Content uploaded by Dragana Filipovic
Author content
All content in this area was uploaded by Dragana Filipovic on Feb 07, 2014
Content may be subject to copyright.
Late Mesolithic lifeways and deathways at
Vlasac (Serbia)
Dus
ˇan Boric
´
1
, Charles A. I. French
2
, Sofija Stefanovic
´
3
, Vesna Dimitrijevic
´
3
,
Emanuela Cristiani
2
, Maria Gurova
4
, Dragana Antonovic
´
5
, Ethel Allue
´
6
,
Dragana Filipovic
´
7
1
Cardiff University, Cardiff, U.K.,
2
University of Cambridge, Cambridge, U.K.,
3
Belgrade University, Belgrade,
Serbia,
4
Bulgarian Academy of Sciences, Sofia, Bulgaria,
5
Serbian Academy of Arts and Sciences, Belgrade,
Serbia,
6
Universitat Rovira i Virgili, Tarragona, Spain,
7
Oxford University, Oxford, U.K.
Recent excavations (2006–2009) at the Mesolithic-Neolithic site of Vlasac in the Danube Gorges region of
the north-central Balkans have focused on a reevaluation of previous conclusions about site formation
processes, stratigraphy, chronology, and the nature of occupation. Mostly Late Mesolithic remains had
been encountered in the preserved portion of the site, but, for the first time, in a restricted zone of the
excavated area, vertical stratification of burial and occupation features yielded evidence about the use of
the site in the period that is contemporaneous with Phase I–II at Lepenski Vir, the Mesolithic-Neolithic
transition phase in this region, ca. 6200–5900 CAL B.C. Various strands of archaeological evidence show
both continuities and discontinuities in Late Mesolithic forager life- and deathways at the start of the
Neolithic in the central Balkans.
Keywords: southeastern Europe, Mesolithic-Neolithic transition, Vlasac, the Danube Gorges, forager-farmer interactions
Introduction: The Research Context
There are few places in Europe, and even fewer in
southeastern Europe, where one can study details of
the Mesolithic-Neolithic transition when autochtho-
nous communities of Late Mesolithic foragers were
affected by the spread of the Neolithic economy and
lifeways. Despite years of research, including targeted
surveys that aimed at identifying pre-Neolithic
occupation in parts of Greece and the Balkans (e.g.,
Chapman et al. 1996; Cherry and Parkinson 2003;
Runnels et al. 2005; Runnels and van Andel 2003), we
still know little about the character of early Holocene
human adaptation here, although the region shows
unquestionable chronological priority in the spread
of the Neolithic way of life into Europe from
southwestern Asia (cf., Bocquet-Appel et al. 2009;
Gkiasta et al. 2003). Researchers have focused their
efforts on countering various biases (e.g., preference
for later prehistoric periods with ceramic finds,
unsystematic surveys, etc.) in the history of research
regarding the pre-Neolithic periods in this part of
Europe (Galanidou 1996) through micro-regional
case studies, which provide substantial evidence for
forager lifeways across the region (e.g., Eichmann
et al. 2010; Galanidou 2011; Galanidou and Perle`s
2003; Gaspari 2006; Komsˇo 2006; Miracle 1997,
2001; Miracle et al. 2000; Mlekuzˇ et al. 2008; Runnels
et al. 2004, 2009; Sampson 2007, 2010).
In contrast to this patchy archaeological record for
foragers is the evidence from the Danube Gorges area
in the north-central Balkans. The Danube Gorges area
is split by the River Danube between the territories of
present day Romania and Serbia (FIG. 1). With the
discovery of more than 20 sites having Mesolithic
deposits in the Danube Gorges area in the mid-1960s,
new data became available that were unprecedented
for other areas with a known Mesolithic presence in
Europe (cf., Bonsall 2008; Boric
´2011; Boroneant¸
2000; Clarke 1976; Jovanovic
´2008; Nandris 1971;
Radovanovic
´1996; Srejovic
´1972; Tringham 2000).
However, excavations conducted at these sites were
rather crude rescue projects because of the threat of
the rising waters of the artificial lake created by the
construction of a hydroelectric dam (Boric
´2011: 159;
Boroneant¸ 2000: 11–15; Radovanovic
´1996: 3–8). This,
in tandem with a controversy about the dating of the
key site of Lepenski Vir (Boric
´1999, 2002a) as well as
the slow pace of the publishing of the primary data
from the 1960s–1980s excavations in this region, left
this rich corpus of archaeological data undeservedly
on the margins of archaeological discussions about the
character and tempo of the Mesolithic-Neolithic
Correspondence to: Dusˇan Boric
´, Department of Archaeology and
Conservation, SHARE, Cardiff University, Colum Drive, Cardiff CF10
3EU, U.K. Email: BoricD@Cardiff.ac.uk
4
ßTrustees of Boston University 2014
DOI 10.1179/0093469013Z.00000000070 Journal of Field Archaeology 2014 VOL.39 NO.1
transition in Europe. This research context signifi-
cantly changed in the mid-1990s with a fresh synthesis
of the available evidence by I. Radovanovic
´(1996),
new absolute dating of the rich corpus of mortuary
remains coupled with stable isotope analyses (Bonsall
et al. 1997), and the reexamination of collections with
human osteological and faunal remains along with
other material remains (Antonovic
´2006; Boric
´1999,
2002a; Boric
´and Dimitrijevic
´2007; Roksandic
´2000;
Stefanovic
´and Boric
´2008). In addition, new field
research was undertaken in the early 1990s at the only
site at the time still available for continuing research
after the rise of the Danube—Schela Cladovei, (FIG. 1)
on the Romanian side of the river (Bonsall 2008;
Boroneant¸ et al. 1999).
In 2004, a collaborative survey and excavation
project, titled ‘‘Prehistory of North-East Serbia,’’ was
initiated by the Department of Archaeology, Univer-
sity of Cambridge, U.K. and the Department of
Archaeology of the University of Belgrade, Serbia,
with Dusˇan Boric
´and MilosˇJevtic
´as the principal
investigators. A part of this wider project was designed
to test the validity of the forager-farmer moving
frontier model of cultural change and its applicability
in this region by reference to known Mesolithic
settlements on the Danube and largely uninvestigated
hinterland areas on the Serbian side of the river. In
2004–2005, the project focused on cave sites across this
karstic area (Boric
´and Jevtic
´2008) but did not confirm
their use during the Mesolithic/Neolithic. However, in
the course of the project the Early/Middle Neolithic
open air site of Aria Babi was discovered, situated on
Kosˇo Hill (FIG. 2) above the site of Lepenski Vir (Boric
´
2007a, 2011; Boric
´and Starovic
´2008). Further survey
efforts in 2006 discovered preserved Mesolithic depos-
its at the previously investigated site of Vlasac (FIG. 3A),
situated in the Upper Gorge or Lady’s Whirlpool
(Gospod
–
in Vir) region of the Danube River (Srejovic
´
and Letica 1978). Due to the endangered nature of the
deposits that were exposed to constant erosion by the
Danube, and the possibility of excavating deposits with
higher standards of recovery and recording than those
applied during the first excavations (1970–1971), the
preserved portion of the site was excavated from 2006
to 2009 and protected from further erosion (FIG. 3B).
This new research at this ‘‘classic’’ Mesolithic site
in the Danube Gorges region made possible a better
understanding of forager-farmer interactions in this
part of the Balkans. New evidence allowed us to
understand in more detail the nature of the Late
Mesolithic (ca. 7300–6200 CAL B.C.) occupation with
regard to the life- and deathways of the generations
of people who had inhabited it. For the first time
Vlasac revealed contexts that are now dated on the
basis of both material culture and radiocarbon assays
to the Mesolithic-Neolithic transition (ca. 6200–5900
CAL B.C.), with evidence of continuous development
from the Late Mesolithic in one vertically stratified
sequence. There is also unequivocal evidence for
Mesolithic forager-Neolithic farmer contact, offering
Figure 1 Map of the Danube Gorges region with the location of principal Epipalaeolithic and Mesolithic sites. Dots5open-air
sites. Half-dots5caves and rockshelters.
Boric
´et al. Late Mesolithic lifeways and deathways at Vlasac (Serbia)
Journal of Field Archaeology 2014 VOL.39 NO.1 5
greater understanding of the nature of this interac-
tion. We also reexamined the larger corpus of
evidence recorded and collected at the site in
1970–1971.
What follows is an interim report on this work,
highlighting both continuities and changes that affected
these foragers during the Mesolithic-Neolithic transi-
tion. We focus on the evidence for subsistence and
material culture manufacture and use, as well as on the
evidence for ritual and other structured deposition.
Vlasac: Research History and Site Setting
Vlasac is one of the Mesolithic-Neolithic sites found
in the Upper Gorge of the Danube (FIG. 2). The
geological history of the Danube Gorges (also known
as the Iron Gates) is marked by different geological
strata visible as the Danube cuts the southern extent
of the Carpathian Mountains. The Danube Gorges
area, connecting the Pannonian and Dacian basins,
is some 150 km long and is composed of three
small valleys and four gorges with distinct geological
histories and irregular riverbeds due to differential
erosion of the underlying rocks (Vulcanescu 1972;
Markovic
´-Marjanovic
´1978). By the end of the last
glacial period, the Danube was a very large meandering
and fast flowing river confined within the limestone/
granitic and sedimentary rock-dominated gorges, with
narrows, cataracts, and terrace remnants on the
floodplain edges. These terrace remnants often occur
as promontories on the valley floor, and are composed
of riverine sand, wind blown loessic silt and/or scree off
the adjacent steep slopes, and they are often re-cut and
re-carved by channel avulsion processes. It was on
these ‘‘tongues’’ of land projecting at near right angles
to the adjacent valley slopes that the Mesolithic peoples
established themselves in settlements with burial sites:
Lepenski Vir on finely laminated riverine sands and
Vlasac on granitic and limestone derived scree. These
floodplain edge ‘‘terrace remnants’’ could be seen as
more accessible, as they were just above the river’s
influence, but not yet covered to the same extent with
the thick and developing woodland that gradually
blanketed the adjacent slopes in the early Holocene.
