these documents is to be located in the
course of Elamite history and culture. In the Anšan view, these documents are “late Middle
Elamite,” products of the last years and immediate aftermath of the Šutrukid monarchy. They
and the material remains with which they are associated represent an enclaved Elamite elite,
the expression of political expansion at the Middle Elamite zenith. In the Susa view, they are
“Neo-Elamite IA,” unique testimony to the first stages of developments that are illuminated
again only after centuries of obscurity, in the eighth century bc, in material from Susa and
texts from Mesopotamia. They and the associated material represent a refuge of an Elamite
monarchy that had its origins and power center in Susiana, perhaps a temporary refuge followed
by a reassertion of power at Susa (Vallat : , but cf. Potts : ), but nevertheless
inaugurating a new historical era arising in unknown political conditions.
In this respect, the Susa view depends especially on Steve’s assessment of the paleography and
syllabary of the Malyan Elamite texts (Steve :  and :  ff.). With greatest respect for
the late Père Steve’s prodigious epigraphic acumen, I cannot give much chronological weight to
this assessment, for reasons that were mostly stated in already in Stolper :
(1) In describing a single writing system with a single history, Steve treated administrative
texts and royal inscriptions as paleographic equivalents. In my view, this stretches evidence
that is incommensurate into a Procrustean system. Hallock’s critique of Poebel’s first dating
of the Persepolis Fortification tablets is apposite here. Having confirmed that the king
of the Persepolis Fortification Archive was Darius I, and not, as Poebel had supposed,
Artaxerxes I, Hallock noted Poebel’s statement that the sign forms of the Persepolis tablets
are intermediate between those of Darius I and Artaxerxes II, and responded that “The
comparison is, of course, between the cursive forms of the tablets and the monumental
forms of the royal inscriptions. It does not occasion any surprise to discover that the
cursive writing shows later forms than the contemporary monumental writing” (Hallock
: ).
This is an even greater problem for Steve’s demonstration of the steadily increasing number
of logograms in the attested Elamite scripts. The relative paucity of logograms in Middle
Elamite and the relative frequency of logograms in Neo- and Achaemenid Elamite do not

otherwise unknown kings, to include the king Hu-[…] named in TTM I :, who may well not be Hu[teluduš-
Inšušin]ak, despite the confident acceptance of this conjecture in, e.g. ElWb  s.v. v.hu-te.lu-du-uš.d.in-su-uš-na-ak;
Carter : , etc.
 sugirs of anšan 

arise from changes in the writing system. They are consequences of differences in con-
tent and style between the Middle Elamite corpus, which is overwhelmingly dominated
by royal inscriptions, and the Neo- and Achaemenid Elamite corpora, which are over-
whelmingly dominated by administrative texts.
(2) Steve minimized the likelihood of regional variation. On the contrary, Steve considered it
likely that the difference between the sign forms of the Malyan administrative texts and the
Šutrukid royal inscriptions reflects contact with Assyria. The Malyan administrative texts
do offer some grounds for conjecture about Mesopotamian contact, but I do not think the
sign forms are among them.
(3) Steve did not deal with the closest comparanda, fragmentary Elamite tablets that are
more likely to have come from the vicinity of Susa than from the vicinity of Anšan. In
fact, Steve acknowledged this difficulty when he remarked that the fragments published
in Walker  were probably to be attributed to the time of the Šutrukids, although
in graphic terms they were not far removed from the Malyan documents (Steve :
).
I have similar reservations about Steve’s cultural interpretation Elamite paleography. The signs
that Steve mentioned as the first steps in a sequential morphological change that continued
gradually but beyond the view of evidence now available until it emerged in eighth-century
Neo-Elamite seem to me to be simply comparable to Mesopotamian sign forms—in some
cases, resembling Middle Babylonian forms more than Middle Assyrian, and in other cases
non-distinctive. There are surely differences between Mesopotamian and Susian and Elamite
scribal training and traditions (e.g., Reiner : , Rutz : ff.), and probably regional
differences within Elamite territories (just as there were notable differences in practice among
those who wrote Persepolis Fortification tablets in regions near Persepolis and in regions to
the northwest of Persepolis, closer to Susa), but (in my view) the sign forms of Middle and
early Neo-Elamite texts, whether royal inscriptions or practical texts, are generally comparable
to contemporary Mesopotamian sign forms in texts of similar kinds. They do not show clear
antecedents of the sharp, systematic changes in sign forms and the innovative sign values
that appear in Elamite texts after the mid-seventh century, which do not have contemporary
Mesopotamian counterparts (Stolper : –).
Of some interest for gauging this late paleographic change are two “Elamite” MSS of Syl-
labary A (Hallock : ff.). Dating them depends on a combination of provenance,
paleographic assessment, and historical conjecture. Hallock says of one MS, from Küyün-
jik (CT  pl. x Sm+) that it “is evidently a trophy of the Elamite wars of Assurbani-
pal and dates to his time or somewhat earlier.” Of the other, from Susa (MDP  ), he
says that it “would seem to belong approximately to the same period as” the first. Steve
cites this dating without objection, but interprets these lists—which include many signs not
attested in Neo-Elamite texts and inscriptions—as meant to serve Babylonians of Susiana
or Elamite Akkadophones (Steve : ). Noteworthy, however, is the close resemblance
between the form of the sign LÚ in the “Elamite” syllabary (CT  x Sm+ rev. iv ) and
the form of LÚ illustrated in Steve :  No.  NIIIB (cited from Lambert :
 Sb :) as the sole Neo-Elamite IIIB (post-Assyrian, pre-Achaemenid) example.
These examples of LÚ are manifestly intermediate in shape between Neo-Babylonian and
Achaemenid Elamite forms. They answer Steve’s observation () that Hallock identified
Achaemenid Elamite LÚ without making a statement about a possible derivation of the form.
They thus eliminate the need for Steve’s reading of the Achaemenid Elamite logogram for
“man” as SIG7/SA7, graphically unmotivated and based on accidental convergence with a
Neo-Assyrian sign-form.
 matthew w. stolper

To my eye, at least, post-seventh century Elamite sign forms are not the outcome of a long
but largely undocumented development. They depart sharply from earlier forms, and in the
case of LÚ and a few other signs (e.g., MA) the ancestral forms are Babylonian rather than
Assyrian. Discontinuities in Elamite paleography do not remedy discontinuities in known
Elamite political history. Our sugirs remain historically rootless.
So when and over what did these sugirs rule?
In the case of the Akšir-x of M-, he was probably king when the tablet was written. This
seemingly self-evident remark anticipates the objection that in legal and administrative texts
it is not usual to refer to reigning kings by name, except in connection with dates, oaths and
such. For this reason, it took some years to determine the date of the Persepolis Fortification
tablets (Poebel : ff., Cameron : , Hallock :  ff.), which refer often to “the
king,” but only rarely to Darius, and almost exclusively in letters and letter-orders.4 If Akšir-x
was approximately contemporary with the tablet, he was approximately contemporary with,
perhaps slightly later than, the Level IVA building and the documents in it.
The situation of Šutruk-ša-x and Akšir-“ditto” in M- is even less clear. If they are the
sources of outlays, then they ruled when the text was written. They ruled simultaneously, and
for that reason they are cited by name, and not just by title. In that case they were not successors
to the Šutrukid kingship of Anšan and Susa, but subordinate or local rulers, perhaps two
among a larger number of such rulers, as Steve/Vallat/Gasche :  suggest.5 In that case
what kind of post-Šutrukid political change led to two simultaneous “kings” with simultaneous
responsibility or authority at Anšan?
On the other hand, if these sugirs are not the sources but the recipients of outlays, then
they were perhaps deceased kings, whether from the recent past in the shadows of post-
Šutrukid Elamite history, or from the distant past, in the almost total obscurity of Old Anšanite
history.

. Conclusion

When I encountered these sugirs thirty-five years ago, I was thrilled by the prospect that a
new chapter of Elamite history was about to open. The years have not brought me clarity or
conviction and the chapter is still unopened.
All too much of Elamite political history consists of nothing more than the names and
epithets of kings, eked out by the surmises of modern historians. This is an extreme case. There
were three kings at Anšan, or maybe four. They were named Akšir-something, Hu-something,
and Šutruk-something. One, and maybe others, ruled—perhaps—around ,bc, give or take
a hundred years or so—but some of them might have ruled earlier. And what they ruled, I
cannot say.

4 PF  (in the last clause, immediately preceding the date); PF  (in the salutation); Fort.  = JNES
 , republished by Arfaee :  ff. (twice, citing royal orders being forwarded); PF-NN  (citing a royal
order being forwarded). The exception, PF-NN , a text of category D, cites the royal name and title in the date
at the end of the text.
5 But no weight can be given to their suggestion that the absence of the epithet “king of Anshan and Susa” from

these damaged administrative contexts is meaningful for Elamite political history. It means no more than does the
absence of the epithets “great king, king of kings” from the Persepolis administrative texts.
 sugirs of anšan 

Abbreviations

EKI König .

ElWb Hinz and Koch .
M Malyan (excavation number).
Mf Malyan find (registration number).
PF Persepolis Fortification texts published in Hallock .
PF-NN unpublished Persepolis Fortification texts cited from draft editions by Richard T. Hallock
and collated by Wouter F.M. Henkelman.
TTM I Stolper .

Bibliography

Arfaee, A. (): Persepolis Fortification Tablets: Fort. and Teh. Texts. (= Ancient Iranian Studies Series
), Tehran.
Basello, G.P. (): “Elam and Babylonia: the Evidence of the Calendars.” In Panaino, A. and Pettinato,
G. (Eds.), Ideologies as Intercultural Phenomena. Proceedings of the Third Annual Symposium of the
Assyrian and Babylonian Intellectual Heritage Project held in Chicago, USA, October –,  (=
Melammu Symposia ), Milan: –.
Brosius, M. (): “Reconstructing an Archive: Account and Journal Texts from Persepolis.” In Brosius,
M. (Ed.), Ancient Archives and Archival Traditions: Concepts of Record-Keeping in the Ancient World,
Oxford: –.
Cameron, G.G. (): “Darius’s Daughter and the Persepolis Inscriptions,” JNES : –.
Carter, E. (): Excavations at Anšan (Tal-e Malyan): the Middle Elamite Period. Malyan Excavation
Reports  (= UMM ), Philadelphia.
Giovinazzo, G. (): “na-áš dans les tablettes de Persépolis,” NABU /.
———. (): “L’Expression “ha duš ha duka” dans les texts de Persépolis,” Akkadica : –.
Hallock, R.T. (): “Darius I, the King of the Persepolis Tablets,” JNES : –.
———. (): “Two Elamite Texts of Syllabary A,” JNES : –.
———. (): Persepolis Fortification Tablets (= OIP ), Chicago.
Henkelman, W.F.M. (): The Other Gods Who Are: Studies in Elamite-Iranian Acculturation based on
the Persepolis Fortification Texts (= Achaemenid History ), Leiden.
———. () “Parnaka’s Feast: šip in Parsa and Elam.” In Álvarez-Mon, J. and Garrison, M.B. (Eds.), Elam
and Persia, Winona Lake: –.
Hinz, W. and Koch, H. () Elamisches Wörterbuch ( = AMI Ergänzungsband ), Berlin.
Jones, C.E. and Stolper, M.W. (): “How Many Persepolis Fortification Tablets Are There?” In Briant,
P. et al. (Eds.), Les Archives des Fortifications de Persépolis, État des questions et perspectives de recherches
(= Persika ), Paris: –.
Khačikjan, M. (): The Elamite Language (= DA ), Rome.
König, F.W. (): Die elamischen Königsinschriften (= AfO Beiheft ), Graz.
Lambert, M, (): “Deux texts élamites de la fin du septième siècle,” JA : –.
Poebel, A. (): “The King of the Persepolis Tablets. The Nineteenth Year of Artaxerxes I,” AJSLL :
–.
Pons, N. (): “Tchogha Zanbil après Untaš-Napiriša.” In Gasche, H., Tanret, M., Janssen, C. and
Degraeve, A. (Eds.), Cinquante-deux réflexions sur le Proche-Orient ancien offertes en homage à Léon
De Meyer ( = MHEO ), Leuven: –.
Potts, D.T. () The Archaeology of Elam. Formation and Transformation of an Ancient Iranian State,
Cambridge.
Reiner, E. (): “The Elamite Language,” in Friedrich, J., Reiner, E., Kammenhuber, A., Neumann, G.,
and Heubeck, A., Altkleinasiatische Sprachen (= HdO /), Leiden: –.
Rutz, M.T. (): “Textual Transmission between Babylonia and Susa: A New Solar Omen Com-
pendium,” JCS : –.
Steve M.-J. (): “La Fin de l’Élam: à propos d’une empreinte de sceau-cylindre,” StIr : –.
———. (): “Le signe /ruh/ du syllabaire néo-élamite,” NABU /.
———. (): Syllabaire Élamite: Histoire et Paléographie ( = Civilisations du Proche-Orient Série II,
Philologie ), Neuchâtel-Paris.
 matthew w. stolper

Steve, M.-J., Vallat, F. and Gasche. H. (): “Suse,” Suppl. DB : –.
Stolper, M.W. (): “Preliminary Report on Texts from Tal-e Malyān –.” In Bagherzadeh,
F. (Ed.), Proceedings of the IVth International Symposium on Archaeological Research in Iran, rd–th
November, , Tehran: –.
———. (): Texts from Tall-i Malyan, I: Elamite Administrative Texts (–) (= OPBF ), Philadel-
phia.
———. (): “Elamite.” In Woodard, R.D. (Ed.), The Cambridge Encyclopedia of the World’s Ancient
Languages, Cambridge: –.
———. (n.d.) “Malyān EDD : Texts.” In Bagherzadeh, F. (Ed.), Proceedings of the VIth International
Symposium on Archaeological Research in Iran (Tehran, ).
Vallat, F. (): “L’expression “E.V.” (En Ville) à l’époque néo-élamite,” NABU /.
———. (): “Le retour de Hutelutuš-Insušnak à Suse,” NABU /.
———. (): “L’Utilisation des sceaux-cylindres dans l’archivage des lettres de Persépolis.” In Gyselen,
R. (Ed.), Sceaux d’Orient et leur emploi (= Res Orientales ), Bures-sur-Yvette: –.
Walker, C.B.F. (): “Elamite Inscriptions in the British Museum,” Iran : –.
Zadok, R. (): “A Tentative Structural Analysis of Elamite Hypocoristica,” BN NF : –.
———. (): The Elamite Onomasticon (= Suppl.  agli Annali /), Naples.
TEXTS
 matthew w. stolper

M- (mf.)

Obverse
() na?-ak-ka4? AN.NIMMEŠ? ZI.[GA ]
AŠ
()  ITI si-ba -ri [ ]
() +! AŠ
ITI še- ru -um [ ]
AŠ
()  ITI še-ru-um x [ ]
() [(x)]+ ITI gam-ma-ma AŠUD..KAM
AŠ
[ ]
() PAP  ME + te-ep-pír DIŠas-si-za-ah [ ]
() x(??) mu-ša-pír [ ]
() []+ su-gìr šu-ut-ru-uk(-)ša x [ ]
Lower Edge
()  su-gìr ak-šir8-KI+MIN ša x [ ]
Reverse
()  ur-ma-ak-ku x [ ]
() [ ]
() PAP + hi-il-la-hi-la [ ]
() PAP PAP  ME  + na-ak- ka4 AN.NIM [MEŠ ]
() ZI.GA +  te-ep [-pír ]
() AŠITI gam-ma -ma AŠUD..[KAM ]
() a-ak [x] x (-)šu-uk?- ka? [ ]
Upper Edge
() li x [ ] su
 sugirs of anšan 

University of Pennsylvania Museum of Anthropology and

Archaeology. Babylonian Collection --–. Collated April .
 matthew w. stolper

No seal impression.
1, 13f. ElWb  cites my reading reading na-ak-qa-an-tum4MEŠ, and my conjecture that this
represents an Akkadian word, nakkantu, ‘storehouse.’ I am reluctant to explain new or
unknown items of Elamite by recourse Mesopotamian languages, but occurrences of
likely Akkadian words (e.g., ribut, ‘quarter-shekel,’ anaku ‘tin’, see Stolper : ), and
likely Kulturwörter shared with Mesopotamian languages (e.g., zabar ‘copper/bronze’,
basbas ‘duck’) in Malyan administrative texts permit conjectures of this kind. Never-
theless, a borrowed term of such general meaning, ‘storehouse’ or even (as ElWb sug-
gests) ‘stores,’ is far less plausible than borrowed or shared terms for objects or materials.
Furthermore, at the head of an administrative text, at the beginning and end of a list,
what is expected is a term for a specific item or material that can be counted or mea-
sured.
An alternative reading, AŠUD ak-ka4 DINGIR NIMMEŠ is equally conjectural and still more
problematic. Interpreting the phrase as ‘days of gods of Elam’ (or: ‘Elamite gods’) presumes
that DINGIR NIMMEŠ represents a phrase like nappip hatamtip, found in passages of the
great stele of Šilhak-Inšušinak I that invoke ‘gods of Elam, gods of Anšan, gods of Susa’
(EKI  §§ and ; or like napir hatamtir, ‘god of Elam,’ as in inscriptions of Šilhak-
Inšušinak and Huteluduš-Inšušinak referring to Šimut (EKI  III and  IV; cf. ElWb
 s.v. te-ip-pír). Reading ‘days of god(s)’ here implies that more than one god at a time
is involved, to allow for the  god-days of month Šerum in line .
This interpretation further presumes that the Elamite relative pronoun akka forms a
possessive-attributive construction in the same way that the particle ša does in many Level
IVA texts and in a few other pre-Achaemenid Elamite texts (see Stolper : ). That is,
it presumes that akka is not only relative, but also “determinative,” as that term is used
to describe the function of Akkadian ša; that akka and ša are near-synonyms (see Stolper
:  and Walker : ). Indeed, determinative usage of ša survived in Achaemenid
Elamite alongside relative-determinative usage of relative pronouns: ANITIMEŠ Šermu ša bel
-ummemanna, “month X of the st year,” PF-NN : vs. ANITIMEŠ appa MN-nama
PF-NN :, bel appa NN-ummena (e.g., PF :, PF-NN :, :, :,
:), see Stolper :.
However, this interpretation presumes still further that AŠUD, representing Elamite na(n),
‘day,’ is grammatically animate, construed with the animate relative pronoun akka, rather
than inanimate appa. Despite frequent writings of na(n) ‘day,’ bel ‘year’ and ITI, ‘month’
with the divine determinative AN (including writings in Achaemenid Elamite, but not at
Malyan), the Achaemenid Elamite passages just cited indicate that ‘month’ and ‘year’ are
inanimate, construed with inanimate appa, and ‘day’ is likely to be inanimate, too.
Hence, it seems to me most likely that the text begins by naming the item issued, but this
leaves the reading uncertain and the translation unknown.
1, 14. ZI.GA, ‘issued,’ perhaps for Elamite šarraka (see Stolper : ), occurs in Level
IVA administrative texts, but apparently nowhere else in pre-Achaemenid Elamite (ElWb
).
4. ElWb  s.v. še-ru-um reads še-ru-um.D[IRIG] and infers from this restoration that
Šerum, though homonymous with Achaemenid Elamite month X, was month VI at
Malyan. Basello (on www.elamit.net) suggests that Šerum was month IX at Malyan, but
also notes the reading DIRIG as a certain occurrence, evidently accepting the possibility of
an intercalary ninth month. I consider both the restoration of D[IRIG] and the assumption
of intercalation at months other than VI and XII to be excessively generous. Despite
Stolper : , seeing “an additional month within the sequence,” I now find it simpler
to suppose that line  refers to a second set of outlays in the same month Šerum.
 sugirs of anšan 

5. The total in line  urges reading  + ! in line , and []+! in line , but collation does
not support either reading. There appears to be insufficient space at the beginning of this
line for the additional numeral. Judging by the usage of the Level IVA texts (otherwise
paleographically similar to this text), the digit  was ordinarily written with three rows of
wedges (++), not two, as the numeral is written here (hence not !+!).
6. This total is consistent with those in lines – (i.e.,  +  =  + ), but inconsistent
with the preceding entries as they appear on the tablet (i.e.  +  +  +  (or: []) = 
(or: )).
7. I cannot read, interpret or even segment the contents of this line with confidence, nor
locate them in structure of the text. The line may be part of the first subtotal, or it may be
a subheading introducing the second set of entries.
8. Collation after the tablet was baked and cleaned shows that the sign following šu-ut-ru-uk-
is identical with ša in Level IVA texts. It cannot be nah (despite the cautious suggestion
in Stolper :  and the confident restoration as šu-ut-ru-uk.nah-hu-[un-te] in ElWb
).
12. ElWb reads hi-du-me ‘Schafe als Sammelbegriff,’ that is, a noun formed from hidu+abstract
formative -me ( s.v. hi-du-me,  s.v. šu-hi,  s.v. ur-ma-ak-ku). This reading is
encouraged by the appearance of hidu in the certain meaning ‘ewe’ in administrative texts
from Level IVA (ElWb ), but not by the actual appearance of the signs on the tablet.
I prefer to interpret hillahila as a substantive cognate with the verb hilla-, ‘take away’ (or
similar), on the model of the substantive huthut and verb hutta-. This reading, also to be
found in ElWb , is encouraged by the appearance of the verb hilla-, well attested in
Middle Elamite inscriptions, in administrative use in administrative texts from Level IVA
(ElWb  s.v. hi-il-la-áš).
16. The sign following the break is identical to the sign following šu-ut-ru- in line , hence
-uk?-. If it is rather to be read -up-, then the conclusion perhaps refers to the ceremonial
or religious observance called šup in Middle and Neo-Elamite, šip in Achaemenid Elamite
(so ElWb  see also Henkelman : ).
 matthew w. stolper

M-

Obverse
() DIŠ[(x)]-za-ba-la-ni- ba? [ ]
() DIŠku-uk-ANman-za- at [ ]
() lu-du a-ni ki? [ ]
() la-ah-ti-iš a- ni? [ ]
() du-un-ti na-áš x [ ]
() la-ah-ti-iš x [ ]
() pi- x -ni-ia [ ]
Lower Edge
() ŠE.BARMEŠ x (x) um? -mi x [ ]
() hu-ma-ti x (-)te-nu-mi(-) x [ ]
Reverse
() ša-ra!-áš na-áš! a-ak ku? -um-[ ]
() du-hi-ma šar-ra-áš na-áš ir [ ]
() nu ki-ni?-ma-ha na-áš  ANŠE? [ ]
() DIŠku!-ma-ma-na  BÁN x -man-na [ ]
() ŠE. BAR MEŠ ú-mi-ni be? -lu a-ak [ ]
() pu-ru si-ma-gi x [ ]
()  ŠE.BARMEŠ hi?-ma?  x [ ]
Upper Edge
() su-gìr ak-šir8- x [ ]
() x x [ ]

National Museum of Iran, Tehran

 sugirs of anšan 

No seal impression.
5, 10, 11, 12. Cf. text-final na-áš in Persepolis Fortification texts.
The proposition of Giovinazzo , that what Hallock read as two signs in these passages
should be read as a single sign, mur, is not supported by collation. In PF  and , text-
final na-áš is clearly distinguished from the sign HAR in line . The locative use of mur …
mur, ‘in one place … in another place,’ may be adequately translated in French as ‘ici … là,’
but that is not a warrant for interpreting the Elamite word as a deictic locative adverb. If
it were a locative but lacked deictic contrast between ‘here’ and ‘there’ (so Vallat  and
Giovinazzo : ) then it is scarcely meaningful as an Elamite counterpart to French
‘En Ville’ (i.e., ‘here,’ not ‘there’). Understanding the text-final expression with a meaning
of this sort—‘recorded for local use only’—is inconsistent with the consensus view of the
flow of documents into the Persepolis Fortification Archive, namely, that most records of
‘memorandum’ form (Categories A–S) were drawn up in the region around Persepolis,
then brought to Persepolis to be compiled in texts of ‘register’ form (Categories V and W;
e.g., Jones and Stolper : f., Henkelman :  ff.). That is, documents marked
as being for local use only are found at Persepolis, not where they were drawn up. (Vallat
 rejects this view of information flow at Persepolis, but some features of the argument
suggest that it was meant as an academic satire or parody.)
It seems far more likely to me that text-final na-áš, whether or not it is to be interpreted
(with Hallock) as ‘he said(?),’ marks information that relied on an oral source, without reg-
ular accompanying documents. Assorted passages in Fortification texts demonstrate that
outwardly fastidious written records sometimes had to be supplemented with oral infor-
mation, fastidiously identified as such (mostly as direct speech marked with quotational
correlatives nanri/nanbi … mara/ma(n)ba; cf. Brosius : ff.).
6. ElWb , s.v. su-gìr, reading with apparent confidence ak-sir.ŠIMUT (i.e., MAN),
presuming a theophoric element written without the determinative AN that is otherwise
regular in the spelling of theophoric names in Malyan texts: with theophoric element
in second position, Danna(n)-dPinigir, TTM I :, :, :0, […]-tir-dHuban TTM I
:0, and Kuk-dManzat M-:, below; and cf. Stolper  glossary s.vv. dHuban-
mirriš, dIGI+DU-unukaš. Possible exceptions are Šušnakiul, q.v. and Hutra[n?-…], M-
: (unpublished). I omit from consideration names in which common nouns (temti-,
kiri-, zana- etc.) take the place of theophoric elements. On the first element of the name,
I can add nothing to comments in Stolper : , Zadok :, ElWb .
19. Conceivably ši-ra [-áš], for šeraš, hence “king Akšir-x orders/ordered”?
 matthew w. stolper

M- (mf.)

Obverse
() TE?.EN?MEŠ [ ] x
() x PI hu- un -[ ]
()  BÁN ú-ka4-[ ]
()  BÁN ak-šir8
()  BÁN su-gìr [- ] x it
()  BÁN ku-x-x -ri
Lower Edge
() [ ] x -ban-na
Reverse
() [ ]  QA? x
() PAP x [PI x BÁN]  QA TE?. EN? MEŠ
ZI.GA

National Museum of Iran, Tehran

 sugirs of anšan 

No seal impression.
2ff. Personal names appear without Personenkeil in some other texts from room , as they
often do in the group of texts published in TTM I.
5. Cf. Zadok :  for names with sunkir-, and Malyan Elamite names Sunkiki (Stolper
: ), and Sunkuku [unpublished], presumably hypocoristics of names with initial
Sunki- (Zadok : f.).
 APPROCHE HISTORIQUE ET PHILOLOGIQUE
DU TITRE ROYAL ‘LIKAME/WE RIŠAKKI’

Stéphanie Anthonioz and Florence Malbran-Labat*

La titulature des souverains de l’époque méso-élamite, rédigée désormais en langue verna-

culaire, se présente selon trois axes essentiels : le familial avec les titres šak, ruhu šak, ou šak
hanik suivi d’un nom royal, le religieux avec le titre libak hanik suivi d’ un nom divin. La
titulature politique, présente, quant à elle, deux formes différentes : l’ une a pour principal ou
unique élément l’expression sunkik Anzan-Šušun.ka, l’ autre, plus complexe, débute par likume
rišakka (ou une variante de cette formule telles likame rišakki ou likame rišari). Si la première
(«roi d’Anzan et de Suse») fait clairement référence à la royauté qui associe le Haut et le Bas
pays élamites, la seconde est plus délicate à interpréter. Cette dernière est portée d’ abord par
Humbanumena, roi de la dynastie igihalkide dans plusieurs inscriptions trouvées à Bender-
Bushir (l’antique Liyan) ainsi qu’à Suse, qui présentent la formulation suivante : Humbanumena
šak Attar-kitah.kik likume rišakka merrik hatamtik katri hatamtik halmenik hatamtik sunkik
Anzan-Šušunka1. Elle est portée par certains Šutrukides, notamment par le dernier d’ entre
eux, Hutelutuš-Inšušinak: Hutelutuš-Inšušinak likame rišari menir hatamtir ak Šušenri šak
hanik Kutir-Nahhuntir ak Šilhak-Inšušinak.ri / šak hanik Šutruk-Nahhuntir Kutir-Nahhuntir ak
Šilhak-Inšušinak.ri2. Nous ne commenterons pas ici l’ apparition de la désignation géo-politique
hatamtir ak Šušenri ni la curieuse expression de filiation qui fait de Hutelutuš-Inšušinak le
«fils3 aimé de Kutir-Nahhunte et de Šilhak-Inšušinak, voire aussi de Šutruk-Nahhunte » ; nous
nous bornerons à examiner le titre likume rišakka (et ses variantes). Celui-là, généralement
traduit par «agrandisseur du royaume», a été mis en parallèle avec une des épithètes des rois
médio-assyriens. En effet, cette période du milieu du xivème siècle au xiième siècle correspond, en
Mésopotamie, à la montée de la puissance assyrienne et l’ épithète « agrandisseur / élargisseur
de la frontière et du territoire» (murappiš misri u kudurri) est couramment portée par les souve-
˙ entre likame rišakki et murappiš misri u kudurri
rains d’Aššur. Le parallélisme qui semble exister
paraît conforter la traduction du titre élamite par « agrandisseur du royaume ». Pourtant˙ cette
traduction ne nous semble s’imposer ni sur le plan historique, ni sur le plan philologique mal-
gré des prises de positions parfois vives (et nous voudrions revenir sur la possible signification
de ce titre)4.
Nous commencerons par examiner le contexte historique (I), d’ une part, le parallèle akka-
dien murappiš misri u kudurri possible et, d’ autre part, l’ emploi du titre élamite, avant que d’ en
˙
présenter une analyse philologique (II).

* Faculté de Théologie de Lille & Institut Catholique de Paris.

1 Malbran-Labat ( : nº ).
2 Malbran-Labat ( : nos –).
3 Ou « le descendant ».
4 Voir entre autres l’ âpre débat entre Vallat (: nº ) et de Meyer (: –) contre Quintana (:

nº  et  : nº ).
 stéphanie anthonioz and florence malbran-labat

. Analyse historique

.. L’épithète akkadienne murappiš misri u kudurri

˙
Un tableau schématique5 permet de visualiser les occurrences élamites et assyriennes et de noter
quelques synchronismes babyloniens :

AŠŠUR-UBALLIT I BURNABURIAŠ II HUMBANUMENA

– ˙ – –
mu-ra-piš mi-is-ri ù ku-du-ri ú hu-ban-nu-me-na (…) li-ku-me
˙ ri-ša-ak-ka4
ENLIL-NÊRÂRÎ UNTAŠ-NAPIRIŠA
– –
mu-ra-piš mì-is-ri ù ku-du-ri
˙
ARIK-DÊN-ILI
–
mu-ra-piš mì-is-ri ù ku-du-ri
˙
ADAD-NÊRÂRÎ I
–
mu-ra-piš mì-is-ri ù ku-du-ri
˙
TUKULTÎ-NINURTA I ADAD-ŠUMA-IDDINA KIDIN-HUTRAN III
– – –
mu-re-piš mì-is-ri
˙
MELI-SHIPAK ŠUTRUK-NAHHUNTE
– –
li-ku-mi ri-ša-[ak-ki]
ZABABA-ŠUMA-IDDINA li-ku-me ri-ša-ak-ka4

ENLIL-NÂDIN-AHI KUTIR-NAHHUNTE II
– –
ŠILHAK-INŠUŠINAK
–
li-ka4-me ri-ša-ak-ki
TIGLATH-PHALAZAR I NABUCHODONOSOR I HUTELUTUŠ-INŠUŠINAK
– – –
li-kà-we ri-ša-ak-ki
li-ka4-me ri-ša-ri

L’épithète murappiš misri u kudurri est bien attestée pour le règne de Tukulti-Ninurta Ier, mais
aussi des rois qui l’ont ˙précédé: Aššur-uballit Ier (–), Enlil-nêrârî (–), Arik-
˙
dên-ili (–) et Adad-nêrârî Ier (–). Elle est caractéristique de la théologie
d’Aššur, qui repose sur le commandement impérieux adressé au roi d’ étendre les frontières
comme en témoigne le rituel du couronnement, daté du règne de Tukultî-Ninurta Ier : « de
ton sceptre juste, élargis ton pays!» (i-na e-šar-te giš.pa-ka kur-ka ra-piš)6. Cette qualification

5 Notre propos n’ est pas ici de revenir sur les datations précises. Ce n’est qu’à titre indicatif que nous indiquons

dans ce tableau les dates proposées dans le Dictionnaire de la civilisation mésopotamienne (p.  pour les rois médio-
assyriens, p.  pour les souverains cassites et p.  pour les Igihalkides et les Šutrukides).
6 Cité par Garelli ( : ).
 approche historique et philologique du titre royal ‘likame/we rišakki’ 

est classique dans le corpus des inscriptions royales où, cependant, elle ne représente qu’ une
épithète parmi d’autres, qui vient conclure l’ évocation des victoires et des conquêtes, ce
qu’illustre cette inscription d’Adad-nêrârî Ier :
. Adad-nêrârî, noble prince, orgueil des dieux,
. Seigneur, responsable devant les dieux, qui établit les sanctuaires,
. Qui frappe les héroïques, l’armée des Cassites,
. Des Qutéens, Luluméens et Šubaréens,
. Qui disperse tous les ennemis d’en haut
. Et d’en bas, qui écrase leur pays,
(…)
. Qui élargit la frontière et le territoire, roi aux pieds duquel
. Anu, Aššur, Šamaš, Adad et Ištar
. Ont fait se soumettre l’ensemble des rois et des princes7.

La titulature se poursuit avec les conquêtes de chacun de ses pères murappiš misri u kudurri,
«qui a élargi frontière et territoire». Ainsi, en tant qu’ « élargisseur des frontières˙ », il se place
dans la continuité de ses pères, Aššur-uballit Ier, Enlil-nêrârî et Arik-dên-ili. Il ne s’ agit donc
˙
pas d’un titre royal si l’on distingue titre et épithète8
: titre désignant un rang, une dignité, une
fonction; épithète, une qualification louangeuse donnée à quelqu’ un.

Mais l’origine de l’épithète n’est peut-être pas aussi assyrienne qu’ elle le paraît : une inscription
babylonienne de Samsu-iluna proclame que ce dernier est bēlum murappiš mātim, « seigneur
qui élargit le pays»9. Se peut-il alors qu’à une époque où la Babylonie s’ éclipse, les premiers
grands rois assyriens aient repris le flambeau de la royauté descendue du ciel et façonné leur
discours en reprenant des éléments du passé, s’ inscrivant ainsi dans une continuité glorieuse ?
Pour l’époque paléo-babylonienne, on retrouve d’ ailleurs cette expression, dans une forme non
figée, utilisée en Mésopotamie, à Ešnunna (murappiš Ešnunna, « qui a élargi Ešnunna »10), à
Mari (mātı̄ urappiš, «j’ai agrandi mon pays »11) et à Babylone (murappiš mimma šumšu ana
Mešlam, «qui a tout accru pour le temple Mešlam (de Nergal) »12), tandis qu’ en Élam, à l’ époque
des Sukkalmah, Siwe-palar-huppak, le seul de ces dynastes à avoir produit des inscriptions
en élamite porte une titulature comportant l’ expression ligawe rišakki ce qui a d’ ailleurs pu
conduire à mettre en doute leur authenticité13. Ces différentes attestations replacent donc
l’épithète assyrienne et le titre élamite dans une certaine continuité et ancienneté, l’ un comme
l’autre, mais n’établissent pas nécessairement un lien entre eux.

.. L’emploi du titre élamite

En ce qui concerne le titre politique élamite, essentiellement à l’ époque méso-élamite, il est
employé en alternance ou conjointement avec le titre de « roi d’ Anzan et de Suse », au sein de
titulatures plus ou moins complexes. Cette alternance ne s’ explique pas par des différences dans
la politique extérieure menée par les Igihalkides puis les Šutrukides. Le titre de likume rišakka
est porté, entre autres, par deux souverains, Humbanumena et Hutelutuš-Inšušinak, comme on

7 D’ après Grayson ( : A..., A..., A...–, A..., A...).

8 Seux ( : –), reprenant la définition du dictionnaire P. Robert.
9 Frayne ( : ).
10 Seux ( : ).
11 Thureau-Dangin ( : –). Voir aussi différentes traductions: Kupper (: –); Durand (:

–).
12 Seux ( : ).
13 Vallat ( : –) ; Reiner ( : ).
 stéphanie anthonioz and florence malbran-labat

l’a vu, qui ne se caractérisent pas spécialement par leur politique expansionniste. Au contraire,
ne portent pas le titre Untaš-Napiriša et Kutir-Nahhunte qui furent des souverains entrepre-
nants et agressifs à l’égard de la Mésopotamie, guerroyant le plus souvent avec succès contre la
Babylonie. Tous furent des souverains puissants ; mais cette puissance repose-t-elle, dans leur
conception, sur l’extension du royaume ? L’ essentiel des inscriptions que nous connaissons sont
des inscriptions de construction, c’est dire que ce sont surtout leurs réalisations cultuelles et non
militaires qu’ ils exaltent dans leurs écrits. Font exceptions les inscriptions de Šutruk-Nahhunte,
qui commémorent les prises mésopotamiennes célèbres (comme la stèle de Narâm-Sîn) et qui
sont aussi les seules à présenter une titulature très complexe, comportant à la fois sunkik Anzan-
Šušunka et likume rišakka14.

Reprenons brièvement les différents grands règnes pour essayer de comprendre la fonction du
titre likume rišakka dans ses différents contextes géo-politiques.
Siwe-palar-huppak est l’un des premiers Sukkalmah. D’ après les inscriptions de Šilhak-
Inšušinak, il est «fils de la sœur» de Siruktuh15 à l’ instar de ses inscriptions. Par les inscriptions
mésopotamiennes et mariotes, nous le savons contemporain des règnes d’ Hammurabi de
Babylone et de Zimri-Lîm: le pouvoir élamite continue de s’ étendre vers le nord à travers le
Zagros et en Mésopotamie le long de la Diyala. Il prend possession d’ Ešnunna. Mais, encore
une fois, si l’inscription élamite de Siwe-palar-huppak est authentique, il faut souligner qu’ il y
dédie ses offrandes à Inšušinak «pour la vie » de sa famille et de sa descendance16. L’ emploi du
titre semble en lien avec un intérêt familial et dynastique beaucoup plus qu’ expansionniste.
On sait peu de choses sur le règne de Humbanumena mais il s’ inscrit dans une époque où
les liens avec la Babylonie sont resserrés par des alliances matrimoniales : Pahir-iššan épouse la
sœur de Kurigalzu Ier, Humbanumena, sa fille et Untaš-Napiriša la fille de Burnaburiaš II. On
sait aussi que Humbanumena vient sur le trône après le deuxième fils d’ Igihalki, Attar-Kittah,
dont il est lui-même le fils. On peut donc supposer que le fils de la branche aînée de Pahir-iššan
était trop jeune pour monter sur le trône privilégiant la branche cadette d’ Attar-Kittah.17 Le
titre likume rišakka que Humbanumena est le seul des Igihalkides à utiliser aurait alors un enjeu
d’affirmation du pouvoir et d’autorité dynastique. Cette insistance sur la continuité dynastique
dans l’exercice du pouvoir royal se retrouve dans la brique de Šilhak-Inšušinak où il est dit ruhu
šak de Silhaha18 et lui-même, dans la dédicace du temple d’ Inšušinak qu’ il bâtit à Suse, insiste
sur cette continuité et son élection par son dieu : « en raison de la continuité par (ma) mère,
le Grand-dieu m’a choisi et m’a aimé ; la prospérité établie, la couronne restaurée, Inšušinak
m’a donné la royauté»19. Puis il rappelle son travail de constructeur pour le dieu, et ce « pour sa
propre vie», et «pour la vie» de deux Dames Mišimruh et Rišap-La dont les rapports familiaux
ne sont pas précisés.
Untaš-Napiriša est le fils de Humbanumena. Des briques ont été retrouvées dans plusieurs
centres du Khuzistan, mais on a très peu de traces de son activité extérieure : juste un fragment
de statue de [T]upliaš qui suggère la pression des armées élamites au-dessus de Der. C’ est
un règne fécond, caractérisé par la construction de Tchoga-Zanbil. Le titre likume rišakka

14 König ( : nos –).

15 König ( : nº ).
16 König ( : nº ).
17 Quintana a proposé de voir dans la variante «king of Susa and Anzan» et «king of Anzan and Susa» la

traduction d’ une maîtrise sur l’ une des deux parties du royaume divisé en deux entités politiques. Cf. Quintana
Cifuentes ( : nº ).
18 König ( : nº  note ).
19 Malbran-Labat ( : nº ).
 approche historique et philologique du titre royal ‘likame/we rišakki’ 

n’apparaît jamais dans ses inscriptions où il est inlassablement « fils de Humbanumena, roi
d’Anzan et de Suse»20. La suite et la fin de la dynastie igihalkide sont mal connues : le trône
revient à la branche aînée, celle de Pahir-iššan, avec Unpahaš-Napiriša mais pour lequel nous
n’avons aucune inscription, puis Kidin-Hutran.
Šutruk-Nahhunte est le fondateur d’une nouvelle dynastie. On connaît, par la célèbre lettre
de Berlin21, son ressentiment de ne pas siéger sur le trône babylonien. Il mène par la suite
une politique expansionniste et une guerre dévastatrice à l’ encontre de la Babylonie. Plusieurs
incursions lui ont permis de rapporter ses trophées historiques. Il met fin à la dynastie cassite et
réussit à faire monter son fils sur le trône babylonien22. De fait, il « élargit son royaume », mais
il porte au moins aussi souvent le titre de « roi d’ Anzan et de Suse » ; celui de likume rišakka
ne fait pas nécessairement référence à ses victoires sur la Babylonie et pourrait bien avoir pour
fonction d’établir son pouvoir et d’affirmer son autorité sur le pays alors même qu’ il fonde une
dynastie nouvelle puisque son fils Kutir-Nahhunte lui succède et apparemment sans problème.
D’ailleurs Kutir-Nahhunte qui élimine définitivement l’ opposition babylonienne et ose
emporter la statue de Marduk de l’Esagil se dit toujours « roi d’ Anzan et de Suse », jamais likume
rišakka.
Šilhak-Inšušinak est le frère de Kutir-Nahhunte. Il a laissé nombre d’ inscriptions qui commé-
morent non seulement de prestigieuses restaurations (à Suse23, Bender-Bushir24, Tchoga Pahn
Ouest25, Tépé Bormi26, Tul-e Sepid27 sans oublier les temples d’ autres villes mentionnés dans
quelques inscriptions) mais aussi ses campagnes mésopotamiennes. Les possessions se seraient
alors étendues au nord-ouest jusqu’à Nuzi et Arrapha, et au sud-ouest en Babylonie peut-être
jusqu’à Nippur28. Il est célèbre pour la création de vingt-deux provinces et l’ installation de leurs
gouverneurs. Grand roi à la politique clairement expansionniste, il ne porte curieusement le
titre likume rišakka que dans certaines inscriptions takkime, offertes « pour la vie » du souve-
rain et sa descendance. Ainsi l’emploi du titre semble montrer le lien avec la notion d’ accession
au pouvoir suprême de celui qui est en droit de transmettre ce pouvoir. Cette volonté soute-
nue de définir la lignée dynastique pourrait bien expliquer la curieuse dédicace sur perle qui
comporte l’épithète importante du point de vue politique29 de pak hanik attribué à Bar-Uli :
() ù msil-ha-ak-d () in-su-uš-na-ak li- () ka4-me ri-ša-ak-ki i () ia-áš-pu ašpur-al-() si-iš hu-
ma-ah hu-ut-tak () ha-li-ik ú-() me a-ha ta-ah a-ak () fba-ar-d() ú-li pa-ak () ha-ni-ik ú-()
ri i-du-ni-ih
Moi, Šilhak-Inšušinak, likame rišakki, j’ai pris le jaspe de Puralsiš, j’ai placé là mon œuvre accomplie
et je l’ai donné à Bar-Uli ma fille bien-aimée!30
On voit que le titre apparaît de manière exclusive alors que Šilhak-Inšušinak évoque sa des-
cendance, Bar-Uli, «sa fille bien-aimée », probablement celle qui transmettra la légitimité
dynastique.

20 Stève ().
21 Van Dijk ( : –).
22 Brinkman ( : s).
23 Malbran-Labat ( : nos –).
24 König ( : nos –) et voir par ailleurs Grillot & Vallat (: –).
25 Stève ( : nº ) et Stolper ( : –).
26 Vallat ( : –).
27 Lambert ( : ).
28 Potts ( : –).
29 La racine hani- a le double sens « aimer » et «choisir», cette deuxième acception pouvant avoir une valeur

quasi-juridique : « [der Ausdrück] dürfte in der Tat jüristisch gemeint sein: der für die Thronfolge ausgewälte»
(Kammenhuber  : ).
30 Sollberger ( : –).
 stéphanie anthonioz and florence malbran-labat

Huteluduš-Inšušinak dont le royaume fut gravement attaqué et amputé se dit likume rišakka,
jamais «roi d’Anzan et de Suse». Sa filiation est en général šak hanik « fils bien-aimé » de
Kutir-Nahhunte et Šilhak-Inšušinak ; cependant, dans l’ inscription d’ une brique de Suse de
langue archaïsante, il se réfère à Silhaha le grand roi épartide : mhu-te-lu-tu-uš din-su-uš-na-
ak li-kà-we ri-ša-ak-ki me-ni-ik ha-ta-am-ti-ik a-kí šu-ù-še-en-ki ru-hu ša-ak msi-il-ha-ha-ki31.
François Vallat a proposé de suivre l’ interprétation de Maurice Lambert32 et de supposer que
le roi élamite s’était retiré au sein de la montagne, dans sa capitale de repli, Anzan, sous les
coups de Nabuchodonosor Ier. C’est alors qu’ il aurait édifié un « temple de l’ alliance » (tarin).
La situation partiellement rétablie, il serait revenu à Suse où, restaurant et embellissant le
kukunum, il aurait fait graver l’inscription archaïsante. Quoi qu’ il en soit de cette recons-
truction historique, la dédicace de cette brique archaïsante fait appel à un mode successoral
ancien, matrilinéaire (ruhu šak); alors que dans ses autres inscriptions il se dit ruhu hanik de
Kutir-Nahhunte et Šilhak-Inšušinak voire aussi de Šutruk-Nahhunte, il se rattache désormais
à l’ancêtre fondateur par excellence, Silhaha. Difficile à interpréter ce choix de titulature pour-
rait être révélateur de querelles familiales dans la succession dynastique. L’ expression likume
rišakka qui se comprend difficilement dans ce contexte historique si elle signifie « agrandisseur
du royaume», est plus en place si elle se réfère à un prince qui affirme être (resté) préémi-
nent dans un royaume mis en péril. La rhétorique pourrait ainsi être le reflet d’ une certaine
désagrégation impériale. Il affirme alors sa légitimité par sa juste place dans la lignée royale,
voire dans la succession des princes qui se sont transmis le pouvoir depuis les temps anciens
des ancêtres fondateurs, plus que par sa maîtrise sur tout le royaume ou par le succès de ses
armes.
Enfin, l’histoire néo-élamite apparaît encore trop fragmentée et fragmentaire pour tenter de
tirer un argument historique de l’emploi ou, au contraire, du non emploi du titre.
Au terme de l’étude du contexte historique, l’ influence qui pourrait expliquer un possible
parallélisme entre le titre élamite (likume rišakka) et l’ épithète assyrienne (murappiš …) est loin
de s’imposer. Les circonstances politiques en Élam au cours des différents règnes ne donnent
pas une justification nette de l’emploi de l’ épithète « agrandisseur du royaume ».

. L’ analyse philologique

Le titre, inauguré par Siwe-palar-huppak33, porté par plusieurs souverains igihalkides et šutruki-
des, puis par quelques souverains néo-élamites, seul ou combiné avec d’ autres titres34, présente
des variations. Celles-ci portent sur la voyelle du premier terme (likume / likame), sur celle du
suffixe du second terme (rišakka / rišakki)35, sur la forme de ce suffixe (rišakka/i / rišari / rišah) :
– ligawe rišakki (Siwe-palar-huppak)
– likume36 rišakka (Humbanumena, Šilhak-Inšušinak37, Šutruk-Nahhunte)38

31 Vallat ( : –).

32 Lambert ( : –).
33 ligawe rišakki menik hatamtik cf. MDP XXXI:sq. = König (: nº  A+B, pp. – et note ).
34 L’ inscription de Humbanumena fait ainsi suivre l’expression likume rišakka de trois autres titres: merrik

hatamtik, katri hatamtik et halmenik hatamtik.

35 Ces deux variations semblent liées : likume + rišakka // likame + rišakki / rišari. Une exception: likume rišakki

cf. König ( : nº  I).

36 li-ku-mi ri-ša-[ak-ka/ki] cf. König (: nº  I).
37 Seulement dans deux inscriptions, cf. König (: nos  et ).
38 [likume rišakka] restitué dans une inscription de Atta-hamiti-Inšušnak, cf. König (: nº ).
 approche historique et philologique du titre royal ‘likame/we rišakki’ 

– likame rišakki (Šilhak-Inšušinak)

– likame rišari (Huteluduš-Inšušinak)
– Anzan Šušun likumena rišah (Hallutaš-Insušinak)
La première variation (likume / likame) pourrait opposer une construction comportant une
référence à la première personne (lik.u.me « le donné à moi = ce qui m’ a été donné ») à une
forme qui en serait dépourvue (lika.me « le donné »). Pour la seconde (rišakka / rišakki),
l’alternance a / i, si elle est significative, pourrait porter sur la caractérisation (ou la non
caractérisation) d’un état considéré comme le fruit d’ une réalisation (« devenu riša »)39. Mais,
pour l’heure, notre analyse porte essentiellement sur une possible remise en question du sens
du second terme construit sur la racine riša-, traditionnellement traduit par « agrandisseur (du
royaume) »40. Pour ce faire, nous verrons successivement les attestations de cette racine (),
l’analyse morphologique de ces formes (), leur construction et la signification qui peut en
être ainsi déduite ().

.. Les attestations de la racine riša- hors de l’ expression likume rišakki

Cette racine apparaît comme épithète, que ce soit dans des noms divins (Napi-riša, Kiri-riša),
des anthroponymes (f Riša.p-dLa), des substantifs (napir rišarra, temti rišar / rišari, rutu rišarra
/ rišarri, zana rišarri, pisan riša).
Il peut être construit avec un déterminant, soit antéposé (Inšušinak riša nappirra, temti rišar
nappirra), soit postposé (dumu lugal rišara).
Dans toutes ces attestations, riša a le sens de « grand » : Napiriša « Grand-dieu », Kiririša
«Grande-déesse», temti rišar(i) «grand seigneur », rutu rišarra / rišarri « grande épouse », zana
rišarri «grande dame», Inšušinak riša nappirra « Inšušinak le (plus) grand des dieux », dumu
lugàl rišara «(des) fils du roi, le grand = l’ aîné des fils du roi », etc.
Dans ces emplois, riša fonctionne comme épithète soit sans marque (cf. Napi-riša), soit avec
la marque de ème personne (-r) qui l’accorde avec le terme qu’ il qualifie (temti rišar). Il peut,
dans cet emploi, être substantivé ce qui a conduit à une traduction par « maître »41. Est aussi
attestée une forme rišehhuna dans une inscription de Šilhak-Inšušinak42.

.. Les différentes formes de la racine riša-

Ainsi, si l’on considère l’ensemble des attestations, la racine riša- apparaît soit nue, soit con-
struite avec différents suffixes:
– riša.ri / riša.r(a)
– riša.p
– riša.kki/a
– riša.h
– riše.hhuna
Les formes comportant les suffixes -r et -p sont clairement des formes nominales de ème
personne, du singulier pour la première (riša.ri / riša.r(a) « (il est) grand »), du pluriel pour
la seconde (riša.p «(ils / elles sont) grand(e)s »).

39 ri-ša-ak-ki : « die Form auf -ki dürfte unvollender Aspekt sein»; ri-ša-ak-qa «die Endung -ka deutet auf
vollendeten Aspekt » (Hinz & Koch  : ).
40 s.v. ri-ša-ak-ki, ri-ša-ak-qa, ri-ša-ri cf. Hinz & Koch (: –).
41 Grillot ( : ) le traduit ainsi à côté d’un autre terme de même construction bahi.r «protecteur».
42 König ( : nº  § ).
 stéphanie anthonioz and florence malbran-labat

Qu’en est-il des formes à suffixe -k? En effet cette marque est à la fois celle du passif /
intransitif (hani.k «aimé», sa.k «il est parti ») et celle de la ère personne dans la conjugaison
nominale, celle que E. Reiner a nommé « locutif » (u … sunki.k Anzan-Šušun.ka « je (suis) roi
d’Anzan-Suse»). Il est peu vraisemblable que riša.kki / riša.kka soit une forme passive, pour
trois raisons:
– les formes passives se rattachent toujours à un verbe de sens transitif, qui présente des
formes «agentives»43 (procès réalisé par un agent sur un patient) face à des formes
«passives» (ex. hutta.k(a) «(il a été) fait » ≠ hutta.š « il a fait »). Ce n’ est pas le cas pour
la racine riša-.
– une forme passive aurait le sens de « (il a été) agrandi », qui ne convient pas au contexte.
– enfin l’existence des formes clairement nominales de ème pers. (riša.r et riša.p) pousse à
analyser le riša.kki / riša.kka comme une ère personne nominale.
L’emploi d’un suffixe nominal entraîne, pour des racines de sens actif, la formation d’ un terme
qui signifie, selon la personne du suffixe : « je (suis) celui qui fait l’ action de … » / « il (est) celui
qui fait l’action de …» / «ils / elles (sont) ceux / celles qui font l’ action de … » : de nombreux
noms de métier sont ainsi formés: kuši.r(a) « (il est) celui qui fait l’ action de construire =
constructeur»; tahhi.ri «(il est) celui qui fait l’ action d’ aider = aide, auxiliaire », hutti.p « ceux
qui font» «faiseurs» «artisans», etc. tipira « scribe », « secrétaire », bakra « gardien », kazira
«forgeron»44, etc. Mais ce sens n’est bien attesté que pour des verbes transitifs, pour désigner le
réalisateur de l’action. Ces verbes ont tous une conjugaison « agentive » attestée. Or ce n’ est pas
le cas pour riša- et rien n’établit la possibilité de passer de « (être) grand » (état) à « agrandir »
(action).

