Dissertation.book

Table of Content

List of figures

p. v

Acknowledgements

p. vi

Vita

p. vii

Abstract

p. x

Conventions

p. 1

Chapter 1. Introduction

p. 4

Chapter 2. The sources, the technical terms

and the structure of the documents

p. 13

Chapter 3. Patterns of Succession

p. 87

Chapter 4. The House of Ur-Nammu

p. 96

Chapter 5. The Ruling Family of Ur III Umma

p. 130

Section 1: Introduction

p. 130

Section 2: The earliest generations

p. 149

Section 3: The governor

p. 153

Section 4: Ur-Lisi

p. 156

Section 5: Ayakala

p. 168

Section 6: Dadaga

p. 175

Section 7: The children of the governors: Lu-Emah, Ikala,

Nin-Ekuta, Namzitara, Gududu, Inim-∑ara, and Lu-∑ulgi(ra)

p. 181

Section 8: Ur-E

’

e, the chief cattle administrator

p. 195

vii

Acknowledgments

Many remarkable people have helped me achieve this milestone in my academic career. I wish

to thank, firstly, my former advisor at the University of Copenhagen (KU), Bendt Alster, and

my current advisor Robert Englund at the University of California, Los Angeles (UCLA). I

can never repay the support both have offered me over the years!

I wish to thank all of my former teachers, both at KU and UCLA, and I mention here, briefly,

Aa. Westenholz, M. Trolle Larsen, and I. Thuesen in Copenhagen, and E. Carter, and G. Buc-

cellati in Los Angeles. Among the many students at the KU and UCLA who have offered their

help to me over the years I whish to thank here, in particular, C. Johnson at UCLA and G.

Barjamovich at KU.

No study such as this appears in a vacuum, and I wish to thank all my colleagues, above all: N.

Vanderoost, M. Widell, N. Koslova, B. Lafont, R. Mayr, R. de Maier, M. Molina, and M.

Sigrist. I wish to thank, in particular, M. Molina and M. Sigrist for allowing me to include

large amounts of unpublished material in this study.

This study would never have materialized had it not been for the economic support which has

been given to me through the years. Firstly, this study is the result of a research stipend from

Forskerakademiet (the Danish Research Academy, now under Forskeruddannelsesrådet). My

thanks to the Danish state are sincere! Secondly, during my years at UCLA, I have received

substantial financial aid, in particular during the final period of study during which I received

the Chancellors Dissertation Year Fellowship.

Finally, it is necessary for me to mention that this study would not have been possible without

the framework of the Cuneiform Digital Library Initiative (CDLI) and without the support

of our German partners at the Max Planck Institute for the History of Science, Berlin. All the

associates of the CDLI, and all of the UCLA staff of the CDLI shall here be rightfully

thanked.

Jacob Dahl

Los Angeles, May 30, 2003

C o n v e n t i o n s

This study is based on the more than 15,000 primary documents recovered from the

ancient city of Umma, as well as the tens of thousands of texts from other sites.

Approximately 90% of all published texts along with several thousand unpublished texts

have been available to me in transliteration (in addition to easily accessible scanned-images of

the hand-copies of ca. 25,000 Ur III texts, and a limited number of images of original

tablets) through the Cunieform Digital Library Initiative (CDLI).

The transliteration conventions applied in this study rely largely on the standards of

the CDLI project, and the Pensylvanian Sumerian Dictionary (PSD). Sign-readings have, as

far as possible, been based on the Borger / Ellermeier leagacy.

The dates of administrative

records, abbreviations, and text ID’s are given here according to CDLI recommendations.

Dates are entered in the form

(uler)

(ame)

(ear)

(=number)

(onth)

# D

(ay)

with "XX" for lost information, "00" when information was not given by the scribe, "?"

where uncertain and "

min

" or "

diri

" where appropriate. Roman numerals are used for

months

Examples:

AS 5 v 12?

Amar-Suen, year 5, month 5, possibly day 12

See http://www.ucla.edu

R. Borger,

Assyrisch-babylonische Zeichenliste

(= AOAT 33-33A; Neukirchen-Vluyn 1981) (abbreviated

ABZ), and F. Ellermeier,

Sumerisches Glossar 1/1/1-2

(Nörten-Hardenberg 1979-1980).

∑

5? - iv

diri

∑ulgi

, possibly his year 5, extra month 4, day 12

∑

S? 5? xiii

diri

Possibly reign of

∑u

-Suen, possibly his year 5,

extra month 13, day 12

XX ix

min

Ibbi-Suen, year name lost, second month 9, no day given

Metrological conventions are adapted from R. Englund,

Organisation und Verwaltung

der Ur III-Fischerei

(= BBVO 10; Berlin 1990) xiii - xvii, whereby notations have been

revised according to the conventions of the CDLI.

Sexagesimal system (used when counting discrete objects, as well as for capacity

notations larger than one gur):

1(Òar

/guru

) = 6 (geÒ

’

u) = 3600

1(geÒ

’

u) = 10 (geÒ2) = 600

1(geÒ2) = 6(u) = 60

1(u) = 10(diÒ/aÒ)

Fractions of are expressed either by using the “gin

” system and divide the 1 into

fractions of 60, or by applying the “igi-

-gal

” system.

igi-6-gal = 1/6

igi-4-gal = 1/4

etc.

Capacity system (sila - gur system):

1(aÒ) gur = 5(barig) = 300 sila

http://cdli.ucla.edu/methods/de/conventions.html

1(barig) = 6(ban

) = 60 sila

1(ban

) = 10(diÒ) sila

The sila

can be subdivided in fractions of 60 by using the “gin

” system:

1(sila

) = 60 gin

etc.

Weights (shekel system):

1 gu

= 60 ma-na

1 ma-na = 60 gin

1 gin

(shekel) = 180 Òe

Area:

1(Òar

) = 6(bur

)

1(bur

’

u) = 10 (bur

)

1(bur

) = 3(eÒe

)

1(eÒe

) = 6(iku)

1(iku) = 100 (sar)

Approximate conversions to metric system:

1 sila

= 1 liter

1 gur = 300 liter

1 gin

= 8 1/2 gram

1 ma-na = 1/2 kilogram

1 sar = 36 square meter

1 bur

= 6, 48 hectare

Bibliographical abbreviations follow the CDLI recommendations (http://

cdli.ucla.edu/Tools/abbrev.html).

C h a p t e r 1 . In t r o d u c t i o n

Tens of thousands of administrative documents have survived from the time of the

Third Dynasty of Ur, making it one of the best documented periods in the early history of

man. In brief, the Ur III empire underwent four stages of development. During the

consolidation period, the local ruler of Ur managed to unite the city-states of Sumer and to

conquer the Akkadian area. Subsequently, during the expansion period, the king was able—

by way of political alliances and military raids—to subjugate neighboring areas to the east

and northeast. The empire flourished for but some few years, during which the

administrative machinery produced thousands and thousands of records. Shortly after the

empire had reached its zenith, it collapsed, and, following a period of territorial losses, the

capital city, Ur, was taken by invading troops from the neighboring, former vassal, Elam.

In spite of the extraordinarily rich documentation describing, in detail, many aspects

of the social history of the period, sources are not clear on the issue of succession.

Consequently, it has been impossible to establish, beyond doubt, a genealogy of the royal

family, the clan of Ur-Nammu. Succession within the elite families in the provincial centers

such as Umma—the topic of the current study—are, however, far better documented.

I divide the period as follows: the reign of Ur-Nammu and the first 20 years of

∑

ulgi’s rule, until his

deification, are considered a period of consolidation.

∑

ulgi’s expansions, recorded in the year-dates from

his 20

year till around his 40

year, and the construction of the administrative center Drehem,

characterize the next period. The final years of

∑ulgi

and the reign of Amar-Suen, with numerous

administrative records, but fewer military achievements, were the apex of the Ur III Empire. From the

middle of

∑u

-Suen’s reign and during the early years of Ibbi-Suen we find the first signs of weakness,

while the last 20 years of Ibbi-Suen’s reign are considered a period of decline.

In consequence, this study seeks to use the daily records of the Umma provincial

court to reconstruct the genealogy, the line of succession for the important offices, as well as

the careers and offices of the members of the ruling family of Umma. This study proposes a

model for understanding other, less well-known, Ur III elite families. This analysis is the first

detailed description of patterns of succession concerning ancient Mesopotamia, references to

previous studies of the same topic are not surprisingly few.

The present study takes as its

starting point the prevailing understanding in the field that restricted primogeniture was, by

far, the most successful pattern of succession in ancient Mesopotamia.

Since this has never

been systematically questioned it has become something of a dogma in the field. This study

seeks to challenge that dogma.

Tthe only study which, to my knowledge, mentions the possibility of fratrilinial succession in non-royal

circles, during the Ur III period is P. Steinkeller, “The Foresters of Umma: Toward a Definition of Ur III

Labor,” AOS 68 (=M. Powell (ed.),

Labor in the Ancient Near East

; 1987) 73 - 115. Steinkeller did not

discuss the sociographical implications of this, and the issue has remained almost entirely untouched.

Since it has long been suggested that Amar-Suen and

∑

u-Suen were brothers following each other on the

throne, some studies have described this as fratrilineal succession; however, there has never been any

discussion of the consequences of having one system of succession for the royal family and another for

the rest of society, if this was in fact the case. Edzard briefly mentioned, but ultimately discarded the

theories of fratrilineal succession in ancient Mesopotamia (see fn. 228 on p. 92 in this study). This

sentiment was captured by D. Snell,

Life in the Ancient Near East

(New Haven 1997) 20. See also N.

Postgate,

Early Mesopotamia, Society and Economy at the Dawn of History

(London 1992) 148 and 270,

and N. Postgate, “Royal Ideology and State administration in Sumer and Akkad” in J. Sasson (ed.)

Civilization of the Ancient Near East

(New York 1995) 397: “...strict primogeniture is not universal:

Sumerian rulers were sometimes succeeded by their brother, and among sons it was not always the

eldest.”

See for example A. Leo Oppenheim,

Ancient Mesopotamia; Portrait of a Dead civilization

(Chicago 1964)

77 and 79.

The founder of the dynasty, Ur-Nammu, ruled for eighteen years, his son and

successor

∑ulgi

for forty-eight years;

∑ulgi

was followed first by Amar-Suen who ruled 9

years, then by

∑u-Suen

who likewise ruled 9 years, and lastly by Ibbi-Suen who ruled 24

years. The latter three rulers were all, perhaps, sons of

∑

ulgi.

According to the now contested middle chronology

the first year of Ur-Nammu fell

around 2112 BC, and the last year of Ibbi-Suen around 2004 BC. The extent of overlap of

the Ur III period with the preceding period, the Second

LagaÒ

Dynasty (the dynasty of Ur-

Baba), is unclear at best. The two dynasties are believed, however, to have substantially

coexisted.

Chronological correlation between the Ur III period and the following Isin-Larsa

period is, on the other hand, relatively clear: the Isin dynasty was founded by

IÒbi-Erra

, a

former ally of Ur

already in the eighth year of Ibbi-Suen.

The reconstruction of the social history of the Ur III period relies heavily on the

analysis of administrative documents. Historical inscriptions of the period contain very little

information about social phenomena. The year names used by administrators to date the

economic records, however, are a useful source for the history of the period. The years were

named according to important events, and the year formulae of the administrative records

recount these names. The historical inscriptions are found on buildings (or rather on

building materials such as bricks and door sockets, etc.), on commemorative objects

(including stelae, statues, etc.), and on objects dedicated to the rulers, primarily seal

J. Reade, “Assyrian King-Lists, the Royal Tombs of Ur, and Indus Origins,” JNES 60 (2001) 1 – 29.

Further see P. Huber, “Astronomical Dating of Ur III and Akkad,” AfO 46 – 47 (1999/2000) 50 – 79.

See fn. 237 on p. 98.

inscriptions and miscellaneous cultic objects.

The literary texts as well as the majority of the

pseudo-historical inscriptions can only be used with great caution to investigate the social

history of the third millennium BC due to their obvious political or literary biases.

Early

Assyriologists placed great hope in the Sumerian King List (SKL)—an Isin-Larsa pseudo-

historical account of all previous kings of Sumer—as a source of the history of the Ancient

Near East.

In later times scholars often discarded the text entirely, deeming it a fictitious

political manifesto of the Isin rulers.

However, since the information given by the SKL

regarding the last hundred years of the third millennium BC is basically correct, we must

interrogate the sources available to the Isin scribe who composed this list.

For the historical inscriptions of the Ur III period see D. Frayne,

Ur III period (2112 - 2004 BC)

RIME 3/2; Toronto 1997).

10.

For the pseudo-historical inscriptions, or the intertextuality between historical inscriptions and the

literary texts see W. Hallo, “The Cultic Setting of Sumerian Poetry,” in A. Finet (ed.),

Actes de la XVIIe

Rencontre Assyriologique Internationale

(= Rencontre 17; Brussels 1969) 117 - 122.

11.

The 1939 edition by Th. Jacobsen remains the standard Assyriological work of reference [published as

AS 11]. Among later, significant, studies are C. Wilcke, “Die sumerische Königsliste und erzählte

Vergangenheit,” in J. Von Ungern-Sternberg (ed.),

Vergangenheit in mündlicher Überlieferung

(Stuttgart

1988) 113 - 140, and P. Michalowski, “History as Charter,” JAOS 103 (1983) 237 - 248. See F. Hrozny,

“Die ältesten Dynastien Babyloniens,” WZKM 26 (1912) 1 - 20, for an introduction to the early debate

on the use of the SKL. See also A. Schneider,

Die Anfänge der Kulturwirtschaft. Die Sumerische

Tempelstadt

(Essen 1920) 3.

12.

See for example P. Michalowski, JAOS 103 (1983) 237 - 248, and C. Wilcke “Genealogical and

Geographical Thought in the Sumerian King List,” in H. Behrens et al. (eds.),

DUMU-É-DUB-BA-A

Sjöberg Fs.; Philadelphia 1989) 557 - 571.

The SKL recounts the third dynasty of Ur as follows:

unug

geÒ

tukul ba-an-sag

Uruk was slain with weapons,

nam-lugal-bi ¿uri

•



-Òe

] ba-DU

and its kingship was carried to Ur;

uri

-ma ur -

nammu] lugal

in Ur, Ur-Nammu was king,

mu 1(u) 8(diÒ) i

-ak]

(he) ruled 18 years.

Òul-gi dumu

ur-nammu-ke

∑ulgi, the son of Ur-Nammu,

mu 4(u) 8(diÒ) i

-¿ak•

ruled 48

years.

amar-

suen dumu

Òul-gi-ke

Amar-Suen the son of ∑ulgi,

mu 9(diÒ) i

-ak

ruled nine years.

Òu-

suen dumu

amar-

suen

∑u-Suen, the son of Amar-Suen,

mu 9(diÒ) i

-ak

ruled nine years.

i-bi

suen dumu Òu-

suen-ke

Ibbi-Suen, the son of ∑u-Suen,

mu 2(u) 4(diÒ) i

-ak

ruled 24 years.

5(diÒ)

lugal

(They were) five kings

mu-bi 1(geÒ

) 4(u) 8(diÒ) ib

-ak

who ruled 108 years.

uri

-ma

geÒ

tukul ba-an-sag

Ur was slain with weapons,

nam-lugal-bi i

-si-in

-Òe

ba-DU

and its kingship was carried to Isin.

-si-in

-na iÒ-bi-er

-ra lugal

In Isin, IÒbi-Erra was king.

The accuracy of this excerpt suggests that the Isin-scribe who wrote the SKL had

access to lists of year-formulae—some such lists have been recovered.

However, it is likely

13.

Col. viii 7 - 23 following Th. Jacobsen, AS 11, 1939, and addendum by C.-A. Vincente, “The Tall

Leilan recension of the Sumerian King List,” ZA 85 (1995) 234 - 270.

14.

See C.-A. Vincente, ZA 85, 266.

15.

See C.-A. Vincente, ZA 85, 266.

16.

Perhaps the commemorative stelae preserved at Nippur, copied and compiled into literary compositions,

served as the sources for the SKL (W. Hallo, Rencontre 17 (1969) 118ff.). See also P. Michalowski, “The

Bride of Simanum,” JAOS 95 (1975) 716.

that the author of the SKL, knowingly or unknowingly, replaced the original Sumerian

hereditary system of seniority and lateral succession with a system of primogeniture and

lineal succession that the rulers of his own time and culture aimed at implementing, resulting

in a simple, repetitive, pattern [PN son of PN son of PN] in our text.

Some historical as well as quasi-historical sources are available for the study of the

ruling family of Ur, but no sources other than the administrative corpus exist which

document the lives, careers and offices of the members of the rural elites. These sources offer

a unique opportunity to use exclusively the primary contemporary records describing social

phenomena. As a result some of the dangers inherent in relying on sources meant as displays

of idealism, faith, and propaganda are eliminated. Although a few of the historical and quasi-

historical documents may indeed hold some historical truth these documents contains little

information on issues such as ethnicity, tribal affiliations, or other vital sociographical

questions. Further, the information in these documents is weak evidence at best without the

support of reliable historical sources, such as the contemporary administrative documents.

The administrative records, on the other hand, represents true reliable first-hand

documentation of the daily lives of the members of Ur III society. These texts have, however,

traditionally been used for the reconstruction of the Sumerian vocabulary rather than for

uncovering sociological information, such as the patterns of succession.

17.

For a critique of the historicity ofthe SKL see C. Wilcke, “Zum Königtum in der Ur III-Zeit,” in P.

Garelli ed., Le palais et la royautée (= Rencontre 19: Paris 1971/75) 180 + fn. 66. See also E. Sollberger,

“Sur la chronologie des rois d'Ur et quelques problemes connexes” AfO 17 (1954-1956) 10-48.

Early in the Ur III period, the formerly independent city-states of Sumer were

subjugated and made provinces of the Ur III empire. Military presence ensured permanent

control over these areas. There is no indication that the supporters of the crown in the old

cities were awarded estates; rather, the local elite families remained the chief administrators.

The Ur III state was therefore not feudal and there was no landed “knight-hood;” instead,

the king relied on a sophisticated administrative system aimed at exploiting the agricultural

production of the core of the state.

The political center of the Ur III state was the city of

Ur, presumably the main residence of the court and a religious center as well, but Nippur

retained a great deal of its former influence as the religious capital of Sumer and Akkad.

During the reign of

∑ulgi

an administrative center was erected close to Nippur, known to us

by its modern name Drehem, and its ancient name

PuzriÒ-Dagæn

. It is uncertain whether

18.

We only know of one Ur III governor who was relocated, the governor of Assur, Zariqum. He is

speculated to have been identical with Zariqum, the later governor of Susa. His case, which was first

discussed by W. Hallo in 1956 (W. Hallo, “Zariqum,” JNES 15 (1956) 220 - 225.); for a more recent

treatment, see B. Foster, “Management and Administration in the Sargonic Period,” in M. Liverani (ed.),

Akkad, the First World Empire. Structure, Ideology, Traditions

(= HANE/S - V; Padua 1993) 28 + fn. 27,

and P. Steinkeller, “The Administration and Economic Organization of the Ur III State: The Core and

the Periphery,” in McG. Gibson & R. Biggs (eds.),

The Organization of Power

, (= SAOC 46: Chicago

1987) 32, citing the same literature as B. Foster. It is likely that the case of Zariqum was both unique

and indicative of the state of affairs in the periphery, and that the ruling families of the old Sumerian

cities, in most cases, were very capable of adapting to changes, and thus remained in office even through

difficult times. The old elite family of Nippur may have been ousted by Amar-Suen as part of a

succession-struggle during his reign; however, that family was perhaps reinstated by

∑u-Suen

, this was

suggested already by W. Hallo, "The House of Ur-Meme," JNES 31 (1972) 94. New evidence from the

so-called

GarÒana

archive (kept at Cornell University, to be published by D. Owen, et. al.) may yet

change our views, since it seems as if Gar

ana, located in Sumer, was indeed a (semi)-private estate (D.

Owen, RAI 47, paper (2001)).

there were more such administrative centers. The peripheral areas, which were conquered by

the military, were subject to tribute, presumably organized through the military presence.

In addition, the king of Ur spun a web of political alliances that covered the entire Syro-

Mesopotamian area and reached well into Iran. Numerous references to the presence of

foreign envoys in the Sumerian heartlands, coupled with the extensive use of dynastic or

political marriages, suggests that a developed diplomatic machinery existed. The king of Ur

also bestowed titles on his semi-independent vassals; as is not uncommon in history, such a

title could later become the title of royalty after that vassal-state had regained its

independence.

The relations between provincial elite families and the royal clan is poorly

understood. The original thesis of this study, that the ruling family mimicked the royal

family, and that they perhaps belonged to an old elite family centered for generations at

Umma, is only partly supported by the data. Certain features of the succession to the highest

offices of the Umma administration suggests an entirely different scenario, namely one in

19.

The system of exploitation referred to here is the so-called bala system. Classically, bala is translated as a

term of office, and this might also be an acceptable interpretation in this case. Each province was said to

be in the period of the bala every year for a specific period of time. Presumably this meant being

responsible for the maintenance of the collective fund of the state, through deliveries to Drehem and

other administrative centers for later redistribution. The periphery paid what was called gun mada,

“tribute of the land,” a kind of tribute or tax delivered by the military stationed there, undoubtedly

calculated on a basis of how much they could collect from the land they controlled. See P. Steinkeller,

“The Core and the Periphery” (1987), which deals extensively with the system of redistribution and

taxation in the periphery during Ur III.

20.

P. Steinkeller, “The Core and the Periphery” (1987) 25 - 26.

which

∑

ulgi may have instrumented a “take-over” by a powerful clan of imperial

administrators in his 33

year.

Figure 1: The extent of the Ur III empire.

21.

Based in part, on P. Steinkeller, “the Core and the Periphery” (1987) 38 figure 6.

Baghdad

Basra

Isfahan

Khabur

Lower

Zab

Upper

Zab

Euphrates

Uruk

Girsu

Nippur

Kish

Babylon

Asmar

Lagash

Umma

Susa

Drehem

Adab

Mari

Tell Brak

Assur

Nimrud

Nineveh

Nuzi

Ancient settlements

Ancient shore

Modern towns

Lake Van

Teheran

Tigris

Karun

Diyala

IRAQ

IRAN

KUWAIT

B A B Y L O N I A

Euphrates

Tigris

Lake Urnia

Kerkha

100

200

300 km

E L A M

Î u ¬ n u r i

∑ i m a Ò k i ?

Z a b Ò a l i

U r b i l u m

S i m a n u m

S i m u r r u m

S a b u m

A d a m d u n

K a r a ¬ a r

A r r a p ¬ u m

U r u a

P a Ò i m e

T h e c o r e

T h e p e r i p h e r y

T h e v a s s a l c o u n t r i e s

C h a p t e r 2 . T h e s o u rc e s , t h e t e c h n i c a l t e r m s a n d t h e

s t r u c t u re o f t h e d o c u m e n t s

Neo-Sumerian administrative documents, which understandably contain a rich array

of technical terminology, can be read only when their inner structure is understood. In other

words, the technical terms will often seem meaningless when the structure of the documents

is unclear. It is deceptive to translate such loaded terms literally, as is most often the case

when translating any ancient language.

In essence, there is no structural difference between a document recording a specific

transaction in the agricultural sector or in the manufacturing sector of the economy, but

there may exist types of documents within one sector that are not mirrored in the other.

Likewise, documents from any level of the administrative hierarchy appear to conform to the

same format and to use the same terminology. In this section I will try to establish the

internal hierarchy of the texts, using that information to better understand the technical

terminology of the documents.

Essentially, all goods were recorded at least twice, encoded as either “debits” or as

“credits” in the accounts of a person or an institution. On another level, this can be described

as the documentation of the

expectations

and the documentation of the

rate of fulfillment

obligations within a “planned” economy.

The Sumerian word for goods, nig

-gur

, is perhaps derived from Akkadian,

makkºru

The distinction between private and public property in 3rd millennium BC

Mesopotamia is not always clear; nig

-gur

could be used to describe private as well as

public property.

Work was thought to be commensurable with movable property, and it

could be translated into movable goods by means of standardized equivalencies. Unparalleled

in ancient history, work in itself was treated as a commodity which could be transferred,

accumulated, or dissipated. The accountants of the Ur III administration were the first in

recorded history to implement a system of man-days tied in to a set of fixed equivalencies.

The work-load of a labourer had by then been standardized and made into a commodity

which could be divided and transferred.

A sophisticated system of equivalences was used throughout the administration.

Some scholars have erroneously equated this system with a monetary system.

However,

since the equivalences were used in a “planned” economy, centrally organized, in a system

lacking most elements of a market economy,

it seems justified to explain the fixed rates,

used by the administrators to estimate the value of a product or a work, as equivalences.

In the following analysis of the Ur III administrative documents—and in particular

while attempting to decode the internal hierarchy of the documents—it seems appropriate to

22.

MSL 17, 190: nig

-gur

| ma-ak-ku-ru.

23.

W. Heimpel, ”Disposition of Households in Ur III and Mari,” ASJ 19 (1997) 63 - 82.

24.

W. Hallo and J. B. Curtis, “Money and Merchants in Ur III,” HUCA 30 (1959) 103-139.

25.

For a discussion of Mesopotamian socities without a market(-place) in the classical sence (~agora) and

the subsequent social and historical implications, see, above all, K. Polanyi, “Marketless Trading in

Hammurabi’s Time,” in K. Polanyi et. al.:

Trade and Market in the Early Empires, Economies in History

and Theory

(Illinois 1957) 12 - 27. See also the discussion in R. Englund,

Ur-III-Fischerei

(1990) 14 -

18.

Outline of the account:

An account was called nig

-ka

(“things counted”) in Sumerian.

References to

accounts occur with some frequency in the economic records documenting different

administrative activities (e.g., nig

-ka

PN-ta, “from the account concerning PN”).

“draw up an account” and subsequently the technical term “account concerning PN”, is

expressed using the verb ak, “to do,” in its finite state /nig.ka(s) ak-a/ > nig

-ka

-ak.

This

term appears as an explicit, or implied subscript in all accounts.

A product or an

administrative function can take the place of the personal name in this formula. Such non-

26.

Earlier read nig

∑ID

or nig

-kas

. nig

-kas

might be prefered over the short reading nig

-ka

, since the

Akkadian borrowing reads nikkassu. ka

is based on a late Babylonian entry in MSL 14 (Ea A = nâqu, Aa

A = nâqu) Ea Tablet 7 line 6’ (p. 451):

ka-a

∑ID

Òa NIG

.∑ID ni-ka-as-su, and

Reciprocal

Ea line 123 (p. 526):

MIN(= ka-a)

∑ID

ni-ka-as-su. See also

R. Englund,

Ur-III-Fischerei

(1990) 26 + fn. 93.

27.

NATN 805 (from AS 7), rev. 21:

ugu

nig

-ka

dam-gar

¿ga

•-ga

-dam

“It is to be placed on the

“debits” (ugu = “front-side” see below for an explanation of this term) of the account concerning the

trade agents”. OrSP 47-49, 411 (from

∑S

3), obv. 1 - 6:

1(geÒ

) 1(aÒ) 3(barig) 1(ban

) Òe gur /

-am

urudu / ugu

-kal-la / ba-a-gar / nig

-ka

KA-guru

-ka / ba-an-zi

, “61 gur and

190 sila of barley, exchange for copper, has been transfered to the debit of Lukala, it has been booked out

of the account concerning the chief of the granary.” TCL 5, 6037 (from

∑S

6), obv. i 4: nig

-ka

siki-ta

“from the account of wool”. SNAT 504 (from

∑S

5 and

∑S

6), obv. 2: nig

-ka

-ta “from the

account concerning silver” the colophon of the same text reads, rev. 29: nig

-ka

-ak dam-gar

a-ia

“damgar account concerning Lu-

aya” See also the parallel text SNAT 518 (from

∑S

7), an “account

concerning Lu-

aya” (colophon, rev. 6: nig

-ka

-ak lu

a-ia

), see below on Lu-

aya (Chapter 5:

Section 9).

personal accounts may have a colophon reading “silver-account concerning the governor”

(nig

-ka

-ak ku

-babbar ensi

-ka),

“wool-account” (nig

-ka

-ak siki).

Several texts can be

quoted that refer to the process of drawing up an account.

It was presumably the job of the

accountant (

Òa13

-dub-ba) to “draw up” the account. The subscripts on all accounts, the

28.

see also R.

Englund, CDLB 2003:1 §11 and fn. 15 (edition of Erlenmeyer 152), where sag-ba is understood as an

abbreviation of sag-nig

-gur

-ra-kam (/sag-<nig

-gur

-ak-am>-bi-ak/).

86.

See for example YOS 18, 123 (from AS 9), rev. viii 33: ugu

a-kal-la ga

-ga

-dam, and Princeton 1, 126

(no date), without enclitic copula (am

), reverse line 2: ugu

-a ga

-ga

-de

, and finally Princeton 1, 404

(from AS 5 viii), obv. 4: ugu

lugal-e

-ma

-e ga

-ga

-dam.

87.

See for example AUCT 1, 324 (from

∑

43), obv. 3: ugu

KAS

ba-a-gar, BIN 5, 322 (from

∑

32? xii),

rev. 8: ugu

ÒeÒ-a-ni-ka ba-a-gar

, L'uomo 47 (from IS 3), obv. 14: ugu

nu-banda

-gu

GAN

gibil

ba-a-gar.

88.

See also MVN 16, 843 (from AS 8), “it shall not be placed on the ‘ugu’” (rev. 1: ugu

-a nu-ga

-ga

89.

See AHw p. 820

pænu(m)

meaning C 2.

90.

See fn. 82 above.

boxes in the text quoted above (SANTAG 6, 20, reverse line 1) we find a particular box

destined for the texts termed “tablets transferred to the debits” (dub ugu

-a ga

-ra).

The sag-nig

-gur

-ra-kam can begin with a remainder from the previous account

(mostly from the previous year), if such a remainder exists.

A remainder is called si-i

-tum.

si-i

-tum is an Akkadian loan-word in Sumerian, presumably borrowed during the Old

Akkadian period.

The “remainder” (si-i

-tum) is always expressed in the medium of

equivalence of the particular type of account in which it appears.

The “remainder” (si-i

-tum) is the “deficit” (Sumerian la

-ia

) from a previous

account carried forward to the next. It has proven possible to demonstrate that the

“remainder” (si-i

-tum) is the “deficit” (la

-ia

) carried forward, using the many cases where

accounts can be strung together forming a sequence of running accounts.

If a “remainder”

(si-i

-tum) is present, it will always occupy the first slot of the sag-nig

-gur

-ra-kam. A

91.

The accounting period was normally one year; irregularities could appear either due to a change of power

in Ur (an external reason), or due to the early settlement of a deficit, either in the case of the death of

“the accounted”, or (presumably) following his wish or ability to settle a large deficit (an internal reason).

Several examples of accounts running for several years, constantly carrying foreward a negative remainder

can be found from the Umma sources; see in that regard D. Snell,

Ledgers and Prices

(= YNER 8; Yale

1982) 103 - 108.

92.

See I. Gelb,

Glossary of Old Akkadian

(= MAD 3; Chicago 1957) 262.

93.

An exception is for example TCL 5, 6036, where no equivalence was used for raw-materials such as

wood-beams that also make up the “remainder”.

94.

See R. Englund,

Ur-III-Fischerei

(1990) 33 - 48, Englund speaks about “hard” and “soft” evidence for

the si-i

-tum being the la

-ia

carried foreward; as “hard” evidence he counts the sequence of damgar

accounts established by D. Snell,

Ledgers

(1982) 34.

number of documents further specify the “remainder” (si-i

-tum) by calling it the “remainder

(of the account) of the Y(ear) N(ame)” (si-i

-tum (nig

-ka

-ak) mu YN),

thus further

clarifying the administrative nature of the si-i

-tum, as the remainder of the account (from

the previous accounting period). The first basket in the receipt quoted above (SANTAG 6,

20) was destined to hold the “deficit” (la

-ia

) tablets, presumably so that these could be

entered into the accounts of the following accounting period. We know that a “la

-ia

” had

to be fulfilled (la

-ia

su-ga), and it is therefore hard to understand it as other than an arrear.

The sag-nig

-gur

-ra-kam was typically made up of stock goods or the work-days of

a regular work crew. In the case of an account concerning the trade agents, or any other

account concerned with staple goods, such as the barley accounts, the first section consists

mostly of durable goods. These goods were transferred to that section by officials of central

households or workshops. In the accounts concerning work, the first section commonly

consisted of the work-days of a permanent work-crew with various additions. Equivalences

95.

See for example the wool-account of the governor, mentioned on pp. 201 - 203: AAICAB 1, 1924-666

(from AS 3), obv. i 2: si-i

-tum mu

amar-

suen lugal-e ur-bi

-lum

mu-

ul. See also Erlenmeyer 152,

obv. i 2: si-i

-tum mu

suen lugal, and finally the “damgar”-account concerning Ur-Dumuzida from

AS 4, Ledgers pl. 4, 3, obv. i 2: si-i

-tum mu gu-za

en-lil

-la

ba-dim

. An account recording a

substantial deficit (more than 11 mana of silver) of Ur-Dumuzida, the trade-agent, from the year AS 3

has recently been published (SANTAG 6, 119 [from AS 3 x]). Since the “remainder” recorded in Ledgers

pl. 4, 3, is less than one-third of a mana, we can speculate that Ur-Dumuzida in the meanwhile had dealt

with his deficit. This is in part proven by the account concerning Ur-Dumuzida from AS 4 i, STA 22,

which reduces a deficit of 2 1/3 mana to the same deficit later entered into the year-end account Ledgers

pl. 4, 3. Several other records deal with the activities of Ur-Dumuzida around this year. See also H.

Neumann, “Ur-Dumuzida and Ur-Dun,” (Istanbul 1999) 48 - 50.

were used throughout the neo-Sumerian accounting system in order to compute totals. We

find the total immediately preceding the header, sag-nig

-gur

-ra-kam. No media of

equivalence was used when the account included very distinct materials such as wood.

Since the sag-nig

-gur

-ra-kam section consisted of the “remainder” (si-i

-tum)

carried forward from the previous account, and the staple goods or work-days of a permanent

work-crew transferred to the account (ugu

-a ba-a-gar) by state officials, it is justified to

propose that the sag-nig

-gur

-ra-kam represented the “debits” of “the accounted”.

96.

TCL 5, 6036 (AS 4), treated below pp. 83 - 86, see also AAS 135 (AS 3 i to xii), where wool of varying

qualities are listed separately.

97.

B. Foster wrote (dealing with the Old Akkadian accounts) (B. Foster,

Umma in the Sargonic Period

Memoirs of the Connecticut Academy of Arts and Science, Vol. 20; Connecticut 1982)11): “In its fullest

form, a balanced account (níg-kas

ak) consisted of an amount on hand (sag níg-GA-ra), disbursements

(zi-ga), and amounts due or carried over(là(l)›u).” See for example the texts MCS 9, 236; Nik 2, 76 and

SP 22, 55, for examples of balanced accounts from the Old Akkadian period. D. Snell believed that the

first section of the large accounts (which he called balanced) was the credit, or as he writes “capital,” of

“the accounted”. He translated the term sag-nig

-gur

-ra-kam as “the head of what has been deposited”,

and based on the similarity of this translation, and the Latin root for capital, he gave another translation:

“It is the capital.” (D. Snell,

Ledgers

(1982) 25). M. Powell, understood the sag-nig

-gur

-ra-kam as

being the property of the merchant, and translated the term as “that which is over and above the basic

makkºru

.” (M. Powell, “Sumerian Merchants and the Problem of Profit,” IRAQ 39 (= Rencontre 23

Birmingham) (1977) 23-29, especially 27). In favor of the viewpoint expressed in this study see R.

Englund,

Ur-III-Fischerei

(1990) 16-17.

In this connection it may be advantageous to consider the following short account

BIN 5, 114 (from

∑

47), in which the remainder is not called the “deficit” (la

-ia

) but rather

the “remainder” (si-i

-tum) of “the accounted”:

1. 4(geÒ

’

u) 6(geÒ

) 2(u) 9(aÒ) 4(barig)

2,789 gur and 4 barig of barley,

Òe gur

(is the

“

debits

”

2. Òa

-bi-ta

Therefrom,

3) 1(geÒ

’

u) 9(geÒ

) 1(ban

)

1,140 gur 1 ban 5 sila barley rations,

5(diÒ) sila

Òe-ba gur

4) 2(geÒ

) zi

-ba-ba] gur

120 gur flour rations,

5) mu en

nanna maÒ-e i

-pad3

(in the)

year: “the En-priest of

Nanna was choosen”.

7. 1(geÒ

’

u) 4(geÒ

) 3(u) 3(aÒ) 3(barig)

873 gur 3 barig 2 ban of barley rations,

2(ban

) Òe-ba gur

8. mu si-mu-ru-um

lu-lu-bu-um

(in the) year:

“

Simurum and Lulubum

a-ra

1(u) la

1(diÒ)-kam ba-

were destroyed for the 9

time

”

9. 2(geÒ

) 6(aÒ) 3(barig) Òe gur

126 gur and 3 barig of barley,

Reverse.

1. kiÒib lugal-uÒur

(LAL

.TUG

)-ra

Sealed by Lugal-uÒura,

2. mu ki-maÒ

ba-

(in the) year:

“

KimaÒ was destroyed

”

3. ∑U+

NIGIN

•

3(geÒ

’

u)] ¿5(geÒ

)• 4(u) Total: 2,140 gur 1 barig 1 ban

1(barig) 1(ban

) 5(diÒ) sila

Òe gur

5 sila of barley.

4. ∑U+NIGIN

2(geÒ

) dabin gur

Total: 120 gur flour.

5. zi-ga-am

is torn out.

6. 8(geÒ

) 4(u) 9(aÒ) 2(barig) 2!(ban

)

529 gur 2 barig, 4 ban and 5 sila barley,

5(diÒ) sila

Òe gur

7. si-i

-tum

(is the)

“remainder”

8. lu

-dingir-ra Òabra

(of the account concerning) Lu-dingira,

the chief administrator.

9. mu-us

-sa ki-maÒ

ba-

Year:

“

after KimaÒ was destroyed

”

In this text the negative balance is called the “remainder”, and not the “arrear” (la

), again reinforcing the position that the first section of the account represented the

“debits” of “the accounted”, and, subsequently that the second section represented his

“credits.”

The second section of an account was called

-bi-ta ... zi-ga-am

(literally “... is

torn out from its middle”). It represented the “credits” of the account.

Òa-bi-ta

introduced

the section and zi-ga-am

terminated it, following immediately upon the total of the section.

zi(g)-a is understood as “to book out”, its Akkadian equivalent

elû

, “to raise,” has in CAD

the meaning “to debit” said particularly about accounts. Numerous Drehem texts have the

subscript ki PN-ta zi-ga-am

, which is interpreted as “booked out from the account of PN.”

However, zi.g can also be equivalent to Akakdian

nasæ

, “to tear out” which may, indeed

make better sense in this respect, since we see, literally, that the goods are torn out of the

middle of the “debits”

98.

The calculations are:

46.29;4,0 ..

-19.00;0,1 5 sila

(-> zi

summarized as dabin) -02.00;0,0 ..

-14.33;3,2 ..

-02.06;3,0 ..

= 08.49;2,2 5 sila

The copy has 8(

geÒ

) 4(u) 9(

aÒ

) 2(barig) 4(ban

) 5(

diÒ

) sila

, this is probably an error for 2(ban

SNAT 395 (from AS 8) is an example of a text recording only the “credits” of an

account. The colophon reads (rev. 13): nig

-ka

-bi nu-ak zi-ga-a

ba-DU, “its account has

not (yet) been made, it has been brought to the ‘credits’.” However, the text is like any other

credit section of a large account; I have even been able to locate one of the receipts used to

draw up the account, MVN 14, 369, (from AS 8 v), which relates to obverse lines 3 and 6 of

SNAT 395.

The “credits” section recorded the actual deliveries, or work-performance of the

work-crew, as these were recorded in the primary documents.

The section was split in two

when the transactions recorded involved a complex production. The second half of the “split-

credits” was used to calculate the equivalencies of the manufactured products. The first half

recounted only the individual receipts. By “booking out” and meeting his requirements, the

agent of the state took part in a “planned” economy. Most deliveries and work-tasks followed

“standing orders.”

100

99.

V. V. Struve, who was the first to realize that the primary documents were summarized in the large

accounts (V. Struve, “Some new data” [1969] 156 - 157, showing the relationship between BIN 5, 262

and BIN 5, 272, obv. vi 17-22), was among the few scholars during the first three quarters of the 20

century who understood the “planned economy” nature of neo-Sumerian accounts. I draw attention in

particular to the tratment of the account TCL 5, 6036 by B. Landsberger,

The Date Palm and its By-

products according to the Cuneiform Sources

(= Archiv für Orientforschung 6; 1967) 7 - 10, as an example

of the misunderstanding by even the leading Sumerologists of the structure of the neo-Sumerian

account.

100. See also D. Snell,

Ledgers

(1982) 96 - 99 + Appendix 2 (270 - 278).

The third and last section of the account, which has already been mentioned in our

discussion of the typology of the accounts, may hold the “operating balance” of the account.

An “operating balance” occurred if the total value of the first and the second section of the

account were not of equal size. If the total value of the first section, the sag-nig

-gur

-ra-

kam, or the “debits” of “the accounted”, exceeded the total value of the second section, the

-bi-ta ... zi-ga, or the “credits” of “the accounted”, a negative “operating balance” would be

recorded. A negative “operating balance”, or a deficit (Sumerian la

-ia

) could as we have

seen be transferred to the “debits” section of the following account. Eventually, “deficits”

could be collected by the state, and we have abundant evidence for the fulfillment, or

redemption, of “deficits” (la

-ia

si-ga or su(-ga)). If the total value of the second section of

the account, on the other hand, exceeded the total value of the first section, then a positive

“operating balance” would be recorded. A positive “operating balance”, or a surplus,

(Sumerian diri(g)) could be transferred to the “credits” section of the following account.

More often than not, however, “the accounted” would presumably treat this surplus as his

private property and it would not appear in any written records. If “the accounted” met the

requirements of the “debits” section, and the total value of each of the two sections were of

equal size, no “operating balance” would be recorded. Since the primary documents testify to

the production related nature of the accounts, there is little evidence to favor a classification

of the neo-Sumerian account as a balanced account. There was simply no attempt made to

balance the accounts; instead only the “debits” and the “credits” tablets kept in different

baskets during the period of accounting were recorded. At the end of the accounting period

the tablet containers were emptied and their contents arranged chronologically (when dealing

The equivalencies:

Equivalencies—the general term used in ethno-economics as well as historical

economics for sets of (fixed) conversion rates—were widely employed in ancient

Mesopotamia, where they are perhaps best known from the “damgar”-accounts.

102

The

equivalencies, however, were not restricted to the “damgar”-accounts; rather, they were

applied throughout the Ur III administration. The implementation of the equivalencies

seems to be the result of a “planned” economy, and can be studied as a characteristic of state

control. As long as the Ur III state was a well functioning administrative machine, the

equivalency rates remained almost completely fixed. Coinciding with the economic recession

early in the reign of Ibbi-Suen, following the breakaway of the provinces, a serious inflation

in the equivalence rates can be observed.

103

The fractional relationship between products (including work) and the medium of

equivalence was expressed in the following terms: “product: its equivalence is x units of y,”

(e.g., 1(

aÒ

)

Òe

gur / ku

-bi 1(

diÒ

) gin

). The rate at which the product was converted into its

equivalence value is easily computed: following the calculations of the Ur III administrators

we shall express this rate as “how much of the product one unit of the medium of

equivalence can fetch.”

104

102. See R. Halperin, “The Concept of Equivalencies in Economic Anthropology,” Research in Economic

Anthropology, Volume 14 (1993) 255 - 298. See also K. Polanyi, “Marketless Trading” (1957) 20 - 22.

103. See Th. Jacobsen, “The Reign of Ibbisin,” JCS 7 (1953) 36-47, and T. Gomi, “On the Critical

Economic Situation at Ur Early in the Reign of Ibbisin,” JCS 36 (1984) 211-242.

The main reason for hypothesizing that the equivalencies were state-imposed—

acknowledging that there exists no hard evidence in the form of conversion tables

105

—lies in

the fact that most of the equivalencies represented a nearly perfect fractional relationship

between product and equivalence. The fact that these fixed conversion rates were employed

by the administrators with a certain rigidity is thought to be good confirmation of this. The

best known neo-Sumerian medium of equivalence, silver, has from time to time been

understood as money.

106

Subsequently, it has been suggested that the silver equivalencies

recorded the amounts of silver used to purchase the products listed in the accounts

concerning the “trade agents.” Silver was only one among many products, that together with

the concept of man-days, could function as a medium of equivalence. These equivalencies

were applied to the counted objects of the account with a double purpose. First of all, the

104. Note, therefore, that the way D. Snell presented the equivalency rates is misleading (D. Snell,

Ledgers

[1982]). In Snell’s system the equivalency rates become prices, since he translated them into the amount

the basic unit of the good equates to, e.g., one unit of product x, equals y units of the medium of

equivalence (in Snell’s analysis mostly silver). D. Snell discussed what he called the two options faced by

the ancient when trying to express what Snell termed “prices” (p. 116). The “a” option conforming with

the one expressed in this study was rejected by Snell as a viable way of presenting the equivalences, and

the “b” option giving the number of units one “unit” of the product/work could fetch. Snell’s argument

for this solution is hard to follow; he seems to base it on the few instances where equivalence does not

represent a nice fraction, and the few cases where the “product” was equated with more than one unit of

the media of equivalence. Such an argument is hardly viable when considered in light of the many

instances of explicitly stated equivalencies in the texts themselves, all of which were expressed as “how

much one unit of the media of equivalence can fetch.” D. Snell’s terminology implies supply and

demand in a price-making market system, which is without foundation in the extant record.

105. It may be possible to consider the first 11 paragraphs op, e.g., the Laws of Eshnunna such a codified list

of equivalencies (See M. Roth,

Law Collections from Mesopotamia and Asia Minor

(Atlanta 1997) 59 - 60.

106. W. Hallo and J. B. Curtis, HUCA 30 (1959) 105, 108 and 112.

equivalencies were introduced to ease the computation of the valuables recorded in the

account. When adding numerous discrete objects, the equivalencies made the calculation of

the total an easy task. Secondly, supplying each product with a set conversion rate, which

presumably resembled an expected value, the administrators hoped to maximize the

production.

Sometimes the equivalencies were explicitly stated in the text, as for example in BM

106050 (unpubl.), rev. vii 5 - 6: 4(

diÒ

) gin

-sig

uÒ

-a 1(u) 5(

diÒ

)-ta / ku

-bi 1(

diÒ

)

ma-na “4 shekel of red gold, at a rate of 15 (shekel silver for each shekel gold) / its silver is 1

mana”. The text Fs. Jones 216 (from AS 3 i), provides a clue to why the accountants would

specify the equivalence rate of gold; we read, obv. ii 7 - 8: 1(

diÒ

) gin

-sig

diÒ

)-ta /

-bi 7(

diÒ)

gin

“1 shekel of gold, at a rate of 7 (shekels silver for each shekel gold), its

silver is 7 shekels.” The rates for gold varied greatly, corresponding to the quality of the gold,

and the specific quality of the gold could be described by its equivalence value.

107

Another

way to translate the previous example is thus “1 shekel of

seven-shekel

gold”, and to treat this

explicit equivalence as a quality marker, comparable to the modern unit karat.

107. When the equivalence value was not specified, gold was normally exchanged at a rate of seven shekels of

silver, or less, to each shekel of gold. See AUCT 2, 173 (from

∑

40 v), where one shekel of gold is

equivalent to 6 2/3 shekels of silver; MVN 4, 147 (from

∑

45), where the rate is at 6 1/2 shekels of silver

for one shekel gold; and MVN 11, 165 (from

∑

39), where the rate is at 7 shekels of silver for one shekel

of gold. This text mentions a wrist-band (HAR) weighing 1 mana (Rev. l., 1) with an equivalence value

of a little more than 7 mana of silver. AAICAB 1, 1911-483 (from

∑S

6 x), (lines 1 and 2) has a very

high exchange rate of more than 30 shekels of silver for one shekel of gold. Finally, note that in SACT 2,

119 (from AS 5), the exchange rate is 8. See line 1 - 2: igi-4(di

)-gal

-sig

/ ku

-bi 2(di

) gin

“1/4

(shekel) gold, its silver is 2 shekels.”

The accountant who wrote MVN 11, 101, an account concerning

Ur-∑ulpa

’e the

trade-agent (dam-gar

), from

∑

43 xii to

∑

44 i, applied a set of explicit conversion rates to

calculate the equivalencies of all the products in the text.

108

It begins with a “debits” section as follows:

1. 3(u) 9(aÒ) 3(ban

) Òe gur lugal

39 gur 3 ban of barley,

according to the royal gur,

2. ku

1(diÒ) gin

-a 1(aÒ) Òe gur lugal

in 1 shekel of silver, 1 royal gur of barley,

3. ku

-bi 1/2(diÒ) ma-na 9(diÒ) gin

its silver is 1/2 mana 9 shekel,

1(u) 8(diÒ) Òe

18 grains.

The first entry in the “credits” section is:

7. 3(aÒ) esir e

-a gur

3 gur of e

’

a bitumen,

8. ku

1(diÒ) gin

-a 1(barig)

in 1 shekel of silver, 1 barig e

’

a bitumen

esir e

-a lugal-ta

each, according to the royal measure,

9. 1 (gur) 4(barig) 2(ban

) 8(diÒ) sila

1 gur 4 barig 2 ban 8 sila of e

’

a bitumen,

esir e

-a gur 1(barig) 3(ban

)-ta

1 barig 3 ban each,

10. ku

-bi 1/3

Òa!

(diÒ) 1(diÒ) 1/3(diÒ) gin

its silver is 1/3 (mana) 1 1/3 shekel,

4(diÒ) Òe

minus 4 grains.

That is, in other words, 3 gur of e’a bitumen, at a rate of 1 shekel silver for each 60

sila

’

bitumen, and 1 gur and 268 sila of e’a bitumen, (at a rate of 1 shekel of silver) for

108. R. Englund already cited the text OrSP 47-49, 196 (from

∑

37 xi), R. Englund,

Ur-III-Fischerei

(1990)

52 + fn. 177. Several other examples are found in the Girsu text TUT 122 (from

∑

39 to

∑

40), see for

example iv 4' - 7’: 8(

diÒ

) sila

Òim

gig 8(

diÒ

) sila

-ta / ku

-bi 1(

diÒ

) gin

/ 1(ban

) 5(

diÒ

) sila

¿ÒaÌ

•

1(ban

) 5(

diÒ

) sila

-ta / ku

-bi 1(

diÒ

) gin

, “8 sila of

Òim

gig, at a rate of 8 sila each (shekel), its silver is

1 shekel. 15 sila lard, at a rate of 15 sila each (shekel), its silver is 1 shekel.”

each 90 sila (of e’a bitumen).

109

It is interesting here to note how the full equivalency rate is

written in the first case, but the second time only an abbreviated statement is written down.

The entire text (MVN 11, 101) states the equivalence rates in a similar fashion.

In some of the texts published in UET 3, the equivalence rates were stated explicitly.

This may be explained by the fact that all of these texts originate from Ur during the reign of

Ibbi-Suen, and therefore fall within the period of time when we expect to find an inflation in

the equivalency rates. The phenomenon of explicitly stated equivalency rates may well be a

symptom of a degrading economic system.

110

All goods could be furnished with equivalencies like the ones described above. These

rates were, however, as a rule not explicitly stated in the texts. The fact that equivalencies

were almost never written in the primary documents suggests that the equivalencies were not

109. Note that this is written /

ku(g) diÒ gin-a(loc.) aÒ Òe gur lugal

/, literally “in one shekel (is) one gur

barley.” Compare to MVN 11, 102 (from AS 3), where the equivalency rate of

’

bitumen is expressed

in a similar fashion. Further TUT 121 (from AS 9), where 5 shekels of silver fetch 1 gur of

’

bitumen

(see reverse, column 3, lines 5’-6’); that text also uses explicitly stated equivalency rates throughout. See

also HLC 398 (pl. 152) (no date), where equivalency rates are used, and the value of

’

bitumen and

dry(?) bitumen is computed together, rev. iii 2-4: 3(

aÒ

) 2(barig) 3(ban

) esir

-a

¿x•

1(barig) 2(ban

ta / 4(u) 9(

diÒ

) gu

esir

had

1(u)-ta / ku

-bi 1(u) 9!(di

) gin

4(u)! 1/2(di

)

Òe

, “3 gur and 210 sila3 of

’

bitumen, at a rate of 80 sila3, 49 talents of dry(?) bitumen, at a rate of 10 talents, its silver is 19

shekel and 4 1/2 grains” (1150 sila e’a bitumen divided by 80, equals 14 3/8. 49 talents of dry(?)

bitumen divided by 10 equals 4 9/10. The total is 14 3/8 + 4 9/10 or 19 11/40; the total recorded in the

text is off by 0.05 shekel). In MVN 17, 62 (no date), column 2, lines 6’ to 9’, and reverse 4 lines 1 to 4,

e’a bitumen and dry(?) bitumen is calculated using explicit equivalence rates, but not computed together.

110. The texts are UET 3, 1165 (from IS 7 xi), 1198 (from IS 3), 1207 (no date), 1309 (from IS 8 vii), 1377

(from IS 6 viii), 1422 (from IS 8 xi), and 1514 (from IS 3).

prices, but rather state-imposed conversion rates, applied with the sole purpose of aiding the

computation of the rate of fulfilment of the obligations, as expressed when subtracting the

“credits” from the “debits.”

Barley could function as a medium of equivalence, and could also be transferred into

workdays. In the accounts concerning overseers of the mill, barley would be converted into

the work-days of female workers.

111

The work-days of “hirelings” who were added to almost

all work-teams concerned with agricultural work were recorded in the books as an amount of

barley—the conversion was always specified according to the category of the “hireling,” who

normally worked at a rate of 6-7 liters per day.

112

Work-days could function as a medium of equivalence, applied in a very similar

fashion to barley or silver equivalencies. The amount of work required as a satisfactory

equivalence of one man-day seems to have been rigorously established. Departures from the

standard equivalences were based on extraneous circumstances such as the condition of the

soil.

113

The complex equivalencies used to calculate the teams who harrowed and plowed the

fields, as well as the complex equivalencies used to compute the work-days used in

manufacturing of goods will be described below.

Equivalencies were also used in the administration of animal husbandry. The herders

were responsible for delivering a certain amount of the products from their flocks each year,

111. See R. Englund,

Ur-III-Fischerei

(1990) 81 - 90.

112. K. Maekawa, “Rations, Wages and Economic Trends in the Ur III Period," AoF 16 (1989) 47 - 48, See

also J. Dahl, AoF 29 (2002) 333.

calculated according to the size of the herd. The products of sheep and goats were wool and

meat, while the products from large cattle were primarily dairy products, meat, horns, hides,

and tendons, not to mention the increase in the herd.

Typically, the administrative records bundled all the animals of a shepherd together

and calculated the amount of wool that he was obliged to produce according to the entire

herd. Accordingly it has been impossible to establish whether the accountants used a specific

set of equivalencies when computing the amounts listed in these texts. The products from

cattle, in particular the dairy production and the increase in the herds, is well understood.

The administration used a set of equivalencies, resembling estimates, when planning the

production and growth of each herd of cattle.

114

Other products from cattle, such as hides,

meat, horns, and tendons, also entered the economy, but no equivalency rates between the

herds and these products have been found.

115

113. A man-day was the equivalent of, e.g., a volume of earth to be excavated of the approximate size of 3 m

See, for example, Aegyptus 8, 264 11 (from AS 3), obv. 4 - 7: 5(u) 4(

diÒ

) sar kin sa

ar /

guruÒ

10 gin

ta / a

-bi u

geÒ

) 2(u) 4(

diÒ

) / i

a-du

-ga ba-al-la “an area of 54 (cubic) sar (ca. 972 m

) to be

excavated, each worker (each day) 1/6 (unit of 36 cubic sar, ca. 3 m

), its work is 324 days, excavating

the

Sweet-water-canal

”. The man-day equivalence rate of hoeing was between 72 and 288 m

It seems as

if the area an unskilled state-dependent worker was required to hoe was between 144 and 180 square-

meters. The hirelings (lu

un-ga

), on the other hand, were apparently required to hoe an area

considerably larger, perhaps corresponding to their larger daily provisions (whereas the

guruÒ

generally

received 2 sila of barley each day, the hireling could receive approximately 6).

114. See R. Englund, OrNS 64 (1995) 377-429.

115. See R. Englund, CDLB 2003:1 §18 - §21.

Silver was furnished with an equivalency in rare cases.

116

In these cases silver is likely

to have existed as a commodity with a set value. An explanation should probably be sought

in the system of obligations and deficits and their replacement or fulfillment. It is likely that

an obligation expressed in any specific product could be redeemed by an equivalence of

silver.

117

The equivalencies of the Ur III administration were an efficient means of exploiting

the producing population and the land. It seems that the equivalencies mostly corresponded

to a rate of production which the producers could not achieve, accumulating deficits each

year.

118

R. Englund has described this process, as well as the liability of the debtor in some

detail. It is most likely that the state employee mentioned in the colophon of an account was

liable for the “operating balance”, in extreme cases threatening his own life as well as that of

his family.

119

The text BE 3-1, 13, cited above, serves to support that observation.

The examples of explicit equivalency rates shown here represent only a sample of

exceptional references.

120

The majority of the equivalencies were simply calculated and

116. See, for example, CHEU 51 (no date), obv. 5: 7(

diÒ

) gin

-babbar siki-bi 1(

diÒ

) gu

1(u) ma-na “7

shekel of silver, its wool is 1 talent and 10 mana.”

117. See, for example, CHEU 51, obv. 6:

-sa

-ge na-gada ... etc. ..., rev. 17 - 19: la

-ia

-ta su-ga / sipa-de

ne / mu-DU, “Hesaga, shepherd, etc., replaced from the deficit, the shepherds brought in.”

118. D. Snell, on the other hand, is of the conviction that "From comparison of prices for different volumes

of the same commodity, one can see that the agency was not interested in maximizing return on the

money spent; it was interested only in making sure that the bureau were properly supplied." (D. Snell,

“The Activities of some Merchants of Umma,” Iraq 39 (= Rencontre 23, Birmingham) (1977) 50)

119. See R. Englund,

Ur-III-Fischerei

(1990) 38-48.

The structure of the primary documents:

Whereas the accounts were top-level administrative texts, summarizing all other

documents,

121

the primary documents, the receipts, were the building-blocks of the

administrative machinery. These documents were, of course, the most numerous of all neo-

Sumerian cuneiform tablets. All of the primary documents were summarized in either the

“debits” or the “credits” of the accounts. The format of the receipts, or the primary

documents, can be defined according to their position and function within in the

administrative system.

Primary documents relating to the “damgar” accounts, and any other account using

silver as the equivalency, or concerned with the transfer of raw materials, generally conform

to the format of the so-called “

Òu

ti-a” documents.

The typical “

u ti-a” document takes the following form:

Standard form:

AAS 72 (from ∑ 46):

Translation:

Product

5(diÒ) 5/6(diÒ) gin

-babbar

5 5/6 shekel of silver,

-ta

ki a-kal-la aÒgab-ta

from Ayakala the leather

worker,

da-da-ga

Dadaga,

Òu ba-ti

received.

(blank space)

Date

mu ki-maÒ

ba-

Year:

“

KimaÒ

was destroyed

”

121. Except for texts such as the “tablet container” tags (pisan-dub-ba), and other inventory texts.

Evidently this simple format could not always be followed.

122

Sealed “

Òu

ti-a”

documents mostly carried the seal-impression of the person who received the goods,

123

but

sometimes another party was introduced in the text, specifically being named as the sealing

party. As a rule, the party who it is said in the document sealed the transaction also did so.

124

Notable deviations to this rule, involving members of the ruling family of Umma, are

discussed below. A few texts mention explicitly that a seal other than that of the receiving

party had been used.

125

122. Grammatical variations were rare. Only a few texts use the ergative case-marker on the agent, and the 3

person marker in the verbal prefix chain (/PN-e

Òu

ba-n-ti/ ; the ergative is predominantly expressed

following the genitive case-maker ak /ak-e/ > -ke

, since most personal names were genitive

constructions), see, for example, AUCT 3, 248/249 (from AS 7 xii), rev. 1-2; MVN 21, 162 (from

∑

ii), obv. 5 - 6; and YOS 4, 39 (from

∑S

9 ix), obv. 6-7.

123. Exceptions are, for example, Aleppo 61 (from

∑

34 vii), obv. 5: a-nu-u

-a

Òu

ba-ti, the document is

sealed with the seal of

Ur-∑ara

the son of Lugal-usur, the captain of the (plow-)oxen of

∑ara

. AnOr 1,

256 (from AS 6), obv. 4 - 5: lu

en-lil

-la

Òu

ba-ti, the documant is sealed with the seal of Lugal-si-

NE-[e] son of Lugalsaga. CHEU 86 (from

∑

39 xi), obv. 5: lu

-i

-zu

Òu

ba-ti, the document is sealed

with the seal of Ur-ama, the

iÒib

-priest of Nin-da-

LagaÒ

. In a few cases the delivering party is also the

sealing party, for example, Aleppo 119 (from

∑

33 vii), obv. 3-4: ki ur-

Òara

-ta / ur-zu

Òu

ba-ti, the

document is sealed with the seal of Ur-

∑ara

, and Aleppo 126 (no date), obv. 2-4: ki lugal-ezem-ta /

lugal-a

-zi-da /

Òu

ba-ti, the document is sealed with the seal of Lugal-ezem (Lugal-ezem, scribe, son of

Lugal-Ema

e, chief administrator).

124. See, for example, AAICAB 1, 1911-489 (from

∑

45 xii), rev. 5 - 7: ki lugal-ezem-ta / lu

-eb-gal

u ba-ti /

ib bi

-du

-ga, the document is sealed with the seal of Biduga.

125. In DC 283 (from

∑S

6 iii), it is directly said that PN sealed instead of PN who received, obv. 5 to rev. 9:

ki KA-guru

-ta / ad-da-kal-la /

Òu

ba-ti / mu ad-da-kal-la-

Òe

kiÒib

ur-kin-na-ke

, the document is

sealed with the seal of Ur-kinake, the son of

amati. See also the parallel text ASJ 19, 220 58 (from

∑S

5 xi), where another (?) son of

amati, Ur-balanga, rolled his seal. See also the similar text AUCT 3,

279 (from AS 7 i).

In the accounts, the “

Òu

-ti-a” documents were entered in the format “

kiÒib

PN”

(sealed tablet of PN).

126

The equivalencies were applied to the valuables only when the

receipts were being quoted in the accounts. The one-to-one correspondence between the

primary documents and the accounts, together with the absence of equivalencies from the

primary documents, is the main argument behind the classification of the large accounts as

records of the central organization.

Primary records concerning work traditionally conform to the following format:

Standard form:

MVN 3, 265 (from ∑S 4):

Translation:

n workers

-Òe

2(u) 7(diÒ) guruÒ Òa

-gu

27 workers, ox-drivers, to one

1(diÒ)-Òe

day, (= 27 man-days

of ox-drivers)

work task

ki-su

ar-u

-u

-ka gub-ba

placed at the threshing-floor

of the Sa

aru-u field.

ugula PN

ugula ur-

nin-su

Foreman: Ur-Ninsu,

kiÒib

kiÒib e

-gal-e-si

Sealed by Egalesi.