Vlasac is located close to the downstream exit of
the Upper Gorge of the Danube, marked by a large
promontory known as Greben. The site is found near
Greben, at the place known as Tahtalija, at the foot
of Boljetinsko Hill (FIG. 2). Before the 19th-century
regulation of the Danube, there used to be a
dangerous cataract at Tahtalija with the strong sound
of roaring water (Petrovic
´1941). Here the channel is
2400 m wide and rocky, which in the past created
navigational problems. Greben is a large promontory
and it narrows the navigation channel to 420 m or,
together with the rocky plateau called Vranj at the
time of low water levels, to 220 m. Immediately after
Greben there is a very strong and deep whirlpool (30
m deep) (Petrovic
´1941). After Greben the Danube
widens again and the depth is ca. 3–9 m; the Upper
Gorge is followed by the Donji Milanovac valley
filled with Miocene sediments (Markovic
´-Marjanovic
´
1978).
D. Srejovic
´and Z. Letica (1978) investigated the
site in 1970–1971, covering an area of 640 sq m along
Figure 2 View of the Upper Gorge of the Danube from the
Neolithic site of Aria Babi on Kosˇ o Hill above the site of
Lepenski Vir. The arrow indicates the location of the
Mesolithic-Neolithic site of Vlasac.
Figure 3 A) View of Vlasac at the beginning of excavations
in 2006, facing west; B) View of Vlasac under excavation in
2007, facing west.
Boric
´et al. Late Mesolithic lifeways and deathways at Vlasac (Serbia)
6Journal of Field Archaeology 2014 VOL.39 NO.1
the riverbank up to an altitude of around 70 masl, the
zone that would have been flooded with the
construction of the Ðerdap (Iron Gates) Dam I some
100 km downstream from Vlasac. The rising water
levels of the artificial lake in front of the first dam in
1971 prevented further work in the area below 70
masl while the Danube slowly eroded away the newly
created riverbank section, continuing to the start of
our work in 2006 (FIG. 3A). Prior to the discovery and
excavation of Vlasac, D. Srejovic
´(1972) excavated
the neighboring site of Lepenski Vir, located approxi-
mately 3 km upstream from Vlasac. Lepenski Vir is
well known for elaborate trapezoidal limestone
building floors and central stone-lined hearths along
with the presence of numerous carved sandstone
boulder artworks, some depicting hybrid human-fish
beings (Boric
´2005; Radovanovic
´1996; Srejovic
´1972;
Srejovic
´and Babovic
´1982). Most of these features
are now dated to the Mesolithic-Neolithic transition
from ca. 6200 to 5900 CAL B.C. (Boric
´2011).
The 2006–2009 field seasons at Vlasac covered an
area of 326 sq m, investigating a 63 m stretch of the
new riverbank section created after 1971 in the (likely
peripheral) southernmost part of the site (FIG. 4). This
new effort was upslope from the area that was
excavated in 1970–1971. In the course of new
excavations at the site, all archaeological deposits
were watersieved using 3 to 5 mm mesh. Flotation
samples were taken from every unit or context,
commonly 20 L of soil but occasionally also 35 L per
unit or more. Heavy residue fractions were sieved
with the addition of 0.3 mm and 0.5 mm mesh,
enabling the recovery of a large number of small
artifacts, including different types of beads.
Formation Processes and Stratigraphy
The first excavators of the site, Srejovic
´and Letica
(1978), on the basis of observations made during
their 1970–1971 campaign at Vlasac, described the
cultural stratigraphy of the site as consisting of four
‘‘cultural’’ horizons or phases: Vlasac I to IV. Vlasac
I was further divided into two subphases: Ia and Ib.
While a detailed ‘‘dissection’’ of stratigraphic pro-
blems related to this older division is provided
elsewhere (Boric
´et al. 2008), it suffices to say that
the formation processes at Vlasac along with its
topography suggest that these horizons cannot be
maintained as defined by the first excavators of
the site, the principal reason being that the site
deposits were formed through complex colluvial
processes on a sloping river terrace. Our recent
fieldwork, which was aided by geoarchaeological and
micromorphological examinations of these compo-
site colluvial deposits, suggests that the horizons
reflect woodland clearance and downslope movement
of scree, and these factors must be incorporated
into an adequate understanding of the site formation
processes.
The stratigraphy of Vlasac largely consists of
hillwash deposits formed on top of the palaeosol at
the base of the stratigraphic sequence. Minor colluvial
episodes continued to affect this soil, alternating with
periods of stability and incipient soil development, and
together led to slow soil aggradation and thickening
throughout the later Mesolithic and into the Early/
Middle Neolithic period. This early Holocene soil is
essentially a cumulative colluvial soil. The composite
hillslope sequence of deposits as seen on the north-
facing, exposed riverbank section of Trench 3/2006 at
Figure 4 Site plan of Vlasac showing only Late Mesolithic pits and burials (indicated with ‘‘F.’’ and ‘‘H.’’ prefixes respectively)
in the central part of the area excavated in 2006–2009 in the southernmost part of the site with Trenches 1–3/2006, 2–3/2007, and
1/2008 (indicated with ‘‘T.’’ prefixes). Context numbers are in brackets. Drawn by Dusˇ an Boric
´and Miroslav Koc
ˇic
´.
Boric
´et al. Late Mesolithic lifeways and deathways at Vlasac (Serbia)
Journal of Field Archaeology 2014 VOL.39 NO.1 7
Vlasac is shown in Figures 5 and 6 and described in the
online supplement.
The complete profile at Vlasac exhibits episodic
deposition of variable mixtures of soil and chalk
rubble hillwash occurring around trees within greater
or lesser amounts of open woodland. Thus, some
areas of the hillside were more intact and stable than
others; some areas were severely affected by overland
flow hillwash processes. There is much ‘‘tree throw’’
activity in evidence, i.e., tree uprooting which causes
soil disturbance features that are visible both in
the field and through the polarizing microscope. In
particular, tree throw may lead to the mixing and
inversion of soil fabrics from different horizons and
Figure 5 North-facing section in Trench 3/2006. A) Section drawing with stratification of context numbers and burials (with
‘‘H.’’ prefixes) and vertical positions of AMS dates and associated context/burial numbers (see online supplement for a
description of the stratigraphic sequence); dotted lines indicate uncertain context boundaries; burnt soil from cremation zones
is light shaded; reddish flooring is dark shaded; visible bones are shaded black; 2785292 marks the basal scree deposit; B)
Cleared section before the start of excavations in 2006.
Boric
´et al. Late Mesolithic lifeways and deathways at Vlasac (Serbia)
8Journal of Field Archaeology 2014 VOL.39 NO.1
the movement of fine silt, organic matter and clay
down profile. Former root depressions and root
disturbed areas are often associated with concentra-
tions of Mesolithic artifacts. Hillwash activity essen-
tially occurred on bare, devegetated slopes; colluvial
slumping may have led to some folding over of
existing deposits on the hillside, such as the upper/
uphill fill of the inhumation burials, and even the
inversion of sediments.
The main archaeological deposits, even though
they occur at different levels on the hillside, are
probably indicative of the same stabilized soil surface
level in the Late Mesolithic, from ca. 7300 CAL B.C. or
perhaps even earlier. Nonetheless, there is little doubt
that this relative stability was broken from time to
time by some downslope soil movement. When the
woodland on the slope above became seriously
disturbed/exploited, hillwash events began in earnest,
and may well have led to the abandonment of parts of
the site at the base of the slope. Archaeologically,
complete abandonment did not occur before the end
of the Middle Neolithic, i.e., sometime between 5700
and 5500 CAL B.C., but parts of the site might have
been abandoned much earlier, i.e. by ca. 6300 CAL
B.C.
At the bottom of the stratigraphic sequence is the
bedrock consisting of gray limestone scree. Above
this level one finds reddish brown calcitic clay, 15–40
cm thick (FIG. 6B) (Boric
´et al. 2008: fig. 11). This is a
palaeosol with some stability, representing incipient
Early Holocene soil formation with woodland cover.
At the upper part of this palaeosol one may see some
anthropogenic activity, possibly related to woodland
clearance, which, with time, intensified downslope
erosion of scree and its deposition in depressions and
tree throws. Cultural activities continued for more
than a millennium (see below) even in those areas
affected by the hillwash accumulation as in the case
of the burial sequence discovered in Trench 3 in 2006
(FIG. 6A).
In sum, on the basis of our understanding of the
colluvial sequence at Vlasac, it is difficult to justify
the assumption that the cultural levels (Vlasac Ia-b to
IV) identified by Srejovic
´and Letica (1978) during
the first excavations of the site were laid down
uniformly across the site; the deposition of anthro-
pogenic sediments at Vlasac must have taken place
at different times in different areas. Two related
processes created the cultural stratigraphy at Vlasac:
on the one hand, the dynamics of hillwash movement
affected by woodland clearance, and, on the other
hand, complex anthropogenic practices of digging
into older deposits and the redepositing of older
layers and materials. These redeposited sediments are
often difficult to notice in the hillwash deposits where
one finds scree of different sizes to be a significant
component of the soil matrix. These observations
were aided by radiocarbon dating of dwelling features
and primary burials found in this complex strati-
graphic matrix.
