Sliced meat: animal processing and butchery technologies at Early Dynastic Abu Salabikh, Iraq

Introduction

Human adaptation in the ancient Near East was expressed not only through large-scale transformations such as urbanization or state formation, but also through everyday technological practices that sustained social life. The butchering of animals - an act both utilitarian and symbolic - embodies these adaptive processes. The zooarchaeological study of butchered remains offers a direct record of decision-making in food production, reflecting choices about efficiency, material availability, and cultural identity. As recent theoretical work emphasizes, human-animal interactions and the technologies mediating them form part of complex adaptive systems in which ecological, social, and cognitive factors co-evolve (Crabtree, 2018; Gaastra et al., 2020a, 2020b). Within this framework, the analysis of butchering mark morphology becomes not merely a technological signature, but evidence of adaptive behaviour – i.e., how communities negotiated innovation, risk, and continuity in changing material worlds.

Across the third millennium BCE, communities of the ancient Near East faced accelerating changes in raw-material lifestyles. In particular, the introduction of bronze metallurgy introduced new possibilities for productive efficiency and status display, yet its adoption in daily subsistence tasks was uneven and delayed (Greenfield, 2021; Manclossi et al., 2019; Yener, 2000). The persistence of chipped-stone butchering tools indicates that technological substitution followed a complex trajectory shaped by accessibility, performance, and embedded cultural practice - as documented from Egypt (De Sousa and Greenfield, 2023; Graves-Brown, 2015; Lund, 2015), the Levant (Greenfield, 2013 Greenfield and Brown, 2016), Anatolia (Greenfield et al., 2022b; Greenfield and Marciniak, 2019), and neighbouring regions such as SE Europe (Eriksen, 2010; Greenfield, 1999, 2017). Rather than a linear replacement of stone by metal, the zooarchaeological record reveals an adaptive coexistence of materials where functional reliability and tradition often outweigh novelty.

Butchering is an everyday (quotidian) activity that affects all levels of society. The study of butchered animal remains, therefore, moves discussions beyond simple economic reconstruction to become a key proxy measure for new forms of human adaptation that appear in complex societies. Such analyses expose how households and workshops adjusted to technological change while maintaining cultural continuity. The rate and context of adoption for metal blades - whether driven by elite demand, craft specialization, or diffusion through trade – can inform our understanding of innovation of groups who are now embedded within hierarchical social and economic networks. Thus, the micro-scale traces of butchering allow for a direct connection to macro-scale patterns of interregional interaction linking the various regions of the Near East and to the lands beyond (De Sousa and Greenfield, 2023; Greenfield, 2017; Greenfield et al., 2022b).

One of the significant questions in Near Eastern archaeology and the evolution of complex societies is the effect on social differentiation when metals begin to appear and widely used (Moorey, 1982; Yener, 2000). To answer how metals have aided the transformation of society over time, one must understand when metals became important. Most studies on early metallurgy have focused on well-preserved remains from tombs and hoards (often associated with higher-status elements in society), whereas relatively few come from domestic contexts within settlement sites (Blackwell, 2011; Nigro, 2015; Philip, 1988; Zimmermann and Yıldırım, 2007). Yet, such studies do little to inform about the role of metal in daily life, which is best studied through its use in food processing and butchering for the majority of people.

The analysis of faunal remains from archaeological contexts permits inquiry into a range of behaviours associated with the interaction between humans and animals. Such inquiry has provided information on aspects of carcass processing, status, diet, ethnic identity, and technology of past communities (e.g., Binford, 1981; Greenfield, 1986, 2002; Lyman, 1987; Potts and Shipman, 1981). In particular, information on the butchering process can inform not only on the type of technology used (e.g., Greenfield, 1999, 2006, 2013; Guilday et al., 1962), but also on the nature of provisioning of urban settlements – i.e., the production and distribution of food (Beller et al., 2022). Such analyses can also be directed toward the study of the nature and adoption rate of new technologies by cultures. Butchering is an everyday activity that affects all levels of society.

Previous research on this subject in other parts (e.g., S. Levant) of the Near East has demonstrated that the Early Bronze Age (EBA) (3rd millennium BCE) was still a time when chipped stone tool technologies were used to butcher animal carcasses for consumption by humans. It was only toward the end of the EBA that metal knives first appeared (Moorey 1994; Philip 1988). This coincides with the appearance of the earliest metal knife cut (slice) marks on bones (Greenfield 2013). Yet, the quantities remain very low during this period. Given that such research was previously carried out largely in the southern Levant, it is unclear whether the same situation applies to the heartland of civilization, southern Mesopotamia (also known as Sumer), during the 3rd millennium BCE, Early Dynastic Period (ED, 2900-2350 BCE) (see Table 1 for a chronological timeline).

Table 1. Chronological timeline of the Near East for the third and second millennium in different geographical regions and their associated cultures.

Names for the countries are based on modern geographic designations.

Southern Mesopotamia occupies a pivotal position in discussions of technological adaptation during the third millennium BCE, as it was both the heartland of early urbanism and a region almost entirely dependent on imported mineral resources (Moorey, 1982, 1988). Despite its reputation as a centre of innovation, the archaeological record demonstrates that the adoption of metal tools for quotidian activities such as butchering was slow, uneven, and task-specific across technological domains.

The Early Dynastic faunal assemblage from Abu Salabikh, located in southern Iraq, provides an exceptional case study for exploring these dynamics. Situated within the Sumerian urban heartland, the site's domestic contexts permit investigation of everyday adaptation at the household level. Through the high-resolution microscopic study of butchering marks, we examine how butchering practices reflect both technological conservatism and incremental innovation. In particular, we investigate the microscopic butchering marks on the animal bones to determine whether the animals were butchered with chipped stone or metal tools.

In this context, the microscopic analysis of butchering (i.e., slicing) marks can provide a unique window into the nature and tempo of technological adoption. High-powered microscopy enables the identification of tool raw materials and production techniques (Aramendi et al., 2023; Courtenay et al., 2018; Egeland et al., 2024; Greenfield, 2006; Maté-González et al., 2023), allowing for the assessment not only of the technology used, but also considerations of why (Courtenay et al., 2019; Maté-González et al., 2019; Yravedra et al., 2017). Slice-mark morphology reflects the interplay of human choice and environmental constraint. These are cultural choices conditioned by cost, maintenance, and availability of metal vs. stone tools.