Seules les deux formes rišah et rišehhuna pourraient attester une conjugaison « agentive ». Or la
forme riša.h, qui pourrait évoquer une ère sg. agentive par son suffixe -h, peut être une forme
nominale de ère pers. où le h représente une forme spirantisée du suffixe -k45. Il s’ agit donc
d’une simple variante de riša.k(ki/a). Reste la forme rišehhuna dans une phrase qui a pour sujet
«moi et Nahhunte-Utu» et dont les prédicats sont à l’ optatif (suffixe -na) : u ak Nah[hunte-
Utu] rišehhuna šatehuna. Cette forme a été traduite par « puissè-je exalter », par « wir möchten
grossmachen», par «afin que nous soyons agrandisseurs( ?) »46. Ces différentes traductions
rattachent cette forme à la conjugaison agentive, le suffixe -hu y étant de fait bien attesté comme
marque de la ère pers. pl. Les deux premières posent, pour la racine, un sens transitif « exalter »
ou «agrandir»; la troisième une base ayant un sens d’ état « être agrandisseur ».
Mais cette analyse peut être remise en question sur deux points :
– le sens donné est loin d’être satisfaisant, ce que souligne la diversité des traductions
proposées.
– l’absence de complément direct n’ est pas normale avec les formes agentives qui réclament
un patient.
– les suffixes de conjugaison I (que nous appelons « agentive ») se réfèrent à un agent et non
à celui / ceux dont on exprime une qualité.

43 Nous appelons conjugaison « agentive », ce qui est la conjugaison I pour Grillot (: ) et Khačikyan (:
).
44 Certains de ces termes ont des constructions avec déterminant antéposé, à la manière d’un mot composé:
lin.huttip /ra « creuseur(s) de canal », bala.hutippe «malfaisants», lalla.rippe «qui rompent la paix», lam.lir(ri)
« célébrant d’ un culte », etc.
45 Reiner ( :  n. ).
46 Hinz & Koch ( : ).
 approche historique et philologique du titre royal ‘likame/we rišakki’ 

Aussi, la forme du suffixe nominal de ère pl. n’ étant pas connue, ne serait-il donc pas possible
de voir dans rišehhuna et šatehuna deux optatifs des formes nominales ?

.. La construction du syntagme likame riša(k)ki / likume rišakka

La traduction «agrandisseur du royaume » s’ appuie sur la présence, dans ce titre, de likume
placé devant riša : cette place est celle du complément direct d’ un verbe transitif (du type : u
siyan kuših), ou celle d’un déterminant dans une construction où le déterminé est un nom
d’agent et le déterminant l’objet de son action47 :
– meten-huhuta.n.ki «(de) victoire(s)-je suis réalisateur » = « je suis celui qui remporte des
victoires».
– hain-kuši.ra «(de) maison(s) -il (est) constructeur » = « il est constructeur de maison(s) ».
– hiš-(h)u .li.ra «(d)un nom-à moi.il (est) donneur » = « il est celui qui m’ a nommé un nom ».
En ce cas le déterminant antéposé ne présente pas de marque de dépendance48. Une telle
construction avec un déterminant antéposé exprimant l’ objet de l’ action réalisée par l’ agent
n’est pas compatible avec une racine qui n’ a pas un sens actif. D’ où le passage supposé de « (être)
grand» à «agrandir» pour riša et la traduction « agrandisseur » pour riša + marque personnelle.
Cependant une autre hypothèse peut expliquer ce syntagme, en tenant compte des arguments
qui font de riša- une racine exprimant un état (« être grand »). Le destinataire et la localisation
sont aussi, en élamite, deux participants possibles pour un prédicat ; il est concevable qu’ un
déterminant soit aussi de cette nature49 : likume rišakka peut être compris comme un syntagme
formé d’un déterminant antéposé représentant le destinataire (ou/et une localisation) : « grand
pour le royaume» ou «grand sur le royaume ». De même hal.meni.k peut-il être compris à partir
d’une racine men(i) et un déterminant antéposé hal « pays ».

. Conclusion

Ainsi le titre de likume rišakki/a peut morphologiquement et syntaxiquement être compris

comme «je (suis) le grand pour / sur le royaume ». Cette notion de « grand » pourrait exprimer
la prééminence de celui qui est devenu souverain. Il aurait alors la même portée que l’ épithète
mah du titre des souverains de l’époque paléo-élamite, sukkal mah, dont nous ignorons la
lecture50. Mais il est notable qu’il est spécifique à l’ Élam51, à une époque où est nettement
attestée l’existence d’un cursus honorum, qui permet aux dynastes locaux de briguer le pouvoir
suprême sur l’ensemble élamite. Cette notion de prééminence est constitutive de la conception
royale en Élam: elle correspond à la structure « fédérative » sur laquelle s’ appuyait la puissance
élamite et dont M. Stolper a décrit la formation avec l’ émergence de la dynastie de Simaški. Elle
rejoint le concept de primus inter pares, proche de celui de « primat » dans l’ expression « primat
des Gaules».

47 En revanche, lorsque le déterminé est un substantif «d’état» (qui n’exprime pas une action), la construction
normale est : déterminé –déterminant –marque du déterminé (ex. u … šak Humbanumena.ki «moi … fils de
Humbanumena »).
48 Cependant une évolution a tendu à introduire une marque de dépendance pour ce déterminant antéposé:

c’ est le cas pour l’ expression que l’ on trouve chez Hallutaš-Insusinak Anzan Susun likume.na risah qui emploie la
particule de génitif.
49 De même avec un verbe intransitif : hal-sa.k «landvertrieben» ou un syntagme nominal: alimelu siyan.me «dans

le temple de l’ Acropole ».
50 Quintana ( : nº ) a proposé de voir dans likame rišakki la lecture de sukkal mah, titre employé

uniquement en Élam (même s’ il est composé de deux idéogrammes d’emploi fréquent en Mésopotamie).
51 Traduit par Durand par « empereur ».
 stéphanie anthonioz and florence malbran-labat

Par ailleurs ce titre royal comporte sans doute une référence à la succession dynastique :
«le grand» est celui qui a reçu (légitimement) le pouvoir suprême et à qui il appartient de
le transmettre tout aussi légitimement. Cela est particulièrement important dans un système
dynastique complexe, où les droits des fils (šak) du souverain régnant pouvaient être contestés
par ceux d’un frère, voire des fils d’un frère ou d’ une sœur du souverain prédécesseur. Rappe-
lons qu’il nous a semblé que le titre likame rišakki apparaissait dans la titulature de souverains
qui avaient rencontré des problèmes de succession dynastique et qu’ il était particulièrement
lié aux inscriptions pour la Vie (takkime) de la famille royale ou à celles qui font référence à
des ancêtres lointains et glorieux (considérés comme des ancêtres fondateurs pour la fonction
royale)52.
Enfin, un dernier point: la titulature méso-élamite se caractérise, nous l’ avons dit, par
l’alternance de deux titres: sunkik Anzan-Šušun.ka et likume / likame rišakki/a. N’y aurait-il
pas lieu de voir dans ces deux titres deux aspects de la souveraineté royale. Le premier (sunkik
Anzan-Šušun.ka) ferait référence à une notion spatiale : la racine sunki-, à laquelle se rattachent
les mots «roi(s)» (sunki.k/r/p) et sunki.me qui signifie à la fois « royaume » et « royauté »
exprimerait la royauté définie par le territoire du royaume. En revanche, le second titre (likame
riša(k)ki / likume rišakka) exprimerait la royauté légitimée par la succession à l’ intérieur d’ une
lignée dynastique53 déterminée par des règles successorales remontant aux ancêtres fondateurs.
En effet, lik.u.me est sans doute à rattacher à la racine li- « donner, remettre » : « la chose donnée
/ remise (voire, «transmise») à moi ». Ce pouvoir royal « remis / transmis » pourrait englober
deux aspects, peut-être d’importance variable suivant les époques : la transmission au sein
d’une lignée dynastique et la reconnaissance par les « pairs » qui constituent l’ entité élamite.
Cette royauté serait légitimée à la fois par le droit successoral et par le choix divin. En
effet, plusieurs inscriptions associent ces deux plans. Ainsi Humbanumena se dit « grand sur le
royaume (likume) en raison de la continuité par lignage féminin » avant d’ affirmer « le Grand-
dieu m’a choisi (un haniš), la prospérité établie, la couronne restaurée, Inšušinak m’ a donné la
royauté (sunkime)», à moins qu’il ne faille traduire de manière plus nette « devenu chef de la
dynastie en raison de la transmission par lignage féminin, Grand-dieu m’ a choisi … Inšušinak
m’a remis le royaume». Deux références divines viendraient alors se compléter ici : Grand-
dieu, divinité du Plateau et de ses populations (semi-)nomades et Inšušinak, qui par son nom
même de «Maître de Suse», se caractérise comme le dieu d’ une entité politico-géographique.
En revanche, sous les Šutrukides, Inšušinak étant devenu le dieu dynastique, l’ élection divine
est exprimée de manière différente: dans deux inscriptions de Šilhak-Inšušinak54 le titre likame
rišakki est remplacé par l’expression likame Inšušinak ir.hani.š.ri « Inšušinak est celui qui l’ a
choisi pour la royauté / dynastie»55.

Ces différentes considérations nous ont semblé permettre de remettre en question le sens et
l’interprétation d’un titre porté par plusieurs des souverains d’ Élam, tout au long de son histoire
ne faisant somme toute que revenir à peu de choses près à la traduction avancée en  par
V. Scheil: li-ku-mi ri-sha-[ak-ki] «grand prince »56.

52 On sait comment les souverains usurpateurs ont souci de proclamer leur légitimité, comme l’illustre en

Mésopotamie la personnalité des deux Sargon («šarru-kēn»).

53 Une inscription de Hanne associe la racine riša- avec le terme «Maison»: a-a-in-u-me.na rišah.
54 König ( : nº  c II et  § ).
55 « Dessen X er liebte » (Hinz & Koch : ).
56 Scheil ( : –).
 approche historique et philologique du titre royal ‘likame/we rišakki’ 

Bibliographie

Brinkman, J.A. (): «A Preliminary Catalogue of Written Sources for a Political History of Babylonia:
–B.C.», JCS , –.
———. (): A Political History of Post-Kassite Babylonia, – B.C. (Analecta Orientalia ),
Rome.
———. (): «Foreign Relations of Babylonia from  to B.C.: The Documentary Evidence»,
American Journal of Archaeology , –.
Carter, E. (): Excavations at Anshan (Tel-e Malyan): The Middle Elamite Period (University Museum
Monograph ), Philadelphia.
———. (): «An Interpretation of the Middle Elamite Remains from Anshan, Tall-I Malyan, Iran»,
XXXIVe Rencontre Assyriologique Internationale (Türk Tarih kurumu yayinlari. Dizi /), Ankara,
–.
Charpin, D. & Durand, J.M. (): «La suzeraineté de l’empereur (sukkalmah) d’Élam sur la Mésopo-
tamie et le ‘nationalisme’ amorrite», Mésopotamie et Élam, Actes de la XXXVI e RAI, – Juillet 
(Mesopotamian History and Environment. Occasional Publications ), Gent, –.
De Meyer, L. (): «Elamite likame risakki: ‘l’agrandisseur du royaume’ ou ‘le grand du royaume’?»,
Akkadica , –.
De Miroschedji, P. (): «La fin de l’Élam: essai d’analyse et d’interprétation», IrAnt , –.
Durand, J.-M. (): «La maîtrise de l’eau dans les régions centrales du Proche-Orient», Annales.
Histoire, Sciences sociales , –.
Eilers, W. (): «Zwei kurze elamische Inschriften», AMI , –.
Farber, W. (): «Eine elamische Inschrift aus der . Hälfte des . Jahrtausends», ZA , –.
Frayne, D. (): Old Babylonian Period (–BC) (RIME IV), Toronto.
Garelli (): «Les temples et le pouvoir royal en Assyrie du xive au viiie siècle», Le temple et le culte.
Compte rendu de la XXVII e RAI, Leiden, – juillet  (PIHANS ), Istanbul, –.
———. (): «La conception de la royauté en Assyrie», Assyrian Royal Inscriptions: New Horizons in
Literary, Ideological, and Historical Analysis, Papers of a Symposium Held in Cetona (Siena), June –,
, Roma, –.
Grayson, A.K. (): Assyrian Rulers of the Third and Second Millennia bc (to ) (RIMA I), Toronto.
Grillot, F. (): «La postposition génitive -na en élamite», DAFI , –.
———. (): Éléments de grammaire élamite, Paris.
Grillot, F. & Vallat, F. (): «Dédicace de Šilhak-Inšušinak à Kiririša», IrAnt , –.
Hinz, W. & Koch, H. (): Elamisches Wörterbuch, Berlin.
Joannès, F. (): Dictionnaire de la civilisation mésopotamienne, Paris.
Kammenhüber, A. (): «Historisch-geographische Nachrichten: Aus der althurrischen Überliefe-
rung, dem Altelamischen und den Inschriften der Könige von Akkad für die Zeit vor dem Einfall der
Gutäer», Acta Antiqua , –.
Khačikyan, M. (): The Elamite Language, Roma.
König, F.W. (): Die elamischen Königsinschriften (AfO Beih. ), Graz.
Kupper, J.-R. (): Inscriptions royales sumériennes et akkadiennes (LAPO ), Paris.
Lambert, W.G. (): «Une inscription de Kutir-Nahunte», JA , –.
———. (): «Hutelutush-Inshushnak et le pays d’Anzan», RA , –.
———. (): «Disjecta membra aelamica (II). Inscriptions du décor architectural construit par Shilhak-
Inshushinak», Arts asiatiques , –.
———. (): «The Akkadianization of Susiana under the Sukkalmahs», Mésopotamie et Élam, Actes de
la XXXVI e RAI, – Juillet , Ghent, –.
———. (): «The God Aššur», Iraq , –.
Liverani, M. (): «The Ideology of the Assyrian Empire», Power and Propaganda. A Symposium on
Ancient Empires, Copenhagen, –.
Malbran-Labat, F. (): Les inscriptions royales de Suse. Briques de l’époque paléo-élamite à l’Empire
néo-élamite, Paris.
Potts, D.T. (): The Archaeology of Elam: Formation and Transformation of an Ancient Iranian State,
Cambridge.
Quintana Cifuentes, E. (): «Humban-numena I, un usurpateur à la royauté en Élam?», NABU , nº
.
———. (): « ‘Yo soy el engrandecedor del reino’: un título real elamita», NABU , nº .
———. (): «Une titulature royale élamite. Una réplica obligada», NABU , nº .
 stéphanie anthonioz and florence malbran-labat

Reiner, E. (): «The Elamite language», in: Altkleinasiatische Sprachen (HdO //–/), Leiden, –
.
———. (): «The Location of Anšan», RA , –.
Scheil (): «Légendes de Šutruk Nahhunte sur cuves de pierre», RA , –.
Seux, M.-J. (): Épithètes royales akkadiennes et sumériennes, Paris.
Sollberger, E. (): «A New Inscription of Šilhak-Inšušinak», JCS , –.
Stève, M.-J. (): Textes élamites et accadiens de Tchoga Zanbil (MDP ), Paris.
———. (): Mission de Susiane: ville royale de Suse (). Nouveaux mélanges épigraphiques: inscriptions
royales de Suse et de la Susiane (MDP ), Nice.
Stève, M.-J. & Vallat, F. (): «La dynastie des Igihalkides, nouvelles interprétations», Archaeologia
Iranica Orientalis, Miscellanea in Honorem Louis Vanden Berghe I, Gent, –.
Stolper, M.W. (): «Inscribed Fragments from Khuzistān», DAFI , –.
———. (): Texts from Tall-i Malyan I: Elamite Administrative Texts (–) (Occasional Publi-
cations of the Babylonian Fund ), Philadelphia.
Thureau-Dangin, F. (): «Iahdunlim, roi de Hana», RA , –.
˘
Vallat, F. (): «Une brique élamite ˘
de Hutelutush-Insushnak», DAFI , –.
———. (): «Un fragment de brique de Tépé Bormi», DAFI , –.
———. (): «Une histoire cinq fois millénaire», Dossiers Histoire et Archéologie , –.
———. (): «Une inscription élamite de Tépé Horreeye», Contribution à l’histoire de l’Iran. Mélanges
offerts à Jean Perrot, Paris, –.
———. (): «Une titulature royale élamite», NABU , nº .
Van Dijk, J. (): «Die dynastischen Heiraten zwischen Kassiten und Elamern. Eine verhängnisvolle
Politik», Or. N.S. , –.
 BEZIEHUNGEN ZWEIER GROßMÄCHTE –
ELAM UND BABYLONIEN IN DER 2. HÄLFTE DES 2. JT. V. CHR.
EIN BEITRAG ZUR INTERNEN CHRONOLOGIE

Susanne Paulus*

Die . Hälfte des . Jt. v. Chr. wird häufig als Zeitalter der internationalen Beziehungen
bezeichnet, worunter man die engen Kontakte der Großmächte des Vorderen Orients ver-
steht, die durch Korrespondenz, Güteraustausch und Handel, diplomatische Ehen, Staats-
verträge, aber auch kriegerische Auseinandersetzungen miteinander verbunden waren1. Der
Fokus der Betrachtung liegt dabei stark auf jenen Mächten, die in der sogenannten Amarna-
Korrespondenz vertreten sind, nämlich Ägypten, das Hethiterreich, Mittani, Assyrien und
Babylonien2. Auch die diplomatischen Beziehungen Babyloniens werden meistens nur für
diese Länder untersucht3, während der östliche Nachbar Elam kaum eine Rolle spielt4. Dieser
Umstand verwundert umso mehr, da die Elamer auch5 in der mittelbabylonischen Zeit, d. h. in
der . Hälfte des . Jt. v. Chr., die Geschicke Babyloniens entscheidend beeinflussten. So brach-
ten sie 6 die kassitische Dynastie7, die mehr als  Jahre Babylonien beherrscht hatte, zu
Fall und verschleppten nicht nur die Mardukstatue, sondern auch zahlreiche weitere wichtige
babylonische Objekte nach Susa8.
Zunächst verdeutlichten nur diese Funde sowie die Erwähnung der elamischen Überfälle in
den babylonischen Texten9 die schwierigen Beziehungen beider Länder. Dann wurde jedoch
 durch die Veröffentlichung der Texte aus Haft Tepe bekannt, dass bereits die frühen
Kassiten, wie in älteren Zeiten, einen engen Botenkontakt zu Elam unterhielten10. Als besonders
aufschlussreich erwies sich die neubabylonische Abschrift eines Briefes aus Babylon, in der ein
elamischer König die diplomatischen Ehen zwischen dem kassitischen und dem elamischen
Königshaus skizzierte, den van Dijk  veröffentlichte11. Ergänzend zu diesem Material

* Universität Münster.
1 Für eine Einleitung in die Problematik siehe Cohen, Westbrook () und Liverani ().
2 Siehe Moran (). Zur Auswertung siehe Cohen, Westbrook () und Liverani (). Kunsthistorisch

findet sich diese Fokussierung auf die westlichen Nachbarn zuletzt bei Aruz, Bezel & Evans ().
3 Vgl. dazu z. B. den Sammelband Leick () wo lediglich die Beziehungen Babyloniens zu Ägypten, den

Hethitern, Assyrien und der Levante betrachtet werden.

4 Siehe hier zusammenfassend Brinkman (: –), wo die Beziehungen Babyloniens zu Elam konse-

quent miteinbezogen wurden.

5 Für die früheren Beziehungen siehe zusammenfassend Potts ().
6 Alle Daten der mesopotamischen Geschichte sind als „v. Chr.“ zu verstehen. Die konkreten Jahreszahlen

folgen denen von Gasche, Armstrong, Cole & Gurzadyan () vorgeschlagenen (vgl. zu diesem Ansatz jedoch die
Diskussion unten). Dabei handelt es sich, was die babylonischen Daten angeht, um eine leicht korrigierte Version
des Ansatzes von Brinkman ().
7 Zu den Kassiten, ihrer Herkunft und Geschichte siehe einleitend Brinkman (–: –), Sommer-

feld (: –) und Zadok ().

8 Siehe zu diesen Ereignissen Potts (: –). Abbildungen der wichtigsten Beutestücke finden sich bei

Harper, Aruz & Tallon (: –).

9 Elamische Überfälle werden für die kassitischen Könige Enlil-nādin-šumi () und Adad-šuma-iddina

(–) in der sogenannten Chronik P erwähnt, vgl. Glassner (: Nr.  iv –). Die entscheidenden
Überfälle fanden jedoch unter Zababa-šuma-iddina () und Enlil-nādin-ahi (–) statt. Vgl. hierzu
Frame (: Nr. .. ’- ’). ˘
10 Herrero (: –).
11 van Dijk (). Die Kopie des Textes findet sich in van Dijk (: Nr. ). Zum Fund- und möglichen

Archivzusammenhang vgl. Pedersén (: Archiv N  Nr. ).

 susanne paulus

können noch einzelne Objekte hinzugezogen werden, wie eine aus Susa stammende kassitische
Statue mit einer Inschrift des elamischen Königs Untaš-Napiriša, die Vallat  wieder ins
Interesse der Forschung gerückt hat12. Aufgrund dieser verbesserten Materiallage sind mehrere
Arbeiten entstanden, die versuchen, die mittelbabylonisch-elamischen Beziehungen chronolo-
gisch zu rekonstruieren, wobei die wichtigsten Ansätze von Steve & Vallat ()13, ausgebaut
von Vallat ()14, von Goldberg ()15, übernommen und erweitert von Potts ()16, und
schließlich von Quintana ( und ) stammen17. Die Problematik dieser Ansätze liegt
jedoch darin, dass sie zwar dieselbe Materialgrundlage verwenden, jedoch auf Grund dieser zu
völlig unterschiedlichen Schlüssen kommen, was sich gut an einem einfachen Beispiel verdeut-
lichen lässt. Fragt man, welcher kassitische König Zeitgenosse des berühmten Untaš-Napiriša,
Erbauer der Zikkurat von Dūr-Untaš-Napiriša (Tchoga Zanbil), war, so erhält man folgende
Antworten:

Steve & Vallat (); Vallat () Burna-Buriaš II. (–)18

Goldberg () Kadašman-Enlil II. (–)19
Quintana ( und ) Variante A: Könige von Kurigalzu II (–) bis Kaštiliaš IV.
(–)
Variante B: Kaštiliaš IV (–)20

Da derartige Unterschiede in der Datierung auch eine große Auswirkung auf die Rekon-
struktion der mittelbabylonischen Geschichte haben, werden die verschiedenen Ansätze im
Folgenden unter Rückgriff auf die Primärquellen kritisch diskutiert, wobei es auf Grund der
schwierigen Quellenlage nicht immer möglich ist, endgültige Aussagen zu machen. Ziel ist es
daher, ein möglichst präzises Bild dessen zu zeichnen, was sicher, was wahrscheinlich und was
unmöglich ist.

. Der „Berliner-Brief“ VAT 

Zunächst ist wichtig festzuhalten, dass es sich bei diesem „Brief “ um ein im Schulunterricht
überliefertes literarisches Dokument handelt21, das im Zusammenhang mit den spätbabylo-
nisch überlieferten sogenannten „Kedor-laomer Texten“ steht22, jenen literarischen Briefen und
Texten, die ebenfalls den Fall der kassitischen Dynastie zum Inhalt haben23. Die für die Bezie-
hungen Babyloniens zu Elam relevante Stelle wird hier vollständig wiedergegeben24, auf kriti-
sche Ergänzungen wird dabei zunächst bewusst verzichtet. In der einleitenden Passage nach

12 Scheil (: – Tafel ).

13 Steve & Vallat (: –).
14 Vallat (: –). Eine Zusammenfassung der wichtigsten Thesen von Steve und Vallat findet sich bei

Steve, Vallat & Gasche (–: –).

15 Goldberg (: –).
16 Potts (: –).
17 Quintana (: – und : –).
18 So Steve & Vallat (: ) und Vallat (: ).
19 Der Synchronismus ergibt sich aus der Angabe von Goldberg (: ), der Untaš-Napiriša um  ansetzt,

und der von ihm verwendeten Chronologie, die Boese (: ) folgt.
20 Quintana (: –).
21 Siehe so bereits van Dijk (: –). Zum Schulunterricht in Babylonien und den dort verwendeten

literarischen Briefen vgl. Gesche (: –).

22 Zum Zusammenhang siehe van Dijk (: –). Zu den Texten siehe mit weiterer Literatur Foster (:

–).
23 Siehe zum Kontext auch Frame (: ).
24 Es fehlen die ersten stark zerstörten Zeilen, die die Anrede, möglicherweise auch den Beginn der Argu-
 elam und babylonien in der . hälfte des . jt. v. chr. 

der nicht erhaltenen Anrede- und Grußformel beschreibt der Autor des Briefes zunächst seine
Abstammung.

Vs." 25 pi-hi-ra-nu-dU? f! X - at -GAL Pahhir-Iššan hat PN, die [älteste] Tochter]26

˘
DUMU.MUNUS GALtum]
šá LUGAL dan-nu mku-ri-gal-zu [i-ta-ha-az] des starken Königs, Kurigalzu, [geheiratet]27.
˘
md
hu-um- ba -an-im-me-ni ma-ra- su] Hum ba n-numena hat seine] Tochter
˘[i-ta-ha-az]28 ˘ [geheiratet.]
˘
m
hu -un-da-šá- d]GAL it-ta-lad U ntaš-Napiriša hat er gezeugt.
˘mhu-un-da-[šá-dGAL] Unta[š-Napiriša]
˘
" DUMU . MUNUS-su [šá] mbur-na-bur-iá-áš hat eine Toch ter des Burna-Buriaš
i-ta-[ha-az] geh[eiratet.]
˘
m
ki-di-nu-[hu]-du- ru]-[diš] it-ta-lad Kidin-[Hu]tran hat er gezeugt. Kidin-[Hutran]
m ˘ hu-du-ru-diš]
ki-di-nu-[ ˘ ˘
˘
DUMU.MUNUS- su šá hat die Tochter von …29 geheiratet.
X X- kar -an-dun-iá-áš i-ta-h[a-az]
˘
md
GAL -[un -da- áš] it-ta-lad a-na-ku Napiriša - untaš] hat er gezeugt. Ich […]30
DUMU […]
DUMU . MUNUS GAL tum šá habe die älteste Tochter des Meli-Šipak
m
mé-le-e-dši-i-pak a-ta-h a-[az] geheira[tet].
˘
Später im Brief führt der Autor diese Filiation als Argument für seinen Anspruch auf den
babylonischen Thron an:

Rs. am-me-ni ia-a- ši LUGAL DUMU LUGAL
NUMUN LUGAL i-lit-tum LUGAL
šá [a-na KUR.MEŠ kar-ra-an-dun-ía-àš u
kur
e-la-[am LUGAL.MEŠ]31
DUMU DUMU.MUNUS GALtum šá LUGAL
dan-nu ku-ri-[gal-zu]
ina gišGU.ZA kurkar-an-dun-ía-àš ul
uš- šá]-[a-ab]
Warum ic[h], König, Sohn eines Königs, Samen
eines Königs, Abkömmling eines Königs,
die [für die Länder Babylonien und Ela[m
Könige waren],
„Sohn“ der ältesten Tochter des starken Königs
Kuri[galzu],
sitze] ich nicht auf dem Thron Babyloniens?

mentation enthalten. Vgl. van Dijk (: ). Für die Möglichkeit, den Text zu kollationieren, sei J. Marzahn
(Vorderasiatisches Museum, Berlin) herzlich gedankt.
25 Die Zeilenzählung folgt der Kopie van Dijk (: Nr. ) (s.o. Anm. ).
26 Zur Problematik der Lesung des Königsnamens vgl. Stolper (–: ) mit weiterer Literatur. In der

Zeile ist, wie van Dijk (: ) vermutet, der Name der Tochter zu lesen, davor steht jedoch nach Kollation ein
weibliches Personenzeichen. Für die Ergänzung des Zeilenendes spricht vor allem Rs.–, siehe dazu auch die
Diskussion im Folgenden. Vgl. Goldberg (: ).
27 Am Zeilenende ist ein Verb zu ergänzen. Sehr wahrscheinlich, wie van Dijk () vorgeschlagen hat, i-ta-ha-

az. ˘
28 Die Ergänzung ist sehr wahrscheinlich, da es in der folgenden Zeile um das Kind aus dieser Verbindung geht.

Vgl. auch ".

29 Siehe die Diskussion unten.
30 Hier sind mehrere Ergänzungen denkbar: „Ich bin [sein] Sohn.“ DUMU[-šú-ma]; „Ich bin sein Enkel.“

DUMU.[DUMU-šú-ma]; „Ich bin der [Schwester]sohn.“ DUMU [DUMU.MUNUS], so van Dijk (: ). Für
andere Vorschläge siehe auch Quintana (). Alle Ergänzungen sind möglich. Denkbar wäre auch ein kurzer
Personenname.
31 Diese Lesung erscheint mir wahrscheinlicher, als der von van Dijk () vorgeschlagene Singular bei LUGAL,

da hier die Vorfahren des Autors gemeint sind.

 susanne paulus

Von den hier genannten elamischen Königen lassen sich Pahhir-Iššan, Humban-numena
und Untaš-Napiriša einfach identifizieren, da alle Herrscher in einer Inschrift Šilhak-Inšušinaks
vorkommen und die dort aufgeführte Filiation mit der im „Berliner-Brief “ übereinstimmt32.
Während diese Zuordnung unumstritten ist, identifiziert Vallat den hier genannten König
mit Kurigalzu I. (vor )33, während Goldberg und Quintana Kurigalzu II. (–)
favorisieren34. Auf den ersten Blick sind nach babylonischen Quellen, in denen es gewöhnlich
äußerst schwer ist, zwischen Kurigalzu I. und II. zu unterscheiden35, beide Varianten möglich.
Der springende Punkt ist jedoch die Nennung von Burna-Buriaš im Anschluss an Kurigalzu. Es
gibt hier nur einen Herrscher, der gemeint sein kann, nämlich Burna-Buriaš II. (–)36,
wodurch Kurigalzu sicher als Kurigalzu I. identifiziert werden kann, da er vor Burna-Buriaš
II. regiert hat, was bei Kurigalzu II. nicht der Fall ist. Goldberg und Quintana behelfen sich
hierbei mit der Lösung, dass Burna-Buriaš kein König, sondern lediglich ein Prinz war37. Eine
derartige Annahme hätte für die Legitimation des Autors des Briefes fatale Folgen, da er betont,
von „Königen“ abzustammen38. Das Fehlen des Titels LUGAL = „König“ vor dem Namen ist
hier nicht ausschlaggebend, da dieser auch vor den elamischen Königsnamen, aber auch vor
Meli-Šipak und Adad-šuma-usur fehlt, beides unzweifelhaft kassitische Könige39. Wäre Burna-
˙ es für die Filiation des Autors entscheidend, den königlichen
Buriaš jedoch ein Prinz, so wäre
Vater zu nennen, was nicht geschieht. Demnach können folgende Verbindungen festgestellt
werden, was die Rekonstruktion von Vallat40 vorläufig bestätigt:
Pahhir-Iššan, Sohn des Igi-halki41 ∞ älteste Tochter Kurigalzu I. (vor ) (wahrscheinlich)
Humban-numena, Sohn des Attar-kittah42 ∞ Tochter Kurigalzu I. (vor )
Untaš-Napiriša, „Sohn“ des Humban-numena ∞ Tochter Burna-Buriaš II. (–)
An dieser Stelle tauchen die erste Lücken in der Filiation im „Berliner Brief “ auf: Für Untaš-
Napirišas Sohn, Kidin-Hutran, wird zwar eine Braut angegeben, der folgende „Name“ X X-
kar -an-dun-iá-áš lässt sich jedoch mit keinem der erhaltenen, babylonischen Königsna-
men in Verbindung bringen. Goldberg schlägt auch hier einen „Prinzen“ vor43, was aus den
eben erwähnten Gründen abzulehnen ist. Vallat möchte EŠ[ŠANA] kar -an-dun-ía-àš lesen44.
EŠŠANA, das elamische Logogramm für „König“, ist jedoch in einem babylonischen Schul-
text, in dem ansonsten die elamischen Personennamen nur mit starken babylonischen Einfluss
auf die Orthographie widergegeben werden45, und ausschließlich das sumerische Logogramm
LUGAL verwendet wird, nicht zu erwarten.
In Vergessenheit geriet bei den jüngeren Diskussionen der Lesungsvorschlag von van Djik,
der die Stelle m.[kur]kar-an-dun-iá-àš46 liest und eine Übersetzung als „Babylonierin“ postuliert47.

32 Zur Inschrift siehe König (: Nr. ).

33 Vallat (: ).
34 Goldberg (: ) und Quintana (: –).
35 Vgl. hierzu Brinkman (: –), Clayden (: –) und zuletzt Bartelmus (: –).
36 Der frühkassitische Herrscher Burna-Buriaš I. kommt aus chronologischen Gründen nicht in Frage.
37 Goldberg (: ) „who could be a prince“ Quintana (: –) umgeht das Problem in Version A und

folgt Goldberg in Version B.

38 Vgl. Rs. –. Vgl. so auch Vallat (: ).
39 Vgl. Vs." und Rs. ".
40 Vallat ().
41 Vgl. König (: Nr. ).
42 Die Filiation von Humban-numena findet sich in den eigenen Inschriften dieses Königs, vgl. König (:

–). Attar-kittah war nach König (: Nr. ) ebenfalls ein Sohn des Igi-halki.
43 Goldberg (: ).
44 Vallat (: ).
45 Vgl. van Dijk (: –).
46 Vgl. van Dijk (: ): „Die Lesung ist sehr wahrscheinlich“.
47 Van Dijk (: ).
 elam und babylonien in der . hälfte des . jt. v. chr. 

Trotz einiger Schwierigkeiten, wie das Fehlen einer Nisbe-Endung, ist m. E. jedoch, auch im
Vergleich mit den vorangehenden und folgenden Zeilen, DUMU.MUNUS- su šá m.kur kar -an-
dun-iá-áš als „Tochter eines Bab yloniers“ zu verstehen, wohingegen van Dijks Interpretation,
dass es sich bei m.kur kar -an-dun-iá-áš um die Mutter handelt, abzulehnen ist. Hier liegt damit
zum ersten Mal keine königliche Verbindung vor. Denkbar ist jedoch, dass Kidin-Hutran eine
königliche Ehe einging, die von babylonischer Seite nicht anerkannt war. Mit aller Vorsicht
ist hier darauf hinzuweisen, dass in Babylonien direkt auf die Regierungszeit Burna-Buriaš II.
(–) eine Zeit der Unsicherheit mit zumindest einem Usurpator folgte48.
Aus der Ehe dieses Königs Kidin-Hutran mit einer Babylonierin geht Napiriša-Untaš hervor,
für den keine königliche Heirat angegeben wird. Es ist schwierig für Kidin-Hutran, Sohn des
Untaš-Napiriša, und Napiriša-Untaš, Sohn des Kidin-Hutran, Entsprechungen in den elami-
schen Texten zu finden. In der bereits zitierten Inschrift Šilhak-inšušinaks sind im Anschluss
an Untaš-Napiriša lediglich ein Unpahaš-Napiriša49, Sohn des Pahhir-Iššan, und ein Kidin-
Hutran, Sohn des Pahhir-Iššan belegt50. Ein weiterer Kidin-Hutran ist als Gegner der baby-
lonischen Könige Enlil-nādin-šumi () und Adad-šuma-iddina (–) aus der baby-
lonischen „Chronik P“ bekannt51. Goldberg nimmt an, dass alle Erwähnungen sich auf den
gleichen Herrscher beziehen52. Quintana setzt neben dem im „Berliner Brief “ erwähnten Herr-
scher einen frühen Kidin-Hutran I., Sohn des Pahhir-Iššan, an, den er zwischen Pahhir-Iššan
und Humban-numena datiert53. Vallat opereriert mit drei Herrschern dieses Namens: Kidin-
Hutran I., Sohn des Pahhir-Iššan, den er, wie Quintana, ebenfalls früh ansetzt, Kidin-Hutran II.,
Sohn des Untaš-Napiriša, aus dem „Berliner Brief “ und abschließend Kidin-Hutran III., dem
Eroberer in der „Chronik P“54.
Geht man zum „Berliner Brief “ zurück, so geht daraus zunächst hervor, dass der Kidin-
Hutran, Sohn des Untaš-Napiriša (um ), aus chronologischen Gründen nicht mit dem
in der Chronik P genannten (um ) identisch sein kann. Es ist theoretisch möglich,
dass er identisch mit dem Herrscher in der Inschrift Šilhak-Inšušinaks ist, da dieser dort
nach (!) Untaš-Napiriša und Unpahaš-Napiriša aufgeführt wird55. Der dort aufgeführte wird
jedoch als „Sohn“ des Pahhir-Iššan bezeichnet, was nicht mit der Filiation des „Berliner
Briefs“ übereinstimmt. Als Zwischenergebnis kann festgehalten werden: Der Kidin-Hutran des
„Berliner Briefs“ ist der frühste bekannte Kidin-Hutran und wird daher hier als Kidin-Hutran I.
bezeichnet. Daraus ergibt sich für die Beziehungen zu Babylonien:
Kidin-Hutran I., „Sohn“ des Untaš-Napiriša ∞ eine Babylonierin
Napiriša-Untaš, „Sohn“ Kidin-Hutran I. ∞ ?
Das bedeutet, dass sich unmittelbar nach Burna-Buriaš II. (–) ein Bruch in den
babylonisch-elamischen Beziehungen und den daraus resultierenden diplomatischen Ehen
abzeichnet. Die nächste derartige Verbindung, die im Brief erwähnt wird, besteht zwischen dem
Autor und der ältesten Tochter des Meli-Šipak (–). Damit ergibt sich eine deutliche
zeitliche Lücke im „Berliner Brief “. Diese lässt sich nicht, wie bereits gezeigt, durch das

48 Sowohl die Synchronistische Geschichte als auch die Chronik P berichten von dieser Episode der babylonischen
Geschichte, vgl. Brinkman (: – und –).
49 Diese Lesung ist für Zeile " ausgeschlossen, vgl. so auch van Dijk ().
50 König (: Nr. ).
51 Vgl. oben Anm. .
52 Vgl. Goldberg (: –).
53 Vgl. dazu Quintana (: –). Die von ihm geführte Diskussion, ob zwischen Kidin-Hutran und einem

Kidin-Hutrudiš, so die babylonischen Quellen, unterschieden werden muss, ist für die hier geführte Diskussion
nicht relevant, wie sich im Folgenden zeigen wird. Vgl. dazu auch Vallat (: ).
54 Vallat (: –).
55 Vgl. König (: Nr. ). Diese Reihenfolge wird weder von Vallat noch Quintana beachtet.
 susanne paulus

Hinabsetzen der dynastischen Ehen in spätere Zeit, wie es Quintana und Goldberg versuchen,
schließen56. Das ist auch nicht nötig, da der „Berliner-Brief “ nicht aussagt (siehe Vs."), dass
der Autor des Briefes ein direkter Sohn des Napiriša-Untaš ist. Die betreffende Stelle lässt sich
auf vielfache Weise ergänzen57. Klar ist lediglich, dass sich der Autor des Briefes auf eben diese
Linie beruft, wobei es das Wichtigste für ihn ist, dass er „Sohn der ältesten Tochter des starken
Königs Kuri[galzu] ist.“58 Damit steht fest:
Autor des Briefs, „Sohn“ des ? ∞ älteste Tochter des Meli-Šipak (–)
Jedoch muss nun folgende Frage geklärt werden: Warum ist die Verwandtschaft des Briefautors
mit Kurigalzu I. so entscheidend, dass er seine Filiation über einige Ecken auf diesen Kuri-
galzu I. zurückführt? Obwohl diese Frage von van Dijk angeschnitten wurde59 und m. E. für
die Interpretation des Briefes von entscheidender Bedeutung ist, wurde sie in der weiteren For-
schung ausgeklammert. Erhellend sind hier die folgenden Zeilen des Briefes:
md
Rs. IŠKUR.MU.ÙRU DUMU Adad-šuma-usur, Sohn des Dunna-[S]ah vom
m
du-un-na-d[sà]-ah60 šá GÚ ˙
Ufer des Euphrats, ˘
i7
BURAN UNki ˘
šá [tal]-[qa -nim-61ma ina gišGU.ZA den [ihr] genommen und auf den Thron
kur
kar-an-dun-iá-àš tu-š[e-ši]-ba Babyloniens ge[se]tzt habt,
Rs.  ki-[i] šu -ú DUMU DUMU.MUNUS wi[e] hat er den Sohn der Tochter
ú-qat- ta] vernic htend geschla[gen].

Der Autor bezeichnet Adad-šuma-usur als Usurpator, der im Gegensatz zu ihm, dem „Sohn der
Tochter“ = „Sohn der ältesten Tochter ˙ des Kurigalzu“ (siehe oben Rs. ), keinen Anspurch auf
den babylonischen Thron hat. Adad-šuma-usur (–)62 kam in den Krisenjahren, die auf
die Eroberung Babyloniens  durch den ˙assyrischen König Tukultı̄-Ninurta I. (–)
folgten63, durch eine Revolte auf den Thron, wobei die betreffende Passage in der „Chronik P“
der im „Berliner Brief “ sehr ähnelt:

IV … EGIR lúGAL.MEŠ šá KUR URIki šá KUR … Später haben die Großen des Landes Akkade
kar-an-dun-iá-àš BALA.MEŠ-ma (und) Babylonien sich empört und
m.d
IŠKUR.MU.ÙRU ina GU.ZA AD-šú Adad-šuma-usur auf den Thron seines Vaters
ú-še-ši-bu gesetzt. ˙

Anders als im „Berliner Brief “ wird hier jedoch betont, Adad-šuma-usur hätte den Thron seines
Vaters eingenommen. Eine Filiation fehlt jedoch an dieser Stelle ebenso ˙ wie in der Königsliste
64
A . Adad-šuma-usur selbst betont jedoch in einer Inschrift auf einem „Luristan-Dolch“, dass er
˙
der „Sohn“ des Kaštiliaš, gemeint ist Kaštiliaš IV. (–)65, des letzten Herrschers vor der

56 Quintana ( und ) und Vallat () (s.o. Anm. ).
57 Siehe oben Anm. .
58 So Rs. , siehe oben.
59 Vgl. van Dijk (: –).
60 Van Dijk (: ) liest Dunna-d[Za]h. Es liegt jedoch m.E. ein kassitischer Name mit dem theophoren

Element Sah (= Šamaš) vor. Für eine mögliche˘Etymologie siehe Balkan (: ) zu tuna-(mi)-Sah.
˘
61 Zur Ergänzung des Zeilenanfangs vgl. Rs.. ˘
62 Vgl. zu diesem König zusammenfassend Brinkman ().
63 Vgl. dazu zuletzt Paulus () mit weiterer Literatur.
64 Königsliste A II (vgl. Grayson [–]: ). Vgl. auch Brinkman (:  Anm. ). Weitere Informa-

tionen zur Regierungszeit Adad-šuma-usurs finden sich in der fragmentarischen Chronik W (Glassner : Nr. 
I–). ˙
65 Vgl. zu diesem Herrscher Brinkman (: –).
 elam und babylonien in der . hälfte des . jt. v. chr. 

assyrischen Eroberung, sei66. Dies steht klar im Widerspruch zur im Berliner Brief angegebenen
Filiation, die ihn als Sohn des Dunna-Sah bezeichnet. Van Dijk versucht das Problem dadurch
zu lösen, dass er Dunna-[S]ah als Mutter ˘ und nicht als Vater des Adad-šuma-usur interpre-
˘
tiert67. Dies würde nicht nur der üblichen babylonischen Orthographie – Dunna-[S]a ˙ h ist als
männlicher Personenname determiniert – sondern auch der babylonischen Thronfolge˘wider-
sprechen, die klar nur über die männliche Linie verläuft68. Die Argumentation des Autors des
„Berliner Briefs“ liegt gerade darin, dass er behauptet, dass Adad-šuma-usur auf Grund seiner
Abstammung keinen Anspruch auf den babylonischen Thron hat. Diesen˙ bestätigen zwar die
babylonischen Quellen, doch ist auch hier deutlich eine Problematik fassbar. Meli-Šipak (–
), der Nachfolger Adad-šuma-usurs, nennt in keiner seiner Inschriften seinen Vater69. Dass
Adad-šuma-usur sein Vater ist, geht˙ lediglich aus der wörtlichen Wiedergabe einer Aussage
˙
einer Privatperson, die auf einem Kudurru dokumentiert ist, hervor70. In einer Inschrift (VA
Bab ) nennt sich Meli-Šipak jedoch „Sohn des Kurigalzu“71, während sein Sohn, Marduk-
apla-iddina I. (–), sich als Nachfahre des Kurigalzu bezeichnet72. Brinkman hat klar
gezeigt, dass kein zweiter Meli-Šipak, Sohn des Kurigalzu, existiert, sondern dass sich Meli-
Šipak auf seinen Vorfahr, Kurigalzu, beruft73. Damit verweist Meli-Šipak, der wie oben gezeigt
wurde, vermutlich ein Zeitgenosse des Autors des „Berliner Briefs“ war, auf denselben Vorfah-
ren wie dieser, um seinen Anspruch auf den babylonischen Thron zu legitimieren. Das bedeutet
jedoch auch, dass der Autor des Briefes die Legitimation von Meli-Šipak, der ein Sohn des
Adad-šuma-usurs ist, ebenfalls in Frage stellt. Sieht man sich nun die Filiation von Meli-Šipak
und dem Autor ˙ im Vergleich an, so erhält man:

VA Bab  " Meli-Šipak, Sohn des Adad-šuma-usur DUMU ku-ri-gal-zu

˙
VAT  Autor, Sohn des ??? DUMU DUMU.MUNUS GALtum šá LUGAL
Rs.  dan-nu ku-ri-[gal-zu]

In beiden Fällen wird DUMU = māru = (wörtlich) „Sohn“, nicht dazu verwendet, um direkt auf
den Vater sondern um auf einen Vorfahren zu verweisen, was in babylonischer Tradition nichts
ungewöhnliches ist74. Um ihre Abstammung von Kurigalzu I. (vor )75, dem zumindest in
kassitischer Sicht wahrscheinlich wichtigsten König der Dynastie, zu beweisen, überschreitet
nicht nur Meli-Šipak, sondern auch der Autor des Briefes die große zeitliche Lücke.

66 Die Inschrift auf dem Dolch lautet: (Vs. ) ša dIŠKUR.MU.ÙRU () LUGAL KIŠ (Rs.) DUMU kaš-til-ia-šu

() LUGAL KÁ.DINGIR.RAki. Vgl. Dossin (:  Nr.  und Tafel XIII) = Brinkman (: Nr. C..).
67 So van Dijk (: ) besonders Anm. . „Für die hier vertretene Interpretation des Briefes ist die

Feststellung wesentlich, dass Dunna-Sah die Mutter ist.“

˘
68 So auch die Argumentation der Babylonier im „Berliner-Brief “, Vs.’ff., siehe ggf. die Übersetzung bei van
Dijk (: ).
69 Vgl. Brinkman (: ).
70 BM  (Vgl. King [: Nr. ]), iv  „Aus diesem Grund hat () Adad-šuma-usur, dein Vater (a-bu-ka),
˙ Zur Bedeutung dieses
() über das Land des Ta[kil-a]na-ilı̄šu, (…) () eine gesiegelte Urkunde ausgestellt“. Vgl.
Texts ausführlich Paulus (: –).
71 (’) [me]-li-ši-pak (’) DUMU ku-ri-gal-zu. Zu diesem Text zuletzt, mit einer Zusammenstellung älterer

Literatur, Stein (: ).

72 Die entsprechenden Belege sind eine Königsinschrift (= VS  Nr. , späte Kopie, vgl. dazu Stein : )

(Vs. ) ŠÀ.BAL.BAL ku-ri-gal-zu, sowie die Kudurrus BM  (King : Nr. ) (i ) ŠÀ.BAL.BAL ku-ri-
gal- zu und NBC  (unpubliziert, Publikation durch die Autorin in Vorbereitung) (ii ’) ŠÀ.BAL.[BAL] (’)
ku-ri-gal-zu. Vgl. auch Brinkman (: ).
73 Brinkman (: ).
74 Vgl. Brinkman (: ): „the simplest unit PN DUMU PN , in which PN can stand for a biological or
2 2
adoptive father, a more remote ancestor, or an eponymous head of a descent line“. Zur weiteren Problematik in
mittelbabylonischer Zeit siehe Brinkman (: –).
75 Vgl. zur Problematik oben Anm. .
 susanne paulus

Bleibt abschließend die Frage, wer der Autor des „Berliner Briefs“ ist. Van Dijk disku-
tierte sowohl Šutruk-Nahhunte I. als auch Kutir-Nahhunte, entschied sich jedoch auf Grund
des sogenannten Kedor-laomer-Texts, der Kutir-Nahhunte zugewiesen wird76, für denselbi-
gen77. Aus chronologischen Gründen haben sich dann jedoch Stève und Vallat für Šutruk-
Nahhunte I., der unter Zababa-šuma-iddina () Babylonien überfiel, entschieden78. Dieser
Meinung folgte Goldberg, während sich Quintana, auf Grund von Überlegungen zur Abstam-
mung von Šutruk-Nahhunte I. für Hallutuš-Inšušinak, den Vater Šutruk-Nahhuntes I., ent-
schied79.
Zunächst lässt sich festhalten, dass im „Berliner Brief “ kein späterer babylonischer König
als Meli-Šipak (–) erwähnt wird. Über den Autor haben wir zusammenfassend fol-
gende Informationen: Er ist mit der ältesten Tochter Meli-Šipaks verheiratet (Vs. ’). Er führt
seinen Anspruch auf den babylonischen Thron auf die Abstammung von Kurigalzu I. zurück
und bezeichnet sich als „Sohn der ältesten Tochter“ dieses Kurigalzu (Rs. –). Er geht in
seinem Brief ausführlich auf zwei aus seiner Sicht illegitime Könige ein, die in den Wirren nach
der assyrischen Eroberung den Thron inne hatten (Rs. –)80. Dabei wird auch erwähnt, dass
Adad-šuma-usur den „Sohn der Tochter“, das ist mit großer Wahrscheinlichkeit der Autor,
˙
vernichtend geschlagen habe (Rs. –). Diese Informationen sprechen sehr dafür, dass der
Autor mit jenem Kidin-Hutran II. aus der „Chronik P“ identisch ist, der Feldzüge gegen die
Könige Enlil-nādin-šumi () und Adad-šuma-iddina (–), die ebenfalls nach der
Eroberung durch Tukultı̄-Ninurta I. regierten, führte81. Es ist chronologisch und inhaltlich
durchaus möglich, dass dieser Herrscher nach seinen Erfolgen gegen diese Könige und sei-
ner im Brief erwähnten Niederlage unter Adad-šuma-usur zur Stabilisierung der Beziehung
eine Tochter Meli-Šipaks heiratete, was ihn jedoch nicht ˙ daran hinderte, weiterhin seinen
82
Anspruch auf den babylonischen Thron zu begründen . Das würde jedoch bedeuten, dass die-
ser Brief zeitlich früher anzusetzen ist als der sogenannte „Kedor-laomer-Brief “, indem der
Autor, Kudur-Nahhunte, ebenfalls in ähnlicher Formulierung seinen Anspruch auf den baby-
lonischen Thron begründet, wobei jedoch auffälliger Weise zwar die Abstammung von einer
Königstochter erwähnt wird, jedoch der Name Kurigalzu, der im hier vorliegenden Brief eine
entscheidende Rolle spielt, fehlt83.
Ein weiteres Argument im Text spricht dafür, dass es sich beim Autor um Kidin-Hutran II.
handelt: Der Autor ist „Sohn der ältesten Tochter des Kuri[galzu]“ (Rs.). Nach der hier
vertretenen Lesung von Vs. ’–’ war die älteste Tochter des Kurigalzu mit Pahhir-Iššan ver-
heiratet. Das würde bedeuten, dass Kidin-Hutran nicht nur der „Sohn“ der Tochter Kuri-
galzu I. ist, sondern gleichzeitig der „Sohn“ des Pahhir-Iššan, wobei hier „Sohn“, wie oben
gezeigt, jeweils als Nachfahre zu verstehen ist. Ein Kidin-Hutran, Sohn des Pahhir-Iššan,
wird, wie oben zitiert, in der Inschrift Šilhak-Inšušinaks im Anschluss an Untaš-Napiriša und
Unpahaš-Napiriša erwähnt, der ebenfalls als Sohn des Pahhir-Iššan bezeichnet wird. Betrach-
tet man daran anschließend eine Übersicht der Abfolge der Könige der Inschrift Šilhak–

76 Siehe oben Anm. .

77 van Dijk (: –).
78 Steve & Vallat (: –). Zum Beleg siehe oben Anm. .
79 Goldberg (: ), ohne die Problematik zu diskutieren. Vgl. die ausführliche Argumentation Quintana

(: –).
80 Vgl. zur Interpretation der weiter erwähnten Ereignisse van Dijk (: –).
81 Vgl. Chronik P iv  ff. Siehe ggf. Glassner (: Nr. ).
82 Für Kidin-Hutran ist dann eine relativ lange Regierungszeit anzusetzen. Da zwischen der Regierungszeit Enlil-

nādin-šūmis und der des Meli-Šipaks  Jahre liegen.

