1
Documenta Praehistorica XLIII (2016)
A Vin;a potscape> formal chronological models for the use
and development of Vin;a ceramics in south-east Europe
Alasdair Whittle 1, Alex Bayliss2, Alistair Barclay 3, Bisserka Gaydarska 1, Eszter Bánffy 4,
Du[an Borić 1, Florin Draşovean5, János Jakucs 6, Miroslav Marić 7, David Orton 8,
Ivana Pantović 9, Wolfram Schier 10, Nenad Tasić 11 and Marc Vander Linden12
1 Department of Archaeology and Conservation, Cardiff University, Cardiff, UK
whittle@cardiff.ac.uk< gaydarskab1@cardiff.ac.uk< boricd@cardiff.ac.uk
2 English Heritage, London, UK< and Biological and Environmental Sciences, University of Stirling, Stirling, UK
alex.bayliss@historicengland.org.uk
3 Wessex Archaeology, Salisbury, UK< a.barclay@wessexarch.co.uk
4 Römisch-Germanische Kommission, Frankfurt am Main, DE< eszter.banffy@dainst.de
5 The National Museum of Banat, Timişoara, RO< fdrasovean2000@yahoo.com
6 Institute of Archaeology, Hungarian Academy of Sciences, Budapest, H< jakucs86@gmail.com
7 The Institute for Balkan Studies, Serbian Academy of Sciences and Arts, Belgrade, RS< dzibril@gmail.com
8 BioArCh, Department of Archaeology, University of York, York, UK< david.orton@york.ac.uk
9 The City Museum of Vr[ac, Vr[ac, RS< arhivana@yahoo.com
10 Institute of Prehistoric Archaeology, Free University of Berlin, Berlin, DE< wolfram.schier@fu-berlin.de
11 Faculty of Philosophy, Belgrade University, Belgrade, RS< ntasic@gmail.com
12 Institute of Archaelogy, University College, London, UK< marc.linden@ucl.ac.uk
ABSTRACT – Recent work at Vin≠a-Belo Brdo has combined a total of more than 200 radiocarbon
dates with an array of other information to construct much more precise narratives for the struc-
tural history of the site and the cultural materials recovered from it. In this paper, we present the
results of a recent attempt to construct formal models for the chronology of the wider Vin≠a potscape,
so that we can place our now detailed understanding of changes at Belo Brdo within their contem-
porary contexts. We present our methodology for assessing the potential of the existing corpus of
more than 600 radiocarbon dates for refining the chronology of the five phases of Vin≠a ceramics
proposed by Miloj≠i≤ across their spatial ranges, including a total of 490 of them in a series of Baye-
sian chronological models. Then we outline our main results for the development of Vin≠a pottery.
Finally, we discuss some of the major implications for our understanding of the source, character
and tempo of material change.
IZVLE∞EK – Nedavne raziskave najdi∏≠a Vin≠a-Belo Brdo zdru∫ujejo ve≠ kot 200 radiokarbonskih da-
tumov z zbirko drugih podatkov z namenom ustvariti bolj natan≠ne zgodbe o strukturni zgodovini
najdi∏≠a in o izkopanem kulturnem materialu. V tem ≠lanku predstavljamo rezultate nedavnih po-
skusov oblikovanja formalnih modelov za kronologijo na ∏ir∏em obmo≠ju kulture Vin≠a, da lahko
umestimo zdaj ∫e podrobno razumevanje sprememb na Belem Brdu v njihove so≠asne kontekste.
Predstavljamo tudi na∏o metodologijo za ocenjevanje potenciala obstoje≠ega korpusa ve≠ kot 600
radiokarbonskih datumov za bolj natan≠no kronologijo vseh petih faz vin≠anske keramike, kot jih
je postavil Miloj≠i≤, v njihovih prostorskih omejitvah, vklju≠ujo≠ ∏e skupaj 490 datumov v seriji Ba-
yesovega kronolo∏kega modeliranja. Nato ori∏emo glavne rezultate razvoja vin≠anske lon≠enine. Na
koncu razpravljamo ∏e o nekaterih glavnih posledicah na∏ega razumevanja vira, zna≠ilnosti in hi-
trosti sprememb v materialni kulturi.
KEY WORDS – Neolithic; Vin≠a ceramics; Bayesian chronological modelling; radiocarbon dating;
network
DOI> 10.4312\dp.43.1 
Whittle, Bayliss, Barclay, Gaydarska, Bánffy, Borić, Draşovean, Jakucs, Marić, Orton, Pantović, Schier, Tasić and Vander Linden
2
The significance of the Vin≠a culture for the
development of the Neolithic in SE Europe
The Vin≠a culture belongs to the latter part of the
sixth millennium cal BC and the first half of the
fifth millennium cal BC (Chapman 1981; Markoti≤
1984; Bori≤ 2009; Por≠i≤ 2011; Orton 2012). Its
broad distribution extends through the river valleys
– the Danube, its tributaries and their catchments –
of the northern and central Balkans, from southern-
most Hungary and easternmost Croatia through Ser-
bia down to Kosovo and parts of Macedonia, and
from Croatia and Bosnia and Herzegovina eastwards
as far as parts of Transylvania in Romania and west-
ern Bulgaria (Fig. 1). The phenomenon presents a
series of significant changes in the character of Neo-
lithic settlement and social relations (Gara∏anin
1979; Chapman 1981; Kaiser, Voytek 1983; Tring-
ham, Krsti≤ 1990a), following the initial establish-
ment of Neolithic existence in the area of its distrib-
ution from the late seventh and early sixth millen-
nium cal BC onwards (Whittle et al. 2002). These
include changing materiality; the expansion of mate-
rial networks; the spread, consolidation and diver-
sification of settlement, involving increased seden-
tism in the form of large settlements and tells; the
intensification of subsistence; the introduction of
copper metallurgy; and the emergence of both larger
communities and distinctive households within them
(Chapman 1981; 2000; Kaiser, Voytek 1983; Tring-
ham, Krsti≤ 1990a; Tripkovi≤, Mili≤ 2009; Orton
2010; Orton et al. 2016). Im-
portant issues of settlement
expansion, population incre-
ase, changes in production, a
greater variety of sites includ-
ing tells themselves, more per-
manence of occupation, the
role of households and the
nature of community have all
been much discussed (see also
Bori≤ 2009; 2015; Borojevi≤
2006; Chapman 1990; Crno-
brnja et al. 2009; Lazarovici
1979; Link 2006.93–6; Orton
2010; 2012; Por≠i≤ 2012;
Schier 1995; Tasi≤ 2011;
Tringham et al. 1992; Tripko-
vi≤ 2011; Whittle 1996.105).
It has been clear since the
first radiocarbon dates (see
Chapman 1981) that these
features developed over a
timescale of several centuries. Explanations of the
radical transformations which both the emergence
and the demise of the Vin≠a culture brought have
varied considerably, from migrations within the cul-
ture history framework (e.g., N. Tasi≤ et al. 1990.32–
33; cf. Hervella et al. 2015), to the increasing impor-
tance as time went on of individual households and
competitive social relations (Tringham, Krsti≤ 1990a;
Chapman 2000). One widely shared view, within a
culture-historical framework, has been of southern
origins for both initiation of the Vin≠a culture and
subsequent changes during its development (such
as the claimed ‘shock of Vin≠a C’ (e.g., Gara∏anin
1951; 1979; G. Lazarovici 2000; Suciu 2009)). So
the Vin≠a phenomenon is central to our understand-
ing of social and cultural change in Neolithic south-
east Europe, and also in surrounding areas. Despite
the host of interpretations and chronological schemes
produced, however, it remains the case that the cal-
endar dating of Vin≠a culture changes – which prin-
cipally boil down to changes in pottery – has not so
far been rigorously or widely established, though
important earlier efforts in that direction should be
noted (Bori≤ 2009; 2015; Orton 2012; Schier 1995;
1996; 2000; 2014).
Periodisation schemes for Vin≠a ceramics: a
brief historiography
Distinctive material culture, including modelled face-
like lids, figurines and copper artefacts but predomi-
Fig. 1. Map of the maximum extent of the occurrence of Vin≠a ceramics
across south-east Europe.
A Vin;a potscape> formal chronological models for the use and development of Vin;a ceramics in south-east Europe
3
nantly the ubiquitous, abundant and high-quality,
dark-burnished pottery, has been the basis for clas-
sic periodisations of Vin≠a material (see Tasi≤ et al.
2016a; Fig. 2). That these in turn have been the spur
for prolonged debate about the nature and source
of origins and subsequent changes underlines their
continuing importance. The practice of researchers
has been to infer chronology from changing typo-
logical phases of ceramics, sometimes placing this
on an absolute scale by reference to the available
radiocarbon dates or by analogy with dated schemes
from neighbouring areas. This approach assumes
that two typologically identical assemblages are
synchronous, no matter their geographical location.
Variation through time and variation through space
are thus conflated by this approach, which has been
the source of much confusion, controversy and
debate.
Gordon Childe (1929.27–32; cf. Menghin 1931) had
enough data to distinguish only two phases (Vin≠a
I and II). This was based mainly on Vin≠a-Belo Brdo,
since this provided the most abundant material. The
next, more developed periodisation of Vin≠a-Belo
Brdo was by Friedrich Holste (1939) based on the
typology of pottery excavated and published by Mi-
loje Vasi≤, and dividing the material into five phas-
es labelled A–E, with phases A–D covering the Neo-
lithic development of the site. Although redefined
and remodelled by later authors, this basic struc-
ture continues to underpin most periodisations of
Vin≠a material as a whole.
In 1943 and 1949 Vladimir Miloj≠i≤ proposed a fur-
ther refinement, using exclusively the published
material from Vasi≤’s Preistoriska Vin≠a1. He basi-
cally reproduced Holste’s scheme but added sub-
phases B1 and B2. His phase B1 denotes a time
when meanders still appear as a decorative motif
and his phase B2 is the time of appearance of cur-
vilinear motifs which were more visible in the fol-
lowing Vin≠a C phase. Miloj≠i≤ also identified a tran-
sitional C–D phase between 4.5–4 m, marking the
transition to the last Neolithic phase at Belo Brdo
(Miloj≠i≤ 1949a.266–267; see also a useful English
summary in Miloj≠i≤ 1949b). He was also the first
fully to incorporate other Late Neolithic sites from
the central Balkans into his chronological system.
Milutin Gara∏anin (1951) was the first to make an
inventory of the entire pottery collection from Va-
si≤’s excavations at Belo Brdo. He systematised the
ceramic material and divided it into two major phas-
es, Vin≠a-Tordo∏ and Vin≠a-Plo≠nik, also drawing
1 Some of his descriptions reveal that he had in-depth first-hand knowledge of the finds.
Fig. 2. Overview of alternative typological schemes for Vin≠a ceramics (after Schier 1996).
Whittle, Bayliss, Barclay, Gaydarska, Bánffy, Borić, Draşovean, Jakucs, Marić, Orton, Pantović, Schier, Tasić and Vander Linden
4
on material from other sites in the wider Vin≠a dis-
tribution. He used the labels Tordo∏ and Plo≠nik to
link his early and late phases of the Vin≠a sequence
with the northernmost and southernmost sites
known at the time. At first he sub-divided both his
Vin≠a-Tordo∏ and Vin≠a-Plo≠nik phases, and later
elaborated this division even further (Gara∏anin
1979; 1993). He also incorporated the views of Bo-
rislav Jovanovi≤ (1994) on the Gradac phase, which
stands for a sudden and significant change in the
pottery forms first attested at the south Serbian site
of Gradac2.
Overlapping with and following the work of Gara-
∏anin, other periodisations essentially also based
on the typology of pottery, such as those by Dumit-
ru Berciu (1961), Stojan Dimitrijevi≤ (1968; 1974)
and Gheorghe Lazarovici (1979; 1981), considered
various areas of the Vin≠a distribution. In his quite
brief treatment, Berciu essentially followed the
structure already established by Miloj≠i≤, using the
A–D terminology and still therefore with ultimate
reliance on Vin≠a-Belo Brdo; his subdivision of Vin-
≠a B differs slightly from that of Miloj≠i≤ and he pro-
posed a tripartite division of C. Vin≠a is correlated
with other cultures of south-east Europe. His scheme
principally covers Banat, Oltenia, Muntenia and Tran-
sylvania. He made some reference to available un-
calibrated radiocarbon dates and suggested a rela-
tively low absolute chronology (Berciu 1961.Fig. 1,
Pl. I).
