Democracy's Ancient Ancestors: Mari and Early Collective Governance

This page intentionally left blank Democracy’s Ancient Ancestors Mari and Early Collective Governance Democracy’s Anc...

Author: Daniel E. Fleming

170 downloads 899 Views 3MB Size Report

This content was uploaded by our users and we assume good faith they have the permission to share this book. If you own the copyright to this book and it is wrongfully on our website, we offer a simple DMCA procedure to remove your content from our site. Start by pressing the button below!
Report copyright / DMCA form

DOWNLOAD PDF

This page intentionally left blank

Democracy’s Ancient Ancestors Mari and Early Collective Governance Democracy’s Ancient Ancestors examines the political landscape of the ancient Near East through the archive of more than 3,000 letters found in the royal palace of Mari. These letters display a rich diversity of political actors, encompassing major kingdoms, smaller states, and various tribal towns. Mari’s unique contribution to the ancient evidence is its view of tribal organization, made possible especially by the fact that its king, Zimri-Lim, was first of all a tribal ruler, who claimed Mari as an administrative base and source of prestige. These archaic political traditions are not essentially unlike the forms of predemocratic Greece, and they offer fresh reason to recognize a cultural continuity between the classical world of the Aegean and the older Near East. This book bridges several areas of interest, including archaeology, ancient and classical history, early Middle and Near East history, and political and social history. Daniel E. Fleming is Associate Professor of Hebrew Bible and Assyriology at New York University’s Skirball Department of Hebrew and Judaic Studies. He is a member of the American Oriental Society, American Schools of Oriental Research, and the Society of Biblical Literature.

Democracy’s Ancient Ancestors Mari and Early Collective Governance

DANIEL E. FLEMING New York University

cambridge university press Cambridge, New York, Melbourne, Madrid, Cape Town, Singapore, São Paulo Cambridge University Press The Edinburgh Building, Cambridge cb2 2ru, UK Published in the United States of America by Cambridge University Press, New York www.cambridge.org Information on this title: www.cambridge.org/9780521828857 © Daniel E. Fleming 2004 This publication is in copyright. Subject to statutory exception and to the provision of relevant collective licensing agreements, no reproduction of any part may take place without the written permission of Cambridge University Press. First published in print format 2004 isbn-13 isbn-10

978-0-511-16515-3 eBook (EBL) 0-511-16515-3 eBook (EBL)

isbn-13 isbn-10

978-0-521-82885-7 hardback 0-521-82885-6 hardback

Cambridge University Press has no responsibility for the persistence or accuracy of urls for external or third-party internet websites referred to in this publication, and does not guarantee that any content on such websites is, or will remain, accurate or appropriate.

To my parents, with love and gratitude

Contents

page ix xi xvii xxi

List of Maps and Tables Preface Acknowledgments List of Abbreviations 1

2

3

Introduction A. The Mari Texts B. A Survey of Mari History C. A Note on Prominent Terms D. The Mari Archives and Political History E. A Text-Based Study: Comments on Methodology The Tribal World of Zimri-Lim A. Tribally Organized Pastoralists and the Amorrites B. The Primary Constituents of the Confederacies: Sim alite gayum and Yaminite limum C. The Local Leader of Tribe and Town: The sug¯agum in Service to the Mari Kingdom D. The Chief of Pasture: The merhûm ˘ E. The “Hana” Tent-Dwellers ˘ F. The Other Confederacy: The Yaminites The Archaic State and the m¯atum “Land” A. Urbanism and Archaic States B. The m¯atum: The Basic Unit of Regional Politics in the Early Second Millennium C. Subdividing the Major m¯atums: The hals.um District ˘ D. Population Terminology Not Tied to Political Entity E. Zimri-Lim and the Land of the Tent-Dwellers (m¯at Hana) ˘ vii

1 1 6 13 14 20 24 26 43 63 76 85 92 104 106 116 133 139 147

Contents

viii

4

5

The Collective and the Town A. The Towns of the Mari Archives B. Corporate Political Tradition C. The Collective Face of Towns or Lands D. Elders E. Heads F. Words for Assembly G. Imar, Tuttul, and Urgiˇs: Old Towns with Strong Collective Traditions H. Mari in Third-Millennium Mesopotamia I. On Explaining Corporate Power Conclusions A. The Political World of the Mari Archives B. Democracy’s Ancient Ancestors

170 172 174 180 190 200 203 211 219 222 229 229 235

Notes Glossary of Ancient Terms Glossary of Proper Names Bibliography Subject Index

243 309 315 325 345

Index of Mari Texts

351

Maps and Tables

maps 1 Ancient Syria and Mesopotamia 2 The Mari Region 3 The Habur River Basin ˘

page xxv xxvi xxvii

tables 1 Towns with Individual sug¯agum Leaders and No Known Tribal Affiliation 2 Letters Sent by Yaminites (List Not Complete) 3 m¯atums Defined by Central Capital 4 m¯atums Defined by Population 5 References to m¯atum Coalitions 6 References to the Collective Kings of m¯atum Coalitions 7 Activities, People, and Materials Organized by hals.um ˘ 8 Towns Identified as Royal Possessions 9 Adjectival Forms of Geographical Names in Individual Identifications 10 Adjectival Forms of Geographical Names in Group Identifications

ix

52 59 122 122 125 125 136 173 182 183

Preface

Ancient Mesopotamia is famous for its kings. Sargon of Agade is said to have built the first empire. Hammurabi of Babylon showed off his authority in a collection of standard˘ law. The shadow of later Assyrian and Babylonian monarchs darkens the prospects of Israel and Judah in the biblical tradition. Karl Marx’s “oriental despotism” began in the ancient world. Democracy, in contrast, belongs to Greece, a world away, facing west toward Europe with its back to the east. Here we find the roots of the political system that Francis Fukuyama placed at “the end of history,” the system that is the life-breath of all modern academic pursuits (Fukuyama 1992). When we scholars study ancient Greece, we study ourselves. When we study ancient Mesopotamia, we explore the “other,” all ethnic identity aside. Reality, as always, resists the tug of our impulse to categorize. In the case of ancient Mesopotamia and the larger Near East, the reputation for authoritarian monarchy has been transmitted to us through Athens and Israel, the two main conduits that carried eastern Mediterranean ideas into western Europe. Both of these faced eastern empires in the crucial periods of their classical writings, Achaemenid “Persia” against Athens, after Assyria and Babylon had dismantled Israel. The Near East did indeed produce a variety of powerful centralized kingdoms, but it is not true that individualizing, authoritarian rule was specially characteristic of Near Eastern political life. Specialists have long recognized the participation of groups in Near Eastern politics at various levels, though these are often relegated to the margins of real power. Closer inspection shows that collective political decision making could be an essential aspect of Near Eastern governance. Outside of the largest kingdoms, political life was much less predictably centralized, with many possible equilibria between various political powers. Two particular locations of strong collective traditions were the town and the tribe, two distinct modes of identity with overlapping use. In much of the Near East, individual rule existed only in dynamic tension with a range of other individual and collective leadership. xi

xii

Preface

Autocratic monarchy dominated the scene more and more with the success of certain large kingdoms, but these should not be considered the essential form of Near Eastern political life. In fact, the collective component of Near Eastern politics appears to be very old and remarkably persistent, deserving more attention than it has attracted until now. We cannot sketch early political history as a development of increasingly complex societies in which individual leaders established ever more effective central control, until the Greeks introduced a radically different system with no relationship to what came before. The political world before democracy is therefore remarkably diverse, with a range of contributing elements that are not so obviously unlike the various constituents of the Greek scene before the process that led to democracy. In Syria-Mesopotamia, the region of the northern Fertile Crescent, the collective political character of the “town” provides an especially interesting backdrop to the development of Athenian democracy around the unit called the polis, again a “town.” It is not my goal in this book to account for the origins of Greek democracy or even to pin down the relationship between predemocratic Greek society and regions to the east. I do, however, find that in discussion of Greece before democracy, the eastern Mediterranean, Anatolia, and the Fertile Crescent are separated from the Aegean in a way that strikes me as artificial and improbable in geographical terms alone. Greece and Mesopotamia are too easily regarded as polar political opposites, one capable of giving rise to democracy and the other constitutionally prone to autocracy. The Near Eastern evidence alone proves this a caricature. When these two worlds are regarded as diametrically opposed, both are immediately misunderstood. Something radically new unquestionably appeared in Athens with the advent of democracy, and its radical novelty calls for explanation by uniquely Greek circumstances. At the same time, however, Greece need not be treated as a separate cultural continent, which shares traits with the Near East only by importation. Surely, what is uniquely Greek or Aegean was embedded in a cultural matrix that had no absolute demarcations between adjacent regions. Syria-Mesopotamia offers us the particular gift of some of the earliest known writing, and this writing allows an inside view of ancient political life in this part of the world. This writing shows us a complex interaction of many different social and political players, including large entities ruled by kings, alliances that acted as a single polity, tribal groups of varying scale and character, and the unit centered at a single settled site called the “town.” Both individual and group leadership are found in every one of these settings, with no single configuration. As our modern world struggles with the invitation to let democracy spread into every political culture, surely it would be thought-provoking to rediscover underlying continuities between the Greek invention and earlier traditions of collective decision making.

Preface

xiii

There is no question that in many ancient Near Eastern settings, powerful monarchies had reduced other traditional political participants to relatively minor roles, subordinate to an administrative structure that served the individual authority of kings. The evidence for these other traditions is more abundant than generally realized, but during the time of early writing, they almost never existed as a wholly separate alternative to kingship. Even where kings were weak or distant, other political players had to cope with their claims and their power. Because of this interaction, it would be futile to investigate ancient collective political life in isolation from the individual rule that most often occupied the highest rung of any regional political ladder. The role of the group can be understood only as part of a system that usually involves kings. In practical terms, this need to think in terms of a political and social system shapes the entire definition of this book. I am interested in the entire phenomenon of group decision making in a world before democracy, but I have chosen to examine this phenomenon from the vantage of the one large archive from a city called Mari. For much of the third millennium b.c.e., Mari dominated the part of the Euphrates River valley just inside Syria’s modern border with Iraq. Mari’s archives of clay tablets with “wedgeshaped” cuneiform writing come almost entirely from a single brief period just before the site was destroyed, most often dated to the early eighteenth century.1 Although Mari served as the royal seat of a modest kingdom, its collection of more than 3,000 letters offers an amazing vista on the whole SyriaMesopotamian landscape.2 Moreover, Mari’s last king, named Zimri-Lim, was a tribal ruler who survived through a carefully cultivated web of alliances with an array of odd bedfellows. These included tribal peoples from both his own and other groups, as well as many small polities in the region north of Mari. His political contacts reached in every direction and involved every type of leadership. Among these, we find a variety of collective forms. In this network of political interconnections, not kingship, tribes, nor towns can be understood except as they relate to the whole. If it is a system that must be understood, and not the isolated evidence for elders, assemblies, and the like, then the question is how to define the bounds of the system in view. Here there is a real advantage to the coherence of a single archive. With the documentation from Mari, we are looking at interlocking political traditions of a single period, across connected space. The data themselves therefore provide us a mandate to explain them not as isolated signals of interesting phenomena but as part of an integrated system. Towns and villages were incorporated into larger entities that were not defined in town terms. How are we to understand how the royal administration at Mari or at other centers in the kingdom, such as Terqa or Saggaratum, related to whatever political traditions accrued to each town as such? Zimri-Lim, the last king of Mari and the recipient of most of the correspondence found there, proclaimed himself “king of Mari and the land

xiv

Preface

of the Hana.” The first title defines him by the walled city that he made his seat of ˘power, and the second associates him with mobile pastoralists who identify themselves by tribe. With such connections in mind, I have incorporated two other major political phenomena into this study, as a framework for understanding the collective traditions of towns. I begin in Chapter 2 with the identification of people by tribal categories, a custom that reaches every level of society, from rulers to villagers. Such identification by kinship may have been common to all early Mesopotamian society, urban and not, but in the Mari archives we find our richest evidence for large groups spread across considerable territory, whose primary political allegiance may be defined by tribe. These commitments then shape the political life of towns and kingdoms, according to the traditions of the dominant tribal population. Zimri-Lim himself was widely considered to be a tribal king. Having addressed the tribe, I turn in Chapter 3 to the categories that define the Mari kingdom, which may fairly be called an archaic state. In the Akkadian language of the letters found at Mari, the highest unit of political organization was the m¯atum, or “land.” The largest Mesopotamian kingdoms were “lands,” but any people capable of maintaining an independent political identity and of negotiating war and peace could call itself a m¯atum. These “lands” were often defined by the names of the towns at their centers, but some were named for tribal populations. Somehow, the political customs of both towns and tribes were incorporated into this larger framework, and it is not possible to understand either one without exploring the m¯atum. At the same time, it is important to realize that the idea of group decision making is not primarily identified with the “land” in the documentation itself. Most often, the collective political voice is defined by individual towns, and these become the ultimate destination of my study in Chapter 4. This voice may be identified simply by named towns as such or by their gathered leadership, as “elders,” “heads,” or the like. In the cases of Imar, Tuttul, and Urgiˇs, the collective decision making is attested in the consistent evidence of multiple texts. What can we call this tradition? Our evidence does not indicate any conscious equalizing of all participants or any conscious attempt to extend such equal participation to a whole “citizenry,” however defined. We cannot call this “democracy,” however primitive, without losing the value of that term. At the same time, however, the world before democracy was not bereft of inclusive political process, and we do well to examine the democratic tradition against these independent customs, if only to understand better where the innovation truly lies. Given the geographical proximity of Greece to the ancient Near East, it is difficult not to wonder what similar collective traditions may have been shared by preclassical Greece. This book offers no answers, though I hope that its contents help illuminate the discussion among specialists.

Preface

xv

In general terms, the Mari archives offer a perspective on the ancient political landscape that invites a reappraisal of some common typologies. Zimri-Lim was both a tribal king and the master of a major city center, and his administration embraced both tribal leaders and Mari officials from the previous regime, enemies of his tribe. This same king identified himself vigorously with the mobile pastoralists of his tribe, who represented a key power base, even as he managed the river-valley core of a preexisting Mari kingdom with the normal settlement-oriented methods for governing farming communities. We will always oversimplify and misunderstand if we reduce our interpretation to oppositional pairs, such as urban and rural places of residence, subsistence by pastoralism and cultivation, settled and nomadic modes of life, tribal and urban affiliations, or hierarchical palace and communal village political systems. At every turn, Mari displays unlooked-for combinations and connections that show the urban, the elite, the royal, and the settled in much more intimate relationship with the rest of the typological spectrum than some might expect. Few ancient voices deny outright the unquestioned authority of the king, but his actual power seems to be a matter of constant negotiation, as he engages a panoply of traditional leaderships, each with its own constituencies and assumed prerogatives. Here again, I find myself standing at the edge of questions and evidence that extend far beyond the material that I have addressed in this book. If my investigation provokes any progress by others toward better answers than my own, the effort will have been well spent.

Acknowledgments

More than any project I have worked on to date, this book was a collaborative effort. To whatever degree it succeeds, this could not have been achieved without the support and assistance of several key people and institutions. I acknowledge them in roughly chronological order. My interest in Mari followed quickly on the heels of my introduction to Assyriology in the early 1980s under the inspired tutelage of William Moran. As he received the first reports of the new French publication team, led by Jean-Marie Durand, he exclaimed that they were essentially equal to fresh excavated finds, as important as any new Mesopotamian evidence. Ultimately, the project began from this judgment. My actual work on the book began with an idyllic year in Paris, funded by a Fulbright research fellowship, along with sabbatical support from New York University (1997–8). During this time, the staff of the Commission Franco´ Américaine d’Echanges Universitaires et Culturels was particularly helpful, especially Pierre Collombert, the director, and Elizabeth Marmot, head of the American Section. I went to France with the sense that I would learn most about the Mari evidence by making the acquaintance of the experts of the Paris group that has been publishing the thousands of tablets from this ancient city. Both during that first year and since then, the project has benefited from repeated exchanges with various participants in the French group, and I have discovered a new world of friendships and contacts in France. M. Durand generously tolerated my attendance through most of his 1997–8 seminar on nomads and tribal peoples, which provided a constant foil for my own explorations. Each key conversation with M. Durand has become a point of reference for the development of my own system, and I have pushed to stake out my own interpretive ground, knowing that he will be the first to perceive its real faults. His interest and help have been indispensable. xvii

xviii

Acknowledgments

Dominique Charpin first welcomed me to the Mari research center at Rue de la Perle in 1991, and it was at his encouragement that I imagined the possibility of a Paris sabbatical. Dominique has read and responded to every generation of my work on Mari, from the earliest notes in 1997 to two full drafts in 1999 and 2001. Perhaps more than anyone, it was Dominique who corrected my gaffes and helped me to distinguish plausible alternatives from impossible mistakes, though of course he cannot be blamed for those that remain. Both personally and professionally, he has shown me enormous generosity and hospitality, and both I and the book are the richer for them. I benefited from contact with many other participants in Mari research and contiguous areas, including Nele Ziegler, Sophie Lafont, and Francis Joannès. The one other crucial contributor to my project has been Bertrand Lafont. Everything that I gave to Dominique to read I also gave to Bertrand, and we have maintained a continuing dialogue over five years. On our arrival in Paris with our three young children, it was Bertrand and Sophie who immediately offered help with mundane needs. Since then, Bertrand has helped with every phase of the book in every dimension of its preparation. In the process, I have gained a rare friendship. Ce que tu m’as donné, tu sais. Although much of this book has revolved around my year in France and subsequent contacts with the Mari publication team, the project began a new stage with my return to the United States. I have since presented several components at annual meetings of the American Oriental Society and at the 2000 Rencontre Assyriologique Internationale in Paris, all resulting in useful feedback. In this stage, when I was working to understand the wider implications of the Mari evidence for politics and society, I made the acquaintance of Tom McClellan and Anne Porter, the excavators of Tell Banat in Syria. We found that our work dovetailed in many ways, and both Anne and Tom have since played an active part in the development and evaluation of the final product. They also read every draft, and in particular, Anne’s running commentary on the 2001 draft supplied the most demanding and fruitful critique of my larger conceptions, as I set out to produce a final text for the publisher. Tom has contributed the invaluable maps that accompany the volume. Here, also, what began as business has come to friendship. In the fall of 2001, just after I had finished a full second draft, I was introduced to Andrew Beck of Cambridge University Press by a colleague named Bill Arnold (thanks, again). To my great pleasure, Andy Beck agreed to have this still rough manuscript reviewed for publication, so that I could incorporate the extensive comments of the Cambridge referees at a stage when the form was still very malleable. Thus, the 2001 draft was perused in full by a large number of generous readers, solicited by me directly and by Cambridge, and the responses gave me much to address and include. These readers included Harry Hoffner of Chicago’s Oriental Institute, Bertille Lyonnet of the CNRS, Kurt Raaflaub of Brown University, my colleague Mark Smith at New York University, and two more nameless but appreciated

Acknowledgments

xix

readers for the Press. Piotr Michalowski and Gunter ¨ Kopcke sent me their own forthcoming articles on pertinent subjects. All of my contact with Cambridge has been a pleasure, thanks mainly to Andy Beck. Stephanie Sakson efficiently shepherded the manuscript through copyediting and production. I cannot compose a narrative of acknowledgments without thinking of my family, who give color and life to everything I do. I have dedicated this book to my parents, Wendell and Florence Fleming, whose love and influence I may only be beginning to appreciate now that I am taking my own turn as parent. Thank you. To Anthony, Elena, and Luc, my children, thanks for what you did not even know you gave up to this work that I thought worthwhile. Finally, everything I do eventually comes back to you, Nancy, who shares every breath and every heartbeat. Again, you have released me to pursue another love, confident that you can never be supplanted.

Abbreviations

A. AAAS AEM AfO AHw AJA Amurru 1

Amurru 2

Amurru 3 AOAT AoF ARET ARM(T) ASOR AuOr BA BaM BASOR BCSMS BiOr CAD

Louvre Museum siglum Annales archéologiques arabes syriennes Archives e´pistolaires de Mari Archiv für Orientforschung W. von Soden. Akkadisches Handwörterbuch, 3 vols. (Wiesbaden: Harrassowitz, 1965–81) American Journal of Archaeology ´ et les Hourrites: dix ans de travaux, J.-M. Durand (ed.), Mari, Ebla ´ première partie (Paris: Editions Recherche sur les Civilisations, 1996) ´ et les Hourrites: J.-M. Durand and D. Charpin (eds.), Mari, Ebla ´ dix ans de travaux, deuxième partie (Paris: Editions Recherche sur les Civilisations, 2001) Compte Rendu de la 46e Rencontre Assyriologique Internationale (Paris, 2000) Alter Orient und Altes Testament Altorientalische Forschungen Archivi Reali di Ebla, Testi Archives Royales de Mari (Textes) American Schools of Oriental Research Aula Orientalis Biblical Archaeologist Baghdader Mitteilungen Bulletin of the American Schools of Oriental Research Bulletin of the Canadian Society for Mesopotamian Studies Bibliotheca Orientalis I. J. Gelb et al. (eds.), The Assyrian Dictionary of the Oriental Institute of the University of Chicago (Chicago: Oriental Institute, 1956– ) xxi

xxii

CANE CBQ CRRAI CRRAI 38

DCM Emar VI.1–4

FM GKC

GN HSAO IBoT JANES JAOS JCS JESHO JNES KBo KTU

KUB LAPO LAPO 16, 17, 18 M. M.A.R.I. MDOG ME MEE Mélanges Birot

Abbreviations J. M. Sasson (ed.), Civilizations of the Ancient Near East, 4 vols. (New York: Scribners, 1995) Catholic Biblical Quarterly Compte rendu de la Rencontre Assyriologique Internationale D. Charpin and F. Joannès (eds.), La circulation des biens, des personnes et des idées dans le Proche-Orient ´ ancien, CRRAI 38 (Paris: Editions Recherche sur les Civilisations, 1992) F. Joannès (ed.), Dictionnaire de la Civilisation Mésopotamienne (Paris: Robert Laffont, 2001) D. Arnaud. Recherches au pays d’Aˇstata: Les textes ´ sumériens et accadiens (Paris: Editions Recherche sur les Civilisations, 1985–7) Florilegium Marianum A. E. Cowley (ed.),Gesenius’ Hebrew Grammar as Edited and Enlarged by the Late E. Kautsch (Oxford: Clarendon, 1910) Geographical name Heidelberger Studien zum Alten Orient Istanbul Arkeoloji Mu¨ zelerinde Bulunan Boˇgazköy Tableteri Journal of the Ancient Near Eastern Society of Columbia University Journal of the American Oriental Society Journal of Cuneiform Studies Journal of the Economic and Social History of the Orient Journal of Near Eastern Studies Keilschrifttexte aus Boghazköi M. Dietrich, O. Loretz, and J. Sanmart´ın, Die keilalphabetischen Texte aus Ugarit, AOAT 24/1 (NeukirchenVluyn: Neukirchener Verlag, 1976) Keilschrifturkunden aus Boghazköi Littératures Anciennes du Proche-Orient J.-M. Durand, Documents e´pistolaires du palais de Mari, Tomes I, II, III (Paris: du Cerf, 1997, 1998, 2000) Siglum for tablets from Mari Mari: Annales de recherches interdisciplinaires Mitteilungen der Deutschen Orient-Gesellschaft P. Fronzaroli (ed.), Miscellanea Eblaitica 2 (Florence: Università di Firenze, 1989) Materiali Epigrafici di Ebla J.-M. Durand and J.-R. Kupper (eds.), Miscellanea Baby´ lonica: Mélanges offerts a` Maurice Birot (Paris: Editions Recherche sur les Civilisations, 1985)

Abbreviations Mélanges Garelli

N.A.B.U. OEANE

OLZ Or OrAn RA RAI RHPR RIMA RIME RlA Römer Festschrift

RSO SEPOA Subartu IV Subartu VII

Syria Tell ar-Rimah ˘

UF WZKM ZA

xxiii

D. Charpin and F. Joannès (eds.), Marchands, diplo´ mates et empéreurs: Etudes sur la civilisation mésopotami´ enne offertes a` Paul Garelli (Paris: Editions Recherche sur les Civilisations, 1991) Nouvelles assyriologiques brèves et utilitaires E. M. Meyers (ed.), The Oxford Encyclopedia of Archaeology in the Near East (Oxford: Oxford University Press, 1997) Orientalische Literaturzeitung Orientalia Oriens Antiquus Revue d’assyriologie et d’archéologie orientale Rencontre Assyriologique Internationale Revue d’Histoire et de Philosophie Religieuses Royal Inscriptions of Mesopotamia Assyrian Periods Royal Inscription of Mesopotamia Early Periods E. Ebeling et al. (eds.), Realexikon der Assyriologie (Berlin: de Gruyter, 1928– ) Manfried Dietrich and Oswald Loretz (eds.), Dubsar anta-men: Studien zur Altorientalistik, fs. Willem H. Ph. Römer, AOAT 253 (Munster: ¨ Ugarit-Verlag, 1998) Ras Shamra - Ougarit ´ Société pour l’Etude du Proche-Orient Ancien Marc Lebeau (ed.), About Subartu: Studies Devoted to Upper Mesopotamia (Brepols: Turnhout, 1998) Olivier Rouault and Markus Wäfler (eds.), La Djéziré et l’Euphrate syriens de la protohistoire a` la fin du IIe millénaire av. J.-C. (Brepols: Turnhout, 2000) Syria: Revue d’art oriental et d’archéologie Texts from Tell ar-Rimah (Qat.t.arâ); S. Dalley, C. B. F. ˘ The Old Babylonian Tablets Walker, and J. D. Hawkins, from Tell al Rimah (British School of Archaeology in Iraq, 1976) Ugarit-Forschungen Wiener Zeitschrift für die Kunde des Morgenlandes Zeitschrift für Assyriologie und Vorderasiatische Archäologie

Fig. 2

Urgisˇ ˇ Aslakk‚ ˇ ˇ Subat-Enlil/Sehna ˇ Harran Asnakkum (Tell Leilan) Hanzat Qirdahat Tell Banat Nagar Karanâ Nineveh Nihriya Alalah ˇ Sudâ Halab / Aleppo Der Shemshara Tuttul Qattunan Andarig Razamâ Imar Ekallatum Abattum Ebla Tabatum Qabrâ Fig. 3 ˇˇ Assur Dur Yahdun-Lim Amurrum Gasur / Nuzi Saggaratum Qatna Terqa Carchemish

Mari Sapiratum Esnunna ˇ

Hìt ˇ Aksak 1500+m

Sippar

1000-1499m

Babylon

500-999m

Kisˇ

Kasallu(k) Agade ˇ Nippur Puzris-Dagan

200-499m

Isin

0-199m

Uruk 0

100 km

Umma Lagasˇ

Larsa Ur

map 1. Ancient Syria and Mesopotamia

Saggaratum Dabisˇ Salt flats

Samanum Depression

Terqa

ˇ Hisamta 300m 250m 240m

Suprum

230m 220m

ˇ Mislan

200m 185m 175m

Mari

170m

0

20 km

map 2. The Mari Region

Talhayûm

Urgisˇ

Aslakkâ ˇ ˇ Asnakkum Zalluhan Chagar Bazar Nahur Hurrâ

Susâ

ˇSunâ ˇ ˇ (Tell Leilan) Ilan-surâ Subat-Enlil/Sehna

Kahat ˇ Suduhhum

Razamâ

Tadum Nagar Qâ-&-Isqâ

Qirdahat

Tillâ

Tell Raqa’i Tell ‘Atij Karanâ 1000-1499m

Kurdâ 500-999m 200-499m

Andarig

0-199m

0

50 km

Qattunan

map 3. The Habur River Basin

˘

Qattarâ Razamâ

Democracy’s Ancient Ancestors Mari and Early Collective Governance

1 Introduction

This book undertakes a study of ancient political life through the lens of one body of evidence: the cuneiform texts from Mari. By this approach, I am accepting the need to straddle two distinct demands that proceed from two different audiences. My focus on the archives, which are far from fully published, and which remain the subject of continuing reevaluation by specialists in the field, invites serious investigation of evidence the interpretation of which is by no means settled. This attention to Mari for its own sake requires that I present evidence and arguments with sufficient technical detail to demonstrate the basis for my ideas to cuneiform specialists. My interest in the larger issues raised by this Mari material, however, has involved me in literature far beyond my own specialization, and I mean to make the book as accessible as possible to the scholars and students whose fields I have trespassed. The introductory material that follows is designed especially for those who know little about Mesopotamia or Mari and who may not be familiar with the conventions of Assyriological study of cuneiform texts. I begin by introducing the Mari archives and offering a historical overview of ancient Mari before addressing the specific issue of collective political forms. I close with observations about my methodological choices, particularly as they result in a text-oriented study.

a. the mari texts Excavations at ancient Mari (Tell Hariri) began in 1933–4 under the leadership of André Parrot, with a French team. A huge palace was discovered in 1935, and large numbers of cuneiform tablets rapidly began to appear (Margueron 1997, 143). By the onset of World War II, the majority of known Mari archives had already been found, though Parrot took up work again after the war and continued until 1974. In recent years, excavations have been led by Jean-Claude Margueron, and the site is still not considered closed. 1

2

Introduction

Tells such as that of ancient Mari are regularly called “cities,” but this term demands careful qualification, inspired partly by what we know and partly by what we do not. The site of Tell Hariri is enclosed by a mound in the form of an arc that represents about one third of a circle, roughly three to four kilometers from the modern channel of the Euphrates, within the flood plain of the river.1 Almost all of the cuneiform tablets found at Mari come from the reigns of the kings who ruled there during the last halfcentury of its existence, conventionally dated to the early eighteenth century b.c.e. Much of what has been discovered within the existing site for that period served royal and ritual purposes: the main administrative palace and a subsidiary palace dominated by the royal harem, various temples, and large residences occupied by key Mari officials.2 Even after centuries of use, portions of the tell appear never to have been built up, and no proper residential quarters have yet come to light.3 Future excavations always yield new finds that embarrass those who argue from silence, but at this point, it seems that the “city” of Mari cannot be assumed to have housed a large population within its walls, beyond the significant number who depended directly on the king.

1. The Texts and Their Publication The cuneiform texts from Mari reflect this public and royal setting. The overwhelming majority represent the palace archives of Zimri-Lim, the last king of Mari, who inherited significant numbers of tablets from his predecessor, a rival from a completely separate dynasty and region. Most of the texts reflect practical use rather than scribal training, and so we have little classical Mesopotamian literature and few lexical collections or texts from specialized scribal ruminations on divination or incantations.4 Instead, we find two main types among the roughly 20,000 registered tablets and fragments: administrative documentation reflecting the daily affairs of various palace agencies and an unprecedented collection of letters. The detailed evidence for royal administration by itself would make the archives an important discovery, but it is the royal correspondence that is unique among cuneiform finds. With over 3,000 letters included, the sheer number is remarkable, but it is their range of interest and origin that represents their particular historical value. We have exchanges between kings of Mari and other rulers or towns and thousands of reports from high palace officials, district governors, generals, tribal leaders in royal service, diplomats and envoys on royal missions outside the kingdom, and miscellaneous others. There are letters between officials and even some intercepted enemy messages. Some missives are terse and purely informational, but many are more conversational, sometimes even verbose, to the modern reader’s pleasure. From the sum of them it is possible to glean knowledge of widely diverse aspects of Mesopotamian society, with the advantage of historical coherence.

The Mari Texts

3

The voices are distinct, but they speak out of a single brief period, in which their varied experiences were ultimately interlocked. Although most of the Mari tablets were discovered decades ago, their impact has been spread over the years of their gradual publication, so that new evidence continues to become available, as if from recent excavation. Even now, far fewer than half of the Mari documents have been published, and much important material has yet to emerge. There have been two main generations of Mari scholarship, and any use of Mari evidence must give special attention to the more recent work, whether textual or archaeological. After the initial discovery of the tablets in 1934, their publication was entrusted first of all to the venerable Assyriologist Fran¸cois Thureau-Dangin, whose leadership soon passed to Georges Dossin. Most of the Mari texts available to the public before 1980 were published by Dossin and his colleagues through a period roughly contemporary with the excavations of Parrot. Impressive as were the tablets made available by 1980, they still represented only a small fraction of the whole, and after a transition aided especially by Maurice Birot, the baton was passed to a younger generation. In 1982, a new research team was formed under Jean-Marie Durand. This change of leadership not only reinvigorated the publication process, but also introduced a completely fresh analytical perspective, driven especially by Durand and Dominique Charpin. The twenty years of Mari research since the early 1980s have produced a deluge of new texts and interpretive comment, and more evidence awaits publication. Much of this new material has not been digested by the larger circle of Mesopotamian specialists, not to mention scholars outside this field, and one goal of my project is to help extend the impact of the new research. My own serious work on the Mari archives began in the 1997–8 academic year, when I had the pleasure and privilege of a Paris sabbatical. During this stay, I benefited tremendously from the hospitality and intellectual vigor of the current group involved with Mari research, including especially (but not only) Durand, Charpin, and Bertrand Lafont. Even as I have actively sought to forge an independent perspective, based on a critical reading of both the textual evidence and current French interpretation, the extent of my intellectual debt to these scholars will be obvious to anyone familiar with their work. Naturally, my analysis diverges from theirs at many points. Nevertheless, I find many of their conclusions compelling, and my text citations rely heavily on the readings of their new editions.5 In some cases, where my French colleagues’ analysis is both important and potentially controversial, I offer my own rendition of their arguments, both for my readers’ convenience and to add another voice in favor of these ideas. The published texts themselves are scattered through a variety of venues that reflect the long history of work on them. Early discoveries by ThureauDangin and then each team that succeeded him were often presented in individual articles that can be difficult to track down. Dossin initiated the

4

Introduction

first regular series of volumes devoted to Mari texts, entitled Archives Royales de Mari (ARM), which now includes up to volume XXVIII.6 As work on the texts was revived under Durand’s leadership, and Mari’s historical situation became clearer, the earlier categories became increasingly problematic, and Durand has undertaken new classifications. In recent years, Mari tablets have been published in smaller blocks, especially in the series Florilegium Marianum (FM). Durand has recently completed three volumes that present new renditions of all of the Mari letters published before his leadership, with translations and notes for new readings based on fresh collation (direct examination) of the tablets. These appear as volumes 16–18 of the series Littératures Anciennes du Proche-Orient (LAPO), entitled Documents e´pistolaires du palais de Mari, I–III. Obviously, any serious use of the written evidence from Mari calls for a working knowledge of French.

2. Navigating the Technical Terrain: The Language and the Writing System Because this study of ancient political life is rooted in the writing that allows us to hear the categories and interpretations of the participants, I have built large parts of the book around evaluation of specific words and the texts that carry them. This means that readers who are not familiar with the ancient languages in play will be invited to tolerate a certain dose of the unfamiliar in order to understand what this evidence offers to the broader study of human society. I offer the following comments with the hope of making this obstacle less imposing. Almost all of the essential evidence for collective political traditions in the Mari archives comes from the letters. As a whole, this correspondence is written in Akkadian, the Semitic language of eastern Mesopotamia, native to Babylon, Eˇsnunna, and Aˇsˇsur during the early second millennium. Akkadian was used for correspondence in this period wherever cuneiform was used. In Iraq and Syria, the heartland of cuneiform writing, even nonnative speakers exchanged written messages in Akkadian, and good Akkadian at that. As in the other Semitic languages, most Akkadian verbs and nouns were derived from triconsonantal roots that were manipulated in various patterns to yield different meanings. For example, the Akkadian noun “counselor” (m¯alikum) is related to the verb imlik (“he/she counseled”) and the noun “counsel” (milkum). The final -m on the nouns disappears soon after the period of the Mari archives, and the -u- before it is a case vowel that varies according to its function in phrases and clauses. Most Syrians of this period spoke varieties of “West” Semitic dialects that were quite distinct from Akkadian, but we have little more than individual words that were rendered as if Akkadian. It is possible to distinguish Akkadian from West Semitic terminology in the Mari texts only by patterns of use as compared with the range of documentation from this period. In

The Mari Texts

5

this book, all Semitic words will be presented in the same italicized form. Words as such will be presented whole, as in sug¯agum (“leader”), while actual citations of text will distinguish the separate cuneiform signs used to write them, each with its own phonetic value (e.g., su-ga-gu-um). Throughout the history of cuneiform, the writing system preserved embedded within it the primary language of its earliest use. Because the earliest cuneiform writing incorporated almost no recognizeable indicators of grammar, and represented simple objects or actions by symbols developed from a pictographic method, it is difficult to demonstrate the language of its first creators, but it came to flower in southern Mesopotamia with speakers of Sumerian.7 Sumerian was an agglutinative language that was not even remotely related either to the Semitic family or to any other known group. It seems to have ceased to be a living language at the end of the third millennium, one casualty of the same upheavals that led to the increased prominence of West Semitic speakers in eastern Mesopotamia during the early second millennium.8 In spite of its recent demise as a spoken tongue, Sumerian enjoyed a tremendous literary popularity in this period, and it became an essential language of cuneiform scholars and scribes for centuries to come. Sumerian always constituted an unavoidable ingredient in writing through its continued use as a scribal shorthand, hiding the underlying forms of the languages actually spoken and read. Modern conventions for untangling the snarls created by this blended system vary, unfortunately. I distinguish Sumerian writings for Semitic words from the Sumerian vocabulary itself by rendering the former in upper case (e.g. URU, for Akkadian a¯ lum, “town, settlement”), and the latter in lower case (simply “uru”). Assyriologists usually cite Sumerian words without italics in order to distinguish them more clearly from italicized Semitic words, in spite of the wider modern convention of italicizing all foreign terms. In this book, the reader will encounter ancient words in two forms: specific vocabulary and proper nouns. When I refer to specific vocabulary, I use a form fully marked for vowel length, according to the standard conventions of Assyriology, such as sug¯agum (“leader”) or merhûm (“chief of pasture”). Long vowels are marked with a macron, and long˘ vowels formed from the contraction of two vowels are marked with a circumflex. Consonants are presented according to the conventions of cuneiform transliteration, and readers should recognize -h- (/kh/), -s.- (/ts/), and -ˇs- (/sh/). The emphatic ˘ -t.- is pronounced /t/ in common use. I always leave as such the laryngeal consonant written as -h-, while recognizing that the cuneiform signs can ˘ represent a variety of other Semitic laryngeals. For example, the noun written as merhûm actually includes the Semitic consonant ayin (transliterated ˘ which cannot be distinguished as such with this writing sysas mer uˆ m), tem, but I render this and other such words in the forms yielded by their cuneiform spellings. This allows words and names of uncertain etymology to be presented consistently as written in the texts themselves.9

6

Introduction

In the case of proper nouns, I have decided to preserve the basic pattern of consonants and vowels, according to the conventions of transliteration, with one exception. Names of places, people, and deities will not be marked with macrons for simple long vowels (e.g., Qat.t.unan for Qat.t.un¯an, Saggaratum for Saggar¯atum). Only final contracted vowels will be marked with the circumflex, because these change the stress in pronunciation to the final syllable (e.g., Kurdâ). Often the quality of the vowels in proper nouns is simply not certain, and my strategy allows me to refrain from forcing an interpretive choice on every name.

b. a survey of mari history To begin at the end, the city of Mari was destroyed by Hammurabi of Babylon in his thirty-second year, conventionally dated 1761.˘The site was never rebuilt, and the political center of this region moved upstream to Terqa, another old city from the previous millennium. At the least, then, it is clear that the history of Mari belongs to the third and early second millennia b.c.e., and no later. Dating the foundation of the site is less straightforward than dating its destruction. At the least, we know that a settlement existed at the beginning of the third millennium in the Early Dynastic period, whether or not earlier occupation might eventually be discovered. The problem is the scope of the earliest site. The current excavator, Jean-Claude Margueron, dates the settlement to the twenty-eighth century, and concludes that its massive enclosing mound must reflect a city at the center of a fully developed state from the very onset.10 The mound would suggest a size of 100 hectares, unprecedented for this upstream region in this period. It is not clear why Mari was founded. The soil is not good, rainfall is far too low to support agriculture, and irrigation was never possible on a scale that would explain a city of the scope suggested by Margueron. Margueron therefore proposes that the city was founded by an existing state-level society (from downstream?) in order to control traffic below the confluence of the Euphrates and the Habur Rivers, with both the city and a 120-kilometer-long ˘ canal built at the same early date.11 At the moment, there seems to be too little evidence to identify how and why such a large settlement was created.12 We must keep in mind that no actual structures have been excavated for this early period, and we have no idea how much of the enclosed space was built up. Less well-known settlements with such circular enclosures often show little sign of a large population.13 It is perhaps most prudent not to speak of early third-millennium Mari as the center of a unique “state” until we have more information. This earliest settlement appears to have been abandoned before the establishment of an entirely new city, perhaps the first that can properly be called such, near the middle of the third millennium. Texts from Ebla, a large kingdom based in western Syria, south of Aleppo, show that Mari had become

A Survey of Mari History

7

the center of a major regional power, whose influence reached as far as the domains of its western rival. For this period, the excavations yielded monumental public buildings, including a palace and temples, along with part of a residential area. Excavations at Mari have unearthed only a few cuneiform tablets from this period from several different buildings.14 Whatever the population at the time, the city was now home for a permanent population, a condition that is difficult to confirm for both the earlier and perhaps even the later periods. I use the word “city” for the Mari site cautiously, recognizing the combination of large-scale public buildings and the strong possibility of a significant permanent population. By these requirements, I cannot confidently call Mari a “city” either before or after the middle of the third millennium. This great Euphrates center was destroyed by one of the early rulers of the Agade (Akkad) empire, which expanded across much of Mesopotamia from a base near later Babylon. An inscription known only from early secondmillennium copies credits this conquest to Sargon, the founder of the Akkadian dynasty, in the mid-twenty-fourth century.15 After this destruction, Mari again lay empty for some uncertain period, to be rebuilt a third and last time toward the end of the millennium. During the period contemporary with the great southern Mesopotamian kingdom of Ur, Mari was ruled by men who called themselves “governors” (ˇsakkanakkum), apparently with real success. Excavations show this to have been a time of major building projects, including a new palace and new temples. Margueron observes that the entire second-millennium site is made ˇ up of structures that were originally built in this “Sakkanakku” period, and Zimri-Lim’s Mari had no properly “Amorrite” architecture. In spite of the intensive excavation of this last stage of Mari occupation, there remain important questions. Margueron reports that he has begun ˇ excavating a residential quarter for the Sakkanakku settlement, but it will be necessary to know its size, the character of its population, and when ˇ exactly it was inhabited. Although the buildings of the Sakkanakku center were reused by the kings of the eighteenth century, it is not clear from the reports whether they were occupied continuously through the one to two ˇ centuries between the last Sakkanakku ruler and the arrival of Yahdun-Lim. If they were not taken over directly from the active institutions ˘of a prior regime, we cannot have confidence that Yahdun-Lim adopted a long-lasting “urban” and administrative tradition. The˘Mari texts from the eighteenth century do not provide clear evidence for any substantial residential quarter, so it remains difficult to judge to what extent the last kings had to deal with the expectations of a long-standing native population.16 There are no urban institutions identified with the city of Mari, such as collective leadership or the merchant community called a “quay” (k¯arum). What was the economic basis for the obvious power flaunted by the enormous structures of third-millennium Mari? Margueron has argued that it

8

Introduction

was ultimately the control of river traffic, especially for shipping wood downstream to southern Mesopotamian Sumer. It is possible, however, that some significant part of the Mari economy always depended on the steppe, the domain of the mobile herdsmen and their flocks. During the eighteenth century, the peoples of the steppe provided the power base for the kings of the “Lim” dynasty of Yaggid-Lim, Yahdun-Lim, and Zimri-Lim, who claimed ˘ We may treat this as a new secondto rule a “land of the tent-dwellers.” millennium phenomenon, coming from “Amorrite” shepherd peoples, but the middle portion of the Euphrates had always flowed through pastoralist country. It may not be necessary to assume a fundamental economic change from the third millennium.17 The end of the third millennium was marked by changes that set the stage for Mari’s last hurrah under the kings who left us the huge archives of the excavated tell. In southern Mesopotamia, the kingdom centered at Ur quickly lost its grip on the region and gave up ground on all fronts before falling to Elam, the major kingdom of southern Iran. The leaders of Ur identified the crisis especially with people identified as “westerners,” or Amorrites (Akkadian Amurrûm, Sumerian Mar-tu), as shown by the “Amorrite wall” that Ur built to stave them off, without noticeable effect. These westerners were stereotyped as uncouth barbarians, but in fact, by the time of Ur’s collapse, these West Semitic speakers were already integrated into the leadership of political centers quite close to Sumer.18 After the destruction of the city of Ur itself, the power vacuum was immediately filled by Iˇsbi-Erra, the commander of Ur’s northern troops, who established his royal seat at the Sumerian city of Isin.19 The evidence for the transition from Ur III to Isin dominance is as fascinating as conclusions are elusive. This political shuffle stands at the center of what modern scholars have regarded as the end of Sumer and the emergence of West Semitic–speaking Amorrites in positions of power across Mesopotamia. Ancient scribes were preoccupied with the transition as well, as seen in the early second-millennium “Old Babylonian” versions of various texts presented as products of this crisis. A letter that purports to be from Ibbi-Sˆın, the last king of Ur, to Puzur-Numuˇsda, governor of Kasallu(k), has the Sumerian ruler castigate Iˇsbi-Erra as “not of Sumerian stock, a man from Mari, with a dog’s intelligence” and “a monkey from the mountains.”20 Other copied texts from advocates of Iˇsbi-Erra and his Isin dynasty embrace this foreign origin. He comes from the mountains, indeed, but as the shepherd appointed by the gods Anu and Enlil.21 In fact, Iˇsbi-Erra seems to have been born to the royal family of Mari, and his influence at Ur reflected a long period of close relations between the two states.22 If this makes him “Amorrite,” it is by a western identity that is already rooted in a major late third-millennium center. It may be that he had nothing to do with the specific tribal groups named two or three centuries later, but the curious identification of a king from the Euphrates


9

valley with the mountains of cedar suggests that later Amorrites, at least, may have made him their own. By the eighteenth to seventeenth centuries, West Semitic royal names appear all over central and southern Mesopotamia, and it seems clear that “westerners” had come to dominate the political scene across all of Syria-Mesopotamia. In southern Mesopotamia of the early second millennium, which Assyriologists call the Old Babylonian period, the Amorrite rulers accommodated themselves comfortably to the admired culture of old Sumer and Akkad, and in the written evidence, their western roots are obscured by the overlay of the eastern languages and ways. We have archives for this period from Isin, Larsa, Babylon, Nippur, Sippar, Umma, Ur, Eˇsnunna, Shemshara, and other southern and central sites, along with northern sites, including Chagar Bazar ˇ hna/Subat-Enlil), ˇ (Aˇsnakkum?), Ishchali (Nerebtum), Tell Leilan (Se and ˘ Tell ar-Rimah (Qat.t.arâ), and westward all the way to Tell Atchana (Alalah) ˘ and then south to Ebla. Along the Euphrates, the largest finds come from Mari, but texts from this period also were discovered further upstream at the old centers of Terqa (Tell Ashara) and Tuttul (Tell Bia). More than any of these, the tablets from Mari display the active role played by social and economic traditions foreign to southern Mesopotamia in societies with such Amorrite roots. On the social side, Mari shows a complex world of far-flung tribal affiliations that pertain to both settled and mobile “nomadic” peoples, and to those both at the fringe and at the hub of political power and its fortified palace centers. On the economic side, the Mari texts indicate the importance of large flocks of sheep and goats, in a pastoralism that was most often carried out in the country steppe by tent-dwelling shepherds who moved seasonally across fairly long distances. Although the Amorrite influence on southern Mesopotamia was already centuries-old by the time of our Mari archives, these texts perhaps offer our best view of the Amorrite culture, as described in ancient writing. The period of our early second-millennium archives was both brief and turbulent. A king named Yahdun-Lim made Mari the center once more ˘ of a large realm, taking advantage of its reputation as the ancient capital of the region. Yahdun-Lim, who mentions only his father, Yaggid-Lim, as a predecessor in ˘this royal line, gained control of a long stretch of the Euphrates River valley, west as far as Tuttul, which he made the second capital of his realm. Downstream, his kingdom immediately abutted the domains of Eˇsnunna, the major power of south-central Mesopotamia. Rather than challenge Eˇsnunna, Yahdun-Lim worked to extend his rule northward into the basin of the Habur˘River, in competition with Samsi-Addu, the king of Ekallatum, on the˘ Tigris River.23 Yahdun-Lim defined both his core population and his conquests upriver ˘ terms that are examined at length in Chapters 2 and 3. He himself in tribal is associated with the Binu Simal, or Simalites, “Sons of the Left (Hand),” while his defeated enemies ruled peoples known to belong to the Binu

10

Introduction

Yamina, or Yaminites, “Sons of the Right (Hand).” The duality of left and right hands by itself shows that these two groups understood themselves to be related, perhaps with some king of geographical basis for distinguishing their territories. These groups may be called “tribal” because their primary definition is by family affiliation under headings not defined by residence in a particular settlement, which is the most common way of identifying people (e.g., “sons of Terqa”). Yahdun-Lim was not able to establish a lasting Simalite base at Mari. ˘ Sumu-Yamam replaced him under less-than-straightforward circumHis son stances, and Sumu-Yamam in turn perished soon afterward, evidently at the hands of his own officials. To some extent, Sumu-Yamam’s fall not only benefited but may have been hastened by the looming power of Samsi-Addu, the long-standing king of Ekallatum.24 Samsi-Addu soon seized Mari and the Euphrates-based dominions that had been held by Yahdun-Lim, which seem ˘ to have had no strong local force left to fend him off. With this achievement, Samsi-Addu could truly claim to rule the lands between both the Tigris and the Euphrates, the first explicit “Mesopotamian” kingdom.25 Samsi-Addu was also an Amorrite, and the kispum ritual text found at Mari shows that his family claimed a heritage both in Sargon’s great dynasty at Agade (Akkad) and in the Numhâ tribe. During the reign of Zimri-Lim, the Numhâ tribal ˘ ˘ people were associated especially with the kingdom of Kurdâ, between the Tigris and the Habur Rivers, and they were not part of the Simalite-Yaminite ˘ duality that dominated Mari affairs under the “Lim” rulers.26 Samsi-Addu identified himself with pastoralist ancestors whom he called “hana (tent˘ who had dwelling?) yarr¯adum,” which Durand understands to be Amorrites 27 “come down” to the Euphrates valley. By this point in his career, Samsi-Addu was fairly old, and he created a clever and successful new structure in order to govern his expanded realm. Samsi-Addu divided the kingdom into an eastern section, with Ekallatum still at its center, and a western section, centered at Mari. He placed his older son, Iˇsme-Dagan, over the east and set up his younger son, Yasmah-Addu, ˘ as king of Mari and the western dominions. The old king himself retained a firm hold over ultimate decision making, but he left his sons their own capitals and took up an intermediate location in the Habur River basin at ˇ hna (Tell Leilan), which he renamed Subat-Enlil ˇ˘ the town of Se (Charpin ˘ 1987b). By this strategy, Samsi-Addu could maintain a direct royal presence in three main parts of his kingdom, and the stability of the arrangement through the last years of his lifetime bears witness to its effectiveness.28 At Mari itself, Yasmah-Addu reigned with full royal status for eight years, ˘ but had already held responsibility for some time for the region that SamsiAddu had taken roughly ten years earlier (Villard 2001, 10–14). After SamsiAddu’s death, however, neither Iˇsme-Dagan nor Yasmah-Addu was capable of maintaining the vast kingdom of their father. Yasma˘h-Addu quickly lost ˘ finally ruled by his Euphrates capital to a revived coalition of Simalites,


11

Zimri-Lim, a kinsman of Yahdun-Lim who at least claimed to be his son.29 The actual conquest of Mari ˘appears to have been achieved not by Zimri-Lim but by another Simalite leader, named Bannum, who could not compete with the dynastic mantle borne by Zimri-Lim.30 Zimri-Lim ruled Mari for slightly more than thirteen years before his realm succumbed to the ambitions of Hammurabi, king of Babylon.31 Iˇsme-Dagan survived in greatly ˘ reduced circumstances at Ekallatum, where he outlasted Zimri-Lim and may even have contributed to his defeat. Hammurabi’s final seizure of Mari took ˘ carefully reviewed the contents of place without a siege, and his scribes his onetime ally’s archives, apparently removing the most important correspondence with Hammurabi himself and with other major rulers, and then ˘ destroy the palace, entombing the thousands of remainfinally deciding to ing tablets in the rubble (Charpin 1995a). Unlike Samsi-Addu, Hammurabi ˘ seems to have had no use for the impressive buildings and elaborate administrative apparatus of Zimri-Lim’s seat of power. Mari was never rebuilt. Through the turbulent generations from Yaggid-Lim to Zimri-Lim, it is important to consider what changed and what remained largely the same. The written evidence highlights the changes. Yaggid-Lim and Yahdun-Lim ˘ represented a new political force at Mari, reigniting its earlier glory. They were newcomers, not the successors of an existing kingdom. With the arrival of Samsi-Addu, not only did the ruling family change but the political geography shifted radically away from Mari, which became the junior partˇ ner in a family business that was run from headquarters in Subat-Enlil and Ekallatum. Zimri-Lim brought the center of gravity back to Mari, and the power base back to the Binu Simal tribal confederacy. As “the king of the Simalite(s)” (see Chapter 2), Zimri-Lim gave direct authority to his Simalite tribal chiefs of pasture, called merhûms, who led the mobile pastoralist component of his people, most often˘ called simply “Hana” (“tent-dwellers”). Under Zimri-Lim, the whole of the “land” (m¯atum) ˘ he ruled was identified by the tent-dwellers of the Simalite tribal popthat ulation, but the kingdom as a whole was divided into two parts. While the Hana were led by their merhûms, the farming country along the rivers was ˘ governed within a system of ˘“districts” called hals.ums, each with its own local ˘ whole package was named palace administrative center and governor. The the Ah Purattim (“Banks-of-the-Euphrates”), and Zimri-Lim took over both ˘ the territory and its essential administrative structure from Yasmah-Addu. Interestingly, the midlevel leaders of both the settled towns and the˘mobile pastoralists had the same title (sug¯agum). The kingdom of Zimri-Lim was never as large as that of Samsi-Addu. Zimri-Lim retook the Ah Purattim, and during the middle part of his reign, he was able to establish a˘ strong influence over the Habur River basin to the ˘ arrangements. These north through a network of alliances and vassalage relationships are displayed in a rich correspondence that permits us to trace the ups and downs of many individual kingdoms, marked by violent

12

Introduction

changes in leadership and constant shifts in the regional balance of power.32 When the great kingdoms of first Eˇsnunna and then Elam tried to expand their Mesopotamian influence, in the fifth and eleventh years of Zimri-Lim’s reign, both of them made the Habur one key part of their strategies. This ˘ region was made up of so many minor kingdoms that an introduction to the names would add little clarity at this point, but two alliances are worth mentioning. In the northeastern reaches of the Habur basin, a shifting coalition ˘ of individual polities united as Ida-Maras., with a preference for Simalite affiliations. Further to the west, in the basin of a smaller tributary of the Euphrates called the Balih River, an alliance called Zalmaqum maintained ˘ confederacy. links with the Yaminite tribal Although the Simalite tribal power base and the reorientation of Mari from secondary to primary capital gave the kingdom of Zimri-Lim a face very different from that of Samsi-Addu and sons, much remained quite the same. If there was deeper change, it would have occurred with the arrival of Simalite rule under Yaggid-Lim and Yahdun-Lim. Margueron observes ˘ that many of the public buildings of Zimri-Lim’s capital, including the main ˇ royal palace, go back to the Sakkanakku period of the late third millennium, but there may have been a gap of a hundred years or more when the great administrative center had fallen out of use (Margueron 1996a, 103; cf. Durand 1985a, 158–9). There is no evidence that Yaggid-Lim even had to conquer Mari, which may have been standing empty or relatively so. Charpin observes that no year name or inscription commemorates a conquest. Yaggid-Lim and his son Yahdun-Lim appear to have continued ˘ the tradition of Mari as an administrative and religious center, without a major residential aspect to the city, judging by the available results of the excavations. With the repair and reuse of the royal palace and other public buildings, we have an institutional bridge of sorts between the two periods of occupation, though the continuity should not be overestimated. The real institutional continuity in fact perseveres through the last generations of Mari kings. Yasmah-Addu, sponsored by his father, Samsi-Addu, may have ˘ to suit the new administration as a province of the larger made some changes empire, but he did not create a new city. The structural changes initiated by Zimri-Lim had to do with the mobile component of his tribal base and would not have affected the essential workings of the palace. On a larger time horizon, even the active, intimate political bonds between Zimri-Lim and the pastoralists of his tribespeople should not be seen as a novelty at Mari, though it is indeed rare to discover the archive of such a ruler. By the time of Yahdun-Lim, Yasmah-Addu, and Zimri-Lim, Mari already had a long ˘ urban history that appears ˘to have stood in a lasting relationship with the pastoralist economy and peoples of the steppe beyond the Euphrates. The so-called Amorrite upheavals at the end of the third millennium may have radically changed the political scene in southern Mesopotamia, downstream

A Note on Prominent Terms

13

from Mari, but it is not clear that the basic relationship between town and countryside, fields and flocks, was fundamentally altered. In the early second millennium, the urban circle of Mari was less an engine of social change, extending its power outward from the settled center, than a target for other people’s regional ambitions. Mari’s relationship with the lands around it consisted of more than just a network of links between city and countryside, joined to face similar city-based states around it. The political power that occupied it always had roots in populations distant from Mari, with both urban and rural bases.

c. a note on prominent terms In the historical sketch just offered, I have had to make several decisions about how to render important terms that will come up repeatedly in this study. Three of these merit specific comment: the kingdom of Samsi-Addu, the Yaminite and Simalite tribal confederacies, and the word “Amorrite.” One of the most notable results of recent Mari study has been the complete reorientation of how Samsi-Addu’s kingdom is viewed. Samsi-Addu began his rule from the city of Ekallatum, north of the city of Aˇsˇsur on the ˇ hna in the upper Habur basin Tigris River. He then took over the city of Se ˘ of the later Assyrian ˘ ˇ and changed its name to Subat-Enlil. In spite claim to Samsi-Addu as an ancestral king of Aˇsˇsur, his kingdom cannot properly be called “Assyrian.” Durand and Charpin have therefore called this realm “the Upper Mesopotamian kingdom,” though I tend to identify it simply by its king Samsi-Addu, including within it the subsidiary domain of his son Yasmah-Addu at Mari. On ˘a less substantial level, I would like to suggest the abandonment of the old tribal terms “Bensimalite” and “Benjaminite” in favor of “Simalite” and “Yaminite.” The French designations were clearly adapted from the Israelite tribal name Benjamin, evidently to underline the obvious connection, but they are quite awkward when considered in light of actual Hebrew patterns. In traditional Bible translation, the adjectival forms of geographical names derive from different Hebrew originals, one of which is the tribal designation “sons of X.” The three that come to mind are the “Israelites” (Bene Yisrael, not BenYisraelites), the “Ammonites” (Bene Ammon, not BenAmmonites), and, more vaguely, the “Qedemites” (easterners, Bene Qedem). Additionally, the Israelite tribe of Benjamin was never regarded in the Bible as a people under this type of name. Benjamin is introduced as an individual personal name, and the tribal name is interpreted solely in that mode, so that in biblical terms “Benjamin” is grammatically parallel to “Israel,” not “the sons of Israel.”33 If we are going to use the familiar biblical renditions, Bensimalite and Benjaminite are redundant, and the “Ben-” should be dropped.

14

Introduction

I have chosen to follow the current Mari editors in their doubling of the -rr- in “Amorrites,” the “westerners” who are identified with the West Semitic–speaking tribespeople of the Mari archives. The doubling is original to the word and is often represented explicitly in Mesopotamian and Mari spellings. The common English form “Amorite” reflects the Hebrew of the Bible, which never marks consonantal doubling for -rr-. By adopting the longer spelling, we distinguish the term from both the biblical word and its southwestern associations. By retaining the vowel -o-, we still acknowledge the derivation of the better-known Hebrew category from this northern term.

d. the mari archives and political history Ancient history has tended to be told from the record of its most spectacular artifacts, whether the pyramids and monuments of Egypt or the massive citymounds of southern Mesopotamia. Both regions left us evidence of the first creation of writing. Such finds tend to come from the centers of ancient power, however, and they easily multiplied themselves into a history of kings and their magnificent palaces and tombs. Political history for the period of early written evidence then naturally gravitates toward the rulers who stood at the center of these societies, perhaps too quickly assuming a topdown style of authority and ignoring the evidence for a broader exercise of political power.

1. Individual versus Group-Oriented Power In the cuneiform evidence, kings are often portrayed as the embodiment of political action. The rulers of the first-millennium Assyrian empires pronounced that they themselves had humbled enemies in devastating defeat and honored gods with new temples. There is another face of power, however, whereby political leadership takes refuge in the identity of the group. This group can be the state, the town, or the tribe, and the language of this collective identity may simply name the group or may specify some plurality that we translate as “elders” or an “assembly.” Regardless of the actual power in play, the very desire to cast some decisions in collective terms demands explanation. Most often, however, these phenomena receive little attention, and when they do, they are often treated in limited terms as suits their limited political importance. The situation is not that different in the broader study of socieites outside the current that produced our modern democracies. After a foray into professional literatures beyond my own field, I remain surprised at how little attention is paid to the collective, cooperative, consensus-building aspects of political life. Evolutionary hierarchies generally relegate group decision making to the smallest and most primitive settings, and even then focus

The Mari Archives and Political History

15

more on the limited power of individual leaders than on the collective process itself. For example, Elman Service (1975, 71–4, passim) discusses the first institutionalization of power in bands and tribes under “bigmen” who still possess no coercive authority, but he has little to say about collective governance. This preoccupation with individual leadership is widespread.34 There are new voices expressing interest in alternative configurations of power, however. Richard Blanton, working from early Mesoamerican archaeology, states bluntly that he views “as problematic the assumption that political centralization is the central process in the evolution of states” (1998, 138). Blanton and three colleagues have proposed a fresh approach that envisions two main patterns of political action, one “exclusionary” and individual-centered, and the other more group-oriented, which they call “corporate.”35 Even more recently, Susan McIntosh has suggested that the anthropological study of Africa has long demanded a shift in the theoretical discussion of how “complex” societies may be structured (1999b, 4): I suggest that Africa challenges deeply embedded evolutionary notions of complexity as differentiation by political hierarchization and provides an instructive counterpoint to formulations that locate power centrally in individuals and focus analysis primarily on the economic strategies used by these individuals to maintain and expand operational power.

2. “Primitive Democracy” Ancient Mesopotamia has long provided its own counterpoint to the prominent powerful individuals, as observed by Blanton (1998, 155): It should be, but unfortunately has not been, of considerable theoretical interest to anthropological archaeology that Mesopotamian social formations appear to have emphasized corporate forms of government from an early period, including forms of assembly government that evidently had developed by the Uruk period.

Blanton’s interest represents an invitation to specialists in the ancient Near East, as well, both to revisit this phenomenon in light of the recent theoretical initiatives just cited and to do so in a way that will make the Mesopotamian evidence accessible to those involved in this wider discussion. Like many outside the field of Near Eastern studies, Blanton cites the work of Thorkild Jacobsen, which is both ground-breaking and out-of-date, as well as somewhat controversial in its basic interpretation of the cuneiform evidence. Sixty years ago, Jacobsen proposed that before the emergence of kings in ancient Mesopotamia, the earliest urban centers had developed political systems that allowed substantial power to reside in popular assemblies. Jacobsen (1943, 159) called this “primitive democracy,” choosing to use the term “democracy” in what he called its classical sense, “a form of government in which internal sovereignty resides in a large proportion of the governed, namely in all free, adult, male citizens without distinction of fortune or

16

Introduction

class.” He contrasted democracy with “autocracy,” a general term for all forms that concentrate political powers in the hands of individuals. Like so much of Jacobsen’s work, this synthesis was brilliantly drawn, and the organization of the gods into a council, without a simple king, remains difficult to explain from the political systems of the states that bequeathed us their texts. The best-known “creation” text, called Enuma eliˇs, presents the kingship of the Babylonian god Marduk as a new achievement rather than as an ancient inheritance. The text itself was composed toward the end of the second millennium, when there had already been human kings in Babylon for centuries, and in southern Mesopotamia for even longer. It is hard to know where to seek the cultural origins of this divine assembly, which recurs often in Mesopotamian literature, but it appears to derive from conditions much earlier than late second-millennium Babylonia. For a variety of reasons, however, the term “democracy” is not well suited to any ancient Near Eastern practice. The existence in ancient Mesopotamia of “popular assemblies” with any substantial political power has been questioned in the years since Jacobsen, and needs at least to be carefully qualified. Classicists reserve the category for the system that emerged at Athens, where power to make decisions and to enforce them was lodged in the body of citizens.36 I find the word “democracy,” however broadly defined, a barrier to understanding the diverse Near Eastern tradition of group-oriented decision making that may somehow stand behind the remarkable development in Athens. Besides the essential anachronism, the idea of primitive “democracy” artificially dissociates the more inclusive meetings from other forms that allowed only more limited participation. In spite of these cautions, Jacobsen’s terminology raises the question of how Athenian democracy was ultimately related – or not – to Near Eastern political patterns. Classicists consider this issue to be far from resolved, and although this study cannot address the specifically Greek situation that defines the origins of democracy, it can perhaps offer a clear picture of the early Near Eastern phenomena. Athenian democracy developed within the context of assorted Aegean traditions of group-oriented leadership, represented even in the earlier use of the word basileus for groups of leading peers rather than as lone “kings” (Drews 1983). Ian Morris describes a further change in the structure of Greek society that arrived in the eighth century, with the idea of a corporate polis community.37 From an outsider’s eastern perspective, the preexisting social landscape for Greek democracy bears unmistakable resemblances to widespread Mesopotamian custom, whatever the particular process that led to such a radical innovation. It is difficult to understand how the various political forms leading to Athenian democracy were not constructed from the raw materials of eastern group-oriented decision making, however unique the application. I leave aside this fascinating problem through the bulk of this study and then take it up once more by way of conclusion.


17

3. Southern Mesopotamia and the Lands Upstream When Jacobsen first suggested the notion of “primitive democracy” in Mesopotamia, he was thinking above all of Sumer, the southeastern part of modern Iraq where the Tigris and Euphrates Rivers meet and empty into the Persian Gulf. By choosing to work from the Mari archives, I have made my focus the parts of Mesopotamia farther upstream, especially to the west. This is not to underestimate the importance of collectivity in the southern Mesopotamian political tradition, and it is my hope that others will continue to investigate ancient Sumerian political patterns. My choice of western and “upper” Mesopotamia may also be provocative in light of the powerful monarchies that also characterize these regions. Piotr Steinkeller has argued convincingly that we can see a striking contrast between the relatively weak kingship of early third-millennium Sumer and a much more authoritarian version associated first with Kiˇs and then with the late third-millennium empire of Agade (Akkad; 1993, esp. 117–18). During the first half of the third millennium, Sumer thrived on the irrigation agriculture that allowed an unusually dense population, with a political system defined by local “city-states” (here, an acceptable term). Each polity had a central city and surrounding countryside, with little neutral space in between. Each was the private domain of an individual divine family, with the whole pantheon united under the god Enlil as its ultimate paterfamilias and final arbiter in border disputes. Territorial expansion was difficult, and unification was considered theoretically unthinkable. The ruler was called an ensi(k), conceived as the steward of the divine proprietor, selected by the god from the whole citizenry and reappointed each year. Community life revolved very much around the central temple of each city (pp. 116– 17). The famous Mesopotamian tradition of powerful kings seems to have come from outside Sumer, evidently from upstream, where Semitic speakers had established strong states at Kiˇs, Akˇsak, and Mari. Steinkeller concludes that this more hierarchical, centralized monarchy was common to all of the Semitic-speaking peoples of the north and west, and he cites Ebla as the prime example. It is here that my entire project stands in some tension with Steinkeller’s analysis, because the Semitic-speaking world upstream from Sumer is the very purview of the Mari correspondence. Is this world in fact dominated by the model of authoritarian kingship? At one level, I accept as possible the hypothesis that the idea of strong centralized monarchy came to Sumer from upstream. It seems to me, however, that Steinkeller has overgeneralized the non-Sumerian world. Although the Mari archives come from the period just after the end of the Sumerian system, they could be understood to represent the victory of western political power and its assumptions. The evidence from Mari, however, does not show the universal dominance of authoritarian kingship, especially outside the major centers of power. The

18

Introduction

collective political traditions of the regions north and west of Sumer were not shaped by the city-state system and influential role for the temple, but they could be both powerful and durable. One goal of this volume is therefore to underscore the variety of collective political forms that survived under the shadow of monarchy in ancient Mesopotamia.

4. Mari? One problem with synthetic studies of broad topics is that they are bound at many points to treat received material as known quantities, already adequately understood and available for application. From the start of this project, I wanted to explore both a particular body of evidence and large questions of social history at the same time, and so this study has always been conceived as building on the primary foundation of one set of archives: the discoveries from the palace complex at Mari. Although the Mari archives were discovered almost seventy years ago, their continued publication gives them a value little different from fresh archaeological discoveries, and the contents and implications of the new evidence are still familiar to few outside those involved with its publication. Mari cuneiform texts are particularly important to the study of Mesopotamia because of their geographical and social range. As observed earlier, the unique feature of the Mari archives is the massive correspondence, constituting thousands of letters, many of which deal with affairs outside the capital, often even outside the kingdom. This correspondence therefore allows a rare panoramic view of ancient Syria and Iraq together, within an interval so brief as to provide an album of all the regional players within a single generation. Both the writers of these letters and their subjects come from a wide social spectrum, at varying distances from the centers of power, with wildly different responsibilities and concerns. The closest comparison might be to the letters from Amarna Egypt, but that archive is roughly one tenth the size of the Mari correspondence (Moran 1992; Cohen and Westbrook 2000). These are therefore “Mari” letters only insofar as they were found at that city. Many of them were sent by officials stationed elsewhere in the kingdom. Others were written by envoys posted abroad, and still more came from foreigners in contact with the Mari court. The entire volume of ARM XXVIII consists of letters to Zimri-Lim from other kings. Zimri-Lim also kept in contact with military and tribal leaders whose business took them outside the immediate bounds of the kingdom. The extensive royal correspondence of Samsi-Addu with his son Yasmah˘ Addu, who ruled Mari on behalf of his father, comes from outside Mari. Another set was sent to Yasmah-Addu by his older brother Iˇsme-Dagan, from his royal seat at Ekallatum. ˘ Much of the royal correspondence that accumulated in the Mari palace had a political interest of one kind or another, touching on the push and


19

pull of kingdoms, towns, or tribal groups as they decided matters of war and peace, economic concerns, or justice. The view of these affairs may be either first- or second-hand, but the officials of the Mari kings knew that their masters would want to hear everything from hard news to rumors. The resulting stream of verbiage provides an ideal resource for study of many aspects of ancient political life. There is indeed a wealth of evidence for group-oriented decision making, but these riches represent only one facet of this larger view of regional politics. This book is constructed in a way that is intended to acknowledge the essential relationship of the collective forms to other elements of ancient political life. Even the mundane process of gathering references in texts to elders or assemblies, or the portrayal of polities in collective terms, yields unavoidable connections. For instance, the terminology of “elders” and “leaders” (sug¯agum) is applied to both tribes and towns. Under the reign of Zimri-Lim, the subdued Yaminite population is frequenty defined by its towns, though we find no comparable definition of Zimri-Lim’s own “Simalite towns.” The most common collective polity is the town, but in a world where many small kingdoms were defined by their central towns, the town and the kingdom may be indistinguishable. I avoid here the modern category of “city-state,” as carrying too many associations with very different political systems. There are towns that serve as settled centers of kingdoms that might be called archaic states, but it is often doubtful whether such states are best defined as extensions of those central “cities.” The Akkadian language provides its own distinction in the words a¯ lum and m¯atum, the first for the settlement and the second for the whole population that defines the kingdom ruled from that central settlement. In light of this kind of integration, it is crucial to address each individual aspect of the system with reference to the others. Ultimately, it is impossible to examine the collective or group-oriented element of ancient political life except as part of a dynamic system that involves both individual “exclusionary” and collective aspects, each with its own ideologies and spheres of activity. My focus on a single archive supplies an ideal real-life setting to demonstrate the advantages of this approach. The Mari archives display a lively, fluid interaction of collective and exclusionary modes, neither one operating independently of the other. To understand group-oriented decision making in towns, where it is displayed most often in the texts, we must understand all the dynamic complements of Mari society: organization by towns and tribes and realms (m¯atum), collective versus hierarchical authority, and the economic interplay of cultivation and pastoralism. The three central chapters of this book are meant to work together in a way that builds a framework for the collective political forms most visible in the town. Chapters 2 and 3, on the tribes and on the m¯atum “lands,” provide a political context for the towns of Chapter 4 without which they would be badly misconceived. In spite of the space devoted to setting up this context,

20

Introduction

it must be clear from the outset that I do not intend a systematic study of every one of these phenomena. My ultimate target at every stage is the political idea of the group in action, especially as expressed in the town. I recognize that by turning my attention to the larger political system, as well, I immediately neglect important components, especially the actual working character of kingship. For better or worse, I must leave such lacunae for others to fill.

e. a text-based study: comments on methodology It will be clear to the reader that this volume is ultimately a text-based study. I do not mean by this choice to present writing as the best or only evidence for ancient society. It is my goal to explore the texts from Mari in a way that invites those more familiar with other Near Eastern evidence, other regions, and other academic disciplines to extend the search into the material they know best. My preoccupation with written evidence carries with it certain opportunities and limitations. With texts, we can see how the people themselves named the categories of their existence. With texts, therefore, we are always addressing the ideologies through which existence was comprehended. At one level, then, my study is infused with collections of terminologies, each of which offers a different view of the society described. I have tried to respect the world embodied in each ancient word, from “town” to “elder” to “king,” recognizing that the reality behind every term is distant from anything in the modern world, so that we are too easily seduced into inappropriate interpretive frameworks. Too often, modern readers have assumed that they are dealing with fixed institutions, when in fact the words represent ways of speaking that have fluid points of reference. It is my hope that the project takes strength from the hard data of these words and their uses, without becoming trapped in the isolated categories as such. The reconstruction of ancient cultures has often divided those who study languages and texts from the archaeologists who work with other artifacts and the larger context of all such finds. Generally, archaeologists tend to work more closely with theory and methods from outside ancient Near Eastern studies than do those who study ancient writing. While this division is regrettable and will surely collapse some day from its own intellectual instability, it is perhaps worth considering that text-based study of the ancient world may have a particular affinity with cultural/social anthropology as opposed to archaeological anthropology.38 Insofar as cultural/social anthropology is focused on living populations, it depends especially on informants from the communities in question, and so on words. Understanding words and their true social context is as crucial to cultural anthropology as its famous models. This is the very claim that is always being advanced by those who work with ancient writing.

A Text-Based Study: Comments on Methodology

21

In his Critique of the Study of Kinship (1984), David M. Schneider makes an impassioned argument for the impossibility of understanding society without listening to language. Regardless of how effective his broader theoretical critique may be, Schneider was deeply dissatisfied with his own analysis of the social structures of the Yap people in the West Caroline Islands until he gave priority to language over stock kinship models. According to standard theory, an entity called the tabinau, or patrilineage, was the basic political unit of which villages were composed (pp. 13–14). On closer examination, however, he found that the tabinau had different meanings in different settings: a dwelling, person(s) related through ties of land, a group living together with different ties to the same land, or a place where something is founded – but never land without people. Naming is critical to the identity of the tabinau, and children receive names that are already in the tabinau, selected by a conference of elders. Those who hold an office, such as elder, act merely as the voice of the land, which is the basis for a corporate mode of political organization (pp. 21–3). I find it interesting that in his first description of the Yap social organization, which he considered too deferential to standard kinship theory, Schneider made no mention of corporate polity, which became more prominent in his second, language-sensitive description. All in all, Schneider demands that his field be more attuned to the traits peculiar to individual peoples, as expressed in their languages. “One of the most important instruments of analysis is the correct translation of Yapese words and sentences and utterances” (p. 37). More recently, Philip Carl Salzman (2000, 1–3 and passim) makes a similar plea for descriptive terminology that reflects the indigenous categories, and Susan Keech McIntosh also states that accurate “reportage” demands description that is as theory-neutral as possible (1999b, 12). Surely, then, our ancient texts offer equal possibility and require equal sensitivity as a basis for reconstructing ancient societies. Likewise, it must be possible to undertake this careful study of words and language in dialogue with theory about the nature of societies. Keeping in mind the advantages of having access to indigenous languages and categories, it is worth acknowledging one more time the pitfalls of translation into modern and English terms. Every social category embodied by an English word carries with it the wrong identity or an inaccurate range of identities. Our “city” and “town” are essentially large settlements, defined especially by their concentrated populations. In common parlance, cities are bigger and more complex than towns, though an urban commuter may “go into town” for work. In American English, villages and hamlets are either relics of the past or quaint tourist destinations in other people’s countries (with some northeastern exceptions). A “town” thus spans everything from a rural crossroads settlement to the center of a major city, so I have preferred “town” as a translation for the Akkadian word a¯ lum, which has similar range. Even so, the English word deceives us. The early built-up centers of northern

22

Introduction

and western Mesopotamia often retained an old function as meeting places for population spread through the surrounding countryside. With accumulating wealth and power, these sites often acquired palaces and temples, along with housing for the personnel who supported them, but they did not necessarily grow up around a burgeoning population. Often excavations are simply too incomplete to demonstrate the presence or absence of significant residential districts. This is true of Mari, after decades of excavations. All the categories of this study suffer the same inadequacies. The “tribe” is a loose term applied especially to groups of non-European origin, except in secondary use to describe large extended families. In theoretical terms, “tribal” social organization must be distinguished sharply from “pastoralist” subsistence by flocks of sheep and goats, and pastoralist subsistence in turn from mobile or “nomadic” patterns of residence. In the Mari archives the ubiquitous word hanûm (Hana) can make it difficult to maintain these ˘ hanûm, ˘ most often written ha-na, probably means distinctions. The word ˘ ˘ nomadic residence. “tent-dweller” (see Chapter 2), indicating mobile or There is another word for “grazing land” and its flocks (nawûm), but in common speech, hanûm referred to those tribesmen who took the flocks ˘ into the nawûm pastureland. Although the word hanûm is applied to differ˘ form Hana often refers ent tribes and is not essentially a tribal name, the to “(our) tent-dwellers,” which in the leadership circle of ˘Zimri-Lim meant Simalites in particular. In certain settings in the Habur region, the Hana ˘ name even seems to indicate a particular political entity or “tribe,” and˘ the common Simalite use carries a strong tribal identification. I can offer no adequate translation for the Akkadian word m¯atum, the political unit that is recognized to make its own decisions in matters of war and peace. It is most often ruled by kings but is not defined thus; it is territorial rather than urban in geographical terms, but is essentially the people, not the place. I translate it as “land,” with the dictionaries, for want of a more accurate term. The political character of the m¯atum suggests the “state,” and the match is attractive in the case of the larger examples. In terms of anthropology and social history, however, the “state” is a contested category, claimed by many to represent a certain baseline in the achievement of political size and complexity, implying baggage that the simple m¯atum cannot carry. It is not clear how small or undeveloped a m¯atum could be, but the category was not measured in terms that keep step with the modern debate. I discuss “states” in Chapter 3, but I use the word with some trepidation. Similar entanglements come with every attempt to translate Akkadian or West Semitic categories of leadership. The texts themselves allow some overlap of the words ˇsarrum and sug¯agum, which I render “king” and “leader,” and closer examination only worsens the confusion. In Akkadian, ˇsarrum is the standard word for “king,” but earlier and western usage indicates a broader meaning, perhaps as “official” or even “leader.” West Semitic usually presents us with forms of the word malkum for “king,” and the sug¯agum

A Text-Based Study: Comments on Methodology

23

seems to have originated as a tribal category (see Chapter 2). Political groups fare no better. Akkadian ˇs¯ıbum does mean “elder,” but as a group, these may represent “leaders” of varied definition. The Akkadian puhrum can be translated literally as “assembly,” as long as one thinks of the˘ generic English “meeting,” with no definite size or level of inclusiveness. In the end, the difficulty itself underscores the benefits of working from the ancient texts that allow us glimpses of the native categories, which themselves were applied over ranges of meaning. For readers who are familiar with the details of either the ancient languages or the modern interpretive categories, I acknowledge here the gravity of the problems. I have chosen my terminology as carefully as possible and offer definitions as I proceed.

2 The Tribal World of Zimri-Lim

During the period of the Mari archives, a large proportion of the population identified themselves by tribal social structures that incorporated both settled and mobile communities, whose location overlapped the boundaries of fixed kingdoms across wide swaths of ancient Syria, Turkey, and Iraq. The tribal peoples within the Mari kingdom belonged mainly to two confederations that together represented a coherent duality: the Simalites of the “left (hand)” and the Yaminites of the “right (hand),” probably directional references to north and south. One remarkable discovery of the current Mari publication team is the depth of continuity between urban and nonsedentary segments of the region’s population. Far from being set in irrevocable opposition as nomads and townspeople, or even in complementary opposition, this large group considered itself part of one social fabric, divided not by mode of life or place of residence but according to traditional associations of kin. It is impossible, then, to understand the kingdoms of our Mari period without also exploring the intersection of tribe and town and “land” (m¯atum). Beyond this value for context, however, tribal social organization is strongly correlated with group-oriented forms of decision making that balance the power of “exclusionary” hierarchies and their individual rulers. The entire Yaminite people are defined as a confederation of five tribes, each with its own ruler. Zimri-Lim seems to have been acknowledged as the one Simalite king, but his actual dealings reveal the monarch’s need to negotiate frequently with his own tent-dwelling (Hana) tribesmen to win their military support. These negotiations generally ˘take the form of large inclusive meetings, attended by one or more representatives of the king. Under the rule of Zimri-Lim, the individual Simalite “divisions” (gayum) retain no political function, which perhaps makes it all the more impressive that this collective identity never submerged beneath the kingdom’s active political categories. We cannot conclude that all group-oriented decision making attested in the Mari archives originated in tribal political traditions, because 24

Tribally Organized Pastoralists and the Amorrites

25

ancient towns may have preserved their own collective practices, but the tribes represented one powerful structural counterweight to exclusionary governance. Zimri-Lim’s kingdom was governed by two separate lines of authority. The districts (hals.um) of the inherited Banks-of-the-Euphrates (Ah Purattim) dominions˘ were overseen by “governors” (ˇsa¯ pit.um), each with ˘his own local palace and “deputy” (laputtûm) to allow centralized administration. Residents of the Ah Purattim districts were liable to census as a basis for ˘ tribespeople who could be classified as Hana (“tentmilitary draft. Simalite ˘ They acdwellers”), however, were free from this administrative hierarchy. knowledged a “leader” (sug¯agum) who was affiliated only by tribe, not town, and two Simalite chiefs of pasture (merhûm) served as respected leaders ˘ over the entire tent-dwelling tribal population. The merhûms reported only to Zimri-Lim and his immediate cabinet of top advisers,˘ and stood outside the district system. These Simalite Hana represented the backbone of Zimri˘ Lim’s army, but he could never simply command their active support, and the merhûms had to join in tribal meetings and await the collective decision (i.e., see˘ the rihs.um). Zimri-Lim did not take the census of his Hana. The ˘ require special attention because their office˘ straddles sug¯agum “leaders” lines of authority defined in tribal and town terms. Although they occupy a key position in Simalite Hana leadership, the sug¯agums are best known at ˘ Mari as town leaders. We cannot understand the larger structure of the Mari kingdom, especially during the reign of Zimri-Lim, without examining this entire tribal political system. There have been numerous studies of the tribal peoples and “nomadism” in the Mari archives, including a long work in preparation by Durand that will expand tremendously our current knowledge of these phenomena, and my own chapter is not intended to cover the same terrain systematically. The central phenomenon of this book is rather the collective political mode that is most often visible in the town, and I find myself addressing tribes and tribal political order because these are woven seamlessly into the urban landscape of the Mari archives. My focus is the political organization of the tribes, especially as this finds expression in Zimri-Lim’s kingdom at Mari (see the next chapter). To this end, after dealing with some introductory matters, I turn to the contrasting tribal terminology and structures of the Simalite and Yaminite confederacies, in order to establish a framework for all further analysis. Then I discuss the key individual leaders who represent the tribe under Zimri-Lim, the sug¯agum and the merhûm. The sug¯agum plays an important role in distinguishing Yaminite and ˘Simalite political structures, at the same time as he is the one individual leader whose very office can be defined by the town he serves. I conclude by considering certain implications of the contrasting Simalite and Yaminite structures, first for understanding the Hana “tent-dwellers” under Zimri-Lim, and then for evaluating specific ˘ Yaminite political terminology in the Mari archives.

26

The Tribal World of Zimri-Lim

a. tribally organized pastoralists and the amorrites Even in times much more recent than those of the Mari archives, tribal peoples are notoriously slippery subjects for academic study, in large part because they are foreign to both the writers and their sources. The sources are mostly written from a distance by outsiders viewing the tribes with hostility or some other bias. . . . They rarely deal specifically or in reliable detail with the basic social and economic organization of tribal communities; and they mention individual tribes only when prominent in supporting or opposing government, when involved in inter-tribal disorders, or when transported from one region to another. (Tapper 1990, 56)

Against this pattern, the Mari archives from the reign of Zimri-Lim offer a rare view of tribal organization that includes perspectives from within the tribal societies themselves. Zimri-Lim’s thirteen-plus–year reign represents the brief success of a tribal chief in reestablishing the power of his tribe over a city center and kingdom that had been lost to his kin some years earlier. He was in effect a tribal king over what might be called a tribal state,1 and though his correspondence and his administrative archives do reflect the world of sedentary and nontribal urban society, the world of the tribes has a high profile in these texts and is often described by people who maintain current ties with its structures. Some of the letters were even sent by the tribal rulers and leaders of mobile pastoralist communities themselves. If we want to learn about tribal society in the ancient Near East, there is surely no better source. It is difficult to begin a study of “tribal” peoples in the Mari archives without some larger sense of the implications embedded in this identification. At the same time, discussion of the “Amorrite” tribes in these texts is usually joined to ideas about their pastoralist or “nomadic” activities, and the entire cluster requires some reflection. As with my previous treatments of issues and literatures that reach beyond Near Eastern studies, I am selective. My goal is to provide a larger context for the specific problems presented by the Mari evidence.

1. Using the Word “Tribe” One cannot use the word “tribe”2 without defining further what one has in mind. Richard Tapper (1990, 50–1) observes three concepts of the tribe in broad academic discourse: r a type of primitive society, attached to specific cultural-linguistic groups and contrasted to “states”; r more precisely, a type of society in a hierarchy from simple bands of hunters to “tribes” to more complex chiefdoms and states, characterized by “kinship” and “descent” in their social-political organization;


27

r political groups defined by territory, divided by segments the descent framework of which is to be distinguished from the political-territorial structure of tribes as such. (Evans-Pritchard) These categories do not exhaust the applications of the term, but they offer an immediate sense of some of the basic divisions. Working from the second concept, which is typical of the evolutionary approaches of Sahlins (1968), Fried (1975), and Service (1975), Norman Yoffee refuses to apply the word “tribe” at all to the Yaminites, Simalites, and other such “ethnic groups” of the Mari archives, because a proper tribe is “a social organization without independent units of authority and in which surplus production and storage are discouraged through mechanisms of wealth and power leveling” (Yoffee 1988, 51; cf. Kamp and Yoffee 1980, 88). a. Tribe and Tribal. I prefer to embrace the words “tribe” and “tribal,” with all of their pitfalls, to take advantage of their familiarity and their wide use. Tapper offers a practical definition for the study of modern Middle Eastern societies: Tribe may be used loosely of a localised group in which kinship is the dominant idiom of organisation, and whose members consider themselves culturally distinct (in terms of customs, dialect or language, and origins); tribes are usually politically unified, though not necessarily under a central leader, both features being commonly attributable to interaction with states. (Khoury and Kostiner 1990, 5)

I use the word “tribe” only to describe the individual units of the Yaminite people, and I propose that the five Yaminite tribes could be designated by the word li mum. On the Simalite side, the comparable unit was called a gayum, but because the term was used for at least two different levels of social organization, I translate it simply as “division.” Following Tapper again, I call the overarching Yaminite and Simalite groups “tribal confederacies,” suitable to both their evident size and the fact that the individual Yaminite tribes are united politically only as a coalition, having no single ruler (Tapper 1990, 53). As such, the confederacy itself is one important expression of what Blanton and his colleagues call “corporate” political structure (see Introduction, p. 15). The adjective “tribal” merits broader application, as reflected in the definition of the confederacy just given. Anthony Giddens provides a fresh approach to “tribal society” that is not bound to an evolutionary scheme of progressive types. Tribal society is a “loosely organized inter-social system” characterized especially by easy and frequent personal contact (high “presence availability”) and relatively little separation by distance (“space-time distanciation”).3 One can see immediately a problem with this formulation as it will relate to the tribally defined Yaminites and Simalites who were spread across most of ancient Syria. As part of an “inter-social system,”

28


tribal peoples need not be divided into strictly bounded units, each a totally distinct “tribe,” a procedure that has never fit well the fluid organization of such groups. Anne Porter has worked to take advantage of Giddens’s progress in regarding tribal societies as a loosely knit system, rather than a collection of neatly bounded units. For Porter, then, the crucial question is how tribal peoples are organized socially and politically in order to transcend the constraints of distance. Tribal organization is not limited to a small scale, whether in terms of population or of geography. Tribal peoples can overcome distance by “knowledge of place,” especially as kept alive through shared ideology and ritual (Porter 2000, 79–92). Porter is concerned especially with the third-millennium site of Tell Banat, which she excavated with Thomas McClellan. This site is far upstream of Mari in the Euphrates valley, north of Imar, but it still belongs to the same broad zone of largescale pastoralism. So far as Banat is concerned, with its massive monument to the community’s ancestors, Porter concludes that in a regional economy based primarily on long-distance pastoralism, such a structure must be explained in the framework of tribal organization and ideology of the sort described explicitly in the early second-millennium Mari correspondence.4 Tribal peoples who spent one part of the year in the steppe with their flocks would have returned to Banat on some regular basis, in part to maintain their collective identification with the common ancestors at a shared burial place, and so to affirm a shared tribal identity as well. b. Tribes and States in the Mari Archives. Rather than abandon “tribal” terminology because of disagreements about how to define its precise use, especially in evolutionary typologies, I prefer to find an application that fits our actual evidence at Mari. In the Mari archives, we find two primary modes of identifying social and political affiliations. One is by place, especially in relation to a settlement, whether a tiny village or a large city. At the same time, however, the Mari letters show us another form of identification that is distinguished from the grid of states. Some peoples are defined by names that simply transcend the boundaries of settled polities. This is most visible with groups affiliated under the Yaminite banner as “children of the right (hand)” (bin¯u yamina), because they do not even share a single king. Under Mari dominion, king Zimri-Lim could require each Yaminite king to supply him men from the Yaminite towns within the Mari districts to fight with the Mari armies.5 Many Yaminites lived far beyond the bounds of Zimri-Lim’s kingdom, however, such as those in the upper Balih who kept close ties ˘ 6 An unpublished with the kings of the Zalmaqum towns, including Harran. ˘ letter defines the Yaminite and Simalite (“Hana”) spheres in terms that reach far beyond the frontiers of individual ˘kingdoms. As in most of the


29

correspondence from Zimri-Lim’s circle, “Hana” refers specifically to mo˘ bile Simalite pastoralists: While the land of Yamhad, the land of Qatna, and the land of Amurrum are the ˘ – and in each of those lands the Yaminites have their fill range(?) of the Yaminites of barley and pasture their flocks – from the start(?), the range(?) of the Hana has ˘ been Ida-Maras..7

Yamhad is the great western Syrian kingdom whose center was at Aleppo, and ˘Qatna and Amurrum were the other major states of western Syria, to the south and west of Aleppo.8 All of these represent political domains separate from both Mari and Zalmaqum, not to mention from the Yaminites themselves. Another letter is sent to the king of Mari by two high-standing Yaminites from the tribe of the Rabbˆ u, after they have received no help from the other regional king whom they serve, Yarim-Lim of Yamhad, the ˘ major realm to the west.9 Elsewhere, this same Yarim-Lim tells an envoy of Zimri-Lim that the Yaminite kings are not residing “in the midst of my land (m¯atum),” explicitly acknowledging their mobility across the boundaries of kingdoms.10 These are not the only Mari examples of such tribal identification, but the Yaminites show the clearest geographical spread. Other examples of tribal identities surviving across territorial political domains include the Numhâ, ˘ who are specially associated with the kingdom Kurdâ, just south of the Jebel Sinjar, but whom Samsi-Addu claimed as his ancestors in the kispum ritual found at Mari.11 Under slightly later Babylonian rule, the twin settlements that constituted ancient Sippar could be named by two of the Yaminite ˇ tribes, as Sippar-Yahrurum for the part devoted to the sun god Samaˇ s, and ˘ as Sippar-Amnanum for the part devoted to the goddess Annunitum.12 In the early second millennium, the southern Mesopotamian kingdom of Larsa came to be identified as Emutbala or Yamutbal, the same tribal group that defined the people of Andarig during the time of Zimri-Lim. This association appears to have arrived with the Amorrite Kudur-mabuk dynasty (Stol 1976, 63–9). Although proper examination of the political category called the m¯atum is reserved for Chapter 3, I need to comment briefly here on its relation to identification by place and people. The m¯atum is the political unit that can make its own decisions about making peace or waging war. At root, the m¯atum was defined not by place but by population, and in practice, it was adapted to peoples who were defined in both tribal and nontribal terms. By the early second millennium, kings and their kingdoms dominated the Mesopotamian landscape, and most often we see the m¯atum as what a king rules. In fact, however, this royal domain was not a tract of territory but the sum of all the people who accepted the king’s leadership. Even aside from the question of nomadic movement or tribal identities, we should not

30


imagine the political map of ancient Mesopotamia to have consisted of states with fully enclosed boundaries of the sort drawn on modern maps. Political control would have been defined in terms of population and routes of communication, with greater and lesser degrees of effective power and certain regions that could not be fit into the framework at all. This definition of a world of m¯atum polities accommodates entirely the possibility of active tribal entities that could straddle the territorial heartlands of two or more m¯atum kingdoms, and such fluid relationship to territory may have been more common in earlier periods. Kings who desired full domination over as large a population as possible would naturally tend to pursue total political control over core territories, and the m¯atum then took on a strong geographical aspect. It must be kept in mind, however, that although the m¯atum is most often translated as “land,” the word does not define identity by place and therefore at some level has much in common with tribal categories. The territorial aspect of the m¯atum “land” is evident in the text just quoted for its description of Yaminite and Simalite pastoral ranges, especially where Yaminite pastures are located geographically by the “lands” of Yamhad, Qatna, and Amurrum. We have an identification that reaches beyond˘ the political definitions of town and kingdom, but why should this mode of identity be called “tribal”? How do we know what words to translate as “tribe,” or with “tribal” associations? The English word “tribe” in common contemporary usage may denote a people whose primary identity is understood to derive from shared family ties, rather than residence or citizenship.13 Our Mari evidence is clearer regarding geographic distribution than regarding the relationship of these peoples by family ties, treated in anthropological dialogue in terms of “kinship” and “descent.”14 In general, however, such social structures that transcend state boundaries and extend over considerable distances are linked to kinship systems.15 Ira Lapidus applies the word “tribe” to peoples with ideologies of common ancestry, where the larger scale involves “political entities that organize fragmented rural populations” (1990, 26). Mari’s archives alone show that we need not limit ourselves to “rural populations” but that these form one dimension of peoples “fragmented” by distance. Tapper suggests that it is best not to apply this combination of distance (or territory) and kinship too narrowly, forcing the native terminology to conform to an “exact terminology for classificatory and comparative purposes.” Actual terminology tends to be more fluid and ambiguous, with the specific words no more precise than the English words “family,” “group,” or the like. “Tribe, I suggest, is rather a state of mind, a construction of reality, a model for organization and action” (1990, 55). Even these difficulties, however, show that the Yaminites and Simalites of the Mari archives fit naturally into the larger discussion of “tribal” peoples. c. Tribes and Kinship in the Mari Archives. It should be clear enough from the previous discussion that tribal identities in the Mari archives represent an


31

alternative to identification by town or even by “land” (m¯atum). The very use of the “tribal” modifier leads to the question of kinship or descent ideology, however. Unfortunately, the evidence is scanty, and we do not have enough information about kinship ideology in nontribal communities to allow a proper comparison. It appears that what we call “tribal” organization is distinguished more by geographical range across more localized political entities than by any greater reliance on the idea of common descent than might be found elsewhere. At the highest level in the social hierarchy, the Yaminite and Simalite confederacies are defined within the family metaphor as “children” of right and left, but the relation between them is based on the duality of directions, not on kinship as such. Durand has proposed a letter from the Yaminite ruler Hammi-iˇstamar as evidence for the larger unity of the two confederacies, but ˘ text does little more than underline the futility of preoccupation with the the local distinctions in the face of Elamite invaders who will mow down one group along with the other. The writer poses the question, Will they ask themselves whether this town is Simalite and that town is Yaminite?16 These outsiders will fail to recognize the basic identities that separate the two tribal rivals, though the very warning calls for a cooperation against foreign assault in a temporary alliance that may assume the overarching unity of Yaminite and Simalite peoples. No actual kinship is invoked as the basis for such common ground, however. We do find occasional references to kinship among members of tribal groups, but far too little to allow reconstruction of one or more systems. In one case, an official from the Yamutbalite kingdom of Andarig writes that his own people, evidently the Yamutbal tribal people, “have shared brotherhood and branches of the Hana (people) since days gone by.”17 Both the words ˘ u) and “branches” (purs¯atum) could reflect a lineage for “brotherhood” (athˆ ˘ the relationship between large tribal groups, in this ideology that applies to case the Simalites and the Yamutbal peoples as whole entities. Within the Simalite and the Yaminite peoples of Zimri-Lim’s Mari kingdom, I am not aware of any kinship language that is used to define relationships between tribes. Below the level of the tribe, the picture is not much clearer. One Yaminite legal document commits 150 iku of agricultural fields to the mobile component (hibrum) of the “children of Awin” (DUMUmeˇsA˘ the Yaminite Rabbû tribe.18 At the very least, wi-in), perhaps a clan among we are dealing with a large extended family. All thirteen parties to the transaction are said to belong to “the house of Awin” (E´ A-wi-in, line 3). There may be much more to a kinship ideology that could well underlie the tribal categories visible in the Mari archives, but if so, these relationships did not need to be invoked in the practical concerns of the letters. Even if this is true, we must not treat descent-based relationships as the sole domain of the tribes. The ideology of kinship suffuses the political relationships between the kings of ancient m¯atums, also. As discussed at length by Bertrand

32


Lafont in his recent study on international relations in the Mari archives, kings always chose between the language of father/son or brother in their formal address. For diplomatic purposes, everyone is family (2001a, 232–8). These are individual relationships between rulers, not between their peoples, but we must also remember that the inhabitants of any town or land may be called its “children” (or “sons”). The main difference between a “son of Simal” and a “son of Ekallatum” is defined by the geographical-political ranges of the two types of social organization, not by the presence or absence of kinship ideology. All in all, there is indeed a contrast between tribal and nontribal organization, but kinship ideology plays some role in both. This fact might lend support to the notion that kinship was the original coin of all social definitions and that identification by place was a secondary development. d. Kinship Ideology and Tribal Identity in Action. I have embraced the terminology of tribe and tribal society for this study of the Yaminites and the Simalites of the Mari archives, but there remains a question about the effective use of these tribal affiliations in real social behavior. Some anthropologists have challenged the very relevance of lineage or descent to understanding how societies function. Because of the barriers to understanding actual kinship ideologies in our material, it is difficult to respond directly to this challenge as it relates to the tribal peoples of the Mari letters. Technical kinship aside, however, the tribal categories of the Simalite and Yaminite confederacies do appear to have played an essential role in shaping social and political action. Emrys Peters (1967) argued that even where lineage identities do exist, the people do not in fact live by them. Their loyalties are not determined by genealogically defined relationships, they do not pursue their economic interests based on solidarity with lineage groups, and their political alliances do not follow these links.19 Some years later, Adam Kuper (1982) offered a bleak critique of the entire “lineage theory” developed by Evans-Pritchard. According to Kuper, the whole model has been “theoretically unproductive,” especially because in his view, there do not seem to be any societies at all in which “vital political or economic activities are organized by a repetitive series of descent groups” (p. 92). Without defending any particular version of lineage theory, I nevertheless find that this critique underestimates the relevance of kinship. Philip Carl Salzman (1978) presents a persuasive alternative. If the lineage ideologies have no concrete impact, why have them? he asks. Salzman reexamines some Middle Eastern tribal peoples, including Peters’s own camel-herding bedouin of Cyrenaica in Libya, and finds that such groups acted according to lineage, especially under crisis. “Lineage ideology is, during times of stability, a social structure in reserve” (p. 627). For example, the Yomut Turkmen of Iran normally live by a combination of pastoralism and rainfall


33

cultivation, occupying stable territories at the local level. Interestingly, they seem more nomadic than their productive activities require. Even those who specialize in cultivation and could build permanent dwellings choose to live in mobile “yurts.” Yomut political relations reflect a typical “territorial” (versus “lineage”) organization, but they maintain a living, active kinship ideology. Why? Salzman finds the answer in Yomut history. At various times in the past, the Yomut found it necessary to pull up their settled roots and retreat together deeper into central Asia. On such occasions, relations between Yomut groups could not be based on stable territorial relations, which were lost. Lineage ideology would have provided a framework for social life during such times. In support of this hypothesis, Salzman argues that the unnecessary level of mobility serves the same purpose, along with herds larger than the normal economy requires. “The Yomut must maintain their nomadic gear, their mobile productive resources, and the mental availability of migration” (pp. 628–9). Roy Rappaport (1968) describes a system that shares Salzman’s emphasis on the concrete applicability of kinship ideology, while suggesting a different scenario for its effective use. The Tsembaga of New Guinea live by supplemented cultivation, with rights to hunting and gathering land shared by all, and cultivated gardens held by clans. The degree to which land sharing actually followed Tsembaga lineage ideology varies. In time of defeat at war, the people do not move to new territory, unlike the Yomut, but disperse to live with people considered to be family. Here kinship ideology allows groups to include refugees as distant relations.20 One study of early North American “tribal” societies, so named by typology rather than known organization, concludes that tribal social integration appears to have become more intense as risks from local environmental unpredictability increased (Braun and Plog 1982, 518). In the Yaminites and the Simalites of the Mari archives, we have further evidence of tribal identities that govern behavior in a way that cuts across local geographical and political lines, even if we cannot define these identities in terms of kinship. Many Yaminites live in fixed settlements within the core districts of the Mari kingdom, especially in the Terqa district. They are sworn to loyalty by the Mari king, Zimri-Lim,21 and they owe him military service.22 At the same time, however, they maintain their distinct Yaminite identities and loyalties. Even in administrative terms, the Mari government is forced to accept the continuing force of Yaminite affiliations. It is the five Yaminite rulers who are held responsible for delivering their promised military contingents to the Mari armies, thus acknowledging an order and an authority completely outside the Mari hierarchy.23 Rather than abandon the terminology of “tribes” and the ideology of kinship as irrelevant, the Mari evidence begs to be understood in a theoretical framework that is adapted to explain the evident reality of these categories in this ancient context.

34


2. Pastoralism and Mobility At least when studying the ancient Near East, it is common to hear about “pastoralist nomads,” “nomadic pastoralists,” and the like, people who tended their flocks of sheep and goats on the fringes of settled society. Ancient farming communities no doubt kept their sheep in local fields, but there was always an alternative way, combining the grazing of flocks with mobility. Mari has long provided some of the richest ancient evidence for this phenomenon, which was unquestionably real, but the interpretation of this evidence has varied radically in eventual response to changes in anthropological theory. A generation ago, Michael Rowton was the principal agent of this fresh application, with his ground-breaking effort to show how deeply enmeshed the pastoralist nomads were in the network of states with their settled populations and agriculture-based economies. I return to Rowton below. It is impossible to deal with the tribal populations of the Mari archives without addressing pastoralism as well, along with the mobility that contributes to the need for kinship identities that can transcend distance. “Pastoralism” itself is not a type of society, and certainly not a form of organization, but rather a way to survive and a form of “production.” Pastoral production, like cultivation of crops, involves a direct appropriation of natural resources that takes place at two social levels: the domestic group or individual that keeps the herds or flocks and the larger pastoral community that shares rights to pasture (Bonte 1979, 203–4). Another important problem shared by pastoralist and cultivation production is access to and distribution of water, and one reason for resorting to pastoralism is its more flexible water needs.24 In actual practice, pastoral production is integrated into working societies in a variety of ways, and discussion of actual pastoralism requires some sort of sliding scale in order to define degrees of blending with cultivation. A. M. Khazanov identified five basic forms (1984, 19–24): r pastoral nomadism proper, without any supplementary agriculture (rare); r semi-nomadic pastoralism, with some supplementary agriculture; r semi-sedentary pastoralism, with a strong role for agriculture but still involving relatively short seasonal migrations; r herdsman or distant-pastures husbandry, without migration but with specialized herdsmen; r sedentary animal husbandry (not characteristic of “primitive” or “traditional” societies). Khazanov also distinguishes a number of regional ecological types, acknowledging that the “Near Eastern” (versus “Middle Eastern”) type is highly diverse, with varieties of semi-nomadism and herdsman husbandry in


35

Palestine, Syria, and Iraq. In this ecological setting, routes of migration are less stable and the migrations themselves less regular than in Eurasia, to the north. There is considerable dependence on agriculture, except among the pure nomads of Arabia and the Sahara (pp. 53–8). For Khazanov, the essence of pastoral nomadism (his term) includes the economic dominance of pastoralism, herds without stables, seasonal mobility with a majority of the population taking part, and an orientation of production toward subsistence (p. 16). These categories provide a useful orientation to the complexities of pastoralism in practice, but the use of “nomads” as the point of reference predisposes the analysis toward an assumption that these people are not going to be the driving force of any regional economy. Khazanov’s expectation that pastoralist production will be focused on basic subsistence – at least in his “traditional” and “primitive” societies – reflects this assumption of small scale. Porter proposes that a cleaner definition concentrates on methods of subsistence and leaves mobility as a separate issue. This choice leads to four types of food-producing subsistence (2000, 30–1): r pure pastoralism r supplemented pastoralism r supplemented cultivation r pure cultivation Detachment from place may occur among pastoralists “even in cases where there is little seasonal movement involved in the upkeep of animals, for it allows for future movement, always a possibility inherent in the instability of pastoral systems” (p. 34). Pastoralism in the Near East appears to occur very early, in a continuing interaction with cultivation, in a range of forms. This range seems to have included mobile expressions even with the first development of agriculture, though the specialization of mobile pastoralist production complemented the early specialization of cultivation. Lees and Bates (1974, 187) suggested some time ago for southern Mesopotamia that “there was a direct relationship between the development of nomadic pastoralism as a specialization involving substantial populations, and the development of canal irrigation.” The labor demands of irrigation farming made it harder for individual households to engage in both cultivation and animal husbandry, so some of the population began to specialize in husbandry. These households were forced to move farther and farther away from the irrigated plains to find land. Buccellati adapts this hypothesis to the Euphrates River valley, with essentially the same mechanism in view. Along the Syrian Euphrates, however, urbanization did not begin until the mid-third millennium, so Buccellati concludes that “nomadization” would have begun only after that, in response to demographic pressure, so that the main change would have occurred just in time to put pressure on the southern Mesopotamian urban

36


society of the Ur III dynasty, at the end of the millennium. In effect, then, the first mobile pastoralists of the Syrian steppe were the people called “Amorrites” by the Sumerians of the late third millennium. Mari-period pastoralism was therefore a relatively new thing (1990, 98–102). Even if this sort of demographic pressure did occur in some parts of SyriaMesopotamia, Lees and Bates’s model seems too narrowly cast to explain the whole phenomenon of Near Eastern pastoralism. In a detailed examination of evidence from earlier periods, Juris Zarins rejects this hypothesis in favor of a more gradual explanation that locates the roots of “pastoral nomadism” in the time of earliest animal husbandry, in the late seventh millennium. He finds that mobile pastoralism had characterized the Syrian steppe and desert long before the advent of archaic states in the river valleys, based on evaluation of sites from those dry regions.25 Joy McCorriston portrays a gradual process that begins in the Neolithic period but intensifies with the development of agriculture. The motive force is not demographic or resource pressure but the increasing availability of products for exchange, based on new agricultural surpluses (1997, 526). By the late fourth millennium, there has already been a major new and widespread dependence on wool-bearing sheep for wool production (p. 521). In one sense, then, an expansion of back-country pastoralism in the third millennium could have resulted from a technological innovation: the production of textiles from wool instead of flax (p. 519). Because this new fiber resource could be produced away from the prime agricultural land, mobile pastoralism may have expanded significantly just to meet the economic demand. Nicholas Kouchoukos now proposes that the early phases of Uruk’s emerging economic power in the late fourth millennium may have been driven in part by cloth production that depended on pastoralism (1998, 252–3).26 The sheep and goats involved could not have been supported in such large numbers in the marshy environment of the Mesopotamian delta (p. 289). This line of analysis will have to face ongoing evaluation by specialists in the period, but it is intriguing to allow pastoralism to intrude into the moment of early urban growth that is often treated as the special preserve of new irrigation agriculture. During the late fourth millennium, the dynamic culture of Sumerian Uruk exerted a powerful influence upstream, across northern and western Mesopotamia. At the turn of the millennium, this influence declined sharply, and northern Syria shows a noticeable cultural shift, in what is called the Ninevite V period, from a striking type of characteristic pottery. The role of pastoralism in early third-millennium Syria has been debated in relation to an intriguing set of sites along the Habur River, including Tell al-Raq¯ai and ˘ for storing grain that are far too large Tell Atij. These sites included facilities for the size of the settlements themselves. Who then created them, and what kind of polity do they reflect? Various hypotheses have been advanced, but Frank Hole proposed that the silos contained grain for the flocks of Habur ˘ idea region pastoralists (1991, 17–19). McCorriston shows sympathy for the


37

that specialized pastoralism with surplus animals ought to be taken into account, even as she considers the evidence for these sites to be inconclusive (1995, 36; cf. Zeder 1995, 29). Hole’s approach is now developed further by his student Kouchoukos, who argues that the silo structures originated in the late fourth millennium, before any of the sharp urban growth in the region during the following period (1998, 406–10). Aside from the hard evidence for one solution or another, the discussion offers a fascinating synopsis of how pastoralism is viewed by most archaeologists. Glenn Schwartz and Hans Curvers observe that the scale of the storage and sophistication of the architecture indicate an association with complex polities of a sort not found near the sites themselves, and assert that we simply lack the ethnographic parallels for such a large-scale system constructed by pastoralists (1992, 417). More strongly still, Michel Fortin declares that such elaborate structures could not have been built by nomadic peoples, who are by definition not versed in the art of construction. Nomads would never store grain for the following year, he says (2000, 122–3, 126). Margueron, the excavator of Mari, asks why we have found no such silos for the early second millennium, when we know that “nomads” were active from Mari textual evidence (2000, 107–8). For the moment, it seems that the Ninevite V storage sites provide a stimulus for alternative historical reconstructions, without sufficient evidence to allow general agreement on any one of them. This brings us back to the evidence for large new towns in thirdmillennium Syria. Specialized mobile pastoralism was firmly established in Syria-Mesopotamia by the beginning of the millennium, as indicated by McCorriston’s work on the preceding period, though it is difficult to judge its impact in individual settings. During the early third millennium, before the urban wave takes firm hold, a number of new settlements appear that are enclosed by large circular mounds. One of these is Mari. In general, these sites are found in or adjacent to the dry steppe, and Lyonnet argues that their form served a pastoralist settlement (1998, 180–3; forthcoming). The backdrop for understanding early second-millennium Mari is this urbanism of the previous millennium, and this society may turn out to be much more like the world of the Mari archives than is generally thought. Pastoralist rulers such as Zimri-Lim may have been quite familiar to the earlier centers, which may have long depended on mobile sheepherding populations for their very urban strength. People such as the “Amorrites” may have represented a fresh threat to southern Mesopotamia of the Ur III kingdom, but the local Syrian scene may not have changed radically. Third-millennium urbanism in Syria was disrupted from the opposite direction, with the westward expansion of the Sargonids of Akkad. The mere turning of the political tables need not mean a total economic change, even if different pastoralist groups dominated the scene before and after the southern Mesopotamian empires of the late third millennium.

38


Harvey Weiss has proposed that the region suffered a devasting shift toward a drier climate, beginning in about 2200 b.c.e., and this would have disrupted the agricultural economies of the Habur basin and beyond (Weiss ˘ and Courty 1993; Weiss 2000). Increased pastoralism would be a natural corollary (Peltenburg 2000, 200). It is not clear how we should read this hypothesis against the evidence of cities such as Mari and Ebla, both of which enjoyed great wealth during the last century of the third millennium, after the collapse of the Habur sites that interest Weiss. Euphrates sites such as ˘ Mari were never dependent on agriculture for their wealth, having neither the rain of more northern regions nor the possibility of large-scale irrigation ˇ such as what was created for southern Mesopotamia. The Sakkanakku period evidence for Mari does not indicate any second-hand impoverishment of the Euphrates valley, caused by wider regional climate change. Again based on the anomalous evidence of Banat, Porter challenges the common assumption that pastoralism and states mix only as one serves the other. Khazanov, for example, supposes that all types of nomadic pastoralism are features of simple societies that produce little more than what they need themselves and for small-scale exchange. In fact, argues Porter, there is no evidence that mobile pastoralism itself inhibits the development of permanent institutions and political hierarchy (2000, 171).27 Recalling the range of new settlement in the early third millennium, she proposes: This settlement distribution, which precedes expansion in the dry-farming areas of the Habur as around Tell-Leilan, is indicative of the probability that at certain times and places many of the settlements we consider the hallmarks of urbanization and state formation, and as embedded in sedentary agricultural systems, are in fact rooted in the economy and society of pastoralism, for the exploitation of cereal production in no way explains the location of these settlements, nor their size and complexity. (pp. 455–6)

Mari’s great archives display an ideal example of the continuing viability of pastoralist states in ancient Syria (pp. 442–6). It is interesting to find that the second-millennium situation of the Mari archives is invoked occasionally in order to explain other aspects of thirdmillennium Mari culture. Certain continuities are considered to indicate the long-term realities of Mari conditions. In his interpretation of the Ninevite V storage sites along the Habur River, Michel Fortin argues that Mari was the one site in the region˘ large enough to explain such construction in the early third millennium. Mari’s reliance on grain from the Habur re˘ (2000, gion is confirmed by later textual evidence for such transactions 28 52). Bertille Lyonnet observes that the lack of early second-millennium “Habur ware” along the southwestern section of the Habur basin parallels ˘ late third millena ˘similar pattern for the Ninevite V pottery, though the nium interrupts this pattern (2000, 245–6). Finally, Jesper Eidem comments that some of the same major “city-states” remain and compete from their


39

appearance in the middle of the third millennium on through the period of the Mari archives, giving the impression of large-scale social and economic continuity.29 It would not be surprising to find that certain economic patterns either persisted or recurred for centuries before the Amorrite kingdoms of Yahdun-Lim, Yasmah-Addu, and Zimri-Lim. ˘ ˘

3. The Amorrites The Akkadian word amurrûm is the equivalent of the Sumerian mar-tu, both meaning “western” and so “westerner,” someone from the west. Unlike the tribal categories bin¯u yamina and bin¯u simal, “children of the right (hand)” and “children of the left (hand),” amurrûm is not a self-given name. The “right” and “left” may even refer to south and north, as they do in ancient Hebrew, facing the rising sun, but these are entirely different names. Individual persons do not identify themselves as “Yaminite” or “Simalite,” but rather by the tribal units below these two umbrella categories. Nevertheless, the attribution of tribes to these two divisions is made by the people themselves. By origin, the word “Amorrite” is entirely different, and this difference should be kept in mind when we follow its common use in modern scholarly literature. Because “Amorrite” designates a category of outsiders, this naming will be unconscious of native identities and therefore both inaccurate in whom it groups together and liable to carry negative overtones. In its primary Mesopotamian use, it describes a certain sort of people who come from regions to the west, evidently somewhere in Syria, and the term has little use in discussion of contemporary peoples and events in areas further west, north, and south. There is no sharp break of unfamiliarity, but rather a gradual loss of precision. The Ebla texts now attest a kingdom named Mar-tu(m) in western Syria ˇ during the twenty-fourth century, and a year name for king Sar-kali-ˇ sarri (ca. 2200) of Akkad boasts the defeat of the Mar-tu in “Basar.” Mount Basar (modern Jebel Biˇsri) was known in this period as “the mountain of Amurrum.”30 The Ebla evidence could suggest that the “western” name may have been borrowed from a local Syrian polity, but this detail offers little help toward understanding the peoples identified under the broad category (see Whiting 1995, 1232–4). As for the narrow Akkadian association of Mar-tu with the Jebel Biˇsri, in the steppe south of the Euphrates River, one must consider the southern Mesopotamian source, which would not have known much about the actual political and social organization of Syrian pastoralists. All of this has been observed before, but it is worth remembering before we get entangled in the affairs of tribal peoples in the Mari archives. The word “Amorrite” carries a subtle trap, always twisting our understanding toward a southern Mesopotamian perspective, even when picked up in the west itself. The entire idea of a great Amorrite disruption derives from

40


the Sumerian victims of attacks from these outsiders, and there is no doubt that the calamity was both real and partly western in origin. When we reconstruct the history of early Syria, however, we are on much less certain ground. Something new happened in Mesopotamia at the end of the third millennium that inspired the “western” category and ultimately stamped it with fearsome meaning, but what was new to Syria? At least the increasing possibility of putting pressure on the southeast, downstream, was new. Perhaps there was a new expansion of mobile pastoralism, possibly with new expressions of tribal identity. It is not clear, all the same, that the basic nature of Syrian society changed or that the “Amorrites” were truly new peoples in any coherent sense. All of this does not strip the word “Amorrite” of relevance for the Mari archives of the early second millennium. The adjective amurrûm does appear in letters written far from southern Mesopotamia by natives of regions upstream, and by this late date, it has come to be accepted as having local value, as a self-given name. Certain kings of Larsa and in the Diyala region east of the Tigris River embrace the title “Amorrite ruler” (rabi¯an Amurrim).31 There is already in this period a kingdom called Amurrum in the far west, evidently along the Mediterranean coast and possibly even the ancestor of the Late Bronze Age kingdom of Amurru, south of Ugarit.32 Also, one of the Simalite tribes took for itself the name Amurrˆ um.33 Late in his reign, Hammurabi of Babylon began to call himself “the king of ˘ the Amorrite land, the king of Sumer and Akkad.” He Babylon, the king of all added the last two titles to his titulary only after he had conquered Mari in his thirty-first year and Larsa in his thirtieth. Here both “Sumer and Akkad” and “the Amorrite land” have directly political points of reference. Hammurabi calls himself “the” Amorrite king in order to celebrate not his˘ own tribal heritage, but rather his new rule of the domains that had belonged to Mari. From this more southern view, Zimri-Lim’s territory had been understood as particularly “Amorrite,” an area of living tribal commitments and perhaps of current use of the Amorrite language, to fit the “Amorrite” identity of Hammurabi’s forebears. ˘ In the Mari archives themselves, “Amorrite” is not used for the kingdom itself. The word “Amorrite” may name a god or a kingdom in the far west, but in more general use, it identifies either a non-Akkadian Semitic language or some group of people.34 In king Zimri-Lim’s circle, it seems that the primary use of the word “Amorrite” is to name the West Semitic language of this circle, contrasted to the Akkadian that serves as the international language, especially the language of correspondence and writing generally. Durand draws together evidence to show that many scribes were truly bilingual, capable of much more than mere copying of messages dictated by the correspondents who employed them (1992b, 124–6). In one case, the merhûm Ibal-el even mentions a scribe who can handle Akkadian, Amorrite, and˘ “Subarian,” which might be Hurrian.35 King Ammi-s.aduqa of Babylon,


41

long after Hammurabi, seems to reflect the language-based divisions when ˘ people as “Akkadian” and “Amorrite” in his royal edict.36 An he identifies unpublished letter to Zimri-Lim from an influential Simalite named RipiDagan even calls his lord “king of the Akkadian and the Amorrite.”37 A wellknown letter to Zimri-Lim from Bahdi-Lim, the governor of the Mari district, ˘ contrasts Akkadian with “Hana” customs, a more limited dichotomy that ˘ may nevertheless borrow the language-based use of “Akkadian(-speaking)” to identify people.38 We return to this text below, but in light of the language categories found by Durand, it appears that at least by the Mari period, the geographical/ethnic distinction between “Akkadian” and “Amorrite” was strongly identified with the broad language division between the eastern and the western Semitic dialects.39 As Durand himself has argued, this duality surely originates on the Akkadian-speaking side, and it reflects at least in part the constant awareness in scribal circles of the contrast between the primary language of writing (Akkadian) and all other languages (1993b, 46–7).40 As part of this duality, the “Amorrite” category derives from the older geographical identification of foreign “westerners” (Mar-tu/Amurrˆ um), who by this time shared many elements of a regional culture that overlapped east and west.41 In spite of the vigorous cultural exchange, the language difference remained, and the language categories of “Amorrite” and “Akkadian” could serve as rough ethnic indicators. Durand treats the duality as essentially geographical, but though a geographical aspect persists, I suspect that language had become the primary identifying trait. In his publication of new evidence for the “Akkadian” category at Mari, Maurice Birot identified three possible modes of reference, following Durand’s choice of the geographical over the political or the linguistic.42 The new text by itself proves that “Akkadian” is not a political identity. A governor of the northern district of Qat.t.unan tells Zimri-Lim: The Numhâ-ite, the Kahatite, the . . . (?), the Eluhutite, the . . . (?), and an Akkadian ˘ oath (together). ˘ ˘ swore a treaty I was not able to identify the Akkadian who swore the treaty oath with them, whether he was the Eˇsnunna-ite or the Babylonian.43

Two major polities (m¯atums) could be identified as “Akkadian”: the rising kingdom of Babylon under Hammurabi and the earlier dominant power of that region, the kingdom˘ of Eˇsnunna.44 When the term “Akkadian” was associated specifically with the states and territories of Eˇsnunna and Babylon, however, the geographical interpretation of the category fails to explain how Zimri-Lim himself could be said to rule a population of both Akkadians and Amorrites or Hana. The clearest bounds appear instead ˘ applied to broad ethnicities or cultural to be rooted in language, then identities. One possible expression of this phenomenon is found in a Carchemish official’s declaration, “I am not [wearing] a Subarˆ u [garment]; (rather,) I am wearing an Akkadian garment.”45 A man from West Semitic

42


(“Amorrite”) – speaking territory speaks of clothing, perhaps the one commercial item that was most strongly associated with visible ethnicity, in terms of the two other language categories observed by Durand: “Akkadian” and “Subarian.” The same pattern appears with the category “Amorrite,” as with the “Amorrite figs” requested in one case by Zimri-Lim.46 If I am right, neither the Amorrites nor the Akkadians of the Mari texts represent the same peoples who were designated by each term in earlier Mesopotamia. The Amorrites who share a West Semitic language may overlap with the earlier Mar-tu/Amurrˆ um, insofar as language also served to identify the western foreigners, but the Mari term probably indicates a smaller or larger group, not defined the same way. There is certainly no reason to link the Mari Amorrites specially to the Jebel Biˇsri or to the region south of the Euphrates. Likewise the Akkadians need not be related in any way to Akkad or even to any southern Mesopotamian “land of the Sumerians and the Akkadians” that was claimed by the kings who dominate southern Mesopotamia after the fall of Sargon’s dynasty based at Akkad (Kraus 1970, 28–32).47 The Akkadian population of Zimri-Lim’s circle seems rather to be identified by their use of the southeastern Semitic language. This definition would naturally still locate the main “Akkadian” population downstream of Mari and would assume a relationship to the current centers of Akkadian speakers in the kingdoms based at Babylon and at Eˇsnunna. Norman Yoffee calls “Amorrite” an “ethnic denotation,” because we find more than one type of social and economic organization within what is nevertheless a “bounded unit,” presumably based on language. Amorrites are both pastoralists and agriculturalists; both live in cities and move through the countryside and share no single socioeconomic status (1988, 50–1). In the final chapter of this book, I offer a somewhat different reading of the relationship between the tribal peoples and those defined by other social categories, but Yoffee’s observation is essentially correct for the Mari period. We should distinguish the idea of Amorrite ethnicity in the second millennium, when the name seems to be accepted by West Semitic speakers to describe their own identity, from the original southern Mesopotamian coining of the term for people outside their own sphere. If “Amorrite” represents a language and an ethnic category in the Mari archives, it should not be understood to correlate exactly with the paired tribal confederacies named for left and right – the Simalites and the Yaminites – as both Durand (1993b, 47) and Buccellati (1992, 88) propose. These two large tribal groups share the region with other, perhaps smaller West Semitic – speaking tribal peoples, including the Yamutbal and the Numhâ, associated with the kingdoms of Andarig and Kurdâ near Jebel ˘ of the Habur River. In sum, I find the Amorrite category to Sinjar, east ˘ poorly understood to be of great use for a targeted be ill-defined and too analysis of our Mari period and region. The Mari texts display the term only

Primary Constituents of the Confederacies

43

rarely, and we are better off occupying ourselves with the various tribal names and with the categories for identifying pastoralists, including the ubiquitous word hana. ˘

b. the primary constituents of the confederacies: sim alite gayum and yaminite li mum I have already introduced the pair of tribal confederacies that make up a large part of the population of the Mari kingdom: the Yaminites and the Simalites. During the reign of Zimri-Lim, the Yaminites are more visible in the texts because they are conquered cousins who constantly have to be named as outsiders.48 The Simalites do not generally need to be named as such, because they are the people of Zimri-Lim. I propose that in spite of the strong similarity between the two tribal groups’ basic ways of life, they appear to have developed distinct social and political identities and organizations. These can be seen first of all in their rulers. Zimri-Lim is king of the Simalites as a whole, whether or not some stragglers refused to accept his leadership. In this unifying role over the Simalite confederacy, he followed in the footsteps of Yahdun-Lim. The Yaminites, in contrast, have five kings over each of five tribes.˘49 Behind this immediate political difference lie separate terminologies for the social and political organization of each confederacy. The primary units of the Yaminite confederacy appear to be called li mums, which I render as “tribe,” as the individual components of a larger confederacy. Individual Binu Yamina are identified by these named “tribes,” not as “Yaminites.” Each li mum “tribe” functions as an active political entity, as reflected in their separate rulers. Zimri-Lim’s Simalites are also distinguished by named tribal subcategories, in this case called gayums, though these have no individual leaders who would correspond to the rulers of the Yaminite li mums. As with the Yaminites, individual Simalites are generally identified by this component of the larger confederacy, but the standing of the gayum is completely different from that of the Yaminite li mum. For one, the gayum category can be used not just for the primary Simalite entities; at least in military settings, the word gayum can also apply to an intermediate division of all Simalites into two groups: the Yabasu and the Aˇsarugayum. None of the Mari evidence indicates any active political function for the gayum, whether by individual leader or by any collective aspect. Confronted with so many differences from the Yaminite li mum, I have chosen to translate the gayum as “division” rather than “tribe,” though both the li mum and the gayum define the first-order components of each tribal confederacy.50 Even the above sketch suggests a formidable difference between how Yaminite and Simalite tribal societies were conceived. It is difficult to judge exactly how deep this difference runs. Both the Simalites and the Yaminites

44


incorporated mixed sedentary and mobile populations with mixed dependence on both agriculture and herding. We cannot demonstrate that either tribal confederacy was any more or less sedentary or that either one stood at any different developmental “stage” in its size, its dependence on agriculture or settled habitation, or its level of tribal social organization. They lived similarly in broad terms, but had developed markedly distinct social and political structures, perhaps partly in response to the different environments of their settlements and grazing lands. The Simalites’ gayum “division” is particularly difficult to comprehend. Not only does it lack any demonstrable political role, but it also is attested only for the mobile pastoralist or “Hana” segment of the Simalite community. This means that ˘ the gayum cannot automatically be designated as “tribal,” because Hana “tent-dwellers” are defined by their mode of life, not by tribe per se.˘ According to our Mari evidence, then, only the mobile pastoralist Simalites are actually identified by tribal “division.” We know, however, that Simalite towns could be populated especially by people identified with these gayum divisions, as shown for the case of Sapiratum, to be discussed below. Did all Simalites identify themselves primarily by the mobile pastoralist part of their communities? Whatever the answer, the role of the gayum reflects a political configuration completely different from that of the coalition of Yaminite kings. Zimri-Lim ruled his kingdom through two largely independent leaderships, neither of which mediated for the whole Simalite population as such. The system of “districts” (hals.um) and “governors” (ˇsa¯ pit.um) managed the Ah ˘ valley domain conquered from Yasmah-Addu and˘ Purattim, the Euphrates ˘ of loweronce the base for Yahdun-Lim’s kingdom. Meanwhile, a hierarchy ˘ level sug¯agum “leaders” served the Hana “tent-dwellers,” under the higher authority of two “chiefs of pasture” ˘called merhûms. Neither of these func˘ merhûms stood above the tioned at the level of the common gayum. The two gayum, and it is not clear that they were associated with˘ the two overarching divisions, the Yabasu and the Aˇsarugayum. The sug¯agums stood below the gayum in structural terms, in that more than one sug¯agum would be found within one gayum. For the Simalites, the merhûm and the sug¯agum were de˘ fined by pastoralist, not strictly tribal categories, and we can be sure only that Simalite herdsmen looked to their authority. It is not clear whether this arrangement reflects a long-standing Simalite tradition or a more recent development under the unifying rule of the “Lim” dynasty of Yaggid-Lim, Yahdun-Lim, and Zimri-Lim, which could have eliminated the political role of ˘the gayum. The contrast between Simalite and Yaminite political structures then appears in how the merhûm and the sug¯agum function. We know little about ˘ the Yaminite merhûms. They probably bore similar responsibility for grazing ˘ lands, but the Yaminites had no single royal administration within which the merhûms could develop the particular power that came to characterize ˘


45

Zimri-Lim’s chiefs of pasture. The Yaminite sug¯agums, unlike their Simalite counterparts, could be identified by their towns, though it is not certain whether all Yaminite towns had a sug¯agum “leader.” My hypothesis of a significant difference between Simalite and Yaminite social structures represents perhaps my one most notable divergence from Durand’s analysis of the Mari evidence. The implications are important, because they shape how we perceive early second-millennium society more generally. If the Yaminite and Simalite confederacies had in fact developed quite separate social structures, then we cannot look for a single “Syrian” or “tribal” social grid into which any new evidence can be made to fit. This one contrast is a harbinger of other contrasts that must be sought out in a mosaic of populations and forms. The purpose of this section is then to work through the evidence for the particular roles of the Simalite gayum and the Yaminite li mum, the primary consituents of the tribal confederacies.

