Additional services for Cambridge Archaeological Journal:
Commercial reprints: Click here Terms of use : Click here
Email alerts: Click here Subscriptions: Click here
Cambridge Archaeological Journal
http://iournals.cambridae.org/CAJ
Memory and its Demolition: Ancestors, Animals and Sacrifice at Umm el-Marra, Syria
Glenn M. Schwartz
Cambridge Archaeological Journal / Volume 23 / Issue 03 / October 2013, pp 495 - 522 DOI: 10.1017/S0959774313000504, Published online: 24 October 2013
Link to this article: http://iournals.cambridae.org/abstract S0959774313000504 How to cite this article:
Glenn M. Schwartz (2013). Memory and its Demolition: Ancestors, Animals and Sacrifice at Umm el-Marra, Syria. Cambridge Archaeological Journal, 23, pp 495-522 doi:10.1017/S0959774313000504
Request Permissions : Click here
Downloaded from http://journals.cambridge.org/CAJ, IP address: 131.111.124.247 on 27 Nov 2013
Memory and its Demolition: Ancestors, Animals and Sacrifice at Umm el-Marra, Syria
Glenn M. Schwartz
At Umm el-Marra in western Syria, a sequence of Bronze Age ritual installations facilitates the investigation of how Syrian elites employed memory, ancestor veneration, and animal (and -perhaps human) sacrifice to reinforce their position, and how others used countermemory to contest it. Relevant data derive from an Early Bronze Age complex of elite tombs and animal interments and a Middle Bronze Age monumental platform and shaft containing animal and human bodies deposited ritually. Analysis of the spatial landscape, with patterns of access or inaccessibility, facilitates additional insights, as does the consideration of the intentionality or lack of it in ancient references to the past.
Ongoing since 1994, archaeological research at Umm el-Marra, an early urban centre of western Syria, has acquired a rich data set on elite mortuary behaviours and subsequent ritual activities localized on the site acropolis. Results stem from a large complex of elite tombs as well as installations for the burial of animals in the mid/late Early Bronze Age (EBA) (c. 2550-2200 bc), and a monumental platform and shaft containing the bodies of ritually killed humans and animals in the Middle Bronze Age (MBA) (c. 1900-1600 bc). From the associated data and their interpretation, we derive new insights on how elites of Syrian complex societies employed memory, ancestor veneration and sacrifice to legitimate their authority, and how others contested such ideologies. In the following discussion, I present the evidence for this centrally situated, ritualized behaviour in Early and Middle Bronze Age Umm el-Marra and discuss its relevance — or lack thereof — to social memory and the uses of the past.
How people in past societies perceived their own past and manipulated it to accomplish aims in their own present has been a subject of extensive interest in recent archaeological research (Van Dyke & Alcock 2003; Yoffee 2007; Mills & Walker 2008; Boric 2010; Lillios & Tsamis 2010). Through reference to past events and individuals, present-day concerns are communicated, negotiated and contested. Although attention to the uses of the past has been extensive,
scrutiny has most often focused on how the past was employed to legitimize and naturalize power and authority. Less common have been attempts to study how perceptions of the past were contested and modified in order to change people's understandings of reality (e.g. Crawford 2007; Nielsen 2008, 210ff.). In this article, I am concerned with how a constructed past was offered and transmitted materially, and how efforts were made to deconstruct it — on forced forgetting, the erasure of memory (Connerton 1989,15).
The perception of the past on the part of large numbers of people is often termed 'social memory' (Chesson 2001; Van Dyke & Alcock 2003). This variety of memory refers to conceptions of what happened in the past held in common by a group of people in a given space and time: 'the construction of a collective notion (not an individual belief) about the way things were in the past' (Van Dyke & Alcock 2003, 2). It differs from 'collective memory' (Halbwachs 1992) in its emphasis on the role of individual actors in the making of memories (Olick & Robbins 1998; Mills & Walker 2008). Social memory need not involve the remembering of past events that were bodily experienced by an individual, although such bodily memories can be involved (Halbwachs 1992; Jonker 1995; Moshenska 2010).1 Instead, social memory entails understandings of the past that people share with one another, regardless of their personal experience of the events or lack of it.
Cambridge Archaeological Journal 23:3, 495-522 © 2013 McDonald Institute for Archaeological Research doi:10.1017/S0959774313000504   Received 24 Jan 2013; Accepted 13 Mar 2013; Revised 19 Apr 2013
Glenn M. Schwartz
The uses of sodal memory to accomplish political goals have been discussed extensively in recent archaeological literature (Mills & Walker 2008; Yoffee 2007; Roddick & Hastorf 2010; Schortman & Urban 2011). Particularly influential has been the work of Connerton (1989), who refers to the inscription (e.g. through monuments, written texts) and incorporation (through habitual, bodily practices) of social memory, although scholars have pointed out that the two categories are by no means discrete and can manifest substantial overlap (Hamilakis 2010,191-2). Modifying this model, Van Dyke (2009) proposes a typology of discursive (intentional) as opposed to practical (habitual) social memory.
Recently, some have cautioned against an archaeological overreliance on social memory (Herzfeld 2004; Berliner 2005; Van Dyke 2009; Moore 2010).
Such critics fault the tendency to see every instance of continuity as memory and every instance of change as erasure of memory. Van Dyke (2009) proposes a solution to this problem by stipulating that social memory is best studied when it entails self-aware, intentional reference to the past. The inclusion of intentionality can effectively distinguish social memory studies from those of tradition or culture change.
Umm el-Marra in the third and second millennium bc: early societal complexity in western Syria
Located in the Jabbul plain of western Syria between Aleppo and the Euphrates valley (Fig. 1), Tell Umm el-Marra is a site of some 20-25 hectares occupied primarily in the third and second millennia bc (the Bronze Age), when Syrian complex societies experi-
496
Memory and its Demolition: Ancestors, Animals, and Sacrifice at Umm el-Marra, Syria
enced their formation and early centuries of development (Curvers & Schwartz 1997; Schwartz et al. 2000; 2003; 2006; 2012).2 Although far larger than any other Bronze Age mound in the vicinity, Umm el-Marra is considerably smaller than the major cities of early urban Syria such as Ebla (c. 56 ha) and Leilan (90 ha) (Schwartz et al. 2000). As a result, we have interpreted Umm el-Marra as a regional centre subservient to larger and more powerful cities such as Ebla or Aleppo. If Umm el-Marra was ancient Dub or Tuba (Matthiae 1979; Schwartz et al. 2003; Schwartz 2010, 376, n. 3), it would have been head of a small kingdom in the Early and Middle Bronze Ages (mid-late third to early-mid second millennia bc). Given these data, work at Umm el-Marra permits the study of a large but 'second tier' settlement and thus provides a complement to research from the amply studied primary
centres. Whether Umm el-Marra is best interpreted as a political and economic centre, or a ritual centre, is explored below.
Topographically (Fig. 2), Umm el-Marra is a relatively low mound (maximum height c. 8-9 m) whose main features are fortifications ringing the edges of the site and a low but distinct 'acropolis' or higher mound in the south centre. The majority of the research discussed in this article focuses on the site acropolis, which has the lengthiest occupation sequence and the most salient evidence of specialized public activities.
The main periods of site occupation at Umm el-Marra coincide with the emergence of large-scale societal complexity in Syria and its early episodes of centralization and decentralization. In the mid-to-late third millennium bc, states, cities, monumental architecture and intensified economic specialization
497
Glenn M. Schwartz
appeared across Syria and upper Mesopotamia in a process sometimes referred to as the 'second urban revolution' (Akkermans & Schwartz 2003, 233-87; Ur 2010). According to the Palace G cuneiform archives from Ebla in western Syria, elites employing aspects of both exclusionary and corporate strategies (Blanton et al. 1996) presided over the newly developing urban societies, with kings, queens and 'elders' overseeing an extensive administrative organization and engaging in frequent inter-polity warfare (Archi & Biga 2003; Matthiae 2008; Porter 2012a).
By the latter centuries of the third millennium, evidence of urban and state disintegration appears throughout Syria and upper Mesopotamia. Although the concept of 'collapse' has been critiqued (Aimers 2007; McAnany & Yoffee 2010; Middleton 2012) and it is clear that some urban centres persevered during this period, major transformations involving decentralization and deurbanization were patently underway (Kuzucuoglu & Marro 2007; Laneri et al. 2012; Schwartz 2007b). Umm el-Marra may have been abandoned for as long as three centuries during this period (Schwartz et al. 2012).
After this period of instability, powerful new polities such as Yamhad, Qatna and Shamshi-Adad's upper Mesopotamian kingdom emerged in the early second millennium bc (Middle Bronze Age). These entities are distinguished from those of the third millennium, among other ways, in that the main political actors are identifiable as Amorites, an enigmatic group that can be defined in ethnic or other terms (Charpin 2004; Fleming 2004; Jahn 2009; Porter 2012a). Archaeological study of the origins and workings of these 'second generation' states has only recently commenced, with one focus being the regeneration of complex societies after periods of decentralization (Schwartz & Nichols 2006; Ristvet 2012a).
The evolution of an Early Bronze Age elite mortuary complex
Umm el-Marra was founded in the early third millennium bc. By at least c. 2500 bc, the centre of the site acropolis was devoted to a mortuary complex associated with elite, possibly royal, individuals. Study of this complex has allowed for documentation of elite mortuary practices and their socio-political implications, patterns of mortuary treatment relative to gender and age, and the mortuary role of animals (Schwartz 2007a; 2012b). In this section, I present a summary of the main results from the complex, a proposal for the history of its development, and a discussion of the diverse ways that social memory was employed in that history.
Human tombs, animal tombs
As exposed thus far, the Umm el-Marra mortuary complex included ten tombs3 containing human interments, ten features ('installations') devoted to the burial of equids located centrally in the midst of the tombs and additional structures. Given available stratigraphic and artefactual evidence, I propose that the tombs and nearby installations were built and employed sequentially from c. 2550 to 2200 bc (EBA III-IVB/Umm el-Marra late period VI to period IV).
The tombs were rectangular and usually had an entry way on the east that had been blocked with stones.4 Most had a substructure of limestone boulders and a superstructure of mudbricks; the tombs were at least partly if not completely above ground.5 Some of the tombs had remarkably well-preserved remains that included human skeletons inside the vestigial remains of coffins. These were often accompanied by ornaments of gold, silver and lapis lazuli suggesting the elite character of those interred. At the same time, nearly all tombs suffered some degree of disturbance and robbery.
In general, we can observe significant commonalities in the material constitution of the complex. Most tombs, with the exception of those in the latest phase (see below), consisted of one-room rectangular structures of mudbrick above stone. Within the tombs, the human bodies were enclosed in coffins and accompanied by objects such as beads, pendants, toggle pins, torques, daggers and pottery. These data imply that there was a common understanding of the proper way to build and outfit an elite tomb. At the same time, variability among all these factors indicates the propensity to display a degree of individuality (Torres-Rouff et al. 2012). For example, Tomb 8 had two rooms, not one, and unique objects such as a slotted bronze spearhead or ivory handle are attested in individual tombs.
When human bodies interred in the tombs could be sexed, adult females usually were associated with more personal ornaments made of costly materials than adult males (Batey 2011; Schwartz et al. 2006). Despite this gender-based differentiation, object types were not usually restricted to one sex or the other (for a similar result in southern Mesopotamian elite tombs, see Torres-Rouff et al. 2012).
Not only were human tombs present, but the complex also included separate interments for animals. Oriented north-south in the centre of the complex is a line of subterranean installations that contained the skeletal remains of equids, other animals and human infants. The equids are often arranged symmetrically, as if to present a tableau, most likely a team intended to pull a vehicle. Jill Weber's work on the zoo-
498
Memory and its Demolition: Ancestors, Animals, and Sacrifice at Umm el-Marra, Syria
archaeological data has revealed that the equids are all male, likely to be donkey-onager hybrids, and are very probably to be identified with the highly-valued kunga (anse-BARAN) equids mentioned frequently in third-millennium Syro-Mesopotamian written sources (Weber 2008; 2012). While diverse interpretations of these installations have been advanced (Schwartz 2012a; Weber 2012), it is at least likely that the presence of elite animals was intended to illustrate the nearby humans' lofty social position and wealth. It may also be the case that the equids were expected to provide transportation for the deceased elite individuals in the afterlife. If so, the ritual killing of younger equids can be understood as 'retainer sacrifice' in which living beings are killed to serve high-ranking persons beyond the grave (Schwartz 2012a), while the interment of aged individuals who died a natural death may signify the inclusion of treasured steeds for use in the afterlife. The presence of the human infants in the installations is more difficult to interpret, and it is not clear if they had been killed or had died a natural death (Schwartz 2012a).