Lithic blades for animal butchery

Evidence of chipped stone tool use has long been identified by the presence of flakes and blades in the archaeological record (Andrefsky, 2005; Rosen, 1997) and/or butchery marks on bone (Braun et al., 2016; Walker and Long, 1977). Chipped stone tools are generally very sharp implements used for processing animal carcasses (e.g., blades and knives) (Crabtree, 1972). Chipped stone tools remained in use for animal butchery well into the second millennium BCE in Egypt, Levant, Anatolia and elsewhere across the Near East, even after the introduction of metal blades for utilitarian activities (De Sousa and Greenfield, 2023; Graves-Brown, 2015, 2010; Luff, 1994; Manclossi et al., 2019; Manclossi et al., 2018; Rosen, 1997). During the third millennium BCE, despite technological adaptations that occurred later in the region, chipped stone tools remained the primary tool for butchery and other quotidian activities. They are only progressively replaced over time in the Middle and Late Bronze Ages by metal knives (Greenfield, 2004, 2013; Greenfield, 2026; Greenfield and Brown, 2016; Greenfield and Chaput, 2021; Greenfield and Greenfield, 2018; Greenfield and Marciniak, 2019; Greenfield et al., 2018; Saidel et al., 2006). Although the theoretical reasons for the rate of adoption of metal blades in animal butchery are beyond this paper, the accessibility, availability, and feasibility of stone over metal allowed chipped stone blades to be the preferred tool for animal carcass processing at most sites throughout the third and second millennia BCE.

It is well recognised that the decline of chipped stone tools is influenced by the adoption of comparable metal tools (Manclossi et al., 2019; Manclossi et al., 2018; Rosen, 1997). This is especially relevant for butchering, since early metal tools are more costly, dull more quickly, and less easily accessible. It takes time for these tools to become cheaper, hold a sharper edge longer, and be more readily available. In other words, while the use of metal spread throughout the Near East, stone tools continued to be used for daily tasks. Only when metal knives become abundant and less expensive do they become more accessible to most of the population and are eventually adopted for quotidian purposes.

Adoption of quotidian metallurgy in the ancient Near East

During the latter half of the 3rd millennium BCE, harder metal alloys such as 10% tin-bronze began to appear across the ancient Near East (Avilova, 2009; Hauptmann and Pernicka, 2004; La Niece et al., 2007; Moorey, 1982, 1988, 1994; Philip, 1991; Rosenfeld et al., 1997; Yener, 2000). Despite the appearance of harder alloyed copper at this time, there is little discussion of its importance in quotidian activities such as meat processing. This is largely due to the lack of metal objects in settlement contexts and a biased archaeological record aimed at mortuary or buried hoard deposits that mostly represents the higher status population through the appearance of prestige, martial, and display goods (Batiuk and Harrison, 2017; Philip, 1988, 1989; Philip et al., 2003). In fact, the likelihood that most copper and bronze objects survive the forces of recycling and degradation is low (Greenfield, 1999; Northover, 2000, 113; Philip et al., 2003, 92), which makes understanding the nature and important use of metallurgy across an entire population difficult to assess.

Metal knife marks on animal bones only begin to appear with some frequency during the Middle Bronze Age (MBA, c. 2000–1550 BCE) of the southern Levant and Anatolia, where such studies have been conducted already (Beller and Greenfield, 2026; Greenfield, 2005, 2013; Greenfield, 2021; Greenfield et al., 2021; Greenfield et al., 2022a, 2022b; Greenfield and Chaput, 2021; Greenfield and Marciniak, 2019; Greenfield et al., 2018). This is not surprising given that 10% tin-bronze objects only appear towards the end of the EBA in these regions. The constituent parts of bronze tools had to be imported and transported over great distances across the Near East. Bronze would be an expensive resource and was likely limited in its distribution to elites and for display and ritual (non-quotidian) activities, such as warfare, status enhancement, ceremonial, and ritual. However, even with the use and development of metal weaponry and tools, the use of stone continued in the MBA and Late Bronze Ages (LBA, c. 1550–1200 BCE). It is not until the LBA that metal slice marks on animal bones dominate assemblages. This pattern suggests that the initial adoption of metal knives for daily activities only occurred during the MBA and became widespread during the LBA (Beller and Greenfield, 2026; Greenfield, 2013; Greenfield, 2021; Greenfield et al., 2022b).

Two types of instruments may have been used for animal butchering: daggers and knives. Daggers and knives have very different functions, and only the latter are more useful for butchering than a stone tool. Daggers are used for stabbing, while knives are used for slicing. Daggers generally have two edges, while knives will have a single slicing edge (Skak-Nielsen, 2009; Zimmermann, 2004, 2005). The cutting edge of daggers is generally too thick to be used for effective and systematic slicing actions. The appearance of knives with thinner, sharper edges at this later date is likely due to the increasing sophistication of manufacturing processes. Early daggers (and knives) had a low tin-copper ratio (2%–5%), while later daggers and knives had a high tin-copper ratio (c. 10%), thereby allowing for a sharper edge to be created (thus, sharpened) (Moorey, 1982, 1994). The softer metal of EBA daggers would not have retained their slicing edge long (even with repeated sharpening), which is why there is little evidence that they were used regularly for butchering. Daggers appear earlier in the EBA (3rd millennium BCE) in the Near East (Arcan, 2012; Montanari, 2014; Nigro et al., 2018), while knives appear much later in the EBA, toward the end of the 3rd millennium BCE (Moorey, 1994).

Bronze knives appear only sporadically toward the end of the Ur III Period (Late EBA), albeit in small quantities (Moorey, 1994). Knives appear more commonly in this region during the second millennium, in the ensuing Old Babylonian Period (aka the Middle Bronze Age/MBA) and elsewhere in the Near East. Most examples are found in funerary or elite contexts (i.e., areas of settlement identified as having more economic or symbolic wealth in society) (Hauptmann and Pernicka, 2004; Moorey, 1994; Philip, 1989).

