How Many Lives for a Mesopotamian Statue?

Abstract

Among the indicators of the value and power ascribed to statues in Mesopotamia, reuse is a particularly significant one. By studying some of the best-documented examples of the usurpation and reassignment of a new function to sculptures in the round from the 3rd and 2nd millennia BC, our study reveals the variety of motives and methods employed. We hereafter explore the ways in which the status of such artefacts is maintained, reactivated, or adapted in order to secure their agency.

1. Introduction

In ancient Mesopotamia, certain statues stood out because they were considered alive thanks to a process of incarnation (or embodiment). These statues underwent a Washing of the Mouth Ritual (mīs pî1) that led them to transcend their status as mere representations and become the divine entity they represented. Their design and creation, both based on human decisions, culminate in this incarnation process, which enables the representations to transcend their status of objects of craft. It is even required that the sculptors concede the kinship of their production to the divine by saying: “I did not make (the statue); (I swear) I did not [make (it)]; Ninildu, who is Ea the god of carpenter [made it…]”2 (Walker and Dick 2001). In this situation—which concerns only divine statues3, contrary to Egyptian practice4—the concept of agency is significant and the power of the statue easy to identify. However, as theorized by Alfred Gell, even if an image embedded in the cult has a central and distinctive place (Gell 1998, p. 96), the statuary representing kings or elite members also have strong agency. Royal statuary received a form of worship through the daily cultural practices it underwent (Winter 1992, pp. 16–17). As for the images of the orants, ordinary members of the elite, their mere access to the temple, close to the divine image, had an impact on their agency. This special value of statues can be illustrated, for instance, by the efforts made to keep them present in one way or another when they no longer fulfil their initial role (Connor 2018, p. 148).5

Reuse consists of continuing the use of an artefact, but with a new function. This change of function may be significant (e.g., through the inclusion within a ‘collection’ of a statue originally dedicated in a temple) or it may be no more than a minor adaptation (e.g., when an image is (re-)dedicated to a new deity). There are many reasons for such a practice: pragmatism related to the cost of raw materials, the impossibility of getting rid of an object considered to be of sacred nature, the desire to assert one’s dominance over others, the passing on of memories, etc.

This practice is understood as an overarching category comprising the sub-sections of reassignment, usurpation, and recycling. In the frame of this article, the latter—namely the recovery of raw materials (metal smelting, use of the statue as a building block, etc.)—is not considered. While the intended function of the artefact does indeed change—e.g., architectonic element—the life of the statue is not extended for what it once was. Rather, everything it was is transformed/forgotten/adapted. Moreover, the information on this subject is rather of a philological nature. Indeed, the material traces of such a practice are rarely perceptible on the objects itself, the main exception being when a statue or one of its fragments is used as a building block or, as well attested in Egypt, as vases or hammers (Junker 1950, pp. 125–26; Eaton-Krauss 2008). In Mesopotamia, the recovery and rework of some raw materials is also attested. Precious metals are melted down (Na’aman 1981) or stone is transformed into gypsum by the effect of heat (Evans 2012, p. 141).

As for the other two sub-sections, they form the basis of the organization of this paper. The first part is devoted to usurpation and the second to reassignment, the one being the appropriation of a statue’s identity and the other its modification.6 Before delving into the case studies, it is essential to define this study. It is part of a wider research project looking at all aspects of the ‘life’ and ‘death’ of Mesopotamian statues. It is, therefore, a work in progress, destined to evolve and, above all, to be compared with other practices, reuse being only one of them.7 To quote Nancy Highcock on votive objects that have been recast, moved, buried, and recovered: “The lives of dedicated objects in ancient Mesopotamia were dynamic” (Highcock 2021, p. 40). Through case studies, we will endeavor to encompass the trajectory of statues to this variety of reuses.

2. Case Studies

Unlike Egypt, there is very little evidence of reuse over the three millennia of Mesopotamian history (to delve deeper into this issue in Egypt: Brand 2010a, 2010b; Eaton-Krauss 2015). Statues with slight modifications, such as the removal of the beard,8 exist in addition to the five usurpation examples studied in this paper. However, nothing can be concluded about the when, why, and how of this change, which could even be the result of a change at the time of production rather than a reuse. As for attestations of reassignments, the corpus presented in this paper is exhaustive as far as archaeological evidence is concerned. Other deportations—involving modifications—are documented, but exclusively by texts and they are therefore not all included here.9 The cuneiform tablets describing Akkadian statues (cf. Section 2.2.5) are discussed here, since they are considered to be ekphrasis of statues that have been exhibited. It is precisely because the documentation is scarce and sometimes makes it difficult to come to precise conclusions that the organization of these case studies is based on their relevance, the first examples being the most revealing in respect to the issues.

Apart from the statues, however, several texts provide information about the practice of reusing artefacts. In the 8th century BC, Sargon II (Neo-Assyrian ruler, ruling between 721–705 BC) had paraphernalia, such as a cultic bed and various gold objects, moved to temples in Dur-Šarrukin (present-day Khorsabad, Iraq) (Parpola 1987, pp. 49–51, 93–96; Reade 2002, p. 200). As it appears, the cult function of these objects remained similar, but they were dedicated to other deities in a different place. Small objects such as beads (Galter 1987, p. 17), cylinder seals (Amiet 1976, p. 60; Thomason 2005, pp. 155–56), or vases (Eppihimer 2010, p. 378) are among the artefacts that have undergone this type of adaptation, as their size makes them easy to usurp. In short, reuse took many different forms and served many different purposes.