1. The Hypothesis of Simalite/Yaminite Difference In spite of the unusually helpful sources, tribes in the Mari archives are most often mentioned only in passing, with more assumed than explained, so that their organization remains poorly understood. Tremendous progress has been made during the past twenty years with the work of the current French publication team under the leadership of Jean-Marie Durand, but many questions remain. Durand himself has been inclined to see the essential structures of the Mari archives’ “Amorrite” tribes as largely homogeneous, especially across the main duality of Simalite and Yaminite peoples. I suggest that the political and social structures of these two groups, with the words that describe them, diverge more sharply than previously imagined. At the heart of my hypothesis stands a pattern in the use of two terms that are associated with the primary tribal units among these two peoples. On the Simalite side, the first-order components of the group were called gayums, a term that is never incorporated into the standard Yaminite hierarchy for tribal identification.51 It is clear that during the reign of Zimri-Lim, there were five first-order components of a Yaminite tribal confederacy52 – the Yarihˆ u, the Yahrurˆ u, the Rabbˆ u, the Uprapˆ u, and the Amnanû – but their ˘ category ˘ has been unknown, perhaps in part because we have so few generic Yaminite texts. I suggest that the kinship term for these first-order groups may have been limum, a rare Mari word with later expressions at Ugarit and Emar, as well as in biblical Hebrew. Whereas the gayum is applied to units of tribal organization on only the Simalite side, the word li mum seems to have this function only among the Yaminites. The hypothesis of heterogeneous structures is sustained in part by the different roles for the sug¯agum “leaders” within the Simalite and the Yaminite tribal groups. The key contribution of these roles to my larger argument forces me to introduce the sug¯agum in this section and to reserve for the next major section the discussion of

46


the facets of the sug¯agum’s position that do not relate directly to the tribal distinction addressed here. It is not possible to talk about the tribal peoples known from the Mari archives without addressing key proposals advanced by Jean-Marie Durand and Dominique Charpin that change radically the entire framework for evaluating the Mari evidence. This leads me to risk another structural redundancy and to offer brief introduction to material that receives fuller treatment later in this chapter and in the next, with focus on the kingdom of Zimri-Lim. While the specific conception of tribal structures advanced here is my own, it is built on two of their ideas that make sense of whole swaths of data that are otherwise incomprehensible. Before the constitution of a new Mari publication team under the leadership of Durand, the landmark study of the tribal peoples known from the archives was Jean-Robert Kupper’s Les nomades en Mésopotamie au temps des rois de Mari (1957). No one doubted certain basic elements of Kupper’s analysis, which offered excellent insight from the evidence available at the time. In particular, the tribal population of the Mari region was divided into not two but three groups: the widely attested Yaminites and “Haneans” (Hana), and the more obscure Simalites. Yahdun-Lim and Zimri-Lim were not˘ known to ˘ have had any personal tribal affiliation, though they clearly had a close relationship with the Haneans.53 Within this framework, the Mari kings were naturally considered in every scenario to represent an urban elite that was fundamentally separate from the tribespeople of village and countryside. Studies of the Mari tribes by Luke, Rowton, Matthews, and Anbar worked within these limits.54 In 1986, Charpin and Durand proposed that king Zimri-Lim was a Simalite. This conclusion makes Zimri-Lim’s early war with the Yaminites a conflict between competing tribes, rather than between an urban king and a tribal people resistant to state control.55 More recently, Durand has explained the vast majority of references to tribal people called hana or hanûm ˘ of as “tent-dwellers” who may have varying tribal affiliations.56˘ With most the Mari texts produced by the administration of the Simalite Zimri-Lim, the word hana most often indicates Simalite “tent-dwellers,” frequently the backbone˘of Zimri-Lim’s military force (1998, 418). Working from these breakthrough ideas, I find that the last conclusion implies a detailed view of Simalite life under the Hana name. When Michael ˘ gayum unit was found Rowton observed over twenty years ago that the only among the “Haneans” and was never associated with the Yaminites, he attributed this disparity to the notion that the Yaminites lagged behind Hanean tribal reintegration, so that they had not yet developed a full tribal structure (1977, 189). This interpretation envisions a world of tribal peoples integrated into settled society only as outsiders to the centers of power in cities such as Mari, rather than as conquerors whose integration began at the center.57 The Yaminite and Simalite systems present very different


47

social structures, but tribal categories are crucial to both, and I see no basis for defining separate stages of tribalization. Instead, we seem to be confronted with two different expressions of tribal construction, which should lead us to watch for other distinct features and terminology.

2. Zimri-Lim and the Hana ˘ Durand’s interpretation of the word hana (or hanûm) requires more ex˘ ˘ tended discussion. I treat here the basic identification of the word and save for below in the chapter a review of the role played by Hana peo˘ at two ple in the Mari evidence (section E). Durand’s hypothesis operates levels that are worth distinguishing. First of all, and most important, he demonstrates that the word hana is used in conjunction with known tribal names in a way that indicates˘a population category used by more than one tribal group. Moreover, the designation of “Yaminite hana” and “Simalite ˘ hana” requires that the hana be a subset of the tribal group in question, at ˘the same time as such a˘ subset can be defined for different named tribal peoples. The term cannot therefore be the proper name for a “Hanean” ethnicity or tribe, which would fit the evidence only if both the Yaminites and the Simalites were considered to belong to a larger “Hana” people. ˘ Somehow the word hana must define a particular type of people among ˘ both of the tribal confederacies. This pattern is derived from considerable new evidence and is surely correct, whether or not we can determine the meaning of the word. As the second component of his hypothesis, Durand does suggest a meaning, which I find very plausible and adopt throughout this work. He interprets the word hana as a West Semitic feminine singular adjective from the root h.n, “to˘ live in a tent.”58 In my use of the term throughout this book, I still treat the word as a proper noun (Hana) ˘ the when it refers to people with assumed tribal identity, especially among Simalites. Durand’s identification of hana as a subgroup within one of the larger ˘ tribal peoples, regardless of etymology, hinges first of all on ARM II 53, a letter from someone named Yasmah-Addu, clearly not the king of Mari. ˘ This Yasmah-Addu can now be identified with confidence as the ruler of ˘ the Yarihˆ u tribe of the Yaminites, attested in numerous other Mari texts. ˘ In the letter, Yasmah-Addu isolates the “hana leaders” among the larger category of “Yaminite˘ leaders.” The sug¯agum˘ leaders of the hana plead before ˘ town bases be the assembly of all Yaminite leaders that the return of their negotiated from Zimri-Lim: Another thing. The sug¯ag¯u of the Yaminites assembled at Zalpah and went to Ahunâ, ˘ going to my ˘lord. (where) S.u[rahammˆ u and Yarim-Lim] sat down and discussed ˘ The sug¯ag¯u of the hana (tent-dwellers) stood up and addressed Yarim-Lim and ˘ “Go to Zimri-Lim in order to ask for our towns. If LahunS.urahammˆ u as follows: ˘ will not go (along), we will either kill him or drive him from his throne.” This ˘ Dagan

48


is what the sug¯ag¯u of the Yaminites declared. Now then, Yarim-Lim, S.urahammû, ˘ and the sug¯ag¯u are coming to my lord. My lord should not refuse them anything they ask. Also, I myself will be arriving after them.59

There is at least one other example of Yaminites speaking of their own “hana” as those who gather in assembly for political decision making.60 At ˘ two letters from the Mari archives display a similar combination of least terms with the Simalite tribal name. In one unpublished text, the writer speaks of “Simalite hana” who are grazing their flocks in the land of the ˘ Yarkab-Addu, the king of Hanzat in the Zalmaqum Yamutbal.61 Elsewhere, coalition, describes to Zimri-Lim the unity of “the˘ Simalite hana and the Yamutbal” (or “the Simalite and the Yamutbal hana”), both ˘tribal groups ˘ outside his own Yaminite affiliation.62 In all of these compound uses, the word hana seems to identify the element of the tribal population that lives with ˘the flocks outside permanent settlements, a use that is consistent with many occurrences in the Mari documentation of hana standing alone.63 Because the word hana designates the ˘ entire category˘ of those who live on the move, at a distance from the permanent settlements specified by the word a¯ lum (“town”), I find Durand’s etymology from the verb “to camp, to live in a tent” to be inherently attractive, however the common form with final -a may be explained. This etymology would allow the word hana to be translated as “tent-dweller.”64 Most often, the word hana seems to˘ need no qualification by tribal affiliation because it is applied ˘by each tribal group to its own population of mobile pastoralists. To use the word hana alone was to say “our hana.” In such use, ˘ (Hana), though the word does take on some of˘ the quality of a proper noun one that is subsumed under the tribal identity of the speaker.˘ We find the use as “our hana” among both Yaminites and Simalites. ˘ ARM II 53, with only his Yaminite kinsmen Yasmah-Addu spoke this way in ˘ in view. Because most of the texts from the period of Zimri-Lim are composed within a circle dominated by the tribal identity of its king, the great majority of references to the hana assume that they are Simalite. A striking confirmation of this equation˘ is preserved in a letter from a military leader named Yatarum, who encloses his report of a recent victory in a literary envelope that uses both tribal identifications (FM III 135). The letter begins, “The Hana are well; the troops of my lord are well” (line 4). It concludes, ˘ “The troops are well and the Simalite(s) are well” (line 5 ).65 Even this ubiquitous association of the word with Simalites specifically confirms the fact that we are not dealing with a tribal name separate from Yaminites, Simalites, Sutˆ u, and so on, as first assumed by Kupper and still maintained by Anbar.66 Durand also observes the two levels of “hana” use, one for “bedouin” gen˘ erally (my “tent-dwellers”), and the narrower one for “Simalite(s)” within the horizons of Zimri-Lim’s kingdom and tribal people, as expressed in his


49

professed title as “king of Mari and the land of the Hana” (see Chapter ).67 ˘ He cites as an analogy the two French uses of “Américain” for both persons from the United States and persons from the American continents (LAPO 17, p. 418). In spite of the acknowledged duality, Durand still prefers to interpret Zimri-Lim’s titulary by the wider use.68 This perspective is clearly visible in a passing comment to a treaty text between Zimri-Lim and Ibal-piel II of Eˇsnunna, where Durand speaks of “the kingdom itself” as separate from the hana bedouin, who may be both Simalite and Yaminite (LAPO 16, ˘ d). It is not clear to me that Zimri-Lim ever developed an ideolp. 456 note ogy of a united mobile pastoralist “hana” that would embrace both Simalites ˘ and Yaminites, though he could advocate a larger identity against Elamite 69 attack. Under normal circumstances, as we have seen, the Yaminites were not even allowed an officially recognized mobile community when they swore loyalty to Zimri-Lim. This is why the Yaminite town of Dabiˇs swears that it has no hibrum in the steppe when it undertakes to join Zimri-Lim’s kingdom under˘ the auspices of the Simalite Nihadû division (A.981, below). Our final interpretation of “Mari and the land˘of the Hana” in Zimri-Lim’s ˘ titulary, the self-definition of his kingdom, will reflect some of our most important decisions about the social and political organization of Zimri-Lim’s realm. Before going on to explore further the use of “hana” terminology in the ˘ proposed by Giorgio Mari archives, I need to mention the interpretation Buccellati, which was formulated without most of the new evidence from Charpin and Durand. In one article, Buccellati makes the Han¯u and the Sutû two subdivisions under the overall heading of “m¯ar¯u Yam¯˘ına” or Yaminites, as the pastoralists of the steppe. The gayum or “tribe” falls below the Han¯u ˘ u> subdivision, and the hibrum “clan” falls below the gayum: Yaminites > Han¯ ˘ ˘ gayum > hibrum. There are “Haneans” who are qualified as Yaminites but no Haneans ˘qualified as Suteans (1990, 104–6). Slightly later, Buccellati adds that the “ban¯u Yamina” and “ban¯u Samal” (Yaminites and Simalites) replace the earlier Amurru (Amorrites), as “sons of the right river bank” and “sons of the left river bank.” The Hana are the rural class of the Euphrates valley, ˘ the Ah Purattim, and the name of this population becomes a geopolitical ˘ term under both Mari- and Terqa-based administrations. This rural class has appropriated the nearby steppe for herding (1992, 88–9, 92). This reconstruction is fraught with difficulties in interpreting numerous Mari texts, including many cited earlier in this volume. One immediate difficulty is the fact that the core territory of the Simalite “Hana,” mentioned frequently during Zimri-Lim’s reign, is Ida-Maras., in the˘ upper Habur, in ˘ them the heart of dry-farming terrain. This makes it impossible to define in terms of the narrow ecological band of the middle Euphrates valley and adjoining steppe (pp. 88–9). Individual texts such as the reference to “the Simalite hana and the Yamutbal” just cited, are difficult to explain as refer˘ ring to peoples who are ultimately Yaminite.

50


3. The gayum Durand’s interpretation of the word hana allows almost all of the old ev˘ to be reapportioned to the other idence for a supposed “Hanean” tribe named peoples, especially the Binu Simal and the Binu Yamina. One of the first implications of this shift is that the tribal structure previously attributed to the Hanean tribe is in fact Simalite.70 In 1953 and 1955, Maurice Birot published a set of supply lists for hana garrisons stationed at Mari and nearby ˘ S.uprum. These troops were organized according to named affiliations that were in some instances given the further classification of gayum, which he translated as “clan.” Philippe Talon added to the collection of known gayums, based on further lists published in ARM XXII and XXIV.71 None of the hana ˘ involved in these lists turns out to be Yaminite, and the breakdown of tentdwellers by gayum appears to be particular to the Simalites. a. The Letter FM III 136: Identifying a Simalite under Zimri-Lim. The role of the gayum within Simalite tribal identity is displayed in two letters that trace the movements of individual tribesmen, one letter from each of the best attested Mari reigns. ARM IV 1 was sent by Samsi-Addu to his son Yasmah˘ Addu, whom he had made king at Mari, and we return to this text after extended discussion of the other letter, which comes from the period of Zimri-Lim. Both letters cite the categories deemed adequate to identify the Simalites in question. FM III 136 was sent by a military leader to Zimri-Lim, and it gave the king every piece of information necessary to track down a departed member of the Hana fighting force. It is unnecessary to identify the man as Simalite, ˘ this is assumed in the complaint about “Hana” desertion: because ˘ If [(my lord?) does not arrest(?)] the one who takes leave (of his troop), the life of the Hana will take leave (as well). He will break the fortune(?) of the Hana. Now then, ˘ ˘ Lawasum, the Yumahammû, has taken leave. His sug¯agum is Dadi-[Lim], and his ˘ property consists of 200 sheep and five donkeys. My lord should confiscate (this).72

The text proceeds by the following line of reasoning: r The departure without leave is stated with the man’s name, Lawasum, and his first-order group identity is defined as Yumahammû, a known gayum, ˘ though the word is not used here. r “His sug¯agum is Dadi-[Lim].” The second tier of identifying detail is provided by the man’s section head, so that each gayum would have some plural number of local sug¯agum “leaders.”73 r He has baˇs¯ıtum property of 200 sheep and five donkeys. His immediate domain is not a “house” but his livestock. The full hierarchy of identifying information therefore covers his inclusion among the tent-dwelling pastoralists (“Hana”) of the Binu Simal, his gayum ˘


51

“division” of Yumahammû, the name of his particular sug¯agum “leader” ˘ perhaps the extent of his personal livestock. Although within that gayum, and the gayum may come to have geographical associations, it is not primarily treated as a geographical name, and so the entire chain of identifying categories for this mobile herdsman omits any reference to the districts (hals.um) ˘ or towns of Zimri-Lim’s kingdom. This does not mean that these categories exclude settled Simalites, but it is important to recognize that in this case, the hierarchy is constructed in a way that has no need of political structures defined by settled life. b. Leadership by sug¯agum for Herding Group and Town. I have translated the word sug¯agum by the rather nebulous term “leader” because of its range of use.74 A sug¯agum may be the ruler of a large population, whether defined by tribe or by m¯atum “land.” Under the umbrella of the Mari kingdom, sug¯agums were commonly granted leadership of individual towns, and as just seen, Simalite herding groups were led by men with the same title. The letter quoted above (FM III 136) reflects the definition of Simalite gayum categories in exclusively mobile pastoralist terms. To my knowledge, no Simalite sug¯agum is identified by sole governance of an individual town, as “the sug¯agum of town X.”75 This stands in contrast to the evident pattern among the Yaminites. There is no question that many Mari towns were led by an individual sug¯agum. A large number of these have no apparent tribal affiliation (see Table 1), but some of them were certainly Yaminite. Based on this body of evidence, Ichiro Nakata has even proposed that all sug¯agums may rule towns, even when they are identified collectively by tribal categories.76 The association of Yaminite sug¯agums with the leadership of individual towns is explicit in at least one case and quite certain in several others. In a note from Kibri-Dagan of the Terqa district to his lord Zimri-Lim, the governor reports, “I have sent to the towns of the Yaminites, and the sug¯agum of (the town of) Dumtên answered me as follows. . . .”77 The Yaminite letter ARM II 53, quoted above, has sug¯ag¯u who are defined as tent-dwellers ask two of the Yaminite rulers to go to Zimri-Lim as their representatives to request the return of “our towns” to their control. The possessive suffix suggests a distributive meaning, with one sug¯agum for each settlement. According to the fascinating new population evidence surveyed by Adélina Millet-Albà (forthcoming), Dumtên was the largest Yaminite town in all of the core districts of the Mari kingdom, with +/− 675 inhabitants. Two other examples come from the district of Saggaratum, including the smaller Yaminite towns of Ziniyan in the Yahrurˆ u tribe (+/− 195 inhabitants) and Sahrˆ u in the ˘ Amnanˆ u tribe (+/− ˘140 inhabitants).78 In Table 1, I list only leaders defined directly as “the sug¯agum of town X,” excluding more indirect associations of town and title. Other sug¯agum leaders are identified with Yaminite towns, but only as resident in them, with a formula used also for some Simalite sug¯agums. A letter from Sumhu-rabi, ˘


52

table 1. Towns with Individual sug¯agum Leaders and No Known Tribal Affiliation Mari district

Terqa district Saggaratum district

Qat.t.unan district Not located (by me) Not in Mari kingdom

Humzan (ARM XVIII 54:2); S.ubatum (ARM XXIV 60); ˇ˘ Sakka (ARM XXII 326:3–5; XXIV 55; 63 ii :3 –4 ); ˇSehrum (ARM XXIV 62 i :12 –13 , without the title); ˘ (ARM XXIV 61 i:4, without the title). Urbat Himaran ([ARM IX 248:8 ]; XXIV 57; 62 i :18 –19 ); ˘ samta (ARM VI 40:5–7; XXVI 5:12–14; 6:54). Hiˇ ˘ Amatum (ARM XXIV 59);a Dur-Yahdun-Lim (ARM XIV ˘ ); Zibnatum ˇ 46:7); Samdanitum (ARM IX 248:10 (ARM XXIV 60). T.abatum (ARM XXVII 107:8–9). Dunuhˇsi (ARM XXIV 61 ii:3 ); Utahum (ARM IX ˘ ; XXIV 61 ii:5 ). ˘ 248:13 Amaz (ARM X 84:10–11, incorporated into the kingdom of Andarig, north of Mari); Der of the Balih (ARM ˘ XXVI 24:16; FM II 63:9–10); cf. a town in Zalmaqum (Arduwan?, A.2995 + M.14337:4, FM I, p. 61).

This was read by Talon as A-ba-tumki , but the copy suits -ma- at least as well; see Durand’s caution not to confuse Amatum with the western Euphrates town of Abattum, near Imar (1990b, 45n32; cf. ARM XXIV 158:5). a

an early governor of Saggaratum under Zimri-Lim, passes on the news that a malformed lamb (izbum) was born in the flocks of “Zazum the sug¯agum” in the Yaminite town of Zarrum Rabbˆ um, a settlement of modest size in the Terqa district (+/− 275 inhabitants).79 An administrative list of silver and livestock payments received includes a sug¯agum in the major Yaminite town of Dabiˇs(an) (+/− 550 inhabitants) in the district of Saggaratum.80 Another Yaminite town sug¯agum is alluded to in the commentary to ARM XXIII, where a man named Iˇshi-ebal is said to be known as the sug¯agum of the ˇ hus.uratim (+/−˘70 inhabitants).81 Although the title sug¯agum village of Sa ˘ is not used directly, a letter from the governor of Saggaratum suggests the same role for “Kaili-ilumma of Barhan,” the largest Yaminite town in the ˘ town leader, who as sug¯agum would district (+/− 650 inhabitants). The have had responsibility for local execution of the royal census, has reported eight men who had managed to avoid the population count at the official date.82 All in all, seven Yaminite towns show either direct or indirect identifiˇ cation with a sug¯agum leader: Dumtên, Ziniyan, Sahrˆ u, Zarrum, Dabiˇs, Sa hus.uratim, and Barhan. Interestingly, two towns that˘ are the tribal seats for ˘ Yaminite rulers are ˘never said to have a sug¯agum, Miˇslan and Samanum. Miˇslan (population +/− 600) was the largest Yaminite town in the Mari district and was the seat of Yaggih-Addu, ruler of the Yahrurˆ u tribe; Samanum ˘ ˘


53

(+/− 300 inhabitants), located in the Terqa district, was the central place of the Uprapû tribe. It is clear that Yaminite sug¯agums, at least, could be freely identified as leaders based in particular towns, even as they maintained ties with a mobile herding element of the community. When forced to abandon their towns, they could follow their hana tent-dwelling people to the steppe. We cannot tell whether all Yaminite ˘sug¯agum leaders were conceived to have such a town base, though such a rigid formulation is probably inappropriate in any case. Not all of the records of sug¯ag¯utum payments refer to leaders defined by town. As already observed by Philippe Talon, some sug¯agums are indentified by gayum, the Simalite “division” for their mobile hana population (1986, 5). ˘ The sug¯agums who are This distinction correlates with a further pattern: 83 identified by their towns generally pay in silver, while those who are identified by named gayum tend to pay in sheep.84 In effect, this means that a special category of sug¯agum financial obligations is reserved for Simalite leaders who are identified by their tribal affiliation within this confederacy rather than by place of settled residence. We have no evidence for a comparable arrangement whereby Yaminite sug¯agums would be named by any category other than town of residence. There is at least one clear example of a leader who is identified by Simalite gayum but who pays in both silver and sheep.85 This combination could represent a hint of the expected connection between Simalite towns and herding groups, reflecting the leader’s responsibilities for both fields and flocks. We have no evidence for a comparable arrangement whereby Yaminite sug¯agums would be named by any category other than town of residence. c. The sug¯ag¯u of the Hana. Under Zimri-Lim, the gathered leaders described as “the (plural) sug¯a˘g¯u of the hana” are only Simalite, never Yaminite, and so I refer to them under the ˘form of the proper name, Hana. While the ˘ um “districts” administrators of Zimri-Lim may gather the sug¯ag¯u of their hals . ˘ for kingdom business, and these leaders often include those of the Yaminite towns, “the sug¯ag¯u of the Hana” serve entirely different masters: the Simalite ˘ Ibal-el and Ibal-pi-el. Zakira-Hammu, the goverchiefs of pasture (merhûm), ˘ nor over the Qat.t.unan district, writes to Zimri-Lim: “On˘ the day I sent this tablet to my lord, Aˇsmad, the aide/adjutant(??) of the merhûm (Ibal-el), en˘ this leadership, tered Qat.t.unan along with 15 sug¯ag¯u of the Hana.”86 Under ˘ the sug¯ag¯u captain military units fighting for their king. A letter to Zimri-Lim from Ibal-pi-el responds to the king’s inquiry about sending troops to the king of Talhayˆ um: “I questioned the sug¯ag¯u (who are) servants of my lord ˘ troops, but I did not send any troops (myself ).”87 Complaints about sending outside the jurisdiction of district officials must be addressed to the same combination of merhûm and subordinate sug¯agum leaders, as when Ibal-el and the sug¯ag¯u with˘ him are asked to come through with their promised deliveries of salt and grain to Nahur and Aˇslakkâ in Ida-Maras.. Itur-Asdu, ˘

54


an official of Zimri-Lim, reports to his lord: I left Nahur and met with Ibal-el at Napt.arum. I brought together Ibal-el and the ˘ laid before them my complaint about barley and salt, about which I have sug¯ag¯u and been writing repeatedly to my lord for some time.88

In the steppe, the Hana sug¯ag¯u offer a legal authority where the district governors (ˇsa¯ pit.um)˘ cannot reach. The diviner and high adviser named Asqudum attributes to them a remarkable independence in a letter to ZimriLim: “Whoever has made plans to cross into the midst of the steppe must be bound, so that they may be brought together to Mari, before the sug¯ag¯u of the Hana.”89 Samsi-Addu of the upper Mesopotamian kingdom recognizes ˘ a similar leadership over hana elements in his armies, though their specific ˘ 90 As in the other evidence treated so far, these tribal identities are not clear. diverse texts display Simalite tribal sug¯agums who have no visible relation to towns. d. The Simalite gayum and the Simalite Town Under Zimri-Lim. Evaluation of the relationship of the sug¯agum to the gayum in FM III 136 requires one final observation. In spite of the ubiquitous use of gayum tribal names to identify individual Simalites, even those who live in towns, I am not aware of any political leadership defined by the gayum, whether individual or collective.91 Zimri-Lim maintained a single circle of senior officials who could take on responsibilities in any domain, but the actual populations of the kingdom were governed under two parallel administrations. When Zimri-Lim took over Mari, he inherited a valley-based realm that had already been divided into “districts” (hals.um) under the reign of Yasmah-Addu. Each district was ˘ defined by its central town (see Chapter 3), which ˘served as the administrative seat of its governor. The economy of these districts, whose core lay in the Ah Purattim (Banks-of-the-Euphrates), was dominated by agriculture. ˘ the mobile herdsmen among the king’s Simalite tribespeople were At least governed with a lighter hand through two chiefs of pasture, called merhûms ˘ (see section D below). The Simalite towns are left with an ambiguous identity, and it is not entirely clear how their populations were governed at the practical level. On one hand, when such towns were located in regions governed by the territorially defined hals.ums, they should have been subject to the district ˘ governors. On the other, however, a text such as the land document for the town of Sapiratum (ARM VIII 85 + A.4304) shows that tribal affiliations could be maintained in these towns, and these affiliations would have provided a link with the alternative political structure that was allowed their kin who moved with the flocks. Sapiratum cannot demonstrate a universal pattern for all of the valley districts, however, because it was located in the Suhˆ um, downstream from Mari, which always held an ambiguous status in ˘


55

Zimri-Lim’s realm. Leaders named Meptûm and Buqaqum are consistently associated with the Suhˆ um, but they are not defined clearly as either gov˘ ernors (ˇsa¯ pit.um) or chiefs of pasture (merhûm).92 The Suhˆ um had earlier ˘ its connection ˘ with Mari was belonged to the kingdom of Eˇsnunna, and never as secure as that of the districts upstream. If the Simalites could be governed under leadership that was identified by relation to the herding groups, even in some cases where they actually lived in towns, the same was not true for the Yaminites. The tribal identities of Yaminite populations could not be obliterated, but Zimri-Lim refused them any authorized governance outside their towns. One component of Zimri-Lim’s subjects was organized by the gayum “divisions” of Simalite herding groups, under the leadership of both merhûms ˘ and sug¯agums, but neither of these offices is defined by the gayum itself. It seems that the Simalite gayum had lost any political function it may once have possessed, perhaps passing it to the centralizing authority of dominant Simalite kings such as Yahdun-Lim and Zimri-Lim.93 In spite ˘ of this loss, the gayum retained a strong social role that is defined in Simalite pastoralist terms, within which it continued to serve group identification. No political decisions are known to have occurred at the gayum level.94 e. The Letter ARM IV 1: Identifying Simalites Under Samsi-Addu. It is interesting to find that the role of the gayum in identifying a Simalite was equally prominent under the earlier rule of the great upper Mesopotamian king Samsi-Addu, and this ruler provides a second example of the hierarchy found in FM III 136. ARM IV 1, a letter from Samsi-Addu to his son Yasmah˘ Addu at Mari, discusses five men who have left their gayum to move to the ˇ royal court at Subat-Enlil: Sakuranu and Manatanu reside in Harradum; Kaila-ilum, Zazunum, and Dadiya ˘ reside in Amatum; their sug¯agum is Hatiku; their gayum (division) is Yumahammû. ˘ ˘ These five men left their gayum (division) and have come to me.95

The identifying information is given in three parts: r Their names are given, followed by where they reside (the verb waˇsa¯ bum), in two separate towns, Harradum and Amatum. ˘ They live in two towns but have one sug¯agum r “Their sug¯agum is Hatiku.” ˘ in authority over them. r “Their gayum is Yumahammû,” the same Simalite tribal division found ˘ in FM III 136. In this text, the men do need to be located by their town, though subordination to one sug¯agum and clan allows residence in more than one settlement. It is intriguing to find that in this case, the authority of the tribal sug¯agum trumps any leadership defined by the individual towns that

56


are named in the letter. Most references to gayums are attached to a variety of named subdivisions of the Simalite tribal confederacy, with little more information. f. The gayum in Simalite Tribal Structure. It appears that the name of a gayum could be applied both to the primary constituents of the Simalite pastoralists and to one of two overarching headings under which these constituents could be grouped. One can speak of men from the gayum of Yabasu and from the gayum of Amurrum together as Yabasu tent-dwellers.96 Durand suggests that the entire Simalite confederacy was divided first of all into two higher level units, the hanû Yabasa and the Aˇsarugayum, with the gayums forming a ˘ these two.97 According to Durand, the Yabasu, the Kasu second tier below . ˆ m, the Amurrum, and the Abi-Nakar gayums fell under the Yabasu heading, while the Yumahammu, the Ibal-Ahum, the Mannapsu, the Weruˆ m, and ˘ under the heading ˘ of the Aˇsarugayum. This analysis is the Nihadûm fell ˘ based at least in part on the letter A.486 + M.5319, where as both army general and merhûm, Ibal-pi-el confronted a political conundrum in the formal reception˘of his troops by king Hammurabi of Babylon. Babylonian expectations about what gifts should be˘provided for military leaders would not match the actual leadership of Ibal-pi-el’s army, so the merhûm adjusted the numbers by promoting two of his captains to a higher rank,˘one defined to be “over the Yabasa hana” and the other “over the Aˇsarugayum.”98 It ˘ this two-part division to the dominance of two might be tempting to relate Simalite merhûms at any given time, but there is no evidence to support ˘ such a connection. In this specific context, one merhûm has responsibility ˘ A second reference for an army that incorporates units from both groups. to these two overarching categories may be ARM XXIV 235, the text just cited regarding the double use of the Yabasu name for both gayum and larger division. The reverse of this tablet is broken, but before the final count of “24 sug¯ag¯u on leave” (from active military duty), another subtotal is rendered as “16 Aˇsarugay¯u” (lines 15–16).99 This division of tribal fighters into Yabasu and Aˇsarugayum sections was already in place during the reign of Yahdun-Lim. A rare letter to the early ˘ Simalite king lists 462 soldiers who escorted a quantity of booty to Mari for Yahdun-Lim. A summary divides them into 172 “tent-dwellers” (hana) and ˘ “townsmen,” while the actual list places the soldiers into three ˘ 290 main groups: r three sections of Simalites, listed as Yabasu and Aˇsarugayum; r the “sons of the land” from four districts: “lower” (=Mari), “Hiˇsamta and ˘ Terqa” (=later Terqa), and two that are lost in breaks; 100 r Yaminites listed by tribe as Uprapˆ u and Yahrurˆ u. ˘ Unfortunately, the breaks in the letter make it impossible to determine how each group is counted, in terms of the summary division into tent-dwellers


57

and townsmen. There are 94 Yaminites (47 for each tribe) and 125 “sons of the land” for the last three of the four districts. Although Charpin concludes that the Yaminites must be included among the hana, or at least that the ˘ indiscriminately, the hana category must mingle Simalites and Yaminites ˘numbers do not confirm this. It would be quite feasible to imagine that Yaminites were counted by their towns, as they would have been under ZimriLim, and that the 172 hana all came from the Yabasu and the Aˇsarugayum ˘ 101 The Yabasu/Aˇsarugayum duality and the four divisions of the Simalites. districts (hals.um) show a striking continuity of administrative organization across the˘reigns of the Yahdun-Lim and Zimri-Lim. ˘ the duality of Yabasu and Aˇsarugayum might Durand wonders whether be the remnant of an older Simalite division into li mums, though I am not aware of any evidence for Simalite use of that term, particularly in application to their own social structure. In fact, the two published references to the pair that I have just cited do not associate the names with either the gayum or the li mum, though the name “Yabasu” belongs also to a Simalite gayum.102 In both cases, the context of the categorization is military, naming the two primary divisions of the Simalite “Hana” in fighting mode. Neither ˘ text uses the word gayum itself for these military units, and ARM XXIV 235 even separates the larger “Yabasu hana” unit from the Yabasu gayum that ˘ contributes four members to it. New evidence will naturally adjust the details of this portrait, and I am sure that Durand and his colleagues will eventually provide further references. In most cases, individual tribespeople were identified not by the umbrella affiliations as “Yaminites” or “Simalites” but by one of the individual subgroups. Among the Simalites, these subgroups were the gayums of the primary constituent level, below any possible bifurcation as Yabasu and Aˇsarugayum. For the Yaminites, the key identifying groups were the five tribes that are discussed below, named the Yahrurˆ u, Yarihû, Amnanû, Rabbû, and ˘ ˘ Uprapû. In practical use, it seems clear that the primary subdivisions of Simalite pastoralists were called gayums, whatever term may have defined the larger military pairing of Yabasu and Aˇsarugayum. The word gayum is frequently translated as “clan” in a way that is intended to place these below the Simalite and Yaminite “tribes.” I prefer to regard the Binu Simal and the Binu Yamina as tribal confederacies. The Yaminites clearly function as a confederacy of five tribal groups, each with its own king, and here it is natural and fitting to call each group a “tribe.” Because the Simalites were unified under the kings of the “Lim” dynasty, who were able to defeat the assembled coalitions of Yaminite rulers,103 it has been too easy to speak of a single Simalite “tribe” and its “clans.” The Simalites could also be divided into separate political entities, as shown by the seven Hana kings who united to fight Yahdun-Lim, ˘ for the foundation of Dur-Yahdun-Lim. ˘ 104 according to the disc inscription ˘ Whatever hierarchy of terminology one prefers, it is important to recognize

58


that both the five Yaminite tribes, which I will tentatively identify as li mums, and the gayums of the Simalite pastoralists represent first-order constituents of the larger confederacies. The evidence of the Mari archives indicates a functioning Simalite tribal ideology, even if the gayum has lost any political role. The gayum does still serve to identify Simalite kinship relations, if not for decisions to be taken as a whole gayum, then nevertheless for groupings within those individual tribes. For the Simalite confederacy, the larger tribal category that remains politically active appears to be defined by Simalite identity itself, under the unifying kingship of Zimri-Lim, as under Yahdun-Lim before him. ˘ g. The gayum Outside Simalite Circles. Within the Simalite tribal division, the use of the word gayum should not be expected to share exactly the same range of meaning as that found in other circles. One text in particular shows a different application under the administration of the Mari king Yasmah-Addu, son of Samsi-Addu. This letter has Sˆın-teri, an official based ˘ h River region, quote his lord Yasmah-Addu as asking, “The tentin the Bali ˘ crossed (the river): Simalites (or) ˘ Yaminites – which is their dwellers who gayum? Send me a full report.”105 Although this use of the word gayum does not define a technical category, we are in strict terms higher in the kinship hierarchy than with the Simalite gayum. The word applies to the entire left-right division of tribal peoples into two broad confederacies. Certainly, the text does not provide a basis for identifying the gayum with “clan” or “tribe,” rather than a more fluid “division.”106 Notice that the statement does, however, resemble the two Simalite identification texts by both the general intent and the use of the possessive pronominal suffix, with focus on members of the gayum rather than on a ruler. Here and elsewhere, the word gayum shares a characteristic of the common Mari terms hals.um and kaprum (plural kapr¯atum), “district” ˘ serve a larger hierarchy of population and and “village.” All of these terms political definitions, but none of them is restricted to a single rank in the relevant hierarchy. Instead, all the words seem to assume some element of subordination. The hals.um is always part of a larger land;107 the kapr¯atum ˘ are always cast as dependent on a central city;108 and now the gayum seems never to be set at the top of any kinship hierarchy. Even in the letter just cited, Yasmah-Addu uses the word gayum to distinguish divisions of what he perceives as a˘ larger mass of tribal tent-dwellers.

4. The li mum Yaminites made up a significant part of the Euphrates valley population, and references to them are sprinkled throughout the Mari documentation. Very few of the texts were composed by the Yaminites themselves, however, and


59

table 2. Letters Sent by Yaminites (List Not Complete) ARM II 53 (LAPO 17, no. 702), Yasmah-Addu, king of the Yarihˆ u, to Zimri-Lim ARM II 54 (LAPO 18, no. 1059), Yasma˘ h-Addu to Zimri-Lim ˘ ˘ ARM II 55 (LAPO 17, no. 705), Yasmah-Addu to Zimri-Lim ˘ ARM II 56 (LAPO 17, no. 562), Yasmah-Addu to Zimri-Lim ˘ ARM II 61 (LAPO 17, no. 703), Dadi-hadun, king of the Rabbˆ u, to Zimri-Lim ˘ 1134), Dadi-hadun to Sibtu, ˇ wife of ARM X 156 (=FM VII 2, =LAPO 18, no. ˘ Zimri-Lim ARM XXVI 39, Nahimum of the Uprapˆ u(?) to Asqudum, high Mari official ˘ ARM XXVI 168, diviners Yams.i-hadnû, Maˇsum, and Hammi-esim to Sumu-Dabi? ˘ ˘ Zimri-Lim) (the leading Yaminite ruler before the revolt against ARM XXVI 169, same Yams.i-hadnu and Maˇsum to same ARM XXVI 170, same as 169 ˘ ARM XXVI 171, diviners Ilum-ma-ahuma, Yams.i-hadnu, and Maˇsum to same ˘ ˘ ARM XXVI 172, same as 171 ARM XXVIII 25, Yahdun-Lim, an early Yaminite ruler?, to Zimri-Lim (see p. 24) A.449, in B. Lafont, ˘Amurru 2, p. 240, two leaders of the Rabbû to Zimri-Lim A.987, cited at length with A.1146 (below), FM I, p. 122n11, Yans.ib-Dagan to his master Yasmah-Addu of the Yarihû ˘ ˘ FM I, pp. 111–25, Hammi-iˇstamar, king of the A.1146, in P. Marello, “Vie nomade,” ˘ Uprapû, to Yasmah-Addu of the Yarihû ˘ ˘ A.2094 (LAPO 17, no. 728), P. Villard, UF 18 (1986), 411–12, Hammi-iˇstamar of ˘ the Uprapˆ u to Ibal-el, one of two Simalite merhûms, under Zimri-Lim A.2526, AEM I/1, pp. 183–4, Yasmah-Addu of the˘ Yarihˆ u to uncited recipient ˘ “Fourmis blanches ˘ et fourmis noires,” in A.3080 (LAPO 17, no. 733), Durand, Fran¸cois Vallat ed., Contribution a` l’histoire de l’Iran, Mélanges Jean Perrot (Paris: ´ Editions Recherche sur les Civilisations, 1990), 101–8, Hammi-iˇstamar of the ˘ Uprapû to Zimri-Lim A.3185, cited in Durand, M.A.R.I. 6, p. 48n45, Dadi-hadun ˘ hrurû (restored from A.3821 (LAPO 17, no. 737), Yarim-Lim, king of the Ya ˘ line 17) to Zimri-Lim

this badly hampers our ability to see their tribal structures from within. In spite of this barrier, the Mari archives did accumulate a handful of Yaminite letters, and two of these offer a term that might represent the first-order constituents of the Yaminite confederacy (see Table 2). Under the title “Vie nomade,” Pierre Marello published a letter from one Yaminite ruler to another, evidently Hammi-iˇstamar of the Uprapû to ˘ complains that his counterpart Yasmah-Addu of the Yarihˆ u. Hammi-iˇstamar ˘ ˘ ˘ has not cooperated in joint military preparations, and he fills a large part of the letter with a condescending tirade against Yasmah-Addu, who supposedly ˘ touch with his tribal spends so much time at his town base that he has lost people. Along the way, Hammi-iˇstamar introduces the term li mum as the ˘

60


population that has been left to make decisions without the presence of its leader: You put your trust elsewhere, (thinking) “I have given silver to my li mum.” What is this silver that you gave? All of your silver that you gave – I know about it. Yesterday, all of your li mum was assembled at Hen, and the one who loves you was saying, “Write to him so he will go,” while the ˘one who despises you was saying, “He should not bother coming.” Now if I did not make a habit of arriving in person, they would never manage to act as one.109

Yasmah-Addu appears to define himself by his li mum, when he calls it “my ˘ people whom he expects to support him. From the perspective li mum,” his of another Yaminite king, the group is likewise “your li mum.” Faced with some collective political decision, the li mum “assembles” (verb pah¯arum), an ˘ at act typical of tribal decision making in the Mari archives.110 The gathering a location other than the residence of their ruler suggests a group defined in terms broader than a private household or immediate clan. In this context, the language of love and hatred expresses political loyalty and opposition in personal terms.111 Hammi-iˇstamar takes an interest in this assembly because ˘ he wants to know whether or not the people of Yasmah-Addu will join his own forces. Based on all of these details, it is simplest˘ to understand the li mum as the entire tribal population governed by Yasmah-Addu, at least in concept. We already know that Yasmah-Addu was the king ˘of the Yarihˆ u, one ˘ so his li mum would be the˘Yarihˆ of the five subdivisions of the Yaminites, u ˘ tribal population as a whole.112 Another Yaminite letter, written by Yarim-Lim, ruler of the Yahrurˆ u, ˘ adopts a similar use of the word, again with the possessive suffix identifying that which he leads. Yarim-Lim asks Zimri-Lim to write to Dadi-hadun, ˘ 113 a rival Yaminite lord, “that they not bring complaints against my li mum.” Again, the li mum appears to be the chief’s people as a whole, the general identity under which specific cases of conflict have arisen. The li mum would be the entire Yahrurˆ u. I am aware of˘ two other Mari uses of the word li mum, though Durand seems to suggest the existence of others in unpublished texts.114 Marello cites a second letter from Hammi-iˇstamar, still unpublished, which appears ˘ framework, with the same personal association to indicate the same Yaminite as “your li mum.” Neither Marello nor Durand elsewhere identifies the letter’s recipient, but the tone suggests another Yaminite ruler, whether or not Yasmah-Addu again: “Your li mum in the midst of Zalmaqum as well as next door to˘ Zalpah heard about my arrival and moved on to lower country.”115 ˘ the groups in question recalls Hammi-iˇstamar’s complaint The mobility of ˘ with his people by refusin the first letter that Yasmah-Addu is losing touch ˘ ing to leave his urban home. The town of Zalpah is the assembly point of the Yaminite sug¯ag¯u in ARM II 53, quoted earlier.˘


61

My last reference to the li mum comes from a damaged document for swearing loyalty to king Zimri-Lim.116 Its first visible section invokes imprecations against the swearer, including that his enemy “extinguish my tribe” (li m¯ı liballi). The text might suit the submission of Yaminites after their defeat early in Zimri-Lim’s reign.117 When the term hana is used later in the loyalty document, it refers to the general division˘ between those who dwell in tents and those who live in towns, without the common Simalite self-reference. Even if the word li mum turns out to have wider use, it does appear to have a particular place in Yaminite vocabulary for tribal structures as the first component element of the larger confederacy. The Yahrurˆ u, the Yarihˆ u, the ˘ a li mum, each ˘ with Uprapû, the Rabbˆ u, and the Amnanˆ u would each be its own king. As the primary constituents of the Yaminite confederacy, the li mums merit comparison with the Simalite gayums, but the two categories seem to be conceived in different ways. From the limited evidence available, we find that the gayum is consistently defined by reference to the Hana, the ˘ tent-dwelling mobile pastoralist component of the Simalite people. Our handful of attestations of the word li mum suggest no such pastoralist connection. Furthermore, the li mum is defined in relation to the individual leadership of each Yaminite tribe in a way that is impossible in the Simalite framework as we know it. The gayum has no leadership of any kind in the Mari evidence, and it functions as a social category without any occasion for the kind of personal attribution that characterizes the Yaminite li mum. Where the word gayum seems to indicate divisions of affiliated herding groups, the word li mum suggests the people bound to a leading patriarch.118 These differences, when correlated to the distinct Simalite and Yaminite settings for their use, underscore the impression that these two tribal confederacies had developed significantly different structures and tribal ideologies. It is worth noting, finally, that the word limum remained active in defining kinship-based social categories during later periods, with particular evidence from western Syria. In the late thirteenth century, the Ugaritic Baal myth and local Emar divine names attest the word li mu, apparently meaning “people” or “tribe.” In the Baal myth, Anat’s hapless opponents at war are l im.119 Anat herself is also identified with “the peoples,” and Baal’s death elicits a lament for “the people of Dagan’s son.”120 At Emar, there are local shrines for the deified Sarta people and for the goddess d NIN.KUR “of the Sarta people’s gate.”121

5. Simalite and Yaminite Social Structures As we have seen, the Simalite gayum and the Yaminite li mum are both first-order constituents of their respective tribal confederacies, but they are sharply different in both conceptual and structural terms. This contrast

62


reflects the broader difference between the political orders of the two confederacies. Yahdun-Lim and Zimri-Lim ruled the large mass of Simalites, ˘ even if there were reluctant stragglers.122 The texts show us no hint of a gayum confederacy, constructed from distinct political entities defined as gayums, each with its own leadership. I have defined the Simalites as a confederacy in part by comparison with the active Yaminite example and in part by the evidence of Yahdun-Lim’s two major royal inscriptions that he had ˘ groups before he could rule a “land of the Hana” to unify separate “Hana” ˘ from Mari (see Chapter 3). Zimri-Lim then both inherited and had to ˘recreate this acceptance of an individual Simalite king. It is possible that among Simalites, constituent political entities may have been defined in relation to herding groups, like the social order just discussed. Charpin now proposes that Mari itself was conquered not by Zimri-Lim but by a Simalite named Bannum, author of two remarkably arrogant letters to Zimri-Lim from the first year of his reign.123 We know that Bannum held the title merhûm (“chief ˘ hierarchy of pasture”; see section D), the same position that stood above the of Simalite tent-dwelling herdsmen throughout Zimri-Lim’s reign. His letters suggest that he was accustomed to acting independently of Zimri-Lim and that even after accepting Zimri-Lim’s kingship, the respectful refrain of “my lord” came reluctantly to Bannum’s lips, and he preferred to address the king as a peer. In contrast to the unitary authority pursued by Zimri-Lim, the Yaminites ˇ are never said to have a single ruler. In Yahdun-Lim’s famous Samaˇ s temple ˘ inscription, the Yaminites fight as an alliance of the Uprapˆ u, the Amnanû, and the Rabbˆ u, and under Zimri-Lim, these same three names appear repeatedly, along with the Yahrurˆ u and the Yarihˆ u, in various combinations.124 ˘ ˘ Each tribe remains a viable political unit, and I have proposed here that this unit was called a li mum. The contrasting political development reflected in the words for individual constituents finds a second expression in the roles of the sug¯agum “leaders” that I mentioned above. Only the Yaminite sug¯agums may clearly rule individual towns. Many Simalites reside in towns, like their Yaminite cousins, but where we find demonstrably Simalite sug¯agums, they have their primary identity in their affiliation with particular gayums. The best example of this is ARM IV 1, where Samsi-Addu singles out five men from two different towns, all of whom nevertheless fall under the authority of a single sug¯agum, identified in turn by leadership in the gayum of the Yumahammû. The population of Sapiratum in the Suhˆ um, ˘ ˘ u downstream from Mari, seems to belong to a single gayum, Yumahammˆ ˘ again. Both the Simalite and the Yaminite confederacies shared a mode of life that preserved a role for large mobile populations at the same time as a major component lived in villages and towns. This two-part existence was expressed in both tribal communities by the same terminology of hana, “tent-dweller,” and nawûm, “steppe,” on one side, and of the town, a¯˘ lum,

The Local Leader of Tribe and Town

63

on the other.125 In spite of their broad commonalities, the Simalite and Yaminite confederacies evolved separate social and political structures with independent (if overlapping) terminologies. The divergence of gayum and li mum, along with the incontrovertible contrast between combined and segmented rule, warn us not to treat all social and political structures from the Mari-period tribes as an undistinguished mass.126 For example, one partially published text describes the formal affiliation of the Yaminite town of Dabiˇs with Zimri-Lim’s kingdom. In renouncing its Yaminite independence, the town leadership proclaims its readiness to perform the ritual of alliance, which will establish not merely a general loyalty to Zimri-Lim but a specific fraternity with the Simalite division (gayum) of the Nihadû: ˘ Another thing. Uranum and the elders of Dabiˇs came, saying: “By extraction, we are among the Yahrurˆ u, but never (as) yarr¯adum; also, in the back country, we have neither hibrum nor˘ kadûm.” We are native(?) to the Yahrurû. Let us (now) come into ˘ of the Simalite(s) as (part of) the Nihadû (tribe), ˘ the midst so that we may slay the 127 ˘ (treaty) donkey.

In the context of this letter, both the hibrum and the kadûm appear to define particularly Yaminite social-political ˘categories. The hibrum appears to be ˘ perhaps politically the nomadic component of the Yaminite population, distinct from what a sug¯agum governs from a town center. This is the part of the tribal people who live with the flocks, outside the settled villages and towns of the kinsmen they support.128 The kadûm is obscure, but may be a lower-level leader among the hibrum, a role that could then complement that of the town-based sug¯agum.129˘ To my knowledge, neither category belongs to Simalite tribal structures, and these phenomena should provide another case of contrasting systems between the Yaminite and the Simalite peoples. Notice, by the way, that when Dabiˇs affiliates itself with Simal, its leaders compare the original identification with Yahrurˆ u to new association with the Nihadû, a known Simalite gayum. If Ya˘hrurˆ u is a li mum, this would ˘ were perceived to occupy add to ˘the sense that the li mum and the gayum comparable hierarchical levels.

c. the local leader of tribe and town: the sug a¯ gum in service to the mari kingdom Because of the contrasting roles of the sug¯agum in the Simalite and Yaminite tribal confederacies, I have had to introduce this leader in the context of defining two distinct forms of tribal organization. The sug¯agum requires attention beyond this one important issue. As Nakata observed, the sug¯agum is indeed the primary individual leader of towns, both within the Mari kingdom and beyond. This function is particularly visible in the

64


Yaminite towns, where tribal and urban structures merge. If the towns of the Mari kingdom were headed by sug¯agums who could nevertheless maintain tribal affiliations, it becomes important to explore the relationship of these leaders to their sovereign. The relationship was quite varied, and the power of the sug¯agum was most often rooted in local support rather than in the imposition of outside royal authority. Finally, we return to the question of the word itself and the original role it defines. Was it originally a tribal term or not?

1. The Town sug¯agum As we search for order in the chaotic glory of the Mari archives, early definitions are bound to have to allow room for further nuance. Patterns that explain one body of evidence as a universal phenomenon may in fact apply to one part of the data only. This seems to be true of the sug¯agum, a title that seems to have been adapted to almost every political system found in the archives. At every point, we need to delineate limited uses of the term, avoiding hasty generalization. a. Reading for the sug¯agum of Towns. To begin, several letters associate the ´ s plural sug¯ag¯u with assisting leaders called lu.meˇ NU.BANDA3 , or laputtû.130 Mari governors from the core districts of Mari, Terqa, and Saggaratum are found to address systematically all the towns of their districts by contacting both the plural sug¯ag¯u and the laputtû (“deputies”?),131 sometimes adding the looser term “elders” for other undifferentiated participants.132 The combined leadership of sug¯agum and laputtûm is only attested for the core districts of Zimri-Lim’s kingdom, in the Ah Purattim (“Banksof-the-Euphrates”). We cannot assume that sug¯agums ˘outside that administrative system had such company, whether within town or herding group frameworks. It is possible that the Sumerian writing NU.BANDA3 covers a West Semitic Amorrite term, rather than the Akkadian laputtûm, but whether or not this is true, the Mesopotamian writing is appropriate to its restricted attestation in the most carefully ordered part of the realm, most attuned to southern institutional traditions. I have never seen the (plural) laputtû in texts written by tribal leaders themselves, Yaminite or Simalite, nor have I found any “laputtû of the Hana.” So far as they were responsible for work crews, these crews were the˘ particular domain of district (hals.um)-level administration.133 This palace-oriented function may also be˘ reflected in the association of NU.BANDA3 meˇs (“deputies”) and lu.meˇ ´ s GAL.KU5 meˇs (“section chiefs”) in a letter from king Yasmah-Addu to his brother Iˇsme-Dagan, where the laputtû deputies work in ˘a military context.134 Because we find many references to plural sug¯ag¯u in tribal gatherings, it is necessary to distinguish those from the assembled district leaders discussed


65

here. When ruled by outsiders, towns frequently deal with external affairs by a collective voice attributed to “elders,” and these are gathered at the towns’ own initiative. The towns do this even where it is clear that they also have individual rulers. In the examples just treated, the royal officials do not assemble the elders of the towns they want to address, but rather they prefer to work through individual leaders, the sug¯ag¯u and their deputies. This difference should reflect the symbolism of power relations. The sovereign both communicates and exercises power by acting through the individual leaders whom he has at least approved, while the subordinate town finds its most powerful stance in presenting a collective front. In his attempt to link all sug¯agums to towns, Nakata observes that appointment is always by such settled “locality.” To demonstrate the point, he must insist that the “clan” (gayum) names, sometimes written with the placedeterminative KI, are in fact “geographical designations” (1989, 117). If so, they only mean “a place where there are people of tribal division X,” and we have no sign that they represent individual towns. The special authority of kings to appoint sug¯agums appears to apply only to the leaders of towns. Under Samsi-Addu and Yasmah-Addu, we have references to assignments at Yail and Tizrah,135 with Dur-Ya˘hdun-Lim, Hiˇsamta, and T.abatum in the districts of the Ah˘ Purattim under˘Zimri-Lim.136 The pattern of royal appointment for town˘ leaders does not, however, prove that the title of sug¯agum was uniquely linked to towns, but rather that kings governed the selection of town leaders in their own domains (m¯atum). It is not clear whether payment of the sug¯ag¯utum fee by pastoralist leaders demonstrates royal appointment; I have found no conclusive proof.137 This royal authority could still operate outside the core districts in lands that had sworn loyalty to the Mari king, as at Amaz in the northern kingdom of Andarig (ARM X 84:10). In the Habur drainage, where Zimri-Lim ˘ maintained strong influence and many vassals, status as “leader” (sug¯agum) or “king” (ˇsarrum) was a matter of negotiation. However small these polities may have been, the translation of ˇsarrum as “king” is certainly appropriate, because it represented a coveted rank that put even a minor local ruler on the same titular plane as the dominant kings of Babylon, Eˇsnunna, Aleppo, Mari, and the like. Ibal-Addu, the king of Aˇslakkâ, one of the more powerful Habur polities, seems to denigrate the ruler of Hurrâ, a town usually subject˘ to the larger kingdom of Aˇsnakkum, when he ˘identifies him as only a sug¯agum in a letter to Zimri-Lim.138 We seem to find a similar situation with Yaminite rulers in a damaged letter that passes on to Zimri-Lim the doings of the Yamhad ˘ court. King Yarim-Lim of Yamhad is reported to have supported Zimri-Lim’s ˘ royal claims, perhaps near the start of his reign, to some ruler who has been addressing Zimri-Lim as diplomatic “brother.” Yarim-Lim instructs this character, who appears to be Dadi-hadun, the Yaminite ruler of the Rabbû at Abattum, to write to Zimri-Lim as ˘“father” and “lord.” He goes on to order

66


that each “sug¯agum” take an oath in the temple of the storm god Addu to fight loyally for Zimri-Lim.139 These sug¯ag¯u appear to be the actual Yaminite rulers, rather than lower-level leaders, and Yarim-Lim would then be using the term in order to emphasize their subordinate status. Even beyond the question of tribal leadership, we should not assume that all towns had a sug¯agum. Perhaps all towns incorporated into the central districts of Zimri-Lim’s kingdom had to be governed this way, and most of our evidence for town-based sug¯agums under the reign of Zimri-Lim comes from these districts. Outside that circle, however, the pattern is much less consistent. The same pre-Amorrite towns that show the strongest tradition of collective leadership may not have had a place for this individual ruler. Tuttul is well known by its Mari royal representative (hazzannum), but we know of no sug¯agum. The same may be said for Imar. ˘Urgiˇs had a “king,” however weak. It appears that the district centers of Terqa, Saggaratum, and Qat.t.unan needed no more than their governors (ˇsa¯ pit.um) and did not duplicate the individual position in charge. The three sug¯ag¯u from Saggaratum whose sug¯ag¯utum fees were recorded in the same administrative document (ARM VII 311) are probably not the joint leaders of this district center, but rather leaders of towns or (Simalite) tribal units in the district. In light of the last text, it is striking to find a letter to a Mari official called “our father and our lord,” sent by three men who identify themselves as “the sug¯ag¯u of Luhayum,” probably in the western Habur region.140 The ˘ of the letter alone addressee is Ibal-el, ˘the Simalite merhûm. The heading ˘ shows that we cannot insist that every town had one sug¯agum, but how can there have been three? It seems most likely that their identities as individual leaders must have been defined elsewhere. An explanation may be found in the unexpectedly forceful declaration of service to the merhûm. If these ˘ titles may are Simalite leaders based in a Simalite town, then their sug¯agum derive ultimately from their position in the Simalite tribal structure, with its hana pastoralist orientation. With the basis for leadership defined within the˘ tribe rather than by town, more than one sug¯agum could inhabit the same settlement. In sum, the connection between town and individual sug¯agum leadership is real, and worth exploring, but it is not universal. If we try to impose this pattern on every element of the Syria-Mesopotamian population, we will surely miss important alternatives to this structure. b. The Yaminite Towns. In the anthropological literature, tribal peoples are well known to be associated with settled life in many cases. With the Mari evidence, however, where we find an obvious association with a mobile pastoralist way of life, it was once natural to identify the tribal groups with a nomadic way of life, so that “tribal” and “nomadic” tended to be conflated into one social pole against “sedentary” and “urban.”141 Against this easy


67

division, the Mari archives have always confronted us with the oddity of “towns,” indicated by the usual Akkadian word a¯ lum, that are also identified with the Yaminite tribal group. Kupper invoked the idea of semi-permanent settlements or something else less than the other towns of the region, but the texts themselves do not distinguish them from the standard item. The substance of real Yaminite towns, even cities, is evident in the two lists of Yaminite soldiers by numbers promised and then either present or undelivered for the expedition to Babylon (ARM XXIII 428 and 429).142 These two lists record all of the Yaminites from Zimri-Lim’s districts of Mari, Terqa, and Saggaratum who are supposed to serve with the king’s armies. The men are divided first of all by responsibility to their recognizable tribal kings, though their title is not mentioned: Yarim-Lim of the Yahrurˆ u, Yasmah-Addu ˘ of the Rabbˆ ˘u, and of the Yarihû, S.urahammˆ u of the Amnanˆ u, Dadi-hadun ˘ ˘ ˘ Hammi-iˇstamar of the Uprapˆ u. Notice that none of the five Yaminite tribes is ˘ omitted, confirming the exhaustive nature of the documents. Within these tribal groupings, however, the Mari administration defines the entire Yaminite population by towns, with no component named by clan. Many of the named towns are known by repeated reference, and towns such as Miˇslan and Samanum are of substantial size. Millet-Albà (forthcoming) now proposes population counts for numerous Yaminite towns in the Mari kingdom. Miˇslan stood at the center of Yaminite resistance in their revolt against Zimri-Lim. The Yaminite letters that come from the time of this episode mention hundreds of troops that gather there,143 and in one case the writer closes with reference to the town’s fortifications: “Miˇslan is well; your brother Yaggih-Addu is well. . . . We, your servants, are well. No one ˘ is neglecting the patrol of the rampart and the gates.”144 Zimri-Lim finally had to “capture” Miˇslan (verb .sab¯atum), an action associated with successful siege.145 Samanum had a shrine for the goddess Annunitum worthy of a visit by Zimri-Lim, and the high official Asqudum once sends a man to Samanum to gather boats.146 These are typical of proper towns. Letters from Zimri-Lim’s district (hals.um) governors show a similar as˘ sumption that the Yaminites within their domains are to be addressed by their towns. While the military lists hold the tribal kings responsible to deliver their promised troops, the Mari governors leave these figures outside their purview and instead consider the Yaminite men in charge to be the (plural) sug¯ag¯u in the towns.147 Kibri-Dagan of Terqa gathers the “sug¯ag¯u of the towns of the Yaminites” to receive orders in tracking down a missing person: When I was staying at Mari (in) the company of my lord, I heard about this matter in my entourage, so I assembled the sug¯ag¯u of the towns of the Yaminites and warned them as follows: “Whoever you are, the one who has (even) one man leave your town for the uplands, and you fail both to arrest him and to conduct him to me, you shall die – you shall not live.”148

68


When this governor has problems with Yaminites not reporting for work on a district canal project, it is again by a¯ lum, from which only half of the promised number are showing up: I gathered the district work crew and the residents of Terqa for the maintenance of the Mari canal. From the crew of the [Yaminite] towns, half did not show up. A town that (owed) a crew of 50 as its quota provided me a work crew of 25, (and) [a town that (owed) a crew] of 30 [as its quota provided me a work] crew of [15].149

He exhorts Zimri-Lim to rouse the Yaminites in his district by writing directly to their sug¯ag¯u (lines 16–18).150 ARM III 70+ appears to define the entire district of Terqa by two entities, all with the same leadership structure: At present, the sug¯ag¯u and the deputies of the Yaminite towns, along with their peers (at) Zurubban, Hiˇsamta, Himmaran, and Hanna, have not delivered the silver of ˘ wool.151 ˘ ˘ their quota to buy

Kibri-Dagan describes a sweeping resistance, attributed to all the Yaminite towns together and to four named towns excluded from the Yaminites that repeatedly represent a major block in the Terqa district.152 Whether or not the governor reserves some minority of compliant non-Yaminite towns, the towns mentioned here are treated as having the same essential form.153 All of this could be attributed to the demands of Yaminite submission to Zimri-Lim’s kingdom, and indeed it excludes any recognition of Yaminite subjects legitimately on the move. The Yaminites themselves, however, embrace their identification with towns. In the letter quoted earlier at length to establish the meaning of “hana,” Yasmah-Addu, ruler of the Yaminite Yarihˆ u ˘ a group of˘ “tent-dwelling” (hana) sug¯ag¯u from ˘ tribe, writes to Zimri-Lim that ˘ so receive back the Yaminites want to be reconciled with the Mari king and their towns. “Go to Zimri-Lim and ask for our towns,” they have said.154 While their current nomadic condition is clear, they are leaders who want their town seats. These rulers do not consider themselves to be an essentially different sort of sug¯ag¯u from their gathered peers, from whom they desire only help in rejoining the rest. Notice that unlike the sug¯agums identified with gayum “divisions” in the administrative texts, these Yaminite leaders are defined by their relationship to a single a¯ lum each. Kibri-Dagan confirms the impression that Yaminite sug¯ag¯u serve one town each when he writes “to the Yaminite towns” and is answered by “the sug¯agum of Dumtên” (III 38:15–18). c. The sug¯ag¯u and Elders Together. Before moving to the position of the sug¯ag¯u under kings, I want to point out an interesting pattern in the combined reference to these “leaders” with elders. In general, the categories of plural sug¯ag¯u and elders were not used in accounts describing a strictly urban


69

setting, where these would be addressed or gathered together “by town” (¯al¯ıˇsam). As observed earlier, elders are occasionally added to the combination of sug¯ag¯u and laputtû when describing town leadership, but the sug¯ag¯u and elders alone are associated only with tribal gatherings. The specific combination of plural sug¯ag¯u and elders represents the full high leadership of the Yaminites when gathered to confirm an alliance with the kings of Zalmaqum (ARM XXVI 24:11). Unlike the sug¯ag¯u of the Mari districts, who are addressed together only to serve kingdom needs, these tribal leaders meet at their own initiative to pursue their own external affairs, like town elders. We see the same combination in a major Yaminite gathering linked to maintenance of the same alliance, with Asdi-takim and his brother kings, again meeting to slay the treaty ass.155 The Zalmaqum kings, their partners, likewise address the outside world with a collective face and also include plural sug¯ag¯u in negotiations, this time with the kingdom of Yamhad.156 Yarim-Lim, the king of Yamhad, tells Mari envoys that ˘ messengers to Zalmaqum with the ˘ following instructions: he has just sent “As long as you stay (there), two of their kings must come to me from their group. Then, if their kings do not come, their sug¯agum should take the lead and you must return with your explanation.”157 I know of three other texts where the plural sug¯ag¯u and the elders appear together, also representing tribal peoples. When the Numhâ tribal leader˘ of Zimri-Lim’s ship of Kurdâ confronts king Simah-ilanê over his acceptance ˘ superiority (FM II 117:30–31), Zimri-Lim’s correspondent identifies them this way.158 Elsewhere, Zimri-Lim asks his officials to persuade the same combination of Mutebal leaders to meet with him and then join an attack on ˇ Subat-Enlil. A Mari official named Sumu-hadˆ u quotes his own letter to two other officials, following this with the plea˘ to fight: My lord’s tablet reached me en route. The present message is (in) response to my lord’s tablet. Let the sug¯ag¯u and the elders of the Mutebal listen. Turum-natke and ˇ the residents of Subat-Enlil keep writing constantly to your lord to open the city of 159 ˇ Subat-Enlil. Let them listen to this [message]. Conduct the sug¯ag¯u and the elders to my lord at [Dur-Yahd]un-Lim. [My lord] can put them at ease and make preparations for a campaign. ˘(lines 23–5)160

It is interesting to notice that we do not find the combination of sug¯ag¯u and elders of the “Hana,” the common designation for Simalite tent˘ alite gathering. ARM XXVI 40 displays the same dwellers, or for any Sim independent function for the lu´ su-ga-gumeˇs u` ma-ah-re-et [ . . . ] of what might ˘ Asqudum. “The sug¯ag¯u be best restored as the Hana, in negotiations with ˘ and the elders(?) (ma-ah-re-et) [of the Hana(?)] arrived. I inquired about ˘ will [assemble(?)] in one month plans for assembling the˘Hana. The Hana ˘ ˘ 161 at best.” If mahrûm means “earlier, older,” it could even be an Amorrite ˘ equivalent for “elder,” though it might also refer to simple priority.

70


We do, however, find the combination of plural sug¯ag¯u and elders in one more text that appears to reflect a Simalite meeting, not defined by the term “Hana.” The letter sent by the three sug¯ag¯u of Luhayum requests the ˘ merhûm Ibal-el to come to Der for a meeting of “elders ˘ Simalite and sug¯ag¯u”: ˘ “Bring yourself to Der. The elders and the sug¯ag¯u must come(?) to us. . . . ”* The western town of Der, near the Balih River, was an old Simalite base, pos˘ 162 It is possible that the Simalite sibly associated with the tribal ancestors. sug¯agum had lower status than those in neighboring political systems, reflected in the reversed order of the combination.

2. The sug¯agum under Kings The term “sug¯agum” in the Mari evidence has attracted considerable interest, in large part because it represents a significant leadership function that does not come from well-known southern Mesopotamian traditions. It appears to be a West Semitic term, whatever its precise relationship to town or tribe.163 In any case, the sug¯agum was a local leader who served as a key intermediary with the royal administration. Indeed, the sug¯ag¯utum fee paid to the king of Mari for the right to hold this title does show that Zimri-Lim had incorporated the sug¯agum into a formal hierarchy that he controlled.164 It is less certain, however, to what extent the sug¯agum role as such can be defined by its relationship to kings. In this section, therefore, I examine some of the relevant evidence. a. The Rowton Model. Any treatment of the sug¯agum under kings works within some larger framework of ideas about tribal peoples related to the Mari state. The early articles on this subject by Talon and Young and Matthews were heavily influenced by the theoretical model of Michael Rowton, which included direct applications to Mari, and this model is worth examination. In the period immediately preceding the formation of the publication team led by Durand, perhaps the greatest influence on Mari studies came from Rowton, who produced a series of articles in the 1960s and 1970s on nomadism (see the Bibliography). Through the years since his work, Rowton’s terminology has become a prominent part of many treatments of ancient Near Eastern nomadism, especially by nonanthropologists. In spite of the genuine progress Rowton achieved, the very center of his analysis was based on one enormous misconception of the Amorrite period and depended on an equally problematic cross-cultural comparison. Rowton distinguished the transhumant nomadism visible in the Mari archives as “enclosed” and “dimorphic.” It was enclosed by its situation in the interstices of a heavily settled region and dimorphic by its direct participation in that sedentary society, with tribal peoples settled in towns ruled *

ARM XXVIII 120:26–9; 26 ˇse-ep-ka a-[n]a Di-i[r]ki 27qu-ur-ri-ba-am ´ gi meˇs 29 a-na .se-ri-n[e] [l ]i-il ! -[li-ku].

lu´ [SU ˇ G]Imeˇs 28 u` lu´ su-ga-

.


71

by their own chiefs (1967, 114–16; 1973b, 201–4). The tribal towns and populations were “autonomous” under larger states, which by their incorporation of these tribal peoples were also dimorphic. Rowton defines Mari itself as an intermediate dimorphic state, a territorial state within which tribal and dimorphic chiefdoms played a prominent role, but whose own authority stood apart as centralized bureaucracy (1973b, 203–4; 1976a, 27–8).165 All of his specific explanations of the Mari evidence assume that tribal peoples interacted with the royal administration as essentially “other.”166 The comparative problem with Rowton’s work lies in the nature of nomadism in the modern world of nation-states. The anthropological comparisons used by Rowton, and in all general studies of Near and Middle Eastern nomadism, involve peoples who are ultimately governed by larger states, whether the modern regimes of Iraq, Turkey, Iran, and the like or by earlier kingdoms, finally going back to the rule of the Ottomans before World War I. Rather than imagine any similarity to the more remote, desert-based kingdoms of Arabia or elsewhere, Rowton assumed that in a world with such a large urban population, with such dependence on agriculture, and every indication of a highly developed culture, the nomads must be sharply distinguished from the state. This entire regional political landscape is mismatched with the actual pattern of early pastoralist societies in the Syrian Euphrates and its backcountry. Anne Porter turns Rowton’s terminology on its head: We do not see “enclosed nomadism,” she says, but rather, “enclosed urbanism” (2000, 440). All political society in the ancient period need not be defined in relation to the city-based states. In spite of Rowton’s influence, there have been critics. Even without the new wave of Mari publications, Kathryn Kamp and Norman Yoffee already observed difficulties with Rowton’s comparisons with modern societies, arguing that “divergences from modern analogies can and must be explained” (1980, 92). Rowton oversimplifies the movements toward and away from sedentary life, and he unnecessarily isolates tribal social organization from urban settings and full integration into the state. After all, the Amorrites also became kings (Kamp and Yoffee, 1980, pp. 93–4, 98–9). Robert Adams likewise responded without new evidence from Mari, with focus on the opposition defined by Rowton’s dimorphism. “I regard the groups immediately concerned as having been more in flux and less polarized” (1981, 136; also Liverani 1997). If possible, the critics underestimated the degree to which Rowton had misunderstood the social and political situation at Mari. Far from representing the ultimate “other,” the power of sedentary civilization set against and striving to control the unruly tent-dwellers, Zimri-Lim governed Mari as a fully integrated tribal kingdom, within which the ruler’s Simalite kinsmen represented his primary base of power. Other Amorrite kings of the early second millennium retained some connection with the tribal social organization of their roots, but in the preserved cuneiform documentation, this

72


tends to be buried under the institutional language of kingdom, palace, and city.167 In the case of Zimri-Lim, the tribal connections are especially visible because they had been his political foundation before he came to power at Mari. The pastoralist orientation of Simalite tribal structure survived in Zimri-Lim’s separate governance under the chiefs of pasture (merhûm), tied ˘ In the also to the major military support provided by the Simalite Hana. ˘ Mari evidence, the encounter of nomad and city is indeed real, but especially during the reign of Zimri-Lim, the power did not lie in the hands of a distant other, and the population was dominated by tribal peoples sharing the same experience in various expressions. In the case of the sug¯agum, the imperial model of a society in two separate forms (dimorphic) is inadequate to the actual range of definition and function in the Mari texts. As we reevaluate the relationship between the sug¯agum and kings, we must shed as far as possible this picture of foreign imperial power. b. Royal Appointment. By the nature of our palace archives, we see mainly the king’s interest in the world surrounding him. In the case of the sug¯agums within each kingdom, this interest shows royal control of the office by power to appoint (verb ˇsak¯anum), even if in practice the nominees may have emerged out of the crucible of local politics. Both Samsi-Addu and Zimri-Lim exercise this authority over the government of towns within their realms.168 Even when the individual towns display a stubborn independence, as when Dur-Yahdun-Lim evicts a current officeholder, it is assumed that the ˘ replacement will have to come from Zimri-Lim. “I can no final decision on longer serve as sug¯agum of Dur-Yahdun-Lim. I have been driven out. Let ˘ 169 them appoint whomever they want.” Against this prominent evidence, another thread must be given serious consideration. It is not clear that Simalite sug¯agums are appointed by the king, and there is a possibility that this role fell rather to the merhûm, at least by tradition. In A.486+, discussed above in connection with˘ the gayum in Simalite tribal structure, Ibal-pi-el takes it on himself to set two section chiefs over clans within his Hana batallion for reception at ˘ Babylon: Bahdi-Addu had assigned twelve men as section chiefs. I questioned (him) as follows: ˘ number of the army is one thousand, and for an army of one thousand (there “The should be) ten section chiefs (rabi pirsim). At present, you have assigned section chiefs (as if) for an army of 1200, and you have assigned 24 lieutenants (laputtûm) (as if) for an army of 1400. If they investigate, what indeed will be the result?” I thought (about it) further, saying, “If I trim away two men (from) among the section chiefs whom he appointed, I will turn (their) opinion against me.” I was afraid of this, so I appointed Sulum over the Hana Yabasa and Bihirum over the Aˇsarugayum.170

˘ ˘ This is not a sug¯agum appointment, but it may follow normal lines of authority. Another text has one Yams.i-hadnu give silver and sheep to the early ˘


73

merhûm named Bannum, with a direct request to be appointed as a sug¯agum: ˘ When Bannum served as merhûm, Yams.i-hadnu presented him 31/3 minas of silver ˘ me (as) sug¯ ˘ agum.” [Bannum] answered as follows: and 300 sheep, saying, “Appoint 171 “[Wait until] my lord arrives. . . . ”

Though Villard explicitly considers and rejects the possibility that the early merhûm Bannum had power to nominate a sug¯agum out of that office, he is ˘ working within an interpretive framework that appears to me to underestimate the presence of a Simalite social and political structure not primarily defined by towns (1994, 297).172 If we accept that the king allowed tremendous autonomy to his Simalite kin under the powerful merhûms, and add that ˘ we have no evidence for royal appointment of Simalite clan-based sug¯agums, it is simplest to accept the text at face value, with its assumption that Bannum could have made the appointment without consulting Zimri-Lim. Bannum appears to have held the title merhûm before Zimri-Lim became king of Mari, ˘ title that he was considered capable of however, and it may be under this appointing a sug¯agum, and not simply by the prestige of his conquest. c. sug¯ag¯utum and Census Oath. Kings could not afford to allow populations within their circle of power to remain uncommitted, and both oaths and tribute formed one element of the bonds created. The sug¯ag¯utum represented both a requirement and a benefit, reflecting the income produced by the people governed, a transaction that confirmed the accountability of local leader to sovereign. Under this bond, the sug¯agums could even be called the “servants” of their ruler, in spite of the fact that they were not ordinary royal officials. Oddly, my only evidence for (plural) town sug¯ag¯u as the servants of a Mari king comes from the Yaminites in the Terqa district: Regarding the news of the deliberations that our lord undertook at Terqa, our lord has been considering the return of the Yaminite sug¯ag¯u to (their) towns. We learned of this idea from the entourage.173

Durand considers that this unusual language may reflect specific reference to the Yaminite rulers who have accepted Zimri-Lim’s sovereignty and may now be allowed to return to their seats of power in the Euphrates valley (Charpin, personal communication). Outside of this text, the plural sug¯ag¯u appear as “servants” in two different letters, with leaders of the Simalite Hana, Zimri-Lim’s core constituency.174 ˘ One critical stipulation of this commitment was to provide men to fight for the suzerain on request. When the occasion arose, this requirement would be undertaken under the strict oversight of the kingdom, by census with accompanying oath (ARM XIV 64:4–9, passim, under Zimri-Lim).175 King Samsi-Addu considers that the central task facing Yasmah-Addu in ˘ undertaking a census is to reach the unregistered sarr¯ar¯u (“resisters”?) in the steppe, using the sug¯ag¯u leaders of the secure Ah Purattim to go after ˘

74


this more slippery population, with the goal to make them swear the n¯ıˇs ilim oath (ARM IV 7+:13–22). This oath should carry with it not just a promise to supply some number of troops, but also a general commitment to submission and loyalty toward Samsi-Addu as lord. Against this norm, it is remarkable to see the same king upbraid his son for even thinking of including the Yaminites in his census. Instead, their sug¯agums should be reminded of their personal bond, engaged under curse, and urged based on this alone to supply the full contingent: The king is going on campaign. Let the whole (count) be absolutely complete, down to the youngest boy. Any sug¯agum whose (drafted) troop is not complete, and who leaves one man behind – he has taken the loyalty oath.176

Under Zimri-Lim, the demand for military service probably took a different form, with the general change in administration. The lists of mustered Yaminites from Mari’s core districts shows registration by towns, an arrangement that would have complemented the responsibility of each sug¯agum for one town (ARM XXIII 428 and 429). As seen in our discussion of the gayum, the Simalite sug¯agums appear to have governed tribal units not defined by town, though they may have resided in settlements. Following the pattern of their sug¯agums, the Simalites may have performed military service only as Hana, in a chain of command and request that was quite independent of the˘districts and their governors. The large Hana fighting forces mentioned ˘ so often (see section E) may represent the entire Simalite military contribution. I am aware of no reference to any census of Simalite Hana under the ˘ rule of Zimri-Lim.177 It appears that Zimri-Lim declined to impose a palacebased draft on his kinsmen, but he left this to the existing lines of tribal commitments. In the pastoralist Simalite context, a draft might be considered unnecessary because participation was automatic, but this is not so. On one hand, Zimri-Lim had to negotiate Hana participation in military cam˘ rihsum assembly (see Chapter 4). paigns, as shown in letters mentioning the . ˘ alite Hana man by gayum and On the other, our text for tracking down a Sim sug¯agum (FM III 136, in section B) has to do with a˘ soldier who has taken leave without permission. d. sug¯agum Activities. Our focus on the political life of towns in the Mari archives does not require an exhaustive review of all the activities of the sug¯agum that are attested in the Mari archives, but it is worth noticing the general patterns. The royal service of the sug¯agums is focused especially on keeping track of people, as the census oaths already demonstrate. This responsibility reached beyond the monitoring of military commitments, however.178 According to one letter from the governor of the Terqa district, personnel must be provided not for the army but for canal work (ARM III 6:5–7). The local sug¯agums are also held responsible for the arrest of men who have left their proper towns.179


75

Hana sug¯agums serve various more direct military functions than those ˘ defined by leadership of towns. For military service under Zimri-Lim, all Simalite sug¯agums could in effect be Hana leaders, and thus automatically they become middle-ranking officers.˘180 One text shows use of a Hana ˘ sug¯agum as a secret messenger, again within a Simalite military structure (X 91:3 ). Local sug¯agums would serve as the immediate legal authorities (ARM V 72:10 ), and they could be intermediaries with the royal administration in coordination of festival celebration (XIV 8:5–8). In general, the word sug¯agum is known almost entirely from Mari, though it occurs as a personal name in other Mesopotamian texts from the early second millennium. One Sippar document from this period identifies a witness by the title sug¯agu.181 A letter found at Tell ar-Rimah offers slightly more ˘ king Aqba-hammu context, saying that the writer had sent a message to the ˘ 182 reand the sug¯agu(s) about the sale of certain cloth. This combination calls the Mari references to a king and elders, the circle of leaders in the royal court, a usage not found in the Mari archives.

3. The Tribal Origin of the sug¯agum Even in the core structure of Zimri-Lim’s administrative districts, the sug¯agum should not be viewed as a royal appointee in the usual sense of the word. Remember the account of how the town of Dur-Yahdun-Lim in ˘ the district of Saggaratum threw out their sug¯agum (ARM XIV 46:7–9). The town has a say in this appointment in a way that does not apply to the officials of Zimri-Lim’s palace, his army, or even Zimri-Lim’s district governors themselves. Among the Simalite pastoralists, it is not even certain that the king appointed sug¯agums. In some circumstances, at least, these Hana sug¯agums acted with considerable independence, especially when they˘assembled for collective business, as is discussed in Chapter 4.183 The primary setting for the role of the sug¯agum in collective decision making seems to be the tribal organization of the Simalite herding groups, rather than the hierarchy of towns in palace-administered districts. “The (plural) sug¯ag¯u of the district” appear then to represent a secondary formulation, molded to the lines of the royal m¯atum.184 The traditional authority of freely assembled sug¯ag¯u is also visible in the coalitions of Zalmaqum and Ida-Maras. in the Balih and Habur river ˘ basins, respectively (see Chapter 4). Zalmaqum conducts˘ negotiations with Yamhad by its kings and its sug¯ag¯u, where the latter may lead the delegation ˘if no actual kings can show up (XXVI 12:2 –4 , 8 –9 ). In the available texts, Ida-Maras. faces the outside world only as an alliance of kings, without reference to plural sug¯ag¯u, but the effect is indistinguishable. The rulers of these small and vulnerable polities coveted the individual status that accompanied kingship even as they depended heavily on the traditional collective mode of decision making in their effort to survive among

76


the major powers. These coalitions were no temporary alliances but were conceived as unified political entities deserving the label of m¯atum (see Chapter 4). ˇ We have no evidence for sug¯agums among the leadership of the Subartum coalition. The alliance of kings found in the Mari archives represents a definable political entity, not a vague description of distant foreigners, and some specific geographic location must underlie this polity. We cannot ˇ assume that the kings of the Subartum alliance in the early second milˇ lennium had deep roots in the region, and the m¯at Subartim could conceivably even have a West Semitic–speaking population. These cautions ˇ acknowledged, one Mari letter mentions a Subarˆ u language, separate from ˇ “Amorrite,” and we expect some relationship between the Subarˆ u language ˇ ˇ and the Subartum land. It is possible that Subartum simply has no sug¯agums, who may pertain especially to the regions visited by the Yaminite/Simalite tribal cluster. After all, Ida-Maras. has strong affinities with the Simalites, and Zalmaqum with the Yaminites. The absence of the word “sug¯agum” ˇ ˇ in Subartum would suit nicely the possibility that the Subarˆ u language is Hurrian (Durand 1992b, 125; Charpin 1994, 181–3).

d. the chief of pasture: the mer h u ˆm

˘ One of the most important roles in the Mari political system was that of the “chief of pasture,” or merhûm,185 but the position remains difficult to ˘ define and deserves a more complete examination than I manage here. There were merhûm chiefs of pasture among the Yaminites, but most of our evidence for˘ the role comes from the Simalite kingdom of Zimri-Lim. Throughout most of Zimri-Lim’s reign, two merhûms named Ibal-el and Ibal˘ pi-el held special authority over the Simalite Hana herding communities. Because their primary responsibility for actual ˘land lay in the steppe, away from the river valley districts (hals.um) of Zimri-Lim’s governors (ˇsa¯ pit.um), ˘ the king and his inner circle. Even when the merhûms reported directly to ˘ their tent-dwelling tribesmen moved through the valleys of the Euphrates and the Habur, however, the merhûms retained authority over their people, ˘ ˘ and the governors could not interfere freely. Like their wide-ranging herdsmen, the merhûms seem to have expected more room to maneuver than their ˘ settled countrymen, physically and politically. The role of the merhûm is especially important for understanding the ˘ place of tribal organization in the kingdom of Zimri-Lim. When we encounter the “Simalite” label, it is almost always in connection with the mobile herdsmen who are normally identified simply as “Hana” (tent-dwellers). ˘ assurance that “the Ibal-el sometimes signs off in his letters to the king with flocks (nawûm) and the Simalite(s) are well,” the two defining categories of his responsibility.186 The only known subdivisions of the Simalite tribal people, the gayum, relate particularly to the Hana herdsmen. Moreover, ˘

The Chief of Pasture

77

we never hear of a “Simalite town” or a town identified by any Simalite gayum, whereas Yaminite towns are part of common parlance. In political terms, these Hana, their gayum divisions, and the sug¯agum leaders under each division ˘all belong to the domain of the merhûm. Zimri-Lim defines ˘ his realm as “Mari and the land (m¯atum) of the Hana” (see Chapter 3), ˘ so that his whole people are identified with the “tent-dwelling” component of the kingdom, the component specially governed under merhûm ˘ leadership. My analysis of the merhûms’ role in the kingdom of Zimri-Lim is not greatly ˘ different from that of Durand and his colleagues, except that my understanding of separate Simalite and Yaminite structures leads me to expect somewhat different adaptations of the merhûm position. Durand discusses ˘ he observes that the Simalite the merhûm at some length in LAPO 17, where ˘ merhûm is assisted by his own sug¯agums, with general independence from ˘ administration, except under the umbrella of sworn loyalty (p. 497).187 royal At the same time, Durand proposes a single pattern in tribal leadership, within which the sug¯agums represent an authority that is based in a domain separate from that of the merhûm. The sug¯agums bear responsibility for the sedentary population (w¯aˇsib¯u˘tum), while two merhûms head the groups in ˘ transhumance (hibrum).188 ˘ This analysis regards the Simalite and Yaminite systems as roughly equivalent, especially in the contrast between w¯aˇsib¯utum and hibrum, and in the ˘ ultimately sedentary identity of all sug¯agums. My observation of separate tribal structures naturally leads me to suggest different relations between their leaders, including the merhûm. Also, I have not seen evidence that the word “hibrum” is used universally˘ for tribal people in the steppe, and it is not clear to˘ me that the verb waˇsa¯ bum (“to sit, stay, dwell”) describes adequately the varieties of encampment or sedentarization among people called hana ˘ or “tent-dwellers.” The interpretation of the substantive w¯aˇsib¯utum as “sedentaries” then confuses the question of how the tribal peoples are identified with “towns” (¯alum), a direct combination that is somehow avoided among the Simal. In spite of these differences, my proposals are built squarely on the discoveries of the Paris team. This is most evident in the idea that the merhûms ˘ lead the mobile pastoralist component of the Simalite population, outside the framework of districts and governors. Likewise, it is the fresh analysis of the deep identification of Zimri-Lim with his Simalite tribal people that prevents us from seeing the merhûm as a bridge between the tribes and the ˘ city-based state.189

1. The Simalite merhûms of Zimri-Lim ˘ The office of merhûm is known both for the Yaminites and under the reign ˘ it is by far most visible in the documentation for the of Samsi-Addu, but

78


Simalites during the rule of Zimri-Lim. It is therefore easier to define the other, fairly rare occurrences against the one well-attested institution. a. The merhˆ ums of the Sim al. I have already observed that the chief of pasture Ibal-el˘regards his primary domains of responsibility to be “the flocks” (nawûm) and “the Simal.” This identity is visible obliquely in the leadership of individual merhûms in actions against hostile Yaminite tent-dwellers. ˘ with Uprapû shepherds (XIV 86: 17–20), and he Ibal-pi-el’s men skirmish warns Qat.t.unan’s governor of further trouble with Yaminites (XXVII 113: 7–11). No Yaminite merhûm serves or is appointed by Zimri-Lim, in spite of ˘ the fact that the Ah Purattim includes a large sedentary population from this ˘ tribe. Often the merhûm is associated with those simply called “Hana.” He is the ˘ one who has authority to assemble the Hana to fight,190 and˘ he leads them ˘ into battle: Right now, my lord should take counsel with his servants, so that according to his deliberations he can have omens taken, and so that in view of good omens, if (free) to go, my lord might come to me. If not, my lord should write to the merhûm so ˘ us that the merhûm can take the lead of one or two thousand Hana and reach ˘ here. . . . 191 ˘

The shepherds of Qarni-Lim, king of Andarig, fear that the merhûm will favor ˘ the Hana in the partition of pasturage: “And the shepherds arrived, saying, ˘ ‘With the wells that we ourselves control, we can take (only) a third of the pasture, and the merhûm will control two thirds for the Hana.’”192 A merhûm’s ˘ ˘ (XIV 53:6–7), and a merhûm may ˘ have an entourage servant is called a Hana ˘ ˘ with a warkûm adjutant and subordinate Hana sug¯ag¯u (XXVII 93:7–11). ˘ b. Separate Authority. The merhûm is responsible to the king and his inner ˘ circle, outside the system of districts and governors. Zimri-Lim is expected to appoint a successor to Bannum, which shows that the king does have this authority over the Hana leaders.193 The king may give his merhûm a ˘ direct command, as the district governors are always requesting, when˘ tent194 This independence from the districts represents dwellers are involved. a severe irritation to the governors, whose area of responsibility effectively overlaps with that of the merhûm when the Hana camp in the valley re˘ gions. Yaqqim-Addu of the Saggaratum district˘begs his lord to keep a close eye on the merhûm in an affair involving grazing at Lasqum (ARM XIV ˘ 81:11–16). If anything goes wrong in the region of his responsibility, the governor does not want to be blamed for the behavior of people beyond his control.195 Other senior officials in Zimri-Lim’s inner circle also handle the merhûms delicately. Itur-Asdu, who had been sent to Nahur in Ida˘ king’s special representative, explains cautiously that ˘ he has Maras. as the gone to great lengths to avoid offending Ibal-el over a problem in salt and


79

grain deliveries to Nahur, but he needs the king to insist, where he himself ˘ cannot: I was not overbearing [toward] Ibal-[el], [saying(?),] “I cannot afford to be too overbearing, so that he acquires a grudge [against] me, [and(?)] never sends barley and salt into Nahur.” Because of this, I held back. Now, my lord should give strict orders to Ibal-el, ˘so he will send in barley and salt to Aˇslakkâ and Nahur. There is no ˘ barley and salt.196

Although not generally associated with the palace or posted at Mari, Simalite merhûms held influence on a par with the highest officials of Zimri˘ Lim’s inner circle. A letter the address of which is not yet published quotes another letter from Bunu-Eˇstar, the king of Kurdâ before Hammurabi, telling how he survived Samsi-Addu’s reign and complaining˘ that ZimriLim now neglects him.197 Durand translates his plea, “Maintenant, est-ce joli ce qu’a fait Zimrˆı-Lim? Quelqu’un qui n’y avait pas de part est entré sur le trône paternel et moi je suis un errant! Fais pression sur le mer uˆ m et les grands dignitaires qu’ils me fassent revenir sur le trône de mon père!” Besides his general authority over the tent-dwellers of the steppe, the military importance of the Simalite Hana makes the merhûm a high-ranking general ˘ as well.198 The two roles are bridged in the merhû˘m’s leadership of the Hana ˘ ˘ go assembly, the puhrum in which these tent-dwellers decide whether to ˘ to war: In my entourage, I heard the following information: Ibal-[Addu] the Aˇslakkâ-ite ˇ and Hamman the sug¯agum of Der took responsibility for leading Sadum-labua the ˘ Aˇsnakkumite and brought him to Ibal-el the merhûm for the assembly of the Hana at ˘ ˘ Siharatâ (of the?). . . . 199

˘ c. The merhûm in the Sim alite Tribal Structure. The most intriguing aspect of ˘ authority is its overlap with that of district governors and milithe merhûm’s ˘ tary generals, in both cases because it is unthinkable to place Simalite Hana ˘ under any leadership but their own. Insofar as tent-dwelling and sedentary Simalites considered themselves part of one coherent tribal entity, one wonders whether the sympathetic and traditional tribal authority of the merhûms may have encouraged Zimri-Lim’s tribesmen to prefer Hana identity to˘one defined by town, at least whenever anything of value ˘was at stake. When Yaqqim-Addu pleads to his king that “my district is (also) the merhûm’s,” he ˘ may refer to more than just Hana tent camps. Some significant fraction of ˘ his population may be looking to an outside authority. This is where the picture becomes hazy. Some significant proportion of the Simalite population appears to be governed by sug¯agums who are assigned by tribe and pay the sug¯ag¯utum fee in sheep. In both administrative and symbolic terms, these people consider themselves “tent-dwellers” (Hana), and under the governance of the two merhûms rather than the ˘ ˘

80


district governors. All of these sug¯agums should likewise fall under the command of the merhûm, not the districts and their governors. The question is, What fraction of ˘the Simalite population fell under the merhûm system, and for what occasions? For instance, do Simalites register for ˘military service only as Hana? So far˘ as the Simalites are organized in tribal terms and recognized as such among the population of Zimri-Lim’s kingdom, it seems to be only as Hana. We hear repeatedly that the governor (ˇsa¯ pit.um) addresses all the ˘ Yaminite sug¯ag¯u of a district. This official never does the same for the tent-dwelling Hana, but also never for the Simalites as a tribal group. It is natural that ˘Zimri-Lim would not need to identify his own tribal people the way he isolates the defeated Yaminites, but if his kinsmen require special administrative treatment as a tribe, how are they to receive it without some sort of terminology for communicating this identity? The only evident category in use recognizes only the mobile pastoralist face of Zimri-Lim’s tribal people, that is as the Hana. As Durand observes, it is clear that there are Simalite Hana sug¯agums˘ under the merhûm, but surely these are more ˘ following, and rather reflect˘the whole political structure in than a personal which the sug¯agums serve within each gayum division.200 d. The Political Place of the Division (gayum). The evidence presented up to this point includes two structures of Simalite organization that do not mesh. On one hand, individual Simalite Hana are identified and traced by gayum ˘ are governed by merhûm and sug¯agum. and sug¯agum, while on the other they ˘ If we know that there are more than one sug¯agum per gayum division, is there some individual who governs the gayum? Is it the merhûm? Surely not, ˘ two Simalite based on our available evidence. Zimri-Lim was served by only merhûms at a time, and there is now evidence to suggest that there may have ˘ only two Simalite merhûms during the reign of Samsi-Addu: Bannum, been ˘ h-Addu after Samsi-Addu’s death, and Yarimwho seized Mari from Yasma ˘ two gayum divisions. Moreover, the merhûms 201 Addu. We have far more than ˘ are never identified by gayum, and even the standard greeting in letters assures the king that “the Simalites” are well, not any individual gayum. The separation of Simalite Hana soldiers into two broad groups as Yabasu and ˘ Aˇsarugayum offers no demonstrable connection to the tradition of having two merhûms. When these units are named as part of Mari military forces, ˘ belong to a larger mix of troops led by a single merhûm, Ibal-pithey both ˘ 202 the el. I am aware of no evidence for the common category underlying Yabasu/Aˇsarugayum division. The Yabasu classification appears to derive from the gayum by that name, but we cannot assume that the military duality reflects actual tribal structures. In the end, I have found no evidence at all that the gayum has any political function, that is, that it has any leadership assigned to govern it, individual or collective. What does this mean? The problem involves more than


81

sedentarization, because there is a vigorous tribal leadership centered outside the towns in the merhûm and focused on the tent-dwelling (Hana) component. It is possible that˘before the Simalites were unified under˘ the family of Yaggid-Lim, Yahdun-Lim, and Zimri-Lim, the gayums had carried active ˘ as constituents of a Simalite tribal confederacy. Even political significance the office of merhûm is difficult to evaluate in historical terms. In the form ˘ found under Samsi-Addu and Zimri-Lim, the position may have been detached from origins in a group-oriented political context and coopted for service to the great regional kings. e. Authority outside the Bounds of Zimri-Lim’s Kingdom. One of the features of political life in the Amorrite period was that tribal groups were not often confined within the proper boundaries of kingdoms, in terms of either their settled communities or their pastoral range. This is most striking in the text quoted earlier that defines the traditional grazing land of the (Simalite) Hana as Ida-Maras., in the upper Habur, and that of the Yaminites as the ˘ countryside of three western states:˘ Yamhad, Qatna, and Amurrum. Large ˘ populations from both tribal confederacies inhabited the Ah Purattim, ˘ the core Euphrates districts around Mari.203 Given this definition of Simalite grazing land as focused in the Habur ˘ drainage, it is no wonder that Zimri-Lim’s kingdom maintained such strong northern connections, including many vassal polities that were not directly incorporated into the Mari state. Following the movement of his pastoralists, the Simalite merhûm exercised some degree of authority in these northern ˘ Mari kingdom, at least in matters of grazing rights. Once regions beyond the in these outside domains, the power of the merhûm was not clear-cut. Ibal-piel is said to allocate pasture land north of Mari˘ between his Simalite Hana and the shepherds of Qarni-Lim, king of Andarig and the Yamutbal ˘tribe (XXVII 48). The same king complains to the district governor of Qat.t.unan about the theft of sheep by one of the Simalite Hana, and the governor ˘ has the man traced. When it is found that he has moved to the kingdom of Kurdâ, however, the matter is passed on to Ibal-pi-el: Then Ibassir went and seized six sheep from one fold, two sheep from a second fold, and a male goat from a third fold. At this moment, Ibal-pi-el the merhûm is in the company of my lord. He should examine the men who have charge of the˘ sheepfolds, and the men who have charge of the sheepfolds can determine who is the one that had these sheep taken away. These men do not operate under my authority.204

The merhûm has authority over the wider steppe, though he may have to negotiate˘ any action with the local authorities in Kurdâ. New insight on this northern overlap of Mari or Simalite authority comes from a letter cited by Villard in his discussion of the early merhûm named ˘ Bannum (1994, 297 and n33). Bannum writes to Zimri-Lim in order to dissuade him from heeding an attempt by rivals to strip Bannum of the merhûm ˘

82


function. These rivals consider him unable to handle the responsibilities of being merhûm now that he is based at Mari. The merhûm, who is serving ˘ second-in-command, offers the king his ˘own idea of how to as Zimri-Lim’s answer: And if the Hana press you to appoint another merhûm, saying, “Now that Bannum, ˘ our merhûm,˘ lives in the Ah Purattim, we should appoint another merhûm”; you – ˘ ˘ ˘ answer them this way: “Previously, he lived in the steppe, and he maintained the status of the Simalite(s), the Numhâ, and the Yamutbal. Then, he left for the Ah Purattim, where he forced open the˘ fortified towns and so has secured your status˘ in the Ah Purattim. Now, because I myself have come here, I have left this man in ˘ the Ah Purattim in order to hold the fortified towns. Now, as soon as I reach (Mari), ˘ I will send you back your merhûm.” Answer them this way!205

˘ The merhûm considers his sojourn in the Euphrates valley (Ah Purattim) to ˘ ˘ rank and its be temporary, and his primary identity to lie in both his original northern location. Unfortunately for his plans, Bannum died shortly after Zimri-Lim’s capture of the throne.206 The continuing northern authority of a Simalite merhûm later in ZimriLim’s reign is demonstrated directly in ARM XXVII 48 and˘ 70, where Ibal-piel is assumed to have influence that overlaps into the kingdoms of Andarig and Kurdâ, respectively. Even here, Qarni-Lim of Andarig reveals a growing tension between the partners, with a fear that the merhûm will favor his own “Hana,” defined in more restrictive Simalite terms. ˘ ˘ The ongoing competition between the Simalites, on one hand, and the Yamutbal and the Numhâ, on the other, would then be displayed in the ˘ fact that Atamrum of Andarig and Hammurabi of Kurdâ end up being ˘ key players in Zimri-Lim’s ultimate defeat. The complaint by Numhâ tribal ˘ “son” leaders that Kurdâ’s king Simah-ilanê should be “brother” and not ˘ 207 Zimri-Lim’s increasing to Zimri-Lim demands treatment as tribal peers. desire to be recognized as the superior party in these old alliances is visible in the addresses of letters sent to him from the kings of Kurdâ, the realm identified with the Numhâ tribe. Simah-ilanê was the first of these, and he ˘ as “brother”˘ (ARM XXVIII 162:4), as FM II 117 did address the Mari king leads us to expect. Qarni-Lim, the king of Andarig early in Zimri-Lim’s reign, called himself “brother,” like Simah-ilanê.208 The letters from Bunu-Eˇstar, ˘ an uncertain picture, with examples the successor of Simah-ilanê, present ˘ addressed to Zimri-Lim as from a “brother” and as from a “servant,” and by his name alone.209 Finally, Hammurabi, the last king of Kurdâ during the period of the Mari archives,˘ found it necessary to call himself Zimri-Lim’s “son,” acknowledging a lower rank. Perhaps Hammurabi never accepted ˘ rejected the alliance with this fate. He was, after all, the ruler who eventually Zimri-Lim and the Mari kingdom (see Durand 1992a, 45–6). Ibal-Addu of Aˇslakkâ, possibly the most powerful of the Ida-Maras. kings, describes at length a series of events related to the confirmation of


83

ˇ Sadum-labua as the new king of Aˇsnakkum, another member of the IdaMaras. confederacy (ARM XXVIII 65). Throughout the episode, one of Zimri-Lim’s merhûms acts as the Mari king’s agent, with the local rulers fol˘ ˇ lowing his instructions. In another letter, Sadum-labua himself mentions the role of “the merhûm” in keeping him safe during problems with his ˘ districts of the Mari kingdom, this overlap of regional neighbors.210 As in the authority based on the pastoralists’ extensive movements provokes tension ˇ with the local powers. Sadum-labua of Aˇsnakkum remarks to Zimri-Lim, “I do not trust the merhûm at all” (XXVII 105:36 ). ˘ authority of the Simalite merhûms extended into two It appears that the ˘ kingdoms of Andarig northern domains, one in the grazing land near the and Kurdâ, between the Habur and the Tigris Rivers, and the other in ˘ Habur basin. In each case, this authority Ida-Maras. of the northeastern was based on a long-standing Sim˘ alite presence that predated Zimri-Lim’s arrival at Mari. In spite of the tribal competition, or perhaps in some cases because of it, the Simalite merhûm could be drawn into the politics of the Kurdâ throne. In a letter cited˘ earlier, Bunu-Eˇstar pleads for one of the merhûms to help him return to his throne over Kurdâ: “Fais pression sur ˘ um et les grands dignitaires qu’ils me fassent revenir sur le trône de le mer’ˆ mon père!”211 The other region of the merhûms’ authority is the countryside of Ida-Maras., where the Simalites have a˘ long-standing claim to pasture rights. This authority is displayed by the unusual role played by the merhûm Ibal-el in overseeing an alliance with the combined m¯atum of Ida-Maras. ˘in a letter to be discussed in Chapter 4 (A.2226). Charpin has argued that the responsibility of concluding an alliance by slaughtering a donkey is normally reserved for kings (Charpin 1990c).

2. Outside the Kingdom of Zimri-Lim There is very little evidence for the merhûm either during the reign of Samsi˘ Addu or outside the Simalites of Zimri-Lim: Durand observes in a Yaminite letter a reference to a full gathering that will include all their sug¯ag¯u and “their two merhûms,” a reading based on new collation (A.2741; LAPO 16 no. 433:10, p.˘ 630). With only two Simalite merhûms attested at any one ˘ both major Amorrite time, this text suggests to Durand a pattern across divisions. To my knowledge, no text directly counts “two” Simalite merhûms, ˘ and the question of numbers, as well as the political relationship between contemporary merhûms, remains poorly understood. I am aware of no text ˘ Ibal-pi-el, the two Simalite merhûms from the last part that shows Ibal-el and ˘ text may mention a of Zimri-Lim’s reign, working in concert. One other Yaminite merhûm, when it refers to a merhûm at Imar, in service to Dadi-hadun. ˘ to Zimri-Lim: “Dadi-hadnu ˘ came The letter was˘ sent by the diviner Iˇshi-Addu ˘ ˘ to me and he has charged his merhûm with this expedition. I took o[mens], I touched the forehead of the mer˘hûm, and my omens were auspicious.”212 ˘

84


Under Samsi-Addu, one letter from the diviner Asqudum to Yasmah-Addu gives the impression of a larger number of merhûms with lower rank.˘ Some ˘ group of these is supposed to meet with Yasmah-Addu at Tuttul: ˘ The merhûms are at Tuttul in the company of my lord. My lord should give them ˘ that they not be negligent about their (military) patrols. Their sheep strict orders are scattered. They are grazing from upstream of Halabit as far as Surman, and they ˘ are (in fact) being negligent.213

The Yaminite examples could also imply a tradition of merhûms of both lower status and larger numbers. In another reference to Tuttul,˘ Samsi-Addu proposes to let Habduma-Dagan hold the office of chief of pasture (merhûm) ˘ ˘ and local governor (ˇsa¯ pit.um) concurrently, an inconceivable combination within Zimri-Lim’s division between Simalite Hana and the core districts of Mari, Terqa, Saggaratum, and Qat.t.unan:214 ˘ You wrote to me about appointing Hab[duma-Dagan son of] Ayala-sumu to serve as merhûm. They should proceed just as˘ you [wrote(?)], to appoint Habduma-Dagan to ˘ as merhûm. What does his governorship amount to? It is not ˘ as if he (already) serve had charge ˘of a vast land! He should have charge of Tuttul as well as serving as merhûm. Then just as his counterparts have charge of a vast land, he also should have ˘ of Tuttul.215 charge

If the first post is meant to be based in the steppe, the distant sovereign has clearly hijacked it for his system of rewards and imperial control. It is clear even from this limited evidence that merhûm status belonged to the wide range of Amorrite pastoralists and was not ˘a purely Simalite phenomenon. The merhûm’s geographical and political domain lay outside the ˘ town, and his relationship to the sug¯agums would therefore vary in the Yaminite and Simalite social structures, with their distinct adaptations to the a¯ lum “town.” We find references to only “Yaminite” towns, never Simalite, and we know that a Yaminite town may have an individual sug¯agum leader who is identified as “the sug¯agum” of that town. Nothing of the sort is mentioned for the Simalites. We cannot yet know what historical change may have contributed to the range of evidence for the merhûm found at Mari. The smaller scope of activity ˘ this is true, may reflect smaller tribal groups among the Yaminities, if indeed or a different distribution of responsibilities. Judging by Bannum’s conquest of Mari, the Simalite merhûms seemed to have played a powerful military role ˘ even without the sponsorship of Zimri-Lim. We have no evidence for any comparable prominence among the Yaminites. It may have been in ZimriLim’s interests to maintain the considerable power of the Simalite merhûms ˘ of in order to protect his pastoralist power base, or conversely, the power the Simalite merhûms may have forced Zimri-Lim to find them an unusual ˘ freedom in his royal administration. In either case, the tribal kingdom of

The “Hana” Tent-Dwellers ˘

85

Zimri-Lim cannot be understood without taking full account of the merhûms’ ˘ importance as leaders of the king’s Simalite tribespeople.

e. the “ hana” tent-dwellers

˘ At the beginning of this chapter, I introduced the basic argument that the ubiquitous hana in the Mari archives are not first of all a “Hanean” tribe, ˘ mobile pastoralists generally. It was not possible to explore but rather are the alternative tribal structures of the Simalites and the Yaminites without defining this term. Although the hana category was not tribal in itself, mobile ˘ consistently in tribal terms that allowed pastoralists were in fact identified bonds to be maintained over distance. The hana pastoralist element of both ˘ an essential part of how each the Yaminites and the Simalites was therefore tribal group described its own and other tribal populations. In the Mari archives, the hana terminology is crucial to understanding the Simalites in ˘ particular, because the overwhelming majority of the texts come from the government of the Simalite king Zimri-Lim. We cannot therefore come to terms with the political situation of Mari under Zimri-Lim without further attention to his Simalite Hana. ˘ The most important social distinction that was significant even for the everyday affairs of Zimri-Lim’s kingdom was between the mobile part of the community that moved through the steppe and the remaining sedentary portion. Simalite tribal categories were identified especially with the mobile component, who were most often called just “Hana,” and any under˘ must deal with this standing of Simalite social structure, sedentary and not, usage. I argue that Zimri-Lim’s circle of kinsmen and officials use “Hana” ˘ consistently to refer to mobile Simalite pastoralists in their various relations to the kingdom, and apply the word to tribesmen outside Simal in specific situations only where it is necessary to distinguish people of the steppe who are outside this family. Our examples of hana terminology that do not refer ˘ to Simalites should all be capable of explanation by the above principle or should come from outside this tribal circle of Zimri-Lim.

1. The Simalite Hana as Army and as Political Unit ˘ Besides the evidence of the titulary itself, the Hana appear often in Zimri˘ understood as specifically Lim’s circle as a political unit that is most easily Simalite, not just vaguely “tent-dwelling.” Also, Zimri-Lim is served by a Hana army that may be better read as Simalite than more widely inclusive. ˘ evidence shows how common is the narrow reading of the Hana duality This ˘ in this period and administration. a. The Hana Army. It has long been observed that Zimri-Lim makes con˘ stant reference to Hana in his military service, sometimes as a distinct ˘

86


battalion, other times performing specific duties in smaller groups.216 This military force needs to be explained in light of the larger problem of Hana ˘ terminology. Durand appears to acknowledge the identification of this force with Simalites, but a brief discussion is perhaps worthwhile (LAPO 17, p. 465). The treaty offered by Ibal-pi-el II of Eˇsnunna presents a useful overview of Mari military divisions, because it includes a clause whereby Zimri-Lim promises to report all movements of his troops. It lists the following units, each rendered as .sa¯ b “X,” “the army of (X),” where “X” includes: Mari, the Hana (lu´ Ha-nameˇs) , Suhˆ um, kin[g and gener]al (LUGA[L u` ra-bi]-im), and ˘and Akkadian. ˘ ˘ 217 The fact that the first two belong to Mari and the Amorrite Hana provides one indication that the titulary reflects entities with a real ˘ administrative aspect, armies in this case. This primary division should be equivalent to the similar one offered by Zimri-Lim in response to Atamrum’s request for troops: The Mari king has at the core of his force 100 Hana and ˘ heads 100 “of my servants of the Ah Purattim,” who together represent “the of my land.”218 Mari and the˘ Ah Purattim describe the administrative entity ˘ dominated by the ˇsa¯ pit.um governor system and the core districts of Mari, Terqa, and Saggaratum, if not also Qat.t.unan. Again, the pairing with Hana ˘ suggests that the latter should be part of Zimri-Lim’s larger government, most naturally the branch led by merhûms. This association also suits the ˘ over Hana troops, as seen in the fact that the merhûms served as generals ˘ ˘ letter recounting Ibal-pi-el’s presentation of his men at the court of Babylon, discussed with the gayum.219 The military report FM III 135 confirms the limited identification of this Hana army as Simalite. An officer named Yatarum sends word that ˘ a campaign in the north has succeeded, based on the efforts of an army likewise divided into two units. The letter opens and closes with a chiastic affirmation of the well-being of the king’s troops, which I quoted earlier without interest in the specific military aspect: line 4: The Hana are well. The armies of my lord are well. ˘ line 5 : The armies are well. The Simalite(s) are well.

The “armies of my lord” (umm¯an¯at b¯el¯ıya) represent an all-inclusive category whose actual makeup is not made clear, but the direct service to the king recalls the Ah Purattim category in A.2730, a letter that refers to the heads ˘ land (see Chapter 4). It is impossible to miss the equation of Zimri-Lim’s of Hana with Simal in the other element, but the implications of the pair ˘ not be restricted to this event and troop, as if a local phenomenon.220 should Zimri-Lim’s Hana troops occasionally show a specific Simalite character ˘ First, the Hana division of the army is organized by pirsum in other contexts. and gayum, the latter term ˘being specific to the Simalites and not part of Yaminite social structure.221 Second, Ibal-el and Ibal-pi-el, the two bestknown Simalite merhûms, lead Hana contingents into battle.222 Ibal-pi-el ˘ ˘


87

also echoes the equation of FM III 135 when he reports that the Simal and other troops are well: DUMU Si-im-a-al u` um-ma-na-tum ˇsa-al-ma.223 b. The Hana as a Political Unity. The Hana may serve as an indispensible ˘ of Zimri-Lim’s army, but they˘ are certainly more than a military loyal unit force. Under the general name “tent-dweller” lies a people of definable bounds, who gather for various purposes, who receive communications as a single decision-making entity, and even for whom omens may be taken.224 The Hana may even be regarded in some circumstances as a separate tribe, ˘ especially when mentioned alongside the other main tribes of the regions directly north of Mari, the Numhâ and the Yamutbal.225 As already observed, the Hana˘ are frequently reported to gather at their ˘ specific decisions, most often having to do own initiative in order to make with war or alliance. In these meetings, we see a people with remarkable initiative under Zimri-Lim’s rule, whom he treats almost as independent allies, rather than as subjects. In the letter from Ibal-el that recounts peace negotiations with Ida-Maras., to be discussed at length with “elders” in Chapter 4, the merhûm tells about an alliance being negotiated between the gathered leaders ˘of the Habur m¯atum and the Hana, in response to aggression by the ˘ of Eˇsnunna.226 Ibal-el ˘ signs off with an affirmation that eastern kingdom the Simalites are well (line 24 ). The Hana sometimes battle their rivals without evident approval by the king, as ˘with the Yaminites of the steppe in one letter from a governor of Qat.t.unan: A contingent of 200 Hana from the area of Qat.t.unan [assembled(?)], and (with) Yatar-Lim in their lead,˘ went to wage war on the steppe (people) of the Yaminites. I spoke to them as follows. . . . 227

In some cases, the Simalite tent-dwellers could flaunt their special status with an arrogance that must have made life more difficult for the king. One governor of Qat.t.unan reports to Zimri-Lim that he has had to imprison a Hana man who killed a slave in a drunken fit. The arrested man then scoffs ˘ the governor’s attempt to interrogate him, while waving a sword in the at Mari official’s face: Yasim-Addu the Hana took bread and beer by force and mortally wounded one of the palace slaves. ˘I (said to him), “Why did you kill the palace slave?,” and this man answered me as follows: “The king is your master. If Sammetar cannot interrogate me, do you think you can interrogate me?” Then he started waving his bronze weapon in my face in the presence of Ripi-Addu and the servants of my lord. Now, I have put this man in prison. My lord should know this.228

It would be surprising but not impossible to find that some elements of the Simalite Hana might even oppose Zimri-Lim in favor of other local leaders. ˘ the general Yassi-Dagan describes opposition to a Zimri-Lim A letter from force in the eastern Habur basin from a king of Kahat who brings with him ˘ ˘

88


a battalion of 500 Hana, and it is not clear whether these sympathetic tribes˘ Simalite or some other northern group. “According men would have been to the command that my lord gave me, I reached Kabkab, and Akin-amar heard (about it), and 500 Hana forced their way out as his vanguard.”229 We find other evidence ˘that the Hana terminology may sometimes as˘ sume a political entity. A number of texts record contacts made with them as a single decision-making body. Zimri-Lim sends to the Hana to ask for ˘ camps(?), I muster of their army: “Before the arrival of the Hana in fortified ˘ had deliberated and sent Ibal-el the Hana to the Hana and Hali-hadun.”230 ˘ ˘ chase ˘ Elsewhere, the king sends to ask for˘ a rihs.um assembly: “Go down ˘ (kaˇsa¯ dum) the Hana and engage in talks (rihs.um), so that the Hana may ˘ the raising(?) of the torch, ˘ all the Hana may˘ come as assemble and [at] ˘ king Ibal-Addu 231 one man to my assistance.” In the royal correspondence, of Aˇslakkâ tells Zimri-Lim that he should write to the Hana (ARM XXVIII ˘ 55:13 ). Other times, messages and leaders come from the Hana. Hana heralds ˘ “The ˘ bearers of spend the night with Asqudum on their way to Zimri-Lim: good news who reached me from the company of the Hana spent the night ˘ reported to have in my company.”232 In one case, four Ida-Maras. kings are met with “the Hana” (XXVII 20:19–21), and elsewhere, “the Hana” bring ˘ is found news of coming˘ war with the Yaminites.233 The same phenomenon at least once in the earlier period, where after a victory, Iˇsme-Dagan sends a message simply to the “Hana” (ARM IV 40:5–9). It is not clear how this address might overlap with˘ the Simalites of Zimri-Lim’s reign, but it should also carry some specific point of reference. One last aspect of the Simalite tent-dwellers’ independence under ZimriLim is their apparent freedom from census. It is clear from Samsi-Addu’s instructions regarding the Yaminites in ARM I 6 that tribal peoples could be expected to provide troops without formal registration. During the reign of Zimri-Lim, the king’s pleas for a Hana muster (e.g., ARM XXVI 27; A.3567, ˘ the system of census and oath. When above) do not seem to depend on the Hana do gather to fight for Zimri-Lim, it is by agreed appointment ˘ (hadannum): ˘ I have been waiting for the Hana for five days, at the arranged time, and (at this rate,) the army will never be ˘assembled. (Some) Hana have (in fact) arrived from 234 ˘ the back-country, and they are living inside the villages.

Our frequent references to the census of Hana peoples come exclusively ˘ have preferred a policy the from the reign of Samsi-Addu, who may in fact reverse of Zimri-Lim’s, allowing the Yaminites to be exempt.235 All of these texts come from regions that might plausibly be dominated by Simalites rather than Yaminites, where they were perhaps treated as one with some Numhâ or Yamutbal kin. ARM V 51 is of particular interest, because it ˘ the census of Hana who are not living in the steppe but who rather records ˘


89

reside (waˇsa¯ bum) in the district (hals.um) of upper Ida-Maras., in Nahur, ˘ a region with strong Simalite ties˘ to Talhayˆ um, Qirdahat, and Aˇsnakkum, ˘ ˘ Zimri-Lim during the succeeding period. You wrote to the king (Samsi-Addu) about the Hana who have made their home in ˘ drawn up a tablet naming every the district of upper Ida-Maras.. Now then, I have Hana resident in the district of Nahur, Talhayûm, Qirdahat, and Aˇsnakkum, (so) ˘ ˘ the census,˘ and I have sent it to [my belonging to the district for which I ˘have taken lord].236

The reference to “settled” Hana would be an oxymoron according to the ˘ common use of the word during the reign of Zimri-Lim, and this gives the impression that the term has been attached to a particular tribal group. In this case, the name could be self-given or assigned as a generality by Samsi-Addu and his circle. I have not found the verb ubbubum (“to take census”) or noun t¯ebibtum (“census”) to be used with the Hana of Zimri-Lim’s reign. The military lists of Simalite soldiers published by˘Birot register them for receipt of rations, not for service itself, and they were not evidently drawn up as the basis for a draft (1953, 1955, 1956).237 They are classified by division (gayum), not by town. In contrast, the Yaminites under Zimri-Lim are promised by precise number assigned town by town in ARM XXIII 428–9. These texts deal systematically with the Yaminite population of Mari’s three central districts, with all five tribes represented, and all defined by town residence.

2. Non-Simalite Hana in Zimri-Lim Period Texts ˘ Many of the most intriguing uses of the word hana as “tent-dweller” come ˘ not or cannot assume the from situations and settings where the writers do common Simalite identity. In general, “hana” used alone tends to mean ˘ Samsi-Addu period are more “our tent-dwellers,” though texts from the likely to use the name “Hana” to refer to a specific people (settled or not) in ˘ the upper Habur. According to the more common pattern, which is found ˘ especially in texts from the reign of Zimri-Lim, we do not usually find the double identification with both tribal name and the word “tent-dweller” when the at-home setting can be taken for granted by writer and recipient. Our limited Yaminite correspondence mirrors in part the Simalite use of hana alone for “our tent-dwellers,” as shown by the sug¯ag¯u of the hana among ˘ assembled Yaminites in ARM II 53.238 Likewise, the Simalite ˘ hana are the ˘ king called such by outsiders such as the Yamutbal state of Andarig or the Samsi-Addu. From Andarig: And since last year, since the Akkadians (Eˇsnunna?) came up (to attack), the Simalite hana have been grazing (their flocks) in the midst of the land of the Yamutbal. No ˘ offense or breach of conduct has ever arisen.

90


The lance of the Yamutbal and the lance of the Simalite hana have met. They ˘ of) you.239 are joined together, and no barrier can come up before (the two We should establish good relations [between us(?)]. [The Yamutbal(?)] and the Simalite(s) have shared brotherhood and branches of the hana (people) since days ˘ gone by. . . . 240

From the kingdom of Samsi-Addu: . . . a troop of 1000 among our Simalite hana. . . . 241

˘

[Regarding the] hana who have crossed to the opposite bank (of the river), about whom my lord wrote˘ to me as follows: “The hana who crossed (the river) – Simalite ˘ (or) Yaminite, which is their division (gayum)?” Send me a complete report about 242 them.

The perspective of Samsi-Addu’s kingdom does not in general reflect the tribes themselves.243 One explanation for the Samsi-Addu period texts that speak of Hana as a particular settled people may be that the tribal population in the˘ imˇ mediate area of Subat-Enlil, the royal capital, called itself Hana. The best evidence comes from the period of Zimri-Lim, in a letter to˘ the Mari king ˇ from Subram, the king of Susâ in Ida-Maras. (ARM XXVIII 95). According to ˇ Subram, a dispute has arisen over possession of a town in the upper Habur ˘ anˇ a and between one group of local people affiliated with the land of Sunˆ other group from the land of Apum. During the reign of Zimri-Lim, after ˇ Samsi-Addu had died and Subat-Enlil returned to being just another city of modest influence in the upper Habur region, Apum was the name of the ˇ m¯atum that had Subat-Enlil at its ˘center.244 ˇ a and Apum are made to undergo the Four representatives from both Sunˆ river ordeal, and the statement sworn by each group is quoted in the letter. ˇ a are required to say, “This town is indeed my town, The people from Sunˆ and from of old, it was duly assigned to the inheritance of the Yabasu. The Apumite did not assign it as a gift” (lines 24–6). The four from Apum swear ˇ similarly: “This town indeed belongs to Subram, and from of old it was duly assigned to the inheritance of the Hana” (lines 29–30).245 As discussed with ˘ the gayum “division,” Yabasu is the name of both a standard gayum and one of ˇ the two overarching military divisions of the Simalite confederacy. Subram is a ruler of variable fortune, who at this time had authority over the land of 246 ˇ Apum, which included Subat-Enlil. Notice first of all that the question of possession is framed in terms of the disputed town’s ancient tribal affiliation, not in terms of the m¯atum to which it belongs, though of course the political realms are in fact in play. The claim ˇ a is based on its own tribal identity as Simalite Yabasu, though not diof Sunˆ rectly incorporated in Zimri-Lim’s kingdom under his merhûms. What affects the issue at hand is the second tribal name, which is given˘as “Hana,” in this ˘ case, just the sort of proper noun that used to be translated as “Hanean.” The


91

ˇ people of Apum (m¯at Apim), which had Subat-Enlil at its center, belonged to what must be considered a local tribe that had come to identify itself as Hana, making the larger descriptive adjective its proper name. The same ˘ appears later in the letter, where Subram ˇ use is quoted as saying, “These men are Haneans (sons of Hana), not sons (i.e., inhabitants) of his land” ˘ ˇ (lines 60–1).247 “His land” refers to the m¯atum of Zimri-Lim, and Subram considers these “Haneans” not to belong to the Mari king’s m¯at Hana. ˘ king of We must remember that none of the voices in this letter from the Aˇsnakkum belongs to Zimri-Lim or anyone from his circle. Even where the local population of the Apum region knows a tribal group named “Haneans,” Zimri-Lim and the Simalites do not use such a term. When they speak of the people of Apum, they use the name of the m¯atum and avoid the “Hanean” label.248 There are a few rare letters where officials of Zimri-Lim call someone hana who is not a Simalite, and these apply the word in its ˘ general descriptive sense, not as the proper name of a tribal people. These texts tend to come from early in the reign and deal with special contacts with mobile Yaminite pastoralists. In one case before the Yaminite war, the earliest example, Zimri-Lim is sent a report on the status of negotiations with ˇ the rival confederacy to join the Simalites in an attack on Subat-Enlil (FM II 116). Throughout the letter, Sumu-hadˆ u, the writer, speaks of both Yaminite and Simalite hana from a neutral ˘perspective, identifying both groups by ˘ and even referring to a letter that “my lord wrote to the the double name Simalite hana” (line 38): ˘ Regarding the plan to attack the countryside (about) which my lord wrote to the Simalite hana, the hana must (first) arrive, so that this plan may. . . . The Simalite ˘ ˘ assembled there (near Mari), and the Yaminite hana should hana will certainly be ˘assemble here (near Saggaratum). Then the Yaminite regular army from ˘ the banks of the Euphrates, as many as I can get ready to go, I will cause to obey(?) my lord.249

It appears that with a joint operation in view, emphasizing the strictly tribal pastoralist composition of the attacking force, the complicated report of communications with the different parties provoked Sumu-hadû to clarify ˘ “Simalite their identity in this unusual style.250 It is also natural to distinguish hana” from “Yaminite hana” when both specific groups are in view. ˘ Two other texts come ˘ from the period of the Yaminite war, not long after Zimri-Lim’s accession to the Mari throne. Cooperative endeavors of the sort proposed in FM II 116 were no longer imaginable. In one letter, Habduma˘ of “the Dagan responds to earlier orders from Zimri-Lim to attack the flocks Yaminite hana” and to drive them across the Balih, away from territory dom˘ the bedouin population inated by ˘the Simalite king.251 The focus falls on specifically, which is certainly not treated as an accepted part of the landscape. Another letter likewise addresses the situation in the western part of the kingdom after the defeat of the Yaminites, marked here specifically by a victory over Miˇslan. The writer, whose name is lost, reminds Zimri-Lim

92


that he had suggested a brutal communication of the result by sending two “hana” to the Yaminites at the western frontier, where the men would be ˘ mutilated: Ever since my lord captured the town of Miˇslan, that is entirely enough for me. Another thing. Previously, I wrote to my lord as follows: “Let them escort their two surviving (Yaminite) hana to the border, and then at the border, mutilate them. Let ˘ Yaminites in order to proclaim that my lord has captured their survivors go to the the town of Miˇslan by force.”252

The text separates the hana from the “Yaminites,” but they are linked in ˘ intent, with an effect similar to that of Habduma-Dagan’s letter. Beyond ˘ these, I have seen no references by Zimri-Lim or his officials to Yaminite hana, and the standard Simalite use of the word dominates these circles ˘ 253 completely. In sum, king Zimri-Lim stood in a delicate and complex relationship with the Simalite mobile pastoralists who formed the backbone of his power. He was certainly their king, and as a fighting force, the Simalite Hana provided a formidable core to his armies. At the same time, however, ˘the active support of his pastoralist tribal kin was not something simply to be commanded. It had to be wooed as a collective body with its own will, and this distinct identity led to frequent reference to “the Hana” as a separate polit˘ ical entity, a group with the capacity to decide whether and how to follow its king. Within the circle of Zimri-Lim’s core leadership, which naturally took on a Simalite perspective, regardless of the origins of each member, “Hana” terminology came to designate Simalite pastoralists almost exclu˘ sively. Outside this circle, we must expect other patterns of use. All of this evidence provides a necessary backdrop to the “land (m¯atum) of the Hana” ˘ we ruled by Yahdun-Lim and Zimri-Lim according to their titulary, and ˘ return to the Hana one more time in Chapter 3 in connection with the ˘ m¯atum.

f. the other confederacy: the yaminites Zimri-Lim of Mari was the king of the Binu Simal, and a huge number of the Mari letters reflect a Simalite perspective at some level. We encounter the actual Binu Simal name so rarely because it was almost always simply taken for granted. To explore the tribal aspect of the Mari kingdom under Zimri-Lim, its administration and its relationships with the outside world, is to enter above all the world of the Simalite people. I have allowed the majority of this chapter to be defined around Simalite tribal terminology and social structure because these dominate the actual evidence. Often, the insider’s view is Simalite. In spite of the Simalite perspective of Zimri-Lim’s kingdom, however, the success of the Yaggid-Lim dynasty was built in part on conquests over

The Other Confederacy: The Yaminites

93

their Yaminite rivals. Numerous Binu Yamina settlements were incorporated into Zimri-Lim’s realm, especially in the Terqa and Saggaratum districts, upstream from Mari.254 Before Zimri-Lim, Yasmah-Addu also had to deal with ˘ a large Yaminite population in the Euphrates valley. The old towns of Tuttul and Imar, even farther upstream, were strongly influenced by Yaminite populations. Because so many Yaminites were ruled by the different Mari governments, we find references to them scattered liberally throughout the Mari letters. In fact, the Binu Yamina name appears much more often than that of the Binu Simal, because it is so often invoked to identify the tribal group that is other, by reference to the Simalites. Our evidence for the Yaminites in the Mari archives is much more limited than for the Simalites, but it is still very rich, and the group was equally important. As I have stated more than once already, one essential part of my analysis is the proposal that the Yaminite and the Simalite tribal peoples were characterized by different social structures and political traditions. Where the Simalites are defined in pastoralist terms with gayum divisions that follow mobile Hana herdsmen, the Yaminites are divided into five ˘ the li mums of their kings. Whether as the li mum of tribes that can be called a ruler with a fortified settlement as his seat of power or by the frequent references to Yaminite towns whose individual leader may be called a sug¯agum, the Binu Yamina seem to have allowed the town a greater defining role in their social categories. We can expect other differences between the Simalites and the Yaminites to follow these, and there were surely more than are visible in the Mari archives or that I have been able to discover. The purpose of this section is to examine the distinct Yaminite traits in the Mari evidence, especially where they find expression in particular terminology. One extremely important letter (A.981) reports the acceptance of the Yaminite town of Dabiˇs into Zimri-Lim’s kingdom under terms that ally its Yaminite population with one of the Simalite gayums. As part of a declaration of readiness to submit to Zimri-Lim’s terms, the leaders of Dabiˇs define their pastoralist associates by the words hibrum and kadûm, which therefore must be significant categories ˘ in the Yaminite social system. The Simalite and Yaminite confederacies of the Amorrite tribal peoples shared in broad terms the same framework of pastoralist existence. Both had highly integrated town-dwelling and mobile herdsman components of their own, and both encroached on the larger regional towns and took them over where possible, while still prizing their own tribal identities. We have seen that there were broad regional patterns in the distribution of Yaminite and Simalite populations (Durand forthcoming). The Simalites considered their primary grazing lands to lie in the Habur River basin, especially ˘ to the south and west, all in Ida-Maras., with the Yaminites oriented more the way to the Mediterranean and southern Syria.255 While this distribution is surely based on actual patterns, we must remember that it follows a

94


Simalite perspective and may be self-serving. In particular, the focus on the Simalite presence in Ida-Maras., which seems a fairly limited geographical claim, allows the writer to emphasize the legitimacy of Zimri-Lim’s ambitions to control this populous and strategic region. At the same time, the description grants the Yaminites their more distant grazing range while entirely ignoring the Yaminite population in the great arc created by the Euphrates as it drops down out of Anatolia and then loops to the southeast across northern Syria. Yaminites were scattered along the Euphrates from Imar to Mari, and then further north, the Zalmaqum coalition of the upper Balih River basin was firmly allied to the Binu Yamina. The steppe that filled the˘ gaps between these valley regions would have represented the natural grazing land of Yaminite herdsmen, but the Simalites do not appear to have been ready to acknowledge the areas that lay closer to territory claimed for the kingdom of Zimri-Lim. It is possible that the structural differences between the Binu Yamina and the Binu Simal derived in part from these geographical trends. We cannot yet trace the antiquity of the Yaminite/Simalite, right-hand/left-hand duality in the tribal definitions, and we cannot tell what older tribal traditions and identities may lie behind this conceptualization of a wide-ranging pair. At least in the area of the middle Euphrates, the Yaminites or their ancestors would have had occasion for participation in the lives of the settlements that already dotted the valley in the third millennium, from the vicinity of Terqa (Tell Ashara) and Mari (Tell Hariri) upstream toward Tuttul (Tell Bia) and Imar (Meskeneh), possibly even as far as Tell Banat. Perhaps this kind of long participation provides the background for the Yaminites’ comfortable integration of tribal and town structures, if the Simalites had not developed this sort of close connection. The historical question cannot be answered from the available evidence, but the explanation for the observed contrasts may have a geographical component. One wonders whether this greater role for the town in Yaminite political ideology is related to the fact that there are decidedly more Akkadian personal names among the Yaminites than among the Simalites.256 If the Yaminites of the Ah Purattim identified more strongly with the settled social ˘ heritage of the Euphrates valley, which itself had a long-standing association with the pastoralism of the steppe, this could have led to a higher degree of contact with the Akkadian language than would have been found among most Simalites. Although both Yaminites and Simalites occupy settlements in the region of Mari, we must not underestimate the regional pattern, with the Simalite Hana more deeply entrenched in the upper Habur.257 If the ˘ then the dominant tribal population of the ˘ Ah Purattim, Yaminites were ˘ that was and they accepted to a greater degree the cultural paraphernalia tied up with the Akkadian language, this would add another potential basis for their sympathy with the rule of Samsi-Addu and Yasmah-Addu. In ˘ that the one letter, Samsi-Addu upbraids Yasmah-Addu for even imagining ˘


95

Yaminites should be included in a regional census. The Yaminites must not be offended: You wrote to me about taking a census of the Yaminites. It is not a good idea to take a census of the Yaminites. If you take their census, their kin, the Rabbˆ u who live across the river in the land of Yamhad, will hear (about it) and become provoked at them, so that they cannot return to˘ their land. You must not take their census at all.258

Instead, Yasmah-Addu should simply inform the Yaminites that he is going ˘ he should remind them that they are bound by oath to on campaign, and provide troops: “Give them this sort of order; by no means take their census at all.”259

1. Yaminite Towns and Steppe There are various signs that Yaminite tribal ideology incorporated more fully both aspects of the hybrid reality of life in both towns and steppe. While we might expect Zimri-Lim’s system of district administration to have imposed on them a false structure that did not fit their own custom, this conclusion is not obvious. We cannot explain the situation by opposing desert and sown, imagining that the state tried to control the nomads by forcing them into settlements. The governors assembled the Yaminites by their town sug¯ag¯u (e.g., ARM II 92), and we have evidence that this tribal confederacy itself met by gatherings of sug¯ag¯u, where even those called “tentdwellers” wanted back their towns (ARM II 53). Kibri-Dagan, whose Terqa district made him a Yaminite specialist of sorts, reported that his whole Yaminite population had gone from their towns upstream (el¯enum) as a single “contingent” (s.a¯ bum).260 A Yaminite letter, however, also divides the whole people by “the .sa¯ bum of the towns” and “the hibrum of the steppe,” all ˘ of whom are “brothers”: Also, the harvest has (just) arrived. My lord must not fail to write to his brothers, so that both their contingent of the towns and the backcountry hibrum can be assembled. ˘ day we hear the word Then, reinforcements can come to us as one man on the (very) 261 of his attack.