83 Vgl. BM  Rs.  ff. Vgl. Foster (: ) mit Kollation.
 elam und babylonien in der . hälfte des . jt. v. chr. 

Inšušinaks, wo die Könige aufgelistet sind, die den Tempel des Inšušinak restauriert haben84,
im Zusammenhang mit den hier diskutierten Synchronismen, erhält man folgendes Ergeb-
nis:

Könige der Inschrift Šilhak Inšušinaks85 Synchronismen nach dem „Berliner Brief “
…86 …
Pahhir-Iššan šak Igi-halki älteste Tochter des Kurigalzu I. (vor ) (wahrscheinlich)
Attar-kittah šak Igi-halki Tochter Kurigalzu I. (vor )
Untaš-Napiriša šak Humban-numena Tochter Burna-Buriaš II. (–)
Unpahaš-Napiriša šak Pahhir-iššan
Kidin-Hutran šak Pahhir-iššan
Šutruk-Nahhunte šak Hallutuš-Inšušinak
Kutir-Nahhunte šak Šutruk-Nahhunte

Den dort genannten Kidin-Hutran hat man bislang entweder mit dem im „Berliner Brief “
genannten Sohn des Untaš-Napiriša in Verbindung gebracht oder zur Wiederherstellung der
chronologischen Abfolge Unpahaš-Napiriša und Kidin-Hutran als Söhne des Pahhir-Iššan
noch vor Untaš-Napiriša eingeschoben87. Alle Annahmen beruhen auf der These, dass šak =
„Sohn“ im Elamischen nur den leiblichen Sohn bezeichnet88. M. E. spricht jedoch vieles dafür,
den Kidin-Hutran, der sich nach dem „Berliner Brief “ als „Sohn“ der ältesten Tochter des
Kurigalzu I. und damit des Pahhir-Iššan bezeichnet, mit jenem Kidin-Hutran, Sohn des Pahhir-
Išsan der Inschrift Šilhak-Inšušinaks gleichzusetzen, der ebenfalls nach der dort überlieferten
Abfolge kein leiblicher Sohn sondern lediglich ein Nachfahre dieses Königs sein kann. So stärkt
auch diese Inschrift die These, dass Kidin-Hutran (II.) der Autor des Berliner Briefs ist.
Abschließend ergibt sich folgendes Ergebnis:
Mittelelamische Könige und ihre Beziehungen zu Babylonien
Pahhir-Iššan*°, Sohn des Igi-halki° ∞ älteste Tochter Kurigalzu I. (vor ) (wahrscheinlich)
Attar-kittah*°, Sohn des Igi-halki°
Humban-numena*°, Sohn des Attar-kittah ∞ Tochter Kurigalzu I. (vor )
Untaš-Napiriša*°, Sohn des Humban-numena*° ∞ Tochter des Burna-Buriaš II. (–)
Kidin-Hutran I.*, Sohn des Untaš-Napiriša*° ∞ eine Babylonierin
Napiriša-Untaš*, Sohn des Kidin-Hutran I.*
Unpahaš-Napiriša°, Sohn des Pahhir-Iššan*°
Kidin-Hutran II.*°, Sohn des Pahhir-Iššan*° ∞ älteste Tochter des Meli-Šipak (–)
* = im „Berliner Brief “ genannt
° = in der Inschrift Šilhak-Inšušinaks genannt
˘
Die Analyse zeigt jedoch auch, dass bislang nichts über die Beziehungen beider Länder im
Zeitraum zwischen Burna-Buriaš II. und Kidin-Hutran II. ausgesagt werden kann.

84 Es werden nur die Herrscher erwähnt, die auch an den Tempeln gebaut haben. Vgl. dazu Steve & Vallat (:
).
85 Vgl. König (: Nr.  a und b).
86 Die früheren Könige der altelamischen Zeit, endend mit Kuk-Našur, werden hier nicht aufgeführt.
87 Vgl. dazu oben Anm. . Ältere Vorschläge finden sich in der Übersicht bei Vallat (: –).
88 Vgl. dazu besonders Steve & Vallat (: –).
 susanne paulus

. Kassitische und elamische Feldzüge

Die Beziehungen Babyloniens und Elams lassen sich außerhalb der bereits diskutierten diplo-
matischen Ehen vor allem durch Feldzüge fassen, die in Königsinschriften und Chroniken über-
liefert sind89. Von elamischer Seite am besten belegt ist dabei der Feldzug Šutruk-Nahhuntes I.,
Sohn des Hallutuš-Inšušinaks, der in seinen Inschriften auf babylonischen Beutestücken von
diesem Ereignis berichtet90. Dabei wird sein Gegner auf mesopotamischer Seite jedoch nicht
genannt. Die Identifikation erfolgt auf Grund der neuassyrischen Abschrift eines Texts, der sich
mit der Rückeroberung Babyloniens unter Nabû-kudurrı̄-usur I. (–) beschäftigt, wo
zunächst Ereignisse unter dem kassitischen König [Zababa]-˙ šuma -iddina ()91 aufgeführt
werden, wobei auch hier dessen Gegner auf elamischer Seite nicht erhalten ist, da die entspre-
chende Stelle stark beschädigt ist. Weil jedoch direkt im Anschluss Kutir-Nahhunte genannt
und als „sein Sohn“ bezeichnet wird92, ist sehr wahrscheinlich davor Šutruk-Nahhunte I. zu
ergänzen, der nach elamischen Quellen der Vater Kutir-Nahhuntes ist93. Das heißt, ein Feld-
zug Šutruk-Nahhuntes I. fand vermutlich während der Regierungszeit Zababa-šuma-iddinas
() statt. Im selben Text ist sicher ein Feldzug Kutir-Nahhuntes gegen den letzten kassi-
tischen König Enlil-nādin-ahi (–) belegt. Möglicherweise sind auch Ereignisse, von
˘
denen die „Chronik W“ berichtet, jedoch ohne dass ein Königsname erhalten ist, in diesen
Zeitraum zu datieren94. So ergeben sich zwei Synchronismen:
Šutruk-Nahhunte I., Sohn des Hallutuš-Inšušinaks – Feldzug gegen Zababa-šuma-iddina ()
(wahrscheinlich)
Kutir-Nahhunte, Sohn des Šutruk-Nahhunte I. – Feldzug gegen Enlil-nādin-ahi (–)
˘
Diese sind deutlich nach dem letzten Synchronismus unter Meli-Šipak (–) anzuset-
zen, den der „Berliner Brief “ bot. Das ist chronologisch unproblematisch, da der Übergang von
Kidin-Hutran II. zu Šutruk-Nahhunte I. nach wie vor unklar ist, wobei evtl. zwischen diesen
Königen noch die Regierungszeit Hallutuš-Inšušinaks anzusetzen ist95.
Ansonsten kann von elamischer Seite lediglich eine in Susa gefundene Statue als Beweis
für einen Feldzug des Königs Untaš-Napiriša gegen Babylonien herangezogen werden96. Als
Diskussionsgrundlage wird zunächst im Folgenden auch dieser Text vollständig, jedoch ohne
die strittigen Ergänzungen, wiedergeben. Bei der Statue handelt es sich um das Unterteil einer
antropomorphen Figur, die über einem langen Untergewand einen Mantel mit Fransensaum
trägt. Die akkadische Inschrift lautet:

89 Vgl. dazu oben bei Anm. .

90 König (: Nr. –), vgl. dazu oben Anm.  mit weiterer Literatur.
91 K  ’[ … dza-ba -ba ]. MU .ŠÚMna. Vgl. zu diesem Text mit einer Zusammenstellung älterer Literatur
4 4
Frame (: Nr. B...), siehe auch Foster (, s.o. Anm. ). Aus chronologischen Gründen ist eine Lesung
[Adad]-šuma-iddina auszuschließen. Vgl. so auch Brinkman (: –), besonders Nr. Z...
92 K  ’ … ku-dúr-na-an-hu-un-di bu-uk-ra-šu. Zum Text vgl. oben Anm. .
˘
93 Vgl. dazu die in den Inschriften Kutir-Nahhuntes angegebene Filiation. Vgl. König (: Nr. –). Siehe
auch so in den oben aufgeführten Inschriften Šilhak-Inšušinaks (= König []: Nr. ).
94 BM :–. Vgl. Glassner (: Nr. ). Zur Problematik der Zuordnung dieser Stelle siehe Brinkman

(: ) und Walker (: ).

95 Vgl. zu dieser Problematik Potts (: –). Quintana (: –) nimmt auf Grund seiner Interpre-

tation des „Berliner Briefs“ einen direkten Übergang an, ebenso Goldberg (: ). Auch Steve, Vallat & Gasche
(–: ) sehen keinen Bruch. Es wurde jedoch oben gezeigt, dass der „Berliner Brief “ sich vermutlich
nicht auf Šutruk-Nahhunte I. sondern Kidin-Hutran II. bezieht und daher keine Information über die Abstammung
Šutruk-Nahhunte I. bietet. In den elamischen Quellen wird Šutruk-Nahhunte I. an keiner Stelle mit der Linie Igi-
halki/Pahhir-Iššan in Verbindung gebracht. Lediglich im Kedor-laomer-Text wird eine ähnliche Filiation verwendet,
jedoch auch ohne auf die Linie Pahhir-Iššan/Kurigalzu I. einzugehen, vgl. dazu oben bei Anm. .
96 Vgl. dazu oben Anm. .
 elam und babylonien in der . hälfte des . jt. v. chr. 

 [a-na-ku]97 [un]-taš-dGAL DUMU [Ich], [Un]taš-Napiriša, der Sohn des

d
hu-ban-nu-me-ne Humban-numena,
˘
[EŠŠANA98šu-ši ù] an-za-an dim-mé-ri-ia šùl-mi [der König von Susa und] Anšan, – Immerija ist
mein Wohlergehen –
[X X X99]-[li 100
-ia-šu lu-ú ah-bu-ut-ma i-na […] habe für wahr geraubt und im
˘
[si-ia -an ku-uk lu-ú ú-še-ši-ib-šu [Sija n-kuk aufstellen lassen.
 [ša i-n a-áš-šu- ma a-na ma-at na-ak-ri [Der, der] sie101 [auf heben wird und in ein
feindliches Land
[ired]-du102 ha-at-tum ša dGAL din-šu-uš-na-ak [füh]rt, der Schrecken von Napiriša, Inšušinak
˘
[ù dki]-ri-ri-ša ša si-ja-an ku-uk i-na [und Ki]ririša des Sijan-kuk soll
[mu-uh-hi]103-šu li-iš-ša-ki-in [auf] ihn gelegt sein.
˘ ˘
[šum-ma104 EŠŠ]ANA e-la-an- ti šu-um-ma [Wenn] er ein elam ischer [Kö]nig ist und sie
i-na-áš-ši-šu-ma aufheben wird,
 [li -iš-ši-šu-ma ma-áš-qa-an li-ib-bi-šu so soll er sie aufheben und an den Ort, der ihm
beliebt,
[li -še-ši-ib-šu soll er sie stellen.

In Zeile  hat Scheil zunächst [bi-ti-l]i-ja-aš gelesen, was heute [Kaštil]iaš entspricht, daraufhin
„Immeriya, la sauvegarde de Bitiliyaš, je pris de force“ übersetzt und einen Synchronismus zwi-
schen Untaš-Napiriša und Kaštiliaš propagiert, wobei er die geraubte Statue als die des Gottes
Immerija interpretiert hat105. Dies hat jedoch Reiner bestritten, indem sie statt [Kaštil]iaš [tup-
li]-ja-aš liest, und daraus folgert, dass es sich hierbei nicht um einen Feldzug gegen Babylonien,˙
106
sondern gegen Tupliaš im Bereich von Ešnunna handelte . Dagegen ist zunächst einzuwenden,
dass Tuplijaš in kassitischer Zeit babylonische Provinz war, und zudem ist auch diese Lesung
nicht unumstritten107. Zudem legt die Kleidung der Statue, die eher für Könige als für Götter
verwendet wird108, und die Tatsache, dass Šutruk-Nahhunte I. ebenfalls ausschließlich Statuen
und Stelen von Herrschern, nicht jedoch von Göttern, und „private Kudurrus“ entsprechend

97 Wie im Elamischen üblich, wird die Inschrift durch ein Pronomen der . Person (im Elamischen ú) eingeleitet.

Vgl. so auch in den elamischen Inschriften Untaš-Napirišas (König []: Nr. –), die stets mit u „ich“ beginnen.
98 Es ist vermutlich das elamische Logogramm für König zu ergänzen, vgl. Zeile .
99 Die Zahl der fehlenden Zeichen wird nach den Zeilen  und  sowie  ergänzt, wo jeweils zwischen – Zeichen

fehlen.
100 Eine Lesung des Zeichens als ŠA ist möglich. Eine Lesung als BUR, wie Vallat ():  sie ohne Kommentar

postuliert, ist weder nach Photo bei Scheil (): Tafel  noch nach Kollation möglich. Das Zeichen wird klar
durch einen senkrechten Keil, der auf dem Photo gut zu erkennen ist, abgeschlossen. Dies ist auch nach elamischer
Paläographie nicht für BUR üblich, vgl. Steve (: Nr. ). Das Zeichen entspricht klar entweder Nr.  (ŠA)
oder falls der Anfang weggebrochen ist, Nr.  (LI).
101 Da es sich um eine Statue handelt, wird hier zunächst das neutrale „sie“ gewählt.
102 Sowohl die Präposition ana als auch das finale DU sowie die Tatsache, dass es sich um eine Statue handelt,

legen nahe, dass zu ireddû zu ergänzen ist. Vgl. zu diesem Gebrauch CAD R: f. s.v. redû A b’.
103 Zeile  und  wurden nach einer weiteren akkadischen Inschrift des Untaš-Napriša ergänzt, vgl. König (:

Nr.  B).
104 Die Ergänzung von šumma bietet sich syntaktisch eher an, als die von ša, wie von Scheil (: ) vorgeschla-

gen, da in Zeile  anders als in  und  kein Subordinativ verwendet wird.

105 Scheil (: ).
106 Anm. von E. Reiner in: Rowton (: ).
107 Kritisch zur Lesung äußert sich Brinkman (: ). Zur Lokalisierung von Tupliaš in mittelbabylonischer

Zeit vgl. Nashef (: ). Vgl. auch Paulus (in Druckvorbereitung).
108 Vgl. dazu die Übersicht bei Stiehler-Alegria Delgado (: –) und Antiquaria Tafel A.
 susanne paulus

beschriften ließ109, nahe, dass es sich hier nicht um die Statue eines Gottes sondern eines Königs
handelt. Vallat schließlich interpretiert die Statue nicht als Beutestück, sondern als Weihegabe
des Untaš-Napiriša für seinen Schwiegervater Burna-Buriaš II. und damit als Beweis für die
freundschaftlichen Beziehungen Elams zu Babylonien in dieser Zeit. Dazu ergänzt er Zeile 
[bur-na]-bur-ia-aš und übersetzt „[Moi Un]taš-Napiriša, fils de Humbannumena, [roi de Suse
et] d’Anzan j’ai rapporté (la Statue du dieu) Immiriya pour le bien-être de [Burna]buriaš (et)
je l’ai installée dans le [Siy]an-kuk.“110
Dazu ist zunächst zu bemerken, dass die Lesung von BUR in Zeile  unmöglich ist111.
Demnach scheint die Lesung von [Kaštil iaš gesichert112, wobei davor sicher, wie Durand
vorgeschlagen hat, salam „Statue“ zu ergänzen ist113. Zudem hat das Verb habātum, das Vallat
mit „nehmen“ übersetzt, ˙ ˘
als Grundbedeutung „rauben“ und wird stets lediglich im negativen
Sinn, d.h. in der Bedeutung „wegnehmen, stehlen“ gebraucht, und kann daher nicht mit
einer freundlichen Weihung verbunden werden114. Zudem ist auffällig, dass der Topos des
„Statuenraubs“ in der Fluchformel (–) ausführlich wiederaufgenommen wird. Daher ergibt
sich eine Übersetzung „[Ich], [Un]taš-Napiriša, der Sohn des Humban-numena, [der König von
Susa und] Anšan – Immerija ist mein Wohlergehen – habe fürwahr die [Statue] des [Kaštil iaš
geraubt“. Die schwierige Phrase Immerija šulmı̄ ist dabei entweder, wie Durand vorgeschlagen
hat, als Name der Statue zu bewerten115, oder, was ebenfalls möglich ist, als Epitheton für
Untaš-Napiriša, da Immerija kein typisch babylonischer Gott ist116. Bei der Statue handelt es
sich demnach sehr wahrscheinlich um eine Darstellung des frühkassitischen Königs Kaštiliaš
III117., da Kaštiliaš I. und II. noch vor der Eroberung Babyloniens durch die Kassiten anzusetzen
sind118, während Kaštiliaš IV. (–) deutlich später als Untaš-Napiriša, von dem gezeigt
wurde, dass er die Tochter Burna-buriaš II. (–) geheiratet hat, regiert hat. Kann
man daher, wie Scheil vermutet hat119, von einem Synchronismus zwischen Kaštiliaš III. und
Untaš-Napiriša ausgehen? M.E. wäre dies jedoch genauso, als würde man davon ausgehen, dass
Šutruk-Nahhunte I. und der altakkadische König Maništušu gleichzeitig regiert haben, nur weil

109 Beschriftet wurden die Narām-Sîn-Stele (Sb , vgl. König : Nr. ), die Statuen des Maništušu (Sb , Sb

 = König : Nr.  a+b), zwei Statuen von Herrschern aus Ešnunna (Sb  = König : Nr.  c und Sb ,
zur Inschrift siehe Harper, Aruz & Tallon : ) und eine kassitische Stele Meli-Šipaks (Sb  = König : Nr.
). Zur Tatsache, dass es sich dabei mit großer Wahrscheinlichkeit nicht um einen Kudurru handelt, vgl. Paulus (in
Druckvorbereitung): Nr. MŠ .
110 Vallat (: –).
111 Vgl. dazu oben Anm. .
112 Brinkman (: ) sieht die Schreibung des Namens kaš-ti-li-ia-šu als problematisch an, da der Name

in Babylonien gewöhnlich anders geschrieben wird. Vgl. ders., ff. Jedoch ist auch auf dem ebenfalls in Susa
gefundenen und entgegen der Meinung von Brinkman sicher zeitgenössischen Kudurru (In Susa wurden keine
nachkassitischen Kudurrus gefunden.) die Schreibung kaš-ti-li-ia-a-šu zu finden, vgl. zu diesem Text Paulus (in
Druckvorbereitung): Nr. Ka IV . Zudem handelt es sich um eine elamisch-akkadische Inschrift, wobei der Lautwert
TIL im mittelbabylonischen Syllabar nicht üblich ist. Vgl. Steve (: Nr. ).
113 Durand (: Nr. , ). Es ist nicht klar, ob salam mit Logogramm oder syllabisch geschrieben wurde. In
˙
den Inschriften Šutruk-Nahhunte I. wird eine syllabische Schreibung bevorzugt, vgl. König (: Nr. ).
114 Vgl. AHw, – s. v. habātu(m) mit der Bedeutung „rauben, plündern“ und CAD h, – s.v. habātu A „to

rob, to take away by force“. ˘ ˘ ˘

115 Vgl. Durand (: ). Zu Eigennamen von Statuen etc. siehe die Zusammenstellung bei Radner (: –

).
116 Siehe zur Problematik Schwemer (:  Anm. ).
117 Dazu passt, dass die Darstellung des Herrschers der des frühkassitischen Königs Kadašman-Harbe I. auf dem

Kudurru YBC  (unpubliziert, Publikation durch die Autorin in Vorbereitung) entspricht. ˘
118 Zur Problematik der verschiedenen Kaštiliaš vgl. Brinkman (: –). Kaštiliaš I. ist der ., Kaštiliaš II.

vermutlich der . König der kassitischen Dynastie. Beide sind deutlich vor der Eroberung Babyloniens anzusetzen,
vgl. dazu unten.
119 Vgl. Scheil (: ).
 elam und babylonien in der . hälfte des . jt. v. chr. 

der Elamer dessen Statuen raubte120. Das heißt aber, dass die Statue lediglich einen Feldzug des
Untaš-Napiriša gegen Babylonien dokumentiert, während sein Gegner auf kassitischer Seite
nicht feststeht.
Untaš-Napiriša, Sohn des Humban-numena – Feldzug gegen Babylonien
Jedoch lässt sich dieser Feldzug gut mit dem oben im „Berliner Brief “ festgestellten Bruch der
diplomatischen Beziehungen nach Untaš-Napiriša und möglicherweise auch mit dem Phäno-
men der „Elamisierung“, die mit Untaš-Napiriša einsetzt und sowohl Sprache der Inschriften
als auch Götterwelt betrifft121, verbinden.
Von babylonischer Seite wird von weiteren Feldzügen vor allem in der sogenannten „Chronik
P“ berichtet. Unbestritten sind dabei die Überfälle des Kidin-Hutran II. unter Enlil-nādin-šumi
() und Adad-šuma-iddina (–)122, die, wie oben gezeigt, mit großer Wahrschein-
lichkeit dem Autor des „Berliner Briefs“ zugewiesen werden können.
Kidin-Hutran II., „Sohn“ des Pahhir-Iššan – Feldzug gegen Enlil-nādin-šumi ()
Kidin-Hutran II., „Sohn“ des Pahhir-Iššan – Felzug gegen Adad-šuma-iddina (–)
Daneben berichtet die Chronik davor von einem einem weiteren babylonischen Feldzug,
nämlich dem des Kassiten Kurigalzu gegen den elamischen König Hurba-tela123. Gewöhnlich
wurde diese Passage Kurigalzu II. (–) zugeordnet124. Steve, Vallat & Gasche haben dem
widersprochen und auf Grund von Schwierigkeiten mit der „Chronik P“ postuliert, der Feldzug
könne auch Kurigalzu I. (vor ) zugewiesen werden125.
Es ist unumstritten, dass die „Chronik P“ ein spätbabylonischer Text mit zahlreichen Über-
lieferungsfehlern und Schwierigkeiten ist126. Jedoch kann m. E. die Frage, welchem Kurigalzu
die betreffende Passage zuzuordnen ist, beantwortet werden. Der erhaltene Teil der Chronik
beginnt zunächst mit einer Passage (I"ff.) zu Kadašman-Harbe, der einerseits korrekt als Sohn
˘ Mulballitat-Šerūa, Tochter Aššur-
des Kara-indaš127, andererseits fälschlicherweise als Sohn der
uballits I. (–), bezeichnet wird. Das Problem liegt darin, dass˙ die Chronik Kara-indaš
˙
(. König) und Kara-har-daš (. König), letzterer wahrscheinlich Kara-kindaš zu lesen und
˘
wirklich ein Sohn Mulballi tat-Šerūas128, verwechselt. Die Passage zu Kadašman-Harbe kann
˙
jedoch jetzt, da sie sich fast wörtlich auf einem Kudurru Kadašman-Harbes I. befindet, ˘ sicher
129 ˘
diesem König zugeordnet werden . Anschließend (ab I’ff.), klar durch arkānu „später“ abge-
trennt, folgt eine Passage, in der beschrieben wird, wie der Sohn der Mulballitat-Šerūa in
einer Revolte getötet und ein Usurpator auf den Thron gesetzt wird, was das Eingreifen ˙ sei-
nes assyrischen Großvaters Aššur-uballits I. nötig machte, der wiederum Kurigalzu auf den
˙

120 Vgl. dazu oben Anm. .

121 Vgl. Steve, Vallat & Gasche (–: –), siehe auch Potts (: –).
122 Chronik P iv –. Vgl. Glassner (: Nr. ).
123 Chronik P iii –. Vgl. Glassner (: Nr. ).
124 Vgl. so Brinkman (: –); ebenso Glassner (: –).
125 Vgl. Steve, Vallat & Gasche (–: ): „Dans ces conditions, cette campagne peut très bien être attribuée

au premier du nom[= Kurigalzu I.].“

126 Vgl. dazu Röllig (: –) und Brinkman (: –).
127 Kara-indaš ist nach der rekonstruierten Abfolge der kassitischen Könige an ., Kadašman-Harbe I. an .

Stelle anzusetzen. Vgl. Brinkman (: ). ˘

128 Vgl. zur Problematik Brinkman (: – und ).
129 Es handelt sich um YBC . Der Text lässt sich aus inhaltlichen (wichtig ist hierbei vor allem die Erwähnung

des Falls der . Dynastie von Babylon unter Samsu-ditāna) und orthographischen Merkmalen klar Kadašman-
Harbe I. zuordnen. Die entsprechende Passage lautet (i ) ka-mar su-ti-i () ra-ap-šu-ú-ti () iš-tu ma-ti () sí-it
d˘UTUši () an ma- ti () e-rib d UTUši () iš-ku-nu-ma () we-de-em () EN NU re-e-eh () ir -si-bu-šu- ˙
˘
nu-ši-im. Nahezu parallel ist „Chronik P“ i ’ff. Siehe auch hier zu einigen offensichtlichen Fehlern der „Chronik P“
Röllig (: ).
 susanne paulus

babylonischen Thron setzt. Diese Episode ist außerhalb der „Chronik P“ noch in der „Syn-
chronistischen Geschichte“ überliefert130. Jedoch weist die „Chronik P“ im Vergleich zu den
anderen Quellen wiederum einige, vermutlich fehlerhafte Unterschiede auf: Der Ursupator
wird in „Chronik P“ Šuzigaš statt Nazi-Bugaš genannt, und der von Aššūr-uballit I. auf den
Thron gesetzte König statt wie in der „Synchronistischen Geschichte“ Kurigalzu II., ˙ Sohn des
Burna-Buriaš II., als Kurigalzu I., Sohn des Kadašman-Harbe I., bezeichnet131. Das bedeutet
˘
jedoch keinesfalls, dass diese Passage in der „Chronik P“ Kurigalzu I. zuzuordnen ist, sondern,
wie die Parallelüberlieferung zeigt, wird hier in der „Chronik P“ lediglich ein bestehender Feh-
ler weitergetragen.
Nach einer längeren, unklaren Passage folgt in III  ff. die „elamische Episode“, die im
Folgenden diskutiert werden soll. Danach schließt „Chronik P“ mit einer Episode ab, die von
einem Feldzug Kurigalzus gegen den assyrischen König Adad-nērārı̄ (–) berichtet (iii
ff.). Auch dieses Ereignis ist in der „Synchronisten Chronik“ und einer weiteren assyrischen
Chronik überliefert. Der Feldzug wird dort jedoch gegen Enlil-nērārı̄ (–) statt Adad-
nērārı̄ geführt, was mit unserer Rekonstruktion der Chronologie dieser Zeit übereinstimmt132.
Das bedeutet jedoch, dass die „Chronik P“ an keiner Stelle Ereignisse aus der Regierungszeit
Kurigalzu I. aufführt. Es ist daher äußerst unwahrscheinlich, dass zwischen zwei Passagen, die
klar Kurigalzu II. betreffen, eine weitere, die Kurigalzu I. zuzuordnen ist, eingefügt wurde. Das
bedeutet weiterhin, dass der Feldzug gegen Elam mit Sicherheit unter Kurigalzu II. (–)
datiert. Es ist dabei durchaus möglich, dass es sich hierbei um eine Reaktion auf den früheren
Feldzug Untaš-Napirišas handelt.
Das Problem ist, dass Kurigalzus Gegner Hurba-tila auf elamischer Seite bislang nicht belegt
ist. Er wird in der Chronik als LUGAL kure-lam-mat bezeichnet133, was von Gassan, ihm folgend
dann auch von Goldberg und Vallat, so interpretiert wurde, dass es sich bei dem genannten
Hurba-tila nicht um einen König von Elam, sondern von dem davon abzugrenzenden Elammat
handelt134. Dabei nehmen beide in Kauf, dass in derselben Passage erwähnt wird, dass Kurigalzu
gegen Elam (kurNIMki) zog135. Das Problem liegt jedoch darin, dass das Logogramm NIM.(MA),
im Babylonischen Elamtu, spätbabylonisch spielerisch auch Elammat gelesen wird136. In der
„Chronik P“ wird abwechselnd das Logogramm und die syllabische Schreibung gebraucht.
Das bedeutet, dass Hurba-tila ein König Elams war, wobei man jedoch durchaus aus den
Erfahrungen mit der „Chronik P“ davon ausgehen kann, dass der Name stark abweichend
geschrieben oder vertauscht wurde. Dennoch kann man von folgendem „Synchronismus“
ausgehen:
Kurigalzu II., Sohn des Burna-Buriaš (–) – Feldzug gegen Hurba-tila
Daran anschließend müssen nun noch zwei kassitische Königsinschriften diskutiert werden,
die ebenfalls die Eroberung von Susa bzw. Elam zum Thema haben. Die erste befindet sich auf
der Schulter einer Statue, die in Susa gefunden wurde137. Sie ist auf Sumerisch verfasst und lautet:

130 Vgl. Synchronistische Chronik I ’ff. (vgl. Glassner [s.o. Anm. ], Nr. ).
131 Vgl. zur Problematik Brinkman (: ff.).
132 Vgl. Synchronistische Chronik I ’ff. und ein Fragment (Glassner : Nr. ). Vgl. Röllig (: –).
133 Vgl. Chronik P III ,  und .
134 Vgl. Gassan (: –); Goldberg (: ) und Steve, Vallat & Gasche (–: ).
135 Vgl. Chronik P III .
136 Vgl. zu dieser Problematik zuletzt Michalowski (: ). Die entsprechenden Belege hat Zadok (: –

) klar zusammengestellt. Es spricht nichts dagegen, dass das Logogramm kurNIM.MAki spätbabylonisch nicht
Elammat gelesen wird.
137 Scheil (: Nr. ). Für weitere Literatur zu diesem Text siehe Stein (: Nr. ), vgl. auch Brinkman (:

Nr. Q..).
 elam und babylonien in der . hälfte des . jt. v. chr. 

 ku-ri-gal-zu Kurigalzu,
lugal-kiš der König der Gesamtheit,
saĝ-ĝiš-ra der geschlagen hat
mùš -e[re]nki Susa
 nim ki-ma-bi-da und Elam
en -na-zag138 bis zur Grenze
[mar]-ha-šiki139 von [Mar]haši.
˘ ˘
Vallat zieht diese Statue als Beweis für eine Eroberung Susas durch Kurigalzu I. (vor )
heran140, dagegen hat unlängst Bartelmus argumentiert, dass die Statue vermutlich Kurigalzu II.
(–) zuzuweisen ist141. Als Grundlage dafür dient, dass in dieser Inschrift der Titel
LUGAL KIŠ verwendet wird, der ansonsten in den von ihr untersuchten sumerischen Königs-
inschriften Kurigalzu I. nicht belegt sei. Dabei übersieht sie jedoch, dass der Titel in einer
akkadischen Königsinschrift, die durch Filiation klar Kurigalzu I. zuzuordnen ist, unzweifel-
haft belegt ist142. Zwar ist der Text nur in späteren Abschriften überliefert, die Authentizität ist
jedoch m.E. gewährleistet143, so dass es wenig Sinn macht, diesen Text nicht in die Betrachtung
der Titel Kurigalzu I. einzubeziehen.
Der zweite Text ist eine in Nippur gefundene Achattafel, die auf der Vorderseite eine Wei-
hung zu Gunsten des neusumerischen Königs Šulgi und auf der Rückseite folgende Inschrift
enthält144:

 ku-ri-gal-zu Kurigalzu,
LUGAL ka-ru-du-ni-ja-aš der König von Babylonien,
É.GAL ša URU ša-a-šaki hat den Palast der Stadt Šâša (= Susa)
ša NIM.MAki von Elam
 ik-šu-ud erobert und
a-na dnin-líl (es) der Ninlil,
be-el-ti-šu seiner Herrin,
a-na ba-la-ti-šu für sein Leben
i-qí-iš ˙ geschenkt.

Bartelmus hat diese Inschrift ebenfalls fragend mit Kurigalzu II. in Verbindung gebracht,
wobei als Argumente die Sprache „akkadisch“ und ihre Zuweisung des oben zitierten Texts
an Kurigalzu II. als Argument dienen145. Es wurde jedoch bereits gezeigt, dass für Kurigalzu I.
ebenso wie andere frühkassitische Könige Abschriften von akkadischen Königsinschriften
überliefert sind146. Eine weitere Inschrift Kurigalzus, möglicherweise Teil einer elamischen
Beute, wurde ebenfalls in Susa gefunden. Die Inschrift auf der Achatperle lautet147:

138 Zur Lesung dieser Zeile vgl. Deller & Postgate (: ).
139 Vgl. hierzu Steinkeller (:  Anm. ).
140 Vallat (: ).
141 Bartelmus (:  Anm. ).
142 BM  (i ) mku-ri-gal-zu LUGAL GALú () LUGAL dan-nu LUGAL KIŠ … () IBILA šá m kad -dáš-man-

har - be . Vgl. zu diesem Text zuletzt mit Angabe früherer Literatur Paulus (: –).
˘ 143 Vgl. die Studie bei Paulus (: –).
144 CBS . Kopie bei Hilprecht (: Nr.  und ). Vgl. Zu diesem Text zuletzt Radner (: –).

Vgl. Brinkman (: Nr. Q..).

145 Vgl. Bartelmus (:  Anm. ).
146 Vgl. dazu neben dem in Anm.  zitierten Text weitere Beispiele bei Stein (: –).
147 Vgl. Scheil (: ). Siehe auch Brinkman (: Nr. Q..).
 susanne paulus

d
 KA.DI Ištaran
ku-ri-gal-zu hat Kurigalzu
BA es geschenkt148.

M.E. ist es bislang unmöglich, die Inschriften definitiv einem bestimmten Kurigalzu zuzu-
weisen. Die beiden ersten können mit der in Chronik P erwähnten Eroberung Elams unter
Kurigalzu II. in Verbindung stehen. Vergleicht man jedoch die in Susa gefundene Statuen-
inschrift mit der Weihinschrift aus Nippur, scheint erstere archaischeren Charakters zu sein.
Dafür spricht vor allem die Erwähnung von Marhaši, das ansonsten, wie Steinkeller gezeigt
˘ Grundlage einen Feldzug Kurigalzus I.
hat, letztmalig altbabylonisch belegt ist149. Auf dieser
nach Elam zu rekonstruieren, ist jedoch reine Spekulation.

. Der „Synchronismus“ von Haft Tepe

Da die Eroberung Elams durch Kurigalzu I. (vor ) nicht nachgewiesen werden konnte,
wurden hier bislang keine Synchronismen diskutiert, die älter als die Hochzeit zwischen
Pahhir-Iššan und der ältesten Tochter Kurigalzus I. datieren. Wie bereits eingangs erwähnt,
˘ ˘ hier vor allem die Quellen aus Haft Tepe150, jedoch auch die unlängst veröffentlichten
deuten
Texte der in Babylonien herrschenden Meerlanddynastie151 auf enge Beziehungen zu Elam hin.
Mit der Veröffentlichung von Steve, Gasche & de Meyer () wurde der Blick auf den Beginn
der mittelelamischen Zeit, der in Susa auf die Schichten XII und XI datiert, gerichtet152. Der dort
identifizierte König von Susa und Anšan, Kidinû, gab der „Dynastie“ den Namen Kidinûiden,
zu der heute verschiedene Herrscher gezählt werden153. Doch lediglich drei der Dynastie
zugeordneten Könige Kidinû, Tan-Ruhuratir II. und Tepti-ahar tragen den entscheidenden
Titel „König von Susa und Anšan“, während für Inšušinak-sunkir-nappipir lediglich der Titel
„König von Susa“ belegt ist154. Sowohl die Abfolge als auch die Datierung dieser Könige sind
stark umstritten und können hier nicht diskutiert werden155.
Von großem Interesse für die zeitliche Einordnung dieser Herrscher war von jeher ein
„Synchronismus“ zwischen Tepti-ahar, welcher als König vor allem auch durch die Bauten von
Haft Tepe bekannt ist, und einem kassitischen König, rekonstruiert aus einem der Jahresnamen
aus Haft Tepe, der sich auf einer Tafel befindet, deren Siegelinschrift den König Tepti-ahar
nennt156. Der Jahresname lautet:

 MU EŠŠANA ka-da-aš-ma-an-dKUR.GAL Jahr: Der König hat Kadašman-KUR.GAL

ú-sà-ah-hi-ru zurückgewiesen/ ausgewiesen hat157.
˘ ˘
Bereits Herrero brachte den dort genannten Kadašman-KUR.GAL in Zusammenhang mit
kassitischen Königen158, nämlich Kadašman-Enlil I. (. König, –) und II. (–

148
Die Kürze der Inschrift ist durch die geringe Größe des Objekts bedingt.
149
Vgl. Steinkeller (: ).
150 Siehe oben bei Anm. .
151 Dalley (: –).
152 Steve, Gasche & de Meyer (: –).
153 Vgl. Steve, Gasche & de Meyer (: –).
154 Vgl. Steve, Vallat & Gasche (–: ).
155 Siehe u. a. Steve, Vallat & Gasche (–: –); Potts (: –); Glassner (:–).
156 Herrero (: Nr. ).
157 Die Übersetzung folgt der Grundbedeutung von sahāru im D-Stamm, vgl. AhW, – s.v. sahāru(m)

und CAD S, – s. v. sahāru. ˘ ˘

158 Herrero (: ). ˘
 elam und babylonien in der . hälfte des . jt. v. chr. 

)159. Nimmt man an, dass es sich um den älteren König dieses Namens handelt, ergibt
sich ein Synchronismus zwischen Tepti-ahar und Kadašman-Enlil I. Das würde bedeuten, dass
Tepti-ahar nach den oben aufgezeigen Synchronismen gleichzeitig mit den frühen Igihalkiden
anzusetzen ist, was unwahrscheinlich wäre160. Die Annahme, dass das Logogramm dKUR.GAL
innerhalb des Namens Enlil zu lesen ist, geht auf einen in Susa gefundenen Bilingue eines
sumerischen Königsbriefes zurück, wo im Sumerischen den-líl im Akkadischen dKUR.GAL
steht161. Daneben ist in Elam die Schreibung Enlil für diesen Gott weit verbreitet162. In zeit-
genössischen mittelbabylonischen Texten steht dKUR.GAL gewöhnlich für den Gott Amurru,
was daran festgemacht wird, dass dKUR.GAL im gleichen Archiv alternierend für dMAR.TU
= Amurru gebraucht wird163. So scheint mir, die Argumentation von Glassner, der für Haft
Tepe eine Lesung dKUR.GAL = Amurru aus dem Grund ausschließt, weil in diesen Texten
ebenfalls dMAR.TU als Logogramm belegt ist, nicht zwingend164. Gerade bei den von ihm
zusätzlich aufgeführten Personennamen aus Haft Tepe, Ibni-dKUR.GAL und Arad-dKUR.GAL,
ist für Ibni-dKUR.GAL in Babylonien sowohl die Schreibung mit dKUR.GAL als auch dMAR.TU
belegt165.
Um dieses chronologische Problemzu umgehen, haben Cole und de Meyer vorgeschlagen,
Kadašman-dKUR.GAL als Kadašman-Harbe zu lesen und diesen mit Kadašman-Harbe I., der
˘
unmittelbar vor Kurigalzu I. regierte, gleichzusetzen166
. Ihre Argumentation beruht˘darauf, dass
der kassitische Gott Harbe in babylonischen Texten mit den Göttern An und Enlil geglichen
wird167, wobei in Texten˘ des . Jt. wiederum die Gleichung dKUR.GAL = An/Enlil belegt ist168.
Oelsner macht zur Lesung von dKUR.GAL folgende Aussage: „Der neue Beleg zeigt, daß
d
KUR.GAL, in der Regel = (Gott) Amurru, offensichtlich für verschiedene Götter stehen kann.
Das erschwert die Lesung im Einzelfall.“169 Kann also dKUR.GAL in Haft Tepe für den Gott
Enlil oder gar für den kassitischen Gott Harbe stehen? Es ist möglich, jedoch sehr unsicher
˘
und m.E. mehr unserem Wunsch, einen Synchronismus zur rekonstruieren, geschuldet denn
der Quellenlage. Sieht man sich den hier bretreffenden Jahresnamen von Haft Tepe nämlich
in seinem Kontext, das heißt im Rahmen der anderen Jahresnamen von Haft Tepe an, ergibt
sich ein völlig anderes Bild. Neben den Jahresnamen, denen kultische Bauaktivitäten zugrunde
liegen170, sind vor allem folgende bemerkenswert:

159 Vgl. zu diesen Königen Brinkman (: –).

160 Vgl. dazu Glassner (: ).
161 Vgl. Edzard (: Nr.  I und I).
162 Vgl. hierzu die Beispiele bei Glassner (:  Anm. ).
163 Vgl. dazu Brinkman (: ). Vgl. auch die entsprechenden bei Hölscher () unter „Amurru“ aufge-

führten Belege. Dies beweisen auch die Variationen in Šurpu, wo ebenfalls dMAR.TU und dKUR.GAL alternierend
gebraucht werden. Vgl. Borger (:  [Tafel VIII ]).
164 Glassner (: –).
165 Vgl. Hölscher (: ). Hölscher führt gewöhnlich auch dKUR-Belege unter Amurru auf, was vermutlich

nicht richtig ist. Vgl. für dieses Logogramm die Diskussion bei Borger (:).
166 Vgl. Cole & de Meyer (: –). Vgl. Dazu auch die Erwiderung Glassner (: ).
167 Vgl. zum Gott Harbe in dieser Tradition Balkan (: –).
˘
168 Vgl. hierzu besonders die unterschiedlichen Lesungsvorschläge von Lambert (: ) mit einem eindeutigen
Beleg für die Gleichung dKUR.GAL = dEnlil und Oelsner (: ), der auf Grund eindeutiger spätbabylonischer
Belege aus Uruk für die Gleichung dKUR.GAL = dAn die Lesung An für einen bei Lambert (: –) aufgeführ-
ten Beleg vorschlägt. Vgl. auch Cole & de Meyer (: –).
169 Oelsner (: ).
170 Vgl. zu den über  Jahresnamen die Übersicht bei Glassner (: –).
 susanne paulus

MU ì-lí-bar-na uš-tu Jahr: Ilı̄-barna171 ist aus

ka-ra-an-du-ni-ia-a-aš Babylonien
il-li-ka gekommen172.
MU at-ta-[…] Jahr: Atta-[…]173
a-na k[a-ra-an-du-ni]- ist nach B[abylo]nien
i-a-aš [il-li-ku] [gegangen]174.
[MU EŠŠANA mNUNUZ]-i-dIŠKUR [Jahr: Per"]i-Adad hat sich beim
i-mi-du-šu [König] aufgehalten175.
MU EŠŠANA i -na ku-tal-li176 Jahr: Der König (hat) a ls Ersatz
m
NUNUZi-dMAR.TU DUMU sí-ip-ri für Per"i-Amurru177, Boten
šà ka-ra-an-du-ni-i-a-aš von Babylonien (…)178

Diese Jahresnamen dokumentieren klar Ankunft und Abreise sowie Aufenthalt von Boten
aus Babylonien und Elam179. In diesem Zusammenhang scheint es unwahrscheinlich, dass
„das Jahr: der König hat Kadašman-dKUR.GAL zurückgewiesen/ausgewiesen“, sich auf das
Zurückweisen von Truppen bezieht, wie Glassner vorgeschlagen hat180, zumal suhhuru vor
˘ ˘ dieser
allem auch für das Zurückweisen/Ausweisen einzelner Personen belegt ist181. M. E. stellt
Jahresname lediglich einen weiteren Beleg für die diplomatischen Beziehungen und keinen
Synchronismus mit Babylonien dar, so dass die Einordnung der Kidinûiden von babylonischer
Seite bislang nicht möglich ist.

. Zusammenfassung

Aus den hier untersuchten Quellen ergibt sich nun zusammenfassend folgendes Bild für die
Beziehungen Elams zu Babylonien in kassitischer Zeit: Für die frühkassitische Zeit können
Beziehungen durch Botenkontakt mit Elam angenommen werden, jedoch ist eine präzise Ein-
ordnung nicht möglich, da der sogenannte „Synchronismus von Haft Tepe“ entkräftet wurde.
Die erste uns bekannte diplomatische Ehe besteht dann zwischen der ältesten Tochter Kuri-
galzu I. (vor ) und dem Igihalkiden Pahhir-Iššan. Anschließend geht Humban-numena
ebenfalls eine Ehe mit einer Tochter Kurigalzus I. ein. Der aus dieser Beziehung hervorge-
hende Untaš-Napiriša ehelicht eine Tochter Burna-Buriaš II. (–), gleichzeitig ist für
diesen Elamer jedoch auch ein Feldzug gegen Babylonien belegt. Für den nachfolgend im
„Berliner Brief “ genannten Herrscher Kidin-Hutran I. ist nur bekannt, dass er eine Babylo-
nierin geheiratet hat. Es kommt zum Bruch der Beziehungen. Von babylonischer Seite ist unter

171 Ilı̄-barna ist ebenfalls in den Texten aus Haft Tepe häufig belegt. Vgl. Glassner (:  Anm. ). Mögli-

cherweise handelt es sich um einen kassitisch-akkadischen Mischnamen.

172 Der Jahresnamen ist in Herrero (: Nr.  und ) belegt.
173 Der Name ist sehr wahrscheinlich elamischen Ursprungs. Vgl. dazu die entsprechenden Belege bei Hinz &

Koch (: –).

174 Der Jahresnamen ist in Herrero (: Nr. ) belegt.
175 Der Jahresname ist bei Herrero & Glassner (: Nr. ) belegt.
176 ina kutalli hat im Zusammenhang mit Personen diese Bedeutung. Vgl. dazu u.a. CAD K:  s.v. kutallu.
177 Per"i-Amurru ist in den Texten von Haft Tepe hinreichend belegt, vgl. Glassner (:  Anm. ). Es handelt

sich um einen in Babylonien gut bekannten Personennamen, vgl. Hölscher (: ).
178 Der Jahresname ist in Herrero (: Nr. ) belegt. Es handelt sich m.E. um einen Kurznamen. Vgl. auch

Glassner (:  Anm. ).

179 So auch Glassner (: ).
180 Glassner (: ): „le roi repoussant par deux fois des troupes étrangèrs.“. Der zweite Beleg ergibt sich, da

ú-sà-ah-hir-ru in einer weiteren Urkunde (HT ) belegt ist.

˘ ˘ oben Anm. .
181 Vgl.
 elam und babylonien in der . hälfte des . jt. v. chr. 

Kurigalzu II. (–) ein Feldzug gegen den elamischen König Hurba-tela überliefert, die-
ser konnte bislang jedoch auf elamischer Seite noch nicht identifiziert werden. Kidin-Hutran II.,
ein Nachfahre Pahhir-Iššans, bezeichnet sich dann im „Berliner Brief “ als Nachkomme jener
ältesten Tochter Kurigalzus I. und begründet so sein Eingreifen in die babylonische Thron-
folge, nachdem Babylonien durch den Assyrer Tukultı̄-Ninurta I. (–) erobert wurde.
Von Kidin-Hutran II. sind zwei Feldzüge unter den Königen Enlil-nādin-šumi () und
Adad-šuma-iddina (–) belegt und selbst nachdem er die älteste Tochter Meli-Šipaks
(–) geheiratet hat, bekräftigt er noch seinen Anspruch auf den babylonischen Thron.
Die Kassitendynastie wird dann durch die Feldzüge der Šutrukiden Šutruk-Nahhunte I. gegen
Zababa-šuma-iddina () und seines Sohnes Kutir-Nahhunte gegen Enlil-nādin-ahi (–
˘
) zu Fall gebracht. Alle weiteren möglichen Kontakte sind nach momentaner Quellenlage
reine Spekulation.

Bibliographie

Aruz, J., Bezel, K. & Evans, J.M. (Hrsg.) (): Beyond Babylon. Art, Trade, and Diplomacy in the Second
Millennium B.C., New Haven – London.
Balkan, K. (): Kassitenstudien. . Die Sprache der Kassiten (AOS ), New Haven.
Bartelmus, A. (): „Restoring the Past. A Historical Analysis of the Royal Temple Building Inscripti-
ons from the Kassite Period,“ KASKAL , –.
Boese, J. (): „Burnaburiaš II., Melišipak und die mittelbabylonische Chronologie“, UF , –
.
Borger, R. (): „Šurpu II, III, IV und VIII in „Partitur““, A.R. George & I.L. Finkel (Hrsg.), Wisdom,
Gods and Literature. Studies in Assyriology in Honour of W.G. Lambert, Winona Lake, –.
———. (): Mesopotamisches Zeichenlexikon. Zweite, revidierte und aktualisierte Auflage (AOAT ),
Münster.
Brinkman, J.A. (): „Ur: ‚The Kassite Period and the Period of the Assyrian Kings‘“, Or. , –.
———. (): „Foreign Relations of Babylonia from  to  B.C.: The Documentary Evidence“, AJA
, –.
———. (): A Catalogue of Cuneiform Sources pertaining to specific monarchs of the Kassite Dynasty
(MSKH ), Chicago.
———. (–): „Kassiten“, RlA , –.
———. (): „Administration and Society in Kassite Babylonia“, JAOS , –.
———. (): „The Use of Occupation Names as Patronyms in the Kassite Period: A Forerunner of
Neo-Babylonian Ancestral Names?“, A.K. Guinan et al. (Hrsg.), If a Man Builds a Joyful House:
Assyriological Studies in Honor of Erle Verdun Leichty (CM ), Leiden – Boston, –.
Clayden, T. (): „Kurigalzu I and the Restoration of Babylonia“, Iraq , –.
Cohen, R. & Westbrook, R. (Hrsg.) (): Amarna Diplomacy. The beginnings of International Relations,
Baltimore – London.
Cole, S.W. & de Meyer, L. (): „Tepti-ahar, King of Susa, and Kadašman-dKUR.GAL“, Akkadica ,
–. ˘
Dalley, S. (): Babylonian Tablets form the First Sealand Dynasty in the Schøyen Collection (CUSAS ),
Bethesda.
Deller, K. & Postgate, N. (): „Nachträge und Verbesserungen zu RGTC “, AfO , –.
Dossin, G. (): „Bronzes Inscrites du Luristan de la Collection Foroughi“, IrAnt , –.
Durand, J.-M. (): „MDP : “, NABU  Nr. , .
Edzard, D.O. (): „Deux lettres royales d’Ur III en sumérien „syllabique“ et pourvu d’une traduction
accadienne“, R. Labat, Textes littéraires de Suse (MDP ), Paris, –.
Foster, B. (): Before the Muses. An Anthology of Akkadian Literature, Bethesda.
Frame, G. (): Rulers of Babylonia. From the Second Dynasty of Isin to the End of Assyrian Domination
(–BC) (RIMB ), Toronto – Buffalo – London.
Gasche, H., Armstrong, J.A., Cole, S.W. & Gurzadyan, V.G. (): Dating the Fall of Babylon. A
Reappraisal of Second Millennium Chronology (MHE/M ), Ghent – Chicago.
Gassan, M. (): „Hurpatila, roi d’Elammat“, AION , –.
 susanne paulus

Gesche, P. (): Schulunterricht in Babylonien im ersten Jahrtausend v. Chr. (AOAT ), Münster.
Glassner, J.-J. (): „Les textes de Haft Tepe, la Susiane et l’Elam au ème millenaire“, L. de Meyer &
H. Gasche, Mésopotamie et Elam Actes de la XXXVIème Rencontre Assyriologique Internationale, Gand,
– juillet  (MHEOP ), Gent, –.
———. (): „dKUR.GAL à Suse et Haft-Tépé“, NABU  Nr. , .
———. (): Mesopotamian Chronicles (SBL Writings from the Ancient World ), Atlanta .
Goldberg, J. (): „The Berlin Letter, Middle Elamite Chronology and Šutruk-Nahhunte I’s Genealogy“,
IrAnt , –.
Grayson, A.K. (–): „Königslisten und Chroniken“, RlA , –.
Harper, P.O. Aruz, J. Tallon, F. (Hrsg.) (): The Royal City of Susa. Ancient Near Eastern Treasures in
the Louvre, New York.
Herrero, P. (): „Tablettes Administratives de Haft-Tépé“, CDAFI , –.
Herrero, P. & Glassner, J.-J. (): „Haft-Tépé: Choix de Textes I“, IrAnt , –.
Hilprecht, H.V. (): Old Babylonian Inscriptions Chiefly from Nippur (BE /), Philadelphia.
Hinz, W. & Koch, H. (): Elamisches Wörterbuch (AMI. Erg. ), Berlin.
Hölscher, M. (): Die Personennamen der kassitenzeitlichen Texte aus Nippur (IMGULA ), Münster.
King, L.W. (): Babylonian Boundary-Stones and Memorial-Tablets in the British Museum [= BBSt],
London.
König, F.W. Die elamischen Königsinschriften (= EKI) (AfO. Beih. ), Graz.
Lambert, W.G. (): „Ancestors, Authors and Canonicity“, JCS , –.
Leick, G. (Hrsg.) (): The Babylonian World, New York – London.
Liverani, M. (): International Relations in the Ancient Near East, –BC, Houndsmills – New
York.
Michalowski, P. „Observation on „Elamites“ and „Elam“ in Ur III Times“, P. Michalowski (Hrsg.), On the
Third Dynasty of Ur. Studies in Honor of Marcel Sigrist (JCS/S ), Boston, –.
Moran, L. (): The Amarna Letters, Baltimore – London.
Nashef, K. (): Die Orts- und Gewässernamen der mittelbabylonischen und mittelassyrischen Zeit
(RGTC ), Wiesbaden.
Oelsner, J. (): „Hunger, Hermann, Spätbabylonische Texte aus Uruk. Teil I“, OLZ , –.
Paulus, S. (): „„Ein Richter wie Šamaš“ – Zur Rechtsprechung der Kassitenkönige“, ZAR , –.
———. (): „Verschenkte Städte – Königliche Landschenkungen an Götter und Menschen“, L. Kogan
et al. (Hrsg.), Proceedings of the e Rencontre Assyriologique Internationale: City Administration in the
Ancient Near East (Babel & Bibel ), Winona Lake, –.
———. (): „Foreigners under Foreign Rulers – The Case of Kassite Babylonia (nd half of the
nd millennium BC)“, R. Achenbach, R. Albertz & J. Wöhrle (Hrsg.), The Foreigner and the Law.
Perspectives from the Hebrew Bible and the Ancient Near East (BZAR), Wiesbaden (im Druck).
———. (in Druckvorbereitung): Die babylonischen Kudurru-Inschriften von der kassitischen bis zur früh-
neubabylonischen Zeit – Untersucht unter besonderer Berücksichtigung gesellschafts- und rechtshistori-
scher Fragestellungen.
Pedersén, O. (): Archive und Bibliotheken in Babylon. Die Tontafeln der Grabung Robert Koldeweys
– (ADOG ), Saarwellingen.
Potts, D.T. (): The Archeology of Elam. Formation and Transformation of an Ancient Iranian State,
Cambridge.
———. (): „Elamites and Kassites in the Persian Gulf “; JNES , –.
Quintana, E. (): „Filiation et accès au pouvoir en Elam (e moitié IIème millénaire a.C.)“, Mundo
Elamita /, – (www.um.es/ipoa/cuneiforme/elamita).
———. (): „Filiacion y acceso al trono en Elam (a mitad II milenio a.C.)“, Mundo Elamita /,
– (www.um.es/ipoa/cuneiforme/elamita).
Radner, K. (): Die Macht des Namens. Altorientalische Strategien zur Selbsterhaltung (SANTAG ),
Wiesbaden.
Röllig, W. (): „Die Glaubwürdigkeit der Chronik P“, Heidelberger Studien zum Alten Orient. Adam
Falkenstein zum . September , Wiesbaden, –.
Rowton, M.B. (): „Ancient Western Asia“, in I.E.S. Edwards, C.J. Gadd & N.G.L. Hammond (eds.),
CAH 3, Vol. I/: Prolegomena and Prehistory, Cambridge, –.
Scheil, V. (): Textes élamites-sémitiques. Troisièmes Série (MDP ), Paris.
———. (): Textes élamites-sémitiques. Quatrième Série (MDP ), Paris.
———. (): Mélanges Épigraphiques (MDP ), Paris.
 elam und babylonien in der . hälfte des . jt. v. chr. 

