2
Introduction
Numerous Sumerian and Akkadian terms relate to some sort of detention function. This material points to the conclusion that imprisonment was multifunctional and rarely related to criminal activity, providing an important context for understanding the intersection between imprisonment and “crime” in early Mesopotamia. To demonstrate this, I will discuss a number of significant examples of terminology related to detention. This terminology will be elucidated by considering the socio-economic status of those detained together with the reasons for their confinement. Beyond the terms related to social-historical practice, a brief section will also be devoted to terms found in literary contexts that relate to imprisonment. This chapter, however, primarily seeks to establish the functional reality of detention in early Mesopotamia, as opposed to the literary and ideological reception of imprisonment, which will be considered elsewhere in this book.
e2-eš2
The e2-eš2, literally “house of the rope,” is one of the earliest attestations of the terms often associated with imprisonment.1 The use of this terminology/sign combination might be inspired by the actual practice of placing a rope on those who were detained. This view is supported by the terminology, nose-rope (eš2-ĝiri17) from the Ur III period, as discussed below, together with the iconography of ropes attached to captives (Figure 3).2 The use of a rope for controlling movement is perhaps also attested in the Proto-Cuneiform record. An earlier sign SAG×MA from Jemdet Nasr seems to represent a human head and a rope (Figure 4).3
Figure 3. Prisoner bound by ropes depicted on an Akkadian vase from Uruk. Paris, Musée du Louvre, AO 5683. Rendering courtesy of Jarett Hall.
Figure 4. Male and female SAG×MA and ERIMa from Jemdet Nasr representing a human head with a rope and a yoke respectively. Drawing by J. Nicholas Reid.
The signs in Figure 4 appear in a text that counts male and females, who are depicted with a head and a rope and a yoke. Robert K. Englund argued that such workers had foreign names and were perhaps prisoners of war.4 Since imprisonment was multicontextual and space was multifunctional, it is possible that the early term e2-eš2 related to the Mesopotamian houses of detention became associated with prisoners, since the rope was an implement that was sometimes used to control the body.
An early attestation of the e2-eš2 occurs in the Early Dynastic royal inscription/edict, the “Reforms of UruKAgina.”5
Reforms of UruKAgina (RIME 1.9.9.1) xii: 12–286
xii: 12. i3-du11
13. dumu lagaški
14. ur5-ra ti-la
15. gur gub-ba
16. še si-ga
17. nig2-zuḫ-a
18. sag geš ra-a
19. e2-eš2-bi
20. e-luḫ
21. ama-gi4-bi
22. e-gar
23. nu-siki nu-ma-kuš2
24. lu2 a2-tuku
25. nu-na-ga2-ga2-a
26. dnin-ĝir2-su-da
27. uru-KA-gi-na-ke4
28. inim-be2 ka e-da-keš2
Translation: (These things) he proclaimed. “As for the citizens of Lagaš – the one living in debt, the one who had set up (a false) gur measure, the one who had (fraudulently) filled up the gur measure with barley, the thief, the murderer – he swept the prison clear (of them) and established their freedom. UruKAgina made a binding oral agreement with the god Ninĝirsu that he would never subjugate the orphan (or) widow to the powerful…”
Among these reforms, UruKAgina declares that he “swept the prison clear (of them) and established their freedom.” The “Reforms of UruKAgina” attests to multi-functional imprisonment in relation to “crimes” and debt at the very early period.
Beyond the “Reforms of UruKAgina,” the term e2-eš2 occurs particularly in the Early Dynastic and Old Akkadian Periods. For example, FAOS 19: Gir 25, cited below, deals with imprisoned runaways. Most of the Early Dynastic texts that deal with the prison include terminology that they were living in prison (e2-eš2-ka i3-ti) but fail to include reasons behind the imprisonment.7
The term e2-eš2 is also attested during the Old Akkadian period, where several texts reference people stationed or living in the prison (i.e. ITT 1, 1287, 1379, 1418).8 The Old Akkadian text, CUSAS 11: 265 seems to be dealing with a dependent or slave who is living in prison because of a deficit owed by another.9
The e2-eš2 is also attested in a recently published text from Lagaš II.
Maiocchi and Visicato 2020: 511 (BM 88538 – Maiocchi and Molina 2018: 82)10
Obverse
1. 1(diš) ki-tuš-lu2
2. dumu geme2-dgiš-bar-e3
3. [1(diš)] ° lu2-zaḫ3
4. dumu im-ta
5. en-DU
6. dumu lugal-en
Reverse
1. šunigin [3 (diš)] ĝuruš
2. nu-˹e3˺
3. e2-eš2-ka
4. i3-se12
(rest uninscribed)
Translation: One Kitušlu, the son of Gemegišbare; [one] Luzaḫ, the son of Imta; EnDU, the son of Lugalen. Total: [3] workers. They did not go out, they are dwelling in prison.
Maiocchi and Visicato (2020: 511) includes a list of three ĝuruš, male workers for the administrative bodies, who have not gone out for work but are living in prison. This imprisonment relates to their social position as “landless” dependents of the administrative body.11 The first prisoner, Kitušlu, is identified by his mother, on the basis of the name Gemegišbare, rather than his father. When ĝuruš were identified by their mother’s names, it is usually because they belonged to families lacking a father.12 Their mothers worked for the administrations in a variety of capacities. Without a means of production, these women appear dependent on the administrative households, mostly temples, and receive rations of 30–60 liters of barley, and sometimes as low as 20 liters per month. The lower end of these are practically starvation rations, as it takes two liters per day to subsist.13 The children of these women were left with little to no choice but to work for these institutions. The second prisoner, named “Fugitive” (Luzaḫ), might be connected to flight, which is well-attested in relation to imprisonment/detention in early Mesopotamia. Given the context and socio-economic status of these dependent workers, their detention was likely intended to force job performance and did not intersect with “crime.” This mechanism of detention for forcing job performance, however, is primarily coercive rather than corrective.
The terminology also occurs in Ur III texts but appears less frequently than the more common term ennuĝ, discussed below. The known Ur III examples of e2-eš2, like the earlier attestations, occur in administrative contexts.14
While the early terminology of the e2-eš2 relates to multifunctional imprisonment for a variety of “crimes” and debt in the “Reforms of UruKAgina,” most examples point to an administrative context of detention in relation to work, which is particularly prominent in early Mesopotamia.
