Introduction
Recent scholarship has revived interest in how households in early Mesopotamia obtained and utilised foreign-born slaves. Steinkeller (Reference Steinkeller2013a: 143–144; Reference Steinkeller, Steinkeller and Hudson2015a: 7–8) argued that the outright slaves in Sumerian temples and royal manors were largely prisoners of war. Women and children toiled in textile workshops while blinded men irrigated orchards. Bartash (Reference Bartash2018) investigated the massive influx of forced labour from northern Mesopotamia (Subartu) to Babylonia during the military expansion of the Akkadian Kingdom in the 23rd century B.C. Rositani (Reference Rositani2018; Reference Rositani2020; Reference Rositani, Notizia, Rositani and Verderame2021) demonstrated that Old Babylonian war captives, many of whom were foreigners, belonged to the ruler. He donated some of them to temples. The rest toiled in royal workshops, on public projects and in the households of palace affiliates. Richardson (Reference Richardson2020) challenged the traditional war versus trade dichotomy. Foreign-born slaves in late Old Babylonian private households came neither from the regions where kings waged wars nor from areas frequented by Babylonian merchants. War, inequality and precarity produced slaves further away, on the periphery, and merchants then brought them to trade hubs, where Babylonians could buy them. Finally, Bartash (Reference Bartash2020) demonstrated how the royal house of the Early Dynastic city-state of Adab used slaves as gifts to maintain relations with local, regional and even Iranian political elites.
The following analysis of Gudea's slave dossier allows all these topics to be addressed at once. The dossier describes a brief inflow of Iranian captives and slaves to the royal household at Girsu, the capital of Lagash on the lower Tigris. This event took place during the reign of Gudea (ca. 2130–2110 B.C.), who invaded Elam to defeat the powerful Anšan state. Although Mesopotamian rulers brought Iranian captives home before and after Gudea, several features make Gudea's slave dossier a unique source for a case study on slave mobility.
First, the slave dossier outlines various strategies that the palace at Girsu used to replenish its supply of foreign forced labour. Surprisingly, it records a single male slave brought by Lagash troops on their way home from Iran. Dozens of Iranian women and children who produced textiles in the households of royal children were probably captives too, but the scribes did not record their delivery.
Another way to obtain foreign slaves was to buy them, a practice that scholars usually link to private households. Gudea's slave dossier shows that the palace did the same. In one instance, the palace resorted to the help of proxies to buy an Iranian man from abroad. In another, a cupbearer, a person very close to the throne, sold his Iranian slave woman to the ruling house. Comparable evidence from other periods confirms that royals routinely bought local and foreign slaves.
The final, unique conduit of Iranian slaves to the palace at Girsu was mašdaria, or “kids led by one's side”. The traditional mašdaria was a tributary delivery of food by local elites to the ruler for cultic feasts, an occasional gift to the royal family, or a votive donation to a god. Two local men and three Iranian polities gave Iranian slaves as “kids led by one's side” in Gudea's slave dossier. The analysis of Gudea's Iranian campaign suggests that these deliveries came from regions affected by his war with Anšan. The scribes classified them as tributary deliveries of animal young to stress the political dependency of the givers upon the Lagash throne.
Second, it is apparent that Iranian slaves were plentiful in royal and private households in Lagash during Gudea's reign. Only the elite and the military received captive Iranians as gifts. Common private households bought Iranians locally from them or from abroad. This paper applies Richardson's model: Gudea's war in Iran created political havoc and social instability there, which stimulated the long-distance slave trade.
Finally, the analysis identified two demographic patterns of foreign slaves. Iranian slaves in royal workshops and in private households were principally women and children. In contrast, the Iranians who arrived as “kids led by one's side” were few, young, of both sexes, and not burdened by offspring. They were luxury gifts rather than rough labour.
Dissecting the forms of slave mobility in the dossier reveals four common elements: slaves, the agents of their mobility, the physical and social spaces in which they moved, and the economic and political forces that pushed slaves from one agent/space to another. De jure immobile and often physically immobilised, slaves moved like an electric current from plus to minus, giver to receiver. Their mobility took place not only between two persons but between their respective spaces, as slaveholders had membership in the palace, temples or private households and integrated their new chattels accordingly.
The twin forces of economy and politics drove slave mobility. Scholarship has traditionally focused on the value of slaves as cheap labour. This paper draws attention to the political significance of giving and receiving slaves. Gifts and donations of slaves elsewhere, the Iranian “kids led by one's side” and the sale of an Elamite woman by a cupbearer to the ruler in Gudea's slave dossier were all political acts, mechanisms to maintain power hierarchies between individuals, households and states on the local and inter-regional level.
This paper expands our knowledge of foreign-born slaves in the early Near East in two key respects. First, the palace was quite creative in obtaining them. Taking captives was probably the boldest strategy, but it was not the only one. Second, foreign slaves were not just cheap labour or attractive servants but political instruments in the hands of their givers and receivers. Local and inter-regional politics should thus be added to the “why?” of slavery in ancient societies.