Schwemer, D. (): Die Wettergottgestalten Mesopotamiens und Nordsyriens im Zeitalter der Keilschrift-
kulturen, Wiesbaden.
Sommerfeld, W. (): „The Kassites of Ancient Mesopotamia: Origins, Politics and Culture“, J.M. Sas-
son (Hrsg.), Civilizations of the Ancient Near East, Peabody, –.
Stein, P. (): Die mittel- und neubabylonischen Königsinschriften bis zum Ende der Assyrerherrschaft
(JBVO ), Wiesbaden.
Steinkeller, P. (): „The Question of Marhaši: A Contribution to the Historical Geography of Iran in
the Third Millennium B.C.“, ZA , –. ˘
Steve, M.-J. (): Syllabaire Elamite. Histoire et Paléographie (CPOP ), Neuchâtel – Paris.
Steve, M.-J., Gasche, H. & de Meyer, L. (): „La Susiane au deuxième millénaire: à propos d’une
interprétation de fouilles de Suse“, IrAnt , –.
Steve, M.-J. & Vallat, F. (): „La dynastie des igihalkides: Nouvelles Interpretations“, in: L. de Meyer
& E. Haerinck, Archaeologia Iranica et Orientalis. Miscellanea in Honorem Louis vanden Berghe, Gent,
–.
Steve, M.-J., Vallat, F. & Gasche, H. (–): „Suse. F. Suse dans l’histoire“, DBS , –.
Stiehler-Alegria Delgado, G. (): Die kassitische Glyptik (MVSt ), München – Wien.
Stolper, M.W. (–): „Pahir-iššan“, RlA , –.
Vallat, F. (): „L’hommage de l’elamite Untash-Napirisha au Cassite Burnaburiash“, Akkadica –
, –.
———. (): „La chronologie méso-elamite et la lettre de Berlin“, Akkadica , –.
Van Dijk, J. (): „Die dynastischen Heiraten zwischen Kassiten und Elamern: eine verhängnisvolle
Politik“, Or. , –.
———. (): Literarische Texte aus Babylon (VS ), Berlin.
Walker, C.B.F. (): „Babylonian Chronicle : A Chronicle of the Kassite Kings and Isin II Dynasties“,
G. van Driel et al. (Hrsg.), Zikir šumim. Assyriological Studies Presented to F.R. Kraus on the Occasion
of his Seventhies Birthday, Leiden, –.
Zadok, R. (): Geographical Names According to New- and Late-Babylonian Texts (RGTC ), Wiesba-
den.
———. (): „Kassites“, EncIr (www.iranicaonline.org).
 DYNAMICS OF THE FALL: ASHURBANIPAL’S CONQUEST OF ELAM

Peter Dubovský*

. Introduction

The main aim of this paper is to reconstruct the facts responsible for the end of the so-called
Neo-Elamite Period II.1 Assyrian annals give the impression that Ashurbanipal’s campaigns
against the king Ummanaldash sufficed to destroy the Elamite kingdom. In this paper I offer
a more nuanced reading of the last days of Ummanaldash’s reign. Military conflicts between
Elamite kings and Ashurbanipal triggered in Elam certain dynamics such as conspiracies, coups
d’état, rebellions, etc., which ultimately helped the Assyrians to triumph over Elam. The capture
of Ummanaldash was the final straw that broke the back of Elamite resistance. After this point
the quantity and quality of extant documents decreased dramatically and from the Assyrian
point of view Elam stopped being a real rival. Between  and bc we can notice a period
of political uncertainty in Elam.2 Even though extant sources permit scholars to reconstruct a
list of the next Elamite kings, it remains an open question whether after bc Elam regained
its previous power or became only a minor player on the stage of Mesopotamian history.3
In this paper I will point out the dynamics present in Elam that I believe brought to an end
Ummanaldash’s kingdom. In the first part, I will analyze the Neo-Assyrian royal inscriptions,
in the second, I will discuss some neo-Babylonian letters dated to this period, and in the last
part I will summarize the phenomena responsible for the downfall of Elam.

. Royal Inscriptions

In order to provide the historical context for the following analysis let us now briefly summarize
Ashurbanipal’s first three military conflicts with Elam.4 The most important written source for
this period is Prism B. This prism relates the Assyrian campaigns against Elam and eastern
Mesopotamia (against Urtak in BIWA B § –; against Te-Umman in BIWA B § –; against
Gambulu in BIWA B §–; against Ummanigash, Tammaritu, and Indabibi in BIWA B § –
), then the campaigns against the Arabs (BIWA B § –), and the prism ends with the
description of Ashurbanipal’s building projects (BIWA B § –) followed by an admonition
not to destroy the inscription (BIWA B § ). Since this prism does not mention Ashurbanipal’s
campaigns against Ummanaldash, most scholars conclude that the prism was composed in
bc.5 Some of the events described in this prism were mentioned or copied word for word
in the later editions of Ashurbanipal’s annals (Prisms C, Kh, G, F, A, and H).

* Pontifical Biblical Institute, Rome.

1 The Neo-Elamite period is traditionally divided into three parts: Neo-Elamite I, II, and III; for the dates see
Potts : –; Waters : .
2 Frame : –.
3 Potts : –; Waters : –.
4 Amiet : – and : –; Carter & Stolper : –: Nadali : –.
5 Gerardi : –.
 peter dubovský

According to these prisms Ashurbanipal’s first military conflict with Elam took place when
the Elamite king Urtak presumptuously invaded Babylonia. The occasion for the Elamite
invasion of Babylonia was Assyria’s involvement with Egypt in bc. The Elamites quickly
overran Babylonia and laid siege to Babylon. In response Ashurbanipal dispatched his troops
to Babylonia and the Elamites were forced to retreat.6
In bc there was a dynastic upset in Elam and Te-Umman took the throne. The usurper
Te-Umman marched against Assyria, but the Assyrian army promptly occupied Der and the
Elamites fled back to Susa. The Assyrian troops pursued Te-Umman. A pitched battle was
fought at Till Tuba on the banks of the River Ulai in which the Elamites were defeated on their
home ground. Te-Umman was beheaded and Ashurbanipal appointed to the throne two sons
of Urtak, who in the meantime had become Assyrian allies. Urtak’s oldest son Ummanigash
was appointed king at Susa and Tammaritu king at Hidali.7
The third Assyrian invasion took place when pro-Assyrian Ummanigash joined Shamash-
shumu-ukin’s revolt.8 Ummanigash’s two generals Nesu and Attamenetu, along with Te-
Umman’s son Undashi, were killed.9 This abortive attack precipitated the revolution in Elam
and Ummanigash was replaced by his nephew Tammaritu.10 According to Prism B the situation
in Elam after the last Assyrian invasion became stable and the Elamite king Indabibi became an
Assyrian vassal. He was required to pay a regular tribute and to maintain requisite diplomatic
relations with Assyria.11 However, soon after Ashurbanipal’s last campaign Elam slipped out of
Assyrian hands. A new Elamite king Ummanaldash revolted against Assyria and Ashurbani-
pal had to intervene two more times. These campaigns against Ummanaldash are preserved in
multiple versions of the Assyrian royal inscriptions. In order to cast light upon this period of
Elamite history I will first discuss Prisms C, Kh, G, T, F, A, H, the Letter to the God Aššur, the
epigraph on slab BM , and the Inscription from the Temple of Ishtar.

Novotny’s analysis of the fragments from the British Museum and the Chicago collection
demonstrated that Prism C finishes with the installation of Ummanaldash on the throne and it
should be dated to  bc. Thus the parts which Borger attributed to Prism C (BIWA C § –)
in reality represent an independent prism—Prism Kh.12 Following Novotny’s reconstruction a
next redaction of Ashurbanipal’s campaigns against Ummanaldash is thus preserved in Prism
Kh dated to  bc (Table ).13 This prism records only one campaign of Ashurbanipal against
Ummanaldash. The events in Prism Kh develop quite straightforwardly. Even though the doc-
uments are damaged, from the context it is possible to deduce that the confrontation between
Assyria and Elam took place in Bit-Imbi.14 The preserved part of the tablet mentions that the
wounded Shamash-shumu-ukin escaped to Elam and his son was deported from Bit-Imbi to
Assyria where he was flayed. A further section of the tablet mentions that Tammaritu betrayed
Assyria and escaped. Ummanaldash, seeing the advancing Assyrian troops, became frightened
and escaped from his royal residence, Madaktu. In a similar way his rival Umbahabua, who
had seized the throne, escaped from his residence Bubilu. There is no description of the very

6 Grayson : .

7 Grayson : .
8 Grayson : .
9 Potts : .
10 Ibid.
11 With some changes these events are also described in Prisms A, C, D, H, the Inscription from the temple of

Ishtar (BIWA –), and the Relief Inscription (BIWA –).

12 Novotny : – and : –.
13 Novotny : .
14 For the reconstruction of this part see ibid.: .
 dynamics of the fall: ashurbanipal’s conquest of elam 

conquest of Elam and Prism Kh offers only the aftermaths of the campaign: Elamite cities were
plundered and their inhabitants deported to Assyria. Prism Kh finishes with a report on the
distribution of the booty.

A few months after Prism Kh had been composed, a new edition of Ashurbanipal’s campaign
against Elam came to light—Prism G.15 The scribes used Prism Kh as a template for this new
edition making some changes and additions.16 For our purposes, the most important additions
to Prism G are the description of the conquest of Bit-Imbi and the capture of Imbappi and of the
rest of Te-Umman’s family.17 This new version of Ashurpanipal’s campaigns served as a Vorlage
for the introduction to Ashurbanipal’s fifth campaign as described in Prism F.

Another report on the Assyrian campaigns against Ummanaldash is preserved in Prism T

(Table ). This summary inscription consists of three parts. The first part represents a long
introduction including a description of some building activities (BIWA T § –), the central
part of the prism relates Ashurbanipal’s campaign against Ummanaldash (BIWA T § –),
and the last part of the prism reports on Ashurbanipal’s building activities (BIWA T § –
). Thus the core of this prism (BIWA T § –) is dedicated to Ashurbanipal’s campaign
against Ummanaldash. This part of the prism, like Prisms Kh and G, only records one cam-
paign of Ashurbanipal against Ummanaldash; however, it offers some new elements absent
in Prisms Kh and G. After Ashurbanipal’s troops entered Elam, Ummanaldash escaped to
the mountains (BIWA T §). The Assyrian troops captured and plundered  fortified cities
together with countless smaller cities. Their gods, inhabitants, and possessions were deported
and salt and cress were scattered on fields (BIWA T § ). After the conquest of Elam Assyr-
ian gods were brought back to Uruk (BIWA T § ). According to Prism T the main goal
of Ashurbanipal’s invasion was to bring Assyrian gods captured by previous Elamite kings
back to Assyria whereas the destruction of Elamite cities and deportation were only side
effects.

Prism F dated to  bc18 represents the next step in the elaboration of the Assyrian campaigns
against Ummanaldash. This Prism combines the contents of Prisms C, Kh, G, and T19 with some
new additions. The major difference between Prism F and the previous edition is that Prism F
distributes the events described in the previous prisms over two campaigns instead of narrating
them in one as was the case in Prisms Kh, G, and T. According to Prism F (BIWA F § ) the
first campaign against Ummanaldash is already Ashurbanipal’s fifth campaign and it contains
texts from Prisms Kh, G, and T (BIWA C ix 0–0 and BIWA T § ). The second campaign
against Ummanaldash, i.e. Ashurbanipal’s sixth campaign, contains a longer version of T § 
and takes over part of T §. Moreover, whereas Prisms Kh and G connect the invasion of Bit-
Imbi with Shamash-shumu-ukin, Prism F connects military operations in Bit-Imbi with the
first campaign against Ummanaldash (BIWA F § –).

15 This prism was known for a long time as Prism K; see Cogan & Tadmor : –.
16 Novotny : –.
17 However, it is impossible to claim that this part is a new addition since the beginning of Prism Kh was not

preserved.
18 Gerardi : .
19 Both Prisms F and T date to bc. Prism T contains the shorter versions and Prism F the longer versions of

the same narrative. From the literary point of view, it is impossible to say whether Prism F served as a Vorlage for
Prism T or vice versa. Not to be excluded is the possibility that there was another literary source, which was not
preserved, serving as a Vorlage for both prisms. In the following paragraphs I will demonstrate how the scribes used
the themes contained in Prisms C and T to edit Prism F.
 peter dubovský

Let us now reconstruct the events according to Prism F.20 The first campaign against Um-
manaldash, i.e. Ashurbanipal’s fifth campaign, can be divided into two military interventions
against Elam. When Ashurbanipal started this campaign, the inhabitants of Hilmu and Pillatu
surrendered and paid tribute to the Assyrians (BIWA F § ). Advancing Assyrian troops met
resistance only in the Elamite outpost Bit-Imbi, which was quickly overrun and its inhabitants
punished as examples. Imbappi, the commander of Elamite archers stationed in Bit-Imbi,
was captured and deported together with the rest of Te-Umman’s family to Assyria (BIWA F
§).21 Seeing the advancing Assyrian troops and the destruction of the Elamite buffer state
Bit-Imbi Ummanaldash escaped from Madaktu to the mountains (BIWA F § ). Similarly
Ummanaldash’s rival Umbahabua, who took advantage of the instability in Elam and seized
control over one part of Elam, escaped from Bubilu22 (BIWA F § ).23 Ashurbanipal thus could
advance into the heart of Elam without further obstacles. After reaching Susa Ashurbanipal
put Tammaritu on the throne, the former Elamite king, who were seeking asylum before
Indabibi escaped to Assyria (BIWA F § ). However, Tammaritu betrayed the Assyrians and
Ashurbanipal had to intervene for a second time. Tammaritu was removed from the throne
and Ashurbanipal captured a large number of Elamite and Babylonian cities and brought much
booty to Assyria on his return (BIWA F § ).24 Prism F leaves the reader with the impression
that after the fifth campaign the Elamite kingdom was utterly destroyed and left without a king.25
Prism F then continues with the description of Ashurbanipal’s second campaign against
Ummanaldash, i.e. Ashurbanipal’s sixth campaign.26 It does not say why Ashurbanipal decided
to intervene once again or what happened to Ummanaldash after the Assyrian troops retreated.
This campaign followed the same course as the previous one. Ashurbanipal once again led his
troops through Bit-Imbi. Then he marched through Rashi and Hamanu. Ummanaldash seeing
the advancing Assyrians escaped to Dur-Undasi.27 Crossing the river Idide, he prepared his
forces for battle using the river as a line of defense (BIWA F § ).28 Meanwhile the Assyrians
plundered the cities belonging to Elam including Susa and Madaktu together with other cities
(BIWA F §–). Once they had crossed the river Idide, there were no obstacles left for the
advancing Assyrian army. Ashurbanipal conquered one region after another and Ummanaldash
escaped to the mountains (BIWA F § –). The Assyrian troops first advanced eastwards
and then turned against Susa.29 Ashurbanipal triumphantly entered Susa and let his soldiers
plunder it. Once the city was captured the Assyrian soldiers destroyed the temples and the
ziggurat, looted the royal treasury, and desecrated the sacred groves and tombs.30 The Assyrian
gods captured by Elamite kings together with huge booty were brought to Assyria (BIWA F

20 Gerardi : –.

21 A similar sequence of events is preserved in the extant parts of Prism G; see BIWA, , .
22 Bubilu is often identified with Susiana; see TAVO , , Waters : .
23 Waters (ibid.) noticed that Umbahabua was never called king even though he did sit upon the throne of Elam

(or, at least, on a throne in Elam).

24 The location of some cities which were identifiable among the twenty-nine listed captured cities indicates that

the Assyrians destroyed the border cities but did not penetrate into the heartland of Elam; see ibid., .
25 The lack of information regarding Tammaritu’s successor underlines the confusion in which the kingdom was

left after the retreat of the Assyrian troops. It has been suggested that this can simply reflect a lack of source material;
see ibid.
26 Gerardi : –.
27 The region is represented on Slab A in room S1 in Ashurbanipal’s North Palace; see Nadali : .
28 Potts : .
29 However, it is not necessary to take the description at its face value. It has been suggested that the placement

of the vivid account of the conquest of Susa could have been due to the scribe’s unwillingness to interrupt the style
of the first half of the narrative; see Gerardi : –.
30 Grayson : .
 dynamics of the fall: ashurbanipal’s conquest of elam 

§–). After this vivid destruction account, similar to that of the defeat of Te-Umman (BIWA
B §–), the reader is left with the impression that Elam was completely ruined.31

BM 32 found in Room M of Ashurbanipal’s North Palace adds another piece of evidence
useful for the reconstruction of the last days of Elam.33 The relief depicts the capture of
Ummanaldash and his deportation on a chariot to Assyria. The epigraph located in the upper
part of the slab reads:
[…] with the weapon of Aššur, my lord,
[…] from the mountain where his refuge
[…] the city of Murubisu
[…] Aššur, my lord,
[…] (he) seized Ummanaldash
[…] (he) took him to my presence […].34
Murubisu is most likely another spelling for Marubištu, the Ellipian royal city, which served
as a place of refuge for Nibe, a nephew of the Ellipian king Dalta, after his rebellion against
Assyria proved to be unsuccessful. This would suggest that Ummanaldash used the Ellipian
city Murubisu (Marubištu) as his place of refuge when pursued by the Assyrians. The name of
the man who captured and brought Ummanaldash to Ashurbanipal was not preserved.35 This
epigraph thus adds a new detail: someone finally captured Ummanaldash and brought him to
Ashurbanipal.

In the recently republished Letter to the God Aššur Borger edited a part of the Letter missing
in Weippert’s edition.36 Lines v –vi  describe the events in Elam after Ashurbanipal’s last
invasion. Since these lines, even though badly damaged, are almost word for word copied in
Prism A, it is possible to reconstruct their content. Based on such a reconstruction we can
point out some new episodes that are not present in Prism F. Lines v – claim that after
Ashurbanipal’s retreat, Ummanaldash once again returned from the mountains and settled
down in Madaktu. This time, however, only to lament the destruction of his kingdom. Then
the Letter continues with the story of Nabu-bel-shumati of which only the first two lines are
preserved (v. –). Lines vi – describe how [Pa"e] and the inhabitants of several Elamite
cities surrendered and paid homage to Ashurbanipal. Lines vi – describe the revolt of Elam
against Ummanaldash. The preserved lines mention twice a revolt in Elam using the same words
(lines vi –! and !–!). These two descriptions are separated by four lines mentioning the
capture of Ummanaldash. Since the tablet is damaged, it is impossible to conclude whether
the Letter to the God Aššur describes two revolts against Ummanaldash or the same revolt is
mentioned twice.

The last fuller elaboration of the annals describing the campaigns against Elam is preserved in
Prism A dated to  bc.37 The chronology of Prism A follows that of Prism F with a few minor
changes (Table ).38 Prism A, however, does not stop at the description of the deportation from

31 Waters : .

32 Barnett (: Pl. LX, LXI) also interpreted the drawings on plates LX and LXI as the surrender of Ummanal-
dash. However, it is difficult to prove that the person on these slabs is really Ummanaldash.
33 Barnett (: Pl. XXXIV); Gerardi : –; Russell : ; Waters : .
34 Barnett : , Russell : .
35 Waters : .
36 BIWA, –; Weippert : –.
37 Gerardi : ; Frame : .
38 Gerardi : –.
 peter dubovský

Elam as was the case in Prism F, but it adds some new details taken from the Letter to the God
Aššur and from the inscription on relief BM . We can discern five episodes that took
place after Ashurbanipal’s last campaign: . After the retreat of Ashurbanipal, Ummanaldash
returned from the mountains and settled in Madaktu (BIWA A § ). . The Assyrians requested
the Elamites to extradite Nabu-bel-shumati. He became afraid and committed suicide. His
corpse was shipped to Nineveh where it was solemnly beheaded (BIWA A § ). . Pa"e,39 the
inhabitants of Bit-Imbi and of several Elamite cities paid homage to Ashurbanipal (BIWA A
§). . Elam revolted against Ummanaldash, who escaped to the mountains whence he was
deported to Assyria (BIWA A §). . After the final victory over Elam and Babylonia three
Elamite kings—Tammaritu, Pa"e, and Ummanaldash—participated at the humiliating march in
Nineveh (BIWA A §). Even though the Letter to the God Aššur is damaged it is still possible
to recognize the differences between the Letter and Prism A. Above all the Letter does not
insert the campaigns against the Arabs into the Elamite story. The narrative in the Letter moves
directly from episode  to episode . Moreover, in Prism A the revolt against Ummanaldash
is described only once, whereas in the Letter there are two descriptions of the same or of two
different revolts.
Moreover, the similarities between Prism A and the epigraph on relief BM  suggest
that the scribe responsible for the redaction of Prism A used, besides Prism F and the Letter
to the God Aššur, this epigraph as his Vorlage. Comparing Prism A with the epigraph on relief
BM , we can notice some changes. First, the scribe changed the rd p. s. on the epigraph
into the st p. s. in Prism A and, thus attributed the capture of Ummanaldash to Ashurbanipal.
Second, the author of Prism A omitted the city of Murubisu as the place of Ummanaldash’s
refuge.
In sum, the Letter to the God Aššur, the epigraph on relief BM , and Prism A shed
a different light upon the spectacular victory described in Prism F. Despite their victories, the
Assyrians even after five campaigns were still unable to capture their arch-enemies Nabu-bel-
shumati and Ummanaldash. Moreover, after the retreat of the Assyrians Elam quickly returned
to its previous state after Ummanaldash had returned from the mountains to Madaktu. The
situation in Elam, however, developed in favor of the Assyrians. Nabu-bel-shumati committed
suicide and Ummanaldash was captured, though the inscriptions do not say who did it and how.
The audience is left with the impression that after the elimination of these two arch-enemies the
Elamite resistance was finally broken down.

The preserved parts of Prism H, dated to bc,40 describe some parts of Ashurbanipal’s
campaigns against Elam (Table ). The first campaign against Elam is preserved in BIWA H
iii 0–0 and BIWA H ii0 –iii0 .41 Ummanaldash is also mentioned in BIWA H 0–0. The
aftermath of the Assyrian campaigns important for our study is described in BIWA H ii0 0–
0. According to this version dated probably to bc42 the defeat of Elam had impact not
only upon Elamite and Mesopotamian cities but also upon Cyrus, the king of Parsumash, and
Pislume, the king of Hudimiri (BIWA H ii0 0–0). This fragmentary Prism suggests that Elam
was at last defeated and its exemplary punishment was successfully used for the purposes of
Assyrian propaganda.

39 For the reconstruction of Pa"e’s role in Elam see Waters : .
40 Gerardi : , Frame : .
41 See also Novotny : –.
42 Ibid.: .
 dynamics of the fall: ashurbanipal’s conquest of elam 

Finally the Inscription from the Temple of Ishtar summarizes Ashurbanipal’s most impor-
tant deeds.43 Among these, the scribes mention the conquest of Elam. The description of
the conquest of Elam is divided into three parts. The first part (BIWA IIT –) presents
the destruction of countless Elamite cities, the return of the goddess Nanaya to Uruk, and
the submission of three Elamite kings (Tammaritu, Pa"e, and Ummanaldash). The second
section (BIWA IIT –) describes the suicide of Nabu-bel-shumati and the shipment
of his body to Assyria. The last Elam section (BIWA IIT –) describes the impact
of the Elamite defeat upon kings Cyrus and Pislume. This section ends with the proces-
sion that concluded the Assyrian military operations in Elam. In this procession Tammar-
itu, Pa"e, and Ummanaldash drew Ashurbanipal’s chariot. In sum, this summary inscription
follows with small changes the events as described in Prisms A and H and confirms that
the Elamite problem was settled only after Nabu-bel-shumati died and Ummanaldash was
arrested.

Let us now summarize the previous analyses. The Assyrian inscriptions claim that the Assyrian
royal campaigns played the most important role in the conquest of Elam. Prisms Kh, G, and
T claimed that one campaign was sufficient for conquering Ummanaldash’s kingdom; Prism F
divides the events described in Prisms Kh, G, and T between two campaigns and thus claims
that Ashurbanipal needed two campaigns to conquer Elam. Finally the Letter to the God Aššur
and Prism A suggest that after the second campaign Ashurbanipal was still not able to capture
his arch-enemies and Ummanaldash returned to his royal residence after the withdrawal of
Assyrian troops. Prisms A, H, and the Inscription from the Temple of Ishtar indicate that the
final victory was achieved only when Ummanaldash was captured. Whereas according to the
inscription on BM  Ummanaldash was captured in Ellipi by someone whose name was
not preserved, according to Prism A it was Ashurbanipal himself who captured Ummanaldash.
The Assyrians celebrated their victory in Nineveh during which they made the Elamite kings
draw Ashurbanipal’s chariot. The defeat of Elam was efficiently used in Assyrian propaganda to
make kings Cyrus and Pislume surrender.

. Letters

The previous analysis has demonstrated that despite their devastating character the Assyrian
campaigns were unable to achieve their main goal—to capture the Babylonian rebel Nabu-
bel-shumati and the Elamite king Ummanaldash. Thus the rebel Nabu-bel-shumati could not
only move freely around, but also mobilize a good part of Elam against Assyria and check
the Assyrians for at least two years.44 Similarly Ummanaldash soon after the last campaign
recovered his previous position. In order to complete the picture drawn on the basis of the royal
inscriptions, in the following paragraphs, I will investigate the correspondence between the
Assyrian royal court and Bel-ibni, an Assyrian official active in Sealand.45 These letters mention
dynamics, which I believe substantially contributed to the final victory of Assyria. The goal of
this analysis is not to tie these letters to concrete periods of Elamite history but rather to capture
some sociological and military developments in Elam during Ummanaldash’s reign that led the

43 BIWA, –.
44 Malbran-Labat : –.
45 For the edition, translation, and chronological order of the letters see PNAE /II, –; Oppenheim :

– no. ; Ahmed ; Malbran-Labat ; Frame : –; de Vaan ; Potts : –;
Waters : – and .
 peter dubovský

Assyrians to triumph over their adversaries. For this purpose, the following analysis of Bel-ibni’s
letters will be divided into two parts. The first part will discuss the Assyrian activities directed
against Elam; the second will deal with the Elamite internal problems.

.. Assyrian Activities Directed against Elam

... Protagonists
When Nabu-bel-shumati had to escape from Sealand, the region became the base for Assyrian
incursions into Elam.46 Bel-ibni, acting as the governor of this region, was the major protagonist
and the coordinator of most Assyrian disruptive activities.47 His network was quite ramified. The
letters mention five of his close collaborators: Mushezib-Marduk, his nephew (ABL :r.–),
Ahumme, the governor of Hindanu, Nabu-sharru-usur, the cohort commander, Nabu-zeru-
ibni (ABL :–), and Belshunu, his older brother, who was imprisoned by Nabu-bel-
shumati (ABL :r.–). This network of officials had at their disposal military forces to
further their interests. In the letters we can distinguish two types of military units: professional
soldiers (in the letters called emūqu) and servants (sābu and ardu described mainly by means of
the logograms ÉRIN.MEŠ and ÌR.MEŠ šá LUGAL EN-ia). ˙ The professional soldiers consisted of
cavalry (sisû described mainly by means of the logogram ANŠE.KUR.RA) and infantry (zakkû).
The infantry contained mostly archers (qaštu described mainly by means of the logogram
GIŠ.BAN) and once the letters mention the shield-bearers (arı̄tu).
The size of the units which were at Bel-ibni’s disposal can be also estimated from the let-
ters.
Archers:  archers (ABL :r.), – archers (ABL :),  archers in reserve
(ABL :r.), [xx] Aramaean archers, who joined Mushezib-Marduk (ABL :r.–
), Bel-ibni asking for more archers (ABL +:).
Infantrymen:  (ABL :).
Shield-bearers: Asking for more shield-bearers (ABL +:).
Cavalry:  horses (ABL :r.–),  horses and asking for another  horses
(ABL +:-r.),  horses (ABL :), – horses (ABL :), Bel-ibni
asking for more horses (ABL :r.–).
Servants:  servants (ABL :),  servants (ABL :).
Even though these numbers cannot be simply summed up, they give a picture of how many
soldiers and servants were involved in Bel-ibni’s disruptive activities. It is possible to estimate
that Bel-ibni was in command of a network of Assyrian officials, who had at their disposal at
certain moments over  archers,  horses,  servants, and some shield-bearers. Since the
Assyrian forces were active in various regions at the same time, the real number could have been
easily twice as high. A body of over , men constituted an army large enough not only to
threaten but also to disrupt Elamite structures especially when the Assyrian royal troops were
deployed in another part of the Empire.

... Disruptive Activities

To demonstrate the Assyrian subversive activities and their diplomatic pressure on Elam I use
letter ABL  as a pattern, which I believe may reveal Assyrian strategy applied against Elam
during Ummanaldash’s reign.

46 Malbran-Labat : –.

47 For the discussion on his identity see Brinkman : ; Frame : ; de Vaan : –.
 dynamics of the fall: ashurbanipal’s conquest of elam 

According to this letter, Bel-ibni ordered  servants of his to plunder Elam (ABL :–
).48 The servants attacked the town Irgidu about  km from Susa. The raid was quite
bloody: the ruling family, that of Ammaladin, was executed together with another potential
ruler Dalajan. Besides the ruling family, the ruling body of the city,  notables, was also
exterminated. Finally Bel-ibni’s servants took  captives from the city. After this raid the city
was ruined and unable to offer any resistance.
The next part of the letter (ABL :-r.) describes the impact of this raid upon Aramaean
tribes: the Lahiru and Nugu"u tribes made a treaty with Bel-ibni’s nephew and thus became
Assyrian allies.49 Their collaboration with Assyria, however, did not stop at the treaty cere-
monial. Both tribes offered their archers and together with the Assyrian troops made another
incursion into Elam. The raid by this new alliance met with success too. The allies took more
war prisoners and deported them to Bel-ibni.
The last part of the tablet reveals a non-military tactic employed by the Assyrians to control
Elam. Bel-ibni interrogated one of the war prisoners who claimed that Elam was in a state of
civil war (ABL :r.–).50
Using this letter as the organizing pattern, we can divide the Assyrian pressure and subversive
activities into four groups:
1. Diplomatic pressure and propaganda: pressure and threat to extradite Nabu-bel-shumati
(ABL :-r.; :r.–; BM ;51 ABL :–); threatening Natan, the sheik
of the Aramaean tribe Puqudu (ABL :r.–); fear instilled by the Assyrian royal
troops (ABL :–; :–); two Aramaean tribes voluntarily submitted to Assyria
after Assyrian raid (ABL :-r.); Elamites deserting to Bel-ibni (ABL :r.).52
2. Military incursions and raids on Elam: the raid on the region on the other side of the
river Marat (ABL :r.–); the raid on the city Irgidu (ABL :–); Mushezib-
Marduk and his new allies raided Elam (ABL :r.–); letter ABL  refers to a
previous raid (ABL :–) probably that mentioned in ABL ; the raid on the cities
Akbanu and Ale, which were burned down (ABL :–); the raid on the city Mahmiti
(ABL :r.–); Assyrian agitation troops sent to Targibatu (ABL :–); Bel-ibni
continued plundering Elam (ABL :).
3. Deportations and booty:  citizens of Irgidu (ABL :–); war-prisoners captured
by Mushezib-Marduk and his allies (ABL :r.–);  people from Akbanu and
Ale, who were later killed (ABL :-r.); , head of cattle, of which , were
slaughtered and thrown into the river Marrat and  brought to Bel-ibni (ABL :r.–
); Bel-ibni planned to bring to Nineveh , captured in Elam (ABL :r.–;  is
confirmed in ABL :r.); the Assyrians captured for each captive Assyrian  Elamites
together with – head of cattle (ABL :r.–).53
4. Executions of the enemies:  members of the ruling families and  notables of Irgidu
(ABL :–); execution of the soldiers in Akbanu and Ale (ABL :–); Bel-
ibni’s servants killed  and wounded about  enemy soldiers in a military conflict
(ABL :r.).
According to this analysis, the Assyrian officials backed up by their military units had sufficient
power to destabilize Elam even when the royal troops were absent. The most important Assyrian

48 Frame : –.

49 Malbran-Labat : .
50 For the Neo-Assyrian intelligence services see Dubovský : –.
51 Waters .
52 For the means and efficiency of Assyrian diplomatic pressure see Dubovský : –.
53 For the analysis of various aspects of deportation see Oded .
 peter dubovský

subversive activities were incursions and raids. From the letters it is possible to estimate that
during Ummanaldash’s reign the Assyrian officials coordinated at least six major military
incursions. During these incursions the Assyrians destroyed  cities, executed  people,
captured , war-prisoners (+  Elamites for each captive Assyrian + Mushezib-Marduk’s
war-prisoners), and , pieces of cattle. Since these numbers are based only on the extant
letters, the real numbers were most likely much higher. According to the letters the local
Assyrian officials were able to destabilize the entire regions of Elam and leave the cities without
rulers. Diplomatic pressure, exemplary executions, and the divulgation of Assyrian victorious
acts had a psychological effect similar to that of the royal campaigns: some tribes after having
heard about the Assyrian raids joined the Assyrians.54
The diplomatic pressure backed up by military power was exercised especially in the case of
Nabu-bel-shumati. All military and diplomatic efforts failed and the mastermind of Assyrian
resistance was not only able to hide in the mountains but also to get new adherents. The
situation changed when Bel-ibni in the name of the Assyrian king vehemently requested the
extradition of Nabu-bel-shumati. The annals report that Nabu-bel-shumati seeing that his end
was nigh committed suicide; his corpse was shipped to Nineveh and solemnly decapitated. Thus
the major instigator of anti-Assyrian rebellions, taking refuge in Elam, lost his head, and that
naturally had to have a strong impact upon all anti-Assyrian factions.

.. Internal Problems

Assyrian royal campaigns combined with the disruptive activities of the Assyrian officials were
not the only cause that brought to an end Ummanaldash’s reign. The military and diplomatic
pressure of Assyria triggered internal dynamics that enfeebled Elam and its resistance.
Assyrian officials reported with great pleasure on the internal problems in Elam, which
favored Assyrian expansionistic policy. The more fragmented the Elamite opposition was, the
less able it became to withstand Assyrian incursions and propaganda. The internal tensions
can be observed on two levels: . disagreements and changing allegiances, . open revolts and
rebellions.55

... Disagreements and Changing Allegiances

The letters allude to some disagreements and open conflicts within the Elamite leading class:
1. Because of tensions between Ummanaldash and the local nobility, Ummanaldash and his
family had to leave Madaktu (ABL :–). The palace herald Ummanshibar and his
adherents abandoned Ummanaldash, went to the city of Shuharisungur (ABL :–),
and started a secret negotiation with Assyria about the extradition of Nabu-bel-shumati56
(ABL :–; for the difference of opinion over what to do with Nabu-bel-shumati see
also ABL :–). As the result of this tension Bel-ibni claimed that all the land had
abandoned Ummanaldash (ABL :).
2. Shuma, a member of Tammaritu’s family, escaped and joined the Assyrian allies, the
Dahhareans (ABL :–).
3. Puqudeans went to Nabu-bel-shumati and by making a treaty with him joined the anti-
Assyrian faction (ABL :-r.).

54 Gallagher .
55 Frame : .
56 This claim is based on the suggestion that Um-man-ši-bir (ABL :) is another spelling of Um-man-ši-bar

(ABL :); see de Vaan : .

 dynamics of the fall: ashurbanipal’s conquest of elam 

The letters thus mention three major disagreements with Ummanaldash’s policy, mainly regard-
ing Elamite policy towards Assyria, Nabu-bel-shumati, and solving domestic problems.57 These
tensions resulted in the creation of local factions that formed an opposition to Ummanaldash
and some discontented groups even joined Assyria. While trying to resolve the problems with
the opposition, Ummanaldash did not hesitate to kill his adversary Umhuluma. Some of the
malcontents were also connected with the high tributes the Assyrian imposed upon Elam that
must have drained Elamite financial and economic resources (ABL :r.–; ; :–).

... Revolts
In some cases the discontent broke out into open rebellion. The letters mention open rebellions
five times.
1. The Palace herald revolted against the king of Elam and the king’s brother was killed; the
revolt then took form of an open combat (ABL :r.–).
2. Ummanshibar and his adherents (see above) talked evil things against the king and as a
result the whole country rebelled against Ummanaldash (ABL :–).
3. Ummanigash revolted against Ummanaldash; the extent of the rebellion: from the river
Hudhud till the city Haidanu. Ummanaldash mobilized his forces to crush the rebellion
(ABL :r.–).
4. As the result of an Assyrian raid, Elam revolted against Ummanaldash who barely escaped
from the hands of the rebels (ABL :–).
5. Dahhasharians and Shallukeans revolted against Ummanaldash because he killed Umhu-
luma (ABL :–).58
These five occurrences of rebellion can be grouped into three rebellions, out of which the last
two took place during Ummanaldash’s reign.
The first rebellion is mentioned in ABL :r.– and it was orchestrated by the palace
herald (nāgiru). If the suggestion that this nāgiru should be identified with Indabibi59 is
accepted, then this rebellion coincides with his coup d’ état (BIWA B §  and C § ) dated
to bc.60 The rebels killed the king’s brother (ABL :r.).
The second rebellion was orchestrated by Ummanshibar. It started when Ummanshibar
and his adherents abandoned Ummanaldash and joined the pro-Assyrian faction. The change
of allegiances was motivated by common sense—Why should the Assyrians kill us also?
(ABL :–). Having joined the pro-Assyrian camp, the rebels started to negotiate the
extradition of Nabu-bel-shumati (ABL :–) and to slander Ummanaldash (ABL :–).
The rebellion reached its climax when the whole land abandoned Ummanaldash (ABL :)
and rebelled against him (ABL :). Arm in arm with Ummanshibar’s rebellion went Ara-
maean unrest triggered by Ummanaldash’s execution of Umhuluma (ABL :–). The
letter explicitly mentions two Aramaean tribes revolting against Ummanaldash.
The third rebellion was the most widespread (ABL  and )61 and its mastermind was
Ummanigash, son of Amedirra. This rebellion took the form of an armed conflict. Ummanal-
dash after having mustered his troops faced the rebels at a river. According to ABL :–

57 Malbran-Labat, “Nabû-bêl-šumâte”: .

58 The killing of Umhuluma is described in ABL .
59 e Vaan : .
60 Ibid., .
61 Both tablets report on massacres performed by the Assyrians, and both tablets report a revolt against Ummanal-

dash. Based on these similarities it stands to reason to claim that ABL  was written before ABL ; the latter
 peter dubovský

the rebellious troops achieved victory and Ummanaldash had to escape to the mountains. Given
the extent of the rebellion—from the river Hudhud to the city of Haidanu (ABL :r.–
)62—we can claim that this rebellion was greater than all the previous ones and we can even
speak about a civil war in Elam. If we want to harmonize the rebellions mentioned in the royal
inscriptions with those mentioned in the letters we can notice that the Letter to the God Aššur
mentioned two rebellions against Ummanaldash as did the letters whereas Prism A mentions
only one rebellion, probably that of Ummanigash which was the largest one.

... Famine
Besides the raids and internal conflicts famine was also ravaging Elam. The tablets mention
famine four times out of which the last two (ABL  and ) could be dated to Ummanal-
dash’s reign:
1. During the revolt of the palace herald (most likely Indabibi; see above), the whole of Elam
was afflicted by famine (ABL :r.).
2. Another famine broke out among the Aramaean tribes so that they were forced to turn
part of their own excrement into food (ABL :–).
3. During the revolt of Ummanshibar a wave of famine and other disasters affected Elam.
This fear of Assyrian troops ravaging Elam worsened the enfeebled morale of the starving
Elamites (ABL :–).
4. Another letter mentions the lack of food and Ummanaldash’s desperate measures to
encourage the people (ABL :r.–).
Generally speaking famine has always been one of the weakest points of enemy resistance.63 On
the other hand, lack of food on its own has never been a sufficient motive to compel people
to surrender. The study of morale among German civilians during WWI indicates that shortly
after the food situation deteriorated the civilians were still able to motivate themselves to high
altruism. Their morale improved because their military position or other motives still gave them
some hope. It dropped rapidly however once these motives disappeared.64 The same logic could
be applied to Elam as well. The waves of famine were often connected with the ravages by enemy
troops, the threat of Assyrian invasions, and the rebellions against the king. Therefore famine
combined with these elements must have had a strong impact upon the morale of Elam and it
stands to reason to consider it as another phenomenon contributing to the downfall of Elam.

. Conclusions

Let us now summarize the results of this analysis. The Assyrian royal campaigns were indis-
putably the main cause of the collapse of Elam. The annals mention five major military cam-
paigns undertaken against Elam. During these campaigns the Assyrians devastated a good part
of Elam, exiled a large number of Elamites, and plundered their riches. Moreover, the Assyri-
ans seriously interfered with the internal politics by placing their own adherents on the throne.
Besides favoring the pro-Assyrian kings, the Assyrians imposed high taxes that drained Elam
financially and economically.

describes the results of massacres and of revolts. However, ABL  speaks about Nabu-bel-shumati and Bel-ibni
suggests the king to write a severe letter in order to get the mastermind of all the rebels in the East.
62 For the localization and the discussion of the toponym see Parpola, Johns & Tallqvist : ; TAVO , .
63 Eph"al : –.
64 Daugherty : .
 dynamics of the fall: ashurbanipal’s conquest of elam 

Even though the royal campaigns severely damaged the stability of Elam, the campaigns on
their own were unable to eradicate the resistance on the part of Elam. The withdrawal of the
royal troops, however, made room for the subversive activities of Assyrian officials and their
military units that ruthlessly furthered Assyrian interests. These subversive groups plundered
cities, deported their inhabitants, and executed the ruling class of anti-Assyrian centers. Both
royal campaigns and raids were successfully used in Assyrian propaganda and made some of
Ummanaldash’s allies change their allegiances.
This Assyrian military and diplomatic pressure generated internal dynamics that seriously
undermined the stability of Elam as well. The disagreements among the leading groups as
regards domestic affairs and the relations with Assyria were the first phenomenon eroding
the internal stability of the kingdom. These disagreements caused the leading groups and even
entire tribes to change their allegiances, join the pro- or anti-Assyrian factions and even resort
to the execution of the opponents. In the worst cases some of these tensions broke out into open
rebellions. The letters mentioned three rebellions out of which one had the characteristics of a
civil war.
Another factor jeopardizing the internal stability of Elam was frequent coups d’ état. Accord-
ing to the royal inscriptions there were five coups d’ état within twenty years. During the reign of
Ummanaldash some local factions opposed Ummanaldash, claimed the throne, and were even
able to exercise power in some parts of Elam for a limited period of time. The annals mention
two kings who tried to compete with Ummanaldash: Umbahabua who settled in Bubilu and
Pa"e.
The last factor contributing to internal unrest was famine. During Ummanaldash’s reign Elam
was afflicted twice by famine and in one case the king had to intervene to calm down the unrest
among the people.
The network of Assyrian officials taking advantage of the internal tensions as well as low
morale in Elam was also responsible for the elimination of two Assyrian arch-enemies. Thus
Bel-ibni’s diplomatic pressure and threats inflicted the last blow on the anti-Assyrian coalition.
Once Nabu-bel-shumati, the mastermind of opposition committed suicide, the anti-Assyrian
front suffered a deadly blow. The epigraph on relief BM  claims that an Assyrian ally
arrested Ummanaldash, and thus eliminated the last representative of anti-Assyrian resistance.

In sum, this analysis demonstrates that the end of Elam was an extremely complex process.
The end of Ummanaldash’s reign can be attributed to the combination of three major factors:
the destructive royal campaigns, the subversive activities of Assyrian officials, and the internal
instability and tensions in Elam. None of these elements on its own was sufficient to bring to
an end the Elamite kingdom. It was rather the combination of these elements that eradicated
Elamite resistance and brought to an end so-called Elamite period II.

Bibliography

Ahmed, S.S. (): Southern Mesopotamia in the Time of Ashurbanipal (Studies in Ancient History ),
The Hague—Paris.
Amiet, P. (): Elam, Paris.
———. (): “Sur l’histoire élamite,” IrAnt , –.
Barnett, R.D. (): Sculptures from the North Palace of Ashurbanipal at Nineveh (–B.C.), London.
Brinkman, J.A. (): “Merodach-Baladan II,” Studies Presented to A. Leo Oppenheim: June , ,
Chicago, –.
Carter, E. & Stolper, M.W. (): Elam: Surveys of Political History and Archaeology, Berkeley.
Cogan, M. & Tadmor, H. (): “Ashurbanipal’s Conquest of Babylon: The First Official Report—Prism
K,” Or. , –.
 peter dubovský

Daugherty, W.E. (): A Psychological Warfare Casebook, Baltimore.

de Vaan, J.M.C.T. (): “Ich bin eine Schwertklinge des Königs”: Die Sprache des Bel-ibni (AOAT ),
Kevelaer—Neukirchen-Vluyn.
Dubovský, P. (): Hezekiah and the Assyrian Spies: Reconstruction of the Neo-Assyrian Intelligence
Services and Its Significance for Kings –, Roma.
Eph"al, I. (): The City Besieged: Siege and Its Manifestations in the Ancient Near East, Leiden—Boston.
Frame, G. (): Babylonia –B.C.: A Political History (PIHANS ), Istanbul.
Gallagher, W.R. (): “Assyrian deportation propaganda,” SAAB , –.
Gerardi, P. (): Assurbanipal’s Elamite Campaigns: A Literary and Political Study, PhD dissertation,
University of Pennsylvania.
Grayson, A.K. (): “Assyria –B.C.: The Reign of Ashurbanipal,” J. Boardman, L.E.S. Edwards,
N.G.L. Hammond & E. Sollberger (eds.), Cambridge Ancient History III/: The Assyrian and Babylo-
nian Empires and other States of the Near East, from the Eighth to the Sixth Centuries B.C., Cambridge,
–.
Malbran-Labat, F. (): “Nabû-bêl-šumâte, Prince du Pays-de-la-mer,” JA , –.
———. (): L’Armée et l’organisation militaire de l’Assyrie: d’après les lettres des Sargonides trouvées à
Ninive, Paris.
Nadali, D. (): “Ashurbanipal against Elam: Figurative Patterns and Architectural Location of the
Elamite Wars,” Historiae , –.
Novotny, J.R. (): “A : A Fragment of Assurbanipal Prism G,” JCS , –.
———. (): “Assurbanipal Inscriptions in the Oriental Institute: Prisms E, H, and J,” Or. , –.
———. (): “Classifying Assurbanipal’s Inscriptions: Prisms C, Kh (= CND), and G,” R.D. Biggs,
J. Myers & M.T. Roth (eds.), Proceedings of the st Rencontre Assyriologique Internationale Held at
the Oriental Institute of the University of Chicago, July –, , (SAOC ), Chicago, –.
Oded, B. (): Mass Deportations and Deportees in the Neo-Assyrian Empire, Wiesbaden.
Oppenheim, A.L. (): Letters from Mesopotamia: Official, Business, and Private Letters on Clay tablets
from Two Millennia. Chicago.
Parpola, S., Johns, C.H.W. & Tallqvist, K.L. (): Neo-Assyrians Toponyms (AOAT ), Kevelaer—
Neukirchen-Vluyn.
Potts, D.T. (): The Archaeology of Elam: Formation and Transformation of an Ancient Iranian State
(Cambridge World Archaeology), New York.
Russell, J.M. (): The Writing on the Wall: Studies in the Architectural Context of Late Assyrian Palace
Inscriptions (Mesopotamian Civilizations ), Winona Lake.
Vallat, F. (): Les noms géographiques des sources suso-élamites (RGTC ), Wiesbaden.
Waters, M.W. (): A survey of Neo-Elamite history (SAAS ), Helsinki.
———. (): “A Letter from Ashurbanipal to the Elders of Elam (BM ),” JCS , –.
Weippert, M. (): “Die Kämpfe des assyrischen Königs Assurbanipal gegen Araber: Redaktionskri-
tische Untersuchung des Berichts in Prisma A,” WO , –.
 Letter to the god
Aššur and the Inscription from the
inscription on Temple of Ishtar
Prism Kh Prism G Prism T Prism F BM  Prism A Prism H (IIT)
[th campaign: th campaign Prism th campaign: th campaign:
conquest of Bit-Imbi] G=GE (BIWA GE the beginning of the beginning of
(Kh ix –; ii0 0) the campaign— the campaign—
Novotny , ) Tammaritu was Tammaritu was
taken with the taken with the
Assyrian troops; Assyrian troops;
tribes Hilmu tribes Hilmu, Pillatu,
and Pillatu came Dummuqu, Sulaja,
to Assyria and and Lahiru-Dibirina
paid homage to came to Assyria and
Ashurbanipal. F §  paid homage to
(iii –) Ashurbanipal. A §
(iv –)
Against Bit-Imbi, Against Bit-Imbi, Against Bit-Imbi,
execution and execution and Sennacherib’s
deportation of deportation of memorial, execution
their inhabitants; their inhabitants; and deportation of
deportation of deportation of their inhabitants;
Imbappi and Imbappi and of the deportation
of the rest of rest of Te-Umman’s of Imbappi,
Te-Umman’s family family. F §  (iii Ummanaldash’s
Prism G=GE, –) nephew, and of the
GD (Novotny rest of Te-Umman’s
, ; BIWA family. A § (iv
GE ii0 0–0; –v )
GD i0 0–0; K
+:–=BIWA
)
Wounded Shamash- […] escaped to
shumu-ukin Elam, his son
escaped to Elam, Barburu was
dynamics of the fall: ashurbanipal’s conquest of elam

his son Barburu deported from

was deported from Bit-Imbi and flayed.
Bit-Imbi and flayed. Prism G=C (BIWA
Prism Kh=CNDB C ix 0–0)
(BIWA C ix 0–0)

 Letter to the god


Aššur and the Inscription from the

inscription on Temple of Ishtar
Prism Kh Prism G Prism T Prism F BM  Prism A Prism H (IIT)
Tamaritu, betrayed Tamaritu, betrayed
Assyria and escaped. Assyria and escaped.
Prism Kh=CNDB Prism G=C, GF,
(BIWA C ix 0–0) A  (BIWA C ix
0–0)
[Ummanaldash] Ummanaldash Assyrian Ummanaldash Ummanaldash
escaped [from] seeing advancing invasion of Elam, seeing advancing seeing advancing
Madaktu. Prism Ashurbanipal’s Ummanaldash Ashurbanipal’s Ashurbanipal’s
Kh=CND (BIWA C troops [escaped escaped to the troops escaped from troops escaped from
ix 0–0) from] Madaktu. mountains. T §  (iv Madaktu to the Madaktu to the
Prism G=C, GF –) mountains. F §  (iii mountains. A § (v
(BIWA C ix 0–0) –) –)
Umbahabua, Umbahabua, Umbahabua, Umbahabua,
[Ummanaldash’s [Ummanaldash’s Ummanaldash’s rival, Ummanaldash’s rival,
rival], escaped from rival], escaped from escaped from his who took control of
his residence Bubilu. his residence Bubilu. residence Bubilu. F Elam, escaped from
Prism Kh=CND, Prism G=C (BIWA §  (iii –) his residence Bubilu.
CNDB (BIWA C ix C ix 0–0) A § (v –)
0–0)
peter dubovský

Installation of Installation of
Tammaritu in Susa. F Tammaritu in Susa.
§  (iii –) A § (v –)
Ingratitude of Ingratitude of
Tammaritu; Tammaritu, direct
Ashurbanipal speech reporting
invokes gods and his evil thoughts;
captures Tammaritu. Ashurbanipal
F §  (iii –) invokes gods and
captures Tammaritu.
A § (v –)
List of Elamite cities List of Elamite cities List of Elamite cities
captured on return of captured on return of captured on return of
Ashurbanipal + list Ashurbanipal + list Ashurbanipal + list
of things deported of things deported of things deported
to Assyria. Prism to Assyria. F §  (iii to Assyria. A § (v
Kh=CNDB, CND  –iv ) –)
(BIWA C ix 0–0)
 Letter to the god
Aššur and the Inscription from the
inscription on Temple of Ishtar
Prism Kh Prism G Prism T Prism F BM  Prism A Prism H (IIT)
Exemplary
punishment
(flaying of people);
Sennacherib’s
memorial. Kh ix
– (BIWA C §)
Deportation from
Babylonia and
execution. Prism
Kh=CNDB (BIWA
C ix 0–0)
th campaign: th campaign:
conquest of Raši, conquest of Raši,
Ummanaldash Ummanaldash
escaped from escaped from
Madaktu to Madaktu to
Dur-Undasi, crossed Dur-Undasi, crossed
the Idide river and the Idide river and
prepared for combat. prepared for combat.
F §  (iv –) A § (v –)
Conquest of Conquest of Elamite
Mesopotamian cities including
and Elamite cities Madaktu. A §
including Madaktu. F (v–)
§  (iv –)
Conquest of Conquest of
Dur-Undasi. F §  Dur-Undasi. A §
(iv –) (v –)
Immediate crossing Fear of the troops,
dynamics of the fall: ashurbanipal’s conquest of elam

of the river. F §  (iv dream and reported

–) direct speech of
Ishtar, change in
troop morale,
crossing of the river.
A § (v –)

 Letter to the god


Aššur and the Inscription from the

inscription on Temple of Ishtar
Prism Kh Prism G Prism T Prism F BM  Prism A Prism H (IIT)
Conquest of Elamite Conquest of Elamite Conquest of Elamite
cities (shorter cities. F §  (iv cities. A § (v
version). T § a (iv –) –)
–)
Ummanaldash Ummanaldash
escaped to the escaped to the
mountains. F §  (iv mountains. A § (v
–) –)
Conquest of Elamite Conquest of Elamite Conquest of Elamite
cities, their gods cities, their gods cities, their gods
and taking booty to and taking booty to and taking booty to
Assyria (shorter Assyria. F §  (iv Assyria. A § (v
version). T § b (iv –) –)
–v )
On return On return
Ashurbanipal Ashurbanipal
conquered Susa, conquered Susa,
plundered it together plundered it together
with its temples and with its temples and
peter dubovský

palaces and the palaces and the

booty deported to booty deported to
Assyria. F §  (iv Assyria. A § (v
–v ) –vi )
Final destruction Final destruction Summary of the
of Elam. F §  (v of Elam. A § (vi destruction of Elam.
–) –) IIT –
Return of Nana. T Return of Nana. F Return of Nana. A Return of Nana. IIT
§ (v –) §  (v –vi ) §  (vi –) –
Distribution of the Distribution of the Distribution of the
booty of Elam. Prism booty of Elam. F §  booty of Elam (more
Kh=CNDB (BIWA (vi –) officers added). A
C ix 0–0) §  (vi –vii )
 Letter to the god
Aššur and the Inscription from the
inscription on Temple of Ishtar
Prism Kh Prism G Prism T Prism F BM  Prism A Prism H (IIT)
Ummanaldash Ummanaldash
returned from the returned from the
mountains, settled mountains, settled
in Madaktu, and in Madaktu, and
grieved. Letter v grieved. A § (vii
– –)
Nabu-bel-shumati. Request to extradite
Letter v – Nabu-bel-shumati,
his suicide and
beheading of his
corpse which was
shipped to Assyria. A
§  (vii –)
Surrender of Elamite Ummanaldash’s Surrender and Tammaritu, Pa"e,
cities and revolt in rival Pa"e seeing tribute. H II0 0–0 and Ummanaldash
Elam. Letter v – the devastation of embraced
Elam embraced Ashurbanipal’s feet;
Ashurbanipal’s feet; Nabu-bel-shumati
surrender of tribes committed suicide
which fled to the and his corpse
mountain Saladri. A was shipped to
§  (vii –) Ashurbanipal. IIT
–, –
Revolt(s) against Revolt against
Ummanaldash, Ummanaldash,
his escape to the his escape to the
mountains and mountains and
his deportation to his deportation to
Assyria. Letter vi Assyria. A § (x
–! Capture of –)
dynamics of the fall: ashurbanipal’s conquest of elam

Ummanaldash in the
inscription on BM
.