In turn, Dimitrijevi≤ concentrated on northern Ser-
bia, Romanian Banat and Croatian Slavonia, primar-
ily focusing on what was then called the Sopot-Leng-
yel culture, and using key sites like Bapska and Vin-
kovci, but with consideration of relationships and
correlations with its neighbours. In this, reference
was still made to Vin≠a-Belo Brdo by depth, with
variation in subdivisions of B, C and D; for the first
time, Vin≠a D is subdivided into D1 and D2, begin-
ning at a depth of 4.5m at Belo Brdo (Dimitrijevi≤
1968.89, kronolo∏ka tabela). Subdivision of A fol-
lowed in a later article (Dimitrijevi≤ 1974). No ref-
erence was made to radiocarbon dates, in a work
completed in 1964 and a very low absolute chronol-
ogy was proposed, underpinned by supposed Vin≠a
links to Northern Greece (Dimitrijevi≤ 1968.121–
122).
There followed further periodisation of the Neo-
lithic ceramic sequence in the Banat of north-west
Romania by Lazarovici (1979), using the strati-
graphy of key sites such as Gornea and Parta, but
yet again with reference to Vin≠a-Belo Brdo. The
basic structure for the Vin≠a sequence remains famil-
iar in outline, but A is now subdivided differently,
C is simplified, a C–D phase is reinstated, and D is
proposed as a single unit; transitions were also
added, as A3 between A2 and B1, and B1/B2 and
B2/C, equivalent to what were seen as hiatuses in
the Belo Brdo sequence (G. Lazarovici 1979. Tab.
7, 76). While a range of other material is considered,
it is the forms and decoration of the pottery which
are described in most detail, phase by phase (G. La-
zarovici 1979.105–139). Potential imports, from
surrounding groups such as the Alföld Linearband-
keramik, Szatmár, Bükk and Petresti, are also noted
(G. Lazarovici 1979.163–168). Detailed attention
was given to small-scale geographical variation with-
in the Banat, explicit distinction was made between
relative and absolute chronology, and comparisons
were made with much further afield, as far as Ana-
tolia and Mesopotamia (G. Lazarovici 1979.Tabs.
17–18, 178–179). A late absolute chronology is
proposed, with the Vin≠a sequence placed between
the late fifth and later third millennia (G. Lazaro-
vici 1979.Tabs. 17–18, 178–179). The origins of the
Vin≠a culture, with differences to the preceding Star-
≠evo-Cris complex, were seen as lying in a complex
set of cultural and ethnic changes and fusions, affect-
ing the Balkans as a whole (G. Lazarovici 1979.221).
Subsequently, a review of the Vin≠a culture across
Romania, summarising chronological trends in Olte-
nia and Transylvania/Siebenbürgen as well as Ba-
nat, reinforced this scheme; the C–D phase was omit-
ted on this occasion (G. Lazarovici 1981.Beilage 1).
There is still no reference to radiocarbon dating, and
the chronological table lacks any absolute timescale.
Famously, Vin≠a C was later characterised as a
‘shock’, involving migration from areas to the south
(G. Lazarovici 1987). Further detail and refinement
for Vin≠a C in the Banat were provided by Florin
Drasovean (1994; 1996; Drasovean et al. 1996).
Other reviews of the Romanian Neolithic have been
published subsequently (e.g., Mantu 2000; C.-M. La-
zarovici 2006; Suciu 2011). By the time of these
papers, calibrated radiocarbon dates, even if rela-
2 Unfortunately, Gara∏anin (1979.152) misunderstood Miloj≠i≤’s subdivision, maintaining that, according to the latter, the transition
between Vin≠a B2 and C would occur around a depth of 5m at Belo Brdo, instead of at 6m as Miloj≠i≤ had clearly stated. This led
to several decades of mismatch between the two typological systems, since researchers more familiar with the Gara∏anin scheme
consistently ‘translated’ the phase Vin≠a-Plo≠nik I back into the Miloj≠i≤ scheme as ‘Vin≠a B2–C’ instead of correctly ‘Vin≠a C’. We
have been aware of this issue when interpreting the published ceramic phasing of dated sites in the analyses below.
A Vin;a potscape> formal chronological models for the use and development of Vin;a ceramics in south-east Europe
5
tively few in number, were incorporated into chrono-
logical periodisations. It is worth noting that terms
such as ‘wave’, ‘colonisation’ and ‘penetration’ are
frequently used with reference to the Vin≠a culture;
for example, “Vin≠a culture is part of the second
great wave of southern origin” (Suciu 2011.75).
On the basis of available radiocarbon dates, geo-
graphical variation in the effect of ‘Vin≠a C type
communities’ is seen, with contrasts between Tran-
sylvania on the one hand and Banat and Oltenia on
the other; in Transylvania, these are seen as deter-
mining the appearance of the Copper Age, whereas
in Banat and Oltenia, the beginning of the Copper
Age is related to new migrations due to the Tiszapol-
gár and Să lcuta cultures (C.-M. Lazarovici 2006.
277). Recently, Drasovean (2014) has argued that
the beginning of the Copper Age in Banat and Tran-
sylvania is tied to the emergence of Foeni-Petresti
communities.
Hermann Parzinger (1993) quantified the varying
occurrence of selective decorative motifs and forms
by depth through the Belo Brdo tell. Following the
common perception at that time, he did not assign
chronological significance to the sequence of mate-
rial from the relative depths in the tell, but instead
gave more weight to what he considered closed
assemblages, like the few house inventories recog-
nised by Vasi≤ (Parzinger 1993.60–63). However,
this resulted in a rather small number of, arguably,
unrepresentative assemblages being used in his
analysis. Parzinger incorporated this in a compara-
tive, typologically based scheme (deliberately with-
out reference to radiocarbon dates: Parzinger 1993.
273) of successive cultural horizons, which covers
the wider west Balkan region, and far beyond. At
Belo Brdo itself he identified a series of Phases (I–
VI), and over much wider areas a succession of hori-
zons (4, 5a and 6–8 running parallel to the Belo Br-
do sequence; Fig. 2).
The A–D division was endorsed by John Chapman
(1981) for pragmatic reasons, to enable comparison
within the wider Vin≠a culture. In his work, phas-
es A–C denoted early Vin≠a and phase D late Vin≠a.
Going beyond Vin≠a-Belo Brdo, and drawing on
Fig. 3. Selected pot forms by depth at Vin≠a-Belo Brdo (after Miloj≠i≤ 1943).
Whittle, Bayliss, Barclay, Gaydarska, Bánffy, Borić, Draşovean, Jakucs, Marić, Orton, Pantović, Schier, Tasić and Vander Linden
6
the then available radiocarbon
dates (Chapman 1981.Fig. 14),
he sketched sequences for five
regions: Serbia and Macedonia,
the Vojvodina, north Bosnia, Ol-
tenia and Transylvania (Chap-
man 1981.22–31). Taking just
the example of decorative mo-
tifs on ‘early Vin≠a’ pottery, he
was able to show significant
regional variation (Chapman
1981.Fig. 11).
The scheme developed by Wolf-
ram Schier consists of eight
main phases and several sub-
phases, and covers the sequence
from the beginning of the Vin-
≠a occupation on the Belo Brdo
tell to 4m from its top (Schier
1995; 1996; 2000). This was
based on detailed analyses of
changes in the form and deco-
ration in a sample of about 2500 mostly unpublished
vessel fragments (mainly bowls) recorded by rela-
tive depth, and its chronological credibility was test-
ed by correspondence analysis. Schier also com-
pared the newly defined phases with data from sev-
eral other sites and concluded that the proposed
division is valid at least for the ‘core’ area of the
Vin≠a distribution, within a radius of some 100km of
Vin≠a-Belo Brdo itself, in the middle Morava, πuma-
dija and the Banat (Schier 1995.251–293; 1996.147).
Finally, since the appearance of Schier’s chronology,
Gara∏anin (2000) has refined his chronological sys-
tem even further, consistent with Schier’s results.
Additional subdivisions have been added, includ-
ing Vin≠a-Tordo∏ IIb 1–3.
Regional differences, imports, imitation and
influence: issues of variability
The brief review above documents how the culture
concept has persisted, perhaps largely for reasons
of convenience. But it is also perfectly clear that
many specialists have sought both intra- and inter-
regional variation within the Vin≠a ‘culture’ (e.g.,
Gara∏anin 1979; G. Lazarovici 1979; Chapman
1981). The notion of a Vin≠a core radiating around
Vin≠a-Belo Brdo has been proposed (Schier 1995),
though this may not necessarily be accepted from
the perspective of other areas (Lekovi≤ 1990; Mak-
kay 1990; Jovanovi≤ 1994). There is clear recogni-
tion, even within the culture-historical framework,
that the affiliation or allegiance of particular regions
may shift through time; cultural boundaries were
not immutable or permanently fixed. Actual imports
from neighbouring or surrounding regions have also
been claimed, as noted above (G. Lazarovici 1979),
and other potential examples were given by Corne-
lia-Magda Mantu (2000.80) and Florin Drasovean
(1994; 1996), among others. Complex, wide-ranging
interactions and fusions have been envisaged. So,
at various levels, many scholars have in practice
broken down a monolithic Vin≠a culture into a more
varied and nuanced network. Yet in many accounts
there is also a reversion to single explanations, espe-
cially when it comes to moments of key transition,
in terms of emergence, possible rupture and end-
ings, and the dominant threads linking these are the
claimed importance of areas to the south, in the
southern Balkans, northern Greece and beyond, and
inferred migrations.
Given developments in aDNA studies (e.g., Brandt
et al. 2013; Szécsényi-Nagy et al. 2015; cf. D. Hof-
mann 2015), we have to deal with population chan-
ges and infiltrations (and see again Hervella et al.
2015; and further discussion below). But what is
missing from the wider field of Vin≠a culture stud-
ies is a more reflective sense of possible diversity
and variability. The task then becomes to attempt to
unpick the complex cultural history of interaction,
innovation, imitation, influence, copying and import-
Fig. 4. Szederkény-Kukorica-dűlő, Grave 2484 accompanied by a Vin≠a
A1 (Schier 1995) bowl (after Jakucs et al. 2016).
A Vin;a potscape> formal chronological models for the use and development of Vin;a ceramics in south-east Europe
7
ing, not excluding the possibility of population move-
ment and replacement but equally not assuming a
priori that those should be the default explanations
for change and variation. This paper proceeds on
this basis, using pottery as its principal evidence.
Ultimately, the whole range of Vin≠a material and
practices should be the goal.
Pottery as proxy dating: absolute chronology
and typochronology
Finally, by way of introduction, it is important to
underline the difference between relative chronol-
ogy, or sequence provided by a combination of typol-
ogy (and on occasion seriation based on correspon-
dence analysis: Schier 1995; 2000) and site stratig-
raphy, and absolute dating, based on formal model-
ling of calibrated radiocarbon dates on short-lived
samples of known taphonomy, that are robustly
associated with the archaeological contexts or diag-
nostic assemblages which they are used to date. Past
generations of researchers, in the absence of large
assemblages of high-quality radiocarbon dates and
adequate statistical methodologies for their rigorous
interpretation, have perforce attempted the syn-
chronisation of relative sequences of pottery types
based on typological considerations (perhaps with
reference to the limited numbers of low-quality
radiocarbon dates available to them). This form of
‘pseudo’ absolute chronology must be distinguished
from what we propose here, which is the synchro-
nisation of such sequences on the basis of robust
statistical models of large numbers of radiocarbon
dates that are directly associated with pottery which
can be fitted within a common typological sequence.
This approach builds upon the careful typological
schemes of past research3, but in proper combina-
tion with scientific dating, adds absolute timescales
to these sequences. Such timings, as we describe
below, allow us to disentangle spatial from chrono-
logical variation in the use of these well established
types.
Aims of this paper: dating a potscape
As we have seen, the Vin≠a culture has long been a
familiar part of the culture-historical vocabulary used
to frame, order, and interpret the Neolithic of south-
east Europe. Many past notions derived from the use
of the culture concept, however, including those of
origins, identity and change, appear to have been
based, in the end, largely or principally on the char-
acter and distributions of pottery styles, and the
case of the Vin≠a complex is no exception. We are
going to use the wealth of knowledge about Vin≠a
pottery which has been built up over now a century
and more of research as a key component in con-
structing a formally modelled chronology for the
development of Vin≠a pottery across its wide dis-
tribution. If the landscape is what can be seen, and
a taskscape what can be heard (Ingold 1993.162), a
Vin≠a potscape could be defined as what could be
held in the hands of the makers and users of the
many ceramic assemblages across that distribution:
borrowing Ingold on the taskscape (1993.158), “the
entire ensemble … in their mutual interlocking …
an array” of related forms. Harnessing, rather than
casting aside, many of the strengths of past research
on the Vin≠a phenomenon, we seek to discuss the
many implications of our model for questions of the
character and tempo of material change, and the
nature of cultural variability. While we do not avoid
the term, ‘the Vin≠a culture’, alternative terminology
could help to recast many traditional debates (Por-
≠i≤ 2012.171; Bori≤ 2015.189–195), and from here
on we use other terms such as network to charac-
terise the potscape under discussion.