Drawing on patterns in architecture and the age of the interred equids, Weber (2008; 2012) has identified four types of equid installations. Type I installations, three in number (A, E and F), consist of a subterranean mudbrick or stone structure containing four young or prime aged equids suggestive of a team of sacrificed animals (on the problems of the definition of sacrifice and its archaeological recognition, see Schwartz 2012a). The three Type II installations (B, C and D) are mudbrick structures with two compartments, each containing a standing, aged equid facing west in addition to a spouted jar and remains of at least one human infant. Only one installation
Figure 3. Acropolis Centre mortuary complex, phase 1 (EBA IWUmm el-Marra VI later, c. 2550-2450 bc) (schematic plan, T = tomb, R = room, S = shaft, MP = mudbrick platform).
(G) falls in the Type III category, which entails four equids each buried in two sequentially deposited pits. Finally, Type IV refers to three individual skeletons found outside of other features, located against the east wall of Tomb 8, south of Installation E, and in a pit beneath Installation G.
The mortuary complex at Umm el-Marra was not a static entity but changed through time. New tombs were added adjacent to pre-existing tombs, expanding in a horizontal fashion, until the latest period of use, when new tomb constructions were dug into earlier ones. Likewise, equid installations were either added near earlier installations or were placed above them. In the following discussion, I endeavour to present an understanding of the evolution of the complex and its immediate environs, reviewing how the mortuary landscape and its visual or emotional effect, its accessibility or openness, changed through time. The four proposed phases are correlated to the Early Bronze Age relative chronology currently in use in western Syria (Matthiae 1981; Akkermans & Schwartz 2003).
499
Glenn M. Schwartz
Figure 4. Tomb 8. Looking east.
Phase 1: the original line of tombs — Early Bronze Age III (Umm el-Marra late VI)
In the earliest period (perhaps c. 2550-2450 bc),6 three tombs, 5, 6 and 8, were established in a southwest-northeast line across the centre of the acropolis, with Tomb 6 at least partly enveloped by structures suggesting an enclosure wall (Fig. 3).
Tomb 6: Given their ceramic contents, it is most likely that either Tomb 6 or 5 was the earliest of the tombs. I have hypothesized that Tomb 6, the largest of all the Umm el-Marra tombs (c. 9.6 x 4.5 m), was the earliest, because of its size and central location (Schwartz 2012b). Also indicative of the special status of this tomb are the constructions to its west and south that enclose the feature. We might construe Tomb 6 as the tomb of the founder of a new social order who became an ancestor of extraordinary significance (Helwing 2012).
Although a large segment of Tomb 6 had been destroyed by the subsequent construction of Tomb 7, still extant were the skeletal remains of an adult male (age probably 45-50 years) partly inside the vestiges of a wooden coffin (Batey 2011 and pers. comm.). Among
the associated artefactual materials were gold and silver toggle pins and beads of lapis lazuli, gold and carnelian (Schwartz et al. 2006).
Tomb 5: Just east of Tomb 6, Tomb 5 was much disturbed but contained the bones of an adult male and an infant (Schwartz et al. 2006).
Tomb 8: West of Tomb 6, Tomb 8 (Fig. 4) is a substantial two-room structure (Schwartz et al. 2012). Remains of an infant and two adult males that were found in the western room probably had been removed from two superimposed coffins whose impressions and bitumen coating were identified in the eastern room.
Equid installations: Between Tombs 8 and 6 were two Type I equid installations (Schwartz et al. 2012)7 Installation F, the earlier of the two, was a stone construction. Installation E (Fig. 5) included four standing equids whose fore and hind limbs had been placed in eight compartments, while their skulls rested on a ledge.8 Toe bones from an additional two equids were also found.
500
Memory and its Demolition: Ancestors, Animals, and Sacrifice at Umm el-Marra, Syria
Figure 5. Installation E. Looking northwest.
Outside the tombs: Excavations beyond the phase 1 tomb area reveal architectural remains with evidence of ritualized activity. Northwest of Tomb 6 were two rooms inside a curvilinear wall, the southern, burned example having the skeleton of a puppy in the southwest corner, another puppy in the northeast corner, and a piglet in the southeast corner. An additional puppy skeleton was located in the southeast corner of the large area west of the burned room, whose south wall contained a human infant inside it. Likewise, the burned space west of Tomb 6 and pre-dating Installation E included a puppy skeleton and two puppy skulls. Such interments could be linked ritually to the animal and infant inclusions in the equid installations and to those that superseded them in later periods. The burning of these rooms with animals in situ might be interpreted as the result of a ritual closure of the space (Oates et al. 2001, 41; Way 2011,138; Walker et al. 2000).
Northwest of Tomb 5, a rectangular platform of mudbricks might have played a role related to the rituals taking place in the tomb proper. To the west and southeast of the tomb area is evidence of multiple
architectural phases of small-scale architecture now attested by fragmentary stone substructures.
Phase 2: tomb additions to north and south — Early Bronze Age IVA (early) (Umm el-Marra early V) In the next phase of activity in the mortuary complex (perhaps c. 2450-2350 bc), tombs and associated features were added to the north and south of Tomb 8, forming a line of early EBA IVA mortuary structures perpendicular to the original southwest-northeast line of the phase 1 (EBA III) tombs (Fig. 6).9 Other tombs were added south of Tomb 6.
Tomb 4: South of Tomb 8 is Tomb 4 (Schwartz et al. 2006), containing two layers of funerary deposits, the earlier of which is datable to this phase on the basis of its ceramic contents. Included were a disturbed adult female and male, both apparently primary burials, and the secondary interment of an adult female (Schwartz et al. 2006). Accompanying them were artefacts of gold, silver, bronze and ivory, numerous vessels, and miniature basalt tables, the latter perhaps used for processing cosmetics.
501
Glenn M. Schwartz
N
A
6    8 10m
-CUT
Tomb 3: Tomb 3 is situated north of Tomb 8 (Schwartz et al. 2006). Although the tomb suffered disturbance, the remains of numerous ceramic vessels were still present, as were the bones of an adult and an adolescent.
Tomb 10: Partly demolished by Late Bronze Age intrusive activities, Tomb 10 to the east of Tomb 4 included at least two individuals (Schwartz et al. 2012). Given the location of Tomb 6 and its southern enclosure wall, Tomb 10 could not have been as large as most of the other tombs and may have had a square shape like Tomb 1 (see below).
Tomb 9: To the southeast is Tomb 9, which had been substantially looted but still contained the remains of at least three individuals (Schwartz et al. 2012). Considering the marginal location of Tomb 9, it is likely that additional tombs or related structures had been built between it and Tomb 6, but the phase 4 (EBA IVB) tombs 7 and 11 obscured or destroyed evidence of them.
Equid installations: I tentatively assign three equid installations to this phase. Installation C (below Tomb 1 on Fig. 6) was a Type II variety.10 To the north was Installation A, an above-ground chamber later converted into a Type I installation. The Type III Installation G, east of Installation A, contained two pits stratified one atop the other, each with a set of four equids and as well as bones from an additional two equids. Linking Installations A and G chronologically and functionally are sherds from a single incised 'cult stand' found in both features.11
Figure 6. Acropolis Centre mortuary complex, phases 2 and 3 (EBA IVAI Umm el-Marra V, c. 2450-2300 bc) (schematic plan, T = tomb, R = room, S = shaft).
Ancillary structures: At least three additional structures might also be related to the mortuary activities of the complex in phase 2. Room 1, south of Tomb 4, duplicates the plan of the latter structure and includes a mudbrick podium against its eastern wall. Another duplicate structure is Room 2, north of Tomb 3. Finally, the 'U'-shaped Room 3 northwest of Tomb 5 contained a bovid skeleton and broken remains of at least two wavy line jugs, perhaps used for ritual purposes (Schwartz et al. 2012). Activities in these structures are likely to have been related to the rituals attending those interred in the tombs and installations nearby.
Also worth noting are incomplete segments of walls in complex sequences of construction in this and other phases (see, for example, in the vicinity of Installation A, Fig. 6). These may have been the remains of temporary emplacements constructed for ritual events that required the building of individual shelters.
502
Memory and its Demolition: Ancestors, Animals, and Sacrifice at Umm el-Marra, Syria
Outside the tombs: Southeast of the tomb complex was small-scale domestic architecture. To the west were small rooms of unclear function.
Phase 3: space limitations and the last equid installations — Early Bronze Age IVA (later) (Umm el-Marra late V) In later EB A IVA (perhaps c. 2350-2300 bc), the upper layer of Tomb 4 was deposited and Tomb 1 was added to the complex (Fig. 6). It is striking that Tomb 1 is constructed in an area otherwise reserved for equid installations, which might suggest that the availability of space for new tombs was growing scarce. Such spatial limitation may also be indicated by the addition of the second layer in Tomb 4.
Tomb 4: In the upper layer of Tomb 4 were the primary interments of an adult female inside the remains of a coffin, a child and the secondary interment of an adult male (Schwartz et al. 2006). As in the lower layer, the woman had more ornaments than the man, including gold, silver and bronze objects.
Tomb 1: The contents of Tomb 1 (Fig. 7) are both the best-preserved and the most resistant to interpretation in the complex (Schwartz et al. 2003; Schwartz 2012a). A small structure, Tomb 1, contained three layers of bodies that had been interred in wooden coffins in two episodes of deposition. In the first interment was an adult of uncertain sex. In the second, two adult males were placed side by side with a baby at some distance, after which two younger women and two babies were put in above the men. While the symmetry of two men, two women and two babies dying simultaneously is itself striking, also notable is the association of the women with far richer and more abundant personal objects than the men, including gold, silver and lapis lazuli jewelry placed either on the body or in a group to the side.
Equid installations: Phase 3 marks the culmination of the construction of equid installations according to the chronology proposed here, with two Type II equid installations added. North of Tomb 1, Installation B is
503
Glenn M. Schwartz
EBIVB Cut
distinctive for the three puppies placed in each of its two chambers. While the relative dating of Installation B could be contested,12 Installation D to the south of Tomb 1 is clearly contemporaneous with the tomb and post-dates Installation C.13
Enclosure wall: Evidence of what could be interpreted as a stone enclosure wall for the complex adjacent to a heap of cobbles (= glacis?) was noted to the east of Tombs 5 and 9, although no trace of such a feature was evident elsewhere. A restriction and limitation of the area available for tomb construction may be implied by the addition of Tomb 1 above the area formerly reserved for equid installations.
Phase 4: ideological disjunctures and the end of the complex — Early Bronze Age IVB (Umm el-Marra IV) Dating to the latter third millennium (c. 2300-2200 bc), the last phase of the complex includes the construction of two new tombs, 7 and 11 (Fig. 8). Unlike the tombs from preceding phases, the new examples were completely subterranean, constructed only of stone, and wrought damage to earlier tombs through their subterranean construction.
Tomb 7: Tomb 7, the earlier of the two, had at least four rooms and included the disturbed remains of three adults and an adolescent (Schwartz et al. 2006; 2012).
Tomb 11: Excavated in 2010, Tomb 11 damaged the architecture of Tomb 7 to its north. The tomb had two chambers with a narrow roofed passageway between them (Fig. 9). Although disturbed, the tomb still con-
Open zone with pits
Figure 8. Acropolis Centre mortuary complex, phase 4 (EBA IVBIUmm el-Marra IV, c. 2300-2200 bc) (schematic plan, T = tomb, R = room, S = shaft).
tained fragmentary remains of two individuals, recon-structible pottery, a gold rosette and a gold perforated (head?) band. A possible architectural comparison can be made to the EBA III—IV monumental Tomb 302 at Jerablus Tahtani on the Euphrates (Peltenburg 1999). This structure consists of a long room with a small chamber protruding to one side, comparable to the two excavated rooms of Tomb 11.
Ash deposit: Phase 4 was distinguished by the deposition of up to one metre of black ash layers in the northeastern part of the complex, located above equid installations from preceding phases. One might suggest that these ash deposits are the result of activities involving ritual purification to prepare for the installation of the EBA IVB phase tombs or to allow for ritual closure of the entire complex (Walker et al. 2000).
Enclosure wall: In addition to tombs in a new architectural style, an important landmark of this phase is the
504
Memory and its Demolition: Ancestors, Animals, and Sacrifice at Umm el-Marra, Syria
Figure 9. Tomb 11, passage between chambers, with roof slabs in situ. Looking southwest.
construction of a rectangular stone wall enclosing the complex. While the west and southern portions were uniformly built and partly sunk into a foundation trench, the northern segment was thinner, had no associated trench, and was difficult to recognize in the northwest. Perhaps the structure was built in different spans of time under the sponsorship of different individuals or factions, rather than as a monolithic enterprise.