Although there is some intersection between criminal behavior and the e2-eš2, the term seems to be related to administrative contexts that had detention functions for the purpose of restricting movement and utilizing labor. Although there is some evidence pointing to the use of the e2-eš2 to hold those associated with criminal activity, a distinction must be drawn between what an institution was and how it could be used. The evidence does not point to a prison in the strict sense that focused on detention in relation to criminal activity. Instead, the terminology is used to refer to administrative bodies with multifunctional detention.
bīt asīrī
The administrative context of detention also occurs with the Akkadian term bīt asīrī, which was an administrative body in the Old Babylonian period.15 The bīt asīrī is attested to in group of more than 320 texts from the reign of Rīm-Anum of Uruk, who rebelled against Babylon around 1742 BC during the reign of Babylonian king Samsu-iluna.16
As Rositani argues, the bīt asīrī was more than a bureau, however.17 It was a physical place where prisoners of war were housed prior to and in between work assignments. While in the bīt asīrī, the prisoners were worked, likely in connection to milling. Rositani suggests that the prisoners milled for their own sustenance and for grain that was to be distributed.18
Numerous scholars have discussed the meaning of the bīt asīrī. Leemans argues that it may be translated “the house of the prisoners of war.”19 Rositani, following Leemans, traces the related terminology and related contexts of the bīt asīrī, drawing a distinction between administrative captivity for the purpose of labor acquisition, which fits the bīt asīrī, and contexts where some sort of punishment also occurred.20
One example of administrative detention with the bīt asīrī relates to runaways. In general, flight in the Old Babylonian period did not usually result in time spent in prison as it did in the Ur III period, but the closest related evidence to the practice is found in texts dealing with the bīt asīrī.21 Nisaba 4 II: 28 deals with Abdi-Ištar who was among thirteen weavers who fled from Larsa.22 Abdi-Ištar was given into the custody of the overseer of the prisoners, Suen-šemi. Another text (Nisaba 4 II: 66) dealing with fugitives has a similar outcome and yet this time deals with two slaves, who were subsequently caught.23
This house of prisoners was a workhouse primarily used to oversee political prisoners and slaves. As such, detainees in this administrative workhouse should be distinguished from prisoners who are considered criminals of sorts.24 The bīt asīrī had a function of detention but does not seem to relate to prisons per se.
e2 ennuĝ = bīt ṣibitti
The most prominent Sumerian term in early Mesopotamia that relates to prison, likely because of its use by the Ur III scribes, is the ennuĝ “guard.” Civil addresses questions about the meaning of the native terminology as follows:
The word en-nu-un has three semantically related meanings: (I) “watch, guard”; (2) “watch” (as a division of night time); and (3) “imprisonment” and, when designating a place, “prison.” An examination of the administrative texts where this word appears shows that although meaning I is known in Ur III texts, meaning 3 is more frequent. Whenever en-nu-un is used in the locative with the verbs ti-(l) or tuš, or the subject is qualified of lú-dab5-ba, and in most instances of the construction šà en-nu-un-gá-(me), the meaning is undoubtedly “detention” or “jail.”25
The OB lexical text Nippur Izi includes the following semantic range for the term ennu(ĝ): guard, watch, prison/prisoner.
Tab.I, 498
en-nu-uĝ3—prison
Tab.I, 499
ki en-nu-uĝ3—watch(house)
Tab.I, 499a
na-kam-tum—storehouse
Tab.I, 499b
eš3-ta-gur-ra—treasury
Tab.I, 499c
en-nu-uĝ3—night watch
Tab.I, 500
en-nu-uĝ3 an-usan—evening watch
Tab.I, 501
en-nu-uĝ3 murub₄—middle watch
Tab.I, 502
en-nu-uĝ3 ten-na—watch of the approach
Tab.I, 503
en-nu-uĝ3 ud zal-le-da—watch during the day
Tab.I, 504
en-nu-uĝ3 šušana—one-third watch
Tab.I, 505
en-nu-uĝ3 sa9—half watch
Tab.I, 506
en-nu-uĝ3 šanabi—two-thirds watch
Tab.I, 506a
en-nu-uĝ3 u—tenth watch
Tab.I, 507
en-nu-uĝ3 ti-la—prisoner
Tab.I, 507a
en-nu iri—city watch
The range of meanings is determined by the lexical context. The first occurrence of en-nu-uĝ3 in Nippur Izi relates to prisons (literally: guardhouse) because it is followed by other places or physical structures like the (watch)house, storehouse, and treasury. The various watches of the day and night determine the transition to en-nu-uĝ3 as a night watch, while the meaning “prisoner” refers to a person living under guard (en-nu-uĝ3 ti-la). The last mentioned features prominently in the Ur III record, often written en-nun-ĝa2 ti-la.26 This again points to the multifunctionality of detention/imprisonment in early Mesopotamia, since the term “living under guard” (en-nu-uĝ3 ti-la) just denotes any person detained in the presence of a guard. Detainment in relation to criminal activity intersects with this practice in the record, but the primary function of such detention appears to be for the purpose of holding debtors or lower stratum workers for the purpose of coercing work.
The following, organized by year, is an extensive but not exhaustive list of Ur III evidence related to ennu(ĝ) (Table 1 and Figure 5).
Table 1. Ur III Prison Texts
Figure 5. Living in Prison. Obverse of Sumerian tablet from Umma, YOS 4, 183. Courtesy of the Yale Babylonian Collection. Photography by Klaus Wagensonner.
While most texts list prisoners by their names, terminology used to designate the detained includes lu2-dab5-ba (seized men) and ĝuruš (generally lower stratum workers).27 Females are similarly described as geme2 (dependent females).
The administrative bodies needed to account for dependent workers, which appears to be the primary reason to mention prisoners in the record. Most of these texts listed above omit reasons for confinement and instead focus on time and are thus likely the product of the administrative desire to keep track of the human resources available to the overseers together with the distribution of rations. This is further evidenced by half-production prisoners mentioned in the record. In SNAT 405, two individuals are listed as half-production and are designated by parents, suggesting these are children.126
A second group of these texts record the provisions, mostly food but sometimes clothing, for prisoners and guards. As with previous terminology, the textual documentation of imprisonment comes in an administrative context of accounting, this time in relation to movable goods.
A third group of texts records the offenses committed by various individuals. The inclusion and omission of information in these different types of texts stem from the overall purpose of writing, resulting in brief glimpses into practices relating to prisons and prisoners during the Ur III period.127
The term ennu(ĝ), as discussed above, can also make reference to a guard, as with the texts discussed below, that relate to the provisions received by the guard(s) at Girsu.128 The guards mentioned in these messenger texts are probably those stationed at the Girsu resthouse, where there was an en-nu gu-la or “big prison.”129 Places of detention appear frequently in messenger texts associated with these resthouses. During the Ur III period, messengers were sent on a variety of missions, and there were seven known resthouses throughout the region. Resthouses are mentioned in the literary text “Šulgi A,” where Šulgi boasts that he made the roads secure for travel and “built there big houses,”130 referring to temporary lodging for travelers. These resthouses, meant to facilitate safe travel and state business, had detention functions, but this detention function does not appear to be the primary reason for them.