In a major project, The Times of Their Lives (see
Acknowledgements), we have obtained substantial
numbers of new radiocarbon dates and formally
modelled the results in a Bayesian statistical frame-
work for four sites with Vin≠a pottery: the tell of
Vin≠a-Belo Brdo, in northern Serbia (Tasi≤ et al.
2015; 2016a; 2016b), the tell of Uivar in western
Romania (Drasovean et al. submitted; Schier et al.
in preparation), and the flat settlements at Szeder-
kény and Versend, in south-west Hungary (Jakucs
et al. 2016; submitted). Encouraged by the coher-
ent results from these sites, we were then interested
to see if we could put these into the broader frame-
work of the development of the Vin≠a network con-
centrating on the potscape of ceramic change. For
this we have collected the published and other radio-
carbon dates available to us in 2015. For our inves-
tigations at Vin≠a-Belo Brdo, we provided formal
estimates for several of the available ceramic schemes
(Tasi≤ et al. 2016a.Figs. 22– 24, Tabs. 8–9; and see
3 But we note the lack of rigorous, quantitative typo-chronological studies on statistically viable samples of Vin≠a ceramics, and the
lack of representative analysis and full publication of excavated assemblages. These deficiencies in the analyses and publication of
Vin≠a assemblages have been the major obstacle in the construction of the chronological models presented below rather than any
lack of good-quality radiocarbon dates.
Whittle, Bayliss, Barclay, Gaydarska, Bánffy, Borić, Draşovean, Jakucs, Marić, Orton, Pantović, Schier, Tasić and Vander Linden
8
Fig. 2), but for this paper we have used, for the sake
of simplicity and wide applicability, the A–D scheme
of Miloj≠i≤ (1943; 1949a). Specifically, we aimed to:
● relate our models for Vin≠a-Belo Brdo, Uivar, Sze-
derkény and Versend to the wider pattern of cera-
mic change, formally modelled within the A–D
scheme for the Vin≠a culture;
● examine critically the temporal and geographical
coherence of the A–D scheme;
● identify temporal and geographical variation in
the adoption, maintenance and abandonment of
ceramic styles within the A–D scheme;
● and by these means investigate the character and
tempo of cultural change.
Scientific dating and Bayesian chronological
modelling for Vin≠a ceramics
The first radiocarbon dates for Vin≠a ceramics were
very much in the tradition of their time — one or two
samples of charred material from a given site. Only
a handful of sites had three to six dates. Vin≠a-Belo
Brdo itself was fortunate to obtain two measurements
on charred grain samples (GrN-1537 and GrN-1546;
Vogel, Waterbolk 1963.183–184), and one measure-
ment on a sample of unidentified charcoal (GrN-
1535; Todorovi≤, Cermanovi≤ 1961.71).
At the beginning of the 1980s, Chapman (1981.17–
31) was able to gather details of 36 radiocarbon
dates from 16 sites from which Vin≠a pottery had
been recovered, and used these uncalibrated radio-
carbon dates in comparison with local stratigraph-
ic sequences. He reached three major conclusions:
that the duration of phases A–C was comparable to
the duration of the entire D phase; that the final
centuries of the Vin≠a culture were not represented
at the type-site; and that the different rates of devel-
opment in the various regions prevent Vin≠a Belo-
Brdo from being used as a ‘type-site’ for the devel-
opment of the Vin≠a culture as a whole.
In the 1990s, Schier (1996.Tab. 1) obtained a series
of 14 conventional radiocarbon dates on worked
antlers and large animal bones that had been re-
covered by Vasi≤ at Belo Brdo and recorded by
depth. These derive from the lower and middle part
of the tell (4.1–9.3m) covered by his seriation of a
sample of the surviving pottery assemblage. Using
these results and the first series of dates produced
in Groningen, Schier (1996.Figs. 11–12) produced
the first Bayesian chronological model for Vin≠a-Belo
Brdo, combining the radiocarbon dates with the
depths through the tell recorded on the samples.
At the same time Roland Gläser (1996) presented a
review of 76 radiocarbon dates from 14 sites con-
taining Vin≠a ceramics. He provided a Bayesian chro-
nological model combining the site stratigraphy with
the corpus of radiocarbon dates for Selevac (Gläser
1996.Fig. 5), but otherwise his analysis is based on
simple calibrated radiocarbon dates and summed
probability distributions of calibrated dates for select-
ed sites, and site and ceramic phases. He suggested
that the Vin≠a culture spanned the period from c.
5200 cal BC to c. 4500 cal BC.
Du∏an Bori≤ (2009) mustered a total of 155 radio-
carbon dates from 27 sites associated with Vin≠a
pottery, including 40 new AMS measurements on
short-life, single-entity samples from seven sites.
Bayesian models were presented for six sites: Belo
Brdo, Rudna Glava, Belovode, Plo≠nik, Gomolava
and Divostin II. This analysis suggested that the
culture dated from 5400–5300 cal BC to 4650–4600
cal BC. Contrary to Chapman, Bori≤ argued (2009.
234) for an approximately similar duration of each
phase (A to D) of 200±50 years. More recently, a set
of 26 new AMS dates on animal bone from Gomola-
va, Opovo and Petnica have been published and
Bayesian models constructed (Orton 2012). These
new dates generally fit the chronological framework
suggested by Bori≤.
The chronological framework and approach
It is now time not only to compare these new site
chronologies with each other, but also to set them
within the wider framework of other dated sites that
formed part of the Vin≠a network. We consider a
corpus of 564 radiocarbon dates that have been pub-
lished as relating to Vin≠a pottery (Tab. 1)4. Full
details of these measurements are provided in Table
2 where we have gathered information from diverse
sources or when measurements are published for the
first time. Where full details of measurements have
already been published (e.g., Bori≤ 2009) references
to sources used for each site are provided in Table 1.
Chronological modelling has been undertaken using
the program OxCal v4.2 (Bronk Ramsey 2009; Bronk
4 Additionally, there are 34 radiocarbon dates from the latest levels of Vin≠a-Belo Brdo in Sector II (Tasi≤ et al. 2015), 101 from the
recently excavated deep sounding (Tasi≤ et al. 2016b), and two from the 2011 profile that are not considered here.
A Vin;a potscape> formal chronological models for the use and development of Vin;a ceramics in south-east Europe
9
Ramsey, Lee 2013) and the calibration dataset of
Reimer et alii (2013). The algorithms used in the
models are defined exactly either by the brackets
and OxCal keywords on the left-hand side of the fig-
ures or by the CQ2 code provided as supplementary
information (http://c14.arch.ox.ac. uk/). In the fig-
ures, the outputs from the models, the posterior
density estimates, are shown in black, and the un-
constrained calibrated radiocarbon dates are shown
in outline; dates on samples of charred plant remains
that may have an old-wood offset are shown in grey;
the parameter names of distributions that have been
excluded from the modelling are followed by ‘?’.
Some distributions do not relate to single radiocar-
bon dates. These other distributions correspond to
aspects of the model. For example, the distribution
‘start Szederkény’ (Fig. 5) is the posterior density
estimate for the time when the settlement at Szeder-
kény was established. In the text and tables, the
Highest Posterior Density intervals of the posterior
density estimates are given in italics. Calibrated
date ranges given in normal type have been calcu-
lated using IntCal13 and the probability method
(Stuiver, Reimer 1993), and do not derive from the
chronological models described. All ranges have
been rounded outwards to five years.
For each measurement, we examine the character of
the dated material and its association with Vin≠a
ceramics. Given the absence of an agreed typological
scheme across the region, assemblages have been
allocated to Miloj≠i≤’s (1943; 1949a) scheme: A, B1,
B2 (incorporating what was later called the Gradac
phase), C (incorporating Miloj≠i≤’s (1949a; 1949b)
phase C–D), and D (incorporating all proposed sub-
divisions) (Fig. 3). Where site stratigraphy is avail-
able, this is used in a Bayesian model to constrain
the calibration of the radiocarbon dates. In some
cases the available published information is not suf-
ficient to demonstrate a robust association between
the dated samples and a ceramic phase(s) (these
results are reported in footnotes).
Ideally, we wish to include in our models only radio-
carbon dates on short-life samples that are directly
associated with typologically diagnostic assemblages
of the relevant pottery. Dates on articulated animal
bones from closed ceramic assemblages in pits, for
example, are ideal (for a demonstration of the po-
tential of this approach see Denaire et al. in press).
Unfortunately, the samples submitted for dating by
past researchers, the published typological and stra-
tigraphic information, and the reporting of the radio-
carbon measurements and their associated details
(Bayliss 2015) are frequently less than ideal. In these
circumstances, we have been forced to make prag-
matic judgements about the information available
to us.
We have only included radiocarbon dates in our
study that are published as having a clear associa-
tion with Vin≠a ceramics that have been assigned to
various typological schemes (cf. Fig. 2), which we
have converted to the Miloj≠i≤ (1943) scheme. This
means that sites often have more radiocarbon dates
than have been used in the modelling, but these are
either associated with activity of other periods or
do not have explicit associations with diagnostic Vin-
≠a material. In many cases it has been impossible
for us to judge the validity of the published ceramic
association, as sites are not yet published in detail.
Sometimes associations cannot be made at the fea-
ture or structure level, but rather an entire site is cat-
egorised as only containing material of a certain
phase. It should be noted that the association
between cultural material and the radiocarbon sam-
Fig. 5. Probability distributions of key parameters from Szederkény-Kukurica-dűlő, derived from the
model defined by Jakucs et al. (2016.Fig. 10).
Whittle, Bayliss, Barclay, Gaydarska, Bánffy, Borić, Draşovean, Jakucs, Marić, Orton, Pantović, Schier, Tasić and Vander Linden
10
ple is critical to avoid circular arguments (by which
a site is assigned to particular ceramic phase on the
basis of the calibrated radiocarbon dates rather than
the typological characteristics of the pottery present).
This legacy dataset is inevitably of variable quality.
Although over 70% of measurements have been
made by AMS, short-life, single-entity samples that
can be confidently associated with the use of the
feature from which the sample derived constitute
less than a quarter of dates (and 80% of those derive
from just two sites, Szederkény and Versend). Over
20% of samples were of unidentified charcoal (or
of charcoal from long-lived species such as oak and
elm) and so may incorporate an old-wood offset.
Other samples consist of disarticulated animal bones
or single carbonised cereal grains, where it is not
known whether the dated material derived from
particular concentrations. Such materials might well
be residual (older than their contexts) or intrusive
(younger than their contexts).
We have attempted to distil reliable chronology from
this body of data, by incorporating each result into
the model in a way that is appropriate for the dated
material.
● Samples of human bone from graves, articulated
animal bone groups, and short-lived charred plant
remains (including short-life charcoal) from fired
features such as hearths or large concentrations,
such as coherent dumps in pits, have been incor-
porated into the models as short-life material like-
ly to be contemporary with the archaeological
activity of interest (n = 134).
● Disarticulated animal bones and short-lived charr-
ed plant material from houses or pits have also
been included fully in the models, although we
consider the archaeological association in this case
to be less reliable (n = 244); in a number of cases
the dates on these samples are clearly much ear-
lier than related dates (usually having a poor
individual index of agreement in the model) and
so they have been modelled as residual termini
post quos (n = 43), and in a few cases they are
clearly intrusive (n = 8).
● Samples of unidentified charcoal (or charcoal
from long-lived species) have been incorporated
into the models as termini post quos (n = 115),
except for a few which are statistically consistent
with results on short-life materials from the same
context (n = 5).
● Five results from Uivar have been excluded from
the model for that site for technical reasons
(Schier et al. forthcoming).
In total, therefore, our models include 490 radio-
carbon measurements (with a further 66 not in-
cluded in the analysis either because there is a sig-
nificant possibility that the measurements are inac-
curate, or because we think the dated material was
intrusive, or because the association with the Vin≠a
ceramics is unclear). A total of 134 measurements
on short-lived samples firmly associated with the
dated context are included fully in the models
(27%); a further 193 measurements on short-life
material that can be associated with the dated con-
text less reliably have also been included fully in the
modelling (40%), with another 43 such samples
modelled as termini post quos as the dated mate-
rial was probably residual (9%); and 120 measure-
ments on charcoal samples of uncertain maturity
are included in the models as termini post quos
(24%). Our analysis suggests that overall there is
approximately a 25% chance that a disarticulated
animal bone, or a short-life charred plant from a
deposit with which it is not functionally related, is
residual or intrusive. This demonstrates the need to
date articulating bone groups or single fragments of
short-lived charred plant material from primary
deposits (Bayliss et al. 2016).