In contrast to the limited distribution of metal knives, chipped stone implements dominate domestic tool assemblages throughout Mesopotamia during the 3rd millennium BCE, testifying to their continued importance during the EBA (Algaze et al., 2021; Frahm, 2014a; Hartenberger et al., 2000; Helms, 2014; Ibragimova, 2014; Postgate, 2023b). This persistence suggests a technological conservatism shaped by both resource availability and functional adaptation.

While metallurgical knowledge existed, the economic cost of copper importation and the relative efficiency and maintainability of stone blades made lithic tools more viable for daily subsistence activities. Yet, it remains difficult to understand the role, nature, and frequency of use of metal knives, given the few examples and the limited contextual distribution (mostly funerary). As well, the different roles played by metal vs. chipped stone knives during these crucial time periods remains uncertain.

Case study: Abu Salabikh

The site of Abu Salabikh offers a uniquely valuable context for investigating the social and technological dimensions of this gradual transition from stone or metal blade tools used for animal butchering. It is located in the southern Mesopotamian alluvial plain on a tributary of the Euphrates River in southern Iraq (Figure 1). It is a moderately sized settlement for this time period c. 20–25 hectares. The archaeological materials derived from excavated private houses and associated burials suggest that the site was primarily inhabited by a non-elite majority. Additionally, there is no evidence indicating it was a religious capital, although a contemporary temple was discovered and excavated (Martin et al., 1985; Matthews and Postgate, 1987; Postgate, 1982; Postgate, 2023a; Postgate et al., 1983; Steele, 1990).

History of occupation

Archaeological evidence indicates continuous, though punctuated, occupation from the Late Uruk (c. 4000–3000 BCE) and Jemdet Nasr (c. 3000–2900 BCE) periods through the Early Dynastic Period (ED, c. 2900–2350 BCE) I-III phases, with its greatest expansion during the ED III before abandonment in the later third millennium BCE. Between the Jemdet Nasr and ED phases, there is a brief hiatus. The site is reoccupied during the ED I Period (c. 2900–2750 BCE), based on material from Area A on the eastern mound. It reaches its largest extent during the ED III Period, until it is abandoned toward the end of ED IIIa (c. 2600–2450 BCE). Archaeological finds at the site include tablets (East Mound, Area E) and structures that may have served religious purposes and housed individuals engaged in such activities (Biggs, 1974, 16–18; Martin et al., 1985; Postgate, 2023a).

Site description

The site is composed of three large mounds– a long mound to the east (Areas A and E) that housed a citadel, temple and administrative area, and elite residences and two smaller mounds to the west (the Uruk and West mounds). The east and west areas are separated by the low depression of an ancient canal. The Area E mound housed a domestic neighbourhood from which the animal bone data for this study derive (Figure 2). The Area E mound is dated exclusively to the ED III Period.

Excavations in domestic sectors, particularly Area E, reveal a non-elite neighbourhood where food production and household activities occurred in direct association with intra-mural burials – contexts rarely preserved in urban Mesopotamian settings.

The House 6H

House 6H is from the SE quarter of the East Mound (Area E), and is dated exclusively to the ED III Period (c.2600–2350 cal BCE). The entire building and surrounding complex were excavated in the 1980's. It includes several rooms, courtyards, pits, burials, and streets/lane. It is an affluent yet non-elite residence (Postgate, 2023a). The house is interpreted as lying within a domestic neighbourhood. It contained at least 18 rooms with a second story, grand reception rooms and both public and private areas (Figures 3a, b). There is an understanding of household activities and associated behaviour of several rooms (Greenfield and Gaastra, 2020, 2023). Some of the more interesting contexts, both within and outside the house, in the study are described next (Table 2).

Table 2. List of selected rooms, graves and other locations of 6H house, their descriptions, loci, SEM specimen number, and deposit description.

The rooms

• Room 55b is undesignated within the house for a particular activity or use.
• Room 58 is regarded as an outdoor space where “messy” activities may have occurred.
• Room 65 is the interior courtyard located at the centre of the house.
• Room 66 is designated as the main entrance of the building.
• Room 67 is a kitchen annex with associated hearths and tanurs.
• Room 70 was originally the main reception area with a central hearth, but it was later divided into smaller sections (Rooms 71 and 72). The purpose of these two rooms is unclear, but there may be continuity in their activities.

The graves

Within the house, there are also human burials found both within the courtyard and spaced throughout the rooms on the main floor. Multiple graves were excavated, from which human and animal bones and grave goods were recovered. Three specific graves are highlighted here since some specimens with butchering data used in this paper were derived from them.

Grave 223 (containing two superimposed graves of adults which were previously robbed and therefore composed of mostly disturbed contexts) was found in Room 67. Both graves appear to have been lined, one with mortared bricks (Grave 223a) and the other with pisé (Grave 223b). The presence of shattered pottery (including many spouted jars and conical bowls), cosmetic shells, fragments of an ostrich eggshell vessel, fragments from stone bowls and copper objects, a fragmentary silver ring, as well as the occasional bead (all lapis lazuli), likely belongs to an original set of jewellery in one or both graves. A stone mortar and pestle were found associated with both graves, as well as a unique ceramic funnel in Grave 223b and a cuneiform tablet. Fragments of a flint sickle blade and one long flint blade were also found (Steele and Luby, 2023).

Grave 229, also robbed in antiquity, was found within Room 58n. This unlined grave contains a single subadult with a shell ring and three conical bowls (Steele and Luby, 2023).

Grave 244 was found within Room 55b. Although robbed in antiquity, it contained enough preserved stratigraphy to indicate that it was associated with the occupation of House 6H. The grave contained the thorax of an equid in one corner, with the fore and hind quarters destroyed by looting pits in antiquity, disturbing the position of the limb bones. Another undisturbed area of the grave contained a set of cosmetic shells on reed matting. Another third undisturbed corner contained a collection of copper pins, cosmetic shells and many beads from a necklace of carnelian, rock crystal, shell, copper alloy and lapis lazuli. A copper axe and a silver finger ring were found in the disturbed fill along with fragments of stone vessels, multiple jars and conical bowls, an ornate limestone bowl, a cuneiform tablet, a clay cone, a clay sickle, a clay chariot wheel, an animal figurine, spindle whorls and additional beads. Four flint blades were also found (Steele and Luby, 2023).