Outside the tombs: Domestic architecture with lime-plaster floors and basins and clay ovens was built directly against the outside of the enclosure wall of the complex. To the southeast, domestic architecture existed south of an open area with pits.
General observations: Noteworthy in this latest phase is the apparent disregard, if not outright hostility, toward architectural features from earlier in the history of the complex. Older tombs were damaged through the subterranean construction of new ones, and the subterranean foundations of the enclosure wall damaged the walls of Tombs 8 and 9. Such damage to earlier tombs may have been intentional, displaying an ideological or dynastic disjuncture between the individuals of the EBA IVB tombs and those of earlier tombs. Alternatively, it is possible that some of the earlier tombs were in ruins and covered with soil by EBA IVB times, with their presence no longer recognized.
Whichever interpretation is preferable, ideological changes were clearly taking place in this latest phase of the mortuary complex. The architectural innovations in tomb construction imply concurrent ideological shifts, and the architectural comparanda to Jerablus Tahtani might even signal new dynastic ties to the Euphrates region. Another disjuncture is the failure to build new equid installations.
The creation of memory
The contents, architecture and locational characteristics of the tombs indicate the special status of the persons (and animals) interred in the tomb complex. Located on a relatively high spot in the centre of the site, with large, imposingly built structures that were at least partly if not completely above ground, the complex dominated the community. Similarly, the tombs with their unusually high stone substructures were distinguishable from other architecture in the settlement. The fact that non-infant humans were customarily buried off-site14 only reinforces the special character of the people whose remains were interred in the centre of the town. Considering these factors and the costly artefacts found in the tombs, one may conclude that these persons had enjoyed high social status. Whether they belonged to groups that held political power, were important members of significant kin groups, or were representatives of other major factions (e.g. religious or military
505
Glenn M. Schwartz
specialists), remains to be definitively elucidated (Schwartz 2012b).
Bioarchaeological analysis of the human skeletal material from the tombs conducted by Ernest Batey (2011 and pers. comm.) provides further support for identification of the interred persons as members of an elite. Stable isotope analyses indicate a diet relatively high in meat, and lifestyle reconstruction (musculoskeletal stress markers) has revealed that the tomb residents lived lives with low levels of physical activity. Likewise, palaeopathological analysis suggests that the individuals interred in the tomb complex were not overburdened with infection or other biological stress and that they exhibited good dental health. Dental non-metric trait analysis has shown individuals within specific tombs to have signs of genetic inter-relatedness, supporting the hypothesis that each tomb contains a discrete family group (Schwartz 2007a).
The visibility of the tombs at Umm el-Marra, which were at least partly above ground and located on a high point, implies that they continued to play an active role in life of the community after the death of the interred persons. Given the central and raised location of the tomb complex, the above-ground character of the tombs, occasional inclusions of pottery and other objects found well above the tomb floors, and textual documentation from Ebla, we have proposed that ceremonies of ancestor veneration were practised here (Schwartz et al. 2006; Schwartz 2007a).
The localization of an elite necropolis in the city centre would have had a profound effect on the life of the community. Making the elite dead spatially and visually central to lived experience compelled the living community members to consistently refer their own existences to the dead. One can imagine the town inhabitants going about their daily lives with the constant awareness of the elite mortuary landmarks, incorporating them into their everyday practices, worldview and habitus (Bourdieu 1977; Nielsen 2008; Fleisher & Wynne-Jones 2010).
By putting mausoleums with dead authority figures on the central, high point of the site, the living authorities were conveying a message — that social hierarchy was central and dominant in the life of the community, that it had always existed, and that it should continue to exist. The tombs provided a constant reminder of how the world should work, with elites in the centre overseeing everything. As a result, the tombs inscribed social memory on the landscape and materialized elite ideology (Connerton 1989; Chesson 1999; Mills & Walker 2008; DeMarrais et al. 1996), legitimizing the social order of the present, whose inequalities were presented as natural and inevitable (Hobsbawm 1983; Yoffee 2005,40; Sinopoli
2003). With deceased elite individuals still residing in a community that constantly acknowledged their presence and honoured them periodically, living members of high-status groups would acquire and maintain prestige through their association with the revered dead (McAnany 1995; Salomon 1995; Siegel 2010,305).
The materiality of the tombs had an effect, not only on the minds of the community inhabitants, but on their bodies. The central location of the tomb complex affected movement within the town, obliging people and animals to move around it (Mack 2004; Ristvet 2011).
Animals also played a prominent role in the elite landscape of death, with equids accorded their own tombs (Weber 2012). Similarities between human and animal tombs — e.g. the east-west orientation of bodies, with heads to the west, and the two side-by-side bodies in Tomb 1 and the Type II installations — also indicate comparability between animals and humans and possibility that animals were considered as other-than-human persons (Latour 1993; Hill 2011; Argent 2010; Losey et al. 2011). But such comparability only goes so far — equid installations were small and subterranean, while the human tombs were large and above ground, claiming the greater share of attention and visibility.
Despite the perceptual and experiential centrality and conspicuousness of the tomb complex within the community, our evidence indicates that access to the precinct was limited. The area available for viewing the activities at the tombs was not extensive: wherever excavation has sampled the areas near the tombs, it has exposed evidence of adjacent structures rather than large open areas. Although people could have watched from the roofs of the surrounding houses, there was no open space to accommodate large crowds. This suggests a restricted audience for the activities taking place within the complex, perhaps of elite or specialized individuals (Hodder & Cessford 2004; Hastorf 2007; Peltenburg 2007/2008; Swenson 2011). Such restriction is especially evident in the latter phases of the tomb complex, when an enclosure wall was built.15
Not only was there little space for an audience in the mortuary complex, it is also unlikely that the mortuary complex would have accommodated large numbers of ritual actors, given its congested nature and lack of significant open area. The relatively diminutive size of the ancillary rooms 1-3 also implies a small number of persons operating there.
Restriction can be used to demonstrate superior access to the supernatural on the part of ambitious elites (Inomata & Coben 2006). Edgar Peltenburg (2007/2008) has recently discussed the increasing
506
Memory and its Demolition: Ancestors, Animals, and Sacrifice at Umm el-Marra, Syria
limitation of access to monumental tombs in third-millennium western Syria, arguing for a steady monopolization of contact with the divine world by an elite whose power was continually increasing. Given the raised and central location of the monuments at Umm el-Marra, their restriction would have been all the more potent, since people outside the complex would have seen it and appreciated the important events taking place inside, while being aware that access to these activities was denied them (Ristvet 2011, 6; Porter 2012a, 188).
The creation of countermemory While social memory can be used to support and materialize elite ideology, it can also be contested and reconfigured (Nielsen 2008; Mills 2008; Van Dyke 2009). As Van Dyke and Alcock (2003, 2) observe, social memory is 'variable by gender, ethnicity, class, religion or other salient factors, allowing for a multiplicity, and perhaps conflict, of memories in any society'. Obliteration of constructed pasts has been termed 'forced forgetting' (Connerton 1989,15), while the construction of new memories to supplant older, undesirable ones has been designated 'counter-memory' (Halbwachs 1992; Hendel 2010).
At Umm el-Marra, almost all the tombs had signs of contents removed and/or damaged, with evidence indicating that this occurred during the period of the tombs' use (Schwartz et al. 2012; Schwartz 2012b).16 One might suppose that the main reason for such activity was the acquisition of valuable objects, but evidence of anger and vandalism suggests that other motivations were also involved. Tomb 9, for example, contained a number of large boulders, sometimes with human bone fragments on top of them, as if the boulders were tossed onto the tomb floor indiscriminately and mixed up with the human remains in the tomb, in an act of deliberate desecration.17
The tomb robbers could have been non-elite persons attempting to counter the production of elite social memory, in acts that we might characterize as resistance (Scott 1990; Van Dyke & Alcock 2003, 3). In such cases, the non-elites may have attempted to construct a countermemory, replacing the original conception of well-respected authorities worthy of veneration with a memory of despised and illegitimate figures.
Another interpretation might posit that the damage was done by rival elites intent on discrediting earlier authorities and severing the connection between the living community and the previously revered ancestors (Chase & Chase 2011; McAnany & Negron 2010, 149; Schwartz 2007a). Tampering with or destroying the resting places of enemy ancestors
is attested in later Near Eastern history, when the Assyrian king Assurbanipal boasted of demolishing the tombs of the enemy Elamite kings. Likewise, Lundstrom (2009, 217) has suggested that the brutal destruction of the royal tombs at the Assyrian capital of Ashur was conducted in retaliation. As with the scenario posited above, the original hegemonic narrative is replaced by one that devalues and disparages the interred individuals.
It is notable that the tomb robbers at Umm el-Marra managed to do their work in a central, visible location with impunity. Bearing this in mind, it may be more likely that the desecrators were persons in authority rather than clandestine robbers or resentful malcontents. Since all the tombs except the upper level of Tombs 4 and 1 were disturbed, we might propose that the tombs deposited prior to phase 3 were looted and violated in a single instance. This may well have been done, not by a newly established local authority with grudges against its predecessors, but by a marauding external enemy intent on destroying the tombs of an adversary's forebears.
Whether non-elite resistance or elite competition is involved, the evidence points to a disruption and reconfiguration of memory through intentional destruction or modification of physical objects that materialized and embodied social memory. But a complete erasure of memory was not effected, since the tombs were not covered over and rendered invisible. It appears that the perpetrators of the disturbances aimed to leave the tombs in ruins, preserving the countermemory of the discredited individuals buried there.18 As Crawford (2007,27) notes, 'the destruction of memory becomes more powerful when the traces of this destruction continue to inhabit the visual landscape'.
Van Dyke (2009) and others have usefully pointed out that similarities or continuities from one period to another need not imply the existence of an intentional reference to the past. I would contend that the evidence from Early Bronze Age Umm el-Marra does indicate the existence of ritual activities making explicit allusions to the past, whether constructively or deconstructively. If evidence of ancestor veneration is persuasive for the Early Bronze Age tombs, there is a clear reference to a shared past, and the violation of tombs in the same period suggests, not only plundering valuable objects, but antipathy and intentional desecration of ancestral figures. The same may apply to the EBAIVB innovations in the mortuary complex, with damage done to preexisting tombs — unless the EBA IVB people were unaware of earlier tombs or damaged them for reasons of convenience as opposed to ire.
507
Glenn M. Schwartz
Above the tombs: Middle Bronze Age ritual practices
Monument 1 and its environs
In the Middle Bronze period of the early second millennium bc, evidence of new ritual activity in the Umm el-Marra acropolis centre abounds. As in the EBA, issues of memory uses of the past, and the nature and scale of the intended audience come into play.
After a period of abandonment of at least a century if not considerably more (Schwartz et al. 2012), Umm el-Marra was extensively reoccupied in the Middle Bronze Age, from c. 1900 to 1600 bc (Umm el-Marra period III). In this period, the special character of the Acropolis Centre is once again evident. At the beginning of the Middle Bronze occupation, a large round stone platform of 37-40 metre diameter termed Monument 1 was built above the zone of the third-millennium tombs (Figs. 10 & 11). The function of this monument is likely to have involved large-scale ceremonial activities, given its extensive, flat and raised character. Indeed, the feature resembles a large stage for performance of rituals or ceremonies.
No evidence of a shrine, altar or other structure built atop the monument has been found, although it is possible that later intrusive activity demolished such a structure.
Constructed at the same time as Monument 1 was a circular enclosure wall 1.4 m wide that ringed the central acropolis, with a gate on the north (Fig. 11). As a result, the acropolis was organized into two concentric circles, with Monument 1 in the centre, the enclosure wall outside it and small-scale architecture in between.
The circular shape of the monument makes it unique among Middle Bronze counterparts in Syria, although the larger, rectangular Monument P3 at Ebla, next to Temple P (Temple of Ishtar') can be compared (Matthiae 1997). The circularity of Monument 1 might be interpreted in numerous ways, such as embodying a group identity or manifesting the 'holy mound' of Mesopotamian mythology linking heaven and earth (Porter 2007/2008).19 Such interpretations of the round monument in terms of sacred geography or cosmography (Steadman 2005; Van Dyke 2008) are necessarily speculative.20
508
Memory and its Demolition: Ancestors, Animals, and Sacrifice at Umm el-Marra, Syria
Figure 11. Acropolis Centre, early Middle Bronze Age, c. 1900 bc (schematic plan).
