See	discussions,	stats,	and	author	profiles	for	this	publication	at:	https://www.researchgate.net/publication/233651242
Late	Chalcolithic	mass	graves	at	Tell	Brak,
Syria,	and	violent	conflict	during	the	growth	of
early	city-states
ARTICLE		in		JOURNAL	OF	FIELD	ARCHAEOLOGY	·	JUNE	2011
DOI:	10.1179/009346911X12991472411123
CITATIONS
13
READS
164
3	AUTHORS,	INCLUDING:
Arkadiusz	Sołtysiak
University	of	Warsaw
181	PUBLICATIONS			207	CITATIONS			
SEE	PROFILE
All	in-text	references	underlined	in	blue	are	linked	to	publications	on	ResearchGate,
letting	you	access	and	read	them	immediately.
Available	from:	Arkadiusz	Sołtysiak
Retrieved	on:	29	March	2016
Late Chalcolithic mass graves at Tell Brak,
Syria, and violent conflict during the growth of
early city-states
Augusta McMahon1
, Arkadiusz Sołtysiak2
and Jill Weber3
1
University of Cambridge, Cambridge, United Kingdom, 2
University of Warsaw, Warsaw, Poland, 3
University of
Pennsylvania, Philadelphia, Pennsylvania
Excavations and surveys carried out from the mid-1990s through 2009 at Tell Brak, northeast Syria, have
focused on reconstructing the socioeconomic complexity and physical growth of one of northern
Mesopotamia’s earliest urban settlements. The recent discovery of several mass graves on the edge of the
city, created at an important threshold in its physical expansion (ca. 3800–3600 B.C.), adds to a longstanding
debate about the connection between the growth of early city-states and violent conflict. These
graves, with their population of as many as several hundred primarily sub-adults and young adults, are
interpreted as the result of large-scale violent events and may provide evidence for the post-mortem
treatment of enemies. They offer a strong counterpoint to the dominant reconstruction of a peaceful
prehistory in the region.
Keywords: Mesopotamia, early urbanism, violent conflict, mass graves
Introduction
The origins of urbanism are traditionally thought to be
located in the alluvial plains of southern Mesopotamia
(south Iraq) in the later 4th millennium B.C. In the most
widely-accepted model, the earliest city is typified by
the massive settlement of Uruk/Warka, with its
monumental religious buildings and evidence for
writing, priest-king leadership, and industrial specialization
from ca. 3500 B.C. (Adams 1972, 1981; Adams
and Nissen 1972; Algaze 2008; Liverani 2006 [1998];
Nissen 2002; van de Mieroop 1997; Yoffee 1995).
Recent site-based and regional surveys in northern
Mesopotamia (northeast Syria and north Iraq) have
revealed the presence of large settlements with networks
of linked rural villages from at least 3800 B.C.,
several centuries before the apparent floruit of Uruk
(Oates et al. 2007; Ur et al. 2007; Wilkinson and
Tucker 1995). Tell Brak and Tell Hamoukar in the
Upper Khabur Plains of northeast Syria (FIG. 1)
experienced rapid physical expansion starting ca.
3800 B.C. or even earlier. Excavations at Tell Brak in
particular have exposed monumental architecture and
evidence of economic diversification and specialization
from the late 5th and early 4th millennia B.C.
(McMahon and Oates 2007; Oates et al. 2007).
Ceramic production was increasingly standardized,
textile production may have moved in part from
households to workshops, obsidian exploitation presented
a more directed acquisition strategy, and clay
container sealings imply a multi-level bureaucracy
with control over resources. By ca. 3500 B.C., the urban
trajectory of southern Mesopotamia had surpassed
that of the north, with the north experiencing varying
degrees of influence and control by southern colonists
during the ‘‘Uruk Expansion’’ (Algaze 1993, 2008;
Rothman 2001; Stein 1999). But, research at sites such
as Tell Brak shows that at least initially, northern
Mesopotamia may have led the way in urban growth
and socioeconomic complexity.
Warfare and the structured use of force are often
linked with definitions of modern and ancient states
(M. Weber 1968), and violent conflict within a circumscribed
territory is proposed as the primary cause
in an often-cited theory of state origins (Carneiro
1970). Discussion of warfare has been an occasional
feature of studies of late 4th millennium B.C. citystates
in Mesopotamia. Artwork from Uruk in the
Late Uruk period (ca. 3200 B.C.) includes cylinder seal
impressions showing a distinctive ‘‘priest-king’’ with
a spear standing before bound captives (Boehmer
1999: Taf. 8–17, Abb. 16–19; Nissen 2002: fig. 8);
additional contemporary glyptic images of bound
prisoners come from Susa and Choga Mish in
southwest Iran, Tepe Gawra in northern Iraq, and
Habuba Kabira in northwest Syria (Boehmer 1999:
Correspondence to: Augusta McMahon, University of Cambridge,
Department of Archaeology, Downing Street, Cambridge CB2 3DZ, U.K.
Email: amm36@cam.ac.uk
ß Trustees of Boston University 2011
DOI 10.1179/009346911X12991472411123 Journal of Field Archaeology 2011 VOL. 36 NO. 3 201
Abb. 122). The lack of historical texts from this
period makes interpretation difficult: the captives
could be foreigners, inhabitants of a rival local city,
or even rebellious city-state citizens. But the display
of warrior-like power and the ‘‘legitimate use of
force’’ aspect to the scenes are clear, suggesting that a
well-developed ideology of warfare existed by the late
4th millennium B.C. The city of Uruk expanded to ca.
250 ha in the Late Uruk period (Finkbeiner 1991;
Nissen 2002), and this expansion may have been
driven in part by immigration, since a depopulation
of its surrounding countryside is visible from ca. 3500
B.C. (Adams and Nissen 1972). This immigration,
which created an empty buffer zone around the site,
plus contemporary developments such as construction
of a wall around the site of Abu Salabikh, may
have been based in a need for defense, whether actual
or symbolic (Pollock et al. 1991; Yoffee 1995). But it
is the evidence of administrative hierarchies, new
urban-hinterland relations, and economic changes
that are most often addressed in studies of southern
Mesopotamia’s city and state origins (Adams and
Nissen 1972; Postgate 2002; Wright 1984, 2006;
Wright and Johnson 1975). Warfare features more
frequently in interpretations of mature Mesopotamian
city-states in the following Early Dynastic period,
especially the mid-3rd millennium B.C. By around 2400
B.C., textually-documented conflicts over land and
water were entangled with the expansion and consolidation
of royal power. The Early Dynastic citystates
and their kings are often characterized as
involved in chronic battles for land and resources
(Nissen and Heine 2009; Pollock 1999, 2007; Yoffee
1995). Yet textual descriptions of warfare, such as the
Umma-Lagash border conflict (Cooper 1983) or MariEbla
power struggle (Archi and Biga 2003), may have
been inflated for ideological reasons, while economic
accounts linked to outfitting expeditions for war are
few (Tonietti 2010). The conflicts the texts describe,
when observed from a long-term perspective, are
episodic and embedded within stronger evidence for
peaceful economic exchanges and regional cultural
interconnection across the 3rd millennium B.C.
Archaeological evidence for warfare in Mesopotamia
from all millennia is thin: site locations are not
defensive, city walls are less common than is assumed,
evidence of violence on skeletons is rarely reported,
and destruction of buildings is essentially unseen.
However, recent excavations in northern Mesopotamia,
as well as forcing regional and temporal shifts in our
Figure 1 Map of northeast Syria with locations of Tell Brak and Tell Hamoukar indicated.
McMahon et al. Mass Graves, Conflict, and Early Urbanism, Syria
202 Journal of Field Archaeology 2011 VOL. 36 NO. 3
reconstruction of the origins of urbanism, contribute
explicit evidence for violent conflicts. Evidence of
conflagration and piles of sling bullets dating to ca.
3500 B.C. have been recovered from Tell Hamoukar
(Lawler 2006), while mass graves dating to ca. 3800–
3600 B.C. have been discovered at Tell Brak.
Tell Brak and Early Urbanism
Tell Brak is one of the largest ancient sites in
northern Mesopotamia, with a height of more than
40 m and a horizontal extent of ca. 130 ha. Brak’s
location within the rainfall agriculture zone and its
strategic placement on trade-routes that criss-crossed
the region contributed to its size and its long and
complex occupational history. The central mound
dates from at least the Neolithic (7th millennium B.C.
or earlier) through the Middle Assyrian period (later
2nd millennium B.C.); the Outer Town saw variable
occupation from the 4th millennium B.C. through the
Abbasid period (FIGS. 2, 3) (Ur et al. in press). The site
experienced episodes of expansion and contraction
linked to regional climatic and political changes and
at times was less than 20 ha, in the Old Babylonian
period for instance. At its maximum of 130 ha in the
Late Chalcolithic 3 period (ca. 3800–3600 B.C.),
occupation covered the central mound and the unmounded
or low-mounded Outer Town. The central
mound has been the subject of years of targeted
excavation, from Max Mallowan’s groundbreaking
work in the late 1930s (Mallowan 1947) through the
research of David and Joan Oates starting in 1976,
Roger Matthews in the mid 1990s, and Geoff
Emberling and Helen McDonald in the late 1990s–
early 2000s. The Outer Town has been explored
through excavation and survey since the late 1990s.
Late Chalcolithic excavations
The Late Chalcolithic levels of the central mound at
Brak are sealed by large-scale 3rd millennium B.C.
constructions. Deep excavations in selected locations
have, however, shown that the central mound was
probably fully occupied during the entire 4th
millennium B.C., with a range of public buildings,
houses, and industrial areas. The Eye Temple at the
mound’s southern edge (FIG. 2), excavated by
Mallowan in 1937–1938, is traditionally dated to just
Figure 2 Topographic plan of Tell Brak and its ring of surrounding sub-mounds (shaded grey), with areas of excavated 4th
millennium B.C. occupation on the central mound indicated (1 m contours).