With these occurrences, it is clear that the ennu(ĝ) was related to guards that had administrative duties in relation to the state. The ability to restrict worker movement and coerce production made this a fitting place to detain any who were coerced to work, whether those accused of “crime,” runaways, or debtors.
The related Akkadian term ṣibittum is also well-attested in the Old Babylonian record. This term is nomen actionis derived from the verb ṣabātum “to seize” and occurs in the Old Babylonian Omen, YOS 10, 33 (Rev. iv: 20–21):131
Rev. Column iv
20. šumma ˹ubānum˺ a-na i-mi-tim ka-˹me-a˺-at a-we-lum i-na a-˹li˺-ka
21. a-na ˹ṣi2˺-bi-it-tim in-na-ad-˹di˺
Translation: “If the finger/toe is attached on the right, a man in your city will be thrown in prison.”
An Old Babylonian omen text (YOS 10, 47) indicates that there was a concern that the prison could revolt. The following occurs in the Old Babylonian Omen, YOS 10, 47 Reverse line 7 reads as follows:
Reverse
Line 7: šumma qa-˹ra˺-an na-ag-la-bi-im ša i-mi-tim a-na 2(diš) pa-ṭi-ir ṣi-bi-˹it-tim˺ ib-ba-la-ka-at
Translation: “If the ridge of the shoulder blade opens on two sides, the prison will revolt.”132
The ṣibittum also appears in numerous letters that deal with people being thrown into prison or being held there typically because of an offense committed or some debt (AbB 2, 121; AbB 6, 144; AbB 11, 60; etc.).133 As will be seen, this was primarily for holding as part of the judicial process or to coerce payment,134 and by the Old Babylonian period, the context of imprisonment in relation to the ṣibittum does not deal with the distribution of rations or workers. This will be discussed more fully below when considering life on the inside and prisoner care.
The Sumerian terminology ennuĝ seems capable of denoting detention, typically administrative, for a variety of reasons. While various offenses appear in relation to this terminology, the majority of examples appear in texts that are primarily concerned with the distribution of rations and human resources in administrative contexts. When offenses are mentioned, this is likely related to the judicial process or to coerce work and payment. By the Old Babylonian period, the ṣibittum deals with those being held as part of a judicial process or to coerce payment. These texts, unlike the Neo-Sumerian, do not focus on the distribution of rations and the movement of human resources. Again, this provides an important social and historical context to understand the intersection between common detention and detention related to criminal activity, together with ideology that developed in literary and ritual contexts.
Other Key Terms
One early institution with some detention function is the e2-keš2/kešda, literally the “house” + “to bind.” Vitali Bartash argued that this administrative institution, where some children were branded, was an enclosed house.135 In the full edition of the text (CUSAS 35: 23) and a more recent article, he connected e2-keš2/kešda to a place of wool and textile production.136 The status of those housed were either slaves or prisoners of war.137 The connection between prisoners of war and slaves with textile production is common in ancient Mesopotamia, as seen below. Whether a labor house or a place of wool and textile production, it was not a prison in any strict sense and appears to have functioned in similar fashion to the bīt asīrī. This evidence again situates most detention in early Mesopotamia in an administrative context of work/labor coercion.
Other relevant Akkadian terminology includes the e2 (bīt) ki-še3-er-tim. Although mostly attested in Old Assyrian texts as kišersu,138 the term appears in an Old Akkadian letter. HSS 10, 10, deals with someone who is detained until an official arrives. This is temporary detention, which fits the overall evidence. Beyond this, the term appears infrequently in the record from early Mesopotamia.
Another generic term for detainment from the Old Babylonian period is the verbal form kalû. This detainment can be for holding distraints (see CAD K: 98) but also can be used to refer to other forms of detainment that may involve fetters (AbB 1, 27), confinement in houses (AbB 11, 25; AbB 8, 12), or in cities (AbB 6, 178). The term kalû also appears in relation to a four-day confinement for speaking against someone that resulted in the captive person crying out, which caused the offended person to have compassion on the captive and release him (AbB 14, 29). In BM 103136 (1910-10-8, 64), cited below, the prison (ṣibittim) is mentioned in relation to a murder, and the text says to detain him (ki-il-la-aš-šu). The terminology seems reflective of the numerous ways in which detainment might be used, the details of which will be further discussed below.139
Another important term for understanding detention in early Mesopotamia is the nupārum (or bīt nupārim). This was a workhouse, where a person could be detained to work in relation to debt and may be compared with the bīt ṣibitti.140
At Mari, the nepārātum (“ergasterions”) are described by Lafont as “large prison-workshops where mass productions were organized.”141 After discussing the challenges of isolating the origin of the non-Akkadian term, nepārātum, together with some possible connections, Jean-Marie Durand notes that it is generally viewed as an ergastule, or bonded workplace.142 Indeed Denis Lacambre and Julie Patrier note that although the term ergastule is not ideal, it brings together two important aspects of the nepārum, confinement and bonded labor.143
These workhouses were situated near palaces and temples and were used to hold any person who might be considered dangerous or a flight risk.144 Prisoners of war, for example, were also locked up in the nepārātum for the utilization of their labor.145 Those held in the nepārātum could be punished with eye-gouging or the removal of the tongue, and they could be held there indefinitely.146 These workhouses often related to the production of textiles and food.147 As Lafont discusses, the use of prisoners of war in the textile industry also occurred in the Ur III period,148 a practice which can be considered universal throughout the ancient Near East and eastern Mediterranean.149
Jack Sasson suspects that the singular nepārum has a jail in view, while noting the that the nepārātum did not necessarily indicate a jail.150 However, it seems better to view these as workhouses where labor can be coerced and movement restricted. Even if a criminal is held in that context, it is likely not denoting a jail or prison as such. Rather, such a criminal presence points to the multifunctional use of existing agents and structures (workhouses, guards, etc.) to meet a diversity of needs and situations. This points to multifunctional detention in administrative contexts, a common practice attested in early Mesopotamia.
While certain “crimes,” as seen below, could result in individuals spending time in the nepārātum, it would be wrong to view this as a prison where punishment was meted out. Though a person could be punished (eye-gouging or tongue cutting, for example)151 and placed in a workhouse to grind flour, this is a very different thing than punishing someone by imprisoning them. Instead, the institution reflects common ancient Near Eastern administrative bodies where detention occurred in labor contexts for a variety of reasons. In fact, the multifunctional and multicontextual use of detainment fits the ways in which administrative bodies functioned.