 Letter to the god


Aššur and the Inscription from the

inscription on Temple of Ishtar
Prism Kh Prism G Prism T Prism F BM  Prism A Prism H (IIT)
Triumphal The impact of the The impact of the
procession after defeat of Elam upon defeat of Elam
the conquest kings Cyrus and upon kings Cyrus
of the Arabs, Pislume. H II0 and Pislume;
Tammaritu, Pa"e 0–0 triumphal procession
and Ummanaldash after the conquest
marched in it. A § of the Arabs,
(x –) Tammaritu, Pa"e
and Ummanaldash
marched in it. IIT
–
peter dubovský
 ELAMITE AND OLD IRANIAN AFTERLIFE CONCEPTS*

Jan Tavernier**

. Introduction: Elamite Funerary Cult and Afterlife Ideas

In ancient civilisations, death was as much part of life as it is in modern societies. Various
sources inform us on the afterlife thoughts of Mesopotamia, Egypt, Greece and Rome, as a
result of which we have a fairly nice image of how these ancient people thought about death
and the netherworld. In striking contrast with this relatively high amount on information
on death and afterlife in the major ancient civilisations, the information on the Elamite “au-
delà” is extremely poor. In other words, very little is known on Elamite afterlife concepts or on
the Elamite funerary cult. Nevertheless, the available information suggests that death was, not
surprisingly, an important preoccupation of the Elamites (Vallat : –; Grillot ;
Henkelman : , with reservations):
(1) Some principal gods played a role as netherworld deity. According to Vallat, three gods are
particularly interesting: Inšušinak, the lord of the netherworld, Išmekarāb and Lāgamāl.
Grillot adds Kiririša. It will be demonstrated, however, that Išmekarāb was not a pure
netherworld god and that his primary task was to uphold justice, also in the nether-
world.
It is also possible that Inšušinak was not the netherworld’s lord in all of Elam, but only
in Susa and that each deity was responsible for the netherworld in his/her own town (ex.
Kiririša in Liyan [EKI ] or Upurkupak in Tchogha Pahn [Stolper & Wright : –
; Vallat a]). This remains, however, hypothetical.
Part of the discussion on the infernal character of Inšušinak and Kiririša are four epithets:1
– DN temti kukunnum lahakra (for Inšušinak; Atta-hamiti-Inšušinak): EKI :.
– DN tepti siyankuk lahakra (for Inšušinak; Untaš-Napiriša): MDP  A:–,,
B:, :–.
– DN zana Liyan lahakra: EKI :,, :–,– (for Kiririša; Šilhak-Inšušinak).
– Te[pti alimelu]2 lahakra: MDP  no. :, (for Inšušinak; Šilhak-Inšušinak).
The crucial word in this phrase is lahak, a participial form of laha-, a stem attested from
the Old Elamite (Siwe-palar-huhpak) till the beginning of the Achaemenid period (Atta-
hamiti-Inšušinak) and whose meaning is still under debate. Various translations have been
proposed:
(a) “Live, reside”: Bork : ; Hüsing :  and : ; Pézard : .
(b) “Die, perish; kill”: Scheil : ; König : ; Khačikyan : ; Grillot :
; Vallat : ; Henkelman :  and : .

* Abbreviations are cited according to the system used in Northern Akkad Project Reports , , –.
**Université catholique de Louvain.
1 The epithet DN temti ahpir lahakrame (for Inšušinak) in an inscription of Šilhak-Inšušinak (IrAnt  –:–

) has another structure. In adition, ahpir is not an indication of a place.

2 Reconstruction by Steve, based on the formula Inšušinak tepti alimelurra “Inšušinak, Lord of the Acropole”

(EKI :–, :–, , i –; all are inscriptions of Šilhak-Inšušinak).

 jan tavernier

(c) “Dominate, rule”: Pézard : –; Steve : –.

(d) “Enter; die”: Hinz :  and ; Lambert :  n. ; Hinz :  n. ;
Hinz & Koch : .
(e) “Remove”: Lambert : .
(f) “Seize”: Grillot : .
(g) “Destroy”: Grillot : .
(h) “Hold, possess”: Grillot : .
(i) “Secret, hidden”: Grillot :  n.  and : ; Vallat & Grillot : ; Malbran-
Labat : –. Steve : .
Forms of laha- are attested in different contexts, which, unfortunately, do not really
exclude one of the above proposals. Anyhow, in some texts the meaning of laha- is quite
clearly “to kill (transitive); to die (intransitive)”:
(1) First of all, in the Elamite Omen Text (Obv. –) eššana lahan is the Elamite equiv-
alent of Akk. lugal úš-ma “The king will die” (Iqqur ı̄puš § :; cf. Tavernier :
–) in the omen itu g[u4] kimin gameššana la-ha-an a-ak gampu-hu gameššana-pé
giš
gu.zameš ad-da a-pi-ri-na-ba mu-ur-da-am-pi “Month Ayaru: The king will die and
the sons of the king will sit on the throne of their father” / diš ina gu4 lugal úš-ma
dumu.meš lugal aš.te ad-šú-nu sá.sá “If it is in the month Ayaru, the king will die
and the sons of the king will take the throne of their father” (Steve : ).
(2) An Old Elamite inscription of the sukkalmah Siwe-palar-huhpak (EKI :; first half
th century bc) has hit lahakna “May the army ˘ be killed” in what is most likely a series
of curses (e.g. line  has petip luk limašpi “May the fire/lightning burn the enemies”).
As the form is passive, a translation “May the army reside” or “May the army enter” is
not acceptable.
(3) A text of Šilhak-Inšušinak (ca. – bc) mentions that someone (text is broken
here) should be kazzak lahak[na] “May he be smitten, may he be killed” (EKI  iv
).
(4) In a curse formula Šutruru mentions hiše Nahhunte lahašni “May Nahhunte destroy
(kill) his name” (EKI  §; end of th century bc).
(5) Not entirely sure, but very likely laha- has the same meaning in an inscription of
Tepti-Huban-Inšušinak (c.  bc). There this king claims that pap  gudmeš pap 
udu.nítameš la-ap-ma hu-sameš-ik-si-na a-hi la-ha-na “Total:  head of cattle (and)
total:  head of sheep/goats, for the officiants, which is to be brought to the grove
and there to be slaughtered” (EKI :; cf. Henkelman : ).
Nevertheless, a meaning “lord of the dead in the kukunnum” does not fit all too well with
regard to the divine epithets. One would rather expect a form tepti lahakpe-ra (plural).
The solution to translate “lord of the netherworld” is not correct either, since forms in -k
normally are past participles. In addition, a substantive “death” would rather be lahame or
lahan.
In other words, laha- does not mean “die” in the divine epithets. It is, however, clear that
in the epithets laha- is always associated with a deity (Inšušinak or Kiririša) and a place
(alimelu, siyan kuk, Liyan, kukunnum). In fact, laha- expresses the relation between the
deity and the place. For this reason Hüsing adopted a meaning “live, reside”, which was
accepted by Pézard (: ). Such a translation can indeed not be excluded, but it is
not the only possible solution. Various authors have proposed a meaning “secret, hidden,
invisible”. Vallat & Grillot (: ) have “Lord of the Renown and of the Secret”, whereas
Grillot herself (: ) translates “Lady of secret Liyan”.
It is Steve (: ) who has proposed another syntactical relation within this group.
He proposes a translation “DN, the lord/lady who is hidden in GN” for the epithets.
 elamite and old iranian afterlife concepts 

Two arguments might corroborate Steve’s preference. First of all, the passive participial or
adjectival character of lahak (literally lahakra means “He/she who is laha-ed”) supports
a rendering “secret, hidden”. Secondly, in two passages of his prism A the Assyrian king
Assurbanipal, who sacked Susa, refers to the secret character of the Elamite religion:
(1) A v –: (I conquered Susa) “the great cult center, the dwelling of their gods, the
place of their secrets” (māhāzu rabû mūšab ilānišun ašar pirištišun).
(2) A vi –: (I took as booty ˘ to Assur) “Inšušinak, their mysterious god who dwells
in seclusion, (the god) whose divine features nobody was allowed to see” (dMùš.šéš il
pirištišun ša ašbu ina puzrāti ša mamman lā immaru epšēt ilūtišu).
In the light of these two textual witnesses it is better to adopt a basic meaning “to hide; to
be secret” for the verb laha-. A secondary meaning “to disappear” may be postulated, out
of which a meaning “to kill; to die” developed. All textual attestations of laha- can now be
comfortably translated.
This implies, however, that the epithets cannot longer be used to corroborate the con-
nection of Inšušinak and Kiririša with the netherworld. Nevertheless, this does not
mean that this connection is impossible. In fact, a further argument in favour of a link
between Kiririša and the netherworld is an inscription of Šilhak-Inšušinak (IrAnt 
–), dedicated to Kiririša. In this text (lines –) this deity is explicitly linked to the
grove, possible symbol of the netherworld: zana takra giš sip kullamara mattari “Living
Lady, who has under her authority the grove, the gate and the prayer”.3 An argument in
favour of the link between Inšušinak and the netherworld is his Neo-Elamite ideogram
d
Mùš.lam, in which the element lam developed to Akk. lammu “netherworld” (Steve
: –).
(2) A very important aspect of the funerary cult, at least according to Grillot (: –
n. ; : – and ; ), Malbran-Labat (: – et –; :
–) and Vallat (: ; a; b; : –; –: –), is the
grove (El. husa; Sum. giš). This is made clear by the inscription in which the Assyrian
king Assurbanipal relates his destruction of the city of Susa and according to which the
grove housed graves. Possibly the grove was related to the passage of the dead into the
underworld. Its entrance was marked by a “temple of the grove” (El. [] siyan husame,
ex. siyan Inšušinakme husame, the grove temple dedicated to Inšušinak [EKI :rev.]; []
El.-Akk. siyan kištumma4 “temple in the grove”, ex. siyan Kiririšame kištumma [MDP 
:–]), whose gate was perhaps a symbol for the transition zone between the two worlds.
Note also the mentioning of dHu-sa (deified grove) in the Naram-Sîn Treaty (EKI  i )
and in an inscription of Tepti-Huban-Inšušinak (EKI :). The grove temples are mostly
dedicated to Inšušinak, others are dedicated to Išme-karāb, Kiririša, Lāgamāl, Napiriša
and Suhsipa, gods considered by Grillot and Vallat to be netherworld deities.5
An inscription of Šilhak-Inšušinak (EKI :–) makes clear that husa-temples
were widespread in the Middle Elamite kingdom. The text mentions  such temples
(siyan husame).

3 By preferring “bestow” as translation for matta/i-, Hinz & Koch (: ) do not accept the translation “have

under his authority”, proposed by Vallat & Grillot (: ) on the basis of EKI , where matta/i- corresponds with
pat rappa- “subject, subdue”. With regard to this parallel Hinz & Koch (: –) propose a reading u mattit
“chasten you”, which is, however, not at all convincing.
4 Kištum: El. transcription of Akkadian qîštum “grove”, the equivalent of El. husa. Used as Akkadian loanword in

MDP   (Steve : ).

5 Gardens not only contained temples. The garden of Nahhunte/Šamaš in Susa functioned as a place of jurisdic-

tion, at least in the sukkalmah-period (MDP  :, :,; cf. Stolper –: ).
˘
 jan tavernier

Despite the arguments uttered by Grillot and Vallat, it is not completely sure that the grove
was per definition linked to the funerary cult (Henkelman : ). The inscription of
Assurbanipal, in which he says that he entered the grove and destroyed royal tombs, may
have a paratactic sense, so that the statements are just consecutive and are not interrelated.
Secondly, some of the gods (ex. Išme-karāb, Manzat, Simut and Suhsipa) to whom grove
temples are dedicated, are not necessarily netherworld gods.
Grillot (: ) believes that the haštu-temple (cf. infra) was related to the husa,
whereby she seems to imply that the haštu was identical to the temple of the grove. This
is wrong, for EKI  clearly demonstrates that the temples of the grove were a structure
different from the haštu-temple. Also Vallat (a and : ), who identifies haštu
as the lower temple of the ziggurat, suspects a relation between haštu and husa, but again,
the only relation between both structures is the funerary cult.
(3) If Vallat’s hypothesis is right (Vallat a), the ziggurat itself was strongly linked to the
funerary cult. It was surrounded by the grove. If kukunnum is indeed the high temple of the
ziggurat, Inšušinak’s epithet mentioned above (temti kukunnum lahakra) may be a further
indication of a funeral character of the ziggurat.
(4) Grillot (: ) and Vallat (a) also believe that the gates (hel / sip) dedicated to
deities such as Inšušinak, Kiririša, Išmekarāb and Lāgamāl had a funeral character. Their
main argument is the fact that the gates are dedicated to deities of whom they believe that
they are infernal gods. Nevertheless, as long as the connection between all these deities on
the one hand and the netherworld on the other hand is not convincingly demonstrated,
this idea has but little value.
(5) According to Carter (:  and n. ) the stele found in the é.dù.a in Haft Tepe (Reiner
) is dealing with expenditures for festivals, among others for the chariot of Inšušinak
and for the funerary ritual (kispum). She also wonders to what extent Inšušinak’s chariot
could be linked to the need for transport to the underworld. Rituals could be performed
in front of statues of the deceased.
(6) There are several burial types, possibly reflecting the social status of the deceased. All
burials were accompanied by offerings, from simple pottery for the poorer people to lavish
furniture for the wealthy ones (Vallat : ). The fact that no graves were dug in the
great aristocracy houses of the sukkalmah-period could point to an Elamite preference, as
˘
this practice is contrary to Suso-Mesopotamian practices (Gasche, apud Vallat : ).
Basically there are three burial types in Susa: (a) tombs in plain earth pits, (b) tombs in jars,
and (c) tombs in brick-built vaulted structures. Most scholars assume that the different
burial types reflect a social aspect (plainest tombs for the lower classes, vaults for the elites),
but recently Elizabeth Carter (: –) has argued that the burial types rather reflect
a multi-phased burial ritual and that the skeletons were moved from one type of tomb to
another (secondary burial).
(7) Finally, there are the clay statues deposed in the tombs. In Shahdad, statues (bust and head)
from the second half of the rd millennium were discovered in various graves.6 From the
second millennium onwards, the statues have the form of human-size heads without bust.
These pieces of art, unparalleled in the Ancient Near East, may be portraits of the deceased,
partly because some of them were found next to the deceased’s skull (Contenau, apud de
Mecquenem, Contenau, Pfister & Belaiew : ; Ghirshman : ; Vallat :
; Spycket : –; Alvarez-Mon : ; Carter : ). Yet, it is also possible

6 Cf. Amiet : figs.  and ; Hakemi : Obj. nn. , G.a, , G.b, , G., , G.,

, G., , G., , G., , G., , G., , G., , G..
 elamite and old iranian afterlife concepts 

that the heads represent close family members of the dead person, for the heads have their
eyes opened (Alvarez-Mon : ). The youngest heads date from the th century bc,
just before the Elamites would be integrated into the Achaemenid Empire. According to
Ghirshman the heads accompanied the deceased person to the netherworld.

. Funerary Texts from Susa: General Description and Translation

Between  and  the French excavators in Susa discovered a large number of tombs,
dated to various periods, to the east of Darius’ palace (de Mecquenem : –). The
archaeologists thought that they had unearthed a real necropolis, but it should be noted that
remnants of other constructions were found in this area as well, inter alia a wall containing
inscribed bricks from Temti-halki, Kutir-Nahhunte II and Šilhak-Inšušinak I.
One of these tombs7 (vault no. ) proved to be highly interesting, at least from a philological
point of view, for in it were found seven remarkable texts (MDP  nos. –, –
and ). The tomb itself measured ,× , meter and was , meter high. According to
de Mecquenem (: ) the tomb contained “cinq à six vases ronds, un anneau de bronze,
une empreinte sur terre crue”. In an access pit of ,× , meter were found several vases and,
protected by a brick compartment, the tablets. In all likelihood, tomb no.  must be dated to
the end of the sukkalmah-period.
The tablets were first ˘published by Scheil (: –), but better copies were provided
by Dossin (: –), who, however, did not include a new transliteration or translation
in his publication. In  Ebeling (: –) again studied these texts, omitting the th
(being in a too poor condition). In his new edition of the texts corpus, Bottéro (: –
) has proposed some new readings. The most recent edition of the texts was undertaken
by Steve & Gasche (: –), who added some emendations to Bottéro’s transliterations
and translations. The translations below are based on this edition.
Interestingly the texts were not written by one scribe: while the texts , ,  and  are written
with a nice hand, no.  is not very well written and no.  is badly written. The syllabary of
the texts resembles the, as yet unpublished, syllabary of the correspondence of Attaruh-Uktuh,
which is connected with Kuk-Našur III, the last sukkalmah, and the syllabary of other texts
mentioning the three last sukkalmahs Tan-Uli, Temti-halki˘ and Kuk-Našur III. Accordingly
the texts can be dated to the end of the˘ sukkalmah-period (Steve & Gasche : –). In
addition, the syllabary is local, with typical values˘such as šà, ší and ka4 (Bottéro : ). The
general character is, however, typically Old Babylonian (nearly no ideograms8 and CVC-signs9).
All this indicates that the texts were written in the Old Babylonian environment of a peripheral
town, being Susa.10

.. Translation of the Texts

 MDP  
(–) Let’s go! My god, my lord, may I present myself before the Anunnakki. May I pass along the
weighing and may I take your hand in the presence of the great gods. May I hear the judgement,
may I grab your feet.

7 For a detailed overview of the few archaeological data on this tomb, see Steve & Gasche (: –).
8 É.gi6 (MDP  :), udu (MDP  :), ìhi.a (MDP  :), esi-di (MDP  :), kaš4 giš.bar
˘
(MDP  :) and udu še.gur (MDP  :).
9 The signs kur (MDP  :), mar (MDP  :) and li[k] (MDP  :).
10 The date (th–th Century bc), proposed by Aynard (: ), can be discarded.
 jan tavernier

(–) You will take me to the House of Darkness, my god. You will make me pace up and down
a swamp of misery and hardship. In a territory of distress you will look for me. You have rarefied
water and pasture in the land of thirst.11

 MDP  
They have taken the road, they go their way. Isme-karāb and Lāgamāl go in front. Inšušinak, in
the pit, will proclaim the judgement. He will stand before the Weigher. He will pronounce his
declaration.

 MDP  
(–) They have placed the recumbent one near important people, those who possessed land, who
possessed sheep and goats and who did not have rivals.
(–) Inšušinak, in the pit, will proclaim the judgement. He will stand before the Weigher. He will
pronounce his declaration: “Look at me, who has descended in a black cloud”!

 MDP  
Alas, short is the joy of life. They will proclaim12 your judgement. [ ] take out/save. For a jar he will
find oils, may he fill your mouth with hilimmu-locust fat. May the god grant (something) to you.
˘
 MDP  
Free is the access to the questioning. A hard(?) stone will be presented to him to lie down. May he
take the rest, for you launce yourself, O Fire, and you extend your terror across the entire land.

 MDP  
He has heard the ca[ll]. How will he pay? [ ] His brazier is destroyed. [ ] He is thrown in a prison. [
] The water of Inšušinak.

 MDP  
Fo[od portions] … Poti[ons] … A unit of oil, pressed beer … May he bring … For food: sheep,
plump barley, may he e[at]!
Most authors believe that these texts have a funerary character (Scheil : ; Ebeling
: –; Kleveta : –; Bottéro :  and ; André-Salvini ; Steve
& Gasche : –; Henkelman :  and : ; Carter : –). There are,
nevertheless, more doubtful and different opinions. According to Landsberger (apud von Soden
: ) the texts are excerpts of mythological stories,13 while in the CAD (E , s.v. erpetu,
b) it is argued that one is dealing here with school texts. These two opinions have led Aynard
(: ) and de Miroschedji (: –) to the belief that the funerary character of the
texts is far from certain. Carter (: –) connects the texts with the different burial types
and concludes that both corpuses reflect the journey of the deceased to the netherworld.

11 Tūqira mê u šamma ina eqil sumāmı̄ti. This phrase has an inverse parallel in the ritual of the Anu-temple at
Uruk (RAcc  obv. ): rîtum u ˙mašqı̄tum udašša “You provide abundantly pasture and watering place” (Steve &
Gasche : ).
12 Steve & Gasche (: ), translating “est prononcé”, apparently consider iq-ab-bu-ú as a passive form (rd

sg.). They give more references for final ú, but in all these case this sign indicates the subjunctive (e.g. MDP 
:: šá … i-qa-ab-bu-ú “He who … will say”, cf. De Meyer : ). More likely the spelling reflects a present
(rd pl.) and may thus be translated “they will”. After Inšušinak has made clear his judgement it is again (officially)
proclaimed by perhaps Išmekarāb and Lāgamāl.
13 This opinion is slightly modified in AHw s.v. mušeqqilu, which is translated (with question mark) “Weigher”.
 elamite and old iranian afterlife concepts 

In her recent study on death and afterlife in Mesopotamia Van der Stede (: –)
argues that these texts have nothing to do with afterlife. In her analysis, she examines text by
text and proposes a new theory concerning these texts. In the first text the principal person,
who speaks in the first person singular, has the desire to receive a judgement, but in Van der
Stede’s eyes this is not a deceased person. Rather it is a living human being who is appealing to
the gods in order to solve problems which are related to the death, for example when someone
suffers from a sickness caused by a dissatisfied spirit.
There is, however, a major problem concerning this analysis: The text does not mention any
spirit or sickness, so, as Van der Stede states, the patient is here someone who simply wants to
improve his daily life. This would be rather exceptional, since the Mesopotamians were mostly
precise in defining their problems. The Susa text does not even mention the character of the
problem. In addition, some ideas do refer to the netherworld: A swamp of misery, the lack of
water, the House of Darkness. Finally, the text as a whole does not remind one of a ritual.
The second text, too, allegedly has nothing to do with the way Mesopotamians thought
about death and afterlife, because of two lexical issues: the word mu-še-qí-li-im-ma, generally
understood to mean “weigher” (derived from the verb šaqālu), is understood by some scholars
(CAD M/, , Van der Stede [: ]) to be derived from ekēlu “to be dark”. Mušēkilu
would mean something like “he who makes dark”. It should be noted, however, that “he who
makes dark” too may be very well connected with the underworld. The second lexical issue is
šuttu, concerning which Van der Stede, again following the CAD, prefers the more widespread
meaning “dream” to the meaning “pit”. Nonetheless “pit” could be specifically used in Elam,
although it is only attested in Susa in this context: The reason for this assumption is the possible
connection between haštu and šuttu.
The Elamite expression haštu occurs in two inscriptions: EKI  iv  (ašdu nun šalhupah[ni])
and EKI : (api sunkip urpuppi haštu dInšušinakni halihši). Various translations have been
proposed for both passages. König (:  and ; “Gründungsort”) and Grillot (: 
n. ) consider the haštu to be a building. Not accepting this explanation, Hinz & Koch (:
 and ) connect haštu with hašt- “to honour” and translate “Ehre”. Nevertheless, there are
two objections against this interpretation:
(1) In EKI  haštu is the object of a verb hali-, which has an explicitly architectural meaning
(Malbran-Labat : ). This suggests that haštu is likely to be a building or a part of a
building.
(2) Hinz’s and Koch’s translations are far-fetched. For EKI  iv  they have “Ehrungen für
dich habe ich wahrhaftig geboten”, whereas they translate the passage of EKI  as follows:
“Diese früheren Könige schufen (Kunstwerke) zu Ehren des (Gottes) Inšušinak”. It would
be surprising to see that the Elamite scribe of an official royal inscription would not have
mentioned the object (“Kunstwerke”) of the verb hali- (EKI ). In addition, the verb
šalhupa- can also mean “to order”, which converts the meaning into “May I order the haštu
for you”.
More probably the idea of Grillot (: n. ) to consider haštu as identical wit Akk. haštu
“hole, pit” is correct.14 As haštu and šuttu are identified in Malku vi , šuttu15 must also ˘mean
“pit, hole” and is clearly a˘variant of šuttatu “pit, grave”, which is equally identified with haštu.16
˘

14 A derived meaning could very well be “tomb” (Henkelman :  and : ).
15 Šuttu is also identified with būru “pit, hole” (CT   ii ; cf. MSL  Aa II/:).
16 Ha-áš-ti = šu-u[t-ta-tum]: Lambert : Pl.  i . Ha-áš-ti = šu-ut-ta-tú: RA   (K. i ), cf. Reiner

(:˘). ˘
 jan tavernier

The only non-lexical occurrence of šuttu “pit” is the texts nos.  and  discussed here. It could
very well be that šuttu is the Susian variant of šuttatu, but as the number of attestations is so
low, this must remain hypothetical.
The two inscriptions (EKI  and ) mention the haštu of Inšušinak. According to Grillot,
El. haštu “pourrait représenter un édicule particulier lié au culte funéraire, voire symboliser la
résidence souterraine du dieu Inšušinak”. Perhaps one can put it even more specifically. Two
times our texts mention “Inšušinak in the pit” (MDP  :: [Šu-ší]-na-ak i-na šu-ut-ti;
MDP  :: Šu-ší-na-ak iš-šu-ut-ti), which could be an epithet of Inšušinak as netherworld
god. The haštu could then be the temple dedicated to this version of Inšušinak.
Conclusively, one can retain the translations “Weigher” and “pit” (for mušeqqilu and šuttu),
although absolute certainty cannot be given.
Finally, Van der Stede does not consider it necessary to discuss the four remaining texts, for
these are badly preserved and do not concern death and afterlife concepts. According to Van
der Stede, the attempts made in this direction are nothing but speculations. Nevertheless, her
own alternative solution cannot be qualified either as “argument probant” (Van der Stede :
).
Despite the objections one may safely assume the funerary character of the texts. First of
all their find spot does certainly not deny it. Secondly, as Bottéro (: –) correctly
points out, it is excluded that the first text could be of a scholarly character or that it could
be a mythological excerpt or even a medley of various myths. The seven tablets constitute a
unit which is unique in the Mesopotamian literature. The bad quality of the tablets and the
handwriting suggests that they were not library copies but instead used for immediate usage.

. Information on the Netherworld Provided by the Texts

The unique character of the texts has expectedly triggered the minds of various scholars, but it
is Bottéro and Steve & Gasche who have studied them more thoroughly. According to Bottéro
they serve as a sort of guide for the deceased’s behaviour in the netherworld, which is indeed the
first impression one gets after reading the texts: they show what the deceased may expect once
dead. In addition, the fact that none of the seven texts is a nicely delineated unit induces one to
think that they are parts of a longer composition. One may thereby think of the fragments of
the Erra Epos, which were put on amulets as a protection against trouble (Bottéro : ).
Various afterlife aspects occur in the texts: in the first text the deceased person wants,
accompanied by his personal god, to present himself to the Anunnaki, the Mesopotamian
infernal gods whose task it was to welcome each ghost, to instruct him the rules of the
netherworld and to assign him his place in the new world. It seems as if the deceased will also
undergo a process of weighing and that finally he will listen to his judgement.17 Yet it is not
certain if the judgement is the direct result of the Weighing.
The second part of his text is not completely clear. Apparently the personal god will put the
deceased in a land of darkness, distress and thirst (i.e. the netherworld). Probably this must be
situated before the judgement will be proclaimed by the “great gods”. The dark land could then
be an illustration of the trip (the procession headed by Išme-karāb and Lāgamāl) to the place
where the judgement will be pronounced or could refer to some kind of waiting room.

17 I will not go into detail here on the precise character of this judgement, concerning which some believe that
it is a real evaluation of the deceased ones. Van der Stede (: –) admits that the deceased persons had to
undergo some procedure, but considers this procedure not to be a ‘judgement’.
 elamite and old iranian afterlife concepts 

The second text, albeit rather short, contains a lot of information. First the procession to the
place where the judgement is proclaimed is described. Two assistant gods of Inšušinak seem to
accompany the deceased to this place. The judgement itself is proclaimed by Inšušinak (awatu
qabû). The text ends with the announcement that the deceased will speak out his declaration
before the Weigher (mušeqqilu). The latter probably plays an important role in the process of
weighing.
The second part of the third text is largely a repetition of the second text. The first part,
however, seems to imply certain egalitarianism in the netherworld: the dead person, the
“recumbent one”, is placed near important and rich people (Steve & Gasche :  n. ).18
The fourth text is rather unclear. Again the judgement is mentioned (to be proclaimed by
either the Anunnaki or Išme-karāb and Lāgamāl), as well as oils and hilimmu-fat. The last three
texts do not reveal much either. The seventh text is probably a list of˘offerings.
In general, a model of what happened after death, as reflected in these texts can now be
reconstructed. First of all there is a procession, in which the deceased person is led by two deities
(Išme-karāb and Lāgamāl) to the “Weighing” ceremony, during which the soul is apparently
weighed. The soul also pronounces its statement (“Look at me, who has descended in a black
cloud”). The ceremony and the judgement proclamation are in all likelihood presided by the
Anunnaki, of whom it is never said that they proclaimed the result of the “Weighing”. The last
stage is the proclamation of the judgement, a task for “Inšušinak in the pit”.

. The Mesopotamian Character of the Texts

As the texts are written in Babylonian, they must certainly be embedded in the Mesopotamian
culture. In order to investigate the Mesopotamian cultural character of the texts and the extent
to which these documents reflect Mesopotamian afterlife thoughts, it is useful to have a look at
the general Mesopotamian attitude towards the netherworld (Bottéro : – and :
–; Scurlock : ; Berlejung : –; Van der Stede ).
The Mesopotamian netherworld was situated below the earth and is known from various
sources: literary texts (the Assyrian prince’s Vision of the Netherworld, Ur-Nammu’s Death,
Nergal and Ereškigal, the Gilgameš Epos, Inanna/Ishtar’s Descent to the Netherworld –, etc.),
prayers to netherworld deities (e.g. Nergal, Gilgameš and Dumuzi), rituals, etc.
These texts display a complex image of the netherworld, which, in any case, remained fully
inscrutable.19 The land of the dead had various names in Mesopotamia, e.g. “Land of no return”
(erset lā târi), “Land of wailing” (erset tānı̄hi), “Remote land” (erset rūqtu) “House of death” (bı̄t
˙ “House of dust” (bı̄t epri), “House
mūti), ˙ ˘ Darkness” (bı̄t ekleti).
of ˙
The main entrances to this place are situated in graves, mountains and the steppe. The place
itself was organized as a parallel world, with infernal gods (Nergal and Ereškigal) reigning as
king over it, a capital with a royal palace, ramparts, gates, etc.

18 This contradicts the discrimination to which Parrot (: ) seems to allude. This scholar assumes that in texts

– the destiny for the wicked ones (who have received a bad judgement) is described, whereas the destiny for the
good ones is described in text .
19 Cf. Emar / ::

[Ki.burù.da].bi na.me nu.u[n.zu.a]

Ki-iš-tu-[bi] na-me nu-un-zu-wa-a
[Ki-ma ˙ šu-pu-ul er-se-ti mam-ma la i-du-ú]
˙
“Like the depth of the earth, no one knows (them)”.
 jan tavernier

The journey towards this world was arduous. The dead person had to pass through the
demon-infested steppe land and to cross the Khubur River (with the assistance of a person
known as Silushi / Silulim or Humut-tabal, “Quick, take me there”). He is the equivalent of the
Greek Charon. After crossing ˘this river, the deceased had to be allowed by the gatekeeper (Bidu
“Open up”) to pass through the seven gates to the netherworld.
The netherworld was considered a gloomy place, shadowy, dark and dry. Still there are
some indications that this negative image must somewhat be modified: the residents of the
netherworld were not always clad with darkness. Each night Šamaš passes by on his circuit
round the sky, giving them light. There are also testimonies (Gilgameš Epos, cf. infra) according
to which they lived a life parallel to that of the people on earth, eating bread and drinking water.
Nevertheless most mythological and other texts display an utterly negative image of the
netherworld, as the following excerpts prove. The main aspects are: dust as food, no water or
dirty water to drink and darkness.

a. Ishtar’s Descent to the Netherworld –

() To the netherworld, land of n[o return],
() Ishtar, daughter of Sîn, [set] her mind
() Indeed, the daughter of Sîn did set [her] mind
() To the gloomy house, seat of the ne[therworld],
() To the house which none leaves who enters,
() To the road whose journey has no return,
() To the house whose entrants are bereft of light,
() Where dust is their sustenance and clay their food.
() They see no light but dwell in darkness,
() They are clothed like birds in wings for garments,
() And dust has gathered on the door and bolt.
() Shall I eat clay for bread, shall I drink dirty water for beer?

b. Gilgameš VII – and Nergal and Ereškigal ii 0—iii 0

These texts passages are very similar to the one in Ishtar’s Descent to the Netherworld.

c. Gilgameš XII
In the Babylonian version of tablet XII Gilgameš interrogates Enkidu on his experiences in
the netherworld. Various types of dead persons (e.g. the one with two sons, the one with
three sons, the one whose corps was left lying in the open countryside, etc.) are presented.
In some cases the netherworld inhabitants could have bread (XII ), water (XII ) or
even clear water (XII ). This text is clearly inspired by the older Nippurean (Sumerian)
version of the epic, where the same examples occur: bread for the one with two sons (XII ),
water for the one with three sons (XII ) and clear water for the one who died a natural
death (XII s ). In addition the Sumerian version has more examples: bread for the one who
has no heir (XII a ), uprooted grass and waste water for the leper (XII i ), bitter bread
and bitter water for the man who made light the name of his god (XII n ). In a Sumerian
manuscript from Ur, there is mention of “water from the place of a massacre, dirty water” for
the sons of Sumer and Akkad and for the mother and father of Gilgamesh (XII x –, y –
).
Tablet XII makes up an interesting text for the Sumerian perception of the netherworld. In
the Nippur manuscript it seems that the fate of the inhabitants of the netherworld depends
on the status while alive and on the way of dying and being buried. The Ur manuscript has a
more negative image with bad water for every Sumerian and Akkadian person, an image which
 elamite and old iranian afterlife concepts 

would be more general in later periods. It should hereby be noted that the Akkadian version of
Inanna/Ishtar’s Descent to the netherworld is more focused on the harsh destiny one will meet
in the netherworld.

d. The Cuthean Legend of Naram-Sin

In line  of the standard Babylonian version of this story one can read that the ghosts of
Enmerkar and his family should “drink polluted water and not drink clean water”.

e. Codex Hammurabi
(Col. :–) May he (= Šamaš) cause his spirit to thirst for water down below in the
netherworld.

The negative character of the netherworld is clearly present in the documents discussed here
(House of Darkness; land of thirst, a swamp of misery). Nevertheless the list of offerings
(MDP  ) may suggest that the afterlife in Susa was a bit more bearable than that in
Mesopotamia (Steve & Gasche : ).
The idea of a dark cloud is probably of Mesopotamian origin, albeit that it is never directly
connected with the netherworld in Mesopotamian texts. In Gilgameš XI  the black cloud
(urpatu salimti) is the announcer of the Flood. Maybe here it is taken as a metaphor of death
(Bottéro˙: ).
Inšušinak’s two assistants, Išme-karāb and Lāgamāl are equally Mesopotamian of origin.
They are two Mesopotamian deities, which are nevertheless most frequently attested in the Susa
region. Išme-karāb is attested in Susa itself, where Temti-Agun claims to have built a temple for
him20 and in Chogha Zanbil, as can be derived from an inscription of Untaš-Napiriša. He was
also popular in personal names and in an oath by Inšušinak and himself. He is basically a deity
of justice21 and not specifically a netherworld deity. In that sense he can be compared to Šamaš,
who also plays a role in the netherworld, but whose primary responsibility was justice (Lambert
–a).
On the contrary, Lāgamāl seems to be a real netherworld deity (Lambert –),
already attested in an Old Akkadian seal inscription and in the Ur-III period. He was very
popular in Old Babylonian Dilbat. His earliest attestation in Elam is precisely the text discussed
above. Under the Šutrukids (th century bc) he had a temple in Susa, where he stayed until
Assurbanipal carried away his statue as booty.
Steve & Gasche (: ) also indicate that the mentioning of the fire in MDP  
is a Mesopotamian concept, because Gibil / Girra (giš.bar), the Mesopotamian Fire God,
was allegedly an assistant of Nergal, Lord of the Netherworld. In order to strengthen their
assumption, they refer to Gilgameš XI –. Nevertheless, there is no indication of any
hierarchic relation between both deities, nor is there any mention of Girra as netherworld
god. The cited Gilgameš reference (XI – in the edition of A.R. George) has nothing
to do with the Mesopotamian god of Fire or with the Netherworld. Both lines occur in the
description of the beginning of the Flood: () dA-nun-na-ki iš-šu-ú di-pa-ra-a-ti () ina
nam-ri-ir-ri-šú-nu ú-ha-am-ma-tu ma-a-tum “The Anunnaki bore torches aloft, setting the
land aglow with their˘ brilliance”.˙ It should be noted here that Girra is not even mentioned

20 Later kings such as Šutruk-Nahhunte I, Šilhak-Inšušinak I and Hutelutuš-Inšušinak also were engaged in

temple (re)building for this deity.

21 In an Old Assyrian royal inscription and in some Middle Assyrian texts he appears as one of the seven divine

judges.
 jan tavernier

in this passage and that George (: ) considers this fragment to deal with lightning,
a plausible suggestion. Consequently, the notion of Fire as an underworld concept should be
discarded.
In any case, the greatest part of the concepts attested in the texts is in full accordance
with the general Mesopotamian afterlife concepts. Nevertheless, there are three elements that
are less Mesopotamian: Inšušinak, Weighing and the Judgement (Steve & Gasche : –
).

. The Non-Mesopotamian Context

.. Inšušinak22
The presence and the important role of the god Inšušinak may be explained in two ways: either
we see here an Elamite concept (Inšušinak as lord of the netherworld), or the Mesopotamians
themselves are responsible for this role: they conferred the kingship of the netherworld to
the deity which the Susians considered their supreme god. In this context, it should not
be forgotten that in the Vision of the Netherworld (line ) three Elamite divinities (Iapru,23
Humpan24 and Naprušu25) are named as protectors of a ghost (etemmu) (Bottéro : –
): ˙

And whose person (i.e. of the ghost of a king) Iapru, Humpan and Naprušu protect, whose seed
they preserved in well-being, whose army and camp they kept safe, lest a charioteer approach him
in battle.26
These three Elamite divinities are also attested, even in the same order (von Soden : ),
in a composition called Šurpu, a corpus of incantations, prayers and instructions for magic
practices. Its second tablet contains invocations to various deities, asking them to forgive and
release the patient who has committed, consciously or unconsciously, a sin (Reiner : ).
In lines – Mesopotamian deities ant places where they were active are mentioned.
Thereafter the text goes as follows:
(–) In Susa, may Inšušinak and Lahuratil27 release.
(–) May Iapru, Humpan and Naprušu release, these sublime gods.28

22 City god of Susa and god par excellence of the Elamite kings. He is amply attested in both Akkadian and Elamite

texts and was identified by Assyrian scholars with Ninurta and Adad (Hinz –: ).
23 Iapru is only attested in Mesopotamian texts. His function is not very clear, but he is identified with Anu (Šurpu.

Comm. B:: [dA]-nu nimki “Anu of Elam”) as well as Enlil (An = Anum vi ; BA  :). In An = Anum vi 
a variant spelling of his name, dIb-nu, is identified with Anu. A town named Iapru was situated in Elam (Edzard
–; Lambert –b).
24 Supreme god of Elam since the nd Millennium bc. His name is derived from the El. root hupa- “to command”.

In a treaty between Naram-Sin of Agade and a king of Awan he was only the second deity, after Pinenkir, the mother-
goddess (Hinz –: ; Koch : ). The Mesopotamian scholars identified him with Enlil (Šurpu.
Comm. B:). He was the lord of Heaven (Koch –: ).
25 Certainly identical with Napiriša (Reiner : ; Koch –: ). This popular Elamite deity, whose

name means “great god”, was the lord of the earth and is identified with Ea in a commentary on Šurpu (Comm. B:),
a rather logical equation. According to Hinz (), Napiriša was during the Middle Elamite period a taboo name
for Humpan, but it has been demonstrated that this is untrue (Koch –: ).
26 [dI]a-ap-ru dHum-ba dNap-ru-šú zu-mur-šú na-as-ru u-šal-la-mu numun-šú ki-kal-su ú-še-za-bu [i-n]a tam-
˘ la iq-ru-ba?-á[š-š]ú?.
ha-ri lú nar?-ka?-ba-ti ˙
˘ 27 Goddess whose name means “Creator of man”, attested from the Ur-III period till the Neo-Elamite period
(Henkelman : ). Identified with Ninurta (Šurpu. Comm. B:).
28 () Lip-tu-ru ina Šu-ú-šiki dMúš.eren () u dLa-hu-ra-til () dIa-ap-ru dHum-ba-[an dNap]-ru-šu lip-tu-
˙
ru () dingir.meš šur-bu-tú. ˘ ˘ ˙
 elamite and old iranian afterlife concepts 

It should be noted that here the Elamite gods are presented in a positive role, as deities who
have the power to release a patient. The scribe nicely defined the territory where the gods were
active, Susa (and vicinity). The positive role of the Elamite gods further implies that Inšušinak
is a genuine Elamite netherworld concept, not that this role was conferred on him by the
Mesopotamians.

.. The Weighing Ceremony

The Weighing ceremony (MDP  –) is nowhere attested in the Mesopotamian doc-
umentation (Van der Stede : ). The texts discussed here give us barely information
on how this ceremony proceeded: there is a person called mušeqqilu “Weigher”, who further
remains unidentified. It is certainly not Inšušinak (contrary to Bottéro’s opinion [: ]
that “Weigher” is an epithet for Inšušinak), Išme-karāb or Lāgamāl. The deceased person makes
an official statement, in which he says that he descended on a black cloud.
The “Weighing” is of course best known from Egypt, where it played a very important role
in the judgement of the deceased persons. From there it penetrated into biblical literature, e.g.
in Job (, : “Let him weigh me with honest scales”), where even the verb šaqal “to weigh” is
used. Two other examples are Proverbs ,  (“All the ways of a man are clean in his own eyes,
but the Lord weighed his spirit”) and , (“Every way of a man is right in his own eyes, but the
Lord pondered the hearts”).
It is not very likely that this aspect, known in the Egyptian afterlife tradition, has come
directly from Egypt to Susa. It would indeed be quite strange to see that this Egyptian concept
would have reached Elam without having left any trace in Mesopotamia. Rather one is dealing
here with an Elamite idea.

.. Judgement
The third non-Mesopotamian idea is that of Judgement (dı̄nu, ex. in MDP  :), executed
by Inšušinak and his assistants Išmekarāb and Lāgamāl. These two assistants assume the role of
lawyer-defender and accuser, as can be derived from the meaning of their names: Išme-karāb
means “He hears the prayer”, Lāgamāl means “(He who has) no mercy”.
Although the Mesopotamian scribes also mention “judgements” of deceased persons, it is
not quite identical with what is going on in our texts. The Mesopotamian judgement should not
be compared with the Egyptian and Christian traditions, where the deceased is really judged.
Admittedly the Anunnaki were sometimes presented as “judges” (e.g. in LKA  iii –:29
“You are the Anunnaki, who deliver judgement to the people in the netherworld and decide
law cases of the entire humanity”), but in Mesopotamia these expressions cover more than a
real judgement. When the Anunnaki “proclaim their judgement” in Inanna’s Descent to the
Netherworld (Sumerian version, line ), this only implies that they will keep the goddess
captured in the netherworld, treating her according to the same netherworld laws as they treat
other mortals. It was the judge’s task to fix the destiny, which was not performed by (judicial)
logics but instead was an act of power (Bottéro : ; Van der Stede : –).

As may be expected, the question of the provenance of these three concepts imposes itself.
According to Bottéro (: ) the texts could reflect “directement ou après syncrétisme avec
l’idéologie mésopotamienne” Suso-Elamite ideas on the netherworld. Let us not forget that the
documents were found in Susa, a melting pot of Mesopotamian and Elamite culture. It is thus

29 And duplicates (cf. Farber :  and –).

 jan tavernier

perfectly possible and even plausible that some Elamite afterlife concepts may be integrated
in these texts. Nevertheless they are written in Babylonian, which indicates that they reflect
Mesopotamian afterlife concepts that perhaps had been mixed with Elamite ideas.
Unfortunately these concepts are hardly known to modern scholarship. Apart from these
texts, there are no Elamite written sources which could provide some direct information on
how the Elamites saw death and the netherworld. Other information on these issues comes
from archaeological sources (e.g. tombs) but is extremely scanty, as shown in the introduction.
As a consequence, it is difficult to judge in what measure Weighing and the Judgement are
Elamite afterlife concepts, although it is probable that they are.
It is thus impossible to compare the non-Mesopotamian concepts with Elamite afterlife ideas.
One could, however, compare them with the concepts believed by the other, albeit later arrived,
population group in Elam: the Iranians. This is what Steve & Gasche have initiated in their study
of the texts.

. An Old Iranian Context?

In their article Steve and Gasche (: –) have compared the triad Inšušinak, Išme-
karāb and Lāgamāl with the Old Iranian triad Mithra, Sraoša and Rašnu. One of their arguments
is the similarity in their names. Although Inšušinak and Mithra’s names have nothing in com-
mon, the names of Sraoša and Išme-karāb both contain the concept of “hearing, hearkening”
(Kreyenbroek : ; Kellens : ).
The meanings of Lāgamāl and Rašnu’s names are also completely different and even con-
tradictory: “(She who has) no mercy” and “Justice”. The meaning proposed by Steve & Gasche
(“The powerful one” > “The hard one, the tough one”) in order to link both names, is clearly
wrong and has no philological basis.
The role of Inšušinak, Išme-karāb and Lāgamāl is unfortunately not as clear as the role played
by Mithra, Sraoša and Rašnu in the Iranian mythology. Anyhow, Steve and Gasche pointed to
the striking similarities between both triads and cautiously wondered if some play of cultural
influence is going on here: Elamite on Iranian or Iranian on Elamite. Nevertheless they humbly
admitted that they too were still far from “le mot de la fin” (Steve & Gasche : ).
We are well informed on the Old Iranian afterlife concepts, which are spread over two
religious systems: () pre-Zoroastrianism and () Zoroastrianism (Boyce : –, –
; Kellens : ). In any case, the concepts themselves are not that different.
According to pre-Zoroastrian beliefs the soul lingered on earth for three days after the death
of someone. After this short period it started its ascent to the “Crossing of the Separator”
(Činvatō Pərətu), a bridge which the deceased had to cross to reach Paradise. Underway it was
met by a female figure (Daēna30), which was young and beautiful for the righteous people and
old and ugly for the evil people. This meeting with the Daēna already indicated what fate the
soul could expect.
At the “Crossing of the Separator” the soul was directed by the Daēna to its future home,31
either Paradise (for righteous people), a place full of light and happiness, or the netherworld (for
evil and wicked people). The netherworld was situated beneath the earth and was a shadowy
and gloomy place where the deceased “enjoyed” a grey continuance.

30 Not to be confused with the yazata Daēna “Religion”. It is generally accepted that there are two nouns Daēna,

both derived from the root dı̄- “to see”. Daēna “Religion” can mean “that which is seen or recognized (by the truth)”.
The name of the female figure of the Činvatō Pərətu means “she who sees or recognizes (the truth)”.
31 This role is similar to the role of Išme-karāb and Lāgamāl in the procession (cf. supra).
 elamite and old iranian afterlife concepts 

Righteousness was not an ethical concept, but a cultic one. Those who had acquired merit in
the sight of the gods, mainly by keeping prescribed observances and by sacrificing, had more
chance to reach Paradise. They succeeded in crossing the Činvatō Pərətu, whereas the evil people
who had neglected their duties towards the gods fell from it into the shadowy netherworld.
When Zoroaster preached his religion, he seems to have adopted and further developed these
ancient beliefs. Two changes are important: a third place, Hell, is now created and the ethic
aspect replaces the cultic one.
Arrived at the Činvatō Pərətu, the soul was subjected to a moral judgement, in which favour
bought from the gods was no longer of any importance: each man’s own words and deeds were
weighed on a balance. Clearly the judgement has become an ethical judgement, because, when
the good actions outweighed the evil ones, the soul could cross the Činvatō Pərətu and continue
its journey towards Paradise. If, however, the bad actions were heavier than the good ones and
the soul appeared to be wicked, it was sent to Hell, a “dwelling place of Worst Purpose” (Yasna
.).
If good and evil were of the same amount, the soul was sent to a third place: the Misvan
Gātu (“Place for the Mixed Ones”), an abode of shadows, comparable to the pre-Zoroastrian
netherworld, a place lacking joy, sorrow or other emotions.
In the Iranian tradition this moral judgement is taken care of by three judges: Mithra, the
lord of the covenant, is the president, while Sraoša “Hearkening”, the yazata of Obedience, and
Rašnu, yazata of Justice, are his fellow judges. It is Rašnu who holds the scales of the balance.
This triad is especially present in the later Pahlavi texts, not in the Avesta itself.
Mithra probably had separate links with Rašnu and Sraoša before the triad emerged. Mithra’s
and Rašnu’s names are attested in “old” hymns, e.g. the Farwardı̄n Yašt (Yt. .,) and the
Bahrām Yašt (Yt. .) and in the Vendidād (Vend. .–), old in content, albeit not in
form. The links between Mithra and Sraoša appear prominently in Yasna . The development
towards a triad of Mithra, Rašnu and Sraoša may possibly be noticed in the confusion seen
between this triad and a triad Sraoša-Aši-Nairiiō.Saŋha.
In all likelihood Sraoša became only closely associated to Rašnu at the time when Zoroas-
trianism reached Pars (Kreyenbroek : –). Nevertheless it may be assumed that the
Pahlavi texts reflect older beliefs. In fact, the oldest traces of the connection between Mithra,
Sraoša and Rašnu are to be found in the Zoroastrian calendar (designed under command of
Artaxerxes II), where they have the day-names of days – (Yasna :; cf. Boyce :
– and Kreyenbroek : ). The occurrence of the Iranian loanword srwšyt “pun-
ishment” in an Egyptian Aramaic document from the late fifth century bc (Tavernier :
 no. ...) could indicate that the judicial connotations of Sraoša were already present
in that period (Kreyenbroek : ). It is not sure either, that the personal name *Çaušaya-
(Tavernier :  no. ..), an Old Persian32 equivalent of *Sraušaya-, refers to the yazata
Sraoša. This name is attested in the Fortification Archive.
In three Fortification texts a place name *Srauša- (Tavernier :  no. ..) appears
as travel destination. Two other toponyms have the Old Persian equivalent of Av. sraoša-, being
*çauša-: *Çauša- and *Çaušaka- (Tavernier :  no. ..–).
These toponyms, however, seem to be named after a person and as srauša- / *çauša- is also
an adjective “hearkening; obedient”, it is not sure that the divine name is intended here. This is
also valid for the personal name cited above. In any case, all these proper names do not inform
us on the cult of Sraoša in Achaemenid times (Kreyenbroek : ).

32 On the relation between Av. /sr/ and OP /ç/, see Kent (: –).
 jan tavernier

It seems that Sraoša was still a minor deity in the later Achaemenid period. Only in the
Parthian period (rd century bc—rd century ad) his popularity increased (Kreyenbroek :
–).33 Admittedly, this may cause some problems for an equation of Sraoša and Išme-
karāb, as the time gap between both divinities is very large.
The same chronological gap exists with regard to Rašnu, the representation of Justice. He
was in all likelihood the last judge who joined the panel, converting it thereby into a triad. As
he appears in the Old Iranian names attested in the Persepolis Fortification Texts,34 according
to Boyce a time when Zoroastrian influence was not yet active in name giving, he must have
enjoyed at that time some popularity in Western Iran (cf. Boyce : ; Kreyenbroek :
; Steve & Gasche : ). It should, however, be noted that he is not attested in the Iranian
names from Neo-Elamite Susa and that the element is only attested in four personal names
from Persepolis: *Rašnubara-, *Rašnuca-, *Rašnudāta- and *Rašnuka- (Tavernier : 
nos. ..–), attested in the period –bc.35 This is extremely little, illustrating
that his popularity may not have been that high. Could it be that these names belonged to
Zoroastrians who did not originate from Western Iran and, as a consequence, that the end of
the th century was the time when the new religion reached Pars? For the sake of completeness,
it should be noted that two Persepolis Fortification toponyms contained the element Rašnu-
(*Rašnuca- and *Rašnuvatı̄š; Tavernier :  nos. ..–). In all likelihood these
places were named after a person (Tavernier : ), so originally the two names are
anthroponyms.
Anyhow, there are indeed similarities between the role of the Old Iranian triad and that of
the Elamite triad, although the Susian texts withhold a lot of precise information. If there is
a link between both triads, this would imply that Lāgamāl is the Weigher (as is Rašnu in the
Old Iranian mythology). Nonetheless, Rašnu could also be identified with Išme-karāb, as both
divinities have justice as their principal responsibility. In addition, there are also differences.
The Iranian texts do not say anything on an accompanying role of Sraoša and Rašnu. They are
nothing more than judges and certainly do not go in front of a procession as described in the
Susa texts. Secondly, contrary to the Susa texts, the Iranian texts do not assign (by the meaning
of the names) a positive and a negative role to Sraoša and Rašnu.