McMahon et al. Mass Graves, Conflict, and Early Urbanism, Syria
Journal of Field Archaeology 2011 VOL. 36 NO. 3 203
before 3000 B.C., but recent excavations in Area TW
indicate that the distinctive ‘‘eye idols’’ that gave the
building its modern name were already manufactured
in the early 4th millennium B.C. It is probable that the
date of the Eye Temple, and of the religion-centered
power it represents, should be pushed into the first
half of the 4th millennium B.C. Late 5th through 4th
millennium B.C. occupation layers were revealed in
nearby Area CH (Oates 2002). At the northern side
of the central mound, excavations in Area HS
(Matthews 2004) identified similar occupation layers
belonging to the late 5th through mid-4th millennia
B.C., including a massive wall and houses. Excavations
in Area TW, near the northern entrance to the
settlement, provide a sequence through the entire 4th
millennium B.C. During the Late Chalcolithic 2 period
(ca. 4200–3800 B.C.) this area was an industrial zone
for centralized manufacture of everyday ceramics,
tools, and textiles, plus items potentially symbolic of
differential wealth and power, such as obsidian beads
and inlays, shell inlays, and an elaborate obsidian and
marble chalice (McMahon and Oates 2007). These
industries were adjacent to, and may have been
controlled by the users of, a monumental building
(Emberling and McDonald 2003; McMahon and
Oates 2007; Oates 2005; Oates et al. 2007). This
building has a huge basalt threshold (ca. 1.561.8 m)
and mud-brick walls (ca. 1.8 m wide). Parts of two
rooms have been cleared, which were clean and give
little evidence for precise function. The scale of the
structure is non-domestic, while the location of the
doorway near a corner implies that it is not religious,
since contemporary temple plans have symmetrical
central doorways. For these reasons, a secular
administrative function is proposed for this building.
During the Late Chalcolithic 3 period (ca. 3800–3600
B.C.), Area TW was taken over by a ‘‘feasting’’
structure and ovens dedicated to processing large
quantities of meat. From ca. 3500 B.C., during Late
Chalcolithic 4–5, the site and region were colonized
by peoples from southern Mesopotamia, and Area
TW’s use shifted again, to domestic buildings and
trash pits (Oates 2002, 2005; Oates and Oates 1993).
Excavations on Brak’s central mound thus offer
evidence for an early thriving urban settlement and
institutions that may have held power over people
and the economy.
Outer Town survey and soundings
Soundings and an intensive survey of the site’s Outer
Town have further contributed to our knowledge of the
site’s extent and variable density of occupation during
the crucial 4th millennium B.C. A ‘‘corona’’ of small submounds
(1–4 ha each) encircling Brak’s central mound
has been a recognized feature of topographic plans from
Mallowan’s research onward (FIG. 2) (Mallowan 1947:
49, pl. LVI). The first organized investigation of Brak’s
Outer Town and these mounds took place in the late
1990s, when a program of strategic soundings was
carried out (Skuldboel 2009; Skuldboel in Emberling
et al. 1999). These soundings showed that the creation of
some sub-mounds belonged within the Late Chalcolithic
2 period (late 5th–early 4th millennium B.C.), but that
Figure 3 Central mound and northern Outer Town of Tell Brak, from Tell Majnuna; Trench EM-1 in foreground.
McMahon et al. Mass Graves, Conflict, and Early Urbanism, Syria
204 Journal of Field Archaeology 2011 VOL. 36 NO. 3
their intensive use was dateable to the Late Chalcolithic
3, with sparse 3rd millennium B.C. and later occupations.
The sounding program was supplemented by an
intensive survey and surface ceramic collection across
the Outer Town during 2004–2006 (Oates 2005; Ur et al.
in press). The Outer Town survey explored the site’s full
occupational history, but developments in the Late
Chalcolithic period proved to be the most significant.
The survey confirmed that the majority of the submounds,
and some of the non-mounded areas between
them, were first utilized in the Late Chalcolithic 2
period. This ring of sub-mounds then experienced
intensified use in the Late Chalcolithic 3 period, and
the ‘‘empty’’ space between them and the central mound
gradually filled in over the course of the several centuries
of this period (Ur et al. 2007, in press). The result of this
Outer Town survey was a new vision of the mechanics of
urban growth in northern Mesopotamia, comprising
development from outside in, through the merging of
small settlements with a larger center. In southern
Mesopotamia, Uruk may have developed from the
merging of two settlements originally separated by a
channel of the Euphrates (Nissen 2002), but the
customary reconstruction of southern Mesopotamian
urban growth comprises inexorable outward expansion
from a core centered on one or more religiousadministrative
institutions. Like Uruk, Brak may have
expanded through immigration as well as internal
growth and settlement merging. The evidence for this
is the emptying of a ring of land up to 3 km from Brak
during the Late Chalcolithic 3 period, while the area 3–
8 km away was only sparsely settled, with contemporary
sites clustered mostly on the opposite bank of the Wadi
Jaghjagh (Wright et al. 2007).
Since completion of the Outer Town survey, excavation
in one of the sub-mounds compels an amendment
to this merged-settlement urban growth model. The
most prominent of the sub-mounds at the site’s northern
edge, Tell Majnuna (FIG. 2), consists not of occupation
layers as expected but entirely of massive quantities of
rubbish, which was deliberately deposited over approximately
200 years and reached 2–3 ha in extent (the
precise area is unclear due to modern damage). This
discovery has important implications for definitions of
‘the city’ and for the calculation of site size and density
of occupation from surface surveys in any region. Traitlists
of ancient cities (e.g., Childe 1950; van de Mieroop
1997) have so far not included rubbish dumps, but the
generation of garbage would have increased in pace with
population expansion and nucleation. A new variable of
space assigned for rubbish deposition and a factor that
represents its growth with population increase should be
added to any calculations of population size from survey
or surface evidence. It must be emphasized that only
urbanism, with its implications for increased population
and industrial scales of production, consumption, and
discard, could have produced enough rubbish to create
Tell Majnuna. It is also urbanism and its corollary of
socio-political organization that embrace the number of
people needed to transport the rubbish to the site’s edge
and the leaders necessary to manage this activity. Even
though it does not directly reflect traditional domestic
occupation, Tell Majnuna (and possibly the other submounds)
retains implications for urban growth at Brak.
Taken together, the Brak excavations and site survey
allow us to build a picture of northern Mesopotamian
settlement expansion, increase in urban spatial structuring,
growth of institutions, and increased complexity in
economy and society some 300–500 years earlier than
our current evidence from southern Mesopotamia.
Tell Brak (Tell Majnuna) Mass Graves
Tell Majnuna lies approximately 450 m north of Brak’s
central mound and is one of seven visible sub-mounds
that form a discontinuous ring around the high mounded
center (FIG. 2). At 7 m high and 2–3 ha in extent, it is one
of the larger sub-mounds; most are 3–4 m high and 1 ha
or less. During his excavations at Brak in 1937, Max
Mallowan briefly excavated two trenches on Majnuna;
this work was abandoned when no architectural remains
were recovered (Mallowan 1947: 50, pl. LVI). No further
excavations were made on Tell Majnuna until the
autumn of 2006, when modern construction led to the
unexpected discovery of the first of a series of mass
graves and a brief exploratory sampling program.
Tell Majnuna became the primary focus of excavations
during 2007 and 2008 (McMahon and Oates
2007). Sixteen trenches were opened across the site’s
maximum diameter, and three more mass graves were
exposed in these, supplemented by several additional
smaller deposits of human bones (FIG. 4). The mass
graves had been made episodically during approximately
two centuries of the Late Chalcolithic 3 period
(3800–3600 B.C.), when survey indicates that Brak
underwent a rapid increase in area and which precedes
the well-known colonization of the region by peoples
from southern Mesopotamia. Samples have been
taken for 14
C dating but have not yet been analyzed;
however, the Late Chalcolithic ceramic sequence and
discrete assemblages from its sub-periods are well
known and dated from excavations at Areas TW and
HS at Brak, Tepe Gawra (north Iraq), and Hacinebi
Tepe (southeast Turkey). Here, we focus on the two
earliest and most complex mass graves, dated to ca.
3800 B.C., in Areas MTW and EM.
The mass grave in Area MTW
Area MTW lies at the southwestern edge of Majnuna
and partly within adjacent agricultural fields. A
modern construction trench associated with a grain
storage area, plus four of our own trenches, opened a
significant window onto the grave and its context.
The pre-grave deposits comprise a waterlogged clay
McMahon et al. Mass Graves, Conflict, and Early Urbanism, Syria
Journal of Field Archaeology 2011 VOL. 36 NO. 3 205
layer over reddish virgin soil, followed by thin
stratified rubbish layers and lenses containing sherds,
animal bones, and scattered disarticulated human
body elements. It appears that this area was
occasionally used for rubbish deposition during the
very early 4th millennium B.C. and had not seen any
previous occupation. However, around 3800 B.C.
this area was chosen for the deposition of a large
pile of human and animal remains. This mass grave
could not be fully excavated, but sections provided
by the modern trench together with our own excafovations
reveal that it was at least 20 m long by
3–4 m wide. The pile varied in density and thickness
over its length, from dense, deep, and jumbled
at the eastern end (FIG. 5), tailing out to single body
portions with short gaps between at the western
end (FIG. 6). The conservative minimum number of
human individuals recovered from our excavations
is 54, while an estimate of the total number in the
pile (based upon number of individuals recovered
in excavation and during random collection of remains
from the modern trench dumps, together with
reconstruction of the pile’s horizontal extent) is
several hundred.
The pile was placed directly on a single consolidated
surface, and it was essentially contiguous along
its width and length; the gaps between bone clusters
were insignificant. The layer of soil and rubbish that
completely sealed the bone pile was also a single
internally-consistent and easily-defined unit with no
later disturbances or additions. Thus, depositional
and stratigraphic evidence points to the grave’s
creation being a single event of short duration,
probably lasting a few days at most.
The skeletons were significantly disarticulated,
although articulated elements such as torso segments
(thoracic vertebrae with attached ribs) and full upper
or lower limbs were frequently found intact. The
bones were mingled, with no attempt to keep the
parts of any individual together or to keep any
individual separate from others. This degree of
disarticulation and mingling, together with the rarity
of smaller body elements such as feet and hands,
implies that this pile is a secondary context. There
Figure 4 Topographic plan of northern Outer Town of Tell Brak and Tell Majnuna with main excavation areas indicated.
McMahon et al. Mass Graves, Conflict, and Early Urbanism, Syria
206 Journal of Field Archaeology 2011 VOL. 36 NO. 3
was some sorting during deposition that may be
related to the mode of collection and transport from
the presumed primary context: the lowest layer of the
pile comprised mostly separate large elements such as
femora, covered by a layer of partly-articulated body
segments, and the top of the pile was made up mostly
of smaller single elements and skulls. The disarticulations
in all cases appear to be accidental, derived
from natural decomposition, as there are no cutmarks
or indicators of dismemberment on bones. The
loss of smaller bones may have occurred during
gathering and transport to the burial site or may be
related to animal scavenging before the bones were
gathered for transport. The damage includes small,
paired, parallel incisions from rodent teeth plus deep
punctures and furrows probably produced by dogs or
even lions, to judge from the scale and depth of marks.