Date

mu bad

mar-tu ba-du

Year:

“

the wall was built

”

Seal

-gal-e-si

Egalesi,

dub-sar

scribe,

dumu lu

Òara

son of Lu-∑ara,

sag-du

-ka

land-surveyor.

126. See for example R. Englund, “The Year: “Nissen returns joyous from a distant island”,” CDLJ 2003:1,

§14 + figure 4 (edition of Erlenmeyer 152).

The same terminology was used when transferring dead animals

127

and products of

animal-husbandry.

128

A slightly different type of document was used to record the fodder for

animals,

129

rations for humans,

130

as well as offerings for the gods.

131

See for example the short text Aleppo 299 (from

∑S

4 vi):

Obv.

1. 1(barig) Òe-ba lugal

1 barig of barley rations

(according to the) royal (gur).

2. ...]

 ... ]

132

3. nu-banda

the captain of the (plow-)oxen,

4. guru

-amar-ra-ta

From the granary of the E

’

amar,

5. ki gu-du-du-ta

From Gududu,

6. kiÒib lugal-

-gal

133

Sealed by Lugal-

egal.

Rev.

1. kiÒib lu

Òara

-de

When Lu-∑ara

’

s seal has been brought,

127. See for example MVN 4, 81 (from AS 6 vii), one of the many receipts of dead sheep and goats, sealed by

Lukala during his tenure as chief administrator of the household of the governor (see below: Chapter 5:

Section 10).

128. See for example MVN 3, 374 (from

∑

38), recording 60 liters of lard from Pada, sealed by Lugal-ezem.

Further, see also JCS 26, 99 1 (from

∑S

1), recording 2 mana of wool from Ur-Dumuzida, sealed by

Nin

eliya, the wife of Ayakala.

129. See below concerning Ir(mu).

130. See, for example, MVN 4, 47 (from AS 8 x), monthly barley rations from

Lu-∑ulgi(ra)

, sealed by

Damu (reverse line 2 - 4:

Òe

-ba iti-da / ki lu

Òul

-gi-ta / ki

da-¿mu•)

131. MVN 3, 337 (no date).

132. This line presumably originally gave the name of the captain of the (plow-)oxen who received the

rations.

133. Here and in line 2 on the reverse, presumably a copying error for gal

2. kiÒib lugal-

-gal zi-re-dam

the sealing of Lugal-

egal shall be removed.

3. iti Òu-numun-na

Month

“Seeding”.

4. mu-us

-sa si-ma-num

ba-

Year:

“

after Simanum was destroyed

”

(Seal not given)

Also the documents recording the provisions for the household of the governor used

a similar format.

134

Another group of primary documents, most frequently found in the Drehem archive

are the “i

-dab

” texts. The administrative term i

-dab

, “seized”, seems to be related almost

exclusively to live animals, although it was sometimes also used for workers.

See for example SAT 2, 1163

135

(from AS 4 viii):

Obverse.

1. 1(geÒ

) 5(u) 3(diÒ) geme

sag-dub

113

“Head-of-Tablet”

136

female slaves

2. 3(u) 1(diÒ) dumu mi

aÒ

female children

137

3. 1(u) 3(diÒ) dumu nita

13 male children

134. See for example MVN 4, 257 (from

∑S

1 viii 4 to 5), one of the many texts recording the daily deliveries

of beer, and sometimes flour, sealed by the governor.

135. A new duplicate to ASJ 7, 191 / RA 15 61 = RA 24 (1927) 44-45 / Ligabue 33 / YOS 4, 67 previously

discussed by M. Yoshikawa, “A New duplicate of YOS IV, No. 67 // V. Scheil, RA 24/1, No.8c” ASJ 7

(1985) 191 - 192. The texts are almost identical, but the line-numbering varies.

136. See the discussion by M. Yoshikawa, ASJ 7 (1985) 191 - 192. This category of females was presumably a

designation valid only within the administrative machinery, it can be interpreted as “the females

mentioned in the beginning of the document” = the most valuable workers.

137. A translation of dumu-mi

aÒ

) as “one year old girls”, is highly unlikely due to the fact that these

appear before the dumu-nita

. Should that translation be correct this text would fail to be in good accord

with the hierarchical structure of Mesopotamian administrative documents which always list the most

important “objects” first; in the case of humans, high out-put workers first.

4. 1(u) 5(diÒ) dumu mi

15 female children

5. ∑U+NIGIN

2(geÒ

) 5(u) 2(diÒ)

Total: 172

“various”

slaves.

sag-

i!-a!

138

Reverse.

1. nam-ra ak a-ru-a]

Òara

-Òe

]

139

Booty (destined) for the

“arua”

of ∑ara,

2. uru Òa-ri

-ip!

140

u-um-ma

(of the) city of ∑arip

umma.

3. ur-

-si

ensi

umma

-dab

Ur-Lisi, governor of Umma,

took administrative control of.

4. iti Òu-eÒ-Òa

Month ∑ueÒÒa.

5. mu en

nanna ba-

Year:

“

the En-priest of Nanna was installed

”

Quite a few documents have the subscript gaba-ri dub-ba, or “duplicate of the

document.” The text just quoted above is one of the relatively few texts, with one or more

duplicates, that have been recovered from Ur III archives.

141

Neither this text, nor any of the

other known texts with duplicates,

142

use the term “duplicate tablet” (Sumerian gaba-ri dub-

ba). Although I have been able to find only one example of a sealed document and its copy

with the sub-script gaba-ri dub-ba, this pair does prove the theory that gaba-ri dub-ba was a

138. SAT 2: DUB; YOS 4, 67: sag

i-a; Ligabue 33: sag

i-[a]; ASJ 7, 191: sag

i-a; RA 15, 61: sag

i!-a!.

139. YOS 4, 67: no breaks; Ligabue 33: no breaks; ASJ 7, 191: no breaks; RA 15, 61: no breaks.

140. SAT 2: IT; YOS 4, 67:

Òa-ri

-¿ip•-Ìu-um-ma

; Ligabue 33:

Òa-ri

-ip-Ìu-um-ma

;

ASJ 7, 191:

Òa-ri

-¿ip•-Ìu-um-ma

;

RA 15, 61:

Òa-ri

-¿ip•-Ìu-um-ma

141. Discounting of course the relatively few examples of “fakes” made from a “mold”; see M. Hilgert, "Notes

and Observations on the Ur III Tablets from the Oriental Institute" JCS 49 (1997) 45 - 50.

142. The 1985 duplicate list of T. Maeda, published by M. Yoshikawa, ASJ 7 (1985) 191 - 192 can now be

expanded with several new examples.

copy of a sealed document. Both texts are housed in the British Museum. The sealed

document was published by T. Fish as MCS 8, 97, BM 113102, the other text remains

unpublished (BM 108081).

143

Both texts are dated to AS 9 month 3:

MCS 8, 97, BM 113102

BM 108081

Translation

Obverse.

1. 1(diÒ) gu

2(u) 3(diÒ)

1(diÒ) gu

2(u) 3(diÒ)

1 talent and 23 mana of

ma-na siki gir

-gul

ma-na siki gir

-gul

girgul wool,

2. udu-bi 4(u) 5(diÒ)!?

udu-bi 4(u) 6(diÒ)?!

its sheep are 46?!

3. udu ba-ur

sheared sheep.

4. ki an-na-

i-li-bi-ta

ki An-na-

i-li-bi-ta

From Ana

ilibi.

Reverse.

1. kiÒib ensi

Sealed by the governor /

gaba-ri dub ensi

copy of the governor

’

s tablet

(blank space)

2. iti Òe-kar-ra-gal

-la*

iti Òe-kar-ra-gal

-la

Month “Having barley at

the harbour”

3. mu en

nanna*

mu en

nanna kar-zi-da

Year:

“

the En-priest of Nanna

kar*-zi*-da

of Karzida (was choosen)

”

Seal

Column 1

amar-

suen

Amar-Suen,

2. lugal kala-ga

The strong king,

3. lugal uri

-ma

The king of Ur,

4. lugal an-ub-da limmu

-ba

The king of the four corners:

Column 2

1. a-kal-la

Ayakala,

143. The text is included here with the kind permission of M. Molina. It will be published in the near future

by M. Molina together with M. Such-Gutierrez, in the Nisaba series.

Administration of work

Labor was, together with livestock, a very valuable resource of the Ur III state. The

administration of both cattle and workers did not differ substantially. The terminology was

largely the same whether describing rations for persons or fodder for animals, and the same

terminology was used to “seize” live animals as well as humans. Still, there is no consensus

among scholars concerning the social status of the dependent workers of late 3

millennium

Babylonia.

144

The following groups of people were recorded in the Umma documents:

The most populous group, according to state records, were the

guruÒ

and the geme

the male and female dependent workers. Next to these we find the “porters” (ug

-IL

), the

“dumu-gi

,” and the “hirelings” (lu

un-ga

It is possible that the people called

guruÒ

formed a distinct social group. Persons

holding different occupations, skilled as well as unskilled workers, could be summarized as

guruÒ

145

In the bookkeeping system the word

guruÒ

served as a term for a work-day when

144. The discussion on the status of the lower strata of the Ur III society ground to a halt after I. Diakonoff

defined the term slave as it was used by Soviet scholars as being fundamentally different from the use of

the same term by Western scholars (I. Diakonoff, “Slave-Labour vs. Non-Slave Labour: The Problem of

Definition," AOS 68 (= M. Powell (ed.), Labor in the Ancient Near East; 1987) 1 - 3). The term “slave”

is avoided in this discussion, and “dependent worker” is used to cover all categories of workers who

received some, or all, of their subsistence from the state in return for their work. Chattle slavery was of

very limited importance to the economies of 3

millennium BC Mesopotamia, and we will not be

concerned with that here.

used in the phrase “n

guruÒ

diÒ

Òe

(or -kam),” or “n workers for one day,” which is

equivalent to “n work-days.”

146

In the primary documents the real amount of time used to

perform a work-task was often recorded.

147

Most

of the people called guruÒ

had either

Sumerian or Akkadian names, suggesting that they did not make up a distinct ethno-

linguistic group.

148

There is no reason to believe that a major component of the

guruÒ

work-

force consisted of prisoners of war.

The female counterpart of the

guruÒ

was the geme

. A few texts explicitly refer to

female captives (geme

nam-ra-ak)

149

brought to work in the great institutions of southern

Mesopotamia. Coupled with the numerous examples of female workers with non-

Mesopotamian names, it makes sense to suggest that female prisoners of war did enter and

represented an important contribution to the Mesopotamian work-force.

150

145. The term

guruÒ

could be used to describe people with different occupations ranking from the simple

unskilled worker, doing manual labor on the fields, to the skilled workers such as “potters” (see, e.g, VO

8/1, 3 [from

∑ 39 x], obv. 1: 4(u) 9(diÒ) guruÒ baÌar

), “carpenters” (see, for example, Aleppo 148

[

∑ 42 viii], obv. 2: 1(diÒ) guruÒ nagar u

2(diÒ)-Òe

), as well as “cultivators” and other

agricultural staff (see, for example, MVN 21, 187 [from AS 2 ii], obv. 1 - 4: 1(u)

guruÒ engar dumu-

ni(*) / 6(diÒ) guruÒ Òa

-gu

Òe KIN-a(*) / 2(diÒ) guruÒ gu

-e-us

-sa / ugula lugal-gu

-e

146. See already V. Struve, “Some new data” (1969) 127 - 171.

147. See already V. Struve, “Some new data” (1969) 149. See, for example, MVN 15, 20 which was the

primary receipt used to draw the account Erlenmeyer 152 rev. i 9-11 (see now also the edition in R.

Englund, “The Year: “Nissen returns joyous from a distant island”,” CDLJ 2003:1).

148. Abstaining from the debate on ethnicity and language, I claim that naming habits among non-elite

groups in pre-modern societies largely followed ethnic lines.

The following text example from MVN 14, 569 (from

∑S

3 vi), can when compared

with SAT 2, 1163, cited above, and Nik 2, 329, cited below, help to visualize the influx of

captive females in the economic system of the core institutions. The text is a list of provisions

for persons travelling to and from the eastern provinces (see Excursus 2 for more examples of

such texts). The fourth person, Ninmar(ka) the cup-bearer,

151

is instructed to escort the

captives:

Obverse

...

13. *4(ban

) 2(diÒ) sila

kaÒ DU

4 ban, 2 sila of regular beer,

152

Reverse

1. 4(ban

) 2(diÒ) sila

ninda

4 ban, 2 sila of bread,

2. 1/2(diÒ) sila

sum

1/2 sila garlic,

149. nam-ra ak is a verbal noun that might translate as “booty”. See

MSL 16 p. 235 (SIG

.ALAN =

Nabnπtu, Tablet xxvii (=M) lines 148 - 149): nam-r]a

Òal-la-tum / nam-ra ak]-¿a•

MIN Òa

-la-lum. Òallatu(m)

“plunder, booty, captives, prisoners of war”

from Òalælu(m)

“to take

people, animals or possesions as booty”.

See for example the first 4 lines of BIN 3, 532 (from AS 5 i 20):

1. 3(u) la

1(diÒ) gu

29 oxen,

2. 3(u) 1(diÒ) ab

31 cows,

3. nam-ra ak uru

nergal

(is the) booty of the city of Nergal(?)

4. giri

ib-ni-

iÒkur Òagina

via Ibni-Addu, general.

150. I. Gelb, “Prisoners of War in Early Mesopotamia,” JNES 32 (1973) 70-98.

151. The personal name Ninmar does not immediately agree with Sumerian naming practice. The cup-bearer

Ninmar of our text is perhaps identical with the cup-bearer Ninmarka. Ninmarka, however, is a typical

Sumerian personal name. Ninmarka’s office in Umma is not clear to me, although his connection with

the administration of the governor during the reign of

∑u-Suen

is clear.

152. The translation “regular beer” for

kaÒ DU

is chosen from an analogy with

kaÒ sag

, the two often

occur together,

kaÒ sag

always being mentioned first of the two.

3. 1/3(diÒ) sila

6(diÒ) gin

naga

1/3 sila 6 shekels alkali;

nin-mar

sagi

(for) Ninmar(ka), cup-bearer.

5. e

Òara

-Òe

sag nam-ra ak-da gen-na

having gone with the slaves taken as booty,

to the temple household of ∑ara.

The text continues with one more entry, a total, and the date.

Nik 2, 329 (from AS 5 vii), lists the rations, in this text called fodder (

Òa

-gal

153

for 39 captive female workers. All the names are non-Sumerian; some are perhaps Akkadian,

others are definitely neither Akkadian or Sumerian.

154

The last entries named ten children (8

“daughters” (dumu-mi

), and 2 “sons” (dumu-nita

)), receiving half-size rations. Only a

short excerpt of the four-column text is given here:

obverse column 1

1. 4(ban

) zi

2(ban

) kaÒ ga-na-na

4 ban of flour and 2 ban of beer: Ganana.

2. 4(ban

) zi

2(ban

) kaÒ kur-ni-tum

4 ban of flour and 2 ban of beer: Kurnitum.

3. 4(ban

) zi

2(ban

) kaÒ nu-ba-an

4 ban of flour and 2 ban of beer: Nuban.

4. 4(ban

) zi

2(ban

) kaÒ an-na

4 ban of flour and 2 ban of beer: Ana.

....

etc.

....

Reverse column 3

7) 2(ban

) zi

1(ban

) kaÒ a-

u-Òe-ni

2 ban of flour, 1 ban of beer: A

uÒeni.

8) 2(ban

) zi

1(ban

) kaÒ

2 ban of flour, 1 ban of beer:

-eb-URU-ba-tal

Ereb-URU-atal.

153. Compare with OrSP 47-49, 342 (from AS 5 iv to viii), a short text computing some barley summarized

Òa

-gal nam-ra-ak.

154. See also TCL 5, 6039 cited below, p. 77.

(blank space)

Column 4

(blank space)

1. ∑U+NIGIN

4(u) la

1(diÒ) geme

Total: 39 female dependent workers

4(ban

) zi

2(ban

) kaÒ-ta

each: 4 ban flour, 2 ban beer.

2. ∑U+NIGIN

8(diÒ) dumu-mi

Total: 8

“daughters”

2(ban

) zi

1(ban

) kaÒ-ta

each: 2 ban flour and 1 ban beer.

3. ∑U+NIGIN

2(diÒ) dumu-nita

Total: 2

“sons”

2(ban

) zi

1(ban

) kaÒ-ta

each: 2 ban flour, 1 ban beer.

4. zi

-bi 5(aÒ) 4(barig) 2(ban

) gur

Its flour is 5 gur 4 barig 2 ban.

5. kaÒ-bi 2(aÒ) 4(barig) 4(ban

) gur

Its beer is 2 gur 4 barig 4 ban.

6. Òa

-gal geme

nam-ra-a ak

It is the fodder for the captive

female dependent workers

(blank space)

7. iti min-eÒ

Month mineÒ

8. mu en-unu

-gal

inanna ba-

Year;

“

Enunugal-Inanna was installed

”

(blank line)

In short, we have documentation for the escort of captives from the eastern provinces

(MVN 14, 569), we have records documenting the transfer of female captives to the great

institutions (SAT 2, 1163), and we find them listed in the fodder texts, as staff in these

institutions (Nik 2, 329).

The female dependent workers mainly worked in the mill and the textile factory, but

they could at any time be transferred to perform seasonal agricultural work.

The “porters” (ug

-IL

) were full time workers, sometimes summarized as

guruÒ

the accounts.

155

Whether they made up an independent social group is uncertain, but it is

unlikely that they made up a distinct ethno-linguistic group.

156

The “porters” worked

primarily in the agricultural sector of the economy, but their work did not otherwise seem to

differ from that of the regular unskilled workers.

157

The “porters” seem to be classified below

the regular

guruÒ

-worker, however, when calculating both work-output and the size of

rations

158

—they were as a rule mentioned after the

guruÒ

whenever the two appeared

together.

The “hirelings” (lu

un-ga

) presumably did not belong to a distinct ethno-

linguistic group either, but they may have made up a distinct social group.

159

The “hirelings”

were classified according to the size of the “wages” they were “paid”. The texts speak of 6 sila

“hirelings,” 5 sila “hirelings,” etc. A “wage” of 6 sila per day for each “hireling” seems to be

155. For the reading of UN.IL

as ug

-ga

see for example R. Englund,

Ur-III-Fischerei

(1990) 29 + fn. 103.

There is no clear correlation between the “porters” and the work of carrying. MVN 1, 87 (from AS 1), is

a rare example of such an association; see lines 1 and 2: 1(u) 3(

diÒ

) dusu (

geÒ

) / ug

-ga

Òu

ba-ab-ti,

“the porters received 13 panniers.” As a rule the “porters” were qualified as full-output workers, although

examples of 1/3, 1/2, and 2/3 output “porters” exists.

156. Personal names of “porters” are not different from personal names of other workers within the state

economy. See also R. Englund,

Ur-III-Fischerei

(1990) 29 + fn. 103.

157. Among the hundreds of references to

guruÒ

workers doing the work of ox-drivers (ox-drivers were

perhaps

guruÒ

in Umma), we find a few references to “porters” working as ox-drivers (see ASJ 9, 242 19

cited below), as well as “porters” working as reed-workers (for example, AAICAB 1, 1911-178 (from

∑S

4 v5 ?), obv. 2: ug

-ga

ad-gub

158. MCS 3, 89, BM 111774 (from AS 8 iv), obv. 1 - 2: 3(

diÒ

)

guruÒ

1(barig) / 9(

diÒ

) ug

-ga

4(ban

), “3

guruÒ

receving a (standard) 1 barig ration (each month), 9 “porters” receiving a 4 ban ration (each)

month.”

159. The personal names of “hirelings” are not different from the personal names of other workers within the

state economy, but their work-task seems to be distinct from other groups.

the average.

160

The work of the “hirelings” is differentiated from the work of other categories

of workers.

161

The “hirelings” were normally employed in the agricultural sector.

The “dumu-gi

” were, as a rule, recorded as half output workers, that is, they were

workers from whom only the half work-load was being recorded.

162

Based on the name of

this group, “dumu-gi

”

163

it has been speculated that they were an ethnically distinct group,

specifically, that they were natives, in contrast to the

guruÒ

. This theory rests solely on the

grapheme gi

which in other connections may indicate a territorial affiliation with the

Sumerian core area.

164

However, any judgement of the social standing of the dumu-gi

must

160. See K. Maekawa, AoF 16 (1989) 47-48.

161. See for example, AnOr 7, 300 (no date), where the work of hirelings is contrasted with the work of ox-

drivers. Compare to ArOr 62, 233 I 865 (no date). The ox-drivers (presumably

guruÒ

workers) carried a

higher workload than the “hirelings” (the “hirelings” hoed a maximum area of 24 m

(4 sar) each day,

whereas the ox-drivers could manage to hoe 30 m

(5 sar) each day). See, however, AAICAB 1, 1911-

205 (from

∑S

5), obv. 1:

6(diÒ) guruÒ Ìun-ga

1(diÒ)-Òe

“

6 work-days of hired guruÒ

,”

and 3:

-Ìun-ga

“

work of hirelings.

”

162. It is unclear whether the dumu-gi

worked half time, or whether their work was worth only one half of a

regular worker. See also R. Englund,

Ur-III-Fischerei

(1990) 75 - 76 + fn. 250, 252 & 253.

163. It was originally the theory of F. Kraus,

Sumerer und Akkader. Ein Problem der altmesopotamischen

Geschite.

Mededelingen der Koninklijke Nederlandse Akademie van Wetenschappen, Afd. Letterdunde,

Nieuwe Reeks, Deel 33, No. 8. (Amsterdam/London 1970) 47 - 60, that dumu-gi

may be translated as

“citizen of gi

”, semantically based on an analogy with, e.g., dumu nibru

, “citizen of Nippur.” Kraus

concluded that the dumu-gi

could well be native Sumerians as he suggested that the gi

in dumu-gi

was the same as the gi

in ki-en-gi, as well as in “Sumerian” eme-gi

, etc. See also C. Wilcke, “Zum

Königtum in der Ur III-Zeit,” in P. Garelli ed., Le palais et la royautée (= Rencontre 19: Paris 1971/75),

220 - 230.

be made on the basis of his socio-economic status, coupled with an analysis of the personal

names of the different groups of workers. Such a study is not yet available. The short text

MVN 3, 370 (no date), is a list of workers summarized as either

guruÒ

dumu-gi

guruÒ

-ga

. The ug

-ga

-workers are clearly marked in the body of text as 1(

diÒ

) ug

, whereas

the dumu-gi

are marked as 1(

diÒ

) only. This in itself is peculiar since most references to

dumu-gi

label them as half output workers. Another oddity is that the first person

mentioned in this text, a dumu-gi

, is identical with the foreman mentioned at the very end

of the text.

165

The text lists seven

guruÒ

dumu-gi

, all with names that we understand to be

Sumerian.

166

The text lists ten

guruÒ

-ga

, all but two, possibly three, with names which

can be demonstrated to be Sumerian.

167

This argues against an interpretation of either group

as being a unique ethno-linguistic group.

A much larger investigation is needed to clarify whether the different categories of

workers derived from different social or ethno-linguistic groups. Such a study is planned

within the framework of the CDLI.

164. See now N. Koslova, “(Selbst) ein freier Mann ist nicht gegen die Fronarbeit gefeit...” (forthcoming).

Koslova is of the oppinion that dumu-gi

were freed former slaves, with a special status.

165. It is not unlikely that a dumu-gi

member of a work-crew also acted as foreman of the crew; see for

example the forester work-crews which were made up of small family units, where the oldest member of

the family also held the title foreman. P. Steinkeller, “The Foresters of Umma: Toward a Definition of Ur

III Labor,” AOS 68 (= M. Powell (ed.),

Labor in the Ancient Near East

; 1987) 80.

166. The names of these are in-sa

-sa

, lugal-ba-ta-e

, ur-

ma-mi, lu

-ulu

, ba-zi, lugal-KA, and gur

-za-an.

167. Their names are lugal-ma

-gur

-re, a-da-lal

, ir

, ku-li, ur-

Òara

, ur-

nin-su, and

a-ni-sag

. The

following three personal names cannot positively be asserted to be Sumerian: bu

-uz-gir, mu-da

, e-la-ni.

I would suggest that

the guruÒ

were a social group of people entirely dependent on

the favors of the state. These laborers could fulfill various duties within the state economy. At

the same time the word “

guruÒ

” was used as a technical term for labor projects in texts that

were not necessarily concerned with the work completed the

guruÒ

In the following text, ASJ 9, 242 19,

guruÒ

is used as a generic term for worker or

work-day; whether this was always the case is unclear. The “debits” section of

ASJ 9, 242 19,

a document which can be classified as an erin

-account covering all of the year ∑ 47 and

concerning Lugal-kuzu, the captain of (plow-)oxen, lists workers from four categories

Obverse column 1

1. 1(u) guruÒ engar dumu-ni

10 cultivators, and their sons,

168

2. 6(diÒ) ug

guruÒ

“porters,”

3. 4 guruÒ dumu-gi

4 dumu-gi

169

4. iti 1(u) 3(diÒ)-Òe

For 13 month,

5. a

-bi u

1(Òar

) 1(geÒ

’

u) guruÒ

Its work is 7,800 work-days.

170

1(diÒ)-Òe

6. iti Òe-KIN-ku

-ta

From the month

“

Harvest

”

7. iti diri-Òe

To the

“extra”

month.

8. 2(u) 4(aÒ) 1(barig) 4(ban

)

24 gur 1 barig 4 ban minus 1 sila

1(diÒ) sila

Òe gur

barley,

168. See also TCL 5, 5676 (from

∑S

2), line 2.

169. Note that in the final section of the “credits,” when listing the “free-days” of the disabled workers, four

dumu-gi

workers were recorded as half-out-put. That leads us to believe that line 3 actually covered 8

half-output dumu-gi

workers.

170.

((10+6+4)x30)x13 = 7,800

. Note that the calculation only works if the dumu-gi

workers were workers

from whom a full-output was expected.

9. a

6(diÒ) sila

-ta

“wage”

of each (hireling) 6 sila,

10. a

-bi 2(geÒ

’

u) 1(u) 6(diÒ)

Its work (equivalence) is 1,216

2/3(diÒ)-kam

2/3 work-days.

171

11. 7(aÒ) 1(barig) 5(ban

)

7 gur 1 barig 5 ban 4 sila barley,

4(diÒ) sila

Òe gur

12. a

5(diÒ) sila

-ta

“wage”

of each (hireling) 5 sila,

13. a

-bi u4 7(geÒ

) 2(u)

442 5/6 work-days.

172

2(diÒ) 5/6(diÒ)-kam

...

23. ki ir

-ta

From Ir(mu).

173

...

26. ¿2(Òar) 5(geÒ

’

u)! 1(u)! ¿la2 1/2(diÒ)•

Total: 10,209 1/2 work-days.

guruÒ u4 1(diÒ)-Òe

column 2.

1. sag-nig

-gur

-ra-kam

It is the “debits”

...

The “credits” section recorded the primary documents and calculated a work-day

equivalency for each assignment. The section concerned with harrowing applied the term

erin

as a technical term for team-work:

8. 1(bur

’

u) 1(bur

) 1(eÒe

) GAN

11 bur and 1 eÒe of land,

geÒ]-ur

-ra a-ra

n-kam] 2(iku)

harrowed n times,

1/2(iku) GAN

-ta]

2 1/2 iku each day;

171.

(24x300)+(1x60)+(4x10)-1 = 7,299÷6 = 1,216 1/2.

Note that D. Snell copied 2/3(

diÒ

), which is also

necessary to make the total in line 26.

172.

(7x300)+(1x60)+(5x10) = 2,214÷5 = 442 4/5

The notation 5/6 was an approximation of 4/5.

173. Ir(mu) (see Chapter 5: Section 11) is the most frequent supplier of “barley for wages” in the Umma

accounts.

9. a

erin

-bi u

...]

Its erin

-work is n work-days.

174

10. 1(bur

’

u) 5(bur

) 2(eÒe

) 3(iku)

15 bur, 2 eÒe, and 3 1/2 iku of land,

1/2(iku) ¿GAN

• geÒ]-ur

-ra

harrowed 1+n days,

a-ra

1+n-kam x x x] GAN

...]

n iku each time;

11. a

erin

?-bi u

2(geÒ

)?+

Its erin

-work is 120? 1/3?

1/3(diÒ)? ...]

work]-days.

175

....

As was explained by T. Maeda in 1995,

176

the calculations of the work involved in

harrowing and plowing assumes that three laborers worked together. A suitable translation of

erin

in this context might therefore be “team.”

The last entries immediately preceding the total of the “credits” recorded the “free-

days” for the workers who were unable to perform their prescribed work:

177

10'. 1/2(diÒ) uru-bar-re

1/2 output: Urubare

11'. 1/2(diÒ) ur-

suen

1/2 output: Ur-Suen,

12'. 1/2(diÒ) a-kal-la

1/2 output: Ayakala,

174. Compare with AnOr 1, 166 (from

∑S

3):

Obverse.

5. 4(bur

) GAN

geÒ-ur

-ra

4 bur land harrowed

6. a-ra

3(diÒ) 4(iku) 1/2(iku) GAN

-ta

3 times, 4 1/2 iku each day,

7. a

erin

-na-bi u

2(geÒ

) 2(u) 4(diÒ)-kam

its erin

-work is 144 days.

4 bur or 72 iku, harrowed three times , at a rate of 4 1/2 iku each time is equal to 48 days. When three

workers worked together as a team, that is 144 work-days.

175. The total work-time of the team might have been as high as 1,026 work-days if they harrowed an area of

the same size as the first plot, three times, each day.

176. T. Maeda, “Three Men of a Gang for Ploughing and four Men for Sowing,” ASJ 17 (1995) 333ff.

177. See also V. Struve, “Some new data” (1969) 138 - 139.

13'. 1/2(diÒ) nimgir-di-de

1/2 output: Nimgirdide,

14'. iti 4(diÒ)-Òe

For 4 months,

15'. a

-bi u

4(geÒ

)-kam

Its work is 240 work-days

16'. a

-du

-a dumu-gi

? Òa

-gu

¿x•

(it is) the

“free-days”

of the dumu-gi

ox-drivers.

17'. 6(u) 6(diÒ) guruÒ u4 1(diÒ)-Òe

66 work-days,

18'. a

-du

-a ug

-ga

Òa

-gu

(it is) the

“free-days”

the “porter”

ox-drivers.

...

The output of the dumu-gi

workers was not recorded in the “debits”, although it is

recorded here.

178

Finally the total production in man-days is calculated and the deficit entered in the

books:

1'. ¿∑U•+NIGIN

¿2(Òar) 4(geÒ

’

u)•

Total: 10,079 5/6 work-days

9(geÒ

) 3(u)] ¿9(diÒ) 5/6(diÒ)

guruÒ u

1(diÒ)-Òe

•

2'. zi-ga-am

has been booked out

3'. la

-ia

2(u) 9(diÒ) 2/3(diÒ) guruÒ

(resulting in a) deficit of 29 2/3 work-days.

1(diÒ)-Òe

Reverse, column 8.

(blank space)

1'. nig

-ka

-ak a

erin

-na

erin

-account concerning

2'. lugal-ku

-zu nu-banda

Lugal-kuzu,

3'. mu-us

-sa ki-maÒ

-Ì

Year:

“

after KimaÒ was destroyed

”

178. For the “free-days” of the dumu-gi

see now N. Koslova, “(Selbst) ein freier Mann ist nicht gegen die

Fronarbeit gefeit...” (forthcoming).

It is very likely that the only use of the term

guruÒ

in this text was as a technical term

for work. However, summarizing the cultivators, the ox-drivers, the “porters,” and the dumu-

-workers as

guruÒ

may have the sociological implication that they all belonged to the

same social class, namely that of dependent workers.

179

Note that no erin

-workers as a

distinct group appear in the text.

The group of workers called erin

is perhaps best known from sources from Girsu.

180

However, this class of workers is also attested in Umma. In Umma the erin

seem to be

“soldiers”, levied by the state for multiple purposes.

181

In addition, the word erin

was

widely applied as a technical term calculating the man-days involved in the team-work of

harrowing and plowing.

182

179. See J. Dahl, AoF 29 (2002) 334, for an attempt to show that the value of the prebend allottments given

to the (semi-)independent workers such as potters or cultivators did little to affect their social status or

mobility.

180. See, for example, K. Maekawa, “New Texts on the Collective Labor Service of the Erin-People of Ur III

Girsu,” ASJ 10 (1988) 37 - 94. According to Maekawa the erin

-people “formed the core of the

population in Ur III Girsu” (p. 37).

181. MVN 4, 25 (from AS 7), rev. 11: erin

bala-

-a-me. OrSP 47-49, 466 (no date), rev. 10:

∑

U+NIGIN

<x> egir erin

, a text mentioning policemen and egir erin

together with captains(?). TCL

5, 6166 (no date), rev. iv 6-7:

∑U

+NIGIN

1(di

Ò)

1/3(di

Ò)

ma-na 1/3(di

Ò)

gin

-babbar / erin

umma

-ke

su-su-dam, a text that calculates the wages for the Umma erin

, each receiving 4 grains (

Òe

)

of silver per day. See also TCL 5, 6038 (from AS 7), mentioned below (pp. 75 - 76).

182. Note that SAT 2, 77 (from

∑

33 vi), discussed elswhere in this study mentions as many erin

-workers as

“regular” agricultural workers. The purpose and social make-up of this addition is unclear to me.

Only a few Umma references can be cited in favor of the Umma erin

-people being a

distinct social group working primarily on the land. AnOr 7, 234 (from

∑S

3), mentions the

erin

-troops of Ipa’e, a well known captain of (plow-)oxen, bundling and carrying reed.

183

BCT 2, 29, a possible Umma text (from

∑

46 i), mentions 18 workers receiving 1 barig each

(per month; qualified as the fodder for hired erin

-troops), having gone to the

canal of the

king

184

OrSP 5, 64 26 Wengler 36 (from AS 2), is a field plan where half of the area is

described as the field of the erin

of Umma.

185

Although no conclusive evidence exists, from

the Umma province, indicating that the erin

-people were a distinct social group.

186

Such

evidence would amount to, for example, a substantial number of references to personal

names followed by the designation “erin

”.

187

183. Obv. 1 - 3:

¿+3(geÒ

)•

sa gi / gu-kilib-ba 1(u) 3(

diÒ

) sa-ta / erin

-pa-e

-ke

/ ga

-ga

184. Obv. 1 -2: 2(u) la

diÒ

)

guruÒ

1(barig)

Òe

lugal-ta /

Òa

-gal erin

un-ga

-lugal-

Òe

? gen-na. See

also MVN 15, 131 (from

∑S

2 ix 3), rev. 21:

∑U+NIGIN

1(u) 2(di

Ò)

guru

erin

diri-me.

185. Obv. x: 6(bur’u) 1(bur

) 1(

eÒe

) 3(iku) GAN

erin

umma

. See also SNAT 502 (from

∑S

5 viii), obv.

ii 12 - 13: 8(

geÒ

) 5(u) 1(diÒ)

¿guruÒ•

/ erin

umma

¿ki•

, and rev. iv 7 - 8:

9(geÒ

) 8(diÒ) 1/2 (diÒ)

/ erin

a-pi

-sal

-la

. TCL 5, 6041 (from AS 2), obv. i 15 - 16: 4(

geÒ

’u) 3(

geÒ

) 2(u) erin

umma

ugula e

-a-i

-li

, and rev. vi 1 - 3:

∑U+NIGIN

Òar

) 3(

geÒ

) 2(u) la

(diÒ)

erin

/ erin

gur

-me /

ugula lu

-dingir-ra. YOS 4, 39 (from SS 9 ix), obv. 5: mu erin

-na-

Òe

. YOS 4, 170 (from AS 6 xii to AS

7), rev. 1:

erin

-Òe

Òu bar-¿re•

. YOS 4, 211 (no date), rev. iv 4 - 6: 1(

aÒ-tenû

) GAN

guruÒ

engar

erin

-na / 1c GAN

guruÒ ÒeÒ

-tab-ba / engar erin

-na-me. Amherst 84 (from IS 5 iv), rev. 14:

∑U+NIGIN

1(u) 1(

aÒ-tenû)

guruÒ

erin

1/2

(diÒ)

1(barig)

Òe

-ta.

186. SACT 2, 272 (no date), suggests that the erin

were a social group, but that the members fulfilled

various duties as state-laborers. Rev. 3: 6

(diÒ)

guru

Òa

-gu

erin

-me

The

guruÒ

were given rations from the state, on average two liter of barley each day

(one barig each month). Additionally, they were given allotments of wool and fish, among

other things, on special occasions. The dependent female workers, geme

, were generally

given rations half the size of their male counterparts, children even less.

188

The non-skilled

worker was apparently socially immobile, whereas the skilled workers could climb the social

ladder to the level of foreman (ugula). Such unexpected social mobility is, however, likely to

be restricted to a few cases.

189

The sources suggest that the

guruÒ

worked all year

190

with only a limited number of

free-days each year. In the accounts concerning work, the foremen would be credited with 1/

10 of the work-time of a

guruÒ

, and 1/6 of the work-time of a geme

, as free-days called “a

187. NYPL 388 (from AS 2), obv. 4:

1(aÒ-tenû)

a-kal-la erin

diri-

. AAICAB 1, pl. 36-37, 1911-228

(from

∑S

5 vii ?), obv. i 15:

1(aÒ-tenû)

GAN

lugal-

maÒ

-zu erin

diri ki

Òara

-a-mu ugula-ta. Torino

2, 703 (from

∑

48 vi), obv. 14':

1(aÒ-tenû)

Òara

dumu lu

-du

-ga erin

diri-

See N. Koslova, “(Selbst) ein freier Mann ist nicht gegen die Fronarbeit gefeit...” (forthcoming), for the

suggestion that all workers in the lists who received 1(

aÒ

-c) of land as allottment were erin

-troops.

188. See I. Gelb, “The Ancient Mesopotamian Ration System,” JNES 24 (1965) 230-243.

189. P. Steinkeller, AOS 68 (1987) 73-115, represents one of the few published attempts at understanding a

particular sector of the economy through the familial structures of the people occupied there. The

results, that each unit (forest) seems to have been managed by one small family group (p. 78) is likely to

be mirrored in the agricultural administration; see for example SAT 2, 77 (from

∑

33 vi), described

elswhere in this study.

190. See for example the sequence of accounts: MVN 1, 232 (from AS 7), and MVN 21, 203 (from AS 8),

concerning the work of potters, controlled by Lukala (see Chapter 5: Section 10, below). The

dependency of the

guruÒ

was first described by V. Struve, see references to earlier works (in Russian) in

V. Struve, “Some new data” (1969) 129.

-du

-a.” Whether these numbers really represented the free-time of the workers is not

certain. Even when the worker was unable to work, the overseer would still be credited for

the full work-time of worker but his “free-days” would be deducted and entered into the

“debits” of the account.

It seems reasonable to divide the documents concerning the administration of work

into four groups, the first group being the inventories and other lists. These texts served a

dual purpose; first, as state inventoriy records tracking which institution or team a worker

belonged to, and secondly, the texts were used for the calculations of rations and other

allotments. The second group of texts specifically calculated the output of a work-crew.

Receipts make up the third group. The fourth were the accounts.

§1. Lists which group workers according to category and institution or team are

classified as inventories. These lists can record the production-rate of the workers or his or

her rations or allotments. The most famous such text is HSS 4, 4, which lists the “rations” for

all the high-ranking members of an institution as well as the agricultural workers, but no

non-skilled workers or craftsmen.

191

A term associated with a group of workers in their

relation to an institution is giri

-se

-ga. This term is always used in connection with either a

191. See R. Englund,

Ur-III-Fischerei

(1990) 60 - 63. It is worth noting that the office of the land surveyor

(sa

-sug

) who appears as number three in rank above the chief of the granary (KA-guru

) was not in

the hands of the ruling family in Umma. Whether this office was important on a city level, or only

within the temple household is unclear. The office of Umma surveyor belonged at all times to the family

of Inim-

∑

ara, the majority of Inim-

∑ara

’s sons seem to have been occupied as mid-level agricultural

administrators (with the title nu-banda

, or ugula).

divine name alone, a name of a temple household, the palace or an associated institution.

192

It is likely that gir

-se

-ga represented the permanent staff of a household. The following text

example, TUT 154 (from AS 2 iii), lists the barley rations for the staff of the

New Mill

. Only

the totals are transliterated and translated here.

193

Reverse, column 8

(blank space)

1. ∑U+NIGIN

2(diÒ) dub-sar 1(barig)-ta Total: 2 scribes, 1 barig (of barley) each,

2. ∑U+NIGIN

5(diÒ) guruÒ 5(ban

)-ta

Total: 5 workers, 5 ban (of barley) each,

3. ∑U+NIGIN

2(diÒ) dumu 2(ban

)-ta

Total: 2 assistants,

194

2 ban (of barley) each,

4. i

-du

-me

They are doorkeepers.

5. ∑U+NIGIN

2 guruÒ 1(barig)-ta

Total: 2 workers, 1 barig (of barley) each,

6. ∑U+NIGIN

1(diÒ) guruÒ 5(ban

)

Total: 1 worker 5 ban (of barley),

7. ad-gub

-me

They are basketry workers.

8. ∑U+NIGIN

1(diÒ) nagar 1(barig)

Total: 1 carpenter, 1 barig (of barley),

9. ∑U+NIGIN

4(diÒ) ba

1(barig)-ta

Total: 4 potters, 1 barig (of barley) each,

10. ∑U+NIGIN

2(diÒ)

AR-gul-gul

Total: 2 millstone operators,

195

1(barig)-ta

1 barig (of barley) each,

11. ∑U+NIGIN

4(diÒ) gi-ze

1(barig)-ta

Total: 4 reed-cutters(?)

196

1 barig (of barley) each,

192. For a discussion of the other cultural institutuion the arua see I. Gelb, “The arua Institution,” RA 66

(1972) 1 - 32.

193. Compare to a group of texts studied by A. Uchitel, “Daily work at the Sagdana Millhouse,” ASJ 6 (1984)

75 - 98, However, A. Uchitel did not discuss this text, nor does it belong to the same archive.

194. The assistants were probably children.

195. See A. Uchitel, ASJ 6 (1984) 78. Reads perhaps /nargulgul/.

196. See M. Civil,

The Farmer's Instructions. A Sumerian Agricultural Manual

(= Aula Orientalis Supplementa

5; Barcelona 1994) 122.

12. ∑U+NIGIN

3(diÒ) guruÒ 1(barig)-ta

Total: 3 workers, 1 barig (of barley) each,

13. ∑U+NIGIN

1(diÒ) dumu

Total: 1 assistant, 1 ban 5 sila (of barley),

1(ban

) 5(diÒ)

14. ∑U+NIGIN

1(diÒ) dumu 1(ban

)

Total: 1 assistant, 1 ban (of barley),

15. gu-za-la

-me

They are the guzala.

16. ∑U+NIGIN

8(diÒ) ma

-la

Total: 8 boatmen, 1 barig (of barley) each,

1(barig)-ta

17. ∑U+NIGIN

1(u) guruÒ 1(barig)-ta

Total: 10 workers, 1 barig (of barley) each,

18. ∑U+NIGIN

2(diÒ) guruÒ 1/2(diÒ)

Total: 2 workers, half out-put,

4(ban

)-ta

4 ban (of barley) each,

19. ∑U+NIGIN

1(diÒ) dumu 2(ban

)

Total: 1 assistant, 2 ban (of barley),

20. ∑U+NIGIN

1(diÒ) dumu

Total: 1 assistant, 1 ban 5 sila (of barley),

1(ban

) 5(diÒ)

21. ma

-gid

-me

They are boat-pullers.

22. ∑U+NIGIN

1(diÒ) Òu dumu 5(ban

)

Total: 1

“old child”

, 5 ban (of barley),

23. ∑U+NIGIN

1(diÒ) Òidim 1(barig)

Total: 1 bricklayer, 1 barig (of barley),

24. ∑U+NIGIN

2(aÒ) guruÒ 3(ban

)-ta

Total: 2 workers, 3 ban (of barley) each,

25. ∑U+NIGIN

1(diÒ)

Total: 1 worker, an old man

guruÒ Òu-gi

4(ban

)

4 ban (of barley),

26. Òe-bi 9(aÒ) 2(barig) 3(ban

) gur

Its barley is 9 gur 2 barig and 3 ban.

27. ∑U+NIGIN

1 munu

-mu

nu-¿x•

Total: 1 malster, not receiving?

(blank line)

29. Òe-ba giri

-se

-ga e

kikken

gibil

The barley rations of the staff

of the new mill,

31. iti ezem

-si

In the month

“

Festival of Lisi

”

32. mu-us

-sa

amar-

suen lugal

Year after:

“

Amar-Suen (became) king

”

Each institution, or household, here the mill, was essentially a self-sufficient unit.

The large institutions, such as the households of the major gods or the king, presumably had

much larger staffs. The gir

-se

-ga was not the only source of man-power available to the

institutions; rather, it represented only the permanent staff concerned with maintaining the

basic functions of the household. The Girsu text ASJ 20, 108 7 (from

∑

48), listed

approximately 10,000 men and women who were categorized as the female textile workers

(

geme

uÒ-bar

), the fullers (

-azlag

), and the staff of the textile factory (

gir

-se

-ga e

uÒ-bar

). By far the largest of these three groups were the female textile workers, who were

not part of the gir

-se

-ga. The fragmentary Girsu text TUT 147 recorded the staff of the

governors household; of the household of the “chief administrator” of the governor; of the

local section of the royal household; as well as the staff of the various Girsu temple

households.

197

Although no parallel texts from Umma have been found, we should not

contemplate that such texts did not exist. The tablet container BRM 3, 179, might imply

that the same structure recorded in TUT 147 also applied to the household of the Umma

governor.

198

TCL 5, 6038, is a good example of an inventory of workers. It lists large numbers of

workers grouped according to temple households,

199

and further divides them according to

profession. The colophon states that the text is a record of the erin

-troops levied (e

-e

) for

197. It is likely that the beginning of (obverse) column 2, should be reconstructed (lines 2’ to 5’): “n workers,

(of the) household of the nin-dingir priestess, (which is the) household of Baba, and the various other

households.”

198. BRM 3, 179, is a tablet container holding the “inspection” tablets (gurum

ak) for the staff of the palace

(the Umma branch), the staff of the governor(‘s household), and inspection tablets concerning

shepherds, cowherds and various other persons.

the bala service (

bala-Òe

), in Apisal, in the year AS 7.

200

The personal names are

overwhelmingly Sumerian. The persons listed in the individual entries are, however, not the

names of the persons who were to go out from Apisal to do corvée labor. Rather, these were

the names of the foremen responsible for sending one or two workers for the bala duty of the

province. The reason behind this conclusion is two-fold. First, the fact that some personal

names are preceded by more than one number (

diÒ

) indicates that the numbers do not

qualify the persons alone or their work-category. Second, many of the persons in this text are

known Apisal foremen and herders.

201

199. The sequence of households in TCL 5, 6038, is similar to the sequence of temple households in Apisal

otherwise known from the documents. The persons mentioned until obv. iv 12, are called giri

-se

-ga

Òara

a-pi

-sal

-me; until rev. vi 21: e

nin-ur

-ra; until rev. vi 24: e

Òul

-gi-ra; until rev. vi 26: e

inanna zabala

; until rev. vii 6: giri

-se

-ga e

amar-sag; until rev. viii 2: giri

-se

-ga e

-kas

a-pi

sal

-me; until rev. viii 33: giri

-se

-ga e

-kas

gaba ba-

Òim

-e

-me.

200. Compare to AnOr 1, 88 (from AS 6 iv), a parallel text listing approximately 1,000

guruÒ

workers for

the bala service. See also AnOr 1, 88 (from AS 5), another list of the staff of the major Umma

households sent to the bala service (rev. vii 18’ - 19’:

¿u

?• ¿bala?•-Òe

-e

/ Òa

umma

)

201. The cow-herders listed in column 5 and 6 are known from many other texts; they were always listed in

the same sequence, and they were always associated with the households in Apisal. Some of the cow-

herders associated with the household of Ninura (obv. v 17 - 30) were, however, in several other texts

associated with the household of

∑ara

of Apisal. On the other hand, the three cow-herders also serving

the household of Ninura (obv. vi 3 - 6) were generally known to have been in this relationship. See R.

Englund, CDLB 2003:1 §9. Is it possible that the herders normally mentioned in relation to the

household of

∑ara

could suddenly have been moved to the household of Ninura, or were they added to

the list of the staff of the household of

∑ara

, but mistakenly placed with the staff of the household of

Ninura? Two of the persons called na-gada (rev. vii 9 - 10) sent two “conscripts” for the bala-duty each.

An almost identical list of of shepherds can be found in MVN 1, 80 (from AS 9). In that text, all the

shepherds are associated with Ur-E’e, a high-ranking member of the Umma ruling clan, and a chief cattle

administrator.

The last text to be mentioned in this discussion of the administrative texts of the

lowest level is the inventory TCL 5, 6039 (from AS 5 ii). It records captured dependent

female workers (geme

nam-ra ak) and some male children. The women, who for the most

part have foreign names, were grouped according to their foremen, and their rations were

calculated. The foremen Lugal-Ema

e, Lu-balasag,

∑ara

-zame, Ur-

∑ulpa

’e

202

, Adu, and Ur-

Nintu, are all known Umma officials who were in charge of either the grain mill or the textile

factory.

203

The function of this text is clear: it was used to calculate the amount of barley that

was, or had to be, distributed to the individuals mentioned in the text. It is likely that the

text was related to the household of the governor, since the overseers mentioned in the text

all seem to be connected with that institution.

§2. The texts that calculated the development of a work-crew over time seem to form

one of the building blocks of the administration of labor. Although no exact match has yet

been found of a work-crew list and the permanent work-crew in the “debits” of an account, it

is beyond doubt that the former was used to calculate the latter.

The first entry of AAICAB 1, 1912-1141,

204

recorded the size of the work-crew

during the first month of the year. Two categories of female workers were involved, full-time

202. CST 628 (from

∑

46) suggests a title ugula kikken(?) for Ur-

∑

ulpa’e, the son of Lu(gal)-kuga(ni).

Nakahara 25 (from

∑S

2) records the same Ur-

∑

ulpa’e as in charge of women milling grain. The same

Ur-

∑

ulpa’e also appears in charge of male dependent workers and in charge of production of reed

products and delivering reeds.

203. See also under Ir(mu), Chapter 5: Section 11.

workers receiving 3 ban of barley each month (ca. 1 liter per day), and half-time workers also

receiving 3 ban of barley each month:

Obverse.

1. 2(geÒ

) 3(u) 3(diÒ) ¿geme

• 3(ban

)]

153 female workers, 3 ban (each),

2. 9 geme

¿a

• 1/2 3(ban

)]

9 female workers, half (out-put), 3 ban,

3. iti Òe-KIN-¿ku

•

Month

“Harvest”

The following eleven entries recorded the size of the crew through each of the

following eleven months. The size of the team decreased somewhat during the year, making

it necessary to calculate an average monthly work-crew for the “debits” of the account of the

team.

Reverse.

8. ∑U-NIGIN

3(geÒ'u) 1(u)

Total: 1,816 female workers, 3 ban.

6(diÒ) geme

3(ban

)

9. ∑U+NIGIN

4(u) 8(diÒ) geme

Total: 48 female workers, half, 3 ban.

1/2(diÒ) 3(ban

)

10. geme

3(ban

) igi-1(u) 2(diÒ)-gal

-bi A 12

of the 3 ban female workers is:

2(geÒ

) 3(u) 1(diÒ) 1/3(diÒ)

151 1/3.

11. geme

1/2(diÒ)

A 12

of the half(-time)

204. See the parallel texts UTI 3, 2282 (from

∑S

3 i to xi), recording the “debits” of the work of females

concerning Ur-Nintu, the foreman of the textile factory (rev. 23: sag-nig

-gur

-ra a

geme

ur-

nin-tu

ugula u

-bar), MVN 21, 212 (from

∑

42 i to xiii or AS 6 i to xiii) the “debits” of Ur-Nintu, Nebraska 42

(from AS 7 i to xii) the “debits” of the foreman Ur-Nintu, AAS 35 (from IS 1 i to xii) (name of the

foreman is lost), PSBA 35, 47 (=AAICAB 1, 1912-1141) (from the 1

to the 12

month, year lost)

(name of foreman lost). OrSP 47-49, 401 (from

∑S

1 i to xiii) contains the “debits” of Adu, another

known textile worker foreman (rev. 20: sag-nig

-gur

-ra a-du

¿Òabra•(PA.¿AL• collated from

photo)

), and lastly AnOr 7, 302 (AS 1 i to vi), very fragmentary.

igi-1(u) 2(diÒ)-gal

-bi 9(diÒ)

female workers is: 9.

12. iti 1(u) 2(diÒ)-¿Òe

•

For 12 months,

13. a

-bi u

1(Òar'u) 5(Òar

)

its work is 56,100 (work-)days.

3(geÒ'u) ¿5(geÒ

)•

14. sag-nig

-gur

-ra a

¿geme

•

(It is the) debits (of the account concerning)

the work of female workers

15. lu

-kal-la]

(concerning) Lukala

205

15. mu  ... ]

The year:

The total can be compared to the first entry of the “debits” of, for example, Aegyptus

21, 159 (from AS 8 i to xi):

Obverse, column 1.

1) 1(geÒ

) 4(u)+] ¿1(u)• 6(diÒ)

116 and 49/180

206

female workers

1/3(diÒ) gin

¿geme

• 3(ban

)

(category) 3 ban per month

2) 5(diÒ) 1(u) 6(diÒ) 1/3(diÒ) gin

5 16 1/3 shekel female workers

geme

1/2(diÒ)

(category) half work,

3) ¿iti• 1(u) 1(diÒ)-Òe

for 11 months,

4) a

]-bi u

9(Òar) 5(geÒ

’

u) 9(geÒ

)

its work is 35,940.

207

The decrease in the work-force recorded in this text, presumably caused by the harsh

work-conditions of the mill or the textile-factories, can be easily compared with the

205. See below pp. 231 - 233.

206. 16 1/3 gin

divided by 60 = 49/180, since the gin

is 1/60 of one in the sexagesimal numerical system.

207.

106 49/180 x 30 x 11 = 35,069 5/6

5 49/180 ÷ 2 = 2 229/36 (2 229/360)x30x11 =869 11/12

35,069 5/6

+869 11/12

= 35,939 3/4 approximated to 35,940

information provided by our next text example UTI 3, 1916. This text recorded the work-

capacity of one male worker, named Atu, during the seventh year of Amar-Suen.

Obverse.

1. tu-ra a-tu

Sick: Atu

2. iti 8(diÒ)-Òe

iti Òe-KIN-ku

-ta

For 8 months: from the month

“Harvest”,

3. iti e

-iti-6(diÒ) mu

u-nu-ri

-Òe

to the Month e-iti-6, in the

year: “

nuri”.

5. uÒ

a-tu

Dead: Atu

Reverse.

1. iti e

-iti-6(diÒ) mu

u-nu-ri

-ta

From Month e-iti-6, the

year: “

nuri”.

2. ugula da-du-mu

Foreman: Dadumu

3. kiÒib lugal-e

-ma

-e

Sealed by Lugal-Ema

Seal

1. lugal-e

-ma

-e

Lugal-Ema

2. dub-sar

scribe,

3. dumu lugal-ku

-ga-ni

son of Lugal-kugani

The reason for recording this development is also connected to the accounting

procedures of the “free-days”: an overseer who possessed such a document would be ascribed

full credit for the workdays of that particular worker. The standard “free-days” of the missing

worker would, however, be subtracted, hence it was recorded in the “debits” of the

accounts.

208

We lack evidence for any punishment of a foreman loosing large numbers of his

208. R. Englund, “Administrative Timekeeping in Ancient Mesopotamia,” JESHO 31 (1988) 172 + fn. 46.

See also R. Englund, “Hard Work - Where will it get you? Labor Management in Ur III Mesopotamia,”

JNES 50 (1991) 278. V. Struve, “Some new data” (1969) 138 - 143. V. Struve, who had calculated the

monthly time off for the female dependent workers (the geme) as 5 days a month, or 1/6 of their time,

gave as a reason for these “days of detachment”, as he called them, that they “correspond to the days of

the women’s usual indisposition when, in accordance with the notions of the Sumerians, women were

impure and could not be admitted to any work” (p. 140).

crew, but we are in possession of a substantial number of texts testifying to the many escapees

from the Ur III state, occasionally brought back to the state and enrolled in the production

anew.

209

MVN 14, 2 (from AS 1) is a rare example of an administrative record where the

information is substantiated by a legal formulae:

Obverse.

1. uÒ

1(ban

) ur-ama-na

Dead: One-ban category worker Ur-amana,

2. iti e

-iti-6(diÒ)

from the Month e-iti-6.

3. uÒ

1(ban

) ur-si-gar

Dead: One-ban category worker Ur-Sigar,

4. iti Òe-KIN-ku

-ta

from the

month “Harvest”.

Reverse.

1. ugula ur-

da-ni-ke

Foreman: Ur-Dani,

2. zi lugal in-pa

He swore by the life of the king.

3. kiÒib lugal-ku

-zu

Sealed by Lugal-kuzu.

4. mu

amar-

suen lugal

Year: “Amar-Suen (became) king”.

Seal

209. See for example AnOr 7, 286 (from

∑S

6 viii 21): 5 workers are called seized fugitives (lu

dab

ba), living in the “prison” (

en-nu-ga

¿til-la•

). See also MVN 14, 120 (from

∑

46 iii 11), a parallell

text. MVN 6, 267 (from XX XX xii 22), is an inventory of 12 females described as seized fugitive females

of the mill (geme

kikken

dab

-ba-me). TCTI 2, 3591 (from

∑S

2 i), speaks about a “knight” (

geÒ

tukul

) of the king dispatched to seize fugitives (

zaÌ

-a dab

-de

>/lu zaÌ-a dab-ed-e/, i

im-gen-na > /î-b-ñen-a/,

literally “to seize men who have fled, they went there”). The text is similar

to MVN 14, 569 cited above, and the texts treated in Excursus 2. See also RTC 355; Rochester 152;

DAS 199; TCTI 2, 4185; DAS 197; RTC 397; TCTI 2, 4161; and RTC 354 (XX XX xii), where it is

specifically said that the “officer” (u

-kul) shall go to GN (

saÌar?

-ÎAR-

geÒ

Òinig

-ka-Ò

) to seize

the fugitive females.

1. lugal-ku

-zu

Lugal-kuzu,

2. dub-sar

scribe,

3. dumu *¿ur•-nigar

¿*gar•

ÒuÒ

]

son of Ur-Nigar, the chief cattle

administrator.

210

§3. The actual receipts recording the work have been described above. Receipts

concerning the production of household utensils as well as tools and other manufactured

products document the transfer of products from one institution to another, conforming

with the standards for primary documents outlined above (pp. 49 - 56). These primary

documents did not include the labor that went into the products.

§4. The account. Most of the large accounts concerning labor followed the same

outline described above, with the sole exception being that work was used as an equivalency.

Accounts concerning manufacturing, however, had a split “credits” section. The reason for

this is the way in which the value of the production was calculated. Where as the first part of

the “credits,” as expected, listed the individual receipts, the second part bundled the products

together. Further, each product was split into its smallest component, which could then be

added together to form a grand total. The production time for each product was calculated

in a similar way.

210. See pp. 250 - 255 below for a discussion of the activities of Lugal-kuzu.

TCL 5, 6036, a 20-column tablet among the largest Ur III texts ever uncovered, will

serve as an example of a complex account concerning the manufacturing of utensils for use in

both agriculture and household production.

211

The “debits” section lists the stock of raw-

materials and work-days available to the foreman of the workshop, Agu.

212

The first part of

the “debits” section records a “remainder” from the previous year (obv. i 1 - 22). The

“remainder” consists of large numbers of raw-materials of the kind normally used in

carpentry and basketry workshops.

213

Following the “remainder” the text lists the regular

work-crew of the work shop. Each member of the crew is mentioned with his work capacity.

The crew was made up by two kinds of workers, “porters” and dumu-gi

. Following the

work-crew we find minor deliveries of bitumen and reed, among other things. Immediately

preceding the total of the debit section, we find the “free-days” deducted, corresponding to

the sick and disabled workers recorded in the “credits”.

The total of the raw materials was computed without using equivalencies. Timber

was usually qualified according to its length, indicating that the scarcity of long pieces of

lumber was very real (there would be no reason to record individual pieces of timber

according to length if it was of no value). The first entry of the “credits” section lists the

211. Collated by the author.

212. Agu’s title in this text is “scribe of the artisans” (dub-sar ga

am) See also ITT 3, 4906 (from AS 8), obv. 1

- 4: pisan dub-ba / nig

-ka

-ak / a

am-e-ne / i

-gal

/ ; and MVN 6, 303 (no date), rev. iv 3’ - 5’:

nig

-ka

-ak /

gaÒam

-ke

-ne / ugula a-a-mu

Òabra.

213. Note that this might be an account fulfilling a deficit, since all the timber is recorded in the remainder

from the previous year.

“free-days” of the workers. Following these we find the individual entries corresponding to

the receipts recording the real production of the workshop.

214

A good example of the production of the workshop can be found on obv. iv 22 - 26,

which recorded the delivery of two doors sealed by the governor:

215

22. 4

geÒ

ig e

<esir

su-ba>

Four bitumen-coated doors for a house.

23. gid

-bi 6(diÒ) kuÒ

-ta

their length is 6 cubit each,

24. dagal-bi 3(diÒ) kuÒ

-ta

their width is 3 cubit each,

25. e

nin-ur

-ra-ka ba-a-de

They have been brought to the household of

Ninura.

26. kiÒib ensi

-ka

Sealed by the governor.

As expected this entry was recounted in the second part of the credit section (reverse,

column 2, lines 7 - 15):

216

214. Several of the primary documents have been found in the published record: The first entry in the

“credits” was recorded on BIN 5, 273 = TCL 5, 6036, obv. iv 10 - 26. JCS 2, 187 (NBC 3500) = TCL

5, 6036, obv. v 25-29 and 32-38. JCS 2, 187 (YBC 767) = TCL 5, 6036, obv. vi 15-16. MVN 14, 87 =

TCL 5, 6036, obv. vii 23-29. SAKF 5 line 2 = TCL 5, 6036, obv. v 35. UTI, 2770 = TCL 5, 6036, obv.

vii 11 -19. JCS 28, 121 15 = TCL 5, 6036, obv. vi 12 - 14.

215. Compare to UTI 4, 2870 (from AS 8

sic!

), a receipt recording the delivery of one door for a good house,

among other poducts 1(

diÒ

)

geÒ

ig e

sig

(obverse line 1), received by Dingira from Agu. See also BIN

5, 274, a similar text from

∑S

1, and JCS 2, 186 (= UIOM 712) (from

∑S

3 xii), a paralell text.

216. The work shop produced also two “regular” doors (obv. v 28 - 31 // total: rev. ii 16 - 24), and one special

“bitumen coated

Kagaltum

door for a good house” (obv. v 20 - 26 // total: rev. i 39 - rev. ii 6). Compare

with the primary document Princeton 2, 392 (unpubl.) (no date), a receipt of 4 different doors, one of

which is described as

geÒ

ig kid sig

esir

su-ba ka-gal-la-tum

(obv. 8-9), which is to be transfered

to the mill of Ur-

∑ulpa

’e (

ur-

Òul-pa-e

ba-a-de

), sealed with the nam-

Òa

-tam seal of Ur-

∑ulpa

’e, from Agu.