Among the testimonies to this habit related to statuary are the curse formulas10 precisely meant to prevent such practices. One of the aims of creating an anthropomorphic representation was to ensure that the memory of the individual represented would live on, presumably forever. It inevitably involved the risk of seeing this person, via the medium of their representation, attacked (Bahrani 1995, p. 381). It is therefore easy to understand the importance of protecting one’s statue to protect oneself. The same applies to the name, which, once inscribed, serves the same function of leaving its mark on history, but carries the same risk of being erased (Radner 2005, p. 252). These curse formulas are attested in Mesopotamia, generally placed at the end of dedication inscriptions, intended to protect a monument, a text, an individual, etc. The curse formulas referring to the practice of reuse are those dealing with the damage inflicted on the text rather than on the image itself.11 Many of them are structured as follows:

(RIME 4.06.11.1 [Old Babylonian Period]—l. 17–19)

“He who removes my inscribed name and has his (own) name ins[cri]bed (…)”

(17. ša šu-mi 18. ša-aṭ-ra-am 19. u₃-ša-sa₃-ku-ma 20. šum-šu u₃-ša-aš₂-[ṭa]-ru)

As far as sculptures in the round are concerned, what are the criteria for identifying these acts of reuse? Ideally, the image has been transformed, whether this concerns the representation (e.g., changing the features of the face) or the inscription (changing or adding a name to provide a new identity). Stratigraphy also provides information: an archaic style discovered in a more recent layer for instance. However, all these criteria must be considered with caution. An adaptation of a figure or text might simply represent a correction that occurred at the time of production. On the other hand, a style that contrasts sharply with the archaeological layer may attest to the persistence of styles though time. This is what Eva Strommenger warns against when she points to an assertion by Henry Frankfort: “Frankfort seems to assume a stylistic influence from the previous period when he says: ‘In Akkadian works of lesser quality the affinities with the older period are so pronounced that it is sometimes only possible to assign a work to the Akkadian Period because an inscription names the reign in which it was made.’ He also turns to the inscription as the final dating authority, although he clearly recognizes the foreignness of this monument within the Akkadian period.”12 (Frankfort 1955, p. 43; Strommenger 1959, p. 34). In delicate situations where a style seems incompatible with the context in which it is discovered, he presents inscriptions as the best means of dating. She rightly insists on the fact that a single method is conclusive, namely the combination of a study of the style, the inscription (when existing), and the historical context (Strommenger 1959, pp. 28, 36, 50).

2.1. Usurpation

Usurpation regularly occurs by modifying or adding an inscription.13 Understandably, these two methods have very different implications, respectively erasure of the previous owner or the cohabitation with the new one, who may then wish to stress their respect or superiority vis-à-vis the previous owner. This investment in the inscription can also be coupled with displacement since a change of context easily leads to a change of function.14 This practice, known as deportation of the statue, served several purposes. The aim was of course to overcome the enemy, but it was also an opportunity to appropriate their greatness and aura. To appropriate the power of an enemy implies to admit said power. Deportation is therefore a practice that walks the thin line between violence and respect, just like reappropriation. Another way to acknowledge this state of fact is the considerable effort required to move a monumental statue15 (and ultimately any looted object). One might rightly argue that such a considerable effort might not have been undertaken if the purpose was simply to break a valuable artefact to deprive the enemy of it.

Besides this difficulty of determining the intention behind a reuse, another difficulty is that examples are not so easy to identify. As Jean Evans pointed out, many unearthed isolated statue heads might constitute evidence of a habit of replacing them in order to reuse a statue’s body and thus save time and resources (Evans 2012, p. 140). In the meantime, if it is enough to add or change a name to completely alter the identity of an image, it is interesting to address the question of portraiture, an overall rather complex concept, the meaning of which has evolved over time. However, it is generally accepted that the situation in Mesopotamia is similar to the one in Egypt. The representations in both cases were intended to emphasize elements that identified and enhanced the status or function of the individual depicted rather than individualizing them through their physiognomy. It appears that the aim was not to copy, but to confer a significant character to the production (Belting 1993, p. 129; Winter 2009, pp. 266–67; Laboury 2010; Asher-Greve and Westenholz 2013, pp. 159–62; Guichard 2019, p. 31). The statue of Puzur-Eštar (šakkanakku16 of Mari (present-day Tell Hariri, Syria), from the 20th century BC), which we will address in detail later on (cf. Section 2.2.2), is an illustration of this understanding of the concept of portrait. It demonstrates that modifications are only necessary in the case of a change in status or function. Indeed, this depiction of a šakkanakku had horns, the divine symbols par excellence in Mesopotamian iconography, carved in his headress to become a divine representation (Blocher 1999).

Regrettably, nothing on the matter of reuse can be said with any certainty, and the same is true of most of the clues that suggest usurpation. For instance, a work in an ancient style, present in a recent stratigraphic layer, may have “lived” for a long time without necessarily having been reattributed (Hauptmann 1989, p. 33). The five attestations studied in this paper are significant due to their context of discovery or the traces of modification they bear, enabling a closer look at their (multiple) lives.

2.1.1. Liebieghaus Orant [1453]17 (Figure 1)

According to Eva Andrea Braun-Holzinger in 1991,18 the only established usurpation of a Mesopotamian statue was identified thanks to the modification of the inscription, namely on an orant statue preserved in the Liebieghaus in Frankfurt am Main (n°1453; H: 37.5 cm) (E. A. Braun-Holzinger 1991, pp. 220, 254). According to her, the only valid clue is the clearly visible ‘palimpsest’ effect. This limestone statuette is dated to the Early Dynastic (ED) III (ca 2600 to 2330 BC) and depicts an orant, hands clasped, wearing a kaunakes.19 Unfortunately, since it was acquired by an American collector before entering the German collection in 1955, its provenance is unknown (Hauptmann 1989, pp. 3–4). The inscription on the right shoulder is inscribed on a polished area: according to Ralf Hauptmann, a first inscription in Sumerian was erased—a few lines of the old inscription are still discernible. The one visible today, in Akkadian, was inscribed at a later date (Hauptmann 1989, p. 5). Though carefully erased, the verb ‘to consecrate’ in the Sumerian text can be identified. It is likely that it was, as the Akkadian one, a dedication to identify the owner of the statue and the god to whom he is offering his representation.