This scavenging implies that the primary context
involved exposure in the open air. In part, scavenging
may explain the absence of smaller body elements,
such as feet and hands, which could easily be
transported away from the death site by such animals.
The presence of at least two almost-complete skeletons
within the pile (missing only heads and arms) implies
that little time passed between death and removal for
burial. Rates of human post-mortem disarticulation
are notoriously difficult to track with precision, as they
depend upon season, temperature, humidity, microclimate,
soil conditions, micro-organisms, clothing or
lack thereof, and other factors (Bass 1996; Forbes
2008; Galloway 1996; Ubelaker 1996). The pattern of
disarticulations in the MTW grave implies that weeks
to perhaps months elapsed between death and burial.
Notably, the disarticulations of the human remains
across the pile are internally consistent. Given that
their placement in the mass grave was a single
depositional event, the equal degree of disarticulation
of the skeletons implies that death was simultaneous
for all, as were their decay and their gathering and
removal. Other possible explanations for the mass
grave, such as relocation of a normative cemetery, are
belied by this internal consistency in degree of
disarticulation observed across all the bodies.
The human skeletal material from MTW was aged
through standard techniques of measurement and
assessment of tooth formation and wear, plus degrees
Figure 5 Mass grave in Area MTW, east end, with dense thick deposit of human and animal bones.
McMahon et al. Mass Graves, Conflict, and Early Urbanism, Syria
Journal of Field Archaeology 2011 VOL. 36 NO. 3 207
of epiphyseal and cranial bone fusion. Assigning ages
was complicated by the fact that we had to rely upon
single age indication factors for many bones, when best
practice in aging should summarize age markers
available across an entire skeleton; however, a significant
number of body elements could be aged with a reasonable
degree of certainty. A normal attritional cemetery
population derived from natural causes should show a
higher relative proportion of older adults and infants
than would have been present in a living population,
because of the greater risk of death in these age groups. A
catastrophic mortality profile, such as that from a rapid,
deadly disease, should show a close relationship to the
living population profile, but again with a higher
proportion of infants, children, and the elderly
(Margerison and Knu¨sel 2002). The Majnuna MTW
mass grave shows neither profile type (FIG. 7). The MTW
grave age range was dominated by adults over age 20 but
younger than 40 (the majority falling in the relatively
young 20–35 age class), thus comprising the portion of
the population who should normally be the least likely to
die from disease or most other life risks. This portion of
any population is the most likely to take part in violent
conflict, which is thus a plausible cause of death for the
individuals in the mass grave. Children between the ages
of 7 and 14 and adolescents 15–20 years old were also
present in significant numbers, while there are a very few
bones from children younger than 6. The presence of
children does not detract from the interpretation of a
warfare cause of death; the violent event must have taken
place within or near the settlement (see below,
Interpretation) with the inevitable involvement of many
members of the community.
The absence of infants is not surprising. Infant
burials in jars or small pits have been found
throughout the 4th millennium B.C. levels in Area
TW, in contexts such as abandoned ovens and near
trash pits. This separate mode of burial is perhaps
appropriate for those who have not yet achieved a
significant social identity, and discrete locations for
infant graves are also used in the preceding 5th
millennium B.C. The elderly may also have been
buried apart, in special situations reflecting their
accumulated identities. It is the simultaneous death of
so many young adults and sub-adults, more than the
low numbers of younger and older individuals,
however, which is the most compelling evidence that
those in the MTW grave died through violence.
The MTW remains were sexed by methods appropriate
to the varied and often unconnected body
Figure 6 Mass grave in Area MTW, central, with dense ceramic sherds and less dense human and animal bone deposit.
McMahon et al. Mass Graves, Conflict, and Early Urbanism, Syria
208 Journal of Field Archaeology 2011 VOL. 36 NO. 3
elements present. The most reliable indicators of
human sex are features of the pelvic bones, but
unfortunately, the recovery rate of complete pelvic
bones in MTW was low. Pelvic bones have relatively
thin cortical layers compared to long bones, meaning
they frequently do not preserve well in archaeological
contexts. In addition, the pelvis is in a high meat-yield
zone of the human body and may have been targeted,
broken apart, and partly removed by scavengers; the
transport from primary to secondary context may
have also introduced damage to this rather fragile
element. Thus we relied, in addition, on the identification
of cranial and mandibular features and robustness,
specific measurements of long bones (e.g.,
epiphyseal measurements of humerus and femur),
and assessment of the general robustness of postcranial
bones. The latter aspect was assessed primarily
through a series of measurements of long bone shaft
diameters or circumferences and epicondylar breadth,
compared to a growing sample of human remains
from sites across northern Mesopotamia dating from
Late Chalcolithic through Early Islamic periods
(Sołtysiak 2010). The MTW mass grave population
includes both sexes, but the ratio is skewed towards
males among those elements that could be sexed with
reliability (ca. 65% male to 35% female). As with the
presence of children, the presence of women in the
mass grave in no way diminishes the possibility of
warfare being the cause of death. The violent event, as
above, must have taken place within or near the
settlement and involved individuals outside the traditional
‘‘standing army.’’
Injuries, weapons, and pathologies
Death from violent conflict should be revealed in visible
injuries, such as parry fractures to arms, crushing or
hacking injuries to skulls, gashes from swords or spears,
and embedded projectile points (Dawson et al. 2003;
Jordana et al. 2009; Liston and Baker 1996; Ortner
2003; Walker 1989, 2001). Although many of the MTW
bones are damaged, it was difficult to distinguish
possible peri-mortem trauma from damage resulting
from processes of decomposition, scavenging, transport,
and secondary deposition. While the MTW
skeletons lack explicit evidence for peri-mortem injuries,
a significant percentage of the remains shows
healed cranial injuries, evidence of previous violent
events. The healed injuries provide some—albeit
limited—support for the theory that violent conflict
may have also been involved in the individuals’ eventual
deaths. Ten cranial elements from Area MTW show
healed injuries, a relatively high percentage of the total
cranial bones recovered. Skulls of both men and women
showed healed injuries of various forms, from small
pointed dents through larger oval depressions up to
2 cm in diameter to longer grooves ca. 5 cm long
(FIG. 8). Other evidence of healed trauma, including
fractures, dislocations, and periostitis, were rare and
may have occurred from accidental injury or violence.
Cranial injuries are a good indicator of interpersonal
violence in societies such as Late Chalcolithic
Mesopotamia, where the primary weapon was probably
the club or mace. Such weapons are effective in
hand-to-hand combat when used to provide a transfixing
or killing blow to the head. Maces are rarely
recovered in excavations (although cf. Emberling and
McDonald 2003: 9, fig. 15 for slightly later examples of
mace-heads from Area TW), but polished stone axes,
which have been recovered from Late Chalcolithic 2
and 3 levels in Area TW, are possible weapons. They
Figure 7 Mortality profile of skeletons from Areas MTW and EM, compared to profiles of a normal attritional cemetery
population and catastrophic event in Model West (Coale et al. 1983).
Figure 8 Healed depressed cranial fracture from Area MTW.
McMahon et al. Mass Graves, Conflict, and Early Urbanism, Syria
Journal of Field Archaeology 2011 VOL. 36 NO. 3 209
seem small and lightweight to have inflicted serious
damage (average dimensions are 563.561.5 cm), but
their pointed ends are of an appropriate size to have
made some of the healed depressions in the MTW
skulls. These axes are made from beautiful dark stones,
such as black marble, and are highly polished, aspects
that might imply that they were symbolic rather than
functional. However, weapons used in creation of a
visible ‘‘warrior’’ identity can incorporate both functional
and sensual qualities. Copper weapons (spears
and swords) were known in the region at this time, but
would have been expensive and rare; bows and arrows
were possibly more common but their wounds leave few
archaeologically-visible traces. There is an additional
possibility that slings were common weapons in this
period. Recent excavations at Tell Hamoukar have
revealed a slightly later destruction layer with hundreds
of clay sling bullets (Lawler 2006). Sling bullet injuries
are likely to have been disabling rather than fatal, and
they are unlikely to be represented in skeletal damage.
The MTW human bones were also studied for
pathologies, general health, and diet. Tooth enamel
hypoplasia was notable; in humans this is evidence of
malnutrition or stress from disease during enamel
formation, mainly between the ages of two and seven
years. Correlation of enamel hypoplasia with degree
of tooth wear indicative of age was carried out for the
MTW remains, mainly on teeth from the betterrepresented
maxilla. Although the sample size was
small, the results indicate the possibility that the
younger individuals had gone through an episode of
nutritional stress lasting for several years shortly
before their deaths. Evidence from older adults
indicated that an earlier episode of stress had
occurred, perhaps a decade or more before the deaths
(Sołtysiak 2010).
Animal Bones in Area MTW
Animal bones are mixed with and overlie the upper
layer of human bones in the MTW mass grave pile.
These bones primarily consisted of disarticulated but
whole skeletons of sheep, goat, and cattle in large
numbers. The remains of at least 25 cattle and 100
sheep and goat skeletons were recovered from the
area excavated. We estimate that we have exposed
only approximately one-third of the complete mass
grave, and thus it is possible that up to 375 animals
were butchered and buried (75 cattle and 300 sheep
and goats). The density of animal remains decreases
towards the western end of the grave, as does the
density of human remains, so these numbers may be
lower. The animal carcasses were buried amid and on
top of the human skeletal remains in the same single
disposal event, sealed by a contiguous and undisturbed
rubbish layer. Some elements remained in
articulation, particularly among the cattle; meat
joints were fairly common and a whole side of beef
was discarded intact. Despite that fact that these
bones were deposited fresh, they exhibit no evidence
for disturbance by carnivores or weathering, indicating
that they were quickly covered.
The animals were slaughtered within a short span
of time, probably as a single event. The age range of
the animals indicates that death was neither catastrophic
nor attritional; animals of all ages were
killed, but certain ages were selected with greater
frequency than others. Among the cattle, younger
animals are three times more common than adults
(older than 4 years). Infants (under 1 year), juveniles
(1–3 years), and sub-adults (3–4 years) occur in equal
numbers. The presence of so many younger cattle is
unusual (at least 17 were recovered, but the total
number may possibly be as many as 51, if the young
to adult ratio was maintained and the density of
animal bones persisted across the unexcavated portion
of the grave). Cattle’s slow reproductive rates
make for slow herd renewal rates, and thus young
animals are rarely slaughtered. Culling of so many
young animals at one time indicates either the
presence of large cattle herds at Brak, or off-take
from many different herds from Brak and its
environs. The adult cattle thus far recovered were
prime-age individuals, which is also unusual as adults
represent large investments of labor, land, and water,
with the potential return of dairy and draft for many
years. That investment balance reduces the frequency
with which healthy, adult animals are slaughtered.