7. ∑U+NIGIN

geÒ

ig e

esir

su-ba

Altogether: four bitumen-coated doors

for a good house,

8. gid

-bi 6(diÒ) kuÒ

-ta

their length is 6 cubit each,

9. dagal-bi 3!(diÒ) kuÒ

-ta

their width is 3 cubit each,

The final task of the accountant was to credit the foreman of the workshop with the

equivalence, calculated in raw materials and work-days, of the 4 doors (reverse, column 2,

lines 8 - 15):

10. esir

-bi 4(ban

)

its bitumen is 40 liters.

11.

geÒ

-su

7(diÒ) kuÒ

-ta-bi 8(diÒ)

The number of 7 cubit fir-poles is eight

12.

geÒ

Òa

-si 2(diÒ) kuÒ

-ta-bi 2(u) 4(diÒ)

The number of 2 cubit

“cross-beams”

is 24,

13. gi-bi 1(geÒ

) 2(u) sa

its reed is eighty bundles,

14. peÒ-bi 1(geÒ

) 2(u) murgu

its murgu of peÒ is 12,

15. a

-bi 3(u) 2(diÒ)

its work is 32 (days)

This unique reference allows us to reconstruct a door that otherwise would be

completely unknown to us, since it would leave no traces in the archaeological record.

The accountant who drew up the record concerning Agu used a set of equivalencies

specific to the production of baskets and mats to calculate the value of the production. The

equivalencies for reed mats were tied to the surface metrology, and based on the use of reed

and work-days. A worker was expected to weave 1/6 of a

Òar

(ca. 6m

) of reed mat (

kid)

each day, using 6 bundles of reed (sa gi) per 1/6

Òar

217

The other products, such as baskets,

217. See also R. Englund, JESHO 31 (1988) 121-185, in particular fn. 43, and A. Goetze, “Umma texts

Concerning Reed Mats,” JCS 2 (1948) 165 - 202.

C h a p t e r 3 . Pa t t e r n s o f Su c c e s s i o n

Mesopotamian systems of succession have never been dealt with systematically. It has

become a dogmatic statement that strictly enforced patrilineal primogeniture was the

prevailing Mesopotamian system of succession. This is apparent in the way in which the

succession pattern described in the Sumerian King List has been readily adopted by

Assyriologists. The obvious problem, posed by the fact that we have no written evidence that

any rule of succession was formulated by the ancients, paired with the fact that, for example,

the sequence of the Ur III rulers cannot easily be established using the extant administrative

records, should arouse our attention. As soon as an investigation concerning succession

patterns is broadened to cover not only the royal family of Ur III, it becomes apparent that

no obvious pattern exists. In this chapter, I will attempt to outline a theory of succession

applicable for the ruling family of Ur as well as for the provincial elite family in Umma,

which presumably mimicked the royal court. As an excursus, a study of the royal family of

Saudi Arabia has also been appended to this study.

In hereditary succession, power passes from one ruler to another according to a

specific set of rules within that social group. Hereditary succession can be lineal or non-

lineal, that is, it can pass from generation to generation, or it can stay within the same

generations and skip generations more or less randomly.

218

Lineal succession can be either

matrilineal or patrilineal. In a matrilineal system, it is not necessarily that women rule; rather,

218. Lineal succession is by far the most successul of the two, securing a smooth transfer of power.

power passes through the matrilineal line. Patrilineal succession, which is by far the most

common system, can find its expression either in primogeniture or seniority. Primogeniture is

when the oldest son, by virtue of being the oldest son, succeeds his father. Seniority on the

other hand is a system in which senior (male) members of the ruling clan (group) feel that

they have a place in the line of succession. Whether they eventually will be asked to rule is

another issue. Seniority is not always equal to lateral succession, since lateral succession

implies that succession will stay within one generation as long as possible, whereas seniority

only implies that the oldest descendant has a quasi-legal claim to succession.

If a ruler nominates an heir, he naturally propagates rivals as well. This problem is

inherent in any system of succession and can be observed in most cultures.

219

There have

been as many attempts to deal with this schism as there have been cultures adhering to

hereditary succession. How can the ruler secure succession if he can’t trust his nominated

successor?

The aging ruler can approach this problem from three angles. He can choose to wait

and nominate his heir at the last minute, or even make it a posthumous decision, that is, to

have the decision sealed until his death. Delayed appointment has the disadvantage of

complicating succession, since the claim of the heir is more likely to be contested, and he

lacks the relations that years as “heir apparent” would secure.

220

Another possibility is to

indoctrinate the offspring to such a degree that they become completely enervated — this

219. The following owes much to theoretical work of R. Burling,

The Passage of Power, Studies in Political

Succession

(New York & London 1974).

method is also referred to as “caging the heirs”; the disadvantages of this system are obvious:

it leaves the successor at the whim of courtiers and palace staff. Yet another option is to

transfer the responsibilities for the election of the heir to either a particular group within the

ruling house or a group outside of the ruling family. Delegating power to groups other than

the innermost circles of the ruling family is, of course, very dangerous, since it easily becomes

a catalyst for usurpation.

221

As we shall see below, at each succession, power is necessarily

delegated away from the ruler and those closest to him.

Clearly, fraternal rivalries have the strength of promoting “succession of the fittest” [J.

Kechichian (2001) 11], whereas strictly enforced primogeniture has the danger of bringing

utterly incompetent persons to the throne. The designated heir himself has many choices to

make before he can sit calmly on the throne. Will he have to kill all his brothers, or can he

get around this quandary by confining them to imprisonment for life? Can his brothers keep

their positions in the hierarchy? Can they be bypassed in succession? Even more distant

relatives such as uncles, cousins, and widows can become powerful adversaries that have to be

reckoned with when securing succession.

222

220. The choice of an heir could be issued as a will, sealed and only opened after the death of the ruler. In the

Saudi system where seniority prevails, the exact age of a contestant for the throne is often surrounded

with secrecy.

221. Already Hobbes discussed this problem in his typology of the different forms of commonwealth—

Hobbes term for a society. Hobbes claimed that the ruler who did not control succession was not

sovereign, but that the person or institution who controlled succession also was the real sovereign of that

society. (See Th. Hobbes,

Leviathan, or the Matter, Forme and Power of a Commonwealth, Ecclesiastical

and Civil

(Basil Blackwell, Oxford: 1957) 125 - 129.)

It is likely that the empire-builder, the strong man who accumulates almost divine

status by virtue of his own dynamism on the battle field, and at court, can enforce his choice

for succession by means of his charisma alone, and that his wishes will be adhered to for

some time after his death. Such powers are not likely to be inherited, however, and the

subsequent generations will have to deal with the problems of succession on their own.

223

222. Although men have restricted the privilegies of rulership for themselves in the majority of cultures,

women, sometimes, have been able to rule through their juvenile sons, essentially bidding for power by

promoting their own offspring.

223. M. Weber wrote extensively on the topic of charisma and control; see M. Weber,

On Charisma and

Institution Building, Selected Papers, Edited and with an Introduction by S. N. Eisenstadt

(Chicago &

London 1968) for an introduction to Weber’s writings on the topic. Weber wote much less on

succession, and remarkably little on hereditary succession; see pp. 54 - 57 in

Selected Papers

1968 (= M.

Weber, “The Routinization of Charisma” from

Theory of Social and Economic Organization),

where his

views are summarized. Following Weber, it would be possible to argue that the nature of the rule of

∑ulgi

or even that of the first king of Saudi Arabia, Ibn Saud, was neither charismatic nor buraucratic,

since both rulers forged their “empires” from a tribal (or in the case of Ur III, clan-like) composition.

However, it is precisely the fact that they were able to rise above petty tribal issues and create a social unit

larger than their own tribe or clan, by means of their own personality and virility, that is in good

agreement with the system of Weber. M. Weber described charisma as “a certain quality of an individual

personality by virtue of which he is set apart from ordinary men and treated as endowed with

supernatural, superhuman, or at least specifically exceptional powers or qualities.” (p. 48 in

Selected

Papers

1968, originally from p. 329 of M. Weber

Theory of Social and Economic Organization

). The

“routinization” of charisma, and the subsequent vesting of power in the head of the clan rather than in a

single charismatic person as described by Weber, see, e.g., p. 50 and pp. 194 - 197 in

Selected Papers

1968, applies to our data-set as well. There is nothing to suggest that the Ur III state ever evolved into a

true bureaucratic state, as this has been formulated by Weber (pp. 66 - 69 in

Selected Papers

1968). For

more on this particular problem see Chapter 5 below. See also p. 11 in J. Kechichian (2001), and R.

Burling (1974) 87-88.

Since no Mesopotamian text concerning rules of succession has ever been

recovered,

224

and whereas it would appear that, although this has never been proven,

primogeniture was the rule, I suggest that the ancient Mesopotamian ruling families used a

system of seniority when dealing with succession within the ruling families.

225

It is very likely, indeed, that the ruler and his clan would project back in time the

system of succession they favored in order to add legitimacy to their legal claims to the

throne. Thus the SKL may be nothing more than a piece of political rhetoric; although

fraternal succession is not altogether eliminated from this text, it tends to suggest an eternity

of patrilineal succession.

226

224. The ancient Indian Hindu text Arthashastra (attributed to the 4

century BC, although the oldest

extant copy dates to c. 150 AD) can be viewed as such a theoretic work, although it is mostly concerned

with the techniques of rulership, and less with the problems of succession. Much later is Machiavelli’s

The Prince

, and

Discources

, both devoted primarily to instructing a ruler in the techniques of rulership,

and only to a lesser degree with succession. Machiavelli in great detail pointed out the dangers of

hereditary succession and the possibility of bringing an incompetent heir to the throne.

225. P. Moorey, while working on the Kish excavation reports (P. Moorey

, Kish Excavations 1923-1933

(Oxford 1978) 165), suggested, in line with I. Gelb, “Sumerians and Akkadians in their Ethno-

Linguistic Relationship,” Genava NS 8 241ff. (= Rencontre 9: Geneva, 1960) 265-67, that regional

differences between the Akkadian North and the Sumerian South existed. Moorey went further than

Gelb and suggested that the “the precocious emergence there [in the North] of a powerful secular

kingship may have derived more immediately from the exploitation of a tribal system by singularly

forceful individuals in the ruling family or group; a political pattern familiar in more recent Arab

history”. P. Steinkeller, “Early Political Development in Mesopotamia, and the Origins of the Sargonic

Empire,” in M. Liverani (ed.), Akkad. The First Worls Empire. Structure, Ideology, Traditions (=

HANE/S V; Padua 1993) 121, and fn. 37, later quoted P. Moorey, although he disagreed with Moorey’s

statement on the “immediate” tribal background of the rulers of Kish, and suggested a city-based

oligarchy as the source of Kishite kingship.

Although treaties from the first millennium BC Mesopotamia concerning succession

exist such as the so-called Vassal Treaties of Esarhadon,

227

it is nowhere stated that succession

was limited to the oldest son, rather the choice of an heir seems to be somewhat

ad hoc

. The

patterns of succession in the dynasty immediately prior to the Ur III, the second Dynasty of

LagaÒ

, is very difficult to understand, and its use as a comparative source must be

excluded.

228

The dynasties following the fall of Ur III, the Amorites of Babylonia, all seem to

226. Compare with J. Le Patourel, “The Norman Succession, 996-1135,” The English Historical Review, Vol.

86, No. 339. (1971) 225-250, and in particular 226, and fn. 1, where Le Patourel discusses a quote

from Orderic [Ecclesiatical History, iii. 242.] which has been taken as a proof of the existence of

primogeniture at the time of William, Duke of Normandy. Le Patourel argues “that Orderic is very apt

to explain events of the eleventh century according to ideas of the twelfth, particularly in matters of

inheritance.” See also p. 230, J. Le Patourel suggests that the Norman Duke was not bound by any

tradition, or popular “law of inheritance”, and further, that it is very diffucult, due to the absence of

written contemporary sources, to find any such rules. R. H. C. Davis, “William of Jumieges, Robert

Curthose and the Norman Succession,” The English Historical Review, Vol. 95, Issue 376 (1980), 597-

606, especially p. 599 for an assessment that Orderic’s account of the same events actually can be used to

unravel the facts.

227. See S. Parpola and K. Watanabe,

Neo-Assyrian Treaties and Loyalty Oaths

(= SSA II: Helsinki 1988), with

publication history. See M. Liverani, “The Medes at Esarhaddon's Court,” JCS 47(1995) 57-62, for an

alternate view on the nature and purpose of the treaties; the author suggested that the main purpose of

the treaties was to serve as a loyalty oath of the Median mercenaries at the court of Essarhaddon. The

Assyrian title for an heir to the throne was m

ar rabu

a bit red

ti, which translates as “the great son

of the king, the one of the house of succession”.

228. Lateral succession might also have preveiled in Girsu during the days of the second Lagash Dynasty; see

especially J. Renger, “The Daughters of Urbaba: Some thoughts on the Succession to the Throne during

the 2. Dynasty of Lagash,” Fs. Kramer (= AOAT; 1976), 367 - 369, and in particular fn. 17, with

reference to D. Edzard, “Sumerer und Semiten in der Frühen Geschichte Mesopotamiens,” Genava NS 8

241ff. (= Rencontre 9; Geneva 1960) 255 and fn. 107. Edzard believed fraternal succession to be

subordinated primogeniture, however, without eliminating an element of fraternal succession

(“Fratriarchat”) from the history of Sumerian hereditary systems altogether.

favor lineal succession with restricted primogeniture, which might be a peculiarity adopted

from the specific Amorite tribe making up the ruling elites of those dynasties. All through

the 3

millennium, the brothers of a ruler seem to have had a bid at succession, verifying my

suspicion.

229

The problems of any prosopographical analysis rests on the degree to which we trust

the data contained in the sources. Whereas early Assyriological scholarship readily accepted

any information written on a cuneiform tablet, modern studies have, advisedly, improved

their historiography and learned the dangers of relying on court poetry as sources of history.

Whether the administrative records are truly reliable historical documents seems to be the

229. The founder of the the Old Akkadian Empire, Sargon, was followed on the throne by two sons; first

Rimu

, then Mani

u. It is possible that the two struggled against each other for the throne (see Aa.

Westenholz,

The Old Akkadian Period, History and Culture

, in P. Attinger et al. (eds.),

Mesopotamien,

Akkade-Zeit und Ur III Zeit

(= OBO 160/3; Freiburg 1999) 41). Eventually,

ManiÒtusu

was able to

name his own son as successor (i.e. favor his own line), and lineal succession prevailed in the last

generations of the Old Akkadian Empire (note that the SKL agrees on this point, column 6, lines 39 -

41;

ma-ni-iÒ-ti-iÒ-Òu / ÒeÒ gal ri

-mu-¿<<uÒ>>•-uÒ / dumu Òar-ru-ki-in,

“

ManiÒtusu,

big-

brother of

RimuÒ,

son of

Sargon

“). The predominant succession pattern in Old Sumerian

LagaÒ

was

not primogeniture, but rather, again, seniority. Ur-

NanÒe

, the founder of the Lagash I Dynasty was

followed by his son Akurgal, who again was followed by his son, Eanatum, who was succeeded by his

own brother, En-anatum (I). En-anatum (I) was followed by a son, En-metena, who in turn was

followed by a son, En-anatum (II). The next ruler, En-entarzid, was perhaps a brother of En-anatum (II).

The dynasty ended with En-entarzid; his successor, Lugalanda, was presumably not related to

Ur-NanÒe

(see J. Bauer,

Der vorsargonische Abschnitt der mesopotamischen Geschichte

, in P. Attinger et al. (eds.),

Mesopotamien, Späturuk-Zeit und Frühdynastische Zeit

(= OBO 160/1; Freiburg 1998), for the most

recent treatment of Old Sumerian history). For the few pre-Lagash I examples of fraternal succession

known to us see D. Edzard, Genava NS 8, 255 + fn. 107; these are in

KiÒ

MelamkiÒi

and BAR-SAL-

NUN-NA, and in Ur: A’annepada and Meskianganunna.

present-day controversy within Assyriology. Are we capable of uncovering fictitious

transactions in the documents? Can we trust our data?

It seems clear that the basic prosopographical information such as familial

relationships as expressed in the seal-inscriptions can be trusted.

230

The perception that the

term dumu (Sumerian for child, but in an Ur III context almost exclusively meaning son)

can refer to a business associate is borrowed from later Assyrian and Babylonian sources,

231

whereas all third millennium BC Sumerian sources point in the direction that this was a

genealogical term for son. Although adhering to such basic rules, Ur III prosopography

remains complex due to the restricted information provided by the sources. Only the

information necessary for the identification of a person was ever recorded in the

administrative documents of the Ur III bureaucracy. It is therefore always necessary to keep

the familial as well as the official standing of any person in sight when disentangling the

many references to his activities. The use of personal names was not restricted on the basis of

social standing or grouping — static social groups based on ethnicity or self-perceived tribal/

clan affiliation, such as those known from, e.g., India, did not exist, as far as we can see, in

230. Basic familial relations frequently expressed in the Ur III administrative record are dumu (“son,”

although dumu is gender neutral and has the meaning “child,” it is in Ur III administrative documents

normally contrasted to dumu-mi

“female child”),

ÒeÒ

(“brother”), ama (“mother”), and dumu-mi

(“daughter”). Of these only the first two are attested with very high frequency. Father (Sumerian ad-da,

ab-ba, a or a-a, presumably vocalized */aya/) was rarely used to describe a familial relation, as was the case

with the more exotic terms such as e-gi

-a, “daughter in law,” etc.

231. M. Larsen,

The Old Assyrian City-State and its Colonies

(= Mesopotamia 4: Copenhagen 1976) 92 - 102,

seems to be of the opinion that familial affiliation was of paramount importance to the the Old Assyrian

trading companies, allowing for the interpretation of kinship terms as both familial and official.

C h a p t e r 4 . T h e Ho u s e o f Ur - Na m m u

The Ur III dynasty lasted three, perhaps four, generations, with five rulers: Ur-

Nammu and his son

∑ulgi

belonged to the first and second generation, respectively. Most

likely three sons of

∑ulgi

— following each other on the throne — made up the third

generation, they were, Amar-Suen,

∑u-Suen

, and possibly Ibbi-Suen.

232

Ur-Nammu,

233

founder of the Ur III dynasty, is not well known, since few

administrative documents from his time have survived. The primary sources for his reign are

so scarce that even the sequence of year-dates for his reign remains to be finalized.

234

According to the Sumerian King List, Ur-Nammu reigned 18 years. Six years are still

unaccounted for in the administrative record, and it remains impossible to place two known

year-formulae within the sequence. The year-name formulae record peaceful activities only,

such as the excavation of canals.

232. Whereas it is likely that Amar-Suen and

∑u-Suen

were brothers, we have no contemporary information

about the ancestory of Ibbi-Suen. The almost purely hypothetical statement that Ibbi-Suen too was a son

∑ulgi

is spun on a theoretical framework generated, in part, from anthropological and sociological

evidence, and supported by comparative evidence from the Ur III provincial elite families (described

below). New evidence from the

GarÒana

archive (D. Owen, RAI 47, paper (2001)), seems to prove this

theory.

233. For the reading of the theophoric element of Ur-Nammu’s name see D. Frayne, RIME 3/2 (1997) 9,

whith references to the literature suggesting a reading namma instead of nammu. Lately E. Flückiger-

Hawker,

Urnamma of Ur in Sumerian Literary Tradition

, (= OBO 166; Freiburg 1999) 8 - 9, again

promoted a reading Namma. However, following D. Frayne, RIME 3/2 (1997), nammu is retained in

this study, albeit only for the sake of convention.

234. See D. Frayne, RIME 3/2 (1997) 9.

Supported by both indirect as well as direct evidence, Ur-Nammu is considered a

member of the ruling family of Uruk. This seems confirmed by the fact that Ur-Nammu

served as a commander in the Urukite army. The exact nature of Ur-Nammu’s familial ties

with the ruling family of Uruk, and its

paterfamilias

, Utu-

egal, is unfortunately

uncertain.

235

Ur-Nammu brought with him the royal ideology of the kings of Uruk when he

founded his dynasty in Ur. The rulers of Ur would claim to have roles in the classical

GilgameÒ

mythology by being siblings of

GilgameÒ

and, to some extent, by taking over his

role altogether. During the formative period of the empire, the clan of Ur-Nammu,

legitimized through both pseudo-historical as well as theological claims of supremacy,

managed to monopolize power, and exclude all other families and cadet branches from

succession. A cadet branch is defined as a distinct branch of the ruling family, related by

marriage or through an uncle to the ruling line. We see the result of this policy in the later

years of the empire when one family alone held all the important offices in the imperial

administration, co-opting all other important clans in their own rule (see, e.g., the discussion

about the sukkalma

below, pp. 113 - 116).

235. One, fragmentary, historical inscription records the familial relationship between Utu-

egal and Ur-

Nammu: Utu-

egal 6 can be used as evidence that the two were brothers (see C. Wilcke, “Zum

Königtum in der Ur III-Zeit,” (1971/75) 180 + fn. 67). M. Sigrist,

Drehem

(Bethesda 1992) 4, suggests

that Ur-Nammu was the son of Utu-

egal of Uruk. Although more Ur III princes may have served in

the army as military commanders (

Òagina

), this cannot be documented for more than a few. The kings

for whom evidence is at hand (

∑ulgi

and Amar-Suen) seem to have installed “foreigners” as commanders

of the army just as frequently as their sons held these powerful positions. Whether this was out of fear for

the creation of powerful factions inside the royal family is unknown. Ur-Nammu’s relation to the royal

house of Uruk can therefore not be proven by his involvement with the military of that city. He could

have been a blood relative as well as a favored allied married into the royal house of Uruk.

The historical sources pertaining to the reign of Ur-Nammu mostly refer to his

building activities, and to his refurbishing of the canal system. The sources are, on the other

hand, silent concerning any military activities.

236

It is possible that Ur-Nammu only ruled a

small territorial state in the southwestern part of Sumer. Further, it is even likely that his rule

was contemporary with the Second Lagash dynasty.

237

Ur-Nammu was well remembered in later literary tradition. Two very famous texts

have come down to us, “The Coronation of Ur-Nammu” and “The Death of Ur-Nammu.”

The latter suggests that Ur-Nammu died on the battlefield, and that his corpse was not

recovered.

238

236. See W. Sallaberger,

Ur III-Zeit

, in P. Attinger et al. (eds.),

Mesopotamien, Akkade-Zeit und Ur III Zeit

OBO 160/3; Freiburg 1999) 135 - 136, for a summary of the canal constructions of Ur-Nammu, and

137 - 139 for a summary of the building activities of Ur-Nammu. See also E. Flückiger-Hawker,

Urnamma

, 1999, 28 - 40 for a survey of the activities of Ur-Nammu and the sources to his reign. Old

Babylonian copies of Ur-Nammu's historical inscriptions record military campaigns against the Elamites

and the Gutians (see, for example, RIME 3/2 1.1.28 and 29).

237. So far no study has conclusively sorted out the chronology of late third millennium BC. The second

Lagash dynasty, also called The Dynasty of Ur-Baba, has traditionally been considered an intermediate

period between the Old Akkadian dynasty and Ur III, but it is now commonly believed to overlap with

Ur III. Our understanding of the internal chronology of

LagaÒ

II is limited because we have failed even

to establish beyond doubt the sequence of rulers. The external chronology of

LagaÒ

II is not well

understood either, since the foreign relations of the Girsu court are known only from the royal

inscriptions. The expressions of Gudea: “He opened the road from the upper to the lower sea,” are

standardized epithets at most, see Gudea Statue B, column v, lines 25-27, and compare to examples from

literature, for example, A song of Inanna and Dumuzid line 44 (Dumuzid-Inanna D1, see http://www-

etcsl.orient.ox.ac.uk/section4/c40830.htm). For a recent discussion of the Lagash II chronology see C.

Suter,

Gudea’s Temple Building

(Groningen 2000) 15 - 28, and E. Flückiger-Hawker,

Urnamma

(1999) 2

- 5, with references to the debate.

Whereas Ur-Nammu was the founder of the Ur III dynasty his son,

∑ulgi

, was the

builder of the Ur III Empire.

239

He ruled 48 years, and was the most important Ur III ruler.

He was succeeded on the throne by two, or possibly three of his sons. As seen throughout

history, a powerful, long-lived founder of an empire would often be ascribed super-human

powers.

∑ulgi

was no exception. He was deified by the middle of his reign, and there is

nothing that suggests that anyone dared contest his rule or his plans for succession while he

was alive.

240

We do not know the names of any siblings of

∑ulgi

. If there were any, they had

been successfully excluded from the line of succession.

With the elevation in

∑

Liwwir-MittaÒu

, his daughter, to queenship of

MarhaÒi

a political entity to the east of Mesopotamia,

241

the expansionist policy of

∑ulgi

had begun.

This policy, which from around

∑

20 included military campaigns, coincided with the first

attestation of the divine title used by

∑ulgi

242

During

∑ulgi

’s 21

year, certain social

reforms were formulated which seem to have eased the conversion of the population at large

238. The theological implications of a burial without a corpse are thought to have been sinister; however, they

lie beyond the scope of the present study, and will not be dealt with here, see D. Frayne, RIME 3/2

(1997) 20, with references. Note that E. Flückiger-Hawker,

Urnamma

, 1999, see for example p. 7, and

94, understands the text differently; according to her, Ur-Nammu is brought back to Ur where he was

burried.

239.

∑ulgi

’s early years might have been contemporary with at least the latter part of the semi-independent

LagaÒ

II dynasty. See fn. 234 above, and in particular P. Steinkeller, “The date of Gudea and his

dynasty,’ JCS 40 (1988) 51-52.

240. The smooth transfer of power between

∑ulgi

and Amar-Suen suggests that

∑ulgi

had appointed Amar-

Suen “heir apparent” at some point.

241. See P. Michalowski, JAOS 95 (1975) 716-719.

242. D. Frayne, RIME 3/2 (1997) 91.

100

into a state of dependence on the crown. It is from that point in time that we, with reason,

can call Ur III a well-functioning administrative machine. Thus, the beginning of the

bureaucratic age of Ur III is introduced by the following year-formula:

nin-urta ensi

-gal

en-lil

-la

-ke

en-lil

nin-lil

-la

-ke

eÒ-bar-kin ba-an-

-ga

Òul-gi lugal uri

-ma- ke

GAN

nig

-ka

Òa

en-lil

nin-lil

-la

-ke

-sa

-a

“Year: Ninurta, the Ensi-gal of Enlil, having taken an omen for the house of Enlil

and Ninlil,

∑ulgi

, the king of Ur, set straight the fields and the accounts for the house of

Enlil and Ninlil.”

243

The exact nature of the so-called “Reforms of

∑ulgi

” is not well understood.

244

There exists no conclusive evidence that

∑ulgi

ever wrote a set of social reforms; rather it

seems plausible that

∑ulgi

, by virtue of his unparalleled reign, was able to usurp a large

portion of the power and possessions previously held by temple households and local elite

families. Bookkeeping in the years following

∑ulgi

21 seems to have aimed, increasingly, at

embracing all aspects of society; whether ownership of the entire state had been transferred to

∑ulgi

is not clear, but possible.

243. See also K. Maekawa, “The “Temples” and the “Temple Personnel” of Ur III Girsu-Lagash”, in K.

Watanabe (ed.),

Papers of the Second Colloquium on the Ancient Near East - The City and its Life held at the

Middle Eastern Culture Center in Japan (Mitaka, Tokyo), Sonderdruck

(Heidelberg 1999) 66 and 68. Th.

Jacobsen translates: “Year when Ninurta, ensi-gal of Enlil ordered an audit for the temples of Enlil and

Ninlil, and

∑ulgi

, king of Ur, straightened out the fields (forming) the core of the accounting for the

temples of Enlil and Ninlil.” Th. Jacobsen, “The Term Ensí,” AulaOr 9 (1991) 115 (see also fn. 16).

244. See, for example, M. Sigrist,

Drehem

(1992) 9, W. Sallaberger,

Ur III-Zeit

(1999) 148, and K. Maekawa,

“The “Temples” and the “Temple Personnel” (Heidelberg 1999) 66 - 68.

101

During the second half of

∑ulgi

’s reign, numerous military campaigns were directed

towards the eastern and north-eastern regions bordering the Mesopotamian plain. The lands

to the north and north-west are likely to have been kept under control more through

alliances than through war.

245

∑ulgi

’s 39

year a building of some significance was constructed at

PuzriÒ-

Dagæn

. The building was so important that the year was named after this event:

246

Òul-gi lugal uri

-ma-ke

lugal

ub-da limmu

-ba-ke

puzur

iÒ

da-gan

mu-

“Year:

∑ulgi

, the King of Ur, King of the four corners, built the house

PuzriÒ-

Dagæn

.”

247

Whether that building activity alluded to the whole city of

PuzriÒ-Dagæn

or only to

a particular temple or building within it is unknown. Drehem (

PuzriÒ-Dagæn

’s modern

245. See W. Sallaberger,

Ur III-Zeit

(1999) 156-161.

246. The 39

year of

∑ulgi

was also called the Year after the year after: “the wall of the land was built”

(Sumerian mu-us

-sa bad

ma-da ba-du

mu-us

-sa-bi). M. Sigrist believes that the house of Drehem was

constructed in

∑

38, so that the next year (

∑

39) could be named after the event (M. Sigrist,

Drehem

(1992) 13). I doubt this interpretation, and suggest that the years were named only some months after

the new-year, when an important event had taken place, as is implied by the fact that many years had

two names, one “the year after” and one new. Chronological evidence from the data set corroborates this

hypothesis only to some extent. Texts dated with the year-formula mu e

puzur

dagan ba-du

occur

only from month 4 (except for some Umma texts with the month name iti

Òe

KINku, a possible Drehem

date in Umma documents?), whereas texts dated according to the year-formula mu-us

-sa BAD

ba-du

mu-us

-sa-a-bi are attested only from the 3 first months of the year, suggesting a naming of the year

∑

to have taken place at the end of month 3.

247. See also D. Edzard, ZA 63, 1973, 202

102

Arabic name is Drehem, a name which for the sake of convention will be used in the

following) has been termed the great animal-pen of the Ur III Empire,

248

suggesting its

redistributive character, but its exact function is still not entirely clear.

The officials in the local administrations of the Ur III provinces had their seal-

inscriptions altered some time after the year when the “House at

PuzriÒ-Dagæn

” was built,

suggesting that the power of the king was being strengthened. Instead of carrying a

dedication to the local governor, the seal-inscriptions now contained either a dedication to

the king, or nothing but the title and perhaps the patronymic of the seal-owner. The exact

date of this change is difficult to pinpoint, but it seems to coincide with the establishment of

Drehem.

249

∑ulgi

had at least one, but possibly as many as three, wives, and several concubines

(Sumerian lukur).

250

The kings of Ur used dynastic marriages extensively as a political tool,

by sending their daughters off to foreign places, and also by marrying daughters from

neighboring chiefdoms and states. The clan of Ur-Nammu secured its supremacy by

marrying daughters of the kings to high-ranking officials of the empire, co-opting these

248. F. Thureau-Dangin, RA 7 (1910), 186.

249. See, for example, the peculiar history of the seal of Gudea, the son of Ur-Nigar, chief cattle administrator

(

ÒuÒ

)

. The same seal which prior to

∑

43 had had a dedication to Ur-Lisi, the governor of Umma, was

recarved to conform with the simple format of a patronymic inscription in style after

∑

40+/- 2. Another

example suggesting the same evolution is the seal of Lugal-inimgina, the scribe, son of Lugal-nesag’e,

which had a destiny similar to the seal of Gudea. This preliminary survey owes much to the unpublished

dissertation of R. Mayr,

Seal Impressions

(1997). See also fn. 402, p. 158 below.

103

important families in the reign of the kings of Ur.

251

The majority of the known royal

women of the Ur III period had Akkadian names.

It is possible that

Abπ-simtπ

was the wife of

∑ulgi

, rather than the wife of Amar-

Suen (see fn. 273 below). She was from northern Mesopotamia as indicated by the seal-

inscription of her brother Babati. A fragmentary genealogy of the family of Babati is

presented below. That family has been included among the cadet branches of the royal family

of Ur (see Figure 2). Based in part on Old Babylonian sources, it has been suggested that

∑ulgi

married

Taræm-Uram

, a daughter of the ruler of Mari,

Apil-Kπn

252

Amat-Suen, also

read GEME

-Suen, is the only person we can say for certain was a wife (Sumerian dam) of

∑ulgi

253

The names of

∑ulgi

’s concubines were

Ea-niÒa

254

Simat-Ea, Suqurtum, Ninkala,

250. The Old-Babylonian Proto-LU (MSL 12) gives three Akkadian readings of lukur (SAL.ME): lines 263-

265 (p.42):

lukur

na-di-tum

, lukur

-di-iÒ-tum

, lukur

ba-tu-ul-tum

. The second of these,

qadiÒtum

translates as a woman of special status; it is also written nin-dingir. batultu is an adolescent girl, also

written ki-sikil in Sumerian. The more traditional Akkadian translation of lukur,

nadπtu

, is also of little

help understanding the Sumerian title; we read in the CAD (N

p. 63) that a naditu is “a woman

dedicated to a god, usually unmarried, not allowed to have children, usually living in a

gagû

,” a

translation contrary to the Ur III meaning of the term. See also Excursus 2.

251. In the end, the Ur III kings had perhaps become ensnared in a web of familial ties which facilitated the

fall of the dynasty. The case of

IÒbi-Erra

, a general of the Ur III army, and a possible blood relative of

the ruling family of Ur who aided the downfall of the Ur empire, exemplifies the dangers of an over

extensive use of political marriages.

252. J. Boese and W. Sallaberger, “Apil-Kin von Mari und die Könige der III. Dynastie von Ur,” AoF 23

(1996) 24-39.

253. See RIME 3/2 1.2.67. RIME 3/2 1.2.68, which has the unusual spelling a-ma-at instead of GEME

104

Geme-Ninlila, and

∑ulgi-simtπ

was perhaps elevated to queenship around

∑

30.

255

Ninkala is also called queen (nin) in a few texts.

256

All of these women, except

Taræm-Uram

and

Abπ-simtπ

, were recorded with their

personal relationship to

∑ulgi

in various inscriptions on seals and votive objects.

Ea-niÒa

, for

example, is called “his concubine” (Sumerian lukur-ra-ni), “beloved concubine” (lukur ki-

254. A person called i-TI-e

-a (presumably /idin-ea/) is mentioned as a brother of Ea-

niÒa

, see RA 73, 191 (=

AAICAB 1, 1971-351) (from

∑

48 viii), and OIP 115, 199 (= A 2949) (from

∑

46 iii 19). A person by

the same name is also mentioned as a man of

∑ulgi

(

Òul-gi

) in RA 59, 111 S 1 (no year), as a chief

administrator (

Òabra

), in JEOL 33, 114 5 (from

∑

47 xi), and TCND 400 (from

∑

48), as chief cattle

administrator (

ÒuÒ

) in PDT 1, 550 (no date). He is also mentioned as the father of

∑º-Ea

in JCS 32,

171 1 (from

∑

47 viii). Note that all references are from the reign of

∑ulgi

255. See MVN 8, 97 (from

∑

32 v); NYPL 235 (from

∑

38 vii); MVN 3, 162 (from

∑

39 iii); AUCT 1, 952

(from

∑

39 iv 6); Princeton 1, 55 (from

∑

35 xii); RT 37, 129 ab. 1 (from

∑

35 ix); TCS 145 (from

∑

iv); and TCS 170 (from

∑

37 ii), all imply a relationship between

∑ulgi-simtπ

and Tezen-mama (te-ze

-ma-ma), who according to D. Frayne, RIME 3/2 (1997) 267, was counted among the daughters of

Amar-Suen. The texts cited here could perhaps aid future studies targeted at disentangling the maternal

descent of the princes and princesses: Tezen-mama is never mentioned together with another royal

spouse; she was perhaps the daughter of

∑ulgi-simtπ

256. See ASJ 11, 129 59 (= BM 29874) (from

∑

48); CT 7, 27 (= BM 18376) (from

∑

42), and MVN 17, 8

(= BM 12237 = ASJ 2, 31 87) (from

∑

42), suggest that Ninkala held the city

urua

a ki

(or

urua

.URUxGU)

) as her private estate (see also the house of

Simat-IÒtaran

documented in the

GarÒana

archive). The three texts all have the similar colophon nig

-ka

-ak

Òe

urua

a ki

/ nin

-kal-la

nin (account of the grain of GN, Ninkala the queen). Only the latter text has been collated; it is the only

one of the two to give Ninkala the title queen (nin). For the fields of various high-ranking members of

the Ur elite including Ninkala see also ASJ 9, 126f 57 (= BM 29860) (no date), and TIM 6, 3 (from AS

1). In TMHC NF 1/2, 204 (from

∑S

1 iv), Ninkala received 5 mana of wool; the tablet was sealed with

the servant seal of

Ummπ-†æb

, the slave-woman of

∑æt

-[Suen], the daughter of the king.

∑æt-Suen

known as a daughter of

∑ulgi

, but her relationship to

∑ulgi

’s concubine Ninkala is unknown.

105

-a), “beloved concubine of the king” (lukur ki-ag

-lugal), “his campaign consort” (lukur

kaskal-la-ka-ni), and “campaign consort, his bellowed concubine” (lukur kaskal-la lukur ki-

-ga

-ni).

257

Geme-Ninlila is called “his beloved” (ki-ag

-ga

-ni).

258

Ninkala is called “his

beloved citizen of Nippur” (dumu nibru

ki-ag

-ga

-ni).

259

Suqurtum is called “his beloved

concubine” (lukur ki-ag

-ga

-ni).

260

It is unknown, at present whether these “titles” were of

any consequence for the standing of the concubine; we would expect the king to favor some

women over others, allowing the formation of factions within the harem.

261

The nature of

Ur III concubinage is, however, not altogether clear. It was not limited to the king. The

governor of Umma, the sukkalma

and governor of Girsu, as well as several generals, all may

have cohabited with a concubine alongside their wife/wives.

The consorts of

∑ulgi

were involved in the economy, to which the Drehem

administration testifies.

262

Contemporary evidence from the provincial capital of Umma,

where the wife of the governor seems to have had her own resources, suggests that the queen

may also have been in charge of a household of her own.

263

257. See RIME 3/2 1.2.71 to 1.2.81 for the titles of

Ea-niÒa

258. See RIME 3/2 1.2.82 for the title of Geme-Ninlila.

259. See RIME 3/2 1.2.83 to 84 for the title of Ninkala.

260. See RIME 3/2 1.2.85 for the title of

∑uqurtum

261. It is also possible that the king divorced himself from the women over the years, or that they perished,

leaving

∑ulgi

with only one wife at a time. Harem is here used strictly to refer to the group of women

with some sort of marital tie to the king, concubines as well as wives. It is unknown whether these

women dwelled in the same house, or whether they othewise can be said to constitute what can be

compared to the classical Ottoman harem.

262. W. Sallaberger,

Ur III-Zeit

(1999) 253 - 260.

106

∑ulgi

had more than 20 children, some of whom held important positions in the

civil administration or the military.

264

No information has survived as to who the mothers of

the different children of

∑ulgi

were, but we may hope that future prosopographical work will

give some clues as to the possible existence of royal factions based on maternal lineage.

Due to the particular nature of the sources—naming royal progeny “son of the king”

rather than “son of PN the king”, and often providing no title at all—it is not possible to

specify a title for more than a few of

∑ulgi

’s sons. A handful served in the military, others

were only mentioned a few times in the texts from Drehem. These sons of

∑ulgi

were Amar-

Suen (the future king, see below),

∑u-Suen

(the future king, see below), possibly Ibbi-Suen

(the future king, see below),

EÒtar-ilÒu,

perhaps the same as the general (Sumerian

Òagina

)

mentioned in SAT 3, 2143 (no date), Luduga, who seems to have been active in the Umma

province, Lu-Sunzida, who, albeit never mentioned with a title, figures among high-ranking

officials of the empire, Lu-Nanna, the general of Nagsu and Zimudar,

265

whose son

Ennam-

263. The references to a chief administrator of

Ea-niÒa

(

Òabra e

-a-ni-Òa

) corroborate this. See, for

example, CT 32, 36 (= BM103403) (from

∑S

2 to 3) (tenure of

∑ulgi-bæni

), PDT 1, 99 (from

∑

47 iii

4), and SACT 1, 131 (from

∑

46 xii 13) (tenure of

Luduga

). For references to a household of

∑ulgi-

simtπ,

see, for example, Orient 16, 107 174 (from

∑

43), and M. Sigirist,

Drehem

(1992) 222 - 246.

The newly discovered archive of the household of

Simat-IÒtaran

from

GarÒana

, supports this

hypothesis. The archive includes references to the chief household administrator (

Òabra

) and the

scribe of the chief household administrator.

264. See figure 4, below, for a complete list of the members of the royal family, in parts adapted from D.

Frayne, RIME 3/2 (1997) (pp. 166 - 170,

∑ulgi

, pp. 266 - 268, Amar-Suen, pp. 336 - 337,

∑u-Suen

and pp. 375 - 376, Ibbi-Suen). Figure 2 represents an attempt to sort out some of the cadet branches of

the House of Ur-Nammu.

265. UET 3, 75 (from

∑

S 1 i).

107

∑ulgi

is also known,

Puzur-EÒtar,

another general,

∑º-Enlil,

who served as general of Uruk

and BADAN,

266

Ur-Suen, who served as general of Uruk

267

and of BADAN, and Ur-Nigar,

who also served as general of Uruk. One could point to the pattern that several royal sons

held the position of governor of Uruk as an important discovery relating to both the self-

understanding of the royal family of Ur, and the mechanics of succession. However, the

sources are too meager to further advance this theory.

268

Several of

∑ulgi

’s daughters are also

known. They were, among others, En-nirzi-ana, Nin-TUR.TUR-mu, En-uburzi-ana,

PeÒ

TUR.TUR

, Dadagu (?),

Taræm-∑ulgi

, and

Liwwir-MittaÒu

, who became queen of

MarharÒi

∑

18. Apart from

Liwwir-MittaÒu

, it is not know if any of these princesses

were married to high-ranking officials of the empire.

269

266. See below, fn. 270.

267. For example, RIME 3/2 1.2.96 (= NBC 1934), and RA 13, 21 7 (from

∑

48).

268. The possibility of Uruk as an Ur III

Dauphinage

was discussed by P. Michalowski in “Durum and Uruk

during the Ur III Period,” Mesopotamia 12 (1977) 83 - 96.

269. Note that D. Owen recently suggested that

Simat-IÒtaran

was a daughter of

∑ulgi

, and married to

∑º-

Kabta

, a high-ranking official of the empire (D. Owen, RAI 47, paper (2001)). Note also the daughter

∑ulgi

who was married to the ruler of

AnÒan

∑

30.

∑ulgi

destroyed this city in

∑

34.

108

∑ulgi

died on the first or the second day of the eleventh month of his 48

regnal

year.

270

He was well remembered in the literary tradition, with more than 20 hymns

composed in his honor, some of them even said to be self-laudatory.

271

Amar-Suen became king after his father, and ruled for nine years. All the year-dates

of Amar-Suen are known. Whether or not Amar-Suen was the oldest son or his father’s

favorite is unknown. Some form of administrative training is believed to have been a

prerequisite for any royal heir prior to his elevation to kingship; surprisingly, Amar-Suen is

never mentioned before his ascension to the throne. There are three possible answers to this

peculiar problem: 1) Amar-Suen never held any important office before he became king. A

very unlikely scenario, except if he were too young to have had the possibility of the vital

training an administrative or military position would have given him. 2) The name Amar-

Suen is only a throne-name. Throne names are only rarely attested during the third

270. See D. Frayne, RIME 3/2 (1997) 110, for a summary of the evidence, with references to previous

literature.

271. There exists, as of today, no conclusive evidence that any of the

∑ulgi

hymns were composed during the

time of

∑ulgi

. That some of the neo-Sumerian royal hymns contain an “historical kernel” is no evidence

for their historicity (so W. Hallo, “The Coronation of Ur-Nammu,” JCS 20 (1966) 135 - 139, followed

by J. Klein,

Three

∑ulgi

Hymns

(Ramat-Gan 1981) 58 and fn. 139). Whether the literary texts contained

an historical kernel or not is of no consequence for their possible use in reconstructing the history of Ur

III. Such an historical kernel can be found among all the legends, myths, and other incorporated

folkloristic material, after the historical events have been sorted out using the primary sources. The royal

hymns are therefore not historical sources; they are primarily literary texts collected by scribes, for use

within the cult and schools of the Old Babylonian cities. The majority of the Ur III royal hymns were

presumably composed long after the fall of Ur to the Elamites, and are perhaps based on Akkadian

proto-types translated into Sumerian.

109

millennium, but they are not unknown. 3) Amar-Suen served his father outside Sumer, and

references to his office remain in the royal archives of Ur, still to be unearthed. In all three

cases, we are looking for a scenario that would present Amar-Suen with the ability to build a

power base that would ensure his place in succession.

Amar-Suen is described in the historical tradition as a weak ruler. Perhaps this is done

only to create a literary antithesis to the strong ruler

∑ulgi

272

According to the historical

sources, Amar-Suen’s rule was a period of few conflicts, a condition that can be seen as both a

weakening and a strengthening of the empire. Following the example of his father, Amar-

Suen had several concubines, including [x]-natum, ZagaANbi, Udad-zenat, and

Puzur-uÒa

It is possible, but not certain, that

Abπ-simtπ

was the wife of Amar-Suen.

273

The brother of

Abπ-simtπ

, Babati, was an important figure from Northern Mesopotamia. Judging from the

impressive array of titles listed in his seal-inscription, it seems justified to claim that Babati

and his “gens” were an influential clan, whom the rulers of Ur co-opted during their reign.

274

Like his father, Amar-Suen had many children. Establishing a genealogy of the Ur III royal

family is hampered by the administrative preference for referring to the title rather that to the

name of a person. For the royal family in particular, reference to royal descent is almost

exclusively expressed in impersonal terms: very few texts name a royal father. Rather the

expression “son of the king” (dumu lugal) was used. This has resulted in difficulties when

attempting to connect an individual prince with either Amar-Suen or

∑ulgi

; the family tree

272. P. Michalowski, “Amar-Su’ena and the Historical Tradition,” in: M. de J. Ellis (ed.), EANEF (= Fs

Finkelstein; Hamden 1977) 155-157.

110

of the royal family presented at the end of this chapter is drawn primarily from the

information in RIME 3/2, joined with the new evidence suggesting that Amar-Suen,

∑u-

273. It is paramount to the whole question of neo-Sumerian royal succession to establish whether

Abπ-simtπ

was the wife of Amar-Suen or

∑ulgi

. No conclusive evidence exists favoring either

∑ulgi

or Amar-Suen

as the husband of

Abπ-simtπ

. She is called queen (Sumerian nin) in a text dating to the final year of the

reign of Amar-Suen (UTI 3, 2003 [from AS 9]) as well as in texts from throughout the reign of

∑u-

Suen

(for example, MVN 16, 713 [from

∑S

4], and MVN 16, 916 [from

∑S

3]), whereas she is not

mentioned during

∑ulgi

’s reign.

Abπ-simtπ

is not the only person with the title queen during

∑u-

Suen

’s reign, she is mentioned together with

Kubætum

(the wife of

∑u-Suen

) in two texts (MVN 9,

165 [from

∑S

5], and MVN 16, 960 [from

∑S

3]; in both texts, both women are called “queen” (nin).

However “queen” (nin) is perhaps used as an honorary title: “queen-dowager”, when referring to

Abπ-

simtπ

. She is mentioned without title during the reign of Amar-Suen, save for the single text cited above.

D. Frayne claimed in RIME 3/2 (1997) (p. 285 - 286) that

Abπ-simtπ

was the wife of Amar-Suen,

based in part on an Old Babylonian copy of the seal of Babati the brother of

Abπ-simtπ

(RIME 3/2

1.4.33). The original Ur III inscription of that same seal (RIME 3/2 1.4.32) can only be used to prove

that

Abπ-simtπ

was the mother of

∑u-Suen

(see PDT 2, 1200 [from

∑S

7 iii], which fills the lacuna in

the first line of RIME 3/2 1.4.32). The Old Babylonian copy, which substitutes ama (the Sumerian word

for mother) for dam (wife), also names

∑u-Suen

as the benefactor of the dedication. The original Ur III

inscription can thus not be used to determine whose wife she was. Recently, D. Owen published a text

with the interesting personal name “

∑u-Suen

born of

∑ulgi

” (

Òu-

suen-walid-

Òulgi)

, which he

argues might be the full name of

∑u-Suen

(NABU 2001/17). No other contemporary evidence links

∑u-Suen

with either

∑ulgi

or Amar-Suen (the name of the son of

∑ulgi

in the seal inscription on BRM

3, 52, reads

Òu

EN.[ ], is not necessarily

∑u

-Suen, but just as likely

∑º-Enlil

another known son of

∑ulgi

, see already J. Boese and W. Sallaberger, AoF 23 (1996) 36 - 37). D. Owen has suggested that

Simat-IÒtaran

, amply attested in the

GarÒana

archive about to be published, was a daughter of

∑ulgi

(D. Owen, RAI 47, paper (2001)). This has implications for the genealogy of the royal house, since she

claims to be the sister (nin

) of

∑u-Suen

and later Ibbi-Suen (after he followed

∑u-Suen

on the

throne). T. Gomi, “Shulgi-Simti and her Libation Place (ki-a-nag),” Orient 12 (1976) 1- 14, rules out

the possibility that

Abπ-simtπ

was the same person as

∑ulgi-simtπ

, an otherwise attractive solution to

the problem.

274. See RIME 3/2 1.4.32. See also R. Whiting, “

TiÒ-Atal

of Nineveh and Babati, Uncle of

∑u-Sin

,” JCS

28 (1976) 173 - 182.

111

Suen

and Ibbi-Suen, with some likelihood, were brothers.

275

Amar-Suen’s sons, like the

majority of his brothers, are each known only from a few references in the Drehem corpus.

Most of the generals known from Ur III sources were not direct descendants of the royal

house of Ur; some of them had married into the royal family, and most of them had non-

Sumerian names.

276

Akkadian names dominate among the generals (

Òagina

) of the Ur III

Empire, but it is important to note the shift in naming-practice that took place within the

royal court itself and to not exclude any persons with an Akkadian name from the potential

list of members of the royal family. A few of the generals had names that are neither

Sumerian nor Akkadian, such as

aÒib-Atal

(presumably a Hurrian name), and

un-∑ulgi

Among Amar-Suen’s sons were Dada, the general of Zabala, a city in the eastern

province of Umma,

277

and

∑º-∑ulgi

(perhaps a captain).

278

Several other sons are known

only by name and affiliation with the royal clan.

279

Among his daughters we find

Taddin-

EÒtar

, Ninlil-tukulti, Geme-Nanna, Pakinana,

∑æt-Mami

, Nin-hedu, Geme-Eana, Tezen-

mama,

280

the unnamed wife of Lugal-magure,

Simat-IÒtaran

281

the unnamed wife of

∑arrum-bæni

, and the unnamed wife of Lu-Nanna the son of Ur-Nigar.

282

Several of his

275. This evidence is above all contained in the

GarÒana

archive kindly made available to me by D. Owen

(through the CDLI project), and including personal communications from D. Owen (see also D. Owen,

RAI 47, paper (2001)).

276. However, see P. Steinkeller, “The Core and the Periphery” (1987) 25 - 26.

277. See, for example, AUCT 1, 26 (AS 3 xii).

278. See RA 49, 86 2 (from

∑S

2 x), a text from Drehem, where

∑º-∑ulgi

is mentioned together with other

members of the royal family, and MVN 5, 116 (from AS 7 iii), another Drehem text, where

∑º-∑ulgi

mentioned as a son of the king.

112

daughters were married to high-ranking officials of the empire, a practice begun by his father

∑ulgi

283

279.

Ibbi-IÒtaran

, for example, SAT 2, 774 (= NBC11644) (from AS 4 ix).

Ur-IÒtaran

, for example,

SACT 1, 153 (from AS 4 viii). Ur-Ninsuna, for example, ITT 3, 5001 (from AS 8 iii).

Taddin-EÒtar

(

da-din-eÒ

-tar

) is mentioned together with

Ur-IÒtaran

and

Abπ-simtπ

in DAS 51 (from AS 8); see

also the parallel text DAS 53 (from AS 8).

Amir-∑ulgi

, for example, OrSP 47-49, 23 (from AS 3 i), and

AUCT 1, 418 (from AS 2). Ur-Baba, for example, MVN 3, 232 (from AS 6), which mentions the dowry

of Ur-Baba, to be entered into the house of

Lu-NinÒubur

the chief administrator of An. A

uni, for

example, TCS 336 (= Aegyptus 10, 274 37) (from AS 2 iv), which mentions the dowry of A

uni, to be

entered into the e

zabar-dab

, and UET 3, 1369 (from IS 1), which mentions the e

-du

-la of A

uni.

Inim-Nanna, for example, TCL 2, 5563 (from AS 1 i), which mentions the dowry of Inim-Nanna, to be

entered into the

Ìu-ba-¿a•

(House of

ubaya?), and MVN 5, 122 (from

∑S

1 ix), which records

the delivery of animals to Inim-Nanna, as well as to the throne of Ur-Nammu,

∑ulgi

, and Amar-Suen.

Lu-∑ulgi

, for example, TCL 2, 5508 (AS 4 i), and TLB 3, 98 (AS 4 iv).

Næbi-∑ulgi

, for example, STA

8 (from AS 5 ii to x), and Zinbun 18, 102 7 (= BM 20262 = MTBM 81) (no year).

∑ulgi-ræma

, for

example, TrDr 88 (from AS 7 iv). References to Mansum and Nanna-manba are very scarce. Mansum,

HLC 75 (pl. 36) (from AS 1), rev. 2 (according to RIME 3/2 268). Nanna-manba, see JANES 9, 21 3

(from AS 5 ii 17), obv. 4 (according to RIME 3/2 268).

280. See fn. 255 p. 104 above. Tezen-mama was perhaps a daughter of

∑

ulgi rather than Amar-Suen.

281. See fn. 256 p. 104 above.

282. Such is the order of daughters of the king as they are listed in CTMMA I, 17 (from AS 4 vii), a text

recording regular deliveries for the cult. The first entry records a delivery for the throne of

∑ulgi

, the

following entries recorded the deliveries to the daughters of the king, followed by the deliveries to the

two wet-nurses (Sumerian ummeda) of the king, and a number of persons called mar-tu, and finally, a

number of high-ranking officials from the north. We know the names of other daughters of the king

mentioned during the reign of Amar-Suen; these were Ninlilemanag, see, for example, AUCT 2, 367

(from AS 6 i),

∑elepputum

, see J. Klein, “Shelepputum a Hitherto Unknown Ur III Princess,” ZA 80

(1990) 20-39, and En-ma

galanna, known from the formula of Amar-Suen’s fourth year.

283.

ulibar the general (

Òagina

) of Umma (?) was married to a daughter (called dumu lugal) of the king

(see MVN 13, 735 [no date], from Girsu).

113

The highest official of the empire, second to the king, was the sukkalma

; his

responsibilities were closely associated with the eastern provinces of the empire. During the

tenure of Ir(-Nanna/mu), the office of the sukkalma

was conjoined with the office of the

governor of Girsu. In this brief survey, I will concentrate only on this latter best known

sukkalma

Ir(-Nanna/mu), whose name was sometimes written ir

, sometimes, ir

nanna,

and sometimes ir

-mu, held the office of sukkalma

from the time of Amar-Suen through

the early years of Ibbi-Suen. He was married to a daughter of Amar-Suen,

∑æt-Mami

284

One of his sons was married to another daughter of Amar-Suen, Geme-Eana. It is uncertain

how many wives Ir(-Nanna/mu) had, and whether he himself was a descendant of the House

of Ur-Nammu.

285

This is in part due to the fact that references were often made to the office

rather than the person in the administrative records of the Ur III empire, making

identifications rather difficult. Several women besides

∑æt-Mami

were called wife of the

sukkalma

(Sumerian dam sukkal-ma

∑uba-dua

is called

wife of the sukkalma

(¿dam• sukkal-ma

) in ITT 5, 6997 (from AS 9 x), a text that carries the seal of a person

called Ur-Lamma the son of ir

-mu, perhaps identical with the son of Ir(-Nanna/mu) the

284. For reference to the dowry (Sumerian nig

-mi

-us

-sa) of

∑æt-Mami

see, for example, MVN 11, 192

(=MCS 3 (1953) 25) (from AS 2).

285. According to D. Frayne, RIME 3/2 (1997) 268, Amar-Suen had a son called ir

nanna, who the

author of RIME 3/2 suggests is identical with Ir(-Nanna/mu) the sukkalma

. The only reference cited in

RIME 3/2 is AnOr 1, 111 (dated to AS 7, however the abbreviated date formula in that text, mu bi

tum, is ambiguous).

114

sukkalma

. In a text mentioning Ninkala as queen (nin), we find a woman called Nin-hedu

as the wife of the sukkalma

(dam sukkal ma

286

Baba-Ea is mentioned as both wife of the

sukkalma

and wife of the governor (of Girsu). A semi-precious stone bears the sole reference

Aman-ilπ

, wife of Ir(-Nanna/mu), the governor of

LagaÒ

287

Of all of these women,

Baba-Ea is perhaps the most important, and perhaps Ir(-Nanna/mu)’s wife in his capacity as

the governor of Girsu. Baba-Ea is also referred to as a nin-dingir-priestess.

288

Several seal-impressions testify that Ir(-Nanna/mu) was the son of

Ur-∑ulpa

’e, who

held the office of sukkalma

prior to Ir(-Nanna/mu).

289

The sons of the sukkalma

were Ur-

Nanna, A

uni (also called brother (

ÒeÒ

) of the sukkalma

290

∑º-∑ulgi, ∑º-ilπ

(also called

brother (

ÒeÒ

) of the sukkalma

291

Nanna-mansum, and Ur-Baba.

292

In figure 2 I have tried

to suggest a genealogy of the family of the sukkalma

, describing it as one of the cadet

branches of the House of Ur-Nammu.

286. See ASJ 9, 126f 57 (= BM 29860), also cited above, fn. 256 p. 104.

287. See RIME 3/2 1.5.2004.

288. See fn. 247.

289. Seal of

Ur-∑ulpa

’e see, for example, NFT p.185 (= AO 4198) (=DynChal 16 V) (no date), and NATN

388 (no date).

See also ITT 5, 8220 (=TCS 1, 183) (from

∑S

6).

290. See TCTI 2, 3711 (no year).

291. See TCTI 2, 4161 (no year).

292.

Lu-∑ara

is called brother (

ÒeÒ

) of the sukkalma

in MVN 17, 12; he is otherwise unknown. A

uni and

∑º-ilπ

could have been brothers of Ir(-Nanna/mu), or his uncles. It is unknown whether Ir(-Nanna/

mu)’s son inherrited his office. The last sukkalma

was Libur-Suen. His familial relations are unknown

to me (UET 3, 826 [from IS 22 vi]). See the illustration figure 2, describing the cadet branches of the

ruling family.

115

Unlike Umma and Nippur where the same families seems to have remained in power

throughout the years of the Ur III kings,

293

the seat of the governor of the most important

province,

LagaÒ

, appears to have been controlled more centrally. Eventually, Ir(-Nanna/mu)

who was married into the royal family, and presumably a close ally to the clan of Ur-

Nammu, was able to add governorship of the province of

LagaÒ

to his impressive array of

titles.

294

Ur-Lamma, the first well attested Ur III governor of

Girsu

(the main city in the

province of

LagaÒ

during the neo-Sumerian period),

295

ruled from as early as

∑

41 to AS 3,

when Nanna-

ziÒagal

succeeded him.

Nanna-ziÒagal

governed

LagaÒ

for only two years; he

was succeeded by

∑arakam,

who governed

LagaÒ

from AS 4 (perhaps contemporary with

Nanna-ziÒagal

) to AS 6. Following the four years when

LagaÒ

was ruled by

Nanna-ziÒagal

and

∑arakam

, the sukkalma

Ir(-Nanna/mu) added governor of the province of

LagaÒ

to his

293. Except, perhaps, for the brief intermediate period of Amar-Suen’s reign when the old elite family of

Nippur was replaced by bureaucrats from Drehem, see also below, pp. 117 - 118 + fn. 299.

294. See RIME 3/2 1.4.13, Ir(-Nanna/mu) is here mentioned as sukkalma

governor of

LagaÒ

, sanga priest

of Enki, general of

UÒar-GarÒana

, general of

AÒime

, governor of Sabum and the land of Gutebum,

general of Dimat-Enlila, governor of

Al-∑u-Suen

(the city of

∑u-Suen

), governor of

amzi and

Kara

ar, general of NI.HI, and general of

∑imaÒki

and the land of Karda, in that order.

295. The title “governor of

LagaÒ

” is only used in Ir(-Nanna/mu)’s official seal-inscription (see below), rather

the commonly used title of the governor of the province of

LagaÒ

was “governor of Girsu”.

116

titles.

296

He governed

LagaÒ

until the end of Ur domination over

LagaÒ

in Ibbi-Suen’s 6

year.

The family of the sukkalma

seems to be the most important among the cadet-

branches of the clan of Ur-Nammu. A cadet branch, as mentioned above, is defined as a

distinct family with only secondary familial ties to the ruling clan, which excludes them from

the line of succession but places them close to the center of power. It is likely, but impossible

to prove, that the ruling house of Mari was even more closely related to the House of Ur than

the family of the sukkalma

297

Due to the lack of contemporary sources the Mari ruling

family has not been included in this study.

296. According to K. Maekawa, “Confiscation of Private Properties in the Ur III period: A study of é-dul-la

and níg-GA,” ASJ 18 (1996) 121 - 122, the sequence of Girsu governors following Ur-Lamma

exemplifies the social unrest erupting in AS 2, and ultimately culminating with the elimination of the

line of Ur-Lamma, and the positioning of two

ad-hoc

administrators from the central bureaucracy as

governor until the final solution to the

LagaÒ

problem was reached: the appointment of Ir(-Nanna/mu)

as governor of that province in AS 6.

297. Any reconstruction of the ties between the Mari ruling family and the clan of Ur-Nammu is based

largely on quasi-historical documents.

117

Figure 2: The cadet branches of the royal family of Ur.

It is likely that

∑ulgi

’s two sons and successors, Amar-Suen and

∑u-Suen

, struggled

for the throne. This is deduced from information from the provinces, since documentation

from the capital is lacking. For instance, texts from Nippur are instructive. A provisional

survey of the terms of office for three of the highest officials in the Nippur administration,

the “foreman of the household of Inanna” (

ugula

inanna),

the “chief administrator of

Inanna

” (Òabra

inanna

and the “governor of Nippur” (ensi

nibru

), suggests that all

three offices were controlled by the family of Ur-Meme (perhaps an old Nippur family,

tracing its origins back to the time before the Ur III empire),

298

until the reign of Amar-

Suen. During the reign of Amar-Suen the three offices were in the hands of persons not

Babati

Bizua
nin

nin

Aham-arÒi

Girni-isa

Abi-simti

~ ∑ulgi(?)