The usage of “slave” and “Iranian” in this article requires explanation. The former were people whom sources identify as “heads” (Sumerian sag), that is, non-persons. They belonged to three legal and socio-economic groups: chattel slaves in private households, captives whom rulers donated to temples and those prisoners of war who remained ruler's property. The legal and socio-economic status of the latter two groups was more akin to the “serfs” of social historians. Depending on circumstances, a person could move between all three social groups and their respective master-households.
The ethno-linguistic situation in Iran during the third millennium B.C. was complex (Desset Reference Desset, Dietrich and Kämmerer2017; Potts Reference Potts2016: 1–12; Zadok Reference Zadok, Álvarez-Mon, Basello and Wicks2018). Numerous peoples who spoke different, unrelated languages lived there. The mountainous regions in Iran's south-west and their populations appear under the designation NIM “upper (land/people)” in third-millennium sources. Scholars often interpret this sign as “Elam” based on later evidence. Consequently, the dominant language in Iran's south-west is called “Elamite” (Krebernik Reference Krebernik, Kogan, Koslova, Loesov and Tishchenko2006: 60). The article deploys a less binding term “Iranian” when referring to captives and slaves in Gudea's slave dossier. These people were “Iranians” for three reasons. First, the sources identify them as “Elamites” (NIM), and Elam was situated in modern-day Iran. Second, some slaves arrived from “core” Elamite territories, such as Huhnuri. Third, numerous elements in the personal names of the captives and slaves belong to the Elamite language. Finally, the historical analysis below confirms that Gudea obtained his captives and slaves in Iran's south-west.
The following sections first set forth the historical background of Gudea's slave dossier then consider the dossier's general characteristics. The core sections analyse the mobility of Iranian slaves in the context of local and transregional historical events. The subsequent synthesis relates the forms of mobility to each other and extracts their basic elements: slaves, agents, spaces and economic and political dynamics.
Historical background
The chronology of the so-called “Gutian period” in Mesopotamia (ca. 2180–2111 B.C.) remains problematic.Footnote 1 Handbooks often omit the Gutian period or regard it as a dark age. However, the cardinal political shifts and massive human mobility of that time affected the entire Middle East. The earliest transregional state in western Asia, the Akkadian kingdom (ca. 2324–2181 B.C.), collapsed; Gutian tribes from the Zagros Mountains invaded Babylonia and created a state they could not defend; Puzur-Inšušinak of Susa founded a massive but ephemeral Iranian state; and the last Sumerian hegemon with a fixation on Iran, the Ur III kingdom (2110–2003 B.C.), emerged.
In Steinkeller's (Reference Steinkeller, Molina and Garfinkle2013b) historical model, the story began in Iran. Puzur-Inšušinak started as ruler of Susa (Fig. 1). He vanquished about a hundred small polities in the Zagros Mountains before subjugating his main targets, Kimaš and Hurti. These lands controlled the Great Khurasan Road, the principal trade route between inner Iran and Mesopotamia. Puzur-Inšušinak then used that road to invade the Diyala region and northern Babylonia, the latter having been part of the Gutian kingdom. Steinkeller suggests that the resulting damage to the Gutians allowed a king of Uruk, Utu-hengal, to strike them from the south and to extinguish their Babylonian dynasty in 2111 B.C. Utu-hengal's general, Ur-Namma, united Sumer, liberated northern Babylonia from Puzur-Inšušinak and founded the Ur III kingdom the next year. On the eastern front, Ur-Namma pillaged Susa (Marchesi Reference Marchesi, De Graef and Tavernier2013). Ur-Namma's son and successor, Šulgi (2092–2045 B.C.), fought for over thirty years in the north and defeated Kimaš and Hurti. Steinkeller regards this as the reversal of Puzur-Inšušinak's accomplishments. The Ur III kingdom now controlled the main trade route with inner Iran.
Fig. 1 Places mentioned in the article (map by the author)
Rather than these dashing conquerors, the main character here is Gudea of Lagash (ca. 2130–2110 B.C.) with his reputation as a pious conservative (Fig. 2). The prosopography of the Lagash archives suggests that the inflow of Iranian slaves took place during his reign (Lehmann Reference Lehmann2016: 230–234). Lagash had regained its independence some fifty years earlier when the Akkadian Kingdom began to crumble. Local governors known as the Second Dynasty of Lagash (or Lagash II dynasty) made it great again, for the last time. Lagash gravitated towards the kingdom of Uruk during the Gutian period (Sallaberger Reference Sallaberger, Sallaberger and Westenholz1999: 133). Gudea is the last ruler named in the Lagash King List, although he had several successors (Sallaberger & Schrakamp Reference Sallaberger, Schrakamp, Sallaberger and Schrakamp2015: 120). The growing Ur III kingdom incorporated his state as a province. The date and reasons for this annexation remain enigmatic. The Ur III state venerated Gudea as a god (Suter Reference Suter and May2012: 60–63). Gudea must have performed a significant favour for the nascent Sumerian hegemon. Steinkeller (Reference Steinkeller, De Graef and Tavernier2013c) argues that he helped Ur-Namma to defeat Puzur-Inšušinak. Yet Gudea mentions neither of them in his inscriptions.