(R. Hauptmann 1989, p. 5; Liebieghaus 1453—l. 1–6)

“For (the god) Lugal-asal, Bazi son of Bēlī-Tišpak, the pašīšu-priest dedicated this.”

(1. ana 2. ^dLUGAL-ASAL_X 3.Ba-zi 4. DUMU be-li₂-Tišpak 5. KI^?/ŠU^?.pa₄-šiš 6. A.MU.NA.RU)

R. Hauptmann suggests that the Akkadian inscription dates back to the very beginning of the 2nd millennium BC due to the text organization and the signs variability (Hauptmann 1989, p. 6). This means there may be three centuries separating the two inscriptions and/or the carving and the addition of the second inscription. What we have here is indisputable evidence of usurpation. It was suspected due to the erasure of an inscription and confirmed thanks to a lack of consistency between the dating of the representation and the dating of the new inscription. These same criteria are relevant in the study of the following case.

Figure 1. Statue of the orant [1453] from the Liebieghaus. Courtesy of the Liebieghaus Skulpturensammlung, Frankfurt am Main.

Figure 1. Statue of the orant [1453] from the Liebieghaus. Courtesy of the Liebieghaus Skulpturensammlung, Frankfurt am Main.

2.1.2. Ešpum [Sb 82] (Figure 2)

An alabaster bust was discovered on the acropolis of Susa (present-day Shush, Iran) in 1906 and is currently displayed at the Louvre (Sb 82; H: 31.2 cm). An inscription was added to the back of this bust (Frayne 1993, pp. 81–82):

(RIME 2.01.03.2001 [Sargonic and Gutian Periods]—l. 1–8)

“Maništūšu, king of the world: Ešpum, his servant, dedicated (this statue) to the goddess Narunte.”

(1. ma-an-iš-tu-su 2. LUGAL 3. KIŠ 4. eš₄-pum 5. IR₁₁-su 6. a-na 7. ^dna-ru-ti 8. A.MU.NA.RU)

On the one hand, the anthroponyms allow this inscription to be dated to the end of the 3rd millennium BC since Maništūšu was an Akkadian ruler (ruling between 2284 and 2262 BC). On the other hand, given the style of the statue, which dates from the middle of the 3rd millennium BC, the cuneiform text was added several centuries after the creation of the sculpture (Spycket 1981, p. 73; Frayne 1993, 2:81; Highcock 2021, p. 42). There is no evidence that a first inscription was erased and/or covered by the new one.20 Apparently this reuse is pragmatic, as there is nothing to suggest a desire to impose itself over the identity of the previous owner. Now, to speak of opportunism, we need to know whether steps were taken to move the object or whether it was already in Susa at the outset. Moreover, the antiquity of the object may have made it valuable to the new owner (Highcock 2021, p. 42).

Agnès Spycket and E. Strommenger argue with confidence that the object, of Susian origin, was usurped after three hundred years and appropriated through the addition of an inscription (Strommenger 1959, pp. 34, 36; Spycket 1981, pp. 73–74). E. Braun-Holzinger questions this proposition, believing instead that it reflects a simple persistence of ancient styles, which is attested in Mesopotamia and Elam (E. A. Braun-Holzinger 1991, p. 220). The distinctive beard leads her to believe that this is a local Elamite production, but designed for the Akkadian ruler (E. A. Braun-Holzinger 1991, pp. 257–58). Due to the lack of evidence of Elamite productions of this type, this remains to be studied with hindsight (Evans 2012, p. 230). E. Strommenger responds that, stylistically, the statue strictly follows the codes of earlier style in the 3rd millennium BC. She also rejects the possibility of an ancient inspiration, as the retouching work carried out on the hair bears witness to a period of modification of the work which supports the idea of an usurpation (Strommenger 1959, pp. 34–36). A similar discussion arose regarding the next representation, discovered in southern Mesopotamia.

Figure 2. Statue of Maništušu [Sb 82] from the Louvre. Front: Courtesy of the Musée du Louvre (1999 Jérome Galland); Back: (Strommenger 1959, 31, pl. IIb).

Figure 2. Statue of Maništušu [Sb 82] from the Louvre. Front: Courtesy of the Musée du Louvre (1999 Jérome Galland); Back: (Strommenger 1959, 31, pl. IIb).

2.1.3. Ekur/Kurlil [BM 114206] (Figure 3)

This statue, discovered at Tell Al-’Ubaid (present-day Iraq) in 1919, is currently stored in the British Museum (BM 114206; H: 34 cm). Although it was discovered in a well-dated context from the 3rd Dynasty of Ur (ca 2100 to 2000 BC) (Hall and Woolley 1927, pp. 19–20) alongside another representation (BM 114207), its identity and chronology raises several questions. At one point, the two sculptures had been identified as an image of the same individual: Ekur (=Kurlil) (Hall and Woolley 1927, pp. 19, 27, 125). However, the other statue, although in an overall good state of preservation, displays a very fragmentary inscription, with only one sign remaining. The sign in question has long been read as E₂ (E. A. Braun-Holzinger 1991, p. 252), but Gianni Marchesi and Nicolò Marchetti have corrected the reading to NIMGIR, proving that it is not the same dedicant (Marchesi and Marchetti 2011, pp. 66, 163). As for the statue we are concerned with, its inscription, unframed but with separating lines, reads:

(Marchesi and Marchetti 2011, 163—l. 1–5)

“Ekur, superintendent of the granaries of Uruk, created a (statue of) Damgalnunak and built (her) ‘house’.”21

(1. e₂:kur 2. KA:GUR₇-unug^ki 3. ^ddam:gal-nun 4. mu-du₂ 5. [e₂] mu-řu₂)22

Due to the archaic nature of the inscription, G. Marchesi and N. Marchetti suggested that it came from an earlier level of the temple in which it was discovered (Marchesi and Marchetti 2011, p. 66). At the time of its publication by the excavators, they had already noticed the unusual nature of the inscription, but nevertheless had no doubt that it was contemporary with the layer where it was found (Hall and Woolley 1927, pp. 19–20, 125). As for the style of the statue, A. Spycket considers it to be Early Dynastic II (ca 2750–2600) based on the large format of the arms and the fact that they are detached from a narrow bust (Spycket 1981, p. 106). Considering both the surrounding stratigraphical data and the style of the statue, this representation would have persisted for around five centuries, a period during which, at an unidentified moment, it was allegedly usurped.