The ways in which the mechanisms for labor acquisition and control functioned can be seen by the evidence from the Ur III period.152 Dahl reconstructed a three-stage process by which prisoners of war were transported; received by the administrative officials; and subsequently employed in a variety of capacities.153 As seen in MVN 14, 569, a royal messenger and a cupbearer were given provisions for transporting human captives to the temple Šara.
MVN 14, 569154
Obverse:
1. 1(barig) 3(ban2) kaš du
2. 1(barig) 3(ban2) ninda
3. 1(diš) sila3 šum2
4. 1(diš) sila3 i3-geš
5. šu-eš18-dar lu2 kin-gi4-a lugal
6. 2(barig) 3(ban2) kaš du
7. 2(barig) 3(ban2) ninda
8. 2(diš) sila3 i3-geš
9. 1(diš) sila3 naga
10. lugal-šu-nir-re sagi
11. 2(barig) 3(ban2) kaš du
12. 2(barig) 3(ban2) ninda
13. 1(diš) sila3 šum2
14. 1(diš) sila3 i3-geš
15. šu-dnin-šubur
16. 4(ban2) 2(diš) sila3 kaš du
Reverse:
1. 4(ban2) 2(diš) sila3 ninda
2. 1/2(diš) sila3 šum2
3. 1/3(diš) sila3 6(diš) gin2 naga
4. dnin-marki sagi
5. e2 dšara2-še3 sag nam-ra-ak-da gen-na
6. 1(diš) kaš dida du 1(ban2) 5(diš) sila3
7. 1(ban2) dabin 1(diš) sila3 i3-geš
8. šu-den-lil2 lu2 kin-gi4-a sig4 ar-ḫa-še3 gen-na
9. šunigin 1(aš) 2(barig) 1(ban2) 2(diš) sila3 kaš du
10. šunigin 1(ban2) 5(diš) sila3 kaš dida du
11. šunigin 1(aš) 2(barig) 2(ban2) 2(diš) sila3 ninda u3 zi3 gur
12. iti šu-numun-na
13. mu us2-sa ma2 den-ki ba-ab-du8
Translation: 90 liters of normal beer. 90 liters of bread. One liter of garlic. One liter of oil. Šu-Eštar, the royal messenger. 150 liters of normal beer. 90 liters of bread. Two liters of oil. One liter of alkali. Lugal-Šunire the cupbearer. 150 liters of normal beer. 150 liters of bread. One liter of garlic. One liter of oil. Šu-Ninšubur. 42 liters of normal beer. 42 liters of bread. One half liter of garlic. 26 (gin2) of alkali. Ninmar, the cupbearer. Having gone with the people (lit. heads) taken as booty to the temple of Šara. One normal sweet beer fifteen liters. Ten semolina. One liter of oil. Šu-Enlil, the messenger, traveling to Arḫa for bricks. Total: 432 liters of normal beer. Total: fifteen liters of normal sweet beer. Total: 442 liters of bread and flour.
Month: Šu-numuna (6). Year after the boat of Enki was caulked (Šu-Suen 3).
In the above text, prisoners of war were brought from the eastern province for the purpose of utilizing their labor in administrative contexts.155
The second stage of the process involved the reception of the human booty by administrative officials, in order to provide an account of the recently acquired commodities. The structure of the text follows the standard format of a goods account for the Ur III period.
SAT 2, 1163156
Obverse:
1. 1(geš2) 5(u) 3(diš) geme2 sag-dub
2. 3(u) 1(diš) dumu-munus 1(aš) gur
3. 1(u) 3(diš) dumu-nita2
4. 1(u) 5(diš) dumu-munus
5. šunigin 2(geš2) 5(u) 2(diš) sag-dub
Reverse:
1. nam-ra-ak [a-ru-a] d[šara2-še3]
2. iri ša-ri2-ib?-ḫu-um-maki
3. ur-dli9-si4 ensi2 ummaki i3-dab5
4. iti šu-eš-ša 1(diš)
5. mu en dnanna ba-ḫun
Translation: 113 female slaves “of the head of the tablet.”157 31 female children at 1 gur. 13 male children. 15 female children. Total: 172 heads of the tablet. Booty dedicated to Šara, (of the) city of Šaripḫumma. Ur-Lisi, governor of Umma, took administrative control of them. Month Šuešša (8). Year: “the En-priest of Nanna was installed” (Amar-Suen 9).
The prisoners of war actively employed by administrations were provided rations and worked in a variety of capacities.
After being enveloped into the administrative bodies, prisoners could be worked, donated, or sold. During the fifth regnal year of Amar-Suen, a group of administrative texts (TCL 5, 6039; SAT 2, 884; Nik 2, 329)158 provide a bit more context for understanding the lives of prisoners of war during the Ur III period. The first of these, TCL 5, 6039, is a well-preserved text involving barley distributions to captured people (še-ba nam-ra-ak; rev. column iv. line 10). The text lists 185 individuals all of whom are women or children.159 Most of the prisoners of war in this text do not have Akkadian or Sumerian names. Among the listed 185 individuals, 56 people are clearly marked as dead (uš2) in the text.160 The dead are not included in any totals but are only listed with their individual crews, since the overseers had to account for what happened to the human resources that were no longer contributing labor nor receiving rations.
Of the remaining 128 survivors, 24 were sick (tu).161 The first five subtotals (šunigin) deal with crews of around twelve people, give or take. When a person died, another person was added to the crew. These added individuals are marked out in the texts as replacements. This indicates that there was also a pool of replacement workers available. For example, the first crew had one sick member and seven deaths. Apparently, the crew consisted of thirteen workers, one of whom was a child. Seven members of the crew died. Six of these were replaced. The replacements (daḫ-ḫu-me) are all listed as old (libir-am3 or libir-me). This is the general pattern for most of the crews listed.
All of this detail provides a bit of context for understanding the lives of prisoners of war in the Ur III period. The text reveals that there was a larger group of displaced prisoners of war. Since the names of the older people used as replacements are similar in kind to the other names found in the text, it is possible the namra’ak listed on the tablet were only a portion of a larger displaced community brought into the region of Umma to live and work. Since not everyone was receiving rations, the able-bodied might have been left with trying to provide for those who were not given work. When someone died, an older female could replace the deceased and receive the rations allotted for the work performed. Such displaced communities might provide further substantiation of Steinkeller’s views about settlements of prisoners of war during the Ur III period.162
Within three months, all but five of the sick individuals in the previous text fall out of the record together with many others. SAT 2, 884, dated to the fifth month of the fifth regnal year of Amar-Suen, lists forty-nine workers, ten of whom are children, receiving flour and beer as rations and categorized like TCL 5, 6039 as še-ba nam-ra-ak. Of these forty-nine workers, at least twenty-seven appear to be found in the previous text, TCL 5, 6039. SAT 2, 884 does not list any individuals as dead or sick. References to replacements and age are also not found in the text. The almost identical text, Nik 2, 329 was written two months later (Amar-Suen Year 5, Month 7). This text does not have any names missing from the second text (SAT 2, 884) and continues with the same amount of rations. The main difference is that Nik 2, 329 refers to provisions for the female booty (ša3-gal geme2 nam-ra-ak).