This analysis provides quantitative date estimates
for Miloj≠i≤’s (1943) ceramic phase boundaries at
different sites that are independent of any synchro-
nisation between alternative typological schemes.
In some cases the character of the dated material
means that samples can only provide termini post
quos for particular ceramic phases, or can only be
allocated to a range of ceramic phases. We start by
considering the chronologies and ceramic associa-
tions of particular sites, and construct chronological
models for sites which have more than three radio-
carbon dates on short-lived material with robust
ceramic associations. We then construct a series of
models for the chronology of each of Miloj≠i≤’s cera-
mic phases, utilising posterior distributions from the
site-based models as inputs for these models where
appropriate and calibrated radiocarbon dates where
not. This ensures that sites which have many radio-
carbon dates (e.g., Uivar or Belo Brdo which between
them have almost half of the radiocarbon measure-
ments considered in this review) do not dispropor-
tionately affect the analysis.
A Vin;a potscape> formal chronological models for the use and development of Vin;a ceramics in south-east Europe
11
The Vin≠a potscape
We discuss the dated sites from west to east, follow-
ing the Danube downstream and examining sites in
the catchments of its tributaries as these join the
river (Fig. 1).
The Danube upstream of the Tisza
A linear strip 1.7km long, comprising 12.5ha, was
investigated at Szederkény-Kukorica-dűlő between
2005 and 2008. The site lies approx. 20km west of
the current course of the Danube in south-eastern
Transdanubia, Hungary. A total of 66 longhouses,
orientated NE–SW and broadly conforming to the
longhouse architecture which is generally perceived
as a hallmark of the central European Linearband-
keramik (LBK), and 50 crouched inhumations were
recovered. Most of the graves are unfurnished,
although some were accompanied by Vin≠a ceram-
ics (Fig. 4). Vin≠a pottery, from Schier’s (1996) A1–
A3 phases, was recovered from settlement features
(Jakucs, Voicsek 2015). This material can be related
to Miloj≠i≤ phase A used in this study.
Forty-one radiocarbon measurements are available
from this site (Jakucs et al. 2016.Tab. 1), of which
19 are from features containing diagnostic Vin≠a A
ceramics. These dominated the assemblages from
the eastern and central areas of the site (the west-
ern part contained mainly Ra∫i∏te-type pottery). Já-
nos Jakucs et alii (2016.Fig. 11) present a model for
the chronology of Szederkény, which is summarised
in Figure 5. This estimates that Vin≠a A ceramics
appeared here in 5360–5305 cal BC (95% proba-
bility; start Szederkény; Fig. 5), probably in 5340–
5315 cal BC (68% probability), and ceased to be
used in 5230–5175 cal BC (95% probability; last
Szederkény eastern; Fig. 5), probably in 5215–5190
cal BC (68% probability).
The site of Versend-Gilencsa lies in southern Hun-
gary, approx. 2km east of Szederkény-Kukorica-dűlő
(Jakucs et al. submitted). A linear strip totalling
2.24ha was excavated along the planned line of the
motorway, divided by the course of the Versend
stream. Features of the Neolithic settlement could
be detected on the floodplains of the eastern and
western side of the watercourse.
On the eastern side, unambiguous traces of 21 long-
houses orientated NE–SW could be identified. These
were arranged in rows, and each was flanked by
longitudinal pits containing mixed assemblages of
early LBK and Vin≠a A ceramics. The density of
archaeological features was higher on the western
side, where the extensive traces of the Neolithic set-
tlement were heavily destroyed by later activity. The
houses were consequently much less identifiable.
Sixty-eight radiocarbon measurements are available
from this site (Jakucs et al. submitted.Tab. 1). Jakucs
et alii (submitted.Figs. 5–6) present a model for
the chronology of Versend, which is summarised in
Figure 6. This estimates that the settlement began
in 5305–5280 cal BC (2% probability; start Versend
settlement; Fig. 6) or 5255–5210 cal BC (93% prob-
ability), probably in 5235–5215 cal BC (68% prob-
ability), and ceased to be used in 5220–5180 cal BC
(93% probability; end Versend settlement; Fig. 6) or
5150–5115 cal BC (2% probability), probably in
5210–5195 cal BC (68% probability).
Several occupation horizons that include assem-
blages of Vin≠a pottery have been recovered from
the tell settlement at Bapska, eastern Croatia, in 13
campaigns of excavation between 1911 and the pre-
sent (Buri≤, Te∫ak-Gregl 2009; Buri≤ 2011). Two
radiocarbon measurements are available from the
1964 trench; one from the foundation of House 2-
A is associated with an assemblage described by
Dimitrijevi≤ (1968.92) as Vin≠a D1 and equated by
him with Miloj≠i≤’s phase C–D. This dating, being on
charcoal from a potentially long-lived species, thus
provides a terminus post quem of 4850–4485 cal
BC (Bln-348; 95% probability) or 4780–4580 cal BC
(68% probability) for the end of Miloj≠i≤ phase C at
Bapska5. A further series of radiocarbon measure-
ments have recently been obtained on short-lived
material associated with Houses 1 and 2 from renew-
ed excavations (Buri≤, Te∫ak-Gregl 2009; Buri≤
2015). These appear to be associated with Vin≠a C
or Vin≠a D ceramics, and can be modelled as span-
ning 4645–4495 cal BC (95% probability; start
Bapska C/D; Fig. 7), probably 4580–4515 cal BC
(68% probability), and 4550–4420 cal BC (95%
probability; end Bapska C/D; Fig. 7), probably
4535–4475 cal BC (68% probability).
Tisza and Mures valleys
The site of Maroslele-Pana is located in south-eastern
Hungary, on the right (northern) bank of the Maros
(Mures) river, northwest of the village of Maroslele,
5 A second sample from the 1964 excavations (Bln-346, 5955±80 BP, 5055–4665 cal BC (94% probability) or 4640–4615 cal BC
(1% probability)) is associated with Sopot II pottery and is thus not included in this study.
Whittle, Bayliss, Barclay, Gaydarska, Bánffy, Borić, Draşovean, Jakucs, Marić, Orton, Pantović, Schier, Tasić and Vander Linden
12
close to the confluence of the Tisza and Maros rivers.
The site was first excavated in 1963; four Neolithic
features (pits 1–4) and five crouched burials were
documented (Trogmayer 1964). The pottery from
this early campaign represented the late phases of
the early Neolithic Körös culture. In 2008 there were
further excavations in advance of road construction.
The excavation took place on a low ridge and cov-
ered 2.7ha. A total of 71 of the 231 excavated fea-
tures can be dated to the Neolithic period (Paluch
2011). Five radiocarbon measurements were obtain-
ed on animal bone from separate pits (Tab. 2). The
dated pits contained diagnostic early Vin≠a A1–A2
(Schier 1996) and early Alföld LBK style pottery, but
also some fragments and altar pieces which showed
clear resemblances to the pottery style of the late
Körös culture. The Vin≠a A1–A2 elements are most
frequent in pit 85, while the other dated pits contain
more equal proportions of early Vin≠a and Alföld
LBK style material. The mixed (Vin≠a and Alföld LBK
with early Neolithic traits) assemblage of Maroslele-
Pana in the Tisza-Maros confluence area (Paluch
2011) can be seen as a similar phenomenon to that
observed at Versend-Gilencsa in southern Transda-
nubia and at Satchinez in northern Banat. The model
for the dated activity from Maroslele-Pana has good
overall agreement (Amodel: 106; Fig. 8) and suggests
that the site was occupied between 5380–5220 cal
BC (95% probability; Start Maroslele-Pana A; Fig.
8), probably in 5325–5250 cal BC (68% probabil-
ity), and 5300–5095 cal BC (95% probability; End
Maroslele-Pana A; Fig. 8), probably 5275–5195 cal
BC (68% probability).
At Ószentiván (site 8) four measurements were taken
on bulk charcoal from a pit containing Vin≠a A pot-
tery that was investigated during rescue excavations
in 1960 (Kohl, Quitta 1970). This material was iden-
tified as oak and elm charcoal, both species that can
live to several hundred years, and so the combined
result provides a terminus post quem for the pottery
of 5305–5050 cal BC (95% probability; Ószentiván 8)
Fig. 6. Probability distributions of key parameters from the settlement at Versend-Gilencsa, derived from
the model defined by Jakucs et alii (submitted.Figs. 5–6).
Fig. 7. Probability distributions of radiocarbon dates from Bapska. Each distribution represents the rel-
ative probability that an event occurs at a particular time. For each of the dates two distributions have
been plotted: one in outline, which is the result of simple radiocarbon calibration, and a solid one, based
on the chronological model used. Distributions other than those relating to particular samples corre-
spond to aspects of the model. For example, the distribution ‘start Bapska D’ is the estimated date when
Vin≠a D ceramics first appeared on the site. Measurements followed by a question mark and shown in
outline have been excluded from the model for reasons explained in the text, and are simple calibrated
dates (Stuiver, Reimer 1993). The large square brackets down the left-hand side along with the OxCal key-
words define the overall model exactly.
A Vin;a potscape> formal chronological models for the use and development of Vin;a ceramics in south-east Europe
13
or 5295–5245 cal BC (19% probability) or 5235–
5200 cal BC (13% probability) or 5170–5075 cal BC
(36% probability). A single sample of unidentified
charcoal was dated from pit 1, recovered in later
excavations at the same site (published as Tiszasi-
get), which was also associated with Vin≠a A ceram-
ics (Trogmayer 1980.298). This provides a termi-
nus post quem for the assemblage of 5465–5445 cal
BC (1% probability; Bln-1631) or 5380–5190 cal BC
(78% probability) or 5185–5055 cal BC (16% proba-
bility), probably 5340–5205 cal BC (68% probabili-
ty).
The site of At is located to the north-west of Vr∏ac
in south-eastern Banat, Serbia. The settlement was
located on a sloping terrace on the north-west edge
of Veliki rit, a marshy area drained in the mid-nine-
teeth century. Several seasons of rescue excavations
revealed the existence of a multi-period prehistoric
site, ranging from the Lower Palaeolithic to the Late
Neolithic (Joanovi≠, Priki≤ 1978). Renewed excava-
tions in 2014–2015 revealed a structure of the Star-
≠evo period, and cut features associated with Vin≠a
C and Vin≠a D ceramics (Chu et al. 2016). Four radio-
carbon dates are available from the Vin≠a layers,
two from pit 1 and two from the infilling of the Star-
≠evo structure. These place this activity between
6040–5960 cal BC (1% probability; Start At C/D;
Fig. 9), or 5700–4660 cal BC (94% probability),
probably in 5050–4720 cal BC (68% probability),
and 4595–3510 cal BC (94% probability; End At
C/D; Fig. 9), or 3180–3090 cal BC (1% probability),
probably in 4535–4190 cal BC (68% probability).
These estimates are extremely imprecise, with very
long tails to the distributions, because there are cur-
rently few dates from this site.
Also in the vicinity of Vr∏ac is the 100ha flat settle-
ment of Potporanj, which has produced early Vin≠a
pottery and the largest quantity of obsidian found
within the Vin≠a network. Provenance studies of
this material show that it came from the Hungarian
and Slovakian Carpathian Mountains (Tripkovi≤,
Mili≤ 2009). Three radiocarbon dates are available
from recent excavations of the site by Vr∏ac Museum.
MAMS-22667, a sample of unidentified charcoal that
may have an old-wood offset, provides a terminus
post quem of 5210–5000 cal BC (95% probability)
or 5210–5160 cal BC (25% probability) or 5120–
5105 cal BC (4% probability) or 5080–5005 cal BC
(39% probability) for an assemblage of Vin≠a A ce-
ramics in the earliest occupation level. MAMS-22666,
also a sample of unidentified charcoal, provides a
terminus post quem of 5215–5015 cal BC (95% pro-
bability) or 5210–5090 cal BC (51% probability) or
5085–5050 cal BC (17% probability) for the use of
Vin≠a B2/C ceramics in the latest habition level of
the site. The third date, MAMS-22668, is on animal
bone from a level associated with Vin≠a B2 pottery
and dates this deposit to 5295–5240 cal BC (9%
probability) or 5235–5190 cal BC (19% probability)
or 5180–5060 cal BC (67% probability), probably
5220–5200 cal BC (13% probability) or 5170–5075
cal BC (55% probability).