Stratigraphy and Radiocarbon Dating
Currently, there are 53 radiocarbon measurements
from Vlasac (Boric
´2011: appendix). There are 17
dates from charcoal (excavated in the 1970s) and 36
are more recent AMS dates from samples of human
bones (14 dates) and animal bones (21 measurements
dating 20 contexts), with the addition of one sample
from a charred Cornelian cherry Cornus mas pit.
Here we discuss only 13 dates obtained from features
excavated in the course of the 2006–2009 seasons at
Vlasac (TABLE 1). This series of dates should by and
large be representative of the most intensive period of
occupation/use of the site from the last century of the
8th millennium B.C. to the first century of the 6th
millennium B.C., with a spread of dates throughout
the duration of the 7th millennium B.C. There are no
dates for possible Early Mesolithic features found
sporadically in the area excavated in 1970–1971
(Boric
´et al. 2008). The reasons for this could be the
small number of currently available dates from the
Figure 6 A) Close-up of the north-facing section in Trench 3/
2006 upon exposure of burial H53; B) Bottom of the colluvial
sequence with the reddish palaeosol (2725291) about 20–
55 cm above the base of the profile in squares 104/98 and
105/98 in Trench 1/2007 (see online supplement for a
description).
Boric
´et al. Late Mesolithic lifeways and deathways at Vlasac (Serbia)
Journal of Field Archaeology 2014 VOL.39 NO.1 9
new excavation area; the restricted zone of the Early
Mesolithic settlement, which might have been con-
fined to a much smaller area in comparison to the
Late Mesolithic settlement; and/or possible devasta-
tion and removal of Early Mesolithic deposits by
Late Mesolithic inhabitants of Vlasac. The estab-
lished deposits of the Early/Middle Neolithic occupa-
tion at the site remain inadequately dated at present.
However, 6006–5838 CAL B.C. (OxA-16544) (at 95%
confidence) (TABLE 1) can be taken as a terminus post
quem for the appearance of Starc
ˇevo pottery at the
site on the basis of the stratigraphic sequence
established in Trench 3/2006 (FIGS. 5,6A) (see below).
Dwelling Features
The 1970–1971 excavations at Vlasac revealed remains
of five dwellings with floors of reddish crushed
limestone mixed with sand (Srejovic
´and Letica
1978). There were also 26 rectangular stone-lined
hearths as well as 17 stone constructions of different
shapes and extent, the functions of which remain
unclear, but many of which might have been used as
stone foundations of huts or acted as retaining walls.
In the area of the site where new excavations have
taken place there is much less evidence of dwellings,
likely due to the peripheral position of this remaining
portion of the site. However, we have been able to
identify portions of two dugout features that might
have served as dwellings. Unfortunately only the back
parts of both features were preserved.
In Trench 3/2006, at the bottom of the burial
sequence, the reddish burnt flooring of a possible semi-
subterranean dwelling had only a partly preserved rear
area and one is left to speculate whether it might have
had a trapezoidal shape like similar features discov-
ered in 1970–1971. Unfortunately, the Danube waters
eroded away the front part of this feature and left the
floor line (context 149) visible in the exposed section
(FIGS. 5,6A). Upon the abandonment of this feature a
layer of sterile soil (context 145) was placed over the
Table 1 AMS dates from new excavations at Vlasac (2006–2009). Calibrated with OxCal v. 4.0 (Bronk Ramsey 1995,
2001).
Lab
number Context and material
14
C age
(B.P.)
d
15
N
(%)
d
13
C
(%)
68.2% probability
(CAL B.C.)
95.4% probability
(CAL B.C.)
OxA-16544 Context 19, red deer
skull (VL50) over Burial
H53, Trench 3/2006
7035¡40 6.8 –21.3 5984–5891 6006–5838
OxA-16542 Burial H63, human rib 7701¡39*17.0 –17.7 6212–6066 6232–6018
7261¡60
a
OxA-16539 Context 40, x.8, Trench
3/2006, square 95/96 (20),
large mammal bone (VL18)
7425¡39 6.8 –21.7 6362–6246 6393–6229
OxA-20762 Burial H81, human femur –
proximal diaphysis,
Trench 3/2006
8125¡45*14.0 –19.3 6590–6468 6639–6440
7685¡64
a
OxA-20702 Charred pit of Cornelian
cherry in context 249
(Feature 26); terminus post
quem for H232
7725¡40 – –22.8 6596–6502 6636–6476
OxA-24769 Context 320, x.17 (tool on
red deer antler with traces
of green pigment)
7738¡35 6.6 –22.0 6602–6506 6640–6482
OxA-16540 Context 118, x.1, bone
projectile point (VL21), above
the floor, context 149 of
Feature 12, Trench 3/2006
7764¡38 7.7 –22.1 6644–6531 6654–6484
OxA-16541 Burial H2, human rib (VL42),
Trench 1/2006
8228¡40*16.3 –18.2 6681–6530 6775–6470
7788¡60
a
OxA-18865 Burial H136, human right tibia
(VL1/2008), Trench 3/2006
8231¡36*16.2 –18.5 6684–6530 6774–6472
7791¡58
a
OxA-24809 Context 282, x.7 (tool of large
mammal bone), Trench 3/2006
7943¡40 6.2 –22.1 7025–6703 7034–6692
OxA-24810 Context 282, x.6 (tool of large
mammal bone), Trench 3/2006
7952¡38 5.2 –21.1 7028–6768 7035–6698
OxA-24811 Context 314, x.54, spit 2 (tool
of large mammal bone),
Trench 1/2009
7905¡39 7.4 –22.6 6898–6659 7028–6648
OxA-21962 Context 314, x.23, spit 1, roe
deer skull (S3/2009),
Trench 1/2007
8050¡40 5.0 –23.0 7075–6840 7131–6823
*Values corrected for ages affected by the aquatic reservoir effect (d
15
N values .z10 %) using Method 2 as suggested by Cook
et al. (2002);
a
5100 % reservoir correction applied (440¡45 years).
Boric
´et al. Late Mesolithic lifeways and deathways at Vlasac (Serbia)
10 Journal of Field Archaeology 2014 VOL.39 NO.1
floor. There are several cremation pits found around
this dwelling floor with traces of intense burning and
containing burnt human remains. These pits were
likely dug at a later date around the abandoned
dugout. Above the backfilled floor area of this feature,
the remains of the earliest primary burial in this
location—H136 (see below)—are dated to between
6774 and 6472 CAL B.C. (OxA-18865) (at 95%
confidence) after the correction for the aquatic
reservoir effect (TABLE 1), representing a terminus ante
quem for the occupation of the dwelling floor. There
are two other dates, coming from two bone tools
found in a concentration of artifacts beneath the floor
of the dwelling (context 282). Their respective date
ranges are: 7035–6698 CAL B.C. (OxA-24810) and
7034–6692 CAL B.C. (OxA-24809) (at 95% confidence)
(TABLE 1). Hence the likely construction, use, and
abandonment of the dwelling floor can be estimated to
fall sometime in the first three centuries of the 7th
millennium CAL B.C., which corresponds well with the
dates of dwelling features from the 1970–1971 excava-
tions at Vlasac (cf. Boric
´et al. 2008). A bead made
from the marine gastropod Columbella rustica was
found beneath this feature, in context 282 (FIG. 5A).
Another possible dwelling feature was discovered
in the easternmost portion of the excavation area in
Trenches 1/2007 and 1/2009. Approximately one-third
to one-quarter of the feature (its backside) was
preserved. The feature was cut into the slope of this
part of the terrace removing the reddish palaeosol and
further cutting into the gray sterile soil. On the basis of
the excavated portion of this feature one may
speculate about the oval shape of the dugout but there
is no way of telling whether it might have had a
differently shaped floor area in the front part which
had eroded away. A large flat stone slab was found
next to the eroded section at the level of the possible
dwelling floor. A relatively large number of artifacts
(flint, quartz, bone and antler tools, ground stones,
and burnt limestone plaques from destroyed stone-
lined hearths) and faunal remains were found in the
feature fill (context 314 and a pit intruding into this fill,
context 228). One radiocarbon date comes from a roe
deer skull with antlers from the upper portion of the fill
and it yielded the earliest currently available date from
the new excavations of the site: 7131 to 6823 CAL B.C.
(at 95% confidence) (OxA-21962). Another bone tool
from the fill of the feature is dated to 7028–6648 CAL
B.C. (OxA-24811) (at 95% confidence) (TABLE 1). One
bead made from the marine gastropod Columbella
rustica was found in the fill of this feature.
Human-Environment Interactions
Faunal remains
Faunal remains collected in the course of the 2006–
2009 campaigns comprise bones of mammals, birds,
tortoises, frogs and fish, as well as land and aquatic
mollusk shells (TABLE 2). Animal remains were
extremely fragmented, and relatively poorly pre-
served. The color of the remains is pale yellow and/
or whitish. Poor preservation is mostly due to
depositional factors, and to a much lesser degree to
weathering. Depositional factors that influenced the
preservation of animal remains include mechanical
damage on bones caused by abrasion with rock clasts
as well as corrosive action of plant roots and liquids
that circulated through the deposits. Taphonomic
changes include pitted bone surfaces or worm-like
traces that were made by plant roots. Yet, because of
careful recovery methods even fragile skeletal ele-
ments or milk teeth of dogs, red deer and wild swine
were collected. Few bone fragments were rounded by
water transport. Fragmentation and preservation
were also influenced by biogenic factors such as
trampling and gnawing. Gnawing traces were
recorded on 1.6% of mammal bones and they are
mostly consistent in their size and distribution,
indicating dogs were the main agents of bone
attrition. A high degree of fragmentation is reflected
in the distribution of the size classes of animal
remains: approximately 73% of mammal remains
have maximal lengths less than 2 cm. Less than 3% of
mammal remains are complete or almost complete
bones or teeth, and they are represented mostly by
short carpal and tarsal bones and isolated teeth.