. Towards a Conclusion

It has been clearly demonstrated that some aspects in the Elamite and Iranian tradition are
similar: () The role of the procession headed by a divine creature (Daēna in the Iranian
tradition, Išme-karāb and Lāgamāl in the Elamite tradition), () The triad, with a justice god
as one of its members, fixing the deceased’s fate (by weighing and judging) and () The names
of Išme-karāb and Sraoša. This leads to the formulation of three possible explanations of this
similarity between the Iranian and the Susian afterlife concepts:
(1) There is no influence at all, the similarity is a mere coincidence.
(2) The similarity is the result of Iranian influence on the Elamites: this is highly improbable,
as was already pointed out by Steve & Gasche, because of the high date. The Iranian people
were simply not yet in Elam at the time the texts were written.

33 Cf. also the emergence in this period (st century bc.) of personal names containing the divine element Sraoša-:

Srwšk, Srwšdtk, Srwšssnk, etc.

34 In an Aramaic text from about bc a person named *Vahyarašnu- occurs (Tavernier :  no. ..).

It is quite likely, however, that rašnu is a simple adjective here. No other names with vahya- as first element have a
divine element in them (cf. Tavernier : ).
35 The Persepolis Fortification Texts do not allow any speculations on a religious role of these four persons.
 elamite and old iranian afterlife concepts 

(3) The similarity is the result of Elamite influence on the Iranians (Henkelman : –
 n. ): if this is true, the non-Mesopotamian afterlife concepts attested in the texts
must be Elamite ones that remained active during the Middle and Neo-Elamite periods.
When the Iranians arrived in Elam they picked up these ideas and used them in their own
pre-Zoroastrian mythology. The small differences between the Elamite and the Iranian
concepts are acceptable, being the result of tiny modifications added by the Iranians.
However, this possibility is not without problems either. Again the chronology could raise
questions. As already said, Sraoša and Rašnu only became popular among the Iranians
rather late. In addition, the triad as well as the judicial functions which were attributed to
Sraoša are not attested in the oldest Iranian texts, contrary to what Steve & Gasche believe.
If these aspects only developed in the Achaemenid period (or even later) there is a time
gap of  years between the Susa Texts and the Iranian concepts. Moreover, the Susa
triad should have been transformed into an Iranian couple (Mithra and Sraoša), only to
become a triad again later on (Mithra, Sraoša and Rašnu).
One could argue that, conversely, if the Elamite concepts were still actively known in the
beginning of the Achaemenid period, they may have influenced the Iranian ideas. Nevertheless,
the Iranians must have developed their own netherworld thoughts before arriving in Elam. It is
therefore more probably to consider the similarity as a coincidence and to hold that the Iranians
brought their pre-Zoroastrian afterlife concepts with them when they arrived in Elam.

Bibliography

Álvarez-Mon, J. (): “Elamite Funerary Clay Heads”, Near Eastern Archaeology , –.
Amiet, P. (): L’âge des échanges inter-iraniens – avant J.-C., Paris.
André-Salvini, B. (): “Funerary Tablet”, P.O. Harper, J. Aruz & F. Tallon (eds.), The Royal City of Susa:
Ancient Near Eastern Treasures in the Louvre, New York, .
Aynard, J.-M. (): “Le Jugement des morts chez les Assyro-Babyloniens”, Le jugement des morts:
Egypte ancienne—Assour—Babylone—Israel—Iran—Islam—Inde—Chine—Japon (Sources Orientales
), Paris, –.
Berlejung, A. (): “Tod und Leben nach den Vorstellungen der Israeliten: Ein ausgewählter Aspekt
zu einer Metaphor im Spannungsfeld von Leben und Tod”, B. Janowski & B. Ego (eds.), Das biblische
Weltbild und seine altorientalischen Kontexte (Forschungen zum Alten Testament ), Tübingen, –
.
Bork, F. (): “Šutruk-Nahhunte A”, OLZ , –.
Bottéro, J. (): “La mythologie de la mort en mésopotamie ancienne”, B. Alster (ed.), Death in
Mesopotamia: papers read at the XXVI e Rencontre Assyriologique Internationale (Mesopotamia ),
Copenhagen, –.
———. (): “Les inscriptions cunéiformes funéraires”, G. Gnoli & J.-P. Vernant (ed.), La mort, les morts
dans les sociétés anciennes, Cambridge, –.
Boyce, M. (): A History of Zoroastrianism. Vol. : The Early Period (HdO ////A), Leiden.
———. (): A History of Zoroastrianism. Vol. : Under the Achaemenians (HdO ////A), Leiden.
Carter, E. (): “Landscapes of Death in Susiana During the Last Half of the nd Millennium B.C.”,
J. Álvarez-Mon & M.B. Garrison (eds.), Elam and Persia, Winona Lake, –.
De Meyer, L. (): L’accadien des contrats de Suse (IrAnt. Suppl. ), Leiden.
Dossin, G. (): Autres textes sumériens et akkadiens (MDP ), Paris.
Ebeling, E. (): Tod und Leben nach den Vorstellungen der Babylonier, Berlin.
Edzard, D.O. (–): “Jabru”, RlA , .
Farber, W. (): Beschwörungsrituale an Ištar und Dumuzi: attı̄ Ištar ša harmaša Dumuzi (Akademie
der Wissenschaften und der Literatur. Veröffentlichungen der orientalischen˘ Kommission ), Wies-
baden.
George, A.R. (): The Babylonian Gilgamesh Epic, Oxford.
Ghirshman, R. (): “Têtes funéraires en terre peinte des tombes élamites,” Mitteilungen der Anthro-
pologischen Gesellschaft in Wien , –.
 jan tavernier

Grillot, F. (): “A propos de la notion de subordination dans la syntaxe élamite”, JA , –.
———. (): “La postposition génitive -na en élamite”, CDAFI , –.
———. (): “Le ‘suhter’ royal de Suse”, IrAnt , –.
———. (): “Kiririša”, L. De Meyer & H. Gasche, Fragmenta Historiae Elamicae: Mélanges offerts à
M.-J. Steve, Paris, –.
———. (): “Le monde d’en bas en Susiane”, RA , –.
Hakemi, A. (): Shahdad: Archaeological Excavations of a Bronze Age Center in Iran, Rome.
Henkelman, W.F.M. (): “De goden van Iran: (breuk)lijnen in een religious landschap, ca. –
v. Chr.”, Phoenix , –.
———. (): “Ruhurater”, RlA , .
———. (): The Other Gods Who Are: Studies in Elamite-Iranian Acculturation Based on the Persepolis
Fortification Texts (Achaemenid History ), Leiden.
Hinz, W. (): “Zum elamischen Wortschatz”, ZA , –.
———. (): “Die elamischen Inschriften des Hanne”, A locust’s leg: Studies in honour of S.H. Taqizadeh,
London, –.
———. (): “The Elamite God d.Gal”, JNES , –.
———. (–): “Humban”, RlA , –.
———. (–): “Inšušinak”, RlA , –.
Hinz, W. & Koch, H. (): Elamisches Wörterbuch (AMI. Erg. ), Berlin.
Hüsing, G. (): Die Sprache Elams, Breslau.
———. (): Die einheimischen Quellen zur Geschichte Elams. I. Teil: Altelamische texte in Umschrift mit
Bemerkungen, einer Einleitung und einem Anhang, Leipzig.
Kellens, J. (): “L’âme entre le cadavre et le paradis”, JA , –.
Kent, R.G. (): Old Persian: Grammar, Texts, Lexicon (AOS ), New Haven.
Khačikyan, M. (): The Elamite Language (Documenta Asiana ), Rome.
Kleveta, A. (): “Le jugement infernal dans les croyances babyloniennes”, ArOr , –.
Koch, H. (): Die religiösen verhältnisse der Dareioszeit: Untersuchungen an Hand der elamischen
Persepolistäfelchen (GOF. Ir. ), Wiesbaden.
———. (–): “Napiriša”, RlA , –.
König, F.W. (): Die elamischen Königsinschriften (AfO. Beih. ), Graz.
Kreyenbroek, G. (): Sraoša in the Zoroastrian tradition (Orientalia Rheneo-Traiectina ), Leiden.
Lambert, M. (): “Epigraphie élamite (II)”, RA , –.
Lambert, W.G. (): Babylonian Wisdom Literature, Oxford.
———. (–a): “Išme-karāb”, RlA , –.
———. (–b): “Jabnu”, RlA , .
———. (–): “Lagamal”, RlA , –.
Malbran-Labat, F. (): Les inscriptions royales de Suse: Briques de l’époque paléo-élamite à l’Empire
néo-élamite, Paris.
———. (): “La fête en Elam dans le culte royal et les cérémonies populaires”, M. Mazoyer, J. Pérez
Rey, F. Malbran-Labat & R. Lebrun (eds.), La fête. La rencontre des dieux et des hommes (Collection
Kubaba. Série Actes ), Paris, –.
de Mecquenem, R. (): “Fouilles de Suse, campagnes des années , , ”, RA , –.
de Mecquenem, R., Contenau, G., Pfister, R. & Belaiew, N. (): Archéologie susienne (MDP ), Paris.
de Miroschedji, P. (): “Le dieu élamite au serpent et aux eaux jaillissantes”, IrAnt , –.
Parrot, A. (): Le “Refrigerium” dans l’Au-delà, Paris.
Pézard, M. (): Mission à Bender-Bouchir: documents archéologiques et épigraphiques (MDP ), Paris.
———. (): “Reconstitution d’une stèle de Adda-hamiti-In-Šušnak”, Babyloniaca , –.
Reiner, E. (): Šurpu. A Collection of Sumerian and Akkadian Incantations (AfO. Beihefte ), Graz.
———. (): “Inscription from a royal Elamite tomb”, AfO , –.
Scheil, V. (): “Textes funéraires”, RA , –.
———. (): “Déchiffrement d’un document anzanite relatif aux présages”, RA , –.
Scurlock, J.A. (): “Death and the Afterlife in Ancient Mesopotamian Thought”, J. Sasson (ed.),
Civilizations of the Ancient Near East, New York, –.
von Soden, W. (): review of Ebeling (), OLZ , –.
———. (): “Die Unterweltsvision eines assyrischen Kronprinzen nebst einige Beobachtungen zur
Vorgeschichte des Ahiqar-Romans”, ZA , –.
˘
Spycket, A. (): “Funerary heads”, P.O. Harper, J. Aruz & F. Tallon (eds.), The Royal City of Susa:
Ancient Near Eastern Treasures in the Louvre, New York, –.
 elamite and old iranian afterlife concepts 

Steve, M.-J. (): Tchogha Zanbil (Dur-Untash). Volume III: textes élamites et accadiens de Tchogha
Zanbil (MDP ), Paris.
———. (): Nouveaux mélanges épigraphiques: inscriptions royales de Suse et de la Susiane (MDP ),
Nice.
Steve, M.-J. & Gasche, H. (): “L’accès à l’au-delà, à Suse”, H. Gasche & B. Hrouda (eds.), Collectanea
Orientalia: histoire, arts de l’espace et industrie de la terre. Etudes offertes en hommage à Agnès Spycket
(CPOA ), Paris, –.
Stolper, M.W. (–): “Nahhunte”, RlA , –.
Stolper, M.W. & Wright, H.T. (): “Elamite Brick Fragments from Chogha Pahn East and Related
Fragments”, F. Vallat (ed.), Contribution à l’histoire de l’Iran: mélanges offerts à Jean Perrot, Paris, –
.
Tavernier, J. (): “Iranian Toponyms in the Elamite Fortification Archive”, Beiträge zur Namen-
forschung N.F. , –.
———. (): Iranica in the Achaemenid period (ca. –B.C.): Lexicon of Old Iranian Proper Names
and Loanwords, Attested in Non-Iranian Texts (OLA ), Leuven.
———. (): “Migrations des savoirs entre l’Élam et la Mésopotamie”, Res Antiquae , –.
Vallat, Fr. (): “Hutelutuš-Inšušinak et la famille royale élamite”, RA , –.
———. (): “Religion et civilisation élamites en Susiane”, Dossiers Histoire et Archéologie , –.
———. (a): “Le caractère funéraire de la ziggurat en Elam”, NABU /.
———. (b): “Inšušinak, Ea et Enzag”, NABU /.
———. (): “Elam. VI. Elamite Religion”, EncIr , –.
———. (–): “Suse. G: La religion suso-élamite”, Supplément au Dictionnaire de la Bible , Paris,
–.
———. (): “Chroniques bibliographiques. . Curiosités élamites”, RA , –.
Vallat, F. & Grillot, F. (): “Dédicace de Šilhak-Inšušinak à Kiririša”, IrAnt , –.
Van der Stede, V. (): Mourir au pays des deux fleuves: L’au-delà mésopotamien d’après les sources
sumériennes et akkadiennes (Lettres Orientales ), Leuven.
 III

GEOGRAPHICAL PERSPECTIVES
 GEOARCHAEOLOGICAL RESEARCH IN
LOWER KHUZESTAN: STATE OF THE ART*

V.M.A. Heyvaert,** P. Verkinderen† and J. Walstra††

. Introduction

Up to now, geoarchaeological research in the Mesopotamian region has primarily focussed on

the evolution of the floodplain of the ‘twin rivers,’ Tigris and Euphrates, and Upper Khuzestan,
because remains of the great ancient civilizations have been discovered in these two areas. For
the Lower Khuzestan plain such information is still lacking. Within the framework of the Bel-
gian Interuniversity Attraction Pole ‘Greater Mesopotamia: Reconstruction of its Environment
and History’ (IAP /, and its predecessor IAP /), research was initiated on the landscapes
of Lower Khuzestan (Fig. ). The main goal of this research project is to investigate the history
of human-environmental interactions, i.e. how humans adapted to and/or changed their envi-
ronment. The multidisciplinary team covers a wide range of research fields, including geology,
archaeology, history and remote sensing.
In , with the cooperation of Iranian colleagues, two field campaigns in the Lower
Khuzestan plain were undertaken to collect geological and archaeological data. In addition,
field control was done to verify the remote sensing data. The data from these surveys were pub-
lished in a number of progress reports in the journal Akkadica (Baeteman et al. /;
Gasche/Paymani ). This paper presents an overview of the research carried out since then,
including new evidence on the Holocene palaeoenvironmental evolution of the plain, in partic-
ular the positions of the Persian Gulf coastline and the main rivers. The reconstruction is based
on the analysis of the geological and archaeological data collected during the field campaigns
in , and new evidence derived from textual sources, maps, satellite images and aerial pho-
tographs. It concerns mainly the integration of the results of three recently completed PhD stud-
ies (Heyvaert ; Ooghe ; Verkinderen ) and additional remote sensing data (Wal-
stra et al. ). By means of a number of case-studies, the different datasets were integrated
with a recently completed geomorphological map (Walstra et al. a–b; Heyvaert et al. ).
The case-studies provide new insights into the complex human-environmental interactions in
the plain, and demonstrate the added value of a multidisciplinary approach in such studies.

* The research was undertaken within the framework of the Interuniversity Attraction Pole “Greater Mesopota-
mia: Reconstruction of its Environment and History” (IAP /), funded by the Belgian Science Policy. All Landsat
and CORONA data are available from the USGS; the CORONA imagery of mission – was provided by the
Center for Ancient Middle Eastern Landscapes, University of Chicago; the SPOT images were provided by the
Belgian Earth Observation Platform. Mina Alizadeh, Beshad Askari, Dariush Baratvand and Abdol Reza Paymani
of the Iranian Culture Heritage Organization in Ahwaz, and Hermann Gasche are thanked for their support during
the two field surveys in . Mark Van Strydonck has provided the calibration of the radiocarbon datings. Cecile
Baeteman and Henk Weerts are thanked for the many discussions. Olivier Wambacq is thanked for the skillful
production of Figures  to . This paper is a contribution to the INQUA Commission on Coastal and Marine
Processes.
** Royal Belgian Institute of Natural Sciences, Geological Survey of Belgium.
† Ghent University and The Netherlands-Flemish Institute in Cairo.
†† Formerly Ghent University, now Royal Belgian Institute of Natural Sciences, Geological Survey of Belgium.
 v.m.a. heyvaert, p. verkinderen and j. walstra

Fig. . Location map of the Khuzestan plain.

. Regional Setting

The Khuzestan plain is located in southwestern Iran and geologically forms the southeastern
extension of the Mesopotamian sedimentary basin. In the north and east the plain is bordered
by foothills of the Zagros Mountains, in the south by the Khor Musa tidal inlet and Persian Gulf,
and in the west by the Tigris and Shatt al-Arab estuary (Fig. ) Subsidence of the Mesopotamian
 geoarchaeological research in lower khuzestan: state of the art 

basin and uplift of the Zagros Mountains are associated with the collision of the Arabian and
Eurasian tectonic plates (Haynes/McQuillan ; Audley-Charles et al. ; Vita-Finzi ).
The orogenesis started during Late Miocene and is still ongoing (Hessami et al. ).
The Khuzestan plain is bisected from northwest to southeast by a series of anticlines: the
Ahwaz, Marun, Agha Jari and Rag-e Safid anticlines. This paper focuses on the extremely flat
lower half of the plain (c. ,km2). Five perennial rivers coming down from the Zagros
Mountains drain into the lower plain: the Karkheh, Karun, Kupal, Jarrahi and Zohreh. Only
the Karun and Zohreh reach the Persian Gulf, while the others empty into the Hawiza and
Shadegan Marshes. The rivers receive most of their discharge from autumn and winter rains
in the mountains, which cause extensive seasonal flooding of the marshes and changes in
vegetation density.
The present coastline of Lower Khuzestan is shaped by a tidal regime. The tidal range averages
c. –m along the coastline, increasing up to – m inside the Khor Musa tidal embayment
(Höpner ; Admirality Tide Tables ). At the city of Khorramshahr, located  km
upstream on the Shatt al-Arab estuary, the tidal amplitude averages c.  m. The coastline is
fringed by large tidal flats, salt marshes and sabkhas. There is no freshwater inflow in the
intertidal area, except in the case of extreme flood events.
The climate of Khuzestan is generally hot and arid, but some climatic division can be made
in relation to the general relief. Lower Khuzestan falls within the arid zone with annual rainfall
below mm; towards the north and east rainfall amounts rapidly increase with height (Potts
). In summer temperatures may rise up to .ºC, while in winter they may fall below zero
(Johnson : –; Potts ).

. Previous Palaeogeographical Research

.. Position of the Persian Gulf Coastline

Since the early nineteenth century, historians and geomorphologists have debated the Holocene
evolution of the Lower Mesopotamian plain, based on archaeological data, historical sources
and surface observations. These early investigators were interested in the changes in the position
of the Persian Gulf shoreline as a result of the postglacial sea-level rise.
The earliest theories claimed that the head of the Gulf shifted far north of its present position,
followed by a gradual retreat of the Gulf caused by delta progradation*1 during prehistoric
and historic ages. Beke () placed the northern limits of the Persian Gulf inland of the
Mesopotamian plain as far as Samara ( km north of Baghdad). Ainsworth () presented
reports of a geological reconnaissance in southern Mesopotamia and suggested that the front
of the delta had prograded over a distance of  km to its present position. De Morgan ()
produced two maps (bc and bc) based on accounts and reports from historical sources
(cf. Fig.  A and B in Baeteman et al. /). The maps show the presumed position of
the former coastline of the Persian Gulf between the cities of Basra and Amara.
Lees and Falcon (), on the contrary, challenged the nineteenth century concepts, and
claimed that there was no evidence for the occurrence of an extensive marine flooding followed
by delta progradation since the early Pliocene. The authors suggested a delicately balanced sys-
tem between subsidence (neotectonic effects) and sedimentation on occasion of local marine
inundations. Nevertheless, they reported sediments containing marine and estuarine shells
found in the subsoil of the plain as far inland as Amara. The same authors also invoked subsi-
dence caused by neotectonic movements to explain the flooding of the Sasanian (–ad)

1 Geological and geomorphological terms marked with * are explained in the appendix.
 v.m.a. heyvaert, p. verkinderen and j. walstra

and Abbasid (–ad) irrigation canals nearby the present-day Khor Zubair (Iraq) and
Khor Musa tidal embayments (Fig. ). The formation of the Khor Musa tidal area was attributed
to local subsidence and interpreted without further precision as being very young. Hudson et al.
() agreed with the views of Lees and Falcon (), contradicting their own identification
of a landward extending Holocene marine unit (Hammar Formation) underlying the fluvial
deposits of the Shatt-el Arab region.
The tectonic scenario as claimed by Lees and Falcon () has been strongly criticised in the
’s (Purser ; Larsen/Evans ; Evans ). These authors asserted that the Shatt-el
Arab region has been more influenced by eustatic sea-level changes and deltaic progradation
than by tectonic events. Macfyden and Vita-Finzi () suggested on the basis of faunal
evidence and the presence of the Hammar Formation that a marine embayment extended as
far inland as Amara, followed by an overall delta progradation over a distance of about 
to km during historical times. Later research carried out in the Persian Gulf area also
supported the view that Holocene sea-level changes controlled the evolution of the Shatt-el
Arab region, rather than tectonics (Rzoska ; Ya"acoub et al. ; Purser et al. ; Al-
Zamel ; Al-Azzawi ; Sanlaville ; Baltzer/Purser ; Aqrawi ; Aqrawi/Evans
; Lambeck ; Aqrawi ; Sanlaville ; Dalongeville/Sanlaville, ).
In literature, little is known about the post-glacial evolution of relative sea level (RSL) in the
Persian Gulf. According to the RSL curve of Dalongeville and Sanlaville () it is assumed that
sea level rose progressively in the Gulf basin from , years bp onwards. Their reconstruction
shows a particularly rapid rise between  bp and  bp, reaching a maximum at c. 
bp of at least one or two meters above the present-day level, followed by a gradual sea-level fall,
upon which some oscillations are superimposed. Dalongeville and Sanlaville () identified
four sea-level highstands (transgressions*) during the period – cal years bc and four
sea-level lowstands (regressions*) during – cal years bc. Moreover they suggested that
the maximum amplitude of RSL change in the period – cal years bc averages . to
m. The indicative meaning and age of the sea-level index points used by Dalongeville and
Sanlaville () for the reconstruction of the fluctuating RSL curve has been critically reviewed
by Heyvaert and Baeteman ().
Sanlaville (, ) and Dalongeville and Sanlaville () proposed a new general
scheme for the evolution of Lower Mesopotamia on the basis of their previously published
RSL curve. The authors produced three palaeogeographical maps showing the position of the
Persian Gulf shoreline at c. bc, during the Hellenistic period (c.  bc) and the Medieval
period (th century ad), respectively. They concluded that the presumed shoreline of the Per-
sian Gulf at the post-glacial maximum (bc) extended as far as the present-day towns of
Nasiriya, al-Amara and Ahwaz; in Lower Khuzestan the marine transgression was halted by the
series of anticlines. They suggested that the post maximum sea-level period was marked by a
rapid progradation of the Tigris-Euphrates-Karun delta, but they also mentioned that they did
not know when the Gulf reached the position of its present shoreline. They proposed that during
the Hellenistic period (–bc) the coastline was located south of the present-day one, with
a RSL at about one meter below the present-day level as demonstrated in Bahrein and Failaka
(Dalongeville, ). Based on the latter, Sanlaville and Dalongeville rejected the map (bc)
proposed by De Morgan (). Already in , Hansman (b) contradicted De Morgan
() and considered that the southern limit of the Mesopotamian delta was very near to the
present one during Hellenistic period. Hansman (b) claimed, on the basis of historical
texts, that the Persian Gulf coastline has not changed appreciable since the Hellenistic period.
On the contrary, Sanlaville () and Dalongeville and Sanlaville () claimed that since
the Hellenistic period the coastline did not remain stable and proposed a Medieval (th cen-
tury ad) RSL high stand, which implies an inland extension of the Gulf as far as the present-day
 geoarchaeological research in lower khuzestan: state of the art 

city of Abadan. Between Basra and Kufa, the authors drew an extensive marsh, which developed
due to a rising groundwater table, associated with the RSL rise. The extent of this marsh and
the landward limit of the Gulf are based on Arabic texts, as analysed by Le Strange ().
It should be mentioned that for Lower Khuzestan, the reconstruction of the coastline by
Sanlaville () and Dalongeville and Sanlaville () is not based on geological data.
Therefore, the lateral extent of the Holocene marine deposits in the subsoil of the vast plain
was not accurately known. The only indication of marine deposits was reported by Thomas
(quoted in Lees and Falcon : ) who found marine and estuarine deposits north of Bandar
Shahpur (now Bandar-e Imam Khomeini).

.. Position of the Rivers

In previous studies, relatively little attention has been paid to the evolution of the alluvial
systems of Lower Khuzestan. A first notion comes from Lees and Falcon (), who during
their geological investigations in Khuzestan noted a “large deltaic fan”, which was deposited
by the Jarrahi river before it was deflected westwards into the Shadegan Marshes. From
aerial photographs they also observed old irrigation networks (described as “long fingers” and
“herringbone patterns”) extending into the tidal flats of the Khor Musa embayment.
Based on aerial photographs and field observations, Hansman (, a) used the posi-
tion of (former) river courses and archaeological sites to identify the position of major settle-
ments recorded in historical sources. He was the first one to describe meander patterns of an
abandoned river across the plain between Ahwaz and the Shatt al-Arab. Hansman attributed the
meanders to a former course of the Karkheh and used it to identify the ancient city of Spasinou
Charax (Fig. ), which according to classical sources was located at the junction of the Karkheh
and the Tigris.
Kirkby () was mainly interested in the palaeorivers of Upper Khuzestan as indicators
for past water resources and included the northern part of Lower Khuzestan in his analyses. He
credited Hansman’s meander traces to a combined Karun-Karkheh flow, based on the relation-
ship between meander wavelength and bankfull discharge (though Hansman’s identification
of Spasinou Charax remained undisputed). Without further reference, Kirkby dated the river
course at some period between bc and  ad, after which the Karun shifted to a posi-
tion at or near the present one and the Karkheh shifted twice towards positions further north.
Hansman and Kirkby both mentioned traces of extensive irrigation canals, which apparently
derived water from the abandoned river course, and dated them to the Sasanian and Early
Islamic periods.
Although strictly outside our area of interest, some of Kirkby’s work in Upper Khuzestan may
be of significance to this study as well. At least for the Karkheh, Karun and Dez rivers in Upper
Khuzestan he provided evidence for a phase of continuous river aggradation* (since c. bc)
followed by a phase of down-cutting (after c.  bc). His evidence was based on a series of
dated levels of cultural material in excavated mounds, which were (partially) buried beneath
alluvium and subsequently incised by rivers. Although this same sequence of events was previ-
ously used as an argument for tectonic movements (Lees/Falcon ), Kirkby attributed it to
increasing aridity and grazing intensity throughout the Holocene.
In more recent work, the present-day Karun channel was recognized as the main channel
of a large alluvial fan with a radius of about  km. It is one in a series of many alluvial fans
flanking the external parts of the Zagros Mountain belt (Baltzer/Purser ).
Final contributions were presented in the Akkadica progress reports. A preliminary analysis
of satellite imagery confirmed the presence of the (known) former Karun and Karkheh river
courses and irrigation canals in the central plain. In addition, three detected alluvial fans
 v.m.a. heyvaert, p. verkinderen and j. walstra

were attributed to the Jarrahi; two being relict and the other currently active (Baeteman et al.
/). A thorough review of pre-Islamic historical sources by Cole and Gasche ()
resulted in the attribution of many ancient toponyms to current or abandoned river courses.
Interestingly, they embraced Kirkby’s idea of a combined Karun-Karkheh flow, as it provided
the perfect explanation for the persistent confusion between Karun and Karkheh nomenclature
throughout history.

. Current Work: Methodology and Datasets

The present research follows a multidisciplinary approach, drawing on data from different
research fields, including remote sensing, geology, archaeology and historical geography. Based
on the interpretation of satellite data, a geomorphological map was drafted (Fig. ), presenting
the distribution of past and present landforms and providing a spatial framework for the
information derived from the other disciplines.
The geological dataset is based on fieldwork carried out during two Belgian-Iranian field
campaigns in  (cf. Baeteman et al. /; Heyvaert ), which involved the facies
analysis of the sedimentary sequence of hand-operated boreholes and outcrops. During the
same missions archaeological fieldwork was carried out, encompassing the survey of ancient
settlements and resolving their age based on datable ceramics. The results of the archaeological
field campaign, combined with an overview of earlier archaeological data, were published
by Gasche and Paymani () and are summarized below for convenience. The presence of
archaeological sites in the surroundings of palaeochannels is a useful tool to obtain a reliable
chronology of channel belts. In principle, the presence of an archaeological site nearby a
channel belt gives an indication of a minimum age for that channel belt. The last data set
consulted for this study consists of historical documents, mainly in the form of () Arabic
historiographical and geographical literature from the th to the th century (Verkinderen
) and () European travel literature and cartography, dated between the th and the early
th century ad (Ooghe ; Verkinderen ). Throughout the research, a Geographical
Information System (GIS) was used for integration and interpretation of the project data.

.. Remote Sensing Data and Methodology

Given the vast size and limited accessibility to the study area, the use of remote sensing
data is crucial for obtaining a full appreciation of the landscape. A variety of resources was
exploited, differing in terms of footprint, ground resolution, spectral capability and acquisition
time. Although aerial photographs provided guidance in the field, an attempt to use them for
systematic archaeological prospection proved not very fertile (unpublished data Dupin). The
available satellite imagery provided insufficient spatial detail for the detection of archaeological
sites, but on the other hand proved very helpful for creating a geomorphological map and as a
tool for quantifying (short-term) landscape changes. All image processing tasks were performed
using standard functions in ERDAS IMAGINE software; image interpretation and mapping
procedures were carried out in ArcGIS.
 geoarchaeological research in lower khuzestan: state of the art 

Table : Remote sensing data used in this study.

Number
Sensor Acquisition date Scene IDs of bands Resolution Comments
CORONA KH-A: // Revolution D,  c. m Acquired from USGS
mission – frames F–
CORONA KH-A: // Revolution D,  c. m Acquired from USGS
mission – frames A–
CORONA KH-A: // Revolution D,  c. m Provided by CAMEL
mission – frames A–,
F–
HEXAGON KH-: // Operation ,  –m Acquired from USGS
mission – frame 
Landsat MSS // & Path , row  m Multispectral Scanner
// /
Landsat TM // Path , row  /m Thematic Mapper
/
Landsat ETM+ / & // Path /, row  //m Enhanced Thematic
/ Mapper Plus; wet
season
Landsat ETM+  & // Path /, row  //m Idem; dry season
/
Landsat ETM+  & // Path , row   //m Idem; combined
SLC-off images
SPOT // & Path  & ,  .m Provided by BELSPO
// row 
Google Earth (QB) –  .m Variable quality
Google Earth –  .m & acquisition date
(SPOT)
Aerial photographs s/ / frames  c. ./.m No stereo-pairs
SRTM –// Tile _ N/A m DEM

... Data Sources

The multispectral Landsat missions are particularly suited for geomorphological mapping at a
regional scale. In spite of recent developments in satellite sensor technology Landsat remains
a significant resource for geomorphologists due to its repeat coverage, large scene size and low
cost (Smith/Pain ). Drainage and vegetation patterns are important indicators to distin-
guish between landform units and these can be best accentuated in the near infrared region of
the electromagnetic spectrum (Lillesand/Kiefer ). Therefore, false-colour composites were
created from the raw image files. The ETM+ scenes were subject to further processing—using
the panchromatic band the image ground resolution was enhanced to  m. Because the images
have a high level of geometric accuracy (Gutman et al. ), no additional corrections needed
to be applied.
CORONA images were acquired by the first generation of US photo-reconnaissance satel-
lites between  and . They were declassified in  and have been successfully used
in geoarchaeological studies throughout the Near East (e.g. Philip et al. ; Ur ; Hritz
). Their main advantage is to provide a record of the landscape before many elements were
 v.m.a. heyvaert, p. verkinderen and j. walstra

destroyed by modern, large-scale cultivation. Drawbacks are the large image distortions due
to the oblique and panoramic camera geometry. Because GPS surveying for precise ground
control is not a realistic option in Lower Khuzestan due to security restrictions, a rigorous geo-
metric correction of the images could not be carried out. Instead, CORONA image patches of
.× .cm (corresponding to c.  ×  km on the ground) were individually geo-referenced,
based on control points obtained from the Landsat imagery. The CORONA images used in this
study are from KH-A missions, with a best ground resolution of c.  m. An image from a later
photo-reconnaissance program (HEXAGON, mission KH-), with lower ground resolution,
was also acquired.
The acquired high-resolution imagery includes a set of digitally scanned aerial photographs
and two SPOT scenes. The aerial coverage consists of black-and-white photographs with a
scale of c. /,, scanned at  dpi (resulting in a ground resolution of c.. m); unfor-
tunately no stereo-pairs were available and the coverage is rather limited. The SPOT scenes
are false-colour near infrared composites with a ground resolution of . m. As an additional
source, Google Earth provides free imagery of the study area; although the imagery has (at least
partly) a high resolution, the image quality is inferior to original source data.
A last relevant data source consists of elevation data produced by the Shuttle Radar Topo-
graphic Mission (SRTM). This is the best resolution digital elevation model (DEM) with world-
wide coverage (ground resolution is  m), and the most reliable elevation data available for the
study area. SRTM data have been used previously for the detection of alluvial ridges in the
Central Mesopotamian plain (Hritz/Wilkinson ).

... Geomorphological Mapping

A geomorphological map presents information about the form, origin, age and distribution
of landforms and their formative processes, rock type and surface materials (Brunsden et al.
). It helps the understanding of individual landforms and the landscape as a whole.
A geomorphological survey traditionally involves a preliminary interpretation based on
aerial photographs, followed by “groundtruthing” in the field. Photo-interpretation involves
the identification of surface features based on their characteristic morphology, vegetation and
drainage conditions. Due to restrictions of available material and field access (observations
were done during the two field campaigns, but not in a systematic way), an unconventional
approach was adopted, based on the interpretation of a variety of remote sensing data that are
easily accessible and inexpensive (notably Landsat & CORONA imagery).
A standardized working procedure was developed for the consistent mapping of the alluvial
landscapes of Lower Khuzestan (Walstra et al. ). In line with the motivations for this
research project, the map legend distinguishes at the highest levels on the basis of landform
genesis and chronology. As the landscapes have been subject to prolonged human activity,
man is considered as an important agent. The original map sheets were produced at a scale
of /, but a downscaled version is included here.

.. Geological Data and Methodology

During two fieldwork campaigns in ,  hand-operated cores were collected to a depth
of –m below the surface and  shallow outcrops were investigated (Fig. ). The location of
individual boreholes was recorded with a handheld GPS device. Elevations were derived from
topographical maps and site-specific measurements obtained at the regional topographic insti-
tute. In the field, a detailed facies description was done on the basis of lithology, sedimentary
structure (massive, tidal bedding, laminated, bioturbation, sharp or gradual contacts), presence
of plant remains, gypsum and salt crystals, and macrofossils. A limited number of subsamples
 geoarchaeological research in lower khuzestan: state of the art 

were taken for laboratory analyses, i.e. palaeoecological analyses (foraminifera and diatoms*)
and radiocarbon dating (organic material and shells).
The facies interpretation (i.e. at environmental level) was done on the basis of its context (in
relation to neighbouring facies) along geological transects. Knowledge of the context of a facies,
that is, the relationship of one facies to another, is essential before proposing an environmental
interpretation as a single facies can occur in different sedimentary environments. The facies
analysis of the Holocene sequence enabled identification of three sedimentary units: fluvial
(unit ), coastal (unit ; sub/intertidal*2 and supratidal* subunits) and brackish-freshwater
marsh (unit ). A more detailed description of the lithological and palaeoecological properties
of these units is given in Heyvaert () and Heyvaert and Baeteman ().
Radiocarbon dates were obtained from organic material (Table ) and provide a chronolog-
ical framework for the palaeogeographical reconstruction. Calibrated dates are given with a 
sigma error range in calendar years before present (cal bp*). Calibration was completed using
the calibration programme of Stuiver and Reimer ().
Table : AMS radiocarbon data and calibrated ages.
Geographical Laboratory Age 14C Calibrated age Sample
Site coordinates code yrs BP yrs cal BP altitude Dated material
B °’” KIA- ± – +. organic gyttja
°’”
B °’” KIA- ± – +. organic gyttja
°’”
B °’” KIA- ± – +. organic material
°’” (reworked)
B °’” KIA- ± – +. peaty mud
°’”
B °’” KIA- ± – +. peaty mud
°’”
B °’” KIA- ± – +. peaty mud
°’”
B °’” KIA- ± – +. peaty mud
°’”
B °’” KIA- ± – +. vegetation remnant
°’”
B °’” KIA- ± – +. organic gyttja
°’”
B °’” KIA- ± – +. peaty mud
°’”
B °’” KIA- ± – +. peaty mud
°’”
B °’” KIA- ± – +. organic gyttja
°’”
B °’” KIA- ± – -. roots
°’”
B °’” KIA- ± – -. fine roots
°’”
B °’” KIA- ± – -. peaty mud
°’”

2 Geomorphological and geological terms marked with * are explained in the appendix.
 v.m.a. heyvaert, p. verkinderen and j. walstra

Fig. . Landsat ETM+ image mosaic of Lower Khuzestan with the location of
archaeological sites, geological boreholes and areas covered by Figures – and
. The imagery was acquired in July/August  and is displayed as a
near-infrared colour composite (band combination //, converted to greyscale).

.. Archaeological Data

Only limited archaeological information about the Lower Khuzestan plain is available (Fig. );
archaeological research in Khuzestan has tended to focus on the upper part of the plain, where
the older settlements can be found.
The Dutch engineer Graadt van Roggen () surveyed the water works of Khuzestan in
preparation of an (aborted) Persian government project that aimed at restoring agricultural
wealth to the impoverished province. Most of these were located in the northern part of the
plain, the southernmost being the dam at Ahwaz.
 geoarchaeological research in lower khuzestan: state of the art 

The most extensive settlement survey in the lower plain was carried out by McCown in ,
who recorded  sites in the vicinity of Ahwaz and Hawiza. These consisted only of sites visible
from motorable roads, and the material was left unpublished for almost four decades (ultimately
published by Alizadeh ). Based on surface finds of pottery most sites were attributed to
Sasanian (c. –ad) or Islamic (after c. ad) times, and a few to the Seleucid (c. –
bc) and Parthian (c. bc – ad) periods.
Further significant information is provided by Hansman, who surveyed the region of the
Jarrahi river (Hansman a), and identified the ruins of Naisan with the ancient city of
Spasinou Charax along an abandoned course of the Karun and the Tigris (Hansman ).
Kirkby () mentioned extensive canal systems of Sasanian or Early Islamic age extending
from the same former Karun course.
The sum total of these investigations provides a rather limited and geographically biased
distribution of archaeological sites. In the course of this project, a limited survey of Lower
Khuzestan was conducted which revisited some of McCown’s sites and noted another  “new”
sites (Gasche/Paymani ). These sites were all occupied between the Seleucid and Islamic
periods, corroborating earlier findings suggesting that the plain had its heyday during the
Sasanian and Early Islamic periods and earlier sites are rare (Adams ; Alizadeh ). A
comprehensive and systematic survey of the Lower Khuzestan plain is still wanting.
The geomorphological map (Fig. ) shows the location of the archaeological sites surveyed
in the framework of this project (after Gasche/Paymani ).

.. Texts and Historical Maps

Textual information on the alluvial landscapes of Lower Khuzestan is very sparse before Islamic
times. No textual data exist prior to  bc, and between this date and the Islamic conquest,
most of the information derives from two short periods of external military expeditions into
Khuzestan.3 These pre-Islamic sources have been studied numerous times, most recently by
Cole and Gasche (). Limited additional information, from the rd century ad onwards, is
furnished by the records of the Oriental Syrian churches (Fiey , ) and a number of
Sasanian seals (Gyselen , ).
From the th century ad onwards, these sources are overshadowed by a large corpus of
Arabic texts (more than  works have been used in our research) from different genres that
provide information about the landscape of Khuzestan, among others:
– “Road books”: geographical works that focus on routes and imperial geography
– works of mathematical geography, continuing the Hellenistic tradition of Ptolemy
– “marvel literature” that focus on wonders, from strange creatures like unicorns and
elephants to volcanoes
– geographical dictionaries commenting on place names mentioned in the most famous
Arabic literary works
– travellers who describe their journeys in varying detail; often first-hand information
– historiographical works; wars and rebellions prove to be especially fecund grounds for the
survival of geographical information

3 The regular military campaigns by the Neo-Assyrian empire against Chaldean and Aramean tribes in Southern
Iraq and Khuzestan (–bc), as known from Assyrian inscriptions, and the activities of Alexander the Great
and the war between his successors, Eumenes and Antigonus (–bc), as documented in the works of
Diodorus Siculus, Quintus Curtius Rufius, Plutarch and Arrian. For a full survey of sources see the appendix in
Cole/Gasche .
 v.m.a. heyvaert, p. verkinderen and j. walstra

– prosopographical works, giving details about generations of scholars, poets, etc., including
the places where they were born, lived and died, often with dates that can help us prove
the existence of a certain place at a given time
– juristic works that contain references to places, watercourses and practices in the first
century after the Islamic conquest, a period underrepresented in other sources
– collections of poems and anecdotes, and other genres
A number of these sources have been used in the past to reconstruct the lands of the eastern
Caliphates (Le Strange ) and Iran (Schwarz ), but these reconstructions suffer from a
number of shortcomings, which render them all but useless for scientific purposes. The most
important problem is the fact that they are very superficial and do not take into account the
shortcomings of these sources:
– uneven distribution of the sources over time: a few of the earliest works date from the th
century, but the bulk of our knowledge comes from th-century works. After the th
century, very few important sources were found. Information from these sources can be
extrapolated to earlier and later times under some conditions. This is made more difficult
by the
– authority-based structure of early Islamic science: information from trusted sources is
quoted time and again for centuries, regardless of the question whether the information
was still valid or not. This problem becomes more and more pronounced in the later
centuries, because layer after layer is added to the accumulation of “knowledge”. This often
gives rise to
– contradictory reports, both inside one work and between contemporary works: these con-
tradictions can point to an evolution in the landscape, or to misinterpretations in the
information chain; sometimes, however, apparently contradictory reports can be recon-
ciled and proven to be complementary rather than mutually exclusive;
– lack of documentary sources: the descriptive nature of the Arabic sources is not only a
blessing, but also a curse: in contrast to other periods in the history of Khuzestan, and
other areas in the world, almost no real-life written documents from early Islamic Iraq
have survived. We only have information that was filtered through the mind of a medieval
author, with all the restrictions this entails.
– selectivity: the corpus is largely urban-centred, and contains little information about rural
areas
A number of European travellers and explorers visited the wider region from the th century
onwards, but it is only in the th century that these provide useful information about Khuzes-
tan, with the rise of interest in Persia of European imperialist powers, especially Great Britain
(Ooghe ; Verkinderen ).
A variety of historical maps were used in this study. The oldest are the th-century regional
maps of the so-called Islam Atlas (al-Istakhri c. ; Ibn Hawqal c. ; al-Muqaddasi c. ),
which are extremely schematic, and cannot be interpreted without reference to the accompa-
nying text. A second set of maps was made by the th-century geographer al-Idrisi (c. )
for the Norman kings of Sicily and these are equally schematic (for reproductions of all these
maps, cf. Miller ). More detailed are the maps used by European ships on the way to India,
although depictions of the study area did not become very realistic until the early th century,
even as these maps only show the coastal strip (reproductions and commentary: Sahab et al.
; Couto et al. ). The first reliable European maps of mainland Khuzestan appear in the
th century. A British expedition surveyed the Euphrates, Tigris and Karun rivers between
 and  in order to assess their suitability as trade routes (Chesney ; Ainsworth
 geoarchaeological research in lower khuzestan: state of the art 

Fig. . Geomorphological map of the study area. The chronological order of the
alluvial units is attributed according to their final stage(s) of activity. Also
shown are the locations of archaeological sites and avulsions. The labels
of the main river belts and fans refer to the units described in the text.

) and produced a detailed map of these rivers. The border area between Iraq and Iran (then
the Ottoman and Persian empires) was the subject of two major international survey campaigns
(in – and –) in an attempt to solve border conflicts (Ryder ).4

4 For a full overview of the textual sources and historical maps of the area used in this research, see the

bibliography in Verkinderen .

 v.m.a. heyvaert, p. verkinderen and j. walstra

. Results

.. The Geomorphology

... General Relief
The Lower Khuzestan plain is extremely flat, in contrast to the rugged terrain of the adjacent
anticlines. Elevation of the plain ranges from sea level to + m along its northeastern limits,
over a distance of –km. The gentle relief is mainly the result of spatial variation in
river sedimentation, with alluvial fans and ridges forming the higher grounds and floodplains
representing the lower areas. All major rivers draining into the plain, except for the Kupal,
have developed distinct alluvial ridges. The large brackish-freshwater marshes are positioned
in the lowest parts of the plain, representing flood basins enclosed by alluvial ridges and coastal
sabkhas. The main landforms are further described below.

... Structural Landforms

In the northeast the plain is bordered by series of hogbacks* formed as a result of steep dipping
anticlines (°–° according to the geological map sheets). The ridges are dissected by erosion
gullies and valleys of mostly ephemeral streams, while their piedmont consists of a more or less
continuous alluvial apron or bajadas*.

... Aeolian Landforms

Along the entire anticlinal front sand dunes occur. They are mostly limited to small, local dune
fields, except for northwest of Ahwaz, where some very large dune complexes are present.
From field evidence they are known to consist of silty fine sand (Baeteman et al. /;
Heyvaert ). The material either derived from the nearby alluvial plain, or was blown in
by dust storms from the Arabian deserts. It is yet unknown whether the dunes formed under
current environmental conditions or represent relics from past times (e.g. Pleistocene), but the
dune field northwest of Ahwaz is already mentioned in a th-century geographical work (al-
Muqaddasi: ).

Table : Morphological characteristics of the alluvial fans in Lower Khuzestan.

Alluvial fan River type Fan area (km) Average gradient

Jarrahi J Meandering >, .
Jarrahi J Meandering/distributary >, .
Jarrahi J Distributary  .–.
Karkheh Anabranching , .
Karun Meandering , .
Kupal Sheetflow, meandering in past , .
Zohreh Meandering , .

... Alluvial Landforms

The Lower Khuzestan plain essentially consists of a series of large alluvial fans* (Baeteman
et al. /; Heyvaert ; Heyvaert/Weerts ; Table ); based on the large size
and low gradient they may be classified as megafans* (cf. DeCelles/Cavazza ; Leier et
al. ). The fans are characterised by meandering river belts that episodically shift across
the fan surface (a process known as avulsion*), thereby creating distinct, diverging alluvial
ridges that correspond to different evolutionary stages of the fan. Meandering river belts exhibit
 geoarchaeological research in lower khuzestan: state of the art 

Fig. . SPOT (a) and CORONA (b) image showing characteristic elements of a
meandering river, belonging to the K and K channel belts respectively: scroll-bars (),
an oxbow lake () and crevasse channels (); in addition, nearby palaeochannel
K are patterns of ancient irrigation canals visible. The SPOT image is
a near-infrared colour composite, acquired on  June  (scene –/,
© CNES , Distribution Spot Image S.A., France, all rights reserved), while the
CORONA image dates from  February  (frame DS–DA).
 v.m.a. heyvaert, p. verkinderen and j. walstra

some typical geomorphological elements, resulting from spatial variations in flow velocity and
deposition, e.g. scrollbars*, meander cut-offs*, levees*, crevasse splays* and floodplains* (see
Fig. ).
The central Lower Khuzestan plain is occupied by the Karun megafan with its apex at Ahwaz,
where the river enters the plain, and its toe  km downstream at the confluence with the
Shatt al-Arab. Besides the present-day Karun (K) two palaeochannel belts (K and K) were
identified and mapped (Baeteman et al. /; Heyvaert ; Heyvaert/Weerts ; Walstra
et al. b):
– Palaeochannel belt K is located in the south-central part of the plain and splits into
two branches. It is unclear whether both K branches were active simultaneously or
one after the other. As the traces of K are less distinct than K, it is assumed to be
older, but this could also be the effect of soil degradation due to frequent flooding and
salinization.
– Palaeochannel belt K crosses the plain in west-south-western direction from the city of
Ahwaz to its confluence with a former Tigris/Shatt al-Arab channel, nearby the archae-
ological site of Spasinou Charax. Its upper section is obscured by the urban sprawl of
Ahwaz, but further downstream scrollbars and crevasse splays are clearly visible and sug-
gest that the river was subject to dynamics similar to the present-day river (cf. Kirkby
).
– The upper and middle sections of the present channel (K) display a dynamic morphology
with winding meanders, abundant scrollbars and meander cut-offs (Fig. ). Large crevasse
splays occur in the middle section (and only there); their absence in the upper section may
be related to the slight entrenchment of the channel. The lower section of the river consists
of relatively straight segments and eventually bifurcates into two branches: the main chan-
nel, Shatt al-Haffar (Kb), discharges into the Shatt al-Arab near Khorramshahr, while
the Shatt Bamishir (Kc) enters the sea independently. The “Blind Karun” (palaeochannel
Ka) also branches off in south-eastern direction, parallel to the Shatt Bamishir; this chan-
nel lines up remarkably with one of the K branches on the other side of the present-day
river channel.
It should be noted that more palaeochannel belts may be present in the subsurface, which were
not detected from the satellite imagery. For example south of Ahwaz, along the Nahr Bahre,
some individual meander traces were noted, largely covered by irrigation patterns and there-
fore without any context, which may belong to a third Karun palaeochannel (K).
The northwestern part of the plain is dominated by the Karkheh river, entering the plain near
Hamidiya. Three main channel belts (Kh, Kh and Kh) were distinguished, two of which are
currently active (Baeteman et al. /; Heyvaert ; Heyvaert et al. ):
– Traces of palaeochannel Kh are clearly visible before it merges with the K channel belt,
indicating that in the past the Karkheh flowed southwards and was a tributary of the Karun
(cf. Kirkby ).
– Another channel belt, known since its abandonment in the s as the “Blind Karkheh”
(Karkheh Kur, Kh), represents a previously abandoned river course that was recently
reactivated through the construction of a bypass canal near Hamidiya. Upstream it follows
a meandering course (Kha—lined up with the earlier Kh), but after a sharp turn it
continues along a rather straight line (Khb) in north-western direction towards Hawiza.
Traces of two abandoned channels diverting from the Khb are noteworthy: the Khc
and Khd.
 geoarchaeological research in lower khuzestan: state of the art 

Fig. . Comparison of satellite imagery, revealing the reactivation of irrigation

canals () along the Karkheh Kur/Khb channel () upstream of Hawiza (H). The
image on the left shows a mosaic of CORONA scenes (mission DS–,
acquired on  September ); the image on the right is a near-infrared colour
composite Landsat ETM+ scene (/, acquired on  August ).

– The present main channel of the Karkheh (Kha, b) turns in north-western direction
shortly after entering the plain and follows a course parallel with Khb and the anticlinal
front, towards Bostan. Many channels/canals (both active and abandoned) branch off from
the main channel, most notably the Khc at Susangerd. The latter seems to have reoccupied
the Khd channel.
Further downstream many channels/canals branch off from the two main streams Kh and
Kh, eventually discharging into the Hawiza Marshes.
The eastern part of the plain is dominated by three alluvial fans that were successively
deposited by the Jarrahi river (J, J and J). Distinction between the fans is primarily based
on the layout of the river channel and irrigation patterns (Baeteman et al. /; Walstra
et al. a):
– The first and largest fan (J) stretches from the location where the Jarrahi enters the plain
to the tidal flats of Khor Musa. A southwards orientated alluvial ridge can be linked to
traces of a palaeochannel that continues into the tidal flats. Towards the northwest and
southeast the fan merges into the adjoining fan surfaces of the Kupal and Zohreh rivers. In
the upper fan section the present river has cut a . km wide valley, up to  m deep into the
fan surface. Within this valley the Jarrahi flows through a highly dynamic anastomosing
river bed, with several meander cut-offs alongside the main channel.
– The second fan (J) is located immediately downstream of the first one. Here the river
has emerged from its entrenchment and continues over an elevated alluvial ridge, raised
–m above the fan’s surface.
– The third and presently active fan (J) displays a typical distributary channel system, with
many bifurcating outlets branching off from the main channel. These branches ultimately
 v.m.a. heyvaert, p. verkinderen and j. walstra

split up into small ditches, separating extremely narrow elongated fields, before draining
into the Shadegan Marshes. Several crevasse splays have been identified alongside the
Jarrahi, in particular near the apex of fan J; some of them have been transformed into
irrigation networks.
– To the west of fan J a meandering palaeochannel (Jx) crosses the Shadegan Marshes. Its
levees are clearly raised above the water, but at both ends the channel is covered by recent
alluvial deposits of the Jarrahi and Karun rivers.
The single meandering course of the Zohreh river is pushed to the very eastern limits of the
plain, along the foot of the Rag-e Safid anticline. Besides the Karun it is the only river that
reaches the Persian Gulf.
The Kupal river has created a modest alluvial fan (Kp) at its entrance to the plain between
the Ahwaz and Marun anticlines. On the plain itself the river does not have a distinct channel,
although some traces of palaeochannels were detected. During wet seasons the Kupal drains
superficially via a broad marshy zone into the Shadegan Marshes. A similar marshy zone is
present in the area between the Jarrahi and Zohreh fans, fed by ephemeral streams issuing from
the Agha Jari anticline.
Towards the southwest the plain is bordered by the Shatt al-Arab estuary, which receives the
bulk of sediment from the Karun river (Lees/Falcon ). In addition to the active channels
of the Shatt al-Arab and the Shatt Bamishir, several palaeochannels are located further to the
east, most notably the Blind Karun (Ka). The lower parts of these outlets now act as intertidal
channels.