Ur-∑ulpa'e

sukkalmaÌ

Ir(-Nanna/mu)

~ ∑at-Mami

dumu-mi

lugal

~ Geme-Eanna

dumu-mi

lugal

∑º-Adad

~ ∑uba-Dua

~ Nin-Ìedu

Ur-Nanna

AÌuni

∑º-∑ulgi

∑u-Ili

Nanna-mansum

Ur-Baba

Ea-niÒa

lukur ∑ulgi

Iti-Ea

∑ulgi / ÒuÒ

∑º-Ea

The family of the sukkalmaÌ

The family of Abi-Simti

The family of Ea-niÒa

Lani

119

Figure 3: Succession of office in Nippur (preliminary survey).

Already from the sixth year of Amar-Suen, certain officials in the Umma province

began to use a seal with a dedicatory inscription naming

∑u-Suen

as the king of Ur. Nothing

∑33
∑34
∑35
∑36
∑37

∑46
∑47
∑48

AS01

AS02
AS03
AS04
AS05
AS06
AS07
AS08
AS09
SS01
SS02
SS03
SS04
SS05
SS06
SS07
SS08
SS09

IS01
IS02
IS03
IS04
IS05
IS06
IS07
IS08
IS09
IS10
IS11

ugula e

inanna

Enlila-maÌ

Lugal-engardu

Ayakala

Lugal-gizkimzi

Sag-Enlil

Òabra

inanna

Ama-zimu

ensi

nibruki

Ur-Meme

Ur-Ninibgal

AÌuma

Lugal-melam

Nam-zitara

Dada

}

Family of Ur-Meme

}

Family of Ur-Meme

120

suggests that Amar-Suen died before late in his 8

or early in his 9

year.

300

Both years were

named throughout the empire according to Amar-Suen year-names.

301

Another

interpretation of the Umma evidence is to suggest that Amar-Suen and

∑u-Suen

ruled

together for some time.

302

When a king died, it was customary to settle accounts and measure the arable land of

the entire kingdom.

303

Perhaps the circumstances surrounding the transfer of power from

one king to the next also prompted the removal of any official whose loyalty lay with a rival

branch of the ruling clan. Any royal grants to these subjects would also likely have been

confiscated.

304

300. No extant text discloses the time of death of Amar-Suen. See D. Frayne, RIME 3/2 (1997) 242, citing

SET 66, with reference to the offerings for the throne of Amar-Suen, which according to W. Sallaberger,

Der Kultische Kalender der Ur III-Zeit, Teil 1 & 2

, (= UAVA 7/1 & 7/2; Berlin/New York 1993) 147,

occurred only posthumously, suggesting that Amar-Suen was dead before AS 9 xi 26.

301.

MVN 13, 739,

can be used to calculate the correct sequence of years following the death of Amar-Suen

early in his 9

year (see forthcoming note by the author).

302. See W. Sallaberger,

Ur III-Zeit

(1999) 166 + fn. 154, for references to the debate.

303. This is above all documented in the land-survey texts (for example, the text MCS 6, 83 [=BM 105334]).

304. The so-called e

-du

-la texts suggest that the favors which the king bestowed upon his subjects, were

reversible at any time. Ur-Lisi, one of the well-known Umma governors, lost all his possessions and

possibly his life in the eighth year of Amar-Suen, perhaps, as claimed by K. Maekawa, in connection

with the struggle for power at the court (K. Maekawa, ASJ 18 (1996) 127). However, see the discussion

concerning the nature of this property in W. Heimpel, ASJ 19 (1997) 63 - 82.

121

∑u-Suen

followed his brother (or father?) on the throne and ruled the empire for 9

years.

∑u-Suen

may have held the position of military governor/general (Sumerian

Òagina

)

of Uruk and BADAN during his father

∑ulgi

’s reign.

305

As a prince,

∑u-Suen

is only known

from texts dated to the reign of Amar-Suen, perhaps an indication that he was preparing

himself for his natural place in the line of succession.

306

As we have seen, it is possible to

argue that

∑u-Suen

was either a son of

∑ulgi

, or a son of Amar-Suen, but the parallel

scenario provided by evidence from the provincial court of Umma, combined with the

305.

∑u-Suen

held the title “general of BAD

.AN

at some point in his career. This can be inferred from the

dedicatory seal on Mesopotamia 12, 93A (from AS 9 iii). Following Michalowski (Mesopotamia 12

(1977) 83 - 96), it seems reasonable to argue that the general of Uruk was also the general of

BAD

.AN

. See above for the seal-inscription RIME 3/2 1.2.94 (= BRM 3, 52 = MLC 2357), believed

by D. Frayne, RIME 3/2 (1997) 188, following J. Boese and W. Sallaberger, AoF 23 (1996) 36, to be an

inscription of

∑º-Enlil

. It is likely, as Michalowski suggested (P. Michalowski, Mesopotamia 12 (1977)

84) that BAD

.AN of the Ur III period was a locality close to Uruk.

306. The earliest clear reference to the prince

∑u

-Suen is CTM 1, 10 (from

∑

43). There exist numerous

references to a person named

∑u

-Suen with the occupation “runner” or “messenger” (sukkal / lu

kas

he was perhaps son of a captain (see, e.g, MVN 13, 689). During the late years of

∑ulgi

and the early

years of Amar-Suen, numerous references to contributions to (or from) the Drehem livestock, made by a

person called

∑u

-Suen, may, in fact, refer to

∑u

-Suen the prince. Compare MVN 13, 113 (from

∑

47),

to RA 62, 8 11 (from AS 1 i). The title prince is only preserved in the example from AS 1, but could

have been present in the first text too. RA 62, 8 11 (from AS 1 i), is therefore the earliest certain

attestation of the title prince (dumu lugal) for

∑u

-Suen.

∑u

-Suen, called “great knight” (lu

geÒ

tukul

gu-

la) in Amherst 68 (from AS 3 xi), might be the prince, but this is not certain at all. An analysis of the

hierarchical standing of the military command is necessary to determine whether a prince could serve as

a “knight,” but that is beyound the scope of this study. Some texts recorded a

∑u

-Suen delivering bear-

cubs to the Drehem pen (see, for example, MVN 11, 140 [from AS 5]); the majority of these deliveries

were sheep and goats, indicating his special position.

122

anthropological evidence presented in Excursus 1 and outlined in Chapter 3, both weigh in

favor of the former solution.

Although the sources were still abundant at the time of

∑u-Suen

’s ascension, they

only mention one wife of the king,

Kubætum

, and one concubine, Ti’amat-

baÒti

Accordingly, we know of only three children of

∑u-Suen

, ignoring Ibbi-Suen who I believe

could have been another son of

∑ulgi

. These were the daughters

Tabºr-Îa††um

∑æt-Erra

and Geme-Enlila.

Tabºr-Îa††um

and

∑æt-Erra

are not attested directly as daughters of

∑u-

Suen, but

circumstantial evidence suggests this relationship.

307

Geme-Enlila was perhaps identical with

the wife of Ibbi-Suen by the same name;

308

however, her title, daughter of the king (dumu-

lugal), need not refer to Amar-Suen,

∑u-Suen

or Ibbi-Suen. The sole reason for her

identification as a daughter of

∑u-Suen

is her attestation as a daughter of the king in

documents dated to the reign of

∑u-Suen

By the time of

∑u-Suen,

the empire began to experience serious hardship. Judging

by the year-date formulae, the south was threatened by a migration from the north, and in

the middle of

∑u-Suen

’s reign, the protective wall called Muriq-Tidnim (“Holding back the

Tid

num”) was constructed.

∑u-Suen

conducted only two campaigns worthy of year

names.

309

Still,

∑u-Suen

was, and is, considered a strong leader and a supporter of the arts.

307. See D. Frayne, RIME 3/2 (1997) 337 for their identification.

308. As suggested by D. Frayne, RIME 3/2 (1997) 337. See p. 132 + fn. 314 below for a description of this

seemingly incesteous relationship.

123

Several self-laudatory hymns concerning

∑u-Suen

, aimed at magnifying his memory, have

survived. These texts have probably been instrumental in creating the image of

∑u-Suen

as a

successful ruler, both in the minds of the ancients as well as the modern reader.

∑u-Suen

died no later than the fourth day of the tenth month of his ninth year.

From that day on he received funerary offerings.

310

Ibbi-Suen, perhaps yet another son of

∑ulgi,

followed

∑u-Suen

on the throne. Ibbi-

Suen ruled 24 years.

311

It is very likely that Ibbi-Suen never ruled an empire, but only Sumer

during his first five years, and perhaps only the capital Ur from then on.

312

The year-formula from Ibbi-Suen’s sixth year mentions the construction of city-walls

surrounding Ur and Nippur. This is believed to indicate fear of an immanent invasion.

When paired with the break-away of the provinces, it seems reasonable to interpret this year

as the beginning of the end of the Ur III state.

i-bi

suen lugal uri

-ma-ke

nibru

uri

-ma-ke

bad

gal-bi mu-du

“Year: Ibbi-Suen the king of Ur built great walls of Nippur and Ur.”

309.

∑S

3: Year: “Simanum was destroyed”, and

∑S

7: Year: “

ZabÒali

was destroyed”.

310. W. Sallaberger,

Ur III-Zeit

(1999) 171

311. The last administrative documents from the Ur III period date to the 12

months of IS 23 (UET 3, 711

and UET 3, 712). SKL gives Ibbi-Suen 24 years on the throne.

312. Umma texts cease from the fourth year of Ibbi-Suen following a rapid decline in the numbers during

Ibbi-Suen’s first three years. The last Ur III text from Girsu was written in IS 5 (8?), from Drehem IS 8,

and from Nippur in IS 8.

124

The destruction of the Ur III Empire was not only visited on Ur by outside forces—

internal problems also facilitated the decline. A shortage of grain and a resulting collapse of

the system of stable equivalencies can be seen as a result of unrest in the northern provinces

as well as an inner problem fueled by that conflict.

313

The collapse of the system of stable

equivalencies began during Ibbi-Suen’s 5

year, and accelerated during the following three

years.

By the time of Ibbi-Suen’s reign, information about the royal family had begun to

decrease dramatically. Only two royal children can be ascribed to Ibbi-Suen: Mammertum

and

∑ulgi-simtπ

, and Ibbi-Suen is mentioned in connection with only one wife, Geme-

Enlila (see also above p. 122), and no concubines. It is not unthinkable that the union of

Ibbi-Suen and Geme-Enlila was incestuous. Whether she was a sister-queen or a niece-queen

is uncertain. Incestuous relationships, inbreeding, is a well known phenomenon in history; it

is usually restricted to stratified societies, and, in particular, well-known from the uppermost

social levels of society, in patrilineal clans, and in royal households where the king had access

to a large harem (polygamy naturally reduces the risks associated with inbreeding).

314

Ibbi-Suen was confined to Ur during the approximately 20 years following the break-

away of the provinces, and the year-names stemming from these years tell the tale of

increasing paranoia, as illustrated in the year-formula from Ibbi-Suen’s 23

year:

i-bi

suen lugal uri

-ma-ra ugu-dul

-bi dugud kur-be

mu-na-e-ra

313. See Th. Jacobsen, JCS 7 (1953) 36-47, and T. Gomi, JCS 36 (1984) 211-242.

125

“Year: the stupid monkey in the foreign land struck against Ibbi-Suen, the king of

Ur.”

Just a couple of years earlier, Ibbi-Suen’s year formula was:

i-bi

suen lugal uri

-ma-ra mar-tu a

- ul-ta uru

nu zu gu

im-ma-an-ga

“Year: the Amorites, the powerful south wind who, from the remote past, have not

known cities, submitted to Ibbi-Suen the king of Ur.”

The large account from Ur, UET 3, 1498, dating to IS 15 (from the first to the last

month), is a good example of the desperate economic situation in Ur after the empire had

deteriorated. The text, an account of the workshops of the royal household, is mainly

concerned with the recycling of precious materials. As in all Ur III accounts, care is taken to

314. Incestuous relationships have a higher sterility rate than non-incestuous relationships. See P. Van Den

Berghe and G. Mesher, “Royal Incest and Inclusive Fitness,” American Ethnologist, Volume 7, Issue 2

(1980) 300 - 317. According to Van Den Berghe and Mesher, royal incest is a “high-risk, but also an

extremely high-gain, strategy” (p. 304) for the female participant. A royal daughter is left with few

alternatives other than to mate with her own brother or father persuing her goal of producing a

successful heir. This is, according to the authors, due to the tendency to hypergyny (hypergamy) among

most females in stratified societies. For the king the gains are obvious: he may produce an heir with 3/4

of his own genes. The consequences of incest becomes tremendous when analyzed over time, “after just

eight generations of full-sibling matings, father and son have an r [relatedness coefficient] = .95.” In

other words, the heir is a regular clone of the father (see p. 305). However, incestuous mating is not to be

encouraged “under monogamy, [when] the risk of not producing a fit heir with a sister or a daughter

would be too high.” (p. 304). See also fn. 4, in cases of absolute hegemony interdynastic marriages are

not seen as an atractive strategy for the royal daughter, who will be more inclined to mate within her

clan. Note, finally, that brother-sister unions are the most common form of incestuous mating, second is

father-daughter unions, whereas mother - son relations are almost unknown (see. p. 305).

127

Figure 4: The royal family of Ur.

Why did the succession proceed as it did? What determined that Amar-Suen

ascended to the throne, although he was apparently not a strong candidate? A prince who did

not enjoy the support of a large segment of both the royal family, as well as the provincial

Dagan-DUni
∑u-Enlill (Òagina of Uruk? and BADAN?)
Lugal-azida
Nabi'um
Lu-Nanna (Òagina of Zimudar and Nagsu)
Nabi-Enlil
Amar-Damu
Puzur-EÒtar (Òagina)
EÒtar-ilÒu
Etel-pº-Dagan
Lu-Sunzida
Lu-Duga
Na-DI
Ur-Suena (Òagina of Uruk and BADAN)
Ur-Nigar (Òagina of Uruk)
∑u-EÒtar
En-nirzi-ana
Nin-TUR.TUR-mu
En-uburzi-ana
Liwwir-MittaÒu
PeÒ

-TUR.TUR

Dadagu
Taram-∑ulgi
Baqartum
Simat-Enlil
Simat-EÒtar
∑æt-Suen
∑æt-∑ulgi (wife of Abya-muti the Martu)

+ Amat-Suen

- Ea-NiÒa (lukur)
- Geme-Ninlila (lukur)
- Ninkala (lukur/nin)
- Simat-Ea (lukur)
- ∑uqurtum (lukur)
- ∑ulgi-simti (lukur/nin)

+ Abπ-simtπ

+ ?

- Udad-zenat

- Puzur-uÒa
- ZagaANbi

- x]-natum

Ibbi-IÒtaran
Dada (Òagina of Zabala)
Ur-Baba
Bayamu
Nammu-maba
Ahuni
Ur-IÒtaran
Ur-Ninsuna
Amir-∑ulgi
Inim-Nanna
Lu-∑ulgi
∑u-∑ulgi
Nabi-∑ulgi
Ur-saga (Òagina)
Mansum
∑ulgi-rama
Taddin-EÒtar
Ninhedu (wife of ÎaÒib-atal (Òagina))
Pakinana
Ninlil-tukulti
Tesin-Mamma
Ninlile-manag
∑elepputum
Geme-Eana (wife of the son of Ir

(sukkalmaÌ))

En-mah-gal-ana (En pristess of Nanna)
Simat-IÒtaran
Geme-Nanna
∑æt-Mami (wife of ∑º-Adad, son of Ir

(sukkalmaÌ))

PN (wife of Lugal-magure)
PN (wife of Lu-Nanna son of Ur-NIgar)
PN (wife of ∑arrum-bæni)

Geme-Enlila

∑æt-erra

Tabºr-Îattum

+ Kubatum

- Ti'amat-baÒti

+ Geme-Enlila

Mammetum

∑ulgπ-‡imtπ

Ennam-∑ulgi

+ Taræm-Uram

Certain familial relation

Uncertain familial relation

Key:

Wife (dam)

Concubine (lukur)

The Royal Family of Ur

∑ulgi

∑u-Suen

Ibbi-Suen

Amar-Suen

Ur-Nammu

Utu-

Îegal

Kingship

Succession

- Òa

-bi-...]

∑º-EÒtar

?+ SI.A-tum

128

governors and other important members of society, is unlikely to be successful, according to

the simple rules of succession outlined above. His claim to succession would be contested,

and another prince whose career had allowed for the formation of connections and the

building of alliances would prevail. It is likely that some sort of principle existed dictating

that the next in seniority was supposed to follow

∑ulgi

, and it is possible that it was precisely

the circumstances of

∑ulgi

’s long rule that had resulted in this system.

317

It is likely that

∑ulgi

outlived any brothers he might have had, thereby eliminating the threat from the

senior generation. Babati’s supposed distinctive position in the social hierarchy can thus be

explained as deriving from his position as a senior member of the royal family exercising his

influence through his sister’s sons.

The genealogy that has been suggested for the royal family of Ur divides most of the

male princes between

∑ulgi

and Amar-Suen.

318

This is done according to the earliest

attestation of the title dumu lugal or “son of the king” for each of these persons. It is

therefore safe to say that the progeny of

∑u-Suen

and Ibbi-Suen was restricted, and that

∑ulgi

, and probably Amar-Suen too, had multiple male heirs. This might have been a

conclusive factor in determining the succession, since Amar-Suen, perhaps himself weak, had

the support of many sons, themselves holding important offices and hoping for a place in the

line of succession, as well as the support of important allies married to his daughters.

319

317. Compare with R. Burling (1974) 7 - 8.

318. This division is based largely on the information given in RIME 3/2.

319. See also Excursus 1.

130

C h a p t e r 5 . T h e Ru l i n g Fa m i l y o f Ur I I I Um m a

3 2 2

Chapter 5: Section 1: Introduction

Umma, located north of Girsu, and north-east of Ur, on the

Umma-canal

umma

)—an outlet from the Tigris

323

—was already an ancient city during the Ur III period

four thousand years ago. Umma has been identified with Tell Djokha, one of the largest sites

322. Some recent prosopographical studies have achieved the reconstruction of extended genealogies for

several important Ur III families. For Umma: T. Jones & J. Snyder,

Sumerian Economic Texts from the

Third Ur Dynasty

(= SET; Minneapolis 1961) 322-344, briefly discussed the ruling family of Ur III

Umma. The first study to be entirely devoted to this family was D. McGuiness, “The Family of Giri-

Zal,” RA 76 (1982) 17-25. That study was based in part on his unpublished dissertation (D. McGuiness

Studies in Neo-Sumerian Administrative Machinery

(unpublished dissertation; UMI 1981)). D. Snell,

Ledgers

(1982) 77-81, Snell only briefly discussed the ruling family; he introduced the term the “Umma

Fiscal Office”. P. Steinkeller, JESHO (1981) 116 - 121, suggested that the office of Lukala—in this

study called the chief household administrator of the governor—was the (Umma) “irrigation office” (p.

121). Steinkeller later adopted the term the Umma “fiscal office”, in P. Steinkeller, AOS 68 (1987) 76. In

1990, T. Maeda described in detail the duration of the tenures of some members of the ruling family of

Umma (T. Maeda, “Father of Akala and Dadaga, governors of Umma,” ASJ 12 (1990) 71-78). A recent

study by W. Yuhong aimed at decoding the career patterns for the members of the ruling family of

Umma: W. Yuhong, “High-ranking "Scribes" and Intellectual Governors during the Akkadian and Ur III

Periods,” JAC 10 (1995). See also T. Maeda for the most recent study devoted to our topic (T. Maeda,

"Ruler's Family of Umma and Control over the Circulation of Silver," ASJ 18 (1996) 254-260). For

Nippur: W. Hallo, JNES 31 (1972) 87-95. R. L. Zettler, "The Genealogy of the House of Ur-Me-me: a

Second Look," AfO 31 (1984) 1-9. For Girsu: K. Maekawa, "The Governor's family and the 'temple

households' in Ur III Girsu," in K. Veenhof (ed.),

Houses and Households in Ancient Mesopotamia

CRRAI 40; Leiden 1996) 171-1. For Gudua: D. Owen, "The Ensis of Gudua," ASJ 15 (1993) 131-152.

323. Earlier believed to be the “Eastern Euphrates” or the Iturungal canal, the recent study by P. Steinkeller,

“New Light on the Hydrology and Topography of Southern Babylonia in the Third Millennium,” ZA 91

(2001) 22 - 84, is used here as a reference to the geography of the Umma province.

132

Only very few Umma texts from the pre-Sargonic period have been published; and

the abundant historical inscriptions from the neighboring city-state of

LagaÒ

has been almost

the only source when attempting to reconstruct the history of that area of Sumer.

325

Umma

texts from the period immediately prior to the Ur III period, the Old Akkadian period, have

long since been known; albeit poorly understood they are an important source for the history

of Umma.

326

All of the thousands of Ur III documents from Umma were excavated during

the early years of the 20

century by the local population, they quickly found their way to

the markets in Europe and North America. Recent illicit excavations at Umma are said to

have produced an abundant record of texts from all periods, even texts from the earliest

periods of Mesopotamian civilization, the Late Uruk period.

Since the archaeological record of Umma does not provide any understandable

description of the site, I will not venture into any discussion of the geographical setting of

that city; rather I will limit this introduction to a description of the administrative layout of

the province as recorded in the cuneiform record, with a particular focus on the ruling elite.

A much broader study of the geography and history of Umma is planned within the

325. The CDLI project counts 17 stone objects from Umma (thanks are due to Klaudia Englund for making

her Umma files available to me) dating primarily to the pre-Ur III period (two Ur III inscriptions). Based

on these few fragmentary historical inscriptions, and the

LagaÒ

historical record, D. Edzard published a

genealogy of the pre-Sargonic rulers of Umma (D. Edzard,

“

Königsinschriften des Iraq Museums II”

Sumer 15 (1959) 22).

326. B. Foster published a genealogy of the rulers of Umma during the Sargonic period (B. Foster,

Umma in

the Sargonic Period

(1982) 154-156). A web-presentation of the Old Akkadian material is planned within

the frame-work of the CDLI project. Note that Umma may have been a stronghold of the Gutian rulers

of the post-Akkadian period; see P. Steinkeller, ZA 91 (2001) 31.

133

framework of the CDLI; see in that regard also the projected study by P. Steinkeller on the

topography and hydrology of the Umma province.

327

The city of Umma was the capital of the province of the same name, and

consequently the only city with a governor. The only other economically important city in

the province of Umma was Apisal. The cities Zabala and KIAN, although frequently

mentioned in the records, were presumably cities of minor economic relevance that held

some religious significance. The districts of Gu’dena

328

and

MuÒbiana

, often grouped

together, were perhaps without any large permanent settlements.

The province of Umma was divided into three agricultural territories:

aÒa Da-

Umma, aÒa Apisal, and aÒa Gu

’

dena and MuÒbiana

. Da-Umma, often simply called

Umma or

aÒa

Umma, was the most important of these three districts, and the one which by

far generated the largest yield of economic documentation as well as the largest cereal

production.

329

The fact that whenever the districts were listed together Da-Umma was

always mentioned first, and Apisal second, followed by Gu’dena and

MuÒbiana

, support this

327. For an outline of Steinkeller’s projected study, “Population Density, Settlement Patterns and Rural

Landscape in Southern Babylonia under the Ur III Dynasty: The Case of the Province of Umma,” see P.

Steinkeller, ZA 91 (2001) 23.

328. Gu’dena of Ur III Umma is perhaps identical with the famous Gu’edena mentioned numerous times in

the historical inscriptions of the pre-Sargonic

LagaÒ

-rulers. In Ur III Umma sources, the area Gu’dena is

written primarily gu

-de

-na, but occasionally gu

-eden-na; it is possible that the former writing is a

phonological variant of the later.

329. See for example BM 110116 (cf. K. Maekawa, Zinbun 22, 25-82), a record of the yield from Da-Umma.

The total area of Da-Umma recorded in that text was 339 bur and 1 iku (2,197 hectares). See also K.

Maekawa, AoF 16 (1989) 49-50 and table 2.

134

observation. The tablet container record Aleppo 433, archiving documents from the two

years

∑

45 and

∑

46, is useful in that it illuminates the structure of the agricultural territories

of Umma:

1. pisan dub-ba

Tablet container;

2. kiÒib] nam-Òa

-tam

tablets of the Òatam-administrators of

3. ugula nam-1(u)-ke

-ne

“foremen of 10”

330

4. da-umma

(concerning the districts of ) Da-Umma,

5. a-pi

-sal

Apisal,

Reverse.

1. gu

-eden-na

’

edena

2. u

muÒ-bi-an

and MuÒbiana,

3. i

-gal

are present,

4. mu 2(diÒ)-kam

from two years.

5. mu ur-bi

-lum

Year: “Urbilum (was destroyed)”

6. u

mu ki-maÒ

and

Year: “

KimaÒ

(was destroyed)”.

MCS 6, 83, BM 105334 (from AS 2), which is likely to have been a survey of all the

lands of the province of Umma, divided it into three areas of almost equal size. The first third

recorded the domain land, divided into 100 units of 6 bur, each managed by a “cultivator”

(engar). With some additional prebend-lands, this was the core of the Umma state lands. An

area of the exact same size was said to be left fallow.

331

The last section computed another

330. For the use of interchangeable titles in the agricultural sector (ugula >< nu-banda

Òabra

) see

fn. 503 p. 206, below.

331. AnOr 1, 303 (no date), supports this interpretation; the first entry in that text is an area of 1200 bur

called the domain land (GAN

). Unfortunately, the size of the fallow lands is not discernible due to

the fragmentary state of this text.

135

third of the total lands, designating it as cultivated prebend-land for allotment holders.

332

The total amount of agricultural land in Umma as recorded in this text comes to more than

2000 bur, making Umma about 1/4 the size of

LagaÒ

insofar as cereal agricultural is

concerned.

333

The expected yield from 600 domain units—using a 20 gur per bur ratio

334

—is

matched in AAICAB 1, 1912-1143 (from

∑

28), recording an expected yield of 15,000 gur

from the domain units of Umma; this product was controlled centrally by the imperial

court.

335

Unfortunately these key texts, so important to the reconstruction of the Ur III

Umma geographical layout, do not inform us about the fraction of Umma domain units

located in any of the main districts, Da-Umma, Apisal, and Gu’dena and

MuÒbiana

The traditional Mesopotamian pantheon of the late 3

millennium was honored in

the city of Umma, and we find references to temples dedicated to local deities as well as gods

from other provinces, and perhaps even to deities from outside Sumer.

336

The chief deity of

332. See K. Maekawa, Zinbun 22 (1987) 38-39, and table 7.

333. K. Maekawa, “Agricultural Production in Ancient Sumer,” Zinbun 13 (1974) 11, estimated that the

Girsu area to be harvested (domain units and the allotments for the cultivators) was ca. 3,664 bur

. This

area produced ca. 24 guru

, 2,691 gur and a few sila

, or more than 22 millions of liter of barley. The

yield of Girsu domain land, therefore, was perhaps 6 times the yield harvested from Umma domain land.

334. In comparison, the yield in Girsu during the latter half of

∑ulgi

’s reign and the first half of Amar-Suen’s

reign averaged 30 gur per bur. See for instance K. Maekawa, Zinbun 13 (1974) 11. However, K.

Maekawa, AoF 16 (1989) 49, seems to suggest a 30 to 24 gur per bur yield in Umma.

335. AAICAB 1, 1912-1143 (from

∑

28), reverse, line 8, reads “copy of sealed tablet exists in the palace,”

gaba-ri

kiÒib

-gal gal

-am

136

the Umma pantheon,

∑ara

, held a pivotal position in Umma, together with his spouse Nin-

ura,

337

and Inanna,

∑ara

’s mother.

338

The main temple of

∑ara

was the Ema

339

The Ema

appears as an element in some Ur III Umma personal names; names formed with the

theophoric element

∑ara

are among the most common in Umma. In the 9

year of

∑u

Suen’s reign, the central administration constructed (anew?) the main sanctuary of

∑ara

Umma.

340

Students of Ur III history have traditionally assumed that the entire body of Umma

texts were from the city of Umma, and, in particular, from the household of the governor.

The first numerous Ur III Umma texts appear in

∑

26, the last texts were written in IS 4.

More than 16,000 tablets has so far been published dating to the approximately 45 years

between those two dates.

336. See for example the reference to

NanÒe

of Umma in NABU 1989, 95 9 (from IS 4). See M. Cohen, The

Gods of Suburban Umma, Fs. Limet (

Tablettes et images aus pays de Sumer et d'Akkad: Melanges offert a

Monsieur H. Limet

) (Liege 1996) 27 - 35, for a study of the province of Umma from the point of view of

regionalism in the cult.

337. See A. Cavigneaux and M. Krebernik, “Nin-ura,” RLA 9, 510.

338. See for example “A balbale to

∑

ara” (

∑

ara A, see http://www-etcsl.orient.ox.ac.uk/section4/c4301.htm)

339. Nin-ura’s main temple was the E-ul(a). It is likely that the many references to the e

Òara

(“house of

∑ara

) alluded to the e

. Likewise, the e

nin-ur

-ra (“house of Nin-ura) may be a reference to the

(scarcely mentioned) e

-u

-la. See A. George,

House Most High, The Temples of Ancient Mesopotamia

(Winona Lake, Indiana 1993) 152.

340. Year: “

∑u

-Suen, the king of Ur, built the temple of

∑ara

in Umma” (

Òu-

suen lugal uri

-ma-

Òara

umma

-ka mu-du

137

The city of Apisal,

341

perhaps identical with the site known as Muhallaqiya, located

downstream from the outlet of the Umma-canal on the Tigris river (“Eastern Euphrates”),

was the second most important city in the Umma province.

342

The substantial tell at

Muhallaqiya has never been excavated. A large number of Umma tablets refers to activities

that take place in Apisal, suggesting that this city was governed entirely from Umma.

343

The

pantheon of Apisal seems to have been a local variant of that from Umma; its main deities

were the Apisalite

∑ara

, and the Apisalite Ninura, among others. Apisal is thought to have

been an important center for herding activities.

The city of Zabala,

344

which seems to have been of minor economic importance,

345

was an important cultic center, in particular in the period after

∑S

346

during which the

“queen-dowager”

Abπ-simtπ

visited the city four times in five years.

347

Zabala was the city of

Inanna of Zabala, and the oldest Ur III sources from Zabala (the records were presumably

kept in Umma) mention a temple of Inanna at Zabala.

348

Prior to the coronation of

∑u-

Suen,

references to Zabala primarily recorded minor deliveries for the cult of Inanna of

Zabala.

349

341. For the reading Apisal see (with reference) P. Steinkeller, ZA 91 (2001) 54 + fn. 127.

342. See P. Steinkeller, ZA 91 (2001) 54-55.

343. See T. Jones & J. Snyder, SET (1961) 337-338, suggesting that Ur-E’e held a position connected with

the rule of Apisal.

138

The only travelers to Zabala mentioned before

Abπ-simtπ

were a number of gods

344. For a possible identification of Zabala with Ibzaykh see now P. Steinkeller, ZA 91 (2001) 54, + fn 124.

There seems to be some confusion concerning the writing of Zabala. Following Ellermeir (F. Ellermeier,

Sumerisches Glossar

[1979 - 1980]), the following values are accepted:

Ellermeier

Borger AbZ

# of attestations in Ur III sources

zabala

ZA.MU∑

.UNUG

zabala (586)

zabala

ZA.MU∑

.UNUG

zabala

(586)

zabala

MU∑

.UNUG

zabala

(103)

zabala

MU∑

.ZA.UNUG

zabala

(103)

zabala

MU∑

.AB

zabala

(103)

8 (9?)

zabala

AB.MU∑

zabala

(128)

zabala

MU∑

.UNUG.ZA

1 (ITT 3, 4954)

zabala

MU∑

.TE.UNUG

zabala

MU∑

.ZA.AB

zabala

ZA.AB

zabala

ZA.MU∑

.AB

UNUG.MU∑.ZA

1 (AUCT 1, 805)

For this survey, 86 attestations from the published record have been checked against published

photograph or autograph (when no such record was available to me, the attestation in question was left

out), zabala

is by far the most frequent writing of the city-name.

345. For a possible reference to a mayor of Zabala see AUCT 1, 225 (from

∑

25 vi), rev. 8 -9:

¿x•

x(=2(diÒ)?)

a-za-an-num

/ zabala

. AUCT 1, 26 (from AS 3 viii to xii), mentiones a general of

Zabala by the name of Dada, obv. 2 - 3:

mu da-da ¿Òagina• zabala

-ka-Òe

] / kiÒib Òu-i

-li

¿x-x•.

MVN 16, 683 (from AS 7), mentions a galama

-priest of Zabala, rev. 3: gala-ma

zabala

346. One reference from AS 9 suggests that

Abπ-simtπ

visited Zabala during that year too; see UTI 3, 2003.

347. See fn. 273 on p. 110 above for a discussion of

Abπ-simtπ

, the wife of either

∑ulgi

or Amar-Suen. Here

she is called “queen-dowager” in opposition to

Kubætum

who is called “queen.”

348. The household had a small permanent staff; see, for example, AAICAB 1, 1911-229 (from

∑

28 viii). See

also AnOr 1, 88 (from AS 5), which mentions conscriptions for bala service from among the staff of

Inanna of Zabala (the structure of the text is similar to TCL 5, 6038, mentioned above on pp. 75 - 76).

139

who received rations during their journey there:

350

TCS 346 (from AS 6 i):

Obverse.

1. 2(diÒ) udu 1(diÒ) maÒ

2 sheep, and 1 goat,

nin-nun-gal

Nin-nungal,

3. zabala

351

(MU∑

.TE.UNUG)

-Òe

having gone to Zabala.

gen-na

4. 1(diÒ) gukkal

1 fat-tail sheep,

igi-zi-bar-ra

Igizi-bara,

6. zabala

-Òe

gen-na

having gone to Zabala.

Reverse.

1. ki a-lu

-lu

-ta

From Alulu

2. zi-ga iti Òe-KIN-ku

booked out.

Month “Harvest”.

3. mu-us

-sa en unu

-gal

inanna

Year after: “Enunugal-Inanna (was installed)”.

The first reference to the “queen-dowager”

Abπ-simtπ

’s yearly visits to Zabala comes

from the account concerning Lu-kirizal, the pig-herder, SNAT 436 (from

∑S

1):

349. See for example

BIN 5, 19 (from ∑

33), a text describing the wool meant for the “lofty garnment” of

Inanna of Zabala, from Ur-E’e (for

tug

maÌ

see H. Waetzoldt,

Untersuchungen zur Neusumerischen

Textilindustrie

(= UNT; Rome 1972) xxi (“Prachtgewand”), and see xxiii fn. 77 for a possible reading of

tug

Òutur

350. See also UTI 4, 2563 (from AS 8), a list of minor food-stuff offerings designated for (obverse line 11)

“Nin-gipar having ascended to Usag(?),” (

nin-gi

-par

-sag-Òe

-a)

, and (rev 8) “Igizid-bara

having gone to Zabala” (

igi-zi-bar-ra zabala

-Òe

gen-na).

The delivery was made by

Ur-

∑ulpa

’

, and sealed by the governor (of Umma, Ur-Lisi).

351. This reading could not be checked agains any graphic representation of the signs; it has not been

included in the previous analysis (fn. 344 on p. 138), and Boson’s reading has been retained here.

140

Reverse.

...

5. 4(diÒ) Òa

nita

Four male pigs,

6. 2(diÒ) Òa

gur

nita

Two male

“gur”

pigs,

7. igi-kar

nin zabala-Òe

gen-na

provisions

352

for the

“queen-dowager”

having gone to Zabala.

8. kiÒib nu-ra-a ensi

-ka

Unrolled seal of the governor.

...

The vast majority of texts mentioning Zabala from the following five years were

concerned with the annual visits by the “queen-dowager.” Although no royal visitor was

mentioned in

∑u-Suen

’s second year, it is still possible to suggest that

Abπ-simtπ

paid a visit

to Zabala that year as well. The fragmentary text MVN 18, 508 (from

∑S

1?), even alludes to

a house of the queen in Zabala.

353

The deliveries for the cult of Inanna of Zabala as well as

the provisions for the “queen-dowager’s” visits were mostly sealed by the governor of Umma

or members of his administration; one of these,

Ur-∑ulpa

’e was perhaps a royal

representative at the governors court.

354

Twelve or thirteen texts from

∑S

1 mention Zabala; the majority of these texts were

not dated by month, but the five that have a month-name were dated to either month three,

four, or five.

355

Seven of the texts from

∑S

1 mention the “queen-dowager” by title, or

352. See P. Steinkeller, “On the Reading and Meaning of igi-kár and gúrum (IGI.GAR),” ASJ 4 (1982) 149-

151.

353. See also MVN 16, 796 (from

∑S

4 vii), obv. 9, which mentions a house of the queen, although the

location of this house is not specified.

354. See, for example, MVN 18, 463 (from

∑S

1). For the role of

Ur-∑ulpa

’e, see also in fn. 495 on p. 199

in this study.

141

specifically

Abπ-simtπ

by name. All other texts from the same year seem in some way to be

related to the “queen-dowager’s” visit to Zabala;

356

it is likely that the governor of Umma

went to Zabala at the same time.

357

The texts from

∑S

1 which do mention the “queen-

dowager” were all related to basic household functions.

358

355. SNAT 436; MVN 18, 508 (from

∑S

1 ?); YOS 18, 93; UTI 4, 2602; MVN 18, 463; Torino 2, 524;

Princeton 1, 243; BM 105353 (unpubl.); ITT 5, 6983 (from month 3); MCS 3, 43 12 (= BM 105502)

(from month 4); SET 288 (from month 4); UTI 4, 2321 (from month 4); and Ontario 2, 38 (unpubl.)

(from month 5).

356. One text (Torino 2, 524) records the oil-rations for two members of the imperial staff (Ennum(mi)-

ilπ,

the equestrian, and and

∑º-Mamπtum

, the door-keeper) and a nin-dingir priestess of Inanna of Zabala

from the sukkal

maÌ

; another text (BM 105353 (unpubl.)) records the royal offerings for Inanna of

Zabala via Ninmar(ka), the cup-bearer (For Ninmarka (the cup-bearer) see fn 151, p. 59 in this study). A

Girsu text from month three (ITT 5, 6983) record the man-power for shipping flour to Zabala from

Girsu; another text recorded the delivery of baskets from Ur-

∑ulpa

’e received by a person named Gurzan

in Zabala. A text recorded the dispatch of three NIG

(perhaps an error for

tug

-lam

?) garments to

Zabala (YOS 18, 93) (sealed by Gurzan son of x-layabi, the cook of the governor), the last text from

∑S

1 not to mention

Abπ-simtπ

(Ontario 2, 38 (unpubl.)) mentioned the beer rations for Urra-il(?), the

general, sealed by the governor (Ayakala), in Zabala.

357. Ontario 2, 38 (unpubl.) (from

∑S

1 v), is a simple receipt for an allotment of ten sila of good beer for

the general, from Alli, Ayakala’s chief brewer, sealed by Ayakala the governor in Zabala.

358. Apart from the pigs mentioned in SNAT 436, one text mentions the work-days of a team of workers

under Lu-balasag concerned with the “queen-dowager’s” journey to Zabala (UTI 4, 2602); another text

records a delivery of sesame(?) oil

geÒ

) for

Abπ-simtπ

going to Zabala, sealed by Ur-

∑ulpa

’e; one

text records pottery booked out (from the account of the “queen-dowager”) while on the way to Zabala

(zi-ga nin zabala

Òe

gen-na-

aÒ

) (Princeton 1, 243). One text mentions 17 hal-baskets for messengers,

filled with leather (eight skins) (1(u) 7(

diÒ

)

al kin-gi

-a

kuÒ

si-ga /

kuÒ

-bi 8(

diÒ

)-am

) for the

“queen-dowager,” while in Zabala (SET 288 [from

∑S

1 iv]); another text records the same bags booked

out of Zabala via the “queen-dowager” (UTI 4, 2321 [from

∑S

1 iv]).

142

Since the only text with a month-name concerned with the “queen-dowager’s” visit to

Zabala dates to month four, it seems likly that

Abπ-simtπ

visited Zabala during the fourth

month of that year. Three texts from

∑S

1 mention

Ennum(mi)-ilπ

, the equestrian; he may

have been the queen’s personal commissary.

359

Only two texts from

∑S

2 mentioning Zabala have been published; neither of these

mention the queen,

360

but both record deliveries for the siskur-offerings

361

of the king to

the cult in Zabala.

362

Eight of the eleven texts from

∑S

3 that mention Zabala also mention the “queen-

dowager”

Abπ-simtπ

363

Two texts (MVN 16, 837, and SNAT 481) record the provisions for

the “queen-dowager” going to Zabala: the first listed gold, the second baskets and garlic.

Both deliveries were made by Lukala, the chief household administrator of the governor; the

first was sealed by the governor (Ayakala), the second by

∑arakam

364

MVN 16, 960,

recorded the garment provisions for

Abπ-simtπ

on the way to Zabala, following a smaller

provision for the child born to the queen,

Kubætum

365

and in turn followed by several

entries recording provisions for gods and imperial officers, from Ikala, sealed by the governor.

359.

Ennum(mi)-ilπ

is known to have been an officer in the imperial administration; see, for example,

MVN 13, 549 (from AS 9 vi); Ontario 1, 115 (from AS 9 xii

min

14). He is perhaps identical with the

well-known officer responsible for multiple deliveries to the Drehem administration (see, for example,

PDT 2, 1135 [from

∑

43 i]; PDT 2, 1184 [from

∑

45 xii 24 to 29]).

360. AnOr 7, 377 (from

∑S

2 to 4), however, does mention a journey by an unknown person (text is broken)

to Zabala.

361. Following W. Sallaberger, UAVA 7/1 & 7/2 (1993) 41-42.

362. MVN 16, 877 (no month name), and MVN 4, 174 (from the first month).

143

While traveling to Zabala, in

∑S

3, the queen received animals from

UÒmu

(the chief

fattener in Umma, see pp. 239 + fn 591, below)—the animals were termed

maÒ-da-ri-

366

—the transaction was sealed by the governor (MVN 16, 916).

Abπ-simtπ

transferred

some animals as siskur-offerings for Inanna of Zabala (BIN 5, 31).

367

The queen herself

received siskur-offerings while in Zabala.

368

The last text from

∑S

3 to be mentioned here recorded the transfer of beer and bread

to Zabala while the queen was traveling there:

363. MVN 16, 960; MVN 16, 916; MVN 16, 837; SNAT 481; BIN 5, 31; Princeton 1, 238; MVN 16,

1092 (from month 2); and AnOr 7, 235 (from month 2). Three texts do not mention the “queen

dowager” but Zabala dating to

∑S

2: One text (JCS 39, 125 13 [month 2]) record the provisions for (the

divine?) Zabala, Enlil, Nin

ursag, and the (divine?) kab

-ku

of en-gaba-ra

, from

∑arakam

—one of

Ayakala’s provisioners—sealed by the governor (Ayakala). It is, of course, not uncommon to find

geograhical names among the deities receiving offerings in Ur III text. The kab

-ku

of en-gaba-ra

, is,

however, never again attested in this context. For the geographical name kab

-ku

of en-gaba-ra

see, for

example, Princeton 1, 477 (from

∑

36), recording work done there, obv. 4: kab

-ku

en-gaba-ra

A.DUN gub-ba. The “the field across from the lord” (a-

Òa

en-gaba-ra

) is well attested in Ur III Umma

sources. Another similar text (Nik 2, 326 [month 9]) recorded provisions for miscellaneous deities.

SNAT 487 (from month 9) recorded the garment rations for the permanent staff of Zabala (rev. 10:

tug

-ba giri

-se

-ga zabala

364.

∑arakam

, Alli, among others, functioned as supply officers of the governor, several hundred tablets

sealed by Ayakala confirms this, the documents all relate to minor deliveries of beer, flour and other basic

commodities.

365. Obv. 3; igi-kar

ku-ba-tum nin-e dumu tu-da.

366. Following W. Sallaberger, UAVA 7/1 & 7/2 (1993) 160 - 170,

maÒdaria

is understood as a sort of

regular delivery for the imperial court and not an offering, although it was destined for the religious

festivals in the capital and Nippur, and sometimes diverted to cult-offerings.

367. Although both MVN 16, 916, and BIN 5, 31, recorded different kinds of livestock there is no direct

indication that the same animals recorded in the one text were the same as in the other.

144

AnOr 7, 235 (from

∑S

3 ii):

1. 2(u) 2(diÒ) guruÒ u

2(diÒ)-Òe

22 workers for two days,

2. kaÒ ninda zabala

-Òe

having brought beer and bread to Zabala,

3. de

-a u

zabala

-a gub-ba

and having stayed in Zabala,

4. a-bi

-si

-im-ti nin i

-im-gen-na-a

while Abπ-simtπ the queen, was travelling.

Reverse.

1. ...]-x

x x

2. ¿kiÒib• lu

]-kal-la

Sealed by Lukala.

3. iti sig

geÒ

-Òub-ga

-ra

Month

“

placing the brick

in the mold

”

(seal)

4. mu-us

-sa ma

en-ki

Year after: “the boat of Enki”.

Seal

1. lu

-kal-la

Lukala,

2. dub-sar

scribe,

3. dumu ur-e

-e ÒuÒ

son of Ur-E

’

e, chief cattle administrator.

Abπ-simtπ

perhaps visited Zabala during month 2 of

∑S

3, the only month attested

in several texts relating to her visit.

Five of the eight texts from

∑S

4 mentioning Zabala also mention the “queen-

dowager”. Three of these texts record provisions for the queen.

369

One text recorded siskur-

368. See Princeton 1, 238, obv: 7(

diÒ

)

-ti-um 5(

diÒ

) sila

esir

su-ub-ba / siskur

nin /

Òa

zabala

/ ki

a-gu-ta, “Seven Patium 5 sila baskets, coated with bitumen, siskur-offering of the “queen-dowager”

(while) in Zabala, from Agu,” and MVN 16, 1092 (from

∑

S 3ii) (not specified as siskur): 1(u) 5(

diÒ

)

kaskal / kun-zi 3(u)-ta / ki nin-

Òe

Òa

zabala

// ki ur-

ba-ba

-ta /

kiÒib

ensi

..., “15 travel-

baskets for fish, 30 kun-zi fish each, for the “queen-dowager” (while) in Zabala, from Ur-Baba, sealed by

the governor ...”.

145

offerings for Inanna of Zabala transferred by the queen.

370

The last text from

∑S

4 to

mention

Abπ-simtπ

was MVN 16, 796, from month seven, recording household items

booked out (for) the “queen-dowager” while traveling to Zabala from Ayakala (the chief of

the leather workers), once again sealed by Lukala.

Two of the three texts from

∑S

4 which mention Zabala, but which do not mention

the “queen-dowager,” were concerned with textile offerings to Inanna of Zabala.

371

The

textiles of these texts were termed

su-si

inanna zabala

-Òe

372

The last text from that year

(UTI 4, 2705) documented field work. Since the only month-name on tablets from Zabala

during

∑S

4 mentioning the “queen-dowager” was the seventh, there is reason to believe that

this was the month she visited Zabala that year.

SAT 3, 1568 (from

∑S

5), is the only extant text following

∑S

4 which recorded the

“queen-dowager’s” presence in Zabala (reverse line 1;

nin zabala

gub-ba

). That text

recorded the labor involved in bringing certain products to Zabala from Umma and back

369. MVN 16, 1330; MVN 16, 713; and BM 106073 (unpubl.) (from month 7).

370. AnOr 7, 241 (from month 7).

371. MCS 8, 90 (=BM 105544), and MCS 2, 76 (=BM 113031).

372. The term su-si occurs a mere nine times in the published Ur III record, and always in connection with a

divine name. It appears to be a term used to describe different fabrics given to gods, such as regular

tug2

uÒ

-bar or

tug

uÒ-bar

with various qualifications. The texts are: MCS 2, 76, BM 113031 (from

∑S

4): Inanna of Zabala; MCS 8, 90, BM 105544 (from

∑S

4): Inanna of Zabala; MVN 14, 244 (from AS

7 to 8): Gula; Rochester 145 (no year, month 9): Gula; TJAMC FM 51 (pl. 47) (from

∑S

5): Inanna of

Zabala; BM 108269 (from

∑

47): Nin-Egal; UTI 3, 2018 (from

∑S

1): Gula of Umma; UTI 3, 2069

(from

∑S

1): Ninsun; SACT 2, 280 (from

∑S

2): for several gods; and MVN 18, 626 (no date): Nin-x.

146

again; the document was sealed by the provisions officer of the Umma governor,

∑arakam

373

There is no indication that

Abπ-simtπ

visited Zabala at a certain time of

∑S

4. In

∑S

1 she presumably went there during month 4, in

∑S

3 during month 2, and in

∑S

4 during

month 7.

The last text to mention the city of Zabala is AAICAB 1, 1911-206 (from

∑S

9),

recording the delivery of a few hides (of sheep?) for Inanna of Zabala. Based on the relatively

low number of texts mentioning Zabala altogether, this text cannot be used as a terminal date

for Ur control over Zabala.

Numerous records mention the city KIAN, and in particular deliveries for the cult of

∑ara

there.

374

The names of shrines for many other deities as well as many cultic activities

are recorded in the extant record. Little evidence exists, however, that links the city of KIAN

with either the royal family of Ur, or the gubernatorial court of Umma in any other way than

basic administrative activities.

In the following, I devote several pages to the earliest generations of the ruling family

of Umma, including Ur-Nigar, who is considered the actual

paterfamilias

, and to GIRINI,

373.

Abπ-simtπ

must have died prior to

∑S

9 xii 17, when her first ki-a-nag is recorded (see ASJ 3, 92 3).

374. See for example the account of deliveries for

∑ara

of KIAN, concerning Dada (perhaps identical with the

sanga-priest called Dada?) BCT 2, 143 (from

∑

48), rev. iv 20:

nig

]-¿ka

• ak mu-DU

¿d•

Òara

•

.AN

149

Chapter 5: Section 2: The earliest generations

In their seal-inscriptions both Nigar-kidu (nigar

(NIGIN

)

gar

-ki-du

)

375

and Ur-

Nigar (ur-nigar

(NIGIN

)

gar

) claimed to be sons of GIRINI (GIRI

-NI).

376

According to

these seal-legends,

377

GIRINI held two titles, chief of the galla (

gal

-la gal

378

and chief

cattle administrator (

ÒuÒ

379

titles which were to be divided between his two sons; Nigar-

kidu became chief of the galla (gal

-la gal), and Ur-Nigar became chief cattle administrator

(

ÒuÒ

). GIRINI had another son named Basag

380

whose title is still unknown.

381

Few

375. U.UD.KID = KWU 512 (ABZ 447a). In Ur III administrative texts, NIGIN

primarily occurs together

with the phonetic compliment GAR. The exceptions, seem mainly to be the consequence of copying or

scribal errors. Following H. Waetzoldt (Bi.Or. 32, 1975, 383), and lately M. Krebernik (Die

Beschwörungen aus Fara und Ebla (Hildesheim 1984) 197 and note 119) a reading nigar

is accepted

here. The word nigar

was used almost exclusively in the personal names geme

-nigar

(NIGIN

)

gar

nigar

(NIGIN

)

gar

- ki-du

and ur-nigar

(NIGIN

)

gar

, except for a few references to an e

nigar

(Nik 2,

290 [from

∑S

6 viii 6 to 7]; SAT 3, 2145 [XX XX iv 12]; and BE 3-1, 165 [no date]).

376. Also read kiri

-zal by some, presumably seen as a phonetic variant of kiri

-zal. A reading GIRI

-NI is

preferred to leave open any interpretation. My suggestion is that the name is an abbreviation of Giri

-ni-

-sa

, a common Ur III name.

377. The seal of Nigar-kidu was rolled on MVN 1, 182 (from

∑

26), together with the seal of Lu-Nin-x the

“grand scribe of the king” (dub-sar ma

lugal-ka). The document recorded the transfer of a large amount

of barley from Ur-Nigar to the palace.

378. The title gal

-la

gal is problematic. Obviously the gal makes for an interpretation as “chief of the gal

”, leaving us with the gal

-la

. An equation with Akkadian gallû(m)—a word for an evil demon—

must be abandoned for obvious reasons; rather, “police chief, gendarme, deputy, or bailif;” is more

reasonable, although the etymology remains obscure.

150

documents are available from the time when we expect GIRINI to have been active, making

it difficult for us to say anything about him. No impression of his own seal has been

recovered, and no text can be ascribed to him with certainty. I therefore find it advisable to

describe GIRINI as a semi-legendary ancestor of the Umma clan rather than as an

administrative figure with his own place in the administration.

Ur-Nigar, one of GIRINI’s two sons, became father of the most powerful generation

of administrators governing Umma, counting among others three well-known governors. Ur-

Nigar’s title, chief cattle administrator (

ÒuÒ

), is almost exclusively known from the seals of

his sons.

NYPL 318 (from

∑

24(?)),

382

an early text dealing with barley, is the only certain

attestation of Ur-Nigar, chief cattle administrator

(ÒuÒ

), and thereby the only certain

379. Ur-Nigar’s seal inscription has not been preserved on any tablet (strike MVN 1, 136, cf. already R. Mayr;

The Seal Impressions of Ur III Umma

(unpublished dissertation; Leiden 1997) catalogue no. 807).

However, an actual seal of Ur-Nigar, son of GIRINI, the chief of the galla, has been preserved; see L.

Speleers,

Catalogue des intailles et empreintes orientales des Musées Royaux du Cinquantenaire

, (Bruxelles

1917) 106-107, seal no. 625 (L. Speleers noted that the title of GIRINI was also found in the Obelisk of

ManiÒtusu

, column 12, line 2, and column 13, line 11 (both times in the phrase

Òi

gal

-la

-gal, “the

one of the chief of the galla”). See R. Mayr,

Seal Impressions

(1997) 149 + fn. 582 for a brief discussion of

the seal of Ur-Nigar. The text FAOS 17, 88, independently supports the point that Ur-Nigar was son of

GIRINI.

380. Spelled ba-sa

-ga in his seal, and sometimes ba-sig

in the texts, clearly supporting a reading sag

SIG

381. A possible further son of GIRINI is Lugal-niglagar’e mentioned in SACT 2, 98 (from

∑

34 vi).

382. The year information reads mu 3(

diÒ

)-kam us

-bi, this is perhaps an abbreviated writing of nig

-ka

-ak

al-la-ka mu 3(

diÒ

)-kam us

-sa-bi =

∑

24 (compare with MVN 21, 272 [from

∑

24 xi])

151

attestation of Ur-Nigar the father of three governors altogether. It is possible to speculate that

a prospering private economic sphere existed, parallel to the centralized economy of the

empire, and that even members of the ruling family of Umma were involved not with state-

affairs but rather dealings of their own, leaving no traces in the records of the central

households. It is also plausible that the administration of the early years of Ur hegemony had

yet to experience the excessive bureaucracy of the late years of

∑ulgi

and the reign of his two

sons Amar-Suen and

∑u

-Suen, and that Ur-Nigar and his predecessors were state-employees

functioning in an administrative machinery based less on written records and more on

personal charisma and will.

Nigar-kidu is an even more elusive figure, partly because several people in Umma

were called Nigar-kidu, and partly because Nigar-kidu’s son Luduga primarily sealed

documents described as sealed by either his brother Dadaga, or his cousin, Dadaga’s son,

Gududu, and finally, because Nigar-kidu’s brother Ur-Nigar also had a son named Luduga.

Nigar-kidu is only attested in MVN 1, 182 (from

∑

26), in which he certifies the

transfer of a large amount of barley from his brother Ur-Nigar to the palace. The document

is sealed with his seal along with the seal of a “grand scribe” of the king.

Luduga the son of Nigar-kidu never appears in any text sealed with his own seal; he

is, nevertheless, attested independently in a few texts;

383

it seems possible to claim that he

383. SA 76 (from AS 4): obv. ii 3; MVN 11, 162 (from

∑

38 xii): rev. 3; CHEU 30 (from

∑

46): obv. 7; BM

106050 (unpubl.) (no date): mentions the wife of Luduga, the son of Nigar-kidu (rev. v 12: dam lu

-ga dumu nigin

-gar-ki-du

153

Chapter 5: Section 3: The governor

When describing the ruling family of Umma it seems reasonable to begin with its

most prominent members, the governors. The office of the governor (Sumerian ensi

) of

Umma belonged throughout the entire duration of our documentation to the ruling family

of Umma. When we research the earliest generations of the Ur III society we are constrained

by the paucity of the sources, and it is speculative to comment on either the lineage or the

administrative activities of any Umma governor prior to Ur-Lisi. One text has survived

documenting the involvement of the ruling family of Umma in the administration of Umma

during the early days of Ur hegemony: DC 236, presumably from

∑

28.

385

Obv.

1. 1(geÒ

’

u) 2(geÒ

) 3(u)

3(aÒ)

?•

753 gur and 1 barig of barley

1(barig)

?•

Òe gur lugal

according to the royal measure,

2. ¿ki ab• -ba-mu ¿ensi

•-ka-ta

From Abbamu the governor,

386

3. ur-

¿li

-si

KA-guru

-ke

•

Ur-Lisi, the chief of the granary,

Rev.

1. Òu ba-ti

received.

2. x-gi-na-¿tum•  ... ]

¿d

Òara

•

385. The full formula of this year is:

mu en nam-Òita

Òul-gi-ra-ke

ba-gub-ba(-Òe

Òud

-sag) en

en-ki eridu

-ga dumu Òul-gi nita kalag-ga lugal uri

-ma lugal an ub-da limmu-ba-ke

ba-a-Ìun,

“Year:

EnnamÒita-∑ulgirake-bagubaÒe-Òusag,

the son of

∑ulgi

, the strong man, the

king of Ur, the king of the four quarters, was installed as en-priest of Enki in Eridu.”

386. One of the many problems with this text is the possibility to read /Abba mu

aldim ensi/ instead of /

Abbamu ensi/ (Abba the cook of the governor, instead of Abbamu governor).

154

3. giri

a-kal-¿la•  ... ]

Via Ayakala ... of

] Abbamu

ab-ba-mu ¿ensi

•

the governor

387

4. iti e

-iti-6(diÒ)

Month e-iti-6.

5. mu en

en-ki ¿x

ba• -

un?]

Year:

“

the En-priest of Enki

x x was installed in office

”

From the same year (

∑

28) another very important document has come down to us

(AAICAB 1, 1912-1143, mentioned above). It is a text computing a large amount of grain

(mainly barley, Sumerian

Òe

), totaling more than 15,000 gur, or more than four and a half

million liters. The first entry of that text, 12,066 gur, 3 barig, 2 ban and 9 sila according to

the

∑ulgi

-measurement,

388

is the exact yield-estimate corresponding to the 100 domain

units recorded in MCS 6, 83, BM 105334 (from AS 2) (see above).

389

The total is said to be

controlled by Ur-Lisi.

390

Since this document is dated to

∑

28, it is my belief that Ur-Lisi

operated in his capacity as chief of the granary (KA-guru

387. The break in the text could give either the official or the familial relationship between Ayakala and

Abbamu. A reading

ÒeÒ]

would conform with our knowledge of the familial structures of the Umma

ruling family, but it is obscured by the lack of any reliable genealogy of Abbamu. The kinship term

“brother” was used only in a limited number of relations, see Excursus 2.

388. AAICAB 1, 1912-1143, obv. 1: 3(guru

) 2(

geÒ

’u) 1(

geÒ

) 6(

aÒ

) 3(barig) 2(ban

) 9(

diÒ

) sila

1(u) 5(

diÒ

)

gin

Òe

gur

Òul

-gi

389. 12,066 gur, 3 barig, 2 ban and 9 sila divided by 20 is approximately 603 1/3. However, these two text

still present some problems, and the reconstruction here is only tentative. The texts mentioned here were

first discussed by K. Maekawa, however for slightly different purposes; see K. Maekawa, Zinbun 13

(1974) 1 - 60.

390. AAICAB 1, 1912-1143, rev. 11: ur-

-si

- dab

155

Abbamu has been mentioned in the preceeding as the earliest Ur III governor of

Umma, but he remains an almost legendary figure.

391

He is attested in very few documents

and no seal-impression of his has survived to our time. Essentially, all of the texts that have

been used as evidence for Abbamu as an early governor of Umma are ambiguous.

392

Other

possible early governors of Umma are A

ua,

393

Kuli,

394

and perhaps Malakum.

395

391. T. Maeda, "Father of Akala and Dadaga, governors of Umma," ASJ 12 (1990) 71, table 1.

392. Abbamu, the governor of Umma, is only mentioned in two texts, these are, NYPL 37 (from

∑

33), and

DC 236 (∑ 28?), mentioned above. NYPL 37

mentions the e

u sum-ma ab-ba-mu ensi

-ka, perhaps a

reference to the left-over account of the deceased governor. In CHEU 94 (from

∑

28 viii), the house of

Ur-saga the “slave” (ir

) of Abbamu is mentioned, also Ayakala is present in this short text.

393. See FAOS 17, 88* (dating to before

∑

33), obv. 26: u

-ba a-

u-a ensi

umma

, “on the day A

ua (was/

became) governor of Umma.” This is the only extant text which mentions A

ua as governor of Umma.

The text is very interesting in that it mentions the purchase of a house by Ur-Nigar, the son of GIRINI,

the chief of the galla. There are no other Umma texts mentioning A

ua dated to the early years of

∑ulgi

Some Umma texts without date mention A

ua; however, these did not mention any title of A

ua. See

for example the undated text TSU 92 (= RIAA 188) which has as a concluding line inim a-

u-a-ta, “on

the command of A

ua”, suggesting a high office for this person. On the terminology, inim PN-ta, see p.

160 below.

394. Only two references to Kuli, the governor of Umma, exists; Nebraska 44 (

from ∑

39 to AS 3), obv. iv

32:

giri

ku-li ¿ensi

•, and

Ontario 2, 425 (unpubl.).

395. Only one questionable reference to a person named Malakum serving as governor of Umma exists; BM

108029 (no date).

156

Chapter 5: Section 4: Ur-Lisi

Ur-Lisi is recorded as governor of Umma for the first time in the 33

year of

∑ulgi

seven years after numerous sources appear for the first time at Umma.

396

Ur-Lisi is attested

in a number of texts before

∑

33, when he perhaps held the position as the chief of the

granary (KA-guru

) of Umma.

397

This is not only inferred from the text given above (DC

236) but also from the administrative activities of Ur-Lisi before

∑

33.

398

Since Ur-Lisi is

mentioned as the son of Ur-Nigar, the chief cattle administrator (

ÒuÒ

399

and since both

Ir(mu),

400

and Ayakala,

401

well-known members of the ruling family, claimed to be

396. See the seal of Ur-Lisi, the governor (BMC Roma 8, 10 3 [from

∑

33 viii]); the seal of Ur-E’e dedicated

to Ur-Lisi, the governor (Aleppo 147 [from

∑

33]); the seal of Lugalebansa(g), dedicated to Ur-Lisi, the

governor (MVN 14, 133 [from

∑

33 xi]); the seal of

abaluke, dedicated to Ur-Lisi, the governor

(Syracuse 485 [from

∑

33 vii]); the seal of Atu, dedicated to Ur-Lisi, the governor (YOS 4, 151 [from

∑

32 ix to

∑

33 i]).

397. See also T. Maeda, "Ruler's Family of Umma and Control over the Circulation of Silver," ASJ 18 (1996)

255.