Monument 1 gives the impression of a vast circular stage, a 'theatre in the round'. In such a performance venue, spectators can be afforded an equally good view of the activities taking place in the centre — there were no bad seats, so to speak, implying a communal arrangement deemphasizing hierarchical distinctions among the onlookers (Tri-adan 2006).21 But while the monument's circularity and large scale may imply communal ceremonies, the acropolis enclosure wall, with its narrow gate in the north, signals a restriction of access to this central zone. Likewise, it appears that areas for spectators were limited; houses and other small-scale architecture are adjacent to the structure on the east and west — although, of course, people could observe from house roofs. The only relatively open zone thus far identified is to the north which, tellingly, is near the only means of entry thus far identified.22 Spectators could have entered through the one-metre-wide northern gate in the enclosure wall, whose narrow dimensions allowed for a tight control of who went in and out, and then assembled in the open zone north of the monument.
Considering the relatively small open area and its controlled access point, it is likely that those in charge of the early Middle Bronze Age acropolis aimed to demonstrate their privileged access to the supernatural and to the powers and knowledge it could provide (Inomata & Coben 2006; Demarest 2004,205-7; Gilibert 2011). While communicating elite ideologies to large groups of people can be effective tool, as in the case of 'theater states' (Geertz 1980), the gathering of large crowds can also be dangerous, as crowd emotions can turn in undesirable directions. It is likely, therefore, that the audience was limited to a fairly small and select group, perhaps individuals of relatively high social status or specialized economic or religious function.
As with the tombs of the Early Bronze Age, the materiality of Monument 1 and its district would have significantly affected the lives of the inhabitants and visitors to the community. Occupying a central and raised location, it would have overseen the rest of the town and affected traffic within it, dominating the everyday thoughts and movements of the population. But unlike the Early Bronze Age mortuary complex,
509
Glenn M. Schwartz
the main visual landmark for persons outside the monumental district was not the monument itself, but the wall that enclosed it.23 In this way, the exclusivity of the monumental area was particularly emphasized to those outside.
Monument 1: later history
In the later Middle Bronze Age, Monument 1 and its enclosure wall experienced significant changes (Fig. 12). Some western and southeastern parts of Monument 1 were covered over with ash or soil, while a sizeable segment on the east was removed and filled in with brown soil containing late Middle Bronze Age sherds. In addition, small-scale structures were erected atop the platform or dug into it, including a burned room in the southwest. The open zones north of the monument accumulated occupational debris up to the level of the extant top of the monument, at which point small-scale architecture was built directly against the face of the monument.24
The acropolis enclosure wall suffered even greater maltreatment after its original construction. Indications are that the wall was in use for a relatively short period of time early in the Middle Bronze Age occupation at Umm el-Marra, after which it was dismantled down to its lowest mudbrick courses, and diverse kinds of architecture were built above the ruins. In most excavation areas where the wall was recorded, some three or four Middle Bronze phases were deposited on top of it. The exception is the wall's gateway in the Acropolis North, which was not built over until the Late Bronze period (Umm el-Marra II).
This history indicates that the monument and its enclosure wall underwent substantial modification and even demolition after their initial period of use at the beginning of the Middle Bronze occupation at Umm el-Marra. The enclosure wall was employed for only a short time, and the monument was partly destroyed and submerged under deposits of midden-like material. But even though the structure had been partly cut into and covered over, a new architectural phase was not installed above the monument until the Late Bronze period. Given the evidence from the area to the north, it seems that the structure had become more of a plaza than a raised platform by late Middle Bronze Age.
The transformation of Monument 1 from a raised stage inside a walled, restricted area to a plaza with open access implies political and ideological changes. In the earlier part of the Middle Bronze Age, Monument 1 displayed a distinct differentiation between performers and audience, providing the setting for an event that 'presents' (Swenson 2011; Handleman 1990). But in the later Middle Bronze Age, the area is transformed into a plaza with open access and no
Figure 12. Monument 1, later phases of use, c. 1800-1600 (schematic plan).
separation of performer and spectator. This development may entail a shift from hierarchical organization to a more communal arrangement (Graves & Van Keuren 2011; Fargher et al. 2011; Joyce 2009).25
Shaft 1: ritual killing of humans and animals In addition to the modifications detailed above, the later Middle Bronze period sees new and extraordinary evidence of ritual activity. The data derive from a series of sacrificial deposits placed in a subterranean circular feature 90-183 cm in diameter designated Shaft l.26 Located in the south-central part of Monument 1, Shaft 1 had a lining of stone boulders extending 3.5 m down to the bottom of the mound deposit, below which it was cut through 2.6 m of bedrock. Inside the shaft were 11 layers of animal and human skeletons that were carefully interred and separated by relatively clean deposits of hard, homogeneous clayey soil, suggestive of ritualized behaviour in its repetitiveness and adherence to rules (Table 1; Fig. 13) (Bell 1997,138; see Schwartz et al. 2012 for details on the upper nine layers).
First to be interred, in level 11 near the bottom of the shaft, were 13 human individuals including men, women and children, without associated grave goods (Fig. 14). The bodies had been placed between two layers of stone cobbles and boulders. Under the lower boulders was a layer of worn sherds and two female figurine fragments. According to preliminary
510
Memory and its Demolition: Ancestors, Animals, and Sacrifice at Umm el-Marra, Syria
Table 1. Contents of the layers of Shaft 1.
© ©
"©'
© ©
©
Layer	Description of contents
1	Adult (aged) female equid; infant sheep/goat; puppy; infant Lepus (hare)
2	Hind end of donkey
3	Adult male equid (horse?); equid fetus
4	Three disarticulated partial sheep, cut and defleshed; equid fetus
5	Adult (aged) female dog
6a	Adult male dog; subadult sheep
6b	Four equid fetuses; sheep/goat fetus/neonate
7	Adult equid (horse?); equid; adult dog
8	Rear half of articulated young equid; complete dog
9	Disarticulated equid, sheep/goat, dog and bird remains
10	Two Gypsfulvus (griffon vultures) missing bones distal to the humerus
11	Thirteen humans; adult dog; small birds in niches
12	Area without ritual depositions below the boulders and sherds of layer 11
f (11)
0     50 cm
Figure 13. Shaft 1 (schematic section).
Figure 14. Shaft 1 level 11, with human skeletal remains. Looking west.
511
Glenn M. Schwartz
Figure 15. Shaft 1 level 10, with two vultures missing their wing bones. Looking west.
analysis of the human skeletal materials by Christopher Brinker (Johns Hopkins University), the skulls, when well-preserved, showed signs of blunt force trauma, probably perimortem in character and indicative of death by a blow to the head.27 In the same level were the bones of an adult dog, and small birds had been placed in small niches in the shaft wall.
Deposited above the level 11 humans were ten layers of animal interments, separated from one another by up to 50 cm of homogeneous clayey soil. Particularly common were equids (horse, donkey, possibly mule) and dogs, including fetal individuals. Level 10, above the humans in level 11, is notable for the presence of two griffon vultures whose wings had been removed (Fig. 15).
A date of later MB II for the shaft's contents can be assigned through consideration of the sherds in layer 11, in addition to a few scattered sherds from the upper layers.28 Despite this general dating, it is unknown whether the interments were deposited in a short time or over a period of months or years. Still, as Jill Weber has observed (pers. comm.), the period of deposition is not likely to have been long, since there is not much evidence of animal burrowing behaviour, which would be the case if a deposit were left exposed for a substantial amount of time, and the individual layers are relatively clean, with very few sherds or indications of materials having fallen in. It is not clear if Shaft 1 was built at the same
time as Monument 1, perhaps as a sacred well and/or conduit to the underworld, or was newly constructed and sunk through Monument 1 in the later Middle Bronze Age.29
Ritual deposits in subterranean shafts are attested elsewhere in Bronze Age Syria, but they do not contain layers of complete animal or human skeletons. The Middle Bronze occupation at Ebla, for example, has revealed several favissae (ritual pits) with specialized contents from sacred contexts, including the vicinity of Monument P3 (Marchetti & Nigro 1997) and above the 'Temple of the Rock' in the southeastern part of the site (Lisella 2010). These shafts had layered fills, indicating separate episodes of ritual deposition, including materials such as clay figurines, ceramic vessels and animal bones. Also comparable is the large subterranean 'abi' or shaft adjacent to the third-millennium bc palace at Tell Mozan (Urkesh) in the upper Khabur plains of northeastern Syria, interpreted as a ritual passage for offerings to underworld deities (Kelly-Buccellati 2002).30 Like the favissae at Ebla, the contents of the Mozan shaft include ceramics, figurines and animal bones but lack complete animal or human skeletons carefully interred in discrete layers.
Burials of humans in second-millennium bc Syria ordinarily are accompanied by ceramic vessels and, not infrequently, personal ornaments, weapons and other objects. Given the lack of grave goods, the uniform mode of killing, the subsequent animal
512
Memory and its Demolition: Ancestors, Animals, and Sacrifice at Umm el-Marra, Syria
depositions and the unusual and central location, it is clear that the humans in Shaft 1 were not interred in a normal fashion and that these data represent an extraordinary event. Likewise, we may reject the possibility that the animals were discarded as trash, since they were carefully deposited, consisted mainly of complete individuals with few signs of animal butchery and included species that would not ordinarily be the contents of midden (e.g. vultures). The repetitive patterns evinced in the mode of death, location and accompanying animal interments indicate that the people, as well as the animals, were the victims of ritual killing. As is well-known, evidence for the ritual killing of humans, or 'human sacrifice,' is exceedingly rare in the Bronze Age Near East (Schwartz 2012a; Porter 2012b). The obvious exception, the Ur Royal Cemetery, contains the bodies of people killed to accompany their social superiors into the afterlife, a scenario that does not apply to Shaft 1.
The event can be understood as a singular occurrence, since no other attestations of it have yet been detected, in contrast to the multiple tombs and animal installations of the Early Bronze Age. Such a procedure could entail a 'high-intensity' ritual (van Baal 1976, 168-78), a unique event taking place in a time of severe stress as an extraordinary gesture to entreat the gods for assistance, as opposed to a low-intensity ritual performed regularly in ordinary circumstances.
We can rule out the possibility that the humans were soldiers killed after their defeat and capture, since women and children are present. Also to be excluded is the possibility that humans were common criminals that had been executed, because the central and special location of their interment and the addition of numerous sacrificed animals argue for the extraordinary character of the event and of the human victims.
At present, I can suggest two hypotheses to explain the contents of the feature. One centres on the practice of sacrificing children in order to gain divine favour in a time of military crisis. Evidence for such behaviour derives from second- and first-millennium bc East Mediterranean sources such as the Incirli stele from southeastern Anatolia (Kaufman 2007), 2 Kings 3:27, and New Kingdom Egyptian representations of besieged Levantines offering children to the gods in order to enlist divine aid (Spalinger 1977/1978; Tatlock 2006; Heagren 2010; but see Burke 2009 for an alternate interpretation). Current evidence indicates that Umm el-Marra suffered a major catastrophe in the late MB II period evinced by the burning of the northwest city gate31 and by the abandonment of diverse structures inside the town with complete ceramic vessels in situ (e.g. Acropolis Centre, Northwest Area A). This event
may be linked to the capture and devastation of the community, perhaps by troops of the Old Kingdom Hittite rulers, as is proposed for the contemporaneous destructions at Alalakh VII and Ebla IIIB. Given this synchronism, one could suggest that the ritual killings of Shaft 1 were perpetrated in order to secure divine assistance in anticipation of attack or during a siege.
If such an explanation is correct, one must explain why the ritual practitioners deposited the bodies of their victims in a subterranean shaft. Perhaps the entreaties were made to underworld deities or to spirits of deceased ancestors below — perhaps even the persons buried in the Early Bronze Age tomb complex penetrated by Shaft 1. Militating against this apotropaic interpretation, however, are discrepancies between the evidence from Shaft 1 and data derived from external sources. Shaft 1 includes adults as well as children and is located far from the fortified edges of the site, the place of sacrifice in the Egyptian and Biblical sources.
An alternative explanation derives from a consideration of the two vultures placed in level 10 of Shaft 1. When vultures appear in ancient Near Eastern art, they are usually depicted feasting on the bodies of enemy dead (Winter 1985; Bahrani 2008, 140-41; Ristvet 2011, fig. 7B; Cooper 2008). The persons interring the individuals in the lower levels of Shaft 1 may therefore have included vultures as agents of dishonour and abuse. We could hypothesize that the deposits in levels 10 and 11 are the result of the killing of vanquished local rulers and their relatives by an enemy such as the Hittites. In this reconstruction, the two vultures deposited above the humans in Shaft 1 were intended to subject the bodies of the dead humans to continual maltreatment in the afterlife, with their departure precluded through the removal of their wings. Their interment, and that of the humans, was followed by the offering of animals, perhaps prized possessions of the executed persons (cf. the 'killing' of a household, as discussed by Pollock (2007)), sacrifices made in gratitude for victory, or pious offerings by surviving inhabitants in honour of the victims. The execution of such people in a central place with a special location would have had a profound effect on the observers and could have been intended as an object lesson for the survivors. Unfortunately, there are no attestations in Hittite or other sources for this treatment of defeated enemies.