A sample of antler was dated from pit 4 at the Neo-
lithic settlement of Satchinez, north-eastern Banat,
Romania, which contained sherds of Vin≠a A2 pot-
tery along with LBK sherds (Drasovean 1993; Hor-
váth, Drasovean 2010.15). This provides a date of
5325–5200 cal BC (Deb-2579; 87% probability) or
5165–5115 cal BC (5% probability) or 5110–5075
cal BC (3% probability), probably 5300–5220 cal BC
(68% probability) for this assemblage. Also in north-
eastern Banat, Romania, two samples of antler have
been dated from pit 4 at Hodoni, a multi-period site
which has produced 20 pits containing Vin≠a C pot-
tery (Drasovean 1994; 1996). These dates place this
assemblage at 4845–4650 cal BC (93% probability;
Hodoni: pit 4; Fig. 32) or 4640–4615 cal BC (2%
Fig. 8. Probability distributions of radiocarbon dates from Maroslele-Pana. The format is identical to that
of Figure 7. The large square brackets down the left-hand side along with the OxCal keywords define the
overall model exactly.
Whittle, Bayliss, Barclay, Gaydarska, Bánffy, Borić, Draşovean, Jakucs, Marić, Orton, Pantović, Schier, Tasić and Vander Linden
14
probability), probably at 4795–4705 cal BC (68%
probability).
The tell site of Uivar in the Romanian Banat was ex-
cavated between 1999 and 2009 by a Romanian-
German team (Schier, Drasovean 2004; Schier
2008; Drasovean, Schier 2009; Schier 2016). It con-
sists of a central tell settlement 4m high and cover-
ing some 3ha, which was surrounded by a series of
substantial ditches, some accompanied by palisades.
These enclose an area of some 12ha. A flat settle-
ment occupied at least some of this, and may have
succeeded occupation on the tell itself. A complete
sequence through the tell was excavated in Trench
1. A recent programme of chronological modelling
utilising a total of 182 radiocarbon measurements
from the site formally estimates the chronology of
this complex (Schier et al. forthcoming; Drasovean
et al. submitted).
Only preliminary analyses of the pottery assemblages
from the site have been undertaken so far. The low-
est deposits in Trench 1 contained pottery predom-
inantly in the Szakálhát tradition. The first Vin≠a-
type sherds appear to be local imitations of Miloj≠i≤
B1 (Schier 1995.phase B). They occur from the con-
struction horizon of H3d in 5045–4990 cal BC (95%
probability; build H3d+c; Fig. 10), probably in 5035–
5000 cal BC (68% probability). The first sherds of
Schier (1995) phase C1 (equivalent to late Miloj≠i≤
B2, see Fig. 2) pottery appear in the construction
horizon of H3c, which unfortunately could not be
dated directly since the structure did not provide
suitable samples. The date of its appearance can be
estimated, however, since this must be between the
construction of buildings H3d and H3b. This model
is shown in Figure 10 and suggests that late Miloj≠i≤
B2 (Schier 1995.phase C1) pottery first appeared in
5030–4930 cal BC (95% probability; start Miloj≠i≤
B2 at Uivar), probably in 5010–4950 cal BC (68%
probability). It becomes predominant on the site
from the construction horizon H3b in 4985–4910
cal BC (95% probability; build H3b; Fig. 10), prob-
ably in 4960–4925 cal BC (68% probability). Miloj-
≠i≤ C (Schier 1995.phase C2) pottery appears from
the construction of the H3a horizon in 4920–4850
cal BC (95% probability; build H3a; Fig. 10), prob-
ably in 4905–4895 cal BC (8% probability) or
4890–4855 cal BC (60% probability). No sherds of
Miloj≠i≤ C–D (Schier 1995.phase D1) pottery have
been recovered from Uivar, and so Miloj≠i≤ C ceram-
ics appear to have been current until the end of the
Neolithic occupation of the tell in 4740–4600 cal BC
(95% probability; end Uivar tell; Fig. 10), probably
in 4725–4645 cal BC (68% probability).
Three samples of bone have been dated from the site
at Tărtăria, Transylvania, Romania (Merlini 2011.
224; Merlini, Lazarovici 2008.156). One is from a
partial female skeleton found disarticulated and
mixed with animal bone in the pit which is thought
to have contained the Tărtăria tablets, one from
cleaning of the 1942–1943 section, and one from
the base of pit B2. Merlini places the pottery from
the pit containing the skeleton in the Vin≠a phase
A2 or A3, although Lazarovici (2010.Fig 8) places
the same assemblage in Vin≠a A3/B1; he also places
the assemblage from pit B2 in this phase. We have
interpreted these classifications to refer to an assem-
blage of Vin≠a B1 pottery, which contained surviving
elements of Vin≠a A3.
The taphonomy of the dated material and its asso-
ciation with the recovered ceramics are problemat-
ic. The animal bone from the base of pit B2 prob-
ably provides a date for the associated Vin≠a B1 pot-
tery (although, of course, it is always possible that
a disarticulated bone is residual). This result cali-
brates to 5315–5000 cal BC (R-1655; 95% probabil-
ity), probably to 5295–5250 cal BC (13% probabili-
ty) or 5230–5195 cal BC (12% probability) or 5180–
5065 cal BC (43% probability). The dated animal
Fig. 9. Probability distributions of radiocarbon dates from At. The format is identical to that of Figure 7
(grey tone indicates the sample has a possible old-wood offset). The large square brackets down the left-
hand side along with the OxCal keywords define the overall model exactly.
A Vin;a potscape> formal chronological models for the use and development of Vin;a ceramics in south-east Europe
15
bone from the section is effectively unstratified and
cannot be securely associated with the pottery6.
Interpretation of the date on the female partial skele-
ton is more complicated, since there appears to be
evidence that the body was exposed for a period of
time before burial in the pit associated with the
tablets. For this reason, in this study we have includ-
ed this date as a terminus post quem for the use of
the Vin≠a B1 ceramics from the pit. This is 5470–
5200 cal BC (88% probability; R-1631) or 5165–
5115 cal BC (4% probability) or 5110–5075 cal BC
(3% probability), probably 5360–5215 cal BC (68%
probability)7.
The settlement at Miercurea Sibiului-Petris, in the
catchment of the upper Mures valley, consisted of
surface houses and pits (Luca et al. 2006). These
appear to be associated with Vin≠a B ceramics. Four
radiocarbon dates are available on short-lived mate-
rials from two pits and a house (Tab. 2)8. Pit 18 is
stratigraphically earlier than pit 3 and house 11,
although GrA-33127 from pit 18 appears to be on a
residual cereal grain and so is incorporated in the
model as a terminus post quem. This model has
good overall agreement (Amodel: 79; Fig. 11), and
suggests that Vin≠a B occupation here occurred bet-
ween 5650–5025 cal BC (95% probability; Start
Miercurea Sibiului-Petris B), probably 5290–5095
cal BC (68% probability), and 5280–4590 cal BC
(95% probability; End Miercurea Sibiului-Petris
B), probably 5170–4970 cal BC (68% probabili-
ty) 9.
Fig. 10. Probability distributions of key parameters relating to the use of Vin≠a ceramics at Uivar. The
format is identical to that of Figure 7. The large square brackets down the left-hand side along with the
OxCal keywords define the overall model exactly (key parameters have been imported from the model
defined by Schier et alii (forthcoming.Figs. 6.9, 6.12–6.22).
6 This sample was an unidentified animal bone recovered from the cleaning of the section of the 1942/3 excavation and produced
a date of 5310–4995 cal BC (95% probability; R-1630, 6200±65 BP), probably 5285–5270 cal BC (2% probability) or 5225–5050
cal BC (66% probability).
7 The radiocarbon dates raise the possibility that Milady of Tărtăria was curated for some time before her bones were buried in this
pit. It is 83% probable that R-1631 is earlier than R-1655, probably by -115–375 years (95% probability; distribution not shown),
probably -10–240 years (68% probability). The negative parts of these ranges reflect the possibility that her bones were freshly
deposited.
8 A further radiocarbon date from Miercurea Sibiului-Petris (GrA-26606, 6180±40 BP, 5285–5270 cal BC (1% probability) or 5230–
5000 cal BC (94% probability)) appears to be on an animal bone that was intrusive in Star≠evo pit 9 (Luca et al. 2006).
9 A cattle rib that was dated from L3 house, square 6–8 (110–130cm), at Limba (GrN-28112, 6290±50 BP, 5375–5200 cal BC (88%
probability) or 5170–5075 cal BC (7% probability), is from a house that was expected to be contemporary with the Cris IV ceram-
ics within it (Biagi et al. 2005.49). The painted vessel has been attributed to Vin≠a A2 (Lazarovici 2009.184–185; 2010.
115–116; 2014.16–18, Figs. 3, 5c), Vin≠a A2–A3 (Mazâre 2005.258, 288), or Star≠evo-Cris IV (Drasovean 2014.footnote 10). It
appears to be a Lumea Noua import. The result is later than expected and has been reassigned to Vin≠a A3, although the publi-
cation does not provide evidence of an association with Vin≠a A3 ceramics. This result is therefore not included in the modelling
presented here (Suciu 2009.37). Similarly, the association between the cattle tibia dated by GrN-28994 (5760±40 BP, 4710–
4515 cal BC (95% probability) and the, probably earlier Vin≠a ceramics, recovered from Cauce Cave is unclear and this result is
also not included in this analysis (Suciu 2009.67–9, Fig. 82, Annex 1). Three radiocarbon dates from Orăstie–Dealul Permilor
(Deb-5762, 5825±60 BP, 4830–4535 cal BC (95% probability); Deb-5765, 6070±70 BP, 5210–4825 cal BC (94% probability) or
4815–4800 cal BC (1% probability); Deb-5775, 5790±55 BP, 4780–4515 cal BC (95% probability)) are again insecurely associated
with Vin≠a pottery and are not modelled here (Luca 2003). Eight radiocarbon measurements are available from Cârcea Viaduct,
although these samples cannot be associated with Vin≠a pottery at all (contra Mantu 2000.99; Biagi, Spataro 2005.37; Bln-1980,
6100±60 BP, 5215–4880 cal BC (93% probability) or 4870–4845 cal BC (2% probability); Bln-2008, 6550±40 BP, 5615–5585
cal BC (7% probability) or 5570–5465 cal BC (88% probability); Bln-2287, 6300±55 BP, 5465–5440 cal BC (2% probability) or
5425–5405 cal BC (1% probability) or 5385–5200 cal BC (86% probability) or 5170–5075 cal BC (7% probability); Bln-2289,
5910±55 BP, 4945–4680 cal BC (95% probability); Bln-2291, 5990±55 BP, 5005–4725 cal BC (95% probability); Bln-2292,
6350±60 BP, 5470–5220 cal BC (95% probability); Bln-2294, 5865±95 BP, 4965–4495 cal BC (95% probability); and Bln-2354,
5860±60 BP, 4850–4545 cal BC (95% probability)).
Whittle, Bayliss, Barclay, Gaydarska, Bánffy, Borić, Draşovean, Jakucs, Marić, Orton, Pantović, Schier, Tasić and Vander Linden
16
Sava valley and tributaries
The site of Lupljanica, with Vin≠a pottery, is located
in the valley of the Bosna River, one of the main
tributaries of the Sava in central Bosnia. A single
sample of unidentified charcoal was dated from the
lowest level of a house structure, 1.3m below the
surface (Breunig 1987.107; Crane, Griffin 1972.
190–191), and provides a terminus post quem of
4910–3985 cal BC (M-2455; 95% probability), prob-
ably 4705–4255 cal BC (68% probability) for the
end of Vin≠a D.
The Neolithic tell at Gornja Tuzla is located on the
slopes of Mt. Majevica in the upper part of the Jala
river catchment. It was discovered in 1949 and is
estimated to be between 12 and 15ha in size. Exca-
vations were undertaken over four seasons between
1955 and 1958 (∞ovi≤ 1961.79–139). Five radio-
carbon measurements associated with Vin≠a period
activity have been published, but only two have
robust ceramic associations10. These are measure-
ments on unidentified charcoal found in association
with Vin≠a C pottery that provide termini post quos
for this pottery of 4535–4330 cal BC (GrN-1974; 95%
probability) or 4460–4355 cal BC (68% probability)
and of 4780–4355 cal BC (Bln-349; 95% probability)
or 4685–4455 cal BC (68% probability).
The multi-period tell at Gomolava is in Vojvodina,
Serbia, on the east bank of the river Sava. It was
first detected in 1898, and small-scale excavations
were undertaken in 1904–1908. The first systematic
excavations took place between 1953 and 1985,
with approximately half a hectare investigated (Gi-
ri≤ 1988.13; Brukner 1980; 1988). This yielded a
number of prehistoric occupation horizons ranging
from the late Neolithic to late Iron Age. However,
the site is best known for its late Vin≠a necropolis
(Bori≤ 1996; Jovanovi≤ 2015), one of only very few
discovered thus far in the Balkans.