The faunal composition (TABLE 2) is similar to the
one recorded for the large collection of bones that
came from the assemblage recovered in 1970–1971 at
Vlasac (Bo¨ko¨ nyi 1978) and from other Mesolithic
sites in the Danube Gorges region (cf. Bo¨ko¨ nyi 1970;
Boric
´and Dimitrijevic
´2005, 2007; Clason 1980;
Dimitrijevic
´2000, 2008). The major difference is that
dog remains are dominant in material from the 2006–
2009 excavations at Vlasac according to the number
of identified specimens (NISP), minimum number of
individuals (MNI) or number of diagnostic zones
(DZ). Yet, one should note the high occurrence of
dogs also in the southern downslope area of the site
excavated in 1970–1971. The reason for the highest
frequency of dog remains in the northern upslope
area excavated in 2006–2009 is possibly related to the
lack of red deer remains, primarily antler fragments
associated with manufacturing (see below), in this
likely peripheral zone of the settlement. The high
representation of red deer antler in the core zone
downslope is the consequence of more domestic
activities there.
The presence of dogs on the site is further inferred
on the basis of the high incidence of gnawing marks.
Moreover, gnawing is more frequent on dog bones
(15% of all dog bones) than on other mammal
remains (1.6% of all other remains have traces of
Boric
´et al. Late Mesolithic lifeways and deathways at Vlasac (Serbia)
Journal of Field Archaeology 2014 VOL.39 NO.1 11
gnawing). A similar pattern of gnawing marks is also
recorded for Lepenski Vir and Padina (Boric
´and
Dimitrijevic
´2005; Dimitrijevic
´2008), illustrating that
leftovers from human consumption of various other
mammalian species rarely reached dogs. This could
also suggest a particular treatment of dog bones as a
reflection of a specific cultural practice. Out of the
total number of dog remains, excluding isolated teeth
and short bones such as carpals, tarsals and
phalanges, which, almost never bear gnawing traces
due to their position in the skeleton and their small
size, one-third of all dog remains exhibit gnawing
marks.
The most frequent dog skeletal elements are cranial
parts, including mandibles and isolated teeth, and
bones of distal extremities. Although this is partly
due to their frequency in the skeleton, it seems that
this is also the consequence of the destruction by
Table 2 Frequencies of faunal remains from the 1970–1971 (after Bo¨ko¨ nyi 1978: table 2 supplemented by Boric
´2003b:
appendix 4) and 2006–2009 excavation seasons at Vlasac expressed by the number of identified specimens (NISP),
minimum number of individuals (MNI), diagnostic zones (DZ), and weight.
Species
NISP
(1970–1971)
NISP
(2006–2009)
MNI
(2006–2009)
DZ
(2006–2009)
Weight (g)
(2006–2009)
Mammals
Aurochs Bos primigenius 54 1 1 – 76
Bos sp. – 1 1 – 16
Dog Canis familiaris 1914 326 20 51 850
Wolf Canis lupus 103 4 1 – 9.2
Canis sp. – 6 2 2 25.8
Roe deer Capreolus capreolus 510 48 2 14 240.4
Beaver Castor fiber 71 6 1 1 6.6
Red deer Cervus elaphus 6732 199 6 20 3548.4
Chamois Rupicapra rupicapra 22 – – – –
Fallow deer Dama dama – 1 1 – 0.1
Hedgehog Erinaceus concolor 1222 1.2
Wild cat Felis silvestris 45 4 1 – 0.7
Hare Lepus europaeus 22 4 1 – 0.9
Lynx Lynx lynx 5 1 1 – 1.4
Pine marten Martes martes – 4 1 3 3.4
Marten Martes sp. 248 4 1 1 0.7
Badger Meles meles. 58 1 1 1 0.6
Weasel Mustelidae sp. 8 – – – –
Squirrel Sciurus vulgaris 5––– –
Wild swine Sus scrofa fer. 1175 77 3 12 487.7
Sus sp. – 6 1 – 7.3
European mole Talpa europea – 1 1 1 0.1
Brown bear Ursus arctos 169 6 1 1 54.6
Fox Vulpes vulpes 30 7 2 – 3
Small-sized carnivore 13 – – – –
Unidentified ? 10,675 – – 5794.7
Total for mammals 11,185 11,384 ––11,128.8
Reptiles
Pond tortoise Emys orbicularis 317 40 – – 29.8
Fish
Catfish Siluris glanis 2283 85 – – 114
Carp Cyprinus carpio 1552 – – – –
Cyprinidae sp. 5230 6566 – – 776.8
Beluga Huso huso 21 – – – –
Acipenseridae sp. 19 7 – – 0.5
Pike Essox lucius 11 1 – – –
Unidentified 8372 16,570 – – 1119.1
Total for fish 17,488 23,229 – – 2010.4
Birds
White pelican Pelecanus cf. onocrotalus 3––– –
Cormorant Phalacrocorax carbo 3––– –
Great white egret Egretta alba 2––– –
Mallard Anas platyrhynchos 2––– –
Teal Anas crecca 2––– –
Ferruginous (?) duck Aythya nyroca (?) 1 – – – –
Black kite Milvus migrans 5––– –
Imperial eagle Aquila heliaca 6––– –
White-tailed eagle Haliaee¨ tus albicilla 27 2 – – 3.8
Tawny owl Strix aluco 2––– –
Jay Garrulus glandarius 1––– –
Magpie Pica pica 1––– –
Raven Corvus corax 4––– –
Unidentified 87 36 – – 10
Subtotal for birds 146 38 ––13.8
Boric
´et al. Late Mesolithic lifeways and deathways at Vlasac (Serbia)
12 Journal of Field Archaeology 2014 VOL.39 NO.1
gnawing of the more nutritive axial elements and the
bones of upper extremities. Cut marks and traces of
burning were also present on dog remains indicating
their slaughter and/or defleshing by humans. Dogs
might have been used for their pelts as well (cf.,
Bonsall 2008). Isotopic values for dog remains from
both the old (Boric
´et al. 2004) and new (Becker 2010)
excavations at Vlasac show elevated d
15
N values over
10 %and up to 13.6 %, suggesting a high protein
intake that might have come from consuming fish
and from consuming the meat of other dogs.
Upper and lower jaws of dogs clearly show jaw
shortening and crowding of teeth rows caused by the
domestication process as previously suggested by
Bo¨ko¨ nyi (1975, 1978) for the collection he examined.
The same process must have caused anomalies in
teeth morphology and distribution (Dimitrijevic
´and
Vukovic
´2013). Apart from adult individuals, there
were remains of puppies. In one instance (context
306) upper and lower jaws with milk teeth of at least
two puppies were found together; they were probably
from the same litter.
On the basis of weight counts, red deer provided
the most substantial contribution to diet while the
bones of this animal were the most important raw
material for tool manufacturing. The most frequent
red deer skeletal elements were the cranial bones and
the lower limbs. From other mammal species, only
wild pig and roe deer are represented with somewhat
higher frequencies of specimens, while less than 2% of
the sample comprises all other species: aurochs,
various carnivores (brown bear, wolf, red fox, marten,
badger, wild cat), hare, and beaver. Butchering marks
were observed on a relatively small number of red
deer, roe deer, and wild swine bones.
Watersieving of all archaeological deposits and
flotation of a large number of soil samples enabled
the recovery of numerous small vertebrates and
invertebrates. There were 38 bird remains in the
sample but only two bird bones have positively been
identified at present and both come from a white-
tailed eagle, Haliaee¨tus albicilla; the bones were found
in the same unit (context 314) of the likely dwelling
feature (see above). In fact, most of the bird bones
from the 2006–2009 excavations at Vlasac come from
contexts relating to the two dwelling features. One
should note that the dominant bird species from the
1970–1971 excavations at Vlasac is also the white-
tailed eagle, followed by the imperial eagle, Aquila
heliaca (TABLE 2). One could assume that these birds,
which can still be found today in the cliffs above the
site of Vlasac (FIG. 3A), were hunted for their feathers
possibly used for body decoration.
Among fish remains cyprinids sp. (various species
of carp) are the most numerous along with the
presence of some catfish bones and scales of different
Acipenseridae (sturgeon) species. Such an assemblage
of fish bones corresponds with the one from the
1970–1971 excavations at Vlasac. Cyprinids bones
were very fragmented and included vertebrae and a
plentitude of pharyngeal teeth, perforated or unmo-
dified, which were often used as ornaments in burials
(TABLE 3).
Isotopic evidence for human diet
Four AMS-dated burials from the recent excavations
have yielded stable isotope evidence, showing ele-
vated d
15
N levels that range from 14.0 to 17.0 %
(TABLE 1). These high d
15
N values are in keeping with
the isotopic evidence from burials excavated at
Vlasac in 1970–1971 and in the region as a whole,
indicating the intake of aquatic resources. The fact
that the diet was based on fish is a characteristic of
the Late Mesolithic in particular (Bonsall et al. 1997;
Boric
´and Miracle 2004; Boric
´et al. 2004; Cook et al.
2009; Nehlich et al. 2010). While at present we have
obtained isotopic ratios for only one individual (H63)
dated to the Mesolithic-Neolithic transition phase at
Vlasac, this individual exhibits one of the highest
d
15
N values (17.0%)(TABLE 1) known for the Danube
Gorges region, suggesting that during that period
there was no change at Vlasac in subsistence
practices, with fish as a staple. In combination with
the faunal evidence (see above), this pattern would
suggest concentrated and predictable resources,
implying residential stability at Vlasac (Heffley
1981; Radovanovic
´1996).