Prisoners of war, however, only made up a portion of the workforce during the Ur III period.163 Lafont rightly states that the Ur III factories included women, for example, who were “war-captives and booty”; “purchased or indebted slaves”; “donated personnel”; “women of impoverished classes and outcasts of society.” Lafont continues:
Yet these categories of women certainly did not represent the entire workforce inside the state-run workshops. They would have been in insufficient number to meet the economic needs of the state. In all likelihood, many of the female workers employed in the mills were simply state-dependents, working part-time, or full-time for rations and salaries, like their male counterparts.164
This shows the ways in which administrative bodies drew upon a variety of persons and types of persons in order to meet labor needs and fulfill state demands. This contextual reality explains the appearance of those accused of “crimes” or being coerced for payment and job performance within administrative bodies that have detention functions. These bodies existed for the purpose of gaining access to and controlling production. With such mechanisms and practices in place, it is very understandable this would be a suitable place to hold those accused of criminal activity.
In fact, the driving force behind such institutions relates less to criminal activity and more to gaining access to production and detaining resources. Bartash builds on the idea of labor shortage models for early polities to explain in part the importation of workers through war and the slave trade.165 This long history of controlling and gaining access to labor explains how detention came to intersect with those accused of “crime,” owing fines, debtors, and runaways. Within these existing social structures that included overseers and guards, it is natural that administrative bodies could be utilized when dealing with people accused of “crime” or when they needed to be coerced to pay debts, fines, or perform work responsibilities.
The Prison in Literary Texts
Nungal is not the only piece of literature that deals with imprisonment. William W. Hallo isolates instances where the Sumerian terms for the “big house,” written e2-gal and e2 gu-la may denote places of detention.166 While these terms more commonly mean “big house,” and by extension may also refer to a palace or temple, the e2 gu-la is associated with the prison in the “Hymn to Nungal” (see lines 32ff., 40ff., 69).167 Hallo used this semantic range of e2-gal and e2 gu-la to reinterpret other occurrences where the term appears in legal contexts with implications for the imagery of the “big house” in relation to prisoners that was present in other literary texts.168 According to Hallo, “In addition to its basic sense of ‘palace,’ é-gal is applied as an epithet to the royal storehouse, (royal) bivouac, the ark, and perhaps even the extended family.”169 Hallo continues, “é-gal can perhaps be translated literally as ‘big house,’ a colloquial equivalent for prison in contemporary American English. For it seems to have the sense of ‘prison’ in the Nungal hymn (lines 32f., 40f., 69) where the synonymous é-gu-la (line 10) is specifically identified as the ‘guard-house, brig’ (en-nu-un).”170 Hallo seeks to apply this potential meaning to the “big house” in the following proverb.
The use of e2-gal associated with the prison goddess occurs in the Sumerian text Proverb 6:3, which states: e2-gal tir-ra-am3 lugal ur-maḫ-e dnun-gal sa-šu2-uš-gal ĝuruš dul-dul-e (“the big house is a forest, the king, a lion, Nungal, a net thrown over man”).171 Here a prisoner was held captive by Nungal like in a net, and finds himself enclosed within the “big house” as if it were a forest. The king, whose responsibility was to execute justice, was like a powerful lion of the forest. Hallo’s argument, however, seems to overinterpret the presence of Nungal. The more immediate reference to the king in relation to the “big house” suggests that the more likely sense of the e2-gal in this text is the “palace.” The mention of Nungal in relation to the palace, as discussed below, likely connects to the king administering justice in the palace, where judicial process could also relate to oaths taken before Nungal and the net.
Hallo and Civil connect this to similar imagery in “Bilgameš and Ḫuwawa A,” which also mentions Nungal, connecting the forest and the “jail.” In this earlier Sumerian recension of “Bilgameš and Ḫuwawa A,”172 the heroes slay Ḫuwawa and present his head before the god Enlil. In the text, Enlil has Nungal imprison some of the magical auras in the story, which belonged to Ḫuwawa. Civil writes:
It appears Enlil wants to neutralize, as it were, the troublesome rays so that no further mischief can be committed with them. The first three are dispersed all over the landscape, fields, rivers and canals, and marshes. One must conclude that the rest, except the seventh, are locked up in jail…The forest, the lion, and the big house are all a metaphorical allusion to imprisonment. The relationship between jail and forest is unclear but textually assured.173
This concept, of course, connects to the idea of control in relation to the prison and provides meaningful connections to the Netherworld serving a similar imprisoning function discussed below.
Conclusion
The lexical and socio-economic information about detention in early Mesopotamia provides the broader context in which to understand the following chapters that deal with the intersection between “crime” and imprisonment in early Mesopotamia.
First, the terminology, much like the evidence of the practice, could serve multiple purposes. So debtors and murderers could be held in the e2-eš2, or prisoners of war could work alongside slaves or criminals (nepārātum). Detention, as such was practical and not tied to a singular function. This cautions against the imposition of modern categories when thinking of the functionality of imprisonment in early Mesopotamia.
Secondly, the overwhelming impression given by the evidence is that detention was used in administrative contexts to coerce and gain access to labor. Existing functioning bodies, such as guards and workhouses could be used to meet various needs throughout the judicial process, as seen below, but the primary function of these administrative bodies appears to have very little to do with “criminal” activity. Even when dealing with runaways, this appears to be related to the need to control movement in order to gain access to labor, rather than a punishment.
Thirdly, when imprisonment in relation to “crime” does occur, this is mostly in relation to the judicial process or to coerce payment and job performance. If imprisonment for the purpose of punishment ever did exist in ancient Mesopotamia, this was surely not the case normally and such evidence remains unavailable to us. Imprisonment was utilized as mechanism of control until punishment rather than a mechanism of punishment, it seems.
This provides an important contrast between the picture offered in the “Hymn to Nungal” and the evidence of everyday life. This contrast will be kept in view throughout the rest of this work. As such, while there is a return to the prison in the “Hymn to Nungal” for the purpose of social cleansing, this does not appear to be the normal function of imprisonment in the documents of practice. As will be seen below, this return to the prison after guilt was established through the River Ordeal in the “Hymn to Nungal” appears to be more about purification by imprisonment through ritual lament.