The lane

There were two walkways around house 6H, a street, and a lane that was possibly situated between two houses (Postgate, 2023a). These areas offer a unique glimpse into what was discarded outside the home, but not in garbage or middens. One specimen from the lane was examined for this paper.

Artefactual assemblage

Nineteen pieces of chipped and ground stone tools (not including mortars and pestles) were recovered from the house. The chipped stone tools include parts and whole sickle blades (N = 10), unmodified flint blades (N = 9), and one ground stone axe head – the head of which is ground to a narrow edge and worn from an indeterminate use (Steele and Luby, 2023).

The larger number of stone tools is contrasted by the scarcity and indeterminate nature of the copper artefacts. These include two pieces of wire, two shapeless lumps of corroded metal, an ovoid fragment, and two nail or pin shafts (Postgate, 2023b, 170–176).

General faunal assemblage

As the faunal assemblage from House 6H is dated to the ED III, it was therefore possible to present it as a single analytical unit. The animal remains discussed here include the butchering data. The general faunal assemblage was recovered from all contexts within and immediately outside the house, including rooms, street, lane, graves, and pits (Figure 4). Preliminary zooarchaeological analyses of the fauna from the earlier excavations has already been completed and published elsewhere (Clutton-Brock and Burleigh, 1978; Greenfield and Gaastra, 2020) and are briefly summarised below.

Faunal quantification

The faunal data can be assessed using different quantification measures. First, there is the traditional measure of simply counting each bone as a separate specimen, but considering obvious articulations (NISP - the Number of Individual Specimens or N for short). Second, the Number of Specimens with Butchering Marks (bNISP) on them is also quantified. Simply put, NISP counts a single specimen once, but corrects for articulations. This is used here for the faunal assemblage in general. In contrast, bNISP is more specific as it counts the number of specimens with butchering marks. This is a subset of the larger NISP (Beller et al., 2022; Greenfield and Brown, 2016; Greenfield et al., 2018; Greenfield et al., 2016).

Domestic vs. wild fauna

The overall assemblage from House 6H is 2520 NISP (Table 3). Less than half of the assemblage was identifiable to some taxonomic level (47.1% animals, plus 0.4% Homo sapiens). When only discrete, taxonomically identifiable domestic or wild species are considered, the total assemblage declines to 37.4% of the assemblage (943 NISP). Of this assemblage, the vast majority consists of domestic taxa, while wild taxa constitute a statistically insignificant proportion. The indeterminate domestication fraction consists of fragments that could not be assigned to a specific taxon, such as those labelled “medium mammal” or “small bovid”. These categories primarily include unidentifiable fragments and have been excluded from further analysis of the faunal remains from House 6H. Slightly under a quarter of the fragmented remains can be attributed to the indeterminate domestic or wild taxa category (n = 243, or 20.5% of the total). This is true for secure depositional contexts (Table 3; Figure 4 and Figure 5) (Greenfield and Gaastra, 2023, table 7.1.4).

Table 3. Overall distribution of animal remains recovered from house 6H at Abu Salabikh (Greenfield and Gaastra, 2023).

The bolds signify totals and subtotals for the various categories.

Taxonomic diversity

The highest frequency of identifiable domestic taxa includes Ovis aries/Capra hircus (n = 463, or 39%), followed by pig (Sus scrofa domesticus) with n = 290, or 24%) specimens, and donkey (Equus asinus?) with the third highest frequency, with n = 27 or 2.3% of the assemblage (Table 3). Cattle (Bos taurus) have a much lower frequency, n = 18, or 1.5%, followed by dog (Canis familiaris), with 16 + one Canis?, for a total of n = 17, or 1.4% of the total corpus of specimens. Of the wild taxa, most are mammals (n = 90, or 70.3% of wild), with the remaining distributed among fish, reptiles, birds, and molluscs (n = 38, or 29.7%) (see Table 3 for all taxonomic categories described here).

Of the rooms with faunal remains available for study, 297 specimens (including four human remains) were identified from discrete rooms only within House 6H (see Figure 5).

General butchered assemblage

An initial total of 52 specimens (52 specimens (54 = butchery NISP, or bNISP) were identified in the field as exhibiting butchery marks of one type or another (e.g., slices, chops, bashes, scrapes, etc.). An additional two butchered specimens were subsequently identified based on microscopic analysis and added to the butchering analysis for a total bNISP=54. The term “cut” mark is avoided here as it may represent any or all types of butchering marks (e.g., slice, chop, bash/batter, saw, etc.). Instead, we use the term slice marks.

Of the butchering marks, the most common were slice marks (57% of bNISP), followed by fine slices (19%), and chop (13%) (Table 4). The slice marks (inclusive of both slice and fine slice marks) were studied more intensively to determine whether stone or metal blades/flakes were used during animal carcass processing.

Table 4. Absolute and relative frequencies of butchered bone types and associated context types within the general faunal assemblage identified in the field (NISP).

The bolds signify totals and subtotals for the various categories.

Butchered frequencies by context

This study was conducted on remains from only the EDIII period, yet clear differences were found within and between the rooms of House 6H (Greenfield and Gaastra, 2023, 32–33). Butchered specimens appeared to be clustered in Rooms 56, 65, and 67 (Table 4). Each of these specimens had one or more slice marks. There was no evidence for bashes, chops, scrapes, or battering on any of the bones. Rooms 65 (the courtyard) and 56 (domestic work area) had the highest frequency, with the latter having the most slice marks in relative total (3 = bNISP, 10% of total NISP).

Butchered taxonomic distribution

Several different domestic and wild taxa were identified with butchering marks (N = 54). The vast majority are domestic and domestic? (87% bNISP, Table 5), followed by an unknown state of domestication (9%), and wild and wild? taxa (4%). This distribution mirrors the taxonomic representation of the faunal assemblage in general (Table 3).

Table 5. Butchering marks by taxon (bNISP). “?” signifies that the taxonomic status is fully uncertain.

The bolds signify totals and subtotals for the various categories.

The frequency distribution of osteological elements demonstrates that almost every major bone displays some evidence for butchering. The entire body is butchered and used for food or some other activity (such as tools) (Table 6).

Table 6. Frequency distribution of osteological elements (bNISP).

The bolds signify totals and subtotals for the various categories.