Perhaps the most formal recognition of two modes of life among the Yaminites is found in a legal document that grants ownership of 150 iku of agricultural land, far more than most individual fields, to the same “hibrum ˘ pay of the backcountry” (ARM VIII 11).262 Both the eight recipients, who ten minas (more than ten pounds) of silver for the land, and the five sellers are said to belong to “the sons of Awin,” defined explicitly as a descent-based group. The very numbers suggest more than a household, and the document itself supposes that the category embraces both a steppe-based and a townbased component of the same descent group, the Awin-ites. Together, they

96


constitute “the house of Awin” (line 3, E´ A-wi-in), which is located in the larger tribe as the “Rabbean Awin.”263 In fact, the only feature that distinguishes those who receive the land from those who give it up is where they live. The five sellers are “inhabitants of (the town of) Appan” (lines 9–10), and the eight buyers are “the hibrum of the backcountry” (line 21). Both groups are defined as leaders ˘ a title that means something like “one who received sashes(?).”264 The by town of Appan is not identified as Yaminite by inclusion in the military lists of ARM XXIII 428 and 429, but it is known to be in the Mari district of Zimri-Lim.265 By the very nature of the transaction, which transfers land for cultivation from the town community to the steppe community, we cannot conclude that the mobile group takes care of the settled group’s flocks while the settled group farms for the mobile group. Both groups are portrayed as self-sufficient, if interdependent, each represented by adult men. It is not clear whether both communities include full households, though the possession of crop land could keep dependents of the hibrum in the river ˘ valley.

2. Outsider Terminology ARM III 12 illuminates the policy of Zimri-Lim toward the Yaminites. KibriDagan describes a problem with Yaminite sarr¯ar¯u who are returning freely to their towns, against the interest of the kingdom: Another thing. In the past, before my lord would go on campaign, individual Yaminites who were not sworn to loyalty were going to and from their towns from the higher country. At present, after my lord went off on campaign, he gave us a strict order (that this cease). I have made (this) a crime, and none of those unsworn may go as before to his town from the higher country.266

By having “their towns” and returning there “from upstream” (el¯enum), the people identified as sarr¯ar¯u (“resisters”?) are behaving like normal tribesmen. The only thing that distinguishes them is that they have not sworn loyalty to Zimri-Lim. What is intriguing is that Zimri-Lim had a policy of excluding these people from the towns of his kingdom. In effect, these were the Yaminite “hana,” and they were driven away. The Zimri-Lim kingdom ˘ allowed acceptable, oath-sworn transhumants under other descriptions but did not generally call these people “hana.” ˘ and others that the sarr¯ar¯u in Mari Durand has concluded from this text texts are most often tribal pastoralists who have not sworn loyalty to the king.267 These people are associated with the grazing lands of the steppe (nawûm), where they cannot be controlled. Their dependence on flocks is evident in a letter to Zimri-Lim from the governor of Saggaratum, who reports that a band of thirty sarr¯ar¯u has been making raids in order to steal sheep.268 Whether or not the term refers specifically to those not sworn to


97

loyalty, it certainly describes those outside royal control, which effectively amounts to the same thing. I suspect that under Zimri-Lim, the word sarr¯arum was used especially for non-Simalite tribespeople, as in ARM III 12 (just cited). Compare also ARM III 16, the letter mentioned earlier that describes men who likewise come down from the higher country (el¯enum) to visit their wives at night. These men are also Yaminites. They are not called sarr¯ar¯u, but the same sort of outsiders seem to be in view.

3. When Yaminite and Simalite Merge: A.981 With my initial presentation of the separate social structures of the Simalites and the Yaminites, I introduced a remarkable text that describes the formal entry of a Yaminite town into Zimri-Lim’s kingdom. (See “Simalite and Yaminite Social Structures,” p. 63, for translation.) The letter in question proposes the affiliation of the town of Dabiˇs with the Simalite gayum division of the Nihadˆ u, without necessarily giving up its Yaminite identification with ˘ u tribe.269 Notice that Dabiˇs does not adopt any “hana” aspect, the Yahrurˆ ˘ which ˘would be inappropriate to a town. One particular contribution of A.981 is that it provides two key social and political categories from a native Yaminite perspective: hibrum and kadûm. Neither word is attested for the ˘ Simalites during the period of Zimri-Lim, and by the fact that the Yaminites of Dabiˇs renounce any connection with either, they are shown to be important. a. The hibrum. The hibrum is a Yaminite word that derives from the verb ˘ which Durand ˘ renders as “to leave one’s house.”270 The verb seems hab¯arum, ˘to refer not to a move from one building to another in the same town but to removal from one political domain for another. The implication may be “refuge” in some cases, but this nuance is not essential to the verbal action. Individual “emigration” is generally in view, not group action, though it is not clear whether such change of residence is viewed as permanent. A sampling of the verb in active use is provided in the accompanying note.271 One use of the noun hibrum with the Sutˆ u nomads appears to prove that ˘ Yaminite, but it forms no part of the Simalite pasthe word is not uniquely toralist terminology for their social and political structures. I have gathered the following texts, knowing that there are others among the unpublished material from Mari. arm xxvi 168. I cited this text earlier for the contrasting locations of the “towns” (¯al¯an¯u) and “backcountry” or “steppe” (nawûm), as well as for the contrasting settled and mobile tribal populations (s.a¯ bum and hibrum). The letter was sent by three diviners who are attested in a series of˘ letters to Sumu-dabi, a leading Yaminite ruler just before the revolt against ZimriLim.272 The combination of hibrum with .sa¯ bum confirms that the hibrum ˘ ˘ is a group of people, not a place, camp, or other aspect of the herding

98


community, a conclusion already suggested by Durand’s etymology. One further implication of the pairing with .sa¯ bum is that the hibrum itself is allmale, another idea proposed by Durand.273 Generally, the ˘word .sa¯ bum refers to a group of men assembled for some organized task. In a military setting, this is a “troop,” and otherwise, it can be a work crew. I am not aware of any use of the word .sa¯ bum that explicitly includes women, though this may just be the conventional pattern.274 Among the Simalite Hana, the military ˘ identity and is aspect represents one expression of the mobile pastoralist naturally male. At least, both categories focus on the group at work, not the family units of a community that lives together. a.2796, in M.A.R.I. 6, p. 288. The writer of this letter is an official from the reign of Yasmah-Addu who managed to survive the transition to Zimri˘ Lim’s rule. In his report to the earlier king, Sumu-hadˆ u describes a troop ˘ that escorted the queen (“my lady”) to Qatna, far south and west in Syria. This troop included both men who have now returned to their towns and ´ s hi-ib-rum ˇsa naothers who are identified as “the hibrum of the steppe” (lu.meˇ ˘ the category to the time˘ before Zimriwi-im, lines 16–17). The text extends Lim’s reign, so that it cannot simply be an expression of his administration. ´ ˇ marking it as a The word hibrum is preceded by the determinative LU.ME S, category of˘ human personnel, possibly (but not necessarily) all-male. Sumuhadˆ u is not known to be a Yaminite, and the letter comes from outside the ˘ Yaminite/Sim alite tribal divisions. The southwestern location would suit well a Yaminite identity, but we have no definite evidence. a.981. The town of Dabiˇs renounces the hibrum and kadûm that it pre˘ sumably has as a Yaminite community, with Yaminite identity in the Yahrurˆ u ˘ tribe stated explicitly. Both the hibrum and the kadûm are in the “steppe” (nawûm), as in the previous two ˘references, but the very renunciation implies a relationship to the town. The kadû appear to be plural, perhaps as leadership of the hibrum. If so, this would contrast with the individual ˘ led Dabiˇs as a town. sug¯agum that would have arm viii 11. This legal document, discussed above in this section, is shown to be Yaminite by its identification of the clan in question with the tribe of the Rabbˆ um. In this case, the contrast between settled and mobile is defined by a further opposition, between the hibrum and the substantive participle waˇsib¯ut-, “those who reside.” The verb ˘“to reside” (waˇsa¯ bum, to sit, to stay) also applies to Simalite Hana who are established in a settled district, so we can associate it exclusively ˘neither with permanent sedentary life nor with temporary encampment.275 These groups do not obviously belong to the ` “field”) seems same household, and the transfer of agricultural land (A.Sˇ A, to imply that the families of both sets of men will be involved with cultivation at some level. We are not dealing with one household.276 As already observed, the hibrum category was allowed to have recognized legal standing ˘ under the reign of Samsi-Addu, as part of normal Yaminite social structure. Only Zimri-Lim appears to have insisted that all Yaminites swear fealty to


99

him as part of settled communities, leaving in question the status of the traditional hibrum. A.2801. ˘A high official stationed at Mari writes to Zimri-Lim about matters of interest to the king during his absence from the capital.277 The text is damaged, but it refers to a hibrum that has arrived at Mari from the company ˘ of Simah-ilanê, the king of Kurdˆ a, who has also just come to Mari. The hibrum ˘ ˘ is identified with two named men, evidently their leaders, and together as Sutû, a category of nomadic peoples from outside the Yaminite/Simalite duality.278 From this reference, it is clear that the word “hibrum” can refer to ˘ have no evidence mobile pastoralists who are not Yaminite, though we still for Simalite herdsmen being identified this way. arm i 119. ARM I 119 is also from the reign of Samsi-Addu. A man named Yatarum was the sug¯agum of the town of Yail and has now died. The town of Yail fell in the district of Saggaratum under the reign of ZimriLim, without evident tribal affiliation.279 Five of their “heads” and someone from the related hibrum came to Yasmah-Addu to request a new sug¯agum: ˘ ˘ Yailites [and . . . (?)] the hibrum “Now five men from(?) the heads(?) of the ˘ 280 have come to me.” Evidently, the hibrum continues to play an active role in ˘ there is a direct relationship between political decisions for the town, so that them. At some level, the town sug¯agum appears to govern the larger clan, both settled and in the hibrum, because the hibrum shares the responsibility ˘ ˘ to choose one. Out of the above texts, three refer to Yaminite pastoralists in the backcountry, and none of them is written from Zimri-Lim’s circle of leadership. A.981 itself places the word in the mouth of the Yaminites themselves, through the leaders of their town of Dabiˇs. ARM XXVI 168 is a letter from Yaminites to another Yaminite. The legal text ARM VIII 11 was produced for Yaminites under the reign of Yasmah-Addu and his father. Only one ˘ other text comes from the period of Zimri-Lim, in this case from an official in the king’s own court (A.2801). That official uses the term hibrum ˘ the for a contingent of Sutˆ u nomads. Our two other references come from reign of Yasmah-Addu and offer no clear tribal association (A.2796 and ARM I 119). ˘ It appears from this evidence that the Yaminites themselves did use the word “hibrum” to describe the part of their population that moved in the ˘ backcountry with the flocks and that this usage persisted untouched by the turnover of Mari administrations. At the same time, however, Zimri-Lim seems to have refused to recognize the legitimacy of this mobile Yaminite population. In his circle, one did not acknowledge a Yaminite hibrum nor did ˘ one use the word “Hana” (“tent-dweller”) for the mobile pastoralists among ˘ these tribal rivals. If Yaminite herding groups wanted formal acceptance as part of Zimri-Lim’s kingdom, they had to get it by swearing loyalty through a town, under threat of census and draft, without any legal right to disappear into the backcountry.

100


I found no “hibrum of the Hana” and only one broken reference to the ˘ ˘ “tent-dwelling hibrum of the steppe” (Hana hibrum ˇsa nawêm, M.5172, un˘ ˘ ˘ This is the same group and published; Charpin and Durand 1986, 154n67). phrase of ARM XXVI 168, so perhaps it is Yaminite terminology. It seems unlikely that this text was produced in the circle of Zimri-Lim’s administration. The words “Hana” and “hibrum” overlap, with “Hana” the broader ˘ ˘ term that can be applied to all mobile pastoralists. In this˘ case, the “hibrum ˘ of the steppe” appears to be in apposition to the general “tent-dwellers,” providing a more precise description of the people in question, especially if the phrase belongs to the Yaminite social system in particular. b. The kadˆ um. This is the only occurrence of the word in the published Mari literature, as far as I know, and Durand compares it with another noun from the same root, kidûtum. There are now two new references to words from this root in ARM XXVIII, the volume of Zimri-Lim’s correspondence with other kings, and Kupper reports (and follows) a change in Durand’s interpretation. In his publication of A.981, Durand suggested that the kadûm may have been a sort of chief, based on comparison with unpublished records in the form in¯uma ki-du-ut PN (personal name), “at the time of the kidûtum of PN” (1992b, 119 and 167). At the time, Durand translated the noun kidûtum as “coronation.” More recently, Durand has changed his mind about the noun, as passed on by Kupper: Dans un premier temps, J.-M. Durand avait estimé qu’il s’agissait d’une cérémonie d’intronisation. Cette interprétation semblait convenir ici [dans ARM XXVIII 147] puisque, dans le même contexte, Ili-Eˇstar fait allusion a` de l’huile, qui aurait e´ té tout naturellement l’huile d’onction. Mais a` présent, J.-M. Durand est d’avis que le verbe kadûm, au lieu de signifier «faire roi, introniser», a le sens d’accorder sa protection, en réponse d’une requête. Sans doute un geste symbolique devait-il manifester cette protection, et l’huile en question pourrait avoir servi dans un rite d’accompagnement.

Here is the actual reference, in a letter to Zimri-Lim from Ili-Eˇstar, king ˇ a in Ida-Maras.. Whatever its precise nuance, the kidûtum in this text of Sunˆ is a rite performed for this local ruler, who addresses Zimri-Lim from the subordinate stance of “son” to “father”: My father wrote [to me] about the kidûtum rite. My father knows that the oil was a problem. Now, Qarni-Lim has anointed me with oil, but no subordinate (s.uh¯arum/lu´ TUR) is at hand to (serve) me. Your god must speak so that I can acquire ˘ a servant who is the servant of my lord. Now, my father must make a ruling on these matters. I will never set myself up to compete with my father’s power. As for me, I am at the ready either as your servant (vassal king?) or as a sug¯agum. I will never let go of the corner of my father’s garment. I am a loyal son of this land.281

The other occurrence of the root is as a verb, with considerably less context to narrow the interpretation. Kupper translates this other text, “après que


101

Hammân de Dêr eut accordé sa protection a` [messire] Samsi-Erah, [ . . . ] ils prirent Samsi-Erah a` Susâ” (ARM XXVIII 91:3 -5 ).282 There may be further unpublished evidence that leads Durand to interpret the verb as “to provide protection,” but I am not convinced that this translation makes the best sense of these two texts. The second text offers little control for the verb. It must describe what a person in power, evidently the ruler of Der, does toward a leader of lesser status. Samsi-erah is a local leader from the vicinity of Talhayˆ um who is eventually captured ˘and killed ˘ by the people of a town called Ulaya.283 The first text offers a quite detailed context for the kidûtum rite. Above all, Zimri-Lim wants to make sure that the rite has been performed for the ˇ a, one of his vassals in Ida-Maras.. It is of diplomatic concern, ruler of Sunˆ even to the point where the Simalite king wrote to Ili-Eˇstar to make sure it takes place. Ili-Eˇstar first reassures Zimri-Lim that an anointing rite has now been performed by Qarni-Lim, king of Andarig and one of Zimri-Lim’s most important allies. The question of why Ili-Eˇstar, as a vassal of ZimriLim, needed a special servant eludes me entirely. Perhaps the kidûtum rite itself had not yet been performed, and this “servant” is the person with the sacred authority to carry it out. In this case, Ili-Eˇstar has raised the issue of the servant as an excuse for apparent disobedience, and he follows this excuse with a litany of promises, all of which seem aimed to convince ZimriLim that he will remain a loyal vassal. Regardless of the exact relationship between the anointing, the servant, and the kidûtum, the final sequence indicates an association with the ritual commitments of vassalage. “I am at the ready either as your servant or as a sug¯agum. . . . I am a loyal son of this land.” As both “servant” and “son,” Ili-Eˇstar identifies himself as Zimri-Lim’s vassal, in the same terms used in addressing letters to the suzerain king. In his initial interpretation of the kidûtum as “coronation,” Durand appears to have read the text similarly, with the particular notion that this was when the suzerain granted his vassal the ˇsarrum title as proper “king” (1997, 467). One letter from the fourteenth-century Amarna correspondence demonstrates that this rite of proclaiming a new vassal king could be accompanied by a rite of anointing with oil.284 Whether or not Ili-Eˇstar’s promise to be available as “servant” in ARM XXVIII 147: 13 implies the title “ˇsarrum,” I find that Durand’s first interpretation should not be lost in the new focus he offers with his second. It is hard to imagine how the long plea of assured loyalty as servant, sug¯agum, and son can be explained by a simple rite of protection, unless this is the specific protection of the suzerain for a vassal. This may be what Durand has in mind. The oil need not be for the kidûtum as such, if anointing belongs to another aspect of the process of confirming vassal status, in this case performed already by Qarni-Lim. All of this brings us back to the distinct form in our Yaminite context, the kadû.285 Our detailed reference to a kidûtum ritual assumes performance as part of undertaking the responsibility of leadership on behalf of a higher

102


authority, and the renunciation of hibrum and kadû in A.981 would make sense as the pastoralist community in˘ the steppe and its plural leadership.286 Surely, the kadû are neither kings nor vassals, and some more general root action is required. If Durand is right that Samsi-erah is “taken under the ˘ the kadû could be protection” of his superior in ARM XXVIII 91, then “those who take under protection” or guardians. If they are indeed the leaders of the hibrum, this suggests that the hibrum is not governed directly ˘ ˘ by sug¯agums, though an individual mobile community is probably subordinate to a sug¯agum through its affiliation with a town, as when the leaders of Yail pursue the appointment of a new sug¯agum in ARM I 119. It is unclear how the hibrum and its kadû would relate to the Yaminite merhûms, ˘ that even merhûms of lower status than those of Zimri˘ but one expects ˘ Lim would still have held the highest position of authority in the grazing lands. Any identification of the kadûm from the published evidence cited here must remain hypothetical. Regardless of the specific meaning, it is possible that the declaration that “we have neither hibrum nor kadû” does not state ˘ that all clan members are in a permanent condition but simply promises residence, are accounted for, and are able to swear the loyalty oath. Perhaps the new affiliation of the Yaminite town with a Simalite tribe provides an acceptable administrative framework for authorizing transhumant movement. Once accounted for by their identification with a Yaminite town that is properly committed to Zimri-Lim, the mobile component of the community could once again head for the steppe under the Simalite association. Why should a Yaminite town with a particular Yaminite tribal affiliation promise to forge a new affiliation with a specific Simalite tribe? Why should this be necessary for entry into Zimri-Lim’s kingdom, when we know that the town of Dabiˇs was governed as part of the nontribal framework of governors, in the district of Saggaratum? Perhaps the Simalite tribal tag served the mobile pastoralist segment of the community in particular. It is possible that any hana element of the Yaminite population in Zimri-Lim’s kingdom had ˘ a Simalite registration. to have Altogether, the prominence of tribal categories in the Mari archives is remarkable if not unique in evidence from the ancient world. In the Mari archives, tribal groups and tribal leaders constitute more than an obscure population that is peripheral to the centers of power that dominate most ancient writing. Under the interrupted dynasty of Yahdun-Lim and ZimriLim, Mari itself was ruled by leaders of the Binu Sim˘al, one of the great tribal peoples of eighteenth-century Syria and Mesopotamia. Whatever we may eventually discover about the role of tribal and lineage identities in other early Near Eastern settings, here, at least, the social structures and political customs of tribal peoples are central, not peripheral. In Zimri-Lim’s realm, tribal populations do not negotiate a relationship with city-centered power. They hold the reins of power and dominate the population. There


103

is indeed a meeting and intermingling of political traditions both from the tribes and from the landscape of kingdoms and towns that is the focus of the rest of the book. For the kingdom of Zimri-Lim, however, the tribal categories represented the primary structures of the ruler’s own people. The Simalite gayums and their sug¯agum leaders shaped Zimri-Lim’s core constituency as “tent-dwelling” hana, still defined by their mobile pastoralist ˘ chiefs of pasture. Even the Binu Yamina, component, with its two merhûm ˘ the cousins and sometime enemies of the Binu Simal, contributed large numbers to the Mari kingdom, so that their own distinct tribal structures add further dimensions to Zimri-Lim’s dominion. It is no wonder that after Hammurabi of Babylon added the Mari kingdom to his own larger realm, ˘ began to call himself king of “the Amorrite land,” a land noted for living he tribal commitments and perhaps current use of the Amorrite language, to match the “Amorrite” identity of Hammurabi’s own ancestors. ˘

3 The Archaic State and the m¯atum “Land”

It is not possible to understand the role of collective decision making in the Mari evidence without taking into account the basic structures of social and political life that are displayed in the archives. Moreover, although group decision making is specially associated with the town, it finds some expression at every level of the ancient political organization displayed at Mari. I have divided these basic structures into two main categories and set out to explore these in Chapters 2 and 3, before reaching the towns in Chapter 4. During the time of the Mari archives, the tribal Binu Yamina and Binu Simal dominated the region of Mari, as well as much of the surrounding area. I began with the tribal world of Zimri-Lim because a large proportion of the kingdom’s population identified itself in tribal terms, ultimately including Zimri-Lim himself. For those with tribal identities, tribe took precedence over both towns and the larger polities called m¯atums (“lands”). Perhaps the large states such as Babylon, Yamhad, and Eˇsnunna dominated regional ˘ the fortunes of their rulers, growing politics, but their identities followed and shrinking with the latest victory or defeat. Towns offered a basis for family identity that could go back generations, but families could always move. Tribal identities, on the contrary, accompanied people across every geographical and social boundary. Philip Carl Salzman relates a story about how his Baluchi friends reacted with stunned silence to his admission that North American society was not based on lineages. “The standard follow-up was, ‘What do you do if you get into trouble; who do you go to?’” (2000, 231). This sense of absolute solidarity survived any short-term dispersal of lineage members. We do not know enough about urban society in the ancient Near East to establish the role of descent-based relationships in towns, but it may well be that these formed the foundation for primary loyalties in most settings. It is this primacy of tribal relations that provides the logic for beginning the book with an examination of “The Tribal World of Zimri-Lim” (Chapter 2). 104

The Archaic State and the m¯atum“Land”

105

The political world of the Mari archives is most obviously dominated by an entirely different form of organization, the large units from which the Mesopotamian political map was drawn. In the idiom of early secondmillennium Akkadian, the language of the Mari letters, the world was divided into m¯atums, populations that were recognized to have the capacity to negotiate their own peace and war. We cannot make sense of the collective political structures of the town without examining the larger polities into which they were imbedded. The towns could not act except as part of or subordinate to a m¯atum or “land,” the focus of this chapter. At the time of our archives, most of Mesopotamia was divided into kingdoms of varying sizes, but the primary units of the political map were not defined in terms of kings. A m¯atum consisted of the people who together represented a decision-making entity with its own voice, its own seat at the table in the community of Mesopotamian powers. Every m¯atum that recognized an individual ruler was a kingdom. We shall see that the cities of Tuttul and Imar exercised considerable independence in their dealings with the great powers without themselves having a local ruler by any title, but both had to submit to Mari’s and Yamhad’s sovereignty, respectively, and the m¯atum ˘ language was never used of single-city polities who were governed corporately. In most cases, m¯atum status was acknowledged for polities governed by “kings,” individual rulers who could claim the title ˇsarrum. Mesopotamian terminology offers no basis for distinguishing “kings” from “chiefs” in a way that would allow isolation of “states” from “chiefdoms,” and the scale of the m¯atum could vary widely. Three important m¯atums, however, were constituted not as kingdoms but as coalitions of several rulers with ˇsarrum status: the lands of Ida-Maras. in the northeastern Habur River basin, of Zalmaqum ˘ the east of the Tigris. ˇ to in the Balih River basin, and of Subartum ˘ Every polity in ancient Mesopotamia of the Mari period was understood to act either as a m¯atum or in subordination to a m¯atum. Within the kingdoms of Samsi-Addu and Zimri-Lim, that subordination could be defined in two radically different ways. On one hand, the kings could define “districts” (hals.um) around the royal governors (ˇsa¯ pit.um) who oversaw them on their˘masters’ behalf. These hals.ums had no political traditions of their ˘ own and no political identities except as extensions of royal authority. On the other hand, however, these kingdoms incorporated into their dominions certain towns that brought with them their own political identities and structures that functioned as they always had, though now subordinate to their sovereign. This was true during the reign of Zimri-Lim for Tuttul, upstream from the main districts of Mari. The primary political categories of early second-millennium Mari were then the town and the “land” (m¯atum), with its subsidiary “districts” (hals.um). ˘ Although the m¯atum was ultimately defined by its people and not its territory, other terms for the people governed by kings lack the political identity of the m¯atum. In particular, the word niˇsu¯ is usually translated as “people” but

106


derives from the household circle of dependents on a patriarchal family head. The word muˇskênum refers to the component of the population that is not directly dependent on income from the palace. Neither term has a political use, in that neither one is identified with a decision-making entity. One of the most intriguing and difficult aspects of Zimri-Lim’s kingdom is his standard royal titulary, which bridges the categories investigated in this chapter and in the previous one. Like Yahdun-Lim before him, Zimri˘ Lim calls himself “the king of Mari and the land (m¯atum) of the Hana.” I ˘ conclude this chapter with sections dealing with the nature of Zimri-Lim’s kingdom, as it joins city center to a population that is defined not merely in tribal terms but by the mobile pastoralist category typical of all Simalite tribal identity, the hana. ˘ Altogether, this study of the m¯atum is necessary to understand how kingship related to the political traditions of tribal social organization and concentrated residence in towns. I propose that in the Akkadian language, emerging complex polities that may be called archaic states were defined not in terms of the settled center or “city” but by the territorial population of the m¯atum. Akkadian does not recognize what generations of modern scholars have pursued as “urbanism,” social and political complexity defined by the city. The basis for the complex m¯atum is thus a population that may be mostly settled or may include a large mobile contingent. It may have a primarily agricultural economic base or depend heavily on pastoralism. It may or may not involve a prominent tribal aspect to its social organization. The use of the word “m¯atum” for the complex polities of archaic states suggests that we not confine our investigation to conventional “cities” and urbanism. It is crucial to observe that although the political landscape of the early second millennium was dominated by kingdoms, none of the primary political categories is defined by or requires a king. The m¯atum comes to identify the realm of a king, or that which he rules, but it can exist apart from the king himself, a population with its own political will, which an effective king must bend to his own. Moreover, the m¯atum can be a coalition, not a kingdom. A second political center is defined by the “town,” or a¯ lum, the focus of Chapter 4. This category carries with it a strong tradition of collective governance that stands as a counterpoint to individual governance and that requires no king. At any rate, the a¯ lum can be a simple village as much as a royal capital. Kingship comes as an overlay to this political vocabulary, and we should see the language and institutions of kingship in the Akkadianspeaking tradition as built onto a variety of prior arrangements with different relationships between individual and group authority.

a. urbanism and archaic states The building blocks of political power in the world of the Mari archives are the towns (¯alum), but the structures built from them are the “lands”

Urbanism and Archaic States

107

(m¯atum). Although a m¯atum could have wildly varying size, in concept, the m¯atum involved a population spread across multiple settlements that required some form of coordinated governance, at least in time of crisis. So far as the Akkadian language recognized the existence of more complex polities that we might call “states,” it was by the word m¯atum. Apart from the tribal social organization that suffused much of Mari-period society at all levels, the political world of Mari consisted of entities called either m¯atums or a¯ lums, “lands” or “towns.” As we seek to unravel the relationships between these terminologies, it is worth beginning with some reference to the larger discussion of “urbanism” and the early “state.” To some extent, the Akkadian terminology cuts across common classifications, but that in itself should be instructive.

1. States and Cities in the Ancient Near East Interpretation of ancient Mesopotamia has followed that of the rest of the world in moving from a focus on defining cities to defining states. Archaeologists have applied the tool of settlement analysis to make judgments about political formations based on survey evidence, rather than relying only on the excavation of major urban centers. In more complex societies, villages tend to be scattered among and thus visibly linked to larger towns. In states, such towns are in turn distributed across the landscape in patterns that suggest further linkage to one or more true cities. This arrangement is called a “settlement hierarchy,” and where three or four hierarchical tiers are visible, archaeologists conclude that they are dealing with a state, whether or not written evidence is available to identify it.1 In the discussion that follows, I take a cautious approach to any hypothesis that proposes a definition of the state that is too rigid and certain. Much as the great Mesopotamian archaeologist Robert Adams has participated in this search for the early urban-based state, he himself warns against overgeneralization from such settlement analysis (1988, 27–8).2 In spite of the barriers both to defining and to conceptualizing complex polities in early Mesopotamia, something new certainly appeared on the scene, and perhaps more than one thing. Southern Mesopotamian cities have long stood at the center of the discussion about early urbanism and state formation, based perhaps on their size and concentration in a single location, together with the appearance of cuneiform writing in late fourth-millennium Uruk. Very different settlements and political arrangements arose further west and north, including the region displayed in the Mari archives, where there is some danger in allowing southern Mesopotamian urbanism to define the terms of analysis. What should we call the political forms that accompanied the new cities and settled centers of Mesopotamia in the third millennium? Archaeologists have long observed that the languages of the cuneiform writing system

108


fail to recognize the arrival of the new urban center by special words for “city.” Adams, for example, once said that the Mesopotamian words for “town,” Sumerian uru and Akkadian a¯ lum, were associated neither with a particular settlement size nor with specific “urban” institutions (1981, 136; also Oates, 1983, 81). The second statement, regarding “urban” institutions, may underestimate the political implications of the words for “town” themselves, and the next chapter is devoted to the town as such. The first remark, meanwhile, is certainly true. Abundant evidence from diverse periods and regions links these terms to everything from major cities to tiny hamlets. For example, texts from the northwest Syrian city of Alalah list all ˘ of the households for individual “towns” (¯alu), actually, villages averaging about twenty-five “houses,” in one case as small as three houses (Liverani 1983, 160–1; 1975, 153). A tablet from Carchemish under the Neo-Assyrian empire lists a small town with sixteen heads of household (Fales 1990, 100; cf. Van De Mieroop 1999, 10). William Hallo observes that Hittite likewise describes the full range of settlements by one term, happiraˇs, in contrast ˘ (1971, 58). It is to other ancient Near Eastern languages, such as Hebrew odd to find that the peoples of the region who gave the world its first great cities never bothered to come up with any special words to recognize their innovation. Did they not see the change they had wrought? I argue that they surely did, but that the Mesopotamian peoples approached the change from a perspective different from our own. First, they continued to apply their words for “village” to the growing towns and cities because the language was tied to a political definition that was not quickly abandoned with the transformation. As is elaborated in the next chapter, the a¯ lum, and evidently the Sumerian uru as well, appears to be defined above all by its people, especially the people as a corporate whole in action. The defining trait of the a¯ lum and uru is then political, and the institutional face of town polity is the collective body. Because the “town” terminology continued to carry a very specific meaning, the political changes naturally inspired a fresh approach that reflected the ancients’ own view of the essential novelty. It appears to me that the Mesopotamian languages recognized the arrival of cities in their real political context, as part of larger networks. The language of urbanization is therefore the language of the state, or at least of emergent complex polities that resulted in archaic states.3 If we return to the question of how to distinguish a true “city” from mere towns, without the help of the Akkadian a¯ lum or the Sumerian uru, it is intriguing to discover that archaeologists will not define a proper “city” except by the matrix of towns and countryside in which it is imbedded. It should be natural, then, when we evaluate the native terminologies, to look for alternatives to words for “big town.” Especially if the “town” tag refers specifically to the corporate political mode that we explore in Chapter 4 – and this mode is understood to pertain to every concentrated settlement from hamlet to metropolis – then urbanism will require an entirely different sort


109

of recognition. Just as the “town” designation is political, the native languages can be expected to have identified the larger political category that the cities represented. In the Akkadian terminology of the Mari archives, that category is the m¯atum.

2. The City Some decades ago, V. Gordon Childe presented a powerful argument for the importance of early Mesopotamian urbanization in the development of human societies. Although his theoretical framework, with its engine of economic change and its triumphal progress from “savagery” through “barbarism” to the “urban revolution” is now dated, we should not underestimate the significance of the underlying reality that he was pursuing.4 Anthony Giddens, for example, declares that his massive treatise on social theory has at its core the centrality of “the city” (1981, 140). In what Giddens calls “class-divided” (vs. later “class” or “capitalist”) societies, “cities are crucibles of power” (p. 145). While the emergence of the city both signals and itself represents massive changes in human societies, most contemporary research looks outside the urban center to the whole territory governed from that base. Giddens himself says that the city is the “power container” that generates the structural nexus of the state form, in a relationship symbiotic with the countryside (1984, 195–6). In his evolutionary approach to “civilization,” Service replaces Childe’s idea of an “urban revolution” with a focus on “the institutionalization of centralized leadership” that came to flower in the state (1975, 7–8). Giorgio Buccellati declares that “instead of urban revolution one might perhaps more properly speak of ‘state’ revolution” (1977, 20).5 Where true cities appeared in the Near East, there is little question that these were associated with the formation of archaic states. It is not obvious, however, that polities of comparable complexity require urbanism for their existence. Moreover, it is not easy to match modern terminology with ancient society in a way that allows intelligible use of words such as “town” and “city.” Both cities and towns are concentrated settlements, defined at least in part by population. Early southern Mesopotamia was indeed characterized by such “cities,” but the large third-millennium sites in northern and western Mesopotamia often suggest a similar level of political complexity without evidence for a large populace. It seems then that archaic states, or complex polities that some may call such, may have formed without urbanism. Anne Porter recently suggests, regarding the middle Euphrates site that she excavated with Thomas McClellan, that Banat remained throughout its history a critical hub in the sociopolitical, and especially ideological-ritual, systems of the pastoralists who founded it, just as

110


pastoralists continued to constitute the economic and demographic base of the town, despite the fact that once established, it might be expected that the town developed as a completely separate entity. But in fact there is no evidence that the population required to sustain a settlement of this size and type, with its heavy investment in monumental architecture and its extensive industrial sector, was ever permanently settled in the area. Rather, like other cities in northern Mesopotamia at this time, Banat was an administrative/ceremonial/manufacturing center, catering to a largely untraceable public. (2002, 28)

3. The State I am not ultimately interested in defending the use of the “state” category to describe ancient complex polities, but I would like to explore how the Mari evidence relates to larger discussion or cities and states. It is necessary neither to my analysis of the m¯atum and the a¯ lum nor to my investigation of Mari-period society more generally to define the “state” in rigid terms. By attending to the corporate aspect of ancient politics, however, we encounter characteristics of ancient societies that do not easily fit the more rigid definitions, and some comment is worthwhile. In his influential work on cultural evolution, Service focuses on the power to compel. The state is distinguished by power “institutionalized,” so as to use or threaten actual use of force (1975, 14). Michael Mann, elaborating on Max Weber and representing a view from social theory, adds the aspect of territory and a degree of precision that narrows the resulting definition: The state is a diffentiated set of institutions and personnel embodying centrality, in the sense that political relations radiate outward to cover a territorially demarcated area, over which it claims a monopoly of binding and permanent rule-making, backed up by physical violence. (1986, 37)

Richard Tapper offers what he regards as an anthropological consensus, reaching beyond power alone to consider the constituencies of the state. The state is “a territorially bounded polity with a centralized government and a monopoly of legitimate force, usually including within its bounds different social classes and ethnic-cultural groups” (1990, 50).6 All of these definitions carry assumptions from more recent settings that may not be necessary to the ancient Near East. For instance, Mann’s notion of territorial demarcation involves ways of conceiving the relationship of political authority to land that simply did not exist in much of early Mesopotamia. If the key elements in Tapper’s definition are the territory, the centralized government, and the monopoly of legitimate force, it is not clear why “different social classes and ethnic-cultural groups” must be essential to archaic states. Even if we accept the boundaries of these definitions, however, the kingdoms of both Samsi-Addu and Zimri-Lim were states, if institutionally primitive ones. Their government was certainly centralized, and


111

each king ruled a considerable domain, though that territory was defined more by nodes of settlement and the routes that connected them than by blocks of two-dimensional space with strictly drawn borders.7 These kings certainly claimed a monopoly of legitimate force. Even if the definitions succeed in rough terms, however, they leave large swaths of the political landscape in the Mari period unexplained or underexplained. How are we to relate the m¯atums to “state” terminology? None of these definitions defines how big a state must be, in terms of either population or territory, and many of the m¯atums were quite small. The three m¯atum coalitions of ˇ Ida-Maras., Zalmaqum, and Subartum were not truly centralized or authoritative, except in the pursuit of foreign affairs. This observation by itself raises the question of what specific areas of activity must be included in a “state”’s monopolized authority. The tribal confederacies of the Binu Simal and the Binu Yamina present even more acute difficulties. Both groups define not just a social but a political reality, each with large population and complex organization, each spread across a wide territory, and each existing as a confederacy in part for the purpose of joint political action. The Simalites and the Yaminites cannot simply be excluded from the discussion of early second-millennium states, because they participated in the community of states. The Simalite tribal confederacy could be said to have been constituted as a state under Yahdun-Lim and Zimri-Lim, so far as their leader was king over a state based at ˘Mari, though the state definition offers a feeble model for understanding the complex political system of tribal organization. We will see that the m¯atum terminology seems to have required a settled center or centers, so that the Simalites could take their place among the “lands” only by claiming such a capital. How then are we to treat the Simalite confederacy as an entity separate from the Mari center? Perhaps it was inconceivable for the confederacy to exist without one or more such bases. The Yaminite tribal confederacy could mistakenly be relegated to a lower rung on the developmental ladder of political structure, when their actual organization was very similar to that of their Simalite cousins and served a community of similar scale. They are never regarded as a single m¯atum, even though they may make joint political decisions. Within this political framework, we must recognize a number of interrelated but distinct phenomena. The densely populated cities of southern Mesopotamia were quite different from the early centers of regions further north and west. We do best to disentangle the emergence of the “city” from the emergence of the “state,” which may need no conventional city for its own existence. The Akkadian word m¯atum will offer an important control for our discussion of complex polities in early Mesopotamia, but it will not line up easily with the terms of our modern analysis. The m¯atum is not simply a “state” nor does it recognize all complex political structures, in that it excludes the Yaminite tribal confederacy. The Sumerian political

112


landscape of the third millennium may be usefully defined as a “city-state” system, with its one-to-one relation between the urban center and the political state, combining town and the surrounding hinterland.8 This compound category, however, carries the double danger of forcing any given settlement or complex polity into these “city” and “state” definitions.9 It would be terribly inappropriate to represent Zimri-Lim’s Simalite kingdom as a city-state, for example, just because it had Mari as its “city” center and displayed “state” complexity. I therefore avoid this category entirely in my discussion of the political world of Mari.

4. The Chiefdom At this point, we should consider one other explanatory category that is intended to address some of the difficulties of complex polities that do not look like “states”: the chiefdom. Service uses the term “chiefdom” to describe societies on the road to statehood that had not quite arrived. Chiefdoms would be centralized and hierarchical, like states, but with “no formal, legal apparatus of forceful repression” (1975, 15–16). He defines chiefdom governance, unsurprisingly, in individual terms. When the “big man” of what Service calls the “tribal” level is able to make his special role an “office,” to be passed on to his sons, the tribe becomes a chiefdom (pp. 72–4), though actual leadership in action is mainly for sporadic group projects (p. 94). The chiefs of chiefdoms tend to be strongly associated with storehouses and the redistribution of resources. Timothy Earle acknowledges considerable diversity in the category and proposes an umbrella division based ultimately on numbers: “a polity that organizes centrally a regional population in the thousands” (1991, 1).10 The “chiefdom” remains a widely used category, but it has nevertheless suffered sharp critique, mainly for being impossible to distinguish with confidence from archaic states. Anthony Giddens acknowledges the type but admits readily how hard it is to identify (1984, 247). His antipathy toward evolutionary explanations lies behind this discomfort. If states do in fact have chiefdoms as their antecedents, he says, the states cannot be said to derive simply from the “expansion” or the “internal differentiation” of chiefdoms. Some archaeologists seem to object to the term because they find it inadequate to explain their evidence. Gil Stein observes the new evidence for heterogeneity and competition in ancient Mesopotamian society, rather than homogeneous centralization. “By counterposing chiefdoms and states as discrete categories,” he complains, “archaeologists have tended, perhaps, to overestimate the degree of centralization and power of the latter.” He suggests that we do better to see states as organizations operating in a social environment that they control only in part (1994a, 12–13). Elsewhere, Stein proposes that this separation of chiefdoms from states leads to overemphasis


113

on the state’s complexity, when in fact early “segmentary states” are quite primitive, with the capacity to exercise control diminishing quickly with distance from the center (1994b, 10–11).11 The point is well taken, though the notions of “distance” and “center” may be inadequate to the actual political configurations of northern and western Mesopotamia. In particular, the tribal organizations of the Yaminite and Simalite confederacies facilitated the exercise of control across great geographical distances. Working on the Central American Maya, Olivier de Montmollin rejects the whole typological approach as doomed to mistake settlement size for political importance. His settlement analysis of the classic Maya indicates more than one scheme of political hierarchy, and by avoiding typology, one can identify a variety of arrangements for the distribution of population with reference to political activity, beyond the restricted assumption that “only locationally efficient arrangements are possible.” This openness maximizes analytical precision and flexibility (1989, 86). De Montmollin’s objections strike at the application of typologies to individual societies under examination and do not address the defense of typology on comparative grounds. According to Earle, “An evolutionary typology appears necessary to control for cross-cultural comparisons, and the type chiefdom is useful to define societies of generally similar scale and organization” (1987, 280). This plea for the necessity of cross-cultural comparison is indeed compelling for the possibility of mutual illumination, but it is not clear to me that this is best accomplished by typologies of the sort initiated by Service. The question is whether such comparison might proceed more productively without the types and on a more individual and fluid basis. As argued by Stein for the state, the types carry with them the danger of incorrect assumptions embedded in the models. One testimony to the need for more adaptable interpretive systems is the remarkable article on the Harappan civilization by Gregory Possehl (1998). Faced with massive cities and a grid of contemporary settlements but lacking palaces or monumental religious architecture, as well as writing and individual portraiture, he still refuses to abandon the typologies that fail his specific material. In the end, Possehl simply calls the Harappan polity a “nonstate,” leaving unanswered the larger question of how this reflects on the typological system as a whole.

5. State Systems I find that the most pointed critique of the state/chiefdom types comes from a direct assault on their value for comparisons. Gary Feinman takes on the criterion of population, which is so central to Earle’s definition of the chiefdom. There is no agreement on the demographic threshold between states and chiefdoms, he says, citing a range from 2,000 to 3,000 for “small states” in ancient Europe to Hawaii’s “complex chiefdom” of 100,000 people.

114


Territorial extent is equally difficult to define in universal terms (1998, 97–8). Then, within individual regions, we encounter cyclical changes of population, territorial extent, and degree of centralization. Such realities “should direct our thinking away from the simple taxonomic dichotomies that have been drawn between weak or segmental states and bureaucracies, and toward more continuous models that recognize preindustrial organizations as dynamic formations that may oscillate in scale, integration, and complexity” (pp. 100–1). One part of the solution, argues Feinman, is to reach beyond analysis of the single polity “to larger landscapes in which clusters of such entities interact” (p. 101) in what might be called a dynamic “state system,” an “interstate system,” or a “peer-polity network” (p. 103). The archaic states in such systems may sometimes be relatively small, overlapping with what are often called “complex chiefdoms,” but even the smaller examples do not fit the usual profile of the chiefdom. Feinman provides a list that is worth repeating, to show what particular features are at stake (p. 104): economically stratified classes (often with slaves); rulers who live in palace structures; dense regional populations; use of writing; some market systems; at least three decision-making levels. Many of these would apply to the small m¯atums north of Mari, for example, and even to the individual city-based kingdoms of Ida-Maras. in the upper Habur. Feinman ˘ their systems observes that the groups of small states that he defines by or networks are often linked in confederacies, such as those we see for ˇ Ida-Maras., Zalmaqum, and Subartum. I have lingered over Feinman’s analysis because it is particularly suited to the small scale of the numerous secondary states of the early secondmillennium Mari archives. The Mari correspondence displays a time when several powerful states of modest proportions pushed and pulled for power, and when no empire could achieve regional dominance. The crowd of m¯atums that jostles for influence across Syria-Mesopotamia evidently shares much with the early urbanization in the Americas contemplated by Feinman.

6. Mari and the Archaic State In the previous section, I explored the recent discussion of states and chiefdoms in cross-cultural terms, not to undertake a detailed analysis of the whole regional political system in second-millennium Syria-Mesopotamia but to provide a framework for understanding the ancients’ own language of m¯atums. This framework will eventually have bearing on interpreting the corporate political strategy of towns and on the specific political situation of the kingdoms based at Mari. To complete this chapter on the m¯atum, I turn finally to the discussion of state formation in the ancient Near East. With Mesopotamia and Egypt setting the standard for pristine states formed in a world that had lacked them, the chiefdom/state dichotomy


115

has played a smaller role in the interpretation of this region than elsewhere. Mesopotamia has naturally played a continuing role in the discussion of early state formation. I touch only briefly on the larger discussion, in order to arrive more quickly at the context for the early history of Mari and cities like it: Syria and northern Mesopotamia in the third millennium. Of particular interest is the recent suggestion by Anne Porter that we should envision pastoralist states in this region and period, states based primarily on a pastoralist economic foundation and not simply created by conquest of a preexisting state or states. If Porter is correct, and she identifies thirdmillennium Mari as one such new state, then the reigns of Zimri-Lim and his tribal forebears may not be out of character in the longer history of the site. In our analysis of the corporate tradition in towns documented in the Mari correspondence, our understanding of third-millennium urbanism will shape our rendition of the post-Amorrite age. The Mari correspondence alone will demonstrate that some expressions of this corporate tradition predate the Amorrites and belong to the early urban custom of the region. We need, therefore, some background. It appears to me that the Mesopotamian languages then recognized the arrival of cities in their real political context, as part of larger networks. The language of urbanization is therefore the language of the archaic state. Giddens comments that “the state, expressing the city-countryside relation, represents a new type of structural principle that is counter to the old while still depending on it” (1984, 196). In Akkadian, the word for the polity that embodies the “city-countryside relation” is m¯atum, the “land.” Outside the loanword ma.da, the Sumerian kalam seems most relevant in that as all Sumer, it seems to be defined politically rather than geographically or otherwise. We have already seen how the m¯atum in the Mari documentation is often simply that which is ruled by a king or a confederacy of kings. It overlaps what the typologies would define as a state or a complex chiefdom, though I would prefer to identify the m¯atum landscape as a whole with Feinman’s “peer-polity network” of archaic states. In its common use, the word “m¯atum” relates the royal center to the governed population and therefore to the territory outside itself. It is difficult to track the early development of all the Sumerian and Akkadian terms that are related to political or territorial entities larger than the town. It appears that the emergence of the word “m¯atum” as a political category was associated with a shift away from the corporate polity intrinsic to the ideology of uru and a¯ lum. This word is comfortably identified with the individual leadership that we translate as “kingship,” a political mode widely found in the company of larger populations in preindustrial societies, with their increased economic specialization and social hierarchy. It is likely that the word “m¯atum” was not created from scratch for this purpose but began with simple reference to a “region,” perhaps outside a central town.

116


b. the m a¯ tum : the basic unit of regional politics in the early second millennium The Mari archives define the primary political units of Mesopotamia and western Syria by the term m¯atum, usually translated as “land.” From the perspective of the sedentary population centers, the entire political landscape was divided among m¯atum units. It appears that the m¯atum was always defined with reference to towns (¯alum), though any number of configurations was possible. The a¯ lum is the building block of the kingdoms that form the power centers described in the Mari archives. A new vassal defines the network that serves Zimri-Lim by its towns: “Like his town Nahur and like ˘ the (other) towns of his dominions, the town of Burundu is (henceforth) 12 Zimri-Lim’s, and Adal-ˇsenni shall be his son.” It is not possible to reconstruct the origin of the word m¯atum, but it somehow reflects the appearance of more complex political organization, which is more recent than the simple settlement or village represented by the word a¯ lum. Although every a¯ lum had a physical aspect, and one could definitely speak of the city, town, or village in concrete terms, the only physical feature that was essential to its definition was the clustering of permanent buildings that served some group of people. An a¯ lum did not have to have fortifications or any particular size, and it is not even clear that an a¯ lum had to include “houses,” buildings whose primary use was for residence. The a¯ lum was therefore the physical expression of a political reality, the clustering of some group of people for the shared use of the a¯ lum structures, whether for homes, mutual defense, celebration, or economic exchange. As a political entity, the group capable of joint action, the a¯ lum seems to have begun with a strong collective aspect that was retained as it grew. Eventually, larger and more complex polities arose from various relationships between people separated by distance. Such relationships may well have included tribal identities of the sort displayed in the Mari archives, but they also involved bonds between multiple settlements as such. In some cases, these polities incorporated larger towns or cities. Rather than find a new name for these grown-up villages, with their new forms of specialized labor and new hierarchies of organization, both Akkadian and Sumerian speakers maintained the old names: a¯ lum and uru. In Akkadian, the word “m¯atum” came to describe the polities created by combining populations across distance. Interestingly, Sumerian has no direct counterpart besides the loan ma-da, which suggests that the Akkadian (Semitic) terminology did not arise in the same social landscape as the river plains of southernmost Mesopotamia. Both textual and archaeological evidence for regions north and west of Sumer indicate different political conditions, and the word “m¯atum” should probably derive from that world. By the time of the Mari archives, in the early second millennium, the word “m¯atum” had acquired a range of uses that prevents one simple political

Basic Unit of Regional Politics

117

definition. Use of the term varied with perspective. First, one must distinguish the m¯atum as it faces inward, toward its ruler, from political entities facing outward, for view of their neighbors. Also, the specific relationship of people to ruler evokes a distinct use of the word. In all cases, however, the m¯atum reaches beyond the bounds of individual towns. When the m¯atum is ruled by a king, as in most cases, it is the people outside the royal capital, the population that is called on to give its unswerving loyalty to the king, but whose loyalty cannot be assumed. Conceptually, the m¯atum is apart from the king and, by its very separation, represents a distinct political force even in its obedient submission. By naming the m¯atum as a collective unit, it is rendered politically corporate, though most often governed by an individual ruler from a fortified center.

1. Viewed from Outside, a Single Block The above distinctions of agricultural countryside against fortified centers, and of ruler against ruled, may fade when viewed from a distance. People go home to their m¯atum, whether as kings or as commoners.13 Crafts may be assigned their origin by m¯atum boundaries, so clothes from “this land” in ARM X 173:16. Even political dealings between lands may be described by this general entity without distinction of rulers or ruled. Messengers from Eˇsnunna and king Iˇsme-Dagan of Ekallatum are reported to be trying to persuade the ruler of Andarig to join them: Except for Zimri-Lim and the town of Mari, Atamrum and the town of Andarig, no other king and town is my enemy. Zimri-Lim and the Babylonian – with which king have they (not) entered alliance, and with which land (m¯atum) have they (not) touched the chin?14

Here, the m¯atum is the undifferentiated unit of international diplomacy. It is interesting to observe that while Yahdun-Lim and Zimri-Lim eschew the ˘ definition of their land by its capital, Atamrum names it the m¯at Mari from an outsider’s view: (The enemy has made raids in my land); so I myself shall instantly travel 40 doublehours overland! I can even go out of my way as far as the land of Mari! (No!) This is not how I will take care of the land.15

Elsewhere, a Mari official quotes his own words to Gaˇsera, the queen mother of Yamhad, offering her a town “in Mari,” seeming to adopt this foreign ˘ perspective for his dealings with the court at Aleppo.16

2. The m¯atum of Agriculture, Outside Fortified Towns For internal purposes, when the domain of political governance is not in view, the m¯atum is the population and territory outside the fortified urban

118


centers. In time of war, this is the region through which an army may move with relative freedom until it attempts the siege of a major city or encounters massed troops. ARM II 24+ appears to generalize, as a servant of Zimri-Lim spars with Hammurabi of Babylon over relations with the major kingdom of Eˇsnunna.˘ If Eˇsnunna breaks the projected peace, “will you besiege city or raid countryside?”17 In a report by a high Mari official named Yasim-el to Zimri-Lim, the libbi m¯atim (“midst of the land”) represents the focus of enemy activity until an individual city is besieged: When Iˇsme-Dagan went out into the countryside (libbi m¯atim), he laid siege to the town of Atmû. He set up an assault tower and began piling up earthworks. Because of this news, Asqur-Addu has delayed going to my lord, saying, “The enemy is right next to my land. I am afraid that when I myself go to my father, the enemy will find out and violate my countryside (libbi m¯at¯ıya). Until my father makes a foray into the countryside, I can go nowhere.”18

Laessøe and Jacobsen propose a similar interpretation for “the land (of) ˇ sarrâ” as “the Suˇ ˇ sarrâ countryside” in a Shemshara letter from the time Suˇ of king Samsi-Addu.19 It appears that the m¯atum excludes not simply the capital city but all fortified centers. The m¯atum may be gathered into the fortified towns (dann¯atum) for safety from marauding armies,20 and during the reign of Samsi-Addu we find mention of omens taken for the m¯atum opposed to those for the fortified town (¯al dannatim): “Zunan took omens regarding the well-being of both countryside and fortified town.”21 This general division does not exclude the identification of whole peoples more narrowly by the governing city alone, along with its m¯atum, where these bear separate names. A letter from Inbatum reassures the king that “the land of Yamutbal is your land, and Andarig is your city.”22 Such dual definitions serve not internal purposes but the land as it interacts with its neighbors. A range of agricultural activities is associated with the m¯atum. It is the location of fields and houses,23 especially of grain: In the countryside, the Hana are taking their fill of barley.24

˘

Hammurabi (of Babylon) has invaded the land of Eˇsnunna, and he has burnt its ˘ barley. He has harassed(?) the land and brought back its loot.25

In northern Mesopotamia, the inhabited m¯atum is evidently more level than the rugged ˇsadûm, “mountain(s),” to which it is contrasted.26 In her lengthy treatment of the royal rites from Ebla published as ARET XI, Anne Porter observes that the king and queen must travel outside the city to sit on “the thrones of their fathers” (2000, 242–6). This primary ritual location is named Nenaˇs (or Binaˇs), and offerings are made there to dead kings at the “house of the land” (é ma-d´ım/tim), a title that must be linked to its location outside the central site of Ebla. Porter emphasizes the fact that


119

the ultimate ritual destination is not a temple but a funerary structure, and the royal couple reaffirm their right to rule by visiting this external site. We give priority to the city, especially given its impressive textual evidence for centralized administration (e.g., Archi 1993, 469), but the Ebla ritual makes the countryside the location of royal legitimation, reflecting the abiding or at least the remembered power of the rural territories ruled by kings (Porter, p. 255). A number of ancient Near Eastern ritual traditions share a similar movement outside the city walls to some sort of nearby shrine that seems to represent the deity’s deep connection with the land beyond his or her urban temple. At the same time, then, the people admit a profound dependence on this rural territory (Fleming 2000, 133–40). In Mari terms, this territory was the m¯atum, which appears to find its original point of reference in the rural population beyond the immediate circle of agricultural fields surrounding the central city.

3. The Ruler and the Ruled The m¯atum is often very simply that which a king rules.27 In this sense, it is not at all the state, but its complement, the collective voice of the ´ governed, all that stands outside the royal state. So far as the French “l’Etat” represented the nation as such and not just its government, this ancient point of view offers a noticeable contrast. Even the most ambitious of kings would never say of his hypothetical “nation,” “c’est moi.” We should keep in mind that we are talking about Akkadian terminology in wide Mesopotamian use, not a western or “Amorrite” category, even if many of the letters that use it come from correspondents with strong tribal ties or other western and West Semitic associations. The West Semitic counterpart of the m¯atum appears to be the namlak¯atum, or “kingdom” (literally, plural “dominions”), a noun derived from the same root as the common western word for “king” (*malkum). It is not clear what range of meaning the West Semitic namlak¯atum may have had, but actual use in the Akkadian documents found at Mari suggests an application that is not simply synonymous with the m¯atum. When Samsi-Addu dedicated a throne to Itur-Mer, the god of Mari, he recalls how the god entrusted to him “the land (m¯atum) of Mari, the Banks-of-theEuphrates, and his (Itur-Mer’s, or Mari’s) dominions.”28 These terms seem to define three distinct parts of the Mari conquest, with “his namlak¯atum” being outside the Ah Purattim (“Banks-of-the-Euphrates”) but still belong˘ Notice that the namlak¯atum ends up being claimed ing to the Mari center. by the great king Samsi-Addu, not his son Yasmah-Addu, who becomes the ˘ local Mari ruler. A newly published letter to Zimri-Lim from Dariˇs-libur, his envoy to the court of Yamhad at Aleppo, likewise puts the m¯atum and the namlak¯atum side by side, as˘if separate categories. Dariˇs-libur quotes YarimLim, the king of Yamhad, who promises to support the Simalite king in his ˘

120


pursuit of leaders who have opposed his rule: From now on, (whether) this year, next year, or ten years (from now), if these men enter my land (m¯atum) or my dominions (namlak¯atum) again, I will arrest them and have them brought to Zimri-Lim.29

In both of these texts, the namlak¯atum seems to represent a wider circle of people ruled by the king, outside the m¯atum. This might be confirmed by another letter that applies the word namlak¯atum to the vassal towns of IdaMaras. that accept the rule of Zimri-Lim. King Ibal-Addu of Aˇslakkâ reassures Zimri-Lim about another vassal: “Like his town of Nahur and like the (other) ˘ towns of his dominions (namlak¯atum), the town of Burundu is Zimri-Lim’s 30 town and Adal-ˇsenni is his son.” The reference to the towns that form the primary units of a kingdom could also have suited the Akkadian term m¯atum. Another Ida-Maras. ruler, Yawi-el of Talhayum, speaks of “the dominions of my lord” (namlak¯at b¯ıl¯aya) in the same˘ breath as Aˇslakkâ and Qarhadum ˘ (=Qirdahat), two other Ida-Maras. vassals of Zimri-Lim.31 ˘ Durand translates namlak¯atum as “kingdom” (royaume) and says that the word occurs frequently in the Mari archives, although I have not encountered many examples.32 Based on the texts cited here, any allusion to the West Semitic notion of the “king” as malkum (vs. Akkadian ˇsarrum) would apply only to suzerains who could claim the support of external vassals.33 This pattern would also fit the reference to the same term in an oath found at Tell Leilan, directed to “the king (ˇsarrum) of the land (m¯atum) of Apum, his sons, his servants, his army (s.a¯ bum), his pastoralists (nawûm), and his dominions (namlak¯atum).”34 I have translated the namlak¯atum as plural “dominions,” though Charpin points out that one unpublished text treats it as a singular noun.35 The focus on the submitted peoples at the periphery of a large kingdom would suit the plural, and perhaps the singular form is a secondary development. Although the m¯atum is most often defined by a dominant center and may also acquire an independent geographical or tribal identity, we find frequent references to the m¯at Hammurabi,36 the m¯at Zimri-Lim,37 the m¯at ˘ rulers and their entourages, it is natural Atamrim,38 and so on.39 Among to speak of “my land” or “his land.”40 The gods bestow on rulers not a city but a m¯atum. Addu of Aleppo claims in a prophet’s voice to have given Yahdun-Lim and Samsi-Addu their kingdoms: ˘ Abiya, the prophet (¯apilum) of Addu, Lord of Aleppo, came to me and spoke as follows: “Thus says Addu: I gave all the land to Yahdun-Lim, and by means of my ˘ weapons, he had no rival. He abandoned me, (however,) and I then gave the land which I had given him to Samsi-Addu.”41

The king of Yamhad says that the storm god has his own m¯atum, pre˘ underlying the human monarch’s realm.42 In a text sumably the reality mentioned earlier, Samsi-Addu in turn displays an ideological imperialism


121

when he credits Mari’s own god Itur-Mer in a throne dedication inscription with giving him rule of the m¯at Mari and the Ah Purattim: “At the ˘ time Itur-Mer heard my prayer and my petition, and granted me fully the land of Mari, the Banks-of-the-Euphrates (Ah Purattim), and his dominions ˘ (namlak¯atum).”43 In the eyes of the leaders, this m¯atum is a delicate and dangerous partner with a will of its own, whether as inhabitants of the immediate domain or as a vassal people. Rulers may issue it orders,44 but they also fear its rebellion,45 and they are constantly taking its emotional pulse. Is it troubled or anxious46 or calm and content?47 Anne Porter (2000, 260–1) analyzes the situation at third-millennium Ebla in terms of dialectics, based on the interplay of the elite authority and its supporting community. Out of this larger interplay come two specific dialectics expressing types of power (individual vs. community) and location of power (city vs. surroundings). As the primary Akkadian word for “kingdom,” though not reserved for kings only, m¯atum expresses all of these dialectics together, with focus on the ruled rather than the ruler.48 In keeping with its animate character, the m¯atum has a voice: “When all the land of Zalmaqum [sees(??)] what becomes of him, they will say. . . . ”49 A letter reports that king Iˇsme-Dagan spoke to his m¯atum and it answered him.50 In a formal meeting, the m¯atum pressures the king of Kurdâ to reject vassal or inferior status toward Zimri-Lim and to claim instead the independent position of a “brother.”51 It may be hoped that the voice of the land is submissive and loyal. Eˇsnunna is supposed to ratify the kingship of S.illi-Sˆın in saying, “may he take charge of our land.”52 Notice, by the way, that the m¯atum can also be identified with a queen, so that the leading wife of a ruler is attributed a parallel relationship to the land governed. Zimri-Lim’s daughter Inib-ˇsarri married king Ibal-Addu of Aˇslakkâ, where her husband can speak of “her throne and her land.”53

4. Naming the m¯atum Most often, as I remarked above, the m¯atum appears to be named for its central a¯ lum settlement. In some cases, it is not possible to distinguish town from land, but there are enough clear examples to demonstrate the pattern. Most of the major players of the region are rendered this way at least some of the time.54 The first six listed in Table 3, like Mari itself, represent large kingdoms with subordinate vassals and numerous fortified cities.55 Several important kingdoms that do have a central royal capital may nevertheless be identified by a second name when addressed as m¯atums. These are found especially in northern and western Mesopotamia, in areas dominated by tribal populations. Ekallatum, Babylon, Eˇsnunna, and Larsa do not show this characteristic. Consider the list in Table 4. At least Yamutbal


122

table 3. m¯atums Defined by Central Capital m¯at Ekallatim m¯at Babilim m¯at Eˇsnunna m¯at Larsa m¯at Qat.anim m¯at Halab ˘ m¯at Andarig m¯at Kurdâ

ARM II 18:4; XXVI 425:8; XXVII 145:25; A.1289+ ii:18 ARM II 25:15 ARM X 155:9; XXVI 376:10; XXVII 141:17; 143:7; A.3669+:22 ARM XIII 27:15; 47:14; XXVI 385:41 ; XXVII 161:8 ARM II 66:16; V 23:12; XIII 46:14 ARM V 63:12a ARM I 132:6; IV 31:9; XXVI 430:33 ARM II 23:11

a This

alternative to the m¯at Yamhad is unusual and seems to reflect an outsider’s perspective, ˘ as with the m¯at Mari under Samsi-Addu just mentioned above. The sender of this letter is Sˆın-teri, a high official who served Samsi-Addu in the upper Balih region; see Charpin and ˘ Durand 1986, 180–2, and Villard 1990, 567 and n. 17.

table 4. m¯atums Defined by Population m¯at Yamhad

capital, Halab/Aleppo ˘ IV 6:6; XXVI 365-bis:3a ARM I 6:11; ˇ ˇ hna capital, Subat-Enlil/ Se ˘ XIV 102:19; 125:16; ARM II 49:4 ; X 122+:11; XXVI 358:18 ; A.1610+:9–10; A.1421:43 and M.15083b

˘

m¯at Apim

m¯at Yamutbalim

capital, Andarig ARM X 84:24; XXVI 383:7; 432:8 ; XXVIII 172:8 –9 Note a second indirect link in the earlier Samsi-Addu period, as the m¯at Razamâ Yamutbalim, II 18:7.c

m¯at Numhâ/Numahˆım

capital, Kurdâ ARM XXVI 358:9 ; 521:11; A.3209d Note a second link with Karanâ in XXVI 412:12.

m¯at Yapturim

capital, Talhayûm ˘ ARM XIII 144:4; cf. I 19+:11

˘

a Even

˘

this identification is not entirely native, since the local Amorrite equivalent to m¯atum appears to be dadmum; see Durand 1989, 29–30. It is worth noting that the other major western power, Qatna, does not have such an alternate name. b The last two are mentioned in Charpin 1987b, 135–6, nn. 35 and 38. Note also a seal of king ˇ hna, as “king of the m¯at Apim” (E4.27.4.1, in RIME 4); cf. Yakun-aˇsar from the Mutiya of Se ˘ time of Samsu-iluna (E4.27.5.1). c ARM XXVIII 172:8 –9 mentions both “the town (URUki ) of Andarig” and “the land of Yamutbalum,” with the first to be “guarded” (verb nas.a¯ rum) and the second “kept safe” (verb ˇsullumum). d Joann` es 1985, 109, a memorandum.


123

and Numhâ are well-known tribes, and in these cases the m¯atums derive ˘ from the dominant population.56 It is natural that the actual their names tribal people are not confined to one kingdom, so that the overlap of use is not surprising. All the m¯atum names are arguably Semitic, three of them with the western verbal prefix Ya-, so the tribal character of Yamutbal and Numhâ may apply to the whole set. In the earlier Yahdun-Lim inscription ˘ Samaˇ ˘ ˇ for the s temple, the three Yaminite states are defined by central towns and a m¯at+tribe (Amnanum, Rabbˆ um, and Uprapˆ um, iii 4–9).57 ARM II 18 mentions the m¯at Yahrurâ, another Yaminite land, this time defined from the view of Iˇsme-Dagan,˘ the older son of Samsi-Addu. To my knowledge, there is never a m¯at Yamina as such, though A.3960 says that “all the Yaminites are on the roam, having left their towns and their land,” to threaten Imar and Tuttul.58 We should notice that when the Yaminites are understood to occupy a m¯atum, that “land” is understood to be defined in some relation to its settlements. In the Mari evidence, the m¯atum category may often be regarded as a tribe, a striking feature of texts that are oriented toward kings and kingdoms in settled seats of power. As political players, however, these tribal peoples seem to require a relationship to settled centers in order to qualify as m¯atums. The geography of the kingdoms with separate m¯atum names should give pause. Mari and the Yaminite lands are in the Euphrates valley. Apum and Yapturum are in the upper Habur basin, and Yamutbalum and Numhâ are ˘ in the Sinjar just east of the ˘Habur. Yamhad, of course, is in northwestern ˘ ˘ Syria, outside Mesopotamia proper. None of these is found in southern Mesopotamia or in the valley of the Tigris River, the eventual centers of Babylonia and Assyria. All of them, however, are in the regions frequented by the Yaminite and Simalite tribal peoples, as charted by Durand in the map for his address to the 46th RAI in Paris (forthcoming). These peoples have bases in the Euphrates valley as far downstream as Mari and even Hˆıt, ˘ and in the northern country, in the Habur basin and eastward to the Sinjar, ˘ but not along the Tigris itself nor even near it once south of Ekallatum, and not in southern Mesopotamia. Though Samsi-Addu of northern Mesopotamia and Hammurabi of ˘ and with the Babylon recall tribal links with the Numhâ in the first case, ˘ Yaminite Amnanˆ u and Yahrurˆ u in the second, both of which belong to this ˘ same set of tribal homelands, neither king identifies his own realm in this way.59 This choice suggests a greater distance from at least the mobile pastoralist traditions of their tribal heritages than what the names indicate for the more western kingdoms. There may also be a geographical aspect to this distance, reflecting the stature of the eastern cultural and political forms. With movement into the ancient urban centers of southern and eastern Mesopotamia, ambitious tribal rulers preferred to claim the mantle of the local traditions, and in so doing, appear to have allowed their tribal identities to recede somewhat into the background.

124


This process may even be visible in the titularies of early secondmillennium kings at Larsa and Uruk in old Sumer, where the rulers of newly victorious Amorrite dynasties celebrate their tribal identities in a way that is soon eclipsed by local epithets. At Larsa, Warad-Sˆın remembers his father, Kudur-mabuk, as “the father of the Emutbal” and “the father of the Amorrite (land),” though he does not claim such titles for himself. When Warad-Sˆın’s brother Rim-Sˆın succeeds him, he soon drops the mention of his Emutbal father.60 At Uruk, Sˆın-gamil keeps the title of his father, Sˆınkaˇsid, who founded a new dynasty there, both calling themselves “the king of the Amnanum,” the Yaminite tribe.61 The difference may be that Rim-Sˆın of Larsa could consider himself the ruler of a major regional kingdom, heir to the earlier Sumerian realms, while Uruk could stake no such claim. Mari represents a special case that must be read against the above patterns. During each of the three principal reigns of the Amorrite period, the terminology shifts. In the titulary of Yahdun-Lim, this Amorrite kingdom is ˘ with “Mari” the royal seat and the treated much like its Yaminite brethren, “m¯at Hana” (equal to “the Simalite(s),” in one seal of his sister) taking a tribal ˘definition.62 Samsi-Addu has grander ambitions and evidently more distant links to his tribal origins. He establishes a literal “Mesopotamian” kingdom, “the land between the Tigris and the Euphrates” or “the land of the Tigris [and] the Euphrates.”63 Yasmah-Addu governs the western or Euphrates component, identified at least ˘once as the m¯at Mari u Ah Purattim (“the land of Mari and the Banks-of-the-Euphrates,” A.2231, above).˘ When Zimri-Lim returns Mari to its place as tribal capital, he reclaims his kinsman’s titulary, and again, a tribal name is chosen to define the m¯atum, rather than the royal seat.64

5. The m¯atum Alliance Most often, m¯atum is the Akkadian term that encapsulates the entity unified under a king or somehow organized to represent a coherent people. It is somewhat disconcerting, then, to find a small number of these “lands” that have no single ruler but exist only as coalitions.65 There are only three: ˇ Zalmaqum, Ida-Maras., and Subartum, all of which appear repeatedly in the Mari correspondence (see Table 5).66 The character of each m¯atum is different, but all share one other trait: in texts from the reign of Zimri-Lim, they are defined also as an alliance of kings. A similar list can be made for ˇ examples of the phrase, “the kings of Zalmaqum/Ida-Maras./Subartum,” seeming to exclude texts from the kingdom of Samsi-Addu (Table 6). During the Zimri-Lim period only, the Mari leadership recognized grouped “kings” who could deal collectively with outside powers.67 The very correlation of the word “m¯atum” with this reference to kings underlines ˇ the political nature of the term. Zalmaqum, Ida-Maras., and Subartum form m¯atums only in order to act in concert in matters external to the coalition


125

table 5. References to m¯atum Coalitions Zalmaqum

ARM I 10:11, 13, 19, 14 ; 53+:29; (II 68:1 , cf. 3 ); XIII 46:15 ; 146:21; XIV 76:20; XXVI 24:13 (cf. 44); XXVIII 60:28; 79:12; no. 72–39+; A.3024:28a

Ida-Maras.

ARM II 21:21; 130:7, 14, 38; VI 66:7; X 34:24; XIV 51:6, 11, 32; 359:5; XXVII 20:15; 89:8; 132:11; XXVIII 40:4 ; cf. 48:26–7 (m¯at restored); 55:11 -12 ; 56:4; 148:9

ˇ Subartum

ARM I 18:26; VI 27:17 ; XIII 117:3 (cf. 9 ); XIV 112:29; XXVI 365-bis:4; 384:20 , 61 ; XXVII 26:17; 45:6; 80:44; 147:9; 151:86; 162:39; A.2119:12; A.3354+:49b

a b

The last two texts appear in Birot 1985, 129–30; and Finet 1966, =LAPO 16 no. 302, pp. 473–4. The last two texts appear in Charpin 1992b, 98 and 101n21.

table 6. References to the Collective Kings of m¯atum Coalitions Zalmaqum

ARM II 68:3 ; XIII 46:15 (kings of the m¯at Z.); XXVI 12:4 ; 24:10, 15; XXVIII 15:9–10; 62:40; A.215:8; B.590:25–6; A.2526:10a

Ida-Maras.

ARM II 35:23; XXVI 347:22, 25–26; 352:13; XXVII 20:15–16 (kings of the m¯at I.-M.); XXVIII 93:15 ; FM II 123:27 ; A.2119:26; A.1610+:4b

ˇ Subartum

ˇ ARM III 37:7 (adjective Subarˆ u); XIII 117:3 (kings of the ˇ 9 ; XIV 112:24 (Subarˆ ˇ m¯at S.), u); XXVI 308:15–16; 309:13–14; ˇ 384:20 , 61 (of the m¯at S.)

a For

the last three texts, see AEM I/1, p. 183; and Finet 1966, 24–6. the last two texts, see Charpin 1992b, 98; Durand 1988, 109–10. Indirect references to the kings of this coalition may be found in ARM XXVI 303, 306, 325, and 329; XXVIII 78:10–11; A.3206 in Charpin 1993b, 176.

b For

peoples. The m¯atums are not defined by geographical, language, or ethnic boundaries, whatever patterns may have existed. In a period noted for fragmentation into many small polities, some of these allied themselves to face the larger powers surrounding them. Together, these coalitions could be treated as equivalent to larger kingdoms, presumably with an attendant increase in respect. Zimri-Lim is informed in one case that the king of Carchemish has sent messengers to the principal players of western Syria in order to raise troops: Yamhad, Qatna, and the m¯at Zalmaqim.68 A gover˘ nor of Qat.t.unan recalls a reference by his lord Zimri-Lim to “the news of ˇ the land of Subartum, the news of Hammurabi the Kurdâ-ite, and the news of Atamrum, about which my lord ˘wrote to me,” where the alliance stands parallel to the individual kings of Kurdâ and Andarig.69

126


Each of the three coalitions has a somewhat different character. Zalmaqum and Ida-Maras. had at least loose affiliations with the Yaminite and Simalite tribal confederacies, respectively, connections that brought with them various attendant relations with Zimri-Lim’s Mari. Ida-Maras. overlapped the Simalite grazing lands, and its peoples were always potential allies with Zimri-Lim, but they could equally turn against him as rivals. One letter from a local Ida-Maras. ruler recalls the abiding bond between the (Simalite) “Hana” people and Ida-Maras. (FM VI 6:10–11). The Ida-Maras. coalition had˘a somewhat fluid constituency, always comprising several more than the four stable city centers of Zalmaqum.70 ARM XXVIII, the collected correspondence between Zimri-Lim and other kings, includes a majority of letters from individual kings in the Ida-Maras. coalition. They never send to Zimri-Lim as a group, and they reveal themselves to be a fractious crew. In one missive, Ibal-Addu of Aˇslakkâ speaks of “when I made peace with IdaMaras.,” recognizing the claim of the confederacy on the participants who keep their distinct political identities.71 Later in the same letter, the king distinguishes “the land of Ida-Maras.” from “the upper land,” as if both were political entities. Ibal-Addu says that he can lead both to reject any alliance with the Elamites, though it is not clear what “the upper land” represents, within the Habur basin.72 Two of the years in Zimri-Lim’s reign are named for victories˘ over Aˇslakkâ, perhaps the most powerful Ida-Maras. center.73 The rivalry between the local powers of the Habur region takes an inter˘ ˇ esting twist when Ibal-Addu accuses Subram, king of neighboring Susâ, of laying claim to the sole kingship of the Ida-Maras. political entity.74 There is no evidence that Zimri-Lim ever allowed Ida-Maras. to serve any one king other than himself, however. It is possible that the Ida-Maras. coalition functioned on fairly local terms. Samsi-Addu never seems to have recognized or negotiated with such an alliance, perhaps in part because he staked direct claim over the upper ˇ ˇ hna, Habur region by locating his personal royal seat of Subat-Enlil at Se ˘ ˘ along one of the Habur’s tributaries.75 It appears that these town centers ˘ were subsumed directly into Samsi-Addu’s kingdom, without being allowed the status of “lands” (m¯atum) with “kings” (ˇsarrum) on any terms. Our entire documentation for the m¯at Ida-Maras. comes from the period of Zimriˇ Lim, unlike that for Zalmaqum and Subartum. One letter (A.1098) coins a completely different terminology for the time when Yahdun-Lim offered ˘ nor the “kings.” gifts to “the fathers of Ida-Maras.,” using neither the “land” “Previously, when Yahdun-Lim would go to that land, he would present gifts ˘ 76 to the fathers of Ida-Maras . .” Remember that during the time of Yahdun˘ be Lim, his own Simalite Hana are said to be led by “fathers” who can also ˘ 77 called “kings.” Durand’s suggestion that kingship had to be approved by the suzerain seems to apply specifically to Ida-Maras. leaders under Simalite Mari rule, and it is not clear to me from the published evidence whether the practice extended further afield (1997, 207, 467).


127

There is some evidence for the continued viability of an Ida-Maras. coalition during the period after our Mari archives. Samsu-iluna, Hammurabi’s ˘ later king successor at Babylon, claimed to rule a m¯at Ida-Maras., and the Ammi-s.aduqa mentions the “sons” of the Numhâ, the Emutbal, and IdaMaras..78 A text from Eˇsnunna, apparently from˘ a period earlier than the Mari archives, describes a treaty between “the Akkadians, the Yamutbal, the Numhâ, and Ida-Maras..”79 The “Akkadians” represent Eˇsnunna itself, while ˘ the other three participants define northern peoples spread across the region near the Jebel Sinjar and the northeastern Habur basin. It is interesting to notice how these peoples are defined, though˘the word “m¯atum” does not appear. As political entities capable of entering into a treaty, each of these would qualify as a m¯atum, according to the usage found in the Mari archives. “Akkadian” Eˇsnunna and Ida-Maras. are known to have been m¯atums during the later period. The other two polities are defined by their tribal names as “Yamutbal” and “Numhâ,” not associated with the towns of Andarig and ˘ Kurdâ, as in the letters found at Mari. The location of the capitals seems to be less essential to defining the political power than the tribal names. Zalmaqum appears in the Mari archives as a fixed alliance of four cities – ˇ a, and Hanzat – though the picture of stability could Harran, Nihriya, Sudˆ ˘ ˘ reflect the ˘ignorance of greater distance and less intimate contacts. The political function of the coalition is displayed in a letter to Zimri-Lim from Amut-pi-el, king of Qatna, who reports that Yarim-Lim of Aleppo has communicated the fact that he has made peace with “the kings of Zalmaqum.”80 ˇ The specific internal workings of the Subartum coalition seem even less ˇ well known to the Mari kings, though the references to Subartum are fairly frequent. Piotr Steinkeller has recently argued that the name “Subartu” ˇ (=Subartum) had two separate applications that go back to the late third millennium. In its broader meaning, Subartu described a vast territory that extended between the upper Tigris and the Balih Rivers, encompassing all ˘ of the Habur River basin. At the same time, however, “Subartu” designated ˘ a specific area to the east of the Tigris River, north of the Diyala and east as far as the Zagros mountains, the rough equivalent of the later Assyrian heartland. Both meanings survived into the early second millennium, when texts from Babylon and Eˇsnunna used the narrow meaning, while Mari evidence supposedly applies it to the Habur region. The broad meaning is ˘ more easily derived from the narrow than the other way around, especially after the appearance of the Hurrians around 2200 b.c.e., quickly spreading through the Habur basin by the end of the third millennium (Steinkeller 1998, 76–8).81˘ In fact, there is no reason to separate the Mari evidence from that of Babylon and Eˇsnunna. The Mari map of the Habur is quite well documented, ˘ alliance. A location east of the ˇ and there is no place there for the Subartum Tigris would suit the evidence quite well. In the early second-millennium ˇ Mari texts, the definition of Subartum as a m¯atum of plural kings indicates a

128


concrete political-geographical entity. Steinkeller observes that this area was inhabited by Hurrians during the early second millennium, a pattern that ˇ makes it likely that the “Subarˆ u” language was Hurrian.82 The pattern of geographical names in the evidence from Ebla and Beydar suggests that in the mid-third millennium (ca. 2400), both of the regions that came ˇ to be identified as “Subartu (Subartum)” had once been Semitic-speaking (Steinkeller 1998, 88–9). In a recent publication of new Mari letters, Michaël Guichard proposes ˇ that “Subartum” can refer to the entire region encompassing the Habur and ˘ Balih River basins (2002). Zimri-Lim has commanded that “the kings of the ˘ ˇ land of Subartum” be given provisions, and his correspondent replies that he has already provided sheep for the kings of Ida-Maras., and of Zalmaqum and for an otherwise unknown entity called “the kings of Adamˆ u” (FM VI 7:6–10). It is not necessary that these three groups be subsumed under the ˇ single category of Zimri-Lim’s reference to Subartum. They may only vaunt the writer’s initiative in already providing for all of the other such coalitions. The specifically political use of the word “m¯atum” makes it extremely unlikely that Ida-Maras. and Zalmaqum would be incorporated as distinct “lands” ˇ within a larger “land” of Subartum.

6. The Sedentary m¯atum In geographical terms, the m¯atum constitutes the people and land outside the major fortified centers, especially the seat of government. At the same time, it is essentially agricultural and a region of settled population. Tribal people on the move may be conceived to move from one m¯atum to another, and in the short or long term, one component may “reside” (waˇsa¯ bum) in this settled domain. Insofar as they occupy a m¯atum, the word incorporates them into a world of definitions by fixed abode. ARM IV 24 explains the movements of Turukkû peoples during the reign of Samsi-Addu. They now reside in Tigunanum, but previously, because of famine, they had gone to Hirbazanum.83 Iˇsme-Dagan promises to let his ˘ next (lines 28–31). In another letter, Samsibrother know where they go Addu tells his son Yasmah-Addu that some Rabbˆ u tribespeople have com˘ municated to him their plans to cross from Yamhad into land held by the ˘ who reside in the land Upper Mesopotamian kingdom: “The Rabbˆ u (tribe) of Yamhad have written to me as follows: ‘We have set our face to cross (the ˘ [there are(?)] no boats for [the crossing(?)].’”84 Both of these river), but texts come from the reign of Samsi-Addu and may identify tribal peoples more strongly with fixed m¯atums than do the officials of Zimri-Lim.85

7. Akkadian m¯atum and Sumerian Terminology The Akkadian word “m¯atum” is already well attested in the mid-thirdmillennium texts from Ebla.86 There is no reason to understand the word


129

“m¯atum” as a loan from the Sumerian language, and Sumerian culture seems not to have had any exact equivalent to this political category. However the Sumerian language adapted to the emergence of complex polities beyond the bounds of individual settlements (uru), it did so in terms that were entirely different from those defined by the m¯atum. Eventually, the Akkadian m¯atum came to be identified with more than one Sumerian term, thus producing a linguistic tension that was relaxed only by the eventual demise of distinct Sumerian-speaking culture in the early second millennium. The distinct Semitic heritage of the m¯atum comes into sharper relief when we examine it against the backdrop of the Sumerian words that covered overlapping but different fields of meaning. Whereas Akkadian denoted every political “land” as a m¯atum, early Sumerian distinguished “our land” as kalam from “their land” as kur.87 The word “kalam” seems originally to have been regarded as a synonym of u˜ ´ g, “people,” with which it shared a cuneiform sign. This need not have been a political category in its earlier use, though at some point it took on the nuance of “us,” versus “them,” the people of the Sumerian low country versus everyone else (Steiner 1982, 636–7). People from lands outside the southern river plains were often identified by a completely unrelated term, “kur,” which referred to the “lands” of outsiders, but the primary meaning of which seems to have been the “mountain” terrain that was foreign to the Sumerian speakers. It is often said that by the ED (Early Dynastic) III period, in the mid-third millennium, the word “kalam” had come to be applied to the entire southern alluvial plain.88 As an expression of the whole population of southern Mesopotamia, the meaning of “kalam” is intertwined with other notions of this nascent unity. Even the word “ki-en-gi,” rendered by later Akkadian as “Sumer,” appears with its inclusive sense only in the later part of the ED III period.89 At about the same time, the god Enlil is first found at the head of a common Sumerian pantheon, a position that is soon associated with the town of Nippur. What is going on here? Can these phenomena be related? There is at least a possibility that they are, as parallel ideological expressions of new efforts toward the political unification of Sumer, during the last phases of the Early Dynastic period. It appears that the identification of the word “kalam” with all of southern Mesopotamia was contemporary with the appearance of the title “king of the land” (lugal kalam-ma(k)), especially in the vase inscription of Lugalzagesi, king of Umma.90 Famous as is the identification of southern Mesopotamia as “Sumer,” actually Kengir (ki-en-gi), I am not aware of this name’s clear association with all of southern Mesopotamia until the same period and setting. In Cooper’s collection of Presargonic royal inscriptions, he suggests two possible restorations of the name “Sumer” (Kengir) in the earlier inscriptions of Eanatum, neither with any certainty.91 The only definite references to “Sumer” appear in two inscriptions of Lugalzagesi of Umma and of Enˇsakuˇsana of Uruk.92 In the

130


Lugalzagesi vase inscription, kingship of this kalam is said to be bestowed by Enlil, who receives worship at Nippur. The identification of all southern Mesopotamia as Kengir seems to derive from a more modest regional asˇ sociation, judging by texts from slightly earlier Suruppak (Fara), which list ˇ workers from Uruk, Adab, Nippur, Lagaˇs, Suruppak, and Umma as from Kengir. Ur, Eridu, and Larsa are omitted (Steible and Yildiz 1993, 25–6). Although each of these ideologies had a prehistory in more limited contexts, their application to a cultural whole called Sumer (Kengir) is linked to the notion of a political unity, whether or not it was achieved by those who invoked it. As “the land” (kalam), as Kengir, or as the domain governed by the divine king Enlil, this ideology of unity is in large part political. Jacobsen connected the notion of Sumer and the Nippur religious center to a “Kengir league,” a political entity rooted not in kingship but in early coalition leadership, probably going back to the ED I period (1957, 106–7). Postgate advocates Jacobsen’s interpretation and adds the term “kalam” to the mix, all as part of a political unity grounded in the Nippur-centered league of cities (1994, 4–5). It is not clear to me that the idea of one “land,” one Sumer, under Enlil must go back as far as Jacobsen proposed, and the early Nippur league remains a conjecture.93 The first attribution of Enlil’s rule to the town of Nippur appears in an inscription of Enmetena of Lagash (ca. twenty-fifth century), not long before Lugalzagesi.94 Walter Sallaberger has recently concluded that Nippur’s symbolic role followed the rise of Enlil to his dominant position in a unified Sumerian pantheon, rather than that Nippur began as a political center. Old Kengir (Sumer) was considered to have its center at Uruk (Sallaberger 1997; cf. Klein 1998–2001). In the end, one wonders whether the whole idea of a unified Sumer did not come from any early league but arose instead with the individual leadership of imperially minded kings. Whether the king is Enlil or one of the ambitious rulers of the last Presargonic age, the unified Sumerian kalam may turn out to be one expression of a political shift from states based in single cities to multicity states. This entire history behind kalam as the unified “land” of Sumer reflects a logic that is completely unlike that of the Akkadian m¯atum, which applies equally to insiders and outsiders, with no tie to a larger linguistic or cultural identity. By the time of the late third-millennium kingdom of Ur, the Akkadian word had acquired a Sumerian equivalent by lending the term in the form ma-da.95 During the Ur III and Isin periods, the phrase “a-ˇsa` ma-da” describes a “field in the countryside” of a given town. Other uses of ma-da apply the term specifically to the area and population surrounding a central town, a use very similar to the definition of m¯atum in the Mari archives for internal affairs.96 We find an interesting expression of the deep dissimilarity behind the Akkadian and the Sumerian political terminologies in the royal titulary of the earliest second millennium. The occasional bilingual


131

Sumerian-Akkadian inscriptions show that the Sumerian title “king of Sumer and Akkad” is translated into Akkadian as “king of the land of Sumer and Akkad.”97 Douglas Frayne then borrows the Akkadian translation for his English rendition of the common Sumerian phrase. The very fact that the Sumerian version used no equivalent for m¯atum, though ma-da or kalam were available, reflects the divergence between the two political traditions. This contrast shows up especially at the very beginning of the second millennium, and eventually the Sumerian kalam is simply inserted for the Akkadian m¯atum, with no remnant of the southern Mesopotamian notion of “our land.”98 If we return to the ubiquitous use of the word “m¯atum” to describe the large-scale political landscape of the Akkadian Mari letters, we find the distinct Semitic meaning alive and well, without noticeable influence from Sumerian political terminology. In spite of the scribal adaptation of the kur and the kalam word-signs to render the Akkadian m¯atum, it was the Semitic Akkadian idea of a people inhabiting a region distinct from any ruling center that remained essential. The durability of the Akkadian notion adds to the impression that the separate political traditions of lands upstream from southern Mesopotamia survived and thrived in the early second millennium, when the Amorrites gained influence in the east.

8. The m¯atum in Hittite Anatolia The word “m¯atum” continued to be used with roughly the same range of meaning through later periods of writing in Akkadian.99 Given that we eventually ask how broadly Near Eastern these notions may have been, especially in the direction of the Aegean, it is worth observing that the Hittite cuneiform tradition of ancient Turkey took it over almost completely during the second half of the second millennium. One could argue that foreign Mesopotamian categories were forced onto Anatolian social patterns that did not really fit them, but even if this were true, the categories seem to have become integral to Hittite expression in the royal capital of Hattuˇsa, where ˘ the huge mass of Hittite cuneiform texts was found.100 The relationship between Mesopotamian language categories (both Sumerian and Akkadian) and the Hittite language can perhaps be traced in part through the phenomenon of administrative “districts” that were subordinate to the larger Hittite kingdom. The Akkadogram HALS.U is so far ˘ of the fourattested only during the New Hittite period, from the middle teenth through the thirteenth century. A likely Hittite equivalent is known for the Middle Hittite period (mid-sixteenth to mid-fourteenth centuries), but this word has not been found in Old Hittite (ca. 1700–1550). One Middle Hittite text shows a hierarchy of political units with native Hittite terminology, from the “land” (KUR-e = utne) to the “district” (telipuri) to the “town” (URU-aˇs = happiraˇs).101 In wider Hittite use, the Sumerian URU-sign ˘

132


may represent both the native Hittite word and a determinative designating a place name as a town, in either case for a settlement of any size (Beckman 1999a, 167). Notice that the scribal categories for “land” and “town” are taken directly from Sumerian, though the Mesopotamian scribes who used them would have read KUR as m¯atu and URU as a¯ lu.102 During the empire of the fourteenth to thirteenth centuries, the HALS.U “district” was usually named for a central town, such as the district˘ of URU Atriya, in the same fashion known at Mari.103 “Lands” were also named most often for a central town, even to the point where the typical formula became “the land of town X” (KUR URU X). This was applied to foreign entities the actual political forms of which did not in the least fit the pattern, including “the land of the town of Egypt” (KUR URU Mis.ri) and “the land of the town of the Kaˇska (tribes).”104 The kingdom of Hatti was not simply identified with its central city of Hattuˇsa, however, but ˘ also be conceived as composed of “towns,” as in˘ the Mesopotamian could Mari evidence (Beckman 1999a, 168). In general, the “town” stands in contrast to what Beckman (162–3) renders as the “noncity” surroundings ¯ (gimra-/gimmara-), the more distant steppe (the Akkadogram S.ERU), or the mountains (HUR.SAG). It appears ˘that the Hittite formula for naming all large polities as if identified by a central town reflects the administrative and diplomatic perspective of the great Hittite empire, which managed its own people on these terms. The Hittites clearly dealt with polities of a wider variety, however, so that the category of the “land” (KUR/m¯atu/utne) was applied to the pastoralist Kaˇska, northeast of Hatti, and to the allied “kings of the land of Nuhaˇsˇse” ˘ use in northern Syria.105˘Both of these examples follow the same range of found in the Mari archives. In the Hittite texts, a “land” is also what a king rules, ultimately by divine appointment. One text declares that “the land belongs to the storm god; heaven and earth and the people belong to the storm god; he has made the Labarna, the king, his regent; to him he has given all the land of Hatti; let the Labarna govern the entire land with his ˘ hand.”106 By the end of the Bronze Age, the great Anatolian kingdom of Hatti had ˘ settled come to share the rough categories of political organization from centers that were typical of Mesopotamia. However deep these similarities may be, they cannot be regarded as narrowly Mesopotamian by this period, and they seem to have adapted to the Anatolian social landscape without difficulty. If we are to define a predemocratic Greek culture that is fundamentally unlike “Near Eastern” culture, it may be necessary to separate it from the Anatolian, as well. Even the collective terminologies of “elders” and generic townspeople have counterparts in the Hittite texts, and these local groups are harder to explain as impositions and innovations from the imperial center.107 The Anatolian evidence is important for judging the geographical scope of “Near Eastern” political culture.

Subdividing the Major m¯atums

133

c. subdividing the major m a¯ tum s: the hals. um district ˘ I have said that the political map of the world described by cuneiform scribes was divided into m¯atums, or “lands.” These were most often governed as individual kingdoms but could also be recognized as coalitions. The largest realms were called m¯atums, but many m¯atums occupied small territories and could not have had large populations. One substantial feature that could distinguish the large “lands” from their smaller neighbors was the administrative subdivision of the larger and more complex kingdoms into “districts” called hals.ums.108 These hals.ums were governed by servants of the king, with ˘ ˘ political and administrative the same conception as the kingdom center, as extensions of royal authority into more local affairs. Like the m¯atums under which they functioned, the hals.ums were defined in terms of regional populations that reached beyond˘ any individual town. Yahdun-Lim, Samsi-Addu, and Zimri-Lim all made use of the hals.um cat˘ and the egory˘ to define administrative subdivisions within their kingdoms, governments of the two latter kings have recently been treated systematically by Pierre Villard (2001) and Brigitte Lion (2001).109 I do not wish to duplicate their excellent work, and the details offered here address mainly the general question of how the hals.um “districts” both serve and reflect the ˘ this added hierarchical level, the larger notion of a m¯atum as kingdom. With m¯atums display another explicit level of administrative complexity and are more easily identified as archaic states. During the periods of our archives, Mari itself fit this category, though under Samsi-Addu, king Yasmah-Addu ˘ ruled Mari as the western “district” (hals.um) of an even more extensive ˘ domain. The hals.um was a political unit that was never conceived to be au˘ It served only the administrative needs of a m¯atum that was ruled tonomous. by a king in a system of individualized (“exclusionary”) power. There was no proper collective or “corporate” aspect to the hals.um “district.” ˘

1. In the Image of the m¯atum One might say that the hals.um was made in the image of the m¯atum. Its very ˘ structure imitates the larger unit in all respects. Like the m¯atum, the hals.um may be defined by either the town at its center ˘ or the person who governs it. Zimri-Lim’s well-known system of hals.um “dis˘ tricts” and their ˇsa¯ pit.um “governors” is perhaps less highly structured than it first appears. There are differences in status between a core of districts close to Mari and two more distant districts, although this distinction is never acknowledged in formal terms. The core consists of the districts administered from Mari, Terqa, and Saggaratum, which are often mentioned together.110 This language is also used of Qat.t.unan, the governors of which report to their sovereign with similar frequency, but the northern district along the

134


Habur is somehow isolated.111 A fifth region is even more distanced from ˘ others. Suhûm is the land along the Euphrates River downstream from the ˘ closer to southern civilization but in fact in close contact Mari, seemingly with steppe to the north between the Habur and the Tigris that was freum calls the Suhˆ um quented by mobile pastoralists.112 One ˘letter to Meptˆ ˘ “your district,” though this turn of phrase is reminiscent of the looser usage under Samsi-Addu.113 This correspondence more often identifies districts by governor alone: r halas Yantakum, ARM IV 44:6; cf. XXVI 268:7 . r ˘halas Sˆın-tiri, V 35:14114 . r ˘halas Saˇ . ˇ sˇsaranum, V 43:11 (at Apqum) r ˘halas Yanuh-samar, V 43:15 (at Sanduwatum) . ˘ r ˘halas Ikˇsud-appaˇ su u Habduma-Dagan, XXVI 87:13 (distributive) . ˘ ˘ The list of associations between hals.um and governor can be expanded con˘ when Kibri-Dagan, Zimri-Lim’s governor siderably by indirect reference, as of Terqa, repeatedly speaks of “my district.” Consider, for example, the stock reassurance: “I am not negligent toward my district and the need(s) of the palace.”115 Under king Samsi-Addu, a man named Tarim-ˇsakim closes a letter with assurance that “the district is well.”116 All of this language follows closely the pattern of kings and m¯atums. Just as the m¯atum is separate from the royal capital, the hals.um surrounds ˘ that both town its central town. Zimri-Lim’s governors are always affirming and district are well.117 With language that echoes the same m¯atum procedure, Bahdi-Lim urges his colleague at Saggaratum to “gather your hals.um ˘ into fortified towns (dann¯atum).”118 In the next line, the population˘ to be protected in these strongholds is said to live in settlements (kapr¯atum), evidently the essential components of the hals.um.119 A governor of Qat.t.unan ˘ into that city.120 Another echo reports that he has “gathered the district” of m¯atum structural assumptions is found in the idea that one “resides” (waˇsa¯ bum) in a hals.um.121 Qat.t.unan’s governor fears that not one man will ˘ remain in the district if a census is taken, that “the district” will thus be displaced (nas¯ahum).122 ˘ One more odd parallel is found in the use of relative definitions around the flow of rivers. Under Zimri-Lim, one may call Terqa “the upper district” (hals.um elûm), a terminology that might be inherited from ˘ 123 Samsi-Addu’s plural “upper districts” must be informed his predecessor. about Mari’s autumn Nergal festival, which should place them upstream from the city, perhaps roughly equivalent to the Terqa reference of ZimriLim.124 The situation is more complicated for the “upper country” (m¯atum el¯ıtum). Durand says that the primary reference is Zalmaqum, perhaps based on ARM XXVIII 60:26–8, where the two are closely affiliated, but the “upper country” is not a political unit, and in this text it stretches across a large swath


135

of the northern foothills.125 Zalmaqum is located in the Balih River basin, but other texts indicate more eastern points of reference. King˘ Hammurabi ˘ of Babylon writes to Zimri-Lim about the Elamite invasion of Mesopotamia by way of the “upper country,” which should be in the direction of the Habur ˘ pasbasin, starting from the east.126 Likewise, Ibal-pi-el, the Simalite chief of ture (merhûm), appears to place the northward incursions of “the man of Eˇsnunna”˘ in the upper Habur, where he locates the coming confrontation in Ida-Maras..127 Earlier,˘ under Samsi-Addu, the upper country seems to be found in the Tigris River drainage.128 Another letter from the period of ˇ Zimri-Lim offers the apposition m¯atum el¯ıtum m¯at Subartim, “the upper land – 129 ˇ the land of Subartum.” This would be east of the Tigris. In the archives of Shemshara, east of the Tigris and dated to Samsi-Addu’s rise and reign, the “upper land” may be even further east, in the Zagros Mountains.130 It appears that “the upper country” could be applied to any of the lands upstream from the speaker’s location, especially to the foothills that stretched east and west across northern Syria-Mesopotamia. All of the rivers and wadis of this region flow from the north, rendering the whole region upstream of both Mari and Babylon, as well as being higher terrain.131 Eyes turn downstream much less often, it seems, but there are both “lower districts” and “lower lands.” From the perspective of Ilan-s.urâ in the upper Habur, “the lower land” is Zimri-Lim’s Mari kingdom downstream, a use that ˘ merges the political with the geographical meanings of m¯atum.132 During the reign of Samsi-Addu, Iˇsme-Dagan speaks of a “lower land” downstream along the Tigris from Ekallatum, perhaps the kingdom of Larsa.133 On the local scale of Zimri-Lim’s kingdom, the “lower district” may be the Suhˆ um, ˘ furthest downstream.134

2. The Administrative hals.um ˘ Although the word is well known from the reigns of both Samsi-Addu and Zimri-Lim, the hals.um as administrative district is visible almost entirely for ˘ the latter only. Under Zimri-Lim, the division of the kingdom into districts serves internal affairs and has its significance only within the m¯atum. Where the m¯atum is at one level a military unit and can field an army,135 the hals.um ˘ way, is an administrative unit that can field a work crew.136 Viewed another the .sa¯ bum of the one is an army and of the other a work crew. The tasks of the district crew, then, include canal137 and dam work,138 other construction projects,139 and harvest.140 I do not wish to duplicate feebly the work that Brigitte Lion has done on the hals.um, but there are a wide variety of nonmilitary and local affairs ˘ at the district level (see Lion 2001, 151–9). References to activmanaged ities and materials organized literally by hals.um are listed in Table 7. All ˘ reign of Zimri-Lim, with his of the evidence in Table 7 comes from the system of districts under ˇsa¯ pit.um governorship. Samsi-Addu already used

136

The Archaic State and the m¯atum“Land” table 7. Activities, People, and Materials Organized by hals.um

˘

Census Taxes ilkum service General labor Report to the king Fighting locust plague Coordination of festival celebrations Identification of workers

Various aspects of grain agriculture

ARM XIV 70:2 –3 ; XXVII 25:7–8 ARM III 70+:5–6, paid in clothes ARM XXVII 37:15, paid in sheep ARM XXVII 107:6 ARM III 6:5 (epiˇstum) ARM XXVII 58:9 ARM XXVII 28:18–21 ARM III 45:18–21 ARM XIV 47:23, carpenters ARM XXVI 398:9–14, carpenters and boat builders ARM II 81 = XXVII 76:10–15, palace grain for planting? ARM III 34:14 and XIII 39:15 , plows FM II 70:9–10, harvesta

a

One might use ARM XXVII 25 as a model for hals.um-level activity in general. The text describes a census of the district to get workers˘ for canal restoration, lack of grain and generally gloomy conditions, and the administrative responsibility of the governor himself.

the terminology of “districts” and “governors,” but he does not appear to have worked with the same formal system. The whole kingdom of Mari was evidently the “district” of Yasmah-Addu under the larger sovereignty of his ˘ in one letter to Yasmah-Addu from his father, Samsi-Addu. This is explicit ˘ older brother Iˇsme-Dagan, who governs the other main segment of their father’s realm from the town of Ekallatum and who speaks of “the troop of the Mari district.”141 Under Samsi-Addu, the “districts” identified by personal names would be the regions of responsibility assigned to lesser officials, not defined in geographical terms (see Villard 2001, 115; Lion 2001, 159). In the Mari kingdom itself, however, Yahdun-Lim’s geographical districts seem ˘ to have been maintained by Yasmah-Addu. These did not necessarily have ˘ a ˇsa¯ pit.um “governor” at their head, however, as came to be the norm under Zimri-Lim (Villard 2001, 62–4).

3. The Subordinate hals.um ˘ The Mesopotamian society of m¯atums in our Mari period was largely a society of kingdoms under the central authority of kings. The complex hierarchy extended both inside and beyond each individual m¯atum. One “land” could be considered subordinate to another, so that both Samsi-Addu and ZimriLim freely acknowledged their vassals by this term,142 and the earlier king even treated the portion governed by his son Yasmah-Addu as a distinct ˘ emphasizes the “land.”143 Unlike the hals.um designation, this terminology ˘


137

sovereignty and dignity of the sovereignty and dignity of the vassal peoples and their rulers, who together share the status of the polity that dominates them. While the m¯atums may be distributed according to rank, the very essence of the hals.um is its subordination. Durand proposes the general Mari defini˘ tion “(zone) délimitée” for any subdivision of space (1997, 120; cf. Charpin 1991a, 12). Indeed the hals.um seems to be a part, so then by nature a fraction of a superior whole.˘ The space divided is not simply geographical but political.144 Even when the word seems to be used more loosely, not within the narrower company of Zimri-Lim’s districts, it is still subordinate. One letter to Zimri-Lim quotes another official: “my lord has left for a province to deal with various affairs.”145 This is still a dependent part of the kingdom. A diplomatic report mentions the hals.u¯ nadûtum within the land of Eˇsnunna, ˘ with geographical rather than administrative focus: In the land of Eˇsnunna, there are uninhabited districts (hals.u¯ nadûtum). (When) he (Hammurabi of Babylon) sends to the land of Eˇsnunna,˘ his rapid couriers cross over ˘to the ruler of Elam where the districts are uninhabited.146

Again, however, the term “hals.um” carries with it the assumption of incor˘ poration into a larger m¯atum. When the word “hals.um” is applied to states outside that of the writer, it un˘ derlines their subordinate status. Under Zimri-Lim, Numhâ and Yamutbal (ARM X 157:9), Der of the Balih (XXVI 145:21–2), Karanâ˘? (XXVI 340:12, restored), Nahur of Ida-Maras.˘ (XXVI 352:18), Andarig (XXVI 416:35), and Talhayˆ um˘ (XXVII 64:9) are addressed by this term, all within the po˘ litical framework of Mari domination. Individual vassal kings volunteer the word “hals.um” when addressing Zimri-Lim as their overlord. Kabiya of Kahat ˘ speaks ˘of “the flocks of the tent-dwellers that graze in my hals.um” (II 59: ˘ 4–6). The ruler of Qaâ-and-Isqâ accepts his vassal status to Zimri-Lim by identifying his realm as “the hals.um of my lord.”147 I have found one unusual example of the term in˘ a letter from Zimri-Lim to king Hatnu-rabi ˘ the claw of Karanâ/Qat.t.arâ, where the Mari king declares, “I will tear out of the Eˇsnunna-ite from the midst of my hals.um, and I will deal with him ˘ halsum.”148 Zimri-Lim is cerso that he will never again come up to my . ˘ tainly implying no subordination to either Eˇsnunna or his correspondent in Karanâ/Qat.t.arâ, but he may use the word “hals.um” to recognize that Mari ˘ is just one “district” in the larger alliance against the intrusion of Eˇsnunna. He is, after all, awaiting troops from his fellow king Hatnu-rabi. ˘ Throughout its use, the subordinate hals.um both serves and is conceived ˘ in the image of the m¯atum kingdom. Where the m¯atum is large enough to be divided into districts, this added hierarchical level provides a basis in native terminology for distinguishing larger states within the m¯atum category. I have not pursued this line of inquiry exhaustively, to identify exactly which kingdoms are defined to include hals.ums in the wider evidence ˘

138


for the early second millennium. Whatever the etymology of the word, it has been adapted to serve this political order and has no place in tribal or village structures nor in any old urbanism.149

4. Town and Tribal Institutions with m¯atum-Level Expressions Before continuing, I need to observe the adaptability of ancient social and political institutions to changing circumstances. The collective leadership that is specially characteristic of towns and tribes appears occasionally in the framework of m¯atum and hals.um, taken over to serve larger hierarchies of ˘ authority. Zimri-Lim’s administrative districts often do business with the leadership of their constituent towns on a collective basis, producing structures that look but may not function like those internal to the towns themselves. At Saggaratum, in particular, two letters from hals.um governors address “the ˘ sug¯ag¯u, the laputtû, and the elders of the district,” where the plural sug¯ag¯u and laputtû are the individual leaders of each town in the district, along with their administrative “deputies.” Both letters have to do with a major census, once requiring all of these leaders to swear loyalty oaths and once to assemble them for some exhortation regarding completion of the task.150 The one reference to “the sug¯ag¯u of the district” also comes from Saggaratum, when these leaders inquire together about when they may proceed with their local town festivals.151 One text specifies “the elders of the upper district” in connection with the district of Qat.t.unan.152 None of these suggests fixed and formal political bodies that would meet on any regular basis, any more than would the elders of towns. We are dealing instead with convenient terminologies for talking about the mass of town leadership insofar as it is accountable to the rule of the larger kingdom. The same is true when we encounter similar constructions with the m¯atum. It is quite common to find the major terms for collective town leadership applied also to a whole “land.” In some cases, the political entity is built around a capital city, so that we might imagine an extension of the town structures. A Mari letter describes a large number of “m¯atum heads” from Imar who represent the town in a legal procedure at Hˆıt: “Eighty Imarite heads of ˘ the land are going(?) to the River God for immersion because of the silver 153 It is harder to judge the “m¯atum elders” of the goddess Baalta-m¯atim.” in Kurdâ and Andarig during the Zimri-Lim period, when these kingdoms of the Numhâ and the Yamutbal may have incorporated aspects of a tribal structure.154˘ Zimri-Lim says of his own kingdom that he has 200 “heads of my m¯atum,” 100 Hana and 100 from the Ah Purattim.155 During the earlier ˘ classified an individual ˘ official as one of the “heads of period, Samsi-Addu 156 the land.” I am not aware of any evidence for this leadership metaphor before the early second millennium, and it may derive from Amorrite usage, rather than older southern Mesopotamian patterns.157

Population Terminology Not Tied to Political Entity

139

As I suggest below, “elders” represent an especially fluid category, and these may be designated even for “the m¯at Ida-Maras.,” a land with a unified face only as allies in matters of war and peace.158 These form an ad hoc gathering that reflects the diplomatic purposes at hand, and they have little to do with the group associated with town leadership. We never find “sug¯ag¯u of the m¯atum,” although the Zalmaqum coalition has “our kings and our sug¯ag¯u” at another negotiation. Zimri-Lim’s envoy quotes the Zalmaqum leadership in its dealings with the king of Aleppo: “We must go. We must send(?) our kings or our sug¯ag¯u to negotiate with you.”159 Here the collective aspect may reflect the tribal institutions of these Yaminite-affiliated states. My point here is that we should recognize the difference between these and the town institutions by the same names. Most likely, the kingdoms and coalitions adopted town and tribal institutions and terminologies and gave them expressions that ought to be regarded as derivative.

d. population terminology not tied to political entity It is not possible to understand any one strand of the political terminology in the Mari archives without exploring the larger web to which it contributes. Collective institutions were preserved in towns and tribes, particularly, but these towns and tribes formed integral parts of “lands” into which all of Mesopotamia was divided. These “lands” or m¯atums were ultimately defined not by land but by people. If understanding polity means understanding people, then we need to address two other words that are often associated with Mesopotamian population, though neither turns out to be a political term at base. These are the niˇsu¯ , a word that appears in Babylonian inscriptions from this period as the “people” ruled by Hammurabi or his successors, and the muˇskênum, who represented the mass ˘of subjects who were not directly dependent on royal financial support. To some extent, the discussion that follows serves to distance both the niˇsu¯ and the muˇskênum from this political study of the Mari evidence. Whereas the m¯atum and the hals.um are defined by decision making or governance, the niˇsu¯ and the muˇs˘kênum are not. The m¯atum and the hals.um are therefore ˘ political categories, while the other two are not. To establish the grounds for excluding these words from my core terminology, as well as to explain how they do fit into Mari period society, I need to define their use.

1. The Dependent niˇsu¯ Like the m¯atum, the niˇsu¯ are a group of people, but there the similarity ends. Whereas the m¯atum is a political entity, both as a kingdom and as the population that can respond collectively to its ruler, the niˇsu¯ never form a decision-making group. The niˇsu¯ are the circle of dependents under a

140


given patriarchal head, and the term functions entirely in the context of that dependency. A ruler has responsibility to care for a niˇsu¯ as for a family; he governs a m¯atum, which has an independent political will and can choose to follow or to revolt. Even in the context of a household, the niˇsu¯ does not speak to its head, so it should not be surprising that the niˇsu¯ “people” have no political voice at any other level. In the Mari evidence, the word “niˇsu¯ ” may be used for a range of social categories, always with reference to their dependence on a leadership outside itself. Often, the niˇsu¯ pertains only to the extended household of a male or female head. A man moves upstream with “his mother and his dependents.”160 Yaqqim-Addu reports the anticipated arrival of a Kurdâ-ite named Yarim-dadu, traveling with his dependents.161 It appears that the “dependents” form a second circle, after the immediate “house,” based on a letter in which both a man’s house and his niˇsu¯ have been sold into slavery.162 They did not find that man, and he had sold his house(hold) and his dependents into slavery. On the next day, a tablet reached me from Yasim-Dagan, saying, “This man has just arrived.” Now my lord should write to me one way or another, whether I should set free his dependents.163

In another text, a high-ranking lady may be responsible for the same two domains: Samiya writes to Sininaya, complaining, “Why have you given up managing your house and sustaining your dependents?”164 The king of Carchemish writes to Zimri-Lim about the potential outcome of a decision by ordeal. If the men in question die, he says, “I will hand over their houses and their dependents for the striking of their heads.”165 A whole community may be addressed by its prominent male leaders, collectively responsible for their dependent niˇsu¯ .166 In one case, a proposed message to Zimranum may address a whole clan by the name of its chief, as it continually uses plural forms of verbs and pronouns.167 They are to be instructed to move “your people” (line 15 ) into the heartland. “Take the hand of your dependents, come together(?), go away into the countryside. Cross over to Abattum and settle in your (own) backcountry with your kinsmen.”168 Another letter describes the creation of a newly fortified town by a leader and “his niˇsu¯ ”: “Zaikum the Yabliya-ite has taken flight with his dependents to Harbˆ u, (where) he has built a village between Harbû and ˘ another Ayabu.”169 King ˘Haya-sumu of Ilan-s.urâ may speak of his people in ˘ 170 town as “my dependents” (ni-ˇsi-ia), and even a temple may be considered to have a niˇsu¯ .171 Such a definition of the noun “niˇsu¯ ” seems to be standard in early second-millennium Mesopotamia, as shown by the murder case in the Code of Hammurabi (law 24) where the niˇsu¯ of a dead man are to receive ˘ the town. silver from The niˇsu¯ have no political or decision-making aspect whatsoever. In grammatical terms, their dependence is expressed by genitive “possession.”172


141

The niˇsu¯ are the human element of a larger domain, all that falls under a leader’s responsibility. The remaining domain may be defined as the baˇs¯ıtum (“possessions”),173 the en¯utum (“furnishings”),174 or both house and grain.175 Where the word “niˇsu¯ ” is used in a political framework, it serves an individualized vision of the state, making the chief or ruler the head of an extended household. This usage seems to be particularly evident in the ideologies of Hammurabi and Samsu-iluna of Babylon, where the m¯atum “land” and niˇsu¯˘ “dependents” become a merism for the population of the kingdom in a vision of beneficent rule.176 Interestingly, a bilingual royal inscription of Hammurabi omits any Sumerian equivalent for the Akkadian ˘ a Semitic social construction that is not simply equivalent niˇsu¯ , suggesting to the Sumerian.177

2. The muˇskênum The word “muˇskênum” defines the collective mass of royal subjects who are not direct beneficiaries of royal largesse. In the Mari evidence, the word belongs to the narrow social perspective of the king’s palace administration, which thus defines what is outside itself. The muˇskênum have no political institutions, because the function of the terminology is apolitical. These are the undifferentiated mass of “subjects,” defined by their submission.178 Even as the term “muˇskênum” serves a palace-oriented point of view, the very isolation of such outsiders by a single sweeping term suggests a domain where royal power intrudes less into the economic life of the population.179 In some respects, the muˇskênum defines for economic life what the m¯atum defines for political life: a population ruled but still distinct from the ruler’s household. For the Mari evidence, Durand has defined the muˇskênum as the free residents of the countryside, not the cities (“villes”), but who do not join the periodic movements of tribal peoples. They do not depend on the palace, and when they take on palace service, it is only to supplement their private labor.180 These ideas echo in part the analysis of Robert Adams, who concluded that “there is very likely to have been a close association of the term muˇskênu not merely with inferior status but with the outlook and cultural patterns that particularly characterized rural life” (1982, 12–13). Adams passes on a communication from John Brinkman, who observes that the most common early opposition is between palace and muˇskênum, not aw¯ılum (“man”) and muˇskênum, and this is a primarily urban opposition. The Mari evidence alone shows that, with extensive royal lands outside the urban captital, the palace/muˇskênum opposition is crucial to rural life. Insofar as the populace of a m¯atum was identified with the land outside the urban centers, the idea of a rural orientation need not conflict with F. R. Kraus’s definition of the muˇskênum as the whole free population of the state, those not directly dependent on the palace.181

142


a. The economic angle. As Durand has argued, the muˇskênum are not the “poor” (1991, 21n18). They are nevertheless defined in economic terms by their separation from the palace, as argued long ago by Kraus (see above). In the Mari letters, this is visible first of all with agricultural land,182 but it ends up applying also to grain,183 livestock,184 boats, wagons, and even slaves.185 Owners of such property clearly need not be poor, though any destitute persons would be found here, outside the institutional net of palace support.186 In one badly damaged letter from Zimri-Lim’s daughter Kirû, if she indeed complains that she is “the slave of a muˇskênum,” she is first concerned not about poverty or even social status but with a sense that she has generally lost the support of her father, the king of Mari.187 A letter to Zimri-Lim from an unidentified leader outside his kingdom complains that even after he helped Zimri-Lim acquire the throne of Mari, he himself cannot inherit the throne of his ancestor. “I am a commoner,” he says.188 By being anyone but king, he is a muˇskênum. Here we are not talking about membership in a higher aw¯ılum class or land ownership. What, then? Perhaps he speaks of separation from palace support, whether Zimri-Lim’s or pertaining to his own proper administration. If we can refuse definition of the muˇskênum as “poor,” and the primary opposition is muˇskênum-versus-palace, there is still some evidence for a hierarchy that includes the aw¯ılum, “men” or “gentlemen.” The higher status of an aw¯ılum in royal perspective is perhaps assumed in a letter to Zimri-Lim from Dariˇs-libur, his envoy to Yamhad, reporting on how three Yaminite ˘ kings have been lodged at a place called Serdâ at “the house of a certain aw¯ılum of the Yahruru(?) (tribe).”189 The status of this man seems to be linked to the size˘ and prestige of his household, in its capacity to house three tribal rulers. A letter addressed to some local king separates the population of a town ´ meˇs (aw¯ılû) who must be maintained in residence (verb sak¯anum) into the LU and the muˇskênum whom the speaker has already calmed.190 The two groups are presented as the responsibility of a b¯el a¯ lim (“town-master”), the leader of a town as such, not of some larger administrative unit: And I myself am determined to keep the gentlemen (aw¯ılû) in settlements until the town-master arrives, and I have (already) calmed the heart of the muˇskênum. Until your son arrives, you must absolutely keep the muˇskênum in settlements.191

It is tempting to suppose that the aw¯ılû are those who do have some permanent stipend or support from the palace, perhaps even by virtue of loyalty oath or other formal commitment. We must keep in mind that the text invokes a distinction that originates in the perspective of the royal palace, which is foreign to the people in question. Whether as aw¯ılum men or as “the town,” these dominant groups could be those whom the palace recognizes as responsible for town decisions. From this view, then, the muˇskênum could be those who are excluded from formal “town” relations with the king.


143

One thinks of the legal exchanges between Zimri-Lim and the Simalite town of Sapiratum, recorded in the document ARM VIII 85+, a text to be discussed further in Chapter 4 under the section “Tribal Heads.” When the document lists all the men of Sapiratum or of its Simalite tribe, perhaps the list cannot be assumed to represent all households, because a muˇskênum element is left out. Another letter speaks of the muˇskênum w¯aˇsib¯ut a¯ lim, those who are “resident in the town” at Qat.t.unan.192 b. Villagers. Whether in political terms only or also by actual residence, the muˇskênum are set apart from major towns. At the same time, however, they are sedentary and identified especially with smaller villages. Aside from their ownership of agricultural land, another letter from the Samsi-Addu period even identifies the mu-uˇs-ke-nu-um ˇsa a-la-ni (“muˇskênum of the towns”), the villagers who will be performing local celebrations of a Nergal festival. In very broken context, this text sets them against the nawûm, the backcountry steppe and its pastoralists.193 The last bifurcation is odd, when set against another text from SamsiAddu’s reign where both the Hana flocks and “the muˇskênum of the Ah ˘ the move. “The sheep of the Hana and˘ Purattim” are reported to be on ˘ crossed of the muˇskênum of the Banks-of-the-Euphrates (Ah Purattim) have ˘ 194 During the reign of Zimri-Lim, one governor (the river) to the wadis.” of Qat.t.unan (Zakira-Hammu) is very worried about the possible departure of muˇskênum from the˘district, even while assuming the legal right to detain them. And the rich aw¯ılum who has barley is staying, while the poor muˇskênum who has no barley has departed toward the (Euphrates) River.195 Whoever among the muˇskênum of the district is about to depart without your permisˇ sion for the land of Subartum, arrest him and conduct him to me.196

He fears they will abandon the region for “the river” or “the land of ˇ Subartum” and quotes their representatives as asking, “How can we stay?” without food.197 Any movement, however, should have been only with the governor’s permission, and otherwise the renegades may be arrested and sent back to Qat.t.unan.198 All this control stands in contrast to the Simalite Hana, whose movements are not restricted in this way under the reign of ˘ their tribesman, Zimri-Lim. The muˇskênum may be free and independent of regular palace support, but they are fully subject to the royal administration. The king can require their military service and may requisition the use of their oxen, although at the same time he must beware of overburdening this backbone of his realm.199 In Qat.t.unan texts, the palace and the muˇskênum work side by side as district work forces.200 One letter from the governor envisions a harvest carried out by the “men of the palace” (aw¯ılût ekallim) and the muˇskênum.

144


The first group may not be full-time palace personnel but local people who are allowed to work the land owned by the palace.201 Another Qat.t.unan letter illuminates the question by listing the categories of royal beneficiaries who receive land from the king. The pihrum troops and ˘ “the sons of ´ (or iku) to farm, while the ˇsu¯ t r¯eˇsim guards are given five GAN meˇs ma-a-tim) are assigned only three each:202 the land” (DUMU Just as (for) the regular (mustered) soldier and the (royal) guard of the Banks-ofthe-Euphrates who receive five ikus of land each and perform hard service for the palace – and the detachment of the Qat.t.unan district does receive five ikus each – likewise I intend to provide you those full five ikus as your land allotments, so that I can assign the remaining fields to the local population (“sons of the land”) at three ikus each.203

It is emphasized that these are not outright grants: “These fields are palace fields.”204 All of the land that is allotted according to this text is evidently palace land, and we expect that all of the rest would be attributed to the muˇskênum. In this text, then, “the sons of the land” appear to be all of the native residents who receive this royal benefit, by definition excluding the muˇskênum.205 It should not be necessary to exclude the muˇskênum universally from the category of “sons of the land,” however, because this text is concerned only with those who receive regular royal support. Another letter from the reign of Zimri-Lim has the governor of the Saggaratum district declare that he, “along with the muˇskênum,” takes responsibility for “the sacrifice of the sons of the l[and],” a connection that would confirm the inclusion of the muˇskênum in the larger definition of native residents.206 c. The political otherness of the muˇskênum. When the muˇskênum act together in the Mari correspondence, it is always from the ruler’s point of view. References to the collective voice of the muˇskênum are found only in the direct reports of royal governors: r ARM XXVI 154-bis:28–32: Hammanum, governor of Yabliya under ˘ Yasmah-Addu, says, “When he (the diviner Zikri-Hanat) has taken omens, ˘ even if they are totally favorable, I will not hand˘over one liter of barley. The muˇskênum of my district will hear (about it), and they will strike their nose(s).”207 r XXVII 1:24–7: Iluˇsu-nasir, early governor of Qattunan under Zimri-Lim, . .. is concerned that when the muˇskênum hear about plans for evacuation of palace personnel in grain shortage, they will say, “How can we stay?”208 r XXVII 27: 27–9, Zakira-Hammu, governor of Qattunan under same, re.. ˘enum are faced with a locust plague and have ports that the local muˇskˆ protested: “Now that the locusts are eating our crop, are we ourselves supposed to be able to stay in this place?”209


145

r XXVII 37: 41–7, same; a long quoted complaint in the voice of the muˇskênum (line 41) about the terms of harvest labor, ends, “the requirement is too severe for us.”210 In similar references to their joint action, the muˇskênum are equally set against the king and his administration.211 In general, muˇskênum is not a word to describe a specific political entity that makes decisions. They do not meet to decide matters of peace or war, of internal law or any external affairs. Indeed, they do not meet at all. They do not receive or send letters. Rather, muˇskênum is a palace word, the terminology of kingship and its economy as its leaders look out on the mass of everything ruled. Even in grammatical terms, muˇskênum is always a singular collective, only occasionally marked by the plural determinative ´ ˇ 212 The limited social perspective of this terminology should warn LU.ME S. us to be careful of universalizing its categories. In a way that too quickly attributes universal range to limited terminology, Liverani identifies two types of property and two types of population in all of Bronze Age Syria. There is family property joined in village communities, distinguished from palace property. Similarly, there is a free population in the villages, contrasted with “the men of the king” in the palace organization (Liverani 1975, 146). The latter opposition sounds remarkably like that of the muˇskênum terminology, but it surely represents an oversimplification, just as the palace organization itself regarded the life of the people it governed in an oversimplified way. The one-size-fits-all approach of the ruling powers is reflected in the instructions of Zaziya, the king of the Turukkˆ u, to Meptˆ um, one of the leaders of the Suhˆ um region, telling him to collect a thousand sheep from “your ˘ undifferentiated mass.213 muˇskênum,” an It is no accident that the word appears especially in the voice of administrators. From this view, the muˇskênum is the core of the kingdom proper, that which is fully administered by the palace, occupies land side-by-side with palace land, and thus owes the king military and corvée service. We should not find the muˇskênum in vassal domains, foreign lands, or the steppe.214 One letter from the negotiations over Zimri-Lim’s acquisition of a town called Alahtum from the kingdom of Yamhad calls the inhabitants muˇskênum: ˘ ˘ Have you never before sold fields or a house that belonged to you? And as for what Nur-Sˆın is saying to the muˇskênum, “Go to the mountains for wood. Send(?) your donkeys. Come to my help.” This is what Nur-Sˆın took(?) and said, so the muˇskênum are complaining to him, saying, “Are we going to serve(?) as slaves just as for Gaˇsera (queen mother of Yamhad)?”215

˘

These people do not take well to the demand that they carry out the work demanded by kings and queens. Although the acquired town is far from Mari, it is the particular domain of king Zimri-Lim, however, and the use of muˇskênum appears to reflect that fact.