Prisons in Ancient Mesopotamia: Confinement and Control until the First Fall of Babylon. J. Nicholas Reid, Oxford University Press. © J. Nicholas Reid 2022. DOI: 10.1093/oso/9780192849618.003.0003
1 See, in particular, Steinkeller 1991.
2 See Hansen 2002: 101 (figures 1–3) and see Ornan 2007: 59–72 for the Assyrian evidence.
3 See these signs in Green and Nissen (1987) = ZATU Sign No. 439, 267. Note too the term šaga (LU2+EŠ2 = man plus rope) in Steinkeller 2013.
4 See interpretation in Englund 2009 §§6.1–6.4. For the Ur III period, many of the names of the prisoners of war in the text TCL 5, 6039, discussed below, have foreign names. It should be noted that most prisoners of war in later periods clearly come from the people who shared cultural and linguistic similarities. For the Old Babylonian period, see Rositani 2020: 202, who writes, “In the Old Babylonian texts we have pieces of information indicating that the prisoners were often taken from people ethno-linguistically closely reated to their captors.”
5 See Steinkeller (1991) dealing with the “Reforms of UruKAgina” and early terminology for “prison” (e2-eš2).
6 Cited after Frayne 1998: 264–65.
7 See, from Girsu CT 50, 35 (Sollberger 1972): Obverse i:8ʹ, iv:4ʹ: v:10ʹ; DP 116 (Allotte de la Fuÿe 1912): Obverse iv:16–17, vi:4–5; DP 117 ( de la Fuÿe 1912): Obverse i:12, iv:10–11, v:17; DP 118 (Allotte de la Fuÿe 1912): Obverse i:12 (a line reconstructed on the basis of the previous text DP 117, Allotte de la Fuÿe 1912 that dates to UruKAgina Year 1 Month 4), iv:16, vi:2; Nik 1, 20 (Mikhail V. Nikol’skij 1908) = FAOS 15/1, 20 (Gebhard Selz 1989): Obverse iv:11–12, vii:13. Each of these examples from Girsu belong to the reign of UruKAgina and is written in an administrative context concerned with recording rations.
8 See Thureau-Dangin 1910 and Steinkeller 1991: 228–30. ITT 1, 1287, discussed in Chapter 4, mentions five men who were brought in by armed guards and living in prison.
9 CUSAS 11: 265 (Giuseppe Visicato and Aage Westenholz) Obverse ii: 4–9: sag ad-da-še3 / e2-˹eš2˺-ka / i3-ti-am6 / GAR.DU 1(diš) ku3 gin2 / ur-gu / i3-la2: “…for the person (slave?) of Adda, who is in prison, Ur-gu paid the deficit of 1 shekel of silver…” Visicato and Westenholz translate the verb i3-la as “paid the bail.” Given the use of the terminology in relation to balanced accounts, together with the known instances of the prison being used as a place for certain types of debtors, it is perhaps best to understand the terms as “paid the deficit” or “paid the amount owed.” However, since the amount taken away is more than the figure paid by Ur-gu, it is possible that the amount is a type of bail or guaranty.
10 Maiocchi and Visicato (2020: 307–8) compare this text with ITT 1, 1287, cited in Chapter 4.
11 Bartash 2020: 44.
12 See discussion of the Sumerian term for children in Bartash 2018a. I would also like to thank Bartash for pointing me to this text and indicating some of its significance. See Bartash forthcoming.
13 See Reid and Spada 2020: 74.
14 The text BPOA, 7, 2125 (Sigrist and Ozaki 2009) from Umma Obverse lines 1–2 state that eight ĝuruš worked for eight days to build a prison—8(diš) ĝuruš u4 8(diš)-še3 / e2-eš2 du3-a. The terminology is also known from Girsu, such as MVN 13, 382 (Sigrist, Owen, and Young 1984), which mention beer provisions for the prison, likely referring to the provisions for the guard. See similar note in Reid 2016: 87 n. 23.
15 See Seri 2013. For a discussion of the etymology and survey of previous literature, see Seri 2013: 7–10, in particular. See more recently, Rositani 2018: 43 and n. 3.
16 See Rositani 2018: 42–43 and Seri 2013: 1–17.
17 Rositani 2018: 46.
18 Rositani 2018: 48–49.
19 Leemans 1961.
20 Rositani 2018: 55–59.
21 See Renger 1972. For flight and the related consequences during the Ur III period, see Reid 2015. For a comparative study of flight for the entire early Mesopotamian period, see Reid 2014.
22 Nisaba 4 II: 28, Rositani 2003: 136–37, see collation of first line in Seri 2013: 139 n. 73.
23 Nisaba 4 II: 66, Rositani 2003: 169–70.
24 See Seri 2013: 110–41. The bīt ṣibittim in Mari is a comparable institution (see Scouflaire 1987).
25 Civil 1993: 75.
26 See, for example, AnOr 7, 293; AnOr 7, 324; BPOA 2, 2539; BPOA 6, 1113; BPOA 7, 2153; etc.
27 While the term ĝuruš from the Old Babylonian period onwards was equated with the Akkadian eṭlum, or “a young, able-bodied man,” the Sumerian word relates to “full output workers disregarding age” (Bartash 2018a: 16). On the terminology and status of ĝuruš, see, for example, Gelb 1965: 240–41; Gelb 1976: 195–207; Gelb 1979: 283–97; Struve 1969: 127–72 (English version of Struve 1947). Postgate compares the muškēnum of the Old Babylonian Period to the ĝuruš of the Ur III period and claims, “the best guess at present sees them as semi-free, tied to the land by the obligations of service to the palaces, but this need not have ruled out access to private sources of income…” (Postgate 1992: 239).
28 Only the final year of a range is indicated here, since that is when the record on the tablet would be completed. This is true also when years span two rulers. In those instances, only the final ruler is indicated in this table.
29 BIN 3, 491 iii: 55 certifies that a guard received four cows. Keiser 1971.
30 Fish 1932.
31 This text deals with a runaway living in prison. Sigrist 1983.
32 This text deals with a seized runaway living in prison during Šulgi 46. Yildiz, Waetzoldt, and Renner 1988.
33 Oberhuber 1958–1960. This text has one ĝuruš living in the house of the prison.
34 BIN 5, 307 details the barley rations for the prison at Umma. Hackman 1937.
35 At least two people are mentioned living in prison in this text. Van Dijk 1962.
36 Sigrist 1990.
37 Sigrist and Ozaki 2009a.
38 This text also mentions a gardener (šandana).
39 Cagni and Pettinato 1976.
40 Gomi and Yildiz 1997.
41 TCL 5, 5675 Reverse v.13 has a2-bi u4 2(geš2). This text deals with labor assignments and workdays. Genouillac 1922.
42 D’Agostino 1997.
43 Sigrist 1983.
44 Mander 1995. Heimpel (1998: 392–94) identifies these as “captives living in prison” (lu2-dab5-ba ša3 en-nu) and “captives stationed at the woodshed” (gan-dab5 ĝa2-nun-geš-ka gub-ba). On the ĝanun particularly related to Garšana, see Heimpel 2009a: 166–67. Note too TCTI 2, 3563 (Lafont and Yildiz 1996) and TUT 261 (Reisner 1901) for other examples of texts dealing with rations for the prison.