Slaughter of greater numbers of males or females
might suggest greater expendability, respectively, for
draft and sire duties or dairy animals. So far,
however, males and females are found in similar
numbers; but further analysis is required to determine
whether there are sex differences between age groups,
which could offer insight into the use-value of the
living animals. Overall, the slaughter of so many
cattle would have been an expensive endeavor unless
distributed over several herds, settlements and/or
people.
The sheep and goats were fairly evenly divided
between males and females. Nearly half were killed as
adults (older than 42 months), and the majority of
these were slaughtered between 4–6 years of age,
when healthy animals are still highly productive. Few
animals of older age were recovered. Another spike
appeared in the 2–3 year old range, after individuals
have already achieved their maximum meat weight
and have many more years of wool, hair, and dairy
production. A few animals were killed prior to
6 months of age, but in general the youngest
recovered were killed in an early stage between 6–
12 months; no newborns were identified, but their
fragile bones do not survive well. However, since the
McMahon et al. Mass Graves, Conflict, and Early Urbanism, Syria
210 Journal of Field Archaeology 2011 VOL. 36 NO. 3
animals were all killed at the same time, the ages
should be found in classes of 1-year interval. There
are a number of animals killed just before and just
after 6 months of age, suggesting that the lack of
newborns is due to the time of year when the
slaughter event occurred. Around Tell Brak, there is
currently a single birthing season that occurs over 2–
3 months in the spring. If this were the same in the
past, given that the youngest animals were just less
than 6 months of age, then we expect the event to
have occurred in the fall.
The animals were all slaughtered for the same
purpose, as indicated by consistent butchery patterns.
For the cattle, the toes (phalanges) and ankle bones
(tarsals) were repeatedly struck with a cleaver-like
implement. This would have been used to hack
through the hoof, sever the strong tendons of the feet,
aid in the disarticulation of the extremities, and
probably also remove the metapodial bones. There is
evidence for further disarticulation in the form of
cleaving marks around the tight elbow joint (humerus,
radius, and ulna), and around the hip (pelvis and
femur). The metapodial bones were split lengthwise
and the shafts of the remaining limb bones were
smashed. This may have occurred prior to cooking in
order to release more juices, a practice common in the
Late Chalcolithic 3 levels in Area TW (Charles et al.
2001; J. Weber in Emberling and McDonald 2003). It
could have happened after cooking in order to access
the marrow. The latter is indicated for many of the
tibia, which bear distinct traces of polishing along the
inner edge of a spiral-pattern break. We associate this
with the repeated scooping out of marrow with a finger
or bone splinter, creating a polished edge where the
grease rubs past the bone. In contrast to full
exploitation of the limbs, the rib cage does not seem
to have been broken-down further despite the presence
of good meat there. There is no sign of meat removal
from any of the bones, indicating it was probably
cooked on the bone. It is also possible that meat
removal was done without leaving marks, but the
heavy cleaving on other bones suggests that such care
was unlikely. Signs of heavy, though unsystematic,
burning were found on numerous elements; we cannot
say whether the burning resulted from the cooking of
meat or from the burning of rubbish. Numerous burnt
horncore fragments were recovered, but there is no
indication if or how the neck and head were prepared.
Sheep and goat were butchered similarly to cattle,
but with smaller, finer tools. Marks of butchering are
found on the extremities and at most limb joints.
These were generally the result of disarticulation and
skinning rather than meat filleting. Again, we suggest
that meat was cooked on the bone. Bone smashing
was quite common, and the edge polish that we
interpret as resulting from marrow removal occurs on
the deliberately broken shaft of all the limb elements
but most frequently on the tibia. Unlike cattle ribs,
those of sheep and goat were broken down, probably
for meat removal. At least one animal had its head
removed from its neck—though this was probably
systematic. All of the skulls were smashed, possibly
for access to the brain, and the horn and horncores
were removed. Burning on the bones was common,
but as with the cattle bones, we cannot determine
from which activities it resulted.
The slaughter and preparation of these animals
occurred as part of a single event, which we interpret
as a feast. The human and animal bones are also
mixed with Late Chalcolithic 3 ceramics, dominated
by large shallow plates and fine-ware bowls such as
might have been used in communal feasting. Many of
these dishes were scarcely worn and had been simply
broken in half, suggesting symbolic destruction after
their use. Butchery was focused on disarticulation of
meaty limb-bones and was consistent throughout the
entire assemblage. Yet individual variation in the
depth, length, or style of cut marks suggests that
the consistency was not the result of a specialist, but
was rather a communal effort of several individuals
following the same general pattern for the desired
result. The breakage of most long-bone shafts for
marrow and the smashing of the skull to access the
brain together further suggest that these animals were
slaughtered on this occasion for mass consumption.
It might also indicate the serving of a dish similar to
the modern mensef, a feast dish in which meat bones
are placed on trays of rice, topped by a split skull
with brain intact but accessible. That this was a major
event is indicated by the size and cost of the meal.
Together, 25 cattle and 100 sheep and goat equate to
roughly 6200 kg of meat. If there were, in fact, 75
cattle and 300 sheep and goats, as estimated, then
there would have been approximately 21,300 kg of
meat. Even if consumption occurred over a three-day
period, for instance, 7100 people could have had 1 kg
of meat each day. The selection of animals seems to
have been done with little regard for cost. Highly
productive and prime-age animals were slaughtered
for this feast; this wastefulness means it is possible
that some of the meat was discarded uneaten.
Mass disposal of butchered carcasses of sheep,
goat, and cattle has been observed in contemporaneous
contexts at Tell Brak in Area TW, though
never in such large quantities. These remains, too, are
identified as the discard from feasting activity by the
signature features of whole carcass disposal of a large
number of sheep, goat, and cattle in a single episode,
consistent butchery, bone-smashing, and meat roasting
(Charles et al. 2001; J. Weber in press). The MTW
feast, so closely associated with human bodies that
may have died in violence, could have been a ritual
McMahon et al. Mass Graves, Conflict, and Early Urbanism, Syria
Journal of Field Archaeology 2011 VOL. 36 NO. 3 211
that either celebrated a victory or commemorated the
dead. As the final deposit of the mass grave, it acts as
a ritual closure event, and its scale implies that the
creation of the mass graves was a public event.
The mass grave in Area EM
Area EM lies at the southwestern edge of the
Majnuna mound, some 10 m northeast of MTW.
Two trenches revealed the full extent of a second pile
of human remains, created slightly later than that in
MTW, according to stratigraphy. In Area EM,
however, there is a rather different situation. Again,
natural soil was reached at the bottom of the trench;
above it there was no material dating to earlier than
Late Chalcolithic 3 (LC 3), confirming our impression
of the light use of this area before the graves. The
layer resting immediately on natural soil was dense in
LC 3 sherds and may be connected to the rubbish and
sherd layer covering the MTW mass grave; this was
followed by a series of very ashy layers, within which
was the second mass grave. Stratigraphically, the EM
grave is later than the MTW grave, but we assume
that the intervening layers were deposited rapidly and
do not represent more than a decade. The ceramic
and clay container sealing assemblages associated
with both graves were very similar.
Like the MTW grave, the body elements in EM
were heaped up in a long narrow pile, but here a
smaller 6 m long by 1 m wide. Most importantly, the
disarticulation of the human remains was more
extensive than that in MTW, with only parts of
spinal columns remaining more or less intact. The
body elements represented had also been specifically
selected, comprising chiefly femora, fibulae, tibiae
and humeri, smaller numbers of skulls and spine
segments, and a few ribs (FIG. 9). Bones of the hands
and feet were almost absent, as were pelvic bones.
Near the top of the pile in particular, there were
visible ‘‘armloads’’ of long bones, lying parallel to
each other. The pile rested on an easily traceable
surface and was contiguous; it was created during a
single short-term event, lasting at most a few days
and was sealed by a continuous, undisturbed layer of
ashy soil. There was significant pre-transport carnivore
damage to the ends of long bones, in the same
pattern and to the same extent as in Area MTW;
again, there was no evidence for dismemberment.
Animal bones are almost absent and the few that
were associated with the human remains are comparable
to the persistent low-level presence of bones in
most occupational and discard contexts at Brak and
other sites in the region. There is thus no feast
associated with this mass grave.
The minimum number of individuals represented in
the EM mass grave is 89. The age profile was assessed
with the same methodology used for the MTW grave
and is distinctively different from the latter, showing a
higher proportion of children, including five children
aged between 1 and 3 years old (FIG. 7). Almost all
categories of ages except infants are represented:
young children (1–6), older children (7–14), adolescents
(15–20), full adults (20–30), and a few older
adults over 40 years of age. This age range does not
match an attritional or catastrophic age profile. The
adults were relatively young, slightly younger on
average than those in MTW (most are 20–25 years
old), while the elderly as well as newborn children are
Figure 9 Detail of internal layer within mass grave in Area EM.
McMahon et al. Mass Graves, Conflict, and Early Urbanism, Syria
212 Journal of Field Archaeology 2011 VOL. 36 NO. 3
missing and very young children remain statistically
under-represented. Again, the absence of infants and
older adults might be explained by separate burial
locations; but the sheer number of young adults and
the selection of particular body elements argue for an
unusual source for this grave. The same methods were
used to identify sex in the EM grave as in the MTW
grave; here, there was a stronger skew in the ratio
towards females (ca. 68% female to 32% male).
The greater degree of disarticulation in the EM
mass grave might be interpreted as evidence that it
was a cemetery clean-up or other collection of bones
from individuals who had died naturally over some
time. However, the limited range of body elements
present and the age profile dominated by young
adults argue against these interpretations. The limited
range of specific elements also implies a different
activity set than that represented by the death, decay,
transport, and disposal trajectory reconstructed for
the MTW mass grave. The selection of body parts
indicates that the EM pile was the result of targeted
processing of other bodies. A possible source is the
MTW mass grave itself, which could have been reexposed
through pits or trenching in areas we did not
excavate; however, the different age profiles make
this reconstruction unlikely. The evidence thus points
to a second violent event as the possible cause of
deaths, but the EM remains may represent mainly a
group of traditional non-combatants. It should be
noted that only one of the skulls in EM shows
possible evidence of a healed earlier injury.