Remembrance/Forgetting
As with the Early Bronze Age mortuary complex, Monument 1 and Shaft 1 can be profitably considered with relation to issues of social memory. Monument 1 is built directly above the Early Bronze Age mortuary
513
Glenn M. Schwartz
complex, its extent approximately coinciding with that of the tomb area below. Likewise, the practice of animal and human sacrifice in this special locus, well-documented in Early Bronze times (if we consider the infants in the equid installations to have been sacrificed), is replicated in Shaft 1 of the Middle Bronze occupation. Further, the most common animal occupants in Shaft 1 and in the Early Bronze installations are equid and dog. The construction of Shaft 1 may have been intended to make a connection to the deceased ancestors below, and it may be significant that Shaft 1 does not damage any of the Early Bronze tombs.
An additional 'citation' of the past (Butler 1993; Jones 2007; Van Dyke 2009) is indicated by the inclusion of deposits of thick body sherds and occasional thick rim sherds of Early Bronze IV date within the matrix of Monument 1. Although the sherds may simply have been retrieved from easily accessible tell deposits, the need to include potsherds as construction material is not immediately apparent, and it is not unlikely that the builders were actively creating a tie to the past, appropriating its 'charisma' (Schortman & Urban 2011; Khatchadourian 2007).32 The use and power of material fragments may be relevant (Chapman & Gaydarska 2006).
Such indications of the past could have been employed by the parvenu Middle Bronze Age Amor-ite elite to legitimate their authority, establishing a connection to the Early Bronze Age rulers buried in the site centre. Ristvet (2012a,b), for example, notes how Amorite rulers of Middle Bronze Age upper Mesopotamia referred back to an Early Bronze 'golden age' prior to the turmoil of the later third millennium in order to strengthen their own rule, using third-millennium titulary and venerating third-millennium rulers. Such connections could have been maintained, not only to sanction elite power, but to provide a common past and reference point for the community after a period of instability (Van Dyke 2009).
In contrast to the above reasoning, however, we might consider whether Monument 1 represented an attempt to upstage the past rather than invoke it. Leaders often seek to induce social amnesia in order to expunge memory of earlier regimes and underline the importance of the new order. Such may have been the case with Monument 1, with the Middle Bronze Age authorities sealing off the third-millennium mortuary complex, breaking with the past and reinforcing the existence of a new social reality.33
At present, it is difficult to determine whether Monument 1 represents an attempt at connecting with the past or severing such a connection. If the Early Bronze Age sherd inclusions were intentional,
a deliberate attempt to establish a link with the past would be the preferable interpretation.
Indeed, the proposed uses of memory in the Middle Bronze period are more difficult to substantiate than those of the Early Bronze Age, since the intentionally of references to the past is more ambiguous. One must ask whether the Middle Bronze Age ritual features at Umm el-Marra are indicative of the uses of social memory or of 'disjunctive mnemonia' (Meskell 2003; Van Dyke 2009; Moore 2010). If we are correct that Umm el-Marra was abandoned for as much as 300 years between the Early and Middle Bronze occupations, can memory be invoked to make sense of the similar uses of space in the two periods? It is impossible to know what the Middle Bronze inhabitants understood of their predecessors, but the close concurrence of Monument 1 and the area occupied by the tombs support the proposal that the Middle Bronze occupants were making direct references to the past.
Socio-political ramifications
Evidence has steadily mounted of a sequence of large-scale ritual activities that took place in the centre of the Umm el-Marra acropolis in the Early and Middle Bronze Age. In these data, structures associated with centralized political or economic activities have been conspicuous in their absence: there is nothing resembling a centre of elite political authority, elite residential architecture, nor even a conventional 'temple'. As Figures 3, 6, 8,11 and 12 indicate, small-scale buildings of possible domestic function adjoin the EBA tombs and MB monument. If elite individuals were buried in the Umm el-Marra tombs or presided over Monument 1, their residences were located elsewhere. Despite expectations of Umm el-Marra being a lower order political and economic centre with smaller scale-palaces and temples like Tilmen Hoyiik in the Islahiye plain north of Aleppo (Marchetti 2008), or Tell Beydar in the upper Khabur (Lebeau & Suleiman 2011), it may be apt to consider alternatives.
One might propose that Umm el-Marra was a specialized ritual centre like Binash (NE-nas) and Darib, mentioned in the third-millennium Ebla texts as loci where the Ebla kings were buried (Biga 2007/2008). In such a reconstruction, the monuments at Umm el-Marra would have been installed under the auspices of powers who resided elsewhere, perhaps even an elite of mobile pastoralists who buried their dead at Umm el-Marra. Anne Porter (2012a) has made a similar proposal for Tell Banat on the Euphrates, suggesting that the settlement began life as a mortuary centre and steadily accumulated a
514
Memory and its Demolition: Ancestors, Animals, and Sacrifice at Umm el-Marra, Syria
population to service the participants and pilgrims involved in mortuary activities.
Such a suggestion remains to be carefully evaluated against the extant data, and there are arguments that can be raised against it. For example, the infre-quency of attestations in the Ebla texts to Binash and Darib, apart from reference to the tombs of the Ebla kings, implies that they were small settlements. Umm el-Marra, on the other hand, is much larger than other contemporaneous sites in its vicinity.
Conclusions
Research on the Early Bronze Age mortuary complex and the Middle Bronze Age monumental platform and shaft at Umm el-Marra has provided a substantial body of coherent evidence for the study of elite ideology, mortuary ritual and social memory in early urban Syria and how those variables were materialized and changed through time. In this study, I have aimed to show that these results compel us to acknowledge that people in the past used social memory not only to establish and maintain power but to contest and destroy it. While the Early Bronze elites of Umm el-Marra inscribed their ideology on the landscape by establishing a monumental funerary complex, there was 'pushback' in which hostile forces tried to disrupt or eliminate the social memory thus created. In the Middle Bronze Age, a new monumental construction was built above the Early Bronze monuments, either to establish a link with the past, or to sever it. But in the later part of the Middle Bronze Age, the monument itself was partly hidden and gouged into in another attempt to reposition the past and to hide it. In the Late Bronze Age after a likely period of abandonment comes the final turn of the wheel, when the Middle Bronze monumental area was unceremoniously covered over with houses and, presumably, forgotten.
Despite the evidence for social memory and its material transmission at Umm el-Marra, this study has also revealed ambiguities and uncertainties. For example, it can be challenging to determine whether similarities between behaviours or material culture in two different periods represent a deliberate reference to the past by the later group or are the product of continued tradition and custom (Herzfeld 2004; Van Dyke 2009). Given the evidence from Umm el-Marra, I would propose that intentional reference to the past can be identified when objects and/or bodies of persons from the past were used and interacted with in the community of the present — in either positive or negative ways.
Another point of uncertainty is the social identity of the actors involved in the ritual facilities at Umm
el-Marra. Were they members of local elites, external elites, non-elites, or a combination thereof?34 It would be satisfying if we could recognize a sequence of complex negotiations between elite and non-elite individuals in the developmental history of the Umm el-Marra acropolis centre, as Joyce (2009) has done for Monte Alban. But, at this stage of research, the evidence of social memory and its materialization at Umm el-Marra primarily points to the actions of the powerful. Similarly, data on access to the monumental and ceremonial precincts under discussion signals a desire to restrict entry to a small and select group of people for all but the latest phase of use.
One final contribution of the Umm el-Marra material is that it compels us to recognize a remarkable variability in the ritual and political lives of the communities of Bronze Age Syria. The tombs, animal installations, circular monument and sacrificial shaft at Umm el-Marra bear some similarities to contemporaneous Syro-Mesopotamian ritual facilities, yet each has unique — not to say bizarre — characteristics. While the communities of Syria's 'second urban revolution' had much in common, they could also exhibit and express a pronounced individuality.
Notes
1. On the issue of whether memory can 'exist outside of the minds of individuals', see Moshenska 2010, 35-6. As Halbwachs (1992) emphasized, individual memory tends to be produced in social contexts.
2. The Umm el-Marra project is jointly sponsored by the Johns Hopkins University and the University of Amsterdam; directors are Glenn M. Schwartz (Johns Hopkins) and Hans H. Curvers (Amsterdam).
3. The tombs are labelled Tombs 1 and 3-11 (Schwartz et al. 2003; 2006; 2012; Schwartz 2007a). The designation Tomb 2 is not employed, since it was prematurely used in the field notes to refer to the feature now designated Installation B.
4. Exceptions to this pattern are Tomb 4, with no obvious entryway and Tomb 1, with an eastern entry unimpeded by blocking. It is not clear where (or if) Tombs 10 and 11 had their doorways, and it is not known if the Tomb 6 doorway only represented by its northern doorjamb, had been blocked or not.
5. The two latest tombs, 7 and 11, are clearly subterranean. Arguments can be proposed both for and against the above-ground nature of the earlier tombs (Schwartz et al. 2006, 628, n. 99). Additional factors in favour of the tombs being above ground include the following:
(1) Although most of the tombs and associated structures were built close together and there were few opportunities to observe their relationships in section, the available stratigraphic profiles disclosed no sign of an intrusive character for the tombs.
(2) A wall abutting the extant top of the stone substruc-
515
Glenn M. Schwartz
ture of Tomb 5 to the east was associated with a surface bearing EBAIVB pottery, which indicates that Tomb 5, datable to EBA III, had to be above ground.
(3) Tomb 3 had elaborate stone foundations for its entryway on the east consisting of a stepped threshold of stones. Tomb 1 had similar stone foundations below its entryway, built atop the extant top of the Tomb 6 west wall stone substructure.
(4) There is no evidence of earlier architecture cut by the tombs' construction. Architecture between the two tombs 6 and 8, for instance, accommodated the tomb architecture and was not disturbed by it.
(5) It is not likely that room 2 was subterranean, since it has a stepped entryway leading to Installation A; since it is a 'twin' of Tomb 3, it is likewise improbable that Tomb 3 was subterranean.
Given the above, as well as arguments presented previously the current weight of evidence supports an interpretation of the pre-EB A IVB tombs as completely above ground. While Tomb 4 had no obvious entryway it may have been entered with a ladder.
6. This period, understood to precede the era when corrugated 'Hama' goblets were common, is attested at other west Syrian sites such as Ebla (Mazzoni 1991; Dolce 2008), Tuqan (Peyronel 2011) and Qatna (Morandi-Bonacossi 2008).
7. Built against the blocked doorway of Tomb 8, the two features must have been installed sometime after the initial use and construction of Tomb 8. But the presence of a globular jar with everted neck in Installation E virtually identical to examples from Tombs 6 and 8 support the dating of Installations E and F to this phase. Installation E post-dates the walls enclosing Tomb 6 on its western side.
8. Despite earlier proposals (Schwartz et al. 2006), it is unlikely that the equids in the complex were decapitated (see also Way 2011). Instead, Weber has concluded that the animals were placed in the installations with their heads resting on a ledge, and after skeletonizing the heads became separated from the rest of the body due to gravity and pressure from other contents of the structure. It is clear that Installation E postdates Installation F, since the former structure is partly built atop the latter.
9. I am grateful to Jill Weber for this observation. In this phase, corrugated goblets become common, especially a short and squat type with a flat base.
10. A painted Euphrates Banded Ware jar possibly belonging to Installation C was found in an area disturbed by the construction of Tomb 1. If the vessel derived from Installation C, it would provide additional evidence for the EBA III/IVA date of that feature.
11. Supporting an (early) EBA IVA date for Installation A are a corrugated Hama Gil type goblet (Schwartz et al. 2006, 625) and a rim sherd from a bowl of a type well-attested in Tomb 3 that were found in its contents. A rim sherd from a painted Euphrates Banded Ware jar found in the upper pit of Installation G suggests a similar date.
12. The dating of Installation B to this phase rests on
flimsier evidence than the other installations, primarily the approximate equivalency between the absolute elevation of the top of the feature and the lower floor of Tomb 1. The only ceramic inclusion was a spouted jar (Schwartz 2007a, fig. 3.7:1).