146


It is interesting that the word is used much more frequently by the governors of Qat.t.unan than in the districts of Mari, Terqa, and Saggaratum. Qat.t.unan was different from the other districts in several respects. Located much further up the Habur than Saggaratum, just southwest of the Jebel ˘ one Mari district with a ˇsa¯ pitum governor that Sinjar, Qat.t.unan was the . stood completely outside the Ah Purattim (“Banks-of-the-Euphrates”) territory that all the Mari kings of˘this period inherited with their conquest of the city itself. The Qat.t.unan district seems to have been more focused on a single central town than were the Euphrates districts, which possibly allowed a sharper contrast between the people of the royal outpost and everyone else.216 The details of the governors’ letters indicate nagging agricultural difficulties, again reflecting conditions that were different from those downstream. Qat.t.unan’s proximity to the upper Habur would have placed it nearer to the stronghold of Simalite presence˘ in the north, where the muˇskênum would be set apart from Zimri-Lim’s mobile “Hana” tribesmen. ˘ a weaker role Perhaps the more frequent reference to the muˇskênum reflects for the district system of administration in this territory. d. Religious affairs. There is one situation where the muˇskênum appear to act as a body: the execution of religious rites. Tarim-ˇsakim, one of Yasmah˘ Addu’s two viziers, writes to his lord, “Regarding the Wagon of Nergal sacrifice, it is to be hoped that the muˇskênum of the villages will not place (it in) the coming month.”217 In another case, Yaqqim-Addu, the governor of Saggaratum, asks Zimri-Lim to approve “the 8th of next month” for performance of pagra¯u sacrifices by the muˇskênum.218 Another letter from YaqqimAddu identifies an individual who, along with the muˇskênum, takes responsibility for “the sacrifices of the sons of the land.”219 Given that the primary purpose of the muˇskênum distinction is economic, this detail may be provided not to name the class of the celebrants, but rather to observe the financial independence of the rites in question. In a letter from two of his officials to Yasmah-Addu, Samsi-Addu refers to the celebration of the râmum festival according˘ to two sets of preparations, one in the Mari palace and one “with the muˇskênum.”220 The separate preparations may reflect separate funding. Two diviners, Asqudum and Ibbi-Amurrum, report omens associated with a “muˇskênum sacrifice.”221 Each need not indicate a private individual’s sacrifice but could simply be a “local festival,” like those mentioned in the three other texts above, privately funded, without royal contribution. These events would never be recorded in the accounting texts for palace disbursements. If this is so for the two omen sacrifices, it also suggests a different reading of ARM XXVI 1, “the diviners’ protocol”: . . . of the muˇskênum . . . is found/placed, and I see (it), [not(?)] good, whatever I see . . . I will hereby speak. I will not hide (it). The bad sign (“flesh”) . . . which is

Zimri-Lim and the Land of the Tent-Dwellers

147

found in the omen(s) of Zimri-Lim . . . , in abnormal birth or . . . , I will tell no one. And the secret message which Zimri-Lim my lord speaks to me, to take omens. . . . 222

Although the opening lines are broken, the remaining text mentions omens of Zimri-Lim (te-re-et m Zi-im-ri-Li-im, lines 7–8, cf. 11–12), and Durand restores [i-na te-re-et mu-u]ˇ ´ s-ke-nim in line 3, “[in the omens of the mu]ˇskênum.” The two omen reports would perhaps better suit the reading, [i-na SISKUR2 .RE (ˇsa) mu-u]ˇ ´ s-ke-nim, “[in the sacrifices of the mu]ˇskênum,” to match XXVI 109, and none of these would show omen proceedings initiated by private individuals. Whether as individual inquiries or local festivals, the term “muˇskênum” should mark the other pole of ritual sponsorship, set against palace funding. The diviner must report omens that appear in either type, and the actual reports by Asqudum and Ibbi-Amurrum show that the alternative is real. The king cannot afford to miss omens in sacrificed animals not supplied by himself. In spite of this evidence for the economic definition of the muˇskênum, we must be careful to avoid treating them as a social “class.” If we could be confident that the royal perspective was the only one that counted, in terms of raw power, perhaps the distinction of muˇskênum from aw¯ılum could provide a substantial starting point for separating functioning groups. Too much of the Mari evidence, however, shows us other ways of defining groups, as by tribe or by settlement. As a m¯atum, the people can speak with a collective voice that is distinct from the king’s. Naturally, the muˇskênum category will overlap with categories based on other perspectives. We must allow the term to illuminate the perspective of kings without forcing it to serve universal analytical ends. Above all, however, the muˇskênum category must be reserved from discussion of the political power of the group, because the word is not political.

e. zimri-lim and the land of the tent-dwellers ( m a¯ t hana)

˘ The overwhelming majority of our Mari correspondence comes from the reign of Zimri-Lim, the last king based in Mari before its destruction by Hammurabi of Babylon. Zimri-Lim defined his kingdom in terms adapted ˘from his tribal kinsman Yahdun-Lim, who had reigned over another ˘ Simalite kingdom from a Mari capital before the period of Samsi-Addu’s hegemony. Both Yahdun-Lim and Zimri-Lim called themselves “the king of Mari and the land of˘ the tent-dwellers (m¯at Hana),” a titulary that combines ˘ by a word for mobile pastoralthe city center with a regional domain named ists. Insofar as the name “tent-dweller” usually referred to one tribal group’s own pastoralist population, the phrase “land of the Hana” joins the large ˘ the interests of this political unit of the m¯atum to a tribal category, bridging chapter and the previous one. Under the rule of Zimri-Lim, the kingdom

148


and the tribe were knit into an inseparable unity. It is not possible to understand the political situation that framed most of the Mari documentation without coming to grips with this royal titulary and with the persistent but evolving political realities that underlay it.

1. The Land of the Hana Tent-Dwellers ˘ I have tried throughout this study to maintain the necessary distinction between “tribal” and “pastoralist” categories, the one involving social and political organization, the other referring to a mode of subsistence. This distinction becomes slippery with the crucial term “hana,” which first of all ˘ designates groups by their role in subsistence as pastoralist “tent-dwellers,” but which seems to take on an ethnic overtone when applied to a given tribal people’s own mobile pastoralists. In the titulary of Yahdun-Lim and ˘ the regional Zimri-Lim, the subsistence term “hana” provides the name for ˘ population ruled as the m¯atum of these kings. These “Hana” seem to be ˘ specifically Simalite, the tribal confederacy of the two Mari kings, and all of the tribespeople are thus identified by their pastoralist affiliations. Even basic identifications demand defense when this multivalent “tent-dweller” terminology is in play, and the following sections explore the relevant range of use and evidence. a. The “tent-dwellers” in the m¯at Hana. We know the reign of Zimri-Lim much better than that of Yahdun-Lim,˘and therefore we do best to begin our examination of the “land of˘ the Hana” with the well-documented period of Mari’s ˘ a hybrid administration that was intended to last king. Zimri-Lim created address the specific needs of his mixed constituency. Not only did Zimri-Lim have to govern both his own Simalite compatriots and a large population of rival Yaminites, but he also had to deal with long-standing settled townspeople with varying tribal affiliations or lack of them, along with various mobile pastoralists who stood in some relationship or other to the king or to the settlements of his core territory.223 In this wider context, the basic division among Zimri-Lim’s subjects fell between his own tribe and the rest. The Yaminites who swore fealty to the Mari king, especially after their defeat early in Zimri-Lim’s reign, did so by town and were incorporated into the hals.um districts of the core Mari realm, the Banks-of-the-Euphrates (Ah ˘ Purattim). These districts were governed by officials called ˇsa¯ pit.ums in a˘ wholly urban framework. At the same time, however, Zimri-Lim made provision for his own tribespeople in terms that did not require commitment to fixed settlements. Simalites could be defined by their mobile hana as˘ title pect, with governance under another old tribal role, the merhûm, whose ˘ reflects leadership according to herding communities. If the word “hana” (“tent-dweller”) acquires a specific application to ˘ to be by use of the word to refer to “our nomads.” Simalites, it seems


149

Viewed from within, the general term has come to name an individual tribe. Durand’s “American” analogy, whereby “America” can designate either the two continents together or the United States more narrowly, is then worth another look (1998, 418). In a technical sense, the French perspective is that of an outsider, naming categories of other populations that will not match their own uses. Usage in the United States is quite different. At home in the United States, an “American” is a person from this country and nothing else – never a Canadian or a Brazilian. Those in the United States would find any other use confusing. Durand’s broader use of “American” to combine both continents, North and South America, belongs only to the outsider looking in. The combination of continents is possible only from the perspective of Europe, which assigned the names in the first place. The second part of the French duality is really borrowed from our narrower reference in the United States, calling us what we call ourselves. This closer look at the American analogy suggests a different approach to hana terminology in both the kingdom of Zimri-Lim and elsewhere. Zimri˘ Lim’s m¯at Hana is “American” not from the French perspective but from ˘ that of the United States. The m¯at Hana is “us” and not “them,” and as such ˘ is much more likely narrow than universal, Simalite than mobile pastoralist in general. In the same way, we in the United States would never imagine including those outside our own people in the category “American,” especially in a formula meant to encapsulate who “we” are, distinguished from all surrounding peoples. My starting point for evaluating hana terminology in Zimri-Lim’s king˘ formal documents. Royal seals that dom is his titulary in royal seals and identify Zimri-Lim by what he rules repeat the same formula, “king of Mari and the land of the Hana (m¯at Hana).” This reclaims the earlier titulary of ˘ ˘ Yahdun-Lim, whose tribal conception of the m¯at Hana is displayed in the ˘ ˘ him as “king of Mari alternative form found in a seal of his sister, who casts and the m¯at Simal” (see Charpin and Durand 1986, 151–2; Charpin 1992d, 72–3). The importance of this official definition of the realm is displayed by its use in treaty documents and oath protocols. Based on four examples, the titulary appears to have been fixed and formal, with exactly the same form in both text types: “Zimri-Lim, son of Yahdun-Lim, king of Mari and ˘ the m¯at Hana.” ˘ In all cases, the formula provides the official identification of Zimri-Lim as party to an arrangement that will have legal force. The claim to be Yahdun˘ Lim’s heir, as his son, is joined inextricably to the titulary that describes 224 the content of his political inheritance. The two treaty texts both pertain to key alliances made during the second half of Zimri-Lim’s reign, one with Atamrum of Andarig, and the other with Hammurabi of Babylon, in ˘ the Hana themselves in opposition to Elam.225 The oath protocols serve one instance, and officials of the vassal state of Karanâ in˘ the other, both by reference to the same definition of their lord.226

150


Charpin and Durand understood the inclusion of a m¯at Hana with Mari ˘ to be an ideological claim to govern all “bedouin,” while Zimri-Lim was fresh from a victory that may have joined both Simalites and Yaminites against the western remnant of Samsi-Addu’s great realm (1986, 153). The word “m¯atum,” however, consistently refers to some real political unity.227 While there were indeed Yaminites in the kingdom, many were explicitly outside, not to mention other tent-dwelling pastoralist groups such as the Sutˆ u. Also, the Yaminites who were successfully incorporated into Zimri-Lim’s kingdom were consistently identified by their towns (¯al¯an¯u), by which they were intrinsically separated from the “tent-dwellers.”228 The separation of Yaminite towns must be the basis for the odd reference to a m¯at Miˇslan in one of the administrative texts for the census in Zimri-Lim’s sixth year. Miˇslan, in the Mari district, was the settled seat of the Yahrurˆ u Yaminite tribal ruler, and ˘ the “land of Miˇslan” includes ten-plus Yaminite settlements, conceived as a distinct political entity from a Mari administrator’s perspective (M.5085+, in Millet-Albà forthcoming). Together, these patterns indicate that Zimri-Lim’s m¯at Hana should ˘ “bedouin” exclude the Yaminites. Knowing the duality already present in terminology, the hana accepted within the kingdom were likely to be the ˘ Simal. king’s own tribe, the b. The Simalite Hana, truly mobile pastoralists. The point of this section is to ˘ appear to be obvious: that the Simalites who are called reiterate what may “hana” in Zimri-Lim’s circle were truly mobile pastoralists. When the word ˘ “hana” became a tribal name, as among the people of the land of Apum, ˘ such “Hana” could be identified with settled towns such as the one disputed ˘ Sunˆ ˇ a and Apum, cited in my discussion of the Hana in Chapter 2. between ˘ Even when the term kept its general descriptive sense, however, and could be applied by a tribal people to its own “tent-dwellers,” there are complications. One may expect a term plausibly defined by its opposition to the town to signify the truly mobile pastoralists, but this conclusion raises difficulties in the evident parallel between Yahdun-Lim’s m¯at Simal and the m¯at ˘ Zimri-Lim in their royal titularHana name used by both Yahdun-Lim and ˘ ˘ ies. However Yahdun-Lim intended the term, the standard use of “Hana” ˘ government carries a tribal assumption but is not in ˘ fact under Zimri-Lim’s a tribal name. Mari and the m¯at Hana are not the same as Kurdâ and the m¯at Numhâ or Andarig and the m¯at ˘Yamutbalim, because the word “hana” first ˘ of all ˘defines the entire mobile way of life, and the tribal association with the Simalites is secondary. Under Zimri-Lim, at least, if the word “hana” ˘ in the titulary is to be read with the common meaning found in his circle, the king identified the tribal element of his realm by its mobile pastoralist “tent-dwelling” component only. The effect of the titulary is to give primacy to the ideology of tribal mobility, in spite of the strong sedentary component to actual Simalite life.


151

The Hana of Zimri-Lim, then, would seem to be identified with the steppe, which is˘ set against the town. The oath protocol for his “Hana” divides the voice of the kingdom’s population between the “Hana ˘of the steppe ˘ (nawûm)” and “the townsmen:” I shall hereby [never] commit offense against Zimri-Lim my lord, and when I hear an unfriendly word, either from the mouth of the Hana of the steppe or in the mouth of the townsmen, saying, “Zimri-Lim [and] his ˘offspring shall not rule us,” I shall hereby not join in bringing trouble on Zimri-Lim [and] his offspring. . . . 229

Their merhûm governs the Hana from the steppe,230 and the nawûm (“flocks” ˘ expression of ˘the Hana may be understood to move in and out or “steppe”) ˘ defined as part of Zimri-Lim’s administraof a fixed hals.um, whether or not ˘ tion of the Ah Purattim: ˘ The flocks of the Hana that are grazing in my district are well. They have received ˘ fair treatment in pasture (and) water, and in justice. For the flocks of the Hana and ˘ for the town of Kahat, (all) is well.231

˘

I am afraid that when the flocks of the Hana graze on the left bank (or “near bank”) ˘ and losses will occur.232 of the river, the enemy will fall (on them)

Based on their home in the steppe, the Simalite Hana are accepted in the ˘ visitors, still under the lands of Zimri-Lim’s northern vassals as legitimate 233 authority of their own king. Just as the Simalite Hana are powerfully ˘ identified with the mobile life of the steppe, so are the “tent-dwellers” of the 234 Yaminites. To illustrate the freedom of movement enjoyed by pastoralists, Charpin cites a remarkable articulation of the mobility associated with the hana tentdwellers, compared with merchants who pass from land to land ˘in war or peace with equal freedom: My lord knows that I govern the Hana, and like the merchant who travels between ˘ Hana travel on foot [between] (zones of) war (zones of) war and (of) peace, the ˘ and (of) peace.235

An important implication for the Hana of Zimri-Lim, who claims that they ˘ political situation is also like that of represent his own m¯atum, is that their merchants, who still serve their native states while in the domain of other rulers. This letter from the Simalite merhûm Ibal-el offers an indigenous rationale for the idea that the Hana did not˘live by fixed political boundaries, ˘ and likewise for Zimri-Lim’s extended influence beyond these. The mobile definition of Zimri-Lim’s Hana is particularly striking in ˘ these examples from the Habur region, where they mingle with popula˘ tions loyal to local kings. A letter sent by one of these local rulers displays several relevant details together. The ruler of a minor northern center called Zalluhan is complaining to Zimri-Lim about how his town is treated, ˘

152


especially regarding its identity as part of Zimri-Lim’s dominion. He appears to quote the Simalite king and then respond: [You said(?),] “Zalluhan is not part of (lit., “a son of”) Ida-Maras.. It is a part of (“a ˘ should (then) write to Ibal-pi-el so that the population of son of”) Simal.” My lord my district can go around with the tent-dwellers (Hana) at the edge of the . . .(?), ˘ and so he can take responsibility(?) for me.236

First of all, we see Zimri-Lim claim a town for his tribal people, the Simalites. Perhaps the proper name is used, rather than “Hana,” because the letter does not come from Zimri-Lim or his own court,˘ even though the words are put in his mouth. In his response to Zimri-Lim’s supposed claim on his town, the ruler of Zalluhan offers a sarcastic comeback, which should not be understood as actual˘ fact, that now the Mari king will be putting the town under the oversight of a merhûm and expecting it to troop around in ˘ the steppe as hana tent-dwellers. The association of Zimri-Lim’s Simalite ˘ Hana with mobile life in the steppe shows up even outside his own circle. ˘ Even more interesting is the fact that the ruler of Zalluhan now wonders ˘ will now have whether his town, which is not tent-dwelling in the least, to be governed through a merhûm. Why say this, unless this is the norm ˘ least, it appears that Simalite towns in for other Simalite towns? At the the upper Habur, outside the districts of the Ah Purattim and Qat.t.unan, ˘ ˘ hûms and their ideology were administered under the hierarchy of the mer ˘ of mobility. Then the Simalite identity is what would place Zalluhan un˘ were der Ibal-pi-el the merhûm, which would not be the case if the town ˘ counted as part of Ida-Maras., in spite of the merhûms’ influence over that ˘ northern alliance. In this region, within the framework of Zimri-Lim’s domination, it seems that one is either Simalite, and Hana by association, or part of Ida-Maras., and permitted to maintain one˘ more degree of independence. It is less clear to me how exactly the “tent-dweller” terminology related to larger Simalite identity in the core districts of Zimri-Lim’s kingdom. Bahdi-Lim, the district governor of Mari itself, observes in one case that the˘ Simalite Hana come from the steppe to take up residence (waˇsa¯ bum) in local towns˘ (ARM II 48:8–9, cited above). They may come freely, but they consider themselves “held” (kalûm) in towns if not able to leave just as freely.237 Hana in military service may be given grain to keep them settled ˘ There are, of course, plenty of Simalite towns, but there are in a town.238 no “towns of the Hana,” a contradiction in Simalite terms. ˘ c. The m¯at Hana terminology before Zimri-Lim. We know little of the administra˘ that governed the realm of king Yahdun-Lim, and we cannot tive structures ˘ systems of Yahdun-Lim assume that Zimri-Lim had simply reconstituted the ˘ our evjust because he took over the earlier king’s titulary. Nevertheless, idence for the m¯at Hana begins with Yahdun-Lim, and this earlier period ˘ ˘


153

provides an important control on our interpretation of Zimri-Lim’s titulary, which therefore cannot be explained from the perspective of Zimri-Lim’s kingdom alone. Before Zimri-Lim, Yahdun-Lim already called himself “king ˘ of Mari and the m¯at Hana,” most famously in his elegant inscription for the ˘ ˇ dedication of a temple to the sun-god Samaˇs (i:17–19, Dossin 1955). Yahdun˘ Lim recounts the victories and other exploits that led him to the point where ˇ he could build this temple for Samaˇs. Like Zimri-Lim later, Yahdun-Lim traveled to the Mediterranean Sea, where he acquired precious˘ woods for his building projects (i:34–iii:2). He then proceeded to fight off the assault of three Euphrates valley kings of tribes known from other evidence to be Yaminite: During the same year, Lauˆ m king of Samanum and the land of the Uprapˆ u, Bahlu˘ kullim king of Tuttul and the land of the Amnanˆ u, (and) Ayalum king of Abattum and the land of the Rabbˆ u – these kings went to war against him. The army of Sumuepuh of the land of Yamhad then came to reinforce them, and they united together ˘ him in the town of˘ Samanum, of(?) the tribal confederacy(?) of the Yaminites, against but with a mighty weapon he defeated these three kings of the Yaminites.239

The text names the enemy by the individual royal seats of their rulers, along with three of the five Yaminite tribes. They are then identified as “Yaminite” as a collected group, using an odd, evidently Akkadianized rendition as “marmûm.”240 None of these kingdoms is defined in terms of its pastoralist “hana.” ˘ After recalling the dismantling of the Yaminite fortifications in the Euphrates valley, Yahdun-Lim goes on to describe a separate success that is ˘ probably in the north: He tore down the town of Haman, of(?) the tribal confederacy(?) of the Hana, ˘ had built, and made it a mound and a ruin. ˘Thus which all the fathers of the Hana ˘ 241 he defeated Kas.uri-hala, its king.

˘ Knowing the affinity of the Hana with the upper Habur, the town of Haman ˘ without being rebuilt, ˘ should lie in that direction.˘If Haman was destroyed ˘ we will not find references to it in the later periods that provide us with most of the Mari texts.242 Given the continuities between the two reigns of Yahdun-Lim and ZimriLim regarding Simalite tribal base, Mari capital, and ˘principal titulary, we must expect that Yahdun-Lim’s m¯at Hana was essentially like that of Zimri˘ Lim. The titles of both kings appear˘ to be grounded in a Simalite tribal population and do not apply to tent-dwellers from the defeated Yaminite ˇ peoples. Yahdun-Lim’s Samaˇ s temple inscription confirms this exclusion of Yaminites˘ from the Hana population, but the text describes victory over ˘ a “Hana” enemy that seems more limited than the king’s whole tribal kin. ˘ If we begin with the hypothesis that the Hana of Haman and Yahdun-Lim’s m¯at Hana are the same, because no effort˘ is made˘to distinguish ˘them, then ˘

154


“the fathers of the Hana” would be the collective leadership of some part of ˘ Yahdun-Lim’s own tribespeople. ˘The situation of the Hana in the Samaˇ ˇ s temple text is clarified by the ˘ second major royal inscription of Yahdun-Lim, the disc inscription for the ˘ founding of the town of Dur-Yahdun-Lim. The king proclaims: “Seven kings, ˘ fathers of the Hana who fought together against me – I defeated them and ˘ restored their land (m¯atum) to my side.”243 Once again, we encounter the “fathers of the Hana,” this time accorded the title “kings” (LUGAL/ˇsarrum). ˘ is no question that these are current pastoralist leaders In this case, there who opposed Yahdun-Lim’s unification of these “Hana” people under his ˘ Hana themselves in rule. It is clearer˘ that Yahdun-Lim had to fight the ˘ ˘at Hana. The Durorder to create the kingdom that he defined as the m¯ ˘ Yahdun-Lim text portrays Yahdun-Lim as facing the allied leadership of an ˘ ˘ existing “land of the Hana,” in that the seven Hana kings together repre˘ ˘ sent a single m¯atum. During the reign of Zimri-Lim, the Hana people are ˘ first of all the king’s own Simalite kin, and given Yahdun-Lim’s claim to ˘ rule the same m¯at Hana, we must expect these earlier Hana to be Simalite ˘ ˘ as well. ˇ It appears therefore that in the Samaˇ s temple inscription, Kas.uri-hala was a rival Simalite king who had made the fortified center of Haman˘ his royal seat, just as Yahdun-Lim was based at Mari. According to˘ the text, ˘ Haman was not a famous old center like Mari but had been built by the ˘ Simalites themselves. By these two military campaigns, Yahdun-Lim would ˘ have established his supremacy over the Yaminites of the middle Euphrates valley and would have eliminated his one evident rival for rule over the Simalite “Hana.” Zimri-Lim would thus claim to have recreated the unified ˘ Simalite kingdom that had earlier been won by his kinsman Yahdun-Lim. ˘ In spite of the continuities between the definitions of the kingdoms of Yahdun-Lim and Zimri-Lim, there is at least one important difference be˘ tween them. Yahdun-Lim actually considered himself to have two major ˘ capitals: Mari downstream and the captured Yaminite center of Tuttul upstream to the west. In the Dur-Yahdun-Lim text, the king calls himself the ˇ s temple text then “king of Mari, Tuttul, and the m¯a˘t Hana.”244 The Samaˇ ˘ makes clear that Yahdun-Lim considered the districts of the Banks-of-the˘ Euphrates (Ah Purattim) to be conquered territory, distinct from both the ˘ m¯at Hana and “Mari.” The Simalite king claims that “he established control over ˘the Banks-of-the-Euphrates” (ki-ˇsa-ad Pu-ra-tim ig-mu-ur), an achievement that he celebrates in his epithet from the Dur-Yahdun-Lim text as ˘ “the one who established control over the Banks-of-the-Euphrates” (ga-me-er 245 ´ ÍD.BURANUN.NA). GU Zimri-Lim shares this attitude toward the Ah Purattim as an entity separate from the m¯at Hana. In the long version of˘ ˘ his royal seal, he calls himself “Zimri-Lim, appointed by Dagan, beloved of Enlil, who established control over the Banks-of-the-Euphrates (ga-mi-ir Ah ÍD.BURANUNki ), king of Mari and the m¯at Hana, son of Yahdun-Lim.”246 ˘ ˘ ˘


155

When Samsi-Addu of Ekallatum took over Mari, all thought of a “land of the Hana” naturally disappeared, though most of the tribespeople would ˘ have remained, along with the common distinction of settled and mobile pastoralist populations. He does, however, give new status to the Ah Purattim, ˘ calling his son Yasmah-Addu’s realm “the land of Mari and the Banks-of-the˘ 247 Euphrates.” Samsi-Addu was not a Simalite, and we must expect the use of the word “hana” to be essentially different in texts from his sphere, including the courts ˘of his sons Iˇsme-Dagan and Yasmah-Addu, set up as kings over ˘ Ekallatum and Mari, respectively. Under the sovereignty of Samsi-Addu, the ruling circle was neither Simalite nor Hana, though by setting up his royal ˘ ˇ seat at Subat-Enlil, Samsi-Addu placed himself in the midst of Simalite and Hana territory. During this period, we do find references to hana people ˘ may not be Simalite, but the use of hana can also resemble ˘ what we that ˘ find with Zimri-Lim. In one letter, the king promises to have “the hana” mustered to raid the livestock belonging to Sumu-epuh king of Aleppo˘ and ˘ um to Samsi-Addu’s son Yasmahto the Yaminite Rabbˆ u.248 A letter from Laˆ ˘ Addu introduces a potential problem with sheep, with some “tent-dwellers” involved: The sheep of the hana and of the muˇskênum of the Ah Purattim have (just) crossed ˘ of Samaˇ ˇ over to the wadis. ˘Eleven herdsmen under the authority s-muˇsallim and all the Yumahammˆ u hana have crossed. They are pasturing (their flocks) (near) Dur˘ and ˘downstream. I hope that the Yaminites who have not sworn loyalty Yasmah-Addu ˘ not quarrel [with them and] so do them harm.249 oaths will

Both of these two letters set the “hana” against the “Yaminites” in a way ˘ similar to the custom of Zimri-Lim, though the specific definitions may not be the same. At one level, the distinction may reflect the affiliation of people who call themselves Hana with the upper Habur, in no case a Yaminite ˘ ˘ dealing with Simalites of stronghold. In the second letter, we are clearly the Yumahammû tribe, and the writer, Laˆ um, is already familiar with the ˘ of “Hana” that will prevail under Zimri-Lim. Simalite use In general, the˘ pattern in the use of hana terminology follows the pres˘ ence, absence, and restoration of Simalites in power at Mari. The main difference between the periods of Yahdun-Lim and of Zimri-Lim may have ˘ a significant component of his kinsbeen that the earlier king had to fight men to establish unified rule over the Simalite Hana, while Zimri-Lim did not. Even though Bannum was responsible for ˘the actual conquest of Mari, Zimri-Lim took the throne without internecine combat. If Haman was indeed somewhere in the Habur region, which must remain a˘ guess for now, then the conflict there˘ would have anticipated a continuing tension between Zimri-Lim and his kinsmen in that area. Perhaps the battle with Kas.uri-hala left scars that could be exploited by Samsi-Addu, who main˘ tained friendly relations with people called Hana in the upper Habur. By ˘ ˇ hna from Ekallatum, Samsi-Addu relocating to Se placed himself˘ right in ˘

156


their midst, to some extent lessening his foreign status. Also, if the family of Yahdun-Lim and Zimri-Lim removed itself to the west and north to wait for ˘ opportunity, many Simalites may have remained comfortably, willing a new to fight for the new king as long as it was not against their own kin. Durand seems to come to a similar conclusion.250 In claiming to rule a m¯at Hana, Zimri-Lim declared himself the heir of ˘ been established by Yahdun-Lim. Surely, he the Simalite kingdom that had ˘ on that mode of was not attempting to unify all mobile pastoralists based life alone, but rather he was founding his kingdom on a more concrete and achievable tribal unity under the banner of a Simalite confederacy. A further question remains, however. Why should Yahdun-Lim and Zimri-Lim choose ˘ tent-dwellers, rather than the to call themselves the kings of the Simalite kings of Simal as such? The only answer I can suggest is that all Simalites in these two kingdoms identified themselves according to an ideology of mobility, which came to give the Hana name a strong tribal association that ˘ took a variety of forms. When Zimri-Lim renewed Yahdun-Lim’s claim to rule all tent-dwelling Hana, he would have been claiming˘ at the same time to lead all the Simalites,˘who now emphasized their nonsedentary heritage by this name. By the prominence he gave to the distinct governance of Simalite pastoralists under two merhûms, Zimri-Lim allowed the mobile hana ideology ˘ a central role founded in ˘concrete political structures. d. Reflections on ARM VI 76, the Hana, and the Akkadians. Zimri-Lim’s official ˘ titulary is not the only duality found in the Mari archives to define his kingship, though it is the most prominent and certainly important for its formal definition of the realm. Another provocative definition appears in a letter from Bahdi-Lim (ARM VI 76), district governor of Mari, who divides Zimri˘ Lim’s people between “Hana” and “Akkadians.” In the relevant passage, ˘ words to the king at an occasion made unforBahdi-Lim recounts his past ˘ tunately obscure by the condition of the tablet (see Charpin and Durand 1986, 143–5; Durand 1998, 485–8): [It is] a known fact(?) (that) when (we were) [in the camp(?) of] Appan, I spoke [to my lord] as follows: “[Today,] the land of the Yaminites(?) [has been handed over to you(??)]. That land is covered(?) with the garment(??) of the Akkadian. [My lord] must honor the head of his kingship. [Just as] you are the king of the Hana, [so] ˘ ride a you are secondly the king of the Akkadian(s). [My lord] must not (therefore) horse. My lord must (rather) ride [on] a litter and mules, if he is to honor the head of his kingship.” I advised my lord in this way.251

Durand interprets this as a meeting at Appan (line 12), after the [ma]-a-at I[a-mi-n]aki became Zimri-Lim’s (lines 15–16).252 After some statement regarding the identification of “this land” with Akkad, Bahdi-Lim encouraged ˘ (lines 17–19, 24– Zimri-Lim to bring honor on “the head of his kingship” 5). Because he is king of both Hana and Akkadian populations, he should ˘


157

participate in processions with a n¯ubalum (litter?) and mules, rather than with a horse. The use of the n¯ubalum would satisfy the western, Amorrite, and tent-dwelling element of his realm, while the riding of donkeys or mules would have been more Akkadian, as seen in texts that depict dignitaries of Eˇsnunna as “donkey-riders.”253 The horse would not have been known as a royal mount. (1) The Akkadians and the Ah Purattim. What is the point of the above passage? The last advice should˘ indeed be to avoid offending the local “Akkadian” population at this occasion by using foreign horses, as argued by Durand. It is not clear to me that the donkey can be considered the particular mount of “Akkad” or Eˇsnunna kings, however. For one, not one of the “donkey-riders” is a king, and only one set of twelve are from Eˇsnunna, captured in a military encounter (ARM XXVII 16). Another example seems to involve a Babylonian army (XXVI 131), but the last requests a representative from Zimri-Lim to be sent to one of the Ida-Maras. kingdoms and has nothing to do with the Akkadian region (XXVI 312). In general, it is not clear that donkeys are specially identified with southern or eastern Mesopotamian leaders. Bertrand Lafont (2000a, 214–15) cites iconographic evidence for donkeys or mules from both an Ur III seal impression (far southeast) and a Byblos dagger sheath (far west).254 In the end, it seems that all of the choices can be western, and none of them makes sense as specifically “Akkadian,” if by that we mean the region that came to be identified with the great kingdoms of Eˇsnunna and Babylon. Somehow, the offending horse should represent another cultural distinction, rather than that of east versus west. The most obvious answer would be not that the horse is western or “Amorrite,” which are not the categories given in the letter, but rather that it is the preferred mount of pastoralist tent-dwelling hana. Bahdi-Lim’s admonition makes it clear that ˘ this animal. ˘ Zimri-Lim himself preferred A letter to Zimri-Lim from another governor assures him that out of the various animals from an arriving caravan, two white horses have been set aside for the king.255 At least the n¯ubalum and mules will be impressive and acceptable to everyone, so that the king will not come across as a wild man from the steppe. If neither the n¯ubalum nor the mules can be associated with Akkad or Babylonia, then these “Akkadians” cannot be identified by any direct link to the original designation. Who are they, then? Durand also addresses the term (LAPO 17, pp. 480–3), rejecting the idea that these might be a remnant pre-Amorrite population of the older towns in the Euphrates valley. A more direct link with “Akkadian” aspirations, he says, can be found in SamsiAddu’s Mesopotamian kingdom, and the present danger would be to irritate holdovers from that previous administration. It does make sense to look for a more current identification of the Akkadians. The double identity as king of the Hana and king of the ˘

158


Akkadians, however, echoes the frequent duality in the constitution of ZimriLim’s kingdom from Simalite Hana and Ah Purattim components, and it ˘ pair here˘ has a more concrete point of is worth considering whether the reference. If Zimri-Lim was both a Hana king and an Akkadian king, it is ˘ with its Ah Purattim domain based also because he now ruled both Mari, ˘ in the Euphrates valley, and a Simalite tribal domain defined by its mobile pastoralist component. Just as the tent-dwelling component of his rule derived specifically from the Simalites, the Akkadian aspect should be the whole kingdom that he conquered from Yasmah-Addu, the Ah Purattim. ˘ a focus on towns ˘ The Akkadian part of his realm would have retained not because these were pre-Amorrite, but because they were the building blocks of the land he had seized. This solution would still take advantage of Durand’s identification of Samsi-Addu’s kingdom with ancient Akkad. (2) The Akkadians and the Yaminites. There should be little question that the direct point of reference for “the Akkadian” in ARM VI 76 is Samsi-Addu’s kingdom or, more precisely, the part of it conquered from Yasmah-Addu. What did this represent in terms of actual population, however? ˘In fact, the Euphrates valley kingdom may in this period have taken on a particularly Yaminite hue. I do not mean to overlook the other tribal peoples, including the Simalites, who occupy the area, but Durand’s suggestion of a Yaminite connection provokes the question.256 Beyond the evidence already presented for the prominence of Yaminite towns in the Terqa district, they are strongly associated with the middle Euphrates, farther upstream. The Yaminites have both plural towns and a collective m¯atum from which they threaten Imar and Tuttul, both towns with a Yaminite heritage themselves, and of course the core of the Yaminite land conquered by Yahdun-Lim lay ˘ in this region:257 All the Yaminites have left their towns and their land, and they are on the roam. Now some of them are scheming as follows: “We should seize [the town of] Tuttul as a fortified base, so it will be . . . , and (then) we should (also) take over Imar.” The Yaminites are contemplating this.258

ˇ Of the Yaminite towns mentioned in Yahdun-Lim’s Samaˇ s temple inscrip˘ tion, only Tuttul has been located precisely, at Tell Bia, near the junction of the Balih and the Euphrates Rivers. Abattum appears to have been fairly ˘ close to Imar, upstream from Tuttul. The exact location of Samanum is unknown, though it was evidently much further downstream. During the reign of Zimri-Lim, Samanum belonged to the district of Terqa.259 It is interesting to observe in this regard that Anbar found a distinctly higher percentage of Akkadian “ethnic” (personal?) names among the Yaminites than among the “Hana,” 23 percent against 7 percent.260 The ˘ number of Akkadian names among the Yaminites might confirm the possibility that a larger proportion were acculturated to the older traditions of


159

the Euphrates than among the Simalites. Another unpublished text divides Zimri-Lim’s domain into groups defined by language tradition, whereby he is “king of the Akkadian and the Amorrite.”261 If the “Akkadian” part of Zimri-Lim’s kingdom was dominated by a subset of the Yaminites that both preserved living tribal custom and identified further with the great kingdoms of the east, this might explain how a n¯ubalum and mules would satisfy the cultural sensitivities of an “Akkadian” element of the population in the letter from Bahdi-Lim. ˘

2. The Tribal Kingdom of Zimri-Lim As kings of “Mari and the m¯at Hana,” Yahdun-Lim and Zimri-Lim stood in ˘ fortified administrative and rita tradition of tribal kings who ˘ruled from ual centers. What makes the Mari archives so unusual and so valuable is the correspondence that displays the active tribal associations of the king at Mari. Cuneiform writing follows institutions that are grounded in sedentary life, most often defined by their service of palaces, temples, or town-based commerce. Tribal definitions are rarely central to the affairs of these institutions. Although kings with West Semitic “Amorrite” names and tribal backgrounds occupied all of the principal thrones of Mesopotamia in the early second millennium, we must not make any one of them the norm for defining the tribal aspects of political organization in this period. Two of the most powerful Amorrite kings left us evidence for their particular tribal heritages that nevertheless suggests alternative forms of acculturation to nontribal political traditions. Zimri-Lim was an active tribal king, a king still defined by his tribal constituency, more visibly than in any other documentary evidence. A number of other Mesopotamian kings from the early second millennium identify themselves or their forebears in Amorrite tribal terms, and these show that Zimri-Lim was by no means unique. What is unusual is the breadth of evidence from a ruler with fresh, first-generation contact with that tribal base. Also, it may be that the Amorrite conquerors of eastern Mesopotamian cities may have acculturated more fully to that powerful tradition. Both Hammurabi of Babylon and Samsi-Addu of Ekallatum and the ˘ Mesopotamian kingdom remember family connections with Amorrite tribal peoples well known from Mari: the Yaminite Amnanû and Yahru(i)rˆ u, and ˘ a (Charpin the Numhâ, the tribe that retained a lesser political seat at Kurdˆ ˘ 1986, last section). Samsi-Addu’s father, Ila-kabkabu, is known, and Durand but this family did not claim a long-standing identification with any one fixed base.262 Perhaps this lack of attachment to a single capital helps explain the freedom with which Samsi-Addu handed off his base at Ekallatum to his son Iˇsme-Dagan, reveled in the prestige of the old city of Aˇsˇsur, and moved his ˇ hna, where he renamed it Subatˇ own central administration to the city of Se ˘ with a stronger attachment Enlil.263 Hammurabi has a very different profile, ˘

160


to a single place, his dynastic seat at Babylon. He is the fifth known king from the same line at Babylon, and his titulary makes no reference to any tribal affiliation.264 I have already reviewed the additional evidence for the dynasties of Kudur-mabuk at Larsa and Sˆın-kaˇsid at Uruk, which were linked to the Yamutbal (Emutbal) and Yaminite Amnanum tribespeople, respectively (see Chapter 2, section A3, and Chapter 3, section B4). The case of Larsa seems to show that the Yamutbal tribal identity gradually lost its prestige and its priority, so that Rim-Sˆın finally gave up any reference to it in his titulary. a. The Simalite king. In the same circle of Amorrite kings, Hammurabi of Babylon calls his colleague not “king of Mari” but “king˘ of(?) the Simalite(s?),” a tribal definition, whether translated “the Simalite king” (Charpin) or “the king of the Simalites” (Anbar): Rim-Sˆın (king of Larsa) sent out these hostile words. Except for the great gods who [came] to my aid, it was Zimri-Lim, the king of the Simalite(s), who put his life on the line with me again and again – there was no one else.265

This title is given in praise, without any hint of denigration, as an identity to be prized, if not as a full description of his rule.266 During the period leading to Babylon’s capture of Mari, king Hammurabi of Kurdâ addresses ˘ Kurdâ, to persuade them to the assembled Numhâ, the tribal people of ˘ abandon any loyalty to Zimri-Lim. At the end, the king of Kurdâ announces ´ s ´ that “the Babylonians” (lu.meˇ KA.DINGIR.RAki ) want him to join them, so he has decided to declare war on “the Simalites” (DUMU Si-im-a-al): “ . . . The king of Babylon [has sent(?)] me [troops(?)]. [(As of) to]day, Zimri-Lim has never, earlier or later, come in solidarity with the Numhâ. At the time of Qarni˘ the time] of Atamrum Lim, he came up (instead) in solidarity with the rebellion. [At he came back once again. Now he is about to go in solidarity with Himdiya also. At this moment, the Babylonians want to unite with me, and I want to˘ declare war on the Simalite(s).” The Numhâ answered him as follows: “Our brother Simalite(s) aside, we are at ˘ peace with Babylon, but time and again we have put our lives on the line with the Simalite(s).” This is how his land [answered him].267

Even where Hammurabi of Babylon may recall a Yaminite heritage, his realm is defined by˘ its capital city, while Zimri-Lim’s kingdom is identified by his tribe. An earlier and friendlier king of Kurdâ, Bunu-Eˇstar, is reported to have rebuffed the advances of a neighbor with the loyal statement, “Without Zimri-Lim and the Hana, I can enter no peace agreement.”268 The Hana ˘ are surely Simalite in˘ this context, and Zimri-Lim is thus regarded as a tribal 269 Eˇsnunna’s king Ibal-pi-el II may assume the same tribal identity king. ˇ for Zimri-Lim’s domain when he names a year for the attack of Subartu and Hana, the latter representing Mari (Charpin and Durand 1986, 148). ˘


161

During the time of the Mari archives, the word “hana” retained an active use for pastoralist tent-dwellers that may eventually ˘have been lost in the name of the “Hana” kingdom of the next period, with its center at Terqa. Closer˘ to home, the formula by which a district governor of Saggaratum swears loyalty to Zimri-Lim supposes that the outcome will have an audience of all Simalites, who will say that he will receive his just deserts if he brings on himself the curses on disobedience: May these gods wipe out all my descendants (“the seed of my name”) and my offspring. May the Simalite(s) see their . . . by [my] example, saying, “Zimri-Lim [showed favor(?)] to Sumu-hadû, and Sumu-hadˆ u . . . [(something bad)].270

˘ ˘ So far as this oath protocol was created in the very court of Zimri-Lim, it represents the king’s own preference in describing his true constituency.

b. Samsi-Addu, king of Akkad. Yahdun-Lim and Zimri-Lim considered them˘ selves to be the heads of a Simalite kingdom of pastoralists who had established a capital in the famous old center of Mari, in the Euphrates valley. Samsi-Addu’s reign presents a completely different political and geographical vision. When he set up his sons Iˇsme-Dagan and Yasmah-Addu to rule ˘ over the eastern and western domains of his kingdom, Samsi-Addu created a realm of the two great Mesopotamian rivers, “the land between the Tigris and the Euphrates.”271 Above all, however, Samsi-Addu appears to have defined himself as the heir to the glorious kingdom of Akkad. Several centuries earlier, during the late third millennium, the dynasty of Sargon had ruled a celebrated Mesopotamian empire from the central Mesopotamian city of Agade (Akkad), near later Babylon and modern Baghdad. By laying claim to the status of Akkad, Samsi-Addu made himself the superior of Eˇsnunna and Babylon, the rivals on his southern frontiers. This provocative hypothesis has been proposed by Charpin and Durand in notes, without full articulation.272 The idea is based in part on Samsi-Addu’s preoccupation with the great kings of Agade. The kispum ritual found at Mari names only two individual ancestors, Sargon and Naram-Sˆın.273 Mari now turns out to have left us a copy of the text known as “The rebellion against Naram-Sˆın,” probably from the Samsi-Addu period as well (Charpin 1997c, 17). In the dedication for a goblet, Samsi-Addu calls himself “king of Agade.”274 Durand will even propose that Samsi-Addu’s family line, going back to Ila-kabkabu, claimed roots in the city of Akkad (Charpin and Durand 1997, 372n36). It is interesting that this identification with Agade has nothing to do with the common boast of southern Mesopotamian kings from this period that they rule “Sumer and Akkad,” a claim never made in Samsi-Addu’s titulary.275 Zimri-Lim also laid claim to a prior realm in taking Mari and the Ah Purattim, but he equally considered himself the king of a m¯at Hana, defined˘ ˘ predecessor by a combination of urban royal seat and tribal population. His

162


could rule all tribal peoples because he laid claim to the whole land of two rivers, but Samsi-Addu felt no need to satisfy a tribal constituency in his essential titulary, in spite of large numbers of such peoples in his domains.276 Durand suggests that Samsi-Addu may even have pushed large numbers of hana pastoralists outside the geographical bounds of the kingdom. This king ˘ pursued a territorial vision of the kingdom, defined by its two rivers, not by its people. Zimri-Lim, in contrast, found it essential to retain a Hana element in his titulary because this was his true base of power. The Ah˘Purattim was just the achievement of his conquest, with Mari its fortified ˘administrative and symbolic center, the requirement of kingship.277 We do not know what settled seat Zimri-Lim occupied before his return to Mari, nor what group or territory he may have claimed to rule. By contrasting Zimri-Lim with SamsiAddu, I do not judge either as the norm. The Mari archives alone show that there were other kingdoms that were defined tribally, such as Kurdâ as the land of the Numhâ tribe, and this pattern was probably common. It remains rare, however, to˘ find the active tribal kingdom so fully documented as in the correspondence of Zimri-Lim. c. The king away from Mari. The hard question that remains is whether the different ideologies of the two kingdoms that give us the bulk of our Mari texts can be shown to be reflected in different realities, in the conduct of their kingships. I offer here only a few suggestions, and this problem deserves much closer examination than I can undertake here. The continuity of palace administration and harems from Yasmah-Addu ˘ this to Zimri-Lim does not prove the likeness of their kingships, because takeover is the natural accompaniment of conquest.278 Moreover, Yasmah˘ Addu’s administration was based in part on the institutions left in place from 279 Zimri-Lim the reigns of Yahdun-Lim and his hapless son Sumu-yamam. ˘ to be taking back his own, as heir to Yahdun-Lim.280 considered himself Samsi-Addu and Zimri-Lim also share a tradition of the˘ king as warrior that might be understood as a purely tribal characteristic, but this is surely known outside these circles as well, wherever states expand their bounds by force. More specific to tribal peoples may be the more limited ideal of life in the steppe, as maintained by their mobile pastoralist components. In the fascinating letter that Marello entitled “Vie nomade,” sent by the Yaminite ruler Hammi-iˇstamar of the Yahrurˆ u tribe to Yasmah-Addu of the Yarihû ˘ ˘ peer for his lack of ˘interest in life in the ˘ tribe, Hammi-iˇ stamar berates his ˘ open air. This failure means that Yasmah-Addu has fallen out of touch with ˘ his own natural power base (A.1146, in Marello 1992). One way to explore a possible difference in the active tribal kingship of Zimri-Lim might be through his frequent absences from the capital city, not just for battle but for a wide range of activities. First of all, the palace archives have some consistent gaps that indicate Zimri-Lim’s absence for extended periods. Durand concludes that while the main Mari palace may


163

have functioned as administrative center for the whole reign of Zimri-Lim, it was used as the residence for the royal family only when the city walls were needed for safety (1997, 36). If the royal family was located here during times of war, those periods would have required the longest absences of the king himself, on campaign. Between the demands of war and the freedom of peace, Zimri-Lim appears to have lived up to the nomadic ideal more than one might have thought possible for a Mari king. Bertrand Lafont has now addressed the question of Zimri-Lim’s absence from Mari in systematic terms, gathering all of the king’s documented voyages. He finds that Zimri-Lim left Mari roughly ten times, most often in order to go to war with his troops. Other kings do not seem to show the same capacity for travel from their secure home bases: Par contraste, quand on regarde l’incursion e´ lamite dans le domaine nordmésopotamien a` partir de l’année ZL 8 [=9], ou les menées de Babylone dans la ´ région, on a en revanche l’impression que jamais le sukkal d’Elam ou Hammu-rabi ne sont personnellement déplacé ( jusqu’à preuve du contraire) pour conduire ou contrôler les opérations menées en leur nom. (2001a, 303)

Lafont assembles the following list of Zimri-Lim’s outings (dated by years Z-L 0–13): r Z-L 0, conquest of Kahat and campaigning in Ida-Maras (in Habur basin); . ˘ Aˇslakkâ of Ida-Maras; ˘ r Z-L 3, campaign against . r Z-L 4, return to Habur region to fight Eˇsnunna; siege of Andarig (east of ˘ Habur); ˘ r Z-L 6, two separate voyages: early in the year, to establish peace between ˇ Andarig and Kurdâ (east of Habur); midyear, to Subat-Enlil (northeastern ˘ Habur basin); ˘ 7, enters Suhˆ r Z-L um, in the southeastern border region; ˘slâ to make offering to storm god of Kummê and meet r Z-L 8, trip to Huˇ with northern ˘kings; r Z-L 9–10, long voyage to the Mediterranean coast, including Ugarit (lasts five months); r Z-L 10, return to Habur to help Razamâ against a siege by Elam and ˘ Eˇsnunna; r Z-L 12, visits to Ilansurâ and Andarig, against threats from Iˇsme-Dagan of . Ekallatum; and at the end of the year, to put down a revolt by Aˇslakkâ. Porter addresses Zimri-Lim’s predilection for travel directly as one of two mechanisms at work during his reign to transcend the effects of tribal dispersal, along with rites for royal ancestors. According to Porter, the king’s expeditions extend space, not only to display royal power, but also to integrate a tribal people whose identity is not founded on shared residence with regular face-to-face contact (2000, 447–51). Porter’s analysis of Zimri-Lim’s Mari

164


kingdom is remarkably close to my own, though it is based mainly on her archaeological evidence for the social framework of the middle Euphrates: The critical point to make is that pastoralist groups in the Mari texts are not external, and intrinsically hostile to the kingdom, . . . but were integrated into its political, economic and social arenas in a series of different, sometimes crosscutting, sometimes intersecting relationships that were contingent on specific operative circumstances in both spheres. (p. 445)

How does Zimri-Lim’s predecessor compare? Yasmah-Addu may have been very young when he took the throne of Mari under˘ his father, SamsiAddu, and his reputation as a helpless hedonist is probably undeserved. He was to lead a campaign to Qatna, until he was called away to join IˇsmeDagan in the fight at Qabrâ (LAPO 17, pp. 9–10). Setting aside the idea that Yasmah-Addu was a weak stay-at-home, it does still appear that he spent much time˘and energy making the palace an attractive place to be during his time there.281 If oversimplification can be avoided, there may remain a real difference between how Yasmah-Addu and Zimri-Lim handled palace life during their respective reigns. ˘ d. The two rivers of Zimri-Lim. King Zimri-Lim saw himself as the heir to the Simalite Hana kingdom that had been unified under Yahdun-Lim, but in ˘ immediately inherited the kingdom of Yasma ˘ h-Addu, son of fact, he more Samsi-Addu, which had been defined in quite different terms.˘ With the Mari center, Zimri-Lim inherited two streams of royal ideology available to define the essence of what he ruled. Yahdun-Lim’s earlier use of the phrase “Mari ˘ identified his kingdom with the Simalite and the m¯at Hana” appears to have ˘ tribal confederacy, with “Mari” as the royal seat, the captured urban prize, and all its heritage of wealth and prestige and human achievement. On a smaller scale, this was the framework of each defeated Yaminite polity, each with its town center and tribal population: “Samanum and the m¯at Uprapˆım,” “Tuttul and the m¯at Amnanˆım,” and “Abattum and the m¯at Rabbˆım.” One could add “Kurdâ and the m¯at Numhê” and others from the period of Zimri-Lim. The capital receives pride of˘ place as the seat of the king, while the m¯atum was regarded as the population surrounding the central city. There is no reason not to understand Yahdun-Lim’s use of the ˘ The Hana need not title “king of Mari and the m¯at Hana” in the same way. ˘ ˘ symbol. Any be all Simalite tent-dwellers in fact, but they become so in 282 people outside the core are set aside as secondary. Under Samsi-Addu, the whole structure changed. A real administrative duality accompanied the ideal of a Mesopotamian kingdom, framed by the Tigris and Euphrates Rivers. The latter was the Ah Purattim, and when Zimri-Lim used this term he consciously took over ˘a whole unit of SamsiAddu’s kingdom. When Zimri-Lim restored the titulary of Yahdun-Lim, he ˘ a kingdom did not entirely erase the influence of Samsi-Addu’s idea of


165

defined by and administered around two river-centered regions. Even ZimriLim’s titulary acquires hints of a geographical duality, based on the Habur ˘ focus of Simalite pasture land and the concentration of Simalite settled populations. One could even imagine another land of two rivers, one divided between the Euphrates and the Habur. Remember that the main body of Zimri-Lim’s vassals were northern,˘in or adjacent to the Habur drainage, and ˘ Moreover, Zimrisome of these had tribal ties to Zimri-Lim’s Binu Simal.283 Lim’s governance of the mobile pastoralist Simalites under the independent leadership of two merhûms was primarily oriented toward the upper Habur, ˘ as well, so that even ˘this administrative duality had a redefined two-river cast to it. By this interpretation, then, Zimri-Lim allowed the old title of Yahdun˘ Lim to work for him in two different ways. Zimri-Lim enjoyed the benefits of the original glorification of his tribal ancestry, as king of a Simalite state, centered at Mari. At the same time, however, he allowed the first element of the old title to acquire a larger significance. “Mari” brought with it the whole concentration of settled population surrounding it in the Euphrates and lower Habur valleys, as the “Banks-of-the-Euphrates,” or Ah Purattim. ˘ It is important to keep in mind that we are dealing not ˘with strictly bounded populations so much as with ideologies that overlap with rough divisions among the people that are always inadequate to social reality. The “Hana” ideology originates among Zimri-Lim’s own tribal people, with its ˘ idealization of the mobile aspect of their lives. Ultimately, the Ah Purattim ˘ long behas roots in the people oriented toward the Euphrates, probably fore the Simalites of Yaggid-Lim and Yahdun-Lim came on the scene. If ˘ the people of the Ah Purattim also had mobile pastoralist connections, they ˘ would have gone back to much earlier times, in the third millennium.

3. Kingship at Mari This study of the political world of the Mari archives is focused especially on those aspects that represent “democracy’s ancient ancestors,” that involve the collective exercise or interpretation of power. Having just explored the basic definitions of the Mari kingdoms, I should say something about kingship itself, particularly as it relates to group-oriented forms of leadership. The following sketch calls for deeper and more systematic evaluation of each feature, and I offer it in order to recognize the larger relationship between any political collective and the king.284 Many of the texts that I cite in the following chapter show how kings may be served by or interact with plural leadership defined as “elders” or in other collective terms, sometimes identified with the town or population as a whole. In these cases, royal or other individual authority comes face to face with the collective political tradition, with results that vary as widely as the possibilities of wielding power in myriad circumstances. To

166


my knowledge, not Zimri-Lim, Yasmah-Addu, nor his father, Samsi-Addu, is ˘ ever portrayed as addressing any gathered council that is identified with the kingdom as a whole or even with the palace administration. It seems, then, that the Mari kings encountered the collective voice of their people only through the mediation of high officials who might meet groups representing subsets of the whole population. At the least, the written expression of Mari royal affairs does not acknowledge such a corporate ideology of Mari leadership. This does not mean that the Mari kings operated autonomously. On the contrary, the Mari correspondence shows a constant dialogue, usually deferential to the king, but full of opinion and advice. A man such as Bannum, who had captured the very city of Mari, could address his lord as just less than a peer and could make decisions so independently that Zimri-Lim would have to remand them.285 In ideological terms, however, each high official who enjoyed direct contact with the king could address his lord only from the stance of an individual political player in formal submission as a royal servant. Although each king appointed a large administrative team, he never addressed them as a team. Members of the palace administrations carried titles that could equate their ranks with those of other officials, but every administrator served the king directly and separately, perhaps under an individual supervisor, but never allowed the force of any collective. We cannot be sure that such collective political forms did not exist, but the very absence of textual references reflects the dominance in written expression of a royal and exclusionary ideology of power. In this key respect, the Mari archives show little difference between the reigns of Samsi-Addu and of Zimri-Lim. Both Yasmah-Addu and Zimri-Lim ˘ ruled from Mari with an inner palace circle of senior leaders, somewhat separate from the “governors” (ˇsa¯ pit.um) who mediated between the palace and local officials of the sedentary population. The merhûm chiefs of pasture were ˘ especially prominent under Zimri-Lim but still served as mediaries between the king and the local leadership of the herding groups. Under Samsi-Addu, an added level was created by the delegation of local rule to Iˇsme-Dagan and Yasmah-Addu, with three full-fledged palace administrations. ˘ power touched many aspects of social and economic life, and there Royal may have been no domain where one could directly refuse royal authority, if the king dared to make a demand. The palace certainly played some role in most group activities within the bounds of the Mari kingdom. Major festivals were centered in Mari, and the only ritual texts found at Mari give the king a prominent role.286 Naturally, the administrative role of the palace intruded into many parts of everyday life, especially through local district officials who could demand labor on work crews or the arrest of men wanted by the king. The king took income from every town and herding group through the sug¯ag¯utum fees paid by the local sug¯agum “leaders” for the stamp of royal approval on their authority.


167

Considerable land was said to be owned by or cultivated for the palace,287 and Durand concludes that the valley floor by the river could be exploited only with palace permission.288 It may be that the king could make land grants as he wished, but there was clearly much farmland owned on terms that the palace acknowledged to be defined as belonging to a muˇskênum class, outside immediate dependence on royal support. Durand warns that palace inventories of fields should not be taken to encompass all agricultural land, but only the fields that required royal intervention in order to arrange their status or their distribution. We do not know the actual percentage of royal land for each location (1990e, 101–2).289 The king did not consider himself to own all land in practical terms, at least. All consideration of land ownership applied only to the river valleys. In the steppe (nawûm), pastoralists moved and grazed their flocks without the restraints of such individual ownership. It is difficult to know the extent of palace funding or administration of local farming infrastructure, such as canal work. Royal officials certainly got involved in construction and maintenance of irrigation systems.290 Outside of farming, the palace was definitely involved in trade and a variety of craft production. Here again, however, it would be dangerous to assume that the merchants and craftsmen and women who were employed by the palace received all of their income from the king.291 One indication that palace employment did not guarantee full financial support is the situation of soldiers stationed in the vicinity of Mari, who were expected to provide for themselves until their actual departure on campaign (Durand 1998, 397). The one major sector of the economy that does not seem to have been dominated by the palace in the Mari archives is animal husbandry. While the district governors and their local palaces were deeply involved in the farming economy of the river valleys, there appears to have been no corresponding palace administration for the grazing of sheep in the nawûm steppe. Bertrand Lafont informs me that few of the palace administrative texts concern the breeding and management of flocks, and there is palace involvement only during the shearing season. One does find reference to animals belonging to the palace in letters, but these seem to be concentrated in the correspondence of Samsi-Addu’s kingdom.292 It is worth noting that even under the famously centralized administration of the late third-millennium kingdom of Ur, only certain aspects of husbandry occupied the palace’s energy. The texts from Puzriˇs-Dagan (Drehem) show the management of extensive herds, but no concern to manage incoming animals and no involvement in the distribution of wool or dairy products (only meat; Zeder 1994, 185–7). Much was left to the herdsmen, who presumably had a long tradition of independence in such matters. Mari’s large archives do not display even this level of interference, however. Given the importance of pastoralism to the region, this omission from large-scale administrative oversight represents a serious qualification to the

168


practical exercise of royal power. It is no wonder that the tribal political structures of the mobile herding groups show the degree of independence that will be apparent with the rihs.um “talks” of the Hana (Chapter 4). Other˘ ˘ wise, the degree of palace involvement in local affairs seems to have been very much a function of distance. When we explore the traditions of collective leadership in towns, we find that it is Tuttul that retains a vigorous sense of local autonomy, in spite of the presence of a royal representative. Tuttul stood at the western end of Zimri-Lim’s kingdom, along the Euphrates River but far upstream of the territory ruled by the system of districts and governors.293 No domain or population was considered free from king Zimri-Lim’s authority. He was, after all, the king of the m¯at Hana, a realm defined by his ˘ own Simalite pastoralist kinsmen, whom he ruled with the least intrusion into their local affairs. Actual control, however, would have been a matter of push and pull with the various constituencies of the realm. Although ZimriLim had his ultimate political base in the Simalite Hana, he inherited from Yasmah-Addu the Ah Purattim domain of the river˘ valley, which could be ˘ ˘controlled through district palaces. Royal control was more systematically exercised on a local basis through towns, as seen by the fact that the Yaminites in Zimri-Lim’s kingdom were recognized only through their settlements, by which these rival tribespeople could be better monitored. Outside the core districts, however, the towns also maintained a greater ability to push back against royal power. One crucial expression of palace control was the census, which was undertaken only in order to raise military troops (Durand 1998, 332–53). Within the core districts, the king demanded the authority to approve all movements of the settled population, even if he could not always exercise this power.294 We have one body of evidence for a strong central control exerted in a particular domain: the flow of information. The very existence of the enormous correspondence of Zimri-Lim reflects his desire to know everything that went on in his kingdom generally and in the specific areas of each servant’s responsibility. Every official felt obliged to pass on to the king every morsel of information that the ruler might one day find significant. A diplomat named Nur-Sˆın defends his dogged commentary with the following reminder: Previously, when I was on duty in Mari, I passed on to my lord any word that a male or female prophet (¯apilum/¯apiltum) might speak. Now I am on duty in another land (m¯atum) – will I not send to my lord what I hear and (what) they say? If in the future any misstep is found to have occurred, will not my lord say as follows? “The word that a prophet spoke to you so as to claim your plot (of land)(?), why did you not send it to me?” Therefore I have (now) sent to my lord.295

This attitude is demanded by the stipulation in almost every extant loyalty oath from the period of Zimri-Lim that every bit of news and every hint of


169

disturbance be passed on to the king. The diviner’s oath promises to pass on every omen finding and every secret, especially where rebellion is in view.296 When the pastoralist Hana, a district governor, or a vassal town swear loyalty ˘ to Zimri-Lim, they promise above all to pass along all information that might be of interest to the king.297 The riches of the Mari correspondence are ours in some part because Zimri-Lim’s administration pursued this control of information with such zeal (see Lafont, 1997b). Within these royal systems, power was always a matter of negotiation, not merely with outside political entities but also with every internal constituency. We do not find, however, that the royal palaces of Zimri-Lim or Yasmah-Addu were settings for explicit collective decision making. For this, ˘ look elsewhere in the political landscape of the Mari archives, and we must this is the subject of the next chapter.

4 The Collective and the Town

At the most superficial level, the political world of the Mari archives may seem to consist above all of a community of kingdoms, some large and incorporating several major centers, some much smaller and dominated by a single town or fortified center. The texts from Mari allow us to see another expression of political organization in the tribe. Tribal structures overlap the landscape of kingdoms, capable of cutting across the bounds of what a single ruler could control, often contributing the kings themselves, who like Zimri-Lim might define their kingdoms by what they could control from their own tribal base. At a deeper level, however, both the kingdoms and even the large tribal groupings, such as the Binu Simal and the Binu Yamina, were superimposed on the smaller communities of ancient subsistence. During the time of the Mari archives and long before them, these consisted of two main elements that were profoundly interdependent: the pastoralists who lived mainly in portable camps and the farmers of permanent settlements whose livelihood was bound to what they could grow. Each group relied to varying degrees on what the other produced, and the Mari texts show that settlement and steppe were often joined within tribal social structures. Even on the small social scale of families, pastoralists and farmers would often have been closely related. Whether part of tribal structures or not, however, the settlement became a focus for identity and exercise of power: fortification or at least clustered dwellings for defense, impressive buildings to display the glory of rulers or gods, concentrations of people to facilitate agriculture and exchange. This is an oversimplified sketch, as I attempt to account for the social phenomenon that stands at the center of archaeological study of the ancient world: the town as hamlet, as village, as city; as political, military, or symbolic center. In every case, it is defined by being fixed to one spot, as well as by the clustering of people. This clustering may be for residence, as most easily imagined, but I do not want to exclude clusters of buildings that served a much larger population than could actually have been 170

The Collective and the Town

171

housed permanently. Some settled centers would have seen large assemblies only on special occasions, whether for celebration or in the face of crisis.1 In any case, what I assign the generic label of the “town” was by its very nature collective, the shared progression of its residents or of all who used it or assembled there. I have already said in Chapter 1 that both the Sumerian and the Akkadian languages identified every type of settlement by a single primary word, Sumerian uru and Akkadian a¯ lum. These Mesopotamian “towns” were not designated by their size, by their fortification, or by the presence of any dominant leader. In political terms, the language of the “town” seems always to have been corporate in nature, even when subsumed into the exclusionary notions of kingship. Much of the political activity visible in the Mari archives revolves around entities defined by their towns, whether or not they have kings. Often this activity is presented in collective terms, not as the affair of rulers and their servants but as the concern of a people who may be represented in either the individual rule of a king or in some form of group. Even when kings are the focus, the collective town plays a frequent role that is often overlooked in modern study of the ancient world. The Mari archives alone show us scores of cases where no individual leadership is mentioned at all, and these beg to be examined more closely. I have made the terminology surrounding the political life of the town my final destination in this study because it receives relatively little attention, in spite of its foundational importance. The Sumerian and the Akkadian words for “town” appear to have been defined above all in political terms as the collective expression of all who lived or gathered there. A town could speak and could act, could negotiate and could fight. Like the m¯atum as “land,” the “town” is finally not a place but a population, the population that has created and uses the physical settlement. When we search for all of the institutions or stock terminologies defined by “town,” we find only towns in action as bodies of people. Kings could not build their governments without contending with what must have been a preexisting landscape of towns, and these towns were defined in deeply collective terms. It is in the political life of towns that we confront a world before democracy, certainly not democratic in the Greek sense but displaying the foundations of collective decision making against which Greek democracy may be profitably examined. In early Greece, it was not the tribe but the town that became the defining center of the first movement toward democracy. It was the idea of the collected citizenry of a town as a political unit that lay deep beneath the particular historical circumstances of Greek history. I do not intend to intrude on the dialogue about democratic origins in any direct way, but I offer this ancient Mesopotamian evidence as a backdrop that may help explain the possibilities that were inherent in the entire ancient political situation.

172


My treatment of the Mari evidence will prove to be heavy with specific terms, and I propose that this focus on indigenous expression is essential. Too often, Mesopotamian references to the collective political activity of towns have been forced into oversimplified, falsely institutionalized notions of a “council of elders” and a “popular assembly,” not recognizing the range and fluidity of the actual words in play. I argue that the collective idea is most crucial in all of the language involved, and that fixed bodies with readily definable constituents will be hard to find. Different nuances do attend the choices of terms, however, and these need to be explored. After introducing some more general ideas about group leadership, I will begin with the most common way of addressing collective town action, which simply describes the town itself, or all of its people, with an explict plural, as “Imarites,” “Tuttulites,” and so on. Several words offer further definition of plural leadership in terms that allow slightly more detail regarding who is actually involved: “elders,” “heads,” and various words for assembly. Having examined the evidence according to terminology, I then focus on Imar, Tuttul, and Urgiˇs (Urkesh), the three towns in the Mari archives that are most strongly identified with collective decision making, all of which had been established long before the arrival of the Amorrites at the turn of the second millennium. I close the chapter with a broader discussion of how we should conceive this evidence for the power of the group in a political landscape dominated by kings.

a. the towns of the mari archives In ancient Syria-Mesopotamia, the town was the raw stuff of kingdoms and the ambition even of mobile pastoralists, who would control the town if not be pinned to it. At least in the period of the Mari correspondence, the town also remains the basic unit of political life, often still regarded as the essential decision-making entity. It is often difficult to judge the level of local independence for specific towns, as well as the power of individual leaders. Most towns were incorporated into or at least subservient to some larger state, and the effectiveness of external control varied considerably. In taking on the towns of the Mari archives, I am not attempting a systematic examination of all features of urban life. In concept, therefore, this chapter is immediately different from Marc Van De Mieroop’s recent overview, The Ancient Mesopotamian City (1999).2 Instead, I have given priority to the writers’ own definition of “towns” as such and the institutions assigned to these towns, as viewed against the backdrop of the larger society. My focus ends up being essentially political, following the usage of the word “¯alum” itself, which shows what activity or structures the people themselves considered to pertain to the town. The a¯ lum is the building block of the kingdoms that form the power centers of Syria-Mesopotamia in the period of Mari’s archives. A new

The Towns of the Mari Archives

173

vassal defines the network that serves Zimri-Lim by its towns: “Like his town Nahur and like the (other) towns of his kingdom, the town of Burundu is ˘ (henceforth) Zimri-Lim’s, and Adal-ˇsenni shall be his son.”3 In general, a m¯atum “land” is understood to be composed of towns: On the side of the land of Yassan and on the side of the land of Apum, there are reliable towns. This is why Qarni-Lim is staying behind to guard the land.4 And my father (Zimri-Lim) himself knows about my seven towns that are located in ˇ a.5 the land of Sudˆ

At least for his commitments to Zimri-Lim, even the Yaminite ruler Yasmah˘ Addu speaks of his people according to their residence in “my towns.”6 The same composition applies to the hals.um or “district,” when it designates an ˘ 7 administrative subset of a kingdom. The same assumptions are evident where towns are the political tender of kings. When he offers Zimri-Lim a treaty, Ibal-pi-el II of Eˇsnunna treats walled towns as the primary unit of exchange in barter over territory.8 In a letter to Zimri-Lim, king Ibal-Addu of Aˇslakkâ refers to the town as the unit that pays tribute.9 At the same time as towns are the unit of political exchange between kingdoms, they may also be used as the ultimate personal gift in negotiation of alliances. In the Mari archives, this phenomenon is most visible in what Durand calls “the Alahtum affair,” in which the king of ˘ his domain to Zimri-Lim.10 For Yamhad gives (or sells) a significant town in ˘ references to the “towns” of individual kings, see Table 8. If the town is the center of wealth and power coveted and counted by kings, it is no surprise, then, that it is the ultimate focus of attack on a kingdom. In the treaty offer mentioned above, Ibal-pi-el II addresses the major centers that must be besieged, but everyday marauding would still involve the pillaging of settlements, in this case the more accessible ones that had no protective walls.11 These smaller towns are regarded as dependent on a central capital, as at Aˇslakkâ.12 Census and accompanying loyalty oaths, as well as agricultural administration, are defined within larger kingdoms by the local town.13 table 8. Towns Identified as Royal Possessions ARM I 2 ARM XXVI 347 ARM XXVI 384 ARM XXVI 406 ARM XXVI 489 A.1025 in Kupper, M.A.R.I. 6, p. 337–9

Vassal towns of Yahdun-Lim ˘ the king of Amaz What revolts against Belonging to Iˇsme-Dagan of Ekallatum in the Zimri-Lim period Belonging to Haya-sumu of Ilan-s. urâ ˘ Ardigandi Belonging to one Belonging to a local king who may oppose Zimri-Lim

174


A town has the decision-making capacity of an individual. In a dispute over water rights, the antagonists are simply Tuttul and Zalpah, and the ˘ opening comment simply treats the towns as two men: Will there ever be a canal that two men control? Once the waters of the Balih are dammed over there (upstream), without water, what can Tuttul do? Indeed my˘lord knows (that) Zalpah has always followed Tuttul. Why should Zalpah now object?14

˘ ˘ Of course, then, a town may have a mouth, by which loyalty may be declared or refused.15 The town may receive mail,16 hold an inherited patrimony (nihlatum),17 celebrate ritual.18 This solitary voice does not derive simply ˘ the convention of individual rule. Zimri-Lim receives a report that from “the town of Amaz, which had been committed to Atamrum, has killed its ˇ a-ite.”19 king and transferred its commitment to the Sunˆ One Mari text records the result of a legal proceeding carried out in the presence of king Zimri-Lim and witnessed by the whole town of Sapiratum, represented by forty-three listed men.20 The case involves palace land that had been claimed by a private person, and the document reaffirms royal ownership. Throughout the text, the town has the capacity to act as a decision-making entity. The town witnesses the royal decision and swears an oath before the king, and by very recognition of the palace-owned land in its midst, the town shows that it does not derive its political being from that palace and kingdom. Sapiratum is in the Suhˆ um, the region downstream from Mari that was not incorporated into the˘ kingdom as a standard district, and its population belonged to Zimri-Lim’s Simalite tribe, which tended to enjoy a greater measure of independence than other components of the kingdom. Although Sapiratum may not represent a typical town in being distinguished from palace property, it illustrates well the town as political entity, always ready to function as a corporate unit when given the opportunity. The Mari archives certainly do not portray, however, a world where all power and population reside in the town. There remain the tribes, who may inhabit towns but do not define their society by them, and the steppe (nawûm), the geographical base of political independence for the mobile pastoralists who form a crucial part of the tribal population. Nevertheless, the wealth of the ancient Near East accumulated in its kingdoms, and these grew out of and remained identified with their towns. What is remarkable is that even when squeezed between the ambitions of kings who conquered from cities or steppe, the audible voice of the town survived.

b. corporate political tradition Considering how frequently group decision making appears in almost every historical setting, it is surprising to me that more has not been written on

Corporate Political Tradition

175

the phenomenon. I am aware of only one extended study in a more general anthropological vein, and this work is now over thirty years old. More recently, a group of archaeologists who specialize in early Mesoamerican civilizations have proposed a stimulating theoretical framework within which “corporate” political strategies offer a counterweight to the “exclusionary” strategies that find their pinnacle in kings. Before examining the specific evidence from Mari, I review some theoretical considerations from these studies outside the ancient Near East.

1. Council Leadership The extended study of group decision making outside a modern and western context was published under the title Councils in Action, and is now somewhat dated (Richards and Kuper 1971). The object of study is “the machinery by which group decisions are made,” no matter the domain, with special interest in the institutionalized processes of joint discussion as distinct from the informal arguments which take place over and above the debates of the recognized deliberative bodies for which we use the term “council.” (Richards 1971, 1)

Even this definition may be too rigid for our archaic towns, but the case studies in the volume include a number of modern preindustrial societies. One general study preceded Councils in Action and is cited repeatedly by the contributing authors: an article by F. G. Bailey (1965).21 According to Bailey, the first significant factor in defining a “council” is size. Consensus cannot generally be reached in a council with more than fifteen active members. A unanimous decision in a larger group is really “an act of acclamation or legitimation,” and increased size leads to majority vote (p. 2).22 Consensus tends to be at a premium in communities with “multiplex” relationships, especially of those who live together, even while this type is characterized by both intense quarreling and readiness to compromise (pp. 5–8). Councils may or may not enjoin sanctions for enforcing their decisions. Where they do not, they will be more likely to seek compromise and consensus (p. 9). Bailey defines two types of councils. “Elite councils” are those that see themselves as a separate ruling body, where the main cleavage lies between the council and the public – “us” versus “them.” “Arena councils,” in contrast, are perceived to represent segments of the public who meet to resolve conflicts between them. Because their constituencies have to deal with each other, arena councils give full play to argument and are driven to compromise, whereas elite councils have an incentive to present a unified front against the public (pp. 10–11). A council of either type will tend to close ranks when acting in opposition to any body outside itself and its public (p. 13).

176


In his introductory chapter in Councils in Action, Adam Kuper elaborates on Bailey’s categories (1971). Elite councils are adapted to political systems with effective centralization of power and differentiation of political roles, so they are characteristic of states (p. 14). These groups are small, with “fixed membership” (p. 13). Kuper distinguishes two types of arena councils, Bailey’s second type. The “community-in-council” or assembly gives all members of the political community full right to participate, though actual attendance fluctuates widely, depending on the issue (p. 14). The community-in-council develops a decision-making elite within it, so that its size does not prohibit it from working as a true council, in Bailey’s terms. These inclusive councils do not usually possess any sanction beyond public opinion. There are also arena councils that are distinct from the public, however. These normally serve a more limited function and may have more specific sanctions. Kuper’s anthropological interest in African tribal peoples leads him to downplay Bailey’s notion of “majority vote” in a way that applies equally to our ancient councils. It is better to set aside the opposition between “vote” and “consensus,” he says, because even consensus may follow “straw votes.” In Africa, those in the minority who cannot face compromise have the option simply to leave (p. 17). Kuper observes further details not addressed by Bailey. Councils sometimes fail to decide or issue ambiguous decisions to buy time. Some decisions are effectively ceremonial, not to be implemented, as when the council knows it will be overruled by a higher authority (p. 21). Besides the useful comparative details, these analyses of “council” function provide some framework for examining the collective decision making encountered in the Mari archives. First, “elite councils” do not seem to be characteristic of archaic states, even where there are three and more layers of authoritative hierarchy. We must avoid importing any of the features of elite councils into our interpretation of Mari period polity unless compelled to admit them by overwhelming evidence. I am not aware of any evidence for fixed membership, for instance. Our entire range of corporate political terminology will fall into the category of the “arena council,” therefore. The more inclusive meetings will be “communities-in-council,” while various limited groups will still belong to the wider type that tends to encourage open debate and effective compromise, whose members represent segments of the larger public rather than being joined against them with the other council members. The question of sanction is perhaps more relevant to councils dealing with internal affairs, such as the specific judicial function that is widely reported for ancient Mesopotamia.23 More often, the perspective of our diplomatic correspondence shows us the town as it faces the outside world, where consensus decision is apparently implemented by the authority of public opinion, which accepts the process by which the decision was made.


177

2. Corporate Polity in a World of Kings Rather than reevaluate Jacobsen’s “primitive democracy” from the familiar southern and eastern Mesopotamian sources, I have gone to Mari to find fresh evidence from a different setting. In this chapter I have gathered some of the specific information about corporate political activity defined by towns or, in some cases, tribes. This is the raw material of Mari’s new contribution to these questions, and in this context we should consider how these phenomena can be understood as part of political traditions before democracy. Jacobsen called the corporate political forms of ancient Mesopotamia “primitive democracy.” He risked the term “democracy” intentionally, invoking what he called its “classical” sense, “denoting a form of government in which internal sovereignty resides in a large proportion of the governed, namely in all free, adult, male citizens without distinction of fortune or class” (1943, 159). As I concluded in the introduction, the ancient Near Eastern phenomena are far from even classical Greek democracy, and we do better to seek other language to describe them. They should neither be minimized as a minor feature of Mesopotamian political life nor explained by reference to the “egalitarian” social attitudes typical of simple societies.24 We need to work within a theoretical framework that recognizes the effective power of groups, especially in more complex societies and polities.25 It appears that such a framework has been missed even in anthropologically oriented archaeology, which tends to be more engaged with current social and political theory than text-oriented ancient Near Eastern studies. In particular, a number of scholars who work with early Mesoamerican civilization have sought to formulate a theoretical foundation capable of explaining their archaeological evidence for group political activity. As observed by Anne Porter, the work of Blanton et al. (1996) provides a powerful foundation for evaluating the societies of ancient northern Syria, where strong traditions for collective decision making survived even in state-level political systems of the third and early second millennia.26 Blanton, Feinman, and their two colleagues propose a political model that they call “dual-processual” in that it envisions two primary patterns of political action that operate concurrently in the same larger systems. One of these centers on individual leadership and is therefore “exclusionary,” and the other is more group-oriented and thus “corporate” (pp. 1–2). In the corporate political strategy, exclusionary power is not eliminated but rather inhibited, so that evaluating them is a question of balance. In ancient Mesoamerica, oddly enough, the smaller polities appear to have been exclusionary, while the larger ones tended to operate from the foundation of some corporate political strategy (p. 3). To explain this situation, Blanton et al. begin with a distinction between “wealth-based” and “knowledge-based” political economies. The first

178


involves getting power by dominating networks of exchange, especially through contacts outside one’s own group (pp. 3–4). To control increasingly distant partners, the actors need to divert “prestational goods” and followers from their competitors. From this need come two network strategies: “patrimonial rhetoric” aimed at mobilizing labor and controling resources and followers, and “prestige-goods systems” to monopolize exchange of valued items (pp. 4–5). The alternative is a corporate strategy. This group emphasis may be achieved in different ways, but it always involves “a cognitive code that emphasizes a corporate solidarity of society as an integrated whole” (p. 6). Ritual tends to be based on inclusive themes that unite the people. Social status is usually defined in a way that limits individual achievement, including achievement by trade. There tends to be less consumption of prestige goods (pp. 6–7). In Mesoamerica, this corporate strategy thrived in regions with greater potential for irrigation agriculture, which likewise developed larger populations and social scale (p. 7). In a separate article, Blanton extends this model (1998). To press further the idea of corporate political strategy, Blanton redefines “egalitarian” behavior to be “any behavior that aims to establish and uphold restrictions on the exercise of exclusionary power,” whether in simple or in complex social settings (p. 151). Corporate political economies tend to involve the following elements: 1. Some form of “assembly” or “commonwealth” government (citing early Mesopotamia as well as Athens, pp. 154–5). 2. Corporate regulation of the sources of power, as in the Greek euergesai, gifts of the wealthy to support the collectivity. Even the earliest Mesopotamian writing may have appeared in a more corporate setting, he says (pp. 156–62). 3. “Reflexive communication.” These societies may be informationintensive, especially information on the compliance of those in power with expected behavior, in some system of balanced roles (pp. 162–3). 4. Ritual sanctification of the corporate cognitive code and ritualization of political communication. Do not assume that ritual and ideology serve only the exclusionary political strategy (pp. 163–5). 5. Semiautonomy of lower-order subsystems, such as households (pp. 166–7). The Mari evidence presented in the main body of this chapter certainly displays corporate political forms, however inclusive may be the meetings of “elders,” “heads,” or the undifferentiated town. As for the remaining categories, I offer them more to show how corporate strategies may function in complex societies than as a grid against which to read the Mari correspondence. For instance, both the king and the corporate leadership of Urgiˇs use the medium of writing, and the particular form of the diplomatic letter,


179

to advance their individual and group interests. The very language of the collective voice of the town belongs to an ideology of corporate action in the common interest, as in matters of war and peace or conflict with neighboring polities. When town-based polities (not necessarily “city-states”) are said to act as “the Urgiˇsites” or “the Ekallatumites,” we are seeing a “corporate cognitive code” at work. At Mari, a whole set of rites revolving around a feast for the goddess Eˇstar seems to serve a corporate code. These events include the Eˇstar feast itself, the râmum rite (marked by a commemorative stone), and rites for Nergal, among others. They encourage broad popular participation and an element of attention to the ancestors not just of the king but of the people more generally.27 The ritual texts from late second-millennium Emar provide more explicit evidence for this “materialization” of corporate political ideology. For instance, the principal festivals are said to be initiated by “the inhabitants (sons) of Emar.”28 Keeping in mind Blanton’s reference to the euergesai, the wealthy private sponsors known from ancient Greece, some Emar festivals that are celebrated ad hoc, and not according to the calendar, receive the largest part of their animals for sacrifice from someone simply described as a “master of the house” (b¯el b¯ıti). This is not a professional title and seems instead to be a wealthy local leader who makes the gift “to support the collectivity,” like the euergesai (Fleming 1992b, 97–8). I do not intend to pursue at length a rigorous application of the model. I want to show only that it provides a theoretical framework for our town political life that requires no form of “democracy” and that can explain the corporate political behavior of archaic states, as well as of simpler societies. Moreover, it can account for a powerful collective mode of political action in the very same systems that are most often dominated by kings, in a dynamic flux.29 Even the m¯atum confederacies of Ida-Maras. , Zalmaqum, and ˇ Subartum match the “commonwealth” type of group governance mentioned by Blanton (p. 154). Even more recently than Blanton, Feinman, and company, Susan Keech McIntosh has suggested still another body of evidence from which to reevaluate common assumptions about early complex societies and kings (1999a). According to McIntosh and her colleagues, Africa provides many examples of political complexity with much less centralization of authority. Roderick McIntosh observes that many large African tells remained unrecognized because past researchers did not find the expected signs of preindustrial urbanism, namely, encircling wall and citadel (or palace or temple) reflecting coercive political organization, elite tombs or residences or other accoutrements as monuments to economic social stratification, and monumentality of architecture as monument to state ideology. (1999, 56)

In recent African settings documented by anthropologists, Susan McIntosh finds that “kings” often had quite restricted roles, with a strong ritual aspect

180


and balanced by associations or a council of lineage heads. One finds “a range of societies with relatively weak vertical political control and extremely complex horizontal integration” (1999b, 15–16). Susan McIntosh finds the theoretical model of Blanton et al. to be helpful, but the exclusionary and the corporate strategies are treated in their work as “relatively unlinked aspects of two different processes of political tradition.” In Africa, the exclusionary and the corporate are part of a single system, and archaeological evidence for earlier societies in the inland Niger delta suggests the same lack of effective central control (1999b, 17–19; 1999c, 75–7). Blanton and company surely share the notion that corporate and exclusionary strategies are often joined in a single political system, but it is well worth pondering to what degree they stand in tension, perceived or otherwise, rather than forming some kind of symbiosis that is less oppositional. In the Mesopotamian setting of the Mari archives, any collective political tradition that was vigorous enough to preserve an alternative to kingship could not survive except in tension with a radically different notion of leadership. We see a particularly vivid expression of this opposition in the situation at Urgiˇs (Urkesh). While the African evidence shows how political complexity can develop without much centralization, it presents scenarios quite different from those of the ancient Near East. After all, this is the world of the “encircling wall and citadel” and all of the monumental architecture one could wish for. In the end, the theoretical approach of the Blanton group remains particularly applicable to the phenomena found in the Mari archives.

c. the collective face of towns or lands Studies of town leadership have often concentrated on two phenomena, defined as “elders” and “assembly.” One usually speaks of a “council of elders” and an “assembly of free men,” “of citizens,” or the like. Both notions carry with them the picture of formal political institutions, with bounds that would have been consistent across their various expressions, at least in one time and region. The actual range of ancient terminology is more diverse than these modern categories suggest, and the application of each word is more fluid. It is more appropriate, therefore, to focus on the collective idea that underlies all of the terminology and that provides the ultimate basis for choosing what texts and terms are pertinent. When we search for the town in joint action, we find that the most common designation simply names the town or land and identifies its collective ´ ˇ 30 This written exprespopulation by the scribal signs for “men” (LU.ME S). sion is grammatically ambiguous, in that it can be read as “the men of GN” or as “the GN-ites,” where GN is the “geographical name” of the town or land. With the first reading, we would have to deal with the implications ´ of a gathered group called aw¯ılû (“men”), underlying the LU.ME Sˇ writing,

The Collective Face of Towns or Lands

181

and this interpretation is often associated with a popular assembly of free “men.” I argue, however, that in most cases, this written expression assumes no such assembly and is only a simple collective reference to the people of the town or land. The second reading as adjectival “GN-ites” is more likely, so that no separate category of citizens is in view. The frequent appearance of “GN-ites” is thus to be rendered in conventional cuneiform transliteration ´ s ´ GN, where LU.ME Sˇ is a silent “determinative.” as lu.meˇ We have in this first terminology, then, the most generic designation of collective political activity. With this form of reference, we usually cannot answer questions about who exactly is involved and in what actual relationship to a king or other individual leader. It remains extremely important, however, to recognize the choice of a collective rather than an individual expression for these actions, which are thus attributed to the town or land as such rather than to a ruler. An ideology of collective identity and action cloaks whatever particular leadership may have a part. In most cases, the references to Terqa-ites, Imarites, Ekallatumites, and so on name the central town that gives its identity to an affiliated people. Sometimes, however, the collective is named by the “land” or m¯atum, as with “the Yamhadites” and the “Zalmaqumites.” Even when the town center ˘ it is impossible to distinguish “town” (¯alum) from “land” is clearly in view, ´ s (m¯atum), where the polity has m¯atum status. This is because the lu.meˇ GN formula is used specifically for real political action, for decisions taken in the name of the collective people. Where that people is recognized as a m¯atum, this higher-order political rank will take precedence over political status as a “town” (¯alum). In practice, then, this generic collective language applies to whatever political category is in play. When this is a m¯atum, the people invoked by the name may be subsumed in the identity of a town, of a regional alliance (e.g., Zalmaqum), or even of a tribal people, as might be implied by the “Yamhadites” of Table 10, if a tribal name has come to be ˘ fully identified with a “land” (m¯atum).

1. The Town Itself in Action In some cases, the collective action of the town is identified simply by the name of the town as subject. One letter to Zimri-Lim describes how Ili-sumu has gathered all of Ida-Maras. behind him to fight for Zimri-Lim.31 Only the kingdom of Aˇsnakkum resists, while “Urgiˇs” has seized half of the possessions of Iˇsme-Addu, the assassinated ruler of nearby Aˇsnakkum. Elsewhere, ˇ letters are addressed to towns by name. Sadum-labua, the succeeding king of Aˇsnakkum, reports to Zimri-Lim, “I did not (previously) write to Eluhut, but now I will write to Eluhut,”32 and then mentions sending a letter˘ to ˘ ˇ h.33 Similar four recipients: two named kings and two towns, Urgiˇs and Sina ˘ phenomena appear in Babylonian and Assyrian texts of the early second 34 millennium.

182


It is my impression that the solitary town name is the simplest way of recognizing the collective political face of the town. Possibly, this collective representation can also be applied to a m¯atum “land,” which thus takes on the ancient ideology of the community in action. I am not aware of any letters addressed to a m¯atum by name alone, rather than to its ruler. In ´ s one case, the “Zalmaqumites” (lu.meˇ Za-[a]l-ma-qa-yu) are reported to have written to Eluhut to negotiate peace (ARM XIII 144:39–41). Here, the plural does reflect a ˘known m¯atum, but it is one that exists only for such collective action, through the collaboration of its four constituent kings.35

2. “Mariotes” and Company: The Adjectival Form of the Town Name Just as it is always said that the Akkadian m¯atum is ultimately a people, and only by extension the land they inhabit, the a¯ lum is first of all a concentration of people, with the capacity to make decisions together. The physical construction is what they have created for their common use. When the town acts, it acts plurally. We have references to elders and assemblies, but the most common description of decision making under town identity uses the ´ formula LU.ME Sˇ Geographic Name (GN), often interpreted as “the men of GN.” In fact, this formula does not indicate an assembly, never mind a gathering of some aw¯ılum class of free or elite men. When it does refer to a town as a whole and is not simply identifying some limited group of its citizens, the formula provides the most amorphous definition of the people, simply in action as a town. This is the corporate ideology in raw form, requiring no mediating institution, however informal. ´ It appears that the standard reading of the formula LU.ME Sˇ GN is adjectival, not genitival. This can be seen, first, in the singular version of the ´ GN, without the plural marker MES. ˇ Adjectival forms of same formula, LU geographic names abound and seem to be at least as common in references to people as the proper noun in standard form. Table 9 provides a sampling,

table 9. Adjectival Forms of Geographical Names in Individual Identifications ki ´ Qa-at.-t.u-na-na-yu LU ´ ki ´ LU Qa-t.a´ -na-yi[ ] ki ´ KA.DINGIR-yu ´ LU ´ Su-ha-yi ki LU ˘ ´ I-ma-ri-i-im LU ´ LU Ia-am-ha-de-emki ˘ ki ´ Kur-da-im LU ´ LU Za-al-ma-qa-i ki ´ An-da-ri-ga-yuki LU ki ´ La-ar-su[L]U ´ u-um ´

the Qat.t.unanite the Qatna-ite the Babylonian the Suhâ-ite ˘ the Imarite the Yamhadite ˘a-ite the Kurdˆ the Zalmaqumite the Andarigite the Larsa-ite

ARM I 7:29 ARM I 15:6 ARM IV 5:20 ARM XIII 83:9 ARM XIV 33:8 ARM XIV 75:10 ARM XIV 76:6 ARM XIV 78:4 ARM XXVI 375:24 ARM XXVI 385:8


183

far from exhaustive, of adjectival forms of geographic names, all of which contain some form of the final “gentilic” -¯ı that makes a place name adjectival. Instead of translating “the man of Qat.t.unan” or “the man of Qatna,” these are “the Qat.t.unanite” and “the Qatna-ite.” This limited selection just illustrates the phenomenon with some of the more common place names from the Mari archives. ´ is standard but by no means required. We find ZaThe preliminary LU ki al-ma-[q]a-ia-am (ARM XIV 75:10) and Ka-ha-ta-yuki (FM II 127:34). The adjectival reading may also be hidden under˘ the remaining noninflected ` sforms, as evident for the Eˇsnunna-ite army when it is written .sa-bu-um lu´ Eˇ ki 36 ´ ˇ nun-na (e.g. ARM I 53+:17). Sloppy use of singular and plural LU(.MES), as well as disagreement of number between subject and verb, add to the likelihood of an adjectival reading, rather than the genitive “man/men of GN.” For example, “the Eˇsnunna-ite” blocks the way in a letter from SamsiAddu to his son.37 Another text presents the collective and the specific uses side by side: A delegation of envoys identified as “the Babylonian” ki ´ ) wanted to camp on their own side of the river, se(lu´ KA.DINGIR.RA parate from “the Qatna-ite” delegation (lu´ Q[a-t.a´ -na]-yi ki , lu´ Qa-t.a´ -na-yuki ). Someone raids the Babylonian camp and kills “one ranking Babylonian.”38 In neither case do we have “a man of Babylon,” as confirmed by the adjectival form of the Qatna group.39 ´ ˇ as shown The same adjectival forms are common with the plural LU.ME S, by the examples in Table 10. The fluidity of spelling with or without the actual adjectival gentilic and case ending is visible in FM III 148 and 149, two letters sent by an administrator named Manatan. The second letter renders ´ s Ia-am-ha-di-i ki (line 5), planning the adjectival form “the Yamhadites” as lu.meˇ ˘ ´ s in advance, while the first tacks˘ it on as an afterthought, lu.meˇ Ia-am-ha-ad ki -i ´ (line 8). The adjectival form of LU.ME Sˇ GN must therefore be ˘read as lu.meˇ ´ s ´ GN, which in cuneiform transliteration treats LU.ME Sˇ as a silent “determinative” for some designated group of men or people, which was not read as the separate noun aw¯ılû (“men”). This adjectival form may also be ´ s a-lu-yu, or “citizens.”40 reflected in the generic lu.meˇ

table 10. Adjectival Forms of Geographical Names in Group Identifications ´ LU.ME Sˇ Ter-qa ki -yu ´ LU.MESˇ Za-[a]l-ma-qa-yu ´ LU.ME Sˇ Ia-am-ha-di-i ki ˘ u´ ki ´ LU.ME Sˇ Aˇs -ˇsu-ru´ ˇ LU.MES I-ma-ru-u´ ki ´ al-la-ta-yi ki ´ LU.ME Sˇ E-k´ ´ ˇ LU.MES Ra-za-ma-yi ´ ˇ I-da-Ma-ra-s.a-yi ki [L]U.ME[ S] ´ LU.ME Sˇ Tu-ut-tu-li-[yi]

the Terqa-ites the Zalmaqumites the Yamhadites ˘ the Assyrians the Imarites the Ekallatumites the Razamâ-ites the Ida-Maras. ites the Tuttulites

ARM II 94:6 ARM XIII 144:39 ARM XIV 75:4 ARM XIV 128:6 ARM XXVI 246:15 ARM XXVI 384:12 ARM XXVI 409:59 ARM XXVII 135:15 A.885:4 (Durand, 1989, p. 33)

184


´ s ´ If the LU.ME Sˇ GN are really the lu.meˇ GN “GN-ites,” as adjectives of place in substantive use, then few if any of these represent “men of GN.” Likewise, ´ does not usually indicate the noun aw¯ılum, from which the determinative LU we could find “free men” for assembly. The same conclusion may apply to ´ without the plural marker MES, ˇ with the same the simple determinative LU collective effect. This alternative is demonstrated by a text that uses both ˇ with the same plural intent: lu´ Ku-b[a]-aforms, with and without MES, ki lu.meˇ ´ s ki 41 Ku-ba-a-yu . This terminology does denote the people [yu] versus of the town as such in its collective voice, but in the most generic sense, not concerned with specific institutions. One of the letters from the governor of Saggaratum to Zimri-Lim displays this perspective with the generic form a¯ luy¯u.42 Atamrum, king of Andarig, has besieged Razamâ and is now negotiating with its representatives, who are initially identified as “elders” (line 11). After this, they are called a¯ luy¯u (“town-ites” or “townsmen”), so the city itself as it acts. They answer Andarig for Razamâ (lines 19, 41), and they dig tunnels (line 27):

When the army reached Razamâ, at the moment of arrival, a town force came out, and it slew 700 Elamites and 600 Eˇsnunna-ites. They let ten days pass, and then the elders came out to Atamrum and said to him, “We want peace. . . .” (lines 8–12)43 The townsmen answered him (Atamrum), “The town belongs to Zimri-Lim, and the regular army has left along with him.” (lines 19–21)44 When the front of the siege mound reached the base of the fortification wall of the lower town, the townsmen dug a tunnel in the (main) town, and they made two holes into the front of the siege mound, on the right and on the left. (lines 26–8)45 ´ s I find that lu.meˇ GN is generally used to indicate not residence in a town, which is defined by “the sons of GN” but the town’s activity for the collective interest. This corporate action never to my knowledge applies to internal town affairs. Reference to “Terqa-ites, Imarites, Ekallatumites,” and so on appears to reflect a standard perspective on the town as a political unit that deals with political units outside itself. Such external affairs seem to inspire this unadorned expression of the ideology of collective town action, regardless of which individual leaders or representative groups are in fact involved. ´ s According to our Mari evidence, the lu.meˇ GN send and receive letters,46 negotiate peace and war with outside forces,47 and keep accounts with overlords.48 They may be credited with the military success of taking a fortified town,49 and one text describes the Hurrâ-ites discussing politi˘ cal strategy.50 The same focus on collective activity lies behind “reports” 51 ´ s GN precede a second on such collective citizenry. Occasionally, the lu.meˇ designation that introduces a more precise definition. In a text cited once already, eighty representative Imarites are also identified as “heads” of the


185

m¯atum.52 Another important letter provides a fairly elaborate picture of group decision making in the twin towns of Isqâ-and-Qâ, and the contents come up repeatedly in this chapter’s discussion: I heard the following news of the Isqâ-and-Qâ-ites. “They have been called up (for service), (with) a ten-day provisioning. They are going to (join) the reinforcements of Hammurabi.” When I heard this information, I wrote to Yamrus. -el and the elders ˘of Isqâ-and-Qâ, and the (household) heads of Qâ-and-Isqâ assembled before me – a group of 200 as one man. In their meeting, I addressed them as follows. . . . 53 ´ s The “news” is defined by the generic lu.meˇ GN, before the text goes on to detail a written exchange with the town’s ruler and elders and at last a full meeting of its “heads.” As may be obvious, the added clarification in these texts pushes past the simple faceless collective because the reader needs to know something about the actual political process in effect. In both of the texts cited here, that process does really involve large plural groups, so that we should not dismiss the corporate ideology as a mere fa¸cade for individualized decision making. ´ s GN does not package for us “an assembly of free Even so, the formula lu.meˇ men.” Some texts allow us to see behind the expression other details regarding the plural representation of a town or a m¯atum coalition. In the case of the m¯atum coalitions, the generic collective likewise serves to describe their ´ s GN can refer to joint action toward outsiders. With such coalitions, the lu.meˇ a meeting of the member kings. Yaqqim-Addu quotes “the Zalmaqumites” as refusing any contribution of soldiers to join Babylon at war,54 and elsewhere they are cited this way in peace negotiations: “The Zalmaqumites keep writˇ ing to Sarraya the Eluhutite to make peace.”55 Usually, the decision-making body for Zalmaqum is˘ its allied kings.56 Another reference may be best understood by the same convention: “All the Ida-Maras. ites are (as good as) dead because of this man. They (might as well) not be alive. For the sake of the Ida-Maras. ite, save the life of this man.”57 The man in question is Ibni-Addu, the deposed vassal of Zimri-Lim who was king of Tadum.58 A Mari official pleads to the king for some effort to save his life, with the argument that the other loyal vassals on the thrones of Ida-Maras. will never survive if this one local king can be cast aside and killed. It should be not the whole population that is threatened by such an assassination but its rulers. ´ Another reference to collective town action allows the LU.ME Sˇ to stand alone as aw¯ılû (“men”), though with the same generic identification of faceless leadership:

Baqqum the Tizrahite has gone to (his) fate. Some men (from) the inhabitants ˘ (“sons”) of Tizrah therefore came to me, (saying) “Appoint Kali-ilumma to the office of our overseer. .˘. . ”59

186


“Men (from) the inhabitants of Tizrah” arrive to speak to Tarim-ˇsakim, a vizier under Yasmah-Addu, about the ˘appointment of a new sug¯agum for ´ their town.60 Here, ˘with a single town in question, the LU.ME Sˇ are a select group responsible for the nomination of a chief or at least for the communication and negotiation of that nomination to the royal administration. The travel suggests some limited number of representatives much smaller than may be envisioned for an “assembly.”61 As the last example shows, the noun aw¯ılû may sometimes be visible be´ ˇ at least as a limited group who may speak for the town as neath the LU.ME S, ´ s GN a whole. Though we should not read this onto the broader use of lu.meˇ ´ ˇ for any collective leadership, the LU.MES do occasionally appear as a proper class, not united for decision making but still identified with whole towns. I already mentioned A.1051 (AEM I/1, p. 169 note a), which distinguishes the aw¯ılû class of “men” from the muˇskênum common folk within a single a¯ lum. The muˇskênum need to be reassured, but the aw¯ılû require a more concrete form of assistance or oversight, to be “settled (in residence).” In this context, Adams’s distinction of the muˇskênum as a rural population, opposed to the high-status urban aw¯ılum, would not seem to hold up (1982, 12). Both belong to a smaller town or village in what Adams probably would have considered part of the “rural” sphere. The concern in A.1051 for housing the aw¯ılû may be illuminated by another town hierarchy, this time for the town of Kasallu(k) in southern Mesopotamia, though one that has a tribal population of Mutiabal. When king Hammurabi of Babylon orders that the whole town be removed, he ˘ is reported to have addressed the “heads of Kasallu(k).”62 The same let´ s Ka-sa-al-luki , line 14). We ter identifies these with “the Kasallukites” ([l ]u.meˇ cannot separate the word “aw¯ılû” from these corporate identifications, but the letter offers a fairly detailed portrait of the social framework into which such “men” would have to fit. Collectively, the Kasallukites have a m¯atum outside the settled town, where sheep graze (line 31). As heads, they possess “houses” (b¯ıtum), which they will be permitted to occupy even when their dependents have departed (line 32), and these houses are agricultural centers for grain production (lines 28, 38). The human dependents of these heads are their “people” (niˇsu¯ , as dependents), whom Hammurabi wants ˘ brought to Babylon by boat (lines 29–30, 37–9): He (the captured Mutiabalite) set out and [went] to Kasallu(k). This [man] (then) delivered the message which the ruler of the Elamite(s) had written. The Kasallukites paid close attention. They discussed the matter and then wrote to the ruler of the Elamite(s). (lines 12–15)63 [When they brought back(?)] a complete confirmation, (Hammurabi) sent [a message] to the heads of Kasallu(k), saying. . . . (lines 19–21)64˘ “Gather . . . , barley, (and) straw, e[very] boy (and) [girl], and bring them [into] Babylon. The . . . of your flocks can still graze in your countryside (m¯atum), while you


187

yourselves should stay in your houses.” The king said these things to them, and they answered as follows: “Our lord has spoken. We shall go. We shall take action.” They answered the king this way, and then they departed. After them, he sent a troop of 6,000 and boats to transport their barley and to remove their dependents.65

Another text may identify the particular word “aw¯ılû” with the position of being heads of households. This letter, from the Samsi-Addu period, reports that “fifty men (aw¯ılû) along with their houses have left Haradum ˘ should for Urban.”66 In this case, the “house” is a mobile unit, but one that have a concrete counterpart when the group is reestablished at a new base. This association of the aw¯ılum with households may be more fundamental to the term than the particular configuration of aw¯ılum, muˇskênum, and palace. If so, then the palace would appear to have hijacked the term and its natural prestige in order to distinguish palace-supported households from unsupported households. In its older sense, the aw¯ılum category would then have included household heads that a royal palace could identify as either aw¯ılum (“gentleman”) or muˇskênum (“commoner”). The preceding texts bring us back to the relationship between the generic ´ s corporate formula lu.meˇ GN and the supposed “assembly of free men.” Although the generic formula may indicate a range of participation, without necessarily involving any assembly, it may assume that only household heads (once, “aw¯ılum”?) will participate in town decision making. In this light, it is interesting to find a reference to 200 Razamâ-ites (not evidently the entire town in action) building houses at the instigation of Atamrum.67 A similar impression of household leadership is given by another letter, which describes the deportation of “the Harbˆ u, along with their movable prop˘ ´ s the lu.meˇ Ur-ba-na-yuki and the erty and everything they own.”68 Elsewhere, lu.meˇ ´ s Mu-ul-ha-yuki had been separated from “their children and their wives,” ˘ a more limited picture of household leadership.69 Neither of the last two texts involves decision making, but even as victims, the people of all the towns involved play a role in external political affairs. In sum, full town participation in collective decision making may be defined in terms of household heads, but it is surely perilous to insert the full group into their every appearance. As seen with Isqâ-and-Qâ in the remark´ s GN formula may lie a variety of able detail of ARM II 75, behind the lu.meˇ decision-making bodies, in that case perhaps even two, with an individual and elders followed by a gathering of “heads.” Recognizing the capacity of this collective language to cover the role of a powerful central individual, one wonders whether the identification of rulers as “the Babylonian” or the like might carry more than the pure information that he governs “Babylon.” Because the collective idea behind the adjectival form of the town name ´ s functions equally with both plural and singular forms (lu.meˇ GN and lu´ GN), it could be susceptible to being coopted by individual rulers. It is possible then that the habit of naming kings as “the Babylonian” or “the Eˇsnunna-ite”

188


derives from the circle of royal domination in which the whole political will of the towns or lands is understood to be embodied in one figure. Insofar as reference to kings by the same adjectival form of the geographic name borrows from the collective decision-making identity of the town, rulers invoke the corporate ideology in service of a very different political order. In linguistic terms, the transition from collective to individual leadership is smooth and easy, but it masks a deep shift of power.

3. “The Sons of GN”: Inhabitants ´ In strict orthographic terms, the LU.ME Sˇ GN or “GN-ites” appear to be almost interchangeable with another, somewhat less common designation, the DUMU.MESˇ GN. The Sumerian writing DUMU means “son” (Akkadian m¯arum), and the two expressions are often read as “the men of Geographic Name” and “the sons of Geographic Name,” imagining little difference. My ´ interpretation of LU.ME Sˇ GN as the adjective “GN-ite,” however, produces a much sharper contrast between the two expressions. DUMU is not used in cuneiform writing as a silent determinative, so that the formula DUMU.MESˇ Geographic Name (GN) can be interpreted only as a genitive phrase, “the sons of GN.” At the same time, the “son” does not represent a broad social category such as aw¯ılum and muˇskênum, and it provides an identity by lineage that must have struck the ancient ear as more sharply different from “men” ´ ˇ than our modern perception assumes. (LU.ME S) “The sons of GN” identify the full population of a town or land, but without a political, decision-making aspect. The place of the category in the kingdom of Zimri-Lim can be seen in a description of plague in the Saggaratum district: The god is striking the upper district (with plague), and I passed by in haste(?) (lit., “hand to hand”). Now, my lord must give orders, so that the village residents (“sons of the villages”) who have the plague cannot enter the villages not yet struck. There is danger that they could spread the plague to the whole land. If, then, there is any trip(?) (planned) by my lord for the upper district, my lord should stop at Terqa. He must not continue on to Saggaratum. The land is contaminated.70

Given that this letter comes from Asqudum, a high Mari official, the first point of reference is naturally a division of the kingdom, “the upper district” (hals.um elûm, lines 20, 27).71 The next breakdown is urban, as is typical of ˘ m¯atum and the halsum, and speaks of the towns (¯al¯an¯u, line 24). Facing the . the threat of deadly˘disease, no other population is considered.72 “The sons of the villages (=towns),” clearly all who inhabit them, are barred from other communities (line 23). The same universality is evident in another letter, when instructions are given to bar all residents from the central kirhum ˘ “citadel” in threatened towns of the land (m¯atum) of Kahat. “Take care that ˘ 73 not a single inhabitant of the city is allowed to enter the citadel.”


189

When not the whole population, but rather an individual man is in view, the point of the definition as “son” is to indicate the man’s original place of residence. In a letter from early in Zimri-Lim’s reign where the senior Simalite leader Bannum tries to discredit the diviner Asqudum, the centerpiece of his case is the fact that the diviner is a “son of Ekallatum,” the heart of Samsi-Addu’s earlier kingdom.74 Bannum invokes his origin to identify him as an enemy, but Zimri-Lim declines Bannum’s advice and never ´ (“man”) would have inappears to have any reason to regret it. Where LU dicated a current affiliation and in some cases a representative role, DUMU (“son”) provides a simple place of origin.75 In this individual use, the two designations fall more closely together than when linked to whole towns.76 ´ s The difference between DUMUmeˇs GN and lu.meˇ GN is also reflected in the activities attributed to them under these terms. Where the identification lu.meˇ ´ s GN is used for the collective voice of the town as it negotiates its place in the wider world, the phrase “the sons of GN” is most often associated with what every man, if not every person, may contribute. A letter from KibriDagan to Zimri-Lim describes the human resources of Terqa available to help with canal work: I am concerned that they will say to my lord, “A large work-crew is available in the midst of Terqa.” Among the 400 (male?) inhabitants (“sons”) of Terqa, 200 are dispersed, and (only) 200 are on hand.77

It is the “sons of the town” (DUMUmeˇs a-limki ) who must be counted in a census78 and who likewise swear a loyalty oath.79 References such as these are consistently understood best under the rubric of residence.80 King IbalAddu of Aˇslakkâ writes to Zimri-Lim that in a hard year, “the muˇskênum (who are) resident in Aˇslakkâ” have crossed the mountains and left the realm.81 A representative of the king of Yamhad goes to the town of Alahtum after it ˘ all of the residents, and˘ repeats the has been sold to Zimri-Lim, assembles 82 instructions sent by his master. Their very departure shows that the term “sons” indicates residence. Where Kraus reads DUMU.DUMU a.ga.dèki in the Sargon inscription as the “Bürger/Bewohner” of the city of Akkad, the Mari pattern suggests “Bewohner” over “Burger” ¨ (1970, 29).83 The pattern of normal use makes the exceptions more striking. It is here, ´ s GN, that we find rare glimpses of a full assembly. At Tuttul, not with the lu.meˇ taxes to Zimri-Lim are the responsibility of the collective populace, represented either as “sons” or “Tuttulites,” where specific action to pay must be taken by the tahtamum, the local pre-Amorrite assembly.84 ˘ Seat the tahtamum-council. Request from them the s¯ırum-tax that is levied on the inhabitants˘(“sons”) of Tuttul.85 Regarding my lord’s s¯ırum-tax that is levied on the Tuttulites, just as I seated the tahtamum-council once, twice, even three times and I made my request to them, ˘ men have written once, even twice, to Imar.86 these

190


When all of Saggaratum is gathered as the DUMUmeˇs a-limki to face the governor’s harangue on census procedure, this represents a proper assembly, though one without any decision-making power.87 The same governor passes on the information that the sons of some Yaminite town have met together and decided to side with Eˇsnunna against Zimri-Lim. In conclusion, we have three ways of identifying the town or land (m¯atum) in action or acted upon: the named polity in simple form (GN), the adjec´ s GN, tival form of the named entity marked to show its plural voice (lu.meˇ meˇs as “the GN-ites”), and “the sons of” the town or land GN (DUMU GN). The first two function as equivalents, both representing the town or land in action as decision maker. “The sons of GN” are simply the inhabitants of the place, usually the objects of some action, not describing the town or land as a political unit. Notice that the “sons of” language is also used for tribal peoples, and this may then merge with the “sons of” a m¯atum, as with the “son(s) of Simal” and the “son(s) of Numhâ” in dealings between the kingdoms centered at Mari (m¯at Hana) and at ˘Kurdâ (m¯at Numhâ).88 ˘ This phenomenon underlines the potential synchronism between˘ the “tribal” and “land” categories that have sometimes been read as mutually exclusive. We come away from these collective expressions with little evidence for specific forms of corporate political life, but it should be clear that the notion of the town or land in action as a group was much more widespread than a limited study of “elders” and “assemblies” might suggest. At the same time, one goal of this section has been to remove both “the men of GN” and “the sons of GN” from sweeping inclusion in discussion of popular assemblies. The one expression is probably misread, and the other was not generally used for the population as town decision makers.

d. elders In the Mari evidence, as perhaps in cuneiform texts more broadly, the most common definition of collective town decision makers beyond the generic “GN-ites” is as “elders.”89 The primary Mesopotamian words actually mean “old man,” literally “elder”: Sumerian ˇsu-gi and Akkadian ˇs¯ıbum. Mari letters ´ and MESˇ status determinative often use the Sumerian spelling with the LU ´ sˇ meˇs ˇ that was addressed in the previous section (lu.meˇ , for SU.GI or lu´ SU.GI the plural ˇs¯ıb¯utum, “elders”). Discussions of such Mesopotamian “elders” often assume that they functioned as part of a “council,”90 a word that calls forth an image of institutions with some standing status, who perform some regular tasks and meet on some predictable basis.91 In fact, “elder” terminology usually represented a broad classification of status, not the designation of a permanent individual title, whose assembled bearers would meet as a “council” of fixed function. Actual reference to the “elders” of a given town shows an enormous fluidity of both written application and political reality.92 There is no question

Elders

191

that “elders” could designate senior leadership whose collective authority was distinct from that of the king. At the same time, however, the elders of a town could be sent as representatives of the king, in direct service of royal authority and administration. It must not be imagined that all of these represent essentially the same group with varying portfolios. We should rather understand “elder” language to indicate senior leadership that could be defined in different ways according to the particular political situation or intent of the writer. As “elders of the town,” they represent the town collectively, but this may or may not be as emissaries of a king.

1. Elder Functions The word “council” evokes a fixed body of elders, whose members somehow carry their right to participate in a title.93 This framework bears a conceptual burden that will constantly hamper our understanding of individual references to town elders. We should instead think in terms of “elder” terminology that was used in certain contexts associated with specific functions, such as diplomatic exchange. These would have been functions that were commonly deemed appropriate for collective leadership, and little more than that collective voice may be intended by speaking of “elders.” The difference between “elder” terminology and the adjectival form of town or land ´ s GN, “the GN-ites”) is that the former emphasizes the represennames (lu.meˇ tative nature of the group, as a subset of the whole population. No particular rank is indicated beyond a loose seniority. ´ s GN, “elders” are always encountered in negotiations with Like the lu.meˇ outsiders. The writer may choose to mention an individual leader at their head94 or to present them as an undifferentiated mass.95 The letters sent by elders are by nature contacts with such outside governments.96 When elders are gathered with other named categories of leadership, it is always for some expression of external affairs.97 These terms also tend to be general classes, rather than titles pertaining to appointed office or fixed institutions: “messenger,” gerseqqum (“palace-born”?), wedûtum (“dignitaries”?).98 A second role even produced a derivative application of the noun, as “witnesses,” and the very phenomenon of such a transfer displays the priority of function over permanent title in use of the word “elder.” When the legal problem reaches beyond the scope of the individual town or political unit, this function overlaps somewhat with the preceding one. One letter describes an unusual procedure in which a vassal king (Bunu-Eˇstar, king of Kurdâ) sends a formal complaint to Zakira-Hammu, governor of Qat.t.unan, who opens it personally and has it read in ˘the presence of several named “district elders” and 100 town elders (or “witnesses”?, ˇsi-bi a-li-im).99 Otherwise, these serve a purpose completely internal to the town in question and already very familiar to the Old Babylonian period.100 On rare occasions, elders without any named leader are attributed a ritual ´ sˇ function. The elders of Terqa (lu.meˇ SU.GI a-limki ) pray to Dagan on behalf

192


of king Zimri-Lim and his armies. “Let my lord come in peace to kiss the foot of Dagan. The town elders enter regularly into the presence of Dagan and they invoke (his) blessing on my lord and on the armies of my lord.”101 A list of deliveries mentions oxen for the elders of [Hurr]â(?), of Urgiˇs, and of ˘ diplomatic aspect to the ˇ h, towns quite far to the north, suggesting some Sina ˘ 102 provision from Mari sources. One oil disbursement covers the anointing of the elders of Urah in their Dagan temple.103 Even if “elder” is˘ not a titulary office, it may nevertheless designate individual status for the administration of dependents who dine in the royal palace.104 Notice that in this aspect the term is directly linked to royal favor and would not appear to be related to the governance of a single town. Altogether, these texts show clusters of “elder” terminology around three functions: outside contacts, legal witness, and ritual.

2. The Variability of the Term Two specific texts show how varied the use of elder terminology may be. In the first (ARM VII 311), we encounter three named men who are identified as “elders” of a given town but who make a payment that defines them more precisely as holding the rank of sug¯agum, the local leader of a town or tribal unit. In the second (A.2226), one of Zimri-Lim’s chiefs of pasture (merhûm) considers it reasonable to call a group of northern leaders “elders” ˘even when they appear to include local “kings” (ˇsarrum). Both texts suggest that “elder” terminology may be used as an inclusive category that communicates little about specific titles. a. ARM VII 311. This legal text has formed a staple of the literature on the sug¯agum, because it records the payment of silver as a sug¯ag¯utum to the royal administration. The key section reads as follows: Is. i-sarae, Yimsi-el, (and) Mut-ramê, elders of Saggaratum who came for the sug¯ag¯utum payment – the silver of their payment has been checked. I myself will hand over to my lord the proceeds of their gifts.105

As with the gathered “sug¯ag¯u, deputies (laputtû), and elders” of Saggaratum mentioned with discussion of the hals.um “district,” the initial phrase should ˘ indicate elders not of the central town but of the whole district, as a division of Zimri-Lim’s administration of the Banks-of-the-Euphrates portion of his kingdom. These named men should not be mere messengers but should apparently bear “their (own) payment(s).” As individuals, they are evidently the sug¯ag¯u of separate towns, while together they are “elders.” There is no indication either that they become sug¯ag¯u only at this moment or that they have ceased to be elders with the delivery of their sug¯ag¯utum payments. The text need not be understood to record a promotion from one title to the

Elders

193

other. Rather, the “elder” terminology indicates a broader classification as district leaders. b. A.2226 (Charpin, M.A.R.I. 7, p. 182). In the hierarchy of status available to individual rulers or leaders, three main titles accrued to different settings. The chief of pasture (merhûm) was defined only by his leadership of ˘ “leader” took his title from tribal politimobile pastoralists. The sug¯agum cal traditions with varying expressions in the Mari period. Local leadership of towns was often identified with a single sug¯agum. In the community of larger polities, however, the most coveted title appears to have been ˇsarrum, which we translate as “king.” The rulers of Babylon, Eˇsnunna, Ekallatum, Mari, Yamhad (Aleppo), and Qatna and the lower ranking players on the ˘ regional scene all claimed to be ˇsarrums, always from a “town” base, whether that town had a substantial population or was rather a fortified ritual and administrative center. Against this backdrop, one letter to Zimri-Lim seems to use the word “elders” for a gathering that includes several kings, in a way that the kings themselves would scarcely have accepted. Ibal-el, one of Zimri-Lim’s two powerful chiefs of pasture (merhûm), speaks to his lord about the collected ˘ of the land” (lu´ SU.GI meˇs ˇ kings of Ida-Maras. as “the elders ma-at I-da-Ma-raki as. , line 4). Ida-Maras. was one of the three m¯atum coalitions discussed in Chapter 3, each of which was normally regarded as an association of independent polities led by independent “kings” (ˇsarrum). In the Mari letters, kings are almost always referred to by personal name, by town or domain (e.g., “the Babylonian”), or by the ˇsarrum title (“king”). We do not generally find kings mentioned by any other title that could designate a particular status. My interpretation of the “elders” in A.2226 as including kings is therefore counterintuitive, and the text deserves some extended consideration. A large group of northern leaders gathers to ratify a peace agreement, and Ibal-el declines to call any of them kings. First, the text itself: Iˇsme-Addu the Aˇsnakkumite, the elders of the land of Ida-Maras. , the elders of Urgiˇs, ˇ h, (and) of Hurrâ, and the elders of Yapt.ur came to Malahatum. Yatar-malik of Sina ˘ Apil-Sˆın the Aˇsnakkumite, along with the ˘ heads of Urgiˇs, ˇ ˘ hhumite and the Sudu ˘ ˘ [met(?)] together and came to me, [saying,] “Let us slay a goat and a puppy so that we [can swear (an oath)].”106

The initial list of leaders may be arranged as follows: lu´ r m Iˇs-me-d ISKUR ˇ Aˇs-na-ak-ki-imki (line 3)107 meˇs r lu´ SU.GI ˇ ma-at I-da-Ma-ra-as. ki (line 4) meˇs r lu´ SU.GI ˇ ˇ Ur-gi-`ıs ki ˇsa Si-na-a hki ˇsa Hu-ur-ra-a ki (lines 5–6) ˘ meˇs r lu´ SU.GI ˇ Ia-ap-t.u-ur (line 6)˘ ´ r Iˇsme-Addu the Aˇsnakkumite,

194


r the elders of the land of Ida-Maras, . r the elders of Urgiˇs, of Sina ˇ h, (and) of Hurrâ, ˘ ˘ r the elders of Yaptur. . After Ibal-el reports the arrival of these participants at the meeting place, he lists three more participants, apparently as the representatives of the larger group who bear the specific offer of alliance: r m Ia-tar-ma-lik lu´ Su-duˇ hi-im (line 8) ˘s-na-ak-ki-im (lines 8–9) ´ r A-pil-d ZU.EN [lu? ]Aˇ r u` qa-qa-da-at Ur-gi-`ıs ki (line 9) r Yatar-malik the Sudu ˇ hhumite, ˘˘ r Apil-Sˆın the Aˇsnakkumite, r and the heads of Urgiˇs. The first list mentions only one individual, the known king of one of the principal towns in the Ida-Maras. coalition. The two individuals listed with “the heads of Urgiˇs” appear to be high-ranking associates of the kings of ˇ Sudu hhum and Aˇsnakkum, two of the member towns of Ida-Maras. . In the ˘ ˘Yatar-malik, Sudu ˇ case of hhum had a king by this name until the year Z-L 4, ˘ ˘ who leads the delegation bear the name of the and it is odd to see the man former king. If it is not a coincidence, which is indeed possible, this could either be a close relative who did not inherit the throne or even the old king himself, who retains a ceremonial and diplomatic function.108 The reference to political negotiations with “the elders of the land of Ida-Maras. ” is unusual. The plural leadership of the alliance is generally identified as “the kings of (the land of) Ida-Maras. .” Durand recently published another example of reference to elders of Ida-Maras. , however, again lodged in a letter from Ibal-el to the king. After Ibal-el chides Zimri-Lim for going too easy on one of the Ida-Maras. rulers,109 he reports a request that had been made to him by “the elders of [the land of Id]a-Maras. ”110 As in A.2226, this group meets with Zimri-Lim’s merhûm, who regards himself as having authority over them. The phrase occurs˘once more in the mouth of one of the Ida-Maras. kings, with a slightly different application. Ibal-Addu of Aˇslakkâ tells Zimri-Lim that “the servants of my lord (Zimri-Lim) and the elders of Ida-Maras. are going to meet in Aˇsnakkum.”111 In the headings of letters from Zimri-Lim’s vassals in Ida-Maras. to their suzerain, each of the subordinate kings presents himself as “your servant.” Ibal-Addu thus appears to distinguish his royal peers from other senior leaders whom he identifies as “elders.” Should we assume that the “elders” of Ida-Maras. cannot include actual rulers, in the letters of Ibal-el? I conclude that “elder” terminology can even include kings in this specific situation, which should demonstrate the danger of regarding “elder” as a strict title for a hierarchical office. The letter from

Elders

195

the king of Aˇslakkâ shows that the kings themselves did make it a practice to attend such meetings in person. A.2226 explicitly names the king whom Ibal-el regards as the leader, and it is simply not clear who else is involved. Ida-Maras. is only a m¯atum when it meets to decide peace and war, and it surely has no “elders” except in the same limited circumstances. If Ibalel does indicate all the leadership, vassal kings included, in his references to “the elders of the land of Ida-Maras. ,” he is both subtly belittling them and emphasizing his personal authority in his dealings with them. Such authority may seem unexpected, but the vassal kings themselves attribute to the merhûm an authoritative role. In another letter, king Ibal-Addu of Aˇslakkâ ˘ to Zimri-Lim how Subram ˇ recounts of Susâ has boasted that Ibal-Addu is confined to his town, and peace has been concluded with “the merhûm,” ˘ all ˇ left nameless.112 Now, says Ibal-Addu, Subram has claimed kingship over ˇ Ida-Maras. . The very presence of the merhûm is assumed to give Subram the ˘ king Sadum-labua ˇ authority to make such a claim. Elsewhere, of Aˇsnakkum complains to Zimri-Lim about the condescending warning issued to him by the unnamed merhûm.113 In light of such interactions, we cannot assume from the merhûm’s˘role that the negotiations in A.2226, our original letter, were carried ˘out at a lower official level.