398. Ur-Lisi is mentioned in several texts from the early days of Ur III, delivering barley for the cult, for

example, delivering small quantities of barley for the regular offerings for

∑

ara in Hirose 341 (from

∑

i to vi); for rations, for example, delivering large quantities of barley as rations for the ox-drivers and the

engineer-troops (

-gu

Òa

-saÌar-ra:

the

ÒasaÌara,

male workers mainly specializing in excavating,

were often grouped together with the ox-drivers) in CHEU 2 (from

∑

31 iv to xi); and receiving barley

from various persons, for example, receiving large quantities of barley from the governor of Adab, on the

order of the governor of Umma in SANTAG 6, 16 (copy of Umma 92 = ICP 422) (from

∑

31 ix),

(sealed by Ir(mu) instead of Ur-Lisi, obv. 3 - 4:

ki ensi

adab

-ta / inim ensi

umma

-ta

, rev. 1 -

ur-

-si

/ Òu ba-ti / mu ur-

-si

-Òe

/ kiÒib ir

-gal

). Compare these examples with

the responsibilities of Ir(mu) described below Chapter 5: Section 11.

157

“brothers of the governor” (

ÒeÒ

ensi

) at the same time as Ur-Lisi held office, Ur-Lisi must

have been a member of the ruling family and a son of Ur-Nigar, the chief cattle administrator

(

ÒuÒ

399. See the seal of Ur-Lisi (ur-

-si

/ dub-sar / dumu ur-nigar

(NIGIN

)

gar

); AUCT 3, 286 (from

∑

31 xi);

AUCT 3, 427 (from

∑

31 xii); and MVN 21, 272 (from

∑

24 xi):

Obverse.

1. 6(geÒ

) Òe gur lugal

360 gur of barley,

2. uri

ki]

-Òe

to Ur,

3. giri

ur-

-si

via Ur-Lisi

Reverse.

1. dumu ur-nigar

(NIGIN

)

gar

ÒuÒ

-ka

the son of Ur-Nigar, chief cattle administrator.

2. iti pa

-u

-e

Month Pau

’

3. mu nig

-ka

-ak al-la-ka

Year:

mu 3(diÒ)-kam us

-bi

Only very few seal-impressions of the simple seal of Ur-Lisi are known since he ruled Umma almost

from the time when the first numerous archives appear. He came to be known primarily as “the

govenor.”

400. See BM 105554 (unpubl.) (from

∑

34 ii), obv. 3: ir

ÒeÒ

ensi

umma

, “Ir(mu), the brother of the

governor”.

401. OrSP 47-49, 500 (no date), obv. ii 15: a-kal-la

ÒeÒ

ensi

, “Ayakala, brother of the governor”, The text

also mentions Ir(mu) with the title “chief of the granary”, indicating that Ur-Lisi had already taken office

as governor, and Nigar-kidu (the uncle of Ur-Lisi) as “chief of the galla” (rev. iv 4: nigar

(NIGIN

)

gar

-ki-

ugula gal

-la

-gal). See also JCS 28, 215 26 (

from ∑

43): 3(u)

guruÒ

diÒ

Òe

/ ku

? / a-

Òa

en-gaba-ra

/ giri

a-kal-la /

ÒeÒ

ensi

kiÒib

a-kal-la / dumu lugal-nesag

-e / ugula da-da /

(blank line) / mu en

nanna

maÒ

-e i

-pa

, “ 30 work-days, carrying? fish to the field engabara, via

Ayakala brother of the governor, sealed by Ayakala son of Lugal-nesag’e, foreman: Dada. Year: “the En-

priest of Nanna was choosen”.” The obvious reason for the sudden need of an otherwise rare kinship

term was to clearly exclude any ambiguity in a text mentioning two persons, both named Ayakala.

158

During the early years of Ur-Lisi’s tenure it was the costom for office-holders in the

administration to have seals dedicated to the governor rather than to the king, as was

common in later years. More than 40 people held seals dedicated to Ur-Lisi.

402

Although

most of these dedicatory seals were used from the accession of Ur-Lisi in

∑

33 until around

the time of

∑ulgi

’s 40

year, it is not possible to say when, exactly, this custom was

abandoned. Some of the problems faced when using the seals of the Ur III administrators to

describe the social system of the state can be seen from the following few examples. Ayakala

had a seal dedicated to Ur-Lisi while he at the same time had a seal with no dedication;

Kugani had a seal dedicated to Ur-Lisi before the latter became governor, and a different seal

dedicated to Ur-Lisi after he had become governor.

403

Some seals dedicated to Ur-Lisi

remained in use long after Ur-Lisi’s death, when Ayakala had become governor. Nevertheless,

402. See for example the seal of Adaga (ur-

-si

/ ensi

/ umma

// a-da-ga / dub-sar / ir

-zu) in use from

∑

42 iv (MVN 2, 343) to

∑

43 xi (MVN 14, 9) (with one possible attestation from

∑S

2); the seal of

Ayakala (ur-

-si

/ ensi

/ umma

a-kal-la // dub-[sar] / dumu

¿ur•

-[nigar

]

¿gar•

[

ÒuÒ(?)

] / ir

-[zu])

in use from

∑

39 i (MVN 14, 3) to

∑

41 (ArOr 62, 238 I 867)), Aykala also had a regular seal without

dedication in use from

∑

33 until he became governor himself; the seal of Atu, son of Lugalsaga (ur-

/ ensi

umma

/ a-tu dub-sar / dumu lugal-sa

-ga / ir

-zu), in use

from ∑

33 (Nik 2, 189) to

∑

(MVN 18, 425)), and the seal of Atu, son of

∑eÒkala

(ur-

-si

/ ensi

umma

/ a-tu dub-[sar] / dumu

ÒeÒ

?-kal-la), in use in

∑

40 (JCS 28, 220 42)), and so forth. Several other examples can be cited, a full

study within the framework of the CDLI is in preparation by the author. See also R. Mayr,

Seal

Impressions

(1997) 117 - 120 and fn. 492.

403. Kugani son of Ur-

∑ulgi

(ur-

-si

/ ensi

umma

/ ku

-ga-ni / dumu ur-

Òul

-gi / ir

-zu) in use from

∑

34 vii (Aleppo 83) to

∑

38 viii (Aleppo 89). Rochester 206 (ur-

-si

/ ku

-ga-ni / dub-sar / dumu

ur-

Òul

-gi / ir

-zu) from

∑

32 ix.

159

it is seen as an attempt by the king to strengthen his influence in the provinces that officials

in the local administration had seals carved with a dedicatory inscription mentioning the

king rather than the local governor around year

∑

41.

404

As mentioned above, it is likely that Ur-Lisi was chief of the granary before he

became governor. At that time this office passed to his brother Ir(mu), who held that office at

least until the end of

Amar

-Suen’s reign, and presumably several years longer before his son

∑ara

-izu succeeded him.

405

It is very difficult to demonstrate any direct administrative contact between the

crown and the provincial administration. Only two letters to Ur-Lisi, the governor of Umma,

from the king have survived; the first (Scheil RA 24, 44)

406

is an order from the king to give

the Lama

ÌÌ

um-priest of Inanna of Girsu 60 gur (ca. 18,000 liter) of barley—it is written in

Akkadian.

407

The other letter (YOS 4, 117)

408

is an order to Ur-Lisi to give the messengers

of the king various objects; this letter is written in Sumerian.

409

It is possible that the first

letter should be dated prior to Ur-Lisi’s tenure as governor when he was chief of the Umma

404. H. Waetzoldt, “Änderung von Siegellegenden als Reflex der ‘großen Politik’,” in U. Finkbeiner (ed.),

Boehmer Fs

(Mainz: 1995) 659-64.

405. For a discussion of Ir(mu)’s tenure see below, Chapter 5: Section 11.

406. Edition by E. Sollberger: TCS 1, 369, see also P. Michalowski,

Letters from Early Mesopotamia

(Atlanta

1993) 55 (text 74).

407. The letter uses the normal Akkadian letter-fomula and introduces the most important person first; obv. 1

- 3:

um-ma Òar-ru-um-ma /

a-na ur-

-si

-na

/ qi

-bi

-ma

. A full presentation of the Ur III

letters in under-way within the framework of the CDLI.

408. Edition by E. Sollberger: TCS 1, 1, see also P. Michalowski,

Letters

(1993) 54 (text 73).

160

granary, however this question cannot be solved at present. The presence of messengers of the

king, and “followers of the crown” (aga

-us

) at the local court in Umma strongly suggests a

very real royal presence in Umma.

410

The direct involvement of the governor in the daily management of the province is

seen, explicitly in the references to “the command of the governor” (inim ensi

-ta). Such a

terminology resembles a practice known from many contemporary middle-eastern societies

where “the order of the king”, or the like, is analogous to a legally binding and written order.

The same phrase could easily be used with the title governor exchanged with the title of

another high-ranking official, however, primarily naming the person rather than the

office.

411

The phrase, “on the command of the governor” was sometimes used in inter-city

relations,

412

and its usage was not restricted to the Umma province.

413

SANTAG 6, 16, is an

instructive text, making use of the phrase “on the command of the governor” applied in an

inter-city operation. It dates to

∑

31 month 9, that is, two years before Ur-Lisi became

governor; hence, the governor referred is likely to have been Abbamu. Ur-Lisi was chief of

the granary already at this time, and apparently Ir(mu) was already his “apprentice.” The text

relates how Ur-Lisi received a large amount of barley and wheat from the governor of Adab;

409. The letter uses the normal third person voice thoughout, except for the phrase (reverse line 4 - 5)

kin-gi

-a-ga

/ Ìe

-na-ab-sum-mu > /lu kiñ.gi-a-mu-ak Ìe-na-b-sum-e/

“he shall give it to

my messengers ...”

410. See for example Nik 2, 287 (from

∑S

4 viii 21 to 25); Nik 2, 290 (from

∑S

6 viii 6 to 7); and Nik 2, 340

(from IS 3 i), all concerned with the provisions for royal messengers and emmisaries at Umma (on their

way to the east or on mission to Umma).

161

the transaction was sealed by Ir(mu) rather than Ur-Lisi, and it was part of the formal

exchange between the provinces otherwise known as the “bala”.

SANTAG 6, 16 (from

∑

31 ix):

Obverse.

1. 1(u) 8(aÒ) 3(barig) 4(diÒ)

18 gur 3 barig 4 2/3 sila of wheat,

2/3(diÒ) sila

gig gur lugal

according to the royal measure.

2. 1(u) 9(aÒ) 1(barig) 5(ban

)

19 gur 1 barig 5 ban 4 2/3 sila

4(diÒ) 2/3(diÒ) sila

ziz

gur lugal

of emmer according to the royal measure

3. ki ensi

adab

-ta

from the governor of Adab,

411. See, for example, AAS 92 (from

∑

39), rev. 3: inim a-kal-la-ta, and compare with BIN 5, 117 (from

∑

48), rev. 3: inim a-kal-la nu-banda

-ta. Other examples include AnOr 1, 75 (from AS 1), rev. 3: inim ur-

-e-ta; AR RIM 4, 22 (from AS 8 xii), rev. 1: inim ur-

nun-gal; and BIN 3, 549 (from AS 9 viii), rev.

2: inim lu

-kal-la. For examples of the same term used with another official than the governor, but

without mentioning the name of the person, see, for example, MVN 1, 173 (no date), obv. 2: inim

¿Òagina•

-ta; NYPL 258 (from AS 6 xii), obv. 3: inim sukkal-ma

-ta; SAT 2, 753 (from AS 3), obv. 6:

inim

Òa

-dub-ba-ta; and TCL 5, 6047 (no date), obv. i 3: inim lu

-kin-gi

-a lugal-ka-ta.

See also the letter: MVN 4, 182 (no date):

Obverse.

1. ur-¿nigar

•(NIGIN

)

gar

-ra

To Ur-Nigar

2. u

-na-du

speak:

3. udu lugal-

geÒ

gigir-re min-a-ba

“

The sheep of Lugal-Gigir, both of them

4. Òu Ìe

-na-a-du

-e

you shall release them to him!

Reverse.

1. inim ensi

-ka-ta-am

On the command of the governor.

”

412. See for example SANTAG 6, 16 (from

∑

31 ix), obv. 4: inim ensi

umma

-ta; TCS 1, 130 (no date),

rev. 5: inim e

-gal-kam; and UTI 4, 2972 (from

∑S

2), obv. 3: inim ba-ba-ti-

Òe

413. See for example TUT 168 (from

∑

44 ix), rev. 1: inim ur-

ba-ba

dumu ensi

-ka-ta; and TBM 1, 308

(from

∑

44 x), rev. 3: inim sanga

nin-gir

-su.

162

4. inim ensi

umma

-ta

on the order of the governor of Umma.

Reverse.

1. ur-

-si

Ur-Lisi

2. Òu ba-ti

received.

3. mu ur-

-si

-Òe

Instead of Ur-Lisi,

4. kiÒib ir

-gal

the seal of Ir(mu) is present.

5. iti ezem-

-si

Month “Festival of Lisi”.

6. Òa

bala-a

In the period of the bala.

7. mu a-ra

2(diÒ)-kam-aÒ

Year: “the second time

kara

ba-

Kara

ar was destroyed”.

When Ur-Lisi became governor he also became the head of a large household, and

from that time on he is almost exclusively referred to as the governor, or the governor of

Umma, and only very rarely with his own name. From the same time we also begin to learn

about the family of the governor.

Ur-Lisi’s wife was named Nin-melam. Ur-Lisi had a son, Lu-Ema

, with Nin-melam;

however, she was not his only consort. Ur-Lisi also had a concubine (lukur)

414

who bore him

another son, Ikala, as well as a daughter, Nin-Ekuta. Nin-melam is not attested prior to Ur-

Lisi’s tenure.

414. Three ration lists concerning the “concubines” of

∑ara

have been published; AAICAB 1, 1924-0668

(from

∑S

2), rev. iv 7:

-ba lukur

Òara

, “oil-rations for the concubines of

∑ara

,” (see also the

parallel text AAICAB 1, 1911-480 [from

∑

42]), AnOr 7, 296 (no date), rev. iv 9: [

Òe-ba

]

¿lukur•

Òara

, “barley rations for the concubines of

∑ara

.” Each text lists approximately 55 women; their

position in society remains obscure. See also Excursus 2.

163

ASJ 18, 163 nr. 6 (= BM 110263) (from

∑S

4), given here in excerpt, might throw

some light on the situation of women in Umma:

Obv.

4. 1(bur

’

u) 4(bur

) 3 (iku) GAN

14 bur 3 iku of land,

5. 2(eÒe

) 3(iku) GAN

kiÒi

(UDgunû) u

2 eÒe 3 iku of land with weed and grass,

6. 1(bur

) GAN

mur

(LAK 193)

1 bur of land of .....,

7. 1(iku) 1/2(iku) GAN

1 1/2 iku of land is dikes,

8. a-Òa

gid

-da

In the long field;

9. nin-me-lam

dam ensi

-ka

Nin-melam, wife of the governor.

10. 6(bur

) GAN

6 bur iku of land,

11. 1 1/2 (iku) GAN

1 1/2 iku of ruin hill land,

12. 1 (iku) GAN

a-muÒ-DU

1 iku of .....,

13. 1 (eÒe

) 1/4 (iku) GAN

a nu-e

1 eÒe 1/4 iku of land where

the water does not come down,

14. a]-Òa

Òara

In the field Du-∑ara;

15. ¿nin

?•

-e

-ku

-ta dumu lukur ensi

-ka

Nin-Ekuta, daughter of the

concubine of the governor.

The text is a calculation of the size of six plots in the province of Umma distributed

to high-ranking members of the local court.

415

The size of Nin-melam’s field is around fifty

times larger than the average field allotments to workers in Umma.

416

The field of Nin-

Ekuta, the daughter of Ur-Lisi’s concubine, is less than half that size.

417

Evidence exists that

Nin-melam also owned a date-palm plantation,

418

and that she continued to possess this

property even after her husband had fallen from power in AS 8.

164

The parentage of Nin-melam is not known. Ur-Lisi’s concubine is not named in the

extant records, nor is it certain that Ur-Lisi had only one concubine. However, the texts

mentioning Nin-melam, the wife of Ur-Lisi, suggest that she headed some kind of

independent household.

419

Nin-melam was not an uncommon name during the Ur III

period. One or more female workers in Umma bore that name, and at least one woman in

Nippur.

420

Ur-Lisi, the head of a large household, was heavily involved in the agricultural

production of the province. It is conceivable that all of the arable land within the province of

Umma was under the direct jurisdiction of Ur-Lisi. The same can be postulated for Abbamu

415. The total and the colophon of ASJ 18, 163 6 reads:

Reverse.

...

7. ∑U+NIGIN

1(bur

’

u) 6(bur

) 1(eÒe

) GAN

Total: 16 bur 1 eÒe of land,

8. ∑U+NIGIN

2(bur

) 1(eÒe

) 2(iku)

Total: 2 bur 1 eÒe 2

1/4(iku) GAN

murgu

1/4 iku of murgu land,

9. ∑U+NIGIN

1(iku) 1/2(iku) GAN

Total: 1 1/2 iku of land is dikes.

10. nig

-gal

-la

“

things present

”

11. lu

-sa

-ga in-gid

Lusaga measured

(blank space)

12. mu-us

-sa si-ma-num

ba-Ìul

Year after: “

Simanum

was destroyed”.

416. See also J. Dahl, AoF 29 (2002) 334.

417. Compare that to the field of the governor, which in one text (AnOr 1, 303) is recorded as more than 60

bur

. Compare with MCS 6, 83, BM 105334.

418. See MVN 16, 742. This text is among the documents reviewed by K. Maekawa, ASJ 18 (1996) 129.

419. YBC 3681 (from

∑

38), suggests that Nin-melam had access to means of her own, and was able to

contribute to the economy. YOS 18, 123 (from AS 9), testifies that she was able to continue these

activities even after the dethronement of her husband.

165

on the basis of AAICAB 1, 1912-1143. For understanding Ur-Lisi’s control of Umma

agricultural production we rely on MCS 6, 83, BM 105334, mentioned above. The text is a

survey of Umma land conducted in AS 2; the colophon reads, “account of the measured

field, concerning Ur-Lisi, governor of Umma.” It also mentions the prebend field of the

governor, which measured 60 bur, an area capable of yielding up to 1,800 gur of barley.

421

From other documents we know that the prebend field of the governor was made up of

smaller lots in many different fields.

422

Ur-Lisi ruled Umma from

∑ulgi

33 until the death of Amar-Suen sometime during

AS 8, in total 23 years. The exact date of Ur-Lisi’s fall from power cannot be established, but

it seems to coincide with the downfall of Amar-Suen.

423

We have already mentioned the

time of the death of Amar-Suen, the evidence of which is closely linked to the political

situation in Umma: from the sixth year of Amar-Suen, texts appear that label Ayakala as the

governor of Umma, and Ayakala used a seal which mentions

∑

u-Suen as the king of Ur in a

420. See, for example, NATN 530 (from AS 9 x 17), with the seal nin-me-lam

/ geme

nin-e

-gal. It is, of

course, intriguing to see the texts as evidence that Nin-melam, the widow of Ur-Lisi, went to Nippur

shortly after the death of her husband. This cannot, of course, be confirmed although another governor’s

wife (Ninheliya) is known to have traveled (see p. 174 below). Six out of the seven Nippur texts

mentioning Nin-melam date to betweem AS 9 vii and xii. Only one text is without year, that text (TMH

NF 1-2, 212) is, however, dated to month 11, day 7. All seven texts were concerned with the delivery or

receipt of wool and garments, and all transactions included a certain Lugal-magure. The texts are (in

chronological order): NATN 542 (from AS 9 vii 7); TMH NF 1-2, 195 (from AS 9 ix); NATN 530

(from AS 9 x 17); TMH NF 1-2, 197 (from AS 9 xi 22); and TMH NF 1-2, 229 (from AS 9 xii 7).

421. See also MVN 16, 934 (

from ∑S

3 iii); UTI 3, 1687 (

from ∑S

3); and MVN 21, 334 (

from ∑S

8),

three texts that all record the subsistence (

Òuku

) grain of the governor (see J. Dahl, AoF 29 (2002) 337 -

338).

166

dedicatory inscription.

424

Ayakala also has a seal dedicated to Amar-Suen, but he used it only

occasionally from AS 8 ix to

∑S

5 vii, that is, after Amar-Suen had already died.

425

The seal-

inscriptions of Ayakala from these years show no simple development; rather, the inscription

freely alternates between mentioning Amar-Suen and

∑u

-Suen and using the correct spelling

a-a-kal-la instead of the reduced form a-kal-la, both presumably nominalized /ayakala/.

426

During the six years following Ur-Lisi’s fall from power, and possible death, a group

of texts were written dealing with his possessions.

427

These texts all have the critical subscript

“the covered house of Ur-Lisi, the governor of Umma” (e

-du

- la ur-

-si

ensi

umma

422. The evidence for Ur-Lisi’s prebend fields is fairly limited, whereas there is abundant evidence for the

plots making up Ayakala’s and Dadaga’s allotment. See, for example, BIN 5, 276 (from

∑S

9):

1. 1(bur

) 1(eÒe

) 2(iku) 1/2(iku) GAN

1 bur, 1 eÒe, and 2 1/2 iku of field,

1(u) la

1/2(diÒ)-ta

with 9 1/2 furrows per ninda.

2. Òe-bi 3(u) 1(aÒ) gur

its barley is 31 gur

(equal to a yield ratio of ca. 21 gur per bur)

3. Òuku ensi

(is the) subsistence of the governor,

4. a-Òa

igi e

-maÌ-Òe

(in the) field

“in-front-of-the-Ema

,”

5. ¿a-Òa

• gid

-da buru

measured field, to be harvested.

6. mu e

Òara

ba-du

Year:

“

the house of ∑ara was built

”

423. The seal of Ur-Lisi dedicated to Amar-Suen was used until the 10

month of AS 8 (MVN 4, 74). See

also K. Maekawa, ASJ 18 (1996) 126ff.

424. A survey of the seal-impressions of Ayakala during the final years of Amar-Suen’s reign suggest that

Ayakala used two seals during this perid both dedicated to

∑u

-Suen, but with a different writing of his

own name (either /a/ or /a-a/). Note also Hirose 365 (from AS 7), with a possible attestation of a seal of

Ur-Lisi dedicated to

∑u

-Suen.

425. A survery of the seal-impressions of Ayakala’s seal dedicated to Amar-Suen shows that these date between

AS 8 month 9 and

∑S

5 month 7, and the Ayakala, almost without exception is written a-a-kal-la.

426. K. Maekawa, ASJ 18 (1996) 128.

167

ka).

428

Several texts from AS 9 suggest, together with the document AAS 81, an account

concerning the confiscated properties following Ur-Lisi’s fall from power, running several

years from the 12

month of AS 8 to the second year of

∑u

-Suen’s reign, that Ur-Lisi had

died some time during Amar-Suen 8. Based on MVN 16, 627 (see below), we may suggest

that Ur-Lisi’s tenure ended in AS 8 month 10 or 11.

427. The sequence of e

-du

-la texts concerning Ur-Lisi runs from AS 8 (AAS 81 is an account of the e

-du

la of Ur-Lisi the governor from the 12

month of Amar-Suen 8 till

∑

u-Suen 2) to

∑S

5 (MCS 6, 10,

BM 106041, and MCAS 1, 54, BM 106045), and perhaps even until

∑S

7 (YOS 4, 237). See pp. 207 -

209 + fn. 511 below.

428. See K. Maekawa, ASJ 18 (1996) 103-168. According to Maekawa the e

-du

-la was the formerly private

property of an official later confiscated due to political misunderstandings. Whereas I agree that the e

-la was indeed confiscated property, I disagree on the nature of this property which I see as an

allotment to an office rather than the private possessions of the officeholder.

168

Chapter 5: Section 5: Ayakala

Ayakala followed his brother as governor of Umma late in the year AS 8. He ruled

until

∑S

7 ii. The earliest credible mention of Ayakala as acting governor of Umma is from

the 11

month of AS 8. The first certain attestation of the transfer of the office of governor

from Ur-Lisi to Ayakala is MVN 16, 627.

MVN 16, 627 (from AS 8 xi):

429

Obverse.

1. 5(geÒ2) 3(barig) Òe gur

300 gur 3 barig of barley,

2. 2(u) 7(aÒ) 4(barig) ziz

gur

27 gur 4 barig of emmer,

3. Òe geÒ e

-a

levied(?) barley,

4. e

Òu sum-ma

of the remainder

430

5. Òa

-gal gu

-eden-na

in the palace of

Gu’dena

Reverse.

1. a-a-kal-la

Ayakala,

2. ensi

umma

governor of Umma.

(blank line)

3. iti pa

-u

-e

Month Pau

’

429. See also the parallel text MVN 16, 621 (from AS 8 xi).

430. The use of the technical term e

Òu

sum-ma is in Umma sources restricted to three chronological

periods, early, middle and late. The majority of the texts from the middle group are from AS 8, some

running from AS 8 to

∑S

2, most texts from AS 8 are dated to month 11. The texts from the early

period form no coherent block, the late texts seem to concentrate around

∑S

7 month 2. Since the texts

mentioning an e

Òu

sum-ma, with few exceptions, were restricted to these times of leadership change, I

find it likely that the term referred exclusively to property transferred from one person to another.

169

4. mu en eridu

ba-

Year: “the En-priest of Eridu was installed”.

Ayakala’s career prior to his tenure as governor is well known.

431

Although Ayakala is

only attested with the titles scribe (dub-sar), captain (nu-banda

), and finally governor

(ensi

), there is evidence to suggest that Ayakala held the office of chief household

administrator of the governor (

Òabra e

) during some part of the tenure of his brother, Ur-

Lisi.

The personal name Ayakala is rather frequently attested in the Umma records;

Ayakala, the chief of the leather workers was, apart from Ayakala, the son of Ur-Nigar, the

best known Ayakala in Umma. Since the chief of the leather workers and the son of Ur-Nigar

operated on entirely different levels of the economy, the two are rather easily distinguished.

Ayakala, the son of Ur-Nigar, is attested from

∑

432

until

∑

S 7—in total 32 years.

Texts from

∑

33 suggest that Ayakala may already have been the chief household

administrator from the year his brother, Ur-Lisi, took office as governor. Several texts testify

to the fact that Ayakala received the “field interest” from agricultural administrators as early

∑

33.

433

This function is understood as an important part of the office of the chief

household administrator.

434

It was only after

∑

36 month 10, however, that Ayakala came to

seal multiple tablets, each recording the delivery of a few dead sheep and goats, texts that has

been used in this study to build the sequence of holders of the office of chief household

administrator of the governor. Ayakala held that office until

∑

39 month 2, when he was

431. See also W. Yuhong, JAC 10 (1995) 130 - 134.

432. See for example MVN 14, 34 (from

∑

33 viii); and MVN 14, 100 (from

∑

33 viii).

170

succeeded by his brother Dadaga. The sequence of people holding the title chief

administrator of the governor’s household will be described below (pp. 226 - 229 + figure 9).

Ayakala used the title “captain” (Sumerian nu-banda

) from

∑

435

until AS 8, when he

became governor of Umma. He used that title primarily in documents concerning work

436

and the administration of fields.

437

While operating as chief household administrator of the

governor, Ayakala used the title captain only once.

438

I suggest, therefore, that while holding

the title captain, Ayakala effectively held the position as a provincial administrator.

439

433. MVN 14, 188 (from

∑

33 xi), obv. 2-4:

maÒ

Òa

-ga / a-

Òa

ka-ma-ri

/ giri

ur-

nin-su; MVN 14,

190 (from

∑

33), obv. 2-4:

maÒ

Òa

-ga / ki lugal-e

-ma

-e

ÒeÒ

a-ab-ba-ta; OrSP 47-49, 163 (from

∑

33), obv. 2-3:

maÒ

Òa

-ga a-

Òa

Òara

/ ki lugal-e

-maÌ-e-ta

; SAKF 58 (from

∑

33 v), obv. 2-3:

maÒ

Òa

lugal-ka / ki ur-

nin-su-ta; BM 108004 (from

∑

33), obv. 2-3:

maÒ

Òa

-ga a-pi

-sal

/ ki

Ur-e11-e-ta; and Aleppo 448 (from

∑

33), obv. 2-3: ku

maÒ

<a-

Òa

>-ga a-

Òa

muÒ

-bi-an-na / ki lugal-

geÒ

kiri

-ta.

434. Compare with P. Steinkeller, “The Renting of Fields in Early Mesopotamia and the development of the

concept of "Interest" in Sumerian,” JESHO 24 (1981) 113-145, in particular 116 - 119.

435. Two texts date to

∑

34 (MVN 18, 395 [from

∑

34 x]; and Syracuse 17 [from

∑

34 v]), and one text

dates to

∑

38 (JCS 40, 112 3 [from

∑

38 vii]).

436. See, for example, Aleppo 187 (from

∑

48 v); MVN 4, 11 (from AS 5); and MVN 16, 818 (from AS 8

iii).

437. See, for example, OrSP 47-49, 467 (from AS 1 i to ii); TCS 333 (from AS 5 v); and Syracuse 384 (from

AS 7 iii).

438. JCS 40, 112 3 (from

∑

38 vii).

171

One of the main theses of this study is that members of the ruling family and other

high-ranking administrators in Umma acted as supervisors of larger units of agricultural

lands; that is units larger than those administrated by either the captain of (plow-)oxen or the

scribe of ten oxen. That office is tentatively here referred to as the office of a provincial

administrator, since all of the people postulated to have held this office occasionally were said

to seal transactions relating to the agricultural administration with a particular seal of

Òatam

ship, a nam-

Òatam seal (kiÒib nam-Òa

-tam)

440

As governor Ayakala sealed transactions regarding the Umma bala contribution,

441

including Umma contributions to the feasts and festivals of the empire, in both the Umma

province and abroad.

442

As a natural consequence, the governor would appear in numerous

records receiving goods, and contributing to the exchange and the production of the

province. There is also evidence to suggest that the governor was the official responsible for

the major temples of the Umma province,

443

and less evidence in support of a vivid temple-

439. This is in part supported by the unpublished document Princeton 2, 497 (from

∑

47). Ayakala is

attested frequently as the supervisor of agricultural lands. In BIN 5, 117 (from

∑

48), Ayakala gave the

order (inim a-kal-la nu-banda

-ta) that Lugina should seal a document regarding field-work as a

provincial administrator. Lugina was member of the family of Dada discussed below, suggesting that

Ayakala was directly involved in the administration of a specific area of the agriculture. Ayakala was the

conveyer of the produce of a number of different tracts of land (CHEU 10 [from AS 1 xii]; OrSP 47-49,

467 [from AS 1 i to ii]; and BRM 3, 80 [from AS 7 x]), and he interacted with the chief household

administrator (first Dadaga, then Lukala) as the administrator of one or more large agricultural units

(BIN 5, 146 [from

∑

44]; and TCS 333 [from AS 5 v]).

172

household economy as seems to have been the case in Ur III Girsu.

444

In a great number of

texts recording deliveries from the governor’s household to members of the imperial court or

to deities, a certain Ur-

∑ulpa

’e was involved.

445

These documents were often sealed by the

440. Based on an analysis of the documents sealed with a nam-

Òatam

seal, it is my belief that that term

referred to the seal rolled on documents relating to a larger argricultural unit. Groups of 5 - 10

agricultural foremen, variably called nu-banda

or ugula, often worked closely together. See, for

example the texts: UTI 4, 2399 (from

∑S

2), a record of work-days weeding plots in Gu’dena and

MuÒbiana

; MVN 11, 164 (

from ∑S

4), a record of levied? barley from Gu’dena; AnOr 7, 313 (no

date), an account of total output of grain from Gudena and

MuÒbiana

; UTI 4, 2864 (

from ∑S

2), a

record of disbursements of wool and hides for the cultivators of Gudena and

MuÒbiana

; UTI 3, 2126

(

from ∑S

4 v), a record of fallen oxen and their fodder. All these texts mention some (or all) of the

members of the following group of agricultural foremen: Ur-Abzu, Ur-Enun(a), Lu-dingira, GuTAR, Ur-

Ninsu, Dada, Ur-Enlila, Ipa’e. These foremen are all attested with the titles foreman (ugula) (in relation

to the cultivators under their command), captain of (plow-)oxen (nu-banda

) (when relating to the

plow-oxen), and administrator (of domain units) (

Òabra

( gu

)) (when relating to the yield). Often the

same person from within the group, or a member of the ruling family sealed the transactions with his

nam-

Òatam

seal, forming the basis of my suggestion. Certain texts clearly relate the ruling family to

these groups (for instance MVN 16, 751 (no date), treated below), others only suggest such a

relationship. In the following, the title

provincial administrator

is used to describe the title of an

administrator of a larger unit of land covering 25 - 50 domain units. There is no concensus as to a

translation of the term

Òatam

(or

kiÒib

nam-

Òatam

), but it may relate to the physical seal or the act of

using another person’s seal.

441. See for example BIN 5, 82 (from

∑S

2), recording deliveries of butter-oil (i

nun) and “gazi-cheese” (ga

gazi) from the chief cattle administrator, Atu, destined for the palace, as part of the bala, sealed by

Ayakala, the governor. See also, for example , Aleppo 421 (from

∑S

2); AnOr 1, 199 (from

∑S

4 vi); and

Babyl. 8 Pupil 13 (from

∑S

5).

442. See, for example, BM 105508 (unpubl.) (from

∑S

2 iii), recording provisions for the sukkal

, and

JCS 40, 113 5 (from

∑S

4), recording deliveries for Babati. See also, for example, AnOr 7, 221 (from

∑S

1); BRM 3, 44 (from AS 8 iv).

443. See for example BIN 5, 2 (from

∑S

4 iv).

173

governor.

446

The governor seems to have sealed tablets relating to both the management of

the empire as well as the cult, and the Umma contributions to both, whereas the chief

household administrator of the governor sealed documents relating to the daily business of

the province.

Ayakala’s wife was named Nin

eliya.

447

It is likely that Nin

eliya headed her own

office, just as Nin-melam, her predecessor as primary consort of the governor had done.

Nin

eliya is never attested before her husband became governor, and when she is mentioned

in the official records it is often together with him.

448

444. As noted elsewhere in this study, the archives of the central administration of Umma seem to suggest that

the major temple-households of Umma were all run from the household of the governor. For example,

the flocks of the temple-households were all managed by a restricted number of chief cattle

administrators directly subordinate to the governor, and not to the chief administrators of any of the

temple households to which these flocks presumably belonged.

445. In transactions involving food-stuffs, the cup-bearer Ninmarka is often mentioned (CST 784 [from

∑S

3]). Ur-

∑ulpa

’e was active prior to Ayakala’s tenure (DC 257 [from AS 6 i]).

446. See, for example, CST 715 (from

∑S

2 vii).

447. According to P. A. Parr, “Nin

ilia: Wife of Ayakala, Governor of Umma,” JCS 26 (1974) 90, Nin

eliya

used two different seals with two different othographies. The first (used on the majority of the tablets)

reads: nin

i-li

-a / dam a-a-kal-la / ensi

umma

-ka, the second (according to P. A. Parr rolled only on

JCS 26, 99 1 [from

∑S

1 viii]. See now also MVN 18, 467 [from

∑S

1 viii]) reads: nin

-li

-a / dam

a-a-kal-la / dumu ur-

da-mu. The line layout of this seal does not, however, conform with what we

would expect, and it may refer to the wife of a person other than the governor named Ayakala (see also

R. Mayr,

The Seal Impressions of Ur III Umma

(unpublished dissertation: Leiden 1997) 106 + fn. 446).

174

One text suggests that Nin

eliya went to Nippur, and since we have witnessed the

“queen-dowager” making multiple cultic voyages, it is equally possible that the wife of the

governor traveled to Nippur for a similar purpose.

AnOr 1, 304 (from

∑S

4(?) ix):

1. ...] ¿x•

2. Òa-ad-da

∑adda.

3. 5(diÒ) gin

gur

-za-an muÌaldim

5 shekels x , Gurzan, the cook,

4. zi-ga nin

-Ìi-li

-a nibru

-Òe

gen-na

booked out for NinÌeliya,

having gone to Nippur

Reverse.

(blank space)

1. iti

-si

Month “Lisi”.

2. mu bad

mar]-tu ba-du

]

Year: “the Amurite wall was built”.

448. Nin

eliya received wool from Ur-Dumuzida (JCS 26, 99 1 [from

∑S

1 viii]; and JCS 26, 101 3 [from

∑S

2 viii]), as well as raisins and spices from Ur-Dumuzida (JCS 26, 102 4 [from

∑S

2]; and JCS 26,

106 8 [from

∑S

4]). She received leather products from Ayakala, the chief of the leather workers (

aÒgab

gal) ( JCS 26, 103 5 [from

∑S

3 iv]; JCS 26, 104 6 [from

∑S

3]; JCS 26, 105 7 [from

∑S

4 xi]; JCS 26,

107 9 [from

∑S

4]; and JCS 26, 110 12 [from

∑S

5]). She received livestock for the cult from Alulu, the

fattener (JCS 26, 108 10 [from

∑S

5 xii]; and JCS 26, 109 11 [from

∑S

5 xii]).

175

Chapter 5: Section 6: Dadaga

The next governor of Umma was also a brother of Ur-Lisi named Dadaga. Dadaga

ruled Umma from

∑S

7, month 2, to at least IS 3, when sources become so scarce that is

uncertain who was governor.

The end of Ayakala’s tenure as governor of Umma in

∑S

7 seems not to have been

related to any dramatic events outside of Umma, and it is therefore likely that Ayakala died

of old age, judging also from his long career in public service. Accepting that the e

-du

-la of

Ur-Lisi was related to his misfortunate fall from power it is likely that exactly the fact that

Ayakala’s end was not related to any crisis explains why no e

-du

-la concerning him was

recorded.

Among the documents registering the transfer of office between Ayakala and Dadaga

we find one that records the transfer of household objects from the former governor to the

new.

Nik 2, 528 (from

∑S

7 ii):

Obverse.

1. 3(diÒ) gal zabar

3 large bronze(-something?)

2. 3(diÒ) za-

um zabar

3 za

of bronze

3. 1(diÒ) Òen a

-la

zabar

1 bronze ala drum

4. 1(diÒ) gi-gid

zabar

1 bronze measure stick

5. ki-la

-bi 5(diÒ) ma-na

its (combined) weight: 5 mana

Reverse.

176

1. 1(diÒ) urudu Òen

1 copper drum

2. ki-la

-bi 2(u) ma-na

its weight: 20 mana.

3. 1(diÒ) urudu Òen-Òu

1 Òu-drum of copper,

4. ki-la

-bi 5(diÒ) 2/3(diÒ)

its weight: 5 2/3

5. e

¿Òu!(KI)• sum-ma

Of the e

-Òu-suma

6. ki a-a-kal-la ensi

-ta

From Ayakala, the governor,

7. da-da-ga ensi

Dadaga, the governor,

8. Òu ba-ti

received.

9. iti maÒ-da

Month “Eating the gazelle”.

10. mu ma-da za-ab-Òa-li ba-

Year: “the district of

ZabÒali

was destroyed”.

Another set of records related to the transfer of office from Ayakala to his brother

Dadaga were subscribed as /sila/. These /sila/-texts were exclusively concerned with livestock

and its products, but they were written at all times and not only coinciding with rulership

changes, although they could be part thereof.

449

One of these texts concerned the transfer of livestock from the old governor, Ayakala,

to the new governor, Dadaga.

YOS 4, 237 (from

∑S

7 ii):

Reverse, column 8.

(blank space)

1. ∑U+NIGIN 1(Òar

) 1(geÒ

’

Grand total: 4,378 assorted sheep,

2(geÒ

) 5(u) 8(diÒ) udu

i-a

2. ∑U+NIGIN 2(geÒ

’

u) 4(geÒ

)

Grand total: 1,488 assorted goats,

449. /sila/ (mostly written si-il

-la) is perhaps related to sila.a = on the road. See R. Englund,

Ur-III-Fischerei

(1990) 41 + fn. 142 quoting K. Butz. The Umma texts mentioning the term /sila/ are few and do not

form any meaningful sequence.

177

4(u) 8(diÒ) ud

maÒ

i-a

3. 1(Òar2) 3(geÒ

’

u) 7(geÒ

) 4(u)

5,866 assorted sheep and goats.

6(diÒ) udu maÒ

i-a

(blank line)

4. udu si-il

(ILxKAR

)-la

“sila”

sheep,

5. ki a-a-kal-la ensi

-ta

From Ayakala the governor,

6. da-da-ga ensi

-ke

-dab

Dadaga the governor seized.

7. Òa

¿umma• 

...]-

ba-ba

Òu-ur

-zi In Umma x-Baba and ∑urzi

8. iti maÒ-da

Month “Eating the gazelle”.

9. mu

Òu-

suen lugal uri

-ma-ke

Year:

“

∑u-Suen, the king of Ur

ma-da za-ab-Òa-li

mu-

destroyed the district of ZabÒali

”

When a new governor took office, perhaps in the middle of a year, it is likely that the

entire “estate” of the governor was inventoried. At the time of a peaceful change of office, it is

likely that the estate was simply transferred to the debits of the new governor. At the time of

a hostile takeover, when the old governor was ousted it is possible that the estate was taken

apart and the “operating balance” of the province settled, through the administrative means

of an e

-du

-la, the confiscated property of the old governor.

The two documents quoted above help us to establish the time of the transfer of

governorship from Ayakala to Dadaga at around the first or the second month of

∑S

Dadaga’s career mirrored that of his brothers, he too held the office of chief

household administrator. And although he is only attested once with the title captain, it is

still likely that he too acted as a provincial administrator.

450

Since Dadaga was the only high-

178

ranking person in the Umma province with that name, it is unlikely that we will ever find

any substantial documentation of Dadaga using any title other than scribe, and eventually

governor, after

∑S

Dadaga is attested often from

∑

30 until

∑

39, month 9,

451

when he became chief

household administrator of the governor, following Ayakala; however, no seal-inscription

mentioning Dadaga is known prior to AS 9 (perhaps AS 8)

452

All of the texts said to be

sealed by Dadaga prior to

∑

43 were sealed with the seal of Luduga, the son of Nigar-

kidu.

453

Texts said to be sealed by Dadaga between

∑

43 and AS 9 were all sealed with the

seal of Luduga, the son of Ur-Nigar.

454

Although it may seem likely that the two people

called Luduga were identical, it still remain pure speculation.

455

Dadaga sealed documents relating to the agricultural administration prior to

becoming chief household administrator, and we should probably include him among the

450. RA 10, 210, BM 103413 (from AS 4 i to xii).

451. See MVN 14, 66 (from

∑

39 ix).

452. Dadaga’s simple patronymic seal reads da-da-ga / dub-sar / dumu ur-nigar

(NIGIN

)

gar

ÒuÒ

. See for

example UTI 4, 2424 (from AS 9); OrSP 47-49, 395 (from

∑S

1); and UTI 4, 2909 (AS 8).

453. For the seal of Luduga, the son of Nigar-kidu, see, for example, MVN 14, 10 (from

∑

42 i): ur-

-si

ensi

/ umma

/ lu

-du

-ga // dub-sar / dumu nigar

(NIGIN

)

gar

¿ki•

-[du

] / gal

-la

-gal / ir

-zu.

454. For the seal of Luduga, the son of Ur-Nigar, see, for example, MVN 10, 201 (from

∑

48 vii): lu

-du

-ga

/ dub-sar / dumu ur-nigar

(NIGIN

)

gar

455. Note in this regard that the seal of Luduga, the son of Nigar-kidu was dedicated to Ur-Lisi, and that it

seems to have been rolled for the last time in

∑

42 month 1 (see MVN 14, 10), corresponding well with

the change in seal-inscriptions noted above.

179

members of the ruling family holding office as

provincial administrator. It is possible that

Dadaga became

provincial administrator

once again after leaving office as chief household

administrator of the governor at the end of the reign of

∑ulgi

456

The wife of Dadaga, Nin-gizkimzi, is attested in very few texts, and never prior to her

marriage to Dadaga. Nin-gizkimzi was a common Umma female name; two concubines of

∑ara

were also named Nin-gizkimzi.

457

The seal of Nin-gizkimzi is preserved on only one tablet, UTI 4, 2898 (from

∑S

7 to

8), a receipt of a number of containers (

pisan) received from Zugali, and entered into the

house of the governor.

458

The seal reads:

459

Seal.

1. nin-gizkim-zi

Nin-gizkimzi,

2. dumu ur-gi

-par

child of Ur-Gipar,

3. dam da-da-ga

wife of Dadaga.

The last two texts that mention Dadaga as governor were both from IS 3. One,

AUCT 1, 304 (from IS 3), is a text dealing with large amounts of barley, presumably the

456. Last text concerning the delivery of dead animals said to be sealed by Dadaga is Aleppo 395 (from

∑

xii).

457. AAICAB 1, 1924-668 (from

∑S

2): obv. i 12 and 24; and AnOr 7, 296 (no date): obv.i 15’ and rev.iii 1.

Since we do not know anything about the social status of either the “concubines of

∑ara

”, or the wives of

the governors prior to the tenure of these men, it is improper to exclude the possibility that Nin-

gizkimzi, the “concubine of

∑ara

”, was identical with the wife of Dadaga.

458. Rev. 4: e

¿ensi

•

(KWU 636)-ra.

459. Since it is not likely that Ur-Gipar (a man) was the wife of Dadaga (another man), we need to reconsider

our rigid interpretation of the structure of the seal-inscriptions and also reconsider our rejection of the

seal of Nin

eliya (cited above), dismissed by R. Mayr, as the seal of the wife of Ayakala the governor.

181

Chapter 5: Section 7: The children of the governors: Lu-E

Ikala, Nin-Ekuta, Namzitara, Gududu, and

Lu-∑ulgi(ra)

The persons discussed in this section are of great importance for understanding neo-

Sumerian rules of succession. Having reviewed the succession of the office of the governor of

Umma we may ask, “why did the brother of the governor, and not his own son succeed

him?” To answer this question it seems reasonable first to limit our question to, in particular,

the cases where the governor had not yielded to political misunderstandings, i.e., supposedly

the cases of Ur-Lamma in Girsu and Ur-Lisi in Umma. In other words, taking into

consideration Maekawa’s interpretation that the fall of Ur-Lamma, governor of Girsu,

affected the social standing of his entire family,

463

it makes sense to suggest that the office of

the dismissed governor, would not be handed on to his sons, but rather passed on to his

brother or even out of the hands of the ruling family of that city altogether. In Nippur, it

seems that the ruling family lost its influence early in the reign of Amar-Suen and regained its

former positions only with the accession of

∑

u-Suen. The majority of succesions to high-

office in Nippur happened not at the times of leadership changes in Ur. In Umma, as we

have seen, the office of the governor remained at all times within one family. Succession

passed from brother to brother regardless of whether each individual case can be viewed as

independent or in relation to succession conflicts at the royal court. The question posed in

463. K. Maekawa, ASJ 18 (1996) 121 - 122.

182

the beginning of this section is therefore justefied by the conditions. To answer that question

we will now describe the careers of the sons of the govrnors of Umma.

It is the thesis of this study that the sons of the governor faced two challenges, first of

all, that the system of succession during the neo-Sumerian period did not, in particular, favor

primogeniture, and secondly, that the private sphere of the economy as well as the state

administration could offer valuable opportunities for a high-ranking member of society.

Should the prospective heir decide to withdraw from the line of succession, he might also

very well have disappeared from the official records of the state. It is equally likely that each

member of the ruling family (and other privileged families), even after venturing into the

private sphere of the economy, could (or were obliged to) re-surface in the state

administration on special occasions to partake in official transactions for whatever purpose.

The extant documents have only revealed the names of three of Ur-Lisi’s children—

whether he had more remains speculation. He fathered these three children with two women.

His wife Nin-melam was the mother of Lu-Ema

, his concubine was the mother of Ikala and

Nin-Ekuta.

Lu-Ema

, the son of Ur-Lisi and Nin-melam, had his own seal with an inscription

calling him the son of Ur-Lisi, governor of Umma. The seal lacks a dedication.

464

This seal

464. See for example, AAICAB 1, 1911-190 (from

∑S

1 vi), or MVN 4, 92 (from AS 7 vii). In this seal-

inscription, Ur-Lisi is sometimes spelled ur-

-si

-na and sometimes only ur-

-si

. Likewise Lu-Ema

is sometimes spelled Lu-Ema

-e. It is impossible to say whether he had more seals, if the inscription had

been recarved, or whether this is the result of inaccurate transliteration untill adequate photo

documentation of Ur-III texts is available.

183

was, however, never used in any transactions where Lu-Ema

was mentioned as the sealing

party in the texts; in fact, Lu-Ema

is never attested in any text outside of his seal-

inscription. Lu-Ema

’s seal was used exclusively on tablets said to be sealed by Lu-

∑ulgi

(ra)

or Luduga. The majority of these seal-impressions are found on tablets mentioning Lu-

∑ulgi

(ra); only three texts from the 8

month of AS 7 were said to be sealed by Luduga, but

rolled with the seal of Lu-Ema

465

Texts sealed with Lu-Ema

’s seal were dated to the years

AS 7, AS 8 and

∑S

1. No texts from AS 9 have been recovered. All these documents recorded

minor transactions.

466

Both Luduga and Lu-

∑ulgi

(ra) can be identified as members of the ruling family of

Umma. Both were probably Lu-Ema

’s uncles, although Lu

-∑ulg

i(ra) might have been his

cousin (for a discussion of Luduga and

Lu-∑ulgi

see below).

467

Lu-Ema

is nowhere

attested outside of his seal-inscription, and he remains a difficult character to understand.

465. MVN 1, 128; NYPL 301; and SACT 2, 229.

466. For example, the delivery of one ewe with fleece (1(

diÒ

) u

bar gal

), and one male sheep with fleece

(1(

diÒ

) udu nita bar gal

from Urru) (see SACT 2, 229 [from AS 7 viii]); or ten work-days (1(u)

guruÒ

diÒ

Òe

) to punt a boat (that is, pushing the boat up-stream with a punting-pole, Sumerian; ma

gid

-da) from ka-i

-da (literally “the mouth of the river”) to Umma, and to unload the barley (

Òe

ba-al-

la, literary “excavate the barley”), credited to the overseer Nigdupa’e (see MVN 16, 1526 [from AS 7]).

467. The seal on AAICAB 1, 1924-698, a document certainly dated to AS 8 viii (and not

∑

28 as suggested

by J.-P. Grégoire in his notes to the text), has the sole attestation of a seal of Lu

-∑ulg

i(ra), son of Ur-Lisi,

the governor of Umma. This reference is excluded from this investigation since the reading of the seal

inscription is doubtful. A person named Lu-

∑ulg

i(ra) is atested as son of both Dadaga and Ur-Nigar.

184

The whereabouts of Ikala,

468

Ur-Lisi’s son by a concubine, is even more obscure.

Ikala the foreman of the Umma textile factory was the highest ranking person by that name

in Umma.

469

His familial relations are not well understood. This Ikala should not be

confused with the agricultural administrator by the same name. Ikala, the chief of the

weaving-mill, seems to have been a high-ranking member of society, fulfilling some of the

duties also carried out by other members of the ruling family such as weighing out wool and

garments (PN in-la

470

but the sources are silent about his familial relations.

It remains speculation at this point to assert whether Ikala, the son of Ur-Lisi and a

concubine, is identical to Ikala the high ranking Umma textile official, or whether he

remained entirely outside the sphere of the state administration.

468. In both UTI 3, 2139 (no date), and SANTAG 6, 192 (from AS 8 vii), we read: i

-kal-la dumu lukur

ensi

. Note SAT 3, 1765 (

from ∑S

6), obv. 3: i

-kal-la <dumu> lukur!

469. According to H. Waetzoldt, UNT (1972) 101, Ikala held the title dub-sar tug

gada, a title translated by

Waetzoldt as “Schreiber für Stoffe und Leinen.”

470. The term PN in-la

was used exclusively about garnments and metal-wares. It is likely that the official

who weighed these objects was responsible for the accuracy of the weights. A limited number of

administrators weighed garnments and metals, for garnments these were: the governor (14 texts from

∑

46 iii to AS 9 vi); Dadaga (2 texts from

∑

47 x to

∑

48 vi) (both texts together with Ikala); Ikala (8 texts

from

∑

47 x to

∑S

2 iii); Ur-E’e from (15 texts from AS 2 viii to

∑S

5 iv); Lukala (7 texts from AS 7 vi

∑S

3 vi); Dingira (one text in AS 7viii); Gududu (2 texts from

∑

S 9 to IS 1 i). For metal wares these

were: Ur-

∑ara

(72 texts from

∑

36 ix to AS 7 vii); the governor (10 texts from

∑

45 xii to AS 1 v);

Ir(mu) (3 texts from

∑

46 vi to AS 1 vii); Lukala (with Ur-

∑ara

) (2 texts from AS 4 xii to AS 5 i); Lukala

(alone) (9 texts from AS 7 xii to

∑S

7 i); Ur-Nungal (the son of Ur-

∑ara

) (3 texts from AS 5 vi to IS 2

vi); Dingira (2 texts from AS 7 v to

∑S

1 vii); and

Ea-Òær

(1 text in

∑S

7 viii). The ruling family of

Umma did not entirely control weights and measures in Umma; rather, members of the archivist family

of Ur-

∑ara

partook in this important administrative activity.

185

Nin-Ekuta, the daughter of Ur-Lisi by a concubine, is known only from the text

quoted above and remains otherwise completely unknown.

The extant documents from Umma mention no important children of Ayakala, the

following governor of Umma.

471

Three people in Umma had seals claiming to be sons of

Ayakala, but these three can be neither excluded, nor can it be positively asserted that they

were the sons of the governor Ayakala.

472

The last governor of Umma, Dadaga, had one son who is particularly well known.

This son, named Gududu, figures among the most active Umma administrators at any time.

He held an important office in the administration of Umma, an office also held by his father,

Dadaga, his uncle Ayakala, as well as by his cousin, Lukala. It is possible that he was

designated as his father’s successor to the governorship of Umma. Gududu will be discussed

at the end of this section.

Another son of Dadaga, Namzitara—known only from two fragmentary seal

impressions naming Dadaga, the governor of Umma, as his father

473

—is scarcely present in

the administrative records. Among the other sons of Dadaga, we find Lu-

∑

ulgi(ra), who like

471. Lu-dingira mentioned by W. Yuhong, JAC 10 (1995) 134, cannot, for certain be included as a son of

Ayakala.

472. Ur-Gipar (see, for example, MVN 1, 85 [from

∑

48]; and MVN 9, 202 [from

∑S

4 i]). Lu-sa’izu (see,

for example, BIN 5, 138 [from

∑S

1]; and MVN 16, 1287 [from

∑S

3]). Abba (see, for example, MVN

21, 158 [from

∑

47 vi]). Abba is only attested in 5 texts from

∑

46 and

∑

47.

473. See SACT 1, 122 (from

∑S

8 x), and SACT 1, 124 (from

∑S

9 i). Another similar seal inscription which

does not include Dadaga’s title is known from SACT 2, 241 (from

∑S

9 iv), a text dealing with the same

commodities as the two previous ones: sheep hides as regular offerings for

∑

ara of Apisal.

186

Gududu had a seal dedicated to the king. Since the inscription of

Lu-∑ulgi(ra)

is from the

time before Dadaga became governor, calling Dadaga scribe, it lacks the second piece of

evidence necessary to rule out entirely any confusion regarding the identity of the seal-holder.

In this study we have already encountered a person called Lu-

∑ulgi

(ra) more than

once, and we can now distinguish three diferent persons with that name: a son of Dadaga

was called

Lu-∑ulgi(ra)

, someone who used the seal of Lu-Ema

, the son of Ur-Lisi, was

named

Lu-∑ulgi(ra)

, and finally Ur-Nigar the chief cattle administrator had a son called

Lu-∑ulgi(ra)

. Therefore,

Lu-∑ulgi(ra)

might hold the key to understanding the practice—

well-known from Ur III sources, and a continuos source of confusion—of people supposedly

altering their lineage/parentage.

All the texts in which

Lu-∑ulgi(ra)

sealed with his own seal or used the seal of Lu-

Ema

, produce, when arranged according to date, a very intriguing sequence. Namely,

Lu-

∑ulgi(

ra) is shown to have used the seal of Lu-Ema

, his cousin, from AS 7 to

∑S

1. In the

following two years,

∑S

2 and

∑S

3, he used a seal in which Ur-Nigar is named as his father,

and finally from

∑S

5 to

∑S

Lu-∑ulgi(ra)

sealed with a simple patronymic seal claiming to

be the son of Dadaga. From IS 1 to IS 3 he used a dedicatory seal mentioning the king at Ur,

still claiming Dadaga to be his father.

The transactions sealed by Lu-

∑ulgi

(ra) do not form a perfect continuum; for

example, it was only when he sealed receipts of dead sheep and goats that he used the seal of

Lu-Ema

. However, if we compare MVN 16, 1526 (from AS 7), UTI 3, 2217 (from

∑S

3),

187

and SNAT 522 (from

∑S

8), three primary receipts concerning work, such a continuum may

become apparent.

MVN 16, 1526

UTI 3, 2217

SNAT 522

(from AS 7)

474

(from SS 3 iv)

475

(from ∑S 8)

476

Obverse.

1. 1(u) guruÒ

1. +3(u) 2(diÒ) guruÒ

1. 3(u) 2(diÒ) guruÒ

1(diÒ)-Òe

n-Òe

]

1(diÒ)-Òe

2. ka i

-da-ta

2. ki-su

lugal

2. kab

-ku

sal

-la

-ka

¿gub•-ba]

gub-ba

3. umma

-Òe

4. ma

gid

-da

5. Òe ba-al-la

Reverse.

1. ugula nig

-du

-pa-e

3. ugula lugal-ku

-zu

3. ugula a-gu

2. kiÒib lu

Òul-gi

4. kiÒib lu

Òul-gi-ra

4. kiÒib lu

Òul-gi-ra

(seal)

Reverse.

5. iti nesag

(Seal)

Reverse.

3. mu

u-nu-ri

]

6. mu si-ma-num

1. mu

Òu-

suen lugal

ba-

uri

-ma ma

-gur

474. Translation: Ten work-days, punting a boat from Ka-id’a to Umma, emptying the barley (from the boat).

Foreman: Nigdupa’e. Sealed by Lu-

∑ulgi

(ra). Year: “

nuri was destroyed”. Seal: Lu-Ema

, scribe,

son of Ur-Lisi, governor of Umma.

475. Translation: Thirty-two (+?) work-days(?), spent at the treshing-floor of the “royal-canal.” Foreman:

Lugal-kuzu. Sealed by Lu-

∑ulgira

. Month “First fruits”. Year: “Simanum was destroyed”. Seal:

Lu-

∑ulgi

, scribe, son of Ur-Nigar, chief cattle administrator.

476. Translation: Thirty-two work-days, spent at the pond of the Salla-canal. Foreman: Agu. Sealed by Lu-

∑ulgira

. Year: “

∑u

-Suen, the king of Ur, fashioned the lofty barge of Enlil”. Seal: Lu-

∑ulgi

(ra?), scribe,

son of Dadaga.

188

en-lil

-la

-ra mu-na-dim

Seal.

1. lu

-e

-ma

1. lu

Òul-gi

1. lu

Òul-gi-ra]

2. dub-sar

3. dumu ur-

-si

3. dumu

3. dumu da-da]-¿ga•

4. ensi

umma

ur-nigar

(NIGIN

)

gar

ÒuÒ

It is, of course, possible to hypothesize that only one

Lu-∑ulgi(ra)

was behind the

four seals, as can be suggested from the three texts just quoted. Indeed, some of the tablets

witnessing

Lu-∑ulgi(ra)

’s involvement in the administration of labor do increase in

complexity, suggesting the prosperous career of a single person.

477

A final investigation concerning the identity of

Lu-∑ulgi

(ra) concerns the spelling of

his name. Is the spelling of the name

Lu-∑ulgi(ra)

—with or without final -ra—dependent

on the person sealing? Is it a result of the language reform which is supposed to have been

implemented with the coronation of

∑u

-Suen, or is it perhaps completely random and an

expression of the character of neo-Sumerian archival orthographic practice? It does seem that

the spelling with a final -ra is more prevalent in the later texts, when

Lu-∑ulgi(ra)

used a

seal naming Dadaga as his father. In the earliest texts, however, as well as in the texts giving

the patronym Ur-Nigar, the chief cattle administrator, in the seal-inscriptions, the writings

with affixed -ra prevail. This spelling is the most correct, according to our understanding of

Sumerian grammar, and conforms with our suggestion that reforms took place around the

477. See, for example, SNAT 516 (from

∑S

7); and BM 105340 (unpubl.) (from IS 1).

189

coronation of

∑u

-Suen, which would enforce a more correct orthography, such as “a-a” over

“a”, supposedly for /aya/.

The evidence is not conclusive, but does advocate, a more complex scenario than

previously considered. It is possible to speculate that

Lu-∑ulgi(ra)

, a son of Dadaga, worked

with his younger cousin, the son of the governor, and used his seal. When Ur-Lisi lost power,

Lu-∑ulgi(ra)

might have had a seal claiming to be the son of Ur-Nigar, the chief cattle

administrator, a seal which would have granted him some immunity and detached him from

the fallen governor. After the clarification of succession, he would at last have been able to

acknowledge his relations and allegiance to Dadaga, the later governor. But it is equally

possible to suggest that two different persons were called Lu-

∑ulgi

(ra): one the son of Ur-

Nigar and the other a son of Dadaga.

Lu-Ema

’s seal was rolled exclusively on tablets commonly sealed by other members

of the ruling family, that is, receipts of a few dead animals—belonging to the same group of

texts mentioned before and treated below—or rather simple primary receipts from the

administration of the agricultural work.

478

In sum, it appears that Lu-Ema

was obliged to

roll his seal on a number of documents, but otherwise did not partake in the affairs of the

state-run administration. The reason behind this practice is unknown, but we might propose

two sollutions. First, that Lu-Ema

was a minor whose responsibilities were taken care of by

his uncles, and that he died before he came of age and was able to conduct his own affairs, or,

secondly, that Lu-Ema

was primarily occupied in the private sphere of the economy, but

478. These documents were of the same sort often sealed with a nam-

Òatam

seal.

190

maintained certain privilegies or obligations for which he made available his seal to his

uncles. Under either scenario, Lu-Ema

had to seal certain administrative activities, mainly

the regular recipts of (a few) dead animals from shepherds, and the records of the same type

as the documents sealed by the provincial administrator, that is, documents relating to a

particular group of agricultural overseers.

R. Mayr has argued for a different solution and has streamlined the genealogies of a

number of families by introducing the possibility of the use of pseodonyms in the Ur III

onomasticon.

479

Mayr’s prime example was the very large and complex family of agricultural

overseers, in this study referred to as the family of Dada. However, the interpretation offered

here eliminates the need for streamlining any genealogies. Attention should be paid to the

system of familial involvement in the state-economy, and especially the system of

apprenticeships amply attested in the administrative sources, and explicitely described in the

key text SAT 2, 77 (from

∑

33 vi), concerning the family of Dada.

480

Also, the possibility

that people would readily vacate their official positions, letting family members take over

their responsibilities, must be further studied in order to understand the complex lineages of

neo-Sumerian families.

The most famous, and certainly the most important among all the sons of the three

governors of Umma was Gududu, the son of Dadaga. Gududu, who is first attested in AS

479. R. Mayr,

Seal Impressions

(1997) for example pp. 140, 147, 149, and 150 - 152.

191

481

operated as a high-ranking state administrator until the end of Ur domination over

Umma. He is mentioned several times in the last 20 texts of Umma documentation dating to

IS 4. In his early years, Gududu used the seal of Inim-

∑ara

, his uncle. Gududu’s own seal was

dedicated first to

∑u

-Suen, then to Ibbi-Suen, in both cases naming him a son of Dadaga—

in the latter seal Dadaga is mentioned as the governor of Umma.