13. If the dating of the equid installations proposed here is correct, it reveals that the Type I installations occurred earlier in the sequence, while Type II installations appeared later. Although it might be expected that a given equid installation was installed to accompany a nearby human tomb, it is not possible to link any installation with a specific tomb, either by virtue of its locational proximity or its contents. Given the distance of the eastern tombs 5 and 9 from the equid installations, one might question whether those tombs were 'serviced' by any of the equid installations, and it is possible that the installations were only associated with the western tombs.
14. This conclusion is based on the absence of any non-infant interment from excavations in Early Bronze Age contexts at the site outside the tomb complex.
15. Although space for an audience was limited, it is nevertheless possible that processions of ritual participants commenced in another part of the site, or even outside the site, and then made their way to the acropolis, affording the larger population an opportunity to witness the earlier stages of the relevant rituals (e.g. Ristvet 2011, 9).
16. In addition to the arguments supplied in the cited publications, we may also note that the disturbances wreaked on Tomb 8 must have occurred prior to the phase 4 (EBA IVB) period. The phase 4 enclosure wall is dug from above, indicating that the tomb had been filled in with soil by that time. There was no evidence of an intrusive robber's pit.
17. See Meyer (1991, 21, 159-60) for similar evidence of deliberate smashing of bones from contemporaneous, apparently elite, tombs at Shamseddin on the Euphrates.
18. Unless, of course, they did not have the time or the resources to completely demolish or cover up the tombs.
19. While not a mound per se, the monument capped the mounded remains of the Early Bronze tomb complex below.
20. Presumably relevant to the function of Monument 1 are six circular pits found outside it to the northeast, arranged in a semicircular pattern imitating the shape of the monument. The pits mainly contained ashy or bricky debris with occasional sherds and diverse animal bones, but one unusual find was a bitumen anthropomorphic figurine with damage to its face, hands and legs. This object may be interpretable as a figure 'injured for apotropaic purposes, representing the evil to be warded off. See Collins 2002 on the practice of digging pits for apotropaic and other purposes to communicate with supernatural entities in the underworld as attested in Hittite sources. Also of possible relevance is an agate bead found near Monument 1 to the north that has the representation of two eyes or celestial symbols (Schwartz et al. 2006, 634, n. 147; Schwartz 2007a, 62,
516
Memory and its Demolition: Ancestors, Animals, and Sacrifice at Umm el-Marra, Syria
fig. 3.13). For an agate object with comparable celestial symbols, see Matoian 2008, pi. 17"6. I am grateful to Sally Dunham for her suggestions on the figurine and the bead.
21. I am very grateful to Alessandra Gilibert for the insights derived from our discussion of Monument 1 and issues of performance.
22. The blank space in the southernmost trench (Fig. 11) is not an open zone but is due to the destruction, effected during a later MBA phase, of the earliest MBA architecture except for a fragment of the acropolis enclosure wall.
23. That is, unless a tall structure originally existed on top of the monumental platform that is no longer extant.
24. The interpretation and dating of these alterations remain open. Although it seems likely that they date to a time when the monument was still in use, it is possible that the accumulation of soil with late Middle Bronze sherds above the stones of the monument reflects 'levelling fill' used to create a flat surface for the Late Bronze architecture above, or that it represents an accumulation of soil that occurred during an abandonment of the site between the late MBA and the resumption of occupation in the Late Bronze Age.
25. When considering the evidence of heterarchical as opposed to hierarchical social structures, or corporate as opposed to network, it is important to bear in mind that these strategies are often not mutually exclusive, and that aspects of both can be in play at the same time. Thus a typological approach seeking to identify either corporate or network strategies should be avoided for a more nuanced perspective seeking to understand the workings of the specific organizations in operation (Yoffee 2005,177; Pool 2007, 30; Porter 2010).
26. I am grateful to Jill Weber, both for excavating Shaft 1, and for providing the data cited here. Given the absolute elevation of the extant top of the shaft, level with the extant top of the stone substructure of Early Bronze Age Tomb 4, upper segments of Shaft 1 must have been removed by the large Late Bronze Age pit in the area. However, the uppermost animal deposits were found 1.34 m below the extant top of the shaft, implying that the excavations had recovered the complete sequence of deposits.
27. When excavated, the human remains were below the water table, requiring water to be pumped out. As a result, the bones were encased in mud, and it was difficult to determine their precise positioning with respect to one other and the matrix they were found in. Nevertheless, Jill Weber's excavation notebook (23 June 2010) reports that:
'there were many parts in articulation, including several "arm" and "leg" joints (tib/fib/fem) (hum/ rad/ulna) in flexed position. Vertebrae were frequently articulated and usually found with ribs. Cervical vertebrae were typically very near to skulls. Mandibles, however, were not frequently in place. The flexed joints were on the large rocks that now lie across the well. ... Probably the bones were put
in complete, but became disarticulated after decomposition when bones slumped together. This is why those directly on the rocks remained in articulation.' Likewise, in his report on the human bones, Christopher Brinker (n.d.) comments that 'the presence of small carpal and tarsal bones suggests that these individuals [in Shaft 1] were probably articulated when deposited and argues against the possibility that this deposition was the result of a secondary interment'.
28. Radiocarbon dates derived from equid bone from levels 3 and 1 yielded a late third-/early second-millennium and mid-second-millennium date, respectively (Schwartz et al. 2012). Given the ceramics in level 11, the level 3 radiocarbon date is not credible.
29. It is more likely that the shaft was constructed at the same time as Monument 1, since it would have been problematic to dig such a feature into a pre-existing monument, with loosened stones from the monument continually falling into the excavated shaft. With respect to Shaft l's original function as a well, the depth of 2.6 m sunk into bedrock seems shallow for the acquisition of subsurface water, but it is unknown if water was available at this elevation in the second millennium bc. Ground water was present at this depth during excavation, requiring the pumping of water from the shaft in order to excavate the human skeletons at the bottom, but the water table of the region has risen dramatically since the large-scale introduction of irrigation water from Lake Assad in the Euphrates region to the east in the later twentieth century ad.
30. On Hittite rituals performed to communicate with chthonic deities through digging of pits, see Collins 2002.
31. The latest phase of the northeast city gate excavated by the Belgian expedition (Tefnin 1983/1984) was also burned. Although the date of this event is not certain, since the associated materials were not published, it is likely that it also dated to late Middle Bronze Age, since the architectural sequence of the northeast gate mirrors that of the northwest gate. If so, then the debacle that befell Umm el-Marra in late MBA is more convincingly to be attributed to a military disaster.
32. At Jerablus Tahtani on the Euphrates northeast of Umm el-Marra, the mid-late third-millennium monumental tomb 302 was covered with a tumulus containing layers of beveled rim bowl sherds (dating to the fourth-early third millennium bc) intentionally inserted (Peltenburg 1999, 431). See also the early second-millennium bc temple at Tell Leilan, which had incised Ninevite 5 sherds on its floors, perhaps intentionally included as references to the third-millennium past (Ristvet 2012b).
33. In a variation on that theme, the monument may have been intended to seal off the Early Bronze Age tombs below as protection against ill-intentioned ghosts of predecessors or spirits resident in the tombs.
34. Ascertaining the gender or age of the relevant individuals is even more problematic. Nor should it be assumed that entire groups always acted in concert (Hodder 2007, 32).
517
Glenn M. Schwartz
Acknowledgements
I am grateful to the Directorate-General of Antiquities and Museums of Syria for its consistent support and encouragement of the Umm el-Marra project, particularly Dr Bassam famous, Director-General, Dr Michel al-Maqdissi, Director of Excavations, Dr Yusuf Kanjo, Director of the National Museum, Aleppo, and the DGAM representatives Muhammad Lahluh, Mahmud Hamud, Nasr Sharaf, Hekmat Awad, Iyyad Ghanem, Shirin Khidr, Rakkan Suleiman, and Salah Shaker. Invaluable to the project is the unstinting assistance of our site guard and house manager Muhammad al-Helu al-Qasem. Funding support for the excavations have derived from the National Science Foundation (BCS-0545610, BCS-0137513, 9818205), National Endowment for the Humanities, National Geographic Society, The Metropolitan Museum of Art, the Arthur and Isadora Dellheim Foundation, the Johns Hopkins University, the University of Amsterdam, and several private contributors.
I am especially grateful to Jill Weber for her work excavating Shaft 1 and the Early Bronze Age equid installations, for providing faunal data, and for insights on the excavated results on the Acropolis Centre. I also wish to thank Sally Dunham for her many important suggestions and ideas. Christopher Brinker generously provided preliminary analysis of the Shaft 1 human remains. Invaluable was the work conducted by the students and specialists who worked at Umm el-Marra, particularly that of Hatem Katkhoda and Kim Benzel, whose participation in the 2006 season I inadvertently failed to acknowledge in previous publication (Schwartz et al. 2012).
I wish to thank Sally Dunham, Anne Porter, Jill Weber and two anonymous reviewers for reading and commenting on preliminary drafts of this article. I am also grateful to Reinhard Bernbeck, Maria Giovanna Biga, Robert Cooke, Gabrielle Castro-Gessner, Marian Feldman, Daniel Fleming, Hartmut Kühne, Nicola Laneri, Paolo Matthiae, Adelheid Otto, Susan Pollock, Anthony Spalinger, James Wright and Norman Yoffee for very helpful suggestions on some of the ideas presented herein. I am particularly grateful to Ales-sandra Gilibert for the insights derived from our discussion of Monument 1 and issues of performance. Christopher Brinker and Harley King assisted with the illustrations. I also wish to acknowledge the ideas and inspiration derived from the students in my graduate seminar on the archaeology of ritual at The Johns Hopkins University conducted in spring 2012: Adam Bean, Joshua Bowen, Maggie Bryson, Catherine Cobb, Ioana Dumitru, Anna Glenn, Caleb Howard, Michael Simone, Jennifer Swerida and John Thames. Finally I would like to acknowledge the assistance of a Topoi Senior Fellowship, Freie Universität Berlin, Research Group B-II-1, in September-November 2011, which allowed me to do the majority of work for this article.
Glenn M. Schwartz Department of Near Eastern Studies The Johns Hopkins University Gilman Hall 113 3400 North Charles Street
Baltimore MD 21218 USA
Email: schwartz@jhu.edu
References
Aimers, J., 2007. What Maya collapse? Terminal Classic variation in the Maya lowlands. Journal of Archaeological Research 15, 329-77.
Akkermans, P. & G. Schwartz, 2003. The Archaeology of Syria: From Complex Hunter-gatherers to Early Urban Societies (ca. 16,000-300 B.C.). Cambridge: Cambridge University Press.
Archi, A. & M.G Biga, 2003. A victory over Mari and the fall of Ebla. Journal of Cuneiform Studies 55,1-42.
Argent, G., 2010. Do clothes make the horse? Relationality roles and statuses in Iron Age Inner Asia. World Archaeology 42(2), 175-87.
Bahrani, Z., 2008. Rituals of War: the Body and Violence in Mesopotamia. New York (NY): Zone Books.
Batey, E., 2011. Tell Umm el-Marra (Syria), seasons 2000-2006. Bioarchaeology of the Near East 5,1-10.
Bell, C, 1997. Ritual: Perspectives and Dimensions. Oxford: Oxford University Press.
Berliner, D., 2005. The abuses of memory. Anthropological Quarterly 78(1), 197-211.
Biga, M.G., 2007/2008. Buried among the living at Ebla? Funerary practices and rites in a XXIV century B.C. Syrian kingdom, in Sepolti tra i vivi: Buried Among the Living, eds. G. Bartoloni & M.G. Benedettini. (Sáenze delľantichitä storia archeológia antropológia 14(1).) Rome: Universitä degli studi di Roma 'La Sapienza', 249-75.
Blanton, R., G. Feinman, S. Kowalewski & P. Peregrine, 1996. A dual-processual theory for the evolution of Meso-american civilization. Current Anthropology 37(1), 1-14.
Borič, D., 2010. Archaeology and Memory. Oxford: Oxbow Books.
Bourdieu, P., 1977. Outline of a Theory of Practice. Cambridge: Cambridge University Press.
Brinker, C, n.d. Preliminary Study of the Human Remains from the 2010 Season of Excavations at Umm el-Marra. Unpublished report on file with the Umm el-Marra Archaeological Project, the Johns Hopkins University.
Burke, A., 2009. More light on old reliefs: New Kingdom Egyptian siege tactics and Asiatic resistance, in Exploring the Longue Durée: Essays in Honor of Lawrence E. Stager, ed. J.D. Schloen. Winona Lake (IN): Eisen-brauns, 57-68.