3. On Kings and Elders The picture of an elder council may be imagined by some as a sort of early democratic form, where a representative body balances to some extent the autocratic power of the king. When scholars speak of a “council of elders,” this is not usually interpreted as a cabinet of senior advisers or a selection of high administrators and royal henchmen. In some cases, however, the term is used with just this sort of application, and “elders” appear to stand in varied relation to the individual leadership of kings. Where the elders appear in combination with the messengers (m¯ar¯u ˇsiprim), the wedûtum (“dignitaries”?), and the gerseqq¯u (“palace-born”?) – all classes of people defined by their relationship to the king – we expect the elders to be in royal service, as well. Several texts present “elders” as royal envoys, who are in this way defined only by the fact that they are senior leaders available to negotiate on the king’s behalf. Often this language comes from the royal court itself. Yawi-el king of Talhayˆ um sends a letter to ˘ “elders” mentioned Zimri-Lim that would evidently have been carried by the in the opening lines. As Talhayˆ um seals a treaty with Mari, this is the team of most trusted servants who˘can settle the details. “I gave complete instructions to elders who are town residents and have (herewith) sent them to my lord.”114 King Ibal-Addu of Aˇslakkâ twice reports sending envoys whom he calls “elders.”115 Elsewhere, Ibal-Addu mentions a visit from the ruler of Qâ-and-Isqâ, called “the Isqâ-ite,” along with some elders.116 The same sort of joint appearance is envisioned by Samsi-Addu in a letter to the king of the

196


Turukkˆ u, found at Shemshara.117 In a letter from Abi-mekim to Zimri-Lim, the m¯atum elders of Kurdâ are reported to gather with “the heads of the Numhâ” as the servants of king Simah-ilanê to execute a siniqtum, some sort ˘ ˘ royal command. “The heads of the of centralized check that proceeds from Numhâ [and] the elders of the land – the servants of Simah-ilanê – have met ˘ ˘ the land.”118 In in assembly, and a systematic check has been imposed on Babylon of Samsu-iluna, the “elders of the land (m¯atum)” likewise represent the inner circle of the king.119 During the siege of Razamâ, men called elders play a central role, perhaps not so completely defined by their relationship to the king. Razamâ responds to an attack by Atamrum of Andarig with a counterattack, after which they wait ten days before elders are sent out to negotiate.120 Atamrum asks how ˇ there can be a true capitulation unless Sarriya the king comes out. From this ´ s a-lupoint on, the text attributes the voice of Razamâ to “the citizens” (lu.meˇ ki yu , lines 19, 27, 39, 41), who nevertheless appear identical to the “elders” of the original contact (line 19). In negotiations, the elders do represent the king, and by implication they are more expendable in their plural anonymity, saving the king from personal risk until it is absolutely unavoidable. While there is no sign that we have here any sort of fixed council, the letter persists in depicting negotiations between Atamrum and the town of Razamâ as such, speaking in collective voice rather than by its king. Even without a formal body, and even if sent by its king, the group leadership is given a noticeable priority in the account. Kingship has not managed to concentrate the whole existence of the town in itself. This occasional autonomy of elders from kingship is reflected in A.2417, the letter from the elders of Talhayˆ um to Zimri-Lim that informs him of ˘ 1988, 98–100). As senders of a diplotheir king’s assassination (Durand matic missive, these men take on themselves the responsibility to speak for the town, but they need not write as a preexisting political body. Rather, it is the occasion, the need to address their suzerain at Mari, that makes it appropriate to call themselves “elders.”121 These senior leaders want a new king, thus demonstrating their service to the old one, but they have a stature that can somehow stand on its own as well, so that they retain interim authority in the selection of a replacement. It appears that they are counting on a continuing service to the new ruler, giving the impression that Talhayˆ um ˘ high kingship does not include free rein in the appointment of advisers and officials. This is not a feature of complete autocracy. The independent voice of elders as royal advisers is particularly evident at Kurdâ. Two letters record separate episodes where the elders exercise pressure on two different Kurdâ kings to distance the town from Zimri-Lim, once under Simah-ilanê and later under Hammurabi: ˘ ˘ Also, regarding the fact that [Simah-ilanê] did not [write] to my lord as son, but ˘ – the sug¯ag¯u and [the elders] of the Numhâ [wrote] instead to my lord as brother

˘

Elders

197

came and spoke as follows to Simah-ilanê: “Why should you write to Zimri-Lim as ˘ son? Just as Aˇstamar-Addu always wrote to Yahdun-Lim as brother, you too must ˘ the advice of the sug¯ag¯u and the always write to Zimri-Lim as brother.” Following elders of the Numhâ, Simah-ilanê wrote to my lord as brother. Our lord should not ˘ us because ˘ be at all angry toward of these things.122 Perhaps Hammurabi (himself) will not (want to) swear an oath, or the elders(?) of ˘ pressure him so he will [not] swear that oath.123 his land will

A more extreme independence may be found at Imar, for which we have no evidence of any king in the Mari period, and where it is the elders whom Aleppo holds responsible to evict any resident Yaminite kings.124 Two preAmorrite towns in the Mari realm display elders whose roles seem to have no direct relation to any sovereign. The Terqa elders appear only in service to Dagan, seeking his favor on behalf of king Zimri-Lim, but these clearly have some sort of local status not deriving from the royal court or administration.125 A letter from the governor of Saggaratum mentions “the town elders,” evidently of Tuttul. The governor quotes Lanasˆ um, the representative of Mari authority at Tuttul, who appears to be talking about his own town. “When my lord went up to Yamhad, because of the sacrifice, he ˘ the town, and before the town was going to undertake to enter the midst of 126 elders. . . .” Of these three towns, Imar and Tuttul stand out immediately as the home of the tahtamum, a proper official body that must be “seated” to decide town ˘ below). Neither town is known to have a sug¯agum or any local affairs (see individual leader, though Tuttul accepts the presence of Zimri-Lim’s royal representative (hazzannum). We should not regard the “elders” as a body ˘ tahtamum, but only as less precise terminology for plural separate from the ˘ an outside point of view, unless there is some reason to local leadership. From deal with the local governing body on its own terms, it is more natural to use the common sweeping designation “elder.” The case of Terqa is interesting, because this was the center of a Banks-of-the-Euphrates district, complete with governor (ˇsa¯ pit.um) and local palace. At the same time, however, its stature as an old regional center and sacred site for the god Dagan seems to have been accompanied by a tradition of collective local leadership that could be called “elders.” Such independent elders also appear outside the Mari archives. Shemshara preserves a letter from Samsi-Addu to the local ruler of the Turukkˆ u people, in which he recounts negotiations with the elders of a town called Sarre in the land (m¯atum) of Qabrâ.127

4. The Elders and Collective Governance at Urgiˇs Finally, the old Hurrian town of Urgiˇs (Urkesh) deserves extended attention. It maintained a strong corporate political tradition that is often described in terms of “elders,” and we also have letters to Zimri-Lim from a king of

198


Urgiˇs named Terru who admits a remarkable degree of helplessness vis-à-vis the plural leadership of the town.128 In one letter, Terru tells the Mari king: “Because I am submitted to my lord’s pleasure, the inhabitants of my town despise me, and my head – two and three times I have snatched (it) back from death by their hand.”129 Terru speaks as one who owes Zimri-Lim his status as king, a rank that is not greatly respected by the town leadership.130 Later in the same letter, Zimri-Lim answers Terru, “I did not realize that the inhabitants of your town despised you on account of me. You belong to me even if the town of Urgiˇs belongs to someone else.”131 Far from considering Mari military presence an intrusion, Terru requests a Mari royal representative (hazzannum) for ˇ h, to quell internal˘ dissension.132 the towns of Urgiˇs and neighboring Sina ˘ (“the mouth of Urgiˇs ”) to be In this affair, Terru considers public opinion against him, and then says, “The elders of Urgiˇs must not hear this tablet of mine.”133 Obviously, the elders do not in this case represent the king’s own circle of royal advisers but are an entirely independent political force. Finally, Terru ends up admitting that he has had to flee Urgiˇs: “Now I have ˇ h as a refugee.”134 The left the wealth of [my] house, and I have gone to Sina combination of evidence in these “royal” letters offers˘ an important witness to the real power of corporate leadership in some towns of this period. On one hand, we have a town, once a great Hurrian capital city, recognized by Zimri-Lim to have a leader with the status of “king” (ˇsarrum). On the other, this king reels from one setback to another in his dealings with an independent collective leadership that cannot be regarded as a minor player in a primarily monarchic framework. A number of other letters add further depth to this picture of a strong corporate polity at Urgiˇs. In diplomatic affairs, the elders seem to take precedence, while the king remains invisible. Ibal-el’s letter A.2226 includes Urgiˇs in both lists of participants, represented generally as elders (line 5) and in the actual proposal of peace as “heads” (line 9), the only collective voice in that party. The letter from the elders of Urgiˇs to Zimri-Lim (FM II 125) has nothing to do with its king, unlike the one from Talhayˆ um (A.2417). The ˘ tablet accompanies an emissary from Urgiˇs who is bringing a ransom to free a woman captured during the looting of Hazzakkannum (lines 17–20). A letter from king Ibal-Addu of Aˇslakkâ shows˘ the whole town of Urgiˇs acting as free agents without Terru’s involvement. Some individual and his sidekick captured an “encampment,” and the town is forced to render a decision in full meeting (puhrum). Ibal-Addu evidently had complained to Zimri-Lim, who ordered the˘release of all they had seized. The king of Aˇslakkâ reports that “Yans. ib-hadnu and Haziran went to Urgiˇs and called for a meeting. ˘ The Urgiˇsites˘then responded [as follows]: We shall release everything from 135 the encampment. . . . ” Notice that when Zimri-Lim orders these two men from Urgiˇs to give up their plunder, they go back not to their king but to the town as such, which then renders a collective decision.136 The word for

Elders

199

“meeting” (puhrum) is used for the gathering itself, which does not speak. ˘ speak, it is as a town, as “the Urgiˇsites.” “The Urgiˇsites” are When the people ˇ the entire subject of another letter, sent by Sadum-labua of Aˇsnakkum, and ˇ a letter from Amut-pa-el of Suduhum mentions another puhrum “meeting” ˘ in Urgiˇs, this time as a council of ˘war.137 There is some possibility of political change, so that Urgiˇs had a king only during the last few years of Zimri-Lim’s reign.138 Our letters from the king ˇ named Terru all come from the years Z-L 11–12, when Sadum-labua was king 139 of nearby Aˇsnakkum. Ibal-el’s letter would have been composed just earlier, during Z-L 10, and FM II 125 a few years earlier still.140 If Terru’s arrival was a late development, it did not dampen much the abiding power of Urgiˇs’s elders. The royal correspondence that mentions the collective political voice ˇ of Urgiˇs comes mainly from the same late period, when Sadum-labua was 141 king of Aˇsnakkum. Like Imar and Tuttul, which we examine later in this chapter, Urgiˇs was already a town by the third millennium, and we cannot assume that the political situation at Urgiˇs reflects only the recent developments of the early second millennium. The tradition of a powerful collective balance to leadership by kings may be the inheritance of a long urban history. Current circumstances in Urgiˇs certainly had weakened the position of the local king, but the push and pull of individual or “exclusionary” and collective or “corporate” leadership may have gone back centuries. At Urgiˇs and elsewhere, the language of collective “elder” leadership appears to derive not from the administrative systems of palaces but from the venerable priority of age. “Elder” terminology surely is founded in the small communities of villages and mobile pastoralists, and it must ultimately predate the appearance of “kings” who ruled more complex archaic states. In the Mari archives, the varying degree of autonomy for collective leadership called “elders” reflects the different adaptations to and histories of this emergence of more complex polities through the third and early second millennia.

5. Tribal Elders Although the focus of this study is the town, not the tribe, I should note that we find tribal elders in the Mari archives, as well. It appears that negotiation with the larger world is a primary inspiration for “elder” terminology in this domain once again. A brief note from Kibri-Dagan, the governor of Terqa, serves only to pass on the information that a chief named Iridanum is on his way to visit Zimri-Lim, with ten elders of the Hana. “Iridanum and ten elders of the Hana are going to the king. Notify ˘the king.”142 Before ˘ with the king of Qattarâ/Karanâ, Atamrum of Andarig killing the treaty ass .. is reported to have insisted to Asqur-Addu that he must still talk to a man named Haqba-hammˆ u and the elders of the Numhâ: ˘ ˘ ˘

200


“You are my son – remain so. I intend to speak with Haqba-hammû and the elders of the Numhâ.” He summoned Haqba-hammˆ u and the˘ elders˘ of the Numhâ and gave ˘ them the ˘following message. .˘. .143 ˘

In the moon-god temple of Harran, the sug¯ag¯u and the elders of the entire Yaminite tribal confederacy ˘meet with the rulers of Zalmaqum to ratify a treaty, as the basis for a joint attack on the Simalite town of Der:144 Asdi-takim and the kings of Zalmaqum, along with the sug¯ag¯u and the elders of Yaminites, have slain the ass in the moon-god temple of Harran. The kings of the land of Zalmaqum kept advancing the following proposal.˘. . .145

Another letter specifies twenty Yaminite elders who have gathered with Atamri-el, a leader among their Uprapˆ u tribe,146 to discuss peace with the 147 Simalite confederacy. Although individual leaders are present and active, the prominence of collective leadership is again impressive. There should be no confusion over the fact that we find “elders” for both towns and tribes. If the use of the word “elders” with towns represented a particular urban political institution that could be distinguished from tribal elders, then we would have to settle the relationship between the two more precisely. In fact, however, “elders” are less an institution than a terminology that can describe a range of corporate leadership, representing either the monarch or the whole community. We need not imagine that the so-called elders of towns and tribes must follow identical modes of operation, though in fact there is little reason to expect great differences between them. Only their constitutions may have varied somewhat differently, according to the distinct lines of social organization.

e. heads A second defining term for the collective leadership of towns, the “heads” (qaqqad¯atum), produces a minor metaphorical confusion. The anthropomorphic image is used most often in the plural, for both the town and the larger m¯atum. We do not generally find an individual leader called “the head” (qaqqadum) of town or land. If no single head is defined by its relation to the town or land, but all heads hold this status together, we must wonder what lesser bodies to join to them. The answer seems to vary according to context, though “head” terminology is linked primarily to collective leadership. First of all, “heads” may be identified with households. “Heads” are also associated with tribal leadership, perhaps sharing the assumption of family structures with the household heads. It is interesting to discover that even in the political context of kingdoms, the language of “heads” is still reserved for the plural, as part of a tradition of collective leadership.

Heads

201

1. Household Heads I have already discussed ARM XXVI 365, the letter in which the Kasallukites are identified with individual “houses.” The full phrase that introduces the ´ s qa-qa-da-at Ka-sa-al-lu-ukki , so that they are more specifipopulation is lu.meˇ cally “the heads of Kasallu(k).” This text offers the identification of these leading men as owners of houses, complete with the dependents and property that fill them. ARM II 75:9 –11 , also discussed above, tells us that the town of Isqâ-and-Qâ has 200 heads, which might suit the household level, if we consider that Terqa was able to provide only 400 resident “sons” for a work crew.148 Durand estimates that a crew of this size would have come from a total population of about 2,000 (1998, 602). In a third text, eighty heads of the “land” (m¯atum) of Imar go to Hˆıt for a case requiring justice ˘ by river ordeal, where the accused or a representative would undergo a test involving some kind of passage through the waters of the Euphrates. None of these examples seems to connect these town heads with any royal court, and the Kasallu(k) text provides the simplest explanation, that these are indeed heads of households. Liverani observed that the Alalah village lists are based on the “house” ˘ 149 References to leadership by heads unit, each defined by a personal name. of household, identified as patresfamilias, are commonplace, and it is worth noting that this category is not easily matched to the various terminologies for collective decision making in the Mari archives. The handful of texts presented in this section do suggest this solution under the “heads” terminology (qaqqad¯atum), but caution is still warranted. Even if such a household-based social structure is the norm, as is surely the case, it is more difficult to establish with confidence who would have qualified as a “head” in town leadership and whether their full number would have been greater or less than the sum of the patresfamilias.150

2. Tribal Heads It is difficult to disentangle the town heads from those of tribes. This is especially evident in a legal text discussed at length by Dominique Charpin: ARM VIII 85+ (1997b, 343–4, text no. 2). In this text, the thirty-seven heads of the Yumhammu clan are described as the town of Sapiratum itself in the ˘ list (lines 50–4):151 tablet’s witness Pulsi-Addu of the Yabasˆ u (division) and his kinsmen made a claim on a palace-owned field, saying, “(This is) our field.” The town of Sapiratum met in assembly, Zimri-Lim issued a decision in the temple of Hanat, and he imposed an oath on the town of Sapiratum. The 37 heads of the ˘ hammû who had sworn the oath in the temple of Itur-Mer affirmed that (it Yum(a) ˘ was) a palace-owned field. 13 rope-measures in Azrahatum, and 7 rope-measures in the village of ˘ Yabruk-ilum.

202


Total: 20 rope-measures of field. Whoever lays claim to the field of the king in the town of Sapiratum and of the kinsmen of Lahmumu must pay 10 minas of silver to the palace.152

˘ In this tribal town, where the population is identified by the names of both town and tribal division, it is interesting to find that the “head” terminology is associated with the tribal definition rather than with the a¯ lum. Perhaps the tribe is still regarded as the more foundational point of social reference in this Simalite community. The question of how “heads” relate to town and tribe comes up again in the case of Isqâ-and-Qâ in ARM II 75. The name of the town is mentioned three times, once with each new definition of the active parties. First with ´ meˇs and then with elders, the town is rendered Is-qa-a(ki ) u` Qa-e-em, adding LU the determinative for place (KI) with the second occurrence. Only with the gathering of the heads is the order reversed, allowing the leadership category to be joined directly to the Qâ element, qa-qa-da-a[t] Qa-a-em u` Is-qé-e-em.153 Among the recently published royal correspondence of Zimri-Lim, there are eleven letters from Yumras. -el, the ruler of Isqâ-and-Qâ (or Qâ-and-Isqâ), and one of these makes clear that we are dealing with paired towns. Zimri-Lim is quoted as having told the local king, “Appoint Bunum over Isqâ, and appoint whomever you wish to appoint over Qâ,” evidently for comparable roles.154 Nevertheless, in ARM II 75, the “heads” are carefully associated with the “Qâ” element only, the element that represents the common basis for both names. If this name were the original tribal designation, the definition of “heads” would once again tend to follow the kinship identity.155 In one last case, discussed above with elders, the word “heads” is preferred for the tribal face of a kingdom with both mobile and sedentary dimensions. Another letter divides the servants of king Simah-ilanê of Kurdâ into two: ˘ first “the heads of the Numhâ” ([lu´ ].meˇs qa-qa-da-at Nu-ma-< ha->ma) and then ˘ ˘ 156 the kingdom “the elders of the m¯atum.” Numhâ is the tribe that dominates ˘ of Kurdâ, which as a settled m¯atum collects its leadership as “elders” rather than as heads. It is not clear whether the paired titles represent truly separate groups or a kind of merism defined by the two primary identities of the kingdom.

3. Senior Officials We should not attempt to unify all uses of “head” terminology in a single social location. Some “heads” represent a much smaller circle at the center of government. In large kingdoms, these are associated with the m¯atum rather than with the a¯ lum, as is natural for such states. Samsi-Addu proposes that a high official at Terqa should be replaced by “a reliable man, (one of) the heads of the land.”157 Later, Zimri-Lim uses similar terminology to count the leadership who were present with him for the alliance oath sworn with the king of Razamâ. Atamrum, the king of Andarig, has asked for military

Words for Assembly

203

aid against Razamâ, and the merhûm Ibal-el counsels that Zimri-Lim refuse, based on the following suggested˘ reply: ˇ Now then, discuss (this) with Hammurabi (of Kurdâ), Hatnu-rabi, and Sarru˘ ˘ kikalima, the kings who are in your (Atamrum’s) presence, and with your servant ˇ Yanuh-samar. There is a blood-bond between me and Sarraya, and a binding agreement˘has been ratified. One hundred Hana and one hundred of my servants from ˘ 200 of my trustworthy servants, the heads the Banks-of-the-Euphrates were with me; of my land, joined (lit. “stood”) in the blood-bond, and indeed I swore an oath to 158 ˇ him (Sarraya).

This division sets a tribal following called Hana against the inherited Mari domain of the Ah Purattim. It is remarkable ˘to see that the specific definition of this “reliable”˘ elite matches almost exactly the one used by Samsi-Addu, using the better attested qaqqadum for the heads. A letter to Zimri-Lim from ˇ Sadum-labua, king of Aˇsnakkum, associates the “reliable man” with a troop of fifty, perhaps not so much in a military context as to indicate a group headed by someone from the king’s trusted leadership circle.159 The 200 “heads of my land” in Ibal-el’s letter do not seem to derive from household headship, though the language may display how kings borrow from the ideology and terminology of corporate polity to define their own royal administrations.160 The same more limited use may be found in smaller states. In the Ibal-el letter A.2226, the heads of Urgiˇs appear as a subset of or as a more specific description of their contingent of elders. These should be senior officials ˇ again, who join the king of Sudu hhum and a representative of Aˇsnakkum in ˘˘ bringing the peace proposal to Ibal-el. On a similar scale, the oath for the vassal kingdom of Karanâ swears loyalty to Haqba-Hammu “and the chief ˘ of Asqur-Addu, their (head) servants” (ÌRmeˇs qa-qa-da-tum), under˘the rule 161 king.

f. words for assembly Three main words are relevant to our discussion of political assembly. The Akkadian puhrum literally refers to “assembly” or “meeting,” and this is the ancient term˘most often treated as the basic designation for popular assemblies. In the Mari evidence, the word “puhrum” refers to generic “meeting” and is not identified with any institution ˘of town assembly. The one word for assembly in Mari evidence that is directly linked to the town is the West Semitic noun “tahtamum” (or tâtamum). This term refers only to gatherings ˘ the old Euphrates valley towns of Imar and Tuttul, a of local leaders in limited use that suggests a local political tradition with fairly specific application. One other West Semitic word, rihs.um, is associated with the assembly of pastoralist hana. The rihs.um is never ˘defined by towns, and it designates ˘ ˘ occur within a meeting of mobile pastoralists, the talks or negotiations that not the “assembly” as such.

204


All of these terms must be addressed as part of any treatment of the assembly in Mari evidence. In the end, however, the entire notion of “popular assembly” may hinder our understanding of the various meetings and structures of collective decision making in the Mari archives. Even the tahtamum ˘ best probably refers to a limited group that is “seated” for its meetings. The examples of more inclusive meetings do not use any of this terminology, and they seem to have been quite ad hoc. More often, collective decision making seems to have operated on the basis of fairly limited representation rather than with full participation.

1. The Mesopotamian Assembly There is a considerable literature on Mesopotamian assemblies, much of it somewhat old or based on old analyses. The modern discussion begins with the two classic articles by Jacobsen, which are still cited frequently both within the field and beyond (1943, 1957).162 Jacobsen gathered evidence for the legal authority of assemblies in both the Assyrian and the Babylonian spheres, especially for the early second millennium (1943, 161–4). Based in part on the persistent notion of a divine assembly, he concluded that the earlier assemblies had wider powers, especially in their capacity to choose and depose kings (p. 165–72). Even the later myth of Marduk’s elevation to rule over all the gods presents this as the collective divine decision.163 Jacobsen envisioned a more inclusive assembly tradition than his own evidence perhaps warranted. The alternation of “assembly” and “town” language says nothing in either case about how much of the public was included (cf. p. 162). He cites a Babylonian proverb to show the inclusive nature of the assembly: Do not go to stand in the assembly; do not stray to the very place of strife. It is precisely in strife that fate may overtake you; besides, you may be made a witness for them so that they take you along to testify in a lawsuit not your own.164

This reference applies more to the responsibility to serve as legal witness than to townwide decision making and offers little help with the makeup of larger meetings. If anything, the proverb shows that the “assembly” in question includes only those who bother to turn up and who are thereby liable for the ancient counterpart of jury duty, viewed in this text to be as attractive as it is for us today. It is clear that such town-based assemblies did exist in early southern Mesopotamia, but it is difficult to judge their actual scope of activity and power, because we have too little evidence. The earliest cuneiform texts from Uruk, dated to the end of the fourth millennium, do mention both an “assembly” (UKKIN) and a “leader of the assembly” (GAL:UKKIN), without further information (Selz 1998, 301; Glassner 2000, 45–7). Glassner observes that there are a few mid- to late third-millennium examples where

Words for Assembly

205

towns are portrayed as choosing (“elevating”) their own kings, including Enmetena and Irikagina of Lagaˇs. In a royal inscription of the Akkadian king Naram-Sˆın (ca. 2200), “his city” is said to express its wish to him (pp. 43–4). This form of expression, which also occurs in the Mari evidence, suggests some level of active political power beyond the limited domain of delegated legal responsibility. We cannot say with Jacobsen, however, that the assembly was the ultimate political authority in earliest southern Mesopotamia (Selz 1998, 291 and n. 40; Heimpel 1992, 17n44). An important corollary discussion has been provoked by Katz’s article on the two “assemblies” described in the Sumerian tale of Gilgamesh and Agga. Katz argues that one can separate “literary materials” from “actual material” within the existing text, and that the “literary” stratum cannot be used to reconstruct ancient society, particularly the assemblies of older and younger men (1987, 105–8, passim; cf. Kuhrt 1995, 31). Long before Katz, Jacobsen had read these as expressions of a single political phenomenon, the citizen’s assembly (1943, 165–6). Aside from the question of how many assemblies ever may have coexisted at Uruk, surely it is impossible to isolate “actual” historically based from fictional literary material in such texts (cf. Van De Mieroop 1999, 124n6). In fact, the whole argument may be founded on a misunderstanding of assembly terminology generally, as well as of the specific text. First of all, the “assembly” or “meeting,” whether Sumerian ukkin or Akkadian puhrum, is not a formal body but a vocabulary for talking ˘ about large public gatherings. Properly, there is not even one “body” that can be separated from the second debated institution. The lines in question are better translated in a way that recognizes “old men” and “young men” as opposed constituencies in the decision-making corporate body: “in the convened assembly, the old men (or ‘fathers’) of the city . . . ,” and “in the convened assembly, the young men (or, ‘able-bodied men’) of the city. . . .”165 While there is no reason to discard the social reality expressed in the tale of Gilgamesh and Agga as fictional, whatever its exact setting, the conflict between old and young among the gathered people does represent a literary topos. In the conflict that follows the death of Solomon according to the Hebrew Bible, his son Rehoboam is portrayed as consulting both the older and the younger generations of leadership, and the elders again prefer a more cautious course of action (1 Kings 12). Interestingly, one of the anthropological studies in Councils in Action observes that the men who meet in the village councils of the Ibo are graded based on age, literally as “old men” and “young men.”166 In this setting, there is only one “assembly” tradition, and the division into two is ideological, without any strict institutional expression. All three of these far-flung examples derive from the same essential phenomenon, the general meeting, loosely defined. Various sources from the early second millennium have confirmed the durability of an assembly tradition across Syria-Mesopotamia. Outside of the

206


Mari archives, Babylonian and Assyrian texts both display some version of this language. In the kingdom of Babylon, it is found especially as a “meeting” (puhrum) for legal judgments (already observed in Oppenheim 1936, 224–35; ˘Driver and Miles 1952, I:78–9). An Old Assyrian text makes “the city” and “the elders” of Aˇsˇsur synonymous in a negotiation with spokesmen for the colonies, and the local colonies may meet for decisions that involve “small and great” (s.aher rabi; Larsen 1976, 162–4; also Michel 2000, ˘ 113–14). The situation in first-millennium Babylonia has been examined by Dandamayev (1995), a situation that impresses Van De Mieroop as more favorable to the citizenry than earlier millennia had allowed (1999, 118– 38). It seems that evidence for assembly in Hellenistic Uruk may still reflect local Mesopotamian patterns with more limited participation, rather than the Greek idea of including all citizens (van der Spek, 1987, 71–2).

2. The puhrum ˘ This Akkadian word for “assembly” may be derived easily from the common Akkadian verb “pah¯arum” (“to assemble”) and appears frequently in the ˘ but it covers a range of gatherings. No text in my Mari correspondence, reading used puhrum with a town name to represent an urban assembly, “the assembly of GN,”˘ and the only puhur m¯atim is not a political body but rather a ˘ all to hear.”167 I have encountered no nontechnical phrase meaning “for puhur a¯ lim, “town assembly.”168 Thus in this mass of evidence, the word “pu˘hrum” is not used to name any ancient town or village institution, beyond ˘ describing a particular meeting under the familiar umbrella of collective polity. No specific membership or bounds of participation is assumed by the term.169 Far from any supposed popular assembly is the puhur ˇsarr¯ı, coined to ˘ may be asdescribe meetings of the kings of Ida-Maras. .170 Tribal gatherings signed this name without limitation to any particular group: the Numhâ, the Hana (under Zimri-Lim), and the Yaminites.171 The assembly of the˘ Hana ˘ ˘ includes invited kings and is not a simple gathering of tent-dwelling army units. Because the Hana form a key part of Zimri-Lim’s military support, ˘ Mari officials are usually interested in their gathering because war is in the 172 air. As shown in its combined use with the tahtamum of Tuttul, puhrum is ˘ be what is going on ˘ also the most general word for “meeting.”173 This may in Urgiˇs, as seen in ARM XXVIII 69:9 (see above section on elders). No institution is named, but a meeting is called for the town as such.174 The most difficult use of “puhrum” occurs in military contexts. To begin ˘ by the fact that the regular army of “muswith, the situation is complicated tered” troops appears to be called the pihrum, from the same verbal root (phr, ˘ the army that assembles, the use ˘ “to assemble”).175 Even when it is clearly of “puhrum” may carry with it a political or decision-making aspect, as well. ˘ For example, a high Mari official named Yasim-el reports to Zimri-Lim the

Words for Assembly

207

attempted rebellion of a general at Qat.t.arâ. This general uses the puhrum ˘ of the army to air his griefs and to persuade it to join him against the king: On the 5th of Lilliatum, (as the day) was getting on, the Numhâ army began to ˘ assemble in the midst of Qat.t.arâ. [When] the army had assembled, Kukkutanum (the general [rab amurri] of Asqur-Addu) [left] his town of Nunasaru, showed up at the assembly of the army, and laid his complaint [before] the army as follows. . . . (The rebellion is not my fault.) Kukkutanum said [this and] many other things to the assembly of the army, and he both put the army in a craze and moved the consensus of the muˇskênum to revolt against Haqba-hammu their [lord]. [So Haqba]-hammu ˘ ˘ carry unknowingly sent Kakiya to the assembly of˘ the army [at] Qat.t.arâ [in˘order to] 176 out deliberations and to launch a military expedition(?). They then killed [that man], while the muˇskênum went over to the side of Kukkutanum, and they (all) began to make an assault on Qat.t.arâ.177

Though this may have begun as no more than a normal military muster, it becomes the domain of momentous political decision. We encounter a similar meeting of the Yaminite hana in a letter from a man named Yahdun-Lim, ˘ ˘ 178 The perhaps a Yaminite leader, likewise to decide between war and peace. puhrum of an army, therefore, is often (if not always) more than a simple ˘ gathering of troops, in that it also incorporates a decision-making aspect.179 Perhaps the simple gathering is intended in one text from the end of ZimriLim’s reign, which reports that the troops have not complained when mustered (ina puhrim).180 At any rate, the various occurrences of “puhrum” for ˘ ˘ toward a predemocratic popular assembly. war do not lead

3. The tahtamum ˘ All the relevant evidence for this word was gathered in a single article by Durand, and I want only to place the tahtamum in the larger framework of my study.181 Most important is the fact˘ that it demonstrates the reality of pre-Amorrite town institutions preserved in the Old Babylonian Mari texts. Beyond the impression given by the context of use, the relation of the tahtamum to the town as such is demonstrated in one occurrence of the phrase˘ “the tahtamum of Tuttul,” where the link is explicit.182 This definition ˘ is what is lacking for any occurrence of the noun “puhrum,” by the town itself and it indicates a genuine town institution, linked to the very identity of˘ what it is to be that town. Unlike any of the terminology encountered so far – the ´ s GN), elders, heads, or puhrum meeting – the tahtamum gives GN-ites (lu.meˇ ˘ identity of a “town” (¯ ˘alum), which specific institutional form to the corporate is essentially political. The word “tahtamum” is evidently local and technical, preserved only ˘ in the middle Euphrates valley and the specific seated at Imar and Tuttul council at which important town business must be decided.183 At Imar, an outsider may refer to “their meeting (puhr¯ıˇsunu) in the tahtamum”: ˘ ˘

208


At the time of my going up to Imar, I sent a tablet to my lord about Is. s. ur-Addu. Now, after I arrived at Imar from Aleppo, and the boats were held up because of Is. s. ur-Addu, I had words with them at their meeting in the tahtamum (council) about ˘ the boats that they had held up.184

Both of these Euphrates towns are known from the Ebla texts and in the Amorrite period look back on several centuries of uninterrupted occupation, at least.185 One Ebla lexical list appears to confirm the antiquity of the term in the entry KA.UNKEN = da-da-mu.186 In the text just quoted, Durand considers the puhrum to be the meeting, ˘ and the tahtamum to be the deliberation itself, and following the contracted ˘ form tâtamum, suggests that the noun may be derived from the Akkadian verb “atwûm,” “to converse.”187 No detail of this or any other published example indicates a special association with speech, however, and the Akkadian etymology does not fit well with the alternative, uncontracted spelling with -h-. Regardless of its etymology, the word “tahtamum” has no other known ˘ ˘ application besides this body and appears to have been coined for it specifically. This contrasts with puhrum, which has broader use. By its absence from Akkadian, tahtamum should˘ be a Syrian Semitic noun, but not Amorrite, if ˘ as recent to the second millennium. One might consider an we define that etymology with the Hebrew and Aramaic root h.tm, “to seal,” where this would be the group with authority to apply the town seal.188 The tapras- form can yield the agent of the verbal action, as in Akkadian tamk¯arum, “merchant,” and taml¯akum, “adviser,” as well as Hebrew tôˇsa¯ b, “settler” (verb yˇsb). Late Bronze Emar (=Imar) shows the use of town seals in this region.189 It is not clear how inclusive is the circle of leadership assembled in the tahtamum. The Ebla text relates it to the en-en, who would be top leaders ˘ At Tuttul in the Mari letters, the link to “men” (LU ´ meˇs )190 or “sons” only. meˇs 191 (DUMU ) suggests a broader corporate base, but the tahtamum may ˘ be conceived only to represent the larger constituency, as defined thus. In one Tuttul letter, “the sons of the town” are said to provide thirty men for a local policing operation, after Lanasˆ um had summoned a meeting of the tahtamum (FM VI 4 = A.402). This language may indicate no more than ˘ representation rather than full assembly of the town’s residents (vs. Lafont 2002). One characteristic of the tahtamum is that it is “seated,” a posture expected for a more limited number˘ with high status.192

4. The rihs.um ˘ As with the word “tahtamum,” the full published evidence for the rihs.um ˘ ˘ appears in a single presentation by Durand (“Le rihs.um des Hanéens,” AEM ˘ I/1, pp. 181–92). Like the word “tahtamum,” rihs.um is a non-Akkadian, West ˘ entirely ˘separate, range of use. The Semitic noun with a limited, though rihs.um never designates a town assembly but rather relates only to gatherings ˘

Words for Assembly

209

of tent-dwelling “Hana” pastoralists, sometimes at the initiative of outside ˘ authorities. This terminology is at home among both the Yaminites and the Simalites (e.g., A.4530-bis and ARM XXVI 45), but I am not aware of its use among other pastoralists such as those of the Sutˆ u. The recent R volume of the Chicago Assyrian Dictionary (CAD) defines the rihs.um as a “gathering” (s.v. rihs.u B) and the cognate verb as “to gather(?)” ˘ rihsum a “palabre,” the French expression of ˘ rahasu D). Durand calls the (s.v. . . ˘ ˘ an African word for discussions between leaders (cf. English “palaver”).193 It seems clear from the texts that the rihs.um is not the assembly but the ˘ talks that ensue. In one case, Lanasˆ um reports from Tuttul that the confederacy of Zalmaqum and the Yaminites are meeting to confirm an alliance: “Zalmaqum and the Yaminites have met, and they have engaged in talks (rihs.um). The object of their talks is alliance.”194 The meeting comes first ˘ pah¯arum, “to assemble”), and the rihsum follows, as contingent on and (verb . separate˘from the act of gathering. The ˘focus of the rihs.um appears to be ˘ conversation. A report on preliminary negotiations for peace between ZimriLim and the Yaminites mentions a request that up to twenty (Yaminite?) elders be sent to the speaker. This unknown person then says, “Let them engage in talks with (some group), so then I can hear their offer (‘words’).”195 Often, the rihs.um requires travel on the part of the participants,196 and the nature of the˘ meeting place may vary. Such rihs.um talks could take place in ˘ um has to leave his base at Tuttul in ora town,197 but in one case, Lanasˆ der to meet for a rihs.um in the steppe with leaders of the Yaminite Uprapˆ u tribe: “Fifty Uprapˆ u˘ (tribesmen) came to me at Tuttul, and I went with the U[prapˆ u] to the steppe (nawûm), where I engaged in talks (rihs.um).”198 ˘ A rihs.um could be initiated in different ways, but the term applies specially ˘ to talks related to interactions between separate groups. In the letter just cited, Lanasûm seems to initiate the talks as a representative of the Simalite king Zimri-Lim, but he meets the Yaminite leaders in their own grazing lands. Often the rihs.um seems to be initiated by the party that is defined by kingdom or land, as˘ it enters negotiations with groups defined tribally, either as Yaminites or as Simalite “Hana.” In the latter case, the Hana category ˘ emphasizes the pastoralist way˘ of life of the people in question. A letter from a Mari official named Aˇsmad quotes Zimri-Lim’s instructions: Go chase down the Hana and engage in talks (rihs.um), so that the Hana may assemble ˘ ˘ ˘ one man to my and [at] the raising(?) of the torch, all the Hana may come as 199 ˘ assistance.

Participation in rihs.um talks may sometimes be quite broad, although the ˘ to evaluate. When Lanasˆ details are usually hard um journeyed from Tuttul to the steppe, it seems that the fifty Uprapˆ u representatives were not sufficient to speak for the tribe in the given circumstances. When Zimri-Lim asked that his Simalite Hana gather in rihs.um so that they would come “as one ˘ participation seems ˘ man,” his hope of full to rest on the assumption that all

210


potential warriors would be present at the talks. The rihs.um seems to provide ˘ a setting for authorized complaint against senior leaders by individuals of lower rank. Zimri-Lim’s trusted aide Asqudum writes to his lord that he anticipates criticism of the chief of pasture Ibal-pi-el in talks by the Hana: ˘ I am afraid that at the time of the talks (rihs.um), Ibal-pi-el may come under criticism ˘ by the (Simalite) Hana. . . . ˘ Now, my lord wrote to me about Itilim before the talks (rihs.um), and I will send ˘ of Ilan-s. urâ), so he Itilim either to (the town of) T. abatum or to Haya-sumu (king ˘ will not be present at the talks. I will keep (here) the others who might stand up and complain about Ibal-pi-el in the talks.200

It is not always clear that the rihs.um is so inclusive, however. The letter that ˘ h describes them as involving “the kings mentions talks at the town of Zalpa of Zalmaqum” and “the kings of the˘ Yaminites,” with no indication of wider participation. This is the perspective of one of the Yaminite kings, Yasmah˘ Addu of the Yarihˆ u tribe.201 ˘ Altogether, the word “rihs.um” does suggest a tradition of assembly among West Semitic speakers, but ˘not related to the collective governance of towns. Unlike the urban tahtamum of Imar and Tuttul, the rihs.um is often defined ˘ tribal peoples, sometimes likely to ˘ be mobile pastoralby the participation of ists. The rihs.um is very much an Amorrite tradition, both in linguistic and in ˘ however old the word and the custom may be. The tahtamum social terms, ˘ Ebla of the towns, in contrast, seems to predate the Amorrite age, if the evidence is rightly read. Any link between the tahtamum and tribal peoples is indirect, by its limited appearance at Imar and ˘Tuttul, two towns with noticeable Yaminite connections. It is difficult to conclude that the tahtamum ˘ the has specific Yaminite origins because of the Ebla attestation, though term seems to have become lodged in terrain with strong Yaminite character during the early second millennium. The rihs.um, in contrast, shows no particular attachment to one of the paired tribal˘ confederacies.

5. Assembly and Individual Leadership Even where collective leadership survives with some genuine authority, it takes its place in the web of political hierarchy that overlies the ancient Near East. Individual rulers retain their superior rank, if not always with autocratic powers. At Tuttul, the tahtamum may be convened by Lanasˆ um, the senior Mari ˘ force a decision and may need their authorization in officer, but he cannot local affairs. In one case, Zimri-Lim wants some looters arrested at Tuttul, but no action can be taken except through the tahtamum (A.402). Another ˘ for labor and cautiously time, the tahtamum refuses to go along with a request ˘ insists that the people are too few. “I seated the tahtamum (to have 1,000 trees ˘ not comply.”202 Durand cut), and I spoke to them (about it), (but) they did

Imar, Tuttul, and Urgiˇs

211

observes that the two Imar texts show the body meeting only at its own initiative, though there is too little data to prove a greater autonomy than at Tuttul (ME 2, p. 39). Neither town has a king in this period. We have already seen the extent to which corporate leadership at Urgiˇs followed its own lead, even directly opposing Terru, the king approved by Zimri-Lim. All three towns – Tuttul, Imar, and Urgiˇs – had to submit to a distant suzerain, whether based at Mari or at Aleppo. Nevertheless, their collective decision making cannot be construed as part of a simple hierarchy of authority under a royal administration. These ancient towns should be regarded as the peers of the many small kingdoms attested in the Mari archives as vassals of Zimri-Lim or of other major kings, and their collective political forms were not narrowly limited to judicial affairs or the concerns of private individuals. On the contrary, the collective terminologies appear most often when describing the actions of towns in contact with the outside world. The world of tribes displays a similar balance of local initiative under an ultimate but usually distant authority. One time, the same Lanasˆ um requests the assembly of Yaminites in tribal rihs.um (A.4530-bis), though I found no ˘ evidence of a Simalite (or Hana) counterpart convened at the initiative ˘ of Zimri-Lim or his administration. Texts frequently mention gatherings of the (Simalite) Hana in connection with the verb “pah¯arum” and noun “puhrum,” but again,˘ never by outside authority. A.3577 has˘ Hammurabi, the king˘ of Kurdâ, address the tribal assembly of the Numhâ (pu˘hur Numhˆım) as ˘ him.˘203 Zimri-Lim’s ˘ “his m¯atum,” though the gathering is not summoned by relationship to the Simalites appears similar, and the tribal base of Amorrite kings seems to retain a higher degree of autonomy than the conquered peoples within their m¯atum borders. In the end, Mari provides us a productive new source for understanding collective leadership in town and tribe but not to demonstrate a full town assembly of free men. The one term that indicates a true town council is “tahtamum,” but this may be a small group, far from a gathering that rep˘ resents every household. More inclusive participation does, however, seem to characterize the tribal rihs.um, especially as associated with the pastoralist ˘ Simalite tribal groups. At the least, any traelements of the Yaminite and ditions of inclusive town assemblies were not preserved in particular terms such as puhrum. ˘

g. imar, tuttul, and urgiˇs : old towns with strong collective traditions It should be clear from the evidence gathered in this chapter that the town was a key expression of collective political life in ancient Mesopotamia. How are we to evaluate the collective tradition against the individual power of kings, however? Must such group exercise of power always function under the ultimate authority of kings and other individual leaders? These questions

212


could be addressed from any number of starting points in the Mari evidence, but three towns display particularly vibrant forms of collective decision making: Imar, Tuttul, and Urgiˇs (Urkesh). Imar and Tuttul are closely related, situated on the same part of the Euphrates River in northwestern Syria, in the same environment, with riverbased cultivation possible only in the narrow valley and close ties with pastoralists from the grazing lands north and south of the Euphrates (Porter 2000, 14–15, 452–6). Both towns had strong affinities with the Binu Yamina tribespeople in the early second millennium, but their common ties go back long before the appearance of the names known from Mari. Imar and Tuttul also belonged to the sphere of West Semitic speakers long before the arrival of the “Amorrite” peoples.204 Urgiˇs reflects a completely different history and environment. Located in the northeastern reaches of the Habur River ˘ basin, Urgiˇs had already been populated by Hurrian speakers since at least 205 the late third millennium. The more northern site enjoyed more rainfall than the middle Euphrates region, and agriculture was managed by dry farming rather than irrigation. All three towns, however, display strong collective political traditions in the second-millennium Mari evidence, with repeated references to a dominant group preference. This evidence is preserved in the texts in part because all three towns were of moderate size and influence. During the reign of Zimri-Lim, Imar and Tuttul were the two important centers on the western segment of the Euphrates that reached nearest the great Syrian kingdoms of Yamhad and Qatna, even further west. They remained crucial to trade ˘ between Mesopotamia and western Syria, as well as with northern and travel sites such as Carchemish. Urgiˇs was no longer the great Hurrian capital that it had been some centuries earlier, but it still represented one of the bridges between the Mesopotamian and Anatolian spheres, at the northern periphery of Mari’s interests. It is striking that all three of these towns, in spite of their separate situations, were significant fixed centers through much of the third millennium. Given the longevity of all three sites, we cannot quickly attribute their collective political traditions to Amorrite tribal newcomers. Archaeological excavations have not yet demonstrated continuous occupation for the three towns. Nevertheless, the combination of the unusually vigorous collective decision making and the persistent prominence of all three settlements through the centuries before our Mari evidence leads me to suspect ancient rather than new Amorrite political practices.

1. Imar The town of Imar was located on the great bend of the Euphrates River, where it turns eastward across inland Syria after descending from the mountains in Turkey. Excavations at Meskeneh produced extensive finds from the


213

Late Bronze II period, when “Emar” was an outpost of the Hittite empire, with the finds including a large number of cuneiform texts (Fleming 2000, 1 and n. 1). To everyone’s disappointment, no remains of the earlier town were discovered, and Margueron concluded that the pre–Late Bronze site must have been located elsewhere, perhaps closer to the river, whose course has now been flooded.206 There was no question that Imar had been a prominent town through much of the third and second millennia, however, and communication between the western Syrian centers of Ebla or Aleppo and the peoples of Mesopotamia commonly passed through Imar. Both the early second-millennium archives of Mari and the mid- to late third-millennium archives of Ebla contain so many references to the town that Durand and Archi contributed a pair of articles to M.A.R.I. 6 (1990) to provide a historical sketch for each period. New excavations under the direction of Uwe Finkbeiner, in cooperation with a Syrian team led by Shawki Sha’ath and Farouk Ismail, have now demonstrated the existence of both Middle and Early Bronze Age construction at Imar, directly below the Late Bronze town.207 The finds from both periods appear to include houses, but it remains difficult to judge to what extent early Imar was a residential town. It does not fit the type of the Kranzhugel ¨ and has no such circular enclosure. During the period of the main Mari archives, Imar represented a unique site in the Syria-Mesopotamian landscape. It stood between the large kingdoms of Mari and Aleppo, considered subordinate to the latter but far enough from both to maintain considerable independence.208 No royal governor or commander appears to have been stationed there, but there was likewise no local king, so far as we know.209 A number of actions are reported to have been undertaken by the collective initiative of Imar, including reference to the tahtamum assembly or council.210 It seems, then, that the corporate mode of ˘political economy dominated Imar during our Mari period, a time of regional balance among the powers of the modest kingdoms, leaving space for the jostling of numerous smaller polities, associated with tribal identities and not. Imar’s distance from the shadow of the major kingdoms allowed its local political traditions to thrive with relatively little interference from outside powers, and these local traditions were markedly collective. Is it possible to explain Imar’s political character in the early second millennium as a new development, the effect of Amorrite domination? I am not aware of any clear gap between the Early and Middle Bronze levels of occupation at Imar (Meskeneh) that would indicate a new population at the beginning of the third millennium. Perhaps the ongoing excavations will clarify this problem. The Ebla evidence for Imar’s political structures indicates a series of individual leaders each identified as an “en,” like the rulers of Ebla (Archi 1990, 24). It is difficult to know what was the relationship of these leaders to corporate political traditions and to what extent the scribes

214


of Ebla’s ambitious en rulers interpreted Imar through the grid of their own palace-dominated system. The particular designation of Imar’s collective political form as a tahtamum is difficult to explain as Amorrite for two reasons. First, the Amor˘ we meet in the Mari texts are known by tribal names and associations rites that show them to have been scattered across all of ancient Syria and beyond. Their social and political jargon followed them, in terms such as gayum and li mum, merhûm and sug¯agum, all of which were adapted to their diverse circumstances.˘ Even the rihs.um seems to come from this Amorrite terminology, all of which shows fairly˘ wide use and is not yet known from the third millennium. The tahtamum, in contrast, is limited to Imar and Tuttul, two old ˘ same small segment of the Euphrates River. This patneighbors from the tern of use does not fit the character of other second-millennium Amorrite political terminology. The second reason for doubting Amorrite origin is the occurrence of the word at third-millennium Ebla, before Amorrite dominance. So far as “tahtamum” and any other West Semitic words from the Mari archives can be ˘shown to have third-millennium Syrian use, they reflect traditions that cannot be explained by the Amorrite upheavals at the end of the third millennium. If the words also have Amorrite use, then they must be rooted in phenomena and populations from earlier times.

2. Tuttul During the period of the Mari archives, Tuttul is the other town that attests the tahtamum assembly. Tuttul was downstream from Imar at the junction of the ˘Balih River with the Euphrates, and it fell within Mari’s control, though outside˘its core districts. Zimri-Lim stationed a royal representative (hazzannum) named Lanasˆ um there, but the town appears to have had no ˘ local individual ruler during the period of his reign.211 As seen with the evidence for the tahtamum, Tuttul also was characterized by a remarkably ˘ strong tradition of collective political life. The very fact that Imar and Tuttul alone share the tahtamum assembly, so far as the available evidence attests, ˘ underlines their cultural kinship. Sophie Lafont observes that only Imar’s town assembly is explicitly said to meet at its own initiative (2002, 94), but this may reflect only the fact that our Tuttul evidence comes from Lanasˆ um, the Mari representative. Both towns had ready access not only to the east by way of the Euphrates, but also to the north by way of the Euphrates and the Balih. Both towns naturally played a lively role in communication ˘ these would have followed their establishment as imporand trade, but tant settlements. According to Porter, the valley of the middle Euphrates is too narrow to sustain major towns by irrigation agriculture alone on an extensive or intensive basis, and the surrounding steppe is too arid to allow the large-scale dry farming that undergirded the first third-millennium


215

states of the Habur drainage (2000, 14–15).212 Porter suggests that only ˘ a primarily pastoralist economy can explain the first appearance of towns with state-level hierarchy and social complexity during the mid-third millennium. Both Imar and Tuttul fall in this pastoralist zone, along with Mari, downstream. Like Imar, Tuttul is mentioned frequently in the early Ebla archives, as a well-established town on the route between Ebla and Mari (Bonechi 1998, 228). The Ebla texts identify a divine “Lord of Tuttul” (d BE Du-du-luki ).213 Recent excavations at Tell Bia have yielded the remains of solidly built public buildings that have been interpreted as palaces, from the same period as Ebla’s Palace G (mid- to late third millennium) and from the last part of the millennium. These suggest to the excavators a degree of wealth and social hierarchy comparable to those found at other major Syrian sites for this period (Strommenger 1997; Strommenger and Kohlmeyer 2000). The archaeological evidence confirms at the least that Tuttul maintained a strong urban tradition through several stages of the third millennium. It is not yet clear from the excavation reports, however, whether Tuttul can be shown to have housed a significant residential population or whether the walls served mainly to enclose public buildings. It is difficult to evaluate the archaeological and textual evidence for thirdmillennium Tuttul against the grid of exclusionary versus corporate political life. We certainly find wealth and monumental architecture, but it is not yet clear what sort of society they served. As a major sacred site for Dagan, the leading god of the middle Euphrates, one part of Tuttul’s constituency would have been regional, not local. We do not know much in any case about residential patterns associated with Tuttul through the third millennium, although continuing excavations may reveal more. Especially as a primary site for the worship of Dagan, Tuttul was old and influential, though perhaps never as a center of political power. Beyond this there is little certainty. At this point, we are left once again with the tahtamum that Tuttul shared with Imar. At least, the archaeological evidence ˘does not demand a second-millennium and Amorrite explanation for this collective political form. According to the Mari evidence, the tahtamum at Imar and Tuttul was an ˘ with the political voice of the “urban” tradition. It was explicitly identified town. We must not assume, however, that the urban setting for tahtamum ˘ decisions separates it from either tribal social organization or pastoralist subsistence. Imar and Tuttul were situated in an environment entirely different from that of southern Mesopotamia, and their tahtamum must have ˘ depended on a very different social and economic foundation from the Sumerian assemblies. Porter argues that Tell Banat, further upstream in similar ecological environment, displays the evident origin of these urban corporate political ideologies in its burials. Banat’s mortuary monuments are the earliest major constructions at the site. The larger, known as the

216


White Monument, went through two enlargements during periods where large-scale public buildings and tombs were found, thus showing the continuing importance of the mound (2000, 347). The mound contained disarticulated human bones in secondary burial, characteristic of grouporiented ideologies, even as the monumental size shows that the ideology retained its force long after the emergence of an urban elite (pp. 364–6, 387–93). Porter’s analysis, then, suggests an interpretation of our evidence for Imar and Tuttul within the framework of the pastoralist economy of the Euphrates valley through the third millennium. The mortuary monuments along the Euphrates at Banat should reflect the tribal territories and social identities of mobile pastoralists who considered the Euphrates a source of water in the summer and a location for supplementary production of crops (pp. 430–2). These centers all reflect a social system that maintained the strong corporate organization and ideology of pastoralists even as they developed increasingly powerful elites and wealthy urban centers.

3. Urgiˇs Urgiˇs (or Urkesh) has long been known from texts, without ever having been excavated. It was known to be one of the earliest urban centers of the Hurrian presence in Mesopotamia, going back to at least the last two centuries of the third millennium (Wilhelm 1982, 13). Recently, Tell Mozan, near the upper reaches of the Habur drainage, has been identified with certainty ˘ be able to sketch a more broad-based picture of as Urgiˇs, and we will now its past. It turns out that Urgiˇs, like Mari, had a vigorous urban tradition reaching beyond the early third millennium, when the city already had a defensive wall, a central temple, and a size of thirty hectares (Buccellati and Kelly-Buccellati 1999, 14–16). By the middle of the third millennium, the city attained enormous size, as much as 150 hectares, before gradually shrinking through the early second millennium. We cannot be sure how much of that area was built up, however. At least during the late third and the early second millennium, significant residential areas have been excavated, indicating use of the city for an actual urban population (Buccellati 1998, 20–31). Royal seals were found for the mid-late Akkadian period (ca. 2200) and use the Hurrian title “endan” for the ruler, confirming the prior written evidence for Hurrians at Urgiˇs in the late third millennium (Buccellati and KellyBuccellati 1995/6, 6; 1996, 65–8, 81). Buccellati and Kelly-Buccellati argue from the depositional continuity below this stratum that there is no reason to doubt the Hurrian identity of Urgiˇs through the entire third millennium (1997b, 93). Like Tuttul, the town of Urgiˇs had lost size and stature by the time of the Mari correspondence. Also like Tuttul, early second-millennium Urgiˇs appears to have inherited a long urban tradition that went back a thousand


217

years and more, without any sharp material break caused by the arrival of Hurrian speakers, if they were indeed newcomers in the later third millennium. In the case of Urgiˇs, the presence of kings is demonstrated explicitly by writing, and once again we are dealing with a town that has an ancient heritage of considerable size and social complexity. It is harder to identify the origins of the strong collective polity at Urgiˇs, but there is no sudden cultural disruption to which we can attribute a radically new political pattern and ideology. The three large towns that best display active and influential collective governance are all therefore long-established urban centers. With all three, it is difficult to explain how totally new practices would have emerged in the Amorrite period. Rather, the institutions defined by “town” identity are most simply understood to belong to the old political culture of each population.

4. Other Third-Millennium Towns Do any other towns that display noticeable corporate activity in the Mari archives also have pre-Amorrite histories? It is now possible to get some idea of third-millennium Syrian geography from the massive Ebla archives, along with smaller contributions from Beydar, Mari, Nagar (Tell Brak), and Urgiˇs (Tell Mozan), a project recently undertaken by Marco Bonechi (1998, ˇ hna (= Subat-Enlil ˇ cf. 1993). In the Habur region, we find Nagar, Se = ˘ s, among others, while the Euphrates ˘ valley already Tell Leilan), and Urgiˇ attests Mari, Tuttul, and Imar, along with Carchemish upstream and probably Terqa, just upstream from Mari. The moon-god center of Harran is already ˘ found in the upper Balih. ˘ Although Terqa was a key part of the Mari kingdom and the center of one of Zimri-Lim’s administrative districts, we see a particular place for old corporate traditions in the religious sphere. In two texts cited earlier, the elders of Terqa are said to offer prayers to Dagan on behalf of the king, evidently a regular responsibility.214 The political role is perhaps preserved only in the ritual function. Terqa is certainly another Euphrates valley town with roots in the third millennium, regardless of the earliest textual evidence. Wilfrid Lambert has established the reading of BAN for/ter/ in Mari’s oldest god list, from the late Early Dynastic period, which mentions a divine “Lord of Terqa” (d LUGAL Ter5 -qa), and Lambert defends at length the reading of the place name in such an early text.215 Based on this breakthrough, Marti Luciani then identifies the name in texts from both Ab¯u-S. al¯ab¯ıh and Ebla, where BAN-ga ki should be read as Ter5 -gaki , and many more˘ references come from the late third millennium (1999, 1–3). The site of Terqa (Tell Ashara) has been excavated fairly recently, and the finds confirm its occupation back through the first half of the millennium.216 Rouault finds that the sheer extent and size of the third-millennium enclosing wall lead one to

218


expect a major center with a hierarchical society of the sort found in the great southern Mesopotamian cities (2000, 266). This conclusion seems somewhat premature. Such a wall must indeed show a complex society, but the original use of the site remains uncertain. A prominent sacred role for the regional god Dagan is likely, but Terqa’s role as political or residential center is still unknown. In the case of this town, proximity to Mari during the time of the powerful kingdoms of the early second millennium seems to have prevented the retention of any political independence, individual or corporate. Only after the demise of Zimri-Lim’s Mari kingdom and the slight retreat of Babylonian power was it possible for a new “Hana” kingdom to be established with a ˘ capital at Terqa (Buccellati 1988; Podany).

5. The Ancient Syrian Tradition of Collective Decision Making The collective political forms of early second-millennium Syria-Mesopotamia are not the product of a single social setting and mode of subsistence. Whatever the extent of their power in earliest southern Mesopotamia, the assemblies and elders examined by Jacobsen and others are rooted in settled towns supported by irrigation agriculture. Joy McCorriston asserts that “kin-based access to communally held land lay at the heart of Mesopotamian society, but as temples acquired land, the terms if not the terminology of access to land were radically transformed” (McCorriston 1997, 529–30). The Mari archives reflect a completely different range of geographical settings and ecological zones, however. Especially for the period of ZimriLim, the archives reveal an unprecedented picture of ancient tribal peoples, whose political life is marked by prominent collective forms and ideology. These tribal peoples display a particular reliance on long-range herding of sheep and goats for subsistence, and their collective traditions can be explained in part by the tendency of pastoralists to manage access to grazing land at the collective level, for whole communities.217 Even if the political customs of Imar, Tuttul, and Urgiˇs come from before the new dominance of Amorrites at the end of the third millennium, these town traditions are embedded in a world of tribal social organization. Must such organization be a new thing, arriving with the Amorrites? Perhaps not. It is possible that the tahtamum of Imar and Tuttul, in particular, ˘ derives from a situation in the middle Euphrates region in which settled centers were common components of a social organization dominated by tribes. Third-millennium Syria remains very much a mystery, as more and more evidence for fortified centers and monumental construction comes to light. We can assume neither that the collective town custom of the Mari archives is Amorrite nor that all tribal organization appeared only with the Amorrites.

Mari in Third-Millennium Mesopotamia

219

h. mari in third-millennium mesopotamia Somehow, the Mari of our second-millennium archives must be placed in a coherent relation to Mari of the third millennium. The fall of Ur, the rise of Isin, and the new appearance of Amorrites in southern Mesopotamia at the start of the second millennium has long provided an interpretive model governed by the idea of cultural disruption. The prominence of tribal peoples in the Mari archives could be understood to confirm the presence of a very new social order. In fact, however, the degree and kind of social change between the late third and early second millennia in the middle Euphrates valley are difficult to judge, and there is evidence for continuity as well as for novelty. I cannot resolve this problem, but in this section I explore the evidence and caution against assuming that the political world of the Mari archives is entirely unlike what had characterized the previous millennium.

1. Third-Millennium Urbanization in Syria Over the last generation, various excavations in Syria and northern Iraq have demonstrated that large-scale urbanism was not confined to southern Mesopotamia during the third millennium. From Ebla in the west to Tell ˇ hna/Subat-Enlil) ˇ Leilan (Se further east in the Habur region, archaeologists ˘ have discovered an explosion of monumental˘ construction and large city sites just after ca. 2500 b.c.e.218 Weiss (1988) has suggested that across the Jezira of northern Syria, this growth can be explained primarily by the surpluses of dry-farming agriculture.219 Although this period certainly represents a new phase of urbanization in northern and western Mesopotamia, many of the sites in question had already existed in preceding periods and simply went through a growth spurt. This is the case for both Tell Leilan and Tell Brak in the Habur ˘ to basin, for example.220 The changes of the mid-third millennium are not be explained by a sudden shift in economic patterns and seem rather to reflect some degree of smooth transition from what came before.221 Are these patterns based entirely on agriculture? There is some possibility that the role of pastoralism has been underestimated. During the early stages of Ebla research, I. J. Gelb once commented that we should not assume an agricultural foundation for all early urban civilizations and that Ebla and northern Syria may have relied more heavily on surpluses of wool than of grain (Gelb 1986). Given the prominence of pastoralism in the Mari archives of the early second millennium, it makes sense to look for possible expressions of a herding economy in archaeological evidence from the third millennium. One example may be the emergence of a new type of circular settlement in early thirdmillennium Syria, somehow coordinated with the burgeoning urbanism of

220


the succeeding period. These were given the name Kranzhugeln, ¨ or crownshaped hills, by Max von Oppenheim (Kouchoukos 1998, 339) and were later documented in surveys by W. van Liere and J. Lauffray (1954–5; van Liere, 1963). Large Kranzhugel ¨ sites include Tell Beydar in the Habur, Tell ˘ perhaps Chuera to the west, between the Habur and the Balih Rivers, and ˘ ˘ even Mari, whose outer mound forms a neat arc for the section that is preserved. The circular enclosures of both Chuera and Mari appear to have been built in the early third millennium, before the boom years of the Ebla archives in the twenty-fifth to twenty-fourth centuries.222 Similar sites from the same early period are found as far south as Jordan, as at Jawa and Tell Um Hammad (Helms 1987, 49).

2. Pastoralism and Third-Millenium Urbanization Bertille Lyonnet observes that all of the Kranzhügel sites are located in a semi-arid zone with too little rain for dry farming, even though they enclose very large areas (Lyonnet 1998, 180). These circular settlements could have fortified central mounds, but it is not clear that the outer ring had a defensive military purpose. Lyonnet regards the most plausible explanation to be that these served pastoralist communities who could bring into them their dependents and their flocks (p. 182). Especially in the context of large communities, it should not be necessary to treat such pastoralism as wholly separate from farming. This means that a pastoralist interpretation of the enclosures would not exclude Thomas McClellan’s proposal that the depressions surrounding the outside of these circular mounds may have been used to store runoff from rain.223 Lyonnet considers Tuttul (Tell Bia) to belong to this group of sites, as well, so that both Mari and Tuttul may have been founded at least in part to serve pastoralist needs (1998, 180n5). Another pastoralist approach to third-millennium urbanism in Syria has been proposed by Anne Porter, in response to the finds from the excavations by McClellan and herself at Tell Banat, fifty kilometers south of Carchemish in the upper part of the middle Euphrates valley. According to Porter, the massive above-ground monumental burials in the Syrian Euphrates valley reflect a society of pastoralists, marking territory by large secondary burials associated with each group’s ancestors (2000, 4–5, 387–93, 430–7). Banat, along with other third-millennium urban centers such as Mari, Tuttul, and Carchemish, is situated in the narrow part of the Euphrates valley within a similar ecological environment. All of these locations lack adequate land to provide an agricultural economic base by either dry farming or irrigation on a scale necessary to support the size and complexity of such sites (pp. 14– 15). The settlement at Banat (only twenty-five hectares) never grew to the scale of these other centers, but both the monumental architecture in palace style and the associated elaborate tombs indicate the presence of an elite that would normally be understood as evidence for “state” polity (pp. 347–8, 360–1, 370–6). These “state” features are found in the period between about

Mari in Third-Millennium Mesopotamia

221

2450 and 2300, roughly contemporary with the early Ebla state (pp. 287– 8, 366). The giant “White Monument” burial mound is even older, with a more modest early phase that goes back before 2600 (pp. 327–45; McClellan 1998). The key problem is how to explain such concentrations of power and wealth in settled centers in a region that shows no sign of anything but a “supplemented pastoralist” economy.224 Porter maintains that large-scale settlements could be generated from a pastoralist economic social system, without a subsequent “wholesale sedentarization and a switch to fixed cultivation, but that the socio-economic systems of supplemented pastoralism were maintained and expanded” (p. 224). The monumental burials show how this could have been achieved. They are often located where the pastoralist steppe and the river valley meet. The Euphrates should be considered a source of water in the summer and a location for supplementary crop production, not the focus of a narrow zone of contained cultural identity (pp. 431–2). Mobile pastoralists need an “elastic system of social integration that transcends dispersal,” and this becomes the function of kinship, fictive or real, and marked by collective burial sites that are tied to territorial resource rights (pp. 435–6). Margueron has based his interpretation of Mari’s foundation on the presence of canals in the valley, which would have carried boats with wood and other commodities up and down the Euphrates (see Chapter 1). He himself concludes that agriculture cannot explain the size and influence of Mari in the middle of the third millennium. The question is whether there was sufficient trade to support such a city. Also, I am not aware of evidence that proves the use of the Euphrates canals for this purpose at the time when Mari first appeared. Porter suggests that pastoralism may offer a crucial economic factor. Mari stood at the pivot of two “pastoral cycles and territories,” linked to the Euphrates and to the Habur (p. 452). Against the theory of Buccel˘ no reason to consider that pastoralism first lati (1990, 98–102), there is arose in the backcountry of the Syrian Euphrates with the advent of the Amorrites. Rather, there is a long continuity between the economies of thirdand second-millennium Mari, always incorporating a strong relationship between valley cultivation and pastoralism in the hinterlands (p. 453).225 Within this framework, the prominent role of pastoralism in the archives of early second-millennium Mari would represent not an Amorrite novelty but one Amorrite expression of a much older and remarkably persistent pattern.

3. Yahdun-Lim and Southern Mesopotamia ˘ By the time of Yahdun-Lim, Yasmah-Addu, and Zimri-Lim, Mari already had a long history˘ as a central site ˘in the Euphrates valley. Real changes appear to have taken place with the rising Amorrite influence across eastern

222


Mesopotamia, but at Mari, these must be balanced against the continuities suggested by the very reuse of an ancient site. Yaggid-Lim and Yahdun˘ Lim established a new dynasty at Mari, whether or not those who preceded them there could have been called “Amorrite,” but they chose a site with at least the physical structures of palace-based administration already in place, ˇ rulers of the late third millennium. inherited from the Sakkanakku We cannot be sure whether these structures were in active use before the new arrivals, but Yahdun-Lim in any case made himself the heir of Mari’s ˘ Whatever the exact economic configuration of trade, age-old regional role. agriculture, and pastoralism that allowed Mari to succeed, the site had long served as the gateway to the middle and upper Euphrates from southern Mesopotamia. It was not the only site near the confluence of the Habur and ˘ one had the Euphrates, but it became the first major center upstream, once left the region of Agade and Babylon, or Sippar and Eˇsnunna. Yahdun-Lim ˘ a bridge seems to have taken very seriously the prestige of Mari’s situation as ˇ to Mesopotamia downstream. His Samaˇs temple inscription celebrates his northwestern conquests at the same time as it sets a very southeastern tone, saluting Sumer’s Enlil as the head of a very downstream Mesopotamian pantheon. Dagan, the leader of the gods in old middle Euphrates tradition, is ignored entirely. In reestablishing Mari as a major administrative center, Yahdun-Lim updated scribal practices, perhaps to match the current norms of ˘Eˇsnunna (Durand 1985a, 170–2). Without solid evidence for the preceding period, it is hard to judge the effect of Yahdun-Lim’s regime,226 but he gives the impression of renewing a neglected˘ tradition that is defined by the old city, rather than of introducing the foreign ruling culture of Amorrite tribespeople. It is not clear why the basic relationships at Mari between town and countryside, fields and flocks, should be regarded as essentially different under Yahdun-Lim, Yasmah-Addu, and Zimri-Lim. In the early second millennium, ˘the urban center˘of Mari seems to have been less an engine of social change, extending its power outward from the settled center, than a target for other people’s regional ambitions. We should not dismiss the possibility that at the deeper levels, the more things changed, the more they stayed the same.

i. on explaining corporate power There is little question that the political landscape of the Mari archives was dominated by kings and kingdoms. If individual and collective political forms have always coexisted and competed, the individuals seem to have come out on top in most of early second-millennium Mesopotamia. At the same time, however, the more we probe the written evidence from Mari, the more pervasive and influential the collective forms seem to be. They are not relegated to the small scale or the peripheral, as might be said of country villages or nomadic clans. Their responsibilities are not limited to local affairs

On Explaining Corporate Power

223

such as legal judgments, and if anything, they appear most prominently in decisions of war and peace, as perhaps suits the interests of this diplomatic correspondence. And yet the collective mode of operation functions in constant tension with individual authority. The evidence from Mari resists constriction into an either-or duality, where either we acknowledge true power to lie only with the individual ruler or we demand that significant collective power can exist only where the individual has been rendered impotent. Ultimately, this study requires some sense of how political power could be expressed collectively in the ancient world before democracy. To this end, I explore here some broader questions related to understanding the exercise of power in early second-millennium Syria and Mesopotamia. First, the interplay of tribes and towns entangles us in issues of early urbanism and the emergence of the state, issues that are often explained within evolutionary frameworks that I find to impede open response to the actual evidence. Then I return one more time to the work of Richard Blanton and his colleagues, who describe a “corporate cognitive code” that operates concurrently with individualizing ideologies of power.

1. On Origins In general, collective forms of decision making are prominent in small communities. This is true of groups across the spectrum of mobile- and sedentary-based subsistence strategies in premodern settings, and is generally true for small modern groups as well, as in the council study of F. G. Bailey (1965). One influential framework for understanding the historical role of collective decision making in small communities has been to view this as essentially primitive. According to the widely quoted hierarchy of Elman Service, societies that are characterized by such organization, with weaker expressions of centralizing power, may be called “band” and “tribal” societies, as opposed to the increasingly complex “chiefdoms” and “states.”227 Service and others have regarded the relationship between these types as evolutionary, with certain conditions in the simpler societies leading inherently to the more complex types. Leaving democracy aside as a separate development, Service treated the more complex political systems of chiefdoms and states as overwhelmingly centered on individual leadership. Indeed, as human population has increased, new forms have appeared. There was a time before monumental building, before fortification, before multitiered administrative hierarchies, not to mention a trail of technologies, including writing. The question is whether historical change can be ordered into an evolutionary system that allows certain forms to be understood as “primitive,” leading inexorably to more advanced stages. According to an evolutionary explanation, the more markedly collective political life of smaller communities should give way to the centralizing power of

224


chiefdoms and states, especially in settings untouched by modern political developments. In practical terms, the evolutionary theorists have had little to say about corporate polities in places such as early Syria-Mesopotamia, because they do not fit the system. I find that by their antievolutionary approaches, Anthony Giddens and Michael Mann embrace without hesitation the diversity of actual societies and offer more powerful frameworks for understanding my specific problem of the surviving corporate polities in states that are generally ruled by kings. Giddens assaults the evolutionary idea that societies change by “adaptation” to the material conditions of the environment, with the objection that historical success does not explain change. In fact, the surviving simple societies never ended up changing (“evolving”), because nothing in their environments required it (1981, 20–4). Mann objects to the formulation of societal evolution as “a single immanent process.” He acknowledges change that enhances “human capacities,” but only as episodes that must be evaluated individually. “Societies are much messier than our theories of them” (1986, 3–4). A certain number of anthropologists wrestle with the vagaries of change in particular societies and press for more nuanced models, with or without the evolutionary tag. Philip Carl Salzman begins his 1980 study of nomadic sedentarization with a warning against the assumption that change is “irreversible, directional, and cumulative.” Society should rather be approached as “fluid, variable, flexible and adaptable.” Sociocultural change “is only in some instances non-repetitive, directional and cumulative, and is often rather in the form of repetitive cycles of alternating phases” (pp. 1, 4, 7). Blanton, Feinman, and two more colleagues call for a more sophisticated account of social evolution in the abstract for their thesis of alternative “exclusionary” and “corporate” political strategies: Current neo-evolutionary theory is inadequate to the analysis of past social change because it lacks a suitable behavioral theory and because its simple stage typology fails to account for variation among societies of similar complexity and scale. (1996, 1)

It appears that the African evidence may also be taken to offer little encouragement for evolutionary explanations.228 There is no question that the first state formation in the ancient Near East represented a genuine enhancement of human capacities, something new and powerful. This change became one element in a cascade of social developments that produced a plethora of corollary phenomena. I want to understand the corporate political life of the Mari period towns in the light of this past and of these changes without forcing it into a simplistic scheme.

2. Collective Power If we are to understand the reality of the collective political power evidenced in the Mari archives, we need a more general model of power that is capable


225

of explaining patterns more nuanced and perhaps more slippery than those of centralized, institutionally based systems with only one mode of operation. Power is commonly defined as something like the “ability, on a continuing basis, to compel compliance with one’s wishes” (Donlan 1997, 40)229 . Such definitions usually distinguish power from authority, more a matter of permission, which also tends to be defined in individual terms. Service thought along similar lines, though he extended the boundaries to acknowledge explicitly the potential power of groups: Power is “the relative ability of a person or group to cause another person or group to obey, or conversely, the ability to ‘not have to give in’” (1975, 11). The interplay of political forces in the towns and kingdoms of the Mari archives requires a more adaptable analysis of power than this focus on compelling action. In fact, the field of political anthropology often simply ignores collective political action and the complications it introduces into the analysis of individual societies. As one example, Georges Balandier’s general text makes no reference whatever to collective political forms (1970). The evolutionary systems of Fried and Service likewise neglect the role of the group, even when they discuss the simple societies where individual leadership is so deeply integrated into corporate action. Michael Mann begins his history of social power with the statement, “Societies are constituted of multiple overlapping and intersecting sociospatial networks of power,” and he attributes the ultimate sources for this social power to ideological, economic, military, and political relationships (1986, 1–2).230 Following the lead of Talcott Parsons, Mann explores the “collective” aspect of power, “whereby persons in cooperation can enhance their joint power over third parties or over nature.” This collective power operates simultaneously with and is intertwined with the “distributive” power of actor A over actor B. “In pursuit of their goals, humans enter into cooperative, collective power relations with one another. But in implementing collective goals, social organization and a division of labor are set up” (pp. 6–7). Mann also distinguishes what he calls “diffused” from the universally recognized “authoritative” power. Whereas authoritative power comprises definite commands and conscious obedience, diffused power “spreads in a spontaneous, unconscious, decentered way through a population, resulting in similar social practices that embody power relations but are not explicitly commanded” (p. 8). This represents just one recent formulation, but it is clear that a complete analysis of social power in all its manifestations requires recognition of the group, both as a potential player in structured governance and as a more nebulous collective force, even in being governed.

3. The Corporate Cognitive Code More recently still, Richard Blanton and his colleagues have proposed an approach to political power that acknowledges explicitly the collective aspect in premodern complex societies, not simply as it functions under individual

226


rule but as an active mode of governance. According to this group, political action involves the interplay of two coexisting power strategies, “exclusionary” and “corporate.” Exclusionary power is built around the monopoly control of its sources, while corporate power “is shared across different groups and sectors of society in such a way as to inhibit exclusionary strategies” (1996, 2).231 Both political strategies are generally present, but either one may dominate. If they are correct, a corporate strategy may dominate the exclusionary strategy in individual political systems. Under the corporate type, “[m]onopoly control of sources of power is precluded by restrictions on the political behavior of those vested with power.” The distribution of power is structured, determined, legitimated, and controlled within a “prevailing corporate cognitive code.” This last notion of a “corporate cognitive code” reminds us that when we work from the Mari archives, we are dealing not just with words but with letters composed from various vantages on a political relationship with the kings of Mari. Regardless of the realities of actual power, which are indeed worth understanding, it is intriguing that the texts themselves characterize certain exercises of power, certain actions or decisions, as proceeding in a corporate mode. The texts thus reflect an ideology of collective political power, where the portrayal of power as collective is somehow desirable, whether consciously or automatically, in certain situations. The use of collective terminologies communicates a particular interpretation of political action. In the study of the Near East, we are more used to recognizing the ideologies of individual power, but in fact, these do not represent the full array of ancient political thinking.

4. Ideology In an article on power strategies, DeMarrais, Castillo, and Earle (1996) examine ideology as an alternative to coercion. Strategic control of ideology “contributes to the centralization and consolidation of political power” (p. 16). The process by which certain ideologies gain primacy involves physical expressions that are given to them, such as ceremonies, symbolic objects, monuments, and writing systems. This “materialization” of ideology confers social power by the ability to pay for the necessary resources and by manipulation of the meaning expressed (p. 17). DeMarrais and her colleagues do not pursue the ways in which materialization could serve collective political power, though festivals are an obvious candidate for such a role. The public rituals from thirteenth-century Emar offer an interesting case in point. Emar’s ritual archive preserves two versions of an event called the zukru, an acknowledgment of Dagan, the region’s chief god. The king of Emar plays a role only in the more elaborate account of rites to be performed every seven years, and then he appears only as benefactor, subsidizing a large percentage of the cost. This consitutes a clear example of the


227

elite promoting its social and political status by using the power it already possesses as leverage to gain further ground. At the same time, however, the actual ideology represented in the zukru festival is decidedly collective in nature, bringing the citizenry as such front and center and excluding any active role for either the king or any elite individual or group.232 This town festival represents one piece of the larger evidence for a strong collective political ideology in thirteenth-century Emar, coexisting with an emergent local monarchy under Hittite imperial sovereignty (Fleming 1992a). The earlier texts found at Mari show that this corporate political tradition was in fact centuries older than those ritual texts and confirm the impression that centralizing royal power was only newly ascendant at Emar.

5. The Ancient Near East It is impossible to deny the existence of corporate forms such as town meetings or groups of elders in the ancient Near East, and these have received occasional attention in the years since Jacobsen’s early studies (see above). It is more difficult, however, to judge the political power of the corporate mode of leadership relative to the power of the individual leaders, the kings. Recently, Marc Van De Mieroop has suggested that “the powers and independence of the citizenry increased over time in Mesopotamian history, rather than the other way around,” as Jacobsen had supposed (1999, 118). He sees the kidinnu grants of freedom from taxation to various cities in firstmillennium Mesopotamia as marking a substantial increase in the collective power of towns, though only in multicity states, where the towns in question are not the royal capital (pp. 135–8). Aside from the issue of trends, which may have developed in more than one direction, this analysis does not explain the basis for whatever actual power could be wielded by the group in relation to kings or even the possibility of governance without any individual ruler. David Schloen offers what he calls a “patrimonial household model” for all ancient Near Eastern politics and society, so that “the same principles of social order may be found among social groups at all points along the continuum from political centralization to decentralization” (2001, 63; cf. Fleming 2002). All Bronze Age societies, he says, share the same ultimate framework, with the king as the ultimate paterfamilias: In a patrimonial regime the ruler stands at the apex of an integrated socioeconomic system that encompasses all of the land and personnel in his domain. Everyone is ultimately a member of the ruler’s household – there is no “free” sector of independent proprietors who enjoy a separate conceptual and legal status from that of “nonfree” palace dependents. (p. 65)

I find Schloen’s larger model of a patrimonial grid for all social order in the ancient Near East to be quite attractive. Again, however, he has no

228


mechanism for understanding collective political power, though the patrimonial framework is perfectly capable of accommodating it. Schloen follows Max Weber’s assumption that all government in this patrimonial system ultimately falls to a single ruler, the “father” at the apex of a family pyramid. Viewed historically, however, the ideology of collective decision making is based on a social geometry that is different from the pyramid of the royal patriarch. Even the village need not see itself solely as a composition of subordinate family members beneath a single father. The idea of the collective decision depends instead on the idea that a community may be made up of households or family clusters that see themselves as peers, not necessarily gathered under one father’s authority. In many settings the individual and the collective ideologies would have coexisted, and we should not understand political power in these circumstances to require complete dominance of one or the other. In a generally patrimonial environment, chiefs and kings could have exploited the metaphor to their advantage by emphasizing the head of household over the coexisting tradition of gathered heads requiring a corporate style of decision making. At the same time, however, the widespread coexistence of individual and collective political traditions, the “exclusionary” and “corporate” strategies of Blanton and company, suggests that ancient Near Eastern kingship may have involved more consensus building than is sometimes supposed. Behind the exaltation of royal power in a text such as the prologue to the famous Code of Hammurabi stands a careful attempt to ˘ cast the Babylonian king as the caretaker of each major city and its deities, including Mari, which he had just taken from Zimri-Lim. Hammurabi is ˘ guardian to not so much a conqueror to be feared as a divinely appointed be welcomed. The text itself represents an effort to build consensus, and we can imagine that the Babylonian king increased his power by a shrewd combination of military force and constant negotiation. Whether at Babylon or at Mari, the actual exercise of kingship requires further study to gain a better understanding of the interplay between individual royal authority and various other loci of political power, individual and collective. Such work extends beyond the range of what I have been able to undertake in the current project, but further reevaluation is needed. It has been observed that early kingship in the parts of Mesopotamia upstream from Sumer was more autocratic than in the south (Steinkeller 1993). This tradition of a strong individual leadership that was centered in king and palace may itself have been adapted to regional traditions of counterbalancing power, including the collective forms of towns and tribes.

5 Conclusions

From the beginning of its preparation, this book has been built on the notion of studying the large from the small, the nature of ancient society from the particular vantage of the Mari archives. This method does not mean that Mari represents only a secondary interest. On the contrary, I have little to say about ancient political life generally except insofar as the Mari evidence has provoked it. Moreover, I will have failed if this volume offers nothing new to our understanding of the Mari archives and the phenomena found in them. The detailed investigation of a single coherent body of historical evidence demands a responsiveness to the actual that will direct theoretical questions and control theoretical speculation. My conclusions naturally follow the separate levels of my method, and I present them separately, as results relating to Mari’s political world, and then as relating to the ancient political world before democracy.

a. the political world of the mari archives Jean-Marie Durand’s team of Assyriologists who have been involved with publication and interpretation of the Mari archives over the last twenty-plus years have made tremendous progress toward understanding their contents and their implications. After their work, all of the classic works on tribes in the society revealed in the Mari evidence are rendered to some extent out of date: Kupper, Luke, Rowton, Matthews, Heltzer, and Anbar.1 I accept many of the new proposals made by Durand and his colleagues, and it could be said that I have accepted too much. Further research will tell, and perhaps this volume will help attract others to the task. Against the backdrop of the debt that I owe to the French team and their ideas, this project has developed several new conclusions regarding the material from Mari itself. r The two “hands” of the dominant Simalite and Yaminite tribal confederacies were characterized by significantly different social and political structures. 229

230

Conclusions

r Zimri-Lim was an active tribal king in a way that was common to the time and region but that distinguishes him from rulers such as Samsi-Addu of Ekallatum and Hammurabi of Babylon. ˘ atum reflects a non-southern-Mesopotamian developr The Akkadian m¯ ment of larger polities, in which the m¯atum is linked to kings but is not ultimately a “kingdom.” r I have found extensive evidence for collective decision making in the Mari archives, especially as linked to identity as a “town” (¯alum). There is only limited evidence for full assemblies of towns, tribal groups, or armies for political action, but it is clear that such assembly did occur in certain situations. r The three towns in the Mari archives with the strongest collective political forms – Imar, Tuttul, and Urgiˇs (Urkesh) – were all significant thirdmillennium centers, and their collective political traditions are probably old.

1. The Simalites and the Yaminites Kings Yahdun-Lim and Zimri-Lim of Mari were both from the Binu Simal ˘ tribal confederacy, and the Mari archives therefore reflect deep associations with Simalite peoples. The Simalites were cousins and rivals to the Yaminites, as “left hand” to “right hand,” and the interests of these two broad tribal groups suffuse the Mari evidence. Durand has interpreted the social and political structures of these peoples as essentially homogeneous, but I have proposed that the Simalites and the Yaminites had quite different social organizations. Under Zimri-Lim, as previously established with the victories of Yahdun-Lim, the Simalites were ruled by a sin˘ had five kings, one for each constituent gle king, whereas the Yaminites tribe. Individual Simalite pastoralists were identified by gayum “division,” a category that had lost all political significance by the time of Zimri-Lim. Within each gayum, smaller groups were identified by their sug¯agum “leader.” We know that many Simalites lived in towns, but Simalite tribal identities were defined only in pastoralist terms, as hana, probably “(our) tent˘ dwellers.” The Yaminites have long been known to have organized themselves into u, the Yarihû, the Yahrurˆ u, the Amnanû, and five named tribes: the Uprapˆ the Rabbˆ u. These tribal names embrace ˘both town˘and steppe populations, and local Yaminite sug¯agums can be identified by their leadership of individual towns, unlike the Simalite sug¯agums. Both the Simalite and the Yaminite pastoralists of the steppe recognized the authority of merhûm “chiefs of pas˘ have related to ture,” though it is not clear how the Yaminite merhûms might ˘ the authority of the tribal kings. Under Zimri-Lim, two merhûms gained enor˘ mous political power, more than anything visible in the available evidence for the Yaminites. The Simalite tribal division called the gayum is never

The Political World of the Mari Archives

231

applied to Yaminite groups, whose five tribes may have been identified as the limums of each ruler. Yaminite pastoralist communities were called the hibrum, which may have had its own local leadership, as suggested by its ˘ association in one text with the word “kadûm.” These proposed distinctions between Simalite and Yaminite tribal societies shape how we approach all of Syrian society in the Mari period. Tribal organization was fully integrated into the political world of kingdoms and cities, or perhaps we must even say that kingship and urbanism were fully adapted to a tribal form of social organization. Tribal society was likewise adapted to a wide range of geographical and ecological zones, with various resulting structures. Certain specific expressions of this diversity are visible in the Mari evidence, and these should lead us to expect more. The conception of a great duality of Yaminite and Simalite tribal confederacies may not have been that old, but the deeper patterns of tribal organization may go back much further.

2. The Tribal King Zimri-Lim took up the title claimed by the earlier Yahdun-Lim, as “the king ˘ and political entity of Mari and the m¯at Hana.” As kings of a population ˘ identified as “Hana,” Zimri-Lim and Yahdun-Lim considered themselves to be tribal rulers˘ of Simalites who took˘ their name from their pastoralist community. The Mari archives show us many such tribal kings who ruled from a fortified administrative and ritual center, such as the three Yaminite ˇ kings defeated by Yahdun-Lim according to his Samaˇ s temple inscription. ˘ and underestimated privilege to have access to the Nevertheless, it is a rare cuneiform archives of a tribal king, as with Zimri-Lim’s dominant fraction of the Mari archives. Whereas the documents from palace administration offer less opportunity to see the tribal aspects of Zimri-Lim’s realm, the letters reveal much that is usually inaccessible to modern eyes. Beyond his official titulary, various other evidence shows Zimri-Lim to have been considered a Simalite king, rather than a king defined by rule of Mari. This dominant tribal character contrasts with two other major figures from the period who are known to keep memories of tribal origins. Neither Samsi-Addu of Ekallatum and northern Mesopotamia nor Hammurabi of ˘ Babylon identified himself as the king of a tribal people. By emphasizing this contrast between Zimri-Lim and the others, I do not mean to exclude the possibility of significant tribal aspects to their political and social situations. Rather, I propose that with the texts from Zimri-Lim’s reign, we confront the undeniable integration of tribal identity into a large city-based kingdom, so that tribes, on the one hand, and their mobile pastoralist herdsmen, on the other, cannot be relegated to the periphery of ancient Mesopotamian politics or society. The tribes of the Mari archives are neither outside nor other to the regime that produced these texts. From the secure phenomenon in

232

Conclusions

the archives of Zimri-Lim, we can then expand the investigation to adjacent settings in place or time.

3. The m¯atum The world of cuneiform writing in the time of the Mari archives was divided at the highest level into political units called m¯atums, most simply (though too vaguely) translated as “lands.” Unlike the West Semitic word for “dominions” (namlak¯atum), which was defined by the “rule” of the individual leader, the Akkadian m¯atum is not essentially a “kingdom.”2 Although the word “m¯atum” could refer simply to a king’s domain or the political home of a traveler abroad, more precise use separated the m¯atum from the central city of its ruler. The m¯atum was ruled as a population distinct from the king and his fortified administrative center. Confirmation of the notion that the m¯atum is not a “kingdom” comes from the three coalition m¯atums of the Mari archives. Ida-Maras. of the upper Habur River basin, Zalmaqum of the upper Balih ˘ east of the Tigris were all considered individual˘ ˇ River basin, and Subartum m¯atums that functioned as the collective political vehicle for groups of local “kings” whose particular realms were sometimes extremely modest. As the population ruled by a king and distinct from him, the m¯atum had to be treated by a king as an entity separate from his own government, a mass that could either revolt or serve contentedly. In contrast to the “people” or “dependents” (niˇsu¯ ) of a king, the m¯atum was a political category, a group that could speak or act as a whole, whether under the leadership of an individual ruler or not. Is it possible, then, that the basic idea of the m¯atum was collective rather than “exclusionary,” somehow a people making decisions together, like the coalitions? One wonders even whether the term might not originally have come from tribal political organization. Many of the “lands” of the Mari evidence seem to be defined by tribal names, joined to a central fortified capital.

4. The Collective Polity and Full Assembly Sixty years after Jacobsen’s first publication on “primitive democracy,” it would be fascinating to see a full collection of all of the cuneiform evidence for collective political activity. The textual base has increased enormously since then, with a much wider geographical range. With this study, I have attempted only a fraction of that task, and Mari alone provides an immense and varied resource that has not previously been examined systematically. I have found that collective political activity and terminology are associated with the three coalition m¯atums, with tribal groups, and especially with towns. Mari offers a particularly rich documentation for the town in action as a collective unit, often referred to by the plurality of its citizens, as “the Tuttulites,” “the Imarites,” and so on. The terminology surrounding this

The Political World of the Mari Archives

233

phenomenon is varied and somewhat fluid, as in the case of “elders.” The extent of public participation in actual decision-making meetings is usually difficult to judge, and the terminology itself does not automatically indicate the scale. We certainly do not find two clear categories in the form of a smaller “council of elders” and a more inclusive “assembly of free men.” In actual use throughout the Mari evidence, the word for “town” has both a physical and a political aspect. A town or a¯ lum was indeed a cluster of permanent structures, but these were simply the concrete expression of what a clustered population had created for its joint use. The actual nature of ancient settlements from the third and early second millennia is problematic. There is no question that many people lived in villages and towns, though relatively few residential districts have been excavated for the larger tells in Syria and northern Iraq. At least in some cases, the fortified center that served as a royal seat seems to have served a more administrative and symbolic ritual purpose, with little population beyond the personnel of palace and temples. The textual evidence nevertheless maintains an essential link between towns and their people, especially as they act together in matters of war and peace. It is possible that all mention of towns in collective action assume an actual population in residence and that forms such as the tahtamum derived only from such residential communities. Alternatively, the ˘idea of the town as a collective political expression of a local population could have been defined by the people that used the site rather than by occupation of houses there. If a “town” physically consisted of a defensive enclosure with sacred and administrative buildings, with a fairly small full-time population, the political “town” could still have included all who would gather there for festivals or markets or in time of war. Most of the references to collective decision making in towns or tribal groups suggest some level of limited representation rather than full assembly of the population, or at least of free men. A few texts, however, make clear that a much more inclusive gathering was in view. In some cases, it seems that an entire town was called together for proclamation of a royal decree. Here, the people are identified as residents (“sons of” the town) rather than as decision makers.3 A similar situation is envisioned in a letter that describes how the governor of Qat.t.unan has a message from the king of Kurdâ read aloud in the presence of several named “district elders” and 100 “town elders.”4 Two more letters describe large gatherings of town leaders under different terminology, both groups taking active responsibility for their town’s affairs and both times called “the heads” of their town. Once, 200 “heads of Qâ-and-Isqâ” assembled in response to a contact from a high Mari official, who wanted to know what support the town would offer Zimri-Lim. It is this collected group that declares, “We are the servants of Zimri-Lim,” speaking for Isqâ-and-Qâ as such.5 Another letter from an official of Zimri-Lim reports that eighty “Imarite heads of the land (m¯atum)” have been sent to

234

Conclusions

undergo the river ordeal on behalf of their town. This is surely something less than the whole population, but it is a remarkably large number, especially if they travel downstream past Mari to the ordeal site of Hˆıt.6 Although the ˘ ordeal had a legal function, the very representation of a whole town makes this group politically responsible for the people of Imar. It appears that our clearest cases of inclusive town assembly with real political responsibility use the word “heads” to define the people involved. The only text that suggests a full “popular” assembly relates to Isqâ-and-Qâ, where the “heads” may best be understood as heads of households. One reference to the rihs.um “talks” displays what may be the assembly ˘ of an entire tribal population. Zimri-Lim’s royal representative at Tuttul recounts the arrival of fifty Yaminites of the Uprapˆ u tribe. The Mari official accompanies the group to the steppe (nawûm), where he engages in talks.7 It appears that the fifty tribesmen return to the steppe so that they can include those who had been left behind. One account of these tribal rihs.um “talks” ˘ anticipates a serious give-and-take, during which one of the Simalite chiefs of pasture (merhûm) may come under criticism, in spite of his rank.8 This ˘ kind of open debate is also mentioned in connection with the full “assembly” (puhrum) of an army (s.a¯ bum) of Numhâ tribesmen at the town of Qat.t.arâ.9 ˘ ˘ Together, these texts do not show widespread resort to full assembly. It does seem, however, that both the pastoralists of the steppe and the residents of towns did sometimes gather, not only to receive word from an outside king but even to speak for the group.

5. Imar, Tuttul, and Urgiˇs The three towns in the Mari archives with the strongest evidence for collective political organization all turn out to have been significant centers in the third millennium, before what is usually called the “Amorrite” upheavals at the start of the second millennium. In Imar, Tuttul, and Urgiˇs, decision making as a group dominates or at least rivals any native individual authority. It is still possible to imagine that the prominent collective leaderships of these towns arose only in the second millennium with the arrival of new Amorrite populations, but the pattern of prior occupation suggests explanation from older custom. At Urgiˇs, it is not clear that there was any gap in occupation between the Hurrian glory of the late third millennium and the Mari period of the second. The tahtamum of Imar and Tuttul is a term known ˘ centuries before the Mari texts. Even if from third-millennium Ebla, several it were true that all three towns were resettled by new Amorrite populations, we know that the Amorrites inherited and embraced the local traditions of southern Mesopotamia. It seems likely, therefore, that the collective political traditions of Imar, Tuttul, and Urgiˇs already had a home in these towns in the third millennium. Naturally, their success would have waxed and waned with local conditions through time. The period of the Mari archives was not

Democracy’s Ancient Ancestors

235

characterized by major empires, and many smaller polities were able to act more independently than in earlier and later periods. This may have been a time when old local collective forms could thrive.

b. democracy’s ancient ancestors In spite of my focus on the single body of evidence from the Mari archives, this study consciously takes up the challenge still posed by Jacobsen’s old definition of a Mesopotamian “primitive democracy.” By choosing the term “democracy,” Jacobsen placed the Mesopotamian assemblies in the same qualitative class as the Greek and the modern phenomena. He emphasized not only the collective character but the inclusive nature of such assemblies, for all free, adult males. I have kept the word “democracy” in my title both to recall Jacobsen and to invite consideration of this Near Eastern evidence as a backdrop to the Greek development, if only to provide a clearer idea of what raw materials were available to any new political developments of succeeding periods. At the same time, the collective political forms of the Mari archives are certainly not “democratic” by any careful definition of the term. They are not “primitive” forms of democracy, but rather belong to the political world before its development in Athens. Mari preserves not the direct antecedents of Greek democracy but a cross-section of its ancestry in the larger region. I conclude with a brief review of how the Mari evidence may relate to the origins of Greek democracy, after commenting on two other central ideas from the book. First, the Mari archives suggest that the ancient Mesopotamian languages did not respond to the phenomenon of urbanism by a new word for “city,” but rather by words for the complex polities in which cities were embedded. Second, collective and individual power must not be relegated to hierarchical levels, usually with the individual authority on top, but must be treated as two competing and perhaps even complementary elements of a single political system.

1. The Language of Political Complexity Archaeologists have often remarked that the Mesopotamian languages do not distinguish different settlement sizes by different words. The Sumerian uru and the Akkadian a¯ lum both designate everything from a tiny village to a major city. This fact is invoked as an introduction to perennial academic inquiry into the origins of cities. Ancient language will not help, it is said, because the ancients did not recognize any essential change by new categories. Opinions vary widely on how to read the archaeological evidence and on the value of distinguishing “cities” from other settlements. One could reevaluate the evidence from language from any number of starting points, but the political terminology from Mari does by itself suggest

236

Conclusions

a different perspective. At one level, the point about the words “uru” and “¯alum” is correct and important. The Mesopotamians themselves did not consider the increased size of settlements to have represented anything essentially new. They had specific words for “fortification,” such as Akkadian d¯urum (“fortification wall”) and dannatum (“stronghold”), but they created no separate word for “city.” In political terms, the collective organization of the a¯ lum was not viewed as different for a village than for a city, however we define major central sites that could bear this title. By the time of the Mari archives, however, the Mesopotamian political map was conceived to be divided first of all between m¯atums, and the political capacities of every a¯ lum had to be defined against the m¯atum framework. The a¯ lum could be the capital and hub of a m¯atum or integrated somehow into a m¯atum with other a¯ lum centers, but regional politics were decided between m¯atums. In fact, then, the Akkadian language had developed terminology that reflected an increasing political complexity. This complexity was defined not in terms of urban centers, but rather in terms of the populations surrounding those centers, whether or not the a¯ lum “towns” were actual residential communities or served mainly a ritual, administrative, or defensive function. The m¯atum is not easily equated with the modern idea of a “state,” certainly not in terms that could distinguish it from a “chiefdom,” but the m¯atum category offers a native engagement with the complex polities that some may call archaic states. As the dialogue continues regarding the formation and definition of complex polities in the early Near East, the cuneiform texts do provide relevant evidence that has perhaps been underestimated. Southern Mesopotamian Sumer developed a noticeably different terminology for polities that could incorporate “towns” (uru), and the differences between the language evidence deserve further attention.

2. Collective and Individual Power The abundant evidence for ancient Near Eastern kings makes it easy to assume that ultimate political power was most often individual. Where the existence of collective traditions is recognized, they are often attributed only to the villages within a larger kingdom. This is Liverani’s perspective in his treatment of Late Bronze Age rural communities (1975). Schloen rejects vehemently the two-sector social and economic system in Liverani’s interpetation, but he insists on the same decisive subordination of all collective forms to individual royal authority.10 The Mari evidence displays many examples of stronger collective forms than are often acknowledged, at the same time as we see a vigorous and successful brotherhood of monarchs. Rather than force all of the collective political activity into strict and complete subordination to kings, which is occasionally difficult, or to isolate rare exceptions of outright collective authority, it seems more productive to explain the evidence in terms of a


237

single system with both forms in play. At one level, the individual and the collective are distinguished sharply by both structure and ideology. Kingship was a solitary profession in the end, based on the simplicity of a single final word, supported though it might be by an elaborate system of advisers and administrators. Much of the ancient royal literature records claims of individual divine election to the care of and authority over the populace. Collective leadership is recognized in the diverse language for the group in action, always leveling the individual players into an indistinguishable mass by calling them “Tuttulites” or “elders” or “heads.” The very language of the town in action proclaims the group’s collective character and interests. In fact, however, the structures and ideologies of individual and group power seem to have blended within a single political system. Absolute authority or power is generally a meaningless category, because individual players and collective bodies interact at every political level, each exerting a distinct political force. Of course, Mesopotamia does offer some of the parade examples of “exclusionary” power in its dominant monarchies, but this seems to have represented a particular historical development rather than the universal inclination of Near Eastern politics. Ultimately, both individual and collective leadership had roots in the small communities of the era before the appearance of the first concentrated populations and more complex societies. Individual leadership acquired various forms, and what we choose to call “kingship” accompanied the formation of these complex polities. The creation of centralizing administrative systems around “exclusionary” royal leadership therefore must be considered in some sense secondary, not because the simple leads logically to the complex in any intrinsic evolutionary process, but by mere recognition that new phenomena came into existence in human history. We should perhaps also consider the parallel political life of collective leadership to have produced new formations to suit more complex societies, such as the alliances of m¯atums or tribal confederacies. The basic idea of the collective “town,” however, seems by nature to take its definition from a time before complex polities. At this point, it may be asked how the above scheme relates to the old question of early political developments in southern Mesopotamia, including Jacobsen’s model of “primitive democracy.” By focusing on evidence from further north and west, I have purposely left the problem of early Sumer, along with the later “Babylonian” periods, unresolved. Evidence for group leadership in Sumer is very old, and it surely coexisted with forms of individual leadership that predate the “kingship” of the later third millennium. It is a mistake to look for some pristine political landscape of purely collective governance in earliest Sumer, as elsewhere. At the same time, however, the dominance of individual rule appears to have been constrained by strong counterbalancing forces in temple and city institutions. There was no “primitive democracy,” but there was a lively interplay of individual

238

Conclusions

and group leadership. This was not the world of the supposed supreme monarch of a fixed Near Eastern type. In the end, I decline to undertake a solution for every body of Mesopotamian evidence through time and hope instead that specialists in each period and place will give new attention to the roles and backgrounds of group decision making in local political life. There will be no single pattern, in spite of the overlap of many common factors.

3. Mesopotamia and Athens Athenian democracy was such a shockingly new political system that it sometimes seems difficult to account for it as a true historical development, having origins in the practices and possibilities of preceding periods. It is not hard to trace Athenian political history back a couple of hundred years or so, and to consider the immediate background of full-fledged democracy, but how do we conceive the relationship of the entire Greek situation to what came before? There is an enormous problem inherent in the discussion of any background or backdrop to democracy.11 From the classicist’s perspective, it seems, democracy was created in large part by rejecting various prior forms of collective or group-oriented governance.12 Greek democracy depended on a particular value placed on the rights of the individual, which could not be shunted aside in the power play of some larger group.13 It is not at all clear that Near Eastern collective political traditions can be considered to anticipate democracy in any way. From the vantage of ancient Syria-Mesopotamia, however, the issue of background needs to be addressed in broader terms. Surely, the creation of this new political system was in some way responsive to the very traditions that were overcome, which formed a crucial part of the cultural raw material available to change. The issue of historical origins should not depend on immediate likeness, such as whether the Mesopotamians debated courses of action or passed laws in an assembly (yes, and no). We need to know how earlier Greece resembled the Near East or did not. Did the Greeks of the basileis rulers debate or pass laws that could be upheld against the power of the few? How old was the Greek notion that a polis “town” was composed of free “citizens” (politai) who had full rights and privileges, not subject to a king? More significant, then, are claims about ways in which earlier Greek society was already different from its Mesopotamian contemporaries. Was justice in earlier Greece already conceived in entirely human terms, with no resort to divine authority? Can any other essential cultural difference be claimed that would separate Greece from the Levant and Mesopotamia?14 In fact, we do have evidence for debate and critique of leaders in the tribal meetings of the rihs.um. Kingship in Syria was no more characterized as divine ˘ Mari’s particular landscape of minor powers jostling for than in Greece, and


239

an advantage belies any idea that the east was the land of the unapproachable monarch. Mesopotamia came to be dominated by centralized powers in later periods, but through most of its earlier history, this part of the Near East was broken up into small polities that were not essentially different from those of preclassical Greece. From an ancient Near Eastern perspective, predemocratic Greece seems to be cut out of the same cultural cloth as its eastern neighbors, in broad political and social terms. I emphasize the word “broad,” in that I acknowledge entirely not only the possibility of finding essential Greek distinctions but the necessity of doing so. At the same time, however, it is not possible to treat cultural geography as if it were divided into nation-states with absolute boundaries. Before Greek democracy, the strength of collective political forms is particularly striking. Archaic Greece seems to have been characterized by even more of a preference for group decision making than we find in the Mari period, which is unusually rich from a Mesopotamian perspective. In his monograph on the basileus, Robert Drews concludes that “during the ninth and eighth centuries, the evidence indicates that during this period most communities – and specifically those which in the Classical period we refer to as poleis – were ruled by small groups of hereditary leaders” (1983, 5). On one hand, he finds no credible evidence for kings in the poleis of Geometric Greece (p. 97), and on the other, he asserts that the early basileis of Geometric and Archaic Greece generally appeared as groups of elite leaders (pp. 101–7). The term “basileus” is not applied to exclusive leadership until the late seventh century, at the earliest. In his initial study of early Greek society, Ian Morris attempted to identify an essential shift toward the communal values of the polis with the arrival of the eighth century, based in changes of burial patterns (1987, 3).15 Even as Morris represents the restriction of formal cemeteries through earlier Greek history as a feature of “aristocratic” society, before a social revolution beginning around the year 750, he acknowledges that the elite constituted a large fraction of the population, between 25 and 50 percent.16 The formal cemeteries for this large group, located outside the clusters of houses in land that is thus defined in collective terms, fit well with the pattern of early Greek leadership proposed by Drews. To a classicist’s eye, such a political system may seem so far from democracy as to bear no relation to it, but a student of the earlier Near East, accustomed to the near-universal rule of kings, sees in preclassical Greece a remarkable contrast to the Mesopotamian pattern. These were, of course, the noninclusive forms of collective political life that had to be rejected in order for democracy to emerge. My question, then, is how the rejected forms differed intrinsically from the collective traditions of earlier Syria and Mesopotamia. Unless it can be shown that the very cultural soil from which Greek democracy sprang was fundamentally

240

Conclusions

unlike that of Syria and the east, it is surely useful to know more about those far-ranging collective political customs. Comparison with Near Eastern practices may even help to define more sharply where lie the key Greek cultural traits that allowed the entirely non-Mesopotamian development called democracy. There is one feature of the Syria-Mesopotamian tradition that seems remarkably like one core ingredient in the Greek political tradition. It is not immediately obvious that the political vehicle for democracy had to be the “town” or polis, but this became the bounding unit for defining the people to be included in Athenian democracy. At some level, identification as a citizen of Athens had to supercede the identification by tribes that still characterized Solon’s “council of four hundred” (above). The polis was not an idea created to serve a democratic political system, but it was rather adapted from existing use. It seems that intrinsic to the Greek polis was the idea of a community of full citizens, who became ideological equals in the collective action of the town. This building block of Greek politics seems to be anticipated in the Mesopotamian notion of the town, whether as Akkadian a¯ lum or Sumerian uru, and even in the Anatolian happira-. In the Mari evidence, ˘ collective political action is the most common setting for the language of the town, which at the foundation is defined in terms that require no chief or king. Is it possible that the polis “town” was one component of the raw materials from which Greek democracy came that was simply common to a broadly Mediterranean and Near Eastern heritage? To accept this would not in any way imply that the town idea was a direct precursor to democracy. Again, identification of shared cultural traits could help to pinpoint the unique Greek traits that shaped this entirely novel invention. I do not believe that I am suggesting anything that has not already been recognized by classicists themselves. Kurt Raaflaub describes a “Homeric” polis as “a community of persons or ‘citizens’ and as such more than an agglomeration of almost autonomous households (oikoi) banding together only in times of emergency.” This polis was loosely organized, without formal institutions, but with key components in place: equality with regard to prowess in war, ability to speak in assembly, and status compared with others. The Homeric polis “did not exclude continuing social, economic, and political differentiation; it remained in place, rooted in assembly and citizen army, but developed differently in each polis . . .” (1996, 150–2). Such an account of the earlier Greek polis presents more of a difference in degree than in kind from the collective tradition of the town in ancient Syria-Mesopotamia. Raaflaub understands Homer to reflect broadly Greek social conditions from the late ninth to the eighth centuries (2000, 26–7). In the end, ancient Mesopotamia may be famous for its kings, but they were the ones who left us their texts. In Greece, it was the democrats who left us theirs. Greece was not, however, a world away from Mesopotamia, and


241

Mycenae was not at all so far from the world known by Mari. It is my hope that all the peoples surrounding the eastern Mediterranean might be examined with certainty of both their cultural continuities and their individuality. Before democracy, there does after all seem to have been a long-standing regional tradition of collective political life, persisting in various degrees of tension with the grandeur of monarchs.

Notes

Preface 1. The Mari kings represented in the archive include Yahdun-Lim (ca. 1810–1794), ˘ as king of Mari), YasmahSumu-Yamam (ca. 1793–2), Samsi-Addu (ca. 1792–82, ˘ Addu (1782–75), and finally Zimri-Lim (1774–62). For the basic chronology of Mari kings, see Ziegler and Charpin 2001, 500. This widely used scheme is called the “Middle Chronology” and has recently been challenged in H. Gasch et al. 1998, which lowers the absolute dates for the early second millennium by 96 years. 2. There is no published count of the letters found in the Mari archives, and in any case, counting clay tablets is always an inexact procedure, when most of them are found in fragments. Bertrand Lafont has helped me arrive at a rough figure, based on his count of 2,550 separate published letters. There are certainly more to be published, mostly from pieces, but probably enough to put the final number at more than 3,000.

1. Introduction 1. See the map in Geyer 1985, 28. 2. For general descriptions of the archaeological finds from Mari, see Aynard and Spycket 1987–90; Margueron 1994, 1996b. 3. Aynard and Spycket (1987–90, 403) address this question directly for the condition of the site roughly fifteen years ago. They identify only area E of the Iˇstar temple in the Pre-Sargonic (mid-third millennium) town, the “red house” dated by the excavator (André Parrot) to the Akkad period (mid-late third millenˇ nium), and traces from the Sakkanakku period through the destruction of the tell (end of third and early second millennium). See Parrot 1936, 12, for a brief description of the situation for the last period. 4. For texts from scribal training with a mathematical bent, see Soubeyran 1984 and Chambon 2002. 5. I introduce new readings only occasionally, usually in response to patterns made visible by my immediate study of political terminology. The Akkadian transliterations provided in this volume likewise follow those of the new editions, except 243

244

6.

7. 8.

9.

10.

11.

12. 13. 14.

15.

Notes to Pages 4–7 where I remain more cautious regarding the proposed readings or restorations. For a review of the ambitious restoration policy of Durand and his colleagues, see my comments on ARM XXVIII in Fleming 1999, 167–8. In the beginning, each volume of this series was published with two separate components, the hand copies of the cuneiform tablets themselves (ARM), and a convenient book of transliterations and translations as Archives Royales de Mari, Textes (ARMT). With the transition to Durand’s leadership, this separation was dropped, and various solutions were adopted for providing photographs or copies of the tablets. For the sake of simplicity, I prefer to cite all texts from this series as ARM and use ARMT only when citing the commentary. ARM XXVI represents a special case, because it was intended as the start of a special col´ lection of letters, published with extensive discussion as Archives Epistolaires de Mari, or AEM, but only volumes I/1 and 2 have appeared. In the ARM series, the work of the new team is represented especially by volumes XXI (Durand), XXIII (Durand’s initial team, together), and XXVI (the team, together). ARM XXVII and XXVIII had already been assigned to senior members of Dossin’s team, Maurice Birot and Jean-Robert Kupper, respectively. See the recent discussion of Englund (1998, 73–81), who favors a non-Sumerian origin. For the opposing view, see Steinkeller 1995, 694–5, and Rubio 1999. See, e.g., Heimpel 1974, 171–4, arguing for the continued use of Sumerian during the Ur III kingdom at the end of the third millennium, and Lieberman, 1977, 20n50, guessing that pockets of Sumerian speakers may even have survived into the beginning of the second millennium, during the Isin-Larsa period. Cooper (1973, 242) has suggested that Sumerian was already dying out during the Ur III dynasty of southern Mesopotamia, replaced by Akkadian as the primary spoken language. The French team uses the unmarked -h- to represent uninterpreted laryngeals that are spelled with the Akkadian/cuneiform -h-. They also do not distinguish ˘ between two types of long vowels, using a circumflex (e.g., aˆ ) in all cases. I follow a convention whereby supposed original long vowels are marked by a macron (e.g., a¯ ), and the circumflex is reserved for long vowels formed from the contraction of two originally separate vowels. Margueron (1996b, 16) calculates that the diameter of the circular mound would have been 1,900 meters. He suggests both a twenty-ninth-century (1996b, 15) and a twenty-eighth-century (2000, 101) date for the time of establishment. See also Lebeau 1990, 283. Margueron 1996b, 15; 2000, 101. The canal on the left bank of the Euphrates would have carried boat traffic, saving forty kilometers of meanders from the winding river bed. For a cautious view, see Schwartz 1994, 158. On such “Kranzhugeln,” ¨ see Kouchoukos 1998, chap. 8. See Charpin 1987a. Twenty tablets were discovered between 1954 and 1974 in Parrot’s excavations, and seventeen more in 1980. They came from a variety of contexts, including buildings identified as diversely as palace, house, and sanctuary. See text E2.1.1.1 (from an Old Babylonian Nippur copy), in Frayne 1993, 10. For the archaeological evidence of this destruction, see Margueron 1996a, 97, 103.

Notes to Pages 7–11

245

16. Margueron (1994, 316) attributes this lack to the effects of erosion on the less important buildings. “Comme il est peu vraisemblable que la ville paléobabylonienne n’ait e´ té composée que de bâtiments juchés sur les points culminants du site, il faut penser que la topographie actuelle est la conséquence d’une e´ rosion différenciée qui a laissé en relief les bâtiments plus massifs et surtout durcis par l’incendie, mais a e´ liminé les constructions plus fragiles et peut-être non incendiées, situées normalement a` des niveaux légèrement inférieurs.” 17. Porter (2000, 456) emphasizes the possibility of a pastoralist component to the regional economy. 18. For discussion of the phenomenon of Amorrites and others cast as “subhuman barbarians,” see Cooper 1983, 30–3. 19. William Hallo provides a fairly long discussion of this transition in his contribution to Hallo and Simpson 1998, 80–7. 20. Lines 1:46 and 17; see Cooper 1983, 33 and nn. 74 and 75; Michalowski forthcoming b. 21. See Iˇsbi-Erra hymn A I 10’–12’ and hymn G 10’–11’, in Michalowski forthcoming b. 22. This relationship has been the focus of a series of articles by Piotr Michalowski (1995, forthcoming a and b). The author kindly provided me with manuscripts for the two forthcoming articles. 23. For the reign of Yahdun-Lim generally, see Charpin and Durand 1985, 293–9; Durand 1997, 43–4;˘ and Ziegler and Charpin 2001, 496–7. On Mari’s relations with Eˇsnunna during the reign of Yahdun-Lim, see Charpin 1992a. ˘ Charpin and Durand 1985, 297–9, and 24. On the brief role of Sumu-yamam, see the letter ARM I 3, reedited on pp. 339–42 (=LAPO 18 no. 931, p. 72). 25. Samsi-Addu claims to be “the one who united(?) the land between the Tigris and the Euphrates,” in the votive inscription A.889:6–8, published in Charpin 1984, 48. The same epithet is found in another text from Aˇsˇsur; see Grayson 1987, RIMA 1, A.0.39.1:5–8. 26. See FM III 4 i:18–21, in Durand and Guichard 1997, 63–70. 27. Durand, in Durand and Guichard 1997, 64. The form of the substantive yar(r)¯adum is not certain. 28. A good discussion of the arrangement is found in Villard 2001, 10–12. 29. Sasson (1998b, 457–8) maintains that Hadni-[ . . . ] in one royal seal should ˘ king’s other royal seals (see below, be the same figure as Yahdun-Lim in the p. 149, on this evidence).˘ Zimri-Lim is identified as the “son” of both named men. 30. See Dominique Charpin, who generously provided a draft of his forthcoming synthesis, “Mari et le Proche-Orient a` l’époque amorrite: Essai d’histoire politique. Le règne de Zimrˆı-Lˆım (1775–1761)” (forthcoming a). According to Charpin, Bannum had already served Yahdun-Lim, if the seal of Baninum in˘ Bannum was one of the Sim alite dicates the same person. Under Samsi-Addu, chiefs of pasture (merhûm) in the region south of the Sinjar, as demonstrated ˘ by two letters from Yarim-Addu to Yasmah-Addu that mention him as such. In his personal seal, Bannum calls himself ˘the one “who restored the scion of Yahdun-Lim to his place” (mu-te-er p´ı-ri-ih Ia-ah-du-un-Li-im [a-n]a iˇs-re-ti-[ˇsu]). ˘ Bannum calls the diviner and future ˘ confidante ˘ When of Zimri-Lim “my booty”

246

31.

32.

33.

34.

35.

36.

37.

38.

Notes to Pages 11–20 (kiˇsid q¯at¯ıya) in an early letter to Zimri-Lim, he can say this because he was the one who had captured Mari (ARM XXVI 5:28). The chronology of Zimri-Lim’s reign has long resisted definitive resolution. When Charpin and Durand established their initial formulation of Zimri-Lim’s chronology, at the beginning of the new generation of Mari studies, they proposed that he ruled for fourteen years, with the first two remaining uncertain (1985, 305–6). Pierre Villard (1993) then offered a confirmation of this essential scheme by clarifying the position of overlapping year names from the first three years of Zimri-Lim’s reign. The exact timing of the calendar during these first years has remained difficult to define, however. Durand proposed an adjustment of the calendar at the beginning of the reign that produced a gap that had to be made up by added intercalary months during later years (Durand and Guichard 1997, 30–1). In his latest work on the history of Zimri-Lim, however, Charpin (forthcoming a) objects that this gap does not provide the best explanation of the evidence and calculates that ZimriLim’s first regnal year was simply not complete, with overlapping year names used until a stable system was achieved. The king appears finally to have ruled for thirteen years and three months. I have therefore adopted the new dates found in Guichard (2002), with Z-L 0 indicating the months of the king’s first accession to the throne and Z-L 1–13 marking the thirteen full years of his reign. Jesper Eidem (2000, 256–7) observes that Zimri-Lim’s goal in the Habur basin ˘ control was different from that of Samsi-Addu. Where Samsi-Addu wanted direct of the region, Zimri-Lim simply wanted to keep the major eastern powers out and used these alliances to that end. There are four or five rare examples of the adjective Y˘emˆınˆı, “Yaminite,” apparently as a secondary formation from Benjamin, rather than directly taken from the older Mari category. See a similar complaint by Elizabeth Stone (1997), who cites in particular the nineteenth-century Yoruba in Africa and late medieval Islamic cities as examples to show that models of state society “should not focus exclusively on coercion” (p. 16). Blanton et al. 1996, 2; cf. Blanton 1998, 135. Blanton attributes this neglect to the idea that “the European enlightenment’s social contract theory is largely a recent and culturally specific phenomenon, with little relevance to non-Western states” (p. 153). See, e.g., Kurt Raaflaub (1996): “Democratic equality is realized by giving each citizen access to political participation and office” (p. 140). “I have argued that there was no democracy in the precise sense of the word before it was introduced in Athens, and at least initially it was tied to conditions that applied uniquely to Athens” (p. 149). Morris 1987, 3. I cite Morris because he proposes that this change can be seen in the archaeological remains alone, especially the burials, aside from the textual record. I do not see how the expectation that Greek democracy had roots in such polities must be anachronistic hindsight, as charged against Morris by Cartledge (1998, 386). In my choice of these specific terms, I am following the categories offered by the Anthropology Department of New York University.


247

2. The Tribal World of Zimri-Lim 1. This term also comes from Tapper (1990, 69). He observes different forms of tribal states, but finds the most common premodern type to be where one tribal (descent-based) elite or dynasty rules a conquered territory and its heterogeneous population. 2. The problems involved with “tribal” terminology and its relationship to pastoralism are immense. For a helpful and provocative review of the “tribe” in anthropological and social research, with a particular orientation toward the ancient Near East, see chapter 2 of Anne Porter’s dissertation (2000, 51–96). 3. For definitions of “time-space distanciation” and “presence availability,” see Giddens 1981, 4–5, 37–40. His definition of tribal society comes from p. 162, cf. 92–4, and 1984, 194–5. 4. She draws this comparison in the second section of her chapter 10, pp. 442– 57. 5. See the draft records of ARM XXIII 428 and 429, which record numbers from each town that were promised, as opposed to those who showed up. These documents reflect the mustering of troops to combat an invasion by the major kingdom of Elam in the eleventh year of Zimri-Lim’s reign (Z-L 10); see Durand 1998, 156–8; Charpin forthcoming a). 6. See, e.g., the alliance made between the Yaminites and Zalmaqum at Harran, as ˘ reported in ARM XXVI 24. ki ki ki 7. ki-ma ma-at Ia-am-ha-ad ma-at Qa-t.a´ -nim u` ma-at A-mu-ri-im ni-iK-hu-um ˇsa ˘ DUMUmeˇs Ia-mi-na˘u` i-na ma-tim ˇse-ti DUMUmeˇs Ia-mi-na ˇse-em i-ˇsa-bi u` na-wa-ˇ sunu i-re-i-em u` iˇs-tu da-ar-ka-tim ni-iK-hu-um ˇsa Ha-nameˇs I-da-Ma-ra-as.. Durand cites ˘ and the social world of the Mari this text in his forthcoming study ˘of nomads archives, and I would like to thank him for graciously allowing me to quote it here. André Parrot alluded to the three domains of Yaminite movement without quoting the text (1950, 7). The letter is sent to Zimri-Lim by one of his Simalite merhûms, and it offers an excellent example of the Hana as “our tent-dwellers” ˘ Simal, to be discussed further in Chapter 5. Two ˘ key terms are evidently for the Amorrite, and my translation of both is inadequate. Durand discusses them at some length. The two complementary regions for tribal occupation are written ni-iK-hu-um, a word that Durand finds once in the form ni-gi-ih-ˇsu, which would ˘ favor a˘ reading in -g-, perhaps from the root ng , “to touch, to reach.” This would suggest a domain defined by what is “reached” in travel, so a “range.” The phrase “iˇstu darkatim” is otherwise unattested even in unpublished Mari texts, according to Durand, and surely refers to long-standing custom. Durand finally prefers a connection to the Hebrew derek (“road”) and imagines reference to a hypothetical separation of the two nomadic peoples at their first split into right and left, an idea he will elaborate on in his own publications. Notice also that I translate the demonstrative pronoun ˇsêti in the phrase ina m¯atim ˇsêti with a distributive sense, “in each of those lands,” in order to recognize the distinct political identity of each m¯atum. For my initial citation of this text, see Fleming 1998, 61 and n91. 8. See my brief discussion of the geography in Fleming 1998, 62 and nn 92–4. For more on this western region, see, esp., Bonechi 1992, on Hazor and Joannès 1997, on Palmyra and Qatna.

248


9. See Lafont 2001a, 240; A.449, unpublished, and cited here in translation only: “Toutes les choses que [Dâdˆı-Hadˆ un] nous a faites, nous les avons (d’abord) rapportées a` Yarˆım-Lˆım en disant: ‘Toi, tu es notre seigneur et tout le pays est entre tes mains. Que notre seigneur e´ crive donc a` Dâdˆı-Hadˆ un afin que l’on ne nous fasse pas de mal.’” 10. FM VII 7:8–10. The verb for temporary residence is waˇsa¯ bum, “to stay.” 11. See Chapter 1, p. 10. 12. See the overview by Charpin and Sauvage (2001). 13. Consider, e.g., the first definition of “tribe” in Webster’s Third International Dictionary, as “a social group comprising numerous families, clans, or generations together with slaves, dependents, or adopted strangers.” 14. For a helpful introduction applied to our historical setting, see Porter 2000, chap. 3. 15. Evans-Pritchard’s definition of tribal territory begins in fairly mechanical terms, sketched as concentric circles, with the “tribe” defined in part by “common and distinct territory” (1940, 114–16 and 122). He saw lineages, however, as a social organization that was maintained even when people were dispersed across different territories, as might occur when groups were not satisfied by current conditions (pp. 209–10). Bonte uses the word “territory” in opposition to identification by a more local place of residence (1979, 204, 212). Khoury and Kostiner comment that most tribes are organized by ideologies of common descent but that there are exceptions (1990, 5). See also Tapper 1990, 50–1. 16. A.3080: 19–21, passim, in Durand 1990d; cf. 1992b, 116. 17. A.3572:2 –4 , in Durand 1992b, 114; cited in full in Chapter 3, in the section on “Non-Simalite Hana in Zimri-Lim Period Texts.” ˘ 1–3, 9–11, 20–4, and 30. 18. ARM VIII 11, esp. lines 19. Peters finds that the lineage model does not explain what transpired when a killing took place among the bedouin (p. 261) and observes that group meetings are never based on lineage beyond the scale of a “tertiary tribal section” (p. 287). See also E. R. Leach, who concluded that for villagers in Ceylon, kinship and descent do not define action. The community is defined by territory and the livelihood earned from it, and descent-based identities are effectively a fiction (1968, 300–2). 20. The application of Rappaport’s work to Salzman’s hypothesis is made by Porter 2000, 114–20. 21. See A.981, in Durand, 1992b, 117–18 (see below), for a specific case of a Yaminite town accepting loyalty to the Simalite Mari king. 22. Again, see ARM XXIII 428 and 429 for military service, counted by town and Mari district (hals.um). ˘ and 429 depend both on Zimri-Lim’s organization of the Mari 23. ARM XXIII 428 kingdom and on the Yaminite divisions into tribes. For example, they begin with named towns in the districts (hals.um) of Mari, Terqa, and Saggaratum, followed by a total commitment under ˘the leadership of Yarim-Lim, ruler of the Yahruru ˘ tribe. 24. For the early Near East, see Diakonoff 1991, 29–33; Adams 1974, 2–3. 25. Zarins 1990, esp. 39–40, 53–4; see also Zagarell 1989, 294–8. Zagarell sees two main early periods of high-level pastoral activity in the Zagros highlands: the


26. 27. 28. 29. 30.

31.

32.

33. 34. 35. 36. 37.

38. 39.

40.

41.

249

Neolithic, with its dispersed, mixed economy and relatively undifferentiated, acephalous society, ending in the early Chalcolithic; and the Late Chalcolithic, with a coexistence of sedentary villages and mobile pastoral camp sites, both sharing similar material culture (pp. 299–300). Frank Hole (2000, 18–20) finds that bone evidence at Feyda for the Pre-Pottery Neolithic (8000–6500 b.c.e.) and at Kashkashok for the Pottery Neolithic (from 6500 b.c.e.) indicates seasonal occupation in the Habur by pastoralists (pp. 18–20). Kouchoukos develops at ˘length the possible roles for pastoralist products in southern Mesopotamia of this period. Porter presents a focused summary of her argument on pastoralism and the state in her forthcoming article. “Il n’est pas défendu de supposer que la situation qui existait au IIIe millénaire ne devait pas eˆ tre très différente de celle du IIe millénaire.” Eidem 2000, 262. He also argues that there is considerable consistency in the names of smaller towns, including those in the Habur basin. ˘ See Buccellati 1966, 236–7; Edzard 1987–90, 438–40. For the Ebla evidence, see Archi 1985, 7–13, and for the Akkadian evidence, see Westenholz 1999, 96–7 and nn 441–2. In Frayne 1990, see Abi-sare (E4.2.6.1:i 27 ; E4.2.6.2004:4), an early king of Larsa; and Sˆın-gamil of Diniktum and Arim-Lim of Mê-Turran in the Diyala (pp. 685, 700). Warad-Sˆın calls his father, Kudur-mabuk, “the father of the Amorrite (land)” (E4.2.13.3:7–8, passim). Durand places Amurrum on the coast, without offering specific evidence (1997, 49). A far-western solution does make sense. Against Dossin (1957, 37–8), a location south of Qatna should be ruled out by the frequent juxtaposition of Qatna and Hazor, the next major kingdom in the direction of Palestine. See Bonechi 1992, 10n7. See Durand forthcoming, as part of the Yabasa group (see ARM XXIV 235:8). For a useful review of Mari evidence, see Sasson 1998a, 121–3. A.109, lines not cited; in Durand 1992b, 125. ´ as a determinative, as Kraus 1970, 28; lu´ Ak-ka-du-u´ u` lu´ A-mu-ur-ru-u, ´ reading LU in the Mari archives. A.489, in Durand 1992b, 113n137; LUGAL Ak-ka-d[i-i]m u A-m[u-u]r-ri-im. See also Charpin and Durand 1985, 323n131, for some detail about the letter’s context, but not quoting this part of the text. ARM VI 76:20–21. Bertrand Lafont (1999, 51) identifies four main groupings of geographical space in early second-millennium Mesopotamia, based especially on population, which is defined by “des réalités ethno-linguistiques.” These are the Assyrians, the Amorrites, the Akkadians, and the Subarians. Durand divides between Akkadian easterners and Amorrite westerners, with the Amorrites then divided between northerners and southerners (Simalites and Yaminites). This hierarchy combines the Akkadian/Amorrite duality, which originates outside the tribal populations, with a truly native tribal duality, when the two pairs should probably be regarded as separate and only secondarily related. This is the main point of Durand (1992b).

250


42. ARM XXVII, p. 224 note b. 43. ARM XXVII 135:27–33, letter of Zimri-Addu; [lu´ Nu-u]m-ha-[y]uki lu´ Ka-ha-[t]ayu[ki ] x-ma(?)-a(?)ki(?) 28 [lu´ ]E-lu-uh-ta-yuki [x x] x 29 u` lu´˘Ak-ka-du-[u]m˘k [i ] 30 ni-[i]ˇs DIN[GIR-l]im `ız-ku-ru 31 u` l [˘u´ ]Ak-ka-du-um ˇsa it-ti-ˇsu-nu 32 ni-iˇs DINGIRki ` s-nun-naki 33 lu-u´ lu´ KA.DIN[GIR.RA] ´ lim [`ı]z-ku-ru lu-u´ lu´ Eˇ pa-ga-ar-ˇsu u-ul u´ ! ´ w[e]-di. 44. On the importance of Eˇsnunna, see Charpin 1992a, 37–8. For “Akkadian” as Eˇsnunna, see A.988:12, in AEM I/1, p. 306 note a. In ARM XXVI 468:13, the m¯at Akkadˆım seems to be Babylon. The m¯at Akkadˆım appears in broken context in ARM XXVI 75:6 , where it cannot be identified. 45. ARM XXVI 549:4 –7 , letter to Asqudum from S. idqum-lanasi; 4 [lu-bu-uˇs-tam] 5 6 !? Su-ba-ri-a-am u-ul [la-ab-ˇsa-a-ku] lu-bu-[u]ˇs -tam Ak-ka-di-a-a[m-ma] 7 [l]a-ab´ ˇsa-a-ku. 46. FM VII 26:49, 52. 47. The first reference to a “land” (m¯atum) of Akkad, as opposed to the city alone, seems to come from the reign of Naram-Sˆın. The title “king of the land of ˇ Sumer and Akkad” is already found in an inscription of Sulgi, during the Ur III dynasty at the end of the third millennium, and repeated by Lipit-Iˇstar of Isin and eventually Hammurabi of Babylon. ˘ the Yaminites may have always been rivals, but during the 48. The Simalites and third year of Zimri-Lim’s reign, tensions broke into outright war between the Yaminites of the Euphrates region and the Simalite king. Zimri-Lim’s victory did not include all Yaminites in some global sense, but the Yaminites who lived in territory dominated by Zimri-Lim had either to flee or to capitulate to Simalite rule. For a summary, see Durand 1998, 420–2. 49. For the Yaminite rulers as actual “kings” (Akkadian ˇsarrum), see ARM XXVI 233:30, 37, in the report of a dream by a man from a Mari area village, spoken by the dreamer, not the Mari official; and FM VII 7:32, in the mouth of the king of Yamhad. ˘ 50. This distinction of terminology was not included in my preliminary work on these tribal terms, published as Fleming forthcoming. 51. This oft-repeated observation was made first, to my knowledge, by Rowton (1977, 189). He and others after him regarded this as a characteristic of the “Hanean” tribe, a category now greatly illuminated by the more recent work of Charpin and Durand (see below). For the Hanean gayum, see, first of all, Birot 1953, 1955; also Kupper 1957, 20; Malamat 1967, 133–4; Rowton 1976b, 243; Talon 1985; cf. Anbar 1991, 78–81. 52. As mentioned earlier, Tapper’s definition of larger groups is useful (1990, 68). A confederacy is a union of tribal groups for political purposes, sometimes an alliance on the basis of imputed common descent, usually but not always with central leadership. In the case of the Yaminites, the confederacy would not have had any central leadership during the reign of Zimri-Lim. 53. Talon even identified Zimri-Lim as a member of the “Hana” tribe (1986, ˘ 5). 54. Luke 1965; Matthews 1978; Anbar 1991. For some of the key articles in a long series, see Rowton, 1967, 1973a, 1973b, 1974, 1976a, 1976b, 1977. 55. Zimri-Lim’s victory provided the name for the fourth year of his reign (Z-L 3); see the chronological chart in Charpin and Durand 1985, 305–6.


251

56. Durand 1992b, 113 and n138; see also Durand 1998, 418; and his eventual study on the social landscape of the Mari archives. This position moves beyond the explanation initially offered by Charpin and Durand (1986, 153–5), where they still understood the name “Hana” to represent a proper noun, if appli˘ tribespeople. Durand now renders the cable to both Yaminite and Simalite term “bédouin,” which he does not intend as a narrow equation with modern bedouin nomads. I appreciate that the use of this contemporary term need not force an anachronistic identity on the ancient hana/Hana, but I have cho˘ ˘ order to avoid any sen the more literal (and less elegant) “tent-dweller(s)” in misunderstanding. 57. Rowton defined Mari itself as an intermediate dimorphic state, a territorial state within which tribal and dimorphic chiefdoms played a prominent role, but the authority of which stood apart as centralized bureaucracy (see 1973b, 203–4; 1976a, 27–8). 58. See most recently 1998, 417–18, which will eventually be superceded by an exhaustive discussion in his projected monograph on the social landscape of the Mari archives. Before this, the key argument was elaborated in 1992b, 113– 14; also 1993b, 47; 1997, 17, 456; Charpin and Durand 1986, 153–5. ´ s 59. ARM II 53:12-31: 12ˇsa-ni-tam lu.meˇ su-ga-gu ˇsa DUMUmeˇs Ia-mi-naki 13 i-na Za-al-paki 14 ki ah ip-hu-ru-ma a-na A-hu-na-a il-li-ku-ma 15 m S.u-[raha-am-mu-u´ u` Ia-ri-im-Li´ lu.meˇ ´ s ˘ 16 uˇ ˘ im] su-[ga]-gu ˇsa ´˘s-bu-ma a-n[a] .se-er˘ be-l´ı-ia a-na a-l[a-k]i-im 17 id-bu-bu ha-[n]a 18 it-bu-ma a-na Ia-r[i-im-L]i-im 19 u` S.u-ra´ ha-am-mu-u´ k[e-e]m iq-bu-u´ 20 um˘ ˘ u` a-la-ni5 -ni5 er-ˇsa 22 [ˇsu]m-ma ma-a-mi ˇsu-nu-ma a-na .se-er Zi-im-ri-Li-im 21 [al]-ka-ma d 23 La-hu-un- Da-gan [la i]-il-la-ak u-lu ´ ni5 -da-ak-ˇsu 24 u´ la-ˇsu-ma i-na giˇs GU.ZA-ˇsu 25 ˘ ´ s nu-da-ap-pa-ar-ˇsu an-né-tim 26 lu.meˇ su-ga-g[u] ˇsa DUMUmeˇs Ia-mi-naki i-da-ab-bu´ s bu 27 [i-n]a-an-na a-nu-um-ma Ia-ri-im-Li-im 28 S.u-rasu-ga-gu 29 ´ ha-am-mu-u´ u` lu.meˇ 30 ˘ a-na .se-er be-l´ı-ia i-i[l-la-ku]-nim ma-li ˇsa i-ri-ˇsu be-l´ı la i-ka-al-la-ˇsu-nu-ˇsi-im u` a-naku 31 wa-ar-ki-ˇsu-nu-ma a-ka-aˇs-ˇsa-dam. For Durand’s use of this text to advance his argument about the meaning of hana, see 1992b, 113. He translates the whole ˘ text as no. 702 in LAPO 17 (pp. 448–9). 60. See ARM XXVIII 25:8–16, evidently another letter from a Yaminite ruler (p. 24 and n43). 61. A.505:22 –23 , in Durand 1992b, 114n146: ha-nameˇs DUMU Si-ma-al ki i-na li-ib-bi ma-a-at Ya-mu-ut-ba-limki i-re-i, “The Simalite˘ Hana are grazing (their flocks) in ˘ the midst of the land of the Yamutbal.” 62. ARM XXVIII 36:12-16, written ha-na DUMU Si-im-a-al u` Ya-mu-ut-ba-lum. The ˘ combination of hana with tribal names appears when there is no obvious refer˘ below in this chapter. My interpretation would apply to either ence, as discussed the limited or the distributed reading of hana. ˘ and Zimri-Lim divide the popula63. Texts from the reign of both Yahdun-Lim ˘ tion between “hana” and “townsmen.” See A.4280:32–3 (cf. 47) for the divi˘ sion of Yahdun-Lim’s military (in Charpin forthcoming c). For the same di˘ vision in an oath sworn to Zimri-Lim, see M.6060:22 –23 (in Durand 1991, 52). 64. It is not clear to me that the dominant ha-na spelling, with final frozen -a, can be explained as a feminine ending, when˘ the case system still seems to be active in West Semitic dialects. Later Ugaritic still retains case vowels and depends on -t- or -at- to mark feminine substantives.