The frequency of linear enamel hypoplasia in the
teeth from EM was high. This feature was also
present in the mass grave in MTW but was
significantly more common in EM, particularly in
the teeth of the younger individuals. Cranial porosities
(including cribra orbitalia) were also more
common in EM than in other contexts and may
indicate that these individuals in particular suffered
anemia, vitamin deficiencies or generalized systemic
and dietary stress.
Human bone tools from Area EM
Interpretation of the EM mass grave is further
complicated by the fact that some of the femora
and tibiae were made into tools, and there is evidence
of polish from handling on other bones and on skull
fragments. These human bone tools are unique in
Mesopotamia. Forty-two human bone tools were
recovered, of which 19 were mostly complete and 23
were fragmentary. The tools were initially shaped by
twisting and snapping the bone roughly in half,
creating angular pointed ends on each half. This
twisting snap of the bones seems to have been
standardized in both technique and strength applied:
the angle of the point preserved in unworked or
lightly worked examples is very similar (FIG. 10A–D),
suggesting a single person or small number of
individuals using a repeated or habitual body
practice.
The joint ends of each half were then removed, and
the shafts were evened by chipping akin to that used in
lithic production (visible on FIG. 10A, 10C, 10F). The
reaction of the bones to this working, and perpendicular
cracking from past changes in humidity, indicate that
the bones were defleshed and dried from the elapse of
probably several years between the death of the
individuals and creation of the tools. After snapping
and rough chipping of the ‘‘base’’ end, the pointed end
was not altered further before the tool was used. There
is a range of degrees of wear visible on the tool points,
from fresh and unused through worn examples, with a
spatula-like form, high polish and a slightly facetted end
(FIG. 10E, 10F). The tools come in two almost standard
sizes, averaging 12–13 and 18–21 cm long; in most cases
the shaft diameter was left complete (FIG. 10A–D), but
some of the larger femora had been split along their
length before use (FIG. 10E, 10F). Both sizes fit easily in
an adult hand. The tools were found throughout the
EM pile, including in its lower layers. The pile,
therefore, represents not just the place of manufacture
of the tools but also the place of their use and post-use
discard, presumably together with the materials on
which the tools were used. Our reconstruction of the
EM mass grave creation describes a several-stage
process, although it must be admitted that evidence
for some of the steps is indirect. It begins with death of
the individuals in the same or a similar conflict as that of
MTW, initial decomposition in situ, transport and
creation of a mass grave similar to MTW, elapse of
some years, re-exposure of the bones and selection of
specific elements, and finally tool creation, use, and
deposition in Area EM.
In addition to the long bone tools, 17 examples of
cranial fragments from within the EM pile bear
scratches and polish. There is a possibility that some
of these modifications may have occurred during
transport or post-depositional processes; however, at
least four fragments bear heavy polish and/or deliberate
directional abrasions, suggesting that skulls with less
regular markings had simply been less intensively
handled. The polish and scratches are concentrated on
the frontal and occipital bones and probably derived
from a combination of handling and resting on the
ground or a slightly abrasive surface. It is difficult to
determine the aim of this skull processing from the
fragments remaining, but the co-incidence of the human
bone tools provides a suggestion. The wear implies that
the tools had been used for some manufacturing
activity, but the objects created with these tools are
actually unlikely to be represented in the EM discard
pile, since they would have been created and then
McMahon et al. Mass Graves, Conflict, and Early Urbanism, Syria
Journal of Field Archaeology 2011 VOL. 36 NO. 3 213
transported away for use. We propose that the tools
were used to create or process trophy skulls. The
abraded and polished skull fragments left behind could
be the mistakes, skulls on which processing work had
begun but which shattered during the task and were
discarded.
Post-Burial Practices and Additional Mass
Graves
The construction of the overlying mound of Tell
Majnuna is the final unusual aspect to the mass
graves. Our 16 trenches dissected the mound at its
maximum diameter, and modern bulldozer trenches
and two sets of unused house foundations also
provided information on its internal stratigraphy.
There were unrelated and minimal traces of architecture
of the late 1st millennium B.C. near the
surface, but there is no evidence of construction nor
of layered and gradually-accumulating occupational
deposits of the Late Chalcolithic period in any trench.
Instead, the entire 2–3 ha and more than 7 m depth
of the mound were created by large-scale rubbish
dumping events. A third mass grave, dated to
approximately 3700 B.C. by associated ceramics and
stratigraphy, was located near the mound center
(Area EME 3; FIG. 4). It was discovered in a deep
sounding and could only be partly excavated. There
were at least 14 individuals, in a similar depositional
pattern and with a comparable degree of semiarticulation
to the grave in MTW, but its full extent
and scale could not be determined. A bag-shaped pile
of disarticulated bones, perhaps from a clean-up
operation, was found not far from Area EM (EMS
on FIG. 4). And there was a final mass grave of ca.
3600 B.C., when the eastern edge of the mound was
used for the simultaneous and careless burial of at
least 35 individual bodies (Area EME 1–2; FIG. 4).
Ceramic assemblages from trenches across the
mound reveal the temporal elapse of approximately
two centuries, during which the dumping happened
episodically. Within each dumping episode, the
height of the mound increased, in some cases by
more than a meter, and the mound grew horizontally;
clear tip lines across the mound slope down from
southwest to northeast, indicating the direction of
dumping. At the southwest, immediately above the
first two mass graves, the primary material was ashy
loose soil. This material persisted across the center of
the mound, but the deposits became less ashy from
the mound center and further northeast. Especially at
the north and northeast edges (Area MTE), in the
final depositional events, a sherd-rich source had
been mined and there was a greater density of sherds
than soil.
The full range of typical Late Chalcolithic 3
ceramic forms is represented. Large plates and
cooking casseroles that dominate these rubbish
deposits are common in domestic contexts elsewhere
at the site, but the sherds in the Majnuna deposits
tended to be larger. This implies that the sherds were
Figure 10 Human bone tools from Area EM.
McMahon et al. Mass Graves, Conflict, and Early Urbanism, Syria
214 Journal of Field Archaeology 2011 VOL. 36 NO. 3
gathered soon after breakage and before trampling
could break them down further. The quantities of
sherds are also far beyond what one would expect of
normal household use and breakage. It is probable
that the sherds were collected from communal
neighborhood dumps and perhaps industrial or
administrative areas where large numbers of vessels
would have been used and frequently broken and
discarded.
Interpretation: Mass Graves, Warfare, and
Corpse Abuse
Burial practices in Mesopotamia in almost every
millennium from the 6th through the 1st consist
mainly of near-immediate burial and single, articulated
interments in purpose-made pits. Secondary
burials, cremations, and burials of two or three
individuals who died simultaneously are known but
uncommon; communal tombs used for individual
burials sequentially over several generations are
present, particularly in the 2nd millennium B.C., but
are a minority. However, the burial practices of the
4th millennium B.C. are among the least known for
the region. Very few adult burials of this era have
been recovered in either northern or southern
Mesopotamia, although single, primary infant burials
are found more frequently, for example in vessels and
small pits in the industrial zone of Area TW at Brak.
However, the similarity of single, primary adult
burials in both the preceding and following millennia
offers the possibility of general continuity of practice
across the 4th millennium B.C.
Near Eastern comparanda for the MTW and EM
mass graves are rare. The few other mass graves known
differ in key characteristics from the Brak examples. A
mixture of disarticulated animal and human bones was
recovered from a pit at Domuztepe, southeast Turkey,
dated to the much earlier Halaf period, ca. 5500 B.C.
(Kansa and Campbell 2004; Kansa et al. 2009). The
Domuztepe assemblage of approximately 40 individuals
represents a wider range of ages, and interpretation
has varied from communal re-burial of individuals
who died peacefully across some years to a mass grave
resulting from organized violence. However, the
association of human and animal bones there provides
a useful parallel for the linked feasting and burial
reconstructed for MTW at Brak, and the evidence of
possible cannibalism (Kansa et al. 2009) is potentially
related to the symbolic tool-making in Brak Area EM.
A group of 19 mostly young male individuals was
recovered at Titrish Hu¨yu¨k, southeast Turkey, in a late
Early Bronze Age context (Algaze et al. 2001: 69–70).
The dead were disarticulated and arranged within a
plaster feature in a house, with skulls surrounding a pile
of long bones; trauma to the skulls in particular
supports an interpretation of massacre (Erdal 2010).
The disarticulation, lack of smaller elements, and
special arrangement provide useful parallels for the
Majnuna graves. Finally, evidence of mass death
associated with the siege and destruction of the NeoAssyrian
Empire in 612 B.C. comes from the gates at
Nineveh (northern Iraq), particularly the Adad and
Halzi gates (Stronach and Lumsden 1992), while the
Urartian sack of Hasanlu (Iran) in 800 B.C. also resulted
in a massacre (Muscarella 1989). But the skeletons in
these latter cases were left where they died (although
many of the Hasanlu bodies were looted and mutilated),
rather than being gathered and moved to a mass
grave.
The non-attritional and non-catastrophic age curves
dominated by young and essentially healthy adults,
together with the simultaneity of deaths in each group,
point to the MTW and EM mass graves being the
outcome of warfare. Warfare here is interpreted
broadly, to involve both deaths of combatants in battle
and the massacre of general populations embroiled in
conflict. Both MTW and EM graves contain remains of
individuals not traditionally thought of as active
combatants: women and children. The temptation is
to assume that these were non-combatants who were
accidentally killed within a larger, uncontrolled conflict
without a well-defined battlefield. The assumption that
women and children were not involved in battle is
supported by Mesopotamian textual and artistic
evidence from the 3rd millennium B.C. onward.
However, these later sources describe and depict formal
battles of territorial expansion. Civil wars and conflicts
between unequal forces are especially liable to subvert
expected sex and age categories among combatants;
guerrilla movements in Colombia and internal conflicts
in Sri Lanka and Northern Ireland have involved
female combatants (Alison 2009), and the Sierra Leone
and Congolese civil wars have used child soldiers. Thus
it is possible that the sex and age mixtures seen in MTW
and EM might reflect mixed combatants killed in action
as well as massacred bystanders.