Fifteen samples are on unidentified charcoal and
may have contained material with an old-wood off-
set; the other 17 samples are on short-lived materi-
als (two bulk samples of carbonised cereal grain,
four human burials, and 11 animal bones; Waterbolk
1988.Beilage 1; Orton 2012.Tab. 2; Bori≤ 2009.Tab.
5).
A model for the dated Vin≠a activity at Gomolava
has good overall agreement between the radiocar-
bon dates and the stratigraphic information (Amo-
del: 80; Fig. 12), only if three dates that fall in the
early Bronze Age are excluded as outliers (all con-
sidered to be intrusive from later phases of activi-
ty: GrN-7373, GrN-7375 and OxA-21132).
The model presented in Fig. 12 suggests that Vin≠a
C started at Gomolava in 5005–4845 cal BC (95%
probability; Start Gomolava C; Fig. 12), probably in
4955–4875 cal BC (68% probability); the boundary
between phases C and D occurred in 4810–4720 cal
BC (95% probability; Gomolava C/D; Fig. 12), prob-
ably in 4790–4735 cal BC (68% probability); and
10 Three further measurements have been obtained from more recent excavations at Gornja Tuzla (Vander Linden et al. 2014).
OxA-23297 (6165±34 BP, 5220–5015 cal BC (95% probability)) on a large mammal bone from layer 15 does not appear to be
securely associated with a diagnostic ceramic assemblage. Two further statistically consistent measurements (OxA-23298, 5827±33
BP, 4785–4590 cal BC (95% probability) and OxA-23299, 5741±33 BP, 4690–4500 cal BC (95% probability): T’ = 3.4; ν = 1; T’(5%)
= 3.8) were obtained on different species of animal bone (Bos taurus and Cervus elaphus) from layer 8. These samples also have
no known direct ceramic associations.
Fig. 11. Probability distributions of radiocarbon dates from Miercurea Sibiului-Petris. The format is iden-
tical to that of Figure 7. The large square brackets down the left-hand side along with the OxCal keywords
define the overall model exactly.
A Vin;a potscape> formal chronological models for the use and development of Vin;a ceramics in south-east Europe
17
phase D ended here in 4670–4500 cal BC (95%
probability; End Gomolava D; Fig. 12), probably in
4635–4550 cal BC (68% probability).
The Neolithic settlement at Petnica is located in the
Kolubara valley area in western Serbia, on a slope
running up to a north-facing cliff. A cave in this cliff
was also occupied in this period. The site has been
excavated on several occasions, yielding three late
Neolithic occupation phases and several Vin≠a-peri-
od wattle and daub structures (Starovi≤ 1993).
Seven radiocarbon dates have been obtained on ani-
mal bone samples from occupation features at Pet-
nica (Bori≤ 2009.Tab. 6; Orton 2012.Tab. 2). The
model for the dated activity at Petnica has good over-
Fig. 12. Probability distributions of radiocarbon dates from Gomolava. Distributions for GrN-7373, GrN-
7375, and OxA-21132, which are considered to derive from post-Neolithic activity, are not shown. Grey
tone indicates that the sample has a possible old-wood offset. The format is identical to that of Figure 7.
The large square brackets down the left-hand side along with the OxCal keywords define the overall model
exactly.
Whittle, Bayliss, Barclay, Gaydarska, Bánffy, Borić, Draşovean, Jakucs, Marić, Orton, Pantović, Schier, Tasić and Vander Linden
18
all agreement (Amodel: 94; Fig. 13), if we interpret
Pit 3 as being cut from above the level of House 3
and if we interpret the animal bone from House 2
as residual. We follow the ceramic phasing for Pet-
nica published by David Orton (2012).
This model suggests that phase B2 at Petnica start-
ed in 5050–4850 cal BC (95% probability; Start
Petnica B2; Fig. 13), probably in 4995–4900 cal BC
(68% probability); the boundary between phases B2
and C occurred in 4965–4850 cal BC (95% proba-
bility; Petnica B2/C; Fig. 13), probably in 4940–
4880 cal BC (68% probability); the boundary bet-
ween phases C and D occurred in 4950–4845 cal BC
(95% probability; Petnica C/D; Fig. 13), probably in
4925–4860 cal BC (68% probability) and the dated
sequence ended in 4935–4735 cal BC (95% proba-
bility; End Petnica D; Fig. 13), probably in 4900–
4800 cal BC (68% probability).
The site of Masinske Njive is located in western Ser-
bia on a tributary of the Kolubara River. The site
was originally recorded in the 1960s and was exca-
vated in full between 2006 and 2009 in advance of
coal mining. A single phase Vin≠a settlement was
excavated, which produced Vin≠a B1/B2 ceramics
(M. Spasi≤, pers. comm.).
Three radiocarbon measurements have been ob-
tained on samples of animal bone from settlement
features. A model for the dated activity at this site
has good overall agreement (Amodel: 100; Fig. 14),
and indicates that this activity began in 5705–5085
cal BC (95% probability; Start Masinske B; Fig. 14),
probably in 5355–5215 cal BC (68% probability),
and ended in 5300–4700 cal BC (95% probability;
End Masinske B; Fig. 14), probably in 5275–5070
cal BC (68% probability).
Barely 500m west of Masinske Njive, on an elevated
terrace above the same tributary of the Kolubara Ri-
ver, lay the early Vin≠a A site of Jari≠i∏te 1. Although
the site was recorded in the 1960s, it was only exca-
vated between 2006 and 2010 and has since been
destroyed by the Kolubara coal mine (Mari≤ 2013).
The early Vin≠a settlement featured several dozen
pits, some containing ovens, although no wattle and
daub structures were detected.
A single flax seed was dated from pit 1.137, which
contained an assemblage of Vin≠a A pottery, indi-
cating that these ceramics date to 5320–5205 cal BC
(88% probability; NOSAMS-78623) or 5165–5135 cal
BC (3% probability) or 5130–5115 cal BC (1% proba-
bility) or 5110–5075 cal BC (3% probability), prob-
ably in 5300–5240 cal BC (53% probability) or 5235–
5215 cal BC (15% probability).
The Belgrade area
The tell site of Belo Brdo at Vin≠a just east of Bel-
grade forms the type-site for the development of
Vin≠a pottery. The 8m of Neolithic deposits have
been excavated almost constantly since 1908, when
Vasi≤ began large scale excavations. The massive
assemblage of pottery from these excavations has
been the subject of a series of seminal typological
Fig. 13. Probability distributions of radiocarbon dates from Petnica. The format is identical to that of
Figure 7. The large square brackets down the left-hand side along with the OxCal keywords define the
overall model exactly.
A Vin;a potscape> formal chronological models for the use and development of Vin;a ceramics in south-east Europe
19
studies (Fig. 2), the sequence through the tell pro-
viding relative chronology for the schemes suggest-
ed. As part of the The Times of Their Lives project,
major new programmes of radiocarbon dating have
been undertaken on the Vasi≤ sequence (Tasi≤ et al.
2016a), the excavations of the upper levels in Sector
II (Tasi≤ et al. 2015) and the sequence from a new
deep sounding excavated between 2004–2014 (Ta-
si≤ et al. 2016b).
Most relevant to the chronology of Vin≠a ceramics
is the model presented by Tasi≤ et alii (2016a.Fig.
17) for the sequence of deposits excavated by Milo-
je Vasi≤, as these excavations covered a larger area
than those of all subsequent excavations combined
and consequently produced by far the largest assem-
blage of Vin≠a pottery. This is a Poisson process age-
depth model with the rigidity of the process defined
as the 10cm spits used by Vasi≤ to excavate the tell
and record the recovered finds (Bronk Ramsey 2008;
Bronk Ramsey, Lee 2013). The typological scheme
for Vin≠a pottery published by Miloj≠i≤ (1943) used
in this study was based on the published ceramics
from this archive recorded by these spits. The age-
depth model thus explicitly estimates their dates.
Miloj≠i≤ phase A occurs between 9.3m and 8.0m,
phase B1 between 8.0m and 7.0m, phase B2 between
7.0m and 6.0m, phase C between 6.0m and 4.0m,
and phase D from 4.0m to the top of Vin≠a cultural
layers at 1.3m.
On this basis, at Belo Brdo Miloj≠i≤ phase A begins
in 5305–5255 cal BC (95% probability; Belo Brdo
start A; Fig. 15), probably in 5300–5270 cal BC
(68% probability). The transition between Miloj≠i≤
A and B1 occurs in 5210–5135 cal BC (95% proba-
bility; Belo Brdo A/B1; Fig. 15), probably in 5200–
5165 cal BC (68% probability). The shift between
Miloj≠i≤ B1 and B2 occurs in 5115–5040 cal BC
(95% probability; Belo Brdo B1/B2; Fig. 15), prob-
ably in 5090–5055 cal BC (68% probability). The
transition between Miloj≠i≤ B2 and C happens in
4935–4850 cal BC (95% probability; Belo Brdo B2/
C; Fig. 15), probably in 4920–4875 cal BC (68%
probability). The transition between Miloj≠i≤ phases
C and D occurs in 4765–4680 cal BC (95% proba-
bility; Belo Brdo C/D; Fig. 15), probably in 4760–
4735 cal BC (12% probability) or 4725–4690 cal
BC (56% probability). Miloj≠i≤ (1943) placed the
end of his phase D at 2.5m at Belo Brdo, but the
accumulation of Vin≠a ceramics within the deposits
of Belo Brdo continues up to a depth of 1.3m, until
4570–4460 cal BC (95% probability; Belo Brdo end
D; Fig. 15), probably in 4550–4495 cal BC (68%
probability).
The presence of Vin≠a D ceramics above relative
depth of 2.5m at Belo Brdo was confirmed by exca-
vations in Sector II undertaken between 1998 and
2009, where such ceramics occur until 4550–4485
cal BC (95% probability; fire 2; Tasi≤ et al. 2015.
Fig. 8), probably in 4545–4505 cal BC (68% prob-
ability). This date estimate for the end of Vin≠a D
at Belo Brdo is completely independent of that pro-
vided by the age-depth model of the Vasi≤ sequence,
but the medians of these probability distributions
vary by just three years. The chronological model-
ling of the new sounding provides much greater
detail on the structural narrative of the tell but,
given the small area excavated and the small quan-
tity of material recovered, this area does not add
materially to our understanding of Vin≠a ceramics.
The late Neolithic flat settlement at Opovo is locat-
ed on a low knoll overlooking a then active mean-
der of the Tami∏ River, about 25km north of Bel-
grade. During the 1980s five seasons of excavation
recorded a total area of 340m2 (Tringham et al.
1985; 1992), yielding three successive building hori-
zons associated with late Vin≠a C and early Vin≠a D
ceramics.
Thirteen radiocarbon measurements are available on
animal bone (Orton 2012.Tab. 2) and unidentified
Fig. 14. Probability distributions of radiocarbon dates from Masinske Njive. The format is identical to
that of Figure 7. The large square brackets down the left-hand side along with the OxCal keywords define
the overall model exactly.
Whittle, Bayliss, Barclay, Gaydarska, Bánffy, Borić, Draşovean, Jakucs, Marić, Orton, Pantović, Schier, Tasić and Vander Linden
20
charcoal (Tab. 2) from a sequence of building hori-
zons (BH2 and 3) and associated pits. All 13 dates
are in good agreement with the site stratigraphy and
phasing, and the model has good overall agreement
(Amodel: 115; Fig. 16).
The model indicates that activity associated with
Vin≠a C pottery at Opovo began in 4960–4805 cal
BC (95% probability; Start Opovo C; Fig. 16), prob-
ably in 4915–4840 cal BC (68% probability), and
that the use of Vin≠a C pottery on the site ended in
4835–4695 cal BC (95% probability; End Opovo C;
Fig. 16), probably in 4820–4740 cal BC (68% prob-
ability). It should be noted that there are currently
no radiocarbon dates for the latest building horizon
(BH 1) at Opovo and so settlement certainly contin-
ued after this time.
Located in the south-east suburb of Belgrade of the
same name, the site of Banjica is a well-known late
Neolithic settlement. It was situated on a plateau
above the Banjica stream and was discovered during
road construction in 1921 (Todorovi≤, Cermanovi≤
1961). The site has yielded five settlement horizons
of wattle-and-daub structures and pits associated
with Vin≠a D ceramics (Tripkovi≤ 2007).