Fig. 2 Head of Gudea, ruler of Lagash, late 22nd c. B.C. The Metropolitan Museum of Art, New York. Photo in the public domain. (https://www.metmuseum.org/art/collection/search/324061)
Gudea's foreign policies focused on south-western Iran. He restored a temple in Adamsul (AdamDUN) and campaigned against Anšan and other, unspecified Elamite city-states.Footnote 2 Sallaberger and Schrakamp (Reference Schrakamp, Sallaberger and Schrakamp2015: 124–125) argue that Gudea's Elamite war had happened before Ur-Namma conquered Susa. I adopt their chronological model, which proposes that Gudea's and Ur-Namma's reigns overlapped only marginally (ibid.: 112).
Approaching Gudea's slave dossier
Nine texts capture the influx of fifty-five Iranian slaves to the palace at Girsu. Twenty-seven more Elamites (NIM) appear as weavers in the household of the royal children (e2-nam-dumu) in three other texts. All twelve texts belong to the palace archive of Girsu, which totals some six hundred documents (Maiocchi & Visicato Reference Maiocchi and Visicato2020).
The nine core texts specify up to four general transactions related to slaves (Tab. 1, I). Most of them were deliveries of slaves.Footnote 3 The administration weighed out silver for them, except when soldiers brought a slave. The scribes never disclose who received this silver. Three documents identify the “conveyors” of slaves. These were mounted couriers (ra2-gaba) in two cases, which makes sense given that their task was to transport slaves. One of the couriers, Bazi, may be identical with Bazige in no. 252. The palace received three slaves from him. These slaves did not belong to Bazi in this scenario. He was merely a palace proxy leading them from A to B. Finally, palace officials took slaves into custody upon delivery. A certain Mes'e “seized” them in all three cases. The verb refers to the taking hold of or transfer of labour (Dahl Reference Dahl2020: 158–159; Steinkeller Reference Steinkeller, Steinkeller and Hudson2015b: 138–139).
Table 1 Transactions in Gudea's slave dossier
Gudea's slave dossier has three types of data. The first dataset consists of the number, personal names, age and gender of slaves. The names, titles and origins of slave-givers are the second set of information. The emic designations of transactions provide the final element (Tab. 2).
Table 2 Data on slaves, slave-givers and transactions
1 Maiocchi & Visicato's (Reference Maiocchi and Visicato2020) interpretation assumes that he was a son of a certain Ir3-ša3-HI. This personal name does not appear elsewhere though.
A comparison of slaves’ number, age and gender with the ways the palace acquired them reveals two patterns. The slaves delivered as “kids led by one's side”, bought and delivered by the troops, were childless adults. Their number never exceeded five slaves per transaction, and was usually limited to one or two. By contrast, palace workshops employed numerous women and children. Private households had up to five women and children (Fig. 3).
Fig. 3 Age and gender of slaves in each type of transaction
The majority of slaves bore personal names with Elamite elements.Footnote 4 However, names alone are a poor guide to their origins. Names with the same or similar elements appear as far away as Kimaš and Šimaški in the Western Zagros and on the Iranian plateau respectively.Footnote 5 Huhnuri is the only foreign locality in Gudea's slave dossier that can be safely localised in “Elam proper” (Khuzestan and Fars).
“Kids led by one's side” from locals
Half of all deliveries of Iranians to the palace at Girsu were mašdaria, “kids led by one's side”. This expression referred to a special delivery to the royal family, which typically consisted of a kid or a lamb in Early Dynastic Girsu (Selz Reference Selz1995). Other goods included bread, beer, vegetables and fish. Non-eatables were rare. The elite and representatives of professional groups brought these goods to the palace or the temple of Baba governed by the queen. As a regular tributary delivery, the mašdaria allowed the ruler to provide for major cultic festivals. Alternatively, the scribes also classified the occasional gifts of the elite to the royals as mašdaria. For example, the wives of several VIPs brought sheep to the palace after the queen gave birth. Mašdaria's dual nature led Prentice (Reference Prentice2010: 187–195) to define it as a forced gift. The mašdaria lost every trace of gift by the Ur III period (21st century B.C.) and became a tax that allowed the king to fund three major feasts in the capital city. Participants in the feasts ate, drank and made offerings with mašdaria-goods (Sallaberger Reference Sallaberger1993: 160–170). A document from the last year of Ur III king Šulgi (2045 B.C.) offers a notable exception: A soldier of general Hubaya delivered two oxen that were part of a mašdaria brought from a military campaign.Footnote 6 Here mašdaria seems to be a synonym of “booty”.
The Lagash II corpus exhibits three functions of the mašdaria. The first, the classical tributary delivery of animals and objects, is surprisingly rare.Footnote 7 The votive gift is the second variety. A ruler's wife dedicated a stone statue of a human-headed bull as a mašdaria to the goddess Baba (Edzard Reference Edzard1997: 10). The deliveries of Elamite slaves as “kids led by one's side” described in Gudea's slave dossier represent the third form. Lagash scribes classified in this manner the deliveries of Iranian slaves by locals and by foreign rulers alike.