Butler, J., 1993. Bodies That Matter: On the Discursive Limits of 'Sex.' New York (NY): Routledge.
Chapman, R. & B. Gaydarska, 2006. Parts and Wholes: Fragmentation in Prehistoric Context. Oxford: Oxbow Books.
Charpin, D., 2004. Histoire politique du proche-Orient Amorrite (2002-1595), in Mesopotamien: die altbaby-lonische Zeit, eds. P. Attinger, W. Sallaberger & M. Wäfler. (Orbis Biblicus et Orientalis 160/4.) Fribourg: Academic Press, 25-480.
Chase, D. & A. Chase, 2011. Ghosts amid the ruins: analyz-
518
Memory and its Demolition: Ancestors, Animals, and Sacrifice at Umm el-Marra, Syria
ing relationships between the living and the dead among the ancient Maya at Caracol, Belize, in Living with the Dead: Mortuary Ritual in Mesoamerica, eds. J. Fitzsimmons & I. Shimada. Tucson (AZ): University of Arizona Press, 78-101. Chesson, M., 1999. Libraries of the dead: Early Bronze Age charnel houses and social identity at urban Bab edh-Dhra', Jordan. Journal of Anthropological Archaeology 18,137-64.
Chesson, M. (ed.), 2001. Social Memory, Identity and Death: Ethnographic and Archaeological Perspectives on Mortuary Rituals. (Archaeological Publications of the American Anthropological Association 10.) Arlington (VA): American Anthropological Association.
Collins, B., 2002. Necromancy fertility and the dark earth: the use of ritual pits in Hittite cult, in Magic and Ritual in the Ancient World, eds. P. Mirecki & M. Meyer. Leiden: Brill, 224-42.
Connerton, P., 1989. How Societies Remember. Cambridge: Cambridge University Press.
Cooper, J., 2008. Incongruent corpora: writing and art in ancient Iraq, in Iconography without Texts, ed. P. Taylor. (Warburg Institute Colloquia 13.) London: The Warburg Institute, 69-94.
Crawford, C, 2007. Collecting, defacing, reinscribing (and otherwise performing) memory in the ancient world, in Negotiating the Past in the Past, ed. N Yoffee. Tucson (AZ): University of Arizona Press, 10-42.
Curvers, H. & G. Schwartz, 1997. Umm el-Marra, a bronze age urban center in the Jabbul plain, western Syria. American Journal of Archaeology 101, 201-27.
Demarest, A., 2004. Ancient Maya: the Rise and Fall of a Rainforest Civilization. Cambridge: Cambridge University Press.
DeMarrais, E., L. Castillo & T. Earle, 1996. Ideology, materialization and power strategies. Current Anthropology 37(1), 15-31.
Dolce, R., 2008. Ebla before the achievement of palace G culture: an evaluation of the Archaic Early Syrian period, in Proceedings of the 4th International Congress of the Archaeology of the Ancient Near East (29 March-April 3 2004, Freie Universität Berlin), vol. 2: Social and Cultural Transformation — The Archaeology of Transitional Periods and Dark Ages, Excavation Reports, eds. H. Kühne, R. Czichon & F. Kreppner. Wiesbaden: Harrassowitz, 65-80.
Fargher, L., R. Blanton, V. Espinoza, J. Millhauser, N. Xiuhtecutli & L. Overholtzer, 2011. Tlaxcallan: the archaeology of an ancient republic in the New World. Antiquity 85,172-6.
Fleisher, J. & S. Wynne-Jones, 2010. Authorisation and the process of power: the view from African archaeology. Journal of World Prehistory 23(4), 177-93.
Fleming, D., 2004. Democracy's Ancient Ancestors: Mari and Early Collective Governance. Cambridge: Cambridge University Press.
Geertz, C, 1980. Negara: the Theatre State in Nineteenth Century Bali. Princeton (NJ): Princeton University Press.
Gilibert, A., 2011. Syro-Hittite Monumental Art and the Archae-
ology of Performance. Berlin: Walter de Gruyter.
Graves, W. & S. Van Keuren, 2011. Ancestral Pueblo villages and the panoptic gaze of the commune. Cambridge Archaeological Journal 21(2), 263-82.
Halbwachs, M., 1992. On Collective Memory, trans. L.A. Coser. Chicago (IL): University of Chicago Press.
Hamilakis, Y., 2010. Re-collecting the fragments: archaeology as mnemonic practice, in Material Mnemonics: Everyday Memory in Prehistoric Europe, eds. K. Lillios & V. Tsamis. Oxford: Oxbow Books, 188-99.
Handleman, D., 1990. Models and Mirrors: Towards an Anthropology of Public Events. Cambridge: Cambridge University Press.
Hastorf, C, 2007. Archaeological Andean rituals: performance, liturgy and meaning, in The Archaeology of Ritual, ed. E. Kyriakidis. Los Angeles (CA): UCLA Cotsen Institute of Archaeology, 77-107.
Heagren, B., 2010. The Art of War in Pharaonic Egypt: an Analysis of the Tactical, Logistic, and Operational Capabilities of the Egyptian Army (Dynasties XVII-XX). Unpublished PhD dissertation, University of Auckland.
Helwing, B., 2012. An age of heroes? Some thoughts on Early Bronze Age funerary customs, in (Re-)Con-structing Funerary Rituals in the Ancient Near East, eds. P. Pfalzner, H. Niehr, E. Pernicka & A. Wissing. Wiesbaden: Harrassowitz, 47-58.
Hendel, R., 2010. Cultural memory, in Reading Genesis: Ten Methods, ed. R. Hendel. Cambridge: Cambridge University Press, 28-46.
Herzfeld, M., 2004. Whatever happened to 'influence'? The anxieties of memory. Archaeological Dialogues 10(2), 191-203.
Hill, E., 2011. Animals as agents: hunting ritual and relational ontologies in prehistoric Alaska and Chukotka. Cambridge Archaeological Journal 21(3), 407-26.
Hobsbawm, E., 1983. Introduction: inventing traditions, in The Invention of Tradition, eds. E. Hobsbawm & T. Ranger. Cambridge: Cambridge University Press, 1-14.
Hodder, I., 2007. The 'social' in archaeological theory, in^4 Companion to Social Archaeology, eds. L. Meskell & R. Preucel. Oxford & Maiden (MA): Blackwell Publishers, 23-42.
Hodder, I. & C. Cessford, 2004. Daily practice and social memory at Catalhoyiik. American Antiquity 69,17-40.
Inomata, T. & L. Coben (eds.), 2006. The Archaeology of Performance. Lanham (MD): AltaMira.
Jahn, B., 2009. The migration and sedentarization of the Amorites from the point of view of the settled Babylonian population, in Representations of Political Power: Case Histories from Times of Change and Dissolving Order in the Ancient Near East, eds. M. Feldman & M. Heinz. Winona Lake (IN): Eisenbrauns, 193-209.
Jones, A., 2007. Memory and Material Culture. Cambridge: Cambridge University Press.
Jonker, G., 1995. The Topography of Remembrance: the Dead, Tradition, and Collective Memory in Mesopotamia. Leiden: Brill.
519
Glenn M. Schwartz
Joyce, A., 2009. The main plaza of Monte Alban: a life history of place, in The Archaeology of Meaningful Places, eds. B. Bowser & M. Nieves Zedeno. Salt Lake City (UT): University of Utah Press, 32-52.
Kaufman, S.A., 2007. The Phoenician inscription of the Incirli trilingual: a tentative reconstruction and translation. Maarav 14(2), 7-26.
Kelly-Buccellati, M., 2002. Ein hurritischer Gang in die Unterwelt. Mitteilungen der deutschen Orient-Gesellschaft 134,131-48.
Khatchadourian, L., 2007. Unforgettable landscapes: attachments to the past in Hellenistic Armenia, in Negotiating the Past in the Past, ed. N. Yoffee. Tucson (AZ): University of Arizona Press, 43-75.
Kuzucuoglu, C. & C. Marro (eds.), 2007. Societes humaines et changement climatique ä la fin du troisieme millenaire: une crise a-t-elle eu lieu en haute Mesopotamia Istanbul: Institut francais d'etudes anatoliennes Georges-Dumezil.
Laneri, N., P. Pfälzner & S. Valentini (eds.), 2012. Looking North: the Socioeconomic Dynamics of Northern Mesopo-tamian and Anatolian Regions during the Late Third and Early Second Millennium sc. Wiesbaden: Harrassowitz.
Latour, B., 1993. We Have Never Been Modern. New York (NY): Harvester Wheatsheaf.
Lebeau, M. & A. Suleiman (eds.), 2011. Tell Beydar: the 2004/2-2009 Seasons of Excavations, the 2004/2-2009 Seasons of Architectural Restoration: a Preliminary Report. (Subartu 29.) Turnhout: Brepols.
Lillios, K. & V. Tsamis (eds.), 2010. Material Mnemonics: Everyday Memory in Prehistoric Europe. Oxford: Oxbow Books.
Lisella, A., 2010. The MB II unbaked clay figurines from the favissa P.9301+P.9308 in the sacred area HH at Ebla, in Proceedings of the 6th International Congress of the Archaeology of the Ancient Near East, eds. P. Matthiae, F. Pinnock, L. Nigro & N. Marchetti. Wiesbaden: Harrassowitz, 821-33.
Losey R., V. Bazaliiskii, S. Garvie-Lok et ah, 2011. Canids as persons: early Neolithic dog and wolf burials, Cis-Baikal, Siberia. Journal of Anthropological Archaeology 30,174-89.
Lundström, S., 2009. Die Königsgrüfte im Alten Palast inAssur (WVDOG 123.) Wiesbaden: Harrassowitz.
Mack, A., 2004. One landscape, many experiences: differing perspectives of the temple districts of Vijayanagara. Journal of Archaeological Method and Theory 11(1), 59-81.
Marchetti, N, 2008. A preliminary report on the 2003 and 2004 excavations at Tilmen Höyük, in Proceedings of the 4th International Congress of the Archaeology of the Ancient Near East (29 March-April 3 2004, Freie Universität Berlin), vol. 2: Social and Cultural Transformation — The Archaeology of Transitional Periods and Dark Ages, Excavation Reports, eds. H. Kühne, R. Czichon & F. Kreppner. Wiesbaden: Harrassowitz, 353-60.
Marchetti, N. & L. Nigro, 1997. Cultic activities in the sacred area of Ishtar at Ebla during the Old Syrian Period: the favissae F.5237 and F.5238. Journal of Cuneiform Studies 49,1-44.
Mato'ian, V. (ed.), 2008. Le mobilier du palais royal d'Ougarit.
(Ras Shamra-Ougarit XVII.) Lyon: Maison de 1'Orient et de la Mediterranee-Jean Pouilloux.
Matthiae, P., 1979. Du-ub di Mardikh IIB1 = Tu-ba di Alalakh VII. Studi Eblaiti 1,115-18.
Matthiae, P., 1981. Ebla: an Empire Rediscovered. Garden City (NY): Doubleday.
Matthiae, P., 1997. Ebla and Syria in the Middle Bronze Age, in The Hyksos: New Historical and Archaeological Perspectives, ed. E. Oren. Philadelphia (PA): The University Museum, 379-414.
Matthiae, P., 2008. Gli archivi reali di Ebla: la scoperta, i testi, il significato. Milan: Mondadori universitä.
Mazzoni, S., 1991. Ebla e la formazione della cultura urbana in Siria. La Parola del Passato 46,163-94.
McAnany P., 1995. Living with the Ancestors: Kinship and Kingship in Ancient Maya Society. Austin (TX): University of Texas Press.
McAnany, P. & T. Negrön, 2010. Bellicose rulers and cli-matological peril? Retrofitting twenty-first century woes on eighth-century Maya society, in Questioning Collapse: Human Resilience, Ecological Vulnerability, and the Aftermath of Empire, eds. P. McAnany & N. Yoffee. Cambridge: Cambridge University Press, 142-75.
McAnany, P. & N. Yoffee (eds.), 2010. Questioning Collapse: Human Resilience, Ecological Vulnerability, and the Aftermath of Empire. Cambridge: Cambridge University Press.
Meskell, L., 2003. Memory's materiality: ancestral presence, commemorative practice, and disjunctive locales, in Archaeologies of Memory, eds. R. Van Dyke & S. Alcock. Oxford & Maiden (MA): Blackwell Publishers, 34-55.
Meyer, J.-W., 1991. Gräber des 3. Jahrtausends v.Chr. im syrischen Euphrattal: Ausgrabungen in Samseddin und Djer-niye. Saarbrücken: Saarbrücker Druckerei und Verlag.