Two statistically consistent measurements (T’ = 0.1;
T’(5%) = 3.8; ν = 1; Ward, Wilson 1978) on bulk
unidentified charcoal (one including charred grain)
were obtained for features from Banjica associated
with Vin≠a D pottery (Tab. 2). These provided ter-
mini post quos for the end of Vin≠a D here of 4770–
4755 cal BC (1% probability) or 4730–4355 cal BC
(94% probability; GrN-1542), probably 4680–4635
Fig. 15. Probability distributions of Miloj≠i≤ (1943) ceramic phase boundaries at Vin≠a-Belo Brdo,
derived from the model illustrated in Tasi≤ et alii (2016a.Fig. 17).
Fig. 16. Probability distributions of radiocarbon dates from Opovo. The format is identical to that of
Figure 7. Grey tone indicates that the sample has a possible old-wood offset. The large square brackets
down the left-hand side along with the OxCal keywords define the overall model exactly.
A Vin;a potscape> formal chronological models for the use and development of Vin;a ceramics in south-east Europe
21
cal BC (13% probability) or 4620–4455 cal BC (55%
probability), and 4795–4320 cal BC (GrN-1536, 95%
probability), probably 4655– 4640 cal BC (2% prob-
ability) or 4620–4365 cal BC (66% probability)11.
Central Serbia12
Located in central Serbia, south-west of Kragujevac,
the settlement site of Grivac lies on the east bank of
the Gru∫a River, on a wide terrace bounded by sev-
eral streams. A chance discovery in 1950, it was exca-
vated episodically between 1952 and 1994 (Bogda-
novi≤ 2004). The earliest phase of occupation has
Star≠evo pottery, and pre-dates a settlement associ-
ated with Vin≠a ceramics of phases A–C, which is
itself sealed by an early Iron Age barrow cemetery
(Bogdanovi≤ 1977).
Six radiocarbon dates, all on samples of unidentified
bulk charcoal or ‘burnt earth’, can be associated with
assemblages of Vin≠a ceramics (Tab. 2). Bln-870 pro-
vides a terminus post quem for the end of Vin≠a A
of 5480–5040 cal BC (95% probability), probably of
5470–5400 cal BC (12% probability) or 5390–5205
cal BC (56% probability). The latest of the five dates
associated with Vin≠a C ceramics13 provides a termi-
nus post quem for the end of Vin≠a C of 4960–4530
cal BC (95% probability; tpq Grivac C; Fig. 17),
probably of 4880–4680 cal BC (64% probability) or
4640–4615 cal BC (4% probability).
The site of Divostin (II) is situated in central Serbia,
west of the town of Kragujevac, and on a slope bet-
ween two streams in a landscape dominated by gen-
tle hills. Small-scale excavations took place between
1967 and 1970 covering 1.65% of the estimated total
site area of 15ha (McPherron, Srejovi≤ 1988). The
earliest occupation on the site is associated with Star-
≠evo pottery, but the majority of the excavated fea-
tures belong to a later settlement defined by wattle
and daub rectangular structures associated with late
Vin≠a pottery.
Eleven radiocarbon dates were obtained on samples
(seven unidentified bulk charcoal, one ‘burnt earth’
and three single animal bones) from a sequence of
houses and pits associated with Vin≠a C and D pot-
tery (Tab. 1). Two dates on bulk charcoal (Bln-867
and BM-574) are far too young for their stratigraph-
ic position within the described sequence, and have
been excluded from the model. These samples prob-
ably contained a component of intrusive charcoal.
The remaining nine measurements include two sets
of replicates (repeat measurements on the same
bulk sample). Two measurements on unidentified
charcoal from pit F117 are statistically consistent
(Bln-865 and Bln-865a; T’ = 0.2; T’(5%) = 3.8; ν = 1),
as are three measurements, also on unidentified char-
coal, from pit 20 (F121) (Z-336a, Z-336b and Bln-
898; T’ = 2.1; T’(5%) = 6.0; ν = 2).
The model which combines these radiocarbon dates
with the ceramic phasing has good overall agree-
ment (Amodel: 113; Fig. 18), although all the sam-
ples that can be associated with Vin≠a C ceramics
are of unidentified charcoal and so only provide ter-
mini post quos for the end of this phase. This model
suggests that the transition between Vin≠a C and D
ceramics at Divostin occurred in 4830–4600 cal BC
(95% probability; Divostin C/D; Fig. 18), probably
4740–4635 cal BC (68% probability) and the use of
Vin≠a D ceramics ended in 4710–4320 cal BC (95%
probability; End Divostin D; Fig. 18), probably 4675–
4535 cal BC (68% probability).
The Morava valley and southwards
The site of Anzabegovo is located in the eastern part
of FYR of Macedonia, in an area known as Ov≠e Pole.
It was discovered by chance during the construction
of a railway line in the early 1960s, with the main
excavations taking place in 1969 and 1970 (Gimbu-
tas 1974). The multi-period settlement was first es-
tablished in the early Neolithic, with activity on site
occurring into the Roman period.
Two statistically consistent measurements (LJ-2329
and LJ-2411: T’ = 1.0; T’(5%) = 3.8; ν = 1) on uniden-
tified bulk charcoal were obtained for features asso-
ciated with Vin≠a B pottery (Tab. 2). Given the poten-
tial for the sample to have contained a component
of old wood, LJ-2329 provides a terminus post quem
for the presence of Vin≠a B ceramics at Anzabegovo
of 5320–5025 cal BC (95% probability), probably of
5300–5205 cal BC (39% probability) or 5165–5115
cal BC (17% probability) or 5110–5075 cal BC (11%
probability).
11 A third measurement is quoted by N. N. Tasi≤ (1988.46) as ‘GrN *** 5320±150 BP’, but no further details can be traced. The
Groningen laboratory has no record of this measurement, and so it was certainly produced elsewhere. It calibrates to 4455–
3890 cal BC (90% probability) or 3885–3795 cal BC (5% probability).
12 Two measurements have been published by Gimbutas (1976.Tab. IV) as being from Medvednjak in central Serbia. Bln-480, how-
ever, is from Ószentiván, Hungary (Kohl, Quitta 1970.411) and LJ-2523 is from πventoji, Lithuania (Linick 1977.27).
13 Z-1507 is clearly much later than the other samples from this phase and has been excluded from this analysis; it is likely that the
bulk sample contained an element of intrusive later charcoal.
Whittle, Bayliss, Barclay, Gaydarska, Bánffy, Borić, Draşovean, Jakucs, Marić, Orton, Pantović, Schier, Tasić and Vander Linden
22
The site of Belovode in Veliko Laole lies to the west
of River Mlava, on one of its smaller tributaries. It
was discovered by chance in the 1960s. The settle-
ment was located on an ellipsoid plateau, barely
10km to the east of the Velika Morava River, and is
cut in half by a small stream named Belovode (πlji-
var, Jacanovi≤ 1996). The site is multi-period, with
the earliest habitation occurring during the Star≠evo
period, followed by a longer phase of late Neolithic
Vin≠a occupation and ending in the Late Bronze Age.
The first systematic excavations took place in 1994
and continued episodically until 2015.
Nine radiocarbon dates were obtained on samples of
single animal bones and antlers from a sequence of
houses and associated features with Vin≠a A to C pot-
tery (Bori≤ 2009; Tab. 2). One date (OxA-14678) from
the top of the sequence is far too late and almost
certainly belongs to a later phase of activity and is
excluded from the model. Three dates (OxA-14680,
–14 683 and –14 700), with poor individual agree-
ment with their relative stratigraphic positions,
work well with the model presented in Figure 19 if
treated as termini post quos (interpreted as residual
animal bone within their respective contexts). If this
interpretation is applied then the model, based on
the stratigraphy, pottery phasing (A to C) and the
radiocarbon dates, has good overall agreement
(Amodel: 102; Fig. 19).
The model indicates that the activity, associated with
Vin≠a A to C pottery, started at Belovode in 5795–
5230 cal BC (95% probability; Start Belovode A;
Fig. 19), probably in 5515–5310 cal BC (68% proba-
bility); that the transition from Vin≠a A to B occurred
in 5365–5075 cal BC (95% probability; Belovode
A/B; Fig. 19), probably in 5320–5165 cal BC (68%
probability); that the transition from Vin≠a B to C
occurred in 5150–4645 cal BC (95% probability;
Belovode B/C; Fig. 19), probably in 5050–4735 cal
Fig. 17. Probability distributions of radiocarbon dates from Grivac. The format is identical to that of
Figure 7. Grey tone indicates that the sample has a possible old-wood offset. The large square brackets
down the left-hand side along with the OxCal keywords define the overall model exactly.
Fig. 18. Probability distributions of radiocarbon dates from Divostin. The format is identical to that of
Figure 7. Grey tone indicates that the sample has a possible old-wood offset. The large square brackets
down the left-hand side along with the OxCal keywords define the overall model exactly.
A Vin;a potscape> formal chronological models for the use and development of Vin;a ceramics in south-east Europe
23
BC (68% probability). The use of Vin≠a C ceramics
at Belovode ended in 4775–4205 cal BC (95% pro-
bability; End Belovode C; Fig. 19), probably in 4705–
4495 cal BC (68% probability).
Located in a village west of the town of Prokuplje,
the site of Plo≠nik lies on an elevated terrace of the
Toplica River. It was discovered in the 1920s during
the construction of a railway line and was excavated
episodically until 2013 (Stalio 1964; 1973). The site
is best known for the find of approx. 17kg of copper
implements excavated from buildings and hoards
associated with a late phase of Vin≠a culture (πljivar
et al. 2010).
Eight radiocarbon measurements, all on single ani-
mal bones, were obtained for features associated
with houses within the settlement (Tab. 2). The
dated features contained Vin≠a B pottery and ‘Gra-
dac’ pottery. In this area the site begins with the use
of Vin≠a B1 (D. πljivar, pers. comm.), and all the
dated samples are from contexts below the ‘Gradac’
layer except for OxA-14685 (Bori≤ 2009.Figs. 22–
23). Samples from sequences of deposits were dated
in trenches 14 and 15, and all the radiocarbon dates
are in good agreement with these stratigraphic se-
quences and the pottery phasing (Amodel: 102; Fig.
20).
The first use of Vin≠a B1 ceramics at Plo≠nik occurr-
ed in 5350–5060 cal BC (95% probability; Start
Plo≠nik B1; Fig. 20), probably in 5240–5110 cal BC
(68% probability). Vin≠a B ceramics ended here in
5030–4755 cal BC (95% probability; End Plo≠nik
B2; Fig. 20) probably in 4985–4860 cal BC (68%
probability). Trench 16 contained features associated
with ‘Gradac’ pottery. A single radiocarbon sample
associated with this ceramic phase provides a termi-
nus ante quem for the start of Miloj≠i≤ phase C and
a terminus post quem for the end of Miloj≠i≤ phase
D at Plo≠nik of 4710–4530 cal BC (95% probability;
OxA-14685; Tab. 2), probably of 4685–4630 cal BC
(31% probability) or 4625–4555 cal BC (37% prob-
ability).
The site of Ore∏kovica lies on a hill dominating the
surrounding territory around the village of the same
name. It is approx. 6ha in size, and is surrounded by
at least two enclosure ditches. It is a single-phase set-
tlement associated with Vin≠a B pottery. Even though
it was known by the mid-1970s it was only excavat-
ed for the first time in 2013 (Bori≤ et al. in prepa-
ration).
Six radiocarbon dates, all on single samples of ani-
mal bone (two of them articulating) are currently
available (Tab. 2). A model combining these results
Fig. 19. Probability distributions of radiocarbon dates from Belovode. The format is identical to that of
Figure 7. The large square brackets down the left-hand side along with the OxCal keywords define the
overall model exactly.
Whittle, Bayliss, Barclay, Gaydarska, Bánffy, Borić, Draşovean, Jakucs, Marić, Orton, Pantović, Schier, Tasić and Vander Linden
24
with the stratigraphic sequence is shown in Fig. 21,
and has good overall agreement (Amodel: 105; Fig.
21). This suggests that the settlement at Ore∏kovica
(and so Vin≠a B ceramics) began in 5400–5080 cal
BC (95% probability; Start Ore∏kovica B; Fig. 21),
probably in 5335–5210 cal BC (67% probability)
or 5165–5155 cal BC (1% probability), and ended
in 5215–4935 cal BC (95% probability; End Ore∏-
kovica B; Fig. 21), probably in 5200–5180 cal BC
(4% probability) or 5165–5030 cal BC (64% prob-
ability).
Selevac-Staro Selo site lies in the πumadija area, to
the west of the lowest part of the Velika Morava
River valley, barely 22km south of its confluence
with the Danube (Tringham, Krsti≤ 1990a). The site
is located 4.8km east of the modern village of the
same name on elevated ground west of the river.