Šudalibi and Sipa'uruna were the locals who delivered Iranians to the palace as mašdaria. They contributed five and one Elamite slave respectively, all of whom were adult males. Judging by their names, both givers were locals, but neither appears elsewhere in the corpus. They may not have belonged to the elite, the traditional mašdaria-givers. The texts do not specify why they gave the Elamites to the palace either. Gelb (1972: 10) drew an analogy between the mašdaria and the donation of unwanted members of private households to temples. However, temples do not feature in Gudea's slave dossier. The case of Šudalibi and Sipa'uruna is reminiscent of the mašdaria in the context of booty in the Šulgi's document mentioned above.
In contrast to this obscure case, the following section reveals that war and diplomacy induced Iranian rulers to send slaves to Lagash as “kids led by one's side”.
Taking captives and “kids led by one's side” from Iranian polities
Two Elamite rulers and another locality sent Iranian slaves as mašdaria to Lagash (nos. 252–253). More accurately, Girsu scribes classified them as tributary deliveries. In so doing they sought to stress that the Iranian senders of slaves were subordinates of the Lagash crown. It was as though these Iranian polities were part of the Lagash state. Their rulers took the place of the Lagash elite who were required to contribute mašdaria-goods to the ruler.
Simhuzi, a ruler of Huhnuri, sent five women (no. 253). It is the largest mašdaria in Gudea's slave dossier. Huhnuri/Huhunri/Hunnir was strategically situated on the road to Anšan.Footnote 8 A city-state of astonishing longevity, it lasted well into the first millennium B.C.
A ruler of Huhnuri, probably the same Simhuzi, appears in the context of a slave woman in another document of Gudea's slave dossier. She either came from his palace or was intended for Gudea's palace.Footnote 9 The same text mentions another slave woman for/from a Lagashite emissary (sukkal). A messenger and a soldier from Marhaši, a powerful state in Iran's southeast, conclude the list of persons.Footnote 10 This document highlights the connection between slave mobility and diplomacy with Iranian states.
Gimani, the other mašdaria-giver in no. 253, provided a single male slave.Footnote 11 This and some other Lagash II texts do not deploy the determinative for geographical locations, so it is hard to tell whether this Gimani was a man or a location. A text from the Diyala region records a similar name.Footnote 12 Alternatively, Gimani could be a locality in Elam. A text from Sargonic Susa mentions a locality Zimani.Footnote 13 If so, G/Zimani lay not too far away from Huhnuri, the hot spot of Gudea's Iranian campaign.
“The troops” (um-ma-num2) gave the last slave in no. 253 as a simple delivery, not a mašdaria. It is the only explicit testimony that Lagash took captives in Iran. Also curious is the use of the rare word um-ma-num2, a synonym for the Gutians.Footnote 14 The buffer state of Umma separated Lagash from the Gutian Babylonian kingdom. We do not know Gudea's attitude towards these invaders, as his inscriptions are suspiciously silent on the matter.
Finally, an anonymous ruler of ElaNIR, probably to be read Elašer, sent a slave boy as a mašdaria to Lagash (no. 252).Footnote 15 Nos. 173–175 record barley distributions to seven women with two children from Elašer. Probably captives, they joined an existing troop of Elamite (NIM) weavers in the household of the royal offspring. Elašer is the only locality in Gudea's slave dossier where Lagash troops took captives and whose ruler was forced to send a mašdaria-slave.
Elašer does not appear outside the Lagash II corpus and is hard to localise. Nos. 173–175 leave no doubt that the scribes considered Elašer a part of Elam. Elašer was likely Huhnuri's neighbour. Two men from Huhnuri and Elašer and a ruler of an otherwise unknown polity, Urna (ensi2 Ur2-naki), receive fat in no. 216. However, the appearance of foreigners from various locations in one document does not always signify the proximity of their homelands. Nos. 290/293 mention Huhnurians alongside people from Simurrum and Lullubum some 800 kilometres away. Furthermore, Neo-Assyrian sources mention a mountain Elani'u and a locality Urina in Zamua/Lullubum.Footnote 16 Elašer in the central Zagros instead of near Huhnuri would make sense in the context of deportees or refuges from Lullubum and Simurrum at Lagash (see next section). This localisation would not disqualify Elašer from being part of “Elam”.Footnote 17
A closer look at Gudea's Iranian campaign may give a hint as to why Iranian polities sent slaves to Lagash. Gudea's Statue B provides the only testimony of his external politics. Its narrative revolves around the building of Ningirsu's temple in Girsu, a project that required the import of luxurious building materials. Gudea claims that Ningirsu “opened for him the roads leading from the Upper to the Lower Sea”, that is, from the Mediterranean Sea to the Persian Gulf. Gudea acquired wood and stone in Lebanon and Syria, mined copper in Kimaš, whose localisation within western Iran is debated (Potts Reference Potts, Backer, Robson and Zólyomi2010), and brought gold and wood from Meluhha (the Indus Valley). This peaceful narrative ends abruptly with a statement that Gudea “defeated the cities of Anšan and Elam and brought their booty” to Ningirsu's temple (Edzard Reference Edzard1997: 33–35). Gudea's inscriptions do not reveal the cause of hostilities with Elam.Footnote 18 On the contrary, his Cylinder A states that “the Elamites came from Elam, the Susians from Susa” to help with the renovation of Ningirsu's temple (ibid.: 78). Gudea's building activity in Adamsul does not necessarily mean hostilities with Susa. The archives suggest Lagash's close military and cultic relations with that centre.Footnote 19
The Lagash archives also exhibit a consistent west-east or Huhnuri>Anšan>Marhaši Iranian policy. Gudea never mentions Huhnuri even though the archives mention Huhnurians more than other Iranians. Another fact suggests the special role of Huhnuri in Gudea's eastern policy. People from Huhnuri appear with people from the east (Anšan, Marhaši) and north (Lullubum, Simurrum) alike. A slave from/for a ruler of Huhnuri appears side-by-side with a soldier and a messenger from Marhaši (no. 524). Men from Huhnuri and Marhaši receive beer in no. 360. The same text mentions six local interpreters, who must have translated the Elamite and Marhašian languages of the visitors, and implies some negotiation. No. 35 states that Lagash dispatched three pieces of weaponry to a ruler of Huhnuri. One does not give weapons to an enemy. Again, Simhuzi of Huhnuri sent five women to Lagash as a mašdaria. These data raise the possibility that Huhnuri was Gudea's voluntary or forced ally in his war against Anšan. Moreover, it seems that Gudea negotiated the support, or at least neutrality, of Marhaši in this conflict.
The data on slaves give hints about the area of Gudea's offensive in Elam. Mašdaria-slaves from Huhnuri, Gimani (near Susa?) and Elašer (further north in the central Zagros?) point to quite a broad area in western Elam. Anšan lay further east, and the archives mention neither captives nor mašdaria-slaves from there. It is thus questionable whether Gudea actually sacked this city as he claims. What is certain is that his crusade against Anšan resulted in an influx of slaves from Elam to Lagash as captives and as mašdaria-deliveries. The true nature of the latter may have varied from diplomatic gifts to tribute from the defeated.
Lullubeans and Simurreans in Lagash
Two nearly identical documents, nos. 290/293, mention several Huhnurians alongside hundreds of Lullubeans and Simurreans. Some 800 kilometres separate Huhnuri from Lullubum and Simurrum in Iraqi-Iranian Kurdistan, which suggests that Lagash's political involvement also extended to the north.
These documents record seven groups of foreigners who received barley allotments. Two gangs of Lullubeans and one gang of Simurreans, each numbering fifty men, were accompanied by an additional seven to fifty babies and children. Their wives remain anonymous. The men received 300 litres per person—a remarkably high amount—while their babies and older children were allotted 75 and 150 litres. For comparison, the female weavers and children from Elašer had exactly 7.5 times less monthly barley in nos. 173–175. Men generally received a salary 1.5 times that of women. Hence, nos. 290/293 may record the remuneration of the foreigners for five months (60 litres barley × 5 months per capita). This time-span may recall the corvée duty that royal dependents (eren2) owed to the crown during the Ur III period (Steinkeller Reference Steinkeller, Molina and Garfinkle2013b: 351–352). Nos. 290/293 also mention a mayor (ha-za-num2), a characteristic of the royal sector, and troops of fifty men (sixty in Ur III). Finally, these documents mention another group of people who lived in a village in Girsu's environs.Footnote 20 These details suggest that the Lullubeans and Simurreans of nos. 290/293 were royal settlers.
Some of the Lullubeans were of high social rank. A foreman, Uariagiar, received wooden tables (no. 528), whereas his mother-in-law and two other Lullubeans, Uariabad and Darzigiar, collected large amounts of copper and tin (no. 41). Zangadar had a double salary (nos. 290/293) and received a large volume of barley and a linen garment of “king's” quality (Pettinato et al. Reference Pettinato, Picchioni and Waetzoldt1978 no. 157). The latter document also mentions a royal messenger, which refers to the king of Uruk. This detail places these events before 2111 B.C., prior to the Ur III kingdom and Ur-Namma's exploits.
Other foreign nobles include “the sons of Šimpišhu(k)” and “the brothers of Aba-AN.HU”. The former were Puzur-Inšušinak's kin, as Šimpišhuk was his father (Steinkeller Reference Steinkeller, De Graef and Tavernier2013c: 299–301). Steinkeller sees the presence of the sons of Šimpišhuk in Lagash as a confirmation of his historical model: Namely that Gudea helped Ur-Namma to eliminate Puzur-Inšušinak. However, Steinkeller cautiously names the foreigners of nos. 290/293 “workers or soldiers” rather than captives or deportees. In fact, the kinsmen of Šimpišhuk and other VIP foreigners in no. 290/293 lived quite comfortably in Lagash with their families. Furthermore, why would the scribe mention Puzur-Inšušinak's father and not the great conqueror himself?
The fact that neither Gudea nor Puzur-Inšušinak nor Ur-Namma claim to have fought Simurrum and Lullubum also challenges Steinkeller's model. Naram-Suen's famous stele celebrates his victory over Lullubum, and his royal inscriptions confirm that he raided Simurrum too. A war against Simurrum and Lullubum is the highlight of Erridupizir's inscriptions. This Gutian king controlled the capital of the former Akkadian kingdom, but the dates of his reign remain unknown.Footnote 21 Erridupizir imitated Naram-Suen closely. He called himself “the king of four quarters”, adopted Ilaba as his clan god, and made a victory over Simurrum and Lullubum the zenith of his career. Erridupizir's inscriptions state that a king of Simurrum, previously loyal to his father, revolted and instigated unrest as far as Lullubum. The goddess Ištar assembled the Gutian hordes (um-ma-num2) in Agade. These fought numerous battles in rough, mountainous terrain, and brought the war to a successful end with the capture of a ruler of Urbilum/Erbil.
These events may have brought the Lullubean and Simurrean nobles and commoners as deportees, refugees or allies to Lagash. The first scenario presumes that Lagash allied with the Gutians and gained a share of the captives. In the second case, Lagash opposed the Gutians and harboured those who fled their oppression. In the third, these were allied troops. Any of these models may apply to the Lullubeans and Simurreans and to the “sons” of Šimpišhuk and to several other Elamites alike. What seems certain is that the resettlement of the foreigners in Lagash happened prior to Ur-Namma's military exploits in Babylonia and Susiana.
Slave trade and local resale
Returning to the Iranians slaves, taking captives and receiving “kids led by one's side” were not the only ways to acquire them. Gudea's slave dossier records two cases when the Girsu palace bought slaves. The scribes differentiated between a “traded” or “imported” Iranian (šu-bala…ak) and a local purchase (sa10 “to buy, sell”).Footnote 22
In the first case (no. 255), two men helped the palace to acquire a male slave. He constituted the “arrears” of one of them, which shows that this person regularly purchased foreigners for the palace. The document does not state where the slave was bought, whether in Elam or a trade hub en route to Sumer.
In the second case, a cupbearer, Ka'a (“Fox”), sold his Iranian slave woman, Huhume, to the palace (no. 251). Cupbearers were top-tier state officials. They served the royals at meals and belonged to their closest, personal circle (Sallaberger Reference Sallaberger2019). The sale of Huhume was, in fact, an inter-elite resale.
Sources from other periods confirm that royals acquired slaves from trusted persons. For example, a purification priest sold his Subarian slave to the palace in Adab around 2300 B.C. She may have belonged to the precariat that emerged due to Akkadian military expansion in northern Mesopotamia. A war-captive scenario is less probable here: Why would the king give a slave woman as a gift to this priest only to buy her from him afterwards? Someone may have enslaved and trafficked her to Sumer, where the priest bought her for himself but later sold her to the palace. The administration put her to work producing textiles in the House of Bondage, a detention centre where captive Subarian women and children produced textiles (Bartash Reference Bartash2018). Huhume may have shared a similar fate. It was not Lagash troops but economic hardship that enslaved her, and head-hunting merchants just needed to take her to Sumer.
Two purchased women among the fifty-five Iranians in Gudea's slave dossier seems quite few. However, Girsu texts from earlier periods confirm that the royal household regularly bought slaves, mostly of local origin. Lugal-ušumgal, who governed Lagash around 2200 B.C., bought multiple Sumerians.Footnote 23 Sixty female weavers joined the temple of Baba under the queen's control in 2319 B.C. (Prentice Reference Prentice2010: 56). These women were purchased (sag sa10-a) by the household, and their Sumerian names suggest that all were locals. Unlike captives, they did not have children. Moreover, the timing was unsuitable for taking captives. The kingdom of Uruk besieged Lagash's capital for three years in Urukagina's fourth to sixth years (Sallaberger & Schrakamp Reference Sallaberger, Schrakamp, Sallaberger and Schrakamp2015: 85). These details suggest that the sixty childless Sumerian “women” were local, girls from Girsu. The dreadful economic conditions under the siege may have forced parents to sell their daughters to the palace and thus save them from starvation. Lagash capitulated only four years later.Footnote 24
Both examples fit Richardson's (Reference Richardson2020) model that war and economic inequality created slaves, who were “consumed” elsewhere. In effect, Gudea's slave dossier adds the practice of buying foreign slaves.
Individuals give their Iranian slaves as gifts (sag-ba + personal name)
The text on the purchase of Huhume from the cupbearer Ka'a also records that three locals provided fifteen Iranians to the palace (no. 251). One of the givers was a soldier.Footnote 25 Each of the givers provided exactly five slaves: five women, eight boys and two girls. The children were never alone, as there was at least one woman in each group.
The scribe recorded the relation between these three Sumerians and their Iranian slaves using the construction sag-ba plus a personal name of one of the Sumerian givers. Its interpretation is challenging.Footnote 26 The easiest one would be “heads distributed/given as gifts to/by such and such”. This usage appears in the case of a ruler of Adab who gave slaves as gifs to rulers and high officials of other Sumerian and Iranian city-states around 2450 B.C. (Bartash Reference Bartash2020). Also, Mesopotamian rulers gave a share of the spoils of war, including captives, to their military as gifts (see Frayne Reference Frayne1997: 66). Recipients sometimes gave these human presents further to other persons, thus allowing captives to circulate within the “private sector” as chattel slaves.Footnote 27 Applying the “ruler's gift” model to no. 251 would mean that these three Sumerians received their fifteen Iranian slaves as gifts from the ruler, and then gave them back. Alternatively, they may have come into possession of these fifteen slaves by other means, such as slave trade.
One document of the Garšana archives from 2026 B.C. demonstrates the use of sag-ba in a similar context of giving one's slave to authorities as a gift: “Geme-Enki, female weaver, the slave given as a gift by Gasusu, wife of Ahu'a. Ana'a was the controller (of the transaction). (The slave) withdrawn. Takil-ilišu withdrew (her) from Iškur-illat”.Footnote 28
The bond between the palace and private households becomes even more evident if we consider the possibility that Zana in no. 251 was identical with Bakzana in no. 258. The latter text lists five male and fifteen female Iranian slaves.Footnote 29 Bakzana was the only adult woman, the rest were girls.Footnote 30 Sargonic archives from Adab document the same demographic pattern. A troop of captive Subarians consisted of four women, one or two boys, and fourteen girls (Bartash Reference Bartash2018). The twenty Iranians of no. 258 likewise represented a single troop of captives.
Transferring Iranians to other royal households
Labour and goods circulated among a triangle of households in Girsu for centuries: the palace, Baba's temple in the queen's hands, and the household of the royal children (Borrelli Reference Borrelli, Notizia, Rositani and Verderame2021). For example, the administration reassigned a team of female weavers from the children's household to Baba's temple around 2323 B.C. (Prentice Reference Prentice2010: 56).
Gudea's slave dossier offers a similar example. The household of the royal offspring employed twenty-seven Elamite (NIM) women with fourteen children (nos. 173–175). The scribes noted seven women and two children from Elašer among them as a recent addition.Footnote 31 These Elamites appear only among the weavers, the typical occupation of captive women, and comprised almost one-fifth of all the weavers in the household. This confirms the economic significance of foreign labour in state-run households.
The circulation of slaves between royal households and their giving to and receiving from palace affiliates show that the mobility of the Iranians did not stop even after they completed their journey from their homelands.
Synthesis and conclusions
This analysis of Gudea's slave dossier expands our knowledge of foreign-born slaves in early Mesopotamia in two respects. First, the palace, the heart of the early state, acquired foreign forced labour by various mechanisms, only one of which was taking captives. Second, slave mobility cannot simply be reduced to the economic factor of demand for cheap, submissive, uprooted labour. Giving and receiving slaves was a political act that served to create or sustain power relations between two agents. These two points can be elaborated by focusing on the forms and elements of slave mobility.
In a mere dozen administrative texts, Gudea's slave dossier captures an influx of Elamite slaves to Lagash following Gudea's campaign against Anšan in the late 22nd century B.C. The philological and historical analysis of the slave dossier relied on the entire Lagash II archival and royal textual corpus and comparative data on foreign captives and slaves in southern Mesopotamia, ca. 2450–1750 B.C. This analysis revealed that the palace and private households at Girsu obtained and mobilised Iranian slaves in a variety of ways. These transactions relocated Iranians between regions, households, and individuals, and may thus be called forms of slave mobility (see Fig. 4).
Fig. 4 Forced mobility of Iranians in Gudea's slave dossier
Surprisingly, only one text states that Lagash troops brought a single male slave, probably a captive. Either Gudea's administration did not normally record the deliveries of captives or these sources did not survive. Only the demographic pattern suggests that the dozens of Elamite women and children who toiled as weavers in the household of the royal children were captives. Likewise, we do not know what happened to the male captive upon his delivery by the troops.
Gudea's war against Anšan led to political instability and socio-economic precarity and inequality in the affected Iranian regions. Following Richardson's (Reference Richardson2020) model, it is conceivable that many Iranians were sold into slavery by their own families. Slave traders relocated them to trading hubs, where Mesopotamians could buy them. Gudea's slave dossier documents one case of the long-distance slave trade, in which several affiliates helped the palace at Girsu to obtain an Iranian male slave.
The local slave trade features prominently in Girsu documents throughout the centuries. Rulers and their wives routinely bought slaves of local origin. The sixty girls purchased during Uruk's siege of Lagash in the late 24th century B.C. are a case in point. Gudea's slave dossier documents one local purchase, but it differs in two respects. First, the purchased slave was an Iranian woman. Iranians were abundant in Lagash following Gudea's eastern campaign. Second, the seller of the woman was a close associate of the royal family. This transaction was essentially a resale of a commodity between two friends.
Mašdaria or “kids led by one's side” is the most frequently documented form of slave mobility in the dossier. Classic mašdaria were deliveries of kids, lambs or food and drink to the ruler by the local elite. This tribute-like institution allowed the ruler to provide for the main yearly festivals. An occasional gift to the royals or a votive gift to the gods were additional functions of the mašdaria.
In Gudea's slave dossier, Elamite slaves substitute for animal young and the deliveries come both from within and without. In the first case, two locals of obscure background gave Iranians as mašdaria to the palace. Either they did it for money—mašdaria-deliveries could be sales of slaves in disguise—or they had an obligation vis-à-vis the ruler.
On the macro level, three Iranian polities sent slaves to Lagash. These states were situated in the region where Gudea waged his anti-Anšan war. His scribes applied the mašdaria-model when describing the tribute and gifts of the Iranian states to stress their dependency upon Lagash. The administration placed Iranian rulers on the same footing as locals who were required to bring kids, lambs, onions, etc. to the palace.
The demographic pattern of the mašdaria-slaves suggests that they were luxurious gifts rather than crude labour for royal sweatshops. Unlike captives, who were mostly women with children, “kids led by one's side” were young men and women not burdened by offspring.
Finally, three individuals, one of them a royal soldier, gave their fifteen Iranian slave women and children to the palace. The construction sag-ba “head given as gift/distributed” may refer to the fact that the palace gave them these Iranians and they returned them for some reason. Alternatively, it may document the fact that these three individuals gave their slaves as gifts to the palace.
The administration organised Iranian slaves into working troops of twenty individuals, mostly women and children, upon delivery. However, their mobility did not cease at that point, as officials could transfer them to other royal households such as the house of the royal children.
Ultimately, the forms of slave mobility described above share the same integral elements: slaves, two agents, two spaces and the mobilising force.
Slaves
The mobility of slaves is a paradox: they were de jure immobile. My argument is that slaves played merely an instrumental role in the transactions between two agents. They were objects but, as with all useful and valuable objects, they possessed characteristics that defined their value: number, origin, language spoken, age, gender and presence or absence of children. The analysis identified two demographic patterns among Iranian slaves. The palace desired numerous females and children to put them to work in textile mills and other workshops or to transfer them to other royal households to fill labour shortages. In contrast, the mašdaria-slaves were few, childless (=young) men and women.
Agents and their spaces
Slaves were mobile because of and between two people—a slave-giver and a slave-receiver—and their spaces. Like everyone in the ancient Middle East, the agents of slave mobility belonged to or even headed one or several households—private, palatial or temple. A change of master led automatically to a change of household for the slave. Spatial mobility and social mobility thus become one. For a chattel, a new household meant new tasks and a new social and, sometimes, legal status. For example, captives donated to temples could not be sold.
Driving force
It is hard to argue with the traditional approach that explains slavery as a source of cheap, submissive labour in societies across time and space. Gudea's slave dossier confirms that royal and private households craved foreign labour who could be readily employed in various tasks. The example of mašdaria-slaves, however, adds a political dimension to the mobility of slaves. Power relations between giver and receiver feature prominently in Gudea's slave dossier. Iranian polities sent slaves to Lagash after their defeat or because they felt themselves threatened. By sending a few young and, probably, attractive males and females, they showed loyalty and respect.
Other forms of transactions with slaves exhibit political ties between giver and receiver. Troops delivered captives because they fought under their sovereign's colours. It was their job to deliver booty to their master. Likewise, Mr Fox the cupbearer is unlikely to have argued when the ruler asked to purchase his woman Huhume. On the contrary, this transaction would only have reinforced the bond between cupbearer and ruler.
In effect, slave mobility was akin to a vehicle with a hybrid engine. Politics accelerated it but the economic interest made it roll. Almost all forms of slave mobility created and sustained political dependencies between those who gave and those who took. Gudea's slave dossier demonstrates that slave mobility was less about slaves than about those who used them for economic and political purposes.
Taking a broader perspective, the large-scale background events make Gudea's slave dossier particularly exclusive. Gudea's building activity in western Iran, his vigorous diplomacy as far as Marhaši near Pakistan, and his anti-Anšan war fit well into Lagash's eastern politics. This powerful and rich Sumerian state lay on the western shore of the Persian Gulf. Its arch-enemy Umma blocked potential expansion upstream along the Tigris. That is why Lagash had fought, cooperated and traded with Iran from time immemorial. Its shrewd rulers even attempted to assert control over western Iranian polities. Gudea followed this ancient Lagashite geopolitical formula. In this respect, south-western Iran was much more than just a slaving zone for Lagash. The mobility of incoming Iranian slaves scrutinised here is merely an example of the traffic in people, goods and ideas between southern Mesopotamia and Iran that persisted for millennia.