Much could also be said about its state of preservation, since it is particularly damaged in an archaeological context where the majority of other artefacts were found in good condition. Thus, Julian Reade proposes a deliberate attack on this particular representation (Reade 2000, p. 83). The fourth case is in a poor state of preservation too, and the lack of diagnostic features such as the head or inscriptions is a limiting factor in the study of reuse and its circumstances.

Figure 3. Statue of Ekur/Kurlil [BM 114206] from the British Museum. © 2023 Imane Achouche.

Figure 3. Statue of Ekur/Kurlil [BM 114206] from the British Museum. © 2023 Imane Achouche.

2.1.4. ‘Statue Cabane’ [Aleppo Museum M 7917/1326] (Figure 4)

Known as the ‘Statue Cabane’ (H: ca 1.10 m, estimated weight 300 kg), it is an acephalous representation discovered in 1933 on the tell of Mari, before the first archaeological excavations on the site (Parrot 1935, p. 1; Thureau-Dangin 1939, p. 157). It was named after the Lieutenant Cabane who took it to the Aleppo Museum (n°M 7917 or 132623). The dedication on the lower half of the garment attributes the statue to Yasmaḫ-Addu (ruler of Mari, from the 18th century BC) (Thureau-Dangin 1934, p. 144; Frayne 1990, pp. 615–16):

(RIME 4.06.11.1 [Old Babylonian Period]—l. 1–15)

“[Ia]smaḫ-Addu, ap[point]ee of the god Enlil, [so]n of Šamšī-Adad, for the god Šamaš, his lord […] [had] (this statue) fashioned in [the city of] M[ari, wh]ich he l[ov]es, and [de]dicated (it) (…)”

(1. [ia-a₂]s-ma-aḫ-^dIŠKUR 2. š[a-k]i-in ^den-lil₂ 3. [DUM]U ^dUTU- ^dIŠKUR 4. a-na ^dUTU 5. be-li₂-šu 6. […] 7. […] 8. [m]u(?)-te-[…] 9. […] 10. […] ni […] 11. [i]-na q[e₂]-r[e-e]b 12. [a-al] m[a-ri.K]I 13. [š]a i-r[a-a]m-mu 14. [u₂-še]-p[i₂-i]š-ma 15. [u₂-š]e-li)

Ursula Moortgart-Correns suggested in 1986 that it was an usurpation, since the overall shape of the body indicates that it must originate from an earlier period, with the treatment of the beard definitely being of the very beginning of the 2nd millennium BC (Moortgat-Correns 1986, p. 185). As for the original identity, she argues that it was a mountain god, given the type of clothing and the posture, appearing to be carrying a load. She therefore considers that it could have been part of a pair framing a door (Moortgat-Correns 1986, pp. 183–84). However, divine statues in stone are rarely attested archaeologically in Mesopotamia, not to mention that there is no evidence of pairs of gods integrated into architecture before the 1st millennium BC. Still, neither the style nor the type of statue correspond to what would be expected of a šakkanakku image from this period.

U. Moortgart-Correns also noticed that no trace of reworking was visible, not even the removal of an earlier inscription that would have been replaced by that of Yasmaḫ-Addu (Moortgat-Correns 1986, p. 186). However, we must not lose sight of the state of conservation of the artifacts examined. Nonetheless, the elements of the head (headdress, facial features) would have been the first to be modified if this were to be done, but many statues—including this one—have been decapitated. However, as we have seen, usurpation does not necessarily require the adaptation of the features of the representation, as the addition of the inscription may be enough. The following statue is of particular interest in this respect.

Figure 4. ‘Statue Cabane’ [M 7917/1326] from the Aleppo Museum. © (Moortgat-Correns 1986, pl. 36.1 & pl. 37.6).

Figure 4. ‘Statue Cabane’ [M 7917/1326] from the Aleppo Museum. © (Moortgat-Correns 1986, pl. 36.1 & pl. 37.6).

2.1.5. Urlammarak [BM 91667] (Figure 5)

This is a very peculiar statue, which entered the British Museum collection (BM 91667; H: 14.6 cm) by purchase in 1854 and whose provenance is therefore uncertain (Reade 2000, pp. 84–85). At first sight, the cohabitation of two inscriptions would suggest that there was no intention to usurp. However, it is difficult to understand the order in which the cuneiform was inscribed or even the exact date of the bust production. It features an inscription on the back, now considered to be the original, which identifies the individual as Urlammarak:

(Marchesi and Marchetti 2011, 168—l. 1–2)24

“Urlammarak, viceroy of AN.PA.x”

(1. ur:^dlammar_x(KAL) 2. NIĜ₂ [PA].SI AN.PA.[x])

This inscription on the back is most probably contemporaneous to the statue’s creation; both the style of the writing and the representation are dated to the Early Dynastic (ca 2900–2330) (Marchesi and Marchetti 2011, p. 130). It is an inscription on the shoulder that attests to the reuse of this object. These are cuneiform signs that are incomprehensible and difficult to date. They are dated to the 1st millennium BC by G. Marchesi and N. Marchetti (Marchesi and Marchetti 2011, p. 168) or to the beginning of the Early Dynastic period by E. Braun Holzinger (E. A. Braun-Holzinger 1977, p. 75). Regardless of their dating, it is quite clear that the two inscriptions were not written at the same time. If the shoulder inscription is correctly considered to be a (much) later addition, and if it is indeed meaningless, then its purpose could be to fulfil an interest in the past rather than to convey a message (Marchesi and Marchetti 2011, p. 130).