The unusual delay and exposure after death, the
haphazard collection of remains that ignored smaller
bones of hands and feet (although these may also
have been affected by carnivore activity), and the
mingling of bodies in piles show an extraordinary
disregard for the retention of the individual body as a
discrete entity, at odds with customary burial
practices in the ancient Near East. The unconventional
disposal within layers of rubbish further
supports an interpretation that the dead were
considered in similar light to animal carcasses.
Their manner of disposal is analogous to that of
kitchen refuse, disarticulated and separated from
living spaces. This disregard implies that the dead
were enemies, conceptually different to those performing
the burials and considered unworthy of
McMahon et al. Mass Graves, Conflict, and Early Urbanism, Syria
Journal of Field Archaeology 2011 VOL. 36 NO. 3 215
normal burial and the preservation of their individual
identities. This proposal is supported by the creation
of tools and trophies, effectively corpse abuse,
represented in the EM mass grave.
Human bone is structurally less strong than animal
bone and is not well suited for creation of tools. Its
utility decreases as it ages and collagen is lost, further
weakening an already fragile material. Human bone is
unlikely to have been a preferred material for functional
tools, especially given the great availability of more
robust animal bones and the presence of tools made
from animal bones in other Late Chalcolithic contexts
(e.g., Area TW). The use of human bones should thus
be viewed as a primarily symbolic act. The artifacts
created remain elusive, but the polished and abraded
skull fragments in association with the tools suggest
that they were used to deflesh and empty trophy skulls.
Similar human bone tools have been recovered from
Pueblo period sites in the U.S. Southwest (A.D. 900–
1300), with admittedly disputed evidence of cannibalism
(Billman et al. 2000; Flinn et al. 1976; Kantner
1999; LeBlanc 1999; Marlar et al. 2000; Potter and
Chuipka 2010; Turner and Turner 1999). While the EM
bones bear some indicators of cannibalism (particularly
the mixing of individuals and selection of elements), the
length of time between death and manipulation makes
this practice unlikely. Key aspects of cannibalistic
activity, such as burning and butchery marks, are also
missing from our graves. The EM tools are related to
cannibalism, in that the human body was viewed as a
resource for exploitation and was subjected to postdeath
processing and modification. Exploitation of
human bones for creation of tools and trophies,
although stopping short of actual consumption, is
within the same ideational category as symbolic
cannibalism. Elsewhere, skulls and other parts of dead
ancestors and especially of dead enemies have been
made into display items or simply gathered as trophies
(see Chacon and Dye 2007 for a detailed summary of
trophy-making in the Old and New Worlds and
examples from United States and South America, plus
more recently e.g., prehistoric California: Andrushko et
al. 2005, 2010; and Nasca and Inka examples; Knudson
et al. 2009). This practice occurs most commonly at
times of social or environmental change or stress and is
associated with deaths by violence (indicated by mass
graves and peri-mortem injuries); the purpose may be to
provide symbols of status, to advertise group victory, to
celebrate personal military success, and/or to further
humiliate defeated enemies. The use of separated crania
as symbolic artifacts is well-known from the Neolithic
of the Levant (plastered skulls at Jericho) and Anatolia
(skull caches at C¸ ayo¨nu¨). These instances are held to be
elements of ancestor veneration, but they could equally
be evidence of trophy-taking in conflict. Collection and
display of trophy heads and other enemy body parts
appear in Neo-Assyrian historical annals and palace
reliefs illustrating the aftermath of battles, but so far the
practice has not been identified archaeologically.
Finally, the creation of the rubbish mound over and
around the graves was a deliberate act, either commemorative
or celebratory. The depth of each deposit
and internal steep tip-lines indicate that the mound was
designed to be visible; the rubbish was piled high rather
than being spread and diffused. Later Mesopotamian
textual traditions (3rd and 1st millennia B.C.) speak of
the construction of burial mounds above the slain
of battles, and the Tell Majnuna mound may indeed
be at an early step towards this later tradition.
This construction is an unusual creative, rather than
destructive, act associated with warfare. There is a
haunting modern analogy to be found in the Fresh Kills
landfill site (Staten Island, New York). Originally a salt
marsh, followed by clay-pits, fringe settlements, and
famously a landfill for city rubbish beginning in 1948,
Fresh Kills became the repository for much of the ruins
of the World Trade Center in 2001–2002 and thus has
become most recently a memorial and gravesite. The
current development of the site into a ‘‘Lifescape’’ park
acknowledges the site’s multiple layers and roles,
embracing the (now covered) trash mounds and their
varied and evocative contents. In a similar way, the
Majnuna mound was an actively-constructed and everchanging
feature of the settlement edge for at least five
generations or more. Its status as commemorative
monument over the initial burials would have waxed
and waned as time passed and rubbish or additional
graves were added.
The identity of the dead and location of the violent
event are unknown. The pre-burial exposure of the
bodies makes it unlikely that the battle took place
within the core of the settlement itself, or a more
rapid clean-up would be expected, but mass graves in
all wars from prehistoric to modern are never far
from battlefields, for logistic and commemorative
reasons. We place the location of the battle somewhere
in the northern Outer Town between Majnuna
and the city’s core or in the fields beyond. The nature
of the dead is more uncertain; their treatment implies
they were conceptually enemies, either external or
internal. It is difficult to imagine an outside force
attacking a settlement as physically imposing, politically
organized, and highly populated as Brak;
nomadic populations would not have had a large
enough fighting force, and adjacent settlements are
dwarfed by Brak’s scale. In addition, Brak may have
had linked settlements in its hinterland, the inhabitants
of which would have given the alarm and/or
attacked the rear of any intrusive force. The best
probability is an internal conflict among the inhabitants
of the city and its dependent hinterland.
Trophy-taking, such as is suggested above for the
McMahon et al. Mass Graves, Conflict, and Early Urbanism, Syria
216 Journal of Field Archaeology 2011 VOL. 36 NO. 3
EM evidence, is most common in external wars
(Chacon and Dye 2007). Creation of such trophies,
however, together with the disregard for corpses,
would have been a useful control device in any
internal conflict derived from power or status
asymmetry (Potter and Chuipka 2010).
Conclusion: Urbanism, Social and
Environmental Stress, and Violent Conflict
Past warfare and interpersonal violence have seen
innumerable studies in the last two decades, spurred
on by thought-provoking syntheses and intractable
Hobbes-versus-Rousseau philosophical approaches of
modern scholars (e.g., Haas 1999; Keeley 1996; Kelly
2000). There is clear evidence for interpersonal
violence by the Palaeolithic (Walker 2001), but we
must nevertheless search for a particular motive or
motives in each case. At Brak, there is a strong
temporal correlation between the mass graves and the
expansion of the settlement and increase in its internal
occupation density, social pluralism and economic
complexity. In Area TW, starting around 4200 B.C., we
see the increasing scale of buildings and elaboration of
industrial production, including decorative or ceremonial
items that could be material expressions of
asymmetries in social identity, status, and power.
Political, ideological, and religious controls are evidenced
by elite spaces such as the public building in
Area TW and the Eye Temple and by an expanding
artistic vocabulary presenting leaders and their close
proxies, lions, on container sealings (McMahon 2009).
Property rights to land and material things become
complicated and exclusionary with the growth of state
structures (Earle 2000) and denial of ownership can
increase the attributed value of things and individuals’
desire for them. Both the increasing complexity of
heterarchy and expanding distance within wealth and
power hierarchies may have provoked intra-community
conflict that broke out into violence.
We must address indications that the dead in MTW
and especially EM had experienced at least two episodes
of malnutrition and stress. There is evidence that a largescale
rapid climate change event, involving cooling and
increased aridity, may have occurred around 4000–3200
B.C. (6.0–5.2 kya) and might have reduced the carrying
capacity of the landscape around Brak just at the
moment of its urban expansion (Bar-Matthews et al.
1997, 2003; Chew 2007; Elmoslimany 1990; Mayewski
et al. 2004; Weninger et al. 2009). The possible actual
and relative food shortage resulting from this event
could have had an impact upon the full population of
the city and region and held the potential to incite
conflict over access to crucial dwindling resources.
Alternatively, the visible expansion of economic and
administrative hierarchies in the mid-Late Chalcolithic
may have created excessive demands on the agricultural
system, resulting in famine conditions, again for almost
the whole urban population. Or, dietary stress could
have been non-systemic and artificial, falling on
individuals of a specific class or profession through
oppression by an administrative authority. Growing
heterarchy, diverging hierarchy, resource demands, and
the precise date and reach of climatic change are thus all
interwoven with the Brak conflicts.
Our reconstruction of the mass graves involves violent
conflict among inhabitants of Tell Brak, a war related to
internal social developments, rather than an external
war of expansion. We recognize that this reconstruction
of civil war is speculative. It is plausible that the graves
may contain foreign enemies from a local battle,
although the unique size of Brak makes an external
attack unlikely. Slaughter of captured prisoners from a
distant battle is possible but less likely. That scenario
would involve bringing the prisoners to the site, an extra
step for which there is no evidence, followed by
slaughter, exposure, and eventual mass burial. Such a
Foucauldian spectacle would display power, but its
performance would not seem necessary within Brak
itself in a scenario of external battle. Isotopic analyses,
which could indicate diet and thus ‘foreign-ness,’ have
thus far proven inconclusive; aDNA analyses comparing
these individuals to contemporary local infant
skeletons in Area TW, a method successfully used to
compare populations in the Titrish Hu¨yu¨k Early Bronze
Age cist and mass graves (Matney et al. 2010), have not
yet been attempted. The temporal co-occurrence of
battles and urban growth, with all the social and
economic changes this growth implicates, should not be
taken as an indication that one caused the other, but
they are inter-related. Finally, we return to Carneiro’s
(1970) theory of warfare’s role in state origins. The Brak
case does not involve his environmental circumscription,
since the northern Mesopotamian plains present few
physical limits on resources or space. Brak could,
however, provide evidence for ‘‘social circumscription,’’
a social inertia and even centripetal force created by the
new institutions and economic efficiencies of urbanism.
This social conscription might have confined and limited
behavior, preventing adaptive strategies such as fission,
even as it created internal unrest and resentment.