480. Dada might have had as many as ten sons, who all, except for

∑eÒani

, functioned as mid-level

administrative overseers. The fact that one, or perhaps two, of

∑eÒani

’s nephews were also named

∑eÒani,

coupled with the intricate chronological distribution of the use of the seals and the complex

relationship between the information in the body of the texts themselves and the inscriptions of the seals,

forced R. Mayr to suggest that not only did

∑eÒani

’s father—according to Mayr Lugal-niglagar’e—use

the pseudonym

∑eÒkala

, but moreover,

∑eÒani

would, towards the end of his tenure, use a seal naming

his grandfather as his patronymic, following a period where he had referred to either Lugal-niglagar’e or

∑eÒkala

as his father. This interpretation is entirely based on the speculation that the term dumu can

refer to anyone but the father of the person in question, and on the possibility of the existence of

pseodonyms in the Sumerian onomasticon, as well as the assumption that the same Lugal-niglagar’e,

who was the son of Dada, was the father of

∑eÒani

The seal of Lugal-ezem and that of (his brother) Ur-Ema

e appears on about fifty tablets from AS 7 to

∑S

6, all dealing with agricultural work. In the same tablets, a certain Agugu is named as the sealing

party. This Agugu can be shown to have belonged to the same social group of persons as Lugal-ezem, Ur-

Ema

e, and the twelve other agricultural overseers mentioned in the same fifty texts. This situation is

not unique to these people; actually, it can be observed frequently with regard to the same group of

tablets. SAT 2, 77 (from

∑

33 vi), describes the structure of one of these units, mostly organized

according to family lines.

It is entirely possible that some of the sons of Dada were occupied with other (private?) business and had

their group of cultivators administered by another family member. See also D. McGuiness, “Ur III

Prosopography: Some Methodological Considerations,” Archív Orientální 50 (1982) 324 - 342, who

suggested that neo-Sumerian naming-practice involved naming after a paternal uncle. This can, in fact,

be observed in several families; see, e.g, the family of Ur-Meme (W. Hallo, JNES 31 (1972) 87-95).

481. See for example SACT 2, 120 (from AS 5), sealed by Inim-

∑ara,

scribe, son of Ur-Nigar, chief cattle

administrator.

192

Since it is credible that Gududu, following his cousin Lukala in

∑S

9 month 3,

482

was the last person to hold the office of chief household administrator of the governor, it is

the working hypothesis of this study that Gududu was aspiring to become the next governor

of Umma, following his father Dadaga. Gududu is likely to have been successful since it is

reasonable to believe that most of the senior members of the ruling family were dead by the

end of the tenure of Dadaga. Ur-Lisi and Ayakala would certainly no longer have presented a

challenge to Gududu’s candidacy. Ir(mu) and Ur-E’e, two other senior members of the

Umma clan, were probably also dead by the end of

∑u

-Suen’s reign; both would have been

serious contestants to the governorship, since the chief of the granry and the chief cattle

administrator were both offices intimately connected with the line of succession. Most of the

other sons of Ur-Nigar were presumably also dead before Ibbi-Suen’s coronation, or had

already, presumably, been bypassed in succession, judging from their careers (see Section 12

below). Lukala and Gududu were the most prominent members of the younger generation,

and the only ones whose career resembled that of the three previous governors. Since Lukala

may have died late in the reign of

∑u

-Suen, it seems reasonable to assume that Gududu had

successfully positioned himself as the next in line for succession to the governorship of

Umma. Unfortunately the sources for this interesting period of time are nowhere to be

found.

Two other people in Umma held seals with the patronym Dadaga, but neither of

them can be shown to have been sons of Dadaga the governor.

483

482. MCS 2, 55, BM 112948 (from

∑S

9 iii).

193

Several texts mention people who claimed to be sons of Gududu.

484

Only one high-

ranking person called Gududu has been recognized in the extant Umma material—this

person is never attested with any title save for the honorary title of his seal inscription, scribe,

used by all persons from all levels of society (save for the dependent workers).

485

Gududu

was never given the title “captain” used by his uncle Ayakala, and perhaps by his father

Dadaga. It is therefore likely, but impossible to prove, that the people in Umma claiming to

be the decendents of Gududu were right.

The fact that none of the children of the three well documented Umma governors,

except for Gududu, followed in the footsteps of their fathers and the fact that none of these

sons were to occupy high ranking positions in the administration of Umma is puzzling. This

483. Inim-

∑ara

(see, for example, BCT 2, 39 [from

∑S

5]; MVN 16, 1532 [from

∑S

5]; and NYPL 364

[from

∑S

6 i to xii]), and Lu-dingira, only attested in one text (MVN 16, 1018 [from AS 8xii]). Wu

Yuhong argued that Inim-

∑ara

was a brother of Gududu who sealed with his seal in

∑S

5 and 6. See W.

Yuhong, JAC 10 (1995) 140 - 142.

484. Ur-Baba is attested with patronym in only two Umma texts and is a likely candidate as a son of Gududu

(see, for example, BM 105339 (unpubl.) [from AS 6], and perhaps the account SAT 2, 423 [from

∑

45]). The father of a Lu-

∑ara

mentioned in MAOG 4, 191 3 (from

∑S

9) is presumably not identical

with our Gududu. He is in that text called messenger of the king (sukkal lugal), a title otherwise never

attested for Gududu. Three persons are mentioned only once as the son of Gududu, neither can be

positively certified to have been related to the ruling family of Umma. They are: Ur-Ninsu (MVN 16,

908 [from AS 8 iii]), Inim-Inanna, and Ur-Suen (Nik 2, 447 [from AS 3 xi]).

485. Other persons named Gududu include: Gududu, the leather worker (

aÒgab

), see SANTAG 6, 26 (from

∑

35 vi); Gududu, the cultivator (engar), see Syracuse 371 (from

∑

44 ix); Gududu, the messenger

(sukkal), see ASJ 19, 228 74 (no date); Gududu, the gudu-priest of

NinÒubur

, see Nebraska 63 (from

IS 2 xii); and Gududu, the shepherd (na-gada), see SAKF 36 (from

∑

46).

195

Chapter 5: Section 8: Ur-E’e, the chief cattle administrator

This chapter will concentrate on Ur-E’e and one of his sons, Lu-

aya, but it will also

include a discussion of the office of the chief cattle administrator (Sumerian

ÒuÒ

486

Ur-E’e

487

is attested in the Umma records from

∑

33(?) to

∑S

8. His tenure which

lasted 33 years, is paralleled by few other administrators. Throughout this period, Ur-E’e

functioned as a state administrator, and there appears to have been no break in his activities;

the last text to mention Ur-E’e included his title, chief cattle administrator (

ÒuÒ

488

Although the first certain identification of Ur-E’e comes with a seal-inscription on a

document dated to

∑

36, month 2,

489

several documents prior to this date suggest that he

had already held the office of chief cattle administrator since

∑

33. A document from

∑

recorded wool disbursements and is likely, therefore, to have constituted part of the regular

486. For the reading

ÒuÒ

for SA

AR see, above all, R. H. Beal, “Is

KU∑

the Reading of

I∑

= kizû ?,”

NABU 1992, 48.

487. The name Ur-E’e is rare in the Ur III record, Ur-E’e, the chief cattle administrator, son of Ur-Nigar, is

the only high-ranking Umma official to bear that name. The following persons in Umma were also

named Ur-E’e: a cowherd (unu

) (see, for example, MVN 3, 208 [from

∑

45 vii], with broken seal of

Ur-E’e the cowherd); a cultivator (engar) (see NATN 376 [from

∑

33 v]); a fisherman (

Òu

-ku

) (see

SNAT 294 [from

∑

45]); a cult-singer(?) (gala) (see TCL 5, 6039 [from AS 5 ii]); and a gudu-priest

(gudu

) (see Princeton 1, 296 [from

∑

30 vi]).

488. See AnOr 1, 234 (from

∑S

8 x).

489. See MCS 2, 57 (=BM113072). The dedicatory seal-inscription recorded on Aleppo 147 (from

∑

33),

mentioning Ur-Lisi cannot be used to identify Ur-E’e since it fails to record any official affiliation or the

familial relationship of the owner.

196

activities of the chief cattle administrator.

490

That year was also the year that Ur-Lisi took

office as governor and perhaps elected his brother Ayakala as chief household administrator

(and heir?).

In juxtaposition to the great frequency with which the title of Ur-E’e is given in the

seal inscriptions of his two sons (Lukala and Lu-

aya), it is striking to observe that he

himself makes use of it a mere seven times.

491

These examples, running from

∑

37 iii to

∑S

4, include the two key-texts SET 273, a wool-account concerning Ur-E’e from AS 3, and

SET 130, a sheep, oil and wool account concerning Ur-E’e from the following year (AS 4),

both texts will be dealt with below. The first attestation of a seal-impression of either of Ur-

E’e’s two sons recording the title of their father is from

∑

42 ix.

492

According to the published record, the vast majority of the seal-inscriptions of Ur-E’e

do not mention the title of his father Ur-Nigar, also chief cattle administrator. Since the

distribution of the few seal-inscriptions copied (or transliterated) with patronymic title is

completely random, it seems likely that the inclusion of this patronymic title can be deemed

a modern copying error. This is likely to be a result of the fact that the most frequently

attested seal-inscriptions—mentioning other members of the ruling Umma family—did

indeed include the title of chief cattle administrator following the name of Ur-Nigar, the

490. BIN 5, 19 (from

∑

33).

491. These are Rochester 187 (no date); MVN 15, 390 (from

∑

37 iii to vii); OrSP 47-49, 201 (from

∑

37 to

39); SET 273 (from AS 3); Aegyptus 26, 158 6 (from AS 5); MVN 16, 668 (from AS 8 iv); and MVN

8, 202 (from

∑S

4).

492. MVN 14, 450 (from

∑

42 ix), see also Aleppo 399 (from

∑

42). Both texts have the seal of Lukala.

197

ancestor of the Umma clan. Had the occurrence of the title of chief cattle administrator,

following Ur-Nigar’s name, appeared in any meaningful sequence—for example after Ur-

Nigar is presumed to have been dead—it would be difficult to exclude the inclusion of a this

title as a copying error. According to Rudi Mayr’s unpublished dissertation on the seal

inscriptions from Umma, the seal of Ur-E’e never included his father’s title.

493

Since we have

been unable to find any two high-level officials named Ur-Nigar operating at the same time

in Umma, I find it likely that some other solution is behind the exclusion of the patronymic

title than to conclude that Ur-E’e belonged to an important Umma family other than the

ruling family. Ur-E’e’s special position as the one who came to inherit his father’s title and

office might have prompted the exclusion of his father’s title from his seal-inscription. A

parallel scenario can be observed for at least two other members of the ruling family; neither

Lu-

aya, Ur-E’e’s own son, or

∑ara

-izu, the son of Ir(mu), the chief of the granary, ever used

the same title as their father’s although they most certainly inherited their offices. Again, the

position held by Ur-E’e’s son Lukala, chief household administrator of the governor

intimately connected him, and his kin, with the ruling family.

Due to the paucity of sources we must at present conclude that it is likely that Ur-

Nigar was dead already during the reign of

∑ulgi

, and that Ur-E’e operated as chief cattle

administrator from

least

∑

37. It is my suggestion, however, that he had already assumed

that title in

∑

33.

493. R. Mayr,

Seal Impressions

(1997) 150.

198

The title of chief cattle administrator was held not only by members of the ruling

family, as was the case with other high offices; still, there is evidence to suggest that Ur-E’e

ranked above the other chief cattle administrators. This will be discussed in greater detail

below.

Ur-E’e is mentioned in more than 350 texts, primarily pertaining to the livestock

sector of the economy: most of these texts were not sealed. The majority of his sealed texts,

on the other hand, recorded activities related to basic agricultural work. Ur-E’e did not use a

nam-

Òatam

seal in witnessing these transactions, although they resembled very closely the

texts that other high-ranking provincial administrators sealed with their nam-

Òatam

seal.

494

In this connection it might also be beneficial to look at two texts demonstrating that

EnKAS, another Umma chief cattle administrator, operated at the same level as a provincial

administrator.

495

EnKAS sealed the record UTI 4, 2895

496

(from AS 9), with his nam-

Òatam

seal. That text is almost identical to MVN 1, 90 (from AS 8), sealed by

∑eÒkala

, a

well-known member of the group of agricultural overseers.

∑eÒkala

did not use a nam-

Òatam

seal to seal this transaction although he would occasionally do so on other texts. Both texts

were concerned with the erin

-work of plowing and harrowing the field Abagal-Enlila; they

each included a section devoted to work performed by hirelings; and, both were sealed with

the regular seals of the two administrators. This strongly suggests, that EnKAS as well as Ur-

494. Compare, for example, the two almost identical texts MVN 14, 215 (from

∑

46), a receipt for the work

of ox-drivers on the GANma

field sealed by Ur-E’e, and AnOr 7, 199 (from AS 2), a similar receipt

concerning work on the Kamari field, sealed by the

Òabra

-administrator Ur-gigir with his nam-

Òatam

seal.

199

E’e operated as provincial administrators aside from their regular activities as chief cattle

administrators.

A handful of texts are said to have been sealed with the nam-

Òatam

seal of Ur-E’e.

However, most of these texts were not related to agricultural work; rather, some of these texts

were records concerning metal and metal objects; three of them dealt with silver for

emblems, and thereby indirectly related to the administration of the agricultural lands.

497

This is in good accord with the fact that Ur-E’e functioned as an Umma metal supervisor,

498

495. The name EnKAS is always written KAS

in the texts themselves but en-KAS

in his seal-inscriptions.

The reason behind this practice remains obscure to me. KAS

has no reading implying that EN could be

a phonetic complement. For the seal of EnKAS see for example MVN 16, 966 (from

∑

42), a transfer of

the deficits of the shepherds of the temple-household of

∑

ara to the account of EnKAS. EnKAS was

active from

∑

42 (41?) to IS 2 (see, for example, L'uomo 65 [from IS 2 xii 13]). It is possible that

EnKAS took office following Lugal-azida (see, for example, Princeton 1, 24 [from

∑

42]; and AAS 59

[from

∑

42]). According to his seal he was a son of Ur-

IÒtaran

, perhaps indentical with the fattener of

the same name who was also the father of Ur-

∑ulpa

’e. Ur-

∑ulpa

’e claimed to be the brother of EnKAS

in an indenpendent context—approximately nine texts attest to this relationship—Ur-

∑ulpa

’e was an

agricultural overseer at the level of a captain of (plow-)oxen; see, for example, the peculiar tablet

container SAT 3, 1368 (from

∑S

3). EnKAS himself seems to have operated at the level of a provincial

administrator

(see, for example, HUCA 29, 87 13 [from

∑S

3]), sealing documents concerning the

activities of the agricultural overseers at his brother’s level (only very few texts give information such as

kiÒib

nam-

Òa

-tam KAS

; see, for example, SACT 2, 19 [from

∑S

4]; and UTI 4, 2895 [from AS 9],

with the title chief cattle administrator (

ÒuÒ

)). EnKAS also transferred silver as payments for

outstanding deficits from individual shepherds and overseers to the chief household administrator of the

governor (see, for example, BIN 5, 329 [from

∑

47]). Nasa, the well-known Drehem administrator also

had a son named EnKAS, but we believe this EnKAS to have been mainly active in Drehem, posing little

difficulty for a proper identification (see, for example, Princeton 2, 340 (unpubl.) [from AS 3 ii]; and

AUCT 3, 298 [from AS X i 27]).

496. See also SNAT 418 (from AS 9).

200

and with his prominent position within the administration of the eastern district of the

province, Apisal.

499

Through much of his tenure, Ur-E’e was known as the conveyor of the

“field interest” of the entire district of Apisal.

500

The collected field interest and outstanding debits of the shepherds presumably

account for the existence of silver-accounts concerning Ur-E’e, and the fact that Ur-E’e

appears in numerous accounts concerning the Umma colony of trade-agents. The “debits” of

TCL 5, 6045 (from AS 8 xii), a silver account of Ur-E’e, suggest that the “field interest” as

well as the collected debts of the shepherds administrated by Ur-E’e were the personal

“debits” of Ur-E’e. The “credits” of Ur-E’e’s silver account were made up of some amounts of

raw-materials received by Lu-

aya, and a substantial delivery (mu-DU) of silver as well as

minor transfers for specific purposes sealed by Lukala. Later, Lu-

aya, the son of Ur-E’e, had

a silver-account drawn up concerning his silver “debits”, testifying to the transfer of Ur-E’e’s

497. The texts sealed with the nam-

Òatam

seal of Ur-E’e are: OrSP 18 pl.14 40 (from

∑S

2), a copper ring;

SAT 2, 321 (from

∑

43), an emblem of Gu’dena; SAT 2, 963 (from AS 6), an emblem of Gu’dena; MVN

5, 12 (from

∑

42), provisions for female workers employed in the agricultural sector; MVN 16, 1554

(from

∑

37 to 45), various metal; Princeton 1, 185 (from AS 8), dogs; SAT 2, 208 (from

∑

39), sesame

seeds; SAT 2, 288 (from

∑

41), an emblem of Gu’dena in Apisal. The emblems (

Òu

-nir) in question were

always related to agricultural districts and mostly mentioned in relation to agricultural administrators.

498. See fn. 470, p. 184 above.

499. Ur-E’e often received silver, perhaps in his capacity as a provincial administrator; the silver seems mainly

to have been from shepherds and agricultural overseers, and destined for the state, suggesting that Ur-E’e

conveyed field interests from the tracts of land in the district he managed.

500. See BM 108004 (from

∑

33); AAS 75 (from

∑

37); Princeton 1, 548 (from

∑

38); BIN 5, 108 (from

∑

44); and Princeton 1, 409 (from AS 2). These texts were first discussed by P. Steinkeller, JESHO 24

(1981) 116ff. See also above, p. 170+ fn. 433.

202

Figure 7: Accounts concerning the office of the chief cattle administrator.

The “credits” section of Ur-E’e’s account again records the same amounts of wool as

is recorded in the governor’s account, now summarized as (received under) the seal of the

governor. This clearly indicates Ur-E’e’s position as a civil servant answering directly to the

wool-account concerning the governor

Ashm. 1 1924-666 from AS 3

wool-account concerning Ur-E'e,

chief cattle administrator

SET 273

from AS 3

"Debits" section

"Credits" section

Remainder:

1st entry:

2nd entry:

3rd entry:

1st entry:

2nd entry:

3rd entry:

1st entry:

379 talents 59 mana 1/2 shekel mixed wool

1) 13 talent 11 2/3 mana wool
2) 22 talent 26 2/3 mana wool

48 talents 38 1/2 mana kura wool

102 talents minus 1 mana yellow(?) wool

4 talents 7 2/3 mana dark(?) wool

48 talents 38 1/2 mana kura wool

102 talents minus 1 mana yellow(?) wool

4 talents 7 2/3 mana dark(?) wool

1st entry:

48 talents 23 1/2 mana kura wool

142 talents 14 1/2 mana yellow(?) wool

19 talents 22 mana mixed wool

2nd entry:

19 talents 22 mana mixed wool

40 talents minus 6 mana yellow(?) wool

3rd entry:

1st entry:

2nd entry:

3rd entry:

4th entry:

5th entry:

6th entry:

2nd entry:

3rd entry:

4th entry:

5th entry:

40 talents minus 6 mana mixed wool

11 talents 6 mana kura wool

77 talents x] mana ? wool

13 talents 35 mana 10 shekel mixed wool

1 talent yeallow(?) wool

1st entry:

From Ur-E'e

From PN

Emblems

Total (B+C):

Total (A+D):

Grand total:

...] +3 1/2 mana ...]-ga

...]

3 talents 55 1/2 mana (wool)

53 mana wool

2 mana wool

43 2/3 mana wool

45 mana wool

422 talent 51 mana 10 1/2 shekel

301 talent 17 mana mixed wool

(its halmutum:) 10 talents 2] mana 14 shekel

734 talent 10! 1/3 mana 4 1/2 shekel mixed wool

21 talent 17 1/3 mana wool

3 talent 50 minus 1 mana wool

From EnKAS

1 talent 5 2/3 mana wool

From ∑akuge

16 talent 17 5/6 mana wool

From the governor

Received by the governor

Received by ∑eÒsag
(see perhaps AAS 135)

Received by ∑eÒsag
(tranfered to the "debits"
of his account)
(= AAS 135, column 1,
line 21 to column 2 line 9)

10 mana wool

From Ur-∑ara

1 talent 22 mana wool

Total:

293 talent 41 1/3 mana mixed wool

etc.

Reverse column 5, lines 4-17

regular deliveries for the gods

(sa2-du11 dingir-re-ne)

rations for the permanent staff

various quantities and qualities of wool

for textiles

various quantities and qualities of wool

for textiles

Received by Ur-Nintu

204

Our knowledge of the activities of Ur-E’e are, above all, derived from the account

SET 130 (from AS 4).

501

The following figure (8) shows the levels of accounting and the

different offices controlled by Ur-E’e according to that text.

501. This multiple level account starts out with a rather substantial “remainder” from the previous year (AS

3), followed by a “debits” section consisting of, first, the expected production of the shepherds (these

were divided into shepherds of native(?) sheep, and shepherds of foreign(?) sheep (sipa udu eme-gi-ra and

sipa udu kur-ra)), followed by the expected production of the goat-herders (sipa ud

-da), the anticipated

production from a number of cowherders (unu

-de

-ne), and finally several minor contributions to the

“debits” consisting of animals delivered for the cult in either Apisal or Zabala. The rather substantial total

of the “debits” section was recorded in the beginning of column six. The “credits” section, consisting of

the actual production of the herders under the supervision of Ur-E’e, begins by listing the animals

expended for the regular offerings for the gods (sa

-du

dingir-re-ne), more specifically, the deliveries for

∑ara of Apisal

, Ninura of Apisal, Lamma-

∑ulgira of Apisal

, Nin-Zabala of Apisal, and Inanna of

Zabala, as well as the dowry of Dumuzi(d). The next brief section recorded animals which had fallen (ri-

ri-ga), followed by a longer section (rev. i 13 torev. ii 4) listing the siskur-offerings for various threshing

floors in the province. The sealing officials in these cases were all known agricultural overseers (GuTAR,

Agugu, etc.), several of whom are mentioned elsewhere in this study. The header (i.e. last line) of this

section reads “shall not be transfered to the debits” (rev. ii 4: ugu

-a nu-ga

-ga

), presumably suggesting

that these deliveries were meant for consumption and not for further disbursement. Following several

minor transfers (some animals were transfered to the “debits” of known shepherds), we find a long list of

animals transferred to various persons, both herders and administrators. A minor “surplus” from the

previous year was recorded right before the complex totals. The complex “operating balance” recorded

both “deficits” and “surplus”. The colophon named the text as a sheep, oil and wool-account concerning

Ur-E’e.

205

Figure 8: The organization of the office of Ur-E’e.

Ur-E’e administered the shepherds who made deliveries to the temples in the two

districts Apisal, Zabala and some parts of Umma.

502

These shepherds were apparently not

bound to one particular household, but rather either permanently bound to several

households, or assigned

ad hoc

to a variety of households. The exchange between the

shepherds and the cultivators was managed by Ur-E’e on the one hand, and a high-ranking

agricultural overseer at the level of a captain of (plow-)oxen, or above, on the other.

503

Likewise, the exchange between the shepherds and the fatteners was managed by the chief

cattle administrator and the fattener.

504

However, the records of deliveries of dead animals

502. People who were called sipa(d) in one text could be referred to as na-gada in another. Whether this is due

to so-called horizontal terminology, that sipa(d) was the general word for any herder, and na-gada the

only specific (Ur III) word for a shepherd, is still uncertain. See MCS 1, 54, BM 106045 (from

∑S

5),

for evidence that persons with the title na-gada could be summarized as either sipa ud

(goat herder) or

sipa eme [gi

-ra] (shepherder of native sheep).

Ur-E'e

chief cattle administrator

District of Apisal

Ninura of Apisal

∑ulgira of Apisal

Inanna of Zabala

Gula of Umma

Inanna Ebgal

∑ara of Apisal

Districts of Zabala and Umma

Shepherds

Cowherds

∑atam-administrators

District of Gu'dena and MuÒbiana

District of Apisal

Partial reconstruction of the responsibilities of Ur-E'e based in part on YOS 4 237, MCS1, 54 / MCS 6, 10, and SET 130

Captain of (plow-)oxen

Cultivator

Ox-drivers

plow-oxen

sheep and goats

(consupmtion and taxes)

Temple Households

sheep and goats

large cattle

Captain of (plow-)oxen

Cultivator

Ox-drivers

plow-oxen

sheep and goats

(consupmtion and taxes)

206

sealed by members of the ruling family, in this study referred to as sealed by the chief

household administrator of the governor, were mostly mediated directly between the

shepherd and the chief administrator.

The relations between the administrators of cows and oxen, who also held the title

ÒuÒ

, and the agricultural units is documented in numerous texts. SAKF 54 (from

∑

45),

may be useful in illuminating this relationship since it rather clearly describes the transfer of

animals:

Obverse.

1. 3(u) 6(diÒ) gu

i-a

36 diverse oxen and cows,

geÒ

apin-ta gur-ra

having returned from the plow,

3. ki Òabra gu

-ke

-ne-ta

from the administrators of oxen,

505

Reverse.

1. KAS

-dab

EnKAS seized.

2. mu us

-sa si-mu-ru-um

lu-lu-bu

Year after:

“

Simurum and Lulubum

503. As was pointed out by K. Maekawa, Zinbun 22 (1987) 39 - 40, an agricultural overseer was called by

different titles according to the function he held in each particular situation. It seems as if high-ranking

members of the ruling family and prominent members from the level of the captains of (plow-)oxen were

sometimes mediators between groups of captains of (plow-)oxen and the other Umma bureaus. It is

possible that the family structure of the agricultural overseers, as supported by SAT 2, 77 (from

∑

33 vi),

is indicative of the command-structure of the agricultural administration, i.e., that a “clan-leader”, given

the formal title dub-sar gu

1(u) or the equivalent, was the overseer of the lands managed by his sons or

younger brothers who held titles at the ranks of the captain of (plow-)oxen).

504. See for example SACT 2, 242 (from

∑

43).

505.

Òabra

is believed to be the correct title of the

Òabra

administrators mentioned in the Umma

agricultural record (see K. Maekawa, Zinbun 22 (1987) 40) these people were otherwise called captains

of (plow-)oxen (nu-banda

), or foremen (ugula).

207

a-ra

1(u) la

1(diÒ)-kam ba-

were destroyed for the 9

time

”

(blank line)

Cattle management in the Umma province seems to have been managed above all by

EnKAS and Atu, another chief cattle administrator.

506

Ur-E’e seems to have been involved,

exclusively, with the management of sheep and goats.

YOS 4, 237 (from

∑S

7 ii), is an account of the livestock transferred between the

governors Ayakala and Dadaga, when Dadaga took office, following the retirement, or more

likely, the death of his brother Ayakala.

507

The account has no division between a “credits”

section and a “debits” section; rather, it is subscribed udu si-il

-la, perhaps suggesting that the

document recorded the sheep and goats “found” with the shepherds while surveying the

livestock of the province.

508

The counted animals were divided into three groups according

to overseer. In this text, as in other texts concerning livestock the title foreman (ugula) was

used where we would expect the title chief cattle administrator (

ÒuÒ

), however, since this

happens, exclusively, in connection with the calculation of the animal belonging to different

506. For a discussion of the office of Atu see, above all, R. Englund, OrNS 64 (1995) 377-429.

507. This text was briefly discussed above, see pp. 176 - 177 in this study. See also M. Stepien,

Animal

Husbandry in the Ancient Near East

(Bethesda 1996) 50 - 53, and in particular the figure on pp. 51-52,

for an analysis of this text.

508. See also Buffalo SNS.11-2, 134 4 (from

∑S

8 iv); SNAT 526 (from

∑S

9 ix); and TIM 6, 46 (from IS 3

viii), three texts counting the large cattle of the temple-households of

∑

ara, Ninurra, and

∑ulgi

(the last

household mentioned only in Buffalo SNS) (see also fn. 505 below, for this sequence of Umma temple-

households). In the last two text, Atu is mentioned as the overseer (no doubt instead of his regular title

‘chief cattle administrator’ following the pattern described for EnKAS and Ur-E’e) of the temple-

household of

∑

ara, and Lu-

aya (the son of Ur-E’e) as the overseer of the cattle of the temple-

households of Ninurra.

208

temple-households, I expect it to be a context-specific terminology meaning “foreman (of the

shepherds) of such-and-such temple-household”. The first overseer mentioned in YOS 4,

237, is EnKAS. In YOS 4, 237, he was in charge of the shepherds and their animals from the

temple-household of

∑

ara (e

ara

509

The second overseer, Ur-E’e, was in charge of the

shepherds and their animals from several temple-households: the household of Ninura,

∑ulgi

, Inanna of Zabala, Gula of Umma, E’e, and of Nin

ilisu (e

[

nin-ur

]-ra, e

Òul

-gi-

ra, e

inanna zabala

, e

gu-la umma

, e

-e, e

nin-

i-li-su

). The last overseer, Ur-

Nungal,

510

was not mentioned in connection with any household. The animals controlled by

509. EnKAS’s affiliation with the household of

∑ara

of Umma is above all documented by the large, but

fragmentary, account MVN 13, 618 (from AS 7), in which EnKAS is mentioned as the administrator

responsible for the animals of the household of

∑ara

(rev. viii 4 - 5: e

Òara

/ ugula KAS

). The

colophon of that text reads:

Reverse, column 10.

10. nig

-ka

-ak sipa udu eme-gi-ra-ke

-neAccount of shepherds of native sheep,

concerning

11. e

Òara

the household of ∑ara,

12. e

nin-ur

-ra

the household of Ninura,

13. e

Òul-gi-ra

the household of ∑ulgir

14. e

inanna

the household of Inanna,

15. u

gu-la

and the household of Gula

16. mu Ìu-uÌ

-nu-ri

ba-Ìul

Year:

“

ÎuÌnuri was destroyed

”

For the sequence of temple households, see also the label (pisan-dub-ba)—recording accounts of

orchards—AAICAB 1, 1911-176 (from AS 8); compare to YOS 4, 214 (from

∑

47), an actual record of

the orchards of the temple-households of

∑ara

of Ninura; and the unpublished account concerning

cattle BM 105329 (from AS 7). Now, see also WAM 2000.47 (from AS 5) (edition: R. Englund, CDLB

2003:1).

209

Ur-Nungal may be related to the animals belonging to the e

-du

-la of Ur-Lisi. Although this

“remainder” of the accounts of the household of the governor Ur-Lisi is not recorded after

∑S

5, there is strong circumstantial evidence which suggests that the animals recorded in this

text belonged to a similar administrative category, and perhaps even the same group.

511

Whereas EnKAS was the overseer of the shepherds of the temple-household of

∑

ara

of Umma, numerous texts mention Ur-E’e in connection with the temple-household of

∑

ara

of Apisal.

512

However, both were, as we have seen (figure 7), directly subservient to the

governor, and not the temple-households. The individual shepherds cannot be used to

determine the relationship with the temple-households, since, as already pointed out by M.

Stepien,

513

the shepherds could be related to several different households at the same

time.

514

In the large account concerning the hides of fallen animals (ku

udu ri-ri-ga),

MCS 1, 54, BM 106045 (and the fragmentary parallel text MCS 6, 10, BM 106041) from

∑S

5, EnKAS and Ur-E’e were again listed side by side.

515

The first section of this text (until

column 2, line 27) recorded the ox hides from the cow-herders, described as “the hides of

510. Ur-Nungal, son of Ur-

∑ara

, the accountant (

Òa

-dub-ba), was himself an accountant; see OrSP 47-49,

412 (from AS 1):

Òa

-dub-ba e

Òara

, “the accountant of the household of

∑ara

”; and UTI 3, 1692

[from

∑S

2 xii]). For the seal of Ur-Nungal; see, for example, MVN 1, 185 (from AS 7 viii); MVN 5, 77

(from

∑S

6 iv); MVN 14, 372 (from

∑S

1). For the dedicatory seal of Ur-Nungal (

dedicated to ∑u

Suen); see, for example, BCT 2, 81 [from

∑S

8]; MVN 15, 355 [from

∑S

9]; MVN 16, 834 [from

∑S

7]). It is likely that Ur-Nungal was the accountant of the governor’s household, and in that regard

perhaps directly responsible to the king rather than the governor.

210

oxen (?) of the pen, from the cow-herders.”

516

The next section listed the fallen animals of

the cultivators, grouped according to captains of (plow-)oxen. The first group of captains of

(plow-)oxen was directed by two overseers, Abbagina and EnKAS. Abbagina who is not

known as a chief cattle administrator may have been a scribe of 10 oxen, controlling a couple

of captains of (plow-)oxen.

517

This section (ending in obv. iv 12-13 (22-23)) is identified as

511. Ur-Nungal is never attested as an overseer of shepherds, nor as a chief cattle administrator, but rather as

an accountant. Ur-Nungal’s role in YOS 4, 237 should be compared to MCS 1, 54, BM 106045 (with

parallel text MCS 6, 10, BM 106041), where the last entry before the total of the shepherds was the e

-la of Ur-Lisi. The same shepherd mentioned in YOS 4, 237, is also mentioned in MCS 1, 54, BM

106045. Ur-Nungal is attested in three other documents relating to the e

-du

-la of Ur-Lisi. (1)

Controlling garment rations for the female dependent workers still associated with the e

-du

-la of Ur-

Lisi, in MVN 14, 564 (from AS 9), rev. 1-3: tug

-ba geme

-du

-la ur-

-si

ensi

-ka /

ki i

-kal-la-ta

kiÒib ur-

nun-gal,

“

garment rations for the dependent female workers, of the e

-du

-la of Ur-Lisi, the

governor, from Ikala, Sealed by Ur-Nungal”. (2) Transferring oil from the e

-du

-la of Ur-Lisi to Ur-

∑ulpa

’e, perhaps identical with the representative of the palace in Umma, mentioned on p. 140, in

Syracuse 448 (from AS 9), obv. 2 to rev. 5: i

-du

-la ur-

-si

ensi

-ka / ki ur-

nun-gal-ta // ur-

Òul

pa-e

3 /

Òu

ba-ti, “Oil of the e

-du

-la of Ur-Lisi the governor, from Ur-Nungal, Ur-

∑ulpa

’e received”. (3)

Transferring animals from the e

-du

-la of Ur-Lisi to

UÒmu,

the Umma fattener, in UTI 3, 2275 (from

∑S

2), rev. 2-5: gu

udu e

-du

-la

¿ur•

-si

ensi

udu bar ku

-ga / ki ur-

nun-gal-ta] / uÒ-mu

-dab

“Oxen and sheep of the e

-du

-la of Ur-Lisi, the governor, and barkuga sheep from Ur-Nungal,

UÒmu

seized”. The term bar ku

-ga, “silvery fleece”, is attested in one other text (TJAMC JOS 21

(pl.54) [from

∑

29 iv], obv. 7-9: 3 udu bar ku

-ga / mu-DU /

Òara

umma

), and in one, perhaps two,

personal names (CT 10, 28, BM 014316 [from AS 2 iii to xii], obv. ii 11: 1(ban

) 5

(diÒ

) <sila

> (

Òe

)

-bar-ku

-ga dumu lu

-bala-sa

-ga, and AUCT 2, 333 [no date], obv. 4: 1(

diÒ

) bar-ku

-ga-ni. This

personal name is mentioned in a context that seems to exclude a reading 1(

diÒ

) bar ku

-ga-ni, as is

possible, and likely, in SNAT 487 [from

∑S

3 ix], obv. 4: 1(

diÒ

) tug

bar ku

-ga-ni ma

-la

, based on

an analogy with the preceeding and following lines, for example, line 5: 1(

diÒ

) tug

bar nin-za-me

211

part of the household of

Gula

(?),

518

and it was controlled by EnKAS.

519

The second section

(until obv. v 9 (24)) recorded the overseers from (the district of ) Gu’dena and

MuÒbiana

520

the third (until obv. v 18 (33)) recorded the overseers from (the field) Menkara,

521

and the

512. See, for example, SAT 2, 1006 (from AS 7 xii), recording wool from

∑akuge

, as a delivery for

∑ara

Apisal, transfered to the “debits” of Ur-E’e; obv. 2-4: mu-DU

Òara

a-pi

-sal

/ ki

Òa

-ku

-ge-ta /

ugu

ur-e

-e ba-a-gar); SANTAG 6, 106 (from AS 1), recording products from

∑akuge

, as deliveries

for ∑ara

of Apisal, received by Ur-E’e, rev. 1-4: mu-DU

Òara

a-pi

-sal

/ ki

Òa

-ku

-ge-ta / ur-e

-e-

Òu

ba-ti, compare to the identical text Syracuse 450 (from AS 4); MVN 14, 584 (from

∑S

2),

recording bran as fodder for the birds of the house

of ∑ara of Apisal,

from Ur-E’e

, obv. 1-5: duÌ

sag

/ Òa

-gal muÒen / e

Òara

a-pi

-sal

513. M. Stepien,

Animal Husbandry

(1996) 52.

514. See for example the shepherds mentioned in TCL 5, 6038 (from AS 7), and compare with the shepherds

mentioned in the texts discussed here.

515. See also M. Stepien,

Animal Husbandry

(1996) 94 - 96.

516. The corresponding line-numbers from MCS 6, 10, BM 106041, are given in parentheses after the line-

numbers from MCS 1, 54, BM 106045, whenever these are different. Obv. ii 26 - 27:

kuÒ

ab e

tur

-ra /

ki unu

-de

-ne-ta.

517. Abbagina, son of Lugal-Maguire, was also known as a provincial administrator; see for example AAICAB

1, 1911-228. This Abbagina should not be confused with the cow-herder by the same name.

518. The prominent position of the household of Gula in this text is not clear to me. Since this section is

otherwise normally occupied by the household of

∑ara

or the Da-Umma district it is possible to

speculate that the household of Gula was in some way or another here a pseudonym for the household of

∑ara

519. Obv. iv 12 - 13: gu

[d]

gu-la / ugula KAS

520. Obv. v 9: gu

-eden-[na] u

muÒ

-[bi-an]-

¿na•

The agricultural overseers mentioned were Ur-Enun,

GuTAR, Ur-Ninazu, Ipa’e, and Lu-dingira, all of whom we are by now familiar with (see above all pp.

250 - 254, in this study).

521. Obv. v 18 (33): gu

me-en-[kara

]. The field Menkara was hardly a district, but perhaps an important

field.

212

last (until obv. vi 8 (24)) recorded the overseers from (the district of ) Apisal.

522

This entire

section is designated as “the plow-oxen, from their cultivators”.

523

The second main section,

approximately corresponding to columns 7 and 8, record the deliveries of the shepherds.

524

The first long list (column 7) records the shepherds under Ur-E’e,

525

the second record the

shepherds under EnKAS,

526

and the last records the animals still coming from the e

-du

-la

of Ur-Lisi.

527

The last section (columns 9 and 10) records transfers of hides from Ur-E’e

(until column 9, line 34 (38)), and from EnKAS (until column 10, line 4 (15)). The

deliveries from Ur-E’e span the period from

∑S

3 through

∑S

Assuming we understand these texts correctly, Ur-E’e was the main overseer of sheep

and goats in the eastern regions of the Umma province, in particular the animals of the

temple-households of Ninura,

∑ulgi

, Inanna of Zabala, and

∑

ara of Apisal, together with the

temple-households of several other less important deities. EnKAS was the person responsible

for the sheep and goats of the household of

∑ara

of Umma, an altogether less powerful

position than that of Ur-E’e. Atu, another chief cattle administrator, seems to have been in

522. Obv. vi 8 (24): gu

a-pi

-sal

523. Column 6 is broken, it has been reconstructed from MCS 1, 54, BM 106045, obv vi 33 - 34: gu

apin /

[ki] engar-e-ne-<<ne>>-ta.

524. Obv. viii (ix) 28 (3): ki sipa nam-en-na-ke

-ne-ta.

525. Obv. vii (viii) 26 (1): ugula ur-e

-e.

526. Obv. viii 17 (28): ugula KAS

527. Obv. viii 36:

udu ¿e

•-du

-la ¿ur•-

d]

-si

¿ensi

•

213

charge of the large cattle of several households, but seems to be unrelated to the

administration of sheep and goats, and its products. During the reign of

∑

u-Suen, Lu-

aya

gradually took over more and more of the responsibilities of his father. According to M.

Stepien, Lu-

aya also held the title chief cattle administrator,

528

and was connected with the

management of the large cattle of the household of Ninura.

528. M. Stepien,

Animal Husbandry

(1996) 61, concludes, after comparing Lu-

aya’s activities with those of

his father Ur-E’e, and the other Umma chief cattle administrators Atu (cattle), Lugal-azida (cattle), and

EnKAS (called Girim by Stepien), that Lu-

aya, “must have had

per analogiam

the same title” (p. 61).

Stepien divided the responsibilities of the Umma chief cattle administrators in the following way; Ur-E’e

and EnKAS were in charge of sheep and goats and Atu, Lugal-azida, and Lu-

aya in charge of large

cattle. As will be apparent from the preliminary investigation in this study it is likely that the

organization was more complex than suggested by Stepien, and that Lu-

aya was in fact about to take

over the entire responsibilities of his father, i.e., both small as well as large cattle.

215

Lu-

aya’s involvement in the activities of the trade-agents is above all ascertained

from two accounts SNAT 504 (from

∑S

6), and SNAT 518 (from

∑S

7). SNAT 504 is a

short text which is not a complete account, but it does record a number of commodities

“booked out from the silver-account for the bala.”

530

This text suggests that Lu-

aya was in

529. See however the following text suggesting that administrators could deposit seals with an institution,

perhaps accounting for the many diferent seals attested for some Umma administrators.

MVN 16, 628 (from

∑S

4 v):

Obverse.

1. 1(diÒ) kiÒib

za-gin

ka-ba ku

-sig

One lapis seal, on its mouth is set gold

gar gu-bi ku

-babbar

its neck is silver

3. lu

Ìa-ia

dumu ur-e

-e ÒuÒ

(inscription(?):)

“

Lu-Îaya, son of Ur-E

’

chief cattle administrator.

”

4. 3(diÒ) u

1(diÒ) udu nita

Three ewes, one male sheep

5. AG aÒgab

? of the leather worker.

6. 1(diÒ) maÒ

geÒ

tukul ur-

-bar-tab

1 goat, the weapon, Ur-Urbartab,

7. 1(diÒ) maÒ

nin-Ìur-sag ur-

a-Òar

1 goat, NinÌursag, Ur-AÒar,

Reverse.

1. ∑U+NIGIN

1(diÒ) kiÒib

za-gin

Total: One lapis seal, on its mouth is set gold

ka-ba ku

-sig

gar gu-bi ku

-babbar its neck is of silver,

3. ∑U+NIGIN

3(diÒ) u

Total: three ewes,

4. ∑U+NIGIN

1(diÒ) udu nita

Total: one male sheep,

5. ∑U+NIGIN

2(diÒ) maÒ

Total: two goats.

6. mu-DU

Òara

Deliveries for ∑ara.

7. iti RI mu

Òu-

suen lugal-e bad

Month RI, Year:

“

∑u-Suen, the king,

mar-tu mu-ri-iq ti-id-ni-im mu-du

built the Amurite wall (called) Muriq-tidnim

”

530.

-babbar / nig

-ka

-ta / .... / bala-

Òe

∑U

+NIGIN

-babbar / zi-ga-am

/ nig

-ka

-ak

dam-gar

a-ia

date

216

some way obliged to transfer funds to the central administration; the second text (translated

here, in full) is essentially an account of the “field interest,” perhaps of the district of Apisal.

SNAT 518 (from

∑S

7), has the format of a regular dam-gar account:

“Debits:”

Obverse.

1. 1(diÒ)] ma-na 3(diÒ) 1/3(diÒ) ¿gin

•

1 mana 3 1/3 shekels 11 1/2 grains of silver,

1(u) 1(diÒ) 1/2(diÒ) Òe] ku

-babbar

2. ku

maÒ

] a-Òa

-ga

(is) silver (for the) field interest,

531

3. 3(diÒ) 2/3(diÒ) gin

pa muÒen

3 2/3 shekels, (is) silver (for the)

bird-feather.

532

(blank line)

4. ∑U+NIGIN

1(diÒ) ma-na 7(diÒ) gin

Total: 1 mana 7 shekels 11 1/2 grains

1(u) 1(diÒ) 1/2(diÒ) Òe ku

-babbar

of silver.

5. Òa

-bi-ta

From that:

“Credits”

6. 1(diÒ) ma-na ku

-babbar

1 mana of silver,

7. kiÒib gu-du-du

Sealed by Gududu.

8. 2(diÒ) 5/6(diÒ) gin

2(u) 2(diÒ)

2 5/6 shekels 22 1/2 grains silver

1/2(diÒ) Òe ku

maÒ

a-Òa

-ga

(for the) field interest,

9. gu-du-du

Gududu

10. 1(diÒ) 1/2(diÒ) gin

Òu-nir

1 1/2 shekels; emblem of Gu

’

edena

-eden-¿na•

Reverse

531. P. Steinkeller, JESHO 24 (1981)113-145.

532. “Silver for the bird-feather” appear mostly in a context similar to “silver for the goat-of-the-field,” but

escapes all other meaningful interpretation. The texts are MVN 21, 344 (from AS 8 to

∑S

3); MVN 16,

910 (from AS 9); Nik 2, 401 (from

∑S

2); SANTAG 6, 315 (from

∑S

7 ix); SNAT 518 (from

∑S

7);

TCL 5, 6045 (from AS 8 xii); and VO 8/1, 67 (from

∑S

8 xi).

217

(blank line)

1. ∑U+NIGIN

1(diÒ) ¿ma•-na 4(diÒ)

Total: 1 mana 4 1/3 shekels

1/3(diÒ) gin

2(u) 2(diÒ) 1/2(diÒ) Òe

22 1/2 grains of silver

-barbar

2. zi-ga-am

is booked out.

“Operating balance”

3. la

-ia

2(diÒ) 2/3(diÒ) gin

Deficit: 2 2/3 shekels minus

1(u) Òe ku

-babbar

10 grains of silver,

(blank line)

“Colophon”

4. nig

-ka

-ak lu

a-ia

Account concerning Lu-

aya.

5. mu

Òu-

suen lugal-e ma-da

Year:

“

∑u-Suen, the king,

za-ab-Òa-li

mu-

destroyed the district of ZabÒali

”

The “debits” of SNAT 518 corresponds approximately to half the “field interest” of

the district of Apisal when Ur-E’e controlled it.

533

The “credits” were in fact received by

Gududu, acting as a chief administrator of the governor, and called “field interest,” as well as

silver for the emblem of Gu’dena.

Although Lu-

aya is mentioned in the wool-accounts of both Luduga, from

∑S

534

and Ur-E’e, from

∑S

535

as well as a short account concerning Lu-

aya himself

written in

∑S

536

it was not before

∑S

9 that Lu-

aya became intricately involved in the

533. See above (pp. 200 - 201) concerning Ur-E’e.

534. Receiving the entire “credits” in MVN 14, 234 (from

∑S

2 to 3).

535. SANTAG 6, 288 (from

∑S

4).

536. AnOr 7, 260 (from

∑S

4), an account recording less than 100 animals and their wool(-production).

218

administration of livestock (see SNAT 526). This coincides fairly closely with the

termination of his father’s tenure as chief cattle administrator, and it is possible that Lu-

aya

inherited his father’s office. Two texts dated to the 9

month of

∑S

9, and the 8

month of

IS 3, respectively, are among the clearest evidence that Lu-

aya held the office chief cattle

administrator. SNAT 526 (from

∑S

9 ix),

537

perhaps related to the accounting processes

connected with the death of the king, is an account of cattle, subscripted /sila/.

538

The

animals belonged to the households of

∑ara

in Apisal and Ninura in Apisal—Atu was

foreman of the first group and Lu-

aya of the second. The list of herders recorded in Lu-

aya’s section, can be compared with a known list of herders associated with that

institution.

539

Other texts can also be cited to the effect that Lu-

aya took over the office of

his father around the time of

∑S

8 or 9.

540

Several texts record that a transaction was sealed with the nam-

Òatam

seal of Lu-

aya (

kiÒib nam-Òa

-tam lu

a-ia

)

541

adding Lu-

aya to the group of high-level

537. See also the paralell text TIM 6, 46 (from IS 3 viii).

538. See p. 176 + fn. 449, above.

539. See R. Englund, CDLB 2003:1 §9, and M. Stepien,

Animal Husbandry

(1996) 54 - 62.

540. Several texts can be cited in favor of this theory, above all texts such as MVN 14, 525 (from IS 2), a text

which lists large numbers of sheep and lambs termed ki-ba ba-a-gar, booked out from Lu-

aya for the

bala, and received by

UÒmu

, the well-known Umma fattener.

541. See, for example, MVN 13, 573 (from

∑S

9 xi 25 to

xii

), a text concerned with fodder for dogs (

Òa

-gal

ur-gi

-r

a) which has the nam-

Òatam

seal of Lu-

aya. See also Princeton 1, 185 (from AS 8), a similar

text sealed by Lu-

aya’s father Ur-E’e.

219

administrators who functioned—Lu-

aya perhaps together with his cousin Lu-

∑ulgi

(ra),

542

the son of Dadaga—as administrators of specific agricultural groupings. The following text

example (MVN 16, 751), with no date, is a list of two administrative groups, one controlled

by Lu-

aya, the other controlled by Lu-

∑ulgi

(ra), both were called foremen (ugula) in this

text. Each of them also figured as the first entry of their group and received the largest rations

(supposing this text to be a record of rations!). The people mentioned in this text are the

same as those mentioned together in several texts as captains of (plow-)oxen.

MVN 16, 751 (no date):

543

Obverse.

1. 3(geÒ

) gur lu

a-¿ia

•

180 gur: Lu-

aya,

2. 1(u) ur-

¿igi•-zi-bar-ra

10 gur: Ur-Igizibara,

3. 4(u) ¿inim•-

d]

Òara

40 gur: Inim-∑ara,

4. 4(u) ...]-a-ga

40 gur: x-aga,

5. 4(u) ?]-mu-U

544

40 gur:

Mu’u’u (?),

6. 1(u) a-a-¿kal•-la]

10 gur: Ayakala,

7. 4(u) lugal-¿ku

•-ga-ni]

40 gur: Lugal-kugani,

542. See, for example, BIN 5, 258 (from

∑S

6), for a reference to the

nam-Òatam

seal of Lu-

∑ulgi

(ra), rev.

kiÒib nam-Òa

-tam lu

Òul-gi-ra.

543. The text transliterated here might very well be a practice tablet since it lacks a date and does not mention

any institutional information

544. See SNAT 531 (from IS 3 i), for a reference to a person Mu’u’u active in the agricultural sector of the

economy:

Obverse.

7'. 4(barig) Òe Òuku Òa

-gu

kiÒib mu-u

-u

4 barig of barley, is the subsistence

of the ox-driver, sealed by

Mu’u’u.

220

8. 6(aÒ) lu

-dingir-¿ra• ¿dumu•

6 gur: Lu-dingira, the son of Udaga,

-dag-¿ga•

9. 4(u) lu

-dingir-ra nu-banda

40 gur: Lu-dingira, the captain

of (plow-)oxen,

10. lu

Òara

nu-banda

- Lu-∑ara, the captain of (plow-)oxen,

(blank line)

11) ¿6(geÒ

)• 4(u) 6(aÒ) ¿gur•

Total: 406 gur,

12) ugula lu

]¿

a•-ia

Foreman: Lu-

aya.

Reverse.

1. 3(geÒ

) lu

Òul-gi-ra

180 gur: Lu-∑ulgira,

2. 3(u) ¿al•-la-palil

(IGI.∑E.DU)

545

30 gur: Alla-palil,

3. 4(u) lu

-dingir-ra

40 gur: Lu-dingira,

4. 4(u)

Òara

-kam

40 gur: ∑arakam,

5. 4(u) gu

-TAR

40 gur: GuTAR,

6. 4(u) e

-¿gal•-e-si

40 gur:

Egalesi,

7. 2(u) ur-...]-na

20 gur: Ur-enun?]na,

546

8. 4(u) ur-

geÒ

gigir

40 gur: Ur-gigir,

9. 2(u) inim-

Òara

20 gur: Inim-∑ara

545. See MVN 21, 114 (from

∑S

6), for a reference to the seal of Alla-palil, note that he himself could seal as

nam-Òatam

Reverse.

...

7. kiÒib nam-Òa

-tam al-la-palil

Seal.

1. al-la-¿palil

•

2. dub-sar

3. dumu ur-a-a-mu

546. The reconstruction is based on the fact that Ur-Enuna is often mentioned within the same

administrative group.

221

10. 2(u)+n] lu

-...]-la

20 + gur: Lu-...]la

(blank line)]

11. ...]

Total: ...]

12. ugula lu

¿Òul•-gi-ra]

Foreman: Lu-∑ulgira.

All of the persons mentioned in this text are known agricultural overseers, sometimes

mentioned with the title captain of (plow-)oxen, sometimes as foremen (ugula). They each

directed a number of cultivators who in turn directed a number of ox-drivers.

Based on a comparison with MCS 1, 54, BM 106045, where some of the same

people were mentioned, it seems likely that the second section of MVN 16, 751, recorded

the agricultural overseers of Gu’dena and

MuÒbiana.

Perhaps the first section recorded the

overseers from the Apisal district? Lu-

aya may have been in control of this group of

agricultural overseers for several reasons: he could have been the administrator of state or

temple lands, or he could have been administrating his own (prebend) lands.

One group of texts, exclusively written on three-sided bullae, were sealed by either

Ur-E’e or one of his sons, Lu-

aya or Lukala, sometimes together with Ur-Nungal, the

archivist.

547

These bullae all recorded the regular provisions (sa

-du

)

548

for the couriers

stationed in Umma or at the “tower”

549

(an-za-gar

) on the Girsu-canal (i

gir-su

550

547. Two texts were sealed by the governor Hermitage 3, 514 (= Erm 15302) (from AS 7 iv), and OrSP 47-

49, 373 (from AS 7 ix 29).

548. Although sa

-du

is usually translated as “regular offerings” it is certain that a translation “regular

delivery” is more correct, since it applies to humans as well as gods.

222

BRM 3, 12 (from

∑S

4 x 30), has been chosen as a typical example of a text from this

group:

551

Side 1.

1. 4(barig) 5(ban

) 5(diÒ) sila

kaÒ sag

4 barig, 5 ban and 5 sila of good beer,

2. 1(aÒ) 4(barig) 5(ban

) kaÒ DU gur

1 gur, 4 barig, and 5 ban

of regular quality beer,

3. 1(aÒ) 4(barig) 1(ban

) ninda DU gur

1 gur 4 barig, and 1 ban

of regular quality bread,

4. 1(ban

) 7(diÒ) 1/2 sila

1 ban and seven and a half sila of best

igi-sag sum gaz

quality crushed garlic,

6. 1(ban

) 1/2(diÒ) sila

-geÒ

1 ban and a half sila of sesame oil,

7. 7(diÒ) sila

4(diÒ) gin

naga

7 sila 4 shekels of crushed alkali,

¿gaz

•(KUM)

8. sa

-du

kas

Òa

umma

regular provisions for the couriers

while in Umma,

9. ¿giri

•

Òara

-za-me

via ∑ara-zame.

Side 2.

1. ¿kiÒib• lu

-kal-la u

Sealed by Lukala and

549. For a location of the tower of the Girsu-canal see W. Heimpel, “Towards an Understanding of the term

siKKum,” RA 88 (1994) 9 and 18.

550. The first example from this group of texts is Nebraska 43 (from AS 5 ix), which was sealed with the seal

of Ur-E’e. The two last texts date to IS 2 month 6, one (CST 872) sealed by Lukala, the other (CST

873) by Ur-Nungal. The last text with the seal of Ur-E’e is AR RIM 4, 28 (from

∑S

4 vi). However, Ur-

E’e is mentioned as late as

∑S

5 viii, in the body of MVN 15, 96. Unfortunately the seal on that text is

illegible. The first text to be sealed by Lukala was Hermitage 3, 512 (= Erm 15010) (from AS 7 i), the

first to have the seal of Ur-Nungal was Hermitage 3, 515 (= Erm 15280) (from AS 8 iii). Ur-Nungal

sealed some documents alone, and several together with Lukala, but never with anyone else. Lu-

aya

sealed only one text: OrSP 47-49, 457 (from

∑S

9 viii).

551. See F. Pomponio, "Lukalla of Umma," ZA 82 (1992) 172 - 179, for a discussion of these texts.

223

2. ur-

nun-gal-ka

Ur-Nungal,

3. iti ezem

Òul-gi u

3(u)-kam

Month

“

Festival of ∑ulgi

”

, on the 30

day,

4. mu

Òu-

suen lugal uri

-ma-ke

Year:

“

∑u-Suen, the king of Ur, built

bad

mar-tu mu-ri-ig-ti-id-ni-im

the Amurite wall (called) Muriq-tidnim

”

mu-du

5. ur-gi

-par

u

] lu

-sukkal?

Ur-Gipar and Lu-Sukkal made it firm.

-gi-ne

]

Side 3.

Seal 1

1. ur-

nun-gal

Ur-Nungal,

2. dub-sar

scribe,

3. dumu ur-

Òara

son of Ur-∑ara,

4. Òa

-dub-ba-ka

the archivist.

Seal 2.

1. lu

-kal-la

Lukala,

2. dub-sar

scribe,

3. dumu ur-e

-e ÒuÒ

son of Ur-E

’

e, the chief cattle administrator.

These deliveries could also include animals. So far, no connection between these texts

and any other group of texts has been observed—the fact that Ur-Nungal seals together with

Lukala on several bullae reinforces our understanding of his position in society as an archivist

of the governor.

224

Chapter 5: Section 10: Lukala, chief household administrator of

the governor

The chief household administrator (

Òabra

) of the governor is an office well

attested in third millennium sources.

552

However, Umma sources from the time of the Third

Dynasty of Ur do not frequently report this title, nor the simpler title chief administrator

(

Òabra

553

Nevertheless, it appears certain that Lukala, and along with him his predecessors,

Dadaga and Ayakala, and his successor Gududu, all at some point in their careers held an

office comparable to that of the chief household administrator of the governor (

Òabra

ensi

). This claim will be substantiated in the following section.

From the entire corpus of Umma texts, I have found only one text in which Lukala is

not merely called scribe (dub-sar). In this text, MVN 16, 1294, Lukala is called chief

administrator (

Òabra

). The title

chief household administrator

is attested frequently in Ur

III Girsu texts, where the office seems to relate to the ruling family and the household of the

governor.

554

Since a few Girsu texts also mention a Lukal(kal)a with the title chief

administrator

(Òabra

) (for example, TCTI 2, 4177 (no date) obverse, column 1, line 6; udu

-kal-kal-la

Òabra

) great caution must be taken when evaluating the single text which

mentions Lukala, the

Òabra,

in an Umma context.

555

552. J.-P. Grégoire,

Archives Administratives Sumériennes

, (= AAS; Paris 1970) xv.

553. For the use of the title

Òabra

in the agricultural record see K. Maekawa, Zinbun 22 (1987) 40.

225

MVN 16, 1294 (from

∑S

3 x):

Obverse

1. 1(diÒ) u

ba-uÒ

One ewe, slaughtered.

2. Òe-ta sa

-a

Bartered for barley,

3. ki lu

-kal-la sipa-ta

From Lukala, the shepherd,

Reverse

1. kiÒib lu

-kal-la Òabra

Sealed by Lukala, the chief administrator.

(seal)

2. iti ezem

Òul-gi

Month

“

Festival of ∑ulgi

”

2. mu si-ma-num

ba-

Year:

“

Simanum was destroyed

”

Seal

1. lu

-kal-la]

Lukala

2. dub-sar

the scribe

3. dumu ur-e

-e ÒuÒ

]

son of

Ur-E’e

chief cattle administrator].

The seal rolled on this text is the standard seal of Lukala which was rolled on more

than 500 texts. Had this text been a random receipt from an unknown person, its relevance

would have been more restricted. This receipt from the rather well known shepherd

554. In Girsu the well-known system of a temple administration headed by a sanga-priest and a

Òabra-

administrator is elucidated by texts such as HSS 4, 4 (see also R. Englund,

Ur-III-Fischerei

(1990) 58 -

63). The existence of an office of chief household administrator of the governor is also amply attested in

the sources, see, for example, TUT 259 (from

∑

48xi), mentioning Ur-Lamma, chief household

administrator. He is otherwise known primarily through his sons. See also TCTI 2, 4226 (from

∑S

1),

mentioning Lu-Ningirsu the chief household administrator. The activities of the Girsu chief household

administrator seem to have been primarily concerned with the daily management of the household. YOS

18, 115 (no date), rev. vii, may allude to the staff of the chief household administrator.

555. Two Umma texts mention a person Lukala with the title

Òabra

, without any positive means of

connecting this person to Lukala the son of Ur-E’e: Georgica 8.3.2 (from AS 7), obv. ii 23; and MVN

21, 343 (from

∑S

3), obv. i 18 and rev. i 1.

226

Lukala

556

is part of a long sequence of receipts of dead animals, sealed by Lukala, son of Ur-

E’e, during his tenure as the chief household administrator of the governor. The phrase

“bartered for barley” appears in three texts sealed by Lukala, the son of Ur-E’e; twice the

shepherd was Imani (title inferred),

557

and twice the shepherd was named Lukala (title

given).

558

All four texts were from either month 10 or month 12 from the year

∑S

3, and

three out of four of these texts recorded only one ewe. The phrase “bartered for barley”

occurs with some frequency in the Umma material altogether, but only in recording reed and

animals.

559

When Dadaga, and before him Ayakala, sealed receipts concerning deliveries of dead

animals from individual shepherds, it is credible that they too held the office as chief

household administrator. However, no shepherds were called Dadaga or Ayakala, eliminating

the need for ever using a title when recording these transactions. It is clear that the reason

behind using the title of Lukala, son of Ur-E’e, in this particular text lies with an urge to

avoid any misunderstanding of who was chief household administrator (

Òabra

)) and who

was the shepherd (sipa(d)) in a record where both were named the same. However, we must

556. See for example NYPL 330 (from

∑

45), obv. 1: 3(barig) zi

-kal-la sipa, and MVN 15, 390 (from

∑

37 iii to vii), obv. i 30: lu

-kal-la sipa.

557. See MVN 16, 1559 (from

∑S

3 x), and MVN 14, 320 (from

∑S

3 xii); both texts record one ewe.

558. See MVN 16, 1294 (from

∑S

3 x), mentioned above, and MVN 14, 389 (from

∑S

3 xii); both texts

record one ewe, MVN 14, 389 adds a male sheep and one goat.

559. “Bartered for barley” can be compared to the phrase “bartered for silver” which is attested even more

rarely, and in an unqualified contect. “Bartered for silver” is used in relation to all aspects of the

economy, including man-days.

227

consider the parallel text MVN 14, 389 (from

∑S

3 xii), which also mentions Lukala, the

shepherd, as the delivering agent of several dead animals, and Lukala as the sealing party.

Although the text is sealed with the same seal of Lukala, the son of Ur-E’e, the text lacks any

identification of Lukala as chief (household) administrator

(Òabra

). There is other evidence,

however, that suggests that Lukala held the office of chief household administrator, above all

his involvement in the business of the province as described below, and the bala-contribution

of the governor.

Close to 300 texts have been published recording deliveries made by several

shepherds, and sealed by either Ayakala, Dadaga, Lukala, or Gududu (some other Umma

administrators also sealed similar tablets from the same shepherds, but these transactions and

their possible interpretation are noted under the respective persons).