45 Mander 1995.
46 Heimpel (1998: 392) identifies the en-nu gu-la as the “big prison.”
47 Uchitel 1992.
48 Mander 1994.
49 Sigrist 1993.
50 Sigrist 1993.
51 Uchitel 1992.
52 Barton 1905–1914. See Mander 1994: 26 (Kennelmen 36).
53 Mander 1995.
54 Lau 1906 = CUSAS 16, 37 (Garfinkle, Sauren, and van de Mieroop 2010).
55 Boson 1936. After each entry, the number of days are counted—a2-bi u4 n(geš2)-am3. The first two entries each relate to 6 months or 180 days. Note, however, that for the second entry there are three winkelhaken present on the tablet and a faint diagonal wedge, which seems to be a partial sign. It seems that the scribe wrote for the second entry, Obv. Line 10: a2-bi u4 3(geš2)-am, rather than the anticipated am3 (only a portion of the -am is visible on the tablet in its current state of preservation). The calculation of 180 days or 6 months supports this view over against the reading a2-bi u4 3(geš2) + 3(u). This view would leave the sign remains unexplained and bring the calculation to seven months. Although an intercalated month is attested for AS 4 at Umma, including a thirteenth month for the final entry would mess up the calculation in relation to the third entry. The third entry after collation by photo should be read as a2-bi u4 5(geš2)-am3 rather than following the handcopy by Boson (1936), which has 3(geš2). This reading is confirmed by the 10-month period that comes to 300 days. The sentence a2-bi u4 n(geš2)-am3 may be translated as “the (literally: its) wages are n days.” On this terminology, see Heimpel 2009a: 81–90. The text deals with three workers living in prison.
56 Grégoire 1996.
57 Unpublished.
58 Yildiz and Gomi 1993.
59 Yildiz and Ozaki 2001.
60 The colon in nun:ĝa2 indicates that the signs are inverted on the tablet. See also CST 556 (Fish 1932).
61 Molina 1993.
62 Waetzoldt and Yildiz 1994.
63 Two workers are listed as living in prison on this tablet. Waetzoldt and Yildiz 1994.
64 Sigrist and Ozaki 2009b.
65 Sigrist and Ozaki 2009b.
66 Gomi and Sato 1990. The second and third entries are half-production persons, who are listed in relation to their fathers.
67 Gomi and Yildiz 1997.
68 Gomi and Yildiz 1997.
69 Waetzoldt and Yildiz 1994.
70 Schneider 1932.
71 CST 645 (Fish 1932) like the above text BIN 5, 307 relates to barley rations for the prison.
72 Sigrist and Ozaki 2009b.
73 DAS 206 (Lafont 1985) deals with the provisions for farmers and prisoners. This text deals with seven prisoners. Among these prisoners there is a priest, some sort of a professional (du3-a-ku5), and a sailor. The prisoners in this text receive one liter of bread per day. See discussion in Heimpel 1998: 392.
74 Sigirst 1983.
75 Nikol’skij 1915.
76 Yildiz, Waetzoldt, and Renner 1988.
77 Sigrist and Ozaki 2009b.
78 Waetzoldt and Yildiz 1994. The person living in prison was at half-production.
79 Two workers are listed as living in prison on this tablet. Waetzoldt and Yildiz 1994.
80 Yildiz and Gomi 1993.
81 Three workers are listed as living in prison on this tablet. Waetzoldt and Yildiz 1994.
82 Alivernini and Greco 2019.
83 Gomi and Yildiz 1997.
84 Lutz 1928.
85 Yildiz, Waetzoldt, and Renner 1988.
86 Sigrist 1990a.
87 Schneider 1932.
88 Gomi and Yildiz 1997.
89 Ozaki and Sigrist 2006.
90 Ozaki and Sigrist 2006.
91 Yildiz and Ozaki 2000.
92 Sigrist 1990.
93 Gomi and Yildiz 1997.
94 Yildiz and Gomi 1993.
95 Schneider 1932.
96 Waetzoldt and Yildiz 1994.
97 Yildiz and Ozaki 2001.
98 In this instance, the prisoner is a slave of a farmer. Keiser 1919.
99 UTI 6, 3752 accounts for the expenditure of “worker days” by the guards. Yildiz and Ozaki 2001.
100 Sigrist and Ozaki 2009a.
101 Unpublished.
102 Fish 1932.
103 Yildiz, Waetzoldt, and Renner 1988.
104 Ozaki and Sigrist 2006.
105 Sigrist 2000a.
106 D’Agostino 1997.
107 Waetzoldt and Yildiz 1994.
108 AnOr 7, 181 deals with a three shekels of silver deficit owed by a fisherman living in prison. See Schneider 1932.
109 CDLJ 2009/2 §2.7 records clothing rations for those living in prison. The phrase “under armed guard” (tukul-e dab5-ba-a) can also be read “seized with weapons.” In the context, the reading “under armed guard” seems preferable. For this terminology in relation to prisoners, see also MVN 7, 275 and Waetzoldt 1972: 79, 88. See further Reid 2015: 597–98 and 598 n. 54. See also discussion in Robson and Clark 2009/2 §2.7.
110 Koslova 2000.
111 Sigrist and Ozaki 2009a. BPOA 6, 874 deals with twenty-three ĝuruš stationed at the prison for one day.
112 AnOr 7, 286 (Schneider 1932) documents the imprisonment of five runaways.
113 Dahl and Hebenstreit 2007.
114 Ozaki and Sigrist 2006.
115 Yildiz and Ozaki 2000. This text deals with rations.
116 Myrhman 1910.
117 Al-Rawi and Verderame 2009.
118 This is in reference to four ĝuruš as seen in the subtotal on Obverse column i, 18.
119 Owen 1975.
120 Dahl 2019.
121 Sigrist 1990b.
122 This text deals with two runaways. See note below for discussion. The text says, “let both of them live in prison.” Szlechter 1963.
123 This text deals with flour rations for guards. Watson and Horowitz 1993.
124 This text mentions the imprisonment of a person for a false accusation of robbery. See original edition by Ozaki and Sigrist (2006). M. Molina (2008) collation. T. Sharlach studied the text in 2009. Most recently, Civil (2011: 254–74) discusses this text with a partial translation and transliteration.