At one point, E. Braun-Holzinger proposed that the inscription on the back was a later addition too. According to her, this sculpture in the round was an imitation of an Early Dynastic Style (E. A. Braun-Holzinger 1977, p. 75). Then she suggested that it was the reuse of a statue that had been reworked to give it an older style (E. Braun-Holzinger 2007, p. 44). These supposed traces of retouching work are to be seen in the holes drilled in the nose and scratch marks on the forehead. She thus suggests that the hair and eyes must also have been modified (E. Braun-Holzinger 2007, p. 44). However, the elements in question could very well be traces of repair, as attested in Mesopotamia. In any case, the reuse of the statue is illustrated by the addition of the shoulder inscription whether the image features have also been retouched or not.

Figure 5. Statue of Urlammarak [BM 91667] from the British Museum. © 2023 Imane Achouche.

Figure 5. Statue of Urlammarak [BM 91667] from the British Museum. © 2023 Imane Achouche.

More than a quarter of a century ago, Eva Andrea Braun-Holzinger considered that a single statue was certainly a case of usurpation (E. A. Braun-Holzinger 1991, p. 220). Nowadays, four statues in addition to this orant statue from the Liebieghaus have been considered. This is an extremely small total, especially given the large amount of statuary discovered in Mesopotamia.25 Marian Feldman explains that: ‘‘The apparent hesitancy to reuse statues may be traceable to the close correspondence posited by the ancient Mesopotamians between an individual and his or her image.” (Feldman 2009, p. 46).

2.2. Reassignment

When the objective is not to completely appropriate the statue by erasing the memory of its past, it is a question of making its new existence coexist with its past identity. This practice may even, sometimes, aim to insist on the original identity of the image. This is for example the case when king Nabonidus (Neo-Babylonian ruler, ruling between 556–539 BC) took care of a representation of Sargon of Akkad26 found in the Ebabbar foundations (Šamaš temple in Sippar, present-day Tell Abu Habbah, Iraq). This ‘antiquarianism’, as described by Melissa Eppihimer, was the result of a “face-to-face encounter with an ancient king” (Eppihimer 2019, pp. 1–2).

When seeking to reassign, as well as to usurp, what is more common is the addition of an inscription. However, in the second case, it cohabits with a previous inscription, or refers to it if it has not been preserved.

2.2.1. Šutruk-Naḫḫunte I’s Loot

There are few examples dating back to Šutruk-Naḫḫunte I (ruler of Elam, ruling between 1185–1155 BC), who had numerous objects carried away from Babylonia to Susa. This deportation policy was massive. Nearly half of the kudurrus/narûs27 excavated so far were found in Susa, although none were produced for the city (Slanski 2003). As for the statues, at least sixteen were discovered in Susa, transported from Babylonian cities (Table 1). On nine of them, traces of an inscription by Šutruk-Naḫḫunte I remain, indicating that he had recovered these representations and taken them to Susa. An example of these inscriptions, all relatively similar, can be found on the statue Sb 61, depicting a seated sovereign:28

“I am Šutruk-Naḫḫunte, son of Hallutuš-[In]šušinak, king of Anšan and Susa, the great likume(=SUKKAL.MAḪ in Akkadian inscriptions), the throne [holder of Ela]m, sovereign of the land of Elam. [Inšušinak], my god, having helped me, [I have destroyed] Eš[nunna]; I have taken away from there [the statu]e and I have brought it [to the country of El]am. [I have placed it before Inšušinak], my god.”

(1. u₂ ^Išu-⸢ut-ru⸣-uk-^Dnah-hu-un-te ša-ak ^Ihal-lu- 2.-du-uš-⸢^D⸣[in]-šu-ši-na-ak-gi-ik su-un-ki- 3.-ik ^AŠan-za-an ^AŠšu-šu-un-ka₄ li-ku-me ri-ša-ak 4.-ka₄ ka₄-a[t-ru ha-tam₅-ti]-⸢ik⸣ hal-me-ni-ik ha-tam₅-ti- 5.-ik ⸢^D⸣[in-šu-ši-na-ak] ⸢na⸣-pir₂ u₂-ri ur-tah-ha-an 6.-ra ^AŠiš-[nu-nu-uk hal-pu-uh za-al-m]u a-ha hu-ma-ah 7. a-ak hal-ha-[tam₅-ti te-en-gi-ih ^Din-šu-ši-na-ak] 8. na-pir₂ u₂-ri [i si-ma-ta-ah])

Regarding those statues with no inscription, chances are that the inscription is simply missing. It is indeed very common that the statues bear an inscription in Mesopotamia, at least the title of the individual represented (Winter 2010, p. 185). Depending on the period, material, and type of object, the location of the inscription may vary, but it is very often the upper back or lower part of the garment—large, flat areas. For example, Sb 60, which was badly damaged, may have carried an original inscription that is now missing from what was excavated. Similarly, the absence of an Elamite (=Hatamtite29) inscription may be linked to surviving evidence. In any scenario, it is worth questioning whether there is a primary and/or secondary inscription, and whether it is due to conservation reasons or a voluntary decision. Indeed, at least Sb 85 is sufficiently well preserved and still shows no inscription, indicating that the addition of a text by Šutruk-Naḫḫunte I was not systematic. Another variable is the appearance of the name of the sovereign initially represented. Indeed, while some of the secondary Elamite inscriptions mention the king who was originally depicted, in other examples a blank space was left, which was almost certainly intended to be filled in with the name of the king, but this was never carried out. These vacats provide a very interesting insight into the process by which stone inscriptions were made. The addition of an inscription, including the name of the sovereign represented by the statue, indicates that there was a possible attempt—and that an effort was made—to read the original inscription. Furthermore, it is essential to remember that the aforementioned rulers were not those who were ruling in the looted cities. As can be seen from the table below, the rulers identified lived between ten and five centuries before Šutruk-Naḫḫunte I.