Macrobotanical remains
All excavated deposits were sampled for flotation or
watersieving. The selection of samples for flotation
was based on context types and on the overall
quantity of charred material visible in the sediments
in the field. Twenty L of soil were collected wherever
possible (except in cases when the deposits seemed
particularly rich and more was taken).
Water from the Danube was used for the recovery
of plant remains. Flotation was the most effective
method for separating residue that floats (charred
plant remains, light bone fragments, and small
mollusks) from residue that sinks in water (stone,
bone). The flotation machine on the riverbank at
Vlasac was a converted plastic wheeled bin; it
consisted of a plastic tank of ca. 240 L in capacity,
with water pipes at the bottom and a removable 2 mm
mesh in the upper half for collecting heavy residue.
The water outlet at the top was used for collecting light
residue (in 0.3 mm mesh).
In total, 74 samples from 40 units excavated in
2006–2008 were floated, with soil volume ranging from
less than 1 L up to 35 L. Of these, 65 samples from 38
units were available for analysis; all light fractions plus
17 heavy fractions were sorted and analyzed. The light
Boric
´et al. Late Mesolithic lifeways and deathways at Vlasac (Serbia)
Journal of Field Archaeology 2014 VOL.39 NO.1 13
fractions were sieved through 2 mm and 0.3 mm mesh;
the 2 mm fraction was entirely sorted for charred
wood fragments and other plant remains. In most
cases, 100% of the 0.3 mm fraction was sorted for
plant remains other than charred wood (a few 0.3 mm
fractions were subsampled to not less than one-quarter
using a sample splitter). Heavy fractions were sieved
for 4 mm, 1 mm and 0.5 mm particle sizes; the 4 mm
and 2 mm fractions were entirely sorted for plant and
other remains (faunal, beads, stone objects); the
0.5 mm fraction was randomly subsampled (to not
less than one-fifth; sample splitter was not used) and
sorted for plant remains and diagnostic bone frag-
ments (and small finds, if any).
In the analyzed samples, a total of 58 identifiable
macroscopic (non-wood) botanical remains were
encountered and ca. 5 ml of parenchyma (root/tuber
tissue)/nut kernel fragments. In terms of the diversity
(i.e., number of identified taxa), density (number of
items per L of soil), ubiquity (number of samples in
which a taxon occurs) and abundance, the Vlasac
assemblage appears relatively poor compared to, for
example, the Franchthi Mesolithic dataset (Hansen
1991). It is, however, consistent with the record from
the contemporaneous Schela Cladovei site in the
Danube Gorges area where, in Mesolithic contexts,
only minute wood charcoal fragments and a few
fragments of possible parenchyma were discovered
Table 3 Summary of primary and secondary inhumations and cremations discovered in the course of new excavations
at Vlasac and ornamental beads found in these burials. Scattered human remains found in various settlement contexts
are not listed in the table.
Context/
feature
numbers Burial type Sex Age
Carp teeth
ornaments*
Cyclope
neritea
ornaments
Spondylus
beads
Discoid
limestone
beads
H2 Primary Female Ca. 40 y 310z255z77
5642
32 – –
H53 Primary Female Ca. 50 y 5z8z3516 –7{21{
H60 (z59) Primary/secondary/
cremation
(973 burnt bone frags.
weighing 1766 g)
Male? 14–16 y 1z3z155–331
H62 Primary ? 38–40 weeks – – 2{5{
H63 Primary/secondary Female 25–30 y 9z113z565178 –24
H69 Primary ? 36–38 weeks – – – –
H81 Primary/secondary Male Ca. 40 y 1 – – –
H136 Primary/secondary Female 40–60 y – – – 10
H153/21 Primary/secondary ? 2–3 y – – – –
H232 Primary Female Ca. 25 y 17z9z1527 –––
H244 Primary/secondary/
cremation
Female 30–40 y – 1 – –
H254 Primary Male 30–50 y 1 – – –
H267 Primary Female Ca. 50 y 15z161z755251 –––
H297 Primary ? 1 y 252z294z155
5701
22 – –
H317 Primary Female Ca. 30 y – – – –
H326 Primary Male Ca. 50 y 1 1 – –
54 Cremation (34
burnt bone
frags. weighing 14.1 g)
na Adult (?) 6z3z4513 –––
87 Cremation (8
burnt bone
frags. weighing 1.8 g)
na Adult (?) 4z22z3529 – – –
96, 97 Cremation (4
burnt bone
frags. weighing 5.4 g)
na Adult (?) – – – –
110, 115 Cremation (1052 burnt
bone frags. weighing
769 g)
na Adult (?) 8z26z21555 1{––
146 Cremation (40
burnt bone
frags. weighing 22.4 g)
na Adult (?) – – – –
242, 261
(Feature 23)
Cremation (468
burnt bone
frags. weighing 158.4 g)
na Adult (?) 1 10{––
260, 249
(Feature 26)
Cremation (520
burnt bone
frags. weighing 598 g)
na Adult (?) – – – –
*The figures show the number of specimens that are: entire and unmodified zentirely perforated zfragmented5totals.
{While the beads were found in the fill of these burials, it is likely that they are in their secondary positions as a consequence of the
disturbance of earlier burials by later interments.
{Burnt specimens.
Boric
´et al. Late Mesolithic lifeways and deathways at Vlasac (Serbia)
14 Journal of Field Archaeology 2014 VOL.39 NO.1
(Mason et al. 1996). At Vlasac, the only taxon that
stands out in terms of quantity and frequency is
Cornelian cherry, Cornus mas L., of which complete
and fragmented pits were discovered. Other taxa were
represented in very small numbers, the majority by
just one specimen per sample. Amorphous fragments
of plant material that could not be identified to
species/genus included parenchyma and nut kernels,
and some unknown fragments.
An MNI total of 38 Cornus mas pits (complete and
fragmented) was encountered in 14 (out of 65)
samples from 11 (out of 38) excavation units; 26
(MNI) pits were discovered in three samples from a
single unit-cremation pit, Feature 26 (context 249)
(see below). In five other cases, Cornus mas remains
(MNI56) were also derived from burial contexts,
often with evidence of burning (contexts 17, 19, 24,
and 146, all in Trench 3/2006).
The study of charcoal remains has yielded 20
different taxa distributed over 40 units. The most
significant taxa according to their relative value in the
entire assemblage are Cornus,Quercus sp. deciduous,
and Prunus. These taxa have values between 20–30%
of the record. There is a group of taxa including
Corylus,Cotinus,Fraxinus, and Maloideae represent-
ing between 2–5%. The rest of the taxa have values
under 1%. There are also undetermined angiosperm
and other undetermined fragments representing very
low values. In the cremation pit, Feature 26 (context
249), there are 5 taxa with Prunus and Quercus ssp.
showing the highest values. In this unit the absence of
Cornus charcoal remains is remarkable, considering
that this context is rich in Cornus mas pits. The burial
units have yielded 196 charcoal fragments with 12
different taxa identified. Concerning the relative
values, the most significant taxon is Cornus. Other
important taxa are Quercus ssp. deciduous, Corylus,
and Cotinus. The rest of the taxa are represented in
low frequencies (Allue´ et al. in press; Filipovic
´et al.
2010).
In sum, the archaeobotanical record of Vlasac
allows us to describe plant communities in the
Danube Gorges region. The identified taxa mostly
indicate forest plant formation with an important
shrubby vegetal cover. Oaks might have formed the
main forest and most of the other taxa might have
formed understory or shrubby vegetation at the
forest edges or on the riverbanks. The Vlasac record
shows the significance of Cornus mas L.It is possible
to relate archaeobotanical remains to firewood in
some ritual practices but this does not necessarily
suggest symbolic preferences for particular species.
Based on the association of Cornelian cherry with
burial deposits (i.e., in burial fill), it is tempting to
suggest that this species had a symbolic meaning for
the Vlasac community and was somehow linked to
their burial rites. A relatively large number of Cornus
mas pits from Feature 26 (context 249) seem to be the
best indication of such an intentional symbolic use of
the Cornelian cherry fruits, since the complete
absence of Cornus mas charcoal (despite a high
frequency of ca. 70 ml of charcoal remains, which is
much more than any of the other analyzed samples)
from this unit seems to exclude the accidental
association of cherries with the branches of burnt
firewood used for keeping the cremation fire. We can
only speculate about the meaning of the Cornelian
cherries’ association with this context. Here we
mention only a few possibilities: they might have
represented remnants from a feast for the dead; the
fruits might have been of symbolic importance due to
their red color, perhaps analogous to ochre and other
red minerals in meaning (cf., Boric
´2002b); and/or the
Cornelian cherry might have been the type of food
discarded in the course of the burial ceremony (as a
sign of the social rejection of the deceased). Cornelian
cherries become ripe in September (Janc
ˇic
´1990: 36)
and we may suggest that the cremation and probably
also the burial event took place in the fall. However,
if meant for consumption, cherries could have been
dried and stored for several seasons (Wiltshire 1995:
385). Palynological analysis of human coprolites
discovered in the 1970–1971 excavations at Vlasac
(Caˆrciumaru 1978) showed the presence of pollen
grains from species with potentially edible parts (e.g.,
Pinus, Quercus, Juglans, Corylus). This could suggest
that these species were particularly selected for food;
if consumed, they would have provided valuable
vitamins and minerals for the diet, complementing
the nutrients available from fish and meat.