252


65. Ha-nameˇs ˇsa-lim um-ma-na-[at be]-l´ı-ia ˇsa-al-ma (line 4); um-ma-na-tum ˇsa-al-ma u` ˘DUMU Si-im-ha-al ˇsa-li-im (line 5 ). These are literally the first and the last lines ˘ letter, formulated with a crisp chiasm. For discussion of the in the body of the name Yatarum, rejecting an identification with one of the governors of Mari’s district centered at Qat.t.unan, see Guillot 1997, 283–4. 66. Kupper 1957, chap. 1 (“Les Hanéens”); Anbar 1991, 80–3, passim (on “Hanˆ um”). 67. Dominique Charpin brings to my attention the fact that French does not offer the possibility of a compound word that follows the proposed etymology, as in the English “tent-dweller.” This means that “bédouin” must serve to designate the same general category, without any forced anachronism by association with peoples from the modern Middle East. 68. See the earlier discussion by Charpin and Durand (1986, 152–5). 69. This is the special interest of A.3080, published by Durand (1990d, 102–5). 70. For examples of how “Hanean” tribal features have been treated, see the literature cited in note 66. Another expression of this dichotomy, which should now include the Simalites, is geographical. Kupper began his ground-breaking study of tribal peoples in the Mari archives with the observation, “Le territoire propre de Mari semble avoir e´ té le principal centre de peuplement des Hanéens . . .” (1957, 1). Based on newer evidence, Moshe Anbar correctly identified specific concentrations of Yaminite and “Hanean” populations: “De même que les Hanûm e´ taient liés tout particulièrement a` l’Idamaras. , les Bini-Yamina e´ taient liés a` la région d’amont, la région située entre le Balih et le Haut Euphrate . . .” (1985, 21). On the geographical aspect of this identification, see also Charpin 1992c. 71. See also Talon 1978, 1982, and 1985. For variants of the y-b-s root as the same Simalite division, see Abrahami 1992. 72. FM III 136:5–15, letter of Yatarum (not the governor of Qat.t.unan) to Zimri-Lim: 5 pa-t.`ı-ir i-pa-t.a` -ru u-ul ´ [i-s.a-ba-at-ma(?)] 6 na-p´ı-iˇs-ti ha-nameˇs i-pa-t.a´ -ar 7 ia-ri-ka-am meˇs 8 i-ˇsa-bi-ir 9 i-na-an-na an-nu-um-ma ˘10 m La-wa-s´ı-im 11 lu´ Yu-ma-ha-miˇsa ha-na 12 ˘ ˇ hi.a yu ip-t.u-ra-am 13 su-ga-ag-ˇsu Da-di-[Li-im] 14 u` ba-ˇsi-su´ 2 me UDUhi.a 15 5 AN˘SE ˘ ˘ be-l´ı li-ir-di. For the probable identification of the writer with the military leader who wrote FM III 135, see Guillot 1997, 287. 73. Guillot does not provide the basis for the proposed restoration of the name Dadi-Lim. 74. The etymology of the noun sug¯agum is uncertain. No verb *sag¯agum is known either at Mari or among the other early Semitic languages. Durand suggests that the root *sgg may be related to the third-weak root ˇsgy in Hebrew and Aramaic, “to become larger, to increase” (1997, 208). 75. If the sug¯agum stands below the gayum (“division”) in the chain of identification, we will expect more than one such leader per gayum. When that division develops a fixed attachment to certain settlements, its sug¯agums may even come to be identified by town, though as individual residents among many, not as the single leader. In his article on the Simalite town of Sapiratum in the region of Suhûm, downstream from Mari along the Euphrates, Charpin cites two ˘ texts that record sug¯ag¯utum payments by separate named persons unpublished who are “Sapiratumites” (1997b, 353n44). In the eyes of the administrators


76.

77.

78.

79.

80.

81.

253

who recorded the receipt of these payments, these leaders were identified by their town residence, but to my knowledge, we have not yet found a title such as “the sug¯agum of Sapiratum” or of any other proven Simalite town. In the texts for sug¯ag¯utum payments for men of Sapiratum, the “Hana” (tent-dweller) mode ˘ of identifying the Simalite population may be reflected in the fact that both leaders pay in sheep, rather than in silver. “We are even tempted to say that to speak of the sug¯ag¯u of the Yaminites or the Hanaeans is synonymous with speaking of the sug¯ag¯u of the cities/towns of the Yaminites or the Hanaeans” (Nakata 1989, 118). Charpin has shown sympathy for this idea, observing the phenomenon of Yaminite towns (1991a, 12), while Bertrand Lafont has reservations, based on the reference to the sug¯ag¯u of Yaminite hana (tent-dwellers) in ARM II 53, along with similar texts (1997a, ˘ 37n25). ARM III 38:15–18, 15 a-na a-la-ni ˇsa DUMUmeˇs Ia-mi-na 16 aˇs-pu-ur-ma 17 lu´ su-ga-guum ˇsa Du-um-te-enki 18 ki-e-am i-pu-la-an-ni; Durand, LAPO 17, no. 686, pp. 428–9. See also ARM II 53:12–31 (above); II 92:12–13; II 103:12; III 21:7–13; III 70 + M.9610 (see Durand, 1990a, 149–51); XXVI 450:7–9. Sahrû is attested to have a sug¯agum in A.2435:20-21, of Charpin 1995b, 53. ˘ Ziniyan (see ARM XXIV 61 ii:4 ; XXVI 462:5–6) appears to be an abbreviation for Ziniyan-Yahapalum. These towns are identified as Yaminite by Millet˘ Albà. ARM XXVI 241:5-7, 4 i-na Za-ar-riki Ra-bi-im 5 i-na UDUhá ˇsa Za-zi-im 6 lu´ su-ga-giim 7 SILA4 iz-bu-um i-wa-li-id-ma, “In the town of Rabbean Zarrum, among the sheep of Zazum the sug¯agum, a malformed lamb was born. . . .” This text does not exclude the possibility that other sug¯agums could reside in this town, though the wider pattern of references to the leadership of Yaminite towns makes this unlikely. The Yaminite identity of this town is proven by its inclusion in the muster lists of ARM XXIII 428 and 429, in the district of Saggaratum and under the leadership of Dadi-hadun, ruler of the Rabbûm tribe (proven in part by the ˘ ARM XXIII 428 and 429). name Zarri Rabbiyum in ARM IX 248:11 . In the muster lists of ARM XXIII 428 and 429, Dabiˇs is responsible to Yarim-Lim, ruler of the Yahrurû tribe. These lists were drawn up in connection with Mari’s contribution to˘the aid of Babylon in the twelfth year of Zimri-Lim’s reign (Z-L 11). They therefore provide a provocative counterpoint to A.981, the letter in which the leaders of Dabiˇs swear loyalty to Zimri-Lim in part by identifying the town with the Simalite tribe of the Nihadˆ u: “Let us enter ˘ li-ib-bi DUMU Siinto the midst of the Simalite(s) among the Nihadˆ u . . .” (a-na ˘ in Durand 1992b, 117–18). In i[m-a]-al i-na Ni-ha-di-i i n[i-r]u-ub-ma; lines 39–40, ˘ light of the continuing identification of Dabiˇs with Yaminite and Yahrurˆ u rule, ˘ identity perhaps this language should be understood not as a transfer of tribal from Yaminite to Simalite, but rather as the forging of a tribal brotherhood between neighbors as prelude to the alliance ceremony of killing an ass (lines 40–1). Pierre Villard, ARM XXIII, p. 496n113. This town should be identified with ˇ MUSEN ˇ há.ki in the Mari district, in ARM XXIII 428:2, the Yahrurˆ u town of Sa ˘ ki ˘ ˇ in spite of the spelling Sa is.-s.u-ra-tim in XXIII 426:7. Compare ARM XXIII ´ 554, a list of ransomed men and women who are all identified by towns shown to be Yaminite by XXIII 428–9: Dimtên (line 4, Terqa district); Dabiˇs (line 6,

254

82.

83.

84.

85.

86.

87.

Notes to Pages 52–53 Saggaratum district); Raqqum (lines 8, 18, Terqa district); Miˇslan (lines 13, 36, ki ˇ hu-s.u-ra-tim 38, Mari district); Sa (line 21, Mari district); and Daˇsran (line 30, ´ ˇ hi-s.u-ra-tim Terqa district). Notice˘the spelling Sa in ARM XXIII 552:3. The word ´ ˘ lexical texts as u-s.u-ru for bird at Ugarit is .sr, vocalized in syllabic and u-s ´ ´ . u-ru ´ ˇ (Huehnergard 1987, 162), equated to MUSEN. See the discussion of Millet-Albà, on ARM XIV 62:4, 9, and 28 (LAPO 17 no. 648, pp. 350–1). Note that the town name was originally read as Karhan, in the ˘ original publication. ˇ See, esp., ARM XXIV 53–63. Payments in silver come from the towns of Sakka (55; 61 ii:6 ; 63 ii :3 -4 , all three by the sug¯agum Namiˇsum, recording the same amount), Himaran (57), Amatum (59, silver and sheep), S. ubatum (60:3–4), ˘ Zibnatum (60:7–8), Dunuhˇsu (61 ii:3 ), Ziniyan (61 ii:4 ), Utahum (61 ii:5 ; ˘ Menihum), and Se ˇ hrum (62 i :12˘–13 ). Dagan63 ii :2 , both by the sug¯agum ˘ in 62 i :10˘ –11 with a payment in aniabi of S. ubatum appears a second time mals, including sheep. In the same text, Yans. ibum of Himaran makes at least ˘ part of his payment in sheep (62 i :18–19 ). The three “elders of Saggaratum” (the district, not the town?) whose sug¯ag¯utum payment is recorded in ARM VII 311 make it in silver; see the text in Talon 1978, 146; after Bottéro 1958, 164–5. In this case, the identification by gayum covers multiple holders of sug¯agum status: Zu-Hanunim the Yakallitˆ u (ARM XXIV 53); Ripi-Addu the Weru ˆ (54, just ˘ ´ as “the man of Ibalan,” LU I-ba-la-an; but in 62 i :15 -16 , as the son of Iba[lan], ´ W]e-er-i-[im(?)]); S. abihum the Yakallitû (56); Ani-Lim and Layasum, both [LU ˘ of Nisiya(?) (58:9); Ayanum of Ibal-[ahum(?)] of Abi-nakar (58:6); Ibal-pi-El (58:12); Mutatar of Abi-nakar (61 iii:4 ); Bahadum the Yabasˆ u (62 i :3 ,˘possibly with a payment in silver, also); Bahlu-gayim˘ the Amurrˆ u (62 i :4 –5 ); Hatnan ˘ of Nahanum (62 i :7 8 , sheep and˘ cattle). All of these make their payments in ˘ animals. ARM XXIV 61 iii:6 –7 , Bahdi-Addu the Yabasˆ u, possibly the same higher ranking leader mentioned as ˘a military commander in A.486 + M.5319:43–50 (Villard 1992, 146–7). A man named Abi-epuh makes a payment in both silver and animals and is identified as “the son ˘of Aˇsmad” (61 ii:9 –10 ). This recalls the pattern with Ripi-Addu, son of Ibalan, who has a tribal affiliation. The entire text ARM VII 227 appears to record transactions of silver and animals involving Simalite leaders, some of whom are familiar from the sug¯agum texts of ARM XXIV, but it is not clear what relation the text has to the sug¯ag¯utum financial obligation. ARM XXVII 93:6–11; 6 u4 -um .tup-p´ı an-né-em a-na .se-er b[e-l]´ı-[i]a 7 u-ˇ ´ sa-bi-lam Aˇs-ma-[a]d 8 wa-ar-ku-u-um ´ ˇsa lu´ me-er-h[i-i]m 9 qa-du-um 15 lu´ su-ga-gimeˇs ˇs[a] l u´ ˘ki 11 i-ru-ba-[am]. See also ARM XXVII [ H]a-[n]ame [ˇs ] 10 a-na Qa-at.-t.u-na-an ´ ˘ 94:8–11, with a similar visit by Ibal-pi-el, “along with the sug¯ag¯u of the Hana” ´ s ˘ (qa-du-um lu.meˇ su-ga-g[i] ˇsa lu´ Ha-nameˇs ). A.1610+:25–27; 25 lu´ su-ga-gemeˇ˘s Ì[R]meˇs be-l´ı-ia 26 aˇs-ˇsum .sa-bi-im .ta` -ra-di-im 27 aˇs-taal-ma .sa-ba-am u-ul ´ at.-ru-ud (Durand 1988, 109–10). See also A.2119:29 (Charpin 1992b, 98), another letter from Ibal-pi-el, where Hana sug¯ag¯u send two rep˘ their eagerness to fight resentatives to Zimri-Lim to bear the good news of Eˇsnunna; also M.9175, sug¯ag¯u traveling in company with Ibal-el (Charpin 1990a, 76–7).


255

88. FM III 20:6–11; 6 i[ˇs-t]u Na-hu-urki u-s ´ . e-em-ma 7 i-na Na-ap-t.a` -ri-imki it-ti I-ba-al´ s AN 8 an-na-me-er m I-ba-al-AN u`˘ lu.meˇ su-ga-gi 9 u-pahi-ir-ma ta-aD-zi-im-ti 10 ˇsa iˇs-tu ´ ˘ u` MUNx [ma-a]h-ri-ˇsu-nu pa-na a be-l´ı-ia a´ ˇs-ta-na-pa-ra-am 11 aˇs-ˇsum ˇse-em ˘ aˇs-ku-un. 7 8 89. ARM XXVI 41:7 –10 ; ˇsa a-na li-ib-bi na-we-em a-na e-te9 -q´ı-im pa-nu-ˇsu ˇsa-ak-nu 9 ´ s li-ik-s[u]-nim-ma a-na pa-an lu.meˇ su-ga-gi ˇsa Ha-nameˇs 10 a-na Ma-ri ki li-pa-ah-hi´ ˘ ˘˘ ru-ˇsu-nu-[ti]. This focus on the steppe recalls the Yaminite “sug¯ag¯u of the Hana” ˘ in ARM II 53, who have lost their towns. 90. ARM I 13:8–11, the sug¯ag¯u are arresting deserters and sending them to the ´ s commander above them; see LAPO 17, p. 27 for collation of line 8, lu.meˇ su-ga-gi ˇsa [ha-n]a(?). In ARM I 128:5, 10, the direct connection is explicit, where once ˘ the issue is management of hana forces who have taken leave from the again ´ s assembled army. The sug¯ag¯u are to ˘help account for the lu.meˇ ha-ni-i pa-t.e4 -ri. ˘ 91. One striking example of Simalites settled in a town who maintain their identification with a tribal gayum is found in ARM VIII 85 + A.4304, the text that provides the starting point for Charpin’s study of the town of Sapiratum (1997b, 343–4). The corporate leadership that is legally responsible before king ZimriLim is identified both as “the town of Sapiratum” in the current transaction (lines 50 and 53) and as a gathering of thirty-seven “heads of the Yum(a)hammû (gayum)” in a prior commitment (lines 54–5). These are not the leaders˘ of the gayum, however, but the leaders of one town that belongs to the Yum(a)hammû ˘ tribe. 92. For Meptûm and Buqaqum, see Joannès 1996, 334. 93. One possible scenario would be that the gayums once represented active political entities like the five Yaminite tribes, each of which had its own ruler. 94. The historical development of the gayum should somehow be understood in light of the strongly political character of tribal identifications generally. A. M. Khazanov remarks that a tribe is never a purely territorial unit: “First and foremost it is a political organization” (1984, 151). 95. ARM IV 1:5–17: 5 m Sa-ku-ra-nu 6 m Ma-na-ta-nu 7 i-na Ha-ar-ra-dimki wa-aˇs-bu 8 m Kài-la-lum 9 m Za-zu-nu-um 10 m Da-di-ia 11 i-na A-ma-timki ˘wa-aˇs-bu 12 m Ha-ti-ku lu´ su-ga˘ i-zi-bu-m[a] ´ meˇs an-nu-tum 15 ga-a-ˇsu-nu ag-ˇsu-nu 13 ga-u-um Yu-ma-ha-mu-[um] 14 LU ´ 16 ˘ a-na .se-ri-ia 17 it-ta-al-ku-nim. ´ ga-yi Ia-ba-si 1 PN5 1 96. ARM XXIV 235:1–9: 1 PN1 1 PN2 1 PN3 [1 PN4 ] 4 LU ´ ga-yi A-mu-ru 6 LU ´ ha-ni Ia-ba-sa. The sum of six men certainly PN6 2 LU seems to encompass the two ˘groups of both gayums. 97. Durand forthcoming, in the proceedings of RAI 46. I would like to express my deep appreciation for M. Durand’s generosity in providing me a copy of his manuscript during the Rencontre. 98. Villard 1992, 146–7, lines 51–2. 99. Talon transliterates lu´ a-x-ru ga-yu, but the reading of the tribal name is clear in his copy. The copy for the preceding number looks like “19,” too many for a total of 24, when added to the six Yabasu. 100. The text is A.4280, to be published in Charpin (forthcoming c); a draft was generously given to me by the author. For the Yabasu and the Aˇsarugayum, see, esp., lines 14–17. 101. For the count of Yaminite soldiers mustered by town, see ARM XXIII 428 and 429.

256


102. Durand now prefers to derive the name Aˇsarugayum from a root ˇs-r-g, with uncertain meaning. The -gayum element led him earlier to suggest “the other clans”; see Villard 1992, 142–3 and note g. Notice that in both A.486+ and ARM XXIV 235, Aˇsarugayum is the second grouping, and this is the name not shared with any gayum. 103. This was true of both Yahdun-Lim and Zimri-Lim. 104. See the edition in Frayne˘ 1990, 602 (E4.6.8.1:15–20). 105. A.2560:6–10, translated by Durand as LAPO 17, no. 731; 6 ha-nu-u´ ˇsa i-bi-ru 7 ˘ 9 .te4 -em-ˇsu-nu DUMUmeˇs Si-ma-a-al DUMUmeˇs Ia-mi-in 8 ga-a-ˇsu-nu ma-an-nu-um ga-am-ra-am 10 a-na .se-ri-ia ˇsu-up-ra-am. The text was first published in Charpin and Durand 1986, 180–2. See also Durand 1990b, 62; Charpin 1987c. 106. No known etymology offers any help with the basic concept underlying the term “gayum.” 107. There are many different expressions of this consistent pattern in the use of the word hals.um in the Mari archives. Outside the normative use as “districts” of the Ah ˘Purattim in Zimri-Lim’s Mari-based kingdom, the hals.um is still in˘ into a larger m¯atum. For example, the hals.u¯ nadˆ ˘utum (“outlying corporated ˘ districts”) in ARM XXVI 373:23–5 are defined by their relation to the land of Eˇsnunna. When the word “hals.um” is applied to states outside that of the ˘ writer, it underlines their subordinate status (e.g., Numhâ and Yamutbal in ˘ ARM X 157:9; Der of the Balih in XXVI 145:21–2; etc.). Individual vassal kings ˘ volunteer the word when addressing Zimri-Lim as their overlord (e.g., Kabiya of Kahat in ARM II 59:4–6). ˘ ARM XIV 121:26–7, where the “district” (hals.um) is to be gathered 108. See, e.g., into fortified towns (dann¯atum), and the population˘ to be protected in these centers is defined as living in settlements (kapr¯atum). The kaprum need not be defined by absence of fortification; though in common use, this was probably the norm. When a clan chief fortifies a kaprum in ARM XXVI 156:5–9, passim, there is no sign that the designation ceases to apply. ` 109. A.1146:20–30: 20 a-ˇsa-ar-ˇsa-ni-ma ta-ak-la-ta um-ma at-ta-ma 21 KU.BABBAR a` ` na li-mi-ia ad-di-in 22 mi-nu-um KU.BABBAR-ka ˇsa ta-ad-di-nu KU.BABBAR-ka ˇs[a] ta-ad-di-nu 23 ka-la-ˇsu a-na-ku-ma i-di {DI} 24 [a]m-ˇsa-li li-im-ka ka-lu-ˇsu ina H[e]-enki ip-hu-ur-ma 25 [ˇs]a i-ra-am-mu-ka ki-a-am i-qa-ab-[bi] 26 um-ma ˇsuma ˇs˘u-pu-ur-ˇsum˘ li-li-i[k] 27 u` ˇsa i-ze-er-ru-ka ki-a-am i-qa-ab-[bi] 28 [um]-ma ˇsuma la i-il-la-kam 29 u` ˇsum-ma-an a-na-ku la a-ka-aˇs-ˇsa-d[am] 30 iˇs-te-et-ma-an i-te-ep-ˇsu. 110. Several examples appear in Durand’s excursus on the rihs.um assembly in ARM ˘ ARM XXVI 43:17; XXVI/1, pp. 181–92: A.954:7 (p. 183); A.3567:7 (p. 184); 45:28; 46:5, 6. 111. This terminology is surely older than the Amarna period, for which see Moran 1963, 77–87; 1992, xxiv and n. 59. 112. On the identity of Yasmah-Addu, see ARM XXIII 428:18 and 429:18, with the discussion on pp. 358–68;˘ and Marello 1992, 122–3. 113. A.3821:34–5, LAPO 17, no. 737; after Dossin 1972b, 60–1: [li-iˇs ]-pu-ur-ma li-mi la u-da-ab-ba-bu, “Let him write, so that they not bring complaints against my ´ tribe.” 114. Durand 1991, 52–3: “Le terme de ‘lˆımum,’ jusqu’ici attesté uniquement dans l’onomastique . . . désigne normalement, d’après les exemples désormais


115.

116.

117.

118.

119. 120. 121. 122.

123. 124.

125.

126.

257

repérés dans les textes de Mari, la réalité sociale représentée par la ‘Tribu’ ou la ‘Famille e´ largie.’” He also observes that the theophoric element “Lim” in personal names is never written with the LIM-sign, but only as li-im. A.2090:6–8: li-im-ka ˇsa i-na li-ib-bi Za-al-ma-q´ı-im u` ˇsa i-ta-at Za-al-pa-ahki ka-ˇsa-di ˘ Marello `ıˇs-me-ma is-su´ ha-am uˇs-ta-ap-p´ı-la-am; cited in Durand 1991, 53; and in 1992, 119. ˘ M.6060:1 –5 , in Durand 1991, 50: 1 [ . . . ] im-ha-as. [ . . . ] 2 [ . . . ]-du u` pa-da 3 4 ˘ [ . . . ] [ . . . ]-tam la na-wi-ir-tam la[ . . . ] [ . . . q]a-at na-ak-ri-ia ri-tam 5 [liqa(?)-a]t-ti u` li-mi li-ba-al-li, “[(If)] . . . strikes . . . not bright, not . . . May the hand of my enemy bring an end to(?) pasture, and may he extinguish my tribe.” The letter that records the request made by the Yaminite town of Dabiˇs for a treaty with its Simalite neighbors, under the rule of Zimri-Lim, must belong to this moment. See A.981 and the discussion by Durand (1992b, 117–20). For further treatment of this period, see Charpin and Durand 1985, 330–1; ˇ Villard, ARM XXIII, pp. 476–94. The year identified by “the throne of Samaˇ s” is Z-L 5, the sixth in Zimri-Lim’s reign. As with the gayum, no secure verbal etymology suggests a clear conceptual base, and the accumulation of cognate nouns does not illuminate the specific meaning of this old West Semitic term. KTU 1.3 II:7-8, paired with adm, “populace.” KTU 1.3 III:12, for Anat; 1.5 VI:23–4; 1.6 I:6, for Baal. See Emar (VI.3) 373:156[163] and 378:14, d Li-i-mi Sar-ta; 373:157[164] and ´ Li-i-mi Sar-ta. 378:13, d NIN.KUR ˇsa KA Reports of various meetings of “the (Simalite) Hana” as a whole almost treat ˘ one case, the merhûm Ibal-el them as independent allies, rather than subjects. In ˘ and the relates to Zimri-Lim the negotiation of an alliance between the Hana ˘ gathered leaders of Ida-Maras. , as one response to aggression by the kingdom of Eˇsnunna (A.2226, in Charpin 1993b, 182–4, based on lines 1 –2 , and the persuasive restoration in lines 10 –11 ). It is conceivable that some elements of the Simalite Hana might even oppose Zimri-Lim in favor of other local leaders, ˘ of no definite evidence. M.7630:4–6 describes opposition to a though I know Zimri-Lim force in the eastern Habur basin from a king of Kahat who brings ˘ but it is not clear whether these ˘ sympathetic with him a batallion of 500 Hana, tribesmen would have been˘ Simalite or other close kin (see Catagnoti and Bonechi 1992, 52; with collations by Guichard 1994, 258). ARM XXVI 5 and 6. See Charpin forthcoming a. For example, Kibri-Dagan, the governor of the Terqa district, announces that ˆ hû), and the Yah(r)urˆ the Uprapˆ u, the Yarihû (=Eri u have met together (ARM ˘ ˘ 17, no. 690,˘ p. 433). Representatives of the XIII 105:5–7, see Durand, LAPO Uprapˆ u, the Yah(r)urû, and the Amnanû come to Kibri-Dagan to negotiate peace (ARM III ˘50:10–17, see LAPO 17, no. 701). See M.6060:21 –23 , an oath text that Durand calls “le protocole des Bédouins,” which includes a promise to report any disloyalty either in “the mouth of the tent-dwellers of the steppe” or “in the mouth of the townsmen”(1991, 50–1). This is perhaps my principal suggestion for developing the work of Durand on the tribal populations. Durand himself has most recently emphasized the similarities between the structures of the two confederacies, so that he sees an

258

127.

128.

129. 130.

131. 132.

133.

134.

135. 136. 137.

Notes to Pages 63–65 essential unity of social framework among both the Yaminites and the Simalites; see LAPO 17, pp. 482–3, 494–8, with reference to the hibrum, the merhûm, and ˘ ˘ the sug¯agum, among other details. A.981:32–41, in Durand 1992b, 117–18 (letter of Sammetar, high official of meˇs? 33 ´ ˇ Zimri-Lim, to the king): 32 ˇsa-ni-tam m U-ra-nu-um u` lu´ SU.GI ˇsa Da-bi-iˇs ki il-li-ku-nim-ma 34 um-ma-mi iˇs-tu > .si-tim 35 i-na Ia-hu-ur-ra u-ul ´ ia-radu-um 36 ni-nu u` i-na na-we-e-em hi-ib-ra-am 37 u` ka-di u-ul ´ ni-ˇsu 38˘zu-ru-ha-tum a-na ˘ Ia-ah-ru-ur ni-nu 39 a-na li-ib-bi DUMU Si-i[m-a]-al 40 i-na Ni-ha-di-i i˘n[i-r]u-ub41 ˘ ˘ for the various ˇ ma ANSE ha-a-ri i ni-iq-t.u-ul. See Durand’s notes to the text ´ ˘ difficult phrases and terms. The noun ia-ra-du-um must be West Semitic and evidently refers to some definition of tribal movement, derived from the verb “to go down.” Durand thinks that perhaps this represents a dependent status that they hope to attain by their submission (note c). Mark Smith points out to me that the famous declaration required of worshipping Israelites in Deut 26:5 begins by identifying “my father” as an Aramean who “went down to Egypt” (verb y¯arad). The feminine noun zu-ru-ha-tum is evidently a hapax legomenon, ˘ and Durand suggests a possible cognate in the biblical Hebrew ezrah., “native.” Individual references to the hibrum are found in ARM I 119:10; VIII 11:20–1; ˘ XXVI 168:20; A.2796:16, in Durand 1990b, 288; A.2801:14, in Dossin 1972a, 118–20; also LAPO 16, no. 268, pp. 418–19; A.981:36 (here); and M.5172, text not given, cited in Charpin and Durand 1986, 154n68 and 175n158. None of these texts makes evident reference to a non-Yaminite population. I return to this text and to both terms in section F. See LAPO 16, p. 202 note b, for a proposed parallel with the Mesopotamian abu b¯ıtim. For the abu b¯ıtim and the ˇsa sikkatim as assistants to the governors themselves, see Lion 2001, 147–51. These are titles for true palace officials. See A.2801:35 (Mari district; Dossin 1972a, 118–20; LAPO 16, no. 268); ARM II 103:12–13 and III 70+:8, Terqa district; XIV 75:8, Saggaratum district. This may be the particular style of Yaqqim-Addu, governor of the Saggaratum district; see ARM XIV 64:5–6; 65:6–7. These leaders are defined collectively as “of the hals.um,” a category that is itself defined by its urban components. See also XIV˘ 8:5, “the sug¯ag¯u of the district.” See Durand, LAPO 16, p. 202 note b. In LAPO 18, p. 208, Durand suggests the particular use of the category in nonmilitary contexts for “chefs de travaux.” ARM V 3:7 and 14, with ˇsa pihrim in the second line, referring to leadership ˘ over regular troops. Another letter from the reign of Samsi-Addu commands a loyalty oath for all royal servants, with sug¯ag¯u and laputtû at the bottom of the list (A.2724:9–10; see Durand 1991, 30–2). This text also underlines their function in service of the king. ARM I 119:5–6 and V 24:20. ARM XIV 46:7; XXVI 5:14; and XXVII 107:9. Durand’s discussion of the relation between the aspects of tax and recognition in the sug¯ag¯utum is appropriate. The royal appointment, indicated by the verb ˇsak¯anum (“to place”), does reflect a substantial outside involvement, but is not attested with clan sug¯agums.


259

138. ARM XXVIII 67:23–4. At least by title, the rulers of Aˇslakkâ and Hurrâ need ˘ rulers in not have been regarded as essentially different. ARM XXIV 287 lists Ida-Maras. from some time after the year Z-L 11, including Ilulla of Hurrâ along ˘ Durand ˇ with the rulers of Sudu hhum, Ilan-s. urâ, Aˇsnakkum, and others (see ˘ ˘ 1987c, 607, for new date and readings). The unpublished letter A.49 mentions Ilullu as the ruler of Hurrâ, but without title (Wasserman 1994, 325). Durand’s idea that “kings” must˘ be anointed by their suzerain seems to apply specifically to these relationships between Zimri-Lim and the rulers of his northern vassals (LAPO 16, p. 467; cf. ARM XXVIII 147:5 –18 , for the anointing of a vassal of Qarni-Lim, king of Andarig). 139. FM VII 1:26 –30 , with discussion in Durand 2002, 6–7. ´ s 140. ARM XXVIII 120, with the lu.meˇ su-ga-gi ˇsa Lu-ha-yi-imki ; see the discussion of ˘ location on p. 170 of that volume. 141. Kupper, for instance, could not imagine the possibility that the Yaminite a¯ l¯an¯u represented real towns with a significant number of permanent residents. “On ne peut vraiment pas admettre que le pays e´ tait couvert de villes bâties par les nomades. Ces , – il faut souligner l’emploi constant du pluriel, – ne peuvent eˆ tre que les installations précaires que les non-sédentaires occupent d’une fa¸con intermettente, a` leur retour de transhumance, ou qui abritent leurs femmes et leurs enfants quand ils partent en rezzou” (1957, 13). Of the same tribal group, he comments, “Leur nomadisme est plus intense que celui des Hanéens, bien qu’eux aussi se laissent tenter parfois par la vie sédentaire” (p. 55). 142. This phenomenon has also been observed by Buccellati (1990, 99n37). 143. ARM XXVI 169:18 and 171:7–8, 18 –19 . ˇ 144. ARM XXVI 170:22 –26 ; 22 Mi-iˇs-la-anki ˇsa-lim 23 [a]-hu-ka Ia-gi-ih-d ISKUR ˇsa-lim 24 meˇs 25 ˘ ` ` ki˘ u` a-bu-ul-la-tim [ . . . ] ni-nu IR -ka ˇsa-al-ma-nu a-na ma-as. -s.a-r[a-a]t BAD 26 a-hu-um u-ul ´ na-di. ˘ 145. See ARM XXVI 282:14–15. 146. ARM XXVI 58:16–18 and 224:6–8. 147. This pattern alone suggests that the authority of tribal kings was allowed to straddle the bounds of the suzerains to whom their populations had to submit when resident in different regions. 148. ARM II 92:10–19; 10 i-nu-ma i-na Ma-riki IGI be-l´ı-ia wa-aˇs-ba-ku 11 a-wa-tam ˇse-ti ´ s i-na a-hi-ti-ia eˇs-me-ma 12 lu.meˇ su-ga-gi4 ˇsa a-la-ni ˇsa DUMUmeˇs Ia-mi-na 13 u-pahi´ ˘ ´ ir-ma ki-a-am aˇs-pu-ut.-ˇsu-nu-ti 14 um-ma a-na-ku-ma ma-an-nu-um at-ta 15 ˇsa 1 L˘U i-na a-li-ka e-li-iˇs 16 it-ta-al-la-ku-ma 17 u` la ta-s.a-ab-ba-ta-ˇsu-m[a] 18 a-na .se-ri-ia la te-re-de-[e]-ˇsu 19 ta-ma-at u-ul ´ ta-ba-lu-ut.. See also ARM II 102:19–23. 149. ARM III 6:5–13; 5 lu´ .sa-ba-am e-pé-[e]ˇs15 -tam[ˇs]a ha-[a]l-s.[´ı-i]m 6 u` DUMUmeˇs Ter˘ ıs 8 i-na lu´ .sa-bi-[i]m ˇsa a-[l]a-ni qaki a-na ˇsi-p´ı-ir ÍD.[D]A 7 ˇsa Ma-riki u-ka-am-m[i]-` ´ i[l-l]i-k[u]-nim 10 [a]-lum ˇsa ki´ ˇsa [DUMU]me [ˇs Ia-mi-na] 9 mu-ut-ta-tum u-u[l] 11 ma 50 .sa-bu-um i-si-ik-ta-ˇsu [25] .sa-ba-am e-p´ı-iˇs-tam i[d-d]i-[na]m 12 [a-lum ˇs]a 30 [s.a-bu-um i-si-ik-ta-ˇsu] 13 [15] .sa-ba-[am e-p´ı-iˇs-tam id-di-nam]. If the tablet does indeed show the traces indicated, the restorations of the last few lines should be quite secure, from the pattern with 50 and 25. The “Yaminites” are likewise restored, but this is the standard subgrouping of “towns” in the Terqa district.

260


150. See also ARM XIV 64:5, 12 (cf. 65:6–7), for the Amnanˆ u Yaminites in the Saggaratum district, viewed from similar perspective. In XXVI 450:7–8, the writer recalls that Zimri-Lim has considered sending “the sug¯ag¯u of the Yaminites” back to their towns (¯al¯an¯u). ´ s meˇs 9 ´ 151. Lines 8–14; 8 i-na-an-na lu.meˇ su-ga-gu u` NU.BAN.DA ˇsa a-la-ni ˇsa DUMUmeˇs Ia-mi-na 10 it-ti Zu-ru-ba-anki 11 Hi-ˇsa-am-taki Hi-im-ma-ra-anki 12 u` Ha-an-naki ah˘su-nu 14 [a-n]a ˘ SÍGhá ˇsa-mi-im u-ul ˘ ` hi-ˇsu-nu 13 KU.BABBAR i-si-ik-ta-ˇ ´ ˘ [i]d-[di]-nu. ˘ ˘ 152. See LAPO 16, p. 202 note c (ARM III 19:10–12; 20:18–19; 38:11–13; with other references to partial combinations). 153. The tendency to refer to the Yaminite towns en masse evidently reflects the fact that they do retain tribal bonds, as shown in the military lists, ARM XXIII 428 and 429. It is possible that the description in ARM III 70+ is intended to cover the whole Terqa district, in which case only the four separated towns are not Yaminite. The Terqa district would then be dominated by its Yaminite population. 154. ARM II 53:20–21; a-na .se-er Zi-im-ri-Li-im al-ka-ma u` a-la-ni5 -ni5 er-ˇsa. As argued more than once by Durand (1992b, 113; 1998, 449), there is no way to read the text without considering these Hana a subset of the assembled Yaminites, ˘ and specifically those who are not currently resident in towns. 155. A.2692+, in Durand 1994, 92 and n24 (no lines). For Asdi-takim as the leading king of Zalmaqum, see XXVI 24:10. There are 100 of these Yaminite leaders present. 156. Khazanov observes that the higher levels of nomadic sociopolitical organization, what he calls tribes and confederations, do not serve economic needs so much as sociopolitical ones. These include the allocation of key resources and the regularizing of migration routes, along with mutual defense and other interactions with outsiders (1984, 148–9). 157. ARM XXVI 12:7 –11 ; 7 a-di wa-aˇs-ba-tu-nu 8 2 LUGAL[meˇs ]-ˇsu-nu i-na bi-ri-ti-ˇsu ´ s nu li-il-li-ku-nim-ma 9 u` ˇsum-ma LUGALmeˇs -ˇsu-nu la il-li-ku-nim 10 lu.meˇ su-ga-gu-ˇsu11 nu i-ta-ru-u-[m]a u` a-na a-wa-ti-ku-nu tu-ra-a-ma. See also lines 4 –6 . ´ 158. Iˇshi-madar is the sole sender, but he often uses the plural “we”/“us”/“our” in ˘ body of the letter (see lines 6, 9, 15–17, and 42). His identity is not certain, the and he may be the same figure known for the previous reign of Samsi-Addu. See Lafont 1994, 212 note a; 215n24. Lafont then observes that the combination of elders and plural sug¯ag¯u effectively equates to the m¯atum of the Numhâ tribal ˘ people. 159. FM II 116:9–14; 9 um-ma a-na-ku-ma .tup-p´ı be-l´ı-ia i-na KASKAL.A im-hu-ra-anmeˇs ni 10 me-he-er .tup-p´ı be-l´ı-ia .te4 -mu-um an-nu-um su-ga-gu {x x[( . . .˘)]} 11 u` ´ ˘ 9 -ba-al li-iˇs-mu-u´ m Tu-rum-na-ak-te{ ˇsa(?) na x(?) e [( . . . )]} ˇSU.GImeˇs Mu-te 12 ´ s d ˇ wa-aˇs-bu-ut {x} Su-ba-atEn-l´ıl ki 13 a-na be-l´ı-ku-n[u k]a-an-tam u` lu.meˇ d ˇ a-na pé-te9 -e a-l`ım Su-ba-atEn-l´ıl 14 iˇs-ta-na-pa-r[u-ni]m. meˇs 24 ˇ 160. Lines 23–25: 23 [t.e4 -ma-am an-n]é-e-em li-iˇs-mu-u´ su-ga-gi u` SU.GI [a-na ki 25 ` BAD -Ia-ah-d]u-Li-im a-na pa-an be-l´ı-ia tu-ra-ni-iˇs-ˇsu-nu-ti [be-l´ı li-ib-ba]-ˇsu-nu li-ni-ih u` .te4˘-em KASKAL.A li-is.-ba-at. 161. Lines˘26–28: 26 lu´ su-ga-gumeˇs u` ma-ah-re-et [Ha-nameˇs ] 27 ik-ta-áˇs-dam .te4 -em pa-halu´ ˘ i-na ITI.1.KAM ˘ ˘ ar lu´ Ha-na[meˇs aˇs-ta-a-al] 28 id-da-an-na Ha-na i-[pa-ha-ar]. The ˘ ˘ ˘ phrase ina danna (or iddanna) means “with difficulty.”


261

162. See the analysis of Durand in Durand and Guichard 1997, 39–41. 163. Philippe Talon considered the sug¯agum to derive especially from tribal organization, but since the publication of the article by Nakata, Charpin and Villard have endorsed the idea that the position is defined by leadership of individual towns or villages. See Talon 1978, 1982; Nakata 1989; Charpin 1991a, 12; Villard 1994. 164. Along with the two articles by Talon listed in the previous note, see Young and Matthews 1977. 165. Rowton acknowledges the possibility of a tribal state, but seems to treat such kings as outside the tribal peoples they rule (1973a, 254). 166. For example, the merhûm is an intermediary between state and nomadic lead˘ ership (1973b, 212–13). 167. There is surely more evidence for the tribal orientation of early second millennium Mesopotamian kings than has been recognized to date. 168. See the texts cited earlier, with “Reading for the sug¯ag¯u of towns”: Yail and Tizrah under Samsi-Addu; Dur-Yahdun-Lim, Hiˇsamta, and T. abatum under ˘ ˘ Zimri-Lim, all within the proper bounds of each˘ kingdom. ` ki -Ia-ah-du-Li-im {x} 8 u-ul 169. ARM XIV 46:7–9; 7 s[u-g]a-gu-ut BAD ´ e-ep-pé{x}-eˇs15 9 ˘ du-pu-ra-ku ˇsa ˇsa-ka-nim li-[i]ˇs-ku-nu. 44 ´ meˇs a-na GAL.KU5 meˇs Ba-ah-di-d ISKUR ˇ 170. Lines 43–52: 43 12 LU i-si-ik a´ ˇs-ta-al˘ 45 u` ˇsa 1 li-im .sa-bi-im 10 ma um-ma a-na-ku-ma ni-bu-um 1 {1} li-im .sa-bi-im lu´ GAL.KU5 meˇs i-na-an-na at-ta ˇsa 1 li-im 2 me-tim .sa-bi-im te-si-ik 46 lu´ GAL.KU5 m [e ]ˇs te-si-ik u` 24 lu´ NU.BANDAmeˇs ˇsa 1 li-im u` 4 me-tim .sa-bi-im te-si-ik 47 wa-ar-ka-tam ´ meˇs i-na i-pa-ra-su-ma {x} 48 ki-i lu-u´ .te4 -mu-um a´ ˇs-ta-al-ma um-ma a-na-ku-ma 49 2 LU ´ lu´ GAL.KU5 meˇs ˇsa `ıs-ku a-ha-ar-ra-as.-ma 50 ˇsa-ap-tam e-li-ia uˇ ´ s-ba-la-ak-ka-at an-ni-tam ˘ Ha-na-a Ia-ba-sa u` Bi-hi-ra-am 52 e-li A-ˇsa-ru-ga-yi-im aˇse-li a-du-ur-ma 51 m Su-u-lam ´ ku-un. Villard discusses the ˘apparent numerical˘ problem in his publication of the text (1992, 146–7). Notice that I translated laputtû as “lieutenants” in this military context, rather than as “deputies,” which I used for the nonmilitary administrative context of Zimri-Lim’s districts. 171. FM II 131:4–11; 4 i-nu-ma Ba-an-nu-um me-er-hu-tam i-pu-ˇsu 5 3 1/3 MA.NA ` KU.BABBAR u` 3 ME UDUhá 6 m Ia-am-s.´ı-ha-ad-nu-˘u´ 7 [u]-[t ´ . ]à-[a]h-hi-ˇsum-ma um˘ ˘ 9 [m Ba-an-nu-um k]i-a-am ˘ ˘ [i-p]u-ulma-a-mi 8 [a-na s]u-ga-gu-tim [ˇs]u-uk-na-an-ni ˇsu 10 [a-di a-l]a-ak be-l´ı-ia 11 [lu-pu-ut]. 172. Villard notes elsewhere (2001, 20–2) that ARM V 24 has the citizens of Tizrah turn to Tarim-ˇsakim to appoint a sug¯agum, and he is one of Yasmah-Addu’s two˘ ˘ viziers, in a role somewhat like that of Bannum. In that case, however, it is a town that approaches Tarim-ˇsakim, with no reference to the merhûm title. Also ˘ note that the Bannum example involves a payment of sheep, which seems to be associated with Simalite tribal sug¯agums. 173. ARM XXVI 450:5–11, letter from Iˇshi-Dagan and Yans. ib-Addu to Zimri-Lim; 5 ˘ aˇs-ˇsum .te4 -em mu-uˇs-ta-lu-tim 6 ˇsa b[e-e]l-ni i-na Ter-qaki iˇs7 -ta-lu-ma 7 lu´ su-ga-gimeˇs meˇs meˇs 8 9 ` [ˇs]a DUMU Ia-mi-na IR -ˇsu a-na a-la-né-e tu-ur-ri-im [be]-el-ni iˇs7 -t[a]-lu 10 ´ [a-w]a-tam ˇsa-a-ti i-na a-h[i-t]im-ma 11 [ni-i]l-ma-as-s[´ı-ma]. ˘ from Ibal-el, on hearing what Hammurabi of Babylon 174. See ARM II 33:10, a letter has written to the Mari king; A.1610+:25, military leaders˘ asked to provide help for Talhayûm.

˘

262


175. For involvement of the sug¯agums in a census under Samsi-Addu, see ARM II 18:7–9, in which they are extremely concerned to meet their precise commitments, even to the point of sending a replacement for one missing man. See also ARM III 21:58 for Yaminite sug¯ag¯u supervising the census of their villages under Zimri-Lim. 176. ARM I 6:16–19; 16 LUGAL KASKAL i-la-ak ka-lu-ma 17 a-di .se-eh-ri-im li-ig-da-mi´ i-iz-zi-bu ˘a-sa-ak LUGAL ir 18 su-ga-gu-um ´ ˇsa .sa-bu-ˇsu la gu-mu-ru-ma 19 1 [L]U i-ku-ul. On the final phrase, see Charpin 1997a. 177. See, e.g., the discussion and texts in Durand, LAPO 16, pp. 332–7, 347–53. 178. For this responsibility, see also ARM VI 32:16 and 40:7. 179. ARM II 92:12–20, involving Yaminites. Compare ARM II 102:19-23, without actual mention of the sug¯ag¯u, and ARM II 103:12–13. 180. See the DU[MU]meˇs lu´ su-ga-gimeˇs in a list of leaders for the army of IˇsmeDagan during the reign of Samsi-Addu, ARM IV 74:21, cf. 29. 181. See CAD s.v. sug¯agu c. 182. Tell ar-Rimah 100:13–17; in Dalley Walker, and Hawkins 1976, 83–4. 183. The D stem ˘of the verb pah¯arum describes transitive “gathering” (summoning to assembly), and texts ˘with this form display the initiative of the royal administration (see ARM II 92:12–13; XIV 65:6–7). The G stem of this verb represents intransitive “gathering,” leaving the initiative to the people themselves. In A.2119:29–31 (Charpin 1992b, 98), the sug¯ag¯u of the Hana gather (iphur¯u, G stem). See also FM II 123:27 –28 . When “the sug¯ag¯u of˘the Suhˆ um ˘ (region)” gather in ARM IV 16:11 –13 , king Yasmah-Addu issues orders,˘ but ˘ by the same verb form, the meeting seems to be accorded some independence and one wonders whether these leaders may also be tribal. For assembly of sug¯ag¯u at their own initiative, see also ARM XXVI 150:11–12. 184. This phrase is applied to the hals.um district of Saggaratum in ARM XIV 8, 64, ˘ and 65. 185. In every part of this study, I have been selective in my presentation of evidence, sometimes offering only a sketch of available data. My treatment of the merhûm ˘ is such a sketch, made all the more incomplete by the fact that a significant part of the merhûm correspondence is still unpublished (see Durand, preface to ARM XXVIII, p.˘ v). For a review of the evidence available before the reconstitution of the Mari publication team under Durand, see Safren 1982. 186. ARM II 33:21 –22 ; 21 na-wu-um u` DUMU Si-im-a-al 22 [ˇs]a-a-lim; see also II 37:25. Instances of the governors’ greetings are countless; see Kibri-Dagan’s in ARM III 12:6; 13:6; 17:6; etc. The word nawûm can refer to the flocks themselves, to the encampments in the high country where the herdsmen are based, or to the high grazing land itself. By “high” country, I distinguish the nawûm from the lower terrain of the river valleys and do not mean hills or mountains. 187. Durand comments that the Yaminite merhûm work within a slightly differ˘ relationships and lines of auent system of several kings, where the exact thority are not clear. See also Durand’s cautious comments in 1999–2000, 191. 188. LAPO 16, pp. 630–1 note d, commenting on A.2741; cf. LAPO 17, p. 471. 189. This seems to be suggested by Buccellati’s interpretation of the merhûm as the ˘ assumes king’s “tribal commissioner” (1990, 103). Like Rowton’s, this analysis


190.

191.

192.

193.

194.

195.

196.

197. 198.

199.

263

a polarity of “urban” and “tribal” or “rural,” where power is concentrated in the urban state. ARM XXVI 389:9–15. It is striking to find here the D stem of pah¯arum, permitting someone the authority to gather the tent-dwelling “Hana,” ˘a right that ˘ even Zimri-Lim himself does not take for granted. A.1025:77–81, letter to Zimri-Lim from Yassi-Dagan, asking for troops; 77 i-naan-na be-l´ı it-ti ÌR-dimeˇs -ˇsu li-iˇs-ta-al-ma ak-ki-ma mu-uˇs-ta-lu-ti-ˇsu 78 te-re-e-tim li-ˇsepi-iˇs-ma a-na zi-im te-re-e-ti-ˇsu ˇsa-al-ma-tim 79 ˇsum-ma ˇsa a-la-ki-im be-l´ı li-il-li-kam 2 li-im Hau-la-ˇ ´ ´ su-ma 80 be-l´ı a-na .se-er lu´ me-er-hi-im li-iˇs-pu-ur-ma 1 li-im u-lu-ma ˘ ˘ na 81 lu´ me-er-hu-um pa-ni-ˇsu li-is.-ba-tam-ma a-na .se-ri-ni li-ik-ˇsu-dam-ma. See Kupper 1990, 337–9.˘ ARM XXVII 48:3 -8 ; 3 u` DUMUmeˇs NA.GADAmeˇs ik-ˇsu-du-nim-ma 4 um-ma-mi i-na bu-ra-ti[m] 5 ˇsa .sa-ab-ta-nu ni-nu ˇsa-al-ˇsa-am {RI TAM! } 6 ri-tam ni-le-qé-ma u` 2 zi-ti 7 lu´ me-er-hu-um a-na Ha-nameˇs 8 i-s.a-ba-at. ˘ ˘ A.1098:7 –8 , in Villard 1994, 297; with ˇsak¯anum, the same verb used for appointment of sug¯agums. Samsi-Addu has power to appoint a merhûm as far west ˘ as Tuttul (ARM I 62:5 –14 ). The verb wuurum (“to command”) is used by Yaqqim-Addu in ARM XIV 121:45, who wants the nawûm moved into towns in the face of a marauding army. The governors of Qat.t.unan likewise have no authority over bedouin pastoralists and must address problems to the king with the hope that the parallel chain of command will be adequate (see XXVII 61:11; 70:22–5). This professional tension seems to spill over into a larger resentment in ARM XXVII 151, a long and bitter complaint from the governor of Qat.t.unan against Ibal-pi-el that is focused not on local administration but on the merhûm’s supposed grab for inappropriate authority over non-Hana troops. All of˘this takes place at the time of the campaign against Elam, in˘ Zimri-Lim’s 11th year (Z-L 10). FM III 20:2 –10 ; 2 [a-na] I-ba-a[l-AN] u-ul ad-ni-in 3 [um-ma a-na-ku-m]a as´ 4 5 su-ur-re a-da-ni-in-ma [e-li]-ia te-ki-tam i-ra-aˇs-ˇsi [u-la]a ˇse-em u´ MUNx a-na ´ ´ Na-hu-urki 6 [u-ˇ ı-tam 7 [i-na]-an-na a-na I-ba-al-AN ´ se-re]-eb-ma aˇs-ˇsum ke-em u-la-p´ ´ 8 ˘ ı da-an-na-tim li-iˇs-ku-un-ma 9 ˇse-em u´ MUNx a-na Aˇ ´ s-la-ka-aki u´ Na-hu-urki be-l´ ˘ of li-ˇse-ri-ba-am 10 ˇse-um u´ MUNx u-ul i-ba-aˇs-ˇsi. Itur-Asdu served as governor ´ Mari before his assignment to Nahur (Guichard 1997, 191–2; cf. Lion 2001, ˘ 184–5). A.1215, cited in Guillot 1997, 276; the text will evidently appear in Durand’s eventual study of the peoples of the Mari archives. For references to the merhûm leading troops, see ARM XXVI 388:19–21; ˘ 151:48–50; A.1025:8–81 (Kupper 1990, 337– 389:10–13; XXVII 132:22–7; 9). FM II 63:6–17, letter from Hadni-ilum to Zimri-Lim; 6 i-na a-hi-ti-ia a-wa8 l u´ ´ ˘ ˇ tam ki-[a-am eˇs]-me 7 um-[m]a-a-mi I-ba-a[l-d ISKUR] [ ] Aˇs-la-ka-aki˘ 9 u` Ha-amlu´ 10 ki 11 ˘ ki ] ˇ ma-an su-ga-gu-u[m] ˇsa De-er pa-ni Sa-du-um-la-ba 12 lu´ Aˇs-na-ak-ki-im[ 13 is.-ba-tu-nim-ma a-na .se-er I-ba-al-A[N] 14 lu´ me-er-`ı(HI)-im a-na Si-ha-ra-ta-aki 15 [t]a(?)-wa-i-{x x}im 16 a-na pu-hu-ur lu´ Ha-nameˇs 17 u-ˇ ´ ˘sa-ak-ˇsi-du-ni-iˇs˘-ˇsu. Durand ˘ ˘ provides hand copy for the difficult line 15, and the text is broken. He tentatively suggests an ethnic identification of Siharatâ (“of the Tawˆ um”), though such a ˘ group is otherwise unknown.

264


200. For sug¯agums under the merhûm, see ARM XXVI 508:7–8; XXVII 93:7–12; 94:8– 11; M.9175, Charpin 1990a,˘ 76–7. The unpublished texts cited by Durand for the same personal following in LAPO 17, pp. 471–2, should sustain the same interpretation (A.333 and A.4332). 201. See Dominique Charpin and Nele Ziegler in FM V, forthcoming. 202. See the previous discussion of the gayum, with A.486 + M.5319. 203. The book by Anbar (1991) was written without much of the new evidence published in the late 1980s, and its general conclusions must be read with this in mind. It contains much useful analysis, however, including detailed information for tracking the location of settlements associated with each tribal group. For more up-to-date evaluation of the geography of Yaminite settlements in Mari’s core districts, see Millet-Albà forthcoming. 204. ARM XXVII 70:17–29; 17 u` I-ba-ás-si-ir 18 il-li-ik-ma i-na 1 ha-s.´ı-ri-im 19 6 UDUhá ˘ ´ S.GAL ˇ i-na ˇsa-ni-im ha-s.´ı-ri-im 20 2 UDUhá i-na ˇsa-al-ˇsi-im ha-s.´ı-ri-im˘ 21 1 MA is.-ba˘ 22 23 ˘ ˘ at i-na-an-na I-ba-al-p´ı-AN me-er-hu-um ma-ha-ar be-l´ı-ia wa-ˇsi-ib I-ba-al-p´ı-AN 24 ˘ LU ˘ há ˇsi-na-ti 26 u-ˇ ´ meˇs be-el ha-s.´ı-ri-im 25 li-se-ni-iq-ma ´ ˇsa UDU ´ meˇs LU ´ sa-at-bi-lu LU ˘ 29 27 28 meˇs ˘ ´ be-el ha-s.´ı-ri-im u-ka-an-nu LU ˇsu-nu i-na qa-ti-ia u-ul ´ ´ i-il-la-ku. The man ˘ is known to belong to the (Simalite) Hana, according to ARM XXVII Ibassir ˘ 69:12–13, 18. 6 lu´ 205. A.1098:6 –15 ; u` ˇsum-[ma ]Ha-na a-na me-er-hi-i]m ˇsa-ni-im-ma ˇsa-ka-nim 7 ˘ ˘ me-er-hu-ni 8 i-na A-ah Pu-raik-ta-ab-ta-ku-um um-ma-m[i iˇs-t]u-ma Ba-an-nu-um at-tim wa-ˇsi-ib [me-er]-he-em ˇsa-né-em-ma ni-ˇsa-ka-an 9 at-t[a˘ k]e-em a-pu-ul-ˇs˘u-nu-ti ˘ um-ma at-ta-ma pa-na-nu-um i-na na-wi-im u-ˇ ´ si-ib-ma 10 iˇs-di [DUM]U Si-im-a-al ki 11 Nu-um-ha-a Ia-mu-ut-ba-al ˇsa-ki-in-ma a-na A-ah Pu-ra-at-tim it-ta-al-kam-ma 12 ki 13 ˘ da-[an]-na-tim i-na-an-na u-ˇ ´ ´ se-ep-ti-ma iˇs-di-ku-nu i-na A-ah ˘Pu-ra-tim u-ki-in ˘ A-ah Pu-ra-at-tim a-na da-an´ ˇsa-a-ti i-na aˇs-ˇsum a-na-ku an-ni-i[ˇs] al-li-ka-am 14 LU ˘ me-er-hi-ku-nu a-t.a` -rana-ti[m] ku-ul-lim u-zi-ba-aˇ s-ˇsu 15 i-na-an-na ki-ma ka-ˇsa-di-ia ´ ˘ da-ku-nu-ˇsi-im an-ni-tam a-pu-ul-ˇsu-nu-ti. For the crucial new interpretation of line 12 as a reference to Bannum’s conquest of Mari, based on a literal reading of the verb uˇsepti as “he caused to open,” see Charpin forthcoming a. 206. See Durand’s discussion in AEM I/1, p. 74. 207. FM II 117. 208. ARM XXVIII 167:3; 168:3; 169:4; 170:4. On the diplomatic language of “brothers,” “father/son,” and “master/servant,” see Lafont 2001a, 232–8. 209. ARM XXVIII 164:4; 165:4; and 163:3, in that order. See Kupper’s discussion of the historical problems (pp. 235–6). Bunu-Eˇstar appears also to have ruled Kurdâ during the reign of Samsi-Addu, only to have been deposed by that king. See A.1215, cited by Durand’s translation in Guillot 1997, 276. 210. ARM XXVIII 100:26, broken text. 211. A.1215, in Guillot 1997, 276. 212. ARM XXVI 114:3–9; 3 m D[a-d]i-ha-ad-nu il-li-kam-ma 4 [ge-er-r]a-am an-ni-a-am 5 ˘ e-pu-uˇ [a-na m]e-er-hi-ˇsu ip-q´ı-id 6 [te-re-t]im ı-it-ma ´ s-ma 7 pu-ut lu´ me-er-hi-i 8 u-la-ap-p´ ´ 9 ˘ ˘ te-re-tu-ia ˇsa-al-ma. Dadi-hadnu is an alternate form of the name Dadi-hadun, ˘ a man known to be the ruler of the Rabbˆ u tribe of the Yaminites; see˘ Denis Soubeyran, ARM XXIII, p. 360; Durand 1990b, 46, 48. ´ s ´ 12 i-na Tu-tu-ul ki 13 ma-ha-ar be-l´ı-ia 213. ARM XXVI 86:11–21; 11 lu.meˇ me-er-hu-u´ {U} 14 ˘ 16 a-na sa-ak-bi-ˇsu-nu 17 a-ah-ˇsu-nu ˘ la i-[n] be-l´ı dan-na-tim 15 li-wa-e-er-ˇsu-nu-ti-ma

˘


214. 215.

216. 217. 218.

219. 220.

221. 222.

223. 224.

225.

226.

227.

265

a-ad-du-u´ 18 UDUhá -ˇsu-n[u] su-puha 19 iˇs-tu e-le-nu Ha-la-bi-[i]t 20 a-di Su-ur-ma´ ˘ ˘ ˘ anki 21 i-re-u´ u` a-ah-ˇsu-nu na-di. ˘ The Suhˆ um represents a special case and stands somewhat outside of Zimri˘ Lim’s administrative system on both counts. ARM I 62:5 –13 ; 5 aˇs-ˇsum Ha-ab-[du-ma-d Da-gan DUMU] 6 m A-ia-la-su-mu-u´ a-na m[e-er-hu-tim] 7 ˇsa-ka-nim˘ ta-aˇs-pu-ra-am ki-ma ˇsa t[a-aˇs-pu-ra-am] 8 m Ha-abd 9 ˘ du-ma- Da-gan-ma a-na me-er-hu-t[i]m ˇsa-ka-nim i-re-ed-du mi-nu-um ˇsa-p´ı-t.u-us-s ´ ˘ u´ 10 11 ki ˘ tu-ˇsa-ma ma-tam ra-pa-áˇs-tam i-ˇsa-ap-pa-a[r] u` Tu-ut-tu-ul li-iˇs-pu-ur u` me ` s er-hu-tam li-pu-uˇ tap-pu-ˇsu ma-tam ra-pa-áˇs-tam i-ˇsa-ap-pa-ru 13 u´ ´ s 12 u` ki-ma lu.meˇ ˘ u´ T ]u-ut-tu-ul ki li-iˇs-pu-ur. ˇs[uSee, e.g., Kupper 1957, 21. Naturally, these “Haneans” were understood to come from an independent tribe. A.361 iii:13 –15 , cf. ii:3 ; in Charpin 1991b, 141–4. Charpin observes that the first two units correspond to the division in Zimri-Lim’s titulary (pp. 146–7). A.2730:8–10, in AEM I/2, p. 33; I discussed this text already with “heads” as collective leadership. For the same two-part division of Zimri-Lim’s armies, see also ARM II 25:13 and 10 (evidently a letter from Ibal-el). A similar twopart structure already existed under Yahdun-Lim; see Charpin forthcoming c. ´ s ˘ The two groups are counted as “tent-dwellers” (lu.meˇ ha-na) and “townsmen” meˇs ˘ ´ (LU ˇsa a-la-ni); see A.4280:32–33, etc. See A.486 + M. 5319, in Villard 1992, 146–7. Guillot 1997, 284, states that the troop in question is composed of Simalites only, bypassing both the umm¯an¯at b¯el¯ıya and the issue of how these Hana relate ˘ to the larger military classification. ARM VI 28:7–8, see LAPO 17, p. 195. Ibal-el, FM III 20:3–4, and the body of the letter that discusses the merhûm; Ibalpi-el, often, e.g., ARM II 27:5; XXVII 132:22–5; 151:53–5; A.486+:18˘ (Villard 1992, 138). A.1610+:19–20 (=A.1212:18–19), in Durand 1988, 109–10. The letter confirms the connection with the Hana in lines 11, 43, and 47. ˘ omens for people are generally taken not for See ARM XXVI 141:17; notice that vague population categories but in terms of concrete political units, the natural interest of the kingdom. Most often, these are towns or districts (hals.um). ˘ 14:9–12, See ARM XXVI 24:23–5, Numhâ; 358:6 , Numhâ as “brothers”; XXVII ˘ ˘ 2,000 Hana, along with 5,000 Numhâ and Yamutbal; compare FM II 117:13–14, ˘ ˘ hâ. for alliance of the Simal and the Num The Hana are evidently the other party˘in A.2226:1 –2 , and Charpin is surely ˘ to restore lines 10 –11 with the same term: s[a-li]-mu-um u` damcorrect qa-tum bi-r[i-it Ha-nameˇs ] u` I-da-Ma-ra-as.ki it-ta-aˇs-k[a-an]. The alliance between Ida-Maras. and ˘a group called the Hana is also known from other texts: ARM II 37:6–8 (also a letter of Ibal-el), “I˘ have established peace between the Hana and Ida-Maras. ”; A.2119:25–6 and M.9623:11–13, in Charpin 1992b, 98;˘ and 1990a, 79, fighting together against Eˇsnunna, and at peace; A.2326:8–10 and M.7421:4–7, 5 –7 , in Charpin 1993b, 175, king of Aˇsnakkum says always united, and Hana should be allowed to enter Ida-Maras. ; ARM XXVII 20:19–21, four ˘ . kings have met with the Hana before setting out to see Zimri-Lim. Ida-Maras ki 6 ARM XXVII 17:5–10; 5 Ha-nameˇs ˇsa˘ i-t[a-a]t [Q]a-at.-t.u-n[a]-an 2 ME .sa-bu´ ummeˇs i[p-hu-ru-nim-m]a 7 ˘m Ia-[t]ar-L[i]-im a-li-ik pa-ni-ˇsu-nu 8 a-na .sa-ba-i-[i]m

˘

266

228.

229.

230.

231.

232. 233. 234.

235.

236.

237.

238.

Notes to Pages 87–89 a-na na-wi-im 9 ˇsa DUMUmeˇs Ia-[mi-n]a il5 -li-ku 10 u` aq-bi-ˇsu-nu-ˇsi-[i]m um-ma a-naku-ma. The Simalite tent-dwellers can undertake this only when there is no clear contradiction of interests, but in this case the action is by their own initiative. The situation is complicated when they succeed and capture four a¯ hiz¯u (Birot prefers as informers, pp. 63–4), one of whom the governor suggests˘should be sent to the king (lines 26–8). The Hana refuse, typical of their independence. 19 lu´ ˇ˘ ARM XXVII 57:18–33; 18 Ia-si-im-d ISKUR Ha-nu-u´ 20 NINDA u` KASˇ im-ˇsu-u5 ˘ a-na-ku-ma ÌR e´-ká[l]-lim 24 am21 22 23 u` 1 ÌR e´-kál-lim a-di na-p´ı-iˇs-tim i-du-uk um-ma 25 m[i]-nim t[a-du-u]k u` ˇs[u]-u´ ki-a-a[m i]-pu-[la]-an-ni 26 um-ma-a-mi L[UG]AL [b]e-el-ka 27 u` Sa-am-me-e-[tar] 28 u-ul ´ u-s` ´ a-na-qa-an-ni 29 u` at-ta ta-sà-an-na-qa-an30 giˇs ˇ u` ÌRmeˇs ni u` TUKUL ZABAR a-na pa-ni-ia 31 it-ta-na-aˇs-ˇsi IGI Ri-ip-i-d ISKUR ´ ˇsa-a-t[u] a-na ne-pa-ri-im 33 u-ˇ be-l´ı-ia 32 i-na-an-na LU ´ se-ri-ib-ˇsu be-l´ı lu-u´ i-de. 4 a-na Ka-ab-ka-ab akM.7630:3–6: 3 [k]i-ma wu-u-ur-tim ´ ˇsa be-l´ı u-wa-i-ra-an-ni ´ m 5 meˇs ma-ah-ri-tu-ˇsu 6 i-na ˇsu-dam-ma A-ki-in-a-mar [i]ˇs-me-ma 5 ME Ha-na ˘ ˘ da-an-na u-s and Bonechi (1992), ´ . e´-e-em. The text was published by Catagnoti with collations by Guichard (1994, 258). On Yassi-Dagan, see also Kupper (1990). ARM XXVI 27:3–8; 3 la-a-ma ka-ˇsa-ad .tup-p´ı be-l´ı-ia 4 ˇsa aˇs-ˇsum pa-ha-ar Ha-nameˇs ˘ .se-er i-na gu-ub-bu-ri-ma 5 aˇs (erased line) 6 aˇs-ta-al-ma I-ba-al-AN lu´ Ha-né˘-em 7 a-na ˘ Ha-nameˇs u` Ha-li-ha-du-un 8 aˇs-pu-ur. ˘ ˘in Durand, ˘ A.3567:6–9; AEM I/1, p. 184; 6 [lu´ H]a-nameˇs ku-ˇsu-ud-ma ri-ih-s.a-am 7 meˇs ˘ s di-pa-ri-im ki-ma 1 LU ´ ˘Ha-na [ri-h]i-[i]s.-ma Ha-na li-ip-hu-ur 8 [a-na ni-i]ˇ ˘ a4 -ri-ra-am.˘ ˘ ka-lu-ˇ˘su-nu 9 [li]-in´ s ARM XXVI 42:5–9; 5 lu.meˇ mu-ba-ás-si-ru 6 ˇsa iˇs-tu ma-ha-ar Ha-na meˇs 7 ik-ˇsu-du-nim 8 ˘ ˘ ma-ah-ri-ia 9 i-bi-DU. ˘ A.2210:5–7, in Durand 1993b, 54 and n77. ARM II 48:5–9, letter to Zimri-Lim from Bahdi-Lim, governor of the Mari ˘ meˇs u-qa-a district; 5 iˇs-[t]u U4 .5.KAM i-na ha-da-nim 6 Ha-na u` .sa-bu-um 7 u-ul ´ ´ 8 meˇs 9 ˘ ˘ i-pa-ah-hu-ra-am Ha-na iˇs-tu na-we-em ik-ˇsu-da-am-ma u` i-na li-ib-bi a-la-ni-ma ˘˘ ˘ wa-aˇs-bu. See ARM I 37:39–41 (at Haˇsˇsum); I 42:4–11, etc. (names written on tablet, line ˘ listed only by tribes (not called Hana), unlike “the 8), and note the Yaminites Simalite Hana” (lines 28–31); I 82:13–15 (at Qat.t.unan); ˘I 87:4–6 (same?); II 1:10, 24˘ (near Gaˇsˇsum); IV 57:8–12 (Iˇsme-Dagan, his region); V 51:5–18 (upper Ida-Maras. ). ARM V 51:5–19, letter of Iˇsar-Lim to king Yasmah-Addu of Mari; 5 aˇs-ˇsum lu.meˇ ´ s ˘ wa-aˇs-bu 8 a-na LUGAL Ha-ni-i 6 ˇsa i-na ha-la-as. I-da-Ma-ra-as.ki 7 e-li-i-im 9 ˘ ˘ a-nu-um-ma 10 .tup-p´ı LU ´ u` ˇsum-ˇsu 11 ˇsa Ha-ni-i-im 12 ta-aˇs-pu-ra-am i-na-an-na ki ˘ ˇsa i-na ha-la-as. Na-hu-ur ki 13 Ta-al-ha-yi-i-imki 14 Q´ı-ir-da-ha-at ki 15 u` Aˇs-na-ak-ki-im 16 17 18 ˘ ˘ ˘ ˘ wa-aˇs-bu ˇsa ha-a[l-s.]´ı-[i]m ˇsa u-[u]b-bi-[bu] u-ˇ ´ ´ sa-t.e4 -r[a-am-ma] a-na .se-er ˘ [be-l´ı-ia] 19 u-ˇ ´ sa-b[i-lam]. Anbar (1991, 183) considers ARM VI 28 to mention a “census” of the Hana, but this text is concerned rather with assignment of provisions, using the˘ verb paq¯adum for the administrative procedure (lines 6, 23). See also ARM II 12:5–6, 28 (LAPO 16, no. 432); and A.2741 (no. 433), both letters from a man named Atamrel (from Atamar-el?; Durand, LAPO 16, p. 628); and ARM XXVIII 25:8, 11, 14, a letter sent by a Yaminite leader named Yahdun˘ Lim. Atamrel was a Yaminite leader from the Uprapˆ u tribe, known elsewhere


239.

240.

241. 242.

243. 244.

245.

246.

247. 248.

267

in ARM XXVI 24:8 and M.6874:17 (AEM I/1, p. 181). ARM II 14 is identified in LAPO 17 (p. 433) as belonging to the reign of Zimri-Lim, and uses hana ˘ alone for Yaminite bedouin who are being invited to join the chiefs Samsi-Addu and Yaggih-Addu at Imar (if the collation is correct, lines 7–10). According to ˘ Durand’s reading, all of this information comes from an individual hana who ˘ to unis to pass on his report directly to the addressee, whom Durand seems derstand as Zimri-Lim. If these Yaminite chiefs really do summon their mobile population as “hana” without further definition, I wonder whether the letter ˘ may rather be Yaminite itself, sent by and to members of this tribe, as Durand proposed for ARM II 12 and A.2741. A.505, letter without address, only lines 21 –24 and 49–51 cited by Durand 1992b, 114n146. 21 u` iˇs-tu ˇsa-ad-d[a]-ag-di iˇs-tu ˇsa Ak-ka-du-u´ 22 i-le-e-em ha-nameˇs ˘ DUMU Si-ma-al ki 23 i-[n]a li-ib-bi ma-a-at Ia-mu-ut-ba-limki i-re-i 24 g[u-u]l-lu-ul-tum 49 giˇs ˇ ki giˇs ˇ SUKUR Ia-mu-ut-ba-lim u` SUKUR ha-nameˇs u` hi-t.`ı-tum u-ul ib-ba-aˇs-ˇsi . . . ´ ˘ DUMU Si-ma-al 50 pu-hu-ur in-né-mi-id-ma pa-ri-ku-um a-na pa-ni-ku-nu 51˘ u-ul ´ ib˘ puhhur as a D stative. ba-aˇs-ˇsi. In line 50, I read ˘ ˘ 114. The letter comes from a man named A.3572:1 –4 , in Durand 1992b, Hittipanum, evidently found also in ARM XXVI 404:4, where he is a servant ˘ the king Atamrum of Andarig. 1 [u] of ` dam-qa-tim a-n[a bi-ri-ti-ni(?)] 2 i ni ki 3 iˇs-ku-un [Ia-mu-ut-ba-lum (?)] u` DUMU Si-ma-al iˇs-tu da-a[r . . . (?) . . . ] 4 athu-u´ u` pu-ur-s` ´ a-at ha-nameˇs . The restoration of “Yamutbal” reflects the origin in ˘ ˘ of a Yamutbal kingdom. The meaning of the noun purs¯atAndarig, the capital is uncertain, but has to do with some division. Notice the spelling Si-ma-al in both this text and A.505, possibly reflecting a shared origin in Andarig. Note that the Simalites are also called hana in A.3572:7 . ˘ to his son, Yasmah-Addu, listing troops; ARM I 42:28, letter from Samsi-Addu meˇs ˘ 1 li-im .sa-b[a]-am i-na ha-na [Si-i]m-a-lu-ni. ˘ ˇ ˇ A.2560:4–10 (cf. lines 2 , 7 ), letter from Sˆın-tiri, governor of SubatSamaˇ s in the Balih River region, during the Samsi-Addu period (in Charpin and Durand ˘ 4 [aˇs-ˇsum ha-n]i-i ˇsa a-na e-bi-ir-tim 5 i-[bi-r]u ˇsa be-l´ı iˇs-pu-ra-am 6 1986, 180); ˘ 7 DUMUmeˇs Si-ma-a-al DUMUmeˇs Ia-mi-in 8 ga-a-ˇsu-nu um-ma-mi ha-nu-u´ ˇsa i-bi-ru 9 ˘ ma-an-nu-um .te4 -em-ˇsu-nu ga-am-ra-am 10 a-na .se-ri-ia ˇsu-up-ra-am. The king sends an order to gather the Yaminite hana (M.8512:5–6, in AEM I/1, ˘ p. 436). See Charpin (1987b), on the land of Apum. FM II 116 recounts the unfolding ˇ of a plan to raid Subat-Enlil, right after Samsi-Addu’s death, with local support (Eidem 1994). 24 URUki ˇsu-u´ lu-u´ a-li 25 u` i[ˇs-tu a]q-da-mi a-n[a] HA.LA [I ]a8 -[b]a-si-imki lu-u´ ˘ su . . . 29 URUki ˇsu-u´ lu-u´ ˇsa [n]a-di-in 26 lu´ A-pa-a-yu[k ]i a-na q´ı-iˇs7 -tim la id-di-nu-ˇ 30 ˇ Su-ub-r[a-am] [u` i]ˇs-tu aq-d[a]-m[i] a-n[a] HA.LA Ha-naki lu-u´ na-di-in. ˘ pp. 127–8. ˘ ˇ ˇ For recent discussion of Subram, see ARM XXVIII, Subram appears in such varied settings that there is some question about how many individual rulers bore the name. 60 ´ meˇs ˇsu-nu DUMUmeˇs Ha-naki 61 u´ -ul DUMUmeˇs ma-t[i]-ˇsu. LU For references to the m¯at ˘Apim from those in service to Zimri-Lim, see ARM II 49:4 –6 ; X 122+:10–11; XIV 102:19; 125:16; A.1421:43 and M.15083 in Charpin 1987b, 135n35 and 136n38. ARM XXVI 358 describes Hana who may oppose Zimri-Lim in the Sinjar, home of the Numhâ tribe, where˘ the Hana are

˘

˘

268

249.

250.

251. 252.

253.

254. 255. 256.

257.

258.

259. 260.

Notes to Pages 91–95 distinguished from the writer’s own land of Apum. Charpin concludes that the writer, Yanuh-samar, was a leader of Apum (AEM I/2, p. 130). ˘ FM II 116:37–43, letter from Sumu-hadû to Zimri-Lim; 37 aˇs-ˇsum .te4 -em ma-tiim ˇsa-ha-t.´ı-im ˇsa be-l[´ı ] 38 a-na ha-nameˇ˘s DUMUmeˇs Si-im-a-al iˇs-pu--ru 39 ˘ s li-ik-ˇsu-dam-ma .te4 -mu-um ˘ ˇsu-u´ li-[x x x]x 40 ul-la-nu-um{x} ha-na ha-nameˇ ˘ ˘ meˇs DUMU Si-im-a-al a-na pa-an be-l´ı-ia 41 lu pa-hi-ir u` an-nu-um ha-na DUMUmeˇs Ia-mi-na li-ip-h[u-ur] 42 u` p´ı-ih-rum˘ DUMU Ia-mi-na ˇsa ki-ˇs˘a-ad Pu43 ra-tim ma-li u-za-ku be-l´ı u-ˇ ´ ´ ˘sa-aˇs(RUM)-ma.˘ The pihrum form the regular army of conscripted troops. Eidem translates the verb at ˘the end as a regular G stem form, “(the pihrum force) will obey,” but this appears to be a causative Sˇ stem ˘ verb. Aside from the question of how the context of the communication shaped the broader use, Sumu-hadû himself was probably not a Simalite. Like Asqudum, Sumu-hadˆ u entered˘ Zimri-Lim’s service after prior employment by Yasmah˘ ˘ Addu (Durand 1990c, 288, on A.2796). A.1086, in Durand, AEM I/1, p. 178 note f; originally in Dossin 1939, 989. ARM XXVI 282:14–25; 14 iˇs-tu be-l´ı a-lam Mi-iˇs-la-anki 15 is.-ba-tu a-ia-{x}-ˇsi 16 mas.u-um-ma ma-s.´ı 17 ˇsa-ni-tam i-na pa-ni-tim 18 [a]-na be-l´ı-ia ki-a-am aˇs-pu-r[a-a]m ´ 19 20 [u]m-ma a-na-ku-ma 2 lu´ ha-ni-i ba-al-t.u-s a-na pa-t.i4 li-ir-du-nim-ma 21 ´ u-nu ´ 22 ˘ i-na pa-t.i4 li-ka-s.´ı-s.u-ˇ u-nu a-na DUM[Um ]eˇs Ia-mi-na 23 li-il-li´ ´ ´ su-nu-ti ba-al-t.u-[s] ku-ma ki-ma be-l´ı 24 a-lam5 (LIM) Mi-iˇs-la-anki i-na e-mu-q´ı-im 25 [is.-ba-t]u li-id-bu-bu. The western focus is suggested not only by the natural center of Yaminite power after defeat closer to Mari, but also by mention of Tuttul in line 28, in broken context. The letter FM II 71 appears to have been sent by a governor of Qat.t.unan early in Zimri-Lim’s reign. The unknown writer mentions Yaminites who have been held at Qat.t.unan for service to the local palace (lines 11–14). Some hana are ˘ likely then said to be “taken” (s.ab¯atum) for similar palace service, a fate more for Yaminites than Simalites under the rule of Zimri-Lim (lines 18–20). This Yaminite population was the subject of a Paris doctoral dissertation by Adélina Millet-Albà, which is summarized in her forthcoming article. See p. 29, with the text cited in n. 7. See Anbar 1991, 90, cited in Chapter 3 (p. 158), where this phenomenon is explored in the context of a letter that calls Zimri-Lim king of both the Hana ˘ and the Akkadians. It would be interesting to discover whether the percentage is higher for Yaminites from the Ah Purattim than from the Balih or the far ˘ ˘ west. It would also be worth knowing whether the percentage of Akkadian names is different for Simalite Hana who are located in the vicinity of Mari, compared with those in the upper˘ Habur. ARM I 6:6–13, LAPO 17, ˘no. 641, p. 342; 6 aˇs-ˇsum DUMUmeˇs Ia-mi-in 7 ta-aˇs-pura-am DUMUmeˇs Ia-mi-in 8 a-na ub-bu-bi-im u-ul i-re-ed-du-u´ 9 tu-ba-ab-ˇsu-nu-ti-ma ´ lu.meˇ ´ s 10 a-hu-ˇsu-nu Ra-ab-ba-yu ˇsa i-na e-bi-ir-tim i-na ma-a-at Ia-am-ha-ad ki 11 wa-aˇs-bu ˘ ˘´ i-tu-ur-ru-nim i-ˇse-em-muu-ma i-ma-ra-s.u--nu-ˇsi-im-ma 12 a-na ma-ti-ˇsu-nu u-ul 13 ˇ mi-im-ma la tu-ub-ba-ab-ˇsu-nu-{SI}-ti. Lines 20–21: 20 ki-a-am ˇsi-ip-t.a` -am i-di-in-ˇsu-nu-ˇsi-im 21 mi-im-ma la tu-ub-ba-a[bˇs ]u-nu-ti. ARM III 58:5–10.


269

261. ARM XXVI 168:16–24, letter from three Yaminites named Yams. i-hadnˆ u, Maˇsum, and Hammi-Esim to the Yaminite leader Sumu-dabi during ˘the re16 ˘ volt against Zimri-Lim; u` e-bu-rum ik-ta-áˇs-dam 17 be-el-ni a-ha-am la i-na-ad-di 18 ˘ ˇsa na-wi-im 21 lu a-na a-ah-he-ˇsu li-iˇs-pu-ur-ma 19 .sa-bu-ˇsu-nu ˇsa a-la-ni 20 u` hi-ib-rum 22 23 ˘ ˘ ˘ ´ [...?...] pa-hi-i[r] a-na u4 -um tu-uk-ki wa-s.i-[ˇsu] ni-ˇse-em-mu-u´ ki-ma 1 LU 24 ˘ né-eh-ra-rum li-ik-ˇsu-dam. ˘ the lot sizes for agricultural land in ARM VIII 3:1 (one-plus iku) and 262. Compare VIII 14:1 (three iku). 263. Line 30, A-wi-in [R]a-[a]b-bi, in the oath formula. Awin should be understood as a clan defined in relation to the Rabbˆ um tribe, not an individual, even if the group took its name from an ancestor (versus the translation of ARM VIII 11:30 as “Awin le [R]abbéen”). meˇs ` 264. Lines 10–11, 22; ˇsa Sˇ A.GA.DU ma-ah-ru. Durand observes that the Sumerian ` ` probably covers the˘Akkadian n¯ebehum (LAPO 18, p. 172; cf. writing Sˇ A.GA.D U CAD s.v. ˇsakattû) and translates “ceinture” (belt). The˘ CAD entry observes this identification based on an unspecified lexical text from Emar, which appears to be Emar VI.4 556:56 –66 (HAR.ra-hubullu XIX). Various types of n¯ebehu are ˘ hu listed, beginning with the basic˘ term, as˘ gad-ˇsa` -ga-du` = ku-u´ ˇsa-ga-da = ni-be˘6 (line 56 ). Types include “white” (babbar = ba-ab-ar = pè-s.u, pes.uˆ ), “black” (ge = ki-ik-ki = .sa-al-mu, .salmu), “red” (sa5 = ˇse = ˇsa-mu, cf. s¯amu), “bicolored” (gun-a ` = gu-na = bar-ru-mu, cf. burrumu, a mixture of two colors), “linen” (gad = ki-te = ki-tu, ´ kitû), “wool” (s´ıg = ˇsi-ki = [ˇs]a-pa-ti, ˇsip¯atu), “blue (wool)” (s´ıg-za--na = za-ki-na = uq-ni, uqnû; see AHw s.v. uqnû(m) Lapislazuli, (grun)blau), ¨ “red wool” (s´ıg-gan-me-ta = hi-mit-ta = na-ba-ˇsi, nab¯asu; line 19 of ˘ local ikit.u), and two more obscure this text equates this Akkadian word with the terms. The material is evidently cloth, not leather. In Emar’s installation ritual ´ ´ ` for the NIN.DINGIR priestess of the storm god, the n¯ebehu is written tug IB.LA ˘ and is clearly a “sash,” covering the head. See Fleming 1992b, 188. 265. See, e.g., ARM XXIII 70:9 and 595:2 (under Samsi-Addu). 266. ARM III 12:16–26; 16 ˇsa-ni-tam pa-na-nu-um la-[m]a be-l´ı 17 a-na KASKAL.A i-la-ku DUMUmeˇs [I]a-mi-na iˇs-ti-na-a 18 sà-ar-ra-ru iˇs-tu e-le-nu-um 19 a-na a-la-né-e-ˇsu-nu i-la-ku-nim-ma 20 u` i-[t]u-ur-ru 21 i-na-an-na iˇs-[t]u ˇsa be-l´ı a-na KASKAL.A u-ˇ ´ se-ˇsi24 ru 22 u` ˇsi-ip-t.a` -am dan-na-ti[m] iˇs-ku-na-an-né-ˇsi-im 23 [a]r-na-am u-[t]e-e[r]-ma ´ ki-ma pa-[n]a-nu-um ma-[a]m-ma-an i-[n]a sà-ar-ra-ri 25 [i]ˇs-[t]u e-[l]e-nu-um a-na a-la-né-e-ˇsu 26 u-ul ´ il-la-kam. 267. See Durand 1987a, 198, referring to ARM IV 7+:13–15; 13 i-na li-ib-bi na-we{EM}-em 14 1 li-im .sa-bu-um ˇsa A-ah ÍD.BURANUN.NA 15 s[à]-ar-ra-ru-um i-ba˘ there is a contingent of those unsworn aˇs-ˇsi. “In the midst of the backcountry from the Ah Purattim (that is) one thousand strong.” See also Durand 1991, ˘ 335. For other examples of this use, see ARM V 81:13–14 (Lauˆ m 64; and 1998, to Yasmah-Addu); XIII 144:42? (Yawi-el, king of Talhayˆ um, to Zimri-Lim); and ˘ ˘ to Zimri-Lim, quoting XIV 104+:42 (Yaqqim-Addu, governor of Saggaratum, the besieged inhabitants of Razamâ; in Charpin 1993a, 199–200). 268. FM II 34:4–8; see Bonechi and Catagnoti 1994, 69 and note a. 269. Durand understands this engagement to involve a real transfer from Yaminite to Simalite tribal identity (1992b, 117–19), but the text never mentions any renunciation of the town’s Yaminite tribe. Toward the end of Zimri-Lim’s reign, Dabiˇs is listed with its fellow Yaminite towns in the military draft accounts of

270

270.

271.

272.

273. 274. 275.

276.

277. 278. 279. 280.

281.

Notes to Pages 97–100 ARM XXIII 428:8–10 and 429:8–10. Jack Sasson (1998a, 105) likewise understands A.981 to describe a permanent change of tribal affiliation. He understands the leaders of Dabiˇs to be abandoning their Yaminite tribe because of already reduced status. See Durand 1992b, 105–6. Marco Bonechi finds various examples of this root (hb/pr), along with the noun habrum (“migration”), in personal names from ˘ (1997, 503 and n207). ˘ Ebla In all cases, the laryngeal consonant is rendered in -H-, either with the sign ˘ distinguish between AH/IH/UH or with HA, and the BU/PU sign does not ˘ labials. ˘ ARM XIV ˘ 50:14, “I left for the land of (geographical name),” the˘ two ah-bu-ur; 78:18, “the one who takes refuge with you (plural),” ha-bi-ru-ku-nu; ˘ XXVI 510:25, “(personal name), who came to take refuge ˘ here from ARM Mari (was put in prison),” ih-bu-ra-am; ARM XXVII 70:17, a man was wanted for questioning, but “this man˘ has left for (emigrated to) Kurdâ,” ih-bu-ur; ´ ARM ˘ kingdom XXVII 116:32, identifying two men by residence, one still in the Mari settled at Saggaratum (verb waˇsa¯ bum), the other “left for Kurdâ,” ih-bu-ra-am; ARM XXVIII 46:6 , the king of Urgiˇs says that he has had to leave his˘residence ˇ h, “I have departed in exile,” a-na ha-bi-ru-tim in that city for neighboring Sina ˘ ˘ at-ta-s.i. For complete discussion of the evidence for this revolt, along with the relevant dossier of letters, see Durand, AEM I/1, pp. 335–53, and texts ARM XXVI 168-72. ´ Personal communication, 1997–98 seminar at the Ecole Pratique des Hautes Études (Paris). See the extensive listings in the CAD s.v. .sa¯ bu, as well as AHw s.v. .sa¯ bu(m). For Hana who “reside” in towns, see ARM II 48:8–9; V 51:5–7, 11–16; and XXVI˘ 508:5. The first and last texts refer to short-term encampment related to military campaigns, while ARM V 51 locates the Hana more generally in specific districts of the upper Habur: Nahur, Talhayˆ um,˘Qirdahat, and Aˇsnakkum. ˘ can cite ˘this text˘to illustrate the ˘ pastoralist practice of It is not clear that Porter dividing family members according to subsistence tasks (2000, 443–4). She is surely correct, however, to find here an example of how pastoralists are involved in both cultivation and herding. This letter was published by Dossin 1972a, 118–20. Durand presents a collated translation as LAPO 16, no. 268 (pp. 418–19). On the Sutˆ u, see Durand, LAPO 17, pp. 505–11. See ARM XXIII 427 iv:42 ; cf. 69:15. The nature of the relationship is somewhat hazy because the text is damaged: 8 ´ s ´ meˇs 9 [i-na(?) re(?)]-ˇsa-tim lu.meˇ i-na-an-na 5 LU Ia-i-la-yuki 10 [ . . . ] hi-ib-ru-um 11 ˘ 16, no. 81, [a-na .se-ri(?)]-ia il-li-ku-nim-ma. For suggested restorations, see LAPO p. 209. ARM XXVIII 147:5 –18 ; 5 aˇs-ˇsum ki-du-ti[m] a-bi iˇs-[pu-ra-am] 6 a-bi i-de4 ki-ma Ì 7 8 lu´ ` ma-ar-s.u´ i-na-an-na Qar-ni-L[i]-im I ip-ˇsu-ˇs[a]-an-ni u` TUR [a-na q]a-ti-ia u-ul ´ si 10 u-lu ´ ÌR a-bi-ia i-na-an-na 11 {I-NA´ qé-ru-ub 9 [A]N-ka li-iq-bi-ma ÌR lu-ur-ˇ 12 AN-NA} a-na an-n[é ]-tim a-bi li-im-li-i[k] e-mu-uq a-bi-ia u-ul ´ a-kà-ˇsa-ad 13 ! 14 15 ` a-na-ku ki-ma IR-di-ka a-ba-aˇs-ˇsi u-lu-ma ki-ma 1 su-ga ´ ´ 14 -gi5 mi-im-ma qa-ra-an ´ a-bi-ia 16 u-ul TUG s-ˇsa-ar 17 DUMU ˇsa ki-na-tim 18 ˇsa ma-tim a-nu-um-mi-im ´ u-wa-aˇ ´ a-na-ku.

Notes to Pages 101–112 282.

283. 284. 285. 286.

271

´ Sa-a]m-si-e-ra-ah ik-du-u´ [x xiˇs-tu Ha-am-[ma-an lu´ Di(?)-r]a-[yuki (?)] 4 [LU 5˘ m ki ˘ m]u(?) [ Sa-a]m-si-e-ra-ah i-na Su-sa-a [i]l-qubased ´ u. ´ Hamman is restored ˘ on his appearance in line 6˘ . ARM XXVIII 61:4–6; see Kupper’s discussion on p. 67. EA 51:6, a letter to the king of Egypt from the king of Nuhaˇsˇse in Syria. ˘ the last of which The form permits an infinitive, an adjective, or a participle, would best suit the likelihood that these are people. Dominique Charpin has pointed out to me (personal communication) the similarity between the noun k/qadûm and the Arabic qâdi, a local chieftain (see Lachenbacher 1987).

3

3. The Archaic State and the m¯atum “Land” 1. For a general statement of the principle, see Flannery 1998, 20–1. One example of its recent application to earliest state formation is found in Alan Lupton (1996, 99), of three-tiered settlement systems already existing in upper Mesopotamia before contact with the Uruk state in the late fourth millennium. 2. “Most current modeling of early state and/or urban hierarchies within this [Marx-influenced] framework implicitly rests on the assumption of rationally selfish and individualistic location decisions, with these in turn requiring information comparable to that provided by market pricing in order to permit minimization of transport costs.” The most formal version is “central place theory,” he says, which is often too simplified for application in practice. 3. Glassner (2000, 38) observes that the first royal inscriptions of early Mesopotamian “states” chose to designate their kingdoms by the name of the city capital, as in “Ur” or “Kiˇs.” Steinkeller (1999) proposes that we can trace a development in the Sumerian political situation from the earlier “en,” a templebased leader, to the later ensi(k), as a separate political “steward” over the god’s people. One fascinating relic of this development, according to Steinkeller, is the frequent existence of twin capitals, such as the famous Lagaˇs and Girsu. It seems that in such cases, a newer political center, usually with a male god over it, was built near the older religious center, usually with a goddess as its divine head. 4. See chaps. 1–3 in his What Happened in History (1942), on “Palaeolithic Savagery,” “The Higher Barbarism of the Copper Age,” and “The Urban Revolution in Mesopotamia.” For more on the urban revolution, see Man Makes Himself (1936), chap. 7. 5. According to Adams (1972, 735), “Truly urban agglomerations depend upon the institution of the state as a political form, and the emergence of the latter is but an aspect in turn of the formation of stratified class societies.” 6. Khoury and Kostiner, the editors of the volume, simplify this to “structures that exercise a monopoly of power in a given territory” (1990, 4). 7. See the thought-provoking analysis of Joannès 1996, 326–8 (“Contrôle des routes et des populations”). 8. This is the definition suggested by Hansen (2000, 15), who cites Sumer as an example, along with Ebla and other Syrian sites (p. 20).

272


9. Interestingly, the article by Glassner in Hansen’s volume resists the application of “city-state” terminology to Mesopotamia and provides a useful critical review of the phrase (2000, 35–6). 10. He expands this definition in a more recent publication: “A chiefdom is a regional polity with institutional governance and some social stratification organizing a population of a few thousand to tens of thousands of people” (1997, 14). 11. For the “segmentary state,” Stein cites Southall (1988), with further reference to Blanton et al. (1981). 12. A.3024 = B.308, lines 11–14; 11 [k]i-ma a-li-ˇsu Na-hu-urki u` ki-ma a-la-né-e 12 [ˇs]a 14 ˘ nam-la-ka-ti-ˇsu a-lum Bu-ru-un-duki13 a-al ki Zi-im-ri-Li-im u` A-dal-ˇse-en-ni ma-ruˇsu. Durand (LAPO 16, p. 475 note c), observes that the plural namlak¯atum consistently represents the singular “kingdom,” but the word seems to indicate the dominions beyond a king’s core domain. 13. Cf. ARM II 13:7 for Yasmah-Addu king of Mari, and ARM IV 76:43–4 for fugitives. 43 ˘ 14. ARM XXVI 423:42–7; 42 ul-la-nu-um [Z]i-i[m]-r[i-Li-im u] ` a-lim Ma-ri ki A-tamri-im 44 u` a-lim An-da-ri-ig ki LUGAL u` a-lum na-ak-ri ˇsa-nu-um 45 u-ul i-ba-aˇs-ˇsi ´ ki 46 ´ Zi-im-ri-Li-im u` lu´ KA.[DINGIR].RA LUGAL a-ye-em it-lu-lu u` su-qa-at ma-tim ´ a-yi-tim 47 il-pu-tu. I insert the negative in my translation to suit the English sense of the question. For “to touch the chin” for making alliances, see note h on p. 318. Charpin gives the evidence for the verb tal¯alum as “faire alliance avec (itti),” in 1997b, 365. The verb appears to be denominative, meaning “to send (auxilliary) tillatum troops.” ` 5 a15. ARM XXVII 167:4 –8 , quoting Atamrum indirectly; 4 a-na-ku 40 bé-ri A.Sˇ A 6 ki ki 7 8 na su-ri ´ lu-ul-li-ik a-di ma-at Ma-ri {ri } a-sa-ah-hu-ra-am u-ul ´ k[a]-ma qa-aq˘˘ qa-ad ma! -tim! lu-ki-il5 . The first two statements appear to be contrary to fact, rhetorical proposals set up to be refused as ridiculous. Samsi-Addu also uses the title in A.2231 (Charpin 1984, 42, just quoted above), but this has the ring of an imperial assignment from a foreign conqueror rather than the tone of local tradition. 16. FM VII 47:58, cf. “land (m¯atum) of Mari” in line 56. 17. Line 27; a-lamki ta-la-wi-i u-lu-ma ma-a-tam ta-ˇsa-ah-hi-it.. ´ ˘ d Da-gan i-na li-ib-bi ma-a-tim 18. ARM XXVI 416:3–4, 6–12 (cf. 417:9 ; 419:4 ); 3 m Iˇ˘s-me4 ki ki 5 giˇs ki-ma wa-s.e-e-ˇsu a-lam At-me-e il-wi di-im-tam uˇs-zi-iz u` a-na e-p´ı-re ˇsa-pa-ki-im 6 7 ` urˇ qa-tam iˇs-ku-un aˇs-ˇsum .te4 -e-mi-im an-ni-im As-q a-na .se-er be-l´ı-ia a-la-kam ´ d ISKUR 9 ´ a-na ma-ti-ia qé-ru-ub as-su-ur-re-e-ma um-ti-iq um-ma-a-mi 8 lu´ KUR i-nu-ma a-na-ku ´ ´ i-ˇse-em-me-e-ma i-na li-ib-bi ma-a-ti-ia 11 u-gal-la-al a-na .se-er a-bi-ia a-al-la-ku 10 lu´ KUR ´ a-di a-bi a-na li-ib-bi ma-a-tim 12 la u-q´ ´ e-er-re-ba-am mi-im-ma [ú-u]l a-la-ak. ˇ sa[r19. Laessøe and Jacobsen (1991, 159), text no. VII:11–12, reading ma-a-tam Su-ˇ r]a-aki . 20. ARM V 36:8; XXVI 437:29–30; 519:21. Perhaps compare the assembly of the m¯atum by fire signals, in ARM XXVI 515:29 and A.1866 (LAPO 17, no. 622). Note ARM I 91+:11–12 for the m¯atum reaching the safety and provision of the (western) town of Ekallatum (LAPO 16, p. 503–5; see Durand 1987a, 178–80). 21. ARM V 65:10–11; 10 te-re-tim a-na ˇsu-lum ma-a-tim u` a-al da-an-na-tim 11 m Zu-na-an i-pu-uˇ ´ s (same text in ARM XXVI 88:10–11); both letters of Asqudum to Yasmah˘ Addu.


273

22. ARM X 84:24–5. 23. ARM V 73:13 ; cf. I 7; II 121. 24. ARM X 31:13–15; 13 i-na li-ib-bi 14 ma-a-tim lu´ Ha-na[meˇs ] 15 ˇse-em i-ˇsa-ab-bi; cf. ARM ˘ I 67; XXVI 491; 511; 515. 25. A.3669+:22 –23 , in Lacambre 1997a, 446–8; 22 m [H]a-am-mu-ra-bi a-na ma ˘ ı-ip. For the final verb, ` s-nun-naki il-li-ik-ma 23 SE-ˇ ˇ su iq-lu ma-a-tam u-t´ a-at Eˇ ´ a-ap-p´ see note d to the text. 26. ARM V 66:8–9. 27. Mario Liverani finds the same consistent association between m¯atu and king in Late Bronze Age Syria (1974, 335). 28. Charpin 1984, 42, no. 1:5 –7 ; 5 ma-a-at Ma-ri ki 6 A-ah ´ı d UD.KIB.N[UN.NA] 7 ˘ u` nam-la-ka-ti-ˇsu. 29. FM VII 8:39–45; 39 [i]ˇs-tu i-na-an-na a-na ˇsa-na-at a-na ˇsi-it-ta ˇsa-na-[ti] 40 [a41 ´ meˇs n]a 10 MUh a´ LU u` na-am-la-ka-ti-ia 42 [l]i-`ıs-hu-ru-nim ˘ 43 ˇsu-nu [a]-na ma-ti-ia 44 li-ru-bu-nim-ma a-ka-as-su-ˇ ´ su-nu-ti-ma [a]-na .se-er Zi-im-ri-Li-i[m] ˘45 [u]-ˇ ´ saar-ra-ˇsu-nu-[ti]. 30. ARM XXVIII 60:11–14. 31. ARM XIII 143:13; LAPO 16, no. 303, pp. 475–8. 32. ARM X 51:12, cited in the CAD s.v. namlaktu, does not mention this term, according to Durand’s reading in his new edition as ARM XXVI 238. This text would not fit the pattern suggested by other published occurrences, so Durand’s collated change makes good sense. Durand’s restoration of na-a[m-la-ka-ti] [liis.]-s.u-ru, in ARM I 113+:66–7, depends on the assumption that the namlak¯atum ´ usually means “kingdom” generally, rather than the vassal “dominions” under a great king’s rule. Yasmah-Addu would be speaking to Samsi-Addu about new ˘ Mari court, hoping that “they may protect my dooutside appointments to the minions” (see Durand 1987a, 174). With such a large-scale restoration and no idiomatic precedent known to me, this reading must be treated with considerable caution. Even if correct, it is possible that any such “dominions” could pertain to Samsi-Addu rather than his son (so, “your dominions”). 33. Durand sees the malkum title behind the noun “namlak¯atum” (1993b, 50). 34. L87-150+:v 24–7, in Eidem 1987/88, 118. Bertrand Lafont seems likewise to read the namlak¯atum as one element only of what a king rules, so that this list forms an ensemble that defines the whole kingdom (2001a, 219n20). 35. Charpin, personal communication. The text is A.4309:20, na-am-la-ka-tam; to be cited in FM V, forthcoming. 36. ARM XXVI 372:30. 37. ARM VI 66:5–6. 38. ARM XXVI 427:26. 39. I wondered whether this might reflect the perspective of military commanders, who would attach towns and their populations to their individual leaders. ARM XXVI 427 is sent by three commanders (AEM I/2, p. 312). In the company of two kings, a third leader in ARM XXVII 72-bis is described as the GAL.MAR.TU s4 -tár, using the royal definition. I have not checked the ˇsa ma-a-at S.u-ba-at-Eˇ ´ contexts of all my collected examples of this format. 40. E.g., XXVI 384 (Iˇsme-Dagan); 385 (Hammurabi of Babylon); 394, 411 and 416 ˘ (king of Karanâ).

274


ˇ 41. A.1968:5–9 (cf. A.4251+); 5 d ISKUR-ma ma-a-tam4 (TUM) ka-la-ˇsa 6 a-na Ia-ah7 giˇs meˇs ˘ du-Li-im ad-di-in u` i-na TUKUL -ia ma-hi-ra-am u-ul ir-ˇsi 8 i-ia-tam i-zi-ib-ma ´ ˇ˘ ma-a-tam ˇsa ad-di-nu-ˇsu[m] 9 a-na Sa-am-si- d ISKUR ad-[di-i]n (see Durand 1993b, 43, 55). ˇ 42. FM VII 8:27, i-na ma-at d ISKUR, “in the land of Addu.” 43. A.2231:4–9 and 1 –8 , parallel statements in a votive text for the dedication of two thrones to Itur-Mer. The reverse is more complete: 1 i -n[u-ma] d 2 3 4 5 [I-tu]r-M[e-er] ik-ri-[b]i-ia u` ta-ás-li-ti iˇs-mu-ma ma-a-at Ma-ri ki 6 A-ah ´ ´ıd 7 8 ˘ UD.KIB.N[UN.NA] u` nam-la-ka-ti-ˇsu u-ˇ ´ sa-ak-li-lam; Charpin 1984, 42. 44. ARM XXVI 523:31. 45. ARM XXVI 382:3 ; 412:23, 31; XXVII 132:11. 46. Verb dal¯ahum, ARM IV 25; XIII 146; XXVI 323, 542, 548; verb hâˇsum, XXVI ˘ ˘ 210. 47. Verb nâhum, ARM I 43; II 16; IV 57; VI 76; XXVI 411, 430, 519. 48. The one˘most likely example seems to be the reference to eighty Imarites who are also “heads of the m¯atum” in representing their town in a major legal proceeding at Hˆıt (ARM XXVI 256:14–15, to be discussed later). If the m¯atum is indeed that ˘ of Imar, this is a polity with no known king, and no king is in view here. 49. ARM I 10:19–20 (=LAPO 17, no. 475, pp. 522–3); 19 ki-ma ma-a-at Za-al-[m]aq´ı-imki ka-lu-ˇsa ˇsa wa-a[r-ki-ˇsu i-ma-ru(??)] 20 ki-a-am i-qa-ab-bu-u´ um-ma-a-mi . . . 50. ARM XXVI 494:6–7, 12. 51. The land “stands up” (tebûm), with direct speech quoted immediately after (umm¯ami); FM II 118:12 . 52. ARM XXVI 377:16; [qa]-qa-ad ma-ti-ni li-ki-il. Cf. also A.230 in Durand 1991, 54. 53. ARM XXVIII 68:11–12. 54. One way to see who are the key political players in the region surrounding Mari is to review the royal correspondence of Zimri-Lim, as published in ARM XXVIII. 55. Oddly, I have not found mention of a m¯at Elam(tim), though the phrase would be natural from a Mesopotamian point of view. Elam is usually described simply as “Elam” (e.g. ARM VI 66:4), their armies as “the Elamite” (e.g., ARM XXVII 132:16), and their ruler “the vizier (SUKKAL) of Elam” (e.g., ARM XXVI 362:3). 56. Joannès proposes that the geographical associations of Yamutbal and Numhâ ˘ have priority over the tribal (1996, 353–4). This might be gainsaid from various evidence. One starting point is the reference to the early merhûm Bannum’s ˘ hâ and the dealings with his own tribe of Simalite(s), in relation to the Num Yamutbal, before he took up his current responsibilities in the Ah˘ Purattim ˘ (“Banks-of-the-Euphrates”; A.1098:10 , discussed with the merhûm). Notice that ˘ when these pastoralists are described, Bannum speaks of “Simalites,” not their geographical range in Ida-Maras. . 57. The abbreviated form m¯at is used for the nomen regens in a genitive chain, so “the land (of),” with these proper names. 58. Durand 1990b, 50–1 and n. 54; 2 [DUMU]meˇs Ia-mi-na ka-lu-ˇsu 3 i-na a-la-ne-ˇsu4 nu u` ma-ti-ˇsu-nu ˇsu-s.uit-ta-na-ag-gi-ˇsu. The text comes from the reign of ´ u-ma ´ Zimri-Lim. Samsi-Addu once upbraids his son Yasmah-Addu for contemplating ˘ return “to their m¯atum” a census of Yaminites, who would not then be able to in territory controlled by Yamhad (ARM I 6:12).

˘


275

59. For Samsi-Addu and the Numhâ, see the Mari kispum ritual, FM III 4 i:21; for the ˘ Finkelstein 1966. Charpin and Durand discuss Amnanˆ u and the “Yahrirû,” see ˘ of perceived tribal ancestry in these dynastic settings the whole phenomenon (1986, 166–8). 60. See the royal inscriptions of these kings in Frayne 1990, 202–37 (Warad-Sˆın), 266–7 (Kudur-mabuk), 270–300 (Rim-Sˆın). 61. See the texts for these two kings in Frayne 1990, 440–66. The tribal affiliations of the Uruk dynasty can still be seen in the later letter of king Dingiram to Sˆınmuballit. of Babylon near the beginning of Rim-Sˆın’s reign at Larsa, where the “troops of Uruk” are joined to the “troops of Amnan-Yahrur” and the “troops of ˘ lines i 28-30; cf. ii 27; Yamutbalum” (of Rim-Sˆın?). See Falkenstein 1963, 56–9, iii 39. Throughout, Amnan-Yahrur is distinguished geographically from Uruk ˘ proper. ˇ 62. See i:19 of Yahdun-Lim’s Samaˇ s temple inscription. The seal that replaces Hana ˘ with “the Sim˘alite(s)” is on ARM XIII 144; see Charpin and Durand 1986, 151–2. 63. [ma]-a-tim bi-ri-it ´ıd IDIGNA u` ´ıd BURANUN.NA (A.889:6–8), and [ma-a]-tim ` ´ıd BURANUN.[NA] (M.5037:2 ). The two texts are found in ˇsa ´ıd IDIGNA [u] Charpin 1984, 47–8, and Charpin and Durand 1985, 296 and n. 16. 64. Zimri-Lim’s use of the title does seem to be colored by Samsi-Addu’s idea of a kingdom of two rivers, built instead around the Euphrates and the Habur. I ˘ expand on this in the final chapter. 65. This phenomenon would not be the same as a single state defined by twin town centers, as may be represented by Hiwilat and Talmuˇs in ARM IV 68 or ˘ Azuhinum and Tupham in XXVI 437. Other such paired towns include Isqâ˘ a, Karanâ and ˘Qat.t.arâ, and Kurdâ and Kasapâ. and-Qˆ 66. The following lists include only the phrases m¯at Zalmaqim/Ida-Maras. / ˇ Subartim, not the many other references to these lands without the explicit designation as m¯atum. 67. For an example of kings uniting for purposes of making an alliance, see the letter of Zimri-Lim found at Tell ar-Rimah, in which he promises to gather ˘ enter into an alliance with king his “brother” kings so that all may together Hatnu-rabi of Karanâ/Qat.t.arâ (Tell ar-Rimah 1:9–11). ˘ ˘ 68. A.715, LAPO 16, no. 346, pp. 538–9. 6 ˇ 69. ARM XXVII 45:6–8; .te4 -[e]m ma-a-at Su-bar-tim .t[e4 -e]m Ha-am-mu-ra-bi 7 [lu´ K]urdaki [u` .t]e4 -em A-tam-ri-[i]m 8 ˇsa be-l´ı ki-a-am i[ˇs-p]u-ra-am.˘ 70. On the four cities and kings of Zalmaqum, see the general description by Kupper, in his ARMT XXVIII, pp. 35–6. 71. ARM XXVIII 55:5–7. 72. Lines 11 –12 . Ibal-Addu consistently reserves the term “m¯atum” for the coalition and does not apply it to his personal domain at Aˇslakkâ. See also XXVIII 48:61–2 and 54:5 . 73. Years Z-L 4 and 13 (the fourth and thirteenth full years of Zimri-Lim’s reign); ˇ see Charpin and Durand 1985, 306. It is perhaps worth noting that Subat-Enlil’s special prominence in the upper Habur did not outlast the reign of Samsi-Addu. ˘ town returned to its old name of Se ˇ hna, the During the reign of Zimri-Lim, the ˘ settled seat of the land of Apum, which was not part of the Ida-Maras. coalition.

276

74. 75. 76.

77. 78.

79. 80. 81.

82.

83. 84.

85.

86.

87. 88.

Notes to Pages 126–129 ˇ hna could have been included, based on From a geographical point of view, Se ˘ ˇ a and Aˇsnakkum. its location at the eastern end of the alliance, not far from Sunˆ ARM XXVIII 65:16–17. For the town itself, see Charpin 1987b. Lines 27–8, in Charpin 1990a, 70; 27 pa-na-nu-um Ia-ah-du-Li-im i-nu-ma a-na matim ˇsa-a-ti i-la-ku 28 a-na ab-bé-e I-da-Ma-ra-as. q´ı-ˇsa-tim i-q´ı˘-eˇs-ma. The letter was sent by the merhûm Bannum, at the very beginning of Zimri-Lim’s reign. ˘ inscription for the foundation of the town of Dur-Yahdun-Lim, See the disc ˘ RIME 4, p. 602, E4.6.8.1:15–16. For Ida-Maras. under Samsu-iluna, see his royal inscription E4.3.7.8:3 –4 (cf. lines 25–6, etc.), in Frayne 1990. This territory would have been lost after the time of Samsu-iluna, so that the reference in the “edict of Ammi-s. aduqa” (sects. 18 –19 ) would not reflect actual control. TA 1930-T575:7–8, cited in Stol 1976, 64. On Eˇsnunna generally, see Charpin 2001. ARM XXVIII 15:9–10. This analysis updates the cautious approach of Michalowski (1986), who concluded that no consistent geographical definition could be established and that the late third-millennium references reflect southern Mesopotamia’s shifting view of its northern periphery. Prechel and Richter (2001) have gathered a whole set of early secondmillennium texts in a dialect called eme-su-bir4 ki -a (Subarian), which appears to be a variety of early Hurrian. I would like to thank Dominique Charpin for the reference. Lines 7–8 (again in 26–7), i-na ma-a-at Ti-g[u-n]a-[n]imki -m[a] wa-aˇs-bu; and then line 9, a-na ma-a-at Hi-ir-ba-za-nimki il-li-ku-ma. ´ s ˘ ARM IV 6:5–11; 5 lu.meˇ Ra-ab-ba-yu 6 ˇsa i-na ma-a-at Ia-am-ha-ad ki wa-aˇs-bu 7 iˇs10 giˇs ˘ ´ há a-na pu-ru-nim um-ma-a-m[i] 8 pa-ni-ne a-na e-bé-ri-im 9 ni-iˇs-ku-un-[m]a MA ˘ 11 [e-bé-ri]-ni u-ul ´ i-[ba-aˇs-ˇse-e]. This is the text read by Durand in LAPO 17, no. 706 (p. 456). The verbs in the last two lines are simply unknown, in spite of these restorations. ARM V 23:5–13 describes a gathering of 2,000 Sutˆ u to raid Qatna, concluding in lines 10–13 that “they have set out to raid the flocks(?) of the land of Qatna,” 10 a-[na] na-[we-e-em] 11 ˇsa ma-a-at Qa-t.a´ -[nimki ] 12 ˇsa-ha-t.i4 -im 13 it-ta-al-ku. Durand’s ˘ precedent in Dossin (1939, restoration in LAPO 17, no. 745 (p. 507) has good 988 and 991), cited in CAD s.v. namû A 1a, where the nawûm is the object of raiding. In this case, however, the tribal identity of the nawûm population is less in view than their “flocks,” treated as the simple possession of the Qatna kingdom. Such a view is typical of Samsi-Addu’s administration. Steiner 1988, 333–4. The full syllabic writing is fairly rare, but the spelling kalamtim appears more often. The singular kur may also represent ˇsadûm (or sádwum), “steppe,” but kur-kur and kur-kur-ri probably indicate m¯at¯atum, “lands.” The writing ki occurs in names, where ma+da unugki alternates with ki unugki , and so on. For the basic contrast of kur “Ausland” from kalam “Inland,” see Wilcke 1990, 470–1; Steiner 1982, 636–7. E.g., Postgate 1992, 34; Limet 1978, 6–8. Steiner (1982, 633 and 637) acknowledges this use, but suggests a more limited use for one’s “own land.” It is not


89.

90.

91. 92.

93. 94. 95. 96.

97.

98.

99.

100.

101.

102. 103. 104. 105.

106. 107.

277

clear, however, when this would refer to any particular state that is perceived as a subset of Sumer. Glassner observes something to this effect, saying that through the last third of the third millennium, ki-en-gi is used only for Sumer’s own land (2000, 39). For translations, see Cooper 1986, 94–5 (Um 7.1-2); Postgate 1992, 35. For the Sumerian text, see Steible 1982, Luzag. 1 and 2. For comment, see Steiner 1982, 638; Selz 1992, 202. See Cooper 1986, 37, 41–2 (Eanatum La 3.1; 3.5). Ibid., pp. 94–5 (Lugalzagesi, Um 7.1, 2) and p. 105 (Enˇsakuˇsana Uk 4.1, 3, as “lord of Sumer and king of the nation”). Krebernik (1998, 242) observes that two Fara lexical texts and several economic texts refer to ki-en-gi, but it is not clear that the whole region is intended. Lieberman cites the Sumerian King List as one example of later Sumerian unity read back onto earlier times (1992, 128). Enmetena 32 1:4 –8 ; see Selz 1992, 202. This term is already found in the old Ebla texts from the mid-third millennium, evidently to render the Semitic m¯atum (Steiner 1988, 333–4). See the texts and discussion in Limet (1978, 2–6), who does not draw any comparison with the Akkadian m¯atum and its use for the land and population of surrounding towns. The Sumerian title is lugal-ki-en-gi-ki-uri(-ke4 ), translated into Akkadian ˇsar m¯at ˇ Sumerim u Akkadim, as in an inscription of Lipit-Eˇstar of Isin; see text E4.1.5.3:14– 16, in Frayne 1990. ˇ See, e.g., the phrase LUGAL KALAM Su-me-ri-im u` Ak-ka-di-im, “king of the land of Sumer and Akkad,” in a royal inscription of Hammurabi of Babylon ˘ (E4.3.6.17:16–17, in Frayne 1990). I would like to thank Harry Hoffner for pointing out to me the continuity between the Hittite evidence and the earlier Akkadian language tradition. This section is built from his comments and the illustrations that he provided me. It would be interesting to know whether careful examination of the documentation by Hittitologists would yield some distinction between Mesopotamian and Anatolian usage that could clarify the relationship. KUB 21.47 + 23.82 (+) KBo 19.58. Beckman (1999a, 161n5) says that the Hittite word for town (happiriya-/happira-) means a “place of trade,” which ˘ ˘ shared by the Sumerian and the Akkadian would be a native Hittite heritage, not words. The final -m was disappearing already during the early second millennium, and its loss does not affect the meaning of the noun. For URU Atriya, see KUB 14.3 i 37, in Sommer 1932. These observations come directly from Harry Hoffner. For the Kaˇska, see von Schuler 1965. The reference to Nuhaˇsˇse appears in the treaty of the Hittite king Mursili II with Tuppi-Teˇsˇsup of ˘Amurru (Beckman 1999a, 59). IBoT I 30:2–8, translation of Harry Hoffner, personal communication. For the text and an older translation, see Goetze 1947, 90–1. On “elders,” see Guterbock ¨ 1954; Klengel 1965. Harry Hoffner observes that ´ meˇs ˇsa a-la-ni) in contexts that there are also “men of the town” (like Mari’s LU

278

108.

109. 110.

111.

112.

113.

114. 115. 116.

117.

Notes to Pages 133–134 indicate “jurisdiction and autonomy.” See the Hittite laws, par. 40, 46, 47; KBo 22.2 obv.10; HKM 52:32. For the last, see Alp 1991, 216ff. During roughly the same time as the Mari archives, the kingdom of Karanâ/ Qat.t.arâ was large enough to have such subdivisions. See Tell ar-Rimah 280:4–5; ˘ and note the ˇsa¯ pit.um “governor” in 296:4. For the division of Yahdun-Lim’s kingdom into hals.um districts, see Charpin ˘ ˘ forthcoming c. References include ARM XXII 127, a list of large numbers of garments for these three districts; ARM XXIII 428 and 429, the oft-cited muster lists for Yaminite tribal towns; XXIII 430, 1,600 men set to fight on behalf of Babylon; and ARM XXV 783, silver defined by these three districts. ARM VII 277 lists together a sequence of local “palaces” at major town centers in Zimri-Lim’s realm. This list includes Qat.t.unan, but also Dur-Yahdun-Lim, which ˘ Dur-Yahdunwas not a district center. The order is Mari, Terqa, Saggaratum, ˘ of Lim, and finally Qat.t.unan (v :2–6). Note that Qat.t.unan becomes a district the Mari kingdom partway through the reign of Yasmah-Addu (Villard 2001, 78, A.1139). On the Suhûm, see, e.g., ARM XXVIII 84:9–10 and p. 119 note a, on Zimri-Lim ˘ Tuttul and the frontier of Suhûm, showing the boundaries of the speaking of kingdom up and downstream along the ˘Euphrates River. Being downstream from Mari, Suhˆ um was liable to be contested by the powers of southeastern Mesopotamia. ˘On the dispute over Suhûm between Mari and Eˇsnunna, see Charpin 1991b, 147; and on the Suhûm˘ region, see 1997b. ˘ um the merhûm of the “province” of Kupper, ARM XXVIII, p. 2, calls Meptˆ ˘ Mari language (see my Suhûm, a combination that does not suit the standard ˘ discussion of the merhûm, below). Both Meptˆ um and a man named Buqaqum are mentioned as key˘ leaders in the Suhûm, without clear title (see Joannès 1996, 334). The prestige of both leaders is˘ seen in the fact that they received letters directly from foreign leaders: Meptˆ um, from Hammurabi of Babylon (ARM ˘ (XXVIII 38), and Zaziya of XXVIII 10), Yarkab-Addu of Hanzat in Zalmaqum ˘ the Turukkû people (XXVIII 179); Buqaqum from Hammurabi of Babylon ˘ as a sug¯ag¯utum fee, (XXVIII 6 and 7). Buqaqum is known to have paid sheep as appropriate to Simalite pastoralist leaders, and he is identified as a Sapiratumite. See M.9881, from Zimri-Lim’s first year, in Charpin 1997b, 353n44. This early status need not define the full scope of Buqaqum’s role by the time Hammurabi writes to him in the tenth full year of Zimri-Lim’s reign (see Kup˘ ARM XXVIII, pp. 1–2). per, ˇ ˇ See LAPO 17, p. 321, for location at SubatSamaˇ s. ARM III 12:8–9; 8 a-na ha-al-s.´ı-ia u` hi-ˇsi-ih-ti e´-[k]ál-l[im] 9 a-ha-am u-ul ´ na-de-e-ku. ˘ ARM V 33:14. Durand˘ comments˘ that˘ such greetings usually come from a governor or someone with primary local responsibility (1994, 83–4). This type of greeting is standard in the letters of Zimri-Lim’s district governors, gathered in ARM III (Kibri-Dagan of Terqa), VI (Bahdi-Lim of Mari), XIV (Yaqqim-Addu ˘ of Saggaratum), and XXVII (Qat.t.unan governors). This Tarim-ˇsakim appears to hold a post in the Suhûm, downstream from Mari, and is not the same as the vizier of Mari (Villard˘ 2001, 20–1). Cf. for Samsi-Addu’s kingdom the words of Yarˇsi-aˇsari, “the district is well, the town is well” (ARM V 69:5–6).


279

118. ARM XIV 121:26; [h]a-[l]a-[a]s.-ka a-na dan-na-tim ki-mi-`ıs. 119. These are also kapr¯a˘tum in XIV 8:5, 17 and in the Mari district (VI 58:21–2). ARM XIV 8:6 also describes these by the general word “¯alum,” as does III 41:12. 120. ARM XXVII 14:23. 121. ARM XXVII 46:3 ; 116:17; 120:8. 122. ARM XXVII 46:2 –12 . 123. The association is clear for Zimri-Lim in ARM XXVI 17:26–8 and 61:8–9, where travellers through the upper district would stop in the town of Terqa (cf. also 233:10–11 and note b, where the same location is confirmed indirectly). ARM XIV 51, 69, XXVI 447, and FM III 60, 67, and 95 offer no barrier to the same identification. For Samsi-Addu, I found the same designation in XXVI 81 and 265, which could have the same point of reference, though without proof. 124. ARM V 25:21. 125. Durand translates the text as LAPO 16, no. 302, and see pp. 74 and 338. The letter was sent by Ibal-Addu, the king of Aˇslakkâ. “I am close by the upper country, so news of Eluhut, Lullˆ u, Hahhum, the land of Zalmaqum, Burundum, ˘ available˘ [to] ˘ ˘ me.” 26 a-na ma-tim e-li-tim [q]é-er-bé-kuand Talhayum is (always) 27 ki lu´ 28 ˘ ma Ha-ah-hi-imki ma-a-at Za-al-m[a-q]´ı-im 29 .te4 -e-em E-lu-hu-ut Lu-ul-li-i ki ˘ ˘ ˘ h-r]i-ia ˇsa-ki-[in]. [B]u-ru-un-di-im[ ] u` Ta-al-ha-[y]i-imki˘ 30 [ma-a ˘ ˘ 126. ARM XXVIII 1:6, cf. 2 . 127. ARM II 21:18 and 21; cf. LAPO 16, p. 543. 128. ARM IV 12. When Iˇsme-Dagan, Samsi-Addu’s viceroy at Ekallatum, speaks of mares from the upper country and mules and donkeys from Andarig and Harbˆ u, an eastern location is more feasible than Zalmaqum. Ekallatum was ˘ the Tigris River, and Andarig was near the Jebel Sinjar, between the Tigris on and the Habur. See ARM I 132:20, vs. LAPO 16, p. 338 note c. Another refer˘ V 68 provides no clear location. ence in ARM 129. A.3669+: 7 , in Lacambre 1997a, 446–9. 130. See Eidem and Laessøe 2001, 84, no. 12:23. 131. This language was naturally not unique to the Mari evidence. For example, the same phrase is used from the perspective of the Qat.t.arâ, between the Habur and the Tigris Rivers (Tell ar-Rimah 132:7–8, etc.). King Dingiram of Uruk˘mentions some undefined “upper land”˘ from his southern vantage (Falkenstein 1963, 56–7, ii 3–4). 132. ARM XXVIII 83:3 , letter of Haya-sumu, king of Ilan-s. urâ. ˘ in van Koppen 1997, 419–21 and note g. 133. A.1333:6 , ma-a-tim ˇsa-p´ı-i[l-tim]; Charpin reports an example of this expression in FM V (forthcoming a), where “the lower land” refers to Larsa, the dominant state in the old Sumerian region before its conquest by Hammurabi of Babylon. 134. See ARM XXVI 47:15, ˘where Asqudum reports omens taken for the ha-al-s.´ıim ˇsa-ap-li-im. Alternatively, this could be the whole Euphrates segment˘of the kingdom, as opposed to the Habur segment to the north. 135. See under Samsi-Addu, ARM˘ I 22:12; 23:11; 42:37; 69+:13, 2 ; II 131:13–14; under Zimri-Lim, XXVI 367:11–12; A.1610+:9–10 (Durand 1988). 136. Where the census is taken by hals.um, it need not be for military draft, but rather ˘ serves local labor needs. In ARM XXVII 25:7–10, the governor declares that he has performed the census (verb ubbubum), so that now he has cleaned the

280

137. 138. 139. 140. 141.

142.

143.

144.

145. 146.

147.

148. 149.

Notes to Pages 135–138 canal and restored the water supply. “My lord’s god has established obedience in this district, so I could take a census of the district, and with the (work-)crew thus available I dredged the river (or, canal) (and) produced water” (7 DINGIR ˇsa be-l´ı-ia te-eˇs-me-em a-na ha-al-[s.´ı-im] 8 ˇsa-a-ti iˇs-ku-un-ma ha-al-s.[a-a]m u-bi-ib-ma ´ 9 ˘ s-ˇsu-u´ 10 ÍD ah-lu-us.x ? (IS) me-e ˘ u-ˇ i-na .sa-bi-immeˇs ˇsa ki-ma i-ba-aˇ ´ sa-ab-ˇsi). A second ˘ to work instead of war: “Now, if example is less clear, but may likewise refer I take a census in this district, not one man will stay settled in this district” ´ i-na (XXVII 46:2 –4 ; 2 i-na-an-na ˇsum-ma ha-a[l-s.a-a]m 3 ˇsa-a-ti u-ub-ba-bu 1 LU ´ 4 ˘ ha-al-s.´ı-im ˇsa-a-ti u-ul ´ uˇ ´ s-ˇsa-ab). ˘ ARM III 5:44 and XXVII 40:19. ARM XIV 13:4–5. ARM II 101:7–12; see LAPO 17, p. 632. ARM XXVII 27:31. ARM II 16:18, .sa-bu-um ˇsa ha-la-as. Ma-ri ki . See also the discussion in Villard 2001, 15–16. In ARM I 113+, it˘ is Samsi-Addu who has called the Mari kingdom a hals.um, whereas his son Yasmah-Addu uses some other term (lost in a break, ˘ proposed as namlak¯atum). ˘ but The m¯atum of Hiwilat and Talmuˇs rebels against Samsi-Addu (ARM IV 68:5– 6). Vassals such˘ as Amaz and Yapt.urum are said to belong to Zimri-Lim as separate m¯atums (X 84:38; XIII 144:3–4). See also II 50:5 for Adna under ˇ Kurdâ, A.1289+ ii:5–6 for Situlum under Eˇsnunna (Charpin 1991b, 149–55). See A.2231:5–6, 5 –6 , Charpin 1984, 42, already cited; “the m¯at Mari and the Ah Purattim.” In this case, the inscription actually defines what Samsi-Addu ˘ himself has conquered, and not what Yasmah-Addu has been assigned to rule. ˘ One indirect confirmation of this political priority may be the taking of omens for the hals.um unit, at royal command (XXVI 81:10; 87:23, both for the period ˘ of Samsi-Addu). FM III 147:10 –11 ; be-l´ı a-na ha-al-s.´ı-im a-na a-ta-pu-ˇsi-im il-li-ik. ˘ ` s-nun-naki 23 ha-al-s.u´ na-du-tum i-ba-aˇs-ˇsu-u´ ARM XXVI 373:22–6; 22 i-na ma-a-at Eˇ 24 meˇs ˘ ` s-nun-naki i-t.a` -ar-ra-ad-ma 25 a-ˇsa-ar ha-al-s a-na ma-a-at Eˇ . u´ na-du-u´ DUMU la26 ˘ si-mi-ˇsu a-na .se-er SUKKAL ELAM-tim i-it-ti-qu. ´ ARM XXVIII 136:5–6. Other rulers from the royal correspondence of ARM ˇ XXVIII do likewise: the king of Zalluhan (79:6, 26) and the king of Sudu hum ˘ ˘ (110:6; cf. Zimri-Lim’s language in 112:12–13). Ibal-Addu of Aˇslakkâ is willing to name other vassal kingdoms as “districts” of the Mari king (62:10, 12, 29). Outside this volume, see also Haya-sumu in FM II 127:7; Zakura-abum in A.2943 ˘ (Durand 1987b, 230); and Yakun-Derum in ARM VI 31 (see LAPO 17, p. 186 comment); cf. ARM II 122:15–16, and collation in LAPO 16, p. 468. lu´ ` Tell ar-Rimah 2:16-19; 16 .su-pu-ur Eˇs-nun-naki 17 i-na li-ib-bi ha-al-s.´ı-ia a-na-as´ 18 ˘ sà-ah u` ki-ma la i-ta-ar-ma a-na ha-al-s.´ı-ia 19 la i-il-li-e-em e-ep-pé-`˘es-su. ˘ derives the noun hals.um˘ by an exchange of the consonants /l/ and Durand /r/ from the verb har¯a.sum, ˘“to cut off,” meaning “retrancher d’un ensemble” ˘ In her thorough treatment of the word, Brigitte Lion at(LAPO 16, pp. 120–1). tempts no etymology (2001, 151–9). The l/r exchange in Durand’s etymology is possible but less than completely satisfying. It is perhaps worth noticing that a hls. root resembles closely the common Semitic root hlq, “to divide” (Hebrew, ˘ Aramaic/Syriac, Ugaritic; cf. Akkadian eqlum “field”?). ˘Semitic roots sometimes cluster with similar meanings developing around shifts in the last consonant.