Middleton, G., 2012. Nothing lasts forever: environmental discourses on the collapse of past societies. Journal of Archaeological Research 20(3), 257-307.
Mills, B., 2008. Remembering while forgetting: depositional practices and social memory at Chaco, in Memory Work: Archaeologies of Material Practices, eds. B. Mills & W. Walker. Santa Fe (NM): SAR Advanced Seminar Series, 81-108.
Mills, B. & W. Walker (eds.), 2008. Memory Work: Archaeologies of Material Practices. (Advanced Seminar Series.) Santa Fe (NM): SAR Press.
Moore, J., 2010. Making a huaca: memory and praxis in pre-hispanic far northern Peru. Journal of Social Archaeology 10, 398-422.
Moshenska, G., 2010. Working with memory in the archaeology of modern conflict. Cambridge Archaeological Journal 20(1), 33-48.
Morandi-Bonacossi, D. (ed.), 2008. Urban and Natural Landscapes of an Ancient Syrian Capital: Settlement and Environment at Tell Mishrifeh/Qatna and in Central-Western Syria. Udine: Forum.
Nielsen, A., 2008. The materiality of ancestors: Chullpas and social memory in the late prehispanic history of the south Andes, in Memory Work: Archaeologies of Material Practices, eds. B. Mills & W. Walker. (Advanced
520
Memory and its Demolition: Ancestors, Animals, and Sacrifice at Umm el-Marra, Syria
Seminar Series.) Santa Fe (NM): SAR Press, 207-32.
Oates, D., J. Oates & H. McDonald, 2001. Excavations at Tell Brak, vol. 2: Nagar in the Third Millennium bc. (McDonald Institute Monographs.) Cambridge/London: McDonald Institute for Archaeological Research/ British School of Archaeology in Iraq.
Olick, J. & J. Robbins, 1998. Social memory studies: from 'collective memory' to the historical sociology of mnemonic practices. Annual Review of Sociology 24,105-40.
Peltenburg, E., 1999. The living and the ancestors, in Archaeology of the Upper Syrian Euphrates: the Tishrin Dam Area. Proceedings of the International Symposium Held at Barcelona, lanuary 28th-30th 1998, eds. G. Del Olmo Lete & J.-L. Montero Fenollós. Barcelona: Editorial AUSA, 427-42.
Peltenburg, E., 2007/2008. Enclosing the ancestors and the growth of socio-political complexity in Early Bronze Age Syria, in Sepolti tra i vivi: Buried Among the Living, eds. G. Bartoloni & MG. Benedettini. (Scienze dell'an-tichitä storia archeológia antropológia 14/1.) Rome: Universitä degli studi di Roma 'La Sapienza', 215^47.
Peyronel, L., 2011. Area P, in Tell Tuqan: Excavations 2008-2010, ed. F. Baffi. Salento: Congedo Editore, 61-139.
Pollock, S., 2007. Death of a household, in Performing Death: Social Analyses of Funerary Traditions in the Ancient N ear East and Mediterranean, ed. N. Laneri. Chicago (IL): Oriental Institute of the University of Chicago, 209-22.
Pool, C, 2007. Olmec Archaeology and Early Mesoamerica. Cambridge: Cambridge University Press.
Porter, A., 2007/2008. Evocative topography: experience, time and politics in a landscape of death, in Sepolti tra i vivi: Buried Among the Living, eds. G. Bartoloni & M.-G. Benedettini. (Scienze dell'antichita storia archeológia antropológia 14/1.) Rome: Universitä degli studi di Roma 'La Sapienza', 195-214.
Porter, A., 2010. From kin to class — and back again! Changing paradigms of the early polity, in The Development ofPre-State Communities in the Ancient Near East: Studies in Honour of Edgar Peltenburg, eds. L. MacGuire & D. Bolger. Oxford: Oxbow Books, 72-8.
Porter, A., 2012a. Mobile Pastoralism and the Formation of Near Eastern Civilizations: Weaving Together Society. Cambridge: Cambridge University Press.
Porter, A., 2012b. Mortal mirrors: creating kin through human sacrifice in third-millennium Syro-Mesopo-tamia, in Sacred Killing: the Archaeology of Sacrifice in the Ancient Near East, eds. A. Porter & G. Schwartz. Winona Lake (IN): Eisenbrauns, 191-216.
Ristvet, L., 2011. Travel and the making of north Mesopota-mian polities. Bulletin of the American Schools of Oriental Research 361,1-31.
Ristvet, L., 2012a. Resettling Apum: tribalism and tribal states in the Tell Leilan region, Syria, in Looking North: the Socioeconomic Dynamics of Northern Mesopotamian and Anatolian Regions during the Late Third and Early Second Millennium bc, eds. N. Laneri, P. Pfälzner & S. Valentini. Wiesbaden: Harrassowitz, 37-50.
Ristvet, L., 2012b. Rituals of Diplomacy: Pilgrimage, Sacrifice, and Statehood in the Early Second Millennium bc.
Unpublished paper delivered at the Archaeological Institute of America annual meeting, lanuary 2012.
Roddick, A. & C. Hastorf, 2010. Tradition brought to the surface. Cambridge Archaeological Journal 20(2), 157-78.
Salomon, F., 1995. 'The beautiful grandparents': Andean ancestor shrines and mortuary ritual as seen through colonial records, in Tombs for the Living: Andean Mortuary Practices, ed. T. Dillehay. Washington (DC): Dumbarton Oaks, 315-54.
Schortman, E. & P. Urban, 2011. Power, memory and prehistory. American Anthropologist 113(1), 5-21.
Schwartz, G, 2007a. Status, ideology and memory in third millennium Syria: 'royal' tombs at Umm el-Marra, in Performing Death: Social Analyses of Funerary Traditions in the Ancient Near East and Mediterranean, ed. N. Laneri. Chicago (IL): Oriental Institute of the University of Chicago, 39-68.
Schwartz, G, 2007b. Taking the long view on collapse: a Syrian perspective, in Societes humaines et changement climatique a la fin du troisieme millenaire: une crise a-t-elle eu lieu en Haute Mesopotamie?, eds. C. Kuzucuoglu & C. Marro. (Varia Anatolica XIX.) Paris: De Boccard/ Institut francais d'etudes anatoliennes Georges-Dume-zil - Istanbul, 45-68.
Schwartz, G, 2010. Early non-cuneiform writing?: third-millennium bc clay cylinders from Umm el-Marra, in Opening the Tablet Box: Near Eastern Studies in Honor of Benjamin R. Foster, eds. S. Melville & A. Slotsky. Leiden: Brill, 375-95.
Schwartz, G., 2012a. Archaeology and sacrifice, in Ritual Killing: the Archaeology of Sacrifice in the Ancient Near East, eds. A. Porter & G. Schwartz. Winona Lake (IN): Eisenbrauns, 1-32.
Schwartz, G., 2012b. Era of the living dead: funerary praxis and symbol in third millennium bc Syria, in (Re-) Constructing Funerary Rituals in the Ancient Near East, eds. P. Pfalzner, H. Niehr, E. Pernicka & A. Wissing. Wiesbaden: Harrassowitz, 59-78.
Schwartz, G. & J. Nichols (eds.), 2006. After Collapse: the Regeneration of Complex Societies. Tucson (AZ): University of Arizona Press.
Schwartz, G., H. Curvers, F. Gerritsen, J. MacCormack, N. Miller & J. Weber, 2000. Excavation and survey in the fabbul plain: the Umm el-Marra project 1996-1997. American Journal of Archaeology 104, 419-62.
Schwartz, G, H. Curvers, S. Dunham & B. Stuart, 2003. A third-millennium bc elite tomb and other new evidence from Tell Umm el-Marra, Syria. American Journal of Archaeology 107, 325-61.
Schwartz, G., H. Curvers, S. Dunham, B. Stuart & J. Weber, 2006. A third-millennium bc elite mortuary complex at Tell Umm el-Marra, Syria: 2002 and 2004 excavations. American Journal of Archaeology 110, 603-41.
Schwartz, G., H. Curvers, S. Dunham & J. Weber, 2012. From urban origins to imperial integration in western Syria: Umm el-Marra 2006, 2008. American Journal of Archaeology 116,157-93.
Scott, J., 1990. Domination and the Arts of Resistance. New Haven (CT): Yale University Press.
521
Glenn M. Schwartz
Siegel, P., 2010. Continuity and change in the evolution of religion and political organization on pre-Columbian Puerto Rico. Journal of Anthropological Archaeology 29, 302-25.
Sinopoli, C, 2003. Echoes of empire: Vijayanagara and historical memory, Vijayanagara as historical memory in Archaeologies of Memory, eds. R. Van Dyke & S. Alcock. Oxford & Maiden (MA): Blackwell Publishers, 17-33.
Spalinger, A., 1977/1978. A Canaanite ritual found in Egyptian reliefs. Journal of the Society for the Study of Egyptian Antiquities 8, 47-60.
Steadman, S., 2005. Reliquaries on the landscape: mounds as matrices of human cognition, in Archaeologies of the Middle East: Critical Perspectives, eds. S. Pollock & R. Bernbeck. Oxford & Maiden (MA): Blackwell Publishers, 286-307.
Swenson, E., 2011. Stagecraft and the archaeology of spectacle in ancient Peru. Cambridge Archaeological Journal 21(2), 282-313.
Tatlock, J., 2006. How in Ancient Times They Sacrificed People: Human Immolation in the Eastern Mediterranean Basin with Special Emphasis on Israel and the Ancient Near East. Unpublished PhD dissertation, University of Michigan
Tefnin, R., 1983/1984. Apercu sur neuf campagnes de fouilles Beiges aux Tells Abou Danne et Oumm el-Marra (1975-1983). Annates archeologiques arabes syriennes 33/2,141-52.
Torres-Rouff, C, W. Pestle &B. Daverman, 2012. Commemorating bodies and lives at Kish's A cemetery': (re) presenting social memory. Journal of Social Archaeology 12(2), 193-219.
Triadan, D., 2006. Dancing gods: ritual, performance and political organization in the prehistoric Southwest, in Archaeology of Performance, eds. T. Inomata & L. Coben. Lanham (MD): AltaMira, 159-86.
Ur, J., 2010. Cycles of civilization in northern Mesopotamia. Journal of Archaeological Research 18, 387-431.
van Baal, J., 1976. Offering, sacrifice, gift. Numen 23,161-78.
Van Dyke, R., 2008. The Chaco Experience: Landscape and Ideology at the Center Place. Sante Fe (NM): School for Advanced Research Press.
Van Dyke, R., 2009. Chaco reloaded: discursive social memory on the post-Chacoan landscape. Journal of Social Archaeology 9, 220-48.
Van Dyke, R. & S. Alcock (eds.), 2003. Archaeologies of memory: an introduction, in Archaeologies of Memory, eds. R. Van Dyke & S. Alcock. Oxford & Maiden (MA): Blackwell Publishers, 1-13.
Walker, W., V. LaMotta & E.C. Adams, 2000. Katsinas and kiva abandonments at Homol'ovi: a deposit-oriented perspective on religion in Southwest prehistory, in The Archaeology of Regional Interaction, ed. M. Hegmon. Boulder (CO): University Press of Colorado, 341-60.
Way, K., 2011. Donkeys in the Biblical World: Ceremony and Symbol. Winona Lake (IN): Eisenbrauns.
Weber, J., 2008. Elite equids: redefining equid burials of the mid- to late 3rd millennium bc from Umm el-Marra, Syria. Archaeozoology of the Near East 8(2), 499-516.
Weber, J., 2012. Restoring order: death, display and authority, in Sacred Killing: the Archaeology of Sacrifice in the Ancient Near East, eds. A. Porter & G. Schwartz. Winona Lake (IN): Eisenbrauns, 159-90.
Winter, I., 1985. After the battle is over: the stele of the vultures and the beginning of historical narrative in the art of the ancient Near East, in Pictorial Narrative in Antiquity and the Middle Ages, eds. H. Kessler & M. Simpson. Washington (DC): National Gallery of Art, 11-32.
Yoffee, N, 2005. Myths of the Archaic State. Cambridge:
Cambridge University Press. Yoffee, N. (ed.), 2007. Negotiating the Past in the Past. Tucson
(AZ): University of Arizona Press.
Author biography
Glenn Schwartz is Whiting Professor of Archaeology at the Johns Hopkins University. He has directed excavations in Syria since 1987. Research interests include early complex societies in Syria/Mesopotamia, societal collapse and regeneration, and archaeology of rural communities. Books include Archaeological Views from the Countryside (1994), The Archaeology of Syria (2003), After Collapse (2006), and Sacred Rilling (2012).
522