Technology
Chipped stone
The chipped stone industry from the 2006–2009
excavations at Vlasac comprises 503 pieces of which
315 (62.6%) are flint, 181 (36%) are quartz and 7
(1.4%) are quartzite. Considering the area excavated,
this is a very modest sample in comparison to the
amount of material found in the course of the 1970–
1971 excavations at the site, which included 31,225
chipped stone artifacts with 7250 pieces of flint
(23.2%), 19,092 pieces of quartz (61.1%), and 243
pieces of quartzite (0.8%) (Kozłowski and Kozłowski
1982; Srejovic
´and Letica 1978). This is despite the fact
that the area excavated in 2006–2009 represented half
of the area excavated during the first excavations and
that a more meticulous recovery methodology char-
acterized the new excavations. Hence the observed
difference must be related to functional characteristics
of the excavated settlement space, i.e., the peripheral
nature of the southern upslope area excavated in 2006–
2009 in comparison to the northern downslope core
Boric
´et al. Late Mesolithic lifeways and deathways at Vlasac (Serbia)
Journal of Field Archaeology 2014 VOL.39 NO.1 15
area excavated in 1970–1971. This conclusion is sup-
ported by the absence of rectangular hearths as
indicators of domestic activities in the more recent
excavations along with the fact that most of the
recovered artifacts in general were concentrated in
those zones where the supposed dwelling features were
found.
A relatively large proportion of quartz in the
assemblage is characteristic of Late Mesolithic
chipped stone industries in the Danube Gorges region
as a whole (Radovanovic
´1996), suggesting a
predominant reliance on locally available lithic raw
materials. This conclusion is strengthened by the
predominance of gray non-transparent flint with poor
technological properties corresponding to type A1
identified by Kozłowski and Kozłowski (1982). This
type of flint is readily available in the immediate
vicinity of the site in the Jurassic geology of the
region. Some other raw materials are found too, such
as beige, non-transparent flint (type A2), gray white
spotted flint (type A8), and occasional radiolarites.
Finally, in both Late Mesolithic and transitional
levels as well as in units with the first Early Neolithic
pottery (see below), there were also several specimens
of the so called ‘‘Balkan’’ yellow white-spotted flint
(type A11), originating several hundred kilometers
away from the Danube Gorges, in northern Bulgaria
(Biagi and Starnini 2010; Gurova 2012).
The Mesolithic technological chaıˆne o pe´ratoire
indicates the use of the splinter technique and bipolar
cores, which might have been an adjustment to the
properties of the raw material used, with the outcome
of a predominantly flake-based industry. The follow-
ing techno-typological categories are present: small
single platform cores for bladelets, retouched blades
made on regular blanks, some with denticulated
retouch, truncations, irregular scrapers, chisel-like
tools, splintered pieces, perforators and microliths,
such as backed pieces, micro-retouched bladelets, and
two trapezes (FIG. 7). The two trapezes are restricted
to Late Mesolithic dwelling contexts. A similar range
of tools was reported for the analyzed assemblage
from the 1970–1971 excavations at Vlasac (Kozłowski
and Kozłowski 1982: plates IX, XXXIII, XXXV). The
presence of all production stages in the assemblage and
the prevalence of expedient over curated artifacts
would suggest the role of Vlasac as a long term base
camp during the Late Mesolithic. A sample of 34
artifacts (8 flakes, 14 blades, 4 chips, and 8 formal tools)
were analyzed for use-wear patterns. The analysis
revealed 9 pieces with use-wear traces from working on
various materials: wood, bone, hide, and meat.
While this Late Mesolithic industry is dominated
by flakes as the consequence of selecting locally
available raw materials which exhibit poor technolo-
gical characteristics, the technological basis of this
industry is in the Balkan Epipalaeolithic or Epigra-
vettian tradition, with its roots in the industry
documented at the rockshelter site of Cuina Tur-
cului (FIG. 1), found on the northern bank of the
Danube downstream from Vlasac. The blade techni-
que, which is dominant in the Cuina Turcului
Epipalaeolithic industry (Pa˘ unescu 1970), is clearly
present at Vlasac, and when used with good quality
flint it allowed the production of regular blades and
trapezes. The preference for locally available materi-
als throughout the Mesolithic at Vlasac and the
Danube Gorges area in general led Kozłowski and
Kozłowski to suggest that this was the consequence
of ‘‘the increasing forestation which blocked the
access to some primary deposits, and … the increas-
ing isolation of human groups in the Early Holocene’’
(Kozłowski and Kozłowski 1982: 100). Yet, occa-
sional pieces of good quality non-local flint, as well as
marine gastropods, such as Columbella rustica and
Cyclope neritea, the closest source for which was at
least 400 km away, were found in the Late Mesolithic
deposits at Vlasac (Cristiani and Boric
´2012).
Moreover, there are striking similarities in both
mortuary rites and the types of body ornaments used
in Mesolithic levels at Vlasac and at Franchthi Cave
in Greece (Cullen 1995; Perle`s and Vanhaeren 2010)
(see below). This would suggest that at the very least
the community at Vlasac and other Late Mesolithic
sites in the Danube Gorges region must have been
part of trade and information networks operating
across the wider region. Hence the argument that
locally available stone raw materials were preferred
throughout the Mesolithic due to the isolation of
human groups in this region must be carefully
reexamined. The key to solving this question would
be to better understand correlations between chang-
ing patterns of the availability of lithic raw materials
and of ‘‘exotic’’ decorative items (Whallon 2006).
Ground stone tools
There are 70 pieces of ground stone from the most
recent excavations of Vlasac in contrast to 131
specimens from the 1970–1971 excavations at the site
(Srejovic
´and Letica 1978: 98–103). These artifacts
were largely made from sandstone boulders (over
50%) and amphibolites (ca. 25%) with several speci-
mens made from aplite, micaschist and chert. The
raw material structure of the assemblage corresponds
to those at other sites in the region, such as the
neighboring site of Lepenski Vir (Antonovic
´2006:
19); the primary source of the sandstone boulders is
in the vicinity of both sites, in the upper reaches of
the Boljetinska River.
Twenty-four specimens had clear traces of use and
fall into the following categories: fish stunners, small
anvils, or hammerstones. The latter were used on one
Boric
´et al. Late Mesolithic lifeways and deathways at Vlasac (Serbia)
16 Journal of Field Archaeology 2014 VOL.39 NO.1
Figure 7 A selection of chipped stone artifacts, some with their use-wear patterns, found at Vlasac in 2006–2009. Provenance
information (context no., x-find no., square no.) of each artifact is indicated. 1) Blade (45.61, sq. 28); 2) Flake (35. 610); 3) Blade
(16. 61, sq. 8); 4) Blade (115. 610); 5) Pointed retouched flake/perforator (40.613, sq. 19); 6) Chip (24); 7) Trapeze (36); 8)
Trapeze (145, sq. 7); 9) Blade (288. 63); 10) Truncation (unstratified); 11) Blade (293.61); 12) Blade (217.61); 13) Blade
(228.683); 14) Blade (309.61); 15) Bladelet (237.61); 16) Blade (251); 17) Blade (308); 18) Backed piece (228.6182); 19)
Retouched flake (228.6143). Drawn by Maria Gurova.
Boric
´et al. Late Mesolithic lifeways and deathways at Vlasac (Serbia)
Journal of Field Archaeology 2014 VOL.39 NO.1 17
or several sides (FIG. 8: 2,3) over a considerable period
of time most often for flintknapping, or to make
borers and shaft-straighteners, which were possibly
used for bone tool manufacturing or bead making.
One large river pebble found in context 235, a layer
above burial H244, has the shape of a massive axe
and has traces of use as a polisher on one face
(FIG. 8:1). Anvils are the most frequent category of
ground stone tools found in both the old and new
excavations at the site.
There are no elements in the assemblage of ground
stone tools from the new excavations at Vlasac with
Neolithic morphologies, such as polished shoe-last
stone adzes and axes like those found associated with
the Mesolithic-Neolithic transition phase at Lepenski
Vir (Antonovic
´2006). At least one small polished
shoe-last stone adze, associated with the Neolithic
occupation at Vlasac, was found during the 1970–
1971 excavations at the site, however. Residue
analyses on two hammerstones from the new
excavations at Vlasac indicate the presence of starch,
suggesting that those specimens were used for
crushing nut kernels; further analyses on those and
other specimens are in progress (Huw Barton,
personal communication 2009). This evidence con-
tributes to our understanding of plant processing at
Vlasac, suggesting the importance of starchy foods
despite the paucity of macrobotanical remains.
Antler, bone, and ivory tools
The whole osseous assemblage is comprised of bone
(N528), antler (N547), and ivory (N515) artifacts.
Pointed and edged tools and artifacts with lateral
cutting edges are the main morphological categories
(FIG. 9). Manufacturing debitage has also been found.
The analysis has revealed the existence of a wide
ranging repertoire of techniques for bone, antler and
ivory tool manufacturing. The relationship between
artifacts and raw materials is evident in the selection
of specific bones for the production of blanks for
specific tools.
The majority of pointed tools (awls and curated
points) and edged bone tools (straight wedges on
bone splinters) were manufactured using mainly
metapodials of red deer, Cervus elaphus. The main
technique for the processing of metapodials was
indirect percussion followed by retouch of the
obtained blanks (‘‘shaft-wedge-splinter technique
followed by counterblow retouch’’ [David 2003,
2009]). Once obtained, the blanks were further refined
by scraping and/or abrasion. For the manufacture of
projectile points, longitudinal grooving was used. The
proximal parts of the points were finished by flint
shaving, in order to produce a stem, or by transverse
abrasion, to produce tapered bases.
Lateral cutting edged tools (knives and gorges) were
mostly manufactured using wild boar, Sus scrofa lower
canines/tusks. The canines were processed by indirect
percussion for separating the main blades, while active
edges were shaped by scraping with flint, and proximal
parts were shaped by abrasion.