Throughout the plain, extensive patterns of relict irrigation systems were mapped, superim-
posed on the alluvial fans (Walstra et al. a–b; Heyvaert et al. ):
– The most impressive network consists of diverging canals on both sides of the present
Karun channel, south of Ahwaz. The canals radiate from two huge feeder canals, and
extend up to km across the plain before ending in distinct “herringbone” field patterns
(Fig. ). The traces of this network clearly intersect the K and Kh palaeomeanders (in
contrast to Kirkby’s interpretation who rather described them as branching off from the
meanders), and completely obscure the upstream part of K. A somewhat different layout
of parallel canals is branching off at sharp angles from the eastern feeder canal (named
Nahr Bahre), apparently overlying “herringbone” patterns.
– A similar, radially diverging network of canals with “herringbone” patterns covers Jarrahi
fan J.
– Dense networks of relict irrigation canals were also identified across fan J, diverging from
its apex on both sides of the incised river valley.
– The Karkheh Khb channel belt is characterised by irrigation canals orientated perpendic-
ular to the main channel and typically extending over a distance of – km. This system
was previously abandoned, but has been reactivated in recent times due to the construc-
tion of a bypass near Hamidiya).
– A broad zone with regular patterns of ridges at – m intervals flanks the Shatt al-
Arab. Concordant field patterns are present further inland, although less distinct, maybe
due to frequent flooding and soil salinization.
 geoarchaeological research in lower khuzestan: state of the art 

Fig. . CORONA images displaying different patterns of ancient to modern irrigation

systems: (a) “herringbone” field patterns south of channel belt K, (b) large parallel canals
branching off from the Nahr Bahre (NB), (c) traces of canals on both sides of the incised
Jarrahi valley (J) downstream of Ja Nishin (JN), and (d) the present-day distributary channel
system of the Jarrahi. Main feeder canals are indicated by () and field patterns by ().
(CORONA scenes: DS–DA, acquired on  February  (a and b),
DS–DF (c) and DS–DF (d), both acquired on  February ).

... Coastal Landforms

The tide-dominated coastline is fringed by a broad zone of tidal flats*, salt marshes and coastal
sabkhas* with saltpans. The latter are inundated only by extreme high water and can be classified
as clastic coastal sabkhas (Heyvaert/Baeteman ). The Khor Musa tidal embayment is partly
protected from the open sea by spits* and barrier islands*.
 v.m.a. heyvaert, p. verkinderen and j. walstra

.. General Palaeogeography

Considering the vast area of the Lower Khuzestan plain and the limited number of geological
boreholes (Fig. ) and radiocarbon-dates (Table ), any attempt to describe the palaeogeograph-
ical evolution of the plain is constrained by the assumptions that are used to interpolate between
dated samples through both time and space.
The infill of a drowned palaeovalley is controlled by many interacting factors, including rate
of relative sea-level (RSL) rise, sediment budget, morphology of the pre-transgressive surface*,
accommodation space*, neotectonic setting and sediment compaction. During the infill of a
palaeovalley, initially caused by RSL rise, the relative importance of these individual factors
changes in the course of time (Baeteman ; Beets/van der Spek ). Knowledge of these
factors is very limited for the Lower Khuzestan plain, but a more detailed discussion is provided
by Heyvaert and Baeteman ().
geoarchaeological research in lower khuzestan: state of the art 
 v.m.a. heyvaert, p. verkinderen and j. walstra
geoarchaeological research in lower khuzestan: state of the art 

Fig. a–g. Reconstruction of the environmental setting of the Lower

Khuzestan plain from  cal bp to present (after Heyvaert/Baeteman
; adjusted in the light of new results, cf. case studies below).
 v.m.a. heyvaert, p. verkinderen and j. walstra

A series of palaeogeographical maps, published earlier by Heyvaert and Baeteman ()

and Heyvaert (), was further refined for the present paper on the basis of the integration
of archaeological and historical data with geological data. The different time slices are chosen
according to the available radiocarbon dates and to some major changes in the development of
the plain.
During the early and middle Holocene, the Lower Khuzestan plain was a tidal embayment
under estuarine conditions. In the early Holocene, the sea invaded the deepest parts of the
antecedent valley of the Shatt-el Arab which changed into a tidal embayment. Our analysis of
the geological data demonstrates a landward extension of the Gulf until at least  km north of
its present-day shoreline at about  cal bp (c.  bc; Fig. a). Due to a high rate of RSL rise,
the area was flooded rapidly and the tidal environments shifted landwards. Certainly until 
cal bp (c. bc), the low-lying areas were essentially wetlands on account of the proximity of
the water table to the surface and/or their susceptibility to inundation by estuarine or riverine
flooding and little evaporation. Salt marshes with reed growth developed, however, not for a
time sufficiently long for peat to accumulate. This, together with the indication of various high
intertidal silting-up phases with vegetation growth, indicate a high supply of coastal sediment
and a rapid RSL rise. Deceleration of the rate of RSL rise after approximately  cal bp
(c.  bc), together with probably more arid conditions, allowed coastal sabkhas to extend
widely and to aggrade while the position of the coastline remained relatively stable. Instead
of salt marshes, the coastal sabkhas developed directly landwards of the high intertidal flat
(Fig. b).
Continued deceleration of the rate of RSL rise initiated the progradation of the coastline from
c.  cal bp (bc; Fig. c). The effect of sediment supply by the rivers became more impor-
tant than the effect of the RSL rise and major parts of the sabkhas were gradually replaced by a
floodplain. An avulsion-controlled Karun megafan developed under a decelerating rate of sea-
level rise, controlling the shifting of the Karkheh and Jarrahi channels and their loci of sediment
input. By approximately  cal bp ( bc), palaeochannels belt of the rivers Karun/Karkheh
and Jarrahi started to fill the tidal environment in the southern part of the central plain caus-
ing progradation of the coastline. However, the tidal environment still continued to expand in
the northern part of the plain, which hitherto had been out of the reach of marine influence.
Between  cal bp and  cal bp ( bc –  ad), the Karun shifted north (Fig. d) and
started to reduce the initial width of the tidal embayment directly to the south of Qurna (Iraq).
In the northern part of the plain, the tidal environment became gradually isolated and changed
into a brackish-freshwater marsh environment by  cal bp ( ad). In the period between
– cal bp (–ad; Fig. e) the Karun avulsed to (or at least nearby) its present-day
position, and started to control the progradation of the coastal area in the southern part of the
plain. In the northern part of the plain, the brackish-freshwater marshes became locally filled
up by the newly developed Karkheh palaeochannel nearby the village of Hawiza. By  cal bp
(ad; Fig. f.), in the southern part of the plain, a brackish-freshwater marsh was present in
the surroundings of the Jarrahi distributary system. The last avulsion of the Karkheh (Fig. g)
happened in  ad, as documented in historical texts (see case study Karkheh).
The palaeogeographical reconstruction of the plain presented above, represents a significant
step forward with respect to earlier reconstructions. However, it only constitutes a broad scheme
and provides a framework for further research. A more detailed discussion of the palaeogeo-
graphical reconstruction of Lower Khuzestan can be found in Heyvaert and Baeteman ()
and Heyvaert ().
The reconstruction of the shifting of the palaeochannelbelts of the river Karun, Karkheh and
Jarrahi and the development of an age model for the channel belts is based on the integration
of multidisciplinary datasets; the case-studies below will illustrate this.
 geoarchaeological research in lower khuzestan: state of the art 

. Human-Environmental Interaction: Case-Studies

.. The Karkheh River System

The objective of this case-study is to investigate the role of human interference in the late-
Holocene floodplain history of the Karkheh river, located in the northern part of the Lower
Khuzestan plain. Below, the main data and results are summarized. A more detailed discussion
can be found in Heyvaert et al. ().

... Geomorphological Data

As already mentioned above (cf. section geomorphological map, Fig. ), three main Karkheh
channel belts (Kh, Kh and Kh) and several diverting branches (e.g. Khd, Khc) were
distinguished. These different courses can be attributed to successive stages in the development
of the Karkheh floodplain.

... Archaeological Data

The survey by Gasche and Paymani () noted  sites in the area of the Lower Karkheh,
 of which could be dated. Seven of these were founded only in the Early or Middle Islamic
periods (c. –ad).
The oldest archeological site discovered in the area is located in the eastern part of the town
Hamidiya (Gasche/Paymani : , no. ), at the place where the river breaks through
the anticline and enters the Lower Khuzestan plain; the site was inhabited from Achaemenid
(–bc) to Islamic times (c. ad-).
No sites firmly associated with the Kh palaeochannel have been found; one site (Gasche/
Paymani’s no. ) is located very close to both the Khb and Kh channels, and could have been
linked with either one of them. Materials dating to the Seleucid/Parthian (c.  bc – ad)
and Islamic periods (after c. ad) have been found here.
Two more pre-Sasanian sites (Gasche/Paymani’s nos.  and ) were found at some distance
from the Kha and Khb channels, but most of the sites along the Khb-d channels date from
islamic times. The most important of these are the ruins of the city Hawiza (Tell Hawiza), where
Seljuq and Mongol pottery was found (c. –ad). No archaeological sites were found
associated with the Kh channel.

... Historical Data

The area of the lower Karkheh was a backwater and is rarely mentioned in historical sources. In
Arabic sources, a town called Nahr Tira is mentioned in this area; the town was located along
a river or canal of the same name, and one of the  kuwar (districts) of Khuzestan was named
after it. The Nahr Tira channel/canal was most probably part of the Karkheh Kh system, and
is last mentioned as an active watercourse in a description of a battle that took place in ad
(Verkinderen : –). Ninth- and tenth-century itineraries suggest that the place must
have been located in the area of Hamidiya; Gasche/Paymani’s site no.  is perhaps to be
identified as Nahr Tira (Gasche/Paymani : ; Verkinderen : ).
From the mid-th century onwards, the area became the center of a powerful tribe, the
Banu Dubais. Their capital Hawiza (Fig. ) was located on a land route from Shiraz to Baghdad
and rose to prominence by the middle of the th century (Isfahani: ; Ibn Battuta: II ).
Cash crops (corn, cotton and especially sugarcane) were cultivated around the town in the
th century (Mustawfi: ), which indicates abundant water supply was available at the time.
The importance of the city increased further in the following period: it became the capital
 v.m.a. heyvaert, p. verkinderen and j. walstra

Fig. . Historical map showing the position of the Nahr Hashem irrigation
canal upstream along the ‘old channel’ of the Karkheh (after Loftus a).

of the province of Arabistan (Khuzestan) and the seat of its great dynasty of governors, the
Musha#sha#a (Layard ).
Few European travellers visited the region, and only from the nineteenth century onwards
(Layard ; Loftus a–b). Loftus describes a major landscape change in the surroundings
of the Arab city of Hawiza. Prior to  the river Karkheh had flowed along Hawiza (i.e. in
its Khb channel), and the region had been intersected by irrigation canals connected to the
Karkheh. One canal, locally called Nahr Hashem, was dug some  miles north of Hawiza
(Fig. ). Because the lands irrigated by the Nahr Hashem canal lay topographically lower
than expected, the canal gained importance and started to carry off exceeding amounts of
water from the river. As a consequence, a dam was constructed at the bifurcation point to
prevent the Karkheh of abandoning its original course along Hawiza. This dam was damaged
in a flood event, and a new, stronger dam was built. Finally, in , this dam was washed
away, and during a single night the entire river changed its course, leaving its original bed to
flow into the Nahr Hashem canal. The area irrigated by the Hawiza channel became largely
abandoned, as was the city of Hawiza itself. Efforts were made to rectify this situation and a
new canal, called the Mechriya, was dug above and opposite to the Nahr Hashem, but had
little effect.

... Geological Data

The geological data comprise of the facies analyses of  hand-operated boreholes and one
outcrop, forming two transects along the Karkheh palaeochannel belts.
A first transect is located along the Karkheh palaeochannel belt (Khd) which at present
time is fed by the Karkheh Khc channel (Fig. ). The sedimentary succession along this
transect consists mainly of fluvial deposits with a thickness of c.  to  m. These fluvial deposits
are interpreted as channel belt/crevasse splay deposits or floodbasin deposits associated with
the Karkheh (palaeo-) channel belt Khd/c. Only in boreholes B and B the fluvial
deposits (channel belt/ crevasse splay deposits and floodbasin deposits) were found overlying
 geoarchaeological research in lower khuzestan: state of the art 

Fig. . Stratigraphic profile with indication of the depositional environments of

cores along (palaeo)channel belt Khd/Khc. Sediment textures are based on field
descriptions. (after Heyvaert et al. ). The location of boreholes is given in Fig. .

a . to m thick layer of organic rich brackish-freshwater marsh deposits, covering coastal
deposits. In borehole B, organic material at the base of the marsh deposits was dated
at – cal bp. This date indicates that a brackish-freshwater marsh, which can be
attributed to a former extension of the present-day Hawiza Marshes, existed from c. –
ad onwards, and that the sedimentation associated with (palaeo-) channel Khd/c
started later.
A second transect (Fig. ) follows an east-west direction parallel with and south of the lower
part of the currently active Karkheh channel belt Khb. The boreholes in this transect show a
similar sedimentary succession to the one found along the first transect.
Coastal deposits, covering the pre-transgressive surface are gradually overlain by brackish-
freshwater marsh deposits and fluvial deposits. Only in core B and outcrop B datable
organic material was encountered in the brackish-freshwater marsh deposits.
In borehole B, two peaty horizons at +. m and +. m were dated at – cal
bp (–ad) and – cal bp (–ad), respectively (Table ). The reworked
organic material at the level of +.m in B was dated at – cal bp (–ad). In
outcrop B, a peaty horizon was found in the brackish-freshwater marsh deposits on a level
of +.m. The base and the top of the peaty horizon were dated at – cal bp (–
ad) and – cal bp (–ad), respectively. It is suggested that the brackish-
freshwater marsh deposits, which underlie the fluvial deposits of the present-day Karkheh
river system at a level of +. to m can also be linked to a former eastern extension of the
Hawiza Marshes. The onset of the formation of the marshes at this location can be estimated at
 v.m.a. heyvaert, p. verkinderen and j. walstra

Fig. . Stratigraphic profile with indication of the depositional environments of cores
located along the downstream part of channel belt Khb. Sediment textures are based on
field descriptions. (After Heyvaert et al. ). The location of boreholes is given in Fig. .
 geoarchaeological research in lower khuzestan: state of the art 

c. – cal bp (–ad). The date of the peaty layer in B, covered by the fluvial
deposits, suggests that sedimentation by the river (Kh), filling in theses marshes, started very
recently at this location, at the earliest at –ad.

... Correlating Datasets—Karkheh Floodplain History

Based on the integration of geomorphological, geological, archaeological and historical datasets
the floodplain history of the Karkheh can be reconstructed.
The Karkheh was once a tributary of the Karun, as is clearly indicated by the joining of
palaeochannel belts Kh and K west of Ahwaz. The K channel was active from at least the
nd century bc onwards (see Karun case-study below, and Heyvaert ; Walstra et al. b).
Moreover, just like the K, the Kh palaeochannel belt is intersected by irrigation systems
that most probably date from Sasanian/Early Islamic times (Walstra et al. b). The latter
gives an indicaton for an end-date of activity of palaeochannel belt Kh. It is possible that
this Kh channel is to be identified with Nahr Tira, which was abandoned by the time that
the first Arabic geographical works were written in the th century. Along the first transect
(Fig. ), located along the palaeochannel belt Khd/c, brackish-freshwater marsh deposits
dated at – cal bp were found covered by fluvial deposits. The latter sets a maximum
age of –ad for the formation of palaeochannel belt Khd, which branches off from the
palaeochannel belt Khb. Indirectly, this indicates that the abandonment of the predecessor of
Kh, i.e. palaeochannel belt Kh, most likely did not occur long before –ad. This is in
agreement with the distribution of archaeological sites, which suggests that the palaeochannel
belt Khb was active at least from the Early Islamic period to Middle Islamic period (c. –
ad).
Historical sources indicate that the area of Hawiza was flourishing in the th and th
centuries (Mustawfi: ), and inform us that the palaeochannel belt Khb, and extensive
irrigation networks connected to it, were fully active until  ad. In that year, the channel
(Khb) shifted to its present-day position (Khb) (Layard ; Loftus , b), leaving
all of the irrigation canals connected to the Khb channel dry. The historical date of the
avulsion event fits well with the geological data (second transect, Fig. ), which show that
near Bostan, sedimentation by palaeochannel belt Khb started after ad (borehole B).
Palaeochannel belts Khc and Khd, which branch off from Khb, must have been abandoned
due to the same event in  ad as well. Khd was later reoccupied by a channel (Khc)
branching off from the Khb channel downstream of Susangerd. The Khb palaeochannel belt
and its irrigation networks were reactivated by the late s, after the construction of a canal
bypassing Hamidiya (progress of construction works is clearly visible on CORONA images
from ).

... Human Impact

The Karkheh avulsion sites  and , indicated on the geomorphological map (Fig. ), are human-
induced avulsion sites and represent the loci of shifts from palaeochannel Kh to a more
northern position Khb, and from palaeochannel belt Khb to position Khb, respectively. The
straight course of the palaeochannel Khb and the absence of palaeomeanders along its course
indicate that most probably the channel was dug for irrigation purposes. Moreover, a network of
irrigation canals perpendicular to the channel Khb was identified (Fig. ), suggesting human
interference. Our historical data (Layard ; Loftus , b) firmly attribute the second
avulsion event (Khb to Khb—avulsion site ) in  to human interference. For a broader
discussion on human-interference on the Karkheh floodplain evolution, we refer to Heyvaert
et al. .
 v.m.a. heyvaert, p. verkinderen and j. walstra

.. The Karun River System

The aim of this case-study is to reconstruct the evolution of the Karun alluvial megafan. As
shown on the geomorphological map, three Karun palaeochannel belts were detected: K, K,
and K, of which the latter one is in use today. Older channel belts may be present in the plain
but are not visible on satellite imagery. Complementary evidence from texts and geological and
archaeological field data allowed the elaboration of a chronology of the channel belts and canal
systems.

... The Ahwaz Dam and Irrigation System

At its entrance to the Lower Khuzestan plain in Ahwaz, the river breaks through the anticline,
reducing the high ridge to a number of rocky shoals in the river bed. On these rocks, traces of an
ancient dam have been found (Graadt van Roggen ). Based on Graadt van Roggen’s sketch
map of the dam area, huge feeder canals were identified on aerial photographs and satellite
imagery [from the s and s, before the area was absorbed by the modern city], and
linked to the vast irrigation network extending on both sides of the present-day Karun. Since
the radially diverging canals clearly cross the palaeochannels, they must postdate channel belts
K and Kh. The “herringbone” field patterns also cover the upper part of palaeochannel belt
K.
Most of the canals belonging to the system east of the K channel do not exhibit the same
typical herringbone field patterns observed to the west, which might indicate their lifetimes
did not completely overlap. Likewise, one of the east bank canals is cut by the K channel.
It therefore seems possible that only part of the canal system predates the present-day channel
belt, while part of it remained/became active during the early days of K. Both networks appear
to have been in use for a long time, since numerous bypasses that point to restorations have
been identified on satellite images (those on the east bank of the Karun were already noted by
Gasche/Paymani : –).
The exact age of the ruined dam has not been established. Traditionally, irrigation works of
this scale are attributed to a major colonization program in Sasanian times (Christensen ),
but there is no material proof for this date. Although archaeological surveys indeed suggest that
the settlement in the plain had its heyday during the Sasanian and Early Islamic periods, the
canals to the east of K are in fact associated with sites from Seleucid/Parthian to Islamic times
(Gasche/Paymani : –), suggesting that at least a precursor of the irrigation system
existed before.
At the end of the th century, the travellers al-Muqaddasi () and Ibn al-Muhalhil ()
still refer to the dam in Ahwaz and at least  important irrigation canals feeding from it: “the
dam holds back the water (of the Karun), and divides it into three streams, which flow to their
agricultural estates and irrigate their fields. They say that without the dam, Ahwaz would not be
inhabited, and its rivers would be of no use”, al-Muqaddasi () writes. A careful comparison
of the th-century descriptions by al-Muqaddasi and Ibn al-Muhalhil with the archaeological
observations of Graadt van Roggen near the dam of Ahwaz proves that at least the east bank
feeder canal was still in use in the th century (Verkinderen : –). For the west
bank canals, the texts are more ambiguous, but the references to major canals and investments
in irrigation in this area at least until the th century suggest that the canal system was active
in the Early Islamic period (Verkinderen ). It is unknown when exactly these irrigation
systems fell into disuse.
 geoarchaeological research in lower khuzestan: state of the art 

... Channel Belts and Embouchure

No archaeological sites or other direct age indicators associated with the K palaeochannel belt
have been surveyed (although some potential sites have been found on topographical maps and
satellite imagery), and no geological corings along this river bed were carried out to this day.
The K system obviously predates at least the latest phase of the large (Sasanian/Early Islamic)
irrigation systems, as its upper section is covered by the “herringbone” patterns. The channel
belt is probably also older than K, although this is less sure.
The founding and abandonment of the ancient city of Spasinou Charax (Naisan) provides
strong evidence for dating channel belt K: Pliny (c.  ad) writes that the city was located at
the confluence of the Tigris and the Karun, which implies the K channel belt was active from
at least the nd century bc (at which time the area was ruled by the Spasines after whom the
city was named), and perhaps even earlier, if Pliny’s attribution of an earlier founding of the
city to Alexander the Great is correct (Hansman ). The city was probably abandoned by
the early th century ad, as its Arabic name Karkh Maisan figures in the accounts of the Muslim
conquests only as a kura (district), not as a city anymore (Schuol ; Verkinderen ). Also,
the channel belt is intersected by canals of the large (Sasanian/Early Islamic) irrigation system,
thereby providing a terminus ante quem.
McCown’s survey identified a number of archaeological sites from the Seleucid/Parthian
period close to the present-day Karun channel (K) in the vicinity of the city Ahwaz (Alizadeh
). Unfortunately, the recent survey by Gasche and Paymani was not able to localize these
sites more accurately (Gasche and Paymani : ), and therefore they cannot be firmly
associated to either the K channel belt, to one of the large canals that derive from the dam
at Ahwaz, or to the K channel belt, all of which pass through the same corridor in this area.
The only archaeological sites securely associated with the K bed are  imamzadehs (local saints’
tombs) on the lower part of the K channel that were described by th century travellers, but
are lost today (Gasche/Paymani : –). Gasche and Paymani suggest a date between
the th and th century for these shrines, based on their characteristic “sugarbread” or
“pinecone” domes, but this date is far from certain, since domes of this type were constructed on
new imamzadehs in Khuzestan at least until the s (Unvala : ). There is little textual
evidence about the upper course of K; only in the th century is there positive evidence that
the Karun flowed in its K bed. The geological information gathered about the K channel is
also not conclusive.
The earliest evidence for the Ka (Blind Karun) can be attributed to th-century Arabic
sources, which depict the Karun entering the sea independently near Abadan (i.e. instead of
discharging into the Shatt al-Arab). The distances cited by the Arabic geographers point to the
Ka as the most probable early Islamic Karun outlet (Verkinderen : –; Walstra et al.
b). A number of canals between the Ka and the Tigris outlet were constructed in the th
and th centuries, and one of these canals (the Haffar canal) eventually became the main outlet
of the Karun as a result of the destruction of a dam on the Karun in  ad, which led to the
abandonment of the Ka branch (Verkinderen : ; Walstra et al. b: ). The Haffar
(or a similar canal that connected the Karun to the Shatt al-Arab) appears to have functioned
as the main outlet of the Karun at least, in the th century (Mustawfi ). By the mid-th
century, the Haffar (for the first time called by this name) is described by the French traveller
De Thévenot as a narrow snaking canal, only a few meters wide, that connected the Tigris to the
main outlet of the Karun, the Ka branch at that time (de Thévenot ; Verkinderen ;
Walstra et al. b).
To conclude, we can say that the exact course of the Lower Karun before the early Parthian
period is not known. The river flowed in its K bed at least from the nd century bc onwards.
 v.m.a. heyvaert, p. verkinderen and j. walstra

At the time, the Karkheh was a tributary to the Karun, by way of its Kh channel, which joined
the Karun west of Ahwaz. Sometime before the mid-th century, the Karun shifted its bed
eastward, most likely to the K or K channel. After the eastward shift of the Karun, the area
previously occupied by the K channel was irrigated by long irrigation canals that derived from
a single feeder canal, which ultimately appears to have taken off from the Karun at the Ahwaz
dam. The huge size of this irrigation system implies that it was constructed in the heyday of
the plain, i.e. in the Sasanian or Early Islamic period. The fact that the upper part of the K
channel is covered by the herringbone patterns of the irrigation system, suggests that the K
either predates this shift away from the K (and therefore, the K channel itself) or was active
in the period between this shift and the construction of the irrigation system. Since there is
evidence of numerous bypasses and repairs to the irrigation system, it cannot be ruled out that
the K was contemporary to an early phase of the irrigation system. Only in the th century
we have conclusive proof that the Karun flowed in its K bed. More information is available
for the final stretch of the river. From the th century onwards, we have proof that the Karun
discharged into the Persian Gulf by way of (or nearby) the Blind Karun (Ka) channel. It is
not clear if the river was flowing in its K or K bed at that time, since both seem to have been
connected to the K channel. Sometime between the end of the th and the middle of the th
century, the Karun shifted its bed to discharge into the Shatt al-Arab by way of the Haffar (or
similar) channel. By the th century, it flowed into the Persian Gulf independently again, by
way of the Blind Karun (Ka) channel. In the s, a dam break caused the final shift of the
river from the Ka to the Haffar channel.

.. The Jarrahi River System

The aim of this case-study is to reconstruct the development of three alluvial fans successively
deposited by the Jarrahi river (Walstra et al. a).
Initially, the Jarrahi built up a large alluvial fan (J) at the foot of the Marun and Agha
Jari anticlines. An important archaeological site, Tell Tendy, is located along an abandoned
palaeochannel belt and provides an indication for its age of activity. The site was populated
at least between the Achaemenid (– bc) and Parthian (bc – ad) periods, but
probably had a much longer history before (Hansman a). Abandonment of the site may
be related to the avulsion of the Jarrahi to its present position and the incision (Fig. ) that
followed. An extensive irrigation network that derived water via intake canals from a dam
nearby the site of Ja Nishin can be considered as an effort to revive water supply across the
fan (unit Ja). The foundation of Ja Nishin was dated by Hansman to Hellenistic times, and the
site was inhabited until the th century (Hansman a). The dam has been attributed to
Sasanian times (Hansman a), although an earlier date cannot be ruled out.
During a later phase, which may have started together with the down-cutting in the first
fan, a second fan (J) developed downstream. The surface of this fan is characterized by
radially diverging canals with “herringbone” field patterns. No known archaeological sites are
associated with the system, but the size and layout are very similar to the patterns identified
along the Karun river (see case-study Karun), and therefore a Sasanian or Early Islamic origin
is suggested.
The last phase regards the deposition of the present-day fan (J), again further downstream
and westwards. Organic material from the marsh deposits underlying the alluvium was radio-
carbon dated at – cal bp (–ad), thereby providing a terminus post quem for
the onset of alluvial deposition (Heyvaert/Baeteman ; cf. Fig. ).
The earliest historical evidence for the J fan is the founding of Fellahiyah/Shadegan in
the s (Layard ), while archaeological remains of the previous capital town Med-
ina/Dawraq were dated to the th–th century ad (Hansman a). These data all suggest a
 geoarchaeological research in lower khuzestan: state of the art 

Fig. . Stratigraphic profile with indication of the depositional

environments of cores located in de Jarrahi fan J. Sediment textures are
based on field descriptions. The location of boreholes is given in Fig. .

very rapid fan formation, which was confirmed by observations from recent satellite imagery
(see Fig. ).
There is no direct evidence for the age of palaeochannel belt Jx, but it may be associated with
fluvial deposits underlying the dated marsh deposits underneath fan J. The palaeochannel is
intersected by the present-day Karun; unfortunately, the date of the shift of the Karun to its
present-day bed is still unknown (see case-study Karun).
Early Islamic texts do not mention the Jarrahi river itself. th- and th-century sources
make clear that the area of the present-day lower Jarrahi was very water-rich; it was easier to
travel there by boat than on horseback, one of the way stations that was located on the route that
crossed the area is described as “located in the middle of the water”, and the area was known as
 v.m.a. heyvaert, p. verkinderen and j. walstra

Fig. . Jarrahi channel incision of fan J (photo by V.M.A. Heyvaert, )

the place where the water of all of Khuzestan gathered (e.g. al-Istakhri: ; al-Mas#udi: II ;
cf. Verkinderen ). This water probably formed a large marsh area, which drained into the
Karun river and the sea. It is not clear if the Jarrahi had a distinguishable channel through these
marshes, and which could be identified with the Jx paleaochannel. The maps accompanying
the works of al-Istakhri, Ibn Hawqal and al-Muqaddasi (Miller ) do indicate a waterway
that joins the Karun at the place where we would expect the Jarrahi. On the other hand, the
same maps also depict a waterway through the marshes of lower Iraq, where no noticeable river
channel existed, and boats navigated through a series of corridors through the reeds linking
bodies of open water (Ibn Rusta: ; Ibn Sarabiyyun: ). Moreover, two canals identified
with present-day Nahr Bahre and Nahr Maleh are said to have reached the sea, which seems to
preclude a Jarrahi channel that reached the Karun (Verkinderen ).
It is interesting to note that the district capital was relocated stepwise in downstream/west-
wards direction, apparently synchronous with phases in the evolution of the Jarrahi:
– The Sasanian and Early Islamic capital, named Dawraq (Fiey , ; Gyselen ,
), was probably located at Ja Nishin, where pottery from these periods was found
(Hansman a). The same site has also been identified as the Hellenistic city Seleucia-
on-the-Hedyphon, but this is more tentative—another candidate is Tell Tendy, located
along a previous course of the same river (Hansman a; Verkinderen ).
– Early Islamic texts suggest that Dawraq had moved downstream by the th century, as
they mention a clear connection between the town, marshes and the sea, which cannot
refer to Ja Nishin (Verkinderen ).
 geoarchaeological research in lower khuzestan: state of the art 

– In the th–th century the capital was located at Medina and eventually it moved to
Fellahiyah/Shadegan, which was founded in the s (Hansman a; Layard ).

... Human Impact—River Incision

The channel incision of fan J (Fig. ) may simply be the result of an avulsion and internal
adjustment of the river to its new gradient, a common process on alluvial fans known as fan-
head entrenchment (Bridge ; Blair/McPherson ; Harvey et al. ). On the other
hand, it can also be explained by several external factors, such as tectonic activity, changes
in base-level, or changes in river discharge and sediment load (cf. Blair/McPherson ;
Jones/Schumm ). Kirkby provided evidence for a similar phase of down-cutting river
systems in Upper Khuzestan after c.  bc, and suggested this was a regional rather than local
phenomenon. Indeed, the construction of the dam at Ja Nishin can be considered as an effort
to restore water supply after it had been cut off by river incision. Remnants of similar dams,
built to raise water up to the level of intake canals, have been attested throughout Khuzestan
(Graadt van Roggen ) and are traditionally attributed to Sasanian times (although some
predecessors might actually predate this period, see case-study Karun). The requirement of such
dams across all rivers in Khuzestan strongly supports Kirkby’s case.

... Human Impact—Crevasse Splays

It is known from ancient times that natural levee breaks and their associated crevasse splays
formed the ideal loci for irrigation (Wilkinson ; Morozova ; Heyvaert/Baeteman
). The same principle is still being applied today across fan J, where levee breaks are trans-
formed into outflow points and the naturally elevated position of the alluvial ridge facilitates
gravity-flow. Maintenance of the resulting distributary channel system prevents avulsions tak-
ing place and spreads out the sediment load over a large area. As a consequence, deposition
takes place at very high rates and along extremely low gradients: the average fan progradation
rate equaled .km2 a-1 over the last  years (based on analysis of satellite data and map sources,
cf. Fig. ), while the vertical aggradation rate was estimated at – mm a-1 for the last four cen-
turies (based on geological coring B); surface gradients range from .–., which is
exceptionally low.

. Conclusion

In this paper the geographical evolution of the Lower Khuzestan plain was reconstructed
based on the integration of geological, textual, archaeological information and remote sensing
imagery.
Geological data show that during the early and middle Holocene the Lower Khuzestan plain
was a low-energy tidal embayment under estuarine conditions. In the Early Holocene, a high
rate of relative sea level rise caused the land to be flooded by sea water, and the coastline moved
up to km further landward than the present-day shoreline, reaching its peak about bc.
A large part of western Lower Khuzestan was covered with supra- and intertidal flats and salt
marshes, where coastal sediments were deposited at a high pace, resulting in aggradation of the
plain. Due to a decreasing rate of relative sea level rise and perhaps more arid conditions after
bc, coastal sabkhas developed along the now more stable coastline.
From c. bc onwards, sediments supplied by the rivers became more important than
the effect of the relative sea level rise, resulting in the progradation of the plain and reducing
the extent of the sabkhas. The Late Holocene progradation of the plain was controlled by the
 v.m.a. heyvaert, p. verkinderen and j. walstra

Fig. . Progradation of the present-day Jarrahi fan (J) between  and , based on a
sequence of satellite images and maps. The dashed line represents Jarrahi palaeochannel Jx,
which is gradually buried by fan deposits. The graph (inset) shows the increase of
areal extent through time. Source data: K map (, © Crown copyright,
reproduced with permission of the Controller HMSO), CORONA (), Landsat MSS
(), HEXAGON (), Landsat TM (), Landsat ETM+ ( and ).
 geoarchaeological research in lower khuzestan: state of the art 

development of a Karun megafan. This megafan was formed by successive avulsions of the
Karun. Two abandoned Karun channels (K and K) have been detected west of the present-
day Karun bed (K); traces of what might be a third Karun palaeochannel (K) are visible
underneath the irrigation canals east of K. The exact order of these palaeochannels is not
clear. K was active at least from the Parthian period ( bc–ad) onwards. Its meanders
are crossed by an extensive canal system (linked to the dam at Ahwaz) that most probably dates
from the Sasanian or Early Islamic periods, which gives an end-date to the activity of K. The
upper part of K is also crossed by these same canals. K’s more faded appearance suggests it
is older than K, but the lower visibility of its traces might be the result of different erosional
circumstances rather than older age. It is not clear when the Karun shifted to its present-day
position; only from the th century onwards we have definite proof that the Karun was flowing
in its K channel.
The avulsive shifting on the Karun fan probably influenced the changing positions of the
Karkheh and Jarrahi channels. The first known Karkheh palaeochannel (Kh) is synchronous
with the K Karun channel. The formation/extension of the Hawiza Marshes between –
ad may be linked with the avulsion of the Karkheh to its Kh bed. Later, in a single night
event in  ad, the Karkheh left its Kh channel, and took to its present-day position (Kh).
A number of attempts to revive the Kh branch have been executed, the latest being a bypass
canal dug in the late s.
Three successive fans formed by the Jarrahi have been detected. The first one (J), associated
with the earliest detectable Jarrahi channel, is linked to the archaeological site of Tell Tendy,
which dates back at least to the Achaemenid period (– bc). The river avulsion to its
present bed, situated further to the north, took place before or during Sasanian times (–
ad) and was followed by a period of river incision. The dam at Ja Nishin existed at least
from Sasanian times onwards, probably to secure water supply across the fan. A second fan
(J) is characterized by canals that resemble the canal systems west of the K channel, and
is therefore assumed to date from late Sasanian or Early Islamic period. The third Jarrahi fan
(J) is formed by a distributary canal system and has expanded rapidly into the surrounding
Shadegan marshes. An abandoned western extension of the present-day Jarrahi channel has
been detected on satellite imagery (palaeochannel Jx). The development of the Shadegan
marshes may have been triggered by the shifting of the Karun to its K bed that cut the Jx
channel.
Given the limited accessibility to the study area, the paucity of data makes that this recon-
struction of the coastal-fluvial evolution of the Lower Khuzestan plain should be regarded as
preliminary results. Further detailed fieldwork and dating of the palaeochannels is necessary
to refine the present knowledge and to assess the interplay between sea-level change, fluvial
variability and human interference.

Appendix: Geomorphological and Geological Terms

Accommodation space: refers to the volume available in a depositional system for sediment accumulation
to occur. It is a function of, among others, antecedent topography and relative sea-level change.
Aggradation: increase in land elevation due to the deposition of sediment.
Alluvial fan: cone-shaped depositional landform, typically located at the mountain front where a river is
released from its confinement and discharge conditions promote frequent avulsions (Bridge ).
Avulsion: shift of a river channel to a new course on the floodplain. It is the combined result of vertical
accretion of a channel system above the floodplain and external factors such as tectonic movements,
changes in base level, discharge conditions and human interference. Avulsions may be initiated
by the development of a crevasse channel, which eventually takes over the entire river flow, or by
reoccupation of a previous river channel. The abandoned river belt remains visible in the landscape
 v.m.a. heyvaert, p. verkinderen and j. walstra

due to traces of its channel fill, scrollbar patterns and/or elevated topography. These traces represent
the final stage of activity of the abandoned river belt and eventually will be removed by erosion or
covered by subsequent deposition (Berendsen ; Goudie ).
Bajada: zone of a more or less continuous alluvial apron lying between the mountain front and the basin
floor (Goudie ).
cal bp: calibrated radiocarbon age, in calendar years before . It is based on calibration curves that
correct for the natural deviations of atmospheric radiocarbon (C) through time (Gornitz ).
Coastal sabkhas: develop in the supratidal zone along arid coastlines and are inundated only by high
water levels (e.g. during spring tides and storm surges). They are characterized by a surface crust of
carbonates and sulphates, which has formed through precipitation from seawater and/or groundwater
due to evaporation (Bird ; Goudie ).
Crevasse splay: fan shaped deposit, resulting from floodwater breaking through a breach of a river levee.
Crevasse channels are typically distributive and/or anastomosing in planform and wash material onto
the floodplain (Goudie ).
Diatoms: unicellular algae, one of the most abundant groups of phytoplankton in a variety of marine and
freshwater environments. Their siliceous cell walls are taxonomically diagnostic, often well preserved
in the fossil record and therefore helpful in palaeoenvironmental studies (Gornitz ).
Floodplain: the low-lying parts of the alluvial plain at some distance from the river channel, which are
regularly flooded and remain submerged for prolonged periods. Due to the slow flow conditions, the
fine suspended particles (clay) finally settle here (Berendsen ).
Hogback: sharp ridge of hard rock, formed as a result of the steep dipping and differential erosion of
alternating hard and soft strata (Goudie ).
Intertidal: coastal zone situated between the mean high water and mean low water levels, that is daily
inundated by seawater.
Levee: wedge-shaped ridge (usually several decimetres or meters high) bordering the river channel and
the result of overbank deposition. Levees gradually build up as sediment deposited nearby the channel
tends to be coarser and thicker than further onto the floodplain. During high flows that do not exceed
bankfull discharge, sediment is deposited in the river bed, resulting in the river surface gradually rising
above the surrounding floodplain level (Berendsen ; Goudie ).
Meander cut-off: a U-shaped bend, cut-off from the main stream; a result of lateral river bed migration.
Initially, the cut-off forms a lake (oxbow lake), which slowly fills up with sediment, although even then
it may remain visible in the landscape for a long time (Berendsen ).
Megafan: a very large alluvial fans (>3 km2) with extremely low gradient (<.), dominated by fluvial
deposition processes. Usually, it is characterised by meandering river belts that episodically shift across
the fan surface, thereby creating distinct diverging alluvial ridges that correspond to evolutionary
stages of the fan (DeCelles/Cavazza ; Leier et al. ).
Pre-transgressive surface: top of the Pleistocene (or older) deposits, subsequently flooded by the Holo-
cene transgression.
Progradation: seaward extension of a coastal area.
Regression is the retreat of a shoreline, exposing previously submerged seafloor above sea-level. In a
vertical succession of sedimentary strata it is characterized by a shift from deeper marine sediments to
terrestrial and fluvial sediments. The position of a shoreline is determined by many interacting factors,
including sea-level change, tectonic movement and changing rates of sediment supply, deposition and
erosion (Lerner/Wilmoth Lerner ).
Scrollbars: distinct patterns of concentric ridges and swales on the inside of river bends. It is related to the
continuous migration of the river bed, with erosion on the outside of river bends and sedimentation
on the inside (Goudie ).
Spit and barrier island: both landforms result from the transport and deposition of sand by longshore
currents. Their outlines above high tide level are shaped largely by the dominant patterns of wave
action. A spit is attached at one end to the mainland while barrier islands are formed offshore across
the mouth of the embayment (Bird ; Goudie ).
Subtidal: coastal zone situated below the intertidal zone, permanently covered by the sea.
Supratidal: coastal zone situated above the mean high water level, only flooded by the sea during spring
tide and storm surges.
Tidal flats: occur along tide-dominated shorelines with high sediment supply, in particular estuaries and
deltas. Most of the sediment of such environment is in the intertidal zone, i.e. submerged and exposed
twice daily. The lower zone is characterized by sandy tidal flats, the middle zone consists of muddy
flats and the upper zone includes (vegetated) saltmarshes (Goudie ).
 geoarchaeological research in lower khuzestan: state of the art 

Transgression is the advance of the sea across previously exposed land surface, accompanied by a
landward displacement of coastal and marine sedimentary environments. In a vertical succession of
sedimentary strata it is characterized by a shift from shallow water and terrestrial sediments to deeper
coastal water sedimentary facies (Lerner/Wilmoth Lerner ).

Bibliography

Adams, J. (): “Agriculture and Urban Life in Early Southwestern Iran.” Science : –.
Admirality Tide Tables, Volume , . Indian Ocean and South China Sea, United Kingdom Hydro-
graphic Office, Somerset.
Ainsworth, W.F. (): Researches in Assyria, Babylonia and Chaldea, London.
———. (): A Personal Narrative of the Euphrates Expedition, London.
al-Idrisi, Abu Abd Allah Muhammad b. Muhammad: Nuzhat al-mushtaq fi-khtiraq l-afaq. Bombaci,
A., Rizzitano, U., Rubinacci, R., Veccia Vaglieri, L. (Eds.) (): Opus geographicum sive liber ad
eorum delectationem qui terras peragrare studeant, Naples-Rome. Jaubert, A. (Tr.) (): Nuzhat
al-mushtaq fi-khtiraq l-afaq, Paris.
Alizadeh, A. (): “Elymaean Occupation of Lower Khuzestan during the Seleucid and Parthian
Periods: A Proposal.” Iranica Antiqua : –.
Aqrawi, A.A.M. (): Recent sediments of the Tigris-Euphrates delta: the southern marshlands (Ahwar),
Unpublished PhD Thesis, University of London.
———. (): “Stratigraphic signatures of climate change during the Holocene evolution of the Tigris-
Euphrates delta, lower Mesopotamia.” Global and Planetary Change : –.
Aqrawi, A.A.M., Evans, G. (): “Sedimentation in lakes and marshes (Ahwar) of the Tigris-Euphrates
delta, southern Mesopotamia.” Sedimentology : –.
Audley-Charles, M., Curry, J., Evans, G. (): “Location of major deltas.” Geology : –.
al-Azzawi, M. (): La sédimentation actuelle sur la plaine de la basse Mésopotamie, Irak. Unpublished
PhD Thesis, University of Paris.
Baeteman, C., Dupin, L., Heyvaert, V.M.A. (/): “The Persian Gulf Shorelines and the Karkheh,
Karun, and Jarrahi Rivers: A Geo-Archaeological Approach. First Progress Report. . Geo-Environ-
mental Investigation.” Akkadica /: –.
———. (). Factors controlling the depositional history of estuarine infill during the Holocene. Actas
do º Simposio Interdisciplinar de Processos Estuarinos, Faro.
Baltzer, F., Purser, B.H. (): “Modern alluvial fan and deltaic sedimentation in a foreland tectonic
setting: the Lower Mesopotamian Plains and the Arabian Gulf.” Sedimentary Geology : –.
Beets, D.J., van der Spek, A.J.F. (). The Holocene evolution of the barrier and the back-barrier basins
of Belgium and the Netherlands as a function of late Weichselian morphology, relative sea-level rise
and sediment supply. Geologie en Mijnbouw/Netherlands Journal of Geosciences /: –.
Beke, C.T. (): “On the geological evidence of the advance of the land at the head of the Persian Gulf.”
The London and Edinburgh Philosophical Magazine and Journal of Science /: –.
Berendsen, H.J.A. (): De vorming van het land: Inleiding in de geologie en geomorfologie. Fysische
geografie van Nederland, Assen (nd ed).
Bird, E.C.F. (): Coastal Geomorphology: An Introduction, Chichester (nd ed).
Blair, T.C., McPherson, J.G. (): “Alluvial fans and their natural distinction from rivers based on mor-
phology, hydraulic processes, sedimentary processes, and facies assemblages.” Journal of Sedimentary
Research /: –.
Bridge, J.S. (): Rivers and Floodplains: Forms, Processes, and Sedimentary Record, Oxford.
Brunsden, B., Doornkamp, J.C., Fookes, P.G., Jones, D.K.C., Kelly, J.M.H. (): “Large scale geomor-
phological mapping and highway engineering design.” Quarterly Journal of Engineering Geology /:
–.
Chesney, F.R. (): Expedition for the survey of the rivers Euphrates and Tigris, carried on by order of
the British government, in the years , , and , London.
Christensen, P. (): The decline of Iranshahr: Irrigation and environments in the history of the Middle
East B.C. to A.D. , Copenhagen.
Cole, S.W., Gasche, H. (): “Documentary and other archaeological and environmental evidence
bearing on the identification and location of the rivers of Lower Khuzestan and the position of the
head of the Persian Gulf c. BC – AD.” Akkadica /–: –.
 v.m.a. heyvaert, p. verkinderen and j. walstra

Couto, D., Bacqué-Grammont, J.-L., Taleghani, M. (): Atlas Historique du Golfe Persique (XVIe–
XVIIIe siècles), Turnhout.
Dalongeville, R. (): “Présentation physique générale de l’île de Failaka.” In Calvet, Y., Gachet, J.
(Eds.), Failaka. Fouilles françaises –, Lyon: –.
Dalongeville, R., Sanlaville, P. (): “Confrontations des datations isotopiques aux données géomor-
phologique à propos des variations relatives du niveau marin sur la rive arabe du Golfe Persique.”
In Aurenche, O., Evin, J., Hours, F. (Eds), Chronologies in the Near East, Relative Chronologies and
Absolute Chronology –BP, Colloque international du CNRS (BAR International Series .
Archaeological series ), Oxford-Lyon: –.
———. (): “L’évolution des espaces littoraux du Golfe Persique et du Golfe d’Oman depuis la Phase
finale de la transgression post-glaciaire.” Paléorient : –.
DeCelles, P.G., Cavazza, W. (): “A comparison of fluvial megafans in the Cordilleran (Upper
Cretaceous) and modern Himalayan foreland basin systems.” Geological Society of America Bulletin
: –.
De Morgan, J. (): “Etude géographique sur la Susiane. Recherches Archéologiques.” Mémoires de la
Délégation en Perse : –.
De Thévenot, J. (): Le voyage de Mr de Thevenot au Levant, Amsterdam.
Evans, G. (): “The development of the Mesopotamian delta, comments.” Geographical Journal :
–.
Fiey, J.M. (): “L’Élam, la première des métropoles ecclésiastiques syriennes orientales.” Melto : –
; reprinted in Fiey, J.M. (): Communautés syriaques en Iran et Irak des origines à , London
(Ch. III).
———. (): “L’Élam, la première des métropoles ecclésiastiques syriennes orientales (suite)”, Patrologie
de l’ Orient : –; reprinted in Fiey, J.M. (): Communautés syriaques en Iran et Irak des
origines à , London (Ch. IIIb).
Gasche, H., Paymani, A.R. (): “Repères archéologiques dans le Bas Khuzestan.” Akkadica : –.
Gornitz, V. (Ed.) (): Encyclopedia of Paleoclimatology and Ancient Environments. Dordrecht.
Goudie, A.S. (Ed.) (): Encyclopedia of Geomorphology, London.
Graadt van Roggen, D. (): “Notice sur les anciens travaux hydrauliques en Susiane.” Mémoires de la
Délégation en Perse VII: –.
Gutman, G., Byrnes, R., Masek, J., Covington, S., Justice, C., Franks, S., Headley, R. (): “Towards
monitoring Land-cover and land-use changes at a global scale: the global land survey .” Pho-
togrammetric Engineering and Remote Sensing /: –.
Gyselen, R. (): La géographie administrative de l’empire sassanide. Les témoignages sigillographiques
(Res Orientales ), Paris.
———. (): Nouveaux matériaux pour la géographie historique de l’empire sassanide: sceaux adminis-
tratifs de la collection Ahmad Saeedi (Studia Iranica Cahier ), Paris.
Hansman, J. (): “Charax and the Karkheh.” Iranica Antiqua : –.
———. (a): “Seleucia and the Three Dauraks.” Iran : –.
———. (b): “The Mesopotamian delta in the first millennium.” The Geographical Journal : –
.
Harvey, A.M., Mather, A.E., Stokes, M. (): “Alluvial fans: geomorphology, sedimentology, dynam-
ics—introduction. A review of alluvial-fan research.” In Harvey, A.M., Mather, A.E., Stokes, M. (Eds.),
Alluvial fans: geomorphology, sedimentology, dynamics (Geological Society Special Publications, ),
Bath: –.
Haynes, S.J., McQuillan, H. (): “Evolution of the Zagros suture zone.” Geological Society of America
Bulletin : –.
Hessami, K., Nilforoushan, F, Talbot, C.J. (): “Active deformation within Zagros Mountains deduced
from GPS measurements.” Journal of the Geological Society of London /: –.
Heyvaert, V.M.A. (): Fluvial Sedimentation, Sea-level History and Anthropogenic Impact in the Great
Mesopotamian plain: A new Holocene Record. Unpublished PhD Thesis, Vrije Universiteit Brussel.
Heyvaert, V.M.A., Baeteman, C. (): “Holocene sedimentary evolution and palaeocoastlines of the
Lower Khuzestan plain (southwest Iran).” Marine Geology /–: –.
———. (): “A Middle to Late Holocene avulsion history of the Euphrates river: a case study from Tell
ed-Dçr, Iraq, Lower Mesopotamia.” Quaternary Science Reviews : –.
Heyvaert, V.M.A., Weerts, H.J.T. (): “Development of the Holocene Karun megafan, Lower Khuzes-
tan, southwest Iran (Shatt-el Arab region).” Quaternary International – (suppl): –.
Heyvaert, V.M.A., Walstra, J., Verkinderen, P., Weerts, H.J.T., Ooghe, B (): “The role of human
 geoarchaeological research in lower khuzestan: state of the art 

interference on the channel shifting of the river Karkheh in the Lower Khuzestan plain (Mesopotamia,
SW Iran).” Quaternary International :–.
Höpner, T. (): “Intertidal Treasure Khowr-e Mussa-Unraised.” Wadden Sea Letter : –.
Hritz, C., Wilkinson, T.J. (): “Using Shuttle Radar Topography to map ancient water channels in
Mesopotamia.” Antiquity : –.
Hritz, C. (): “Tracing Settlement Patterns and Channel Systems in Southern Mesopotamia Using
Remote Sensing.” Journal of Field Archaeology /: –.
Hudson, R.G.S., Eames, F.E., Wilkins, G.L. (): “The fauna of some recent marine deposits near
Basrah, Iraq.” Geological Magazine : –.
Ibn al-Muhalhil, Abu Dulaf Mis#ar: al-Risala l-thaniya. Minorsky, V. (Ed. and tr.) (): Abu Dulaf Mis #ar
ibn Muhalhil’s travels in Iran (circa A.D. ), Cairo.
Ibn Battuta: Tuhfat al-nuzzar. Defremery, C., Sanguinetti, B.R. (Ed. and tr.) (–), Paris.
Ibn Hawqal, Abu l-Qasim b. Ali: Kitab surat al-ard. Kramers, J.H. (Ed.) (): Surat al-ard, Leiden.
Kramers, J.H., Wiet, G. (Tr.) (): Configuration de la terre, Paris-Beirut.
Ibn Rusta, Abu #Ali Ahmad: Kitab al-a #laq al-nafisa. de Goeje, M.J. (Ed.) (), Leiden.
Ibn Sarabiyyun: Kitab #aja" ib al-aqalim al-sab #a. von Mžik, H. (Ed.) (), Leipzig.
al-Isfahani, #Imad al-Din: Kharidat al-qasr wa-jaridat al- #asr. al-Athar, M.B., Sa#id, J. (Eds.) (),
Baghdad.
al-Istakhri, Abu Ishaq Ibrahim: Kitab al-Masalik wa-l-mamalik. de Goeje, M.J. (Ed.) (), Leiden.
Johnson, G.A. (): Local Exchange and Early State Development in Southwestern Iran (Anthropolog-
ical Papers of the Museum of Anthropology ), Ann Arbor.
Jones, L.S., Schumm, S.A. (): “Causes of avulsion: an overview.” In Smith, N.D., Rogers, J. (Eds.),
Fluvial Sedimentology VI (Special Publication of the International Association of Sedimentologists,
), Oxford: –.
Kirkby, M. (): “Land and water resources of the Deh Luran and Khuzistan plains.” In Hole, F. (Ed.),
Studies in the archaeological history of the Deh Luran Plain: the excavations of Chagha Sefid (Memoirs
of the Museum of Anthropology ): –.
Lambeck, K. (): “Shoreline reconstructions for the Persian Gulf since the last glacial maximum.”
Earth and Planetary Science Letters : –.
Larsen, C.E., Evans, G., (): “The Holocene geological history of the Tigris-Euphrates-Karun Delta.”
In Price, W.C. (Ed.), The environmental history of the Near and Middle East, London: –.
Layard A.H. (): “A Description of the Province of Khúzistán.” Journal of the Royal Geographical
Society of London : –.
Le Strange, G. (): The Lands of the eastern Caliphate, Mesopotamia, Persia and Central Asia from
conquest to the time of Timur, Cambridge.
Lees, G.M., Falcon, N.L. (): “The geographical history of the Mesopotamian Plains.” The Geograph-
ical Journal : –.
Leier A.L., DeCelles P.G., Pelletier J.D. (): “Mountains, monsoons, and megafans.” Geology /:
–.
Lerner, K.L., Wilmoth Lerner, B. (Eds.) (): World of earth science. Farmington Hills, Michigan.
Lillesand, T.M., Kiefer, R.W. (): Remote sensing and image interpretation, New York.
Loftus, K. (a): “On the Determination of the River “Euloeus” of the Greek historians.” Journal of the
Royal Geographical Society of London : –.
———. (b): Travels and Researches in Chaldaea and Susiana, New York.
Macfadyen, W.A., Vita-Finzi, C. (): “Mesopotamia: The Tigris-Euphrates Delta and its Holocene
Hammar fauna.” Geological Magazine : –.
Mas#udi, Abu l-Hasan Ali: Muruj al-Dhahab. Barbier de Maynard, A.C., Pavet de Courteille, A. (Eds.)
(–), revisé par Ch. Pellat, Beirut.
Miller, K. (): Mappae Arabicae, Stuttgart.
Moghaddam, A. (in press): “A note on the Gargar Irrigation System.” In Abdi, K. (Ed.), Mazdesn Shapur
ke chihr az yazdan: Essays in Memory of A. Shapur Shahbazi, Tehran-Persepolis.
Morozova, G.S. (): “A review of Holocene avulsions of the Tigris and Euphrates rivers and possible
effects on the evolution of civilizations in lower Mesopotamia.” Geoarchaeology /: –.
al-Muqaddasi, Shams al-Din Muhammad: Ahsan at-taqasim fi ma #rifat al-aqalim. de Goeje, M.J. (Ed.)
() Descriptio Imperii Muslemici (Bibliotheca Geographorum Arabicorum ), Leiden. Collins, B.
(Tr.) () The Best Divisions for Knowledge of the Regions, Reading.
Mustawfi, Hamd Allah: Nuzhat al-Qulub. Le Strange, G. (Ed. and tr.) (–) The Geographical Part
of the Nuzhat-al-Qulub of Hamd-Allah Mustawfi of Qazvin (= EJW Gibb Memorial Series ), London.
 v.m.a. heyvaert, p. verkinderen and j. walstra