Butchering mark type

Butchering tool types were identified based on their morphology. Three different butchering marks were identified in the assemblage. These include bash, chop (13%), scrape (2%), and slice marks (85% bNISP, see Table 7). This indicates that a variety of tools were utilised in the butchering process.

Table 7. Frequency distribution of butchering mark type (bNISP).

The bolds signify totals and subtotals for the various categories.

Taphonomy of butchered assemblage

The taphonomy of only the butchered assemblage (N = 54) used in the butchering mark study is important in the analysis of butchering marks because various attritional forces can affect the assemblage and thereby any interpretations on the material. Those identified in the assemblage include weathering, gnawing, cooking and cultural modifications. Some of these are described next.

Weathering

The four main stages of weathering are fresh, light, medium and heavy. Fresh bones do not exhibit any erosion, cracking or exposure to the elements. Light weathering is when there are small cracks that are mostly parallel to structure of the bones. Bones are classified as medium weathered when there is flaking and/or loss of bone surface with enlarged cracks. Heavy weathering is exhibited when there is considerable bone surface loss and splintering, and the bone is structurally falling apart (Behrensmeyer, 1978; Greenfield, 1986; Meadow, 1978).

The vast majority of specimens exhibited only light wear (91% bNISP – Table 8). A few specimens reflecting characteristics of medium (4 = bNISP, 7%) and heavy (1 = bNISP, <2%) wear. This pattern suggests a high level of preservation at the site.

Table 8. Butchered specimens by stage of weathering (bNISP).

The bolds signify totals and subtotals for the various categories.

Gnawing

There is almost no evidence for gnaw marks on any of the butchered specimens which indicates that carnivores, such as canids (dogs), omnivores (pigs), or rodents, can be removed as a source of attrition (Atici, 2006; Brain, 1969; Greenfield, 1988; Payne and Munson, 1985). A single specimen appears to have been gnawed at one end of the bone (See Figure 6 right end of bone).

Burning

The type of cooking exhibited on bones was recorded based on a few variables – colour and texture. Burned bones are those displaying charring or calcination and were in direct or nearby contact with an open flame. In contrast, boiled bones are those that fall into the tan and light brown range as they were likely cooked as part of stews. Bones encased in meat cooked over a flame would take on a similar appearance, as they were not directly exposed to the flame and were not super-heated. They are indistinguishable and treated as boiled (Buikstra and Swegle, 1989; Greenfield and Beattie, 2017; Koon et al., 2010; Shipman et al., 1984).

A relatively high number of butchered specimens displayed evidence for heat or thermal alteration, such as burning or boiling (bNISP=14, 25.93%, Table 9) (Greenfield and Beattie, 2017). This suggests that food was cooked with both types of heat alteration.

Table 9. Frequency distribution of cooked butchered bones (bNISP).

The bolds signify totals and subtotals for the various categories.

Cultural modifications

Cultural modification, such as shaping, of bones for use as tools, ornaments, or playing pieces, is a common feature of bone assemblages. Some are morphologically identifiable as modified by their shape. However, most are identified based on surface use wear polish (Dandoy, 2006; González-Urquijo and Ibáñez-Estévez, 2003; Olsen, 1989; Plug, 1987).

A relatively large number of the butchered specimens were modified into tools, gaming pieces, or ornaments (35%; 19 = bNISP - Table 10). In many cases, the shaping of the bone smoothed out the bone surface to such an extent that it is not possible to determine if the grooves were made by stone or metal tools (Figure 6). These are classed as indeterminate in the technological analysis (below).

Table 10. Frequency distribution of culturally modified (e.g., tools, ornaments) butchered remains (bNISP).

The bolds signify totals and subtotals for the various categories.

The high frequency of culturally modified remains raises the larger issue of why there are so many modified bones on the floors of houses. These fragments are the broken pieces of the bone industry and probably reflect their use rather than being subsistence. This is a common pattern in early urban sites (Richardson, 2024, 2025).

Method of analysis

A great deal of experimental work has been conducted over the past 50 years to define the relationship between slicing groove morphology and the nature of butchering technology. The raw material of the butchering implement and the type of tool can be reconstructed from the shape of the depressions and grooves with fair certainty (Binford, 1981; Fisher, 1995; Gifford-Gonzalez, 2018; Greenfield, 1999, 2006; Lyman, 1994; Olsen, 1988; Semenov, 1964; Shipman, 1981; von Lettow-Vorbeck, 1998; Walker, 1978; Walker and Long, 1977). This method has been productively utilised to quantify the relative importance of metal and stone tools over time and for various functions for a number of regions in the Near East and Europe (Greenfield, 1999, 2013; Greenfield, 2017; Marciniak and Greenfield, 2013). These laid the groundwork for a suite of new techniques using a variety of technologies that have largely supported the method of analysis (Abe et al., 2002; Aramendi et al., 2023; Bello, 2011; Bello and Soligo, 2008; Cifuentes-Alcobendas, 2019; Costamagno et al., 2019; Courtenay et al., 2020; Dominguez-Rodrigo et al., 2020; Maté-González et al., 2017; Maté-González et al., 2018a, 2018b; Maté-González et al., 2019; Moretti et al., 2015).

Microscopy

The presence of butchering marks on the bones is initially identified using a combination of the naked eye, a hand-held magnifying glass, and low powered microscopic instruments, such as a binocular light optical microscope (25–40x) or Dino-Lite. Unfortunately, it is not possible to identify the type of raw material (e.g., stone vs. metal) that made the butchering mark by the naked eye with any certainty. As a result, various microscopy techniques are used. With training, light optical microscopes can be used to do tentative identifications.

As demonstrated by all the above-cited studies (and others that may have inadvertently been missed), it is essential to ultimately conduct a microscopic study of the butchering marks to determine whether the butchering marks were made by stone or metal tools, type of tool, raw material, and production type (e.g., uni-/bifacial production or retouch). For example, what may seem like a clear-cut case of a single sharp metal slice mark when viewed in a light optical microscope may actually be two or more intersecting grooves made by a chipped stone tool that smooths out the lateral striations in the grooves to mimic a metal blade.

The characteristics that distinguish one butchering technology from another (e.g., stone from metal, production type, etc.) can be most reliably identified using high-power microscopy. While it is possible to use a low-power light optical binocular microscope for preliminary identification, these should be verified under high-power microscopes, such as a SEM.