Table 1. Statues deported to Susa from Mesopotamia,30 with details of the inscriptions (ordered according to the chronology of the rulers initially represented). Based on Melissa Eppihimer’s work (Eppihimer 2010). [? = information unavailable; no = sufficiently preserved to confirm absence; / = not sufficiently preserved to confirm absence].

Museum Number(s)	(H × L × Th)	Sovereign	Reign Dates	Origin	Original Inscription	Secondary Inscription
Sb 47 + Sb 9099	100 × 58 × 48 cm	Maništusu	23rd century BC	Akkad (?)	/	Claim of the deportation by Š.-N.
Sb 51	20 × 31 × 36 cm	Maništusu		Babylonia	Extract of the standard inscription	/
Sb 15566	H: 5 cm	Maništusu		Babylonia	Extract of the standard inscription	/
Sb 49 + Sb 50 + Sb 9097	99 × 100 × 56 cm	Maništusu		Akkad (?)	/	Claim of the deportation by Š.-N.
Sb 52	59 × 47 × 64 cm	Naram-Sîn	2261–2206 BC	Babylonia	Dedication of the statue	/
Sb 53	18 × 15 × 11 cm	Naram-Sîn (by Šu’āš-takal)31		Babylonia	Dedication of the statue	/
Sb 57	50 × 23 × 14 cm	Ur-Ningišzda	20th century BC	Ešnunna	Dedication of the statue, deliberately erased	Claim of the deportation by Š.-N.
Sb 56	62 × 26 × 17 cm	?	Early 2nd mill. BC	Ešnunna	Illegible, deliberately erased	Claim of the deportation by Š.-N
Sb 58	40 × 21 × 26 cm	?	Early 2nd mill. BC	Ešnunna	Illegible, deliberately erased	Claim of the deportation by Š.-N.
Sb 61	89 × 52 × 55 cm	?	Early 2nd mill. BC	Ešnunna	Illegible, deliberately erased	Claim of the deportation by Š.-N.
Sb 5932	24 × 17 × 19 cm	?	Early 2nd mill. BC	Ešnunna	/	Claim of the deportation by Š.-N.
Sb 85	20 × 11 × 7 cm	?	Early 2nd mill. BC	?	no	no
Sb 141	30 × 33 × 20 cm	?	Early 2nd mill. BC	Babylonia	/	/
Sb 163	31 × 27 × 18 cm	?	Early 2nd mill. BC	Ešnunna	/	Claim of the deportation by Š.-N.
Sb 95	15 × 9 × 11 cm	Hammurabi (?)	1792–1750 BC (?)	Babylonia	/	/
Sb 60	66 × 35 × 20 cm	?	?	?	/	Claim of the deportation by Š.-N.

Table 1. Statues deported to Susa from Mesopotamia,30 with details of the inscriptions (ordered according to the chronology of the rulers initially represented). Based on Melissa Eppihimer’s work (Eppihimer 2010). [? = information unavailable; no = sufficiently preserved to confirm absence; / = not sufficiently preserved to confirm absence].

Museum Number(s)	(H × L × Th)	Sovereign	Reign Dates	Origin	Original Inscription	Secondary Inscription
Sb 47 + Sb 9099	100 × 58 × 48 cm	Maništusu	23rd century BC	Akkad (?)	/	Claim of the deportation by Š.-N.
Sb 51	20 × 31 × 36 cm	Maništusu		Babylonia	Extract of the standard inscription	/
Sb 15566	H: 5 cm	Maništusu		Babylonia	Extract of the standard inscription	/
Sb 49 + Sb 50 + Sb 9097	99 × 100 × 56 cm	Maništusu		Akkad (?)	/	Claim of the deportation by Š.-N.
Sb 52	59 × 47 × 64 cm	Naram-Sîn	2261–2206 BC	Babylonia	Dedication of the statue	/
Sb 53	18 × 15 × 11 cm	Naram-Sîn (by Šu’āš-takal)31		Babylonia	Dedication of the statue	/
Sb 57	50 × 23 × 14 cm	Ur-Ningišzda	20th century BC	Ešnunna	Dedication of the statue, deliberately erased	Claim of the deportation by Š.-N.
Sb 56	62 × 26 × 17 cm	?	Early 2nd mill. BC	Ešnunna	Illegible, deliberately erased	Claim of the deportation by Š.-N
Sb 58	40 × 21 × 26 cm	?	Early 2nd mill. BC	Ešnunna	Illegible, deliberately erased	Claim of the deportation by Š.-N.
Sb 61	89 × 52 × 55 cm	?	Early 2nd mill. BC	Ešnunna	Illegible, deliberately erased	Claim of the deportation by Š.-N.
Sb 5932	24 × 17 × 19 cm	?	Early 2nd mill. BC	Ešnunna	/	Claim of the deportation by Š.-N.
Sb 85	20 × 11 × 7 cm	?	Early 2nd mill. BC	?	no	no
Sb 141	30 × 33 × 20 cm	?	Early 2nd mill. BC	Babylonia	/	/
Sb 163	31 × 27 × 18 cm	?	Early 2nd mill. BC	Ešnunna	/	Claim of the deportation by Š.-N.
Sb 95	15 × 9 × 11 cm	Hammurabi (?)	1792–1750 BC (?)	Babylonia	/	/
Sb 60	66 × 35 × 20 cm	?	?	?	/	Claim of the deportation by Š.-N.

2.2.2. Puzur-Eštar [EŞEM 7813] (Figure 6)

As we mentioned in the introduction, this is an excellent example of what can be performed to adapt a pre-existing statue. On 22 April 1914, a pair of life-size statues (H: 1.7 m) were unearthed in Nabuchadnezzar II’s Babylon (Neo-Babylonian ruler, ruling between 605–562 BC). They are preserved in the Eski Şark Eserleri Müzesi in Istanbul under numbers EŞEM 7813 and EŞEM 7814, except for the head of the best-preserved statue, which is in the Vorderasiatisches Museum in Berlin under inventory number VA 08748. Dated to the 20th century BC, they are identified as representations of Puzur-Eštar, thanks to the inscription on EŞEM 7813. This statue is also the one featuring a pair of horns, a divine attribute in Mesopotamian iconography. The presence of this feature led scholars to consider for some time that it was a representation of a deified ruler. A representation of a god would have had to feature more than a simple row of horns, and two-dimensional examples of deified kings were discussed as parallels (Nagel 1959, pp. 262–65).