Acknowledgments
The 2007–2008 fieldwork at Tell Brak was funded by
the British Institute for the Study of Iraq; the
McDonald Institute for Archaeological Research,
Cambridge; Newnham College, Cambridge; and the
Society of Antiquaries of London. Thanks are due to
the entire Tell Brak team, especially the residents of
modern Tell Brak, Joan Oates, Agata Chilin´ska,
Fahd Sbahi, Tim Skuldboel, Ahmed Slevi, Adam
McMahon et al. Mass Graves, Conflict, and Early Urbanism, Syria
Journal of Field Archaeology 2011 VOL. 36 NO. 3 217
Stone, and Matt Williams. We also are most grateful
to Bassam Jamous and Michel al-Maqdissi of the
Directorate-General of Antiquities and Museums in
Damascus and M. Abdel-Massih Baghdo of the
Department of Antiquities in Hasseke for their
generous support of the research at Tell Brak. This
article has been greatly improved by the suggestions
of Guillermo Algaze, Graham Philip, and an
anonymous reviewer.
Augusta McMahon (Ph.D. 1993, University of
Chicago) is a Senior Lecturer in the Department of
Archaeology, University of Cambridge, and has
been Field Director of the Tell Brak Project
since 2006. Her research interests include settlement
archaeology, site and artifact biographies, urbanism,
warfare, and human adaptations to political and
climate change.
Arkadiusz Sołtysiak (Ph.D. 2001, University of
Warsaw) is the head of the Department of Bioarchaeology,
Institute of Archaeology, University of
Warsaw, Poland, and the editor of Bioarchaeology
of the Near East. He has studied human remains from
more than 20 archaeological sites in Syria, Iran, Iraq,
and Kuwait.
Jill Weber (Ph.D. 2006, University of Pennsylvania) is
a Consulting Scholar of the Near East Section of the
University Museum of Archaeology and Anthropology,
University of Pennsylvania. Her research focuses
broadly on the practical and symbolic interactions
between humans and animals.
References
Adams, R. McC. 1972. ‘‘Patterns of Urbanization in early
Southern Mesopotamia,’’ in P. J. Ucko, R. Tringham, and
G. W. Dimbleby, eds., Man, Settlement and Urbanism.
London: Duckworth, 735–749.
Adams, R. McC. 1981. Heartland of Cities. Chicago: University of
Chicago Press.
Adams, R. McC., and H. J. Nissen. 1972. The Uruk Countryside:
The Natural Setting of Urban Societies. Chicago: University of
Chicago Press.
Algaze, G. 1993. The Uruk World System: The Dynamics of
Expansion of Early Mesopotamian Civilization. Chicago:
University of Chicago Press.
Algaze, G. 2008. Ancient Mesopotamia at the Dawn of Civilization,
The Evolution of an Urban Landscape. Chicago: University of
Chicago Press.
Algaze, G., G. Dinckan, B. Hartenberger, T. Matney, J. Pournelle,
L. Rainville, S. Rosen, E. Rupley, D. Schlee, and R. Vallet.
2001. ‘‘Research at Titirish Ho¨yu¨k in Southeastern Turkey:
The 1999 Season,’’ Anatolica 27: 23–106.
Alison, M. 2009. Women and Political Violence: Female Combatants
in Ethno-National Conflict. London: Routledge.
Andrushko, V. A., K. A. S. Latham, D. L. Grady, A. G. Pastron,
and P. L. Walker. 2005. ‘‘Bioarchaeological Evidence for
Trophy-Taking in Prehistoric Central California,’’ American
Journal of Physical Anthropology 127: 375–384.
Andrushko, V. A., A. W. Schwitalla, and P. L. Walker. 2010.
‘‘Trophy-Taking and Dismemberment as Warfare Strategies in
Prehistoric Central California,’’ American Journal of Physical
Anthropology 141: 83–96.
Archi, A., and M. Biga. 2003. ‘‘A Victory over Mari and the Fall of
Ebla,’’ Journal of Cuneiform Studies 55: 1–44.
Bar-Matthews, M., A. Ayalon, M. Gilmour, A. Matthews, and C.
Hawkesworth. 2003. ‘‘Sea-land Oxygen Isotopic Relationships
from Planktonic Foraminifera and Speleothems in the Eastern
Mediterranean Region and their Implication for Paleorainfall
during Interglacial Intervals,’’ Geochimica et Cosmochimica
Acta 67: 3181–3199.
Bar-Matthews M., A. Ayalon, and A. Kaufman. 1997. ‘‘Late
Quaternary Paleoclimate in the Eastern Mediterranean Region
from Stable Isotope Analysis of Speleothems at Soreq Cave,
Israel,’’ Quaternary Research 47: 155–168.
Bass, W. M. 1996. ‘‘Outdoor Decomposition Rates in Tennessee,’’
in W. D. Haglund and M. H. Sorg, eds., Forensic Taphonomy.
The Postmortem Fate of Human Remains. Boca Raton, FL:
CRC Press, 181–186.
Billman, B. R., P. M. Lambert, and B. L. Leonard. 2000.
‘‘Cannibalism, Warfare, and Drought in the Mesa Verde
Region during the Twelfth Century A. D.,’’ American Antiquity
65: 145–178.
Boehmer, R. M. 1999. Uruk: Fru¨heste Siegelabrollungen.
Ausgrabungen in Uruk-Warka Endberichte 24. Mainz: P. von
Zabern.
Carneiro, R. 1970. ‘‘A Theory of the Origin of the State,’’ Science
169/3947: 733–738.
Chacon, R., and D. Dye, eds. 2007. The Taking and Displaying
of Human Body Parts as Trophies by Amerindians.
Interdisciplinary Contributions to Archaeology. New York:
Springer.
Charles, M., M. M. Hald, J. Weber, and H. T. Wright. 2001. ‘‘The
forecourt of the Niched Building (Tell Brak TW Level 18):
Preliminary note on an interdisciplinary study,’’ in G.
Emberling and H. McDonald, ‘‘Excavations at Tell Brak
2000: Preliminary Report,’’ Iraq 63: 21–54.
Chew, S. C. 2007. The Recurring Dark Ages: Ecological Stress,
Climate Changes, and System Transformation. Lanham, MD:
AltaMira Press.
Childe, V. G. 1950. ‘‘The Urban Revolution,’’ Town Planning
Review 21: 3–17.
Coale A. J., P. Demeny, and B. Vaughan. 1983. Regional Model
Life Tables and Stable Populations. New York and London:
Academic Press.
Cooper, J. 1983. Reconstructing History from Ancient Inscriptions:
The Lagash-Umma Border Conflict. Sources from the Ancient
Near East 2/1. Malibu, CA: Undena Publications.
Dawson, L., T. E. Levy, and P. Smith. 2003. ‘‘Evidence of
Interpersonal Violence at the Chalcolithic Village of Shiqmim
(Israel),’’ International Journal of Osteoarchaeology 13: 115–
119.
Earle, T. 2000. ‘‘Archaeology, Property and Prehistory,’’ Annual
Review of Anthropology 29: 39–60.
Elmoslimany, A. P. 1990. ‘‘Ecological Significance of Common
Nonarboreal Pollen. Examples from Drylands of the Middle
East,’’ Review of Palaeobotany and Palynology 64: 343–350.
Emberling, G., J. Cheng, T. E. Larsen, H. Pittman, T. B. B.
Skuldboel, J. Weber, and H. T. Wright. 1999. ‘‘Excavations at
Tell Brak 1998: Preliminary Report,’’ Iraq 61: 1–42.
Emberling, G., and H. McDonald. 2003. ‘‘Excavations at Tell Brak
2001–2002: Preliminary Report,’’ Iraq 65: 1–75.
Erdal, O¨ . D. 2010. ‘‘A Possible Massacre at Early Bronze Age
Titrish Ho¨yu¨k, Anatolia,’’ International Journal of
Osteoarchaeology n/a. doi: 10.1002/oa.1177.
Finkbeiner, U. 1991. Uruk. Kampagne 35–37, 1982–84. Die
archa¨ologische Oberfla¨chenuntersuchung. Ausgrabungen in
Uruk-Warka Endberichte 4. Mainz: P. von Zabern.
Flinn, L., C. G. Turner, II, and A. Brew. 1976. ‘‘Additional
Evidence for Cannibalism in the Southwest: The Case of LA
4528,’’ American Antiquity 41: 308–318.
Forbes, S. L. 2008. ‘‘Potential Determinants of Postmortem and
Postburial Interval of Buried Remains,’’ in M. Tibbett and D.
O. Carter, eds., Soil Analysis in Forensic Taphonomy. Boca
Raton, FL: CRC Press, 225–246.
Galloway, A. 1996. ‘‘The Process of Decomposition: A Model from
the Arizona-Sonoran Desert,’’ in W. D. Haglund and M. H.
Sorg, eds., Forensic Taphonomy. The Postmortem Fate of
Human Remains. Boca Raton, FL: CRC Press, 139–150.
Haas, J. 1999. ‘‘The Origins of War and Ethnic Violence,’’ in J.
Carman and A. Harding, eds., Ancient Warfare: Archaeological
Perspectives. Stroud: Sutton Publishing, 11–24.
McMahon et al. Mass Graves, Conflict, and Early Urbanism, Syria
218 Journal of Field Archaeology 2011 VOL. 36 NO. 3
Jordana, X., I. Galte´s, T. Turbat, D. Batsukh, C. Garcia, A. Isidro,
P-H. Giscard, and A. Malgosa. 2009. ‘‘The Warriors of the
Steppes: Osteological Evidence of Warfare and Violence from
Pazyryk Tumuli in the Mongolian Altai,’’ Journal of
Archaeological Science 36: 1319–1327.
Kansa, S. W., and S. Campbell. 2004. ‘‘Feasting With the Dead?—
A Ritual Bone Deposit at Domuztepe, Southeastern Turkey (c.
5550 cal B. C.),’’ in S. J. O’Day, W. van Neer, and A. Ervynck,
eds., Behaviour Behind Bones: the Zooarchaeology of Ritual,
Religion, Status and Identity. Proceedings of the 9th ICAZ
Conference, Vol 1. Oxford: Oxbow Books, 2–13.
Kansa S. W., S. C. Gauld, S. Campbell, and E. Carter. 2009.
‘‘Whose Bones are Those? Preliminary Comparative Analysis
of Fragmented Human and Animal Bones in the ‘‘Death Pit’’
at Domuztepe, a Late Neolithic Settlement in Southeastern
Turkey,’’ Anthropozoologica 44: 159–172.
Kantner, J. 1999. ‘‘Anasazi Mutilation and Cannibalism in the
American Southwest,’’ in L. R. Goldman, ed., The
Anthropology of Cannibalism. Westport, CT: Bergin and
Garvey, 75–104.
Keeley, L. H. 1996. War Before Civilization: The Myth of the
Peaceful Savage. Oxford: Oxford University Press.