Red deer antler beams were the main raw material
used for the production of numerous types of edged
tools, such as blade axes, axes, and short and long
intermediate pieces. Tines were utilized as chisels
(FIG. 9: 6). Antler processing is characterized by
techniques different from those used for manufactur-
ing bone and ivory. The beam was at first sawed and
then snapped or was partitioned by indirect percus-
sion followed by snapping. The separation of the
tines from the main antler beam was carried out by
means of nicking and subsequent snapping. The
definition of the functional area was carried out by
means of scraping with a knapped stone tool as well
as by chopping and abrasion.
Based on tool morphologies and use-wear traces,
bone and antler edged tools were used for some heavy
duty processing of hard materials such as wood; they
were likely used as axes and wedges (FIG. 9: 4). Traces
of use on some ivory burins and knives (FIG. 9: 8–10)
indicate that these categories of tools were likely used
for woodworking. Bone projectile points were used as
hunting and fishing gear, but they might also have
been used for tribal or interpersonal combat, as
suggested by a bone point embedded in the pelvic
Figure 8 Examples of ground stone tools found at Vlasac in
2006–2009. Provenance information (context no., x-find no.)
of each artifact is indicated. 1) Massive ‘‘blunt axe’’ of
amphibolite (235.61); 2) Sandstone anvil used on several
sides (304.65); 3) Sandstone anvil-hammerstone used on
several sides (320.67). Photographs by Dragana Antonovic
´.
Boric
´et al. Late Mesolithic lifeways and deathways at Vlasac (Serbia)
18 Journal of Field Archaeology 2014 VOL.39 NO.1
bone of a skeleton from the 1970–1971 excavations at
Vlasac (Roksandic 2004). There were similarly
embedded bone points in skeletons excavated at the
contemporaneous Late Mesolithic site of Schela
Cladovei found 80 km downstream from Vlasac
(Bonsall 2008; Boroneant¸ and Nicolaescu-Plopsor
1990).
In sum, the sample of osseous artifacts coming
from the most recent excavations at the site is
significantly smaller than the preserved assemblage
of osseous tools coming from the area excavated in
1970–1971 where close to 3000 were discovered
(Cristiani and Boric
´in press; Srejovic
´and Letica
1978). The reason for this discrepancy relates to
differences in the character of the settlement in each
of the excavated zones. Yet, the repertoire of artifacts
recovered in the two zones of the settlement is
uniform and is characteristic of the typical European
Mesolithic technological chaıˆne ope´ratoire with a
range of morphologies for foraging activities as well
as intensive wood processing. Such a repertoire also
characterizes other Late Mesolithic sites in the
Danube Gorges region (e.g., Beldiman 2005; Dinu
et al. 2007). The abundant traces of woodworking are
most likely indicative of the long term occupation of
the site in the Late Mesolithic. Woodworking with a
range of tools might be related to the construction of
dwellings, and also possibly to make canoes for water
transportation. As mentioned above, there is clear
evidence of the clearing of vegetation from the slope
above the settlement, which caused the intensification
of hillwash episodes in the Late Mesolithic. Contrary
to some other elements of material culture at this site,
the osseous industry from Vlasac does not exhibit any
Figure 9 A range of typical osseous tools found at Vlasac in 2006–2009. Provenance information (context no., x-find no.) of
each artifact is indicated. 1) Edged tool–chisel (314.654 directly dated with OxA-24811); 2) Pointed tool–awl (215.61); 3)
Pointed tool–awl (282.66 directly dated with OxA-24810); 4) Edged tool–wedge (214.62); 5) Edged tool–chisel (320.622); 6)
Edged tool–chisel (320.617); 7) Short intermediate piece (314.621); 8) Cutting edged tool–knife (314.663); 9) Cutting edged
tool–gorge (214); 10) Cutting edged tool–knife (228.64). Photographs by Dusˇ an Boric
´and Emanuela Cristiani.
Boric
´et al. Late Mesolithic lifeways and deathways at Vlasac (Serbia)
Journal of Field Archaeology 2014 VOL.39 NO.1 19
signs of change related to the introduction of a new
technological chaıˆne ope´ratoire or new morphologies
during the Mesolithic-Neolithic transition. Yet, this
period saw the introduction of new tool shapes and
techniques in the osseous industry during Phase I-II
at the neighboring site of Lepenski Vir (Boric
´and
Cristiani in press). The likely reason for the lack of
change at Vlasac lies in the changing nature of the site
at the end of the 7th millennium B.C., at which time
the site started to be primarily used as a burial
ground (see below).
Death and Body Decoration
In the course of the 1970–1971 excavations, 87 graves
containing either 119 individuals (Nemeske´ri 1978) or
164 individuals (Roksandic
´1999, 2000) were exca-
vated at Vlasac. Our new excavations furnished
evidence of at least 16 individuals from both primary
and secondary inhumations along with at least seven
clearly defined cremation burials (TABLE 3). In six
instances we found primary inhumations that could
be associated with some form of secondary burial
practice, including skull removals and disturbance of
parts of primary inhumations by placing new
interments in the same locations. This resulted in
the displacement of cranial and postcranial skeletal
elements of older burials; the disturbed bones of older
burials were sometimes burnt either in situ or in
nearby oval pits (Boric
´2010; Boric
´et al. 2009). In one
instance, there is evidence that the removed skull was
reburied in a structured manner (a child skull from
burial H21 [FIG. 10] probably related to articulated
burial H153). In several instances disturbed cranial or
postcranial bones were placed within the same
location either in the fill of a new interment in no
particular order, or were piled on the side of the same
burial place.
There were two main zones of human burials in the
area excavated at Vlasac from 2006 to 2009. The first
zone is characterized by exclusively Late Mesolithic,
mid-7th millennium B.C. burials in Trenches 3/2007
and 1/2008 in the western part of the excavated area
(FIG. 4). Five primary burials (H244, H254, H267,
H317, and H326) were distributed over this area
largely following the same position and orientation,
that is, they were placed parallel to the Danube with
their heads pointing downstream. In addition, burial
H2 was found in this general area in front of the
eroded riverbank section at the level of the beach
gravel (Boric
´2006). This burial was, at the time of the
discovery, partly damaged and exposed by river
erosion. It dates to 6775–6470 CAL B.C. (OxA-16541)
(at 95% confidence) after the correction for the
aquatic reservoir effect (TABLE 1). The burial included
642 perforated, unmodified, and fragmented phar-
yngeal carp teeth once attached to some sort of cloak
placed on the back of the deceased. There were also
32 Cyclope neritea marine gastropod beads, 15 of
which were part of a closely knit line of beads found
lying beneath the upper femurs of burial H2 (Cristiani
and Boric
´2012). There was also a large stone block
placed over the lower legs of this individual. The
dating of this burial, the range and quantity of
ornaments and the pattern of their distribution in the
burial as well as the placement of a large stone block
over the lower legs are all strikingly similar to child
burial H297 found in the second burial concentration
in Trench 3/2006 (FIG. 10). These suggest standardized
Late Mesolithic burial customs.
While the other five burials in this part of the site
have not been dated directly, their general orientation
and positions (in some cases with stone construc-
tions) as well as their range of ornaments suggest that
they might also be dated to the Late Mesolithic of the
7th millennium B.C. For instance, a concentration of
251 carp teeth ornaments was found in burial H267
(FIG. 11). In nearby burial H244, the torso and
mandible were disturbed by a later intrusion, with
possible burning of the disturbed bones, which left
only one fragment of a carbonized right humerus in
situ, while most of the other disturbed and likely
burnt bones were removed. Possibly connected with
this exhumation and burning is a cremation pit,
Feature 23 (context 242), (TABLE 3) found in the
vicinity (FIG. 4). Apart from burnt bones found in this
cremation pit, there were 10 carbonized Cyclope
neritea beads, suggesting that the fragments of burnt
bone and ornaments relate to the disturbed torso of
burial H244 (Boric
´et al. 2009). A few Late Mesolithic
cremations also contained broken and burnt projec-
tile points, which might have been comingled and
burnt intentionally with disarticulated and fragmen-
ted pieces of human bone. Cremation associated with
the practice of disturbing parts of older burials is
what characterizes mortuary rites at Vlasac during
the Late Mesolithic (Boric
´et al. 2009). It seems that
this practice also remained vital and relevant for the
inhabitants of Vlasac throughout the Mesolithic-
Neolithic transition as suggested by similar cremation
events in the burial sequence in Trench 3/2006 (see
below).
The second zone of burials at Vlasac is found in
Trench 3/2006. There burials from the Mesolithic-
Neolithic transition are superimposed on Late
Mesolithic ones (FIG. 10). This represents the first
unequivocal evidence for the existence of this transi-
tional phase at Vlasac, significantly contributing to
our understanding of the influence of the earliest
farming and stock breeding Neolithic groups from
the surrounding areas of the Balkans of indigenous
foragers at the end of the 7th millennium B.C. The
sequence of interments in this zone started upon the
Boric
´et al. Late Mesolithic lifeways and deathways at Vlasac (Serbia)
20 Journal of Field Archaeology 2014 VOL.39 NO.1
abandonment of a dwelling. It seems that this
location was first used in the first centuries of the
7th millennium B.C., followed by the construction of a
floor area. This floor area was abandoned by
depositing a layer of sterile soil over it. Among the
first burials found on the same level in this location
were adult burial H136 and child burial H297
(FIG. 10), with the individual in burial H136 dated to
6774–6472 CAL B.C. (OxA-18865) (at 95% confidence)
after the correction for the aquatic reservoir effect
(TABLE 1). Undisturbed one year old child burial
H297 was associated with 701 perforated, unmodified
and fragmented pieces of carp (Cyprinidae sp.)
pharyngeal teeth and 22 beads made of Cyclope
neritea marine gastropod shells, most likely originally
on some kind of burial cloak (Cristiani and Boric
´
2012). Also in child burial H297, there was a large
intentionally fashioned stone block placed over the
lower legs of this individual, similar to the one found
in burial H2 in the burial zone in the western part of
the excavated area (see above).