Scanning electron microscopy (SEM)

SEM technology has long been used to distinguish the type of raw material and production type of the tool that created grooves (Potts and Shipman, 1981; Shipman, 1981). A number of studies have since successfully focused on the differentiation of metal and stone tools using SEM technology (e.g., Alunni-Perret et al., 2005; Beller et al., 2021; Costamagno et al., 2019; Greenfield, 1999, 2006; Greenfield, 2000; Greenfield and Marciniak, 2019; Jones, 2011; Kooi and Fairgrieve, 2013; Olsen, 1988; Pollio, 2009; Shipman and Rose, 1988; von Lettow-Vorbeck, 1998).

Since most bones and fragments are too large to fit into a SEM chamber for viewing, high quality silicone moulds are made of the slice mark. Most SEMs have only a small chamber that prevents the analysis of archaeological materials directly. Others may have larger chambers. Silicone moulds can capture the microscopic diagnostic characteristics of slicing marks when applied to the surface of the bone (Frahm, 2014b; Greenfield, 1999; Olsen, 1988; Rose, 1983). The silicone mould creates an inverted or reverse image of the groove which makes visible diagnostic criteria that are often difficult to observe inside the groove because of the narrow field of view. If the mould is viewed from above, the shape of the apex and the diagnostic criteria on both sides of the groove can be observed and measured. This multi-level microscopic analysis ensures that the diagnostic characteristics of stone vs. metal tool marks are visible. For each butchering mark, the type, number, location, orientation, and representative activity are also documented (skinning, disarticulation, etc.).

When a silicone mould is viewed in a SEM, one sees an inverted image of the original groove and both sides of the apex can be viewed and compared simultaneously. Chipped stone tools have two very different sides – a smooth ventral and uneven dorsal face – which display diagnostics typical of their surfaces. The ventral face of a unifacially produced unretouched flakes/blades will produce a relatively even surface. In contrast, the dorsal face will display a more uneven surface with ridges or lateral striations below the apex. The more the dorsal face is chipped, the more uneven it will become. Retouching adds more variability to the face of the tool and creates a rougher surface with more striations. In unifacially retouched flake/blades, one side is rougher than the other. In bifacially retouched flake/blades, both sides are very rough in profile (see Supplementary Appendix 1).

The silicone moulds were analysed with the Scanning Electron Microscope (Zeiss EVO10) in the Near Eastern and Biblical Archaeology Lab of St. Paul's College, University of Manitoba (by HJG and ES).

Butchering mark diagnostics

Since in-depth descriptions of the diagnostics for each type of butchering mark are found in numerous previous studies (see above), these will not be repeated in detail here. The criteria for distinguishing between stone and metal tool slicing marks are widely accepted and summarised next.

Metal knives leave slicing grooves that are deep and symmetrically V-shaped in profile with steep sides. They culminate in an apex that has a sharp point. The groove can be deep and narrow or wide depending upon the nature of the blade. Metal knives can create a flat-bottomed |_|-shaped profile when they are dull. The sides will be relatively smooth and uniform, without ancillary striations. They leave either no striations or striations of a more uniform depth and spacing than when stone tools are used. In general, metal knives produce a cleaner and more even groove (except for scalloped-edge knives and saw-like blades).

Chipped stone slicing tools (whether flakes or blades) produce a more shallow, and more variably shaped groove. Because of the sinuosity of their cutting edges, chipped stone tools tend to produce wider and more irregular grooves. Chipped stone tool slice marks are always uneven in cross-section, with one side rising relatively steeply to the apex, then descending gradually or in a series of ancillary ridges. Generally, on a unifacially produced, but not retouched flake or blade, there will be one or at most two lateral striations. They are found lateral to the apex of the groove and are of uneven length and thickness.

In addition to slicing marks, two types of impact marks were identified in the assemblage: chops and bashes. Chopping with an axe leaves a deep V-shaped imprint in the bone. Unfortunately, only half of the imprint of the axe is left on the bone when it is completely severed. Stone and metal axe marks can be distinguished by the shape of their imprint (Okaluk, 2020; Okaluk et al., 2026).

Bash marks, by contrast, result from blunt percussion, typically caused by hammer stones during marrow extraction. These are recognizable through conchoidal or scalloped depressions, commonly located at the fractured edges of long bone shafts (de Gruchy and Rogers, 2002; Maltby, 2007; Tucker et al., 2001). Together, the range of mark types reflects multiple stages of animal processing, from initial carcass division to secondary exploitation of resources such as marrow and grease.

These morphological contrasts are critical for reconstructing the type of implement used and, consequently, for evaluating the tempo of technological change in butchering practices across time and space.

Microscopic analysis of selected butchered samples with slice marks

Within the larger butchered assemblage subset, 16 (bNISP) specimens were selected for detailed microscopic and SEM analysis based on the clarity and preservation of single or non-overlapping slicing grooves (also referred to as slice marks). See Supplementary Appendix 1 for images of each bone with butchering marks used in this microscopic study and for SEM pictures of silicone images of butchering marks.

Slice marks

Such specimens, which have preserved discrete and undisturbed striations, are particularly suitable for identifying tool type and reconstructing butchering behaviour. Overlapping or repeated slicing obscures original morphology, as later incisions can alter or mimic metal-like profiles, thereby complicating interpretation (Greenfield, 2021). The 16 specimens derive from a range of domestic, burial, and exterior contexts within House 6H and reflect activities that encompass daily subsistence processing, secondary refuse disposal, and post-depositional mixing (Table 11). The largest number of butchered remains is found in the Courtyard (Room 65, N = 5) and Room 58 (N = 3). In some of the other rooms and graves, and the lane, there are only 1-2 butchered specimens in each. However, most of the rooms in House 6H do not contain any butchered remains. These distributions provide a useful cross-section of the domestic environment and allow exploration of how butchering technologies were deployed in varied settings within an ED III household.

Table 11. Selected SEM analysed specimens (N = 16) with slice marks from house 6H including rooms within and outside the structure, the graves, and the lane.