Ooghe, B. (): Off the Beaten Track: Travellers, maps and the landscapes of Ottoman Mesopotamia
Unpublished PhD thesis, Ghent University.
Persian Gulf. Atlas of Old and Historical Maps (B.C. – A.D.), Centre for Document and
Diplomatic History, Tehran.
Philip, G., Donoghue, D., Beck, A., Galiatsatos, N. (): “CORONA satellite photography: an archae-
ological application from the Middle East.” Antiquity : –.
Pliny the Elder (c.  ad): Historia Naturalis. Rackham, H. (Ed.) (—): Loeb Classical Library
, , , , , , , , , Cambridge, Massachusetts.
Potts, D.T. (): The Archaeology of Elam: Formation and Transformation of an Ancient Iranian State,
Cambridge.
Purser, B.H. (): The Persian Gulf, Holocene Carbonate Sedimentation and Diagenesis in a Shallow
Epicontinental Sea, Berlin.
Purser, B.H., Ya"acoub, S.Y., Al-Hassni, N.H., Al-Azzawi, M., Orzag-Sperber, F., Hassan, K.M., Plaziat,
J.C., Younis, W.R. (): “Charactères et évolution du complexe deltaique Tigre-Euphrate.” Mémoire
de la Société Géologique de France : –.
Rzoska, J. (): “Euphrates and Tigris, Mesopotamian ecology and density.” In Illiesled, J. (Ed.),
Monographiae Biologicae , London.
Ryder, C.H.D. (): "The demarcation of the Turco-Persian boundary in —." The Geograph-
ical Journal /: -–.
Sahab, M.R., Nazarahari, R., Sarvestani, K.K., Safinejad, J., Mahmoudian, A.A., Vatandoust, G., Tehrani,
F.F., Qassemi, H. (): Persian Gulf: Atlas of Old and Historical Maps ( B.C.– A.D.), Tehran.
Sanlaville, P. (): “Considération sur l’évolution de la basse Mésopotamie au cours des derniers
millénaires.” Paléorient : –.
———. (): “The deltaic complex of the Lower Mesopotamian plain and its evolution through millen-
nia.” In Nicholson, E., Clark, P. (Eds), The Iraqi Marshlands, London: –.
Schuol, M. (): Die Charakene: ein mesopotamisches Königreich in hellenistisch-parthischer Zeit,
Stuttgart.
Schwarz, P. (): Iran im Mittelalter nach den arabischen Geographen, Leipzig.
Smith, M.J., Pain, C.F. (): “Applications of remote sensing in geomorphology.” Progress in Physical
Geography /: –.
Stuiver, M., Reimer, P.J. (): “Extended C data base and revised CALIB . C age calibration.”
Radiocarbon : –.
Unvala, J.M. (): “The Origin of the Pine-Cone Decoration of the Imamzadehs of Khuzistan.” Bulletin
of the School of Oriental Studies /: –.
Ur, J. (): “CORONA Satellite Photography and Ancient Road Networks: A Northern Mesopotamian
Case Study.” Antiquity : –.
Verkinderen, P. (): Tigris, Euphrates, Karun, Karhe, Jarrahi: Tracking the traces of five rivers in Lower
Iraq and Khuzistan in the Early Islamic Period. Unpublished PhD Thesis, Ghent University.
Vita-Finzi, C. (): “Rates of Holocene folding in the coastal Zagros near Bandar Abbas, Iran.” Nature
: –.
Walstra, J., Heyvaert, V.M.A., Verkinderen, P. (a): “Assessing human impact on alluvial fan devel-
opment: a multidisciplinary case-study from Lower Khuzestan (SW Iran).” Geodinamica Acta :
–.
Walstra, J., Verkinderen, P., Heyvaert, V.M.A. (b): “Reconstructing landscape evolution in the Lower
Khuzestan plain (SW Iran): integrating imagery, historical and sedimentary archives.” In Cowley, D.C.,
Standring, R.A., Abicht, M.J. (Eds.), Landscapes through the Lens: Aerial Photographs and Historic
Environment (Occasional Publication of the Aerial Archaeology Research Group, ), Oxford: –
.
Walstra, J., Heyvaert, V.M.A., Verkinderen, P. (): “Mapping Late Holocene landscape evolution and
human impact—a case-study from Lower Khuzestan (SW Iran).” In Smith, M.J., Paron, P., Griffith,
J.S. (Eds.), Geomorphological Mapping: Methods and Applications (Developments in Earth Surface
Processes, ): –.
Wilkinson, T.J. (): Archaeological Landscapes of the Near East, Tucson, Arizona.
Ya"acoub, S.Y., Purser, B.H., Al-Hassni, N.H., Al-Azzawi, M., Orzag-Sperber, F., Hassan, K.M., Plaziat,
J.C., Younis, W.R. (): Preliminary study of the Quaternary sediments of SE Iraq. Joint project between
the Geological Survey of Iraq and University of Paris XI, Orsay, Unpublished report.
al-Zamel, A.Z. (): Geology and Oceanography of Recent Sediments of Jazirat Bubiyan and Ras As-
Sabiyah, Kuwait, Arabian Gulf. Unpublished PhD Thesis, University of Sheffield.
 INDEX OF DIVINE NAMES

Adad, , , , , , , ,  Išme-karāb, , , , , , , , ,
Addu, see Adad , , , 
Amurru,  Išni-karāb, see Išme-karāb
Annunı̄tum,  Ištar, , , , , 
Anu, , , ,  Ištar kittitum, , 
Anunakki, , , ,  Ištar šarbat, , 
Arkayı̄tu,  Ištaran, 
Assur, , ,  Itur-Mer, 
Aši, 
Aya,  Kilah-Šupir, , , , 
Kiririša, , , , , , , , ,
Ba#al,  , , , , , 
Bau,  KUR.GAL, , 
Bēlat-Šuhnir, , 
Bēlat-Teraban, ,  Lāgamāl, , , , , , , , ,
, , 
Daēna, ,  Lahuratil, see Ruhurater
Dagan,  Lali, 
Dilbat, ,  Laliya, , 
Dumuzi, , ,  Laqipum, 

Ea, ,  Manzat, , , 

Eanna (temple), ,  Marduk, , , 
Egalmah (temple),  MAR.TU, 
Ekur (temple),  Mashti, 
Ehilianna (sanctuary),  Mešlam (temple), 
Ellil, see Enlil Mithra, , , 
d
Enki,  Mùš.lam, 
Enlil, , , , , , , 
Enzak, ,  Nabu, 
Ereškigal,  Nahhunte, , , , , 
Esagil (temple),  Nairiiō.Saŋha, 
Nanaya, , , 
GAL, see Napiriša Nanna, , , , 
Gibil/Girra,  Nanšak, 
Gilgameš,  Napiriša, , , , , , , , ,
Gula,  , , , , , , 
Napratep, , , , 
Harbe,  Nergal, , , , 
˘
Hendursağa,  NIN-ali, , , , 
Hišmitik, , ,  Ninazu, 
Humban, , , ,  Ningal, 
Hutran,  Ningirsu, , 
Ningišzida, , 
Iapru,  Ninhursag, 
IM, , ,  Ninlil, , 
Immiriya, ,  Nintinuga, 
Inan(n)a, , ,  Nintu, 
Inšušinak, , , , , , , , , Ninurta, , 
, , , , , , , , , , Nisaba, 
, , , , , , , , , ,
, , ,  Parti, 
Pini(n)kir, , , , , , , , 
 index of divine names

Rašnu, , , ,  Šati, , , 

Ruhurater, , , , ,  Šimut, see Simut
Šulpae, 
Simut, , , , , , ,  Šušnakiul, 
Sîn, , , 
Suen, see Sîn Tepti, 
Sraoša, , , ,  Tišpak, , , , 
Suhsipa, ,  Tuzi, 
Šah, 
˘ , , 
Šala, Upurkupak, , 
Šamaš, , , , , , , , , , Usur-amassa, 
 ˙
Šašum,  Zababa, 
 INDEX OF PERSONAL NAMES

[…]-tir-Humban,  , , , , , , , , , ,
, 
A"abba,  Assurnasirpal II, 
Abi-ešuh, , , , , , , , , Ašiši, , 
, , ,  Aššur-bēl-nišēšu, 
Abi-ili, , , ,  Aššur-nārāri, 
Abi-Samar,  Aššur-rabi, 
Abi-simti,  Aššur-rêm-nišešu, 
Abum-waqar, , ,  Atamrum, , , , , , 
Abraham,  Atanah-ili, , , , , , , 
Adad-bāni,  Athibu, ˘ , 
Adad-dumqi,  Atta-[…], 
Adad-ereš,  Atta-hamiti-Inšušinak, , , 
Adad-nêrârî I, , ,  Atta-hušu, , , , , , , , 
Adad-rabi,  Atta-menetu, 
Adad-šuma-iddina, , , , , , Atta-puni, 
 Attar-kittah, , , 
Adad-šuma-usur, , , ,  Attaruh-Uktuh, 
Adallal,  ˙ Ayadaragalama, 
Addaten, , 
Ahumme,  Babati, , 
Akšir-x, , , ,  Bahram II, 
Alexander the Great, ,  Bar-Uli, , , 
al-Idrisi,  Bēl-ibni, , , , 
al-Istakhri, ,  Bēlı̄-qarrād, 
al-Muqaddasi, , ,  Bēlšunu, 
Amaladin,  Bidu, 
Amar-Suen, see Amar-Sîn Bilalama, , , , 
Amar-Sîn, , , ,  Burnaburiaš I, 
Amar-Zu"ena(k), see Amar-Suen Burnaburiaš II, , , , , , , ,
Amedirra,  , , , 
Amenemhat II, 
Amma-haštuk,  Charon, 
Amma-tedak,  Cyrus (Parsumash), , 
Ammaten,  Cyrus II, 
Ammiditana, , , ,  *Çauša-, 
Ammisaduqa, , , , ,  *Çaušaka-, 
˙ 
Amraphel, *Çaušaya-, 
Amudpa"el, , 
Amud-pî-el, see Amudpa"el Daduša, 
Annubanini,  Dalta, 
Antigonus,  Danna(n)-Pinigir, 
Apil-Kūbi,  Darius I, , , , 
Aqba-ahum, , ,  Delilah, 
Arad-dKUR.GAL,  Diodorus Siculus, 
Arik-dên-ili, ,  Dunna-Šah, , 
Arrian,  Dununu, ˘
Arriyuk, , , 
Artaxerxes I,  Ebarat, , 
Artaxerxes II, ,  Ebarat I, , 
Assur-uballit I, , , ,  Ebarat II, , , , , , , , 
Assurbanipal, ˙ , , , , , , , Ekiba, 
 index of personal names

Elamatum,  Igihalki, , , , 

Enkidu, ,  Igihalkids, , 
Enlil-nādin-ahi, , , , ,  Ikšud-lâ-šêmêšu, 
˘ 
Enlil-nādin-apli, Ikun-pî-Sîn, 
Enlil-nādin-šumi, , , , ,  Ikūnum, 
Enlil-nêrârî, , , ,  Ilân-sûrâ, 
Enmerkar,  ˙
Ili-barna, 
Etel-pî-Nabium, ,  Ilı̄šu-ilı̄ya-Nurahum, 
Eumenens,  Ilušuma, 
Imazu, 
Gamāl-ili, , , ,  Imbappi, , 
Gilgameš, ,  Indabibi, , , , , 
Gimilli-Sin, , , , ,  Indada, 
Gimillum,  Inšušinak-abi-enši, 
Gudea, , , , , , , , ,  Inšušinak-gāmil, , , , , , , 
Gulkišar,  Inšušinak-kı̄nam-idi, , , , 
Gungunum, , ,  Inšušinak-sunkik-nappipir, , , 
Inšušinak-šar-ilani, see Inšušinak-sunkik-
Haldu-muluk,  nappipir
Hallutaš-Inšušinak, , ,  Inšušinak-šemi, , , , , , 
Hallutuš-Inšušinak, , ,  Ipiq-Adad II, 
Halu-rabi, , , ,  Iqišuni, , , , 
Hammi-ištamar,  Isi-Qatar, 
Hammi-šagiš,  ˙ ˙ , 
Išbi-Erra,
Hammurabi of Aleppo, ,  Išîm-Addu, see Išme-Addu
Hammurabi of Babylon, , , , , , Išme-Addu, 
, , , , , ,  Išmeanni, , , 
Hammurabi of Kurdâ, ,  Išme-Dagan, , 
Hanni, , , , , ,  Išme-Sîn, 
Haštuk,  Itti-ili-balitu, , , , , , , 
Haya-Sumu,  Iturı̄ya, ˙
Hita, 
Huban-ahpi, ,  Kadašman-Enlil I, , 
Huban-mirriš,  Kadašman-Enlil II, , 
Huban-Šuturuk, , , ,  Kadašman-Harbe I, , , , , 
Humban-haltaš III, , ,  ˘
Kadašman-KUR.GAL, , 
Humban-nikaš II, ,  Kara-har-daš, 
Humban-numena, , , , , , , ˘
Karaindaš, , 
, , , , , , , , , , Kaštiliaš I, 
, , ,  Kaštiliaš II, 
Humur-tabal,  Kaštiliaš III, 
Hunnili, , ,  Kaštiliaš IV, , , 
Hurba-tela, , ,  Kayyayya, 
Hutelutuš-Inšušinak, , , , , , , Kidin-Hutran I, , 
, , , ,  Kidin-Hutran II, , , , , , , ,

Ibal-pi-El II, , , ,  Kidin-Hutran III, , , 
Ibbi-Sîn, , , , ,  Kidin-Hutrudiš, 
Ibbi-Suen, see Ibbi-Sîn Kidinu, , 
Ibn al-Muhalhil,  Kikia, 
Ibn Hawqal, ,  Kindattu, , , 
Ibni-Adad, ,  Kirikiri, , 
Ibni-dKUR.GAL,  Kirname, , 
Idaddu I, , , , ,  Kitten-rakittapi, , , 
Idaddu II, , , ,  Kudu-Zuluš I, , , , , , , , ,
Iddin-Dagan,  , , , , , 
Iddin-Sîn,  Kudu-Zuluš II, 
d
IGI+DU-unukaš,  Kudur-Mabul, 
 index of personal names 

Kudur-Nahhunte, see Kutir-Nahhunte II Naram-Suen, see Narām-Sîn

Kuk-Adar, , , , , ,  Narubtu, , , 
Kuk-Kirwaš,  Natan, 
Kuk-Manzat,  Nazi-bugaš, 
Kuk-Nahhunte,  Nebuchadnezzar I, , , , , , 
Kuk-Našur II, , , , , , , , Nesu, 
, , , , , ,  Nibe, 
Kuk-Našur III, , ,  Nūr-Adad, 
Kuk-sanit, , , ,  Nūr-ahum, , 
Kuk-Simut,  Nūr-ilı̄šu, , , , 
Kuk-šarrum,  Nūr-Šamaš, 
Kuk-Tanra,  Nūr-Sîn, 
Kunnam, 
Kunnaman, see Kunnam Pa’e, , , , 
Kurigalzu,  Pahir-Iššan, , , , , , , , ,
Kurigalzu I, , , , , , , , , , , , , , 
, , , , , , , ,  Pala-Iššan, , , , , , 
Kurigalzu II, , , , , , , , Per#i-Adad, 
 Per#i-Amurru, 
Kuter-Šilhaha, , , ,  Pešgaldarameš, 
Kutir-Nahhunte I, , , , , , , , Pislume, , 
, ,  Pliny, 
Kutir-Nahhunte II, , , , , , , Plutarch, 
, , , , , , ,  Ptolemy, 
Kuya,  Puzriš-Dagan, , 
Kuyyayya, see Kayyayya Puzur-Aššur, 
Puzur-Inšušinak, , , , , , , ,
Lali-sunki,  , , , , , , , 
Lalintaš,  Puzur-Mama, 
Liririša,  Puzur-Šulgi, 
Liwwirum, , ,  Puzur-Teppuna, , , , , , 
Lugalannemundu, 
Raši, 
Mamannuwiš,  *Rašnubara-, 
Maništušu, ,  *Rašnuca-, 
Mannu-lū-jā"u,  *Rašnudāta-, 
Marduk-apla-iddina I,  *Rašnuka-, 
Marduk-nāsir, , , , , ,  Razama, 
Mekubi, ,˙,  Rı̄m-Adad, , , 
Meli-Šipak, , , , , , , , , Rı̄m-Sîn, , 
,  Rimuš, 
Meptûm, , ,  Rišap-La, , 
Milki-El, 
Mišimruh,  Samiya, 
Mulballitat-Šerua,  Samson, 
˙
Mušēzib-Marduk, , ,  Samsi-Addu I, see Šamši-Adad I
Samsu-ditana, , , 
Nabû-bēl-šumati, , , , , , , Samsu-iluna, , , , , , , , ,
 , , , , , 
Nabuchodonosor I, see Nebuchadnezzar I Sargon I, , 
Nabû-šar-usur,  Sennacherib, 
˙ 
Nabû-zēr-ibni, Siluši, 
Nahhunte-Utu, ,  Simah-ilânê, 
Napir-asu, , , ,  Simma-ila, see Simah-ilânê
Napirašu, see Napir-asu Simut-wartaš, , 
Napiriša-untaš, , ,  Sîn-erı̄bam, 
Narām-[ ],  Sîn-iddinam, , , , , , , ,
Narām-Sîn, , , , ,  , 
 index of personal names

Sîn-muballit,  Šutur-Nahhunte, 

Sîn-tayyar,  ˙ Šuzigaš, 
Sir-ahu-pitir, 
Siruktuh, , , , , ,  Tahhi, 
Siwe-palar-huhpak, , , , , , , Takil-ana-ilišu, 
, , , , , ,  Tammaritu, , , , , , 
Srwšdtk,  Tan-Ruhurater I, , , , , 
Srwšk,  Tan-Ruhurater II, 
Srwšssnk,  Tan-Uli, , , , , , , , 
Sulgi(r), see Šulgi Tattâ, , , , , 
Sumu-abum,  Tehip-Tilla, 
Šadununu, ,  ˘
Tem-Enzak, 
Šalmaneser III,  Tem-sanit, 
Šamaš-rabi,  Temti-Agun, , , , , , , , ,
Šamaš-šemmê, , , ,  , , , 
Šamaš-šum-ukin, ,  Temti-halki, , , , , , , , 
Šamši-Adad I, , , , ,  Tepti-Ahar, , , , , , 
Šamši-Addu I, see Šamši-Adad I Tepti-Huban-Inšušinak, , 
Šarraya,  Tept-raptaš, , , , , , 
Šati-Dudu,  Tetep-mada, 
Šati-duš,  Te-Umman, , , , , , 
Šati-hupiti, ,  Tiglath-phalazar I, 
Šati-Humban, ,  Tiš-Ulme, 
Šati-kitin,  Tirigan, , 
Šati-Napiriša,  Tukulti-Ninurta I, , , , , 
Šati-šilhak,  Turunkuz, 
Šati-te[ ], 
Šatin-Humban,  Ukal, 
Šatu, , ,  Umbahabua, , , 
Šelebum,  Umhuluma, 
Šeplarpak, see Siwe-palar-huhpak Ummanaldaš, , , , , , , ,
Šeraš,  , , , , 
Šerum-bāni, ,  Ummanigaš, , , 
Šerum-ili,  Ummanšibar, , , 
Šilhaha, , , , , , , , , Ummanunu, 
, , ,  Unpahaš-Napiriša, , , , 
Šilhak-Inšušinak I, , , , , , , Untaš-Napiriša, , , , , , , ,
, , , , , , , , , , , , , , , , , , , ,
, , , , , , , , , , , , , , , , , , , ,
, ,  , , , , , 
Šilhak-Inšušinak II,  Ur-Namma(k), see Ur-Nammu
Šilhina-hamru-Lagamar,  Ur-Nammu, , , , , , , ,
Šimbi-išhu, see Šimpi-išhuk , , 
Šimpi-išhuk, ,  Ur-Ningirsu II, 
Širuk-tuh, see Siruk-tuh Urtak, , 
Šu-ilı̄ya, ˘, ,  ˘ Ururu, , , 
Šulgi, , , , , , , , , , Ur-Utu, , 
, ,  Utu-hegal, , , , 
ŠulŠI-kudur, 
Šuma,  *Vahyarašnu-, 
Šu-Baba, , 
Šu-Suen, see Šu-Sîn Warad-Ištar, , , , , 
Šu-Sîn, , , , ,  Warad-Kūbi, , 
Šutruk-Nahhunte I, , , , , , , Winnirke, 
, , , , , , , , , ,
, , ,  Yagihhaddu, 
Šutruk-ša-x, ,  Yahdun-Lim, , 
Šutruru, ,  Yarim-Addu, 
 index of personal names 

Yarim-Lim, ,  Zababa-šuma-iddina, , , , , 

Yasim-Dagan, ,  Zangadar, 
Yasmah-Addu, ,  Zimri-Lim, , , , , , , , ,
, , , , , , , , , 
Zoroaster, 
 INDEX OF TOPOGRAPHICAL NAMES

Abadan, ,  Babylon, , , , , , , , , ,
Ab-e Diz, see Idide River , , , , , , , , , ,
Abu Habbah, see Sippar-Jahrurum , , , , , , , , 
Abu Salabikh,  Babylonia, , , , , , , , , ,
Abullat, , ,  , , , , , , , , , ,
Adab, ,  , , , , , , , , , ,
AdamDUN, , , , ,  , 
Adhaim River, ,  Bactria, , , , 
Afghanistan,  Badrah, see Der
Agha Jari anticline, ,  Baghdad, , , 
Ahmad al-Hattu,  Bahar, 
Ahwaz, , , , , , , , , Bahrain, see Dilmun
, , , ,  Bakhtiari Mountains, , , 
Akbanu,  Balikh River, , 
Akkad, , , , , , , , ,  Baluchistan, 
Akshak,  Bampur, , , 
Alborz,  Banesh, , 
Ale,  Bandar Shahpur, see Bandar-e Imam Khomeini
Aleppo, , , , , , , , ,  Bandar-e Imam Khomeini, 
Amanum,  Bani Surmah, , , 
Amara, ,  Barahšum, 
Amarna, ,  ˘ 
Barbar,
Amirabad, ,  Basra, , 
Amlash,  Behbehan, , 
Amorites, , , , , , ,  Benjaminites, , , , , 
Anarak,  Bensim"alites, 
Andarig, ,  Bishapur, 
Anshan, , , , , , , , , , , Bit-Imbi, , , , 
, , , , , , , , , , Bostan, 
, , , , , , , , , , Bubilu, , , 
, , , , , , , , , , Bushehr, , , , , , , , 
, , , , , , , , , , Byblos, 
, 
Anzan, see Anshan Cappadocia, , , 
Arabia,  Central Asia, , , 
Arabistan, see Khuzestan Chavar, 
Aratta,  Chicago, 
Arbela,  Choga Gavaneh, 
Arisman,  Chogha Mish, , , 
Arjan,  Chogha Pahn, , , , 
Arrapha,  Chogha Zanbil, see Dur-Untash
Asampi,  Crete, 
Asia Minor,  Ctesiphon, 
Assur, , , , , , , , , ,
, , , , ,  Da-o Dukhtar, , , 
Assyria, , , , , , , , , , Dahhara, 
,  Dahhashara, 
Awal, , , , , , , ,  Darband-e Gawr, , 
Awan, , , , , , , , ,  Darvazeh Tepe, , 
Ayanharishda,  Dasht-e Fahliyan, 
Ayapir, , , ,  Dasht-e Kavir, 
Dasht-e Nurabad, 
 index of topographical names 

Dasht-e Rustam-e Do,  Gaziantep, 

Dasht-e Rustam-e Yek, ,  Ghabristan, 
Dawraq, , ,  Ghaggar, 
Deh Bala, see Der GIRkal, 
Dehloran, see Deh Luran Girsu, , , , , 
Deh Luran, , , , , , , ,  Gisat, , , , 
Deh-e Now,  Giyan, 
Der, , , , , , , , , ,  Godin Tepe, , , , 
Deshawer,  Gonur Depe, , , , 
Dez River,  Great Khurasan Road, , , , , ,
Dilbat, ,  , 
Dilmun, , , , , , ,  Greece, 
Diniktum, ,  Guabba, 
Diyala, , , , , , , , , , , , Gutium, , 
, , , , , , , , , ,
 Habuba Kabira, 
Dur-Abi-eshuh, ,  Haffar Canal, , 
Dur-Kurigalzu, ,  Haft Tepe, , , , , , , , , ,
Dur-Sharrukin, ,  , , , , , , , , 
Dur-Undasi, see Dur-Untaš Haidanu, , 
Dur-Untash, , , , , , , , , Halil, , 
, , , , , , , , , , Hamadan, , , , , , , 
, , , , , , , , , , Hamanu, 
, , , , , , , , ,  Hamazi, , 
Durham,  Hamidiya, , , 
Hamrin Valley, , , , , , , , , ,
Ebla, , , , ,  , 
Egypt, , ,  Hamun, , , 
Ekallatum, ,  Hana, 
Elam, , , , , , , , , , , , , Harshi, , , 
, , , , , , , , , , Harunabad, see Islamabad-e Gharb
, , , , , , , , , , Hasanlu, 
, , , , , , , , , , Hatti, 
, , , , , , , , , , Hawiza (marshes), , , , , , ,
, ,  , , 
Elammat,  Hidali, , 
Elanir,  Hilmand, 
Elašir, see Elanir Hilmu, 
Ellipi, ,  Hindanu, 
Emar, , , , , , ,  Hiritum, , 
Eshnunna, , , , , , , , , , , , Hit, , , , 
, , , , , , , , , , Hubutish, 
, , , , , , , , , , Hudhud River, , 
, ,  Hudimiri, 
Euphrates, , , , , , , , , , Huhnur(i), , , , 
,  Hulilan, 
Hung-i Nauruzi-i, 
Fahliyan, , , , ,  Hursagkalama, 
Failaka, , , ,  Hurti/um, , , , , , , , ,
Farhardgerd,  , , 
Fars, , , , , , , , , , , Husainabad, see Ilam
, , , , , , , , , 
Fasa,  Iapru, 
Fellahiyah, see Shadegan Ida-maras, , 
Firuzabad,  ˙ , , 
Idide River,
Ilam, , , 
Gambulu,  Ilan-sura, 
Gas(s)ur, see Nuzi Imar,˙
 index of topographical names

Indus Valley, ,  Kinunir, 

Irbil,  Kish, , , , 
Irgidu,  Kismar, 
Irkume,  Konar Sandal, , , , , 
Irnina River,  Kufa, 
Isfahan,  Kuh-e Dasht, 
Isin, , , , , , , ,  Kuh-e Pataweh, 
Ishchali, see Neribtum Kuhgiluyeh-Boraihmed, , 
Islamabad-e Gharb, , , , ,  Kuh-i Alwand, 
Izeh, , , , , , , , , , Kuh-i Damergil, 
,  Kuh-i Manisht, 
Kul-e Farah, , , , , , , ,
Ja Nishin, , ,  , , , , , , , , ,
Jajrood,  
Jalalabad, , ,  Kummama, 
Jazira, , ,  Kupal River, , , , 
Jarrahi River, , , , , , , , Kur River, , , , , , , , 
,  Kurangun, , , , , , , , ,
Jebel Aruda,  , , , , , , , , , ,
Jebel Bishri,  , , 
Jebel Hamrin, , ,  Kurda, , 
Jemdet Nasr,  Kutha, 
Jinjun, , 
Jiroft,  Lagash, , , , , , , , ,

Kabir Kuh,  Lahiru, 
Kalhu, , ,  Lama, 
Kalleh Nissar, , ,  Lara, 
Kar Tukulti Ninurta, ,  Larsa, , , , , , , , , , ,
Karaj,  , , , , , 
Karind,  Levant, 
Karkheh River, , , , , , , , Liyan, , , , , , 
, , , , , ,  Lordegan, 
Karkh Maisan, see Spasinou Charax Lower Sea, 
Karun River, , , , , , , , Lower Zab River, , , , 
, , , , , , , , , , Lullu, 
,  Lullubi/um, , , , , , , , 
Kashan,  Luristan, , , , , , , 
Kazalu,  Lut, 
Kazerun, 
Kerman, , , , ,  Madaktu, , , , , , 
Kermanshah, , , , , ,  Madga, see Hit
Khabur,  Mahidasht, 
Khafajah, , , ,  Mahmiti, 
Khak-i Rustam,  Makhoul, 
Kheit Qasim,  Makkan, , , 
Khor Musa tidal inlet, , , , ,  Makran, 
Khorramshahr, ,  Malamir, , , , , , , 
Khorsabad, see Dur-Sharrukin Malayer, 
Khor Zubair tidal inlet,  Malgium, , 
Khubur River,  Mamasani, , , , , , , , ,
Khuzestan, , , , , , , , , , , , , , , , , 
, , , , , , , , , , Manisht Kuh, see Kuh-i Manisht
, , , , , , , , , , Mankisum, , 
, , , , , , , , , , Marat River, 
, ,  Marda, 
Kimash, , , , , , , , , Margiana, , , 
, , , , ,  Marhashi, , , , , , 
 index of topographical names 

Mari, , , , , , , , , , , Pusht-i Kuh, , , 
, , , , , , , , , , Puqudu, , 
, , , 
Marrish,  Qalat al-Bahrain, 
Martu,  Qaleh Kalli, , 
Marubishtu,  Qaleh Rostam, 
Marun anticline, ,  Qaleh Surkheh, see Surkhegan
Marv Dasht, , , ,  Qaleh-ye Tol, 
Mashhat-sharri, see Mashkan-sharrum Qatna, , , , 
Mashkan-sharrum,  ˙ 
Qom,
Matezzish, 
Medina, see Dawraq Rag-e Safid anticline, 
Mehran, , ,  Ram Hormuz, , , , , , 
Meluhha,  Rannakarra, 
Mesopotamia, , , , , , , , , , , Rashi, 
, , , , , , , , , , , , *Rašnuca-, 
, , , , , , , , , , *Rašnuvatı̄š, 
, , , , , , , , , , Ray, 
, ,  Razamâ, 
Meturan,  Reshahr, 
Mianab, ,  Rome, 
Mitanni, , ,  Rud-e Fahliyan, 
Murubisu,  Rumishan, 
Musasir, 
Musiyan, , ,  Sa’adiya, 
Nahavand,  Sabum, , 
Nahr Bahre, ,  Sabzevar, 
Nahr Hashem,  Samara, 
Nahr Maleh,  Samati, , 
Nahr Tira, ,  Sapiratum, 
Naisan, see Spasinou Charax Sar, 
Naqsh-e Rustam, , , , ,  Sar-i-Pol, 
Nasiriya,  Sar-i-Pul-i-Zohab, 
Neribtum, , , , , , , ,  Sarab-e Bahram, 
Nigin,  Sealand, , 
Nimrud, see Kalhu Seleucia-on-the-Hedyphon, 
Nineveh, , , , ,  Shadegan (marshes), , , , , 
Nippur, , , , , , , , ,  Shaduppum, , , , 
Nugu"u,  Shah Shavar, 
Nurabad-e Mamasani, , ,  Shahabad(-e Gharb), see Islamabad-e Gharb
Nuristan,  Shahdad, , , , , , , , , , , , ,
Nuzi, , , , , , , , , , , 
, , , , , ,  Shahr-i Kord, 
Shahr-i Sokhta, , , 
Oman, see Makkan Shalukku, 
Shashila, 
Pai-i Taq,  Shatt al-Arab, , , , , , , ,
Parsumash,  , 
Pasargadae,  Shatt al-Haffar, 
Pashime, ,  Shatt Bamishir, , 
Persepolis, , , , , , , , , Shaushanush, 
, , , ,  Shehna, 
Persia,  Shekaft-e Salman, , , , , , ,
Persian Gulf, , , , , , , , , , , , 
, , , , ,  Shemshara, , , 
Pillatu,  Shigrish, 
Pol-e Pirim,  Shimashki, , , , , , , , , ,
Pul-i Zal,  , , , , , , , , , 
 index of topographical names

Shimurrum, see Simurrum Tal-e Mushki, , 

Shinear,  Tal-e Nakhodi, 
Shiraz, , , ,  Tal-e Qaleh, 
Shishil,  Tal-e Shoga, , 
Shour,  Tal-e Teimuran, , 
Shubat-Enlil, ,  Tang-e Shib, 
Shumurtan-Duri,  Tappeh Deylam, , 
Shunuhum,  Tappeh Dozak, 
Shurbu,  Tappeh Gotvand, 
Shusharisungur,  Tappeh Horreeye, 
Shushtar, , ,  Tappeh Sorna, 
Sicily,  Taq-i Girreh, 
Simashki, see Shimashki Targibatu, 
Simurrum, , , , , , ,  Tarhan, 
Sippar, , , , , ,  Tartin, 
Sippar-Amnanum, , , , , ,  Tartin-Ammak, 
Sippar-Jahrurum, , ,  Tawuq, 
Sippar-rabum,  Tchoga Zanbil, see Dur-Untash
Sirwan River,  Tell Abqa, 
Spasinou Charax, , , ,  Tell Abraq, 
*Srauša,  Tell Agrab, 
Subartu, , ,  Tell Ahmar, see Til Barsip
Suhûm, , ,  Tell Asmar, see Eshnunna
Sulaimaniya(h), , ,  Tell Aqar, 
Sumer, ,  Tell Bazi, , 
Surkh Dum,  Tell Bi"a, see Tuttul
Surkhegan,  Tell Brak, , 
Tell ed-Dēr, see Sippar-Amnanum
Susa, , , , , , , , , , , , , ,
Tell Deymam, 
, , , , , , , , , , , , ,
, , , , , , , , , , , Tell Dhiba"i, 
, , , , , , , , , , Tell Fecheriyeh, , , 
, , , , , , , , , , Tell Gubbah, 
, , , , , , , , , , Tell Halawa, 
, , , , , , , , , , Tell Hamidiya, 
, , , , , , , , , , Tell Harmal, see Shaduppum
, , , , , , , , , , Tell Hassan, , , 
, , , , , , , , , , Tell Maddhur, 
, , , , , , , , , , Tell Meskene, see Emar
, , , , , , , , , , Tell Razuk, 
, , , , , , , , , , Tell Rimah, 
, , , , , , , , , , Tell Suleimeh, see Awal
 Tall Tendy, , , 
Susangerd, ,  Tell Tābān, see Tābētu
Susiana, , , , , , , , , , , ˙
Tell Yelkhi, ˙ , 
, ,
, , , , , , , , , , Telloh, , , , , , 
,  Tepe Bormi, , 
Syria, , , , , ,  Tepe Djamshidi, , 
Tepe Farukhabad, , , , 
Takab, ,  Tepe Giyan, , , , 
Tal-e Bakun,  Tepe Hissar, 
Tal-e Ghazir, ,  Tepe Sharafabad, , 
Tal-e Iblis,  Tepe Sialk, , , , 
Tal-e Jari, ,  Tepe Sofalin, , , , , , , 
Tal-e Kaftari,  Tepe Sohz, 
Tal-e Kamin,  Tepe Surkhegan, , 
Tal-e Malyan, , , , , , , , , , Tepe Survan, 
, , , , , , , , , , Tepe Yahya, , , , , , , , , , , , 
, , , ,  Terqa, , 
 index of topographical names 

Tigris, , , , , , , , , , , Uruk, , , , , , , , , , ,
, , , ,  , , , , , , , 
Til Barsip,  Urusagrig, , , 
Till Tuba, 
Tilmun, see Dilmun Varamin, 
Tôd, , , ,  Veshnoveh, 
Tol-e Afghani, , 
Tol-e Bashi,  Yagheh Sangar, 
Tol-e Bormi,  Yamhad, 
Tol-e Nurabad, , , , , , , , Yamutbal, 
,  Yasuj, , , 
Tol-e Peytul, , , 
Tol-e Spid, , , , , , , , , Xong-e Azdar, 
, , , , , , , , ,
 Zabshali, 
Tupliash, ,  Zagros Mountains, , , , , , , , ,
Turukkum,  , , , , , , , , , ,
Tut(t)ul, , , ,  , , , , , 
Tutub,  Zanu, 
Tābētu, , , , ,  Zariq, 
˙ Zidahri, 
Udda-zamin,  Zidanum, , , 
Udman,  Zila-Humban, 
Ulai River, ,  Zimudar, , 
Ur, , , , , , , , , ,  Zippa, 
Urshu,  Zitian, see Zidanum
Urua, , ,  Zohreh River, , , 
 INDEX OF TEXT CITATIONS

--, ,  ARM XXVI/ , 

ARM XXVI/ , 
A ,  ARM XXVI/ , 
A ,  ARM XXVI/ , 
A , ,  ARM XXVI/ , 
A ,  ARM XXVII , 
A ,  ARM XXVIII , 
A ,  ARM XXVIII , 
A ,  ARM XXVIII , , , 
ABL , , ,  Ass , see RIMA  A...
ABL , , , ,  Assurbanipal Inscription Temple of Ishtar, ,
ABL , ,  
ABL ,  Assurbanipal Letter to the god Assur, , ,
ABL , , ,  , , 
ABL , , ,  Assurbanipal Prism A, , , , , ,
ABL , ,  
ABL , ,  Assurbanipal Prism B, 
ABL +,  Assurbanipal Prism C, , , 
ABL ,  Assurbanipal Prism F, , , , , 
ABL , , ,  Assurbanipal Prism G, , , , 
ABL ,  Assurbanipal Prism H, , , , 
ABL , , ,  Assurbanipal Prism Kh, , , , 
ABL ,  Assurbanipal Prism T, , , 
ABL ,  AUCT  , 
ABL ,  AUCT  , 
ABL ,  AUCT  , 
ABL , ,  AUCT  , 
A father and his perverse son,  AUCT  , , 
Amarna Correspondence,  AUCT  , 
An = Anum vi , 
An = Anum vi ,  BA  , 
André/Salvini  Sb  // Sb ,  BA  , 
Anzu Myth,  Babylonian King List A, 
AO ,  balağ composition, 
AO ,  BBSt , 
ARM II , BBSt , 
ARM IV ,  BBVOT  , 
ARM X ,  BDTNS no. , 
ARM XIV ,  BE / , 
ARM XXIII ,  BE / , 
ARM XXIII –,  BE / , 
ARM XXIII ,  BE / , 
ARM XXV ,  BM , 
ARM XXV ,  BM A, 
ARM XXV ,  BM , 
ARM XXV ,  BM ++, 
ARM XXV ,  BM , 
ARM XXVI/ ,  BM , 
ARM XXVI/ ,  BM , 
ARM XXVI/ ,  BM , 
ARM XXVI/ ,  BM , 
ARM XXVI/ ,  BM , , 
 index of text citations 

BM ,  DPh, 

BM ,  DSaa, 
BM ,  DSz, 
BM ,  DT , 
BM ,  Dumuzi’s Dream, 
BM , , 
BM ,  EKI , 
BM ,  EKI , , , 
BM ,  EKI , 
BM ,  EKI , , 
BM ,  EKI , , 
BM ,  EKI , , 
BM A,  EKI , 
BM , , , ,  EKI , 
BM ,  EKI , 
BM ,  EKI , , 
BPOA  ,  EKI , 
BPOA  , ,  EKI , , 
BPOA  ,  EKI , , 
BPOA  ,  EKI , 
Braun-Holzinger G , see CBS  EKI , , , , 
Braun-Holzinger G , see CBS  EKI , 
BWL Pl.,  EKI , , , , , , , , ,
, 
CBS , ,  EKI , , 
CBS , ,  EKI , 
CBS , ,  EKI , , , , , 
CDAFI  ,  EKI , 
CDAFI  ,  EKI , , 
CDAFI  ,  EKI , 
CDAFI  ,  EKI , 
Chronicle P, , , , ,  EKI , 
Chronicle W,  EKI , 
CIRPL , Ean. ,  EKI , , 
Cmc,  EKI , 
Codex Hammurabi,  EKI , 
Codex Hammurabi Col.  :–,  EKI , 
CT  Sm+,  EKI , , 
CT  c,  EKI , , 
CT  a,  Emar / , 
CT  b,  Enlil and Sud, 
CT  a,  Erra Epos, 
CT  c,  ES , see RIMA  A...
CT  ,  ˙ Epic, 
Etana
CT  , 
CT  ,  FM VII , , , , , 
Cuthean Legend of Narām-Sîn ,  FM VII , 
Curse of Agade, ,  FM IX , 
CUSAS  ,  Fort. , see JNES  
Cylindroïde Atta-hušu, , 
Cyrus Cylinder,  Gelb/Kienast  Elam , , , , 
Gelb/Kienast  Elam , , 
Date List B, ,  Gelb/Kienast  Elam , , 
DB,  Gelb/Kienast  Elam , , 
Debate between bird and fish,  Gelb/Kienast  Elam , , , 
Debate between date palm and tamarisk,  Gelb/Kienast  Elam , 
Debate between sheep and grain,  Gelb/Kienast  Elam , 
Di , ,  Gelb/Kienast  Elam , , 
 index of text citations

Gelb/Kienast  Elam ,  IrAnt  nr. , 

Gelb/Kienast  Elam ,  IRS , , 
Gilgameš Epos,  IRS , 
Gilgameš Epos VII –,  IRS , 
Gilgameš Epos XI ,  IRS , 
Gilgameš Epos XI –,  IRS , 
Gilgameš Epos XII ,  IRS , 
Gilgameš Epos XII ,  IRS , 
Gilgameš Epos XII ,  IRS , 
Gilgameš Epos XII ,  IRS , 
Gilgameš Epos XII ,  IRS , 
Gilgameš Epos XII a ,  IRS , 
Gilgameš Epos XII i ,  IRS , 
Gilgameš Epos XII n ,  IRS , 
Gilgameš Epos XII s ,  IRS , 
Gilgameš Epos XII x –,  IRS , 
Gilgameš Epos XII y –,  IRS , 
Gilgameš and Huwawa A,  IRS , 
Glassner  no.,  IRS , 
Glassner  no.,  Ištar’s Descent to the Netherworld –, 
Glassner  no. , , ,  ITT  , 
Glassner  no. ,  ITT  , 
GTZ , 
Gudea Cylinder A,  JBVO  , 
Gudea Statue B, , , ,  JCS  , 
JNES  , 
Herrero  no. ,  Job , , 
Herrero/Glassner  text , 
Herrero/Glassner  text ,  K , see CT  
Herrero/Glassner  text , ,  K , 
Herrero/Glassner  text ,  K , 
Herrero/Glassner  text ,  K , 
Herrero/Glassner  text ,  K , 
Herrero/Glassner  text ,  K , 
Herrero/Glassner  text ,  Keder-Laomer Texts, , , 
Herrero/Glassner  text ,  Keš Temple Hymn, 
Herrero/Glassner  text ,  King list of Awan and Šimaški, 
Herrero/Glassner  text ,  Kodex Urnamma –, 
Hirose ,  Königsliste A, 
HT , see Herrero/Glassner  text 
HT , see Herrero/Glassner  text  LAPO  , 
HT , see Herrero/Glassner  text  LAPO  , 
HT ,  LAPO  , 
HT ,  LAPO  , 
HT , see Herrero/Glassner  text  LAPO  , see ARM IV 
HT , see Herrero/Glassner  text  LAPO  , 
HT , see Herrero/Glassner  text  LAPO  , see ARM X 
HT , see Herrero/Glassner  text  Lau , 
HT , see Herrero/Glassner  text  Lipit-Ištar B, 
LKA , 
Inanna’s Descent to the Netherworld, , ,
 M , 
Inanna and Ebih,  M , 
Inanna Hymn C,  M-, , , , , , 
Instruction of Šuruppag,  M-, , , , , , 
Iqqur ı̄puš,  M-, 
IrAnt   no. ,  M-, 
IrAnt  –, ,  M-, , 
 index of text citations 

M-,  MDP  , 

Magic Text,  MDP  , 
Malbran-Labat  no ,  MDP  , 
Malbran-Labat  no ,  MDP  , 
Malbran-Labat  no ,  MDP  , 
Malbran-Labat  no ,  MDP  , , 
Malku vi,  MDP  , 
Maul  text ,  MDP  , 
MDP  –,  MDP  , , 
MDP  ,  MDP  , , 
MDP  ,  MDP  –, 
MDP  ,  MDP  , 
MDP  ,  MDP  , 
MDP  ,  MDP  –, see Gelb/Kienast  Elam 
MDP  ,  MDP  , 
MDP  , ,  MDP  , 
MDP  ,  MDP  , 
MDP  ,  MDP  , , 
MDP  ,  MDP  , 
MDP  ,  MDP  , 
MDP  ,  MDP  , 
MDP  ,  MDP  , 
MDP  ,  MDP  , 
MDP  ,  MDP  , 
MDP  ,  MDP  , 
MDP  , ,  MDP  , 
MDP  ,  MDP  , , 
MDP  ,  MDP  , , , , 
MDP  ,  MDP  , , , , 
MDP  ,  MDP  , , 
MDP  ,  MDP  , , , 
MDP  ,  MDP  , , 
MDP  ,  MDP  , , , 
MDP  ,  MDP  , , , , , 
MDP  ,  MDP  , , 
MDP  ,  MDP  , , , , 
MDP  ,  MDP  , , , , 
MDP  ,  MDP  , 
MDP  ,  MDP  , 
MDP  ,  MDP  , 
MDP  , ,  MDP  , , , 
MDP  ,  MDP  , 
MDP  ,  MDP  , 
MDP  ,  MDP  , 
MDP  ,  MDP  , , 
MDP  ,  MDP  , 
MDP  ,  MDP  , 
MDP  ,  MDP  , , , 
MDP  ,  MDP  , , , , 
MDP  ,  MDP  , , , , 
MDP  ,  MDP  , , 
MDP  ,  MDP  , 
MDP  ,  MDP  , 
MDP  ,  MDP  , , 
MDP  ,  MDP  , , 
MDP  ,  MDP  , , 
MDP  ,  MDP  , , , , , 
MDP  ,  MDP  , , 
 index of text citations

MDP  , ,  MDP  , , 

MDP  , ,  MDP  , 
MDP  , ,  MHET II , 
MDP  , ,  MHET II , 
MDP  , ,  MHET II , 
MDP  ,  MHET II , 
MDP  ,  MHET II , , 
MDP  ,  MHET II , 
MDP  , , ,  MHET II , 
MDP  ,  MLC , 
MDP  ,  MS , 
MDP  ,  MSL  Aa, 
MDP  , , , ,  MSL  , see Proto-Ea 
MDP  ,  MSL  , see Proto-Diri Oxford 
MDP  ,  MVN  , , 
MDP  ,  MVN  , 
MDP  , , ,  MVN  , 
MDP  ,  MVN  , 
MDP  , , , , , ,  MVN  , 
MDP  , ,  MVN  , 
MDP  ,  MVN  , 
MDP  ,  MVN  , 
MDP  ,  MVN  , 
MDP  ,  MVN  , 
MDP  , , , , , , ,  MVN  , 
MDP  ,  MVN  , 
MDP  ,  MVN  , 
MDP  ,  MVN  , 
MDP  ,  MVN  , 
MDP  ,  MVN  , 
MDP  ,  MVN  , 
MDP  ,  MVN  , 
MDP  ,  MVN  , , 
MDP  ,  MVN  , 
MDP  ,  MVN  , , 
MDP  , 
MDP  ,  Nergal and Ereškigal, 
MDP  ,  Nergal and Ereškigal ii ’–iii ’, 
MDP  ,  Nesbit  , 
MDP  ,  NG , 
MDP  , ,  Nik  , 
MDP  ,  Nineveh Letters, 
MDP  –,  Nin , 
MDP  ,  Nin , 
MDP  ,  Nisaba A, 
MDP  A,  NISABA  , 
MDP  B,  NISABA  , 
MDP  ,  NISABA  , 
MDP  no. ,  NISABA  , 
MDP  no. ,  NISABA  , 
MDP  no. ,  NISABA  , 
MDP  no. , 
MDP  no. ,  OIP  , 
MDP  no. ,  OIP  , , 
MDP  ,  OIP  , 
MDP  ,  OLA  , 
MDP  ,  Omen Text, 
MDP  ,  Orientalia  –, , , 
 index of text citations 

PBS / ,  PF , 

PDT  ,  PF , 
PDT  ,  PF , 
PDT  ,  PF , 
PDT  ,  PF-NN , 
Persepolis Fortification Texts, , , , , PF-NN , 
, ,  PF-NN , 
PF ,  PF-NN , 
PF ,  PF-NN , 
PF ,  PF-NN , 
PF , ,  PF-NN , 
PF ,  PF-NN , 
PF ,  PF-NN , 
PF ,  Prism Annal A, 
PF ,  Prism Annal F, 
PF ,  Prism Annal T, 
PF ,  Proto-Diri Oxford , 
PF ,  Proto Ea, 
PF ,  Proto-Ea , 
PF ,  Proverbs, , 
PF ,  Proverbs , , 
PF ,  Proverbs , , 
PF , 
PF ,  RA  , 
PF ,  Racc , 
PF ,  RIMA  A..., 
PF ,  RIMA  A.., 
PF ,  RIMA  A.., 
PF ,  RIMA  A..–, 
PF ,  RIMA  A.., 
PF ,  RIMA  A.., 
PF ,  RIMA  Tukulti-Ninurta , 
PF ,  RIMA  ilušuma , 
PF ,  RIMB  .., 
PF ,  RIME  p. , 
PF ,  RIME / Ibbi-Sîn , 
PF ,  RIME / Ibbi-Sîn , 
PF ,  RIME / Ur-Namma , 
PF ,  RIME / Ur-Namma , 
PF ,  RIME / Ur-Namma , , 
PF ,  RIME / Ur-Namma , 
PF ,  RIME / Ur-Namma , 
PF ,  RIME / Ur-Namma , 
PF ,  RIME / Šulgi , 
PF ,  RIME / Šu-Sîn , 
PF ,  RIME / Šu-Sîn , 
PF ,  RIME / Rimuš , 
PF ,  RIME / Rimuš , 
PF ,  RIME / no. , see UET  
PF ,  RIME / no. , see CBS 
PF ,  RIME / no. , see CBS 
PF ,  RIME / no. , see CBS 
PF ,  RIME  .., 
PF ,  RIME  p. , 
PF ,  Royal Correspondence : Puzur-Šulgi to Ibbi-Sîn,
PF ,  
PF ,  Royal Correspondence : Šulgi to Aradmu, 
PF ,  Royal Correspondence : Šulgi to Išbi-Erra, 
 index of text citations

Royal Correspondence : Šulgi to Puzur-Šulgi,  TS.XI., 

RTC ,  TS.XII., 
RTC ,  TS.XII., 
RTC ,  TS.XII., 
RTC ,  TS.XII., 
RTC ,  TS.XII., 
RTC ,  TS.XV., 
RTC ,  TS.XV., 
RTC ,  TS.B., 
RTC ,  TRU , 
RTC , ,  TTM I , 
RTC ,  TTM I , 
RTC ,  TTM I , 
RTC ,  TTM I , 
RTC ,  TTM I , 
RTC , 
RTC ,  UDT , 
UET  , 
S,  UET  , 
Sargon Geography,  UET  , 
SAT  ,  UIOM , 
SAT  ,  Urnamma A, , 
SAT  ,  Ur-Nammu’s Death, 
Sb , 
Sigrist  ,  VA Bab. , 
Sigrist  ,  Vendidad, 
SMM , see BDTNS no.  Vision of the Netherworld, , 
Sollberger/Kupper  IVOa,  VS  , 
Sumerian King List,  VS  , , 
Susa Acropole Texts, ,  VS  , 
Syllabary A, ,  VS  , 
Syllabe Alphabet A,  VS  , , 
Synchronistic Chronicle, 
Šulgi A, ,  Yasna , 
Šurpu,  Yasna , 
Šurpu –,  Yasna , 
Šurpu Comm. B,  Yašt , 
Yašt , 
T III-–, see Maul  text  YBC , , 
TCL  ,  YOS  , 
TCL  ,  YOS  , 
TCL  A,  YOS  , 
Torino  , , ,  YOS  , 