150.

151.

152.

153. 154. 155. 156. 157.

158. 159.

160. 161. 162. 163.

164.

165. 166. 167. 168.

281

For Biblical Hebrew examples, see GKC, p. 101, roots from q-s.- related to “cutting.” ARM XIV 64:5–8 and 65:6–7. ARM XIV 65 defines these leaders by their plural hals.a¯ n¯u, though the same single district of Saggaratum appears to be in view. ˘ Evidently, the hals.um is subdivided in schematic, not likely formal terms, into another level of˘ subordinate parts. ARM XIV 8:5–8. For indirect indications of the same group communication at the hals.um level, see also ARM II 103:12 and XIV 75:8–9 (sug¯ag¯u and la˘ puttû, Saggaratum) and A.2801 (same, at Mari; see LAPO 16, no. 268). The situation at Terqa is somewhat different, because the majority of district towns are Yaminite. There, when the governor addresses similar Yaminite leadership, the effect is also hals.um-wide, though applied to only one major segment. Con˘ III 6:17; 21:7; 70+:8–9; XXVI 450:7–8, all sug¯ag¯u, with III sider II 92:12–13; 70+ adding laputtû. ARM XXVI 447:7. Ordinarily, one would conclude that the “upper district” was that of Terqa, but the phrase is introduced by the statement that “my lord commanded (the following) about the district of Qat.t.unan.” ´ s ARM XXVI 256:14–17; 14 80 lu.meˇ I-ma{R[U]}-ru-u´ ki 15 qa-qa-da-at ma-tim 16 aˇsd ` ˇsum KU.BABBAR ˇsa il-tim Ba-a4 -al-ta-ma-tim 17 [il-l]a-ku a-na d ÍD ˇsa-li-im. ARM XXVI 393 and 463, vs. XXVI 438. A.2730, in AEM I/2, p. 33. ARM I 9:17, using r¯eˇset m¯atim instead of the more common qaqqad¯at m¯atim. In wider Akkadian evidence, the noun “r¯eˇsum” is not generally used for leadership, unlike its synonym qaqqadum; see CAD s.v. qaqqadu 3, “head of an organization, leader.” See A.2226, in Charpin 1993b, 182, no. 7. lu´ ARM XXVI 12:3 –5 ; 3 i ni-il-li-ik 4 LUGALmeˇs -ni-ne u-lu-ma su-ga-gi-ni i ni-it.-r[a´ 5 ad]-ma u` it-ti-ku-nu li-ta-ap-lu. The verb “to send” is uncertain; .tar¯adum should take the form lit.rud, not lit.rad, though Durand observes that the alternation of u/a is attested elsewhere (note b to the text). ARM XIII 108:1 ; [u]m-mi-ˇsu u` ni-ˇsi-ˇs[u]. ARM XIV 107:4 –6 ; qa-du-um ni-ˇsi-ˇsu. See also XXVII 2:31; A.2432:7 –8 , in Dossin 1972a, 125–7. ´ ˇsa-a-ti u-ul ´ su´ u` ni-[ˇs]e20 -ˇs[u] 61 a-[na ARM XXVI 199:60–3; 60 LU ´ i-mu-ru-ma E-ˇ Ì]R-du-ti[m i]d-di-in i-na ˇsa-ni-i-im u4 -mi-im .tup-p´ı Ia-si-im-d Da-gan ik-ˇs[u-da]m 62 ´ ˇsu-u´ ik-ta-áˇs-dam i-na-an-na an-ni-tam la an-ni-tam be-l´ı li-iˇs-pu-ra[u]m-ma-mi LU [am] 63 ni-ˇse20 -ˇsu lu-wa-aˇs-ˇse-er. ´ am-mi-nim ˇsa E-ti-ki e-pé-ˇsi-im u` ni-ˇsi-ki bu-ul-lu-t.im te-zi-bi-ma, ARM X 166:8–9. Durand translates the verb bullut.um (“to make live”) as “contrôler” (LAPO 18, no. 1268 and pp. 502–3). ARM XXVIII 20:26–8, letter of Yatar-Ami to Zimri-Lim; 26 E´ há -ˇsu-nu ni-ˇsi-ˇsu-nu ˘ 27 a-na ma-h[i-i]s. qa-qa-di-ˇsu-nu 28 a-na-ad-di-in. ˇ See ARM I˘ 22:34–8, the Ita-ites; IV 77:13, the Sinamumites; XXVI 365:38, the heads of Kasallu(k). ARM I 91+; see lines 10, 11, 8 , 10 , 15 –19 . ARM I 91+:15 –19 ; 15 qa-at ni-ˇse20 -ku-nu .sa-ab-t[a-nim] 16 q´ı-ih-la-nim a-na li-ib17 ki 18 ˘ na-wi-ku-nu it-ti bi ma-a-ti[m] at-la-ka-nim i-na A-ba-[a]t-tim[ ] eb-ra-ma i-na ah-he-k[u-nu] 19 [s]u-uk-na. My translation follows Durand’s interpretation of

˘˘

282

169. 170. 171.

172.

173. 174. 175. 176. 177. 178. 179.

180. 181.

182.

Notes to Pages 140–142 the verb qah¯alum as “to assemble,” comparing the Hebrew root qhl (LAPO 16, p. 504). ˘ ARM XXVI 156:5–9; 5 m Za-i-ku-um lu´ Ia-ab-li-ia-yuki 6 qa-du-um ni-ˇse20 -ˇsu a-na Ha˘ ar-bé-e ki 7 it-ta-bi-it ka-ap-ra-am 8 bi-ri-it Ha-ar-bé-e ki u` A-ia-bi-i ki 9 e-p´ı-iˇs. ˘ ARM II 62:6 . ARM XXVI 419:17 ; i-na-an-na mu-u´ ˇsa iˇs-tu .si-it ni-ˇsi i-na E´ DINGIR ˇse-tu u-ul ´ i-baaˇs-ˇsu-u, ´ “Now, (I have restored) in this temple the water that was lacking since the departure of the (its?) dependents.” This translation and interpretation follows the one first proposed by Joannès: “Maintenant, l’eau qui manquait depuis le départ des gens, de ce temple, moi je [l’ai rétablie]” (AEM I/2, p. 307). Durand has now proposed to translate, “En réalité, en faisant apparaˆıtre l’eau dans ce temple, ce qui ne s’était jamais produit depuis la création du genre humain . . .” (1993a). He argues that nothing shows the abandonment of the temple by its personnel, and nothing in the text mentions the restoration of an old well. Indeed, nothing has been said (before this?) about the personnel leaving the temple, but without water, this would not have been surprising. Whether or not the well is new, the temple surely must have had a prior water supply. I would like to see a precedent for niˇsu¯ with the verb was.uˆ m, referring to creation of humanity. I found only one form without this genitive of possession: ARM IV 24:13, where the context gives a clear point of reference. An attacker has killed every male in a village, but taken “the dependents [and] its property” (ni-ˇsimeˇs [u` b]a-ˇsi-is-su). ´ ARM IV 24:13. ARM XXVI 126:18, 22. ARM XXVI 365:38. See in Frayne 1990, texts E4.3.6.2:16/17; E4.3.6.17:20–3 (both Hammurabi); ˘ and E4.3.7.2:41–4; E4.3.7.5:20–1 (both Samsu-iluna). ki See E4.3.6.2, where the Sumerian has zimbir (line 16), where the Akkadian has ni-ˇs`ı ZIMBIRki (line 17); just “Sippar” vs. “the people (dependents) of Sippar.” The noun is derived from the verb ˇsukênum (ˇsuhehunum), “to prostrate oneself, ˘ 1, ˘ 2). to submit, to do obeisance” (see CAD s.v. ˇsukênu While the muˇskênum do indeed confirm Steinkeller’s definition of a stronger kingship in Semitic-speaking Mesopotamian traditions, the very category acknowledges a significant population free of the king’s direct economic control (1993, 121n38). AEM I/1, p. 186n25, and LAPO 16, p. 221. See also LAPO 17, pp. 523–4. Kraus 1958, 144–55; 1973, 92–125. In general, Kraus’s definition is widely accepted: see, recently, Stol 1997, 492–3; Schloen 2001, 285–7. There have been dissenters, including an early attempt by von Soden to maintain the identity of the aw¯ılum as a universal “Burger,” ¨ so that the muˇskênum “stellt eine Sondergruppe in der wirklichen oder postulierten Gesellschaftsordnung dar” (1964, 134). Buccellati offers a reading of the muˇskênum and the aw¯ılum as “homesteader” versus holders of property beyond their inherited family plot, who are aw¯ılum (1991, 92–3). This analysis appears to be too narrowly drawn, however. See, esp., ARM XIII 117+:7, where the whole district of Terqa is divided between palace and muˇskênum. We find the same universal division for the town of T. abatum in the district of Qat.t.unan in XXVII 101:16–17. Notice also ARM II


183. 184. 185. 186. 187.

188.

189.

190. 191.

192. 193. 194.

195. 196. 197. 198. 199. 200. 201. 202.

203.

283

61:25–6; VI 3:10–11, 17; XIV 81:38–9; XXVII 14:42–5; 26:11–16; and 102:13– 14, 40–1. All of this evidence comes from the reign of Zimri-Lim. ARM II 80:9–10; XXVI 181:13–15; XXVII 30:10 -11 ; 43:9-10; cf. FM II 69: 12–15. A.3051:5 (Guichard 1997, 306), “sheep of the muˇskênum.” One text even distinguishes between muˇskênum and palace dogs (ARM XIV 39:13–16). ARM XXVI 58:12–13 and A.687 (Villard 1990, 574), boats; ARM XXVI 115:3 – 4 , slaves; XXVII 37:32–5, wagons for harvest. Without cooperation between all parts of the population and government, it is the muˇskênum who will go hungry (XIV 14:20). ARM X 33:16, GEME2 mu-uˇs-ke-ni-im. This reading comes from Durand, who translates, “qui suis-je, moi? la servante d’un particulier?” (LAPO 18, no. 1230, p. 444). ARM XXVIII 77:28, mu-u[ˇ ´ s]-ke-né-ku. For the unidentified sender, not IbalAddu, see Fleming 1999, 171. The letter probably comes from one of the kings of Zalmaqum. ´ FM VII 6:5, [E]-it 1 a-wi-lim lu´ I[a]-hu-raki , with the reading of E´ (“ house”) ˘ ˇsa-ti. The reading of the tribal name is ´ a-wi-lim confirmed by line 13, i-na E-it uncertain; the first sign is unclear, and the text would not provide the normal form of the name. AEM I/1, p. 169 note a. Durand does not identify the sender or the recipient. ´ meˇs 33 su-ku-nu-ma u-sa-ka-an A.1051:32–5; 32 u` a-na-ku a-di be-el a-limki i-la-kam LU ´ 34 h a-di DUMU-ka i-la-kam 35 mu-uˇ u` li-ib-bi mu-uˇ ´ s-ke-na-am su-ku-nu´ ´ s-ke-nim u-ni-i ma su-ki-in. Durand gives the ˘two segments of text from lines 32–4 and (34?)–35 separately, so it is not clear whether any text comes between the two sentences of my translation. It appears not. ARM XXVII 1:24. ARM XIV 121:39 and 43. ARM V 81:5–7; 5 UDUhá ˇsa Ha-nameˇs u` mu-uˇs-ke-nim 6 ˇsa A-ah Pu-ra-at-tim 7 a-na ˘ ˘ LAPO 17, p. 473 na-ha-li i-te-bi-ir. Understand ˘na-ha-li as “wadis,” with Durand, ˘ ˘ (no. 723), note b (vs. CAD s.v. nah¯alu B, “to inherit”). ˘ ´ da-an-nu-um-ma ARM XXVII 25:12–14; 12 LU ˇsa ki-ma ˇse-em i-ˇsu-u´ wa-ˇsi-ib 13 lu´ muuˇ ´ s-ke-nu-um en6 -ˇsu-um ˇsa ki-ma la i-ˇsu-u´ 14 a-na na-ri-im it-ta-al-kam. ARM XXVII 26:15–19; 15 ˇsa i-na mu-uˇ ´ s-ke-nim ˇsa ha-al-s.´ı-im 16 ba-lum ˇsa-li-ka a-na 17 ˇ 18 ˘ .se-ri-ia 19 ˇsu-re-eˇs-ˇsu. ma-a-at Su-bar-tim it-ta-al-la-ku .sa-ba-as-su-ma a-na ´ ARM XXVII 27:28–9; cf. 30:6 –9 . ˇ ARM XXVII 26:17–19. Lines 20–7 elaborate on why people flee for Subartum, especially at night, showing again their need for authorization. ARM XIV 48:27–9. ARM XXVII 39:20 –1 , 28 ; cf. XXVII 37 and 38. ARM XXVII 100, esp. lines 6–8 and 22–3. ARM XXVII 107:1 –9 . In lines 14 –20 , the second military category is called ´ ˇsu-ut SAG, who carry for the king the heavy bronze spear (giˇs SUKUR ˇ LU ZABAR dan-nam). ´ BURAARM XXVII 107:2 –9 ; 2 ki-ma lu´ p´ı-ih-ri-i[m] 3 u` lu´ ˇsu-ut SAG ˇsa GU 4 ˘ ´ ˇ ` ` NUN.NA [ˇs]a 5 GAN A.SA.AM .sa-ab-tu u` il-kam dan-nam 5 a-na e´-kál-lim i-il-la ki ´ A.Sˇ A. ` AM ` 7 .sa-ab-tu ki-ma ˇsu-nu-ti 5 [k]u u` a-lik-ti 6 ha-la-as. Qa-at.-t.u-na-an 5 GAN ´ ´ A.Sˇ A. ` AM ` ˘ˇsu-ku-sa-ti-ku-nu 8 lu-ma-li-ku-nu-ti-ma ˇsa-p´ı-il-ti A.Sˇ A ` 9 DUMUmeˇs GAN

284

204. 205.

206. 207.

208. 209. 210. 211.

212.

213. 214. 215.

Notes to Pages 144–145 ´ A.Sˇ A. ` AM ` lu-uˇs-tam-li. On pihrum as the regular troops who ma-a-tim 3 GAN ˘ are “mustered” (or “conscripted”) from the general population, see LAPO 17, p. 349 note a, as well as p. 362. For the ˇsu¯ t r¯eˇsi as “royal guards,” see ARM XXVII, p. 186 note j. ` há ˇsu-u´ A.Sˇ A ` e´-kál-lim-ma. Line 22 : A.Sˇ A ˘ Notice that under the earlier Yahdun-Lim, the “sons of the land” category is ˘ serve the Mari king from various hals.um already used to define soldiers who ˘ “districts.” FM II 38:10–11, from Yaqqim-Addu, a badly broken text. 28 i-nu-ma te-re-tim i-p´ı-ˇsu-ma 29 i-ˇsa-ru-um-ma i-ˇsa-ra 1 QA ˇse-em 30 u-ul ´ a-na-ad-di-in lu.meˇ ´ s mu-uˇs-ke-nu 31 ˇsa ha-al-s.´ı-ia i-ˇse-mu-ma ap-pa-ˇsu-nu 32 i-ma-ha-s.um. ´ I include ˘ ˘ this text for the interesting turn of phrase at the end, though there is no direct speech. For identification of Hammanum, see the “Note” to the text. The sender of the letter is lost but is˘ bringing the palace demand to the governor. Notice also line 26, where this palace official anticipates the complaints: “ZikriHanat should take omens so I can take the 50 ugar of barley and we can put ´ s ˘ end to the complaints of the muˇskênum” ([t]a-zi-im-ti lu.meˇ an mu-uˇ ´ s-[ke-ni]m i nu-ha-al-li-iq). 27 ˘ ni-nu wa-ˇsa-ab-ni. Notice the western-style first-person-plural suffixed form of the verb, which is quoted directly. 26 i-na-an-na mu-uˇs-ke-nu-um [x x]-x-ma 27 u-ul ´ i-ˇsa-al um-ma-mi i[ˇs-t]u er-bu-um 28 ˇ SE-ti-ni i-ik-ka-lu ni-nu an-ni-ki-am 29 wa-ˇsa-ba-am ni-le-i. Line 47, bi-il-tum ik-ta-ab-da-an-ni-ˇsi-[im]. See ARM XIV 48:45–50, on whether they will accept demand to use their oxen; XXVII 39:8 , the governor boasts of his prowess in negotiations with them; XXVI 412:17–18, 418:11 –12 , FM II 71:22, and FM II 88:28–30, for their capacity to rebel or accept rule peaceably. See ARM II 80:10; XIV 12:3 ; XXVI 154-bis:10, 26, 30. It is not clear whether the occasional omission of mimmation indicates a proper plural. One evident plural appears when the term is used to identify the class of two specific men, object of an anticipated royal decision: “be they (palace?) servants or muˇskên¯u, they should be brought to me” (lu-u´ .se-he-ru u-lu ` mu-uˇ ´ s-ke-nu li-it-ru-ni-ˇsu-nu-ti, ˘ uˇ ARM XIII 141:22–3). The odd DUMU mu´ s-k[e-nim] is found only in similar circumstances, to designate the class of an individual who faces accusation of a crime against king Zimri-Lim (XXVI 44:10). Notice that aw¯ılum (“gentleman”) is not generally collective but is given a plural form. ARM XXVIII 179:18–20. This excludes the muˇskênum who are proper to those outside lands, viewed as the subjects of those kings, as in ARM XXVI 412:17–21 at Qat.t.arâ. ˇ FM VII 47:46–52, letter from Zimri-Lim’s secretary Sunu hra-halˆ u, reporting to ˇ ˘A-k]i ` ˘ u` E´ há -ki -ma-a the king from his mission in Aleppo: 46 pa-na-nu-um A.S[ ˘ 47 ma-ti-ma ta-ad-di-ni u` aˇs-[ˇsum ˇsa Nu-u]r´ d SUEN 48 a-na mu-uˇ ´ s-ke-nim i-qa-ab49 h´ a h a ´ ˇ bu-u´ [um-ma-m]i a-na KUR-i a-na GISˇ al-ka [AN]SE[ ˘ ˘] -[k]u-nu [ˇsu-up]-ra 50 tap-pu-ti al-ka an-ni-tam Nu-ur´ d SUEN il-q[´ı-ma i]q-bi-ma 51 u` mu-uˇ ´ s-ke-nu-um k[i-a-a]m i-da-ab-ba-ab-ˇs[u-um um-ma-mi] 52 ki-ma [ˇs]a f Ga-ˇse-ra-ma re-ˇsu-ta-am n[ila-ak(?)]. I present here the published text and restorations by Durand, but the tablet is badly damaged. The closing quote is not clear. The shift to the verb


216. 217. 218. 219. 220. 221. 222.

223.

224.

225.

226.

227. 228. 229.

230. 231.

285

dab¯abum suggests a shift of tone, perhaps as a complaint, because the noun r¯eˇsu¯ tum means “slavery,” not just royal service. There was one other major town in the district, T. abatum, further upstream, just before the juncture of the Habur with its tributaries. ˘ see Villard 2001, 20–1. ARM V 25:5–10; on Tarim-ˇsakim, ARM XIV 12:3 –5 . On the pagrau¯ as rites of offerings for the dead, see Durand in Durand and Guichard 1997, 35. FM II 38:9–11, cf. line 4. A.2819:6–8, in Durand and Guichard 1997, 34 (cf. 32). ARM XXVI 85:9–10, by Asqudum (ni-q´ı-im ˇsa mu-uˇ ´ s-[ke-nim]); and XXVI 109:4– 5, by Ibbi-Amurrum (SISKUR2 .RE ˇsa lu´ mu-uˇ ´ s-ke-nim). ARM XXVI 1:3–12; 3 [ . . .i-na te-re-et mu-u]ˇ ´ s-ke-nim 4 [i-na ne-pé-eˇs15 -tim ma-li i]ˇs5 ˇsa-ka-nu-ma a-am-ma-r[u] [UZU le-em-na-am u` la da]m-qa-am ma-li a-am-ma-ru 6 [a-na m Zi-im-ri-Li-im be-l´ı-ia] lu-u´ a-qa-ab-bi la a-ka-at-ta-mu 7 UZ[U l]e-[em-na-am u` la dam-qa-a]m ˇsa i-na te-re-et 8 m Zi-im-ri-L[i-im be-l´ı-ia i-n]a uzu iz-bi-im u` i-na uzu IZMI-im 9 iˇs-ˇsa-ak-ka-na-ma a-am-ma-ru 10 a-na DUMU a-wi-lu-tim ˇsum-ˇsu la a-qa-abbu-u´ 11 u` a-wa-tam na-s.[´ı-i]r-tam ˇsa a-na te-re-e-tim e-pé-ˇsi-im 12 m Zi-im-ri-Li-im b[e]-l´ı i-qa-ab-bé-e-em. This is the basic text of Durand’s publication, with restorations left as he proposed them. Adélina Millet-Albà (forthcoming) calculates that the census of Zimri-Lim’s sixth year counted roughly 7,620 Yaminites in the kingdom, and reflected a total Yaminite population of about 10,000, perhaps one-fourth of the entire Mari population. This conscious recall of Yahdun-Lim’s Mari kingdom is all the more evident ˘ of Zimri-Lim that casts him instead as the son of when set against the early seal Ha-at-ni-d [x] (Charpin 1992d, 72). ˘ A.96:6–7, 10-11, in Joannès 1991, 167–8; M.6435+:17–18, 25–6, cf. 12, in Durand 1986, 111–14. The Hammurabi text appears to be a Mari draft, not ˘ 115). the actual treaty document (p. M.6060:16 –18 , “the oath that I have sworn to Zimri-Lim my lord, son of Yahdun-Lim, king of Mari and the m¯at Hana”; M.7259:12–15, “if Asqur-Addu ˘ son of Yahdun-Lim, [king] of Mari my˘ lord rebels, I will surely tell Zimri-Lim, and the m¯at Hana”; both in Durand 1991, 50–2, 48. ˘ An oath text ˘for people defined simply as “Hana” may be coherent only at a time when they constitute a m¯atum, a political˘ unity. See the “protocole des bédouins” in Durand 1991, 50–2, for the opposition of “the hana of the steppe” to “the men of the towns” (M.6060:22 –3 ). ˘ M.6060:20 –7 , in Durand 1991, 50–2; 20 [a-n]a Zi-im-ri-Li-im be-l´ı-ia 21 [la u]´ 22 lu´ meˇs ga-la-lu u-lu [i-na(?)] p´ı-i Ha-na ˇsa na-we-e-em 23 ´ a-wa-tam la da-mi-iq-tam ˘ ´ meˇs ˇsa a-la-ni 24 [ˇsa e-ˇse]20 -mu-u´ um-ma-mi [u-lu]-ma i-na p´ı-i LU Zi-im-ri-Li-im 25 ´ 26 [u` p´ı]-ri-ih-ˇsu u-ul i-ˇsa-ap-p[a]-r[u]-né-[ti] [i-n]a ma-ru-uˇs-ti Zi-im-ri-Li-[im u] ´ ` 27 ˘ su u-ul [p´ı-i]r-hi-ˇ ´ uˇs-ta-mar-[ra-as.-ma] . . .(broken). Thus, the˘merhûm Ibal-el reports that nawûm and Simal are well, e.g., ARM II 33:21 –2 ; see ˘also A.1098:6 –8 , in Villard 1994, 297 and n. 33. ARM II 59:4–13, letter to Zimri-Lim from Kabiya, king of Kahat in Ida-Maras. ; 4 7 ˘ na-wu-u-um i-na ri-tim ´ ˇsa Ha-na 5 [ˇs]a i-na ha-al-s.´ı-ia 6 i-ik-ka-lu ˇsu-ul-mu-um 8 9 10 11 12 ˘ ˘ me-e u` i-na di-nim i-ˇsa-ri-iˇs ap-lu a-na na-we-e-em ˇsa Ha-na u` a-na a-lim

˘

286

232.

233.

234. 235.

236.

237.

238. 239.

Notes to Pages 151–153 Ka-ha-at ki 13 ˇsu-ul-mu-um. Notice the use of the noun ˇsulmum (“well-being”) ˘ of the verbal form ˇsalim (“is well”), perhaps by local preference. instead ARM III 15:9–15, letter to Zimri-Lim from Kibri-Dagan, governor of the Terqa 10 11 district; 9 as-su-ur-re i-nu-ma na-wu-u-um ´ ´ ˇsa lu´ Ha-nameˇs 12 aq-da-ma-tam ˇsa 15 ˘ ib-ba-aˇs-ˇsi. ÍD.DA 13 i-ka-lu 14 na-ak-rum i-ma-aq-qu-ut-ma hi-t.i4 -tum ´ ˘ In ARM XXVI 358:13 –14 , This is true of the king of Kahat in ARM II 59, above. ˘ Zimri-Lim is supposed to decide whether the Hana are to be removed from ˘Se ˇ ˇ hna. It is not clear whether the land of Apum, whose capital is Subat-Enlil/ these Hana are Simalite, and the writer appears to˘ come from Apum itself. ˘ See A.4530-bis:6 , in AEM I/1, p. 182; ARM XXVI 220:7–10; XXVII 17:8–9. A.350+:4–8, letter to Zimri-Lim from the merhûm Ibal-el (in Charpin 1990b, ˘ 5 u` ki-ma lu´ tam-ka-ri-im ˇsa bi120–2); 4 be-l´ı i-de-e ki-ma Ha-nameˇs ˇsa-a[p-ra-ku] 6 7 ˘ r[i]-i[t] nu-ku-ur-tim u` sa-li-[mi-im] i-la-ku-ma{me} Ha-name i-na ˇse-pa-a-t[im ´ ˘ bi-ri-it] 8 nu-ku-ur-tim u` sa-li-mi-im i-la-[ku]. ´ ARM XXVIII 79:35–9; 35 [um-ma at(?)-t]a-ma Z[a-al]-lu-ha-an u-ul DUMU I´ da-Ma-ra-as. 36 [DUMU Si]-im-a-al be-l´ı a-na m I-ba-al-p´ı-el 37 ˘[l]i-iˇs-pu-ra-am-ma itt[i] ha-nameˇs 38 DUMUmeˇs ha-al-s.´ı-ia a-na IGI pa--im li-it-ta-la-ku 39 u` ri-it˘ idiom, rittam + eli (with a person) + ˇsak¯anum, ta-ˇsu ˘e-li-ia li-iˇs-ku-un. The last seems equivalent to the same idiom with q¯atum, “hand,” in ARM X 78:25–6 and 153:17–18. In both cases, an administrator takes final responsibility for decisions regarding some individual’s situation (see CAD s.v. ˇsak¯anu 5a q¯atu c), and rittu A 1a7 , human hand (for reference to this text with literal translation). For the text, see also Durand 1987b, 230. A letter from Yaqqim-Addu, the governor of the Saggaratum district, cites a previous letter from Zimri-Lim that said, “The Hana have been held in the ˘ The king appears to be villages for the corvée of oxen” (ARM XIV 80:5–6). responding to a complaint from the Hana leadership that the Terqa governor ˘ should have no right to keep them from leaving with their merhûm for their seasonal movements (cf. Durand, LAPO 17, no. 742, p. 504). ˘ ARM XXVII 1:33, 45–6; perhaps cf. XXVII 16:43–4. Lines iii:3–21; 3 i-na ˇsa-at-tim-ma ˇsa-a-ti 4 m La-u-um LUGAL Sa-ma-nimki 5 u` ma´ 6m at Up-ra-p´ı-im Ba-ah-lu-ku-li-im LUGAL Tu-tu-ul ki 7 u` ma-at Am-na-ni-im ki 8 m ki 9 ˘ A-ia-lum LUGAL A-ba-at-tim u` ma-at Ra-ab-bi-im 10 LUGAL an-nu-tu-un 11 i-ki-ru-ˇsu-ma 12 a-na ti-lu-ti-ˇsu-nu 13 .sa-ab Su-mu-e-pu-uh 14 ˇsa ma-at Ia-am-ha-ad ki 15 18 ˘ ˘ s ip-huil-li-ka-am-ma 16 i-na a-li-im Sa-ma-nimki 17 um-ma-at DUMU-mi-im iˇs-ti-ni-iˇ 19 20 meˇs 21 ˘ ru-ˇsum-ma i-na ka-ak-ki-im da-an-nim 3 LUGAL an-nu-ti-in ˇsa DUMUmi-im ik-mi. See Frayne 1990, pp. 604–8, E4.6.8.2, for bibliography and the full text. Abraham Malamat (1979) connected the unfamiliar substantive umm¯atwith the Hebrew ummâ, a tribal unit of the Midianites (Numbers 25:14) or the Ishmaelites (Genesis 25:16). He is surely correct, in that the word appears also in Ugaritic with some such meaning, in the form umt. He then translates umm¯at Marmˆım in the Yahdun-Lim inscription as “the mother city of the Yaminites.” ˘ Hebrew and the Ugaritic examples, however, the cognate In both the biblical words refer to people, not places, so “mother city” is somewhat unexpected. It is possible to read the whole phrase in genitival relation to the central town, rather than in apposition, so that the noun ummat(um) need not be the town itself.


287

240. Lines iii:17 and 21. On the Marmûm and the common writing of Binu Yamina with DUMU for “son” (Akkadian m¯arum), see Durand, LAPO 17, p. 418. 241. Lines iii:28–32; 28 a-lam Ha-ma-anki um-ma-at Ha-na 29 ˇsa a-bu-u´ Ha-na ka-lu˘su-ma 31 a-na ti-li u` ka-ar-mi ˘ iˇs-ku-un-ˇsu 32˘ u` ˇsar-ra-ˇsu ˇsu-nu i-pu-ˇsu-ˇsu 30 iq-qu-ur-ˇ ´ Ka-s.u-ri´ ha-la ik-mi. ˘ 242. The location of Haman is still completely unknown; see Charpin forthcom˘ ing c. 243. Frayne 1990, E4.6.8.1:15–20; 15 7 LUGALmeˇs 16 ab-bu-u´ Ha-na 17 ˇsa uq-ta-ab-bi-lu20 ˘ a-na i-di-ia u-te-er. nim 18 ak-mi-ˇsu-nu-ti 19 ma-at-su-nu ´ ´ 244. See lines 1–5. ˇ 245. Samaˇ s temple inscription, line iv:4; and Dur-Yahdun-Lim inscription, lines 7–8. ˘ 246. RIME 4, p. 626, E4.6.12.4. 247. See his dedication of two thrones to Itur-Mer in Charpin 1984, 42–4 (Grayson 1987, A.0.39.4:5–6; cf. A.0.39.5:3–9). 248. ARM I 24+:43–5. 249. ARM V 81:5–15; 5 UDUhá ˇsa ha-nameˇs u` mu-uˇs-ki-nim 6 ˇsa A-ah Pu-ra-at-tim 7 ˘ ˘ ˇ d UTU-mu-ˇsa-lim 9 ˘u` ha-nameˇs Yua-na na-ha-li i-te-bi-ir 8 11 lu´ NA.GAD NÍG.SU 10 12 ˘ ` ki Ia-ás-[m]a-ah-d ISKUR ˇ ma-ha-mu-um u` ˇs˘a-ap-li-iˇs-ma ka-lu-ˇsu i-te-bi-ir 11 BAD 13 meˇs 14 ˘ ˘ DUMU Ia-mi-na sà-ar-ra-ru la is.-s.a-ba-t[u-ˇsu-nu-ti-ma] 15 la i-re-u´ as-su-ur-re ´ u-ga-la-lu. See Wasserman 1994, esp. p. 328, for negatives with assurre. ´ 250. “[L’absence des Bensimalites] des textes d’époque e´ ponymale pose donc un ˇ ˇ véritable problème car beaucoup de textes ont e´ té e´ crits depuis SubatSamaˇ s [sic]. La fa¸con de voir les choses change cependant lorsqu’on accepte de considérer comme Bensimalites tous les prétendus ‘Hanéens’ de ces régions” (LAPO 17, p. 477). 251. ARM VI 76:11–25; 11 [lu-u´ i]t-tum i-nu-ma 12 [i-na ka-ra-áˇs Ap]-pa-anki 13 [a-na bel´ı-ia k]i-a-am aq-[b]i 14 [um-ma a-na-ku]-ma 15 [u4 -ma-am ma]-a-at I[a-mi-n]a ki 16 18 [a-na qa-ti-k]a n[a-ad-na-at] 17 [u` ma]-a-tum ˇsi-i .s[u-ba-a]t [Ak-k]a-di-im-ma ´ 19 20 hu-l[u-pa]-at [be-l´ı q]a-qa-ad ˇsar-ru-ti-ˇs[u l]i-ka-bi-it [ki-ma] LUGAL Ha-nameˇs 21 ˘ ˘ há la ˇ at-ta [u` ˇs]a-ni-iˇs LUGAL Ak-ka-di-im at-ta 22 [be-l´ı] i-na ANSE.KUR.RA ˘ i-ra-ka-ab 23 [i-na] giˇs nu-ba-lim u` anˇse.há ku-da-ni-ma 24 [b]e-[l´ı] li-ir-ka-am-ma qa-qa˘ 25 ad ˇsar-ru-ti-ˇsu li-ka-bi-it an-né-tim a-na be-l´ı-ia ad-bu-bu. See Durand, LAPO 17, no. 732, with notes on l¯u ittûm (line 11) and n¯ubalum (line 23). For the n¯ubalum as “chaise a` porteurs,” or “litter,” see LAPO 16, pp. 236–7. Charpin proposes that the seat may be carried by animals such as the mules of ARM VI 76 (1993b, 170n35). This interpretation is distinguished from the identification as a type of wagon or chariot, as still proposed by Groneberg 1990, 164–5. 252. This designation is not otherwise known to me, and Durand cites no reference, though it is not impossible that the tribal confederacy would be given such a definition. The reading must remain tentative. 253. Durand, LAPO 17, pp. 486-7, citing ARM XXVI 131 and 312; ARM XXVII 16. Only the last text identifies these with Eˇsnunna in particular, which could have been identified with Akkad. 254. Lafont identifies the donkeys or mules by their long ears. 255. FM VII 20:6 –7 , letter from Yaqqim-Addu, governor of Saggaratum; 1 ˇ ANSE.KUR.RA pé-s.u-um 1 MUNUS.EME5 .KUR.RA pé-s.e´-tumx (TAM), male and ´ female.


288

256. The idea that the Yaminites are in view could be supported by the reference to Kulhitum in line 6, an Amnanˆ u Yaminite town in the district of Terqa (see ˘ ARM XXIII 428:20 and 429:20). [Ap]pan in line 12 would be the Yaminite town of the “Awin” clan in the Rabbˆ u tribe of ARM VIII 11. 257. See A.3960:2 –4 , in Durand 1990b, 50–1 and n. 54. 258. A.3960:2 –12 , letter evidently from Lanasˆ um, the military deputy of Zimri-Lim at Tuttul (in Durand 1990b, 50–1 and n. 54); 2 [DUMU]meˇs Ia-mi-na ka-lu-ˇsu 3 4 5 i-na a-la-ne-ˇsu-nu u` ma-ti-ˇsu-nu ˇsu-s.uit-ta-na-ag-gi-ˇsu i-na-an-{NA}na i-na ´ u-ma ´ l[i]-i[b-b]i-ˇsu-nu ki-a-am i-ka-ap-pu-du 6 um-ma-m[i a-lam T]u-ut-tu-ul ki 7 a-na da a[n-na-tim] i ni-is.-ba-[a]t-ma 8 lu-u´ i-n[a] x x x x ˇsu-u´ 9 u` i-na mu-uh-hi I-ma-a[rki ] 10 giˇs ˘ ˘ su u-ˇ KAK i ni-im-ha-[a]s. 11 an-ni-tam DUMUmeˇs Ia-mi-na 12 i-na li-ib-bi-ˇ ´ sa-am. ˘ 259. ARM XXIII 428:34 and 429:35. 260. Anbar 1991, 90; the precise percentages seem rather artificial, and it is not necessary to accept his interpretation of the Hana. ˘ 261. A.489, in Durand 1992b, 113n137; LUGAL ak-ka-d[i-i]m u` a-m[u-u]r-ri-im. No lines are cited. 262. This family history will be considerably illuminated by Durand’s study of the “Eponym Chronicle” of Samsi-Addu, a badly broken but fascinating text that has still not been mined for its historical content (cf. Birot 1985). The text was ´ the subject of several meetings of Durand’s seminar at the Ecole Pratique des ´ Hautes Etudes in the fall of 1997. 263. Eidem observes that Samsi-Addu’s administration left little impact on local social structures (2000, 256). 264. See the section on Babylon in Frayne 1990, pp. 323–70. 265. ARM XXVI 385:3 –8 , letter with missing address; 3 a-w[a-a-tim] 4 an-né-e-[tim] d 5 meˇs la da[m-qa]-tim Ri-im- SUEN uˇs-[te-s.´ı] ul-la-an DINGIR r[a]-bu-tim ˇsa til-lu-ti [il-li-ku] 6 u` Zi-im-ri-Li-im LUGAL DUMU Si-im-a-al 7 ˇsa ba-la-t.am u` ba-la-t.am-ma 8 it-ti-ia i-pu-ˇsu ˇsa-nu-u-um u-ul ´ ´ i-ba-aˇs-ˇsi. This text was edited and translated by Charpin. Anbar (1996) proposes the alternative translation in order to reserve the possibility that Zimri-Lim is not himself a Simalite, against the interpretation of Charpin and Durand, but he only rules them. From the perspective of the social structures and the political affiliations that frame the realities of Zimri-Lim’s rule, the choice makes little difference. Nevertheless, it seems far simpler to suppose that Zimri-Lim shares the tribal heritage of his Simalite people. 266. A Mari official who announces victory to Zimri-Lim mentions only the prowess of the Hana, the Simalite bedouin contingent, as if no other Mari forces ˘ contributed (ARM XXVI 386:3 –4 ); cf. AEM I/2, p. 213, on no. 389. This concrete point of reference would contribute to the orientation of Babylon’s king. ki ´ 267. A.3577:20 –33 , in Durand 1992a, 45 and n. 39; 20 LUGAL KA.DINGIR.R[A s.a-ba-ˇsu(?)] 21 [a-na .s]e-ri-ia [iˇs-pu-ra-am] 22 [u4 -m]a-am ma-ti-ma iˇs-tu pa-na a-na wa-a[r-ka-ma] 23 [m ]Zi-im-ri-Li-im tap-pu-ut Nu-um-h[a-aki ] 24 [u-u]l il-li-ik ´ 25 ˘ i-nu-ma Qar-ni-Li-im tap-p[u-tam] [ˇsa n]a-ba-al-ku-[tim] i[l-l]i-i 26 [i-nu-m]a A-tam-ri-i[m iˇs-n]é-em i-tu-ur ´ i-na-an-n[a] 27 [tap-p]u-ut Hi-im-di-[i]a-[m]a i-la ˘ ´ s ´ ak 28 [i-na-a]n-na lu.meˇ KA.DINGIR.RAki qà-aq-qa-di 29 [uˇ s-te]-em-me-ed-ma it-ti 30 DUMU Si-im-a-al a-na-ki-ir [N]u-[u]m-hu-um ki-a-am i-pu-ul-ˇsu um-ma-a-m[i] e ki 32 ˘ ´ zu-ub 31 a-hi DUMU Si-im-a-al it-ti K[A].DINGIR.RA a-sa-al-li-im it-ti DUMU

˘


268.

269.

270.

271. 272.

273. 274.

275.

276.

277.

278.

279.

289

Si-im-a-al-ma a-ba-a[l-lu-ut.] 33 u` a-ba-al-lu-ut. ki-ma an-ni-tam ma-as-su´ [i-pu-lu-ˇsu]. I follow Durand in translating the Akkadian first person singular with the plural. ARM XXVII 19:14–16, ba-lu-um m Zi-im-ri-Li-im u` Ha-nameˇs u-ul a-sa-lim. The ´ ˘ writer is the Mari governor of Qat.t.unan, so we cannot be sure whether the terminology reflects his own or Bunu-Eˇstar’s words. Kurdâ maintained close relations with both Mari and Babylon, who did not oppose each other until the end of Zimri-Lim’s reign, and king Simah-ilanê even visited Babylon during the year Z-L 1 (see ARM XXVIII, p. 235). ˘ M.6182:17–26, in Durand 1991, 26; 17 DINGIRmeˇs an-nu-tum ma-l[a] 18 p´ı-ir-ih 22 i-na˘ ˇsu-mi-i[a] 19 u` tu-s.a-[ti] 20 li-ha-al-li-[qu] ´ 21 DUMU Si-im-a-al ru-[bu-su-nu] ´ 23 24 25 ˘ qa-ti-[ia] li-mu-[ur] um-ma-a-mi Zi-im-ri-L[i-im] a-na Su-mu-ha-du-u´ u-[t ´ . `ıib(?)] 26 u` Su-mu-ha-du-u´ ba-[ . . . ]. This is the only loyalty oath for ˘a high Mari ˘ I have followed Durand’s restorations, except in the last official. In general, line, where I hesitate to assume two separate spellings of the governor’s name. ´ belongs to the name, then we must be even less sure about the end If the U of the line. Durand restores u-ba-[ar-ˇ su] and translates, “et Sˆ umu-Hadu s’est ´ révolté contre lui.” It is not clear what title or titles Sˆ umu-hadˆ u may have held, ˘ at Mari. See the though he seems to have served briefly as district governor discussion in Lion 2001, 176–7, 183–4. See Grayson 1987, pp. 47–51, ma-tim bi-ri-it ÍD.BURANUN.NA, A.0.39.1, a text found at Aˇsˇsur (cf. A.0.39.7, on an object dedicated to Dagan found at Mari). See for now Durand in Durand and Guichard 1997, 28; Durand 1998, 108– 9; the paragraph on “l’époque d’Akkad” in Charpin, RAI 43, 106–7; Charpin forthcoming b; and Eidem and Hoejlund 1999. FM III 4 i:5–6, 18–19; see Durand’s discussion on p. 43. LUGAL A-ga-dè ki ; A.4509:7, in Charpin 1984, 44–5. When Samsi-Addu restored the Emenue shrine in the Eˇstar temple of Nineveh, he shows concern for maintaining Sargonid honor, by recognizing that it was originally built by Maniˇstuˇsu, son of Sargon (RIMA 1, A.0.39.2:i 7–13). Compare, for example, from the royal inscriptions compiled in Frayne 1990, the Isin kings from Iddin-Dagan on; many Larsa kings; and the Babylonian kings after Hammurabi (who only claimed this with some reason after conquering Larsa, ˘but whose successors continued in any case). See, e.g., the titles claimed in Samsi-Addu’s votive inscriptions (Charpin 1984, texts 1–4 [pp. 42–8], and the seals presented on pp. 50–1, with a chart of his titulary on p. 52). Cf. Grayson 1987, section A.0.39.1. Durand describes the kingship of both Yahdun-Lim and Zimri-Lim in terms ˘ Der, the “campement nomade” of their conquest of Mari, when he discusses where both kings go to affirm their power, with their kingship “définie a` la fois par la possession de Mari et l’autorité sur des tribus bédouines” (in Durand and Guichard 1997, 40). See Durand 1985b, 421. Nele Ziegler has now studied this phenomenon in greater detail and finds that under Zimri-Lim, the women from Yasmah-Addu’s ˘ (1999, harem tended to be kept apart from those who arrived with the new king 36–8 and table on p. 37). Villard 2001, 13. Mari territory under Yasmah-Addu was defined around the Banks-of-the-Euphrates, beginning with the ˘districts of Mari, Terqa, and

290

280. 281.

282. 283.

284. 285. 286.

287. 288. 289.

290.

291.

292.

Notes to Pages 162–167 Saggaratum. The young king wanted to add the Habur district of Qat.t.unan, ˘ By this time, Ida-Maras. but his father did not allow it until later in his reign. may also have come to be responsible to Yasmah-Addu. ˘ article in RAI 43, esp. pp. 96– Charpin discusses several expressions of this in his 7, 109. His allies and vassals viewed things the same way. For reference to major renovations at the palace during Yasmah-Addu’s first years at Mari, see e.g., Charpin’s comments in AEM I/2, p. 10, on˘the letters of Us. ur-awassu. Even as the m¯at Hana, this kind of titulary is not inclusive but exclusive. ˘ distribution of Zimri-Lim’s royal correspondence in ARM This is visible in the XXVIII. After numbers 1–25 cover the great kingdoms of Babylon, Qatna, Yamhad, and Carchemish, numbers 26–176 are organized to move from west ˘ across the upper country from Zalmaqum in the Balih to the Sinjar to east region. Most of these come from Zimri-Lim’s vassals, especially˘ in the Habur ˘ and Sinjar areas. After these we have only numbers 177–81, from the eastern tribal people called the Turukku, ´ and the great kingdom of Elam. It is possible that some of the correspondence involving the great powers was culled out by Babylonian-employed scribes before the palace was destroyed, but it may be best to accept the existing distribution as the natural result of Zimri-Lim’s day-to-day political contacts. See Charpin 1995a, 37–8. For a broad overview of royal activities at Mari, peruse the explanatory introductions in Durand’s three LAPO volumes. See ARM XXVI 5 and 6. See Durand and Guichard 1997, 52–71, texts FM III 2–5. These rites show no sign of having been carried out at Mari, and all come from the reign of SamsiAddu (Fleming 1999, 160–1). The texts still show the central ritual role of the king assumed by the whole royal system under Samsi-Addu. ` e´-kál-lim, in ARM XXVII 36:9, in the district E.g., “the fields of the palace,” A.Sˇ A of Qat.t.unan. LAPO 17, pp. 513–16, on the hamqum. It does not seem possible to˘ assume that Mari matches Piotr Steinkeller’s model of a standard upper Mesopotamian kingship, in which the rulers and their extended families were the chief proprietors of agricultural land (1993, 124–5). See, e.g., Lafont 1992b, 93–101, for problems with a canal near Mari that were addressed by a high palace official named Sumu-hadˆ u. Lafont (2000b, 138) underlines the centralized power that seems to have˘ been required for the new irrigation projects of the early second millennium, though the dates of canals discovered for this period cannot be determined precisely. I find the analysis of David Schloen (2001) to be quite helpful with regard to both land and employment. Schloen discusses palace employment in terms of part-time service (pp. 223–5), and land in terms of a “hierarchy of rights” (pp. 230–1). I would like to thank M. Lafont for his helpful response to my inquiry regarding my general observation. Out of the five letters that he cited to me, four of them come from the time of Samsi-Addu (ARM I 118; IV 80; V 1; and V 15). Only one (ARM XIII 36) was written to Zimri-Lim, regarding the flocks of the palace. For the role of the palace in shearing, see Durand, LAPO 17,


293. 294.

295.

296.

297.

291

pp. 670–6 (three letters, from both reigns). This question deserves further study. It would be interesting to know how Yahdun-Lim handled Tuttul, which he ˘ made his western capital. According to ARM XXVII 26:15–19 (see pp. XXX, above), the governor of Qat.t.unan could not carry out Zimri-Lim’s order to keep all muˇskênum in the district, unless they had special permission to leave. They continued to escape during the night. A.1121+:34–45, in Lafont 1984, 9–10; 34 pa-na-nu-um i-nu-ma i-na Ma-ri ki wa-aˇsba-ku 35 lu´ a-p´ı-lum u` a-p´ı-il-tum mi-im-ma a-wa-tam 36 ˇsa i-qa-[ab-b]u-nim a-na be-l´ı-ia 37 i-na-an-na i-[n]a ma-a-tim ˇsa-ni-tim wa-aˇs-ba-ku 38 ˇsa {x} e-ˇse-em-muu-ta-ar ´ a-ˇsa-ap-pa-a-ar 40 ˇsum-ma ur-ra-am ˇse-ra-am u´ u` i-qa-ab-bu-nim 39 a-na be-l´ı-ia u-ul ´ 41 mi-im-ma hi-t.[`ı-tu]m it-ta-ab-ˇsi be-l´ı ki-a-am u-ul ´ i-qa-ab-bi-i um-ma-a-mi 42 a-wa-tam ˘ iq-bi-kum u` ma-aˇs-ka-an-ka 43 i-na-az-za-ar am-mi-nim a-na .se-ri-ia 44 la ˇsa lu´ a-p´ı-lum ta-aˇs-pu-ra-am a-nu-um-ma a-na .se-er be-l´ı-ia 45 a´ [ˇs-p]u-ra-am. For all of the loyalty oaths, see Durand 1991. The diviner’s oath is ARM XXVI 1, and see 1991, 14. News of revolt is also demanded in an oath for some kind of subordinates to the king (M.7964, p. 24). See M.6060, in Durand 1991, 50–2 (the Hana); M.6182, p. 26 (a governor); ˘a). and M.7259, pp. 48-50 (the town of Karanˆ

4. The Collective and the Town 1. For the idea of the royal-ritual city, see Fox 1977. 2. My work also has a noticeably different geographical definition. Van De Mieroop’s map of Mesopotamia on p. xi extends no further west than Mari ˇ ˇ and the Habur River, and north only to Subat-Enlil and Dur-Sarrukin. 3. A.3024 =˘ B.308, lines 11–14; 11 [k]i-ma a-li-ˇsu Na-hu-urki u` ki-ma a-la-né-e 12 [ˇs]a 14 ˘ nam-la-ka-ti-ˇsu a-lum Bu-ru-un-duki 13 a-al ki Zi-im-ri-Li-im u` A-dal-ˇse-en-ni ma-ruˇsu. See also Chapter 3, note 12 (p. 272). 4. ARM XXVII 72-bis:24 –27 , 24 a-na i-di ma-a-at Ia-ás-sa-anki 25 u` a-na i-di ma-a at A-p´ı-im[ki ] 26 a-la-nu ta-ak-lu-tum i-ba-aˇs-ˇsu-u´ aˇs-ˇsum ki-a-a[m] 27 m Qar-ni-Li-im a-na ma-a-tim na-s.a-ri-im in-né-ez-ze-eb. ki ˇ 5. ARM XXVIII 80:8–9, 8 u` 7 a-la-nihá -ia ˇsa i-na ma-a-at 9 [S]u-da i-ba-aˇs-ˇsu-u´ a-bi˘ ma i-de. A.315+:16, cited in Charpin 1990a, 71, mentions “the towns that are in the land of Kahat” (a-la-nu ˇsa i-na ma-a-at Ka-ha-at ki ) and then lists them, concluding in line˘ 20 that “all of these towns are ˘in bad shape” (a-la-nu ka-luˇsu-nu pa-as-su; ´ the verb pas¯asum, “to be dilapidated”). 6. ARM II 56. See also the plural towns and “their m¯atum” for the Yaminites in A.3960, cited by Durand (1990b, 50–1 and n. 54). 7. Saggaratum under Zimri-Lim; see FM II 35 and 46, in Bonechi and Catagnoti 1994, 69 and 81. 8. The towns in question are Haradum and Bab-nahli, A.1289+: iii:8–23; see the ˘ 149–55. ARM I 39 ˘shows the addition of a town discussion in Charpin 1991a, to a m¯atum, with the verb ˇsu¯ rubum. For exchange of towns, see also ARM XXVI 405:9 –10 ; 449:58; 468:15 . 9. ARM XXVIII 50: 19 .

292


10. See FM VII, pp. 59–172. See also Jesper Eidem (in Subartu VII, p. 258) for ˇ ˇ hna a letter from Sukram-Teˇ sˇsup, king of Eluhut, to Till-Abnˆ u, king of Se ˘ (Tell Leilan), offering him the possession˘ of a town (Leilan 87–939:36– 41). 11. ARM XXVI 384, Iˇsme-Dagan complains to Hammurabi of Babylon that the Turukkû have taken three or four; XXVI 512, ˘Haqba-hammu seized five in the ˘ ˘ land of Hatnum. ˘ 12. ARM X 74. 13. For census and oath, see FM II 46 and ARM XXIII 86 (list of men swearing oath); for administration of fields, ARM XXIII 69; for grain loans by town, ARM XXIII 70. 14. A.1487+:30–7, from Villard 1987, 591–2 (cf. LAPO 17, no. 788); 30 i-ba-aˇs-ˇsi-i ÍD.DA ˇsa 2 LU ´ meˇs 31 i-ˇsa-ap-pa-ru-ˇsi iˇs-tu me-e 32 ´ıd Ba-li-ih ul-la-nu-ma sà-ak-ru ba-lum ˘ hki 35 iˇs-tu pa-na wa-ar-ki me-e 33 Tu-ut-tu-ul ki mi-nam i-ip-pé-eˇs 34 u` be-l´ı i-di Za-al-pa-a ki 36 37 ki ˘ Tu-ut-tu-ul -ma i-la-ak i-na-an-na Za-al-pa-ah am-mi-nim u-da-ab-ba-bu-ˇ su. ´ ˘ ˇ a). 15. ARM II 33:16 . Without the mouth, see also ARM X 121 (Urgiˇs and Sunˆ 16. ARM XXVI 310 and A.250, FM I, p. 94. 17. ARM XXVI 443. 18. ARM XXVI 215, and perhaps FM III 60. 19. ARM XXVI 430:18–20; 18 a-lum A-ma-azki ˇsa i-na qa-at 19 m A-tam-ri-im i-la-ku ki ˇ LUGAL-ˇsu i-du-uk-ma 20 a-na Su-na-yi-im qa-as-su´ id-di-in. 20. ARM VIII 85+, in Charpin 1997b, 343–4. I include all the first names listed before Pulsi-Addu, who represents a separate party to the legal proceeding. 21. Bailey admits that his interest derived from watching the workings of Indian village “panchayats,” as well as of university committees. 22. It appears to me that cultures without any voting tradition must resort to “acclamation,” which could be achieved only through persuasion by the more influential parties most affected. 23. See, e.g., Dandamayev 1982. For references to the legal role of councils in Mesopotamian tradition, see various articles in Joannès 2000: Lafont, 16–17; Charpin, 97–8; Michel, 113–14; Villard, 195–96. 24. I use the term “simple societies” to include the bands and tribes of the evolutionary systems found in much (but not all) anthropological theory. The “bands” and “tribes” are the most widely used categories, and come from Service 1975, chap. 3. 25. This is the purpose of the theoretical sketch in Chapter 1. 26. “Of particular interest is the further development of the idea of the corporate state by Blanton et al.” (2000, 263). I would like to thank Dr. Porter for providing me a copy of her dissertation, from which I learned much, including the Blanton et al. 1996 reference. 27. For Durand’s complex reconstruction of these events, see Durand and Guichard 1997, 32–44. It is not clear to what extent these really constitute a single religious event. 28. See Emar (VI.3) 369:1, the installation of the storm god’s high priestess; 373:169 ˇ (174), the zukru festival; and 385:2, “the inhabitants of Satappi,” for the kissu festival of Dagan. 29. This complex mix of strategies is what Blanton et al. describe in their application of the dual-processual model to Mesoamerica (pp. 7–8).


293

´ indicates the Akkadian 30. When used to render a separate word, the Sumerian LU aw¯ılum (“man” generally, or a man of a certain status). The MESˇ marks the ´ and LU.ME ´ plural. LU Sˇ can also be placed before or around other words in order to designate a group of people (especially “men”) who are identified by the marked word, often to be understood as a status, office, or profession. 31. ARM XXVIII 98:13–20. 32. ARM XXVIII 104:33–6. 33. ARM XXVIII 105:29 –30 . 34. For example, in law 24 of the Code of Hammurabi, “the town” (URU) and the ˘ rabi¯anum (ruler) are called on to pay silver to the dependents of a murdered man. The Assyrian letters from Anatolian Kaneˇs depict the mother city of Aˇsˇsur in similar collective terms. We hear of the “tablet of the city” (t.uppum ˇsa a¯ lim), “envoys of the city” (ˇsipr¯u ˇsa a¯ lim), the “verdict of the city” (d¯ın a¯ lim), and “city hall” (b¯ıt a¯ lim); see Larsen 1976, 161. 35. For the reading of the name Zalmaqay¯u, see the section that follows. ` s-nun-naki , written as a genitival ´ Eˇ 36. Otherwise, one would expect a *s.a-ab LU phrase, instead of the apposition, which suggests the common adjectival reading. ` s-nun-naki pa-ar-ku. 37. ARM I 37:26, lu´ Eˇ ki ´ 38. M.6159:9 -10 , 1 lu´ KA.DINGIR.RA GAL. 39. For comparison, it is interesting to note that F. R. Kraus identifies the earliest ` case of a-GA-dè/DI-um, in a text from Gasur, as meaning a “Burger ¨ der Stadt Akkade,” an adjectival use applied to the city itself (1970, 27). 40. See ARM XXVI 422:29; 424:29; XIV 104+:19, 27 (Charpin 1993a, 199); A.2728:15 (Dossin 1972a, 125; LAPO 17, no. 515). These must be distinguished from the “sons of the town” (e.g., ARM IV 73:7, DUMUmeˇs a-limki ), who represent all the inhabitants (see below). Based on these comparisons, the restoration ´ a-l]a-nuki in ARM XXVIII 48:45 does not appear likely. The pattern in of [LU slightly earlier Cappadocian (Anatolian) texts from Kaneˇs, a colony of Aˇsˇsur, seems somewhat different. Larsen (1976, 160) observes that a citizen is called a “son of Aˇsˇsur,” and one does not use the word “Assyrian” (Aˇsˇsur¯ıyum) for persons. The word a¯ l¯ıyum is not used for “citizen.” 41. ARM XXVIII 42:2 and 14 , with the plural shown by line 3 , a-l[i-ˇs]u-n[u], “their town.” 42. ARM XIV 104+. For the new edition, see Charpin 1993a, 199–202. 43. 8 i-nu-ma a-na Ra-za-ma-aki 9 .sa-bu-um ik-ˇsu-du ki-ma ka-di-im-ma .sa-bu-um ` s-nun-naki i-du-uk 11 a-lu-yu-umki 10 u-s ´ . e-em-ma 7 ME .sa-ba-am ELAMki u` 6 ME lu´ Eˇ meˇs ˇ u4 -mi U4 .10.KAM ip-ru-su-ma lu´ SU.GI a-na .se-er A-tam-ri-im 12 u-s ki-a´ . u-nim-ma ´ am [id]-bu-bu-ˇsum um-ma ˇsu-nu-ma sa-al-ma-nu. ´ s 44. 19 lu.meˇ a-lu-yuki 20 ki-a-am i-pu-lu-ˇsu um-ma-a-mi a-lumki ˇsa Zi-im-ri-Li-im 21 u` .sa-buum p´ı-ih-rum wa-ar-ki-ˇsu il-li-ik. ´ s ` ki a-da-aˇs-ˇsi-im ik-ˇsu-du-ma 27 lu.meˇ 45. 26 pa-an˘ e-pé-ri a-na .s´ı-ti[m] ˇsa BAD a-lu-yuki {KI} ki ki 28 ` a-lam ip-ru-us-ma BAD a-na pa-an e-pé-ri i-mi-it-tam u` ˇsu-me-lam hup-pé-em i-pu´ ˘ ˇsu-ma. 46. See ARM XIII 144:39 (Zalmaqum, send); XXVI 246:15 (Imar, send after receiving); XXVI 365:20 (Kasallu(k), receive); XXVI 409:59 (Razamâ, receive); A.2993+:42–3 (Urban and Mulhum, receive; in Michel 1992, 128); A.2417:9 (Luhâ, receive; in Durand 1988, ˘98–100).

˘

294


47. See XIII 144:39 and XXVIII 19:15–16 (Zalmaqum); XXVI 144:18 (Larsa); ˇ hpad); XXVIII 61:4 XXVI 365:14, 20, 34 (Kasallu(k)); XXVI 409:5, 26 (Su ˘ 30; LAPO 17, no. (Ulâ); XXVI 435:11 (Amaz); A.2769 (Tuttul, in Dossin 1974, 789); cf. A.2459:8 (Hurban in Charpin and Durand, 1997, 387, after peace treaty between Aˇsˇsur, ˘Ekallatum, and Eˇsnunna). 48. This includes the loyalty oath itself (A.2993+:42–3, Urban and Mulhum); pay˘ Durand, ment of taxes (A.885:4, Tuttul, in Durand 1989, 33; M.8884:2, Dabiˇs, in 1992b, 119–20); and management of palace fields (XXVI 62:28, etc., Terqa). Outside these categories, notice also the appointing of guards (XXVI 156:11, Yabliya); detaining merchants (TH72–16:9, Has. or, in Durand 1990b, 46 and n. 42); and choosing or accepting a king (A.257, Eˇsnunna, LAPO 16, no. 300). 49. “Now the Kurdubahites have breached the town of Kalmatum” (ARM XXVIII ˘ 91:8–9). 50. ARM XXVIII 104:30–3. 51. ARM II 75:2 (Isqâ-u-Qâ em); III 83:5 (Kulhitum); XXVI 409:56 (Razamâ); cf. ˘ XIII 139:17 (Admum, after aˇsˇsum). 52. ARM XXVI 256:14–15. ´ s 53. ARM II 75:2 –12 (see LAPO 17, no. 557, pp. 171–3); 2 .te4 -em lu.meˇ Is-qa-a u` Qa 3 4 e-em eˇs-me um-ma-a-mi .s´ı-di-it U4 .10.KAM ra-ag-mu a-na né-eh-ra-ar 5 m Ha 6 ˘ am-mu-ra-bi i-la-a-ku an-ni-tam a-wa-ta[m e]ˇs-me-e-ma a-na Ia-am-ru-us.x (IS. )-AN˘ 7 meˇs ´ 2 me-tim .sa-bu-um ˇ Is-qa-aki 8 u` Qa-e-em aˇs-pu-ur-ma ki-i-ma 9 1 LU u` lu´ SU.GI qa-qa-da-a[t] 10 Qa-a-em u` Is-qé-e-em 11 [ip]-hu-ru-nim-ma i-na pu-uh-ri--nu 12 ˘ ˘ [ki-a-a]m ad-bu-ub-ˇsu-nu-ˇsi-im. 54. ARM XIV 76:22. ´ s ˇ 55. ARM XIII 144:39–41; 39 lu.meˇ Za-[a]l-ma-qa-yu a-na sa-li-mi-im 40 a-na ze-er Sar-ra-ia lu´ E-lu-uh-ta-yi 41 iˇs-ta-na-pa-ru. See also ARM XIV 78:7, where the same group is ˘ to track down four Hana. supposed 56. See above, with the discussion˘ of m¯atum coalitions. Durand proposes a similar interpretation of the text in LAPO 16, p. 408 note b. This rare parallel in the plural should be compared with the more common reference to kings by the formula lu´ GN, as in “the Babylonian.” As far as I can tell, the substantive is always applied by outsiders, not by the kings themselves. Actual rank is invoked ´ Nu-um-ha-a[ki ]” in for ideological advantage, from within. Notice that the “LU ˘ and note FM III 131 may not represent the difficulty supposed by Guillot (p. 274 a). If this is simply “the Numhaean” (Numhâ-ite), it should not be necessary to ˘ roi du Num˘hâ.” “The report of the Numhaean” translate as “l homme qui sera ˘ and would address the condition of the throne, ˘able to cover both a seated king the kingship in transition (“Regarding the throne of Numhâ . . .”). ´ s ˘sum LU ´ ˇsa-a-ti mi-tu 57. ARM XXVI 312:32 –34 ; 32 lu.meˇ I-da-Ma-ra-as. ka-lu-ˇsu 33 aˇs-ˇ 34 lu´ ´ u-ul aˇs-ˇsum I-da-Ma-ra-as. LU ˇsa-a-ti tu-ba-al-la-at.. Note that both the ´ ba-al-t.u´ singular and the plural forms of the determinative serve the same collective meaning. 58. See ARM XXVI 310, and AEM I/2, p. 40. On the location of Tadum, see Guichard 1994, 242–4. ´ meˇs 59. ARM V 24:5–12; 5 Ba-aq-qum lu´ Ti-iz-ra-ahki 6 a-na ˇsi-im-tim 7 it-ta-la-ak 8 u` LU meˇs ki 9 10 11 ˘ DUMU Ti-iz-ra-ah i[l]-li-ku-ni-im-ma um-ma-a-mi Ka-a-li-AN-ma a-na ˇsa˘ Note the comment on this letter with LAPO 16, no. 80, p´ı-ru-ti-ni 12 ˇsu-ku-un. p. 206.


295

60. For Tarim-ˇsakim as one of Yasmah-Addu’s two viziers, see Villard 2001, ˘ 21. ´ meˇs DUMUmeˇs GN should be separated from 61. One may ask whether the form LU the simpler formula, as specifying a subset. So far as these “men” represent the whole town as a town, however, it seems to me that such a distinction would be meaningless. ´ s 62. lu.meˇ qa-qa-da-at Ka-sa-al-lu-ukki , ARM XXVI 365:20. 12 ´ ˇsu]-u´ .te4 -ma-am ˇsa 63. i[t]-bé-e-ma a-na Ka-sa-a[l-l]u-ukki [il-li-kam-ma(?)] 13 [LU 14 l u.meˇ ´ s SUKKAL ELAM.MA-tim iˇs-pu-[ru] [iˇs-ku-un-ma(?) ] Ka-sa-al-luki i-´ı - `ı-[du] 15 [t.e4 -em-ˇs]u-nu iˇs-ta-lu-ma a-na SUKKAL ELAM.MA-tim iˇs-[pu-r]u. lu.meˇ ´ s 64. 19 [i-nu-ma] ta-ki-it-tam ga-me-er-tam 20 [u-te-ru-nim] qa-qa-da-at Ka-sa-al-lu´ ki 21 uk [i-na a-wa-a-tim i]s.-ba-ás-su-nu-ˇ si-im um-ma-mi (then broken). ´ 65. 28 [ . . . ] u` ˇse-em IN.NU.[D]A .se-eh-ra-am 29 [s.e-he-er-tam ka-la-ˇsu-n]u ki-im-sa-nim-ma 30 ki 31 ˘ ´ [a-na K]A.DINGIR.RA [ . ˘. . ]-re-et UDUhá -ku-nu i-na ma-ti-kuˇsu-ri-ba-nim ˘ 32 h´ a nu i-ka-la [u` at]-tu-nu i-na E´ -ku-nu lu wa-aˇs-ba-tu-nu 33 [an-né]e-tim LUGAL ˘ id-bu-ub-ˇsu-nu-ˇsi-im-ma 34 [ki-a-a]m i-pu-lu-ˇsu um-ma ˇsu-nu-ma iˇs-tu be-el-ni iq-ta-bu-u´ 35 i ni-il-li-ik .te4 -em-ni i ni-is.-ba-at 36 an-ni-tam LUGAL i-pu-lu-ma it-ta-al-ku 37 wa´ há 38 a-na SE-ˇ ˇ su-nu ˇsu-ur-ku-bi-im ar-ki-ˇsu-nu 6 li-mi .sa-ba-am u` giˇs MA u` ni-ˇsi-ˇsu-nu ´ ˘ 39 na-sa-hi-im it.-ru-d[a-ma]. ˘ ´ meˇs qa-du-um 66. M.10991:7 –9 , letter from Abi-epuh to Yasmah-Addu; 7 50 LU ´ ˘-ba-anki u-s ´ há ]-ˇsu-[nu] 8 iˇs-tu Ha-ra-di-imki 9˘ [a]-di Ur text no. 9 E[ ´ . u-ni[m]; ´ ˘ in Charpin 1997b, 361. ˘ 67. ARM XXVI 409:63–4. ´ s 68. ARM XXVI 410:22 –23 ; 22 lu.meˇ Ha-ar-b[é-eki ] 23 [qa]-du-um ba-ˇsi-ti-ˇsu-nu u` mi˘ im-ma ˇsu-um-ˇsu-nu. 69. A.2993+:36–9, in Michel 1992, 127–8. 70. ARM XXVI 17:20–30; 20 DINGIR-lum i-na ha-al-s.[i-im e-l]i-im 21 u-[l]a-ap-pa-at-ma ´ 23 ˘ qa-tam a-na qa-tim-ma 22 e-ti-iq u` be-l´ı li-wa-e-er-ma [DUM]Umeˇs a-la-ni ˇsa ki-ma la-ap-tu 24 a-na a-la-ni la la-ap-tu-tim 25 la i-ru-bu as-su-ri ´ ma-ta[m] ka-l[a]-ˇsa 26 u-la´ ap-pa-tu u` ˇsum-[ma ge-r]i be-l´ı-ia 27 a-na ha-al-s.´ı-im e-li-im i-ba-aˇs-ˇsi 28 be-l´ı i-na Ter-qaki 30 ˘ -m[a] li-ik-ka-li 29 a-na Sa-ga-ra-timki la i-ti-qa-am ma-a-tum lu-up-pu-ta-at. 71. For Asqudum the diviner, see Durand, AEM I/1, chap. 1 (pp. 69–228). 72. There is no reference to mobile groups who might be defined as “tent-dwellers” (Hana) or “backcountry (people)” (nawûm). ˘ 73. A.315+:29–31; in Charpin 1990a, 74; 29 nu-uh-hi-id-ma i-na DUMUmeˇs a-limki 30 ˘ ˘ Cf. also ARM XIII 146:14– ´ 1 LU a-na ki-ir-hi-im a-na e-re-bi-im 31 la i-na-ad-di-nu. ˘ 17 and 147:32–33 (Talhayûm); XXVI 144:4 , 6 , 9 , 13 (Rapiqum); XXVI ˘ 320:14 (DUMUmeˇs sa-al-hu-um, note problem with nominative case); XXVI 462:9 ˘ (Ziniyan); A.4513:21 (Alatrˆ u, AEM I/1, p. 309). ´ al-la-timki , ARM XXVI 5:20. 74. DUMU E-k´ 75. Cf. also ARM XXVII 99:21–2, two men originally “sons” of one a¯ lum; M.7630:7 (Guichard 1994, 258) and A.4687:5 (Charpin 1993b, 181), the origin of men who are away from home. ARM XXVI 495:17–19 contains the report of IˇsmeDagan’s death, with the writer’s comment that he has sought information from all “sons of Ekallatum” who came his way. Other larger groupings include A.285:22 and 25 (Charpin and Durand 1997, 385), for people of Ekallatum origin allowed to enter Karanâ; and A.4535-bis:2 (same, p. 389–91), Aˇsˇsur merchants complaining about their treatment.

296


76. Notice that there is also a West Semitic word for “inhabitants” that does not tend to be linked to place names: maskan¯u, from the verb sak¯anum, “to take up residence.” See ARM V 78:6 in Durand 1987b, 227; and FM VII, pp. 77–9. 10 77. ARM III 3:9–17; 9 as-su-ur-r[e] a-na be-l´ı-ia ki-a-am 11 i-qa-ab-bu-u´ 12 um-ma-a´ 13 mi .sa-bu-um ma-d[u-um] i-na li-ib-bi Ter-qa ki 14 i-ba-aˇs-ˇsi 15 i-na 4 ME DUMUmeˇs Ter-qaki 16 2 ME lu´ .sa-bu-um ZI.GA 17 u` 2 ME ba-ˇsi-tum. See also ARM III 6:5–6; 26:12–13; XIII 123:24; and A.2757:9, in Villard 2001, 64. 78. ARM XIV 61:7, etc. In FM VII 30:41, the “sons of Alahtum” are then residents who will have to perform military service, the draft ˘that results from census taking. 79. M.5157+:19 , in Durand 1991, 53, to fight with the Hana. For other examples ˘ I 10:6 (Till-Abnim); II of the “sons of the town” as a generic term, see ARM 137:9 (Tuttul); IV 11:12 (a city in the Balih); IV 73:7–8 (some town under ˘ Yasmah-Addu, from which “Mariotes” are removed); XIII 148:3 (Talhayˆ um); ˘ ˘ XXVI 215:28 (Tuttul?); A.315+:25, 29 (towns in the m¯at Kahat). ˘ 80. It is interesting that the DUMUmeˇs m¯atim represent all the inhabitants of a “land,” meˇs ´ but there is never a *LU m¯atim, which would perhaps be seen as redundant. See ARM I 43:7 (army); I 118:15 (shepherds); IV 17:18 (taken for military service); IV 78:11 (population, at war?); XXVII 23:6 (recruiting for bazah¯atum ˘ service); XXVII 27:15 (in describing hals.um population); XXVII 28:22 (owning sheep); XXVII 107:18 (given land ˘by palace). Notice also the “sons of the district” in A.2943:38 (Durand, 1987b, 230). ´ s 81. ARM XXVIII 51:9–10, lu.meˇ mu-uˇ ´ s-ke-num DUMUmeˇs Aˇs-l[a-k]a-aki . 82. FM VII 36:8–9, 51–3, 55. 83. Likewise, the dumu-nita-dumu-nita-eb-laki should be “the sons (inhabitants) of Ebla,” rather than “children, citizens of Ebla” (vs. Pettinato 1991, 78). 84. A.2951:7 and A.885:23, in Durand 1989, 33–5. In the first text, see line 6 for the tahtamum. Both letters were written by the Mari hazzannum, Lanasûm. ˘ in Durand 1989, 34–5; 6 ta-ta-ma-am ˇsu-ˇsi-ib 7˘ [si]-ra-am ˇsa e-li DUMUmeˇs 85. A.2951:7, T[u]-tu-ul[ki ] 8 ˇsa-ak-nu e-ri-`ıs-su-nu-ti-ma. ´ ´ s Tu-ut-tu-li-[yi] 86. A.885:4–7, in Durand 1989, 33; 4 aˇs-ˇsum si-ri-im ˇsa be-l´ı-ia ˇsa e-li lu.meˇ 5 ´ meˇs ˇsa-ak-nu ki-ma 1-ˇsu 2-ˇsu u` 3-ˇsu ta-ta-ma-am u-ˇ ´ se-ˇsi-[ib] 6 u` a-ˇsa-al-ˇsu-nu-ti LU ˇsu-nu 7 a-na I-ma-ar ki 1-ˇsu 2-ˇsu iˇs-pu-ru. 87. ARM XIV 61:7, etc. 88. See FM II 117:12. 89. There is a considerable bibliography of references to “elders,” generally rather old. These include Walther 1917, 52–63; Oppenheim 1936; Klengel 1960, 1965; Schlechter 1968; Reviv 1989, 137–77 (on Syria-Mesopotamia); Anbar 1991, 150–4. 90. Klengel 1960, 364–5, describes this as a “Stammesverfassung,” based on ARM II 75 (Isqâ-and-Qâ). He divides collective governance into a hierarchy of “Vorste¨ her der Gemeinde,” “Altestenrat,” and “Volksversammlung.” Mario Liverani calls the body a “college” of elders (1975, 154). 91. Such images actually reflect what F. G. Bailey called “elite” councils, which are not characteristic of archaic societies (1965). 92. The discussion of the tabinau of the Yap islanders by David Schneider offers a useful methodological comparison. The tabinau can represent a house, persons


93. 94.

95.

96. 97.

98. 99.

100.

101.

102.

297

related through ties of land, or a place itself, depending on context. Schneider observes that careful description of culture has to precede the assumptions of a model, and the variety of meaning for this key social category is visible only when such sensitivity is a conscious priority (1984, 21–2). For example, Klengel describes the Sumerian ab-ba as a “title” (1960, 358). See ARM II 75:6 –7 (with Yamrus. -el, no title; Isqâ-and-Qâ); VI 12:5–6? (with Yaqqim-Addu, no title); XXVI 48:5–6 (with Niqma-el, no title; Qâ); 503:5–6 (letter from Kibsi-Addu and elders of Uruban and the Dunnayu); A.981:32 (with Uranum, no title; Dabiˇs). It is interesting that none of the individual leaders, who might otherwise be supposed to be sug¯agums, is assigned a title when joined by “elders.” Notice ARM III 65:5–7, with “Iridanum and 10 elders of the Hana,” for the same phenomenon in a purely tribal and nonurban context. ˘ first of all the various elders of A.2226. See also ARM IV 29:22 (Qat.t.arâ?); Recall XIII 148:3 (Talhayûm); XIV 104+:11 (Razamâ); XIV 114:8–9 (Kurgiˇs); XXVI ˘ XXVI 411:73 (Samaz?); FM II 122 (Iluna-ahi); FM II 125:3 391:7 (Razamâ); ˘ (Talhayûm, in (Urgiˇs); A.2567:16 (Urgiˇs, in Charpin 1993b, 178); A.2417:3–4 ˘ Durand 1988, 98). For similar terminology with tribal Hana, see M.6874:17 . ˘ s; A.2417, Talhayûm. ARM XXVI 503, Uruban and the Dunnayu; FM II 125, Urgiˇ See ARM IV 68:7–8, elders and w¯edûtum of Hiwilat and Talmuˇs, come to˘ Iˇsme˘ Dagan for diplomatic gifts; XIII 145:5–6, messengers and elders of Yarkib-Addu, a king of Zalmaqum, visit a vassal of Zimri-Lim; XIV 64 and 65, sug¯ag¯u, laputtû, and elders of individual towns in the Saggaratum district gathered to receive kingdom orders; XXVI 438:8 , the gerseqq¯u and elders of Andarig attend a ceremonial feast; XXVI 463:6, heads of the Numhâ and m¯atum elders of Kurdâ ˘ gather to discuss foreign policy with the king. Notice that the tribal sug¯ag¯u and elders of the Yaminites meet with similar purpose to make peace with Zalmaqum (XXVI 24:11). On the last two titles, see LAPO 16, pp. 85–6 and 457. For “messengers” (m¯ar ˇsiprim) as high-ranking diplomats, see Lafont, 1997b, 317; 1992a. ARM XXVII 67:13–22. The text is severely damaged in this section, and some caution is appropriate, though it appears that mainly personal names are missmeˇs ˇ ing. If correct, the restoration lu´ SU.G[I ], [ˇsa ha-al-s.]´ı-im in lines 14–15 would ˘ accept designation as a vassal “district” in a common expression of subordination. See ARM XIII 106:11 (just “elders”; stolen goods); XXVI 206:13, 32 (just “elders”; witnesses of prophecy, at Saggaratum); XXVI 479:10 (just “elders”; unmeˇs ˇ certain purpose); XXVII 60:9 (lu´ SU.GI a¯ lim; harimtum-women sent to ZimriLim’s palace, evidently Qat.t.unan); A.2500+:14, ˘in Durand 1990b, 77n199 (five “elders of Der”; sealing ceremony). Charpin observes that the elders appear frequently in Old Babylonian texts along with the individual rabi¯anum who governs a town, and cites example from a wide range of towns, including Sippar, Aˇsdubba (near Larsa), Burzibiˇsur (near Kiˇs), Larsa, and Babylon (2000, 99). ARM III 17:14–20; 14 i-na ˇsu-ul-mi-im 15 be-l´ı li-il5 -li-kam-ma 16 ˇse-ep d Da-gan li-iˇs-ˇsi´ sˇ iq 17 ka-a-ia-an-tam lu.meˇ SU.GI a-limki 18 a-na IGI d Da-gan i-ru-bu-ma 19 a-na be-l´ı-ia See also ARM XIII 117+:6 . u` um-ma-na-tim ˇsa be-l´ı-ia 20 u-ka-ar-ra-bu. ´ ˇ a ARM XXIII 504:1–9. The original publication suggests the restoration of Sunˆ for the first town, based on ARM X 121, but A.2226 (above) places Hurrâ in the triad instead. There is not yet any published copy for ARM XXIII ˘504.


298

103. FM III 28; see also the summary text III 60:35–6, which omits mention of the temple. 104. See Lafont, 1985, 163, and ARM XXI 388, for .sa¯ bum lists with this term. meˇs ˇ 105. Lines 12–21; 12 m I-s.´ı-sà-ra-e 13 m Yi8 -im-si-èl 14 m Mu-ut-ra-me-e 15 lu´ SU.GI Sa-gaki 16 17 ` ra-tim ˇsa a-na su-ga-gu-tim il-li-k[u] KU.BABBAR su-ga-gu-ti-ˇsu-nu 18 `ıs-ni-qu´ 19 ˇsi-im q´ı-[ˇs]a-t[i-ˇsu-nu] 20 a-na be-l´ı-ia a-na-[ku] 21 lu-ud-di-in. lu´ meˇs ˇ ˇ 106. A.2226:3–12; 3 m Iˇs-me-d ISKUR Aˇs-na-ak-ki-imki 4 lu´ SU.GI ma-at I-da-Ma-raki 5 lu´ ˇ meˇs ki ki 6 ki lu´ ˇ 7 ˇ as. aSU.GI Ur-gi-`ıs ˇsa Si-na-ah ˇsa Hu-ur-ra-a u` SU.GImeˇs Ia-ap-t.u-ur ´ ki 8 ˘m lu´ ˇ d ˘ na Ma-la-ha-tim il-li-ku-nim-[m]a Ia-tar-ma-lik Su-du-hi-im u` A-pil- SUEN 9 lu´ ˘ [ ]Aˇs-na-ak-ki-im u` qa-qa-da-at Ur-gi-`ıski 10 [pa-ni]-ˇsu-nu˘is.-ba-tu-nim-ma il-li-ku` u` mé-ra-nam a-na [za-ka-r]i-ni 12 [i ni-i]q-tu-u[l]. nim 11 [um-ma-m]i UZ 107. Iˇsme-Addu was king of Aˇsnakkum only briefly, overlapping the years Z-L 11–12, immediately dating the letter (Charpin 1993b). ˇ 108. For Yatar-malik and the chronology of the Sudu hhum throne, see ARM XXVIII, ˘˘ p. 161 and n. 166. ˇ 109. Subram, king of Qirdahat and then Susâ; see ARM XXVIII, pp. 127–8. meˇs ˇ 110. A.609:26–7, in Durand˘1999–2000, 191, lu´ SU.GI [ma-at I-d]a-Ma-ra-as.ki . 111. ARM XXVIII 59:9–13. 112. ARM XXVIII 65:15–17. 113. ARM XXVIII 103:4–6. me ˇs ˇ 114. ARM XIII 148:3–5; 3 lu´ SU.GI[ ] DUMUmeˇs a-limki 4 [a-na ga-a]m-ra-tim u-te-e´ [ra]-aˇs-ˇsu-nu-ti-ma 5 [a-n]a .se-er be-l´ı-ia at.-t.[à]-ar-da-ˇsu-nu-ti. See also LAPO 16, p. 462 note b. 115. ARM XXVIII 50:25 and 52:22 , only one in the second text. In the first, we should probably translate “elders of (from) Aˇslakkâ,” not “the elders of Aˇslakkâ,” as if we were dealing with all who hold such a rank. 116. ARM XXVIII 64:4–7. See also 67:23–4, for the sug¯agum of Hurrâ and five elders; ˘ “elders” of two Idaand 103:23–4. In 91:11–12, Zimri-Lim assigns what he calls Maras. kings to undergo the river ordeal. These are really just “representatives.” ARM XXVIII 95 similarly requires each of two kings to bring “his elders” before Haya-sumu king of Ilan-s. urâ so he can render judgment. Here the language ˘ overlaps with the purely legal use of the term for “witnesses.” 117. Eidem and Laessøe 2001, 91, no. 18:8–10, 25–6, as “the elders of the land” (ˇs¯ıb¯ut m¯atim). 118. ARM XXVI 463:5–10, letter to Zimri-Lim from Abi-mekim, a high official who is ´ sˇ serving as envoy to Kurdâ; 5 [lu´ ].meˇs qa-qa-da-at Nu-ma--ma 6 [u] SU.GI ` lu.meˇ ˘ı-ni-iq-tum i-na ma-[tim] ma-tim ÌR-du 7 [ˇsa] Si-ma-ah-i-la-a-né-e 8 [i]p-hu-ru-ma 9 s´ 10 [i]ˇs-ˇsa-ki-in-m[a]. Notice ˘the phrase “your ˘elders” used with reference to the king of Aˇsnakkum, in A.2326 (Charpin 1993b, 175). 119. TCL XVII 76:21, in Klengel 1960, 367. 120. ARM XIV 104+:11–14; for the join, see Charpin 1993a, 199. 121. If ARM XXVIII 111:9 –12 has been restored correctly, the word “elders” can be used for the group who puts the king on his throne. Hammi-kun, king of ˇ hhum, writes to Ibal-Addu as a peer: “Another thing.˘ If you are truly my Sudu ˘ ˘ and you love me, the elders will restore me to the throne of [my] brother, father’s house”; 9 ˇsa-ni-tam ˇsum-ma i-na ki-na-tim a-hi at-ta 10 u` ta-ra-ma-an-ni lu.meˇ ´ sˇ 11 giˇs 12 ˘ ´ S[U.GI] a-na GU.ZA E a-bi-[ia] u-ta-ar-ru-ni-in-[ni]. ´ 122. FM II 117:27–42; 27 u` aˇs-ˇsum ma-r[u-tam Si-ma-ah-i-la-a-né-e] 28 a-na be-l´ı-ia la i[ˇs´ s ´ sˇ pu-ru-ma] 29 a-hu-tam-ma a-na be-[l´ı-ia iˇs-pu-ru] 30˘lu.meˇ su-ga-gu u` [lu.meˇ SU.GImeˇs ]

˘


´ 33 ˇsa Nu-ma-he-e i[l-li-ku-nim-ma] 32 a-na Si-ma-ah-i-la-a-né-e ki-a-am iq-[bu-u] ˘ ˘ um-ma-mi am-mi-nim a-na Z[i-i]m-ri-L[i]-im 34 ma-ru-tam ta-ˇsa-ap-pa-ar 35 ki-ma ˇ Aˇs-ta-mar-d ISKUR at-hu-tam 36 a-na Ia-ah-du-Li-im iˇs-ta-ap-pa-ru 37 at-ta a-na Zi-im38 ´ s meˇs ˇ ri-Li-im at-hu-tam ˇs˘i-ta-ap-pa-ar 39 i-na ˘qa-bé-e lu.meˇ su-ga-gi u` lu´ SU.GI Nu-ma40 m 41 ˘ ha-a Si-ma-ah-i-la-a-né-e at-hu-tam a-na .se-er be-l´ı-ia iˇs-pu-ra-am a-na an-né-tim 42 ˘ ˘ ˘ la i-ra-aˇs-ˇse-né-ˇsi-im. Notice that in a letter from mi-im-ma ki-s.´ı-ir li-ib-bi be-el-ni Simah-ilanê to Zimri-Lim, the king of Kurdâ calls himself “your brother,” just ˘ as demanded in FM II 117 (ARM XXVIII 162). ARM XXVI 393:7–9; 7 mi-id-di Ha-am-mu-ra-bi ni-iˇs DINGIR-lim 8 u-ul ´ i-za-ak-kalu´ ˇ ar u-lu-ma S[U.GImeˇs m]a-ti-ˇsu 9˘ i-da-an-ni-nu-ˇsu-um-ma ni-iˇs DINGIR-lim ˇs[a-a-ti ´ u-ul ´ i-z]a-ak-ka-ar. According to another letter, the same Hammurabi is said to have met his own Numhâ tribal population as his “land”˘ (m¯atum), a distinct ˘ will, whom he must persuade to join him in abanpeople with its own political doning his alliance with Zimri-Lim. See A.3577:10 , in Durand 1992a, 45 and n. 39. A.3347, originally published in translation only in Durand 1990b, 64–5. The text now appears as FM VII 7:30–3. There are Yaminite rulers living at Imar, but the king of Yamhad regards them as foreign to the town in the sense that ˘ they are subject to eviction by Imar’s own “elder” leadership. ARM III 17 and XIII 117+, above. ARM XIV 55:19–22; 19 i-nu-ma be-l´ı a-na Ia-am-ha-ad ki 20 i-lu-u´ aˇs-ˇsum SISKUR2 meˇs ˘ 22 [l]a-ma lu´ SU.GI ˇ a-na li-ib-bi 21 a-limki a-na e-re-bi-im is.-s.[a]-b[a-a]t-ma a-limki (break . . . ). Eidem and Laessøe 2001, 93, no. 19:14–18. ARM XXVIII 44, 44-bis, 45, 46; and see p. 55 for discussion of Urgiˇs and its subordination to Aˇsnakkum. For the title “king,” see A.2939:3, in Charpin 1993b, 188; m Te-er-ru LUGAL Ur-gi-iˇs ki . ARM XXVIII 44-bis:7–11; 7 aˇs-ˇsum a-na ˇsi-ir be-l´ı-ia 8 ma-aq-ta-ku DUMUmeˇs a-li-ia 9 i-ze-ru-ni-in-ni u` qa-qa-di 10 i-na qa-ti-ˇsu-nu 2-ˇsu 3-ˇsu i-na mu-tim 11 u-b´ ´ a-al-liit.. Note “inhabitants” as “sons.” The address is broken, but the reference to ˇ Sadum-labua, king of Aˇsnakkum, as the sender’s master, fits Urgiˇs, along with the direct mention of the town in line 21. The traces for the sender are suitable: [T]e-er-[ru] (line 3). Durand, LAPO 16, pp. 207 and 467, has suggested that all second-rank kings were approved and anointed by a higher suzerain. However widespread this practice may have been, it does seem to apply in Zimri-Lim’s relationships to his vassals in the upper Habur. Lines 18–22; 18 a-na-ku u-ul ´˘ i-de! -e 19 ki-ma DUMUmeˇs a-li-ka 20 aˇs-ˇsu-mi-ia i-ze-ru21 ka at-ta i-yu-um u` a-lum Ur-gi-iˇs ki 22 ˇsa ˇsa-ni-im-ma. I read “even if” for “and,” following Kupper. ARM XXVIII 45:7–9, p. 61 note a; “dissension” is bus.a¯ rum (see ARM II 38:18 and LAPO 16, p. 521. Lines 4 , 11 –13 . ˇ ARM XXVIII 46:4 –6 ; 4 i-na-[an-n]a du-mu-uq bi-t[i-ia] 5 e-zi-im-ma a-na Si-naki 6 ah a-na ha-bi-ru-tim at-ta-s.´ı. The noun “h¯abir¯utum” refers to the state of being ˘ ˘ home for any reason, and˘ my translation as refugee status redisplaced from flects Terru’s particular circumstances. See Durand, AEM I/1, p. 428 note d; ARM XIV 50:14 and p. 228 (Birot); XIV 72:18; ARM XXVI 510:25; ARM XXVII 70:17; 116:32. 31

123.

124.

125. 126.

127. 128.

129.

130.

131.

132. 133. 134.

299

300


135. ARM XXVIII 69:8–10; 8 [m ]Ia-an-s.´ı-ib-ha-ad-nu u` Ha-zi-ra-an a-na Ur-gi-iˇs ki 9 [il´ s ˘ sa-yuki 10 [ki-a-am ˘ l]i-ku-ma pu-uh-ra-am id-du-ma lu.meˇ Ur-gi-ˇ i-p]u-lu um-ma-mi ha˘s-ˇsu-u´ nu-wa-aˇs-ˇsa-a[r]. Note the verb “nadûm” as “initiate,” with ˘ ia-tam ˇsa i-ba-aˇ the object puhrum, “assembly.” For the suggestion that hayatum be translated as ˘ meaning both the people and the livestock ˘ “encampment,” in it, see Fleming 1999, 170–1. Elsewhere, we hear about a Mari official who has gone to Urgiˇs and “called for an assembly” (puhram nadûm, again). “They responded “yes” ˘ a unit (ARM XXVIII 99:11 –13 ). regarding troops” – again, acting as 136. It is not necessary to identify Haziran as an unknown ruler of Urgiˇs, as Kupper ˘ appears to do (p. 68). 137. ARM XXVIII 107, and 113:7–11. 138. Dominique Charpin (personal communication) observes that some towns may have had kings only for a short period. He suggests Tillâ as another, though this town left us no royal letter in Zimri-Lim’s archives. Joannès identifies Samsierah as “king” of Tillâ in ARM XXVI 417:15 –16 , where he is identified as “the˘ Tillâ-ite” (lu´ Ti-il-la-aki ). This letter does portray Samsi-erah as the ruler of Tillâ, though he does not receive the title “ˇsarrum.” The text ˘dates to Z-L 13. Samsi-erah is also known from the year Z-L 11 (ARM XXVI 313:42, 78), and as ˘ 7–8 (XXVI 357:10 ). The Samsi-erah whose death is reported by early as Z-L ˘ 11–12) must be a separate Ibal-Addu of Aˇslakkâ in ARM XXVIII 61:4–6 (Z-L figure (cf. also 62:18, 53 and 91:4 –5 ). 139. See ARM XXVIII, pp. 55–6. 140. See Charpin 1993b, 169, for A.2226, when Iˇsme-Addu was still king of Aˇsnakkum; and Guichard 1994, 254–5, for a possible date of FM II 125 in the year Z-L 6. ˇ 141. ARM XXVIII 99 was sent by Sadum-labua, and XXVIII 107 and 113 come from the same period. Only XXVIII 69 is without clear date, but it could also be fairly late. ´ sˇ 142. ARM III 65:5–9; 5 m I-ri-da-nu-um 6 u` 10 lu.meˇ SU.GI[m ]eˇs 7 ˇsa Ha-na 8 a-na .se-er ˘ LUGAL i-la-ku 9 a-na LUGAL bu-ur-ri. 35 143. ARM XXVI 404:35–8, a long letter of Yasim-el to Zimri-Lim; DUMU-ri at-ta ´ s ke-em i-zi-iz it-ti Ha-aq-ba-ha-am-mu-u´ 36 u` lu.meˇ ˇsi-bu-ut Nu-um-ha-aki lu-ud-bu-ub Ha´ s ˘ u´ 37 u` ˘lu.meˇ ˘ aq-ba-ha-am-[mu]ki-a-am˘ 38 ˇsi-bu-ut Nu-um-ha-aki `ı[s]-si-ma a-wa-t[am] ˘´ ˘ is.-ba-su-nu-ˇ si-im um-ma-a-mi. 144. Durand says of Der, “Il faut donc supposer que les cérémonies de Dêr jouaient un rôle déterminant parmi les rites fondateurs de la monarchie bensimalite, sinon on ne comprendrait pas cette exigence impérative de présence faite aux vassaux” (in Durand and Guichard 1997, 39). Durand promises here to publish the archive of materials related to the great Eˇstar festival at Der in ARM XXVI/3, and meanwhile cites ARM XIV 66 for the requirement of attendance. The duplicate Der near Mari was home specially to Simalite tribespeople, and this town stands in for Der of the Balih, in the Simalite “fatherland.” Zimri-Lim ˘ from this region (p. 40). prefers to marry his daughters to rulers 10 m ´ 145. ARM XXVI 24:10–13; As-di-ta-ki-im u` LUGALmeˇs ˇsa Za-al-ma-q´ı-imki u` 11 lu´ sumeˇs ˇ ga-gumeˇs u` lu´ SU.GI [ˇ s ]a DUMUmeˇs Ia-mi-na 12 i-na E´ d SUEN ˇsa Ha-ar-ra4 ki anˇse 13 ˘ nim ha-a-ri iq-t.u[LUGALm ]eˇs ˇsa ma-a-at Za-al-ma-q´ı-im k[i]-a-am ´ u-lu-[n]im ´ ˘ i-da-ab-bu-u-b[u]. Asdi-takim is the king of Harran, which is one of the four towns ´ ˘ of the Zalmaqum coalition; see ARM XXVIII, p. 35.


301

146. Atamri-el is known from the great Yaminite gathering at Harran in ARM XXVI ˘ 24:8, just quoted. 147. M.6874:17 –18 ; in Durand AEM I/1, p. 181. A similar gathering of Yaminite sug¯ag¯u and elders, numbering 100, also accompanies the killing of a treaty ass with Asdˆı-Takim in A.2692+ (Durand, 1994, 92 and n. 24, no line numbers). ARM II 83:18–19 speaks of Yaminite elders who are to gather for some discussion with Zimri-Lim. Finally, note in FM II 116:10–11 and 23–4 the sug¯ag¯u and the elders of the Mutebal, who appear to designate Yaminites; see p. 208 note d; Durand, AEM I/1, p. 175 and n. 20, for ARM XXVI 39. 148. ARM III 3:15. Thirty-seven heads of the Yumhammu clan represented the town ˘ of Sapiratum in the oath sworn for the statement preceding the present text (ARM VIII 85+:54–7, in Charpin 1997b, 344). This could represent a structure of the same scale in a smaller town. 149. Liverani 1983, 157; generalized to all Late Bronze Syria in 1974, 329–30. 150. The most complete case for this general social structure is now found in Schloen 2001. 151. Charpin argues persuasively that the list of forty-three names in the current tablet should include the thirty-seven heads of the Yumhammu who are said ˘ known to be local to have sworn the earlier oath (p. 347). Because two men members of the Yumhammu gayum are found in lines 20 and 39 of the list, and one expects all of˘the division to be listed together, Charpin then suggests that the six superfluous names should belong to a second gayum. He proposes that this would be Bi-ha-la-nu, the sixth name, which might be equated to the once-attested gayum˘ of Ba-ar-ha-la-nu (A.3592, unpublished). The phono˘ sight, but Charpin has suggested to me logical match is not strong at first that an equation could be maintained if we reconstructed alternate forms of the name as *Bin-halanu and *Bar-halanu. “Bin-” would represent the known ˘ West Semitic “son,” ˘ Amorrite form of the while “Bar-” would be a pre-Aramaic equivalent. It would be useful to know whether there is any other evidence for the Aramaic-type vocabulary in Mari or other Amorrite texts from this period. Also, Barhalanu is not included in Durand’s list of gayums, and without a better idea of˘ its associations, the evidence remains inconclusive. From the perspective of this text alone, it would be simpler to allow that the whole town was identified with the Yumhammu clan, and the number of participants ex˘ panded naturally with the growth of the town during the time between the two oaths. ˇ ` e´-kál-lim ib-qu152. Lines 48–63; 48 m Pu-ul-s´ı-d ISKUR Ia-bu-su-um u` a-ah-hu-ˇsu 49 A.Sˇ A ´ ´ 50 ˘ ˘hu-ur-ma di-nam Zi-im-ri-Li` a-lum Sa-p´ı-ra-timki 51 ipru um-ma ˇsu-nu-ma A.Sˇ A-ni ˘ im {nu} 52 i-na E´ d Ha-na-atki i-di-in-ma 53 a-na ni-iˇs DINGIR-lim a-lam Sa-p´ı-r[aki 54 lu.meˇ ´ s 55 ˘ tim ] i-[di]-[i]n 37 qa-qa-da-at Yu-[u]m-ha-am-mi ˇsa ni-iˇs DINGIR-lim 56 58 ` LUGAL ˘u-ki-in-nu ` i-na i-na E´ d I-tur-Me-er ih-su-s 13 aˇs-li A.Sˇ A ´ ´ u´ 57 ki-ma A.Sˇ A ´ 59 ˘ ˇ Az-ra-ha-tim u` 7 aˇs-li i-na ka-ap-ri-im Ia-ab-ru-ki(?)-AN(?) 60 SU.NIGIN 20 ` ba-q´ı-ir A.Sˇ A ` LUGAL 61 i-na a-lim Sa-p´ı-ra-timki 62 u` a-ah-he La-ah-mu-mi aˇs-li A.˘Sˇ A 63 ˘˘ ˘ ` ´ Ì.LA.E. ´ i-ba-qa-ru 10 ma-na KU.BABBAR a-na E.GAL 153. ARM II 75:2 and 7 –10 . 154. ARM XXVIII 141:7–9, with the whole collection being nos. 134–44; 7 m Bu-naam i-na .se-er Is-qa-aki 8 ˇsu-ku-un u` i-na .se-er Qa-aki 9 ˇsa ta-ˇsa-ak-ka-nu ˇsu-ku-un-ma. The title of the office is not given, though one imagines the sug¯agum role.


302

155. The name Isqâ should be derived in some way from the Qâ element that both town names have in common, evidently through the prefixed is-. It seems simplest to imagine that Isqâ meant “belonging to (the) Qâ (people),” with iscorresponding to the Phoenician form of the determinative pronoun (eˇs), as opposed to Akkadian ˇsa. Both of these meant “the one of.” Amorrite normally expresses original /ˇs/ (“sh”) as /s/ (“s”). For the Phoenician pronoun, see Hoftijzer and Jongeling 1995, s.v. ˇs10 , pp. 1089–94, especially in the Zinjirli Kilamuwa inscription, KAI 24:4, 6, 14, 15, and 16 (ca. 800 b.c.e.). 156. ARM XXVI 463:5–6. ´ ta-ak-la-am re-ˇse-et ma-a-tim, ARM I 9:16–17. This text from the earlier reign is 157. LU the only example I found that uses r¯eˇsum instead of qaqqadum. In wider Akkadian use, r¯eˇsum may refer to a slave, but it does not generally identify a leader (CAD s.v. r¯eˇsu). This substitution therefore suggests West Semitic influence. 158. A.2730:4–11, in Charpin, AEM I/2, p. 33 and n. 24; 4 a-nu-um-ma 5 m Ha-am6 ˘ hˇ mu-ra-bi m Ha-at-nu-ra-bi m Sar-ru-ki-ka-li-ma LUGALmeˇs ˇsa ma-ah-ri-ka u` Ia-nu-u 7 mˇ 8 ˘ ˘ ˘ ` sa-mar IR-ka ˇsi-ta-al bi-ri-ti-ia u` bi-ri-it Sar-ra-ia da-mu u` dan-na-tum ˇsa-ak-na meˇs 9 10 ` ` 1 ME Ha-na u` 1 ME IR-du-ia ˇsa a-ah pu-ra-an-tim it-ti-ia 2 me-tim IR-du-ia ˘ ta-ak-lu-tum qa-qa-[d]a-at ma-ti-ia 11 i-na ˘da-mi iz-zi-zu u` ni-`ıˇs DINGIR-lim [l]u-u´ za-ak-ra-ak-ˇsum. 159. ARM XXVIII 106:34–5. 160. The same terminology is used more indirectly for officials of Zimri-Lim’s kingdom in ARM XXVI 35:13 (of Asqudum and Aˇsmad) and XXVII 67:21 (“servants of my lord, heads”). In the first text, the status is attached to the writers, whom Durand identifies especially by comparison with XXVI 38. 161. M.7259:7, cf.1–2; in Durand 1991, 48. 162. Note the follow-up article by Evans (1958), focused on why Sumerian society became more “autocratic.” 163. En¯uma eliˇs ii:16–26; iv:3–10, 13–16, 28–32. 164. K.8282 obv. i 25–9, Jacobsen’s translation (pp. 163–4 and n. 23). 165. This translation follows Wilcke 1998, 460: “In der einberufenen Versammlung ¨ der Altesten/wehrhaften Männer seiner Stadt antwortete man Gilgameˇs.” See also Römer 1980, 38 and 50, for a similar translation. My reading of a single overall collective follows the approaches both of Römer (1980) and of Falkenstein 1966, 47. For “able-bodied men,” see Van De Mieroop 1999, 125. 166. Jones 1971, 68–9. The criterion of age did not prevent younger men to achieve “elder” status by dint of their personal achievements. 167. ARM XXVI 371:19, 31, referring to public prophecies by the a¯ pilum prophet of Marduk in front of Iˇsme-Dagan’s palace gate. “Just as before, he stood in the gate of Iˇsme-Dagan and kept crying out in the assembly of the whole land, saying . . . ” (lines 18–20); 18 ki-ma pa-ni-ˇsu-nu-ma i-na ba-ab Iˇs-me-d Da-gan 19 iz-ziiz-ma i-na pu-hu-ur ma-a-tim ka-li-ˇsa 20 ki-a-am iˇs 7 -ta-na-ás-si um-ma-a-mi. ˘ a similar observation for first-millennium Babylonia, where 168. Larsen makes there is a puhru “assembly” but no example of an assembly for the city (2000, ˘ 125). 169. It is interesting to note that the well-known Babylonian myth called the Enuma eliˇs recounts a puhrum meeting of only three gods (I 34–55); Cassin 1973, 111. ˘ 170. ARM XXVI 347:5–8, 16–19. “All the kings gathered in Nahur in the presence of Haya-sumu, and in their meeting, they said as follows .˘. .”; 5 LUGALmeˇs

˘


171. 172.

173.

174. 175. 176.

177.

178. 179.

303

7 ka-lu-ˇsu-nu i-na Na-hu-urki 6 ma-ha-ar Ha-ià-suip-hu-ru-ma i-na pu-uh-ri´ u-[m]u ´ ˘ u´ um-ma ˇsu-nu-ma; ˘ ˘ “These kings said ˘ this in their meet˘ ˇsu-nu 8 ki-a-am iq-buing, but Sammetar did not come to the meeting of these kings”; 16 an-ni-tam LUGALmeˇs ˇsu-nu i-na pu-uh-ri-ˇsu-nu 17 iq-bu-u´ u` Sa-am-me-e-tar 18 i-na pu-hu-ur ˘ ˘ VI LUGALmeˇsˇsu-nu-ti 19 u-ul See also ARM XXVI 306:42; 352:17; FM ´ il-li-kam. 9:11 . See in the same order, A.3577:9 (Durand 1992a, 45 and n. 10); FM II 63:16; and ARM XIV 84+:4–5 (Charpin and Durand 1986, 176–8). For use of the verb “pah¯arum” to describe the mustering of the Hana to fight, ˘ see: ARM I 24+:43 (Samsi-Addu); II 48:12–13; XXVI 27:4, 15;˘30:8; 31:14– 16; 40:27; 43:62–4; 45:28; 46:3; A.3567:7 (in AEM I/1); XXVII 14:9; FM II 116:40–1; 123:28 –9 . In two texts, the word “puhrum” precedes the use of tahtamum (tâtamum in Tuttul ˘ general to specific, sweeping ˘ to technical. Lanasûm texts) in a progression from writes to Sammetar, aˇs-ˇsum pu-uh-ri-im na-da-nim m Na-ab-d UTU ta-at.-ru-da-am ta˘ ˇ ta-ma-am ˇsa Tu-tu-ul ki [u]p-ta-hi-ir-ma, “You sent me Nab-Samaˇ s in connection ˘ with a meeting to set up. I assembled the tâtamum of Tuttul, (and he said . . . )”; A.3243:6–9, in Durand 1989, 35. Warad-iliˇsu writes home to Mari from Imar that he has had a problem getting boats. He says, “So, at their meeting in the tahtamum, I gave them a piece of my mind on the subject of the boats that had ˘ detained . . . ”; 11 u` i-na pu-uh-ri-ˇsu-nu i-na ta-a-ta-A-mi-im 12 aˇs-ˇsum giˇs MAha been ˘ ˘ si-im-[m]a (A.228:11–13, in same, ˇsa ka-lu-u´ 13 [i-na a-w]a-[t]im as.-ba-as-su-nu-ˇ p. 36). See also ARM XXVIII 99:12 –13 ; 113:11. See Durand, LAPO 17, p. 362. The reading of this noun has long been uncertain. For association of the “peg” with military campaigns, see Charpin 1993/4, 17. Charpin cites both this text and M.11070:11 for the phrase “sikkatam ep¯eˇsum,” “to make a military expedition,” as well as M.13217 for the phrase “ana sikkatim al¯akum,” “to go on a military expedition.” ARM XXVI 412:6–10, 16–22; 6 iti Li-il5 -li-ia-tim U4 .5.KAM is-su-u[ ´ h-m]a 7 lu´ .sa-buki 8 ˘ .sa-b]u-[u]m um Nu-um-ha-yu-um i-na li-ib-bi Qa-t.a` -ra-a ir-t.u-up pa-ha-ra-am [ki-ma ´ ˘ ip-hu-ru Ku-uk-ku-ta-nu-um iˇs-tu a-li-ˇsu Nu-na-sa-riki 9 ˘[u-s ´ . i-ma a-n]a pu-uh-ri-im ˇsa 10 ˘ ˘ .sa-bi-im ik-ˇsu-dam-ma m[u-ru-us.x (IS. ) li-ib-bi-ˇsu [ma-ha-ar .sa]-bi-im ki-a-am iˇs-ku16 ˘ un um-ma-a-mi . . . [an-né-tim u] ` ma-da-tim Ku-uk-ka-ta-nu-um a-na pu-uh-ri-im ˘ e]-li ´ s-ke-ni-im ˇsa .sa-bi-im id-bu-bu-ma 17 [t.e4 -em .sa]-bi-im uˇs-ta-an-ni u` p´ı-i lu´ mu-[uˇ Ha-aq-bu-Ha-am-mu 18 [x x be-l´ı]-ˇsu uˇs-ta-ba-al-ki-it (erasure) [u` Ha-aq-ba]-Ha-am˘ u´ 19 [a-na ˘ mu-u]ˇs-ta-lu-tim u` giˇs KAK e-pé-ˇsi-[im ki-ma] la i-du-˘u´ 20 [m Ka]-ki-ia ˘ muki 21 ´ a-na pu-uh-ri-im ˇsa .sa-bi-im a-na Qa-t.a` -ra-a it.-ru-ud-ma [LU ˇsa-a-tu] i-du-ku ˘ a-na[i-d]i-im ˇsa Ku-ku-ta-nim `ıs-hu-ur 22 [Qa-t.a` -r]a-aki ir-t.u-pu u` lu´ mu-uˇ ´ s-ke-nu-um ´ ˘ .sa-ba-tam. ARM XXVIII 25:8–16, esp. lines 11 and 14. The puhur Hana in ARM II 33:12 moves as a military force and results in ˘ but in its immediate context it accompanies consultations with military˘action, the merhûm Ibal-el (see LAPO 17, p. 217 and note a). If the restoration [a-n]a ˇsu-ku-un˘ pu-[uh-ri-im] ip-hu-ru-ma is correct, ARM V 23:9–10 would record the ˘ Sutû for ˘ raids against Qatna, but such a tribal assembly may gathering of tribal include the decision to go to war. King Asqur-Addu of Karanâ mobilizes the


304 lu.meˇ ´ s

180. 181. 182. 183.

184.

185. 186. 187. 188. 189. 190. 191. 192.

193. 194.

195. 196. 197. 198.

pu-uh-ru-um against the threat of attack (XXVI 523:32), but this may simply ˘ for the pihrum, the general muster of regular troops, as suggested be a variant ˘ to me by Dominique Charpin (personal communication). A.3976:8 , see Durand 1991, 64–5. Durand 1989. I use the fuller form “tahtamum” even though the more common ˘ spelling does not mark the -h- (see A.623+:11, 14, from Imar). ˘ ki . A.3243:6, ta-ta-ma-am ˇsa Tu-tu-ul The texts are A.402, A.885, A.1230, A.2951, and A.3243 for Tuttul; and A.228, A.623+, and possibly A.2428 for Imar. It is possible that the younger(?) town of Abattum may have a similar institution under a different name. In M.13096:12– 13, the DUMUmeˇs A-ba-at-timki are assembled ([na]-ap-hu-[ru], in a broken setting; Durand 1990b, 46 and n. 42. I did not find any˘ name in Bonechi’s list of Ebla town names that would match Abattum (1993). On one occasion, Tuttul’s tâtamum (=tahtamum) is said to provide “sons of the a¯ lum” to ˘ um (A.402:17–19), and my only example of help the Mari hazzannum Lanasˆ meˇs meˇs ˘ ´ LU equated to DUMU likewise comes from Tuttul, where responsibility for taxes falls to the citizens under either terminology (see above). A gathering of town “sons” at Abattum could represent the same sort of collective political framework. ˇ FM VII 18:5–13 (A.228); 5 i-na e-le-ti-ia i-na I-ma-ar ki 6 aˇs-ˇsum I-s.ur-d ISKUR .tup7 8 ki 9 pa-am a-na .se-er be-l´ı-ia u-ˇ a-na I-ma-ar ki ´ sa-bi-lam i-na-an-na iˇs-tu Ha-la-ab giˇs ˘` a-na pu-uh-ri-ˇsu-nu i-na ˇ ´ há ka-le-e 11 u ak-ˇsu-dam-ma 10 aˇs-ˇsum I-s.ur-d ISKUR MA ˘ 12 giˇs há 13 ´ ta-a-ta-mi-im aˇs-ˇsum MA ˇsa ka-lu-u´ [a]-w[a-ti]m as.-ba-as-su-nu-ˇ ´ ˘ si-im-[m]a. ˘ The local council is insisting that the Mari official pay some quantity of silver because of a problem concerning Is. s. ur-Addu. For the Ebla evidence, along with evidence from Beydar, Brak, Mozan, and Mari, see Bonechi 1998, 224–5, 227 (Imar), 228 (Tuttul). MEE 4.4–6 III :2 ; Durand 1989, 27; Selz 1998, 301. FM VII, p. 46 note b. Durand 1989, 39–40, does not consider this possibility, and leaves the question unresolved. On the seal of the town god d NIN.URTA (Il Imari?) representing the corporate town, as distinct from the local king, see Yamada 1993 and Fleming 2000, 93–5. A.885:4–5. A.402:17–19; A.2951:6–7. See A.402:9; A.885:5; A.1230:10; and A.2951:13, all from Tuttul, in letters from Lanasˆ um. This detail may be the gift of the writer, and not a true difference between the institutions of Tuttul and Imar. AEM I/1, p. 183. A.954:7–9, p. 183; 7 Z[a-a]l-ma-qum u` D[UMUmeˇs I]a-mi-naki ip-hu-r[u-m]a 8 ri9 ˘ ih-s.a-[am] ir-hi-s.upa-an ri-ih-s.´ı-ˇsu-nu a-na s[a]-li-mi-im-ma. Note the use of ´ u-ma ´ ˘ ˘ ˘ the talks. p¯anum plus ana for the “object” of M.6874:18 –19 , p. 181; 18 it-ti lu´ [ . . . ] 19 li-ir-hi-s.a-ma a-wa-ti-ˇsu-nu lu-uˇ ´ s-me. ˘ p. 183; A.3567:6–7, p. 184. See also A.4530-bis:6 –8 , p. 182; A.2526:13–15, For example, Zalpah, in a meeting between Zalmaqum and the Yaminites, ˘ A.2526:10–17, p. 183. ´ s A.4530-bis:5 –7 , p. 182; 5 50 lu.meˇ Up-ra-pu-u´ a-na Tu-ut-t[u-ul ki ] 6 il-li-ku-nim-ma a-na na-we-em it-ti lu´ U[p-ra-p´ı-i] 7 al-li-[i]k-ma ri-ih-s.a-[am] ar-ha-as..

˘

˘


305

199. A.3567:6–9, p. 184; 6 [lu´ H]a-nameˇs ku-ˇsu-ud-ma ri-ih-s.a-am 7 [ri-h]i-[i]s.-ma Ha´ Ha-na˘ ka-lu-ˇsu-nu˘ 9 nameˇs li-ip-hu-ur-ma 8 [a-na˘ ni(?)]-iˇs di-pa-ri-im ki-ma˘ 1 LU ˘ ˘ [li]-in-a4 -ri-ra-am. 4 5 200. ARM XXVI 45:4–6, 18–27; as-su-ur-ri i-nu-ma ri-ih-s.´ı-im a-na p´ı-i lu´ Ha´ meˇs 6 m 18 ˘ na I-ba-al-p´ı-AN i-na-ad-di . . . i-na-an-na 19 be-l´ı aˇs˘-ˇsum I-ti-li-im 20 iˇs-puki 22 ra-am la-ma ri-ih-s.´ı-im-ma 21 m I-ti-li-im a-na T Ha-i` au-lu-ma a-na s e-er a -ba-tim ´ ` . . ˘ ˘ su-muu´ 23 a-ˇsa-pa-ar-ˇ su-ma i-na ri-ih-s.´ı-im 24 u-ul ´ ´ iz-za-az 25 ul-lu-tim ˇsa i-te-bu-m[a] 26 aˇs-ˇsum I-ba-al-p´ı-AN i-na ri-ih-s.´ı 27˘ i-da-bu-bu a-ka-la-ˇsu-nu-ti. ˘ 201. A.2526:10–13, p. 183. 202. A.1230:9–11; 9 ta-ta-ma-am 10 u-ˇ ´ se-ˇsi-ib-ma aq-bi-ˇs[u-n]u-ˇs[i] 11 u-ul ´ im-gi-ru-[ni]ni. 203. Durand is surely right to anticipate an intransitive form of the verb after the nominative, Nu-um-hu-um ka-lu-ˇsu i-na Ka-s[a-pa-aki ip-hu-ur], line 8 (1992a, 45 ˘ ˘ and n. 10). 204. Bonechi observes that while the personal names from mid-third-millennium texts show a difference between the Semitic dialects of the Ebla, Mari, and Nagar (Habur River) regions, the more conservative geographical names do ˘ 236–7). The shared Semitic toponymy may go back to the beginning not (1998, of the millennium, if not before. 205. For the Hurrian arrival ca. 2200, see Steinkeller 1998, 88–90. Buccellati and Kelly-Buccellati themselves, the excavators of Tell Mozan (Urgiˇs), believe that the continuity of material culture going back through the entire third millennium suggests that the city may have been Hurrian-speaking from much earlier times (2001, 93). There is some tendency, however, for new arrivals to take on the material cultures of their adopted homeland. 206. See Geyer 1990 for the hypothesis of an earlier site closer to the river. 207. See Finkbeiner 1999–2000, 14–7; Finkbeiner et al. 2001, 74–5, 80–1, 88–94; Finkbeiner et al. 2002, 11–15. I received copies of these reports from Anne Porter and would like to thank both her and Uwe Finkbeiner for access to the recent finds and analysis. 208. See Durand 1990b, 42. Charpin observes that Imar had to pay tribute to three different kings, based at Mari, Aleppo, and Carchemish (FM V, forthcoming). This reflects the distance of Imar from any single center of power. 209. Durand, 1990b, 55–6. A man named Dadi-hadˆ un had considerable influence ˘ at Imar but seems to have been a Yaminite ruler (in the Rabbˆ u tribe), with his particular base at nearby Abattum (pp. 45–6, 55–6). 210. See above, with the tahtamum, and ARM XXVI 256. Durand discusses the gen˘ 55–7). eral situtation (ibid., pp. 211. See Durand’s brief review of the situation (1990b, 45) and Lafont (2002, 93). Charpin observes that whereas Yahdun-Lim had conquered Tuttul from the ˘ Yaminite king Bahlu-kullim of the Amnanum tribe and then made it his western ˘ left it as a separate entity with a “regulated autonomy” (see capital, Zimri-Lim FM V, forthcoming). 212. For the idea of a northern “dry-farming” region, see Weiss 1988. 213. Archi 1979/80; Pettinato and Waetzoldt 1985, 237, 243. The title is also written with d LUGAL in place of d BE. 214. ARM III 17:14–20 and XIII 117+:7 –10 , with Dagan restored in the second text.

306


215. Lambert 1985, 531, with discussion of the value of BAN as ter5 in the lengthy note 14. 216. Rouault 1993, 186. For a basic statement with bibliography, see Buccellati and Kelly-Buccellati 1997a, 188–90. 217. See, e.g., Bonte 1979, 203–4. It is within the residential group that pastoral work is effectively carried out and that “communally held resources are really appropriated” (p. 212). This phenomenon was observed long ago by EvansPritchard 1940, 17. 218. Schwartz 1986 concludes that there is little evidence for indigenous, stratified society in northeastern Syria before this date. 219. For a more fully developed version of the same essential approach, see Wilkinson 1994, 502–4. 220. Tell Brak is a special case, because it had been a major city with significant ties to southern Mesopotamia at the turn of the fourth to third millennia, the period of Uruk. After centuries of more modest scale, Tell Brak became perhaps the most important city in northeastern Syria during the second half of the third millennium (Eidem 1998–2001, 75–7; Oates and Oates 2000). 221. Roaf assumes just this scenario in his article on the preceding Ninevite V culture of the Habur basin, which he understands to have been based entirely on “rainfed˘cereal farming” (1998–2001, 438). 222. See Kouchoukos, 1998, 421. For Chuera, see Orthmann 1986. 223. See McClellan 1995; McClellan, Grayson, and Ogleby 2000. This hypothesis is based on an interpretation of radial lines or tracks that reach outward from many of these sites, and which would carry such runoff. Wilkinson objects directly to McClellan’s approach, saying that the topography of the tracks is too uneven to allow water flow (1994, 492), though McClellan argues that they are pitched downhill consistently enough to produce this effect. 224. Porter defines a spectrum of four subsistence practices, from pure pastoralism to pure cultivation, with supplemented pastoralism and supplemented cultivation in between (p. 30). For one statement of the notion that pastoralist “nomads” do not acquire states except by conquest, see Khazanov 1984, 228– 30, cf. Ernest Gellner in the Foreword, p. xiii. 225. Within Buccellati’s argument itself, and aside from the plausibility of the model as such, I find particularly problematic the idea that a whole region of inexperienced pastoralists would be signaled by less than “professional” culling practices at Old Babylonian Terqa (p. 99). In general, Frangipane considers the early third millennium a period of increased pastoralist subsistence (1998, 199–201). ˇ 226. Durand wonders whether the regional capital between the Sakkanakku rulers and Yahdun-Lim may have been at Terqa and considers the possibility that the ˘ of Eˇsnunna may even have kept a garrison at Mari (idem). kingdom 227. Service 1975, 47–64 and 71–94 (cf. Fried 1967), for “egalitarian” and “ranked” societies, versus “stratified” and “state” societies. Naturally, an enormous literature on such peoples precedes and follows them. 228. The entire volume edited by Susan Keech McIntosh as Beyond Chiefdoms (1999) rejects the categories of the evolutionary approach. See, e.g., Vansina 1991, 171, “The confrontation between neo-evolutionary theory and the evidence from equatorial Africa leads to disastrous results for the theory. There is simply no evidence for an overall increase in centralization.”


307

229. This definition follows Fried, who calls power “the ability to channel the behavior of others by the threat or use of sanctions” (1967, 13). Earle measures power “by the mastery that a leader exercises over others,” with at least an implied threat (1997, 3). 230. Giddens likewise reaches beyond the focus on power as it relates to compulsion, which Marx conceived as bound to social conflict. For Giddens, power is simply “the capacity to achieve outcomes” (1984, 257). 231. Giddens does not address the possible predominance of corporate political power, but he offers a useful perspective on the “dialectic of control” between the rulers and the ruled. He says that, “all forms of dependence offer some resources whereby those who are subordinate can influence the activities of their superiors” (1984, 16). 232. For extended discussion of Emar’s zukru rites, see Fleming 2000, chap. 3.

5. Conclusions 1. I have mentioned all of these but Heltzer 1981. 2. The West Semitic terms form a cluster including the noun “malkum” (“king”), the verb “mal¯akum” (“to rule as a malkum”), and the namlak¯atum “dominions” (what a king rules). 3. There are at least two clear examples. In one, the governor of the district of Saggaratum gathers “the sons of the town” (DUMUmeˇs a-limki ) to pass on royal orders about an upcoming census (ARM XIV 61:7). The second text belongs to the series of letters relating to Zimri-Lim’s purchase of the town of Alahtum ˘ from the king of Yamhad. A man named Yasmah-Addu, not the same as either ˘ king, assembles “the sons ˘ of Alahtum” to be informed the Mari or the Yaminite ˘ of the town’s transfer of dominion (FM VII 36:8–9). 4. ARM XXVIII 67:13–22; the “elders” could also be understood as “witnesses” for a procedure with legal overtones. 5. ARM II 75:17 –18 , a letter with unknown sender and recipient. 6. ARM XXVI 256:14–15; on the universal use of Hˆıt for the river ordeal, see ˘ Durand, AEM I/1, pp. 521–3. 7. A.4530–bis:5 –7 , in Durand, AEM I/1, p. 182. 8. ARM XXVI 45:4–6, 18–27; letter of Asqudum to Zimri-Lim. 9. ARM XXVI 412:6–10, 16–22; letter from Yasim-el to Zimri-Lim. 10. See especially his discussion of “the elders of Emar” (2001, 309–10). 11. Morris seems to be particularly tenacious in his insistence on relating Greek democracy to the earlier political history of the region (1991 and 1996). 12. Before the arrival of proper Athenian democracy, the archai magistrates were picked by and from the eupatridai (“well-born”), of leading clans. The most powerful magistrates were “nine archons,” the chief of whom gave his name to the year. Another important political assembly was the naukrariai, the board of chairmen of the forty-eight. In the sixth century, Solon created the “council of 400,” with 100 from each of four tribes. For all of these, see Hansen 1991, 27–31; and on early classes before Solon, see Ober 1989, 55–6. In even earlier days, there could be several basileis per town (Drews 1983, 113, passim). 13. Raaflaub, 1996, 140: “Democratic equality is realized by giving each citizen access to political participation and office.” On the individual of Athenian democracy, see Lloyd 1987.

308


14. I appreciate the insight with which Gunter ¨ Kopcke penetrates to the heart of this question in his paper, “Mycenaean Kingship, or Why Mycenaean Greece Is Different” (2002). Kopcke concludes that Mycenaean art shows a distinct division between the palace style that derives from Crete by way of conquest and a truly native vase style. He finds that the native style is characterized by a sensuousness and individuality that distinguishes a basic Greek culture from any Near Eastern or other neighboring region. I would like to thank the author for giving me an early copy of this study. 15. On the polis, see also Hansen 1997, 11. A polis is found to wage war, make peace, enter alliance, strike coins, defray expenses, repair walls, pass laws, and so on. The Greek polis comes to do more than the Mesopotamian “towns,” but the political and collective aspects are strikingly similar. 16. This conclusion is based on the comparison of burial plot sizes from the eleventh through the fifth centuries (p. 94).

Glossary of Ancient Terms

This glossary includes words from the Sumerian, Akkadian, West Semitic, Hittite, Hurrian, and Greek languages. All of these except the Sumerian are italicized, with the apparent source language identified in parentheses. Identification of words as West Semitic rather than (eastern) Akkadian is not always secure and can become at some level an arbitrary distinction, where both languages were used in close contact. Sumerian words are rendered without italics, following the convention of Assyriology, which must distinguish between Akkadian and Sumerian words. Sumerian writings that were not necessarily pronounced as Sumerian words are presented in uppercase letters (e.g., DUMU, “son”). abu b¯ıtim (Akkadian). “Steward.” Administrative aide to the district governor (ˇsa¯ pit.um), in a monarchic setting. a¯ lum (Akkadian). “Town.” Covers all sizes from major cities to villages and hamlets; strong political aspect, as people who act. amurrû (Akkadian). “Amorrite.” A “westerner” from eastern Mesopotamian perspective; adapted as both adjective and proper name with various associations. archai (Greek). “Magistrates.” As a plural group of authorities, before full Athenian democracy. athû (Akkadian). “Brotherhood.” Referring to the affiliation of Yamutbal ˘and Simalite tribal groups. basileus (Greek). “Ruler, senior leader.” Eventually “king,” but earlier as part of small groups of ruling leaders. bin¯u yamina and bin¯u simal (West Semitic). “Sons/children of the right hand” and “of the left hand.” See DUMU, “son.” The two overarching tribal confederacies that dominated Syria of the Mari period. dannatum (Akkadian). “Fortified settlement.” Plural dann¯atum. darkatum (West Semitic). “Start(?).” Uncertain meaning, somehow derived from the spatial noun *darkum, “path, way.” 309

310


DUMU (Sumerian writing). “Son,” with the Sumerian writing covering either Akkadian m¯arum or West Semitic binum. As a plural, DUMU.MESˇ means “sons” or “children,” and when identified by a geographic designation, refers to the inhabitants of that place (e.g., m¯ar¯u Terqa, “children of Terqa”). With a tribal designation, it refers to the members of that group (e.g., bin¯u yamina, “children of the right hand”). E´ (Sumerian writing). “House,” for Akkadian (cf. West Semitic) b¯ıtum. el¯enum (Akkadian). “Higher country.” Not clear whether upstream or away from the valleys. en (Sumerian). “Ruler. “ A third-millennium term with varying use in east and west; originally a sacred figure, and coming to refer to a high priest, in second-millennium southern Mesopotamia. endan (Hurrian). “Ruler.” From earliest Hurrian texts at Urkesh (Urgiˇs). ensi(k) (Sumerian). “City ruler.” A Mesopotamian king conceived as the appointee of the ruling deity. eupatridai (Greek). “Well-born.” The larger class from which the archai leadership were chosen, before Athenian democracy. lu´ GAL.KU5 (Sumerian writing). See rabi pirsim. gayum (West Semitic). “Division.” Category of tribal social organization; among the Binu Simal, the first-order designation of tribal groups under the Simalite umbrella. gerseqqum (Akkadian). “Palace-born(?).” A class of people in the royal palace, perhaps those born to service there. hadannum (West Semitic form of Akkadian adannum). “Agreed date, ˘ appointment.” hals.um (Akkadian, from West Semitic?). “District, subsidiary region.” The ˘ subsidiary administrative districts in the larger states, representing an added level in political hierarchy. hana (West Semitic). “Tent-dwelling(?)” (perhaps from root h.ny, “to camp”). ˘ Defining the mobile pastoralist components of tribal peoples; as “(our) hana,” most often the Simalite pastoralists. The standard form hana is not ˘ easily explained; the appropriate Akkadian-style adjective hanû˘m is much ˘ rarer. happira- (Hittite). “Town.” Roughly equivalent to Sumerian uru and ˘ Akkadian a¯ lum, referring to settlements of every size. Originally a “place of trade”? hayatum (West Semitic). “Encampment.” Referring to both the people and ˘ the livestock in it. hazzannum (Akkadian?). “Royal representative.” Often translated as “mayor,” ˘ but this is a representative of an outside royal power in a conquered town, not a local leader. hibrum (West Semitic). “Nomadic community.” Especially among the Binu ˘ Yamina. iku (Akkadian). “Plot of land. ” As a measurement of area, adjectival ikû.