560

The first of these

texts in chronological order was Aleppo 368 (from

∑

37 iii); it recorded the delivery of one

dead ewe (further classified as foreign (kur)) from Ur-

IÒtaran

, sealed by Ayakala. Ayakala

sealed approximately 16 such texts from

∑

37, month 3, to

∑

39, month 2,

561

with his

regular seal (see above); all of these animals were classified as “slaughtered” (

ba-uÒ

), and all

were delivered by either Ur-

IÒtaran

or Urru, two well-known shepherds. Dadaga who was

mentioned as the sealing party in 34 texts took over after Ayakala in

∑

39 month 9.

562

560. These texts should not be confused with other receipts for animals, see, for example,

CST 781 (from

∑S 2

), recording animals for siskur-offerings, received by Dadaga, and sealed with Dadaga’s own seal.

This document falls within the period of Dadaga’s nephew Lukala’s tenure as chief household

administrator.

561. MVN 14, 5 (from

∑

39 ii), and OrSP 47-49, 206 (from

∑

39 ii).

228

Dadaga received the animals from the shepherds Ur-

IÒtaran

, Urru and Lugal-azida. Dadaga

is mentioned in this connection for the last time in the 12

month of

∑ulgi

’s last year.

563

Dadaga, however, never used his own seal to seal these tablets. Rather, all of the texts that

mentioned Dadaga as the sealing party had the seal of a person named Luduga rolled upon

them. During the first half of the period when Dadaga held office as chief household

administrator, the seal he used was that of Luduga, the son of Nigar-kidu. During the latter

half he used the seal of Luduga, the son of Ur-Nigar. No texts have been found dating to the

first year of Amar-Suen. From AS 2 the same sort of deliveries were sealed by Lukala.

564

Lukala sealed more than 200 documents recording the delivery of one or several dead

animals. Lukala was presumably followed by Gududu, in

∑

S 9, as the Umma official sealing

receipts of dead sheep and goats. The last text mentioning Lukala as the sealing party in this

is from

∑S

6, month 13,

565

the first to mention Gududu is from

∑S

9, month 10.

566

I have

no explanation for this break in the textual record. Only a few texts from this group sealed by

Gududu have survived, perhaps due to the late date of his tenure and the corresponding

decrease in records altogether. Below (figure 9) is a graphic representation of the sequence of

people holding the office of governor and those holding the proposed office of chief

562. MVN 14, 66 (from

∑

39 ix). One text sealed by Dadaga; Akkadica 7 pl. 1 2, falls within the sequence

of Ayakala.

563. Aleppo 395 (from

∑

48 xii).

564. See Ligabue 15 (from AS 2 vii). No texts have been found from AS 3, but the sequence restarts with BM

107994 (from AS 4 i).

565. SANTAG 6, 307 (from

∑S

6 xiii); see also CST 793 (from

∑S

7).

566. MVN 14, 493 (

∑S

9 x).

230

It cannot be proven that Lukala, the son of Ur-E’e, was active in the administration

of Umma prior to

∑

48. A person named Lukala with the title of overseer appears in a

number of texts from before this date, but there is no reason to believe that he was identical

with the son of Ur-E’e by the same name.

567

Between Amar-Suen’s first year and the end of

Ur domination over Umma, only one person named Lukala carried a seal. Thus it is likely

that all attestations of a high-ranking Lukala—a Lukala entitled scribe—referred to Lukala,

son of Ur-E’e, the member of the ruling family of Umma. Nothing is known about Lukala’s

own family. Since he is never mentioned with a title specific to any office (save for the one

example quoted here), it is impossible to ascertain whether any of the people claiming to be

sons of Lukala were, indeed, sons of Lukala, the chief household administrator.

568

567. Note, however, the dedicatory seal rolled on BIN 5, 232 (from

∑

37): ur-

-si

/ ensi

umma

/ lu

kal-la //

¿x•

-[x] / ir

-[zu].

568. Among the other people in Umma called Lukala we find Lukala the cultivator (engar) (for example, AAS

83 [no date]). The text Aleppo 51 (no date), has a peculiar seal inscription mentioning Lukala, the son of

Lugal-ezem. Compare this text to Aleppo 132 (no date), which is said to be sealed by Lukala, but carries

the seal of Lugal-ezem, the well known agricultural overseer and son of Lugal-Ema

e. This Lukala, son

of Lugal-ezem, then, is likely to be identical with the cultivator by that name; he was perhaps a junior

member of the family of Dada. See, in that regard, SAT 2, 77 (from

∑

33 vi), obv. ii 9, which offers

conclusive evidence in support of this reconstruction. Among the other people named Lukala, we find

Lukala, the son of Lugal-Suen (see for example, Aleppo 62 [from

∑

35vii], with seal inscription); Lukala,

son of Uludi (ulu

-di) (see for example, UCT 3, 334 [no year, month 11], with seal inscription); Lukala,

the shepherd (see, for example, NYPL 121 [from AS 6]; MCS 1, 54, BM 106045 [from

∑S

5] (title na-

gada); MVN 15, 390 [from

∑

37 iii to vii] (title sipa)); Lukala the oil-presser (for example, BIN 5, 277

[AS 8?]). Lukala, the forester (see, for example, BM 104774 (unpubl.)); Lukala the police-officer (see for

example, MCS 7, 22 BM 105330 [from

∑

43]); Lukala, the overseer (ugula) (see, for example, MVN 13,

316 [from

∑

42 ii]; Rochester 173 [from

∑

37 v]); Lukala, the scribe of beer (for example, Nebraska 39

[no year, month 11]).

231

As long as Lukala is attested in the extant sources, he also held the office of chief

household administrator of the governor. Lukala very seldomly sealed documents which seem

to relate to the activities of a provincial administrator, and no certain use of a nam-

Òatam

seal can be ascertained for Lukala.

569

All the activities of Lukala can, therefore, with some

justification be claimed to represent those of a chief household administrator.

Lukala not only oversaw agricultural work done by workers from the teams of

agricultural workers as well as workers from the workshops of the Umma households,

570

also functioned as a foreman of large work-crews himself.

571

Relatively few accounts of the work of female dependent workers (nig

-ka

-ak a

geme

) have been recovered.

572

Three of these mention Lukala in the colophon: Aegyptus

569. One text concerned with labor, SNAT 495 (from

∑S

4), directly states that it has been sealed with the

nam-

Òatam

seal of Lukala; however the seal rolled on that document was published as: lu

-kal-la / dub-

sar / dumu ur-nigar

(NIGIN

)

gar

ÒuÒ

(see also MVN 15, 83 [from

∑S

4]). This unique reference

might, however, be the result of a copying error. The text was unavailable to me, and the seal is not

included in the catalogue of R. Mayr,

Seal Impressions

(1997). Note that the parallel text AAS 26 (from

∑S

6 vii to viii), which failed to use the phrase ‘

kiÒib

nam-

Òa

-tam’, was sealed with a seal of Lukala

where the patronym is apparently missing.

570. See MVN 16, 1047 (from

∑S

6 iii), recording workers from Agu’s basketry workshop working in the

fields, sealed by Lukala, Nik 2, 103 (from

∑S

6), recording leather workers likewise doing field-work

sealed by Lukala, and credited to Ayakala (presumably the chief of the leather-workers). Whereas one

account concerning Agu has been found (TCL 5, 6036 [from AS 4]), no such text has ever been

discovered concerning Ayakala.

571. It is possible that the work-crew of Lukala was previously managed by Dadaga when he held the office of

chief household administrator; see, in particular, CHEU 49 (from

∑

48 ii 1), where Dadaga is listed first

among the well-known Umma milling-foremen as being responsible for a work-crew of his own.

232

21, 159 (from AS 8 i to xi); MVN 21, 200 (from AS 2); and MVN 21, 201 (from AS 5).

573

The amount of work-days recorded in each of these three texts surpasses, by far, those of the

other accounts of the work of female dependent workers. Only the account concerning Lu-

dingira, the scribe of flour, rivals that of Lukala.

The first entry in the “debits” section of all three accounts (in MVN 21, 201

following a “remainder”) is an artificial computation of the average number of female

dependent workers available each day during the period of accounting. The calculation of the

work-days of the crew was presumably aided by the work-crew inventories mentioned above,

pp 74 - 78. A normal work-crew of female dependent workers seems to have averaged 30

women in total, Lukala’s crew totaled more than one hundred dependent female workers.

574

It seems reasonable to assume that this work-crew was the permanent staff of the mill

of the governor’s household.

575

Producing for the governor and his bala-obligations.

Three large accounts have survived that closely tie Lukala and the household of the

governor to the production of pottery.

576

Whereas the summary statement “pottery-worker-

572. JNES 50, 255-80 (from

∑S

4), the name of the responsible person is broken; MVN 21, 204 (from

∑S

7?), account concerning Lu-balasaga; STA 2 (from AS 4), account concerning Lu-dingira, scribe of flour;

TCL 5, 5669 (from

∑

48 i to xii), account concerning Lugal-inimgina.

573. In both texts, from MVN 21, almost the entire “credits” section is devoted to milling grain for the bala.

574. See also AnOr 7, 226 (from

∑S

6), recording a transfer of 3 female milling workers from the crew of Ur-

Nintu to the crew of Ur-Suen and Dagu, sealed by Lukala (with regular seal of Lukala).

575. See, for example, YOS 18, 115 (no date), rev. i 2-3 and 14, perhaps to be understood as an inventory of

the female workers of the mill of the governors household, among other staff.

576. MVN 1, 232 (from

∑

43 iii); MVN 21, 203 (from AS 8); and MVN 1, 231 (from AS 4).

233

account concerning Lukala” is only attested in the recently published text MVN 21, 203

(from AS 8), it can be suggested that at least one of the other texts contained such a

statement, namely MVN 1, 231. This text is an exact parallel from AS 4; the work-crew of

the two texts is largely identical, as is the production according to standing orders.

577

The pottery-workshop of the governor’s household is the only such workshop

attested in Umma; however, the production figures of that workshop suggest that it had

monopolized Umma pottery production. Contrary to what has been suggested in some

recent studies,

578

there is now evidence to suggest that the potter—like other work-groups—

worked full time for the state, and with only a limited number of “days off ” (a

-du

-a).

579

Lukala did not personally manage the delivery of all the finished products (which

were credited to his account); rather, a member of his work-crew (presumably an overseer)

was responsible for transferring most products from Lukala’s workshop to other Umma

households.

580

577. D. Snell,

Ledgers

(1982) 96 - 99 + appendix 2, 270 - 278. Snell did not describe standing orders within

the production of household items. Both the workshops discussed here, as well as the basketry workshop

documented in the account TCL 5, 6036, produced according to standing orders.

578. P. Steinkeller, “The Organisation of Crafts in third Millennium Babylonia: The Case of Potters,” AoF 23

(1996) 232-253. W. Sallaberger,

Der Babylonische Töpfer

(= UHEM 3; Ghent 1996).

579. The work-crew in MVN 1, 231, and MVN 21, 203, is almost identical. In both texts the foreman is

debited with the full work-time of the workers, that is, all 12 month of the year. See the fortcoming

study by the author. Compare this with V. Struve, “Some new data” (1969) 139.

580. My interpretation differs here substantially from that of W. Sallaberger, UHEM 3 (1996) 35.

234

Since Lukala appears very frequently as the contributor, or recipient, of large (as well

as small) amounts of silver, and since the silver involved in the so-called balanced trade-agent

accounts have received enormous attention from historians of late third millennium

economic history, the office headed by Lukala was named the fiscal office of Umma, or the

“irrigation office,” and Lukala himself, the “silver comptroller” of Umma.

581

The amounts of silver debited or credited to the account of Lukala did not, as a rule,

represent real metal, but rather obligations ennumerated in terms of silver equivalencies. This

can be inferred from the fact that it is impossible to trace any particular amount

(representing a piece) of silver from one account to another, which would have happened had

the metal been a physical commodity. The amounts of silver received by Lukala from a wide

range of individuals represented mostly arrears from shepherds, cultivators or foremen.

Amounts of silver were also entered into the books as payments for the so-called “field

interest” the

maÒ-aÒaga

. It is likely, but not certain, that Lukala was transferring this metal

into the “debits” section of the accounts of the trade-agents.

Lukala was presumably also the overseer of the Umma colony of trade-agents, as

suggested by TCL 5, 6037, the colophon of which is given here:

TCL 5, 6037 (from

∑S

6):

Reverse column 10.

...

581. See P. Steinkeller, JESHO 24 (1981) 121. P. Steinkeller, AOS 68 (1987) 76 + fn. 17. T. Maeda, ASJ 18

(1996) 254-260. Snell gave Lukala the title “Comptroller,” and devoted a section to his office, see, D.

Snell,

Ledgers

(1982) 77 - 81.

235

10. nig

-ka

-ak dam-gar

-ne

Account of trade-agents,

11. lu

-kal-la

concerning Lukala.

12. mu

Òu-

suen lugal uri

-ma-ke

Year:

“

∑u-Suen, the king of Ur

na-ru

-a ma



en]-lil



nin]-lil

-ra erected the lofty stela of Enlil and

mu-ne]-¿ru

•

Ninlil

”

Although no such account has survived it is the hypothesis of this study that Lukala

managed the flow of wealth from Umma to the imperial center, as is indicated by the tablet

container Hirose 405 (from

∑S

to 8):

1. pisan dub-ba

Tablet container:

2. nig

-ka

-ak bala-a

Bala-accounts

3. lu

-kal-la

(concerning) Lukala,

4. mu ma-da za-ab-Òa-li

from the year:

“

ZabÒali (was destroyed)

”,

5. u

mu ma

-¿gur

• ¿ma

•

and the year:

“

the lofty barge

(was fashioned)

”

Reverse.

1. i

-¿gal

•

are present.

He certainly also managed the wealth of the governor, which is above all ascertained

from the colophon of the short account Ledgers pl. 23 13 (from

∑S

1 to 5):

Reverse.

...

9. nig

-ka

-ak ku

ensi

-ka

Silver-account concerning the governor,

10. giri

-kal-la

conveyor: Lukala.

....

As a collector of field interest and a foreman of the Umma trade-agent colony Lukala

naturally transferred large amounts of silver from one account to another. The economic

237

Chapter 5: Section 11: Ir(mu), chief of the granary, brother of the

governor

Ir(mu),

583

the brother of the governor, held the office “chief of the granary”

(Sumerian KA- guru

) from the middle of the reign of

∑ulgi

(

∑

33) until at least the end of

Amar-Suen’s reign, and perhaps considerably longer. I have already referred to the texts

concerning Ir(mu)’s accession to office when describing Ur-Lisi, Ir(mu)’s predecessor as chief

of the granary (p. 156 + fn. 398). In this section I will try to describe his office in greater

detail.

584

Little is known about Ir(mu)’s own family; the few extant references to his wife do

not reveal her name.

585

We only know, for certain, the name of one of Ir(mu)’s sons,

∑ara-

izu, although he may have had more than one.

586

This

∑ara-izu

was certainly identical with

the person mentioned in the colophon of TCNU 468 (from IS 3), a text which seems to be

583. Ir(mu)’s name is almost exclusively spelled ir

in the body of the texts, but always ir

-mu in his seal-

inscription.

584. A guru

is both a very large unit of meassurement in the capacity system (equal to 3,600 gur or

approximately one million liters), as well as a word for the physical structure of a granary. The sign KA

may refer to the opening (mouth) of the silo, analogous to the Sumerian word for the opening of a canal,

ka i

-da. Thus, the KA-guru

may be interpreted as the one who controls the opening of the granary.

585. See for example MVN 16, 908 (from AS 8 iii), obv. 10: dam ir

KA-guru

, and perhaps also MVN 2,

176 (from

∑

48 vii), rev. viii 9': dam ir

-mu.

586. Seal of

∑ara

-izu see, for example, MVN 2, 128 (from

∑

S 7):

Òara

-i

-zu/ dub-sar / dumu ir

KA-guru

238

an almost exact parallel to an earlier account concerning Ir(mu) (ASJ 19, 226 72 (no date),

see below). If that observation is correct, then there is reason to believe that towards the end

of Ur domination over Umma, Ir(mu) was succeeded by his son,

∑ara

-izu, as chief of the

granary.

587

From the end of Amar-Suen’s reign, the name of the chief of the granary ceases to be

recorded. The last explicit reference to Ir(mu), holding the title of chief of the granary, is

from AS 8, in a text mentioning his wife.

588

However, the absence of references to Ir(mu)’s

name together with the title of chief of the granary is not sufficient evidence to show that

that office was not in his hands any longer, since throughout the history of Ur III, Umma

references to the name of the chief of the granary were relatively rare, whereas the title was

mentioned alone very frequently. Consequently, it has proven quite difficult to reconstruct

the sequence of people holding that office. Here I will tentatively suggest that Ur-Lisi was

chief of the granary until

∑

33, when Ir(mu) took over; he was in turn followed by his son

∑ara

-izu some time after AS 8, and before IS 3.

It should be noted that

∑ara

-izu is never mentioned with the title of chief of the

granary—this resembles the situation at the time of the transfer of office between Ur-E’e and

his son Lu-

aya, and perhaps even between Ur-E’e and his father Ur-Nigar.

587. There are other indications that

∑ara

-izu did, indeed, follow in his fathers footsteps as chief of the

granary, notably the many primary documents from the time of

∑u

-Suen’s reign until the end of Ur

domination over Umma dealing with activities previously managed by Ir(mu). The earliest certain

references to

∑

ara-izu are MVN 13, 353, and SAT 3, 1734, both from

∑S

588. MVN 16, 908 (from AS 8 iii), cited above.

239

In his capacity as the chief of the Umma central granary, Ir(mu) supplied a number

of other institutions with barley for fodder and rations, hence we find him supplying the

fatteners, the foremen of the mill, and foremen of other work-crews. In addition we find

Ir(mu) supplying the trade-agents with barley for barter.

589

As a member of the ruling family

of Umma, Ir(mu) also partook in the agricultural administration and occasionally sealed

documents as a provincial administrator.

590

In addition, Ir(mu) sometimes sealed receipts for

dead animals for the chief household administrator. Ir(mu), in his capacity as chief of the

granary, supplied the animal fatteners with barley, on what seems to be a regular monthly

basis. Although the structure of the Umma animal fattening institution lies beyond the scope

of this study, I will briefly survey the activities of three leading Umma fatteners, Inim-

∑ara

Bida, and Ana

ilibi.

591

The fattened oxen were controlled by several people, but Ir(mu)

seems to be the main supplier of fodder for these animals as well.

592

The interaction between

fattener and chief cattle administrator has been dealt with above.

Ana

ilibi’s herd of approximately 100 animals supplied the provincial court, and

presumably also the imperial court, with fattened animals. Ana

ilibi received the barley and

bran for the fattening directly from the governor and his chief household administrator. On

the other hand, Inim-

∑ara

593

one of the main animal fatteners in Umma, received fodder

589. See in particular SA 76 (from AS 4), and OrSP 47-49, 411 (from

∑S

3).

590. Only one text is sealed with the nam-

Òatam

seal of Ir(mu) (MVN 18, 543 [no date]). Some of Ir(mu)’s

sealed documents (which do not mention the use of a nam

-Òatam

seal) do, however, resemble

documents otherwise normally sealed by provincial administrators, suggesting that Ir(mu) too held that

position.

240

for his animals directly from Ir(mu).

594

Inim-

∑ara

was perhaps connected with the

591. The most important sheep-and-goat fattener during the latter part of

∑ulgi

’s reign, and presumably

during the reign of Amar-Suen as well (and perhaps beyond), was Inim-

∑ara

. See, for example, Syracuse

169 (from

∑

48 v 5), where Inim-

∑ara

figures as the most important among the three well-known

fatteners Inim-

∑ara

, Bida, and Ana

ilibi:

Obverse.

1. 4(geÒ

) 5(u) 3(diÒ) sa gi zi

293 bundles of zi reed;

2. inim-

Òara

kuruÒda

Inim-∑ara, fattener.

3. 4(geÒ

) 4(u) 4(diÒ) sa gi zi

284 bundles of zi reed;

4. bi

-da kuruÒda

Bida, fattener.

5. 1(geÒ

) 4(u) 5(diÒ) sa gi zi

105 bundles of zi reed;

6. an-na-Ìi-li-bi kuruÒda

AnaÌilibi, fattener.

Reverse.

1. u

5(diÒ)-kam

On the fifth day.

2. gu-kilib-ba 6(diÒ) sa-ta

In each sheaf there are 6 bundles.

3. giri

igi-peÒ

Via IgipeÒ,

4. kiÒib Ìu-wa-wa

Sealed by Îuwawa.

5. iti RI mu-us

-sa ki-maÒ

mu-us

-sa-a-bi

Month RI, Year after:

“

KimaÒ

(was destroyed), the year after (that)

”

Seal

1. lu

-eb-gal

Lu-Ebgal,

2. dub-sar

scribe,

3. dumu ur-gi

-par

son of Ur-Gipar.

UÒmu

and his circle perhaps operated at a higher level of the administrative hierarchy than the three

fatteners mentioned here.

592. Nigar-kidu, an important fattener of oxen in Umma, received fodder from Ir(mu) in Aleppo 302 (from

∑

45 xii); note that Atu the chief cattle administrator received fodder for calves from Ur-E’e in AnOr 1,

51 (from

∑

44).

593. For the seal of Inim-

∑ara,

see for example MVN 21, 156 (from

∑

46 xiii):

inim-

Òara

/ dumu ur-

dingir-ra / kuruÒda

Òara

For the seal of Inim-

∑ara

’s son and successor as fattener see CST 797

(from IS 2):

a-lu

-lu

/ dumu inim-

Òara

/ kuruÒda

Òara

-ka.

241

household of

∑ara

, Bida—another fattener—with the Umma bala account.

595

The chief of the granary also supplied the foremen of the workshops and the

agricultural foremen (these are the foremen called “captains of (plow-)oxen” in some contexts

and “foremen” in others) with barley rations for the dependent workers under their care and

control.

596

Fodder (Sumerian

Òa

-gal) was also given to the draft donkeys (

anÒe

kunga

597

other donkeys,

598

and plow oxen (Sumerian gu

apin), and, not to be forgotten, the gods of

the Mesopotamian pantheon.

599

594. See, for example, AnOr 7, 182 (no date); ASJ 19, 226 72 (no date), the account concerning Ir(mu)

discussed below (p. 242) (see rev. vii 2’-3’: 1(u) 1(

aÒ

) gur /

Òa

-gal udu niga /

kiÒib

inim-

Òara

); and

CHEU 27 (from

∑

39 iv).

595. See also the milling-accounts below, p. 244 + table 2, in which Bida contributes to the “debits” of three

accounts.

596. Rations for humans were often called “barley-rations” (

Òe

-ba), but could be called “fodder” (

Òa

-gal). See

for example Orient 16, 71 96 (from

∑S

1 x), concerning rations (

Òe

-ba); AAICAB 1, 1911-170 (from

∑

2 x to xi), concerning rations (

Òe

-ba); MVN 13, 276 (from

∑S

6 i), concerning fodder (

Òa

-gal); and

MVN 16, 1272 (from

∑S

3 xii), concerning fodder (

Òa

-gal).

597. Delivery said to be from Ir(mu); see, for example, AnOr 7, 174 (from

∑

44 xii), and AnOr 7, 179 (from

∑

38 x). Delivery said to be from the chief of the granary (no mention of personal name); see, for

example, ArOr 25, 561 20 (from

∑S

6 iv).

598. Delivery said to be from the chief of the granary (no mention of personal name); see, for example, ASJ

19, 220 58 (from

∑S

5 xi); and BCT 2, 256 (from

∑S

3 ix).

599. Rations for the gods were mostly called sa

-du

, which seems to be the correct term for describing

“rations” for deities; see, for example, ASJ 19, 220 56 (from

∑S

5 iv): sa

-du

for Nin-Egal; MVN 3,

269 (from

∑S

5): sa

-du

for the divine Amar-Suen; and Orient 16, 73 103 (from

∑S

5): sa

-du

for

∑ara

. However, see AnOr 7, 309 (from

∑S

9), rev. 3:

Òa

-gal udu niga sa

-du

Òara

dingir-re-ne,

clearly advocating a distinction between sadu and

Òagal

242

Ir(mu) appears in numerous texts as the supplier of amounts of barley converted into

work-days at fixed rates, mostly at an equivalence of between 6 and 8 sila to one day. These

were the “wages” for the hirelings (Sumerian a

un-ga

600

ASJ 19, 226 72, dating to the earliest part of Ur III Umma history,

601

is a 10-column

account (column 10 was perhaps blank), concerning Ir(mu). The “debits” section recorded

the deliveries of barley to the granary by temple-household administrators; the “credits”

section, on the other hand, recorded the deliveries of barley and wheat to a number of

administrators and foremen working for one or more households.

602

TCNU 468 is a shorter

account recording a smaller amount of barley, but otherwise an exact parallel to ASJ 19, 226

600. See, for example, the two accounts TCL 5, 5675, and TCL 5, 5676. In TCL 5, 5675 (from AS 4 i to

xii), Ir(mu) delivered 5 gur, 2 barig and 4 ban of barley to be converted at a rate of 7 sila per day (= 237

work-days), and 7 gur, 4 barig, 1 ban and 8 sila of barley to be converted at a rate of 8 sila per day (= 393

work-days) to the debit section of the account of Lugal-gu’e the captain of (plow-)oxen. In TCL 5, 5676

(from

∑S

2), the chief of the granary delivered 36 gur and 80 sila of barley which was to be converted at

a rate of 6 sila per day (= 1813 1/3 work-days), to the account of Ur-Ninsu the captain of (plow-)oxen.

Primary documents were, as we have seen, used to draw up the accounts. However, none of the primary

documents recording transfers to the “debits” of the two accounts cited above have been found. As an

example of such a document see AUCT 1, 681 (from

∑

45).

601. Ir(mu) is mentioned in the colophon as Ur-Lisi’s brother, and Ur-Lisi is not (yet?) called governor

advocating an early date for this text.

602. See, for example, rev. vii 2’ - 14’: 1(u) 1(

aÒ

) gur /

Òa

-gal udu niga /

kiÒib

inim-

Òara

/ 8(

aÒ

) 2(ban

)

gur /

kiÒib

uÒ

-mu / 2(

aÒ

) gur /

Òa

-gal

anÒe

kiÒib

ur-dingir-ra / 3(barig)

Òa

-gal

muÒen

kiÒib

-lu

/ 1(u) 4(

aÒ

) 2(barig) gur /

Òe

-ba-

Òe

kiÒib

lugal-nig

-lagar-e ugula

uÒ

-bar, “11 gur as fodder for

the fattened sheep, sealed by

Inim-∑ara

, 8 gur and 2 ban, sealed by

UÒmu

, 2 gur as fodder for the

donkeys, sealed by Ur-dingira, 3 barig as fodder for the birds, sealed by

alulu, 14 gur and 2 barig as

rations, sealed by Lugal-niglagar’e, foreman of the weaving mill.”

243

72. Perhaps due to a separation of many years very few persons were still in office when

TCNU 468 was written.

603

Barley was processed in the mill (Sumerian kikken

604

). In the following I will briefly

discuss the administration of the work and the products of the mill. One group of six

foremen of the mill (ugula kikken

) seems to have operated together as one unit. They were

perhaps all associated with the same household. The summary document AAICAB 1, 1911-

485 (from AS 1), recorded the “arrears” from the accounts of these foremen, the colophon of

that text reads:

Reverse.

(blank space)

1. ∑U+NIGIN

-ia

5(geÒ

’

u) 1(geÒ

)

Total deficits: 3060 1/3 work-days,

1/3(diÒ) geme

1(diÒ)-Òe

2. ∑U+NIGIN

diri 4(geÒ

) 1(u) gin

Total surplus: 240 1/6 work-days,

geme

1(diÒ)-Òe

3. diri la

-ia

geme

-ka

(It is) the surplus and the deficits of the

work of female dependent workers,

4. ugula kikken

-na-ke

-ne

of the foremen of the mill.

5. mu

amar-

suen lugal

Year:

“

Amar-Suen (became) king

”

(blank space)

603. See, for example,

UÒmu

: ASJ 19, 226 72, rev. vii 6’ // TCNU 468, rev. ii 8.

604. A reading kikken

AR.

AR is supported by numerous Ur III attestations of kikken

1-2

-na, a reading

ar-

ar-ra, or perhaps ar

-ar

-ra cannot be excluded for the title of the foreman of the mill; see MVN 6,

84 (from

∑

46), obv. 5; ab-ba-mu

AR-

AR-ra, as well as for the verb “to mill, grind, etc.” ar

-ra

(

†

u) ar

-ar

-ra (

marû

244

From the same text we are able to produce this table:

Four of the same foremen are also known to us from their milling-accounts,

recording a “debits” and a “credits” of both female dependent workers and grain (except for

∑ara

-zame’s account MVN 21, 204, which does not include grain).

The following table (2) suggests that the same foremen participated in a “planned”

economy, where grain as well as labor was distributed to the mill-foremen centrally, and

according to fixed rates:

Table 1: Foremen of the mill (from AAICAB 1, 1911-485)

a. Dingira, Ur-Nintu,

∑ara

-zame, and Lu-balasag are also known from CHEU 49 (from

∑

48 iii

1).Lugal-Emahe, Lu-balasag,

∑ara

-zame, Ur-

∑ulpa

’e, Adu, and Ur-NIntu are also attested in

TCL 5, 6039 (from AS 05 ii), see p. 77 above.

Foreman

Lu-sa’izu

Ur-Nintu

Ur-∑ara

Lu-balasag

Dingira

∑ara-zame

“

remainder

”

Surplus;
240 1/6
work-days

Deficit;
506 1/6
work-days

Deficit;
620 1/6
work-days

Deficit;
971 work-
day

Deficit;
959 1/2
work-day

Deficit;
3 1/2
work-days

245

It seems appropriate to interpret the data presented above as describing the mill of

one administrative unit, perhaps one household. All four, and perhaps even all six, foremen

mentioned in table 1 worked in close cooperation with the chief of the granary, who supplied

them with the grain to be processed. Further, it seems likely that they all worked closely

together with, and perhaps directly under the administrative control of the chief household

administrator of the governor. Bida, who is mentioned in three of the accounts presented in

Table 2: Foremen of the mill; the “debits”

a. The interaction between the three accounts cited above was discussed by R. Englund,

Ur-III-

Fischerei

(1990) 78 + fn. 263.

Foreman

Lu-sa’izu

(Erlangen 1
from ∑ 48 to
AS 1])

Ur-
Nintu

Ur-∑ara
(TCL 5, 5670

from AS 1])

b. R. Englund,

Ur-III-Fischerei

(1990) 79-90.

Lu-
balasag

Dingira
(TCL 5, 5668
from ∑ 48])

∑ara-zame
(MVN
21, 204
from ∑S 6
to 7])

From
Ir(mu):

216 gur 230
sila of barley,
35 gur emmer?,
16 gur wheat,

193 gur, 25 sila
of barley,
38 gur of
emmer,
33 gur and 40
sila of wheat,

184 gur, 280
sila of barley
35 gur of
emmer,
16 gur 110 sila
of wheat,

From
Bida:

c. Other people contributed to the “debits” as well.

3 gur of barley

3 gur of barley,

Work-
crew

d. The work-crews did not work all year, but rather between 8 and 10 months.

36 females

e. See MVN 21, 212 (from

∑

42 i to xiii), a calculation of Ur-Nintu’s work-crew of female dependent

workers; see also Nebraska 42 (from AS 7).

36 females

246

table 2, was, as we have suggested, connected with the Umma bala, and it would be

interesting to see this group of millers as connected with the Umma bala as well. It is,

however, also possible to see these six overseers as connected directly with the household of

the governor.

Some texts indicate that Ir(mu) too, from time to time, received dead animals

following the pattern defined above. Further, it can be shown that Ir(mu) sealed these

documents almost exclusively when Lukala was not available to do so. That is, in particular,

during the 10

month of Amar-Suen 5.

605

The texts are entirely similar to the texts

described above under Lukala, and the shepherds delivering the animals were the same.

605. One of these texts (Aleppo 381 [from

∑

39 v]), falls right in between the tenure of Ayakala and Dadaga,

the next two texts (Aleppo 391 [from

∑

47 i]; and MVN 14, 62 [from

∑

48 ii]), are from within the

tenure of Dadaga. One text is from month 9 of AS 5 (Syracuse 85), contemporary with several

documents sealed by Lukala. Six texts date to month 10 of AS 5, when Lukala sealed no records (BM

105419 [unpubl.]; MVN 5, 36; MVN 18, 471; MVN 18, 474; SNAT 361; and SNAT 362. Although

Lukala began sealing these documents again in the 11

month of AS 5, Ir(mu) still sealed two more

texts (MVN 4, 85, and Syracuse 86).

247

Chapter 5: Section 12: The other sons of Ur-Nigar

At least two high-level administrators in the Ur III empire were called Ur-Nigar: Ur-

Nigar, the chief cattle administrator of Umma, and the prince Ur-Nigar, a son of

∑ulgi

Early Umma texts mentions Ur-Nigar, the fattener (

kuruÒda

), who is also attested in the

seals of his three sons.

606

The best known sons of Ur-Nigar, the Umma chief cattle

606. The seal rolled on AAICAB 1, 1912-1148 (from AS 5), probably reads: ur-

dumu-zi-da / dub-sar / ur-

nigar

(NIGIN

)

gar

¿kuruÒda•

; compare with Torino 2, 645 (from

∑S

2), and TIM 6, 55 (from IS 2

ix), rev. v 12: ur-

dumu-zi-da engar dumu ur-

¿nigar

•

gar

kuruÒda

. Two other sons of Ur-Nigar, the

fattener, are known from their seals; Lu-Suen and Ur-Halmutum. For the seal of Lu-Suen; see, for

example, MVN 14, 453 (from

∑S

1). For the seal of Ur-Halmutum; see, for example, MVN 16, 1201

(from

∑S

1). The majority of the documents that mention Ur-Nigar, the fattener, date from the 10 year

span between

∑

34 and

∑

44. Ur-Nigar seems to have been in a position with direct responsibilities to

both the imperial administration (based in Drehem), as well as the local court in Umma. Two interesting

documents describes his interaction with a certain Basa and the bala account of Umma: BIN 5, 80 (from

∑

43): 1(

geÒ

) 2(u) 3(

diÒ

) udu bar su-ga / 1(

geÒ

) 7(

diÒ

) sila

bar gal

/ 4(u) 6(

diÒ

)

maÒ

/ zi-ga bala-a

Òa

nibru

// ki ba-sa

-ta / giri

ur-nigar

(NIGIN

)

gar

kuruÒda

/ mu en

nanna

maÒ

-e i

-pa3, “83

sheep without fleece, 67 lambs with fleece, 46 goats, booked out of the bala, in Nippur. from Basa, via

Ur-Nigar the fattener. Year: “the En-priest of Nanna was installed”.” MVN 2, 301 (from

∑

35 xi):

geÒ

) 3 (u) 2(

diÒ

) [udu] bar su-

¿ga•

/ 1(

geÒ

) 3(

diÒ

)

maÒ

/ udu didli u

udu a-[GAR]-ra u

udu

bala-a / ki ur-nigar

(NIGIN

)

gar

kuruÒda

-ta // ba-sa

-dab

/ iti pa

-u

¿e•

/ mu-us

-sa an-

Òa

ba-

ul, “92 sheep without fleece, 63 goats, they are the various sheep, and the xx sheep, and the bala sheep,

from Ur-Nigar, the fattener, Basa seized. Month Pau’e. Year after: “

AnÒan

was destroyed”.” See also Nik

2, 371 (from

∑

41), and SACT 2, 242 (from

∑

43), records of the “debits” (sag-nig

-gur

-ra-kam

Òa

-bi

su-ga) of Ur-Nigar, the fattener; and TLB 3, 37 (from

∑

42), a record of the “debits” of Basa, in which

Ur-Nigar, the fattener, figures among the people contributing to the “debits”.

248

administrator, have been mentioned above; in the following I will try to list and discuss the

remaining sons of Ur-Nigar.

It is important to try and understand the career of these members of the ruling family

since they seem to have been by-passed in the line of succession for the four leading offices in

the Umma central administration. This chapter will therefore seek to answer what became of

these family members.

The following 7 persons were all called “son of Ur-Nigar, the chief cattle

administrator”: Luduga, Lugal-kuzu, Inim-

∑ara, Lu-∑ulgi(ra),

Atu, Lugal-ezem, and

Gudea. The following 5 persons were all called “son of Ur-Nigar” without any mention of his

title: Lu-dingira, Mansum, Lu-

egal, Lugal-

egal,

607

and Ur-Nisaba.

Lu d u g a :

The seal of Luduga;

-du

-ga / dub-sar / dumu ur-nigar

(NIGIN

)

gar

ÒuÒ

was used exclusively on tablets said to be sealed by Dadaga, Luduga’s brother. This paradox is

not easily resolved; the suggestion that Dadaga was another name for Luduga is not

convincing.

608

Although the names of Dadaga and his son Gududu are the only names of

607. Only the text MCS 3, 91, BM 112993 (from

∑S

3), is said to be sealed with a seal including the title of

Ur-Nigar, the father of Lugal-

egal. Note in this regard also SAT 2, 315 (

∑

43 iv), which has the title

-sub

Òara

for Ur-Nigar, the father of Lugal-

egal!

249

members of the Umma ruling family which are not readily explicable according to our

understanding of Sumerian, they still conform nicely with 3

millennium naming practices.

The explanation that Luduga rolled his seal on tablets for a minor brother, fails to account

for why no single tablet bears both the information “sealed by Luduga”

(kiÒib lu

-du

-ga

)

and the seal of Luduga.

However, after scrutinizing the approximately 34 texts that contained the seal of

Luduga, the following picture emerges: 22 texts are receipts of dead animals of the type

mentioned several times before. The first text is from

∑

36 and the last from

∑

48; a majority

of the texts are from

∑

48 (12). The names of the shepherds and the number of dead animals

are similar to all the other texts from the same group. These texts were used above (pp. 224 -

226) in defining the time-frame for Dadaga’s tenure as chief household administrator.

Four texts, from

∑

45 (1 text), 46 (2 texts), and 48 (1 text), were concerned with

work and resemble other texts from the same period sealed by Dadaga. The remaining 12

texts record various minor transactions, with one exception: Ontario 2, 312 (unpubl.), which

is an administrative document referring to the fulfillment of the “debits” of a silver-account

concerning the trade-agents. This suggests that Luduga was not active in the state

administration, but that he occasionally sealed tablets for his brother Dadaga while the latter

608. Although no conclusive evidence from the Third Dynasty of Ur in favor of this theory can be offered, it

remains a topic for further investigation. The theory in question here was put forward by R. Mayr,

Seal

Impressions

(1997) 141 - 143. The use of double-names is, as far as I know, only attested for Neo

Babylonian Sippar (see most recently M. P. Streck, “Das Onomastikon der Beamten am

neubabylonischen Ebabbar-Tempel in Sippar,” ZA 91 (2001) 110 - 111).

250

served as the chief administrator of the household of the governor (

Òabra-e

ensi

). See pp.

151 - 152 above for a brief discussion of Luduga, the son of Nigar-kidu, whose career in

many ways resembled that of his cousin and name-sake, Luduga, the son of Ur-Nigar.

Lu g a l - k u z u :

Lugal-kuzu was active from

∑

44 to

∑S

9, exclusively sealing documents relating to

the agricultural administration. He operated at the level above the captains of (plow-)oxen,

perhaps with the title provincial administrator. Lugal-kuzu sealed more than 30 documents

after AS 6, all relating to the district of Gu’dena and

MuÒbiana

, a district that was managed

by a number of agricultural administrators mentioned earlier in this study. It may be

appropriate to compare UTI 4, 2569 (from

∑S

2), MVN 11, 164 (from

∑S

4), AnOr 7, 313

(no date), UTI 4, 2864 (from

∑S

2), and UTI 3, 2126 (

∑S

5 iv), four texts concerned with

Gu’dena and

MuÒbiana

UTI

4, 2399

609

is a list of work-days expended weeding (

kiÒi

-a

) six

different units in the

MuÒbiana

and Gu’dena area. The fields, here termed GAN

610

were

each qualified according to a personal name; the majority of these names correspond to the

609. Compare to the parallel text UTI 4, 2569.

610. A brief investigation of the GAN

units worked by the individuals in this text shows that these were not

static; rather, each person worked on a number of different fields with-in the greater area of Gu’dena and

MuÒbiana

251

names in the other texts to be discussed here (see table 3, p. 253). The next text, MVN 11,

164, is a list of amounts of barley levied from the threshing floor of Gu’dena (

reverse lines 6

- 7; Òe geÒ e

-a

/ ki-su

-eden-na

)

. The personal names in MVN 11, 164, are almost

identical with the ones from the first text, the same group of people were, however, in this

text referred to as captains of (plow-)oxen (nu-banda

). MVN 11, 164, should be

compared to AnOr 7, 313 (no date), a much more detailed record with the same subscript as

MVN 11, 164. In AnOR 7, 313, the same persons were listed, but the amounts of grain

recorded for each person was approximately 20 times larger than the amounts recorded in

MVN 11, 164, which therefore might only be a partial record of the product of the fields. In

AnOr 7, 313 the farmers who worked on the different units were also listed. Some of the

people in this text were called (agricultural) administrators (

Òabra

). UTI 4, 2864, records the

disbursement of wool and hides for the cultivators of the same units—the same people

previously called “captains of (plow-)oxen” were here called “foremen” (ugula). It is likely that

the reason behind this change in titles reflected the different functions of the same persons.

In UTI 4, 2864, the captains of (plow-)oxen acted as foremen in relation to the cultivators

who were to receive the wool and hides. The titles given in AnOr 7, 313, might reflect the

hierarchical standing of the persons.

The reason for this change of title, which seems to be determined by the function of

the person, is even more confusing when studied chronologically, and inter-regionally. It

appears that the original title for the same official might have been scribe of 10 oxen (dub-sar

1(u)).

611

The last text to be mentioned here, UTI 3, 2126, is a list of the ox-carcasses,

252

hides, and barley. The animals, which were referred to as “fallen” (gu

ri-ri-ga) were

presumably the plow-oxen used on the same units as described above; the barley recorded

here is likely to be the barley that these creatures had not eaten since the time of their death.

The fact that the number of hides does not reflect the number of dead animals may indicate

that they died an untimely death, which could have left some hides useless.

612

In this text the

persons are once again referred to as captains of (plow-)oxen.

613

611. See, in particular, Syracuse 356 (from

∑

35 iv), a text which has the same format as UTI 3, 2126,

mentioned here, and according to which the foremen were referred to as follows: “foreman: Lugal-

kugani, scribe of 10 oxen” (obv. 9: ugula lugal-ku

-ga-ni dub-sar gu

1(u)); “foreman: Dada, scribe of 10

oxen” (rev. 2: ugula da-da dub-sar gu

1(u)); and “foreman: Lugina, scribe of ten oxen” (rev. 11: ugula

-gi-na dub-sar gu

1(u)). These three foremen were all attested with the title captain of (plow-)oxen

(nu-banda

), and it is entirely possible that the title scribe of 10 oxen went out of use during the

later years of

∑ulgi.

612. ri-ri-ga is always contrasted to ba-

uÒ

, which is understood as “slaughtered”.

613. Compare this text to MCS 1, 54, BM 106045, obv. iv 14 (24) to v 9 (24), where the same units are

listed. It is entirely possible that UTI 3, 2126, mentioned here was the actual receipt used when

composing the account MCS 1, 54, BM 106045, however, due to minor discrepancies this cannot be

proven at present.

253

Several of the persons in the table above may be related to one another. Egalesi a

member of the family of Lu-

∑ara,

the land-surveyor (sa

-sug

), sealed documents very

similar to those sealed by Lugal-kuzu concerning the fields Gu’dena and

MuÒbiana

614

Table 3: captains of (plow-)oxen under the command of Lugal-kuzu

UTI 4, 2399

(record of work-

days weeding

plots in Gu’dena

and

MuÒbiana

)

(

∑S

a. Compare to UTI 4, 2569, a parallel text where the sequence of persons is

Ur-Abzu, Lu-Utu, Ur-

Enun(a), Lu-dingira, GuTAR, Ur-Ninsu.

MVN 11, 164

(record of levied?

barley from

Gu’dena)

(

∑S

AnOr 7, 313

(Account of total

output of grain

from Gu’dena

and

MuÒbiana

)

(no date)

UTI 4, 2864

(disbursements

of wool and

hides for the

cultivators of

Gu’dena and

MuÒbiana

)

(

∑S

UTI 3, 2126

(record of fallen

oxen and their

fodder)

(

∑S

4 v)

Ur-Abzu

Ur-Ninsu
(nu-banda

Ipa’e

(ugula)

Ur-Enun(a)
(nu-banda

Ur-Enun(a)

Dada
(nu-banda

Ur-Ninsu
(ugula)

GuTAR
(nu-banda

Lu-dingira

GuTAR
(nu-banda

Luduga
(Òabra)

GuTAR
(ugula)

Ur-Ninsu
(nu-banda

GuTAR

Ur-Enuna
(nu-banda

Ur-Enuna
(Òabra)

Ur-Enun(a)
(ugula)

Ipa’e

nu-banda

Ur-Ninsu

Lu-dingira
(nu-banda

Ur-Enlila
(Òabra)

Dada

GuTAR
(ugula)

614. Compare to P. Steinkeller, JESHO 24 (1981)116 - 119.

254

It seems that Lugal-kuzu operated as a provincial administrator supervising a group

of approximately six captains of (plow-)oxen, perhaps making up an entire district. The

following example supports this interpretation:

CHEU 46 (from

∑

47 ix):

Obverse.

1. 8(geÒ

) Òe gur lugal

480 gur of barley, according to

the royal measure,

2. ma

-a si-ga

loaded in the boat,

3. nibru

-Òe

for Nippur.

4. guru

lugal-ka-ta

From the granary of

“the canal of the king,”

5. ki lugal-ku

-zu-ta

From Lugal-kuzu

6. Òe gu

-na Òabra-e-ne

(it is the) gun

615

barley of the

Òabra-administrators

Reverse.

1. kiÒib lugal-uÒur

(LAL

.TUG

)-ra

Sealed by Lugal-uÒur.

2. iti

-si

Month

“

Lisi

”

3. mu-us

-sa ki-maÒ

ba-

Year after:

“

KimaÒ was destroyed

”

(blank space)

From this text we can infer that Lugal-kuzu was supervising a large area, managed by

Òabra

administrators, who were later called captains of (plow-)oxen.

Most of the sealed documents of Lugal-kuzu contained the phrase “sealed with the

nam-

Òatam

of Lugal-kuzu,” in the body of the text. It is uncertain whether this expression

referred to a specific seal of the person, or to an act of sealing as someone else; however, the

615. The “gun” tax was paid by the agricultural overseers in both barley and livestock.

255

most likely option seems to be that

kiÒib

nam-

Òatam

implied that the sealing party sealed in

his capacity as a provincial administrator.

There is no consensus as to how to translate

Òatam

nor how to understand the office of the

Òatam

616

and it is still too premature to directly

connect the office of Lugal-kuzu (here tentatively designated as a provincial administrator)

with the title

Òatam

In i m -

•ara

Inim-

∑ara

is well attested in the Umma records, sealing close to one hundred tablets.

The majority of these tablets were sealed with a regular “dub-sar” seal, and a smaller number

were sealed with a seal with a simple two line inscription naming the holder of the seal and

the name of the father of the seal-holder.

617

At first there seems to be only a very minor

difference between texts sealed with either seal, and in fact several texts are almost identical,

although they are not duplicates.

618

Inim-

∑ara

’s regular seal was used from

∑

47 to

∑S

616. The corresponding Akkadian

Òatammu(m)

is a loan from Sumerian, AhW (p. 1199) translates,

“Verwalter; Verwaltungsdirektor.” Neither the office of the

Òa

-tam, nor its Akkadian equivalent, the

Òatammu

, has to my knowledge been discussed in the literature: M. Sigrist,

Drehem

(1992) 66, 68, and

122 does not translate the term, nor does he give any suggestions to its functions. W. Sallaberger,

Ur III-

Zeit

(1999) 266, translate “Verwalter” without further comments.

617. See, for example, ASJ 19, 214 39 (from

∑

48 xi), with the seal-inscription: inim-

Òara

dub-sar /

dumu ur-nigar

(NIGIN

)

gar

And, see, for example, MVN 16, 866 (from

∑S

1), with the seal-

inscription: inim-

Òara

dumu ur-nigar

(NIGIN

)

gar

256

This seal was rolled on several tablets said to be sealed by Gududu, whereas his simple seal

(used from AS 7

619

(5?)

620

∑S

1) was rolled exclusively on tablets said to be sealed by

Inim-

∑ara

himself. Eight texts were rolled with the seal of Inim-

∑ara,

the son of Dadaga (all

dated to

∑S

4 xi and

∑S

6 xii)—all were said to have been sealed by Gududu.

621

The

frequency with which Inim-

∑ara

sealed for Gududu increased over time and peaked around

AS 8. The first occurrence of the second, simpler seal belonging to Inim-

∑ara

coincides with

this development.

The activities recorded in texts sealed with either seal are, as already stated, almost

identical. The overwhelming majority of the texts record agricultural activities at a particular

set of fields.

622

Six texts, all sealed with Inim-

∑ara

’s regular seal but said to have been sealed

by Gududu mention the Apisal granary, and the work of bringing barley to the granary from

the threshing floors of the province, etc. The rest of the 16 texts sealed by Inim-

∑ara

rather

than Gududu are also concerned with granaries or barley for the bala contribution of the

province.

618. UTI 3, 2099 (from AS 7) // Princeton 1, 516 (from AS 7); UTI 3, 1785 (from AS 7 v) // UTI 4, 2912

(from AS 7).

619. UTI. 4, 2912 (from AS 7).

620. UTI 3, 2094 (from AS 5).

621. BCT 2, 39 (from

∑S

5), seal: inim-

Òara

/ dub-sar / dumu da-da-ga. See also MVN 16, 1532 (from

∑S

5); NYPL 364 (from

∑S

6 1 to xii); OrSP 47-49, 429 (from

∑S

5); SAT 3 ,1525 (from

∑S

4 xi); SAT 3,

1718 (from

∑S

6 xi to xii); SNAT 501 (from

∑S

5); and UTI 4, 2927 (from

∑S

5).

622. For example the a-u

-da field, the nin

-nu-du

field, and the na-ga-ab-tum field.

257

There is compelling circumstantial evidence that Inim-

∑ara

sealed, using a distinct

seal, instead of Gududu only insofar as certain particular administrative functions were

concerned, primarily restricted to the provincial district of Apisal. In fact Gududu rarely

sealed documents relating to field-work; rather, the seal of Inim-

∑ara

was rolled on the

tablets said to have the seal of Gududu. Gududu never sealed any documents during the

reign of Amar-Suen, although he was active at that time. Gududu was active during the same

years as an administrator (perhaps based in Umma), receiving animals and silver (from field

interest), among other things.

All of the foremen and captains of (plow-)oxen appearing in the texts sealed by Inim-

∑ara

were known members of the same group of agricultural administrators, and there is

therefore good reason to assume that also Inim-

∑ara

, as well as perhaps Gududu, held the

position of a provincial administrator, with Gududu having his uncle (and cousin?) seal the

documents relating to his office,.

Lu -

•ulgi

( r a ) :

Lu-

∑ulgi

(ra) has been mentioned several times above; two members of the ruling

family of Umma were named Lu-

∑ulgi

(ra), the son of Dadaga, and Lu-

∑ulgi

(ra), the son of

Ur-Nigar. Both were active primarily outside of the state-administration, retaining, however,

certain privileges or obligations for which they deposited a seal with a close relative, or did in

fact seal in person, either as an aid to this relative or fulfilling an obligation.

258

At u , Lu g a l - e ze m a n d Gu d e a

Two persons were named son of Ur-Nigar, the chief cattle administrator, in only one

text each: Atu and Lugal-ezem. The seal of Atu, the son of Ur-Nigar, the chief cattle

administrator,

623

is recorded only on Aleppo 176 (from AS 3).

624

The seal of Lugal-ezem,

the son of Ur-Nigar, the chief cattle administrator, is recorded only on Aleppo 371 (from

∑

37 viii). The document is a simple “

Òu-ti-a

receipt” relating to the delivery of dead animals

(of the type described above, pp. 224 to 226). Ayakala is named as the recipient in the body

of the text; the herder delivering the animals, called Ur-

IÒtaran

, is also known from other

similar texts. It is possible that Lugal-ezem was another son of Ur-Nigar, but it is equally

possible that we are dealing with a simple copying error.

Gudea, the son of Ur-Nigar, was briefly mentioned in footnote 246 of this study;

there is no indication that Gudea ever participated in the state adminstration.

625

A number of other persons attested in the Umma material appear as descendants of a

person called Ur-Nigar. They are listed here, although their affiliation with the ruling family

623. As already noted (fn. 384 on p. 152), yet another text has the seal of Atu, the son of Nigar-Kidu, the

chief of the galla.

624. The text is a simple receipt relating to the administration of work.

625. Princeton 1, 414 (from AS 3); SAT 2, 665 (from AS 1); and YOS 4, 84 (from

∑

48 v), without the title

chief cattle adminstrator.

259

cannot be proven, since they do not record the title of their father or any other indication of

their familial relationship.

Lu g a l -

e g a l :

Nineteen of the texts that have the subscription “sealed by Lugal-

egal” were receipts

of work, mostly work done by male workers (both

guruÒ

and ug

-ga

)—all of the recorded

work was agricultural field-work. One text (MVN 16, 1500 [from

∑S

1 vi]) was a receipt of

three reed-mats from Lu-Ebgal. Three texts mention the granary of Apisal,

626

two texts

mention a canal outlet (kun-zida) at Apisal,

627

and three texts mention the field APIN-bazi,

located in the Apisal district.

628

There appears to be some substance to the claim that Lugal-

egal was indeed associated with one area in particular: the granary of Apisal, and the field

APIN-bazi. Therefore, it may be argued that Lugal-

egal was a member of the ruling family

of Umma who was not active in the state-run administration, but rather in the private sector

of the economy, while retaining certain ties to the state; in this regard, he had a career

analogous to Inim-

∑

ara and several other well-known members of the ruling family.

626. MVN 14, 65 (from AS 1 ix); MVN 16, 1427 (from AS 7 iii); and UTI 3, 1777 (from AS 7).

627. UTI 3, 1669 (from

∑S

4), obv. 2; kun-zi-da a-pi

-sal

; MVN 16, 775 (from

∑S

4), obv. 2; kun-zi-da a-

-sal

628. MVN 21, 152 (from

∑S

6); SAT 3, 1696 (from

∑S

6); UTI 4, 2509 (from

∑S

7 i). This field was

located in the Apisal district, see CST 539 (from IS 3).

260

Lu - d i n g i r a , Ma n s u m , Lu -

e g a l a n d Ur - Ni s a b a

Lu-dingira’s seal (lu

-dingir-ra /

dub-sar / dumu ur-nigar

(NIGIN

)

gar

)

appears

on very few texts; Lu-dingira is always mentioned as the sealing party in these texts.

The texts sealed with Lu-dingira’s seal deal mainly with labor, a few with household

items, and one 6-column text concerns a multitude of commodities, which Lu-dingira sealed

instead of Ur-Gipar, the

iÒib

-ma

priest (AAICAB 1, 1911-240 [from

∑S

5]). All the texts

sealed by Lu-dingira fall between AS 9 and SS 7. Lu-dingira cannot be included, for certain,

in the family of Ur-Nigar; he remains almost entirely outside the state-administrative records.

Mansum, the son of Ur-Nigar, is known from his seal-inscription found on one,

possibly two, tablets.

629

Mansum’s seal had a dedication to Ur-Lisi, and we can speculate that

he was active very early, before the administrative expansion following the construction of

Drehem.

The seal of Lu-

egal, son of Ur-Nigar is attested only once, on a tablet published by

M. Touzalin (Aleppo 257 [from AS 1 v]).

630

Ur-Nisaba’s seal (ur-

nisaba / dub-sar / dumu ur-nigar

(NIGIN

)

gar

) was rolled on

very few tablets. The majority of these documents recorded the production of simple

household items.

631

Ur-Nisaba was active between

∑

38 and

∑

44. There seems to be no

629.

Nik 2, 188 (from

∑

35 vii), and BIN 5, 183 (no date).

630. This text should be collated; lu

is perhaps a mistake for lugal?

263

(described in Excursus 1), to quote only one modern example, also lacks a codified plan for

succession.

Two offices held by members of the ruling family of Umma were handed down

laterally, but two other offices were inherited vertically within that very same family,

suggesting that the system of succession was more complex than a simple opposition between

fratrilineal or non-fratrilineal might suggest. Although it is impossible to determine the exact

sequence of the sons of, for example, Ur-Nigar—head of the Umma ruling family—and

therefore impossible to determine whether Ur-Lisi was his oldest son, it remains certain that

primogeniture did not determine the line of succession, since we know that both Ur-Lisi and

Ayakala, his successor, had sons who were by-passed in succession. However, it may be

correct to describe the system of succession as patrilineal, since, as we shall see, inclusion in

succession was based on patrilineal decent only.

In our analysis of the ruling family of Ur III Umma [Chapter 5] we began by

investigating the earliest generations [Sections 2 and 3]. This investigation was in some

regards unproductive, since it is impossible, at present, to connect the earliest rulers of Ur III

Umma with the clan of Ur-Nigar, although this family was clearly involved in the

management of the province from the time of the earliest documentation. The next sections

[4 to 6] dealt with the three well-known governors of Umma, Ur-Lisi, Ayakala, and Dadaga,

who were shown to be brothers. Succession to the highest of offices in Umma thus remained

within one generation of brothers, and succession can be described as fratrilineal. This was

exactly the basis for the question posed in the beginning of the next section [7]. Why did the

264

brother of the governor succeed him and not his own son? Only Gududu, the son of Dadaga,

seems to have been aligning himself in the line of succession; his cousin Lukala, however,

may have done the same. The many sons of the three Umma governor’s Ur-Lisi, Ayakala, and

Dadaga, who have left little trace in the public records of the period must all have been by-

passed, either willingly or not, in the line of succession so as to allow for the fratrilineal

succession; whether this was out of concern for tradition or due to a desire to follow a quasi-

imperial ideology will be discussed below. The following sections [8 to 10] concerning Ur-E’e

and his sons were aimed at connecting this group with the ruling family. This was

successfully carried out based on the existing evidence associating Ur-E’e and his sons with

the ruling family, as well as on the basis of the strong circumstantial evidence showing that

the careers of Ur-E’e and his sons were closely connected to the ruling family. Section [11]

which dealt with Ir(mu) and his son shows that the patterns of office succession in these two

family-lines were identical; that both Lu-Haya, Ur-E’e’s son, and Ir(mu)’s son

∑

ara-izu

inherited their father’s office without using his title. Based on this analogy it is strongly

suggested that the rules of office inheritance were more complex than hitherto believed. In

the last section [12] the remaining sons of Ur-Nigar were surveyed, and their careers

described. The fact that only very few of the many remaining sons of Ur-Nigar ever pursued

a career within the state-administration is indicative of the fact that succession was limited to

a restricted sequence of candidates.

The result is a system not unlike that of Saudi Arabia, where the succession pattern

within the royal family is largely fratrilineal, but not systematized, and where, generally

speaking, seniority seems to be the dominant pattern. A member of that family described

265

succession as a system in which an heir is chosen among the most fit senior princes by means

of a set of unspecified “coincidences”.

632

Even if a law of succession is defined as a common set of rules adhered to by a

majority of the population, we are still forced to admit that we do not know of any such law.

We note, in passing, the results of Steinkeller’s investigation of the social structure of the

family-groupings making up the forester work-crews—similar to our results concerning the

ruling elite—and suggest that even the much larger groupings working the domain lands of

Umma were arranged in a similar way.

633

Figure 10 below, demonstrates the sequence of persons holding the four most

important offices in Umma (it should be compared to the preliminary survey of the ruling

family of Nippur presented elsewhere in this study). Here it is worthwhile to note, in

particular, three different developments. First, at the time of the take-over by the clan of Ur-

Nigar in

∑ulgi

33 the family went from controlling only the offices of the chief of the

632. See p. 280 fn. 658.

633. See in particular P. Steinkeller, AOS 68 (1987) 80 - 81, and figure 2 - 5. The family of Dada was

mentioned above (fn. 480, p. 191). SAT 2, 77 (from

∑

33 vi), which is a summary of the work-force

associated with the ca. 10 domain units controlled by Dada (who in this text is given the title scribe of

10 oxen (dub-sar gu

1(u)), clearly advocates that this group (consisting of 9 administrators and 40

permanent workers + ca. 20 additional workers) was made up of an extended family. Four generations

are mentioned in this text. Note for example the following: Lukala the cultivator (obv. ii 9) was the son

of Lugal-ezem (obv. i 2), who is here mentioned as the assistant to his father Dada (obv. i 1). Moreover,

Lukala’s own son, Ur-Ninpirig, is mentioned right after Lukala as a member of his work-team (perhaps

as an ox-driver,

Òa

-gu

266

granary and that of the chief cattle administrator, to having full control over the city of

Umma.

634

Second, after the take-over, it appears that succession within those two offices

became entirely lineal, passing from father to son. It is unclear whether the power of the two

offices also diminished after the take-over—Lukala’s high-ranking position within the local

hierarchy seems to counter this suspicion. Likewise, it is probable, but impossible to prove,

that the office of the chief cattle administrator and that of the chief of the granary were in

some way closely related to the imperial court, therefore we might see the events of

∑

33 as a

take-over by a family of imperial administrators. Lastly, it is important to note that four sons

of Ur-Nigar, simultaneously, took over the four most important offices in Umma in that

same year,

∑

33.

The historical sources do not relate any events dating to

∑

33 which could be seen as

an external reason for a hypothesized “take-over” by a clan of imperial administrators.

Drehem, the administrative center of the empire was supposedly not founded until

∑

coinciding with the wide-spread, if not universal change in the dedications of seal-

inscriptions away from the local governor and unto the king.

∑

ulgi was deified already

around his 21

year, almost contemporary with the so-called reforms of

∑

ulgi. In

∑

34 we

learn about the sack of

AnÒan

, a city in Iran where

∑

ulgi had sent his daughter (unnamed)

only four years earlier to become the “consort” of the ruler of that political entity (the ruler

634. Of course, we do not known if the governor prior to Ur-Lisi (perhaps Abbamu) was a member of the

same family that we have called the ruling family of Umma. The fact, however, that there exists no

reference to a familial relationship of the clan of Ur-Nigar with the governor of Umma prior to

∑

33 is

indicative of the fact that that office was not yet in their hands.

267

of An

an was subsequently bequeathed with the title ensi

, “governor’ by

∑

ulgi). This

campaign is hardly the reason for the “take-over” by the clan of Ur-Nigar, but it may be

possible to reconstruct the period as one of contraction and consolidation.

635

As such it may

be possible to understand the “take-over” in the light of long-term trends in the political

history of the Ur III period. It is equally difficult to find any internal reason for the take-over;

although absence of the name of the governor of Umma prior to

∑

33 is perhaps indicative of

a very weak governorship prior to that date. Although it may be possible to speculate about

several reasons for the situation in

∑

33, it is better to await the publication of further

records, and perhaps even renewed excavations at the city of Umma, before any further

suggestions are made.

Figure 10: The succession of office in Ur III Umma.