125 Ozaki and Sigrist 2006.
126 Note the half-production worker also found in MVN 16, 1578.
127 Not everything in Mesopotamia was written down. Socio-economic documents were written when certain conditions gave rise to the need to document transactions, see Van De Mieroop 1997a: 7–18 and more summarily in Van De Mieroop 1997b: 301–2. See also Steinkeller 2004: 95–96. These conditions had a direct impact on the content included in the documents themselves. For this reason, these socio-economic texts that provide information for the modern scholar about prisons were not written for that purpose. These texts primarily recorded financial transactions that could be contested and might need to be proven.
128 Heimpel 1998: 392.
129 Heimpel 1998: 392.
130 Kramer 1977: 65. Hallo 1979: 162 n. 17.
131 Goetze 1947. Renger (1977: 77) notes that both ṣibittum and kīlum are nomina actionis that do not have the ma-prefix used for localities.
132 Goetze 1947 plate XCIX. See also YOS 10, 11 ii: 30 (Goetze 1947 plate VIII).
133 For AbB 2, 121, see Frankena 1966: 82–83. See handcopy in King 1910, pl. 2. Transliteration and translation can also be found in Ungnad 1914, text 100. For AbB 6, 144, see Frankena 1974: 92–93. See handcopy in Schroeder 1917. For AbB 11, 60, see Stol 1986: 34–35. On the text ARM 26, 524, which mentions the ṣibittum, see edition in Lafont 1988: 500–502 (in particular 501 n. h); see too translation in Heimpel 2003: 402–3. This text is cited and discussed below. On the bīt ṣibittim at Mari, see Scouflaire 1987.
134 Lafont and Westbrook 2003: 221.
135 Bartash 2015.
136 Bartash 2017: 90–91. Bartash 2018b: 268.
137 Bartash 2018b.
138 See CAD K: 450–51.
139 As discussed above, Renger (1977: 77) states: “It is unlikely that terms like bīt asīrum [sic], bīt kīlim, or bīt ṣibittim will introduce the notion of prison or imprisonment as a form of punishment through the backdoor.”
140 See Kraus 1959–1962: 25–29. See CAD nupāru. At Mari, see further Scouflaire 1987, 1989. See BM 80448 = CT 6 32, Pinches 1898. For transliteration and translation, see Kraus 1959–1962: 27–28 and Frankena 1966 (AbB 2, 114). See also, the Old Babylonian texts, AbB 14, 128 (Veenhof 2005: 118–19); Kraus 1959–1962: 29 no. 6. This school text mentions a wife, the sons, and slave girls being placed in jail. The letter calls for the recipient of the letter to come quickly to get them out of jail. The cause behind the confinement is unstated, though this seems to fit the treatment of distrainees. Note the distinction between the nepārātum and bīt ṣibitti at Mari in Scouflaire 1989: 157–60 and Durand 2000: 250. See further discussion in Maul 1997: 762–63, who describes the nepārātum as workhouses.
141 See Lafont 2016: 161.
142 Durand 2000: 250.
143 Lacambre and Patrier 2016: 178.
144 Durand 2000: 250.
145 See discussions in Lafont 2016: 161 and Renger 1977: 77 n. 34.
146 Durand 2000: 67–68, 250.
147 Durand 2000: 67–68, 250.
148 Lafont 2016: 161.
149 Bartash 2018b: 271.
150 Sasson 2015: 223–24; Sasson 1977: 104–5.
151 See, for example, ARM 14, 78 (Birot 1974 and Durand 2000 no. 929, LAPO 18). See too the translation in Sasson 2015: 226–27. Note further discussion below. In this text, the punishments were to humiliate and then utilize their labor. This was in lieu of selling them to distant tribesmen, never to be heard of again.
152 Lafont connects the practice of utilizing of prisoners of war in the Ur III period to the practice of utilizing prisoners of war in workhouses at Mari. Lafont 2016: 161.
153 Dahl 2003: 59–61.
154 See Yidiz, Waetzoldt, and Renner 1988. MVN 14, 569 (Umma; Šu-Suen Year 3 Month 6; Ist Um 569).
155 Dahl 2003: 59–61.
156 SAT 2, 1163 (Puzriš-Dagan; Amar-Suen Year 9, Month 8; YBC 128, Sigrist 2000b).
157 The term denotes a full output worker. See discussion and bibliography in Dahl 2020: 169 n. 385.
158 Gelb’s article on prisoners of war offers a prosopography of these texts (1973: 74–76).
159 This summary is based on my collation of this tablet (Reid 2020).
160 One further individual should be included among the dead from the damaged bottom lefthand corner of the tablet, since only one of the two individuals is included in the subtotal (12) and again in one of the grand totals. See Reid 2020.
161 It is possible there are only twenty-three sick individuals. It depends on whether tu is included in part of the name 4(ban2) tu-mi-na-a-u2 or as in indication of health 4(ban2) tu mi-na-a-u2 (Rev. ii:11). The sign tu, for sick, can appear before (Obv. iii:3–4) or after (Obv. i:1) the rations in this text. There is also one additional sick person not shown on the original handcopy. The final line of Obv. iii begins with the sign še in the handcopy but should be a tu.
162 Steinkeller (2013b: 354, 357) writes: “The land was situated in the rural settlement of Lullubu(na) (in the province of Girsu/Lagaš), which, as its name indicates, must have been settled with the prisoners-of-war from Lullubum. […] While a majority of these individuals seem to have come from Northern Babylonia, a very significant number of them were foreigners, quite often prisoners-of-war. Such undoubtedly was the background of the settlements of Lullubu(na), Ebiḫ, Eduru-Elamene, and Šimurrum, all in the province of Girsu/Lagaš.” Perhaps TCL 5, 6039 represents one such settlement that was being worked in Umma.
163 Lafont 2016: 161–62.
164 Lafont 2016: 161–62.
165 Bartash 2018b: 277. See Richardson 2012: 8, 11; Steinkeller 2015: 5–6.
166 Hallo 1979: 161–65. See also Civil 2003: 84–85.
167 Hallo 1979: 162.
168 See Hallo’s discussion of NSG 1, 140 n. 3 (Falkenstein 1956) in Hallo 1979: 163–64. The text is dealing with an accusation about diverting water.
169 Hallo 1979: 162–63.
170 Hallo 1979: 163.
171 On this proverb and its interpretation, see Alster 1997: 147; Hallo 1979: 161–66; and Civil 2003: 84–85. See also, Stadhouders and Panayotov who render the lines as follows: “The é.gal is a thicket, the king a lion: Ninegal (var. Nungal) is a net large enough to enmesh many men at a time” (Stadhouders and Panayotov 2018: 636 n. 15).
172 For “Bilgameš and Ḫuwawa A,” see Edzard 1990: 165–203; 1991: 165–233. On the Sumerian name, Bilgameš, see George 2003: 71–90.
173 Civil 2003: 85.