The inscription on the right forearm identifies their dedicators as being Puzur-Eštar, šakkanakku of Mari, as well as his brother, Ṣilla-Akka,33 while that on the lower garment identifies another dedicator, Tūra-Dagan, also šakkanakku of Mari and father of Puzur-Eštar (Gelb and Kienast 1990, pp. 363–64; Colonna d’Istria 2022, p. 180). It is suggested that the inscription on the forearm was added later, perhaps after the father’s death, by the two sons (Nassouhi 1926, p. 109). The other statue, EŞEM 7814, also featured inscriptions, but they were damaged, like most of the statue itself. Those artefacts were found in what was once thought to be a palace-museum, but is rather a place where “monuments fulfilling both an information and a propaganda function” (Joannès 2011, p. 118) are concentrated. In the Nabuchadnezzar II case specifically, the purpose seems to have been to expose war spoils. Their exposition seems to have been more intended as an effort for public demonstration of power, or even as a magical assemblage aiming to diminish or humiliate the enemies’ gods, rather than an exhibition accessible to all to educate or decorate, as a museum would do (Klengel-Brandt 1991, p. 45). In parallel, the addition of the horns, which turn out to be a modification of the pre-existing headdress of the šakkanakku, have been part of a desire to deify this ancestor, or rather because the identity of the character had been lost.

Figure 6. Left: Statue of Puzur-Eštar [EŞEM 7813]. © Laurent Colonna d’Istria (2007). Right: Statue of Puzur-Eštar [EŞEM 7814]. © Laurent Colonna d’Istria (2007).

Figure 6. Left: Statue of Puzur-Eštar [EŞEM 7813]. © Laurent Colonna d’Istria (2007). Right: Statue of Puzur-Eštar [EŞEM 7814]. © Laurent Colonna d’Istria (2007).

2.2.3. Puzur-Sušinak34 (Figure 7)

Françoise Tallon published an article in 1993 in which she suggested that several of the artefacts (table Sb 17, limestone lions Sb 98 and b 99, pebble Sb 6) attributed to Puzur-Sušinak (ruler of Elam, ca 2100–2050 BC), though—potentially, as the depths are not systematically documented—discovered in a more recent level, may have been a reuse (Tallon 1993, p. 107). She includes in this list the statue Sb 48, discovered in the area of the temple of Insušinak. According to her, this sculpture is also a representation of Puzur-Sušinak rather than a sovereign of the Akkad dynasty, deported by Šutruk-Naḫḫunte I, as is commonly accepted (Tallon 1993, pp. 105–6). In this scenario, this representation of the Elamite ruler may have been recovered as a pivot stone, given the traces it bears on the front (Tallon 1993, p. 108). The stratigraphic argument is valid: the statue Sb 48, when compared to some looted items, is neither in the same place (temple area versus acropolis), nor at the same level (3.2 m above level II as opposed to the 6 m to 7 m of some of the deported objects). However, the context of the objects deported by the Šutrukids (Elamite dynasty, from the 13rd and 12th centuries BC) is not systematically documented (Harper 1994, p. 161), and the statue Sb 48 is more similar to Akkadian productions than to surviving representations of Puzur-Sušinak. Javier Álvarez-Mon, based on this argument, proposed an Akkadian inspiration for this image of Puzur-Sušinak (Álvarez-Mon 2020, p. 278).

In the end, this sculpture was reassigned, either as a result of a deportation from Mesopotamia, when it became part of the spoils of war, or as a result of a change of context within the city of Susa itself. Indeed, the traces it bears most certainly testify to wear related to a use that is difficult to identify, the pivot stone being one of several possibilities. The other way round is also possible, since Sb 55 (H: 84 cm) is often considered to have been usurped by Puzur-Sušinak, as the type of clothing appears older, probably from the period of Maništušu (rigid, formal, wavy folds) (Strommenger 1959, p. 37; Amiet 1972, p. 103). Although J. Álvarez-Mon also acknowledges the foreign features—mainly the presence of the sandals—he is again more inclined to the theory of Akkadian inspiration, rather than the result of looting (Álvarez-Mon 2020, p. 281). When the initial origin of the representation is identified by an inscription and differs from the place where it was found, the study is more straightforward. This is the situation presented next.

Figure 7. Left: Statue of Puzur-Sušinak [Sb 48]. Courtesy of the Musée du Louvre (Raphaël Chipault 2023). Right: Statue of Puzur-Sušinak [Sb 55]. Courtesy of the Musée du Louvre (Raphaël Chipault 2023).

Figure 7. Left: Statue of Puzur-Sušinak [Sb 48]. Courtesy of the Musée du Louvre (Raphaël Chipault 2023). Right: Statue of Puzur-Sušinak [Sb 55]. Courtesy of the Musée du Louvre (Raphaël Chipault 2023).

2.2.4. Gudea

Statues of Gudea (ruler of Lagaš, ruling between ca 2130–2110 BC) were centralized in Girsu (present-day Tello, Iraq) by Adad-nādin-aḫḫe (Seleucid ruler, from the 3rd century BC). Gudea has passed through history mainly thanks to his large production of statues representing himself (twenty-six in total, whose identification is the subject of current consensus, labelled from A to AA, minus the letter L) (Colbow 1987). Another striking feature is that almost all are manufactured in diorite. In the case of those bearing—and having preserved—an inscription, one understands that they were originally intended to stand in a temple next to a deity (Suter 2000, pp. 57–59). The divinity is not always the same depending on the statue, and the size of the latter can vary considerably. It is therefore quite clear that these statues were not designed to operate together.