635. See M. Stolper, “On the Dynasty of

∑imaÒki

and the Early Sukkalma’s,” ZA 72 (1982) 52 - 53.

GIRINI

Ur-Nigar

Ur-E'e

Ir(mu)

Gududu

Lukala

Lu-Haya

Ur-Lisi

Ayakala

Dadaga

∑ara-izu

Chief household administrator

of the governor
(*Òabra e

ensi

Governor

(ensi

)

Chief of the granary

(KA-guru

)

Chief cattle administrator

(ÒuÒ

)

∑ulgi 33

268

The fact that the son of the chief cattle administrator as well as the son of the chief of

the granary rarely mentioned themselves with the titles of their predecessors has been noted

above

In particular on pp. 196 - 197 and p. 238, where it was suggested that the absence of

the title of Ur-Nigar in the seal of his son Ur-E’e was indeed not an accident, but rather

inspired by a social practice “forbidding” the mention of the title of one’s father when

inheriting his office. Whether this has any importance for our understanding of the system of

succession is not certain, but it could indicate that fratrilineal succession was more accepted

within the ruling elite than primogeniture; this remains, however, largely speculative.

This, of course, leads us to the second question of whether the provincial ruling

family mimicked the royal court. Since we initially investigated the Umma family as a

comparative case in order to help us understand succession within the royal family, it appears

unreasonable to ask whether the Umma ruling family mimicked the royal family. However,

the fact that the patterns of succession observed within the Umma ruling family seem to be

restricted to that family and perhaps only to a limited extent influenced the lower strata of

society means that it may still be possible to suggest that the ruling family of Umma

mimicked the clan of Ur-Nammu when emulating their practice of lateral succession.

In other areas as well it seems likely that the Umma elite tried to mimic the royal

court, and we see that concubinage became (or was) an institution in Umma as well as at the

court. The governor of Umma was, however, completely subordinate to the king of Ur.

636

Unfortunately, very few documents provide evidence for direct correspondance between the

269

royal and the provincial court, although two letters testify to its existence (see pp. 159 - 162

in this study). The accounts concerning the governor, as well as the military presence in the

Umma province clearly show the degree of royal control.

It is, of course, peculiar that the fratrilineal kinship terms (

ÒeÒ / nin

) were used very

infrequently in the Ur III administrative records since we have shown that fraternal lineage

was indeed very important for succession.

637

The distribution of this term, as shown in

Excursus 2, suggests that it was primarily used by the brothers of women connected through

marriage or marriage-like liaisons with members of the ruling elite. This, of course, has a

bearing on our understanding of neo-Sumerian system of succession since it is clear that

paternal descent was considered the most important factor for success in society, whereas

fraternal affiliation would have been useful only to people for whom this proved the only way

to affiliate themselves with the ruling elite. Babati may be the best example of this. He had

no other way of associating himself with the royal family than by claiming to be the in-law of

the king, and later the maternal-uncle of the new king. The sukkalmah, on the other hand,

may have belonged to an important cadet branch co-opted into the ruling family through

marriage with a royal daughter. His familial relations to the royal family are rarely mentioned

in the extant documents.

636. This relationship is seen clearly in the documents describing the yearly visits of the “queen-dowager”

Abπ-simtπ

to Zabala. It is documented that the household of the governor of Umma supported her

visits in every way.

270

It is unknown at present whether the system described here applies to all of

Babylonia during the period in question, or whether regional differences existed. It may be

opportune at a later date to extend this investigation in both time and space in order to

637. According to Yu. Yusifov, “The Problem of the Order of Succession in Elam again,” a paper published in

Wirtschaft und Gesellschaft im alten Vorderasien

(Budapest 1976) 321 - 331, a fratrilineal order of

succession is not present when the successor does not base his claim to the throne on a fratrilineal

affiliation with the previous ruler (p. 327). The case of Elam, studied by Yu. Yusifov, is of great interest

for our present study, although it cannot be fully discussed here due to the restrictions of this limited

study.

Yusifov claims that the order of succession in Elam was not fratrilineal, as previously thought, but rather,

“that in the history of Elam there existed more than one order of succession to the throne. At first there

predominated the order of succession to the throne through the female side which subsequently gave

place to the patrilineal principle of succession.” (p. 331) The cases surveyed by Yu. Yusifov of brother

succeeding brother all seem to have related to extreme situations and nowhere but once did the new king

use the “title” brother of the king. However, it is my interpretation of the evidence put forward by

Yusifov that the system of succession in Elam might have changed from matrilineal to patrilineal, but

that primogeniture was hardly the rule. Yusifov does points out that the title “beloved son” (

Òak hanek

)

perhaps functioned as a title comparable to “heir apparent”, whereby, as he writes, “this fact in its turn

excludes succession to the throne in the order of seniority” (p. 328).

The only example (according to Yu. Yusifov) for the use of the title “beloved brother” used by a successor

to the throne, about his brother, a preceding ruler, was

∑ilhak-InÒuÒinak

the brother of Kutir-

ÌÌ

unte. The titles used by

∑ilhak-InÒuÒinak

were i-ke

a-ne-ek kuti-ir-

nap

u-un-te, “beloved

brother (of ) Kutir-Na

ÌÌ

unte, and i-ke

a-ne-ek u

-ri-me, “my beloved brother” used about Kutir-

ÌÌ

unte.

∑ilhak-InÒuÒinak

was in time followed by Kutir-Na

ÌÌ

unte’s son

ÎuteleduÒ-InÒuÒinak

who claimed to be the son of both his father and his uncle. Yusifov explains the usage of this title, as well

∑ilhak-InÒuÒinak

’

use of a matrilineal claim to succession as the claims of an usurper (p. 325), and

concludes that, “in that period the succession from the father to the son was a firmly established rule” (p.

325).

Although the overall structure of Elamite succession seems clarified by the article of Yu. Yusifov, much

still remains to be done (compare F. W. König, “Geschwisterehe in Elam” RLA 3 (1964) 224 - 231),

particularly with regard to the theoretical issues raised by the definitions of Yusifov. They seem to

question the nature of seniority rather than the question of fratrilineal succession.

271

uncover whether inheritance patterns were in any way connected to ethnicity. No such study

is yet available, nor is the primary data necessary for such an investigation similar to the one

conducted here available.

Based on what we know about other powerful patriarchs, such as

∑ulgi

—or Abd al-

Aziz, a more contemporary model—it seems very plausible that Ur-Nigar, the head of the

Umma clan, fathered multiple male heirs. Furthermore, it is very likely that only a handful of

these sons embarked on a career in the state-administration, whereas the rest would have

been able to profit from their connection with the powerful family of Ur-Nigar and venture

into the private sphere of the economy, leaving no traces behind them in the official record.

It is only affirming to see these persons on an occasional basis fulfilling certain obligations

within the state-administration. Another reason for the irregular and rather peculiar, and

seemingly infrequent, appearance of the sealings of several high-ranking members of society

is presumably connected to the familial structures of the clan of Ur-Nigar. Some members

perhaps fulfilled obligations for others. Therefore, it is both possible and likely that people

would deposit their seal with other administrators or perhaps households for extended

periods of time.

The relation between the members of the ruling family and the agricultural lands

have been mentioned several times in this study. MVN 21, 343 (from

∑S

3), a text recording

the yield and field interest of various plots in different fields, gives interesting clues as to the

administration of land. It can be suggested, on basis of this text, that individuals had certain

272

rights (or obligations depending on the point of view) to tracts of land in different fields,

sometimes corresponding to the same plots of land mentioned in their sealed documents

concerning labor.

638

Therefore, it is safe to say, in answer to the third question, that we have modified the

understanding of public and private Mesopotamia, since we have demonstrated (with

numerous examples from the primary sources) the likelihood that some members of the most

prominent family of Ur III Umma were not present in the public domain pursuing a career,

but that they remained liable for some service to the state, and therefore were present in the

records. And we have also suggested that the relations between high-ranking members of

society and the estate-land of Umma be re-condsidered.

639

This seems to be true regardless

of the fact that no private economy has ever been attested in the documents from the Ur III

period. It is not the suggestion here that a high-risk, elite economy, independent of the state,

existed; rather, private is here used only in opposition to official, and may refer to the lack of

any career as well.

640

638. See in this regard the field Abagal-Enlila and EnKAS mentioned above (see p. 198 - 199, above).

639. See also P. Steinkeller, JESHO 24 (1981)118 - 119. Steinkeller associates some of the same persons

which I have identified with the title district administrator, with individual temple households. However,

aside from the fact that Steinkeller describes the Umma domain land as essentially tenant-land, his

outline of the administrative command structures are not far from the one suggested in this study.

640. Gelb, as is well known, argued for the existence of a private economy, coexisting side by side with the

state sponsored economy. This cannot at present be proven or disproven. See I. Gelb, "On the Alleged

Temple and State Economies in Ancient Mesopotamia," Studi in Onore di Edoardo Volterra, Vol. 6

(1971) 138-154.

274

Exc u r s u s 1 : T h e Ho u s e o f Sa u d

In 1926 after he had conquered most of the Arabian Peninsula and secured his own

rule, Abd al-Aziz ibn Abd al-Rahman (c. 1880 - 1953), head of the House of Saud (from

1902 to 1953),

641

assumed the title “king of the Hijaz and sultan of Najd and its

dependencies” (Malik al-Hijaz wa-sultan Najd wa-mulhaqatihah) at a ceremony in Mecca —

the most important city of Saudi-Arabia.

642

Although “ibn Saud,” as he was often referred to

by Westerners, had accumulated immense respect and an irrefutable reputation as a strong

and charismatic ruler, he had not secured the line of succession. After his death, sovereignty

over the House of Saud could easily pass to his brother Muhammad ibn Abd al-Rahman,

born almost at the same time as Abd al-Aziz.

643

Muhammad’s son Khalid ibn Muhammad,

almost of the same age as Saud, the oldest surviving son of Abd al-Aziz, was also a contestant

to the throne. Shortly after the proclamation of the kingdom of Saudi Arabia (1932), the

king Abd al-Aziz challenged his own generation and nominated Saud his heir.

641. The House of Saud is named after Saud ibn Muhammad, who was a local Shaykh in the central Arabian

province of Najd. Abd al-Aziz was the true head of the clan following his father’s 1900 abdication,

although his father kept the title imam (a primarily religious title) for himself, and left Abd al-Aziz with

the title hakim (arbitrator). Following Abd al-Aziz’s successful conquest of Riyadh in 1902, he was

granted the title amir by his father. Abd al-Rahman died in 1928.

642. Abd al-Aziz was not crowned king of (the unified) Saudi Arabia until 1932. Saud was announced “heir

apparent” in 1933 (not by public statement). Oil was discovered in Saudi Arabia for the first time in

1937; however, large oil-revenues were unknown before the Second World War.

643. Both Abd al-Allah ibn Abd al-Rahman and Muhammad ibn Abd al-Rahman presented a challenge to the

lineal succession when this was first introduced by Abd al-Aziz ibn abd al-Rahman in the 30’s. The

importance of the decendants of the brothers of Abd al-Aziz and the other cadet branches of the House

of Saud is now dying out (J. Kechechian,

Succession in Saudi Arabi (

New York 2001) 32).

275

The House of Saud had risen twice before to rule the Arabian Peninsula. The first

great Saud leader, Muhammad ibn Saud (1742 - 1765), was able to secure succession for his

own son, and lineal succession lasted for two generations when the House of Saud again

collapsed due to internal rivalries. The second rise of the Al Saud, under Turki ibn Abd al-

Allah (1824 - 1834), was even less successful in establishing lineal succession. Turki’s son was

deposed by the Ottoman Sultan and his Egyptian allies, and imprisoned in Cairo. He was

later to return to Riyadh and regain power over central Arabia, but he was unable to secure

succession. Lateral succession was as common as lineal succession in the House of Saud, as it

was among their immediate competitors on the Arabian Peninsula, the Al Rashid.

644

Abd al-Aziz, being the strong charismatic ruler that he was, fathered at least 35 sons

with numerous wives and concubines. He had, however, never more than four legally

married wives at the same time, following the example of the prophet.

645

Abd al-Aziz’s first

son was born in 1900, his last in 1947, resulting in a generation stretching an entire century.

The oldest sons of Abd al-Aziz—some of whom rode with the Ikwan,

646

uniting the Arabian

Peninsula—had strong investments in the creation of the kingdom of Saudi Arabia. The

House of Saud faced widespread rebellion when they were rethinking succession, and

644. Internal rivalry was one of the primary reasons for the final fall of the Al Rashidi (J. Kechechian (2001)

14-15).

645. Abd al-Aziz would take full advantage of the possibilities of divorce sanctioned by the Quran; however,

he would continue to support his divorced wives, and those of them who had born him children would

remain living at the court. The exact number of Abd al-Aziz’s sons remains unknown as does the number

of wives he married over time.

646. The Ikwan, or brotherhood, was Abd al-Aziz’s creation: a Beduin army fueled by a desire for plunder and

religious fanatism.

276

confronted the cadet-branches of the House of Saud with the new centralization of power

within one line of the family. Subsequently, Abd al-Aziz was forced to fight members of his

own family, who were to be known as the “araif ” (an Arabic word used for redeemed camels,

applied to this family faction since Abd al-Aziz had pardoned them after their first defeat).

Abd al-Aziz managed the problem caused by the rebels, first by co-opting these cadet

branches in his rule and finally by executing those who continued their rebellion.

647

The basis of the Saudi rulership over the larger part of the Arabian Peninsula military

power and the religious fanaticism of the Beduin soldiers assembled by Abd al-Aziz rather

than any legal claim. The Saudi theological claims of religious supremacy, and custodianship

of the holiest places in Islam, negotiated with their allies, the Al Shaykh (descendants of the

reformer Muhammad ibn Abd al-Wahhab), are easily contested from within Islamic circles.

These theological claims are not founded on any genealogical claims, as was the case when

the Hashemonites ruled the Hijaz, but exclusively on reformist, puritan arguments of the

Wahhabi branch of Sunni Islam. This schism remains the

Achilles heel

of the Al Saud who

have failed to eliminate all opposition to their religious claims.

The rulers of central Arabia prior to the successful conquest of Abd al-Aziz, the Al

Rashid, were only one of the tribes of the Arabian Peninsula. Abd al-Aziz incorporated many

of these tribes into his realm either by making them join his family (co-opting them in the

rule of Al Saud) or by vanquishing them. Other tribes formed smaller “nation states”

647. A. Bligh,

From Prince to King, Royal Succession in the House of Saud in the Twentieth Century

(New York

& London 1984) 17 - 18

277

bordering the Persian Gulf, some of which relied on primogeniture for the selection of

heirs.

648

The rulers of Saudi Arabia have tried several times to expand their territory at the

expense of these states.

After the premature death of his first son Turki, the head of the Al Saud, King Abd

al-Aziz, named his second son, Saud, “heir apparent” and his third son, Faysal, number two

in the line of succession.

649

Saud has been described as everything, ranging from the

flattering image of a true son of the desert, a replica of his father, to being denounced as a

dilettante without any vision.

650

Whatever the viewpoint, it remains a fact that Saud ibn

Abd al-Aziz had little training in foreign affairs and only limited official schooling.

651

His

648. The rulers of both Yemen and Oman have been chosen according to the system of primogeniture,

whereas the ruling families of Kuwait, Abu Dhabi, and Umm al-Quwayn all seem to have relied on

seniority (A. Bligh, 1984, 9-10).

649. No public statement was issued in 1933 announcing the appointment of Saud as “heir apparent”. The

first such statement on the order of succession was issued in 1992, during King Fahd’s rule, when it was

then explicitely stated that Abd al-Allah is the “heir apparent”, and implicitly that Sultan is number two

to the throne. This is seen as a bargain between the powerful faction of princes known as the Sudayri

Seven, after their maternal descent, and the other senior princes. The Sudayri Seven are: the present king

(Fahd), Sultan, the minister of defence and civil aviation (number two to the throne), Nayif, the minister

of the interior, Ahmad, the deputy minister of the interior, Salman, the governor of Riyadh, Turki,

Sultan’s deputy defence minister, and Abd al-Rahman, a successful business leader (see J. Kechechian

(2001) 5ff.). The statement of 1992 is not a law of succession. However, even if it was considered a rule

of succession, it is limited by physical conditions such as the longevity of the first generation following

Abd al-Aziz. It follows that the question of succession will have to be addressed more seriously in the

near future.

650. See for example D. Holden and R. Johns,

The House of Saud

(London 1981) 176ff., W. Powell,

Saudi

Arabia and its Royal Family

(Secaucus 1982) 222ff.

651. A. Bligh, 1984, 58-59.

278

traditional education, unlike the more westernized education of his brother Faysal, centered

around virtues of a Beduin shaykh more than those of a ruler of a rich oil-state. Saud

therefore seemed utterly unprepared to take over at the time of his father’s death in 1953. His

reign became one of the most troubled in modern Saudi history. He was

de-facto

deposed

after six years (in 1958), and was finally forced to sign a letter of abdication in 1964 (actually

Faysal ibn Abd al-Aziz signed in his place, but the take-over remained peaceful).

652

Faysal,

supported by a large number of senior Saudi princes, in turn nominated his younger brother

Khalid as “heir apparent”. A cousin, Faysal ibn Musaid ibn Abd al-Aziz, assassinated Faysal

after only nine years as king of Saudi Arabia.

When Faysal replaced Saud he had to deal with the many grown sons of Saud, some

of whom saw a potential spot in the line of succession slip out of their hands with the

dethronement of their father Saud.

653

To this end Faysal could rely on the support of not

only his own sons, but also his younger brothers and to some extent their sons.

654

Faysal

managed to built considerable support for his rule, and it was he who finally secured lateral

succession.

655

Khalid took over, following the death of Faysal, and nominated prince Fahd as

652. The struggle between the two brothers was seen by J. Kechechian (2001) 40ff. as amounting to a family

feud lasting most of Saud’s rign, in particularly picking up momentum after Sauds involuntary

funnelling of executive power to Faysal in 1958. Saud regained power in 1960 and threw the country

into one of the worst conflicts in its history.

653. J. Kechechian (2001) 41, 42 and again 43, suggested that Saud had tried to secure succession for his own

progeny. See also A. Bligh, 1984, 58.

654. The alliances between the more than 300 sons and grandsons of Abd al-Aziz are, of course, much more

complex than this simplistic outline suggests.

655. J. Kechechian (2001) 47.

279

crown prince and Abd al-Allah as number two to the throne. Today king Fahd is the aging

ruler of Saudi Arabia, promoting more than ever before speculations over the future of the

Saudi monarchy.

656

This particular system of succession, plotted by the old king Abd al-Aziz to prevent

fraticide, has so far been successful, discounting of course the deposing of king Saud and the

seemingly unconnected assassination of king Faysal.

657

As the charismatic leader of the House of Saud for almost 50 years, and a successful

leader in war and politics, Abd al-Aziz managed to concentrate power within his branch of

the family. From the middle of his reign he began securing the succession of power for his

own line only, excluding the line of his brothers and their families. When Abd al-Aziz

nominated Saud and Faysal as number one and two to the throne, he did not change any

656. Decendents of Abd al-Aziz on the Saudi Arabian throne are, Saud (1902 - 1969, ruled 1953 - 1964

(deposed)), Faysal (1906 - 1975, ruled 1964 - 1975 (assassinated by nephew)), Khalid (1912 - 1982,

ruled 1975 - 1982), and Fahd (born 1921, ruler since 1982). Abd al-Allah (born 1923) is the designated

heir, and Sultan (born 1924) is considered number two to the throne. Fahd was the real ruler during

most of his brother Khalid’s reign, Khalid himself had apparently little interest in the duties of the

throne; his reign has been seen as part of a deal between two powerful factions within the royal family.

The succession paradigm of the Saudi royal family is by Western observers often described as a nest for

future troubles. With graduadly aging heirs waiting for their bid at the throne, this could result in a long

list of elderly rulers in office for only a few years prior to their death, unless a generational change takes

place.

657. Faysal’s assassin seems not to have been motivated by a quest for the throne, it was perhaps an act of

revenge by a young man following the excecution of his brother and removal of his father from the line

of succession.

280

traditional Arabic rules of succession, for no such rules exists. Rather he did what any strong

Arab leader will try to do, to secure his own line over that of his brothers.

658

Primogeniture exists along with seniority as a system of succession among the ruling

families of the post-Hijrat Arab world, but seniority has been traditionally, and by far, the

most common of the two.

659

Looking to the Qur’an and the examples of the prophet

Muhammad (

hadith

), even within the first four generations after Muhammad, Arab leaders

are without any set of rules guiding succession.

660

During the early generations of the House

of Saud (ca. 1700 to ca. 1900) all succession was presumably determined through the

principle of seniority, and it was not before Abd al-Aziz that any ruler was able to favor his

own line as decisively as he did.

661

So far Abd al-Aziz has been followed on the throne by

four sons, and several well-educated aspirants to the throne follow the “heir apparent”,

Sultan, in seniority, but in a system such as that found in Saudi Arabia, where the ruler

propagates heirs over a period of fifty years, it is likely that succession will pass some younger

658. “For a variety of reasons, chiefly because of religious and tribal traditions, primogeniture has not

developed among Arabian dynasties in quite the same way [as in Europe], because under Shariah law, all

sons of a man are equal and legitimate, even if they were born from illegitimate marriages. Moreover, in

pre-Islamic tribal norms, while the throne could have passed from one generation to the next within a

particular family, it was not necessarily passed from father to son. Rather, authority fell to a ruler’s

brother, uncle, or cousin, depending on which of these oldest male relatives was seen to possess “the

qualities of nobility; skill in the arbitration;

hazz

or ‘good fortune’; and leadership”[quoted from an

interview with prince Sultan bin Salman, quoted in J. Kechechian (2001) see fn. 23]”, p. 10 in J.

Kechechian, 2001. Compare this autobiographical statement from a member of the royal family with the

theoretical writings on the charismatic leader by M. Weber, see fn. 223 p. 90.

659. See A. Bligh, 1984, 9 - 10.

281

brothers by and eventually pass on to the next generation, the “grandson generation”, where

the most senior member suddenly is to be found.

662

Abd al-Aziz seems to have been a strong and charismatic ruler, to use the

terminology of M. Weber, whose decisions were not contested.

663

It is interesting, however,

660. See A. Bligh, 1984, 6, and J. Kechechian (2001) 11-12. Harun al-Rashid, the calif of Baghdad, revealed

his plans for succession inside the Ka’ba in 802. Al-Rashid had decided, late in his reign, to nominate his

son al-Amin as his first successor, and another son, al-Ma’mun, his second successor. Following the death

of al-Rashid in 809, al-Amin perhaps tried to form his own line of succession; however, soon after their

father’s death, al-Ma’mun succeeded in forming his own power-base in a distant province and finally in

812 to take Baghdad and have al-Amin excecuted. T. El-Hibri writes (p. 463) ; “Al-Amin’s execution, the

first regicide in the Abbasid house, shook the caliphate’s legitimacy ... In time it also gave rise to an

apologetic historiography that sought to legitimate al-Ma’mun’s overthrow of an incumbent caliph.”

According to the decree of 802, al-Amin was supposed to nominate the successor of al-Ma’mun, but in

805 al-Rashid nominated a third son, al-Mu’tamin, the successor of al-Ma’mun. The nomination of both

al-Ma’mun and al-Mu’tamin were accompaigned by land-allotments, or rather gubernatorial posts in

border regions. T. El-Hibri (p. 475) concluded that “the caliph’s purpose on the pilgrimage of 802 was

simple. Having nominated al-Amin (in 792) and al-Ma’mun (in 799) during their minority [al-Amin’s

nomination of a minor as heir following the death of al-Rashid was used by al-Ma’mun as the juricical

reason for the war], it was timely in 802 to confirm the succession with binding oaths on the princes in

their majority.” ( T. El-Hibri, “Harun al-Rashid and the Mecca Protocol of 802: A Plan for Division or

Succession?” International Journal of Middle East Studies, Volume 24, Issue 3 (1992), 461-480).

661. 19

century Ottoman tradition of succession, although far removed in space from the Arabian

Penninsula, favored linial succession, passing the office from father to son. Traditionally the Ottoman

ruling family did not enforce strict primogeniture; rather, the strongest and most cunning of the sultans

sons would take hold of the throne, eliminating all of his rivals. Fraticide became institutionalized, early

on, at the Ottoman court. Among the Arab rulers from the House of Saud there were those who had

firsthand knowledge of the Ottoman traditions from their “imprisonment” in Istanbul. See J.

Kechechian (2001) 12-13, and A. Bligh, 1984, 9.

662. A. Bligh, 1984, 53.

663. See A. Bligh, 1984, 103. We exclude, of course, the above-mentioned contestants to succession who,

ironically, in their defeat eventually contributed to the un-rivaled position of Abd al-Aziz.

282

to note that not even Abd al-Aziz’s plans for succession were adhered to for very long, and

that his heir Saud was deposed only nine years after his death. Internally, the House of Saud

is divided into competing factions and the balance of power is very fine, thus, the danger of

relapsing into a system of fraticide is never far away, and in every designation to a high office

a number of political factors are taken into consideration.

The prospective heirs to the Saudi throne are chosen—at the time being—exclusively

from among the sons of the late king Abd al-Aziz.

664

Maternal descent as well as the career

and education of the individual princes are of importance for the selection. On the other

hand, it is imperative for the brothers never to delegate too much power to one of their

brothers, a faction of full brothers, or to anyone from the next generation.

665

It might be

beneficial for our investigation of the system of succession within the elite families of late 3

millennium BC Mesopotamia, to look at the credentials that may help a prospective heir

within the House of Saud.

Succession within the House of Saud has bypassed several of the sons of Abd al-Aziz,

either with the consent of the neglected (perhaps as an act of support for a full brother

(brothers with the same mother) more fit for government) or for reasons such as mental

664. This was the topic of a paragraph in a royal decree, see J. Kechechian (2001) 210, quoting “Royal Decree

Number A/90, Dated 27 Shaaban 1412H/1 March 1992” Chapter 2, article 5b.

See also A. Bligh, 1984, 53-55, who attempted to group the heirs according to their prospects at

succession.

665. This can be seen for example in the Al Saud management of the oil-production and the government of

the oil-producing eastern provinces that are never concentrated in the hands of one heir alone. J.

Kechechian (2001) 87.

283

disabilities or weak health. Several of the sons of Abd al-Aziz have not taken part in any

aspects of the administration of the kingdom, a fact that would disqualify them in the

struggle for succession at once. Most of these sons have ventured into the sphere of private

enterprises, heavily subsidized by their affiliation to the ruling family, where they have

become very wealthy.

The use of dynastic marriages—the co-optation of rivaling families by the ruling

house—is widely employed in Saudi Arabia.

666

Abd al-Aziz would marry the daughters of

opponents, or even rebels, as well as the daughters of his allies, to secure the line of ibn

Saud.

667

Among the many children of Abd al-Aziz, the sons from the political or dynastic

marriages, as well as sons from marriages within the extended clan of Saud have been the

most successful. The children of his concubines have been successful too, albeit not

ascending to the highest offices. This is perhaps only by chance, since according to Shariah

law all the sons of a man are equal. It is debated weather the tribal background of the mother

is important for the success of her son. For example, the mother of the present “heir

apparent”, Abd al-Allah ibn Abd al-Aziz, was from the Shammar tribe, an influential tribe

from the central Najd, supplying many beduin-conscripts for the national security forces,

which also has Abd al-Allah as their chief.

668

In addition, it is likely that the closeness of the

mother to the king is important when later the son makes his bid for power.

669

The number

of full brothers is another very important factor determining the standing of a Saudi prince,

666. J. Kechechian (2001) 4.

667. R. Burling (1974) has pointed out some of the possible problems related to dynastic marriages, see p. 68.

284

since brothers with the same mother tend to support each other.

670

Members of the

grandson generation (grandsons of Abd al-Aziz) have also formed factions, mainly based on

father - son relations, and largely dependent on the success of their fathers.

671

Saud ibn Abd

al-Aziz had 53 sons many of whom ascended to high offices in the state when their father

became king. Exactly the fact that Saud had multiple successful sons is a possible reason for

his relative success, being as he was, an otherwise unlikely candidate to the throne of Saudia

Arabia. When Saud fell from power his sons fell with him; when Faysal ibn Abd al-Aziz rose

to power his sons too assumed high offices, however, two of them remained in power after

the assassination of their father, perhaps because of their long and sound bureaucratic

training.

672

Since succession in the House of Saud is based on seniority and restricted to the

descendants of Abd al-Aziz, rulership will, at some point, be transferred to the grandson

generation following a non-legalized scheme where both the individuality and familial

connections of the contestants are likely to play a role. The education and offices of the sons

668. However, precisely the fact that the familial structures of the Al Saud were so intricate perhaps weakens

the argument that the tribal maternal affiliation of an heir is of much importance. See fn. 674 p. 286.

The sister of Abd al-Allah’s mother was the wife of his own brother Saud.

669. A. Bligh, 1984, 40 - 41. This is an interesting point with regard to the Ur III royal family. As seen above

(pp. 104 - 105) the concubines of the Ur III ruler often employed a terminology invoking their closeness

to the king

670. J. Kechechian (2001) 26 - 28.

671. J. Kechechian (2001) 28-30.

672. For the careers of the sons of Faysal see J. Kechechian (2001) 29. Note that Saud al Faysal was co-opted

in the reign of Khalid, as a minister of foreign afairs (A. Bligh, 1984, 90).

285

and grandsons of Abd al-Aziz, which has been studied by J. Kechichian and others,

673

might

point to any one descendant of Abd al-Aziz as the next in line for the throne, but with the

large number of princes and the constant formation of factions the hierarchy of the House of

Saud seems fluid.

During any struggle for succession, the contestant must delegate power to supporters

in exchange for their support, endangering his future position. Is it possible for the new ruler

to maintain dynastic power? Will his dynasty slowly disintegrate? Centralized power is

perhaps at its peak in present day Saudi Arabia, with a minimum of cadet branches

competing for the throne, but when power passes to the next generation, it might result in

the loss of controll recoverable only through another strong man.

The many similarities between the ancient ruling family of the Ur III empire and the

contemporary royal family of Saudi Arabia has made this modern clan a valuable model for

studying patterns of succession during the time of Ur III. Many post-Islamic Arab ruling

houses could be cited as useful comparative material for our study, but the House of Saud is

more applicable for our case due to its unprecedented success and richness. Consider the

following complex family structure: “Abd Allah was born in 1921, when his father was about

fifteen years old. His mother, Faysal’s first wife, was Sultanah bint Ahmad al-Sudayri, whose

older sister Hassah was married to Abd al-Aziz. Hassah gave birth to the first of her seven

673. Data concerning the political history of the House of Saud, from the early 18

century till today is not

always readily availably, and many studies have been based, in part, on interviews with anonomys Saudi

personalities.

287

Exc u r s u s 2 : So n o f t h e k i n g v s . b r o t h e r o f t h e k i n g

In a society which favors lineal succession, it is expected that a contestant to the

throne would express his close familial ties to the ruler, and in a society with strictly enforced

primogeniture, the heir would hold a title such as crown-prince. But would a competitor call

himself brother of the (ruling) king in a society where fratrilineal succession exists together

with patrilineal succession? In the following brief excursus I will list and discuss all the

examples of the kinship terms “brother”and “sister” used by persons claiming a close

affiliation with the ruling clans of Sumer, as found in the extant texts from the Ur III period.

Only one person used the kinship term “brother” in relation to the king. This person

is otherwise unknown, and the affiliation is expressed in only two texts.

Likewise, only four persons specified their familial affiliation with the second most

important person of the empire, the sukkalma

, in terms of brotherhood.

Table 4: “brother of the king”

date

BM 103420 (unpubl.)

AS 8 ix

ur-

en-lil

-la

ÒeÒ lugal

MVN 8, 129

AS 4 iv 5

ur-

en-lil

-la

ÒeÒ lugal

Table 5: “brother of the sukkalma

”

date

JCS 10, 30 9

∑S 8 ii 05

ur-

nanna sukkal ÒeÒ ir

nanna sukkal-ma

CT 3, 21 (= BM 18957)

∑ 43 to ∑ 46

Òara

ÒeÒ sukkal-ma

TCTI 2, 3711

XX XX iv 21

u-ni ÒeÒ sukkal-ma

288

A few persons called themselves “brother of the governor” to specify their affiliation

with the head of an important family:

Babati, the brother of queen

Abπ-simtπ

, whose seal was copied in the Babylonian

school, was an important person in the empire (see fn 273, on p. 110 above). Strangely,

however, only one text includes the information, known from his seal inscription, that Babati

TCTI 2, 4161

XX XX x 21

Òu-i

-li

ÒeÒ sukkal-ma

Table 6: “brother of the governor”

date

BM 105554 (unpubl.)

∑ 34 ii

ÒeÒ ensi

umma

(Umma)

ASJ 19, 226 72

(no date)

ÒeÒ ur-

-si

(Umma)

JCS 28, 215 26

∑ 43

a-kal-la ÒeÒ ensi

(Umma)

OrSP 47-49, 500

(no date)

a-kal-la ÒeÒ ensi

(Umma)

SAT 2, 1078

AS 8

utu ÒeÒ ensi

adab

(Adab)

AUCT 3, 31

∑S 7 x

utu ÒeÒ ensi

adab

(Adab)

ITT 2, 4090

∑ 42

nin-Òubur ÒeÒ ensi

(Girsu)

LB 610 (unpubl.)

(no date)

ur-ba-gara

ÒeÒ ensi

(Girsu)

NATN 123

∑ 40

1 ur-x-x] ÒeÒ ¿ensi

• (Nippur)

Table 5: “brother of the sukkalma

”

date

289

was the brother of the queen. Aside from Babati, we know of only two persons who mention

their relationships to a named royal consort.

Some persons in texts from Girsu were referred to as “brother or sister of the nin-

dingir-priestess.” The nin-dingir-priestess is presumably identical with the wife of the

governor of Girsu, an office held by the sukkalma

after AS 6.

Table 7: “brother of the queen”

date

BCT 1, 126

AS 3 iii 19

ba-ba-ti ÒeÒ *¿nin•

CT 3, 35 (= BM 21335)

(no date)

-mu ÒeÒ nin

-kal-la,

OIP 115, 199

∑ 46 iii 19

i-di

-e

-a ÒeÒ e

-a-ni-Òa

RA 73, 191

XX XX ix to
XX XX i

i-di

-e

-a ÒeÒ e

-a-ni-Òa

Table 8: “brother / sister of the nin-dingir-priestess”

date

UNT 26

a. See also Amherst 27 (from

∑

37), where the same Ur-saga is called a foreman (ugula), and the

very fragmentary text MVN 2, 283 (no date).

∑ 47 xii

ur-sa

-ga ÒeÒ nin-dingir-ra

BAOM 2, 28 45

b. See also TUT 112 (no date).

∑ 48 xi

-la-la-a nin

nin-dingir-ra-ke4

MVN 00 BM 013925
(unpubl.)

c. See also BAOM 2, 24 14 (from

∑

39 vi).

∑ 43

-AB.DI ÒeÒ nin-dingir

290

Only one person used the kinship term “sister of the queen,” Bizua; she is only rarely

attested:

Only two texts mention anyone who claimed fraternal affiliation with a general:

Apart from the examples we have seen here, there few references to un-named wives,

brothers, and sisters of high-level officials of the empire (see for example, OIP 115, 74 [from

∑

43 ix 22], which mentions both the sister of

ΩelluÒ-Dagæn

(obverse line 7;

nin

-lu-

uÒ-

da-gan

) and the wife of

∑

arakam (obverse line 5;

dam

Òara

-kam

), among others).

The single most numerous group of people claiming affiliation with the royal clan,

through brotherhood, is a group of approximately 40 persons, each of whom claimed to be

the brother of a (royal) concubine. This group of people, who can only be associated with the

royal concubines through circumstantial evidence, are all mentioned in texts similar to the

Table 9: “sister of the queen”

date

ASJ 3, 74

AS 4 viii 16

-zu-a nin

nin

Fs. Jones 68

∑S 6 i 04 to
xii 30

-zu-a nin

nin

TCL 2, 5484

AS 5 viii 20

-zu-a nin

nin

Table 10: “brother of the general”

date

SNAT 333

AS 2 xi

ur-lugal ÒeÒ Òagina

RTC 331

∑ 35 vii

da-da-a ÒeÒ lu

utu Òagina

291

simple messenger texts. They all appear together with military envoys traveling to and from

the eastern provinces. The texts are all written records of the provisions for these persons.

Table 11: “brother of the concubine”

date

HLC 132 (pl. 99)

XX XX iii

a]-mur-

utu ÒeÒ lukur

OMRO 66, 55 20

XX XX v

a-bu-ni ÒeÒ lukur

MCS 5, 27 (= HSM 6367)

XX XX xii

u-a ÒeÒ lukur

MVN 9, 135

XX XX vii

a-na-ti ÒeÒ lukur

OMRO 66, 55 20

XX XX v

-bi

-li

-a ÒeÒ lukur

RA 59, 145 (= FM 56)

XX XX iii

ba-ba-a ÒeÒ lukur

RTC 349

XX XX viii

ba-lu

-a ÒeÒ lukur

MVN 7, 377

XX XX viii

da-gu ÒeÒ lukur

RTC 347

XX XX xi

e-la-<ak?>-nu-id ÒeÒ lukur

CT 10, 30 (= BM 014612)

∑ 43

-ni-bi ÒeÒ lukur

ABTR 13

XX XX vii

¿er

•-ra-LUGAL ÒeÒ lukur

ASJ 2, 9 15

XX XX iii

a-ti ÒeÒ lukur

HSS 4, 66

XX XX x

i-din-

utu ÒeÒ lukur

MVN 9, 136

XX XX viii

i-din-e

-a ÒeÒ lukur

TBM 1, 148

XX XX vii

i-ti-iÒ ÒeÒ lukur

TBM 1, 122

XX XX viii

iÒ-me-a ÒeÒ lukur

SAT 1, 157 (= BM 21022)

XX XX vi

ku-da-num

ÒeÒ lukur

TBM 1, 148

XX XX vii

ku-da-num

ÒeÒ lukur

TBM 1, 112XX XX i

la-bi-ru-um ki ÒeÒ lukur mu

a-gar

gibil-Òe

SET 212

(no date)

la-la-a <ÒeÒ> lukur

292

TUT 211

(no date)

la-la-a ÒeÒ lukur

TBM 1, 122

XX XX viii

la-ni ÒeÒ lukur

TBM 1, 138

XX XX iv

nanna ÒeÒ lukur

ABTR 13

XX XX vii

-gu-la ÒeÒ lukur

HLC 168 (pl. 104)

XX XX iii

nig

-e

-Òa-Òu ÒeÒ lukur

TBM 1, 145

XX XX i

nu-a-a-ti ÒeÒ lukur

A 824 (unpubl.)

(no date)

nu-ur

-i

-li

ÒeÒ lukur

TUT 211

(no date)

puzur

-¿eÒ•-dar x] ÒeÒ lukur]

SET 212

(no date)

puzur

-eÒ

-dar ÒeÒ lukur

TBM 1, 106

XX XX vii

puzur

-ga-ga-u

a-da-lal

ÒeÒ lukur

RA 19, 42 89

XX XX iv

puzur

-Òu ÒeÒ lukur]

HLC 106 (pl. 94)

XX XX iii

puzur

-Òu ÒeÒ lukur

TBM 1, 148

XX XX vii

Òu-

utu ÒeÒ lukur

MVN 9, 136

XX XX viii

Òu-e

-a ÒeÒ lukur

TUT 212

XX XX viii

Òu-eÒ-dar ÒeÒ lukur

RTC 355

XX XX i

Òu-eÒ-dar ÒeÒ lukur

RA 19, 40 21

XX XX xx 19

Òu-i

-li

ÒeÒ lukur(?)]

HLC 161 (pl. 103)

XX XX v

Òu-i

-li

ÒeÒ lukur

TBM 1, 148

XX XX vii

ud-du-Òa ÒeÒ lukur

Table 11: “brother of the concubine”

date

293

All of the texts mention deliveries for persons traveling to and from cities in the

eastern provinces. Some of these people are called couriers (lu

kas

) or messengers (sukkal),

but others have military titles, such as captain (nu-banda

), chiefs of police (aga

-us

gal,

literally “chief follower of the crown”), or simply knights (lu

geÒ

tukul). See for example SAT

1, 157 (= BM 21022) (no year, month 6):

Obverse

1. 1(ban2) kaÒ DU lugal

1 ban of regular standard quality beer,

2-kam Òa

uru

for 2 days while in the city,

2. 1(diÒ) dug dida kaskal-Òe

-la-a

1 jar of dida (beer) for the road: Ωelaya,

aga

-us

gal sa-bu-um

-Òe

DU-ni

police-chief, traveling to Sabum.

a. Perhaps the same person who is mentioned without title or familial relation in PDT 1, 434,

together with Ninkala (spelled nin-kal-la),

∑ulgi-simtπ

, and other important figures from the

empire as recipients of pairs of du

Òi

a boots (

kuÒ

ÒuÌub

Òi

-a e

-ba-an). The text also

mentions the father of

∑ulgi-simtπ

; DIB

-ib-si-na-at. The personal name

I†πb-sinat

mentioned in seven Ur III texts altogether. In all seven texts he appears as a high official together

with other high-ranking officials and members of the clan of Ur-Nammu (see BCT 1, 38 [from

∑

48 viii 19]; BIN 3, 603 [from IS 2 xii 1]; OIP 115, 293 [from

∑

XX XX 7]; Princeton 1, 81

[from AS 4 ix 26]; SAT 2, 774 [from AS 4 ix 25]; TBM 1, 451 [from

∑

46 xi

min

]; and PDT 1,

434 discussed here).

b. Only the PN e-la-ak-nu-id is known to me.

c. Perhaps the same as the military person mentioned in MVN 6, 215 (from

∑

34 vii), with the

title “police officer” (aga

-us

) and captain (nu-banda

d. Includes the provisions for the daughter of the king (dumu-mi

lugal).

e. Perhaps identical with the Idin-Ea (spelled i-DI-e

-a) who was a known brother of

Ea-niÒa

mentioned above fn. 254 on p. 104.

f. The reconstruction is suggested by the following reference combined with the circumstantial

evidence that only men appears as recipients of rations in these texts.

294

3. 2(ban

) kaÒ 2(diÒ) dug dida kaskal-Òe

2 ban of beer, 2 jars of dida (beer) for the

road,

en-ki u

-kul u

nu-ur

-NE

Lu-Enki,

ukul

, and Nºr-NE, captain

nu-banda

dumu nu-banda

son of the captain,

a-dam-dun

-Òe

DU-ni

travelling to Adamdun.

4. 3(diÒ) sila

kaÒ iÒ-me-a lu

-kas

3 sila of beer. IÒme-Ea, the runner,

Reverse

1. ÒuÒin

-ta DU-ni

travelling from Susa.

2. 5(diÒ) sila

kaÒ dan-i

-li

sukkal

5 sila of beer, Dæn-ilπ, messenger,

3. 3(diÒ) sila

kaÒ a-a-kal-la lu

-kas

3 sila of beer, Ayakala, runner,

sa-bu-um

-ta DU-ni

travelling from Sabum.

4. 2(ban

) kaÒ 2(diÒ) dug dida kaskal-Òe

2 ban of beer, and 2 jars of dida (beer) for

the road,

ku-da-num

ÒeÒ lukur u

i-din-

suen

Kudænum brother of the concubine,

and Idin-Suen,

sukkal a-dam-dun

-Òe

DU-ni

messenger, travelling to Adamdun.

5. 1(ban

) kaÒ da-da-a aga

-us

gal

1 ban of beer, Dadaya, police chief,

Òu-i

-li

sukkal ÒuÒin

-ta DU-ni

and ∑º-ilπ, messenger, travelling from Susa.

6. iti Òu-numun

Month

“Seeding”

The other people mentioned in this and similar texts all seem to belong to two

different social strata. The elite group of military staff and “diplomats”, on the one hand, and

the mid-level group of couriers and messengers, on the other.

The first person,

Ωelaya

, has the title police-chief. Nothing is known about this

person other than this reference. Sabum, the destination of his travel, was located on the

route to Susa, east of Sumer.

295

The next entry is for a pair of people, both traveling to Adamdun, an area south of

Sabum, southeast of Sumer.

675

The title of Lu-Enki (u

-kul) is presumably related to the

military, since it appears exclusively within this group of texts. The title of

Nºr-NE

, Lu-

Enki’s companion is usually understood as a military title comparable to a captain. Lu-Enki

is not otherwise attested with the title u

-kul. A messenger by the name of Lu-Enki, is

attested frequently as member of the diplomatic corps; see, for example, MVN 1, 142 (from

AS 5 viii 24), obv. 7 - 10:

1 gu

5(diÒ) udu ÒimaÒki(LU

.SU) / 5(diÒ) maÒ

gal ÒimaÒki

/ li-ba-an-aÒ-gu-bi lu

kin-gi

-a li-ba-nu-uk-Òa-ba-aÒ lu

mar-

a-Òi

/ giri

en-

ki sukkal

, “1 ox and 5

ÒimaÒki

sheep, 5 large

ÒimaÒki

goats;

LibanaÒ-Gubi

, the envoy of

Libanuk-∑abaÒ

, the man of

Mar

aÒi

676

via Lu-Enki, the messenger.” See also CT 10, 30

(= BM 014612) (from

∑

43), where a certain Lu-Enki, son of the captain (nu-banda

), is

mentioned. A police chief is called

Nºr-NE

in TBM 1, 122 (no year, month 8), traveling

from Urua (

urua

675. See W. Sallaberger,

Ur III-Zeit

(1999) plate 4, p. 157.

676.

Libanuk-∑abaÒ

is called governor (ensi

) of

MarhaÒi

in TCL 2, 5508 (from AS 4 i 6). Although it is

likely that /libana/ was a

MarhaÒi

PN theophoric element, the two names recorded here are the only

known names of persons from

MarhaÒi

where such an element is detectable, except for persons with

names such as

ÎaÒib-Atal

(BIN 3, 12 [from

∑

46 ix 12]), who probably was not a native of

MarhaÒi

and Amur-DINGIR, an envoy of

Libanuk-∑abaÒ

(see, for example, AUCT 2, 278 [from AS 4 ii 25]).

Other names of persons from

MarhaÒi

are:

ArÒik-u

(see, for example, CST 235 [from AS 1 viii 17]);

Banana (see, for example, CST 436 [from

∑S

6 iii 29]);

Arbi-maÒbi

(MVN 1, 124 [from AS 1 viii 17]);

Mar

uni (see, for example, MVN 13, 636 [from AS 8 i]); PN the envoy of

ArÒik-ug(u)bi

(see, for

example, MVN 15, 194 [from AS 1 v 29]);

BariaÒum

(see, for example, OrSP 47-49, 24 [from AS 3 vi

2 to 30]); and PN the envoy of Arwil-ug(u)bi (see, for example ASJ 19, 204 13 [from AS 1 vi 21]).

296

The next persons subsequently mentioned,

IÒme-Ea

677

Dæn-ilπ

, and Ayakala, were

all lower level officials. Their rations were considerably smaller than those of the first group.

Kudanum, the brother of the concubine, is only mentioned in this text and in TBM

1, 112. The last pair of persons, Dadaya and

∑º-ilπ

, are recorded as coming from Susa.

∑º-

ilπ,

the messenger, is recorded frequently in the published record. Dadaya travels to Susa with

Apilænum

in HSS 4, 67 (no year, month 8), and to Adamdun together with Idin-Ea in

MVN 9, 136 (no year, month 8).

The term lukur qualifies two groups of women, one being the prosperous group of

royal consorts, and to a lesser extent consorts of other high-ranking officials, the other being

a group of temple staff. The group of men with some affiliation to the military or the

diplomatic corps discussed in this excursus is the only closed group distinguishing their status

through a laterally defined genealogical association. Whether this relation was with either of

the two groups of concubines, outlined above, is less important than the sociological

conclusions that can be drawn from this observation, namely that in a patriarchial society a

person’s rank is presumably defined through his patrilineal descent only, but persons without

any relation to the ruling group except the marital link of their sister are forced to use this as

their claim of affiliation.

677. See for example SET 212, which mentions a person called

IÒme

-Ea, son of the captain, travelling from

Sabum, obv. 15-16:

iÒ-me<-e

>-a dumu nu-banda

/ sa-bu-um

-ta

DU-ni

297

Bi b l i o g r a p h y

W. Andrae,

“Aus einem Berichte W. Andrae’s über seine Exkursion von Fara nach

den südbabylonischen Ruinenstätten,” MDOG 16 (1902-03) 16 -

24.

J. Bauer,

Der vorsargonische Abschnitt der mesopotamischen Geschichte

, in P.

Attinger et al. (eds.),

Mesopotamien, Späturuk-Zeit und

Frühdynastische Zei

t (= OBO 160/1; Freiburg 1998)

R. H. Beal,

“Is

KU∑

the Reading of

I∑

= kizû ?,”

NABU 1992:48

A. Bligh,

From Prince to King, Royal Succession in the House of Saud in the

Twentieth Century

(New York & London 1984).

J. Boese

and W. Sallaberger,

“Apil-Kin von Mari und die Könige der III. Dynastie von Ur,” AoF

23 (1996) 24-39.

R. Borger,

Assyrisch-babylonische Zeichenliste

(= AOAT 33-33A; Neukirchen-

Vluyn 1981).

R. Burling,

The Passage of Power, Studies in Political Succession

(New York &

London 1974).

A. Cavigneaux

and M. Krebernik,

“Nin-ura,” RLA 9, 510.

298

M. Civil,

The Farmer's Instructions. A Sumerian Agricultural Manual

(= Aula

Orientalis Supplementa 5; Barcelona1994).

M. Cohen,

“The Gods of Suburban Umma,” Fs. Limet (

Tablettes et images aus

pays de Sumer et d'Akkad: Melanges offert a Monsieur H. Limet

) (Liege

1996) 27 - 36.

G. Contenau,

Contribution a l'histoire économique d'Umma

(Paris 1915).

J. Dahl,

“Land Allotments During the Third Dynasty of Ur, Some

Observations,” AoF 29 (2002) 330 - 338.

I. Diakonoff,

“Slave-Labour vs. Non-Slave Labour: The Problem of Definition,"

AOS 68 (= M. Powell (ed.),

Labor in the Ancient Near East

; 1987) 1 -

R. H. C. Davis,

“William of Jumieges, Robert Curthose and the Norman Succession,”

The English Historical Review, Vol. 95, Issue 376 (1980), 597-606

Stable URL: http://links.jstor.org/sici?sici=0013-

8266%28198007%2995%3A376%3C597%3AWOJRCA%3E2.0.C

O%3B2-9

D. Edzard,

Die “Zweiten Zwischenzeit” Babyloniens

(Wiesbaden 1957).

D. Edzard,

“Königsinschriften Des Iraq Museums II,” Sumer 15 (1959) 19 - 28.

D. Edzard,

“Sumerer und Semiten in der Frühen Geschichte Mesopotamiens,”

Genava NS 8 241ff. (= Rencontre 9; Geneva 1960) 241 - 258.

300

B. Foster,

Umma in the Sargonic Period

(= Memoirs of the Connecticut

Academy of Arts and Science, Vol. 20; Connecticut 1982).

B. Foster,

“Management and Administration in the Sargonic Period,” in M.

Liverani (ed.),

Akkad, the First World Empire. Structure, Ideology,

Traditions

(= HANE/S - V; Padua 1993)

E. Flückiger-Hawker,

Urnamma of Ur in Sumerian Literary Tradition

(= OBO 166; Freiburg

1999).

D. Frayne,

Ur III period (2112 - 2004 BC)

(= RIME 3/2; Toronto 1997).

I. Gelb,

“Sumerians and Akkadians in their Ethno-Linguistic Relationship,”

Genava NS 8 241ff. (= Rencontre 9; Geneva 1960) 258 - 271.

I. Gelb,

“The Ancient Mesopotamian Ration System,” JNES 24 (1965) 230-

243.

I. Gelb,

"On the Alleged Temple and State Economies in Ancient

Mesopotamia," Studi in Onore di Edoardo Volterra, Vol. 6 (1971)

138-154.

I. Gelb,

“The arua Institution,” RA 66 (1972) 1 - 32.

I. Gelb,

“Prisoners of War in Early Mesopotamia,” JNES 32 (1973) 70-98.

I. Gelb,

Glossary of Old Akkadian

(= MAD 3; Chicago 1957).

303

J. Klein,

Three

∑ulgi

Hymns

(Ramat-Gan 1981).

J. Klein,

“Shelepputum a Hitherto Unknown Ur III Princess,” ZA 80 (1990)

20-39.

F. W. König,

“Geschwisterehe in Elam” RLA 3 (1964) 224 - 231.

N. Koslova,

“(Selbst) ein freier Mann ist nicht gegen die Fronarbeit gefeit...”

(forthcoming).

F. Kraus,

Sumerer und Akkader. Ein Problem der altmesopotamischen Geschite

Mededelingen der Koninklijke Nederlandse Akademie van

Wetenschappen, Afd. Letterdunde, Nieuwe Reeks, Deel 33, No. 8

(Amsterdam/London 1970) 319 - 415 (1 - 99).

M. Krebernik

Die Beschwörungen aus Fara und Ebla

(Hildesheim 1984).

B. Landsberger,

The Date Palm and its By-products according to the Cuneiform Sources

(= Archiv für Orientforschung 6; 1967).

M. Larsen,

The Old Assyrian City-State and its Colonies

(= Mesopotamia 4;

Copenhagen 1976).

M. Liverani,

“The Medes at Esarhaddon's Court,” JCS 47(1995) 57-62.

J. Le Patourel,

“The Norman Succession, 996-1135,” The English Historical

Review, Vol. 86, No. 339. (1971) 225-250.

305

K. Maekawa,

"The Governor's family and the 'temple households' in Ur III Girsu,"

in K. Veenhof, ed.,

Houses and Households in Ancient Mesopotamia

CRRAI 40; Leiden 1996) 171-179.

K. Maekawa,

“The “Temples” and the “Temple Personnel” of Ur III Girsu-Lagash”,

in K. Watanabe ed.,

Papers of the Second Colloquium on the Ancient

Near East - The City and its Life held at the Middle Eastern Culture

Center in Japan (Mitaka, Tokyo)

, Sonderdruck (Heidelberg 1999) 61-

95.

R. Mayr,

The Seal Impressions of Ur III Umma

(unpublished dissertation;

Leiden 1997).

D. McGuiness,

Studies in Neo-Sumerian Administrative Machinery

(unpublished

dissertation; UMI 1981).

D. McGuiness,

“The Family of Giri-Zal,” RA 76 (1982) 17-25.

D. McGuiness,

“Ur III Prosopography: Some Methodological Considerations,”

Archív Orientální 50 (1982) 324 - 342.

P. Michalowski,

“The Bride of Simanum,” JAOS 95 (1975), 716 - 719.

P. Michalowski

“Durum and Uruk during the Ur III Period,” Mesopotamia 12

(1977) 83 - 96.

P. Michalowski,

“Amar-Su’ena and the Historical Tradition,” in: M. de J. Ellis, ed.,

EANEF (= Fs. Finkelstein; Hamden 1977) 155-157.

306

P. Michalowski,

“History as Charter,” JAOS 103 (1983) 237 - 248

P. Michalowski,

Letters from Early Mesopotamia

(Atlanta 1993).

P. Moorey,

Kish Excavations 1923-1933

(Oxford 1978).

H. Neumann,

“Ur-Dumuzida and Ur-Dun,” in J. G. Derckssen (ed.),

Trade and

Finance in Ancient Mesopotamia, Leiden 1997

(Istanbul 1999) 43 -

53. 30

A. Leo Oppenheim,

Ancient Mesopotamia; Portrait of a Dead civilization

(Chicago 1964)

D. Owen,

"The Ensis of Gudua," ASJ 15 (1993) 131-152.

D. Owen,

“The Household of Princess Simat-I

taran, Sister of

∑

u-Suen”

unpublished paper read at the RAI 47 (2001).

D. Owen

“On the Patronymy of

∑

u-Suen.” NABU 2001/17.

S. Parpola

and K. Watanabe,

Neo-Assyrian Treaties and Loyalty Oaths

(= SSA II; Helsinki 1988).

P. A. Parr,

“Ninhilia: Wife of Ayakala, Governor of Umma,” JCS 26 (1974) 90 -

111.

J. N. Postgate,

Early Mesopotamia, Society and Economy at the Dawn of History

(London 1992).

307

N. Postgate,

“Royal Ideology abd State administration in Sumer and Akkad” in J.

Sasson (ed.)

Civilization of the Ancient Near East

(New York 1995)

395 - 411

M. Powell,

“Sumerian Merchants and the Problem of Profit,” IRAQ 39 (=

Rencontre 23 Birmingham) (1977) 23-29.

W. Powell,

Saudi Arabia and its Royal Family

(Secaucus 1982).

K. Polanyi,

“Marketless Trading in Hammurabi’s Time,” in K. Polanyi et. al.

(eds):

Trade and Market in the Early Empires, Economies in History and

Theory

(Illinois 1957) 12 - 27.

F. Pomponio,

"Lukalla of Umma," ZA 82 (1992) 169 - 179.

J. Reade,

“Assyrian King-Lists, the Royal Tombs of Ur, and Indus Origins,”

JNES 60 (2001) 1 – 29.

J. Renger,

“The Daughters of Urbaba: Some thoughts on the Succession to the

Throne during the 2. Dynasty of Lagash,”

Fs. Kramer (= AOAT;

1976), 367 - 369.

M. Roth,

Law Collections from Mesopotamia and Asia Minor

(Atlanta 1997).

W. Sallaberger,

Der Kultische Kalender der Ur III-Zeit, Teil 1 & 2

(= UAVA 7/1 & 7/

2; Berlin/New York 1993).

W. Sallaberger,

Der Babylonische Töpfer

(= UHEM 3; Ghent 1996).

308

W. Sallaberger,

Ur III-Zeit

, in P. Attinger et al. (eds.),

Mesopotamien, Akkade-Zeit und

Ur III Zeit

(= OBO 160/3; Freiburg 1999).

A. Salonen,

Die Hausgeräte der alten Mesopotamier nach sumerisch-akkadischen

Quellen 1-2

(=AASF B 139, 144; Helsinki 1965-1966).

A. Schneider,

Die Anfänge der Kulturwirtschaft. Die Sumerische Tempelstadt

(Essen

1920).

M. Sigrist,

Drehem

(Bethesda 1992).

D. Snell,

“The Activities of some Merchants of Umma,” IRAQ 39 (=

Rencontre 23, Birmingham) (1977) 45 - 50.

D. Snell,

Ledgers and Prices

(= YNER 8; Yale 1982).

D. Snell,

Life in the Ancient Near East

(New Haven 1997).

E. Sollberger,

“Sur la chronologie des rois d'Ur et quelques problemes connexes”

AfO 17 (1954-1956) 10-48.

L. Speleers,

Catalogue des intailles et empreintes orientales des Musées Royaux du

Cinquantenaire

(Bruxelles 1917).

P. Steinkeller

“The Renting of Fields in Early Mesopotamia and the development

of the concept of "Interest" in Sumerian,”JESHO 24 (1981) 113 -

145.

309

P. Steinkeller,

“On the Reading and Meaning of igi-kár and gúrum (IGI.GAR),”

ASJ 4 (1982) 149-151.

P. Steinkeller,

“The Administration and Economic Organization of the Ur III State:

The Core and the Periphery,” in McG. Gibson & R. Biggs (eds.),

The

Organization of Power

(= SAOC 46; Chicago 1987)

P. Steinkeller,

“The Foresters of Umma: Toward a Definition of Ur III Labor,” AOS

68 (=M. Powell (ed.),

Labor in the Ancient Near East

; 1987) 73 - 115.

P. Steinkeller,

“The date of Gudea and his dynasty,’ JCS 40 (1988) 47 - 53.

P. Steinkeller,

“Early Political Development in Mesopotamia, and the Origins of the

Sargonic Empire,” in M. Liverani (ed.),

Akkad. The First Worls

Empire. Structure, Ideology, Traditions

(= HANE/S V; Padua 1993)

107-129.

P. Steinkeller,

“The Organisation of Crafts in third Millennium Babylonia: The

Case of Potters,” AoF 23 (1996) 232-253.

P. Steinkeller,

“New Light on the Hydrology and Topography of Southern

Babylonia in the Third Millennium,” ZA 91 (2001) 22 - 84.

M. Stepien,

Animal Husbandry in the Ancient Near East

(Bethesda 1996).

M. Stolper,

“On the Dynasty of

∑

ima

ki and the Early Sukkalma

s,” ZA 72

(1982) 42 - 67.

310

M. P. Streck,

“Das Onomastikon der Beamten am neubabylonischen Ebabbar-

Tempel in Sippar,” ZA 91 (2001) 110 - 119.

V. Struve,

“Some new data on the organization of Labour and on Social

structure in Sumer during the reign of the third dynasty of Ur,”

(originally written in Russian in 1949), in I. Diakonoff (ed.),

Ancient

Mesopotamia: Socio-Economic History, A Collection of Studies by Soviet

Scholars

(Moscow 1969) 127 - 172.

V. Struve,

“The Accounts of Work-Team Overseers on a Royal Estate under the

Third Dynasty of Ur,” in:

Papers presented by the Soviet Delegation of

the 23

International congress of Orientalists

(Cambridge 1954) 43 -

51.

C. Suter,

Gudea’s Temple Building

(Groningen 2000).

F. Thureau-Dangin,

RA 7 (1910), 186. 99

A. Uchitel,

“Daily work at the Sagdana Millhouse,” ASJ 6 (1984) 75 - 98.

P. Van Den Berghe

and G. Mesher,

“Royal Incest and Inclusive Fitness,” American Ethnologist, Vol. 7,

Issue 2 ( 1980), 300 - 317.

Stable URL: http://links.jstor.org/sici?sici=0094-

0496%28198005%297%3A2%3C300%3ARIAIF%3E2.0.CO%3B

2-J

M. van de Mieroop, “An Accountant's Nightmare: the Drafting of a Year's Summary,”

AfO 46-47 (1999/2000) 111-129.

311

C.-A. Vincente,

“The Tall Leilan recension of the Sumerian King List,” ZA 85 (1995)

234 - 270.

H. Waetzoldt,

Untersuchungen zur Neusumerischen Textilindustrie

(= UNT; Rome

1972)

H. Waetzoldt

Bi.Or. 32, 1975, 383.

H. Waetzoldt,

“Änderung von Siegellegenden als Reflex der ‘großen Politik’,” in U.

Finkbeiner (ed.),

Boehmer Fs

(Mainz: 1995) 659-664.

Aa. Westenholz,

The Old Akkadian Period, History and Culture

, in P. Attinger et al.,

(eds).,

Mesopotamien, Akkade-Zeit und Ur III Zeit

(= OBO 160/3;

Freiburg 1999).

M. Weber,

On Charisma and Institution Building, Selected Papers, Edited and with

an Introduction by S. N. Eisenstadt

(Chicago & London: 1968)

R. Whiting,

“

TiÒ-Atal

of Nineveh and Babati, Uncle of

∑u-Sin

,” JCS 28 (1976)

173 - 182.

C. Wilcke,

“Zum Königtum in der Ur III-Zeit,” in P. Garelli ed., Le palais et la

royautée (= Rencontre 19: Paris 1971/75) 177 - 232.

C. Wilcke,

“Die sumerische Königsliste und erzählte Vergangenheit,” in, J. Von

Ungern-Sternberg (ed.),

Vergangenheit in mündlicher Überlieferung

(

Stuttgart 1988) 113 - 140.

Document Outline