The site was recognised in 1968, with small-scale
excavations taking place until 1970 and again in
1973. The Selevac Archaeological Project ran from
1976 until 1981, and yielded settlement remains of
the Late Neolithic Vin≠a period.
Twelve radiocarbon dates for Selevac were obtained
on samples (bulk charcoal and bulk cereal grain)
from a sequence of houses and associated features
with Vin≠a B1 to C pottery (Tab. 2). We have used
the ceramic phasing for this sequence presented by
Schier (1995.Figs. 144–5), although unfortunately
all the dates associated with Vin≠a B2 and C ceram-
ics are on unidentified charcoal and so can only be
included in the model as termini post quos. This
model is shown in Fig. 22, which has good agree-
ment (Amodel: 100). Generally, measurements from
the Rudjer Bo∏kovi≤ Institute, Zagreb, are likely to
Fig. 20. Probability distributions of radiocarbon dates from Plo≠nik. The format is identical to that of
Figure 7. The large square brackets down the left-hand side along with the OxCal keywords define the
overall model exactly.
Fig. 21. Probability distributions of radiocarbon dates from Ore∏kovica. The format is identical to that of
Figure 7. The large square brackets down the left-hand side along with the OxCal keywords define the
overall model exactly.
A Vin;a potscape> formal chronological models for the use and development of Vin;a ceramics in south-east Europe
25
be on average 77 years too young (Horvatin≠i≤ et
al. 1990.299). In this case, however, it can be noted
that the two measurements in question (Z-233a–b)
are statistically consistent with a third from La Jolla
(LJ-2521; T’ = 0.4; T’(5%) = 6.0; ν = 2) and so they
are probably accurate.
The model indicates that this activity, associated with
Vin≠a B1 pottery, started in 6800–6685 cal BC (2%
probability; Start Selevac B1; Fig. 22) or 6340–4960
cal BC (93% probability), probably in 5525–5010
cal BC (68% probability). The transition between
Vin≠a B1 and B2 occurred here in 5210–3585 cal
BC (95% probability; Selevac B1/B2; Fig. 22), prob-
ably in 5165–4655 cal BC (68% probability). The
dated occupation at Selevac ended in 4580–2460
cal BC (78% probability; End Selevac C; Fig. 22) or
2195–1495 cal BC (17% probability), probably in
4565–3190 cal BC (61% probability) or 1770–1495
cal BC (7% probability).
The site of Predionica is located in a suburb of Pri∏-
tina in Kosovo. It was excavated in 1955 and 1956
in advance of the construction of a factory building.
It yielded a late Neolithic settlement with three build-
ing horizons (Galovi≤ 1959). A single sample of oak
charcoal was dated from a fire installation associat-
ed with Vin≠a A ceramics, and provides a terminus
post quem of 5470–5400 cal BC (6% probability;
Bln-435) or 5390–5045 cal BC (89% probability),
probably of 5360–5205 cal BC (57% probability) or
5165–5135 cal BC (5% probability) or 5130–5115
cal BC (2% probability) or 5110–5100 cal BC (1%
probability) or 5095–5075 cal BC (3% probability)
for the end of Vin≠a A at this location.
The site of Vala≠ is located in the north-west part
of Kosovo, near the town of Kosovska Mitrovica.
The site is located on a terraced hill on the edge of
the Ibar River valley, and it appears to have been
defended by a palisade and a stone wall (N. Tasi≤
1960). It was excavated briefly in 1957. A single mea-
surement on a bulk sample of charred acorns and
peas from a pit associated with Vin≠a D ceramics
dates this assemblage to 4960–4545 cal BC (95%
probability; Bln-436), probably to 4900–4865 cal BC
(7% probability) or 4855–4685 cal BC (61% proba-
bility).
Beran Kr∏ is located on a dominant ridge on the left
bank of the river Lim, approx. 3km from the town
of Berane in Montenegro. Small test trenches were
dug in 1961, but the main research excavations were
undertaken in 1975 and 1976. The settlement con-
sisted of several building horizons belonging to the
Vin≠a period (Markovi≤ 1985).
Two measurements on unidentified bulk charcoal
are available from two stratigraphically related depo-
sits (Tab. 2). The published contextual details, how-
Fig. 22. Probability distributions of radiocarbon dates from Selevac. The format is identical to that of
Figure 7. Grey tone indicates that the sample has a possible old-wood offset. The large square brackets
down the left-hand side along with the OxCal keywords define the overall model exactly.
Whittle, Bayliss, Barclay, Gaydarska, Bánffy, Borić, Draşovean, Jakucs, Marić, Orton, Pantović, Schier, Tasić and Vander Linden
26
ever, differ. Z-491 is attributed to Level VII and Z-
492 to Level XIII (Srdo≠ et al. 1977.474), but also to
Levels III and IV/Horizon IIa (Markovi≤ 1985.70).
Here, we follow ∞edomir Markovi≤ (1985.72) in in-
terpreting both dates as termini post quos for the
end of Vin≠a C ceramics at this location, of 5205–
5165 cal BC (2% probability; Z-491) or 5075–4765
cal BC (93% probability), probably of 5000–4840 cal
BC (68% probability), and of 5210–5160 cal BC (1%
probability; Z-492) or 5080–4440 cal BC (92% prob-
ability) or 4425–4370 cal BC (2% probability), prob-
ably of 4930–4920 cal BC (1% probability) or 4910–
4545 cal BC (67% probability). It should be noted
that both measurements are from the Rudjer Bo∏ko-
vi≤ Institute, Zagreb, and it is likely that the mea-
surements are on average 77 radiocarbon years too
young (Horvatin≠i≤ et al. 1990.299).
The site of Donje Vranje in the southern Morava val-
ley was excavated in advance of road construction
in June 2013 (Kapuran et al. 2016). Several pits con-
taining Vin≠a B2 ceramics were recorded. A single
radiocarbon date on a disarticulated animal bone
was dated from pit 1, providing a date of 5295–5240
cal BC (9% probability; SUERC-57927) or 5235–5050
cal BC (86% probability), probably of 5220–5200 cal
BC (10% probability) or 5175–5070 cal BC (58%
probability) for the use of Vin≠a B2 ceramics on the
site.
The Danube downstream of Vin≠a-Belo Brdo
A single sample of unidentified charcoal has been
dated from pit 21 at Gornea, providing a terminus
post quem of 4895–4865 cal BC (2% probability;
BM-1124) or 4850–4585 cal BC (93% probability),
probably of 4825–4815 cal BC (3% probability) or
4805–4685 cal BC (65% probability) for the associ-
ated assemblage of Vin≠a A ceramics (Burleigh, Hew-
son 1979.350). A bulk sample of charred grain from
Liubcova (László 1997) has been dated from a con-
text associated with Vin≠a B2 pottery, dating it to
5320–4905 cal BC (95% probability; Bln-2133),
probably to 5225–5000 cal BC (68% probability).
The site of Rudna Glava is located in the village of
the same name on a steep slope above it. The loca-
tion is about 20km south of the Danube in the area
of the Iron Gorge (Jovanovi≤ 1982). Rudna Glava is
best known for its early copper metallurgy and was
the first copper ore mine to be excavated in the Bal-
kans. Discovered in the mid-1960s, the first archaeo-
logical survey took place in 1968, followed by sev-
eral seasons of excavations which uncovered multi-
ple prehistoric shafts cut vertically into the cliff face.
Fourteen radiocarbon measurements are available
on samples from Rudna Glava (Bori≤ 2009.Tab. 1;
Burleigh et al. 1982.255), but only eight of these
appear to be associated with Vin≠a ceramics14. These
have been assigned to the Gradac phase (Jovanovi≤
1982; 1994) which here we equate broadly with Mi-
loj≠i≤ phases C and D. The relationships between the
dated material and the ceramics, however, are not
straightforward. The deposition of hoards in which
the best preserved and diagnostically most sensitive
vessels were found was probably structured and
possibly ritualised. This may have taken place after
a long period of use of the mining shafts, and it is
also possible that the hoards represent an accumu-
lation over time rather than single events. In these
circumstances, we have interpreted the start of the
period when Vin≠a ceramics were deposited at Rud-
na Glava as a terminus post quem for the start of
Miloj≠i≤ phase C, and the end of the period when
Vin≠a ceramics were deposited at Rudna Glava as a
terminus post quem for the end of Miloj≠i≤ phase D.
A model for the mining activities at Rudna Glava
associated with Vin≠a C and D pottery is shown in
Figure 23. A weighted mean has been taken on the
statistically consistent measurements on antler RG8
before incorporation in the model (OxA-14624–5;
T’ = 0.2, T’(5%) = 3.8, ν = 1), which has good over-
all agreement (Amodel: 97; Fig. 23). This suggests
that Vin≠a C ceramics appeared here after 5745–
5240 cal BC (95% probability; tpq Rudna Glava C;
Fig. 23), probably after 5505–5325 cal BC (68%
probability), and suggests that Vin≠a D ceramics
were last deposited here after 4920–4445 cal BC
(95% probability; tpq Rudna Glava D; Fig. 23), prob-
ably after 4870–4685 cal BC (68% probability).
14 Three measurements (BM-1589, 6900±1000 BP, 8640–3780 cal BC (95% probability); OxA-14623, 7198±36 BP, 6205–6135 cal BC
(8% probability) or 6115–5995 cal BC (87% probability); OxA-14627, 6665±36 BP, 5645–5520 cal BC (95% probability)) appear
to indicate use of the mining complex in the earlier part of the seventh millennium cal BC. The collagen in all these samples, how-
ever, was poorly preserved and so there must be some doubt over the accuracy of these dates. One further sample appears to
indicate further use of the mines in the Copper Age (OxA-14676, 4273±32 BP, 3005–2990 cal BC (1% probability) or 2930–
2865 cal BC (91% probability) or 2805–2775 cal BC (3% probability)). Two of the Vin≠a period dates do not appear to be associ-
ated with ceramics and are not included in the model presented here (OxA-16585, 5816±35BP, 4780–4550 cal BC (95% proba-
bility); OxA-14699, 5974±39 BP, 4975–4765 cal BC (94% probability) or 4755–4740 cal BC (1% probability)).
A Vin;a potscape> formal chronological models for the use and development of Vin;a ceramics in south-east Europe
27
A timescape for Vin≠a pottery
We begin by constructing a chronological model, in
which the use of Vin≠a ceramics forms a continuous
and relatively constant phase of activity, opening
with the start of Miloj≠i≤ phase A and ending with
the end of Miloj≠i≤ phase D (Fig. 24). The start of Mi-
loj≠i≤ phase A is constrained to be earlier than the
start of Miloj≠i≤ phase B1 in this model; the start of
B1 is constrained to be earlier than the start of B2;
the start of B2 is constrained to be earlier than the
start of C; and the start of C is constrained to be ear-
lier than the start of D. The start of phase B (sites
where we are unsure whether the ceramics are phase
B1 or B2) is also constrained to be later than the
start of Miloj≠i≤ phase A. Since we are uncertain
whether the transition from one phase to the next
happened at the same time everywhere, we have
not constrained the end of phase A to be earlier than
the start of phase B1, for example. We have howev-
er constrained the end of Miloj≠i≤ phase A to be ear-
lier than the start of Miloj≠i≤ phase C and the end of
Miloj≠i≤ phases B1 and B2 (and the combined phase
B) to be earlier than the start of Miloj≠i≤ phase D.
We believe these constraints to be reasonable, since
there appear to be no closed assemblages containing
both Vin≠a A and Vin≠a C pottery, or both Vin≠a B
and Vin≠a D pottery15.
This approach means that we can examine whether
there is any overlap between the use of ceramics
assigned to successive phases in the Miljo≠i≤ scheme.
Where there are more than three radiocarbon dates
for a site, key parameters from the site-based mod-
els described above are imported as likelihoods in
the synthetic models. For example, in the model com-
ponent relating to the currency of Vin≠a A ceramics
(Fig. 25), Vin≠a Belo Brdo is represented by two
parameters – Belo Brdo start A and Belo Brdo A/B1
– which means that the 19 radiocarbon measure-
ments from samples recovered between the depths
of 9.3m and 8.0m in the Vasi≤ sequence are not dis-
proportionally weighted in the model in comparison
with sites with fewer dates. Where there are one or
two dates on short-lived materials securely associat-
ed with a ceramic phase, these are included fully in
the model (e.g., Deb-2579 from Satchinez; Fig. 25).
In other cases, where there are one or two results
on charred plant material that might have included
a component of old wood, the dates are modelled as
termini post quos for the end of the phase. These
distributions are shown in grey (e.g., Bln-435, a sam-
ple of oak charcoal from Predionica; Fig. 25). In a
few cases, dates are clearly misfits, and have been
excluded from the modelling (e.g., BM-