Probably somewhat later was an adult inhumation,
H232, found in the southern part of the burial area in
Trench 3/2006 (FIGS. 10,12A). In contrast to burial
H297, only 27 carp teeth ornaments were found in
this inhumation (TABLE 3). The date of 6636–6476
CAL B.C. (OxA-20702) (at 95% confidence) (TABLE 1)
provides a terminus post quem for this inhumation.
The AMS sample was from a charred cornelian
cherry, Cornus mas, pit (Allue´ et al. in press; Filipovic
´
et al. 2010) found in the fill of the cremation pit,
Feature 26 (context 249), which was directly beneath
burial H232 and which contained large amounts of
charcoal, burnt human bones, and bone projectile
points (FIG. 12B). The cremation event and the
interment of H232 might have been related, occurring
one after the other. Some of these burnt human bones
might have come from disturbed burial H136 (Boric
´
et al. 2009). A large part of H136 might have ended
up in one or more cremation pits made at this
location; only the feet and partly preserved lower legs
of this individual were found in primary articulation.
This disturbance of older burials and the subsequent
fragmentation and burning of bones seems to have
been a recurrent practice at Vlasac as stratigraphi-
cally later burials in this location exhibit the same
pattern of manipulation and secondary deposition of
bones from earlier burials (Boric
´2010). The described
cremation event(s) happened prior to the interment of
burial H81, dated to 6639–6440 CAL B.C. (OxA-20762)
(at 95% confidence) (TABLE 1). Primary inhumation
burial H81 is found slightly displaced to the south
and the west of H136 and at a higher level, but along
the same axis and with the same orientation and
position as H136 (FIG. 10). This more complete
Figure 11 A) Extended burial H267; B) Numerous carp
(Cyprinidae sp.) pharyngeal teeth ornaments found primarily
in the chest area of this individual.
Figure 12 A) Extended burial H232 placed above a crema-
tion pit (Feature 26); B) cremation pit containing charred
human bones, bone projectile points, and numerous
Cornelian cherry Cornus mas pits. One of these pits was
AMS-dated to 6636–6476 CAL B.C. (OxA-20702) (at 95%
confidence).
Boric
´et al. Late Mesolithic lifeways and deathways at Vlasac (Serbia)
22 Journal of Field Archaeology 2014 VOL.39 NO.1
inhumation was clearly disturbed by the placement of
primary burial H63, which was placed along the same
axis following the same position and orientation of
burial H81. Furthermore, in the fill of H63 one finds
several disarticulated bones of H81 placed alongside
burial H63 (FIG. 10). However, on the basis of the
radiocarbon date for burial H81 and the date of
burial H63 with the range 6232–6018 CAL B.C. (OxA-
16542) (at 95% confidence) (TABLE 1), it seems that
these two interment events are at least three centuries
apart. Disarticulated and fragmented bones of Late
Mesolithic burial H81 were placed along the legs and
in the fill of burial H63 which was later.
Such continuity in the use of the same location
and characterized by the same mortuary practice
(extended burial, oriented parallel to the Danube and
with the head pointing downstream) is typical of the
Late Mesolithic mortuary canon, which also refer-
ences the position of older burials, suggesting the
longevity of social memory, possibly related to claims
made by particular lineages to older burials. It also
indicates the stability of mortuary rites over long
periods of time at the site and across the Danube
Gorges region as a whole in the course of the
Mesolithic, or at the very least an intentional attempt
to reference ‘‘the old ways’’ (Boric
´2003a, 2010; Boric
´
et al. 2009). On the other hand, there are also clear
deviations from the typical Late Mesolithic mortuary
practice starting at the level of burial H63. These
changes mainly relate to burial furnishings and body
decoration. For instance, two fashioned flat stone
plaques were found placed over the lower legs of this
individual (FIG. 10); the lower legs remained undis-
turbed by later exhumation events and interments. It
must be said that this practice of covering the lower
legs of the deceased with stone has its roots in the
Late Mesolithic at Vlasac where one finds several
burials (H2, H297 and H326) (Cristiani and Boric
´
2012) with large, mainly unmodified stones placed
over the legs of these individuals. Yet, in the case of
burial H63 and later burials in this location,
fashioned stone plaques rather than unmodified large
stones were used for covering parts of the body of the
deceased upon burial. It has been suggested that
the practice of placing heavy stones over the legs of
the deceased might have been related to attempts to
restrict the movement of the dead body and its
possible intentions to harm. Other practices evi-
denced at Vlasac including wrapping/tying the body,
and subsequently exhuming, fragmenting, and burn-
ing the remains of the dead, might have been
motivated by the same concerns (Boric
´et al. 2009).
The clearest indications of both continuities and
dramatic changes starting at this time, ca. 6200 CAL
B.C., in the Danube Gorges region come from the style
of body decoration. Associated with the individual in
burial H63, on both sides of the neck and below the
shoulder blades were 178 carp teeth ornaments
(FIG. 13), most of which were perforated in order to
be sewed onto a cloak or headdress. As previously
mentioned, this type of ornamentation was abun-
dantly used during the course of the Late Mesolithic
at Vlasac and at Schela Cladovei (Bonsall 2008;
Cristiani and Boric
´2012). Yet, with burial H63, for
the first time comes evidence of the use of these
ornaments in relation to the head of the deceased.
Since no major disturbance characterizes these
ornaments and the neck bones, it is possible that
the head was removed upon the decomposition of
soft tissues. On the left side of the neck of the
individual in burial H63, within the concentration of
carp teeth ornaments, there was an ovoid shaped
Spondylus bead (FIGS. 13B,14: 7). This bead, clearly
associated with the body decoration worn at the time
of the burial, is possibly the earliest securely dated
item made from Spondylus in southeastern Europe.
Moreover, not only is the new exotic marine material
(completely replacing previously used exotic materi-
als such as Cyclope neritea or Columbella rustica,
marine gastropods) used for the production of this
bead, but it is also the morphology of the bead and
other similar beads (FIG. 14: 7–9) that signifies foreign
Figure 13 A) Perforated carp teeth applique´s found on both
sides of the neck and beneath the shoulders of the headless
adult female in burial H63; B) In situ Spondylus bead (larger
than carp teeth beads) found next to the neck.
Boric
´et al. Late Mesolithic lifeways and deathways at Vlasac (Serbia)
Journal of Field Archaeology 2014 VOL.39 NO.1 23
cultural influence. Hence, the combination of differ-
ent cultural elements within the same burial feature
illustrates the contact between the foragers in the
Danube Gorges area with the expanding network of
Early Neolithic communities in the north central
Balkans during the last two centuries of the 7th
millennium B.C.
There were several other inhumations that prob-
ably soon followed the interment of the individual in
H63 in the same location. Two juvenile/subadult
individuals (H60 and H153) (TABLE 3) were placed
directly atop H63. Further, digging into the same
location at a later date and removing the left pelvis,
femur and forearm of H63, which were all kept for
later use, and damaging the articulated limbs of the
2–3 year old child in burial H153 (FIG. 15) placed over
H63, two neonates (H62 and H69) were interred one
on top of the other, possibly as part of the same
burial event. At some point, the skulls of individuals
in burials H60, H63, and H153 were removed. While
the skull of the child in burial H153 was kept for later
use, skulls from H60 and H63 and a large number of
the postcranial bones from H60 were burnt atop
these burials, creating a cremation zone with in situ
fire that also partly affected the undisturbed upper
torso of the juvenile in burial H60, which was lying
directly atop the torso of the adult in burial H63. The
burning of likely ‘‘dry’’ bones of these individuals was
followed (perhaps as part of the same burial
ceremony) by the interment of an old woman (burial
H53), the last inhumation in this burial place. While
in the same position as the other burials and
paralleling the Danube, H53 is different from all
other burials in this burial place as the woman was
oriented with her head pointing upstream, not
downstream. The disarticulated left femur from
Figure 15 A) In situ limestone reddish and whitish discoid
beads found next to the damaged femur in child burial H153;
B) Detail of H153 with the beads.
Figure 14 Types of beads found at Vlasac in 2006–2009.
Provenance information (context no., x-find no.) of each
artifact is indicated. 1) Unmodified carp tooth used as a bead
(burial H297); 2) Perforated carp tooth used as a bead (burial
H297); 3, 4) Cyclope neritea applique´s (burial H297); 5)
Columbella rustica bead (228.627); 6) Columbella rustica
bead (314.614); 7) Ovoid Spondylus bead (burial H63.618);
8) Ovoid Spondylus bead (16); 9) Ovoid Spondylus bead (44.
62); 10) Discoid red limestone bead (burial H63, 64.650); 11)
Discoid green stone bead (8). Photographs by Emanuela
Cristiani.
Boric
´et al. Late Mesolithic lifeways and deathways at Vlasac (Serbia)
24 Journal of Field Archaeology 2014 VOL.39 NO.1
burial H63 that must have been kept in the burial
area for some time, was now placed along the axis of
the extended body in H53 between her legs (FIG. 10),
with a likely intention of producing the appearance of
an erect penis. The left pelvic bone and several other
disarticulated bones from H63 were also placed over
the lower legs of H53. Next to the right hip of this
individual a flat stone plaque was inserted vertically,
and, on the side of this plaque, two ulnae and a radius
coming from older burials H63 and H81 were stacked
up together. The pelvis and the head of H53 were
then covered by two flat stone plaques. These stone
plaques were similar to the ones that were used to
cover the lower parts of the legs of the individual in
burial H63. A red deer skull was placed atop the
stone plaque covering the pelvis in burial H53. This
red deer skull is dated to between 6006–5838 CAL B.C.
(OxA-16544) (at 95% confidence) (TABLE