In the field, using a hand-held magnifying lens and a Dino-Lite, many of the slice marks appeared to be made by metal tools. However, when multiple slice marks overlap, the subsequent ones destroy or modify the morphology of the preceding ones. In such cases, chipped stone tool slice marks can be transformed into what appear to be metal tool marks (Greenfield, 2021). Hence, the situation changed when the moulds were viewed in an SEM, and the intersecting grooves became visible (See images in Supplementary Appendix 1). This is one of the dangers of relying in the field on magnifying glasses and Dino-Lites where the depth of field is minimal and relief is flattened. A similar situation was encountered previously elsewhere, particularly when obsidian is the primary raw material for slicing actions (Greenfield and Chaput, 2021; Greenfield and Marciniak, 2019).

Sixteen specimens were ultimately identified as exhibiting high quality butchering marks deemed suitable for SEM analysis (See images in Supplementary Appendix 1). These were specimens with either single slice marks or multiple slice marks without interceptions. They are the most optimal specimen type for analysing the slice mark morphology to determine the raw material type employed in the animal butchery process.

Based on the criteria described above for distinguishing metal from stone slice marks, the specimens identified in this study demonstrated that only stone tools were utilised in the butchering process. This is evident by the nature of the groove created by the various tools. All 16 specimens selected for SEM analysis displayed morphology consistent with chipped stone use. No specimens displayed morphology diagnostic of metal tool marks within the analysed sample. Each of the slice marks displays a lopsided V-shaped groove with ancillary striations consistent with marks produced by chipped stone tool blades/flakes (See images in Supplementary Appendix 1). As a result, there is no evidence for the use of metal-based butchering tools at ED Abu Salabikh.

Chipped stone blade/flake production type

It is impossible to distinguish between slicing grooves made by unretouched chipped tool blades and flakes, as they share the same basic edge morphology (Greenfield, 2006).

All slice marks, except one (15 = bNISP) display the characteristics of a unifacially produced unretouched chipped stone blade/flake. One side of the apex is generally smooth, while the other side displays only one or two unevenly spaced lateral striations (see images in Supplementary Appendix 1).

Only a single specimen displayed any evidence that it may have been a unifacially retouched chipped stone blade/flake (Specimen 8281-9 in Supplementary Appendix 1). None were identified as produced by a bifacially produced chipped stone blade/flake. These data demonstrate that unretouched unifacially produced chipped stone tool blades/flakes were the primary tool type utilised for animal butchery at Abu Salabikh.

Discussion on the results

The results from Abu Salabikh reinforce a broader Near Eastern pattern in which technological adaptation in animal processing unfolded gradually rather than abruptly across the third millennium BCE. Despite the emergence of copper metallurgy in elite and ceremonial contexts, the evidence from both tool assemblages and butchery mark morphology indicates that lithic implements remained the dominant technology for quotidian activities. This pattern, also seen in Egypt, the southern Levant, and Anatolia, illustrates a fundamental principle of human adaptation: technological innovation is mediated by local resource constraints, cultural preference, and the perceived reliability of established practices rather than by the simple availability of new materials. Within a complex adaptive system such as ED Mesopotamia, decisions about tool use were shaped by feedback loops between ecological cost, social hierarchy, and the everyday demands of household production.

The data analysis of butchering marks from Abu Salabikh demonstrates that unretouched unifacially produced chipped stone tool blades/flakes were the primary tool type utilised for animal butchery during the ED III occupation of the site. At Abu Salabikh, the absence of metal slicing signatures corresponds directly with the archaeological record, where copper-based artefacts are scarce and confined to small items such as pins and chisels (Baker et al., 1993; Martin et al., 1985) No evidence suggests that copper or bronze blades were employed in animal butchery. Instead, the microscopic analysis of slicing marks reveals the continued dominance of chipped stone tools for carcass processing. This persistence underscores not technological stagnation, but strategic adaptation. Stone knives were inexpensive, easily resharpened, and well suited to local subsistence needs, whereas metal tools were rare, costly, and embedded in social or symbolic spheres—used for adornment, weaponry, or ritual deposition (Hayden, 1998; Joyce, 2005; Moorey, 1982, 1988). Such selective adoption is typical of many societies, where effective traditional technologies are retained within dynamic social systems, even amid the introduction of novel materials (Bentley et al., 2008; Rosen, 1997).

The delay in the integration of metal tools into the fabric of daily life highlights the socioeconomic dimensions of technological choice. In Mesopotamia, as in neighbouring regions, the high cost of copper - an imported commodity given the region's lack of native ore sources - restricted its accessibility. Written and archaeological evidence from the late fourth millennium BCE attest to long-distance exchange networks supplying copper from Oman, Anatolia, and Iran (Englund, 1998; Moorey, 1994; Muhly, 1973). This reliance on external sources meant that metal objects carried both material and ideological value, circulating primarily among elite groups who controlled trade and production. Consequently, the adoption of metal for subsistence activities such as butchering would have been economically inefficient for most households. The continued use of chipped stone, therefore, represents a rational adaptation within the constraints of a resource-limited and socially stratified economy in the absence of local sources (Payne 1980).

From a technological systems perspective, the Abu Salabikh assemblage contributes to understanding how early urban societies managed innovation within stable cultural frameworks. The continued use of lithic butchering tools in a context otherwise characterized by administrative and architectural complexity demonstrates that technological modernization did not progress uniformly across sectors of production. Instead, it was negotiated within the micro-scale dynamics of household economies. These decisions were not simply conservative but functionally and socially embedded - illustrating the interplay between efficiency, tradition, and symbolic meaning that defines the evolution of technological systems in complex societies (Hodder, 2012). Similar adaptive equilibria are evident in contemporary contexts across the ancient Near East, including the Nile Delta, where flint remained a primary butchering material well into the Old Kingdom despite state-sponsored metallurgical development (De Sousa and Greenfield, 2023; Graves-Brown, 2015), and in Anatolia and the Levant, where the dual use of metal and stone persisted for centuries (Greenfield, 2021, 2013; Greenfield and Marciniak, 2019).

Thus, the evidence from Abu Salabikh reflects not technological lag, but a deliberate and contextually rational pattern of adaptation. Stone tools continued to fulfill essential economic and culinary functions within an environment that demanded reliability, accessibility, and efficiency over novelty. The absence of metal butchering marks on faunal remains aligns with both the limited presence of copper artefacts and the broader socioeconomic realities of ED southern Mesopotamia, where metallurgy was a symbol of elite identity rather than a transformative force in domestic production. In this sense, the butchering marks from Abu Salabikh capture a subtle but critical dimension of human adaptation: the negotiation between innovation and tradition within the everyday material practices of complex early states.