311

izbum (Akkadian). “Malformed lamb.” Newborn lamb associated with divine messages, when malformed, and the point of reference for formal divination. kadûm (West Semitic). “(Local leader).” Some kind of leader in the pastoralist community of the Yaminites (hibrum)? Possibly to be read as ˘ qadûm. kalam (Sumerian). “Land (of Sumer).” In the second millennium, equated with Akkadian m¯atum, but not at all the same original concept; limited to Sumerians. kaprum (Akkadian). “Village.” Subordinate settlement, in the vicinity of some larger center; plural kapr¯atum. k¯arum (Akkadian). “(Commercial) quay.” Merchant community or leadership, defined by the mooring places for boats. ki (Sumerian). “Place, location.” Used to designated place names as such. kidûtum (West Semitic). “(A rite of vassalage).” kirhum (West Semitic?). “Citadel.” The high fortified section of a larger town. ˘ kispum (Akkadian). “Rite for the dead.” kur (Sumerian). “Land (outside Sumer); mountain, highlands.” Contrast with kalam, the Sumerian land that is dominated by irrigated lowlands. laputtûm (Akkadian). “(District) deputy, (military) lieutenant.” Assistant to a primary leader, the ˇsa¯ pit.um in a district (hals.um) context. ˘ the population outside the libbi m¯atim (Akkadian). “Countryside.” Especially central capital. limum (West Semitic). “Tribe, people.” The first-order designation of tribal groups under the Yaminite umbrella. ´ (Sumerian writing). “Man.” Used as a silent “determinative” to desigLU nate all classes of male personnel or mixed-gender groups, as well as for ´ ˇ Akkadian aw¯ılum. As plural, written LU.ME S. lugal (Sumerian). “King.” In contrast to “en,” which was originally linked to sacred leadership in connection with the temple cult, this is originally a “great man,” without the temple role. ma-da (Sumerian). “Land.” A loan from Akkadian m¯atum into Sumerian. mahrêtum (West Semitic?). “Elders(?).” Somehow those having priority, ei˘ of rank or of age; a rare term in this use. ther malkum (West Semitic). “King, ruler.” The western counterpart of the Akkadian ˇsarrum. m¯ar ˇsiprim (Akkadian). “Envoy, ambassador.” A high-level diplomat. mar-tu (Sumerian). “Amorrite.” See amurrû. The Sumerian may be the original expression. m¯arum (Akkadian). “Son.” See DUMU. maskan¯u (West Semitic). “Inhabitants.” Plural, from the verb “sak¯anum,” “to take up residence.” m¯atum (Akkadian). “(Political) land.” The essential unit of political division in ancient Mesopotamia, most often but not always as kingdoms.

312


merhûm (West Semitic). “Chief of pasture” (from root ry, “to pasture”). Major ˘ leader of mobile pastoralist communities, often serving as intermediary with kings. MESˇ (Sumerian writing). Plural marker, not generally pronounced when used in Akkadian texts. muˇskênum (Akkadian). “Commoners.” The collective mass of royal subjects who are not direct beneficiaries of royal largesse. namlak¯atum (West Semitic). “Dominions.” Derived from the verb mal¯akum, “to rule”; referring to populations ruled outside the core m¯atum in larger kingdoms. naukrariai (Greek). An assembly before Athenian democracy. nawûm (West Semitic). “Steppe or backcountry highlands; flocks (of the steppe); pastoralist community (of the steppe).” Basic definition of pastoralist domain. niKhum (West Semitic). “(Pastoral) range(?).” Uncertain meaning, referring to˘ the territory covered by the pastoralist movements of the Binu Simal and the Binu Yamina tribal confederacies. n¯ıˇs ilim (Akkadian). “Oath.” Meaning “life of a god.” niˇsu¯ (Akkadian). “Dependents.” Often translated “people,” but not a political term. Unrelated to the above term, n¯ıˇs ilim (“oath”). n¯ubalum (West Semitic). “Litter(?).” Some kind of “carried” transportation for dignitaries. oikos (Greek). “House, household.” The basic social unit. pah¯arum (Akkadian). “To meet, assemble (intransitive); to gather, assemble ˘ (transitive, in form puhhurum).” ˘ ˘administer.” This word has a very wide range of paq¯adum (Akkadian). “To application. pirsum (Akkadian). “(Military) section.” A lower-order battalion. polis (Greek). “City.” Commonly associated with the “city-state”; the defining unit of democratic Athens. politai (Greek). “Citizens.” Derived from the word for “town” (polis). puhrum (Akkadian). “Meeting, assembly.” Commonly regarded as a formal ˘ institution, but most often a generic term for any meeting. purs¯atum (West Semitic). “Branches.” Referring to lines of relationship between large tribal groups. qaqqadum (Akkadian). “Head.” Plural qaqqad¯atum, for gathered “heads” of a polity; perhaps heads of households. rabi¯anum (Akkadian, from West Semitic?). “Town ruler, mayor.” A title also taken by certain early Amorrite rulers who had seized southern Mesopotamian urban centers. rabi pirsim (Akkadian). “(Military) section chief.” A lower-level military commander, on the order of a battalion commander. Written lu´ GAL.KU5 . r¯amum (West Semitic). “Stone monument.” Used in commemorative rites.


313

rihs.um (West Semitic). “Talks.” Political negotiations, especially involving ˘tribal peoples. .sa¯ bum (Akkadian). “(Military) troop, army; (work) crew.” Gathered contingents, generally (and perhaps always) of men only, performing both military and nonmilitary tasks. ˇsadûm (Akkadian). “Mountain, highland steppe.” sak¯anum (West Semitic). “To take up residence.” ˇsakkanakkum (Akkadian). “(Appointed) ruler.” Title taken by Mari kings during the late third millennium, contemporary with the Ur III dynasty. ˇsa¯ pit.um (West Semitic). “District governor, judge.” Responsible for some district (especially hals.um) under the authority of a king. ˘ sarr¯arum (West Semitic?). “Resister(?).” Someone who has evaded registration by census and loyalty oath in settled communities. ˇsarrum (Akkadian). “King.” The most coveted individual political status in Akkadian terminology, especially as ruler of a m¯atum. ˇsa sikkatim (Akkadian). “Registrar(?).” See also abu b¯ıtim; an administrative aide of the district governor (ˇsa¯ pit.um). ˇs¯ıbum (Akkadian). “Elder, witness.” As “elder,” refers to generic senior leadership in a group context; idea of “witness” derives from that. Written lu´ ˇ ˇ for Akkadian ˇs¯ıb¯utum. SU.GI, with plural MES, siniqtum (West Semitic?). “Systematic check(?).” An administrative control initiated by the palace. ˇsiprum (Akkadian). “Message; messenger, envoy.” s¯ırum (West Semitic). “Loyalty tax.” Paid by a vassal or subordinate polity to the suzerain. sug¯agum (West Semitic). “Leader.” Local leaders of towns, villages, and small mobile tribal units, with broad variation in status and scope. sug¯ag¯utum (West Semitic). “Leadership; leadership payment.” The status of sug¯agum and the fee owed the king for formal recognition of it within a monarchical political context. lu´ ˇ SU.GI (Sumerian writing). “Elder, witness.” See ˇs¯ıbum; plural written with ˇ MES. s.uh¯arum (Akkadian). “Subordinate.” Written lu´ TUR. ˘ tahtamum (West Semitic). “Council.” Particular to the towns of Imar and ˘Tuttul; a “seated” group. t¯ebibtum (Akkadian). “Census.” telipuri (Hittite). “District. ” Similar to the Akkadian hals.um. ˘ t.uppum (Akkadian). “Tablet.” The primary writing medium for cuneiform. ubbubum (Akkadian). “To take census.” u¯ g˜ (Sumerian). “People.” Written with a sign very similar to that of kalam, for the Sumerian “land.” ukkin (Sumerian). “Assembly.” See puhrum. In early writing, UKKIN is the ˘ be the “leader of the assembly.” assembly and GAL:UKKIN appears to

314


uru, written URU when the underlying reading is uncertain (Sumerian). “Town.” Essentially like the Akkadian a¯ lum. utne (Hittite). “(Political) land.” Roughly equivalent to the Akkadian m¯atum, written with Sumerian kur. warkûm (Akkadian). “Adjutant.” Meaning an aide of lower rank. waˇsa¯ bum (Akkadian). “To take up residence.” Basic meaning, “to sit, stay”; applied to anyone who changes lodgings or place of residence. As a participle, the w¯aˇsib¯utum are “those who take up residence,” and are thus at least temporarily sedentary. wedûtum (Akkadian). “Dignitaries(?).” A general class of people found in palace service. yarr¯adum (West Semitic). Uncertain meaning. Apparently those who “come down,” perhaps pastoralists to the river valley from higher country.

Glossary of Proper Names

This volume is full of names: names of regions, names of peoples, names of gods, names of kings, and so on. Most will be unfamiliar to most readers, as they were to me when I began this project. I intend this glossary not to be exhaustive, but rather to serve as a basic point of reference for some of the more important names. With the hope of improving convenience, I have divided this glossary into a number of smaller sections. In some cases, the crucial detail requires not a separate definition but the gathering of separate names into groups.

a. geography (see maps) 1. Regions Ah Purattim. “Banks-of-the-Euphrates.” The region of the Euphrates River ˘valley immediately surrounding Mari and the name for the main settled population ruled from Mari. Anatolia. Ancient Turkey, north and west of Mesopotamia. Jezireh. The region of northeastern modern Syria that stretches across the upper Habur River valley. ˘ Mesopotamia. “Between the (Two) Rivers.” The land defined by the Tigris and the Euphrates Rivers. These rivers descend from the mountains of east-central Turkey into Syria and Iraq, joining before they empty into the Persian Gulf. Sometimes “Mesopotamia” is used as if it referred primarily to ancient Iraq, though in fact it should include much of northern Syria and some of southern Turkey. Near East. Scholars who study ancient Mesopotamia still use the somewhat archaic (and “Orientalist”) category of “Near East” to combine all of the lands of southwest Asia, including mainly the regions now occupied by the Arabian states and Iraq, Israel, Jordan, Lebanon, Syria, and Turkey. Egypt and Iran would be separate. 315


316

Sumer. Ancient “Kengir,” the southeastern Mesopotamian region distinguished by shared use of the Sumerian language.

2. Third millennium (not exhaustive, but by reference) Southern Mesopotamian Sumer Lagaˇs, Larsa, Nippur, Puzriˇs-Dagan (Drehem), Sippar, Umma, Ur, Uruk East-central Mesopotamia Agade (Akkad), Akˇsak, Kiˇs Northern Mesopotamia Aˇsˇsur, Gasur (=Nuzi), Nineveh, Tell al-Raq¯ai, Tell Atij, Tell Leilan, Urkesh (Urgiˇs) Western Syria Ebla, Tell Banat

3. Period of the Mari archives (early second millennium) Southern Mesopotamia Isin, Larsa, Nippur, Sippar, Uruk East-central Mesopotamia Babylon, Eˇsnunna, Shemshara East-northern Mesopotamia Andarig, Aˇsˇsur, Ekallatum, Karanâ/Qat.t.arâ, Kasallu(k), Kurdâ, Niniveh, ˇ Qabrâ, Razamâ, Subartum (as a coalition of kings), Tillâ Middle Euphrates valley Abattum, Hˆıt, Imar (Emar), Mari, Suhûm, Terqa, Tuttul ˘ ˘ Habur River basin ˘ ˇ ˇ hna, Tell Leilan), Tadum, Urgiˇs Apum, Chagar Bazar, Subat-Enlil (Se ˘ *Ida-Maras. (coalition): Aˇslakkâ, Aˇsnakkum, Hurrâ, Ilan-s.urâ, ˘ Isqâ-and-Qâ (or Qâ-and-Isqâ), Kahat, Nahur, Qirdahat, Susâ, ˘ ˘ ˘ ˇ ˇ a, Talhayˆ Sudu hhum, Sunˆ um, Tamarzi, Zalluhan ˘˘ ˘ ˘ Northwestern Mesopotamia (Balih River basin, Euphrates, and west) ˘


317

Alalah (=Alahtum?), Carchemish, Der, Harran ˘ ˘ ˘ ˇ a *Zalmaqum (coalition): Hanzat, Harran, Nihriya, Sudˆ ˘ ˘ ˘ Western Syria Aleppo (Halab; kingdom of Yamhad), Amurrum, Qatna ˘ ˘

4. The Mari kingdom of Zimri-Lim The district of Mari Mari, Miˇslan (Yaminite), S.uprum The district of Terqa Dabiˇs (Yaminite), Hiˇsamta, Samanum, Terqa ˘ The district of Saggaratum Dur-Yahdun-Lim, Saggaratum ˘ The district of Qat.t.unan Qat.t.unan, T.abatum The Suhˆ um ˘ Sapiratum

b. populations, social groupings, language categories 1. Tribal names gayum “divisions” of the Binu Simal (Simalite) tribal confederacy (selective) Amurrum, Nihadˆ u, Yabasu, Yumahammˆ u ˘ ˘ *As military duality: Aˇsarugayum, Yabasu “Tribes” of the Binu Yamina (Yaminite) tribal confederacy (all five) Amnanû, Rabbˆ u, Uprapˆ u, Yahrurˆ u, Yarihˆ u ˘ ˘ Tribal groups outside the left/right hand duality Mutiabal (Mutebal), Numhâ (center at Kurdâ), Turukkˆ u (center at Shemshara, east of Tigris),˘ Yamutbal(um) (center at Andarig; =Emutbal of Larsa)

318


2. Nontribal social categories Amorrite. “Western(er).” First a description of people from the west, from the perspective of southeastern Mesopotamia (especially Sumer); later adopted to describe various political and cultural entities by those of the “west” themselves. Hana (hana). “Tent-dwellers(?)” as a general group, no matter the tribe or ˘ political ˘ affiliation. Sutˆ u. Nomadic peoples of the deserts south of the Euphrates; outside the Syrian bounds conceived for the Yaminite/Simalite duality; not a single political entity (unlike Turukkˆ u).

3. Languages Akkadian. The “Northeastern” Semitic language defined in terms of thirdmillennium Agade (Akkad), the first regional empire. Later includes the dialects of eastern Mesopotamian Aˇsˇsur, Babylon, and Eˇsnunna, along with the regions they controlled. Takes over the old region of Sumerian speech. Hittite. The Indo-European language of ancient Anatolia, with the kingdoms and cultures based there. Hurrian. The language of a population first proven to inhabit Mesopotamia during the late third millennium, especially centered at Urkesh (Urgiˇs). Hurrian speakers spread across most of northern Mesopotamia. Unrelated to any of the other Mesopotamian or Anatolian languages. Old Babylonian. The dialect of Akkadian dominant in southeastern Mesopotamia during the early second millennium; serves as the scribal standard across most of western and northern Mesopotamia as well, even where other languages are native. Semitic. The family of languages that includes Arabic and Hebrew, as well as ancient Akkadian, Aramaic, Phoenician, Ugaritic, and various dialects related to these languages. Subarian (Subarˆ u). The Mari-period name for a language described as separate from Sumerian, Akkadian, and “Amorrite”; probably a dialect of Hurrian. Sumerian. The early language of the southeastern Mesopotamian river plains; unrelated to any other known language. West Semitic. A broad term that I use to designate the set of early secondmillennium dialects that show features eventually shared by “western” languages such as Ugaritic, Hebrew, and Aramaic. This West Semitic language group includes but is not limited to dialects that the people of the Mari archives would have identified as “Amorrite.”


319

c. individuals 1. Deities Addu. The West Semitic form of the storm god, identical to Akkadian Adad. Annunitum. A goddess worshipped especially among the harem of the Mari palace; identified with Eˇstar. Dagan. The chief god of the middle Euphrates pantheon during the third and second millennia, especially from Mari upstream through Terqa, Tuttul, and Imar. Enlil. The chief god of the southern Mesopotamian pantheon during the late third and early second millennia. Based at Nippur. Eˇstar. The primary Semitic name for the divine type of goddess as young woman, associated with both love and death, equated with Sumerian Inanna. Marduk. The chief god of Babylon, suddenly prominent with the ascendance of the first dynasty of Babylon. Nergal. Mesopotamian god of war and death. ˇ Samaˇ s. The Akkadian sun god, responsible especially for justice. Worshipped under some form of this name throughout the Semitic-speaking world, ˇ su at Ugarit. though female as Sapˇ

2. Rulers (dates from DCM, according to the so-called middle chronology) Adal-ˇsenni. A king of Burundum during the reign of Zimri-Lim. Ammi-s.aduqa. Second to last king in the first dynasty of Babylon; ruled ca. 1646–1626. Amut-pi-el. King of Qatna during the reign of Zimri-Lim. Asdi-takim. King of Harran and a leading ruler in the confederacy of ˘ reign of Zimri-Lim. Zalmaqum, during the Asqur-Addu. King of Qat.t.arâ/Karanâ during the later part of Zimri-Lim’s reign. Aˇstamar-Addu. A king of Kurdâ during the reign of Yahdun-Lim. Atamrum. Briefly king of Andarig during the later ˘part of Zimri-Lim’s reign. Bunu-Eˇstar. King of Kurdâ during the middle part of Zimri-Lim’s reign, between Simah-ilanê and Hammurabi. ˘ of the Rabbˆ ˘ u tribe of the Binu-yamina during the time Dadi-hadˆ un. Ruler ˘ of Zimri-Lim. Based at Abattum, near Imar. Enmetena. A king of Lagaˇs during the mid-third millennium. Hamman. The local ruler at Der (of the Balih), called a sug¯agum. ˘ Hammi-iˇ stamar. Ruler of the Uprapˆ u tribe of˘ the Binu-Yamina during the ˘ time of Zimri-Lim.

320


ˇ Hammi-kun. King of Sudu hhum in Ida-Maras. after Yatar-malik, during most ˘ of Zimri-Lim’s reign. ˘ ˘ Hammurabi of Aleppo. The last king of Aleppo/Yamhad during the reign ˘ of Zimri-Lim. ˘ Hammurabi of Babylon. King of Babylon, ca. 1792–1750. Forced Zimri˘ Lim from the throne of Mari and subsequently destroyed the city’s public structures. Hammurabi of Kurdâ. The last king of Kurdâ during the reign of Zimri-Lim, ˘ who turned against the Simalite king at the end, in favor of Babylon. Hatnu-rabi. King of Qat.t.arâ/Karanâ during the early part of Zimri-Lim’s ˘ reign. Haya-sumu. King of Ilan-s.urâ in Ida-Maras. through the entire reign of ˘ Zimri-Lim. Ibal-Addu. King of Aˇslakkâ, one of the primary centers of Ida-Maras., through most of Zimri-Lim’s reign. Ibal-pi-el II of Eˇsnunna. King of this major south-central Mesopotamian kingdom, ca. 1779–1765. ˇ ˇ hna in the Habur River Ibni-Addu. King of Tadum, south of Subat-Enlil/ Se ˘ ˘ basin, during the reign of Zimri-Lim. Ila-kabkabu. The father of Samsi-Addu of the northern Mesopotamian kingdom; ruled from Ekallatum on the Tigris River. Irikagina. A king of Lagaˇs during the mid-third millennium. Iˇsbi-Erra. The first king of the Isin dynasty (ca. 2017–1985), after the fall of the Ur III kingdom of Sumer. Iˇsme-Addu. Briefly king of Aˇsnakkum in Ida-Maras. during the Elamite invasion of northern Mesopotamia, toward the end of Zimri-Lim’s reign. Iˇsme-Dagan. King of Ekallatum and the eastern portion of Samsi-Addu’s northern Mesopotamian kingdom, as Samsi-Addu’s older son; survives the death of his father in 1775, to maintain a truncated kingdom at Ekallatum. Kabiya. Briefly king of Kahat in Ida-Maras. during the later part of Zimri˘ Lim’s reign. Kudur-mabuk. Ruler of the Yamutbal tribal people who take over Larsa, ca. 1835. Father of Warad-Sˆın and Rim-Sˆın, important Larsa kings. Lugalzagesi. King of the Sumerian city of Umma shortly before the empire created by Sargon of Agade, ca. twenty-fourth century. Qarni-Lim. King of Andarig during the first part of Zimri-Lim’s reign, always a faithful vassal. Rim-Sˆın. Second king of Larsa from the Yamutbal family of Kudur-mabuk; brother of Warad-Sˆın and son of Kudur-mabuk; ruled ca. 1822– 1763. ˇ Samsi-Addu. King of Mari, ca. 1792–1782, from a base at Subat-Enlil ˇ (Sehna/Tell Leilan); ruler of an extensive northern Mesopotamian kingdom˘ that began from a base at Ekallatum on the Tigris River (until 1775).


321

Samsi-erah. Ruler of Tillâ, northwest of the Jebel Sinjar, during the reign of ˘ Zimri-Lim. Samsu-iluna. Successor to king Hammurabi of Babylon, at the height of the first dynasty of Babylon; ruled˘ ca. 1749–1712. Sargon. Founding king of the empire based at Agade (Akkad), ca. 2335– 2279. Simah-ilanê. King of Kurdâ during the early part of Zimri-Lim’s reign. ˘ Sˆın-gamil. Second king after Sˆın-Kaˇsid, from the Yaminite Amnanum dynasty at Uruk; ca. 1826–1824. Sˆın-kaˇsid. Founder of a new dynasty at Uruk, from the Yaminite Amnanum tribe; ruled ca. 1865/60–1833(?). Sumu-dabi. A Yaminite ruler before the revolt against Zimri-Lim, early in his reign. Sumu-epuh. King of Aleppo/Yamhad during the reign of Yahdun-Lim of ˘ ˘ Mari. ˘ Sumu-yamam. King of the Binu Simal at Mari, ca. 1793–1792. S.urahammˆ u of the Binu Yamina. Ruler of the Amnanû tribe during the time of ˘Zimri-Lim. ˇ Sadum-labua. King of Aˇsnakkum in Ida-Maras. for the last few years of ZimriLim’s reign. ˇ Sar-kali-ˇ sarri. Last king of Agade (Akkad), ca. 2217–2193. ˇ Subram. King of Qirdahat and then Susâ in Ida-Maras., among other roles, ˘ during the reign of Zimri-Lim. Terru. King of Urgiˇs during the last part of Zimri-Lim’s reign; quite weak, dependent on the support of the Mari king. Warad-Sˆın. First king of Larsa from the Yamutbal family of Kudur-mabuk, ca. 1834–1823. Yaggid-Lim. Father of Yahdun-Lim, the king of the Binu-Simal. Yahdun-Lim. King of the˘Binu Simal at Mari, ca. 1810–1794. ˘ Yarim-Lim of Aleppo. King of Aleppo/Yamhad overlapping the reigns of ˘ Yasmah-Addu and Zimri-Lim. ˘ Yarim-Lim of the Binu Yamina. Ruler of the Yahrurˆ u tribe during the time ˘ of Zimri-Lim. Yarkab-Addu. King of Hanzat, one of the centers of Zalmaqum, during the time of Zimri-Lim. ˘ Yasmah-Addu of the Binu Yamina. Ruler of the Yarihû tribe during the time ˘ ˘ of Zimri-Lim. Yasmah-Addu of Mari. King of Mari and the western portion of Samsi˘ northern Mesopotamian kingdom, as Samsi-Addu’s younger son Addu’s (ca. 1782–1775). Defeated soon after the death of his father. Yatar-ami. Briefly king of Carchemish during the last part of Zimri-Lim’s reign. ˇ Yatar-malik. King of Sudu hhum in Ida-Maras. during the early years of Zimri˘˘ Lim’s reign.

322


Yawi-el. King of Talhayˆ um during the middle part of Zimri-Lim’s reign. Zaziya. King of the˘Turukkˆ u tribespeople, east of the Tigris River, during the reign of Zimri-Lim. Zimri-Lim. King of the Binu Simal at Mari, ca. 1774–1762.

3. Other important men and women Asqudum. Trained as a diviner under the reign of Yasmah-Addu, brought into Zimri-Lim’s inner ruling circle against the wishes of˘ Bannum. Bahdi-Lim. Governor of the Mari district under Zimri-Lim. ˘ Bannum. A Simalite merhûm before Zimri-Lim becomes king; the ac˘ of Mari, but allows Zimri-Lim the throne, tual conqueror of the city becoming his vizier. Dies within the first two years of Zimri-Lim’s reign. Buqaqum. One of the two principal leaders of the Suhˆ um region during the ˘ reign of Zimri-Lim. Dariˇs-libur. High official of Zimri-Lim, sent as envoy to the kingdom of Aleppo/Yamhad. ˘ of Aleppo/Yamhad, wife of Yarim-Lim and queen mother Gaˇsera. The queen during the reign of Hammurabi˘(of Aleppo). ˘ leader of the Simalite Hana during the early part Hali-hadun. A prominent ˘ of Zimri-Lim’s ˘ ˘ reign. Haqba-Hammu. The second-ranking leader and brother-in-law of king ˘ Asqur-Addu ˘ of Qat.t.arâ/Karanâ, during the reign of Zimri-Lim. Ibal-el. One of the two Simalite “chiefs of pasture” (merhûm) during the later ˘ part of Zimri-Lim’s reign. Ibal-pi-el. One of the two Simalite “chiefs of pasture” (merhûm) during the ˘ later part of Zimri-Lim’s reign. Iluˇsu-nas.ir. A governor of Qat.t.unan under Zimri-Lim. Itur-Asdu. A member of Zimri-Lim’s inner ruling circle. Kibri-Dagan. Governor of the Terqa district under Zimri-Lim. Kirû. Daughter of Zimri-Lim, married to and then divorced from Haya˘ sumu, king of Ilan-s.urâ. Lanasûm. Zimri-Lim’s royal representative (hazzannum) at the western town ˘ of Tuttul. um region during the Meptˆ um. One of the two principal leaders of the Suhˆ ˘ reign of Zimri-Lim. Ripi-Dagan. A Simalite of high status under the reign of Zimri-Lim. Sumhu-rabi. An early governor of the Saggaratum district under Zimri-Lim. ˘ hadˆ Sumuu. A high official during the reign of Zimri-Lim, with responsibilˇ ities ˘at Mari and at Saggaratum. Tarim-ˇsakim. One of Yasmah-Addu’s two viziers at Mari. ˘


323

Yaqqim-Addu. Governor of the Saggaratum district for roughly the second half of Zimri-Lim’s reign. Yarim-Addu. Possibly the name of a second Simalite merhûm, with Bannum, at the end of Yasmah-Addu’s reign, overlapping into˘ the start of Zimri˘ Lim’s reign. Yasim-el. A high official based at Mari during the reign of Zimri-Lim. Zakira-Hammu. A governor of the Qat.t.unan district under Zimri-Lim. ˘

Bibliography

Abrahami, Philippe. 1992. “Yabisa/Yabasa/Yabusu.” N.A.B.U., no. 29, pp. 25–6. Adams, Robert McC. 1972. “Patterns of Urbanization in Early Southern Mesopotamia.” Pp. 735–49 in Peter J. Ucko et al. (eds.), Man, Settlement and Urbanism. London: Duckworth. . 1974. “The Mesopotamian Social Landscape: A View from the Frontier.” Pp. 1–11 in Charlotte B. Moore (ed.), Reconstructing Complex Societies. Cambridge, Mass.: American Schools of Oriental Research. . 1981. Heartland of Cities: Surveys of Ancient Settlement and Land Use on the Central Floodplain of the Euphrates. Chicago: University of Chicago. . 1982. “Property Rights and Functional Tenure in Mesopotamian Rural Communities.” Pp. 1–14 in J. N. Postgate (ed.), Societies and Languages of the Ancient Near East: Studies in Honour of I. M. Diakonoff. Warminster: Aris and Phillips. . 1988. “Contexts of Civilizational Collapse: A Mesopotamian View.” Pp. 20– 43 in Norman Yoffee and George L. Cowgill (eds.), The Collapse of Ancient States and Civilizations. Tucson: University of Arizona Press. Alp, Sedat. 1991. Hethitische Briefe aus Mas¸at-Höyük. Ankara: Türk Tarih Kurumu Basimevi. Anbar, Moshe. 1985. “La distribution géographique des Bini-Yamina d’après les archives royales de Mari.” Pp. 17–24 in Jean-Marie Durand and Jean-Robert ´ Kupper (eds.), Miscellanea Babylonica: Mélanges offerts a` Maurice Birot. Paris: Editions Recherche sur les Civilisations. . 1991. Les tribus amurrites de Mari. Göttingen: Vandenhoeck und Ruprecht. ¨ Tunca . 1996. “L’origine tribale de Zimri-Lim, roi de Mari.” Pp. 7–10 in O. and D. Deheselle (eds.), Tablettes et images aux pays de Sumer et d’Akkad, Mélanges offerts a` Monsieur H. Limet. Liège: Université de Liège. Archi, Alfonso. 1979/80. “Les dieux d’Ebla au IIIe millénaire avant J. C. et les dieux d’Ugarit.” AAAS 29–30: PP. . 1985. “Mardu in the Ebla Texts.” Or 54: 7–13. . 1990. “Imâr au IIIe` me millénaire d’après les archives d’Ebla.” M.A.R.I. 6: 21–38. . 1993. “Fifteen Years of Studies on Ebla: A Summary.” OLZ 88: 461–71. 325

326

Bibliography

. 1995. “Polity Interaction in the Age of Ebla.” Pp. 13–17 in Farouk Ismail (ed.), Proceedings of the International Symposium on Syria and the Ancient Near East 3000–300 B.C. Aleppo: University of Aleppo. . 1996. “Chronologie relative des archives d’Ebla.” Amurru 1: 11–28. Aynard, J.-M., and Agnès Spycket. 1987–90. “Mari.B.Archäologisch.” RlA 7: 390– 418. Bailey, F. G. 1965. “Decisions by Consensus in Councils and Committees: With Special Reference to Village and Local Government in India.” Pp. 1–20 in Political Systems and the Distribution of Power. London: ASA Monographs no. 2. Balandier, Georges. 1970. Political Anthropology. New York: Random House. Beckman, Gary. 1999a. “The City and the Country in Hatti.” Pp. 161–69 in Horst Klengel and Johannes Renger (eds.), Landwirtschaft im˘alten Orient. RAI 41. Berlin: Dietrich Reimer. . 1999b. Hittite Diplomatic Texts, 2nd ed. Atlanta: Scholars Press. Birot, Maurice. 1953. “Trois textes e´ conomiques de Mari (I), (II).” RA 47: 121–30, 161–79. . 1955. “Trois textes e´ conomiques de Mari (III).” RA 49: 15–31. . 1985. “Les chroniques ‘assyriennes’ de Mari.” M.A.R.I. 4: 219–42. ¨ Tunca . 1990. “La lettre de Yarˆım-Lim No 72–39 + 72–8.” Pp. 127–35 in O. (ed.), De la Babylonie a` la Syrie, en passant par Mari. Mélanges Kupper. Liège: Université de Liège. Blanton, Richard E. 1998. “Beyond Centralization: Steps Toward a Theory of Egalitarian Behavior in Archaic States.” Pp. 135–72 in Gary M. Feinman and Joyce Marcus (eds.), Archaic States. Santa Fe: School of American Research. Blanton, Richard E., Stephen A. Kowalewski, Gary M. Feinman, and Laura M. Finsten. 1981. Ancient Mesoamerica: A Comparison of Change in Three Regions. Cambridge: Cambridge University Press. Blanton, Richard E., Gary M. Feinman, Stephen A. Kowalewski, and Peter N. Peregrine. 1996. “A Dual-Processual Theory for the Evolution of Mesoamerican Civilization.” Current Anthropology 37: 1–14. Bonechi, Marco. 1992. “Relations amicales syro-palestiniennes: Mari et Has. or au XVIIIe siècle avant J.-C.” FM I: 9–22. . 1993. I nomi geografici dei testi di Ebla. Wiesbaden: Dr. Ludwig Reichert. . 1997. “Lexique et idéologie royale a` l’époque proto-syrienne.” M.A.R.I. 8: 477–535. . 1998. “Remarks on the III Millennium Geographical Names of the Syrian Upper Mesopotamia.” Pp. 219–41 in Subartu IV, vol. 1. Bonechi, Marco, and Amalia Catagnoti. 1994. “Compléments a` la correspondance de Yaqqim-Addu, gouverneur de Saggarâtum.” FM II: 55–82. Bonte, Pierre. 1979. “Pastoral Production, Territorial Organisation and Kinship in Segmentary Lineage Societies.” Pp. 203–34 in P. C. Burnham and R. F. Ellen (eds.), Social and Ecological Systems. London: Academic Press. Bottéro, Jean. 1958. “Lettres de la salle 110 du palais de Mari.” RA 52: 163–76. Braun, David P., and Stephen Plog. 1982. “Evolution of ‘Tribal’ Social Networks: Theory and Prehistoric North American Evidence.” American Antiquity 47: 504– 25. Buccellati, Giorgia. 1966. The Amorites of the Ur III Period. Naples: Istituto Orientali di Napoli.

Bibliography

327

. 1977. “The ‘Urban Revolution’ in a Socio-Political Perspective.” Mesopotamia 12: 19–39. . 1988. “The Kingdom and Period of Khana.” BASOR 270: 43–61. . 1990. “ ‘River Bank,’ ‘High Country,’ and ‘Pasture Land’: The Growth of Nomadism on the Middle Euphrates and the Khabur.” Pp. 87–117 in Seyyare Eichler, Markus Wäfler, and David Warburton (eds.), Tall al-H . am¯ıd¯ıya 2. Göttingen: Vandenhoeck und Ruprecht. . 1991. “A Note on the muˇsk¯enum as ‘Homesteader.’ ” Maarav 7: 91–100. . 1992. “Ebla and the Amorrites.” Eblaitica 3: 83–104. . 1998. “Urkesh as Tell Mozan: Profiles of the Ancient City.” Pp. 11–34 in Buccellati and Marilyn Kelly-Buccellati (eds.), Urkesh and the Hurrians: Studies in Honor of Lloyd Cotsen. Bibliotheca Mesopotamica 26. Malibu, Calif.: Undena. Buccellati, Giorgio, and Marilyn Kelly-Buccellati. 1995/6. “The Royal Storehouse of Urkesh: The Glyptic Evidence from the Southwestern Wing.” AfO 42–3: 1–32. . 1996. “The Seals of the King of Urkesh: Evidence from the Western Wing of the Royal Storehouse Ak.” WZKM 86: 65–97. . 1997a. “Terqa.” Pp. 188–90 in OEANE. . 1997b. “Urkesh: The First Hurrian Capital.” BA 60: 77–96. . 1999. “Das archäologische Projekt Tall Mozan/Urkeˇs.” MDOG 131: 7–16. ¨ . 2001. “Uberlegungen zur funktionellen und historischen Bestimmung des Königspalastes von Urkeˇs. Bericht uber ¨ die 13. Kampagne in Tall Mozan/Urkeˇs: Ausgrabungen im Gebiet AA, Juni-August 2000.” MDOG 133: 59–96. Cartledge, Paul. 1998. “Writing the History of Archaic Greek Political Thought.” Pp. 379–99 in Nick Fisher and Hans van Wees (eds.), Archaic Greece: New Approaches and New Evidence. London: Duckworth. Cassin, Elena. 1973. “Note sur le ‘puhrum’ des dieux.” Pp. 111–18 in André Finet ˘ ´ (ed.), La voix de l’opposition en Mésopotamie. Brussels: Institut des Hautes Etudes Belgique. Catagnoti, Amalia, and Marco Bonechi. 1992. “Le volcan Kawkab, Nagar et problèmes connexes.” N.A.B.U., no. 65, p. 52. Chambon, Grégory. 2002. “Trois documents pédagogiques de Mari.” FM VI: 497– 503. Charpin, Dominique. 1984. “Inscriptions votives d’époque assyrienne.” M.A.R.I. 3: 41–81. . 1987a. “Tablettes présargoniques de Mari.” M.A.R.I. 5: 65–127. ˇ . 1987b. “Subat-Enlil et le pays d’Apum.” M.A.R.I. 5: 129–40. . 1987c. “Mallanum et Mallanate.” N.A.B.U., no. 38, pp. 21–2. . 1990a. “A Contribution to the Geography and History of the Kingdom of Kahat.” Pp. 67–85 in Tall al-H . am¯ıd¯ıya 2. Göttingen: Vandenhoeck und Ruprecht. . 1990b. “Conclusions et perspectives: Tell Mohammed Diyab, une ville du pays d’Apum.” Pp. 117–22 in Jean-Marie Durand (ed.), Tell Mohammed Diyab, Campagnes 1987 et 1988. Cahiers de N.A.B.U. 1. Paris: SEPOA. . 1990c. “Une alliance contre l’Elam et le rituel du lipit napiˇstim.” Pp. 109–18 in Fran¸cois Vallat (ed.), Mélanges Jean Perrot, Contribution a` l’histoire de l’Iran. Paris: ´ Editions Recherche sur les Civilisations. . 1991a. “Les mots du pouvoir dans les archives royales de Mari (XVIIIème av.J.-C.).” Pp. 3–17 in Cahiers du Centre G. Glotz II. Paris: De Boccard.

328

Bibliography

. 1991b. “Un traité entre Zimri-Lim de Mari et Ibâl-pˆı-El II d’Eˇsnunna.” Pp. 139–66 in Mélanges Garelli. . 1992a. “Les champions, la meule et le fleuve, ou le rachat du terroir de Puzurrân au roi d’Eˇsnunna par le roi de Mari Yahdun-Lim.” FM I: 29–38. . 1992b. “De la vallée du Tigre au ‘triangle du Habur’: Un engrenage géopolitique.” Pp. 97–102 in Jean-Marie Durand (ed.), Recherches en Haute Mésopotamie. Mémoires de NABU 2. Paris: SEPOA. . 1992c. “Hanéens et Simalites.” N.A.B.U., no. 31, pp. 26–7. . 1992d. “Les légendes de sceaux de Mari: Nouvelles données.” Pp. 59–76 in G. D. Young (ed.), Mari in Retrospect. Winona Lake, Ind.: Eisenbrauns. . 1993a. “Données nouvelles sur la poliorcétique a` l’époque paléobabylonienne.” M.A.R.I. 7: 193–203. . 1993b. “Un souverain e´ phémère en Ida-Maras. : Iˇsme-Addu d’Aˇsnakkum.” M.A.R.I. 7: 165–91. . 1993/4. “Compte rendu de: CAD volume S (1984).” AfO 40/41: 1–23. . 1994. “Compte rendu de: Mario Liverani (ed.), Akkad: The First World Eurpire.” RA 88: 181–3. . 1995a. “La fin des archives dans le palais de Mari.” RA 89: 29–40. ` propos des erreurs de scribes dans les lettres de . 1995b. “ ‘Lies natürlich . . .’: A Mari.” Pp. 43–56 in Manfried Dietrich and Oswald Loretz (eds.), Vom Alten Orient zum Alten Testament. Festschrift Wolfram von Soden. Neukirchen-Vluyn: Neukirchener. . 1997a. “Manger un serment.” Méditerranées 10–11: 85–96. . 1997b. “Sapˆıratum, ville du Suhˆ um.” M.A.R.I. 8: 341–66. . 1997c. “La version mariote de l’insurrection générale contre Narâm-Sˆın.’” FM 3: 9–18. 1998. “L’évocation du passé dans les lettres de Mari.” Pp. 91–110 in Jiri Prosecky (ed.), Intellectual Life of the Ancient Near East. CRRAI 43. Prague: Oriental Institute. . 2000. “Lettres et procès paléo-babyloniens.” Pp. 69–111 in Francis Joannès (ed.), Rendre la justice en Mésopotamie. Saint Denis: Presses Universitaires de Vincennes. . 2001. “Eˇsnunna (rois).” Pp. 314–18 in Francis Joannès (ed.), DCM. . forthcoming a. “Mari et le Proche-Orient a` l’époque amorrite: Essai d’histoire politique. Le règne de Zimrˆı-Lˆım (1775–1761).” In FM V. . forthcoming b. “Mari und die Assyrer.” CDOG. . forthcoming c. “Nomades et sédentaires dans l’armée de Mari du temps de Yahdun-Lˆım.” CRRAI 46. Amurru 3. Charpin, Dominique, and Jean-Marie Durand. 1985. “La prise du pouvoir par ZimriLim.” M.A.R.I. 4: 293–343. . 1986. “ ‘Fils de Simal’: Les origines tribales des rois de Mari.” RA 80: 141–81. . 1997. “Aˇsˇsur avant l’Assyrie.” M.A.R.I. 8: 367–91. Charpin, Dominique, and Martin Sauvage. 2001. “Sippar.” Pp. 782–4 in DCM. Childe, V. Gordon. 1936. Man Makes Himself. London: Watts. . 1942. What Happened in History. London: Max Parrish. Cohen, Raymond, and Raymond Westbrook, eds. 2000. Amarna Diplomacy: The Beginnings of International Relations. Baltimore: Johns Hopkins University Press. Cooper, Jerrold S. 1973. “Sumerian and Akkadian in Sumer and Akkad.” Or 42: 239–42.

Bibliography

329

. 1983. The Curse of Agade. Baltimore: Johns Hopkins University Press. . 1986. Sumerian and Akkadian Royal Inscriptions, I: Presargonic Inscriptions. New Haven: American Oriental Society. Dalley, Stephanie, C. B. F. Walker, and J. D. Hawkins. 1976. The Old Babylonian Tablets from Tell al Rimah. British School of Archaeology in Iraq. Dandamayev, Muhammed. 1982. “The Neo-Babylonian Elders.” Pp. 38–41 in Dandamayev et al. (eds.), Societies and Languages of the Ancient Near East. Warminster: Aris and Phillips. . 1995. “Babylonian Popular Assemblies in the First Millennium B.C.” BCSMS 30: 23–9. DeMarrais, Elizabeth, Luis Jaime Castillo, and Timothy Earle. 1996. “Ideology, Materialization, and Power Strategies.” Current Anthropology 37: 15–31. Diakonoff, I. M. 1991. “General Outline of the First Period of the History of the Ancient World and the Problem of the Ways of Development.” Pp. 27–66 in Diakonoff (ed.), Early Antiquity. Chicago: University of Chicago Press. Donlan, Walter. 1997. “The Relations of Power in the Pre-State and Early State Polities.” Pp. 39–48 in Lynette G. Mitchell and P. J. Rhodes (eds.), The Development of the Polis in Archaic Greece. London: Routledge. Dossin, Georges. 1939. “Benjaminites dans les textes de Mari.” Pp. 981–96 in Mélanges syriens offerts a` M. René Dussaud. Paris. . 1955. “L’inscription de fondation de Iahdun-Lim, roi de Mari.” Syria 32: 1–28. . 1957. “Kengen, pays de Canaan.” RSO 32: 35–9. . 1959. “Le palais – documents et monuments.” In MAM II/3. . 1972a. “Adaˇsˇsum et kirhum dans les textes de Mari.” RA 66: 111–30. . 1972b. “Le madârum dans les archives royales de Mari.” Pp. 53–63 in Gesellschaftsklassen im Alten Zweistromland. CRRAI 18. Munich. . 1974. “Le site de Tuttul-sur-Balih.” RA 68: 25–34. Drews, Robert. 1983. Basileus: The Evidence for Kingship in Geometric Greece. New Haven: Yale University Press. Driver, G. R., and John C. Miles. 1952. The Babylonian Laws. Oxford: Clarendon. ˇ Durand, Jean-Marie. 1985a. “La situation historique des Sakkanakku: Nouvelle approche.” M.A.R.I. 4: 147–72. . 1985b. “Les dames du palais de Mari a` l’époque du royaume de HauteMésopotamie.” M.A.R.I. 4: 385–436. . 1986. “Fragments rejoints pour une histoire e´ lamite.” Pp. 111–28 in Léon de Meyer et al. (eds.), Fragmenta Historiae Elamicae. Mélanges M.-J. Stève. Paris: ´ Editions Recherche sur les Civilisations. . 1987a. “Documents pour l’histoire du royaume de Haute-Mésopotamie I.” M.A.R.I. 5: 155–98. . 1987b. “Villes fantômes de Syrie et autres lieux.” M.A.R.I. 5: 199–234. . 1987c. “Questions de chiffres.” M.A.R.I. 5: 605–10. . 1988. “Les anciens de Talhayˆ um.” RA 82: 97–113. . 1989. “L’assemblée en Syrie a` l’époque pré-amorite.” Pp. 27–44 in Miscellanea Eblaitica 2. Florence: Università di Firenze. ¨ . 1990a. “ARM III, ARM VI, ARMT XIII, ARMT XXII.” Pp. 149–77 in O. Tunca (ed.), De la Babylone a` la Syrie, en passant par Mari. Mélanges Kupper. Liège: Université de Liège. ´ d’Imâr a` l’époque des rois de Mari.” M.A.R.I. 6: 39–92. . 1990b. “La cité-Etat

330

Bibliography

. 1990c. “Documents pour l’histoire du royaume de Haute Mésopotamie II.” M.A.R.I. 6: 271–301. . 1990d. “Fourmis blanches et fourmis noires.” Pp. 101–8 in Fran¸cois Vallat ´ (ed.), Mélanges Jean Perrot, Contribution a` l’histoire de l’Iran. Paris: Editions Recherche sur les Civilisations. . 1990e. “Problèmes d’eau et d’irrigation au royaume de Mari: L’apport des textes anciens.” Pp. 101–42 in Bernard Geyer (ed.), Techniques et pratiques hydroagricoles traditionelles en domaine irrigué. Paris: Paul Geuthner. . 1991. “Précurseurs syriens aux protocoles néo-assyriens: Considérations sur la vie politique aux Bords-de-l’Euphrate.” Pp. 13–71 in Mélanges Garelli. . 1992a. “Espionnage et guerre froide: La fin de Mari.” FM I: 39–52. . 1992b. “Unité et diversités au Proche-Orient a` l’époque amorrite.” Pp. 97– 128 in Dominique Charpin and Francis Joannès (eds.), La circulation des biens, des ´ personnes et des idées dans le Proche-Orient ancien. CRRAI 38. Paris: Editions Recherche sur les Civilisations. . 1993a. “Mythologèmes d’époque amorrite.” N.A.B.U., no. 114, pp. 96–7. . 1993b. “Le mythologème du combat entre le Dieu de l’Orage et la Mer en Mésopotamie.” M.A.R.I. 7: 41–61. . 1994. “Administrateurs de Qat.t.unân.” FM II: 83–114. . 1997, 1998, 2000. Documents e´pistolaires du palais de Mari, Tomes I, II, III. LAPO 16–18. Paris: du Cerf. . 1999–2000. “Apologue sur des mauvaises herbes et un coquin.” AuOr 17–18: 191–6. . 2002. Le culte d’Addu d’Alep et l’affaire d’Alahtum. FM VII. . forthcoming. “Les BenSimalites dans la documentation de Mari.” CRRAI 46. Amurru 3. Durand, Jean-Marie, and Michaël Guichard. 1997. “Les rituels de Mari.” FM III: 19–78. Earle, Timothy. 1987. “Chiefdoms in Archaeological and Ethnohistorical perspective.” Annual Review of Anthropology 16: 279–308. . 1991. “The Evolution of Chiefdoms.” Pp. 1–15 in Earle (ed.), Chiefdoms: Power, Economy, and Ideology. Cambridge: Cambridge University Press. . 1997. How Chiefs Come to Power: Political Economy in Prehistory. Stanford, Calif.: Stanford University Press. Edzard, Dietz Otto. 1987–90. “Martu (Mardu). B. Bevölkerungsgruppe.” RlA 7: 438– 40. Eidem, Jesper. 1987/88. “The Tell Leilan Tablets 1987 – A Preliminary Report.” AAAS 38/39, p. 118. . 1994. “Raiders of the Lost Treasure of Sams¯ı-Addu.” FM II: 201–8. . 1998–2001. “Nagar.” RlA 9: 75–7. . 2000. “Northern Jezira in the 18th Century BC: Aspects of Geopolitical Patterns.” Pp. 255–64 in Rouault and Wäfler (eds.), Subartu VII. Eidem, Jesper, and F. Hoejlund. 1997. “Assyria and Dilmun Revisited.” Pp. 25–31 in H. Waetzoldt and H. Hauptmann (eds.), Assyrien im Wandel der Zeiten. CRRAI 39. Heidelberg: Heidelberger Orientverlag. Eidem, Jesper, and Jorgen Laessøe. 2001. The Shemshara Archives 1: The Letters. Royal Danish Academy of Sciences and Letters.

Bibliography

331

Englund, Robert K. 1998. “Texts from the Late Uruk Period.” Pp. 15–233 in Pascal Attinger and Markus Wäfler (eds.), Mesopotamien: Späturuk-Zeit und Frühdynastische Zeit. Göttingen: Vandenhoeck und Ruprecht. Evans, Geoffrey. 1958 “Ancient Mesopotamian Assemblies.” JAOS 78: 1–11. Evans-Pritchard, E. E. 1940. The Nuer. Oxford: Clarendon. Fales, Frederick Mario. 1990. “The Rural Landscape of the Neo-Assyrian Empire: A Survey.” State Archives of Assyria Bulletin 4: 81–142. Falkenstein, Adam. 1963. “Zu den Inschriftenfunden der Grabung in Uruk-Warka 1960–1961,” BaM 2: 1–82. . 1966. “Zu ‘Gilgameˇs und Agga.’ ” AfO 21: 47–50. Feinman, Gary M. 1998. “Scale and Social Organization: Perspectives on the Archaic State.” Pp. 95–133 in Feinman and Joyce Marcus (eds.), Archaic States. Santa Fe: School of American Research. Finet, André. 1996. “Adalˇsenni, roi de Burundum.” RA 60: 17–28. , ed. 1973. La voix de l’opposition en Mésopotamie. Brussels: Institut des Hautes ´ Etudes Belgique. Finkbeiner, Uwe. 1999–2000. “Emar and Balis 1996–1998. Preliminary Report of the Joint Syrian-German Excavations with the Cooperation of Princeton University.” Berytus 44: 5–34. Finkbeiner, Uwe, et al. 2001. “Emar 1999 – Bericht u¨ ber die 3. Kampagne der syrischdeutschen Ausgrabungen.” BaM 32: 42–120. . 2002. “Emar 2001: Bericht uber ¨ die 4. Kampagne der syrisch-deutschen Ausgrabungen.” BaM 33: 2–41. Finkelstein, J. J. 1966. “The Genealogy of the Hammurapi Dynasty.” JCS 20: 95–118. Flannery, Kent V. 1998. “The Ground Plans of Archaic States.” Pp. 15–57 in Gary M. Feinman and Joyce Marcus (eds.), Archaic States. Santa Fe: School of American Research. Fleming, Daniel E. 1992a. “A Limited Kingship: Late Bronze Emar in Ancient Syria.” UF 24: 59–71. . 1992b. The Installation of Baal’s High Priestess: A Window on Ancient Syrian Religion. Atlanta: Scholars Press. . 1998. “Mari and the Possibilities of Biblical Memory.” RA 92: 41–78. . 1999. “Recent Work on Mari.” RA 93: 157–74. . 2000. Time at Emar: The Cultic Calendar and the Rituals from the Diviner’s Archive. Winona Lake, Ind.: Eisenbrauns. . 2002. “Schloen’s Patrimonial Pyramid: Explaining Bronze Age Society.” BASOR 328. . forthcoming. “The Simalite gayum and the Yaminite li mum in the Mari Archives.” CRRAI 46. Amurru 3. ´ Fortin, Michel. 2000. “Economie et société dans la moyenne vallée du Khabour durant la période de Ninive 5.” Pp. 111–36 in Rouault and Wäfler (eds.), Subartu VII. Fox, Richard G. 1977. Urban Anthropology: Cities in Their Cultural Settings. Englewood Cliffs, N.J.: Prentice-Hall. Frangipane, Marcella. 1998. “Changes in Upper Mesopotamian/Anatolian Relations at the Beginning of 3rd Millennium B.C.” Pp. 195–218 in Subartu IV, vol. 1. Frayne, Douglas. 1990. Old Babylonian Period (2003–1595 BC). RIME 4. Toronto: University of Toronto.

332

Bibliography

. 1993. Sargonic and Gutian Periods (2334–2113 BC). RIME 2. Toronto: University of Toronto. Fried, Morton H. 1967. The Evolution of Political Society. New York: Random House. . 1975. The Notion of the Tribe. Menlo Park, Calif.: Cummings. Fukuyama, Francis. 1992. The End of History and the Last Man. New York: Free Press. Gasche, H., J. A. Armstrong, S. W. Cole, and V. G. Gurzadyan. 1998. Dating the Fall of Babylon: A Reappraisal of Second-Millennium Chronology. University of Ghent and the Oriental Institute of the University of Chicago. Gelb, I. J. 1986. “Ebla and Lagash: Environmental Contrast.” Pp. 157–67 in Harvey Weiss (ed.), The Origins of Cities in Dry-Farming Syria and Mesopotamia in the Third Millennium B.C. Guilford, Conn.: Four Quarters. Geyer, Bernard. 1985. “Géomorphologie et occupation du sol de la moyenne vallée de l’Euphrate dans la région de Mari.” M.A.R.I. 4: 27–39. . 1990. “Une ville aujourd’hui engloutie: Emar. Contribution géomorphologique a` la localisation de la cité.” M.A.R.I. 6: 107–20. Giddens, Anthony. 1981. A Contemporary Critique of Historical Materialism, vol. 1: Power, Property and the State. Berkeley: University of California Press. . 1984. The Constitution of Society: Outline of the Theory of Structuration. Berkeley: University of California Press. ´ Glassner, Jean-Jacques. 2000. “Les petits Etats mésopotamiens a` la fin du 4e et au cours du 3e millénaire.” Pp. 35–53 in Mogens Herman Hansen (ed.), A Comparative Study of Thirty City-State Cultures. Copenhagen: C. A. Reitzels. Goetze, Albrecht. 1947. Review of H. Bozkurt et al., IBoT, 1944. JCS 1: 87–92. Grayson, A. Kirk. 1987. Assyrian Rulers of the Third and Second Millennia BC (to 1115 BC). RIMA 1. Toronto: University of Toronto. Groneberg, Brigitte. 1990. “La culture matérielle a` Mari, II: ‘Der n¯ubalum und seine Objekte.’ ” M.A.R.I. 6: 161–80. Guichard, Michaël. 1994. “Au pays de la dame de Nagar.” FM II: 235–72. . 1997. “Le sel a` Mari (III): Les lieux du sel.” FM III: 167–200. ˇ . 2002. “Le Subartum occidental a` l’avènement de Zimrˆı-Lˆım.” FM VI: 119– 68. Guillot, Isabelle. 1997. “Les gouverneurs de Qat.t.unân: Nouveaux textes.” FM III: 271–90. Güterbock, Hans G. 1954.“Authority and Law in the Hittite Kingdom.” JAOS Supplement 17/3: 16–24. Hallo, William W. 1971. “Antediluvian Cities.” JCS 23: 57–67. Hallo, William W., and William Kelly Simpson. 1998. The Ancient Near East: A History, 2nd ed. Fort Worth: Harcourt Brace. Hansen, Mogens Herman. 1991. The Athenian Democracy in the Age of Demosthenes. Oxford: Basil Blackwell. . 1997. “The Copenhagen Inventory of Poleis and the Lex Hafniensis de Civitate.” Pp. 9–23 in Lynette G. Mitchell and P. J. Rhodes (eds.), The Development of the Polis in Archaic Greece. London: Routledge. . 2000. “Introduction: The Concepts of City-State and City-State Cultures.” Pp. 11–34 in Hansen (ed.), A Comparative Study of Thirty City-State Cultures. Copenhagen: C. A. Reitzels. Heimpel, Wolfgang. 1974. “Sumerische und addadische Personennamen in Sumer und Akkad.” AfO 25: 171–4.

Bibliography

333

. 1992. “Herrentum und Königtum im vor- und frühgeschichtlichen Alten Orient.” ZA 82: 4–21. Helms, Svend W. 1987. “Jawa, Tell Um Hammad and the EB I/Late Chalcolithic Landscape.” Levant 19: 49–81. Heltzer, Michael. 1981. The Suteans. Naples: Istituto Universitario Orientale. Hoftijzer, J., and K. Jongeling. 1995. Dictionary of the Northwest Semitic Inscriptions. Leiden: E. J. Brill. Hole, Frank. 1991. “Middle Khabur Settlement and Agriculture in the Ninevite 5 Period.” BCSMS 21: 17–29. . 2000. “The Prehistory of the Khabur.” Pp. 17–27 in Rouault and Wäfler (eds.), Subartu VII. Huehnergard, John. 1987. Ugaritic Vocabulary in Syllabic Transcription. Atlanta: Scholars Press. Jacobsen, Thorkild. 1943. “Primitive Democracy in Ancient Mesopotamia,” JNES 2: 159–72. . 1957. “Early Political Development in Mesopotamia,” ZA 52: 91–140. Jeffery, L. H. 1976. Archaic Greece: The City-States c. 700–500 B.C. New York: St. Martins. Joannès, Francis. 1985. “Nouveaux mémorandums.” Pp. 97–113 in Jean-Marie Durand and Jean-Robert Kupper (eds.), Miscellanea Babylonica: mélanges offerts a` ´ Maurice Birot. Paris: Editions Recherche sur les Civilisations. . 1991. “Une mission secrète a` Eˇsnunna.” Pp. 185–93 in Mélanges Garelli. . 1996. “Routes et voies de communication dans les archives de Mari.” Amurru 1: 323–61. . 1997. “Palmyre et les routes du désert au début du IIe millénaire av. J.-C.” M.A.R.I. 8: 393–408. Joannès, Francis, ed. 2000. Rendre la justice en Mésopotamie. Saint Denis: Presses Universitaires de Vincennes. Jones, G. I. 1971. “Councils among the Central Ibo.” Pp. 63–79 in Audrey Richards and Adam Kuper (eds.), Councils in Action. Cambridge: Cambridge University Press. Kamp, Kathryn A., and Norman Yoffee. 1980. “Ethnicity in Ancient Western Asia during the Early Second Millennium B.C.: Archaeological Assessments and Ethnoarchaeological Prospectives.” BASOR 237: 85–104. Katz, Dina. 1987. “Gilgamesh and Akka: Was Uruk Ruled by Two Assemblies?” RA 81: 105–14. Khazanov, A. M. 1984. Nomads and the Outside World. Cambridge: Cambridge University Press. Khoury, Philip S., and Joseph Kostiner. 1990. “Introduction: Tribes and the Complexities of State Formation in the Middle East.” Pp. 1–22 in Khoury and Kostiner (eds.), Tribes and State Formation in the Middle East. Berkeley: University of California Press. Klein, Jakob. 1998–2001. “Nippur.A. I.Sum. Nibru, Akk. Nippuru.” RlA 9: 532–9. Klengel, Horst. 1960. “Zu den ˇs¯ıb¯utum in altbabylonischer Zeit.” Or 29: 357–75. ¨ ´ ˇ . 1965. “Die Rolle der ‘Altesten’ (LU.ME Sˇ SU.GI) im Kleinasien der Hethiterzeit.” ZA 57: 223–36. Kopcke, Gunter. ¨ 2002. “Mycenaean Kingship, or Why Mycenaean Greece is Different.” October Melammu conference. von Koppen, Frans. 1997. “L’Expédition a` Tilmun et la révolte des bédouins.” M.A.R.I. 8: 417–29.

334

Bibliography

Kouchoukos, Nicholas. 1998. “Landscape and Social Change in Late Prehistoric Mesopotamia.” Ph.D. dissertation, Yale University. Kraus, F. R. 1958. Ein Edikt des Königs Ammi-s.aduqa von Babylon. Leiden: E. J. Brill. . 1970. Sumerer und Akkader: Ein Problem der altmesopotamischen Geschichte. Amsterdam: North Holland. . 1973. Vom Mesopotamischen Menschen der altbabylonische Zeit und seiner Welt. Amsterdam: Netherlands Royal Academy. Krebernik, Manfred. 1998. “Die Texte aux F¯ara und Tell Ab¯u S. al¯ab¯ıh.” Pp. 235– ˘ 427 in Pascal Attinger and Markus Wäfler (eds.), Mesopotamien: Späturuk-Zeit und Frühdynastische Zeit. Göttingen: Vandenhoeck und Ruprecht. Kuhrt, Amélie. 1995. The Ancient Near East c. 3000–330 BC. London: Routledge. Kuper, Adam. 1971. “Introduction: Council Structure and Decision-Making.” Pp. 13– 28 in Audrey Richards and Kuper (eds.), Councils in Action. Cambridge: Cambridge University Press. . 1982. “Lineage Theory: A Critical Retrospect.” Annual Review of Anthropology 11: 71–95. Kupper, Jean-Robert. 1957. Les Nomades en Mésopotamie au temps des rois de Mari. Paris: Les Belles Lettres. . 1990. “Une lettre du général Yassi-Dagan.” M.A.R.I. 6: 337–47. . 2002. “Dans les jardins de Carkémish . . .” FM VI: 195–200. Lacambre, Denis. 1997a. “La bataille de Hirˆıtum.” M.A.R.I. 8: 431–54. . 1997b. “La gestion du bronze dans le palais de Mari: Collations et joints a` ARMT XXII.” FM III: 91–123. Lachenbacher, Sylvie. 1987. “madinâtum.” N.A.B.U., no. 81, p. 43. Lafont, Bertrand. 1984. “Le roi de Mari et les prophètes du dieu Adad.” RA 78: 7–18. . 1985. “Le .saˆ bum du roi de Mari au temps de Yasmah-Addu.” Pp. 161–79 in Jean-Marie Durand and Jean-Robert Kupper (eds.), Miscellanea Babylonica: mélanges ´ offerts a` Maurice Birot. Paris: Editions Recherche sur les Civilisations. . 1992a. “Messagers et ambassadeurs dans les archives de Mari.” Pp. 167–83 in Dominique Charpin and Francis Joannès (eds.), La circulation des biens, des personnes ´ et des idées dans le Proche-Orient ancien. CRRAI 38. Paris: Editions Recherche sur les Civilisations. . 1992b. “Nuit dramatique a` Mari.” FM I: 93–101. . 1994. “L’admonestation des Anciens de Kurdâ a` leur roi.” FM II: 209–20. . 1997a. “Corps social et pouvoir politique dans les sociétés de l’ancienne Mésopotamie.” Méditerranées 12: 25–48. . 1997b. “Le fonctionnement de la poste et le métier de facteur d’après les textes de Mari.” Pp. 315–34 in G. D. Young et al. (eds.), Crossing Boundaries and Linking Horizons, Studies in Honor of M. C. Astour. Bethesda, M.D.: CDL Press. . 1999. “Le Proche-Orient a` l’époque des rois de Mari: Un monde sans frontières?” Pp. 49–55 in L. Milano et al. (eds.), Landscapes: Territories, Frontiers and Horizons in the Ancient Near East. CRRAI 44. Padova: Sargon srl. . 2000a. “Cheval, aˆne, onagre et mule dans la haute histoire mésopotamienne: Quelques données nouvelles.” Topoi Suppl. 2: 207–21. . 2000b. “Irrigation Agriculture in Mari.” Pp. 129–45 in R. M. Jas (ed.), Rainfall and Agriculture in Northern Mesopotamia. Te Istanbul: Nederlands HistorischArchaeologisch Instituut.

Bibliography

335

. 2001a. “Relations internationales, alliances et diplomatie au temps des royaumes amorrites.” Amurru 2: 213–328. . 2001b. “Ur III.” Pp. 878–82 in DCM. Lafont, Sophie. 2000. “Considérations sur la pratique judiciaire en Mésopotamie.” Pp. 15–34 in Francis Joannès (ed.), Rendre la justice en Mésopotamie. Saint Denis: Presses Universitaires de Vincennes. . 2002. “Un cas d’exécution sommaire a` Tuttul.” FM VI: 89–101. Lambert, Wilfrid G. 1985. “The Pantheon of Mari.” M.A.R.I. 4: 525–39. Lapidus, Ira M. 1990. “Tribes and State Formation in Islamic History.” Pp. 25–47 in Philip S. Khoury and Joseph Kostiner (eds.), Tribes and State Formation in the Middle East. Berkeley: University of California Press. Larsen, Mogens Trölle. 1976. The Old Assyrian City-State and Its Colonies. Copenhagen: Akademisk. . 2000. “The City-States of the Early Neo-Babylonian Period.” Pp. 117–27 in Mogens Trolle Hansen (ed.), A Comparative Study of Thirty City-State Cultures. Copenhagen: C. A. Reitzels. Leach, E. R. 1968. Pul Eliya: A Village in Ceylon. A Study of Land Tenure and Kinship. Cambridge: Cambridge University Press. Lebeau, Marc. 1990. “Esquisse d’une histoire de la Haute Mésopotamie au début de l’Age du Bronze.” ZA 80: 241–96. . 2000. “Les voies de communication en Haute Mésopotamie au IIIe millénaire avant notre e` re.” Pp. 157–62 in Subartu VII. Lees, Susan H., and Daniel G. Bates. 1974. “The Origins of Specialized Nomadic Pastoralism: A Systemic Model.” American Antiquity 39: 187–93. Lieberman, Stephen J. 1977. The Sumerian Loanwords in Old-Babylonian Akkadian. Missoula, Mont.: Scholars Press. . 1992. “Nippur: City of Decisions.” Pp. 127–36 in Maria deJong Ellis (ed.), Nippur at the Centennial. CRRAI 35. Philadelphia: University Museum. van Liere, W. J. 1963. “Capitals and Citadels of Bronze-Iron Age Syria in Their Relationship to Land and Water.” AAAS 13: 109–22. van Liere, W. J., and J. Lauffray. 1954–5. “Nouvelle prospection archéologique dans la Haute Jézireh syrienne.” AAAS 4/5: 129–48. ´ Limet, Henri. 1978. “Etude sémantique de ma.da, kur, kalam.” RA 72: 1–12. Lion, Brigitte. 2001. “Les gouverneurs provinciaux du royaume de Mari a` l’époque de Zimrˆı-Lˆım.” Amurru 2: 141–209. Liverani, Mario. 1974. “La royauté syrienne de l’âge du Bronze Récent.” Pp. 329–56 in Paul Garelli (ed.), Le palais et la royauté. CRRAI 19. Paris: Paul Geuthner. . 1975. “Communautés de village et palais royal dans la Syrie du IIe` me millénaire.” JESHO 18: 146–64. . 1983. “Communautés rurales dans la Syrie du IIe millénaire a.c.” Pp. 147–85 in Les communautés rurales. Paris: Dessain et Tolra. . 1997. “ ‘Half-Nomads’ on the Euphrates and the Concept of Dimorphic Society.” AoF 24: 44–8. Lloyd, G. E. R. 1987. The Revolutions of Wisdom: Studies in the Claims and Practice of Ancient Greek Science. Berkeley: University of California Press. Luciani, Marta. 1999. “Zur Lage Terqas in schriftlichen Quellen.” ZA 89: 1–23. Luke, J. T. 1965. “Pastoralism and Politics in the Mari Period”. Ph.D. dissertation, University of Michigan.

336

Bibliography

Lupton, Alan. 1996. Stability and Change: Socio-Political Development in North Mesopotamia and South-east Anatolia, 4000–2700 B.C. Oxford: Tempus Reparatum. Lyonnet, Bertille. 1998. “Le peuplement de la Djéziré occidentale au début du 3e millénaire, villes circulaires et pastoralisme: Questions et hypothèses.” Pp. 179–93 in Subartu IV, vol. 1. . 2000. “Méthodes et résultats préliminaires d’une prospection archéologique dans la partie occidentale du Haut-Khabur, depuis le Néolithique jusqu’à la fin du IIe millénaire av.n.è.” Pp. 241–53 in Rouault and Wäfler (eds.), Subartu VII. . forthcoming. “Le nomadisme et l’archéologie: Problèmes d’identification. Le cas de la partie occidentale de la Djéziré aux 3e et début du 2e millénaire avant notre e` re.” CRRAI 46 (Amurru 3). Malamat, Abraham. 1967. “Aspects of Tribal Societies in Mari and Israel.” Pp. 129– 38 in André Finet (ed.), Actes de la 15e Rencontre Assyriologique Internationale. Liège: Les Belles Lettres. . 1979. “Ummatum in Old Babylonian Texts and Its Ugaritic and Biblical Counterparts.” UF 11: 527–36. Mann, Michael. 1986. The Sources of Social Power, vol. 1: A History of Power from the Beginning to A.D. 1760. Cambridge: Cambridge University Press. Marello, Pierre. 1992. “Vie nomade.” FM I: 115–25. Margueron, Jean-Claude. 1990. “L’aménagement de la région de Mari: Quelques considérations historiques.” Pp. 171–91 in Bernard Geyer (ed.), Techniques et pratiques hydro-agricoles traditionelles en domaine irrigué. Paris: Paul Geuthner. . 1994. “Mari au IIe millénaire.” Pp. 313–20 in H. Gasche et al. (eds.), Cinquante-deux reflexions sur le Proche-Orient ancien, offertes en hommage a` Léon de Meyer. Leuven: Peeters. ¨ Tunca and . 1996a. “Mari a` l’époque des Shakkanakku.” Pp. 95–103 in O. D. Deheselle (eds.), Tablettes et images aux pays de Sumer et d’Akkad, Mélanges offerts a` Monsieur H. Limet. Liège: Université de Liège. . 1996b. “Mari, reflet du monde syro-mésopotamien au IIIe millénaire.” Akkadica 98: 11–30. . 1997. “Mari.” Pp. 3.413–17 in OEANE. . 2000. “Mari et le Khabur.” Pp. 99–110 in Rouault and Wäfler (eds.), Subartu VII. Matthews, Victor H. 1978. Pastoral Nomadism in the Mari Kingdom (ca. 1830–1760 B.C.). Cambridge, Mass.: American Schools of Oriental Research. Matthiae, Paulo. 1997. “Tell Mardikh, 1977–1996: Vingt ans de fouilles et de découvertes: La renaissance d’Ebla amorrhéenne.” Akkadica 101: 1–29. McClellan, Thomas L. 1995. “Irrigation in Dry-Farming Syria: Agricultural Intensification in the Bronze Age.” Pp. 51–76 in Farouk Ismail (ed.), Proceedings of the International Symposium on Syria and the Ancient Near East 3000–300 B.C. Aleppo: University of Aleppo. McClellan, Thomas L., Rodger Grayson, and Cliff Ogleby. 2000. “Bronze Age Water Harvesting in North Syria.” Pp. 137–55 in Subartu VII. McCorriston, Joy. 1995. “Preliminary Archaeobotanical Analysis in the Middle Habur Valley, Syria and Studies of Socioeconomic Change in the Early Third Millennium BC.” BCSMS 29: 33–46. . 1997. “The Fiber Revolution: Textile Intensification, Alienation and Social Stratification in Ancient Mesopotamia.” Current Anthropology 38: 517–49.

Bibliography

337

McIntosh, Roderick. 1999. “Western Representations of Urbanism and Invisible African Towns.” Pp. 56–65 in Susan Keech McIntosh (ed.), Beyond Chiefdoms: Pathways to Complexity in Africa. Cambridge: Cambridge University Press. McIntosh, Susan Keech, 1999a. ed. Beyond Chiefdoms: Pathways to Complexity in Africa. Cambridge: Cambridge University Press. . 1999b. “Pathways to Complexity: An African Perspective.” Pp. 1–30 in McIntosh (ed.), Beyond Chiefdoms. . 1999c. “Modeling Political Organization in Large-Scale Settlement Clusters: A Case Study from the Inland Niger Delta.” Pp. 66–79 in McIntosh (ed.), Beyond Chiefdoms. Michalowski, Piotr. 1986. “Mental Maps and Ideology: Reflections on Subartu.” Pp. 129–56 in Harvey Weiss (ed.), The Origins of Cities in Dry-Farming Syria and Mesopotamia in the Third Millennium B.C. Guilford, Conn.: Four Quarters. . 1995. “The Men from Mari.” Pp. 181–8 in K. van Lerberghe and A. Schoors (eds.), Immigration and Emigration within the Ancient Near East: Festschrift E. Lipinski. ´ Leuven: Peeters. . forthcoming a. “The Ideological Foundations of the Ur III State.”. . forthcoming b. “Literary Works from the Court of King Ishbi-Erra of Isin.” Festschrift for Jacob Klein. Bar Ilan University Press. Michel, Cécile. 1992. “Les ‘diamants’ du roi de Mari.” FM I: 127–36. . 2000. “Les Litiges commerciaux paléo-assyriens.” Pp. 113–39 in Francis Joannès (ed.), Rendre la justice en Mésopotamie. Saint Denis: Presses Universitaires de Vincennes. Millet-Albà, Adélina. forthcoming. “La localisation des terroirs benjaminites du royaume de Mari.” CRRAI 46 (Amurru 3). de Montmollin, Olivier. 1989. The Archaeology of Political Structure: Settlement Analysis in a Classic Maya Polity. Cambridge: Cambridge University Press. Moran, William L. 1963. “The Ancient Near Eastern Background of the Love of God in Deuteronomy.” CBQ 25 : 77–87. . 1992. The Amarna Letters. Baltimore: Johns Hopkins University Press. Morris, Ian. 1987. Burial and Ancient Society: The Rise of the Greek City-State. Cambridge: Cambridge University Press. . 1991. “The Early Polis as City and State.” Pp. 25–58 in John Rich and Andrew Wallace-Hadrill (eds.), City and Country in the Ancient World. London: Routledge. . 1996. “The Strong Principle of Equality and the Archaic Origins of Greek Democracy.” Pp. 19–48 in Josiah Ober and Charles Hedrick (eds.), D¯emokratia: A Conversation on Democracies, Ancient and Modern. Princeton, N.J.: Princeton University Press. . 1997. “An Archaeology of Equalities? The Greek City-States.” Pp. 91– 105 in Deborah L. Nichols and Thomas H. Charlton (eds.), The Archaeology of City-States: Cross-Cultural Approaches. Washington, D.C.: Smithsonian Institution Press. Nakata, Ichiro. 1989. “A Further Look at the Institution of sug¯ag¯utum in Mari.” JANES 19: 113–18. Oates, David, and Joan Oates. 2000. “Recent Excavations at Tell Brak.” Pp. 91–7 in Subartu VII. Oates, Joan. 1983. “Urban Trends in Prehistoric Mesopotamia.” Pp. 81–92 in La ville dans le Proche-Orient ancien. Leuven: Peeters.

338

Bibliography

Ober, Josiah. 1989. Mass and Elite in Democratic Athens. Princeton, N.J.: Princeton University Press. Oppenheim, A. Leo. 1936. “Zur keilschriftliche Omenliteratur.” Or 5: 199–228. Orthmann, Winfried. 1986. “The Origins of Tell Chuera.” Pp. 61–70 in Harvey Weiss (ed.), The Origins of Cities in Dry-Farming Syria and Mesopotamia in the Third Millennium B.C. Guilford, Conn.: Four Quarters. Parrot, André. 1936. “Les fouilles de Mari: Deuxième campagne (hiver 1934–35).” Syria 17: 1–31. . 1950. “Les tablettes de Mari et l’Ancien Testament.” RHPR 30: 1–11. Peltenburg, Edgar. 2000. “From Nucleation to Dispersal: Late Third Millennium BC Settlement Pattern Transformations in the Near East and Aegean.” Pp. 183–206 in Rouault and Wäfler (eds.), Subartu VII. Peters, Emrys L. 1967. “Some Structural Aspects of the Feud among the CamelHerding Bedouin of Cyrenaica.” Africa 37: 261–82. Pettinato, Giovanni. 1991. Ebla: A New Look at History. Baltimore: Johns Hopkins University Press. Pettinato, Giovanni, and Hartmut Waetzoldt. 1985. “Dagan in Ebla und Mesopotamien nach den Texten aus dem 3. Jahrtausend.” Or 54: 234–56. Pinnock, Frances. 2001. “The Urban Landscape of Old Syrian Ebla.” JCS 53: 13–33. Podany, Amanda. 2002. The Land of Hana: Kings, Chronology, and Scribal Tradition. Bethesda, Md.: CDL Press. Porter, Anne. 2000. “Mortality, Monuments and Mobility: Ancestor Traditions and the Transcendance of Space.” Ph.D. dissertation, University of Chicago. . 2002. “The Dynamics of Death: Ancestors, Pastoralism, and the Origins of a Third-Millennium City in Syria.” BASOR 325: 1–36. . forthcoming. “The Urban Nomad: Countering the Old Clichés.” CRRAI 46 (Amurru 3). Possehl, Gregory L. 1998. “Sociocultural Complexity without the State: The Indus Civilization.” Pp. 261–91 in Gary M. Feinman and Joyce Marcus (eds.), Archaic States. Santa Fe, N.M.: School of American Research. Postgate, J. N. 1992. Early Mesopotamia: Society and Economy at the Dawn of History. London: Routledge. . 1994. “In Search of the First Empires.” BASOR 293: 1–13. Prechel, Doris, and Thomas Richter. 2001. “Abrakadabra oder Althurritisch: Betrachtungen zu einigen altbabylonischen Beschwörungstexten.” Pp. 333–71 in Richter et al. (eds.), Kulturgeschichten: Altorientalistische Studien für Volkert Haas zum 65. Geburtstag. Saarbrucken: ¨ Saarbrucker ¨ Druckerei und Verlag. Raaflaub, Kurt A. 1996. “Equalities and Inequalities in Athenian Democracy.” Pp. 139–74 in Josiah Ober and Charles Hedrick (eds.), D¯emokratia: A Conversation on Democracies, Ancient and Modern. Princeton, N.J.: Princeton University Press. . 2000. “Poets, Lawgivers, and the Beginnings of Political Reflection in Archaic Greece.” Pp. 23–59 in Christopher Rowe and Malcolm Schofield (eds.), The Cambridge History of Greek and Roman Political Thought. Cambridge: Cambridge University Press. Rappaport, Roy A. 1968. Pigs for the Ancestors: Ritual in the Ecology of a New Guinea People. New Haven: Yale University Press. Reviv, Hanoch. 1989. The Elders in Ancient Israel. Jerusalem: Magnes.

Bibliography

339

Richards, Audrey. 1971. “Introduction: The Nature of the Problem.” Pp. 1–12 in Richards and Adam Kuper (eds.), Councils in Action. Cambridge: Cambridge University Press. Richards, Audrey, and Adam Kuper, eds. 1971. Councils in Action. Cambridge: Cambridge University Press. Roaf, Michael. 1998–2001. “Ninive-5-Kultur (Ninevite 5).” RlA 9: 434–9. Römer, Willem H. Ph. 1980. Das sumerische Kurzepos “Bilgameˇs und Akka.” AOAT 209. Neukirchen-Vluyn: Neukirchener. Rouault, Olivier. 1993. “Tell Ashara – Terqa.” Pp. 185–90 in Rouault and Maria Grazia Masetti-Rouault (eds.), L’Eufrate e il tempo: Civiltà del medio Eufrate e della Gezira siriana. Milan: Electa. . 2000. “Quelques remarques sur la société de Terqa.” Pp. 265–77 in Subartu VII. Rouault, Olivier, and Markus Wäfler, eds. 2000. La Djéziré et l’Euphrate syrien de la protohistoire a` la fin du II e millénaire av. J.-C. Subartu VII. Turnhout: Brepols. Rowton, Michael B. 1967. “The Physical Environment and the Problem of Nomads.” Pp. 109–21 in André Finet (ed.), Actes de la 15e Rencontre Assyriologique Internationale. Liège: Les Belles Lettres. . 1973a. “Autonomy and Nomadism in Western Asia.” Or 42: 247–58. . 1973b. “Urban Autonomy in a Nomadic Environment.” JNES 32: 201–15. . 1974. “Enclosed Nomadism.” JESHO 17: 1–30. . 1976a. “Dimorphic Structure and Topology.” OrAn 15: 17–31. . 1976b. “Dimorphic Structure and the Tribal Elite.” Studia Instituti Anthropos 28: 219–57. . 1977. “Dimorphic Structure and the Parasocial Element.” JNES 36: 181–98. Rubio, Gonzalo. 1999. “On the Alleged Pre-Sumerian Substratum.” JCS 51: 1–16. Safren, J. 1982. “Merhûm and merhûtum in Mari.” Or 51: 1–29. ˘ Tribesmen.˘ Englewood Cliffs, N.J.: Prentice-Hall, 1968. Sahlins, Marshall. 1968. Sallaberger, Walter. 1997. “Nippur als religiöses Zentrum in historischen Wandel.” Pp. 147–68 in Gernot Wilhelm (ed.), Die Orientalische Stadt: Kontinuität, Wandel, Bruch. Saarbrucken: ¨ Saarbrucker ¨ Druckerei und Verlag. Salzman, Philip Carl. 1978. “Ideology and Change in Middle Eastern Tribal Societies.” Man 13: 618–37. . 1980. “Introduction: Processes of Sedentarization as Adaptation and Response.” Pp. 1–20 in Salzman (ed.), When Nomads Settle: Processes of Sedentarization as Adaptation and Response. New York: Praeger. . 2000. Black Tents of Baluchistan. Washington, D.C.: Smithsonian Institution Press. Sasson, Jack M. 1998a. “About ‘Mari and the Bible.’ ” RA 92: 97–123. . 1998b. “The King and I: A Mari King in Changing Perceptions.” JAOS 118: 453–70. Schlechter, E. 1968. “Gouvernés et gouvernants en Mésopotamie depuis les origines jusqu’à la fin de la Ie` re dynastie de Babylone.” Recueils de la Société Jean Bodin 23: 72–86. . 1970. “Les assemblées en Mésopotamie ancienne.” Pp. 3–12 in Liber Memoralis Georges de Lagarde. Louvain: Nauwelaerts. Schloen, J. David. 2001. The House of the Father as Fact and Symbol: Patrimonialism in Ugarit and the Ancient Near East. Winona Lake, Ind.: Eisenbrauns.

340

Bibliography

Schneider, David M. 1984. Critique of the Study of Kinship. Ann Arbor: University of Michigan Press. von Schuler, Einar. 1965. Die Kaskäer. Berlin: Walter de Gruyter. Schwartz, Glenn M. 1986. “Mortuary Evidence and Social Stratification in the Ninevite V Period.” Pp. 45–60 in Harvey Weiss (ed.), The Origins of Cities in DryFarming Syria and Mesopotamia in the Third Millennium B.C. Guilford, Conn.: Four Quarters. . 1994. “Before Ebla: Models of Pre-State Political Organization in Syria and Northern Mesopotamia.” Pp. 153–74 in Gil Stein and Mitchell S. Rothman (eds.), Chiefdoms and Early States in the Near East. Madison, Wis.: Prehistory Press. Schwartz, Glenn M., and Hans H. Curvers. 1992. “Tell al-Raq¯ai 1989 and 1990: Further Investigations at a Small Rural Site of Early Urban Northern Mesopotamia.” AJA 96: 397–419. Selz, Gebhard J. 1992. “Enlil und Nippur nach präsargonischen Quellen.” Pp. 189– 225 in Maria deJong Ellis (ed.), Nippur at the Centennial. CRRAI 35. Philadelphia: University Museum. ¨ . 1998. “Uber Mesopotamische Herrschaftskonzepte. Zu den Ursprüngen mesopotamischer Herrscherideologie im 3. Jahrtausend.” Pp. 281–344 in Manfried Dietrich and Oswald Loretz (eds.), dubsar anta-men. Studien zur Altorientalistik, Festschrift Willem H. Ph. Römer. AOAT 253. Münster: UgaritVerlag. Service, Elman R. 1975. Origins of the State and Civilization: The Process of Cultural Evolution. New York: W. W. Norton. von Soden, Wolfram. 1964. “muˇsk¯enum und die Maw¯al¯ı des frühen Islam.” ZA 56: 133–41. Sommer, F. 1932. Die Ahhijava-Urkunden. Munich: Verlag der Bayerischen Akkadamie der Wissenschaften. Soubeyran, Denis. 1984. “Textes mathématiques de Mari.” RA 78: 19–48. Southall, Aidan. 1988. “Segmentary State in Africa and Asia.” Comparative Studies in Society and History 30: 52–88. van der Spek, R. J. 1987. “The Babylonian City.” Pp. 57–74 in Amélie Kuhrt and Susan Sherwin-White (eds.), Hellenism in the East. Berkeley: University of California Press. Steible, Horst. 1982. Die altsumerischen Bau- und Weihinschriften. Wiesbaden: Franz Steiner. ˇ Steible, Horst, and Fatma Yildiz. 1993. “Kiengi aus der Sicht von Suruppak: Eine fruhdynastische ¨ Regio nach Fara-zeitlichen Urkunden.” Istanbuler Mitteilungen 43: 17–26. Stein, Gil. 1994a. “Introduction Part II. The Organizational Dynamics of Complexity in Greater Mesopotamia.” Pp. 11–22 in Stein and Mitchell S. Rothman (eds.), Chiefdoms and Early States in the Near East: The Organizational Dynamics of Complexity. Madison, Wis.: Prehistory Press. . 1994b. “Segmentary States and Organizational Variation in Early Complex Societies: A Rural Perspective.” Pp. 10–18 in Glenn M. Schwartz and Steven E. Falconer (eds.), Archaeological Views from the Countryside: Village Communities in Early Complex Societies. Washington, D.C.: Smithsonian Institution Press. Steiner, Gerd. 1982. “Der Gegensatz ‘Eigenes Land’: ‘Ausland, Fremdland, Feindland’ in der Vorstellungen des alten Orients.” Pp. 633–4 in Hans-Jörg Nissen and Johannes Renger (eds.), Mesopotamien und Seine Nachbarn. CRRAI 25. Berlin: Dietrich Reimer.

Bibliography

341

. 1988. “Die Bezeichnungen für den Begriff ‘Land’ in den Texten aus Ebla.” Pp. 333–43 in Hartmut Waetzoldt and Harald Hauptmann (eds.), Wirtschaft und Gesellschaft von Ebla. Heidelberg: HSAO 2. Steinkeller, Piotr. 1993. “Early Political Development in Mesopotamia and the Origins of the Sargonic Empire.” Pp. 107–29 in Mario Liverani (ed.), Akkad: The First World Empire. Padova: Sargon srl. . 1995. Review of M. W. Green and H. J. Nissen, Zeichenliste der Archaischen Texte aus Uruk. BiOr 52: 689–713. . 1998. “The Historical Background of Urkesh and the Hurrian Beginnings in Northern Mesopotamia.” Pp. 75–98 in Giorgio Buccellati and Marilyn KellyBuccellati (eds.), Urkesh and the Hurrians: Studies in Honor of Lloyd Cotsen. Bibliotheca Mesopotamica 26. Malibu, Calif.: Undena. . 1999. “On Rulers, Priests and Sacred Marriage: Tracing the Evolution of Early Sumerian Kingship.” Pp. 103–37 in Kazuko Watanabe (ed.), Priests and Officials in the Ancient Near East. Heidelberg: Carl Winter. Stol, Marten. 1976. Studies in Old Babylonian History. Te Istanbul: Nederlands Historisch-Archaeologisch Instituut. . 1997. “Muˇsk¯enu.” RlA 8: 492–3. Stone, Elizabeth C. 1995. “The Development of Cities in Ancient Mesopotamia.” Pp. 235–48 in Jack M. Sasson (ed.), CANE. . 1997. “City-States and Their Centers: The Mesopotamian Example.” Pp. 15– 26 in Deborah L. Nichols and Thomas H. Charlton (eds.), The Archaeology of City-States: Cross-Cultural Approaches. Washington, D.C.: Smithsonian Institution Press. Stone, Elizabeth C., and Paul Zimansky. 1992. “Mashkan-shapir and the Anatomy of an Old Babylonian City.” BA 55/4: 212–18. Strommenger, Eva. 1997. “Bi a/Tuttul.” Pp. 112–13 in Harvey Weiss, “Archaeology in Syria.” AJA 101. Strommenger, Eva, and Kay Kohlmeyer. 2000. Ausgrabungen in Tall Bi a/Tuttul, Bd. III. Die Schichten des 3. Jahrtausends v. Chr. im Zentralhügel E. Saarbrucken: ¨ Saarbrucker ¨ Druckerei und Verlag. Talon, Philippe. 1978. “La Taxe sug¯ag¯utum a` Mari.” RA 73: 143–51. . 1982. “La sug¯ag¯utum a` Mari.” Pp. 54–68 in André Finet (ed.), Les pouvoirs ´ locaux en Mésopotamie. Brussels: Institut des Hautes Etudes Belgique. . 1985. “Quelques réflexions sur les clans hanéens.” Pp. 277–84 in Miscel´ lanea Babylonica: Mélanges offerts a` Maurice Birot. Paris: Editions Recherche sur les Civilisations. . 1986. “Les nomades et le royaume de Mari.” Akkadica 48: 1–9. Tapper, Richard. 1990. “Anthropologists, Historians, and Tribespeople on Tribe and State Formation in the Middle East.” Pp. 48–73 in Philip S. Khoury and Joseph Kostiner (eds.), Tribes and State Formation in the Middle East. Berkeley: University of California Press. Van De Mieroop, Marc. 1999. The Ancient Mesopotamian City. Oxford: Oxford University Press. Vansina, Jan. 1999. “Pathways of Political Development in Equatorial Africa and Neo-Evolutionary Theory.” Pp. 166–72 in Susan Keech McIntosh (ed.), Beyond Chiefdoms: Pathways to Complexity in Africa. Cambridge: Cambridge University Press. Veenhof, Klaas R. 1972. Aspects of Old Assyrian Trade and Its Terminology. Leiden: E. J. Brill.

342

Bibliography

Villard, Pierre. 1986. “Un roi de Mari a` Ugarit.” UF 18: 387–412. . 1987. “Un conflit d’autorités a` propos des eaux du Balih.” M.A.R.I. 5: 591– 96. . 1990. “Documents pour l’histoire du royaume de Haute-Mésopotamie III.” M.A.R.I. 6: 559–84. . 1992. “Parade militaire dans les jardins de Babylone.” FM I: 137–51. . 1993. “La place des années de ‘Kahat’ et d’’Adad d’Alep’ dans la chronologie du règne de Zimri-Lim.” M.A.R.I. 7: 315–28. . 1994. “Nomination d’un scheich.” FM II: 291–7. . 2000. “Les textes judiciaires néo-assyriens.” Pp. 174–200 in Francis Joannès (ed.), Rendre la justice en Mésopotamie. Saint Denis: Presses Universitaires de Vincennes. . 2001. “Les administrateurs de l’époque de Yasmah-Addu.” Amurru 2: 9–140. Walther, Arnold. 1917. Das altbabylonische Gerichtswesen. Leipzig: J. C. Hinrichs. Wasserman, Nathan. 1994. “The Particle assurre/¯e in the Mari Letters.” FM II: 319– 35. Weiss, Harvey. 1986. “Introduction: The Origins of Cities in Dry-Farming Syria and Mesopotamia in the Third Millennium B.C.” Pp. 1–6 in Weiss (ed.), The Origins of Cities in Dry-Farming Syria and Mesopotamia in the Third Millennium B.C. Guilford, Conn.: Four Quarters. . 1988. “The Origins of Tell Leilan and the Conquest of Space in Third Millennium North Mesopotamia.” Pp. 71–108 in Norman Yoffee and George L. Cowgill (eds.), The Collapse of Ancient States and Civilizations. Tuscon: University of Arizona Press. . 1990. “ ‘Civilizing’ the Habur Plains: Mid-Third Millennium State Formation at Tell Leilan.” Pp. 387–407 in Paolo Matthiae, Maurits van Loon, and Harvey Weiss (eds.), Resurrecting the Past: A Joint Tribute to Adnan Bounni. Te Istanbul: Nederlands Historisch-Archaeologisch Instituut. . 2000. “Causality and Chance: Late Third Millennium Collapse in Southwest Asia.” Pp. 207–17 in Rouault and Wäfler (eds.), Subartu VII. Weiss, Harvey, and Marie-Agnès Courty. 1993. “The Genesis and Collapse of the Akkadian Empire: The Accidental Refraction of Historical Law.” Pp. 131–55 in Mario Liverani (ed.), Akkad: The First World Empire. Padova: Sargon srl. Westenholz, Aage. 1999. “The Old Akkadian Period: History and Culture.” Pp. 15– 117 in Pascal Attinger and Markus Wäfler (eds.), Mesopotamien: Akkade-Zeit und Ur III-Zeit. Göttingen: Vandenhoeck und Ruprecht. Whiting, Robert M. 1995. “Amorite Tribes and Nations of Second-Millennium Western Asia.” Pp. 1231–42 in Jack M. Sasson (ed.), CANE. Wilcke, Claus. 1990. “Orthographie, Grammatik und literarische Form. Beobachtungen zu der Vaseninschrift Lugalzagesis (SAKI 152–156).” Pp. 455–504 in I. Tzvi Abusch et al. (eds.), Lingering over Words. Festschrift William L. Moran. Atlanta: Scholars Press. ¨ . 1998. “Zu ‘Gilgameˇs und Akka.’ Uberlegungen zur Zeit von Entstehung und ¨ Niederschrift, wie auch zum Text des Epos mit einem Exkurs zur Uberlieferung ˇ von ‘Sulgi A’ und von ‘Lugalbanda II.’ ” Pp. 457–85 in Manfried Dietrich and Oswald Loretz (eds.), dubsar anta-men: Studien zur Altorientalistik, Festschrift Willem H. Ph. Römer. AOAT 253. Münster: Ugarit-Verlag.

Bibliography

343

Wilhelm, Gernot. 1982. Grundzüge der Geschichte und Kultur der Hurriter. Darmstadt: Wissenschaftliche Buchgesellschaft. Wilkinson, T. J. 1994. “The Structure and Dynamics of Dry-Farming States in Upper Mesopotamia.” Current Anthropology 35: 483–520. Yamada, Masamichi. 1993. “Division of a Field and Ninurta’s Seal: An Aspect of the Hittite Administration in Emar.” UF 25: 453–60. Yoffee, Norman. 1988. “The Collapse of Ancient Mesopotamian States and Civilization.” Pp. 44–68 in Yoffee and George L. Cowgill (eds.), The Collapse of Ancient States and Civilizations. Tuscon: University of Arizona Press. Young, Dwight W., and Victor H. Matthews. 1977. “The raison d’être of the sug¯agum in Mari.” Or 46: 122–6. Zagarell, Allen. 1989. “Pastoralism and the Early State in Greater Mesopotamia.” Pp. 280–302 in C. C. Lamberg-Karlovsky (ed.), Archaeological Thought in America. Cambridge: Cambridge University Press. Zarins, Juris. 1990. “Early Pastoral Nomadism and the Settlement of Lower Mesopotamia.” BASOR 280: 31–65. Zeder, Melinda A. 1994. “Of Kings and Shepherds: Specialized Animal Economy in Ur III Mesopotamia.” Pp. 175–91 in Gil Stein and Mitchell S. Rothman (eds.), Chiefdoms and Early States in the Near East. Madison, Wis.: Prehistory Press. . 1995. “The Archaeobiology of the Khabur Basin.” BCSMS 29: 21–32. Ziegler, Nele. 1999. Le harem de Zimrˆı-Lˆım. FM 4. Paris: SEPOA. . 2002. “Le Royaume d’Ekkalâtum et son horizon géopolitique.” FM VI: 211–74. Ziegler, Nele, and Dominique Charpin. 2001. “Mari (rois).” Pp. 496–501 in DCM.

Subject Index

Abattum, 158 Africa, 179–180 Agade, 17, 161 Ah Purattim (Banks-of-the-Euphrates), ˘administrative entity, 86, 119, 154–155, 161–162, 164–165 Akkadian: and Ah Purattim, 157–158; as ethnic ˘ 41–42, 127, 156–159; indicator, language, 4, 40–42, 94; and Yaminites, 158–159 Alahtum, 173, 189 ˘ alliances, 83, 87 a¯ lum, 21, 108, 116, 171, 172–173 Ammi-s. aduqa, 127 Amnanˆ u, 123, 124 Amorrite: as ethnic indicator, 41–43, 159; language, 40, 103 Amorrites, 14, 39–43; definitions, 39–40, 41 Amurrum, 29, 40 Andarig, 82, 127, 138, 150, 163, 184, 196 anointing, 101 anthropology, cultural/social, 20 Apum, 90, 123, 150 Aˇslakkâ, 126, 163, 173 Asqudum, 189 assembly as puhrum meeting, 199; collected ˘ populace, 189–190, 211, 233–234;

definitions, 203–204; Hana, 79, 87, ˘ 88, 206, 209, 211; and individual leaders, 210–211; military, 206–207; in southern Mesopotamia, 204–206; tribal, 209–210, 211 Aˇsˇsur, 206 Atamrum, 82, 184, 196 Atij (Tell), 36 aw¯ılum, 142–143, 185–187 Baal myth of Ugarit, 61 Babylon, 41, 104, 121, 157, 160, 161, 206 Banat (Tell), 28, 94, 109, 215–216, 220 Bannum, 11, 62, 73, 80, 81–82, 155, 166, 189, 274 basileis, 238, 239 Bunu-Eˇstar, 82, 83, 160 census, 52, 73–74, 88–89, 95, 168, 173 chiefdom, 112–113 Childe, V. Gordon, 109 city and town, definitions, 21, 107–110, 170–171, 235–236 city-state, 17, 18, 19, 112 collective decision making, 104, 171, 174–175, 177–180, 218, 222–223; terminology, 180, 187–188 collective power, 224–225, 236–238 corporate cognitive code, 225–226 corporate/exclusionary politics, 177–180, 225–226, 228 345

346

Subject Index

councils, 175–176; “arena,” 175–176; “elite,” 175–176, 296 Dabiˇs, 93, 97 Dagan, 191, 197, 215, 217–218, 222, 226 democracy: ancestors, 171; Athenian, 16, 238, 240; definitions, 15–16, 235; Greek, 171, 238–241; primitive, 15–16, 177, 235, 237 district governors (ˇsa¯ pit.um), 54, 67–68, 76, 78, 86–87, 144 districts, 58, 66, 76, 84, 86, 105, 133–139; definition, 133; subordinate, 136–138; town and tribal leadership, 138–139 donkeys, 157 Ebla, 118–119, 121, 128, 208, 210, 213–214, 215, 219 Ekallatum, 10, 11, 13, 121, 159, 163 Elam, 8, 126, 163, 274 elders, 68–70, 190–200; actually kings, 193–195; as negotiators, 191; as royal advisers, 195–197; as witnesses, 191; at Urgiˇs, 197–199; definitions, 190–191; fluid category, 190, 194; with sug¯agum title, 192–193; tribal, 199–200 Emar (13th century; = Imar), 61, 179, 208, 213, 226–227 Emutbal, 124, 127, 160 Enlil, 129–130, 222 En¯uma eliˇs, 16, 204 Eˇsnunna, 41, 49, 55, 87, 104, 118, 121, 127, 137, 157, 161, 163, 222, 257 Eˇstar festival, 179 euergesai, 178, 179 evolutionary models, 223–224 gayum, 27, 50–51, 54, 55–58; definition, 43, 45, 57; military, 86, 89; non-Simalite, 58; Simalite use, 50, 54–55, 56–58, 80–81, 86; Yabasu/Aˇsarugayum division, 56–57, 80, 90

Gilgamesh and Agga, 205 Greece (archaic), 238–240 hals.um, see districts ˘ Haman, 153, 154, 155 ˘ Hammi-iˇ stamar, 162 ˘ Hammurabi of Babylon, 6, 11, 40, 103, ˘ 123, 141, 159, 160, 250 Hammurabi of Kurdâ, 82, 160, 211 ˘ hana, 22, 47–50, 85–92; “tent-dwellers,” ˘ 46–47, 89, 148–150; under Samsi-Addu, 90; Yaminite, 91–92 Hana: ˘ as Terqa kingdom, 161, 218; combining pastoralist and tribal, 148; fathers, 153–154; independent, 87–88, 92, 167; military force, 50, 74, 85–87, 92; mobile, 150–152, 156; a non-Simalite entity?, 89; Simalite pastoralists, 85, 89, 148–149, 150–152; a Simalite political unity, 87–89; under Samsi-Addu, 90; in Zimri-Lim’s kingdom, 79–80, 85, 92, 143, 146, 147 Harappan civilization, 113 Harran, 200 ˘ hazzannum, 197, 198 ˘heads (leaders), 138, 186–187, 200–203; as high officials, 202–203; definitions, 200; of households, 201; tribal, 201–202 hibrum, 49, 63, 77, 95–96, 97–100 ˘ Hittite “land,” 131–132 horses, 157 household, 140–141, 186–187, 201, 227–228 Hurrians, 127–128, 198, 212, 216–217 Huˇslâ, 163

˘

Ibal-Addu, 82, 126, 194, 198 Ibal-el, 76, 83, 86, 87, 194–195 Ibal-pi-el of Eˇsnunna, 49, 86, 160, 173 Ibal-pi-el the merhûm, 80–81, 83, 86 Ida-Maras. , 49, 75,˘ 76, 82–83, 89, 124–127, 128, 152, 194–195; linked to Zimri-Lim, 89, 126; Simalite grazing range, 29, 93, 126

Subject Index ideology and collective power, 226–227 Ila-kabkabu, 159, 161 Ilans. urâ, 163 Imar (=Emar), 29, 40, 64, 66–82, 93, 94, 105, 121, 124, 138, 197–209, 210, 211, 212–214, 234 information control, 168–169 Iˇsbi-Erra, 8 Isin kingdom, 8, 250 Iˇsme-Dagan, 10–11, 123, 136, 155, 159, 161, 163 Isqâ-and-Qâ, 185, 202 Jacobsen, Thorkild, 15–16, 130, 177, 204, 235 kadûm, 63, 100–102 Kahat, 163 ˘ kaprum, 58 Kasallu(k), 186–187 Kas. uri-hala, 153, 154, 155 Kengir,˘129–130 king (ˇsarrum), 105, 193, 198 king and collective, 165–166 kingship at Mari, 165–169 kinship ideology, 30–33 kinship language in diplomacy, 66–82 Kiˇs, 17 kispum ritual, 161 Kranzhugel, ¨ 213, 220 Kudur-mabuk, 124, 160 Kurdâ, 82–83, 127, 138, 150, 159, 160, 162, 163, 196, 202, 211 Lanasûm, 93, 94, 105, 138, 197–209, 210, 211, 214 laputtû (deputies?), 64, 207 Larsa, 29, 40, 121, 124, 160, 210, 212–214, 234 li mum, 58–61, 93; definition, 43, 45, 61; later use, 61; people of a ruler, 60–61, 93; Yaminite term, 57, 61 lineage theory, 32–33 loyalty oaths, 61, 74, 96–97, 102, 161, 173 Lugalzagesi, 129–130

347

´ LU.ME Sˇ + Geographic Name lu.meˇ ´ s ( GN), 180–181, 182–188; acting as a polity, 184–185; not an assembly, 181, 182, 184, 185, 187 Marduk, 204 Mari (Tell Hariri), 1–2; archives, 1–4, 17, 18–19, 229, 243; district, 52, 133; first settlement, 6, 221; and the land of the Hana, 106, 147–148, 149–150, ˘ 156, 159, 161, 164–165; 152–154, letters, 2–3, 4, 18–19; third millennium, 6–8, 220, 221 Mar-tu, 39 m¯atum (land): as a block, 117; as coalition, 124–128, 179, 185; as countryside, 117–119; definition, 22, 29–30, 105, 107, 115, 116–117; of a god, 120; names, 121–124; with own will, 121; ruled by kings, 105, 119–121; sedentary, 128; and towns, 116, 236; and urbanism, 106, 115, 116, 235–236; of Yaminites, 123 Maya, 113 merhûm (chief of pasture), 76–85; ˘ among Simalites, 76–83, 151, 152; authority outside kingdom, 81–83, 152, 194–195; in charge of pasture, 78, 81; military leader, 78, 79, 84, 86; non-Simalite, 83–85; over sug¯ag¯u group, 53; and Simalite towns, 152; and sug¯agum appointment, 72–73 Mesoamerica, 178 mules, 157 muˇskênum, 139, 141–147; as other, 144–146; as villagers, 143–144, 186; vs. awilum, 142–143, 186, 187; defined by palace, 141–144, 145, 167, 187; not “poor,” 142; religion, 146–147 namlak¯atum (dominions), 119–120 Naram-Sˆın, 161 nawûm (steppe), 151, 167, 174, 262 Nergal festival, 134, 143, 146 Ninevite V period/pottery, 36–37

348

Subject Index

Nippur, 129–130 niˇsu¯ (dependents), 139–141, 186; not political, 139 nomadism, 34–35, 66, 70–72 n¯ubalum (litter?), 157 Numhâ, 82, 87, 88, 121–123, 127, ˘ 150, 159, 160, 162, 196, 202, 138, 211 palace and economy, 166–167 pastoralism: associated with urbanism, 36–39, 66, 70–72, 214–215, 220–221; definition, 34–35; emergence, 35–37; Habur ˘ the River, 36–37; mobility, 34; and state, 38, 70–72, 115, 215 patrimonial household model, 227–228 polis, 238, 239, 240 puhrum (meeting), 198–199, 206–207 ˘ s-Dagan (Drehem), 167 Puzriˇ Qarni-Lim, 82 Qatna, 29 Qat.t.unan, 87, 133, 138, 143–144, 146 Raq¯ai (Tell al-), 36 Razamâ, 163, 184, 196, 203 Rehoboam, 205 rihs.um, 88, 208–210 ˘ h (Tell ar-), 75 Rima ˘ın, 124, 160 Rim-Sˆ Rowton, Michael, 34, 46, 70–72 .sa¯ bum, 97–98 ˇ Sadum-labua, 83, 195, 199 Saggaratum (town of), 190 Saggaratum district, 51–52, 133, 138 ˇ Sakkanakku, 7, 222 Samanum, 158 Samsi-Addu, 10, 13, 123, 124, 126, 135–136, 155, 159, 161–162, 164, 166; administrative organization, 84, 88–89, 133, 134, 135–136; heir of Akkad, 161; “Mesopotamian” kingdom, 124, 161 Samsu-iluna, 127, 141

Sapiratum, 54, 62, 143, 174, 201–202, 252–253 ˇsa¯ pit.um, see district governors Sargon of Agade, 161 sarr¯ar¯u (resisters?), 73, 96–97 Semitic languages, 4 Service, Elman, 15, 110, 112, 223 settlement hierarchy, 107 Shemshara, 197 Simah-ilanê, 82 ˘ Simalites, towns, 54–55 Simalites/Yaminites: as tribal confederacies, 57–58, 61–62, 93–94, 111; definition, 9, 10, 28, 39; different social structures, 45–47, 61–63, 77, 93, 230–231 Sˆın-gamil, 124 Sˆın-kaˇsid, 124, 160 Sippar, 29 sons of Geographic Name (GN), 188–190; as residents, 188–190; not political, 188, 190 state, 107, 110–112 state formation in the Near East, 114–115 state systems, 113–114 ˇ Subartum, 76, 124–125, 127–128, 135, 143, 160 ˇ Subarˆ u language, 76, 128 ˇ ˇ hna), 10, 90–91, 126, Subat-Enlil (Se 155, 159, 163˘ ˇ Subram, 90–91, 126, 195 ˇ hhum, 194 Sudu sug¯ag¯u˘ ˘groups, 53–54, 67–70, 73, 75, 138, 139 sug¯agum (leader), 51–54, 55, 63–76; activities, 74–75; and census, 73–74; definition, 51; grouped with elders, 68–70; leader within a gayum division, 51, 55; royal appointment, 72–73; ruling towns, 51–53, 64–70, 99; tribal leaders, 69–70; tribal origins, 75–76 sug¯ag¯utum payment, 53, 73, 79, 166, 252 Suhûm, 54, 134, 145, 163, 174 ˘ and Akkad, 131, 161, 250 Sumer

Subject Index Sumer as unit, 129–130 Sumerian “land,” 128–131 Sumerian language, 5, 129, 244 Sumu-Yamam, 10, 28 ˇ a, 90 Sunˆ Sutˆ u, 99, 209, 303 tabinau, 21, 296 tahtamum, 189, 197, 207–208, 210, 211, ˘213, 214, 215 Talhayˆ um, 195, 196 ˘ (town of), 94, 189, 197, 217–218 Terqa Terqa district, 51, 52, 133 Terru, 198, 199, 211 town: acting by name, 181–182; as political unit, 172; politically corporate, 171, 174, 232–234 tribal identity, 104 tribe and state, 111 tribe, definitions, 26–28, 30 Tsembaga of New Guinea, 33 Tuttul, 66, 93, 94, 105, 158, 168, 189, 207, 208, 210–211, 212, 214–216, 220, 234 Ugarit, 163 Ur (kingdom), 7, 8, 130 urbanism: southern Mesopotamia, 107; and the state, 106–115; third-millennium Syria, 36–39, 114–115, 218, 219–221 Urgiˇs, 66, 178, 180, 197–199, 211, 212, 216–217, 234 Urkesh, see Urgiˇs Uruk, 36, 124, 130, 160, 205, 206 vassalage, 101 Warad-Sˆın, 124 West Semitic languages, 4–5

349

Yaggid-Lim, 8, 12 Yahdun-Lim, 7, 8, 9–10, 12, 56, 124, ˘ 136, 152–154, 164, 221–222; 126, ˇ military organization, 56–57; Samaˇ s temple inscription, 123, 153–154, 158, 222; two capitals, 154 Yahrurû, 123, 159 ˘ had, 29, 65, 104, 123, 145 Yam ˘ Yaminites: and Akkadian language, 94; and Imar, Tuttul, 212; revolt, 67, 91–92, 148, 250; towns, 51–53, 66–68, 93, 94, 95–96, 102, 150; tribal character, 28–29, 33; under Samsi-Addu, 88, 93, 94–95; under Yahdun-Lim, 153, 154, ˘ 158; under Zimri-Lim, 89, 91–92, 93, 95–96, 99–100, 102, 158–159 Yamutbal, 82, 87, 88, 123, 127, 138, 150, 160 Yap people, 21 Yapturum, 123 Yarim-Lim of Aleppo/Yamhad, 65, ˘ 119 Yasmah-Addu of Mari, 10, 12, 124, 133, 136,˘155, 161, 162, 164, 166 Yasmah-Addu the Yaminite, 162 ˘ Yatar-malik, 194 Yomut Turkmen, 32–33 Zalluhan, 151–152 ˘ Zalmaqum, 75, 76, 94, 124–126, 127, 128, 134, 139; linked to Yaminites, 94 Zimri-Lim, 11, 71–72, 156; as Simalite, 46, 92; away from Mari, 162–164; chronology, 246; district system, 133, 135; realm of two rivers, 164–165; tribal king, 26, 71, 152, 156, 159–161, 231–232; two parallel administrations, 54, 78–80, 84, 156, 158, 159

Index of Mari Texts

Note: In this index, text citations that include translations of more than one or two words are marked with an asterisk (*) before the page numbers in question. ARM I 2 I3 I6 I7 I9 I 10 I 13 I 15 I 18 I 19+ I 22 I 23 I 24+ I 37 I 39 I 42 I 43 I 53+ I 62 I 67 I 69+ I 82 I 87 I 91+ I 113+ I 118 I 119 I 128 I 132

173 245 *74, *88, *95, 122 182, 273 202, *281 *121, 125, 296 255 182 125 122 279, 281 279 287, 303 183, *266 291 *90, 266, 279 274, 296 125, 183 *84, 263 273 279 266 266 *140, 272, 281 273, 280 290, 296 *99, 102, 258 255 122, 279

ARM II 1 II 12 II 13 II 14 II 16 II 18 II 21 II 23 II 24+ II 25 II 27 II 33 II 35 II 37 II 38 II 48 II 49 II 50 II 53 II 54 II 55 II 56 II 59 II 61 II 62 II 66 II 68 351

266 266 272 267 274 122, 123, 262 125, 279 122 *118 122, 265 265 261, 262, 285, 292, 303 125 262, 265 299 *88, 152, 270, 303 122, 267 280 *47–48, 59, 89, 95, 253, 260 59 59 59 137, 151, *256, 286 59, 282 282 122 125

352 ARM II 75

Index of Mari Texts

*185, 187, 201, 202, 294, 296, 297, 301, 307 II 80 283, 284 II 81 136 II 83 301 II 92 *67, 95, 253, 262, 281 II 94 183 II 101 280 II 102 259, 262 II 103 253, 258, 262, 281 II 121 273 II 122 280 II 130 125 II 131 279 II 137 296 ARM III 3 *189, 301 III 5 280 III 6 *68, 74, 136, 281, 296 III 12 *96, 134, *262 III 13 262 III 15 *151 III 16 97 III 17 192, *262, 299, 305 III 19 260 III 20 260 III 21 253, 262, 281 III 26 296 III 34 136 III 37 125 III 38 *51, 68, 260 III 41 279 III 45 136 III 50 257 III 58 268 III 65 *199, 297 III 70+ *68, 136, 253, 258, 260, 281 III 83 294 ARM IV 1 50, *55, 62 IV 5 182 IV 6 *122, 128 IV 7+ *269 IV 11 296 IV 12 279 IV 16 262 IV 17 296 IV 24 282

ARM IV 25 IV 29 IV 31 IV 40 IV 57 IV 68 IV 73 IV 74 IV 76 IV 77 IV 78 IV 80 ARM V 1 V3 V 15 V 23 V 24 V 25 V 33 V 36 V 51 V 63 V 65 V 66 V 68 V 69 V 72 V 73 V 78 V 81 ARM VI 3 VI 12 VI 27 VI 28 VI 31 VI 32 VI 40 VI 58 VI 66 VI 76 ARM VII 227 VII 267 VII 277 VII 311 ARM VIII 3 VIII 11 VIII 14

274 297 122 88 266, 274 275, 280, 297 293, 296 262 272 281 296 290 290 258 290 *122, 276, 303 185, 258, *261 279, 285 278 272 88, *89, 266, 270 122 *118 273 279 278 75 273 296 143, *155, *269 282 297 125 265, 266 280 262 52, 262 279 125, 273, 274 156, 158, 249, 274 254 258 278 66, *192, 254 269 95, 98, 99, 248, 258, 288 269

Index of Mari Texts ARM VIII 85+

54, 143, *201–202, 255, 292, 301 ARM IX 248 52, 253 X 31 *273 X 33 283 X 34 125 X 51 273 X 74 292 X 78 286 X 84 52, 122, 273, 280 X 91 75 X 121 292, 297 X 122+ 122, 267 X 153 286 X 155 122 X 156 59 X 157 137, 256 X 166 *140 X 173 117 ARM XIII 27 122 XIII 36 290 XIII 39 136 XIII 46 122, 125 XIII 47 122 XIII 83 182 XIII 105 257 XIII 106 297 XIII 108 281 XIII 117+ 125, 282, 297, 299, 305 XIII 123 296 XIII 139 294 XIII 141 *284 XIII 143 273 XIII 144 122, 183, 185, 269, *275, 280, 293, 294 XIII 145 297 XIII 146 125, 274, 295 XIII 147 295 XIII 148 *195, 296, 297 ARM XIV 8 75, 258, 262, 279, 281 XIV 12 284, 285 XIV 13 280 XIV 14 283 XIV 33 182 XIV 39 283 XIV 46 *52, 72, 75, 258

ARM XIV 47 XIV 48 XIV 50 XIV 51 XIV 53 XIV 55 XIV 61 XIV 62 XIV 64

353

136 283, 284 270, 299 125, 279 78 *197 296, 307 254 73, 258, 260, 262, 281, 297 XIV 65 258, 260, 262, 281, 297 XIV 66 300 XIV 69 279 XIV 70 136 XIV 72 299 XIV 75 182, 183, 258, 281 XIV 76 125, 182, 294 XIV 78 182, 270 XIV 80 286 XIV 81 78, 283 XIV 84+ 303 XIV 86 78 XIV 102 122, 267 XIV 104+ *184, 269, 293, 297, 298 XIV 107 281 XIV 112 125 XIV 114 297 XIV 121 134, 256, *263, 283 XIV 125 122, 267 XIV 128 183 ARM XVIII 54 52 ARM XXI 388 298 ARM XXII 127 278 XXII 326 52 ARM XXIII 69 270, 292 XXIII 70 269, 292 XXIII 86 292 XXIII 426 253 XXIII 427 270 ARM XXIII 428/429 67, 74, 89, 96, 247, 248, 253, 255, 256, 260, 269, 278, 288 XXIII 430 278

Index of Mari Texts

354 ARM XXIII 504 XXIII 552 XXIII 554 XXIII 595 ARM XXIV 53–63 XXIV 53 XXIV 54 XXIV 55 XXIV 56 XXIV 57 XXIV 58 XXIV 59 XXIV 60 XXIV 61 XXIV 62 XXIV 63 XXIV 158 XXIV 235 XXIV 287 ARM XXV 783 ARM XXVI 1 XXVI 5 XXVI 6 XXVI 12 XXVI 17 XXVI 24

XXVI 27 XXVI 30 XXVI 31 XXVI 35 XXVI 38 XXVI 39 XXVI 40 XXVI 41 XXVI 42 XXVI 43 XXVI 44 XXVI 45 XXVI 46 XXVI 47 XXVI 48

297 254 253 269 254 254 254 52, 254 254 52, 254 254 52, 254 52, 254 52, 253, 254 52, 254 52, 254 52 56, 57, *249, 255, 256 259 278 *146–7, 291 52, 246, 257, 258, 290, 295 52, 257, 290 *69, 75, *125, 139 188, *279 52, 69, 125, *200, 247, 260, 265, 267, 297, 301 *88, 303 303 303 302 302 59, 301 69, 303 *54 *88 256, 303 284 209, 210, *256, 303, 307 256, 303 279 297

ARM XXVI 58 XXVI 61 XXVI 62 XXVI 75 XXVI 81 XXVI 85 XXVI 86 XXVI 87 XXVI 88 XXVI 109 XXVI 114 XXVI 115 XXVI 126 XXVI 131 XXVI 141 XXVI 144 XXVI 145 XXVI 150 XXVI 154-bis XXVI 156

279, 283 279 294 250 279, 280 285 *84 280 272 147, 285 *83 283 282 157, 287 265 294, 295 137, 256 262 *144, 284 140, *256, 294 XXVI 168–172 270 XXVI 168 *59, 95, 97, 99, 100, 258 XXVI 169 59, 259 XXVI 170 59, *67 XXVI 171 59, 259 XXVI 172 59 XXVI 181 283 XXVI 199 *140 XXVI 206 297 XXVI 210 274 XXVI 215 292, 296 XXVI 220 286 XXVI 224 259 XXVI 233 250, 279 XXVI 238 273 XXVI 241 *253 XXVI 246 183, 293 XXVI 256 *138, 274, 294, 305, 307 XXVI 265 279 XXVI 282 *92, 259 XXVI 303 125 XXVI 306 125, 303 XXVI 308 125 XXVI 309 125 XXVI 310 292, 294

Index of Mari Texts ARM XXVI 312 XXVI 313 XXVI 320 XXVI 323 XXVI 325 XXVI 329 XXVI 340 XXVI 347 XXVI 352 XXVI 357 XXVI 358

157, 185, *287 300 295 274 125 125 137 125, *173, 302 125, 137, 303 300 122, 265, 267, 286 XXVI 359 125 XXVI 362 274 XXVI 365 *186–7, 201, 281, 282, 293, 294 XXVI 365-bis 122, 125 XXVI 367 279 XXVI 371 *302 XXVI 372 273 XXVI 373 137, *256 XXVI 375 182 XXVI 376 122 XXVI 377 *121 XXVI 382 274 XXVI 383 122 XXVI 384 125, 173, 183, 273, 292 XXVI 385 *122, 160, 182, 273 XXVI 386 288 XXVI 388 263 XXVI 389 263 XXVI 391 297 XXVI 393 197, *281 XXVI 394 273 XXVI 398 136 XXVI 404 200, *267 XXVI 405 291 XXVI 406 173 XXVI 409 183, 293, 294, 295 XXVI 410 *187 XXVI 411 273, 274, 297 XXVI 412 122, *207, 274, 284, 307 XXVI 416 *118, 137, 273 XXVI 417 300 XXVI 418 284

ARM XXVI 422 XXVI 423 XXVI 424 XXVI 425 XXVI 427 XXVI 430 XXVI 432 XXVI 435 XXVI 437 XXVI 438 XXVI 443 XXVI 447 XXVI 449 XXVI 450 XXVI 462 XXVI 463 XXVI 468 XXVI 479 XXVI 489 XXVI 491 XXVI 494 XXVI 495 XXVI 503 XXVI 508 XXVI 510 XXVI 511 XXVI 512 XXVI 515 XXVI 519 XXVI 521 XXVI 523 XXVI 542 XXVI 548 XXVI 549 ARM XXVII 1 XXVII 2 XXVII 14 XXVII 16 XXVII 17 XXVII 19 XXVII 20 XXVII 23 XXVII 25 XXVII 26 XXVII 27 XXVII 28 XXVII 30

355 293 *117 293 122 273 122, 174, *274 122 294 272, 275 281, 297 292 279, 281 291 *73, 253, 260, 281 253, 295 196, *281, 297, 302 250, 291 297 173 273 274 295 297 270 270 273 292 272, 273 272, 274 122 274, 304 274 274 *41 283, 286 281 265, 279, 283, 303 157, 286, 287 *87, 286 *160 88, 125, 265 296 *136, *143, 279 *125, 143, 283, 291 *144, 280, 283, 296 136, 296 283

356 ARM XXVII 36 XXVII 37 XXVII 38 XXVII 39 XXVII 40 XXVII 43 XXVII 45 XXVII 46 XXVII 48 XXVII 57 XXVII 58 XXVII 60 XXVII 61 XXVII 64 XXVII 67 XXVII 69 XXVII 70

Index of Mari Texts

290 *144–145, 283 283 283, 284 280 283 125 279, *280 *78, 81, 82 *87 136 297 263 137 297, 302 264 *81, 82, 263, 270, 299 XXVII 72-bis 173, *273 XXVII 76 136 XXVII 80 125 XXVII 89 125 XXVII 93 *53, 78, 264 XXVII 94 254, 264 XXVII 99 295 XXVII 100 283 XXVII 101 283 XXVII 102 283 XXVII 105 83 XXVII 107 52, 136, 258, 283, 296 XXVII 113 78 XXVII 116 270, 279, 299 XXVII 120 279 XXVII 132 125, 263, 265, 274 XXVII 135 *183 XXVII 141 122 XXVII 143 122 XXVII 145 122 XXVII 147 125 XXVII 151 125, 263, 265 XXVII 161 122 XXVII 162 125 XXVII 167 *117 ARM XXVIII 1 279 XXVIII 6 278 XXVIII 7 278 XXVIII 10 278

ARM XXVIII 15 XXVIII 19 XXVIII 20 XXVIII 25

125, 276 294 *140 59, 251, 266, 303 XXVIII 36 251 XXVIII 38 278 XXVIII 40 125 XXVIII 42 293 XXVIII 44 299 XXVIII44-bis *198, 299 XXVIII 45 299 XXVIII 46 198, *270, 299 XXVIII 48 125, 275, 293 XXVIII 50 291, 298 XXVIII 51 *189 XXVIII 52 298 XXVIII 54 275 XXVIII 55 88, 125, 275 XXVIII 56 125 XXVIII 59 298 XXVIII 60 *125, 134–135, 273 XXVIII 61 271, 294, 300 XXVIII 62 125, 280, 300 XXVIII 64 298 XXVIII 65 83, 276, 298 XXVIII 67 259, 298 XXVIII 68 274 XXVIII 69 *198 XXVIII 77 283 XXVIII 78 125 XXVIII 79 *125, 152, 280 XXVIII 80 *173 XXVIII 83 279 XXVIII 84 278 XXVIII 91 *101, 102, 294, 298, 300 XXVIII 93 125 XXVIII 95 90, 298 XXVIII 98 293 XXVIII 99 300, 303 XXVIII 100 264 XXVIII 103 298 XXVIII 104 293, 294 XXVIII 105 293 XXVIII 106 302 XXVIII 107 300

Index of Mari Texts ARM XXVIII 110 XXVIII 111 XXVIII 112 XXVIII 113 XXVIII 120 XXVIII 136 XXVIII 141 XXVIII 147 XXVIII 148 XXVIII 162 XXVIII 163 XXVIII 164 XXVIII 165 XXVIII 167 XXVIII 168 XXVIII 169 XXVIII 170 XXVIII 172 XXVIII 179 FM II 34 II 35 II 38 II 46 II 63 II 69 II 70 II 71 II 88 II 116 II 117

II 118 II 122 II 123 II 125 II 127 II 131 FM III 2–5 III 4 III 20 III 28 III 60 III 67 III 95

280 *298 280 300, 303 *70, 259 280 *202 *100, 101, 259 125 82, 299 264 264 264 264 264 264 264 122 278, 284 269 291 284, 285 291, 292 *52, 79, 303 283 136 268, 284 284 *69, *90, *91, 267, 301, 303 69, 82, *197, 264, 265, 296, 299 274 297 125, 262, 303 199, 297, 300 183, 280 *73 290 245, 275, 289 *54, 265 298 279, 292, 298 279 279

FM III 131 III 135 III 136 III 147 III 148 III 149 FM VI 4 VI 6 FM VI 7 VI 9 FM VII 1 VII 6 VII 7 VII 8 VII 18 VII 20 VII 26 VII 30 VII 36 VII 47 A.49 A.96 A.215 A.228 A.230 A.250 A.257 A.285 A.315+ A.333 A.350+ A.361 A.402 (=FM VI 4) A.449 A.486+ A.489 A.505 A.609 A.623+ A.687 A.715 A.885 A.889 A.954

357 294 *48, *86, 87, 252 50, *55, 74 *137 183 183 208 126 128 303 259 *142 248, 250, 299 *120, 274 *208 287 250 296 296, 307 *145, 272 259 285 125 *303, 304 274 292 294 295 *188, 296 264 *151 265 208, 210, 304 59, 248 56, *72, 254, 256, 264, 265 249, 288 *90, 251, 267 298 304 283 275 *183, 189, 294, 296, 304 245 209, *256

358 A.981

A.987 A.988 A.1025 A.1051 A.1086 A.1098 A.1121+ A.1139 A.1146 A.1212 A.1215 A.1230 A.1289+ A.1333 A.1421 A.1487+ A.1610+ A.1866 A.1968 A.2090 A.2094 A.2119 A.2210 A.2226 A.2231 A.2326 A.2428 A.2432 A.2435 A.2459 A.2500+ A.2526 A.2560 A.2567 A.2692+ A.2724 A.2728 A.2730 A.2741 A.2757 A.2769 A.2796 A.2801

Index of Mari Texts *63, 93, 97, 98, 99, 100, 102, 248, 253, 257, 258, 270, 297 59 250 *78, 173, 263 *142, 186 268 *82, 126, *263, 285 *168 278 *59, 60 265 263, 264 *210, 304 122, 280, 291 279 122, 267 *174 *53, 122, 125, 261, 265, 279 272 *120 *60 59 125, 254, 262, 265 266 83, 192, *193–195, 198, 257, 265, 281, 297, 300 *119, *121, 124, 280 265–280 304 281 253 294 297 59 *58, *90 297 260, 301 258 293 86, *203, 265, 281 83, 262, 266 296 294 98, 99, 258, 268 99, 258, 281

A.2819 A.2939 A.2943 A.2951 A.2993+ A.2995+ A.3024 A.3051 A.3080 A.3185 A.3206 A.3209 A.3243 A.3347 A.3354+ A.3567 A.3572 A.3577 A.3592 A.3669+ A.3821 A.3960 A.3976 A.4251+ A.4280 A.4309 A.4332 A.4509 A.4513 A.4530-bis A.4535-bis A.4687 B.590 M.5037 M.5085+ M.5157+ M.5172 M.5319 M.6060 M.6159 M.6182 M.6435+ M.6874 M.7259 M.7421 M.7630 M.7964 M.8512

285 299 280, 296 *189, 296, 304 293, 294, 295 52 *116, 125, *173 283 59, 248, 252 59 125 122 *207, *303, 304 299 125 *88, *209, 256, 303, 304 *90, 248 *160, 211, 299, 303 301 *118, 122, 279 59, 256 *123, *158, 288, 291 304 274 251, 255, 265 273 264 289 295 *209, 211, 286, 304 295 295 125 *124 150 296 100, 258 254 *151, 251, *257, 285, 291 293 *161, 291 285 *209, 267, 297, 301 285, 291, 302 265 *88, 257, 295 291 267

Index of Mari Texts M.8884 M.9175 M.9610 M.9623 M.9881 M.10991 M.11070 M.13096 M.13217 M.15083

294 254, 264 253 265 278 *187 303 304 303 122, 267

Dur-Yahdun-Lim inscription (Yahdun˘ Lim)˘ 287

359

Fleming 1998, 61 and n. 91 *29 RIMA 1, A.0.39.4 287 RIMA 1, A.0.39.5 287 RIME 4, E4.6.8.1 (= Dur-Yahdun-Lim ˘ inscription) *154 RIME 4, E4.6.12.4 287 ˇ Samaˇ s temple inscription (Yahdun-Lim) 123, *153, 275, 286, 287 ˘ TH 72–16 294 TH 72–39+ 125

Democracy's Ancient Ancestors: Mari and Early Collective Governance

Mari Stregati

Mari invenţii

Mari Stregati

Mari d'elle

Mari Stregati

Mari stregati

Mari Stregati

Sui Mari Del Fato

Mari (Women's West Series)

Sui Mari Del Fato

Mari, une ville perdue

Terrorism and Collective Responsibility

Ancestors of Avalon

Sociodrama and Collective Trauma

Collective Preference and Choice

А.П.Чехов. Mari d'elle

Enrique Mari 1928

Ancestors and Relatives: Genealogy, Identity, and Community

Shadows of Forgotten Ancestors

The Collective

The Collective

The Collective

Collective Memory

Sui Mari Del Fato

Collective Memory

The Collective

The Collective

The Collective

Collective Memory

Collective remembering

Democracy's Ancient Ancestors: Mari and Early Collective Governance

Mari Stregati

Mari invenţii

Mari Stregati

Mari d'elle

Mari Stregati

Mari stregati

Mari Stregati

Sui Mari Del Fato

Mari (Women's West Series)

Sui Mari Del Fato

Mari, une ville perdue

Terrorism and Collective Responsibility

Ancestors of Avalon

Sociodrama and Collective Trauma

Collective Preference and Choice

А.П.Чехов. Mari d'elle

Enrique Mari 1928

Ancestors and Relatives: Genealogy, Identity, and Community

Shadows of Forgotten Ancestors

The Collective

The Collective

The Collective

Collective Memory

Sui Mari Del Fato

Collective Memory

The Collective

The Collective

The Collective

Collective Memory

Collective remembering

Recommend Documents