However, most of those sculptures, in the round, were discovered grouped in an installation made nineteen centuries after they were produced in Gudea’s time. According to Sebastian Rey, this initiative was part of the Seleucids’ political desire to spread the cult of ancestors (Rey 2020, pp. 78–79). To achieve this, it does not appear to have been necessary to modify the statues or even to add an inscription testifying to the reassignment. Only the rest of the archaeological context made it possible to identify Adad-nādin-aḫḫe.

2.2.5. Akkadian Statuary

Some five hundred years after the demise of the Akkadian Empire (ca 2330 to 2100 BC), representations of 23rd century BC rulers Sargon of Akkad, Rimuš, and Maništušu were still on display. Archaeologically, no evidence of this assemblage has been uncovered. However, texts do tell us that these statues existed in the courtyard of the Ekur (Enlil temple in Nippur). Indeed, Old Babylonian tablets discovered at Nippur (present-day Nuffar, Iraq) are copies of the inscriptions on these Akkadian statues35 installed in the open air (Spycket 1968, pp. 44–45; Gelb and Kienast 1990; Buccellati 1993; Thomason 2005, p. 100). These copies are accompanied by short ekphrasis. Those sources allow us to determine that the function and location of the monuments differed from what they had originally been produced for, as they were assembled in the courtyard of the Ekur as witnesses to a glorious past (Feldman 2009, p. 42)—or to avoid offending the gods (Eppihimer 2019, p. 15). As such, although they were reused in a temple, and perhaps had the status of ex-voto, they fulfilled at the very least the additional function—inevitably attributed to them in a second phase of their existence—of witnesses to the past.

In this situation, we suggest using the term ‘collection’, not in the sense of some sort of museum, which would be a misnomer, but in order to refer to groupings of monuments that were intended to expose traces of the past. The purpose of these ‘collections’ was manifold: to pass on memories, to establish a political tradition, and so on. The practice of collecting has been documented over almost three millennia of Mesopotamian history (Thomason 2005, p. 9). On the other hand, the gathering of statues in a given place is often difficult to distinguish from a purely cultic installation of orants within a temple. Indeed, although difficult to identify, these assemblages may have been used for memorial worship.36

Certainly, in this case in particular, the process of reattribution is vague. What is undeniable is that having survived for almost half a millennium, these artefacts were no longer perceived in the same way as when they were first erected. Although little information has come down to us, mainly because of the lack of associated archaeological remains, there is no doubt that this installation was remodelled over time. However, the impossibility of identifying and quantifying these changes greatly reduces the scope for further study.

This desire to preserve the memory of an original function or context, that is to say of a previous life of the statue, is the major difference from the practice of usurpation. As we have seen, although Šutruk-Naḫḫunte I was most likely responsible for erasing the original inscriptions, he preserved the memory of the sovereign represented via his own dedication. In the case of statues reinstalled in the 1st millennium, such as Puzur-Eštar or Gudea, it is not certain that the memory maintained is specifically that of the sovereign represented, but rather that of the ancestor and what he represents. On the other hand, cases such as that of Puzur-Sušinak—because the reconstruction of its history is complex—do not allow us to clearly identify the motivations.

3. Conclusions

This paper aimed to demonstrate the power of statues through the study of investment in their reworking to make them convey a new message. As a complex artefact, the statue allows stories and ideas to coexist. Their reuse illustrates their capacity to adapt; in addition to the efforts made by societies to change their status or function, there is a belief in their intrinsic malleability. If its context influences the statuary, it also influences its environment.

In addition to reuse, the significance of the image is also evident in other actions carried out on the statues. As we have seen, the effort invested in the deportation reveals the importance attached to the statue by individuals outside the community that produced it. The mutilation of statues, mostly carried out by enemies of the community as well, could also be explored in greater depth (Brandes 1980). It would demonstrate that the opinion of the target community was being considered: there is no need to be convinced of the agency of a statue in order to damage it; it is enough to know that its ‘owner’ will be disempowered when confronted with this violence. When faced with their damaged statues, communities either repair them or get rid of them. However, in both cases, the practice must be codified and supervised—in theory anyway. The mīs pî ritual mentioned in the introduction provides information on the principles to be observed if a divine image is to be repaired or disposed of by returning it to the deities (Walker and Dick 2001, pp. 228–29). Reuse, examined as an isolated phenomenon in this paper, may have been a way of keeping a statue active and within the community that produced it. However, the change of function, or even context, implied by this practice puts it more likely on the enemy side of the statue: an identity is replaced, appropriated, or imposed. As we have seen with cases such as the Gudea corpus (cf. Section 2.2.4) and the statue of Sargon of Akkad restored by Nabonidus (cf. Section 2.2), reuse can sometimes be part of a process of respect and commemoration. As such, while practices can easily be classified as respectful (burial, reparation) or disrespectful (deportation, mutilation), reuse is an isolated case that, depending on the methods and motivations, can slide from one end of the spectrum to the other.

Combining usurpation—as the practice of appropriation of a statue—and reattribution—as the alternative of the coexistence of identities—very few attestations are identified. As we have seen, it may be due to the poor attitude towards reuse; since curse formulas guard against reuse, how wrong was it considered? Furthermore, as far as the actors of these processes are concerned, the new owners faced specific issues as they could not conduct the commission process as a usual one. Indeed, the latter depended on criteria chosen by the previous owner of the image, sometimes long before. As for the sculptor, his role in the documented examples was mainly related to the inscription: to erase and/or add. Were the new owners interested in the statue’s anteriority? What message did the choice to usurp send? As the summary of this special edition points out: ‘Commissioners simultaneously innovated and looked to the past, sometimes with the same social goals in mind’.

In the end, the power of statues, or agency, is also perceptible in their ability to transmit and make explicit the innovations and archaisms of the civilizations in which they evolved, down to the present day.