Copper-based butchering implements (such as knives) are largely absent from most archaeological records of the Near East during the third millennium BCE except for a few elite contexts in burials/tombs and hoards (De Sousa, 2020; Montanari, 2015; Moorey, 1994; Philip, 1988). Metal objects (e.g., pins or chisels) during this period are largely for display (e.g., pins, jewelry), symbolic (grave goods), and warfare (daggers, swords, etc.). Even though metal knife blades first appear in the late EBA of the Near East, they are not utilised in butchering activities. It can be concluded that they are not a significant part of the toolkit used in food processing during the third millennium BCE. It is only in the MBA and later periods that metal knife blades become a regular part of the butchering technology (Greenfield, 2026; Greenfield et al., 2022b).

The lack of metal knives for quotidian activities is a common theme among metallurgical studies of the Near East during the Early Bronze Age. In fact, it is only in the Middle and Late Bronze Ages, and eventually in the Iron Age, that metal knives became more widely adopted and nearly replaced stone tool technology for butchering (Greenfield, 2013). Thus, it would appear that the adoption of metal in Mesopotamia followed a similar pattern of development as in other areas of the Near East during the third millennium BCE (e.g., Egypt, Levant, Anatolia). Our study highlights the continued importance of chipped stone tools for food processing during the third millennium BCE and the lack of a direct relationship between bronze quotidian tools and increasing social differentiation in early Near Eastern complex societies.

Conclusion- Abu Salabikh in a wider context

At ED III Abu Salabikh, in House 6H, there appears to be some evidence for spatially distinct patterns of butchering activities that distinguish rooms, even though each room is clearly used for different activities, such as outdoor, indoor courtyard, grave, etc. From the general butchering analysis, it is evident that the entire body of each taxon is utilised by the inhabitants. When this is coupled with only chipped stone butchered material, it suggests that the non-elite residents of House 6H had access only to chipped stone tools for animal butchering and food preparation.

The absence of metal slice marks in the ED III faunal assemblage at Abu Salabikh from House 6H suggests that metal blades were not yet adopted for animal butchery. It further suggests that stone tool technology remained the primary method for animal butchery during this period of occupation at the site. While it is not surprising that only chipped stone materials were used for animal butchering in a domestic non-elite household, it is important to note the lack of difference between the butchering marks and stone tools (blades and flakes) found within graves and those in the reception room or kitchen. This suggests that butchering of animals was conducted with stone tools.

The transition from stone to metal butchering technology likely occurred later than the stratigraphic sequence associated with the ED III faunal assemblage at Abu Salabikh. Based on studies from other regions, it is likely that this transition happened during the MBA (De Sousa and Greenfield, 2023; Greenfield, 2013, 2017; Greenfield et al., 2021; Greenfield et al., 2022a, 2022b; Greenfield and Brown, 2016). Therefore, it post-dates the archaeological sequence at Abu Salabikh by several hundred years and more than 1000 years after bronze technology was introduced to the region.

Abu Salabikh provides a critical lens through which to examine the rate and nature of technological adoption in early complex societies. As a mid-tier urban settlement without the elite industrial infrastructure of major centres such as Ur or Nippur, it reflects how technological innovation diffused through the broader social fabric. The predominance of chipped stone tools, coupled with the scarcity of copper artifacts apart from small pins and chisels in the Abu Salabikh artefactual assemblages, underscores that the transition from lithic to metal technology was uneven and socially differentiated. As Abu Salabikh's House 6H was relatively affluent and likely participated in wider exchange networks, the continued use of stone for carcass processing cannot simply be attributed to poverty or lack of access. Instead, it signifies that technological adoption was task-specific and that lithic tools remained effective even among households with access to prestigious materials.

The butchering marks preserved on the faunal remains from its domestic contexts illuminate how non-elite households negotiated innovation – retaining familiar and effective stone tools while metal remained a prestige or ritual medium. By focusing on the microscopic and contextual evidence from Abu Salabikh, it becomes possible to trace not only the mechanical signatures of tool use, but also the adaptive decisions underlying them, thereby situating Mesopotamian households within a wider Near Eastern pattern of delayed yet ultimately transformative adoption of metallurgical technologies.

Postscript – thoughts for the future

The authors (and reviewers) recommend that the next stage is to begin to define potential cultural markers of butchering, just as there are for bone taphonomy since there are now sufficient data from a variety of time periods and cultures across Europe and the Near East. It is necessary to determine if there are cross-cultural commonalities that apply to the size of the carcass (e.g., Maté-González et al., 2019), cooking styles, butchering technology, bone morphology and animal physiology, in other words are the butchering patterns unique to each culture? This type of investigation will require a metadata analysis and is something for the future. This idea is indeed food for thought! It is also important to investigate the nature of lithic and metal procurement systems further to more fully understand the interplay of availability and cultural choice.

Generative AI statement

The author(s) declared that generative AI was not used in the creation of this manuscript.

Any alternative text (alt text) provided alongside figures in this article has been generated by Frontiers with the support of artificial intelligence and reasonable efforts have been made to ensure accuracy, including review by the authors wherever possible. If you identify any issues, please contact us.

Publisher's note

All claims expressed in this article are solely those of the authors and do not necessarily represent those of their affiliated organizations, or those of the publisher, the editors and the reviewers. Any product that may be evaluated in this article, or claim that may be made by its manufacturer, is not guaranteed or endorsed by the publisher.

Supplementary material

The Supplementary Material for this article can be found online at: https://www.frontiersin.org/articles/10.3389/fearc.2026.1812460/full#supplementary-material

Supplementary material Supplementary material at the source

Footnotes

Edited by: Lloyd Austin Courtenay, Université de Bordeaux, France

Reviewed by: Ruth Blasco, Institut Català de Paleoecologia Humana i Evolució Social (IPHES), Spain

Paula Mateo Lomba, UMR5199 De la Prehistoire A l'actuel Culture, Environnement et Anthropologie (PACEA), France