Kelly, R. C. 2000. Warless Societies and the Origin of War. Ann
Arbor: University of Michigan Press.
Knudson, K., S. Williams, R. Osborn, K. Forgey, and P. R.
Williams. 2009. ‘‘The Geographic Origins of Nasca Trophy
Heads using Strontium, Oxygen Carbon Isotope Data,’’
Journal of Anthropological Archaeology 28: 244–257.
Lawler, A. 2006. ‘‘North Versus South, Mesopotamian Style,’’
Science 312: 1458–1463.
LeBlanc, S. A. 1999. Prehistoric Warfare in the American
Southwest. Salt Lake City: University of Utah Press.
Liston, M., and B. Baker. 1996. ‘‘Reconstructing the Massacre at
Fort William Henry, New York,’’ International Journal of
Osteoarchaeology 6: 28–41.
Liverani, M. 2006 (Originally published 1998, Uruk: La Prima
Citta`). Uruk, The First City. London: Equinox.
Mallowan, M. E. L. 1947. ‘‘Excavations at Brak and Chagar
Bazar,’’ Iraq 9: 1–259.
Margerison, B. J., and C. Knu¨sel. 2002. ‘‘Paleodemographic
Comparison of a Catastrophic and an Attritional Death
Assemblage,’’ American Journal of Physical Anthropology
119: 134–143.
Marlar, R. A., B. L. Leonard, B. Billman, P. M. Lambert, and J. E.
Marlar. 2000.‘‘Biochemical Evidence of Cannibalism at a
Prehistoric Puebloan Site in Southwestern Colorado,’’ Nature
407: 74–78
Matney, T., G. Algaze, M. C. Dulik, O¨ . D. Erdal, Y. S. Erdal, O.
Gokcumen, J. Lorenz, and H. Mergen. 2010. ‘‘Understanding
Early Bronze Age Social Structure through Mortuary Remains:
A Pilot aDNA Study from Titrish Ho¨yu¨k, Southeastern
Turkey,’’ International Journal of Osteoarchaeology n/a. doi:
10.1002/oa.1213.
Matthews, R. 2004. The Excavations at Tell Brak 4: Exploring an
Upper Mesopotamian Regional Centre, 1994–96. Cambridge
and London: McDonald Institute Monographs and British
School of Archaeology in Iraq.
Mayewski, P., E. Rohling, J. C. Stager, W. Karle´n, K. Maasch,
L. D. Meeker, E. Meyerson, F. Gasse, S. van Kreveld, K.
Holmgren, J. Lee-Thorp, G. Rosqvist, F. Rack, M.
Staubwasser, R. Schneider, and E. Steig. 2004. ‘‘Holocene
Climate Variability,’’ Quaternary Research 62: 243–255.
McMahon, A. 2009. ‘‘The Lion, the King and the Cage: Late
Chalcolithic Iconography and Ideology,’’ Iraq 71: 115–124.
McMahon, A., and J. Oates. 2007. ‘‘Excavations at Tell Brak, 2006
and 2007,’’ Iraq 69: 145–171.
Muscarella, O. 1989. ‘‘Warfare at Hasanlu in the Late 9th Century
B. C.,’’ Expedition 31(2–3): 24–36.
Nissen, H. 2002. ‘‘Uruk: Key Site of the Period and Key Site of the
Problem,’’ in J. N. Postgate, ed., Artefacts of Complexity;
Tracking the Uruk in the Near East. London: British School of
Archaeology in Iraq, 1–16.
Nissen, H., and P. Heine. 2009. From Mesopotamia to Iraq: A
Concise History. Chicago: University of Chicago Press.
Oates, D., and J. Oates. 1993. ‘‘Excavations at Tell Brak 1992–93,’’
Iraq 55: 155–200.
Oates, J. 2002. ‘‘Tell Brak: The 4th Millennium Sequence and Its
Implications,’’ in J. N. Postgate, ed., Artefacts of Complexity;
Tracking the Uruk in the Near East. London: British School of
Archaeology in Iraq, 111–122.
Oates, J. 2005. ‘‘Digging Deeper at Tell Brak,’’ Proceedings of the
British Academy 131: 1–39.
Oates, J., A. McMahon, P. Karsgaard, S. al-Quntar, and J. Ur.
2007. ‘‘Early Mesopotamian Urbanism: A New View from the
North,’’ Antiquity 81: 585–600.
Ortner, D. 2003. Identification of Pathological Conditions in Human
Skeletal Remains. San Diego, CA: Academic Press.
Pollock, S. 1999. Ancient Mesopotamia. Cambridge: Cambridge
University Press.
Pollock, S. 2007. ‘‘The Royal Cemetery of Ur: Ritual, Tradition,
and the Creation of Subjects,’’ in M. Heinz and M. Feldman,
eds., Representations of Political Power; Case Histories from
Times of Change and Dissolving Order in the Ancient Near East.
Winona Lake, IN: Eisenbrauns, 89–110.
Pollock, S., C. Steele, and M. Pope. 1991. ‘‘Investigations on the
Uruk Mound, Abu Salabikh, 1990,’’ Iraq 53: 59–68.
Postgate, J. N., ed. 2002. Artefacts of Complexity; Tracking the
Uruk in the Near East. London: British School of Archaeology
in Iraq.
Potter, J. M., and J. P. Chuipka. 2010. ‘‘Perimortem Mutilation of
Human Remains in an Early Village in the American
Southwest: A Case for Ethnic Violence,’’ Journal of
Anthropological Archaeology 29: 507–523.
Rothman, M., ed. 2001. Uruk Mesopotamia and Its Neighbors:
Cross-Cultural Interactions in the Era of State Formation.
Advanced Seminar Series. Santa Fe, NM: School of American
Research.
Skuldboel, T. B. B. 2009. Earliest Urbanism in Northern
Mesopotamia. Ph. D. dissertation, University of Copenhagen.
Copenhagen: University of Copenhagen.
Sołtysiak, A. 2010. Death and Decay at the Dawn of the City:
Interpretation of Human Bone Deposits at Tell Majnuna.
Warsaw: Institute of Archaeology, University of Warsaw.
Stein, G. 1999. Rethinking World Systems; Diasporas, Colonies and
Interaction in Uruk Mesopotamia. Tucson: University of
Arizona Press.
Stronach, D., and S. Lumsden. 1992. ‘‘UC Berkeley’s Excavations
at Nineveh,’’ Biblical Archaeologist 55: 227–233.
Tonietti, M. V. 2010. ‘‘The Expedition of Ebla against Ashdar(um)
and the Queen of Harran,’’ Zeitschrift fu¨r Assyriologie und
vorderasiatische Archa¨ologie 100: 56–85.
Turner, C. G., II, and J. A. Turner. 1999. Man Corn: Cannibalism
and Violence in the Prehistoric American Southwest. Salt Lake
City: University of Utah Press.
Ubelaker, D. H. 1996. ‘‘Taphonomic Applications in Forensic
Anthropology,’’ in W. D. Haglund and M. H. Sorg, eds.,
Forensic Taphonomy. The Postmortem Fate of Human Remains.
Boca Raton, FL: CRC Press, 77–90.
Ur, J., P. Karsgaard, and J. Oates. 2007. ‘‘Early Urban
Development in the Near East,’’ Science 317/5842: 1188.
Ur, J., P. Karsgaard, and J. Oates. In press. ‘‘Spatial Dimensions of
Early Mesopotamian Urbanism: The Tell Brak Suburban
Survey, 2003–2006,’’ Iraq 73.
van de Mieroop, M. 1997. The Ancient Mesopotamian City.
Oxford: Clarendon Press.
Walker, P. L. 1989. ‘‘Cranial Injuries as Evidence of Violence in
Prehistoric Southern California,’’ American Journal of Physical
Anthropology 80: 313–323.
Walker, P. L. 2001. ‘‘A Bioarchaeological Perspective on the
History of Violence,’’ Annual Review of Anthropology 30: 573–
596.
Weber, J. In Press. ‘‘Patterns of Animal Production and Harvesting
in the 4th and 3rd Millennia B. C. at Tell Brak,’’ in J. Oates,
ed., Excavations at Tell Brak, Volume III: Early Urbanism and
Prehistory. Cambridge and London: McDonald Institute
Monographs and British School of Archaeology in Iraq.
Weber, M. 1968. Economy and Society. Edited by G. Roth and C.
Wittich. New York: Bedminster.
Weninger, B., L. Clare, E. Rohling, O. Bar-Yosef, U. Bo¨hner, M.
Budja, M. Bundschuh, A. Feurdean, H-G. Gebel, O. Jo¨ris, J.
Linsta¨dter, P. Mayewski, T. Mu¨hlenbruch, A. Reingruber, G.
Rollefson, D. Schyle, L. Thissen, H. Todorova, and C.
Zielhofer. 2009. ‘‘The Impact of Rapid Climate Change on
Prehistoric Societies during the Holocene in the Eastern
Mediterranean,’’ Documenta Praehistorica 36: 7–59.
McMahon et al. Mass Graves, Conflict, and Early Urbanism, Syria
Journal of Field Archaeology 2011 VOL. 36 NO. 3 219
Wilkinson, T. J., and D. J. Tucker. 1995. Settlement Development
in the North Jazira, Iraq. A Study of the Archaeological
Landscape. Iraq Archaeological Reports. Warminster: Aris and
Phillips.
Wright, H. T. 1984. ‘‘Prestate Political Formations,’’ in T. Earle,
ed., On the Evolution of Complex Societies: Essays in Honor of
Harry Hoijer 1982. Malibu, CT: Undena Publications, 41–78.
Wright, H. T. 2006. ‘‘Early State Dynamics as Political Experiment,’’
Journal of Anthropological Research 62: 305–319.
Wright, H. T., and G. Johnson. 1975. ‘‘Population, Exchange, and
Early State Formation in Southwestern Iran,’’ American
Anthropologist 77: 267–289.
Wright H. T., E. Rupley, J. Ur, J. Oates, and E. Ganem. 2007.
‘‘Preliminary Report on the 2002 and 2003 Seasons of the Tell
Brak Sustaining Area Survey,’’ Annales Arche´ologiques Arabes
Syriennes 49/50: 7–21.
Yoffee, N. 1995. ‘‘Political Economy in Early Mesopotamian
States,’’ Annual Review of Anthropology 24: 281–311.
McMahon et al. Mass Graves, Conflict, and Early Urbanism, Syria
220 Journal of Field Archaeology 2011 VOL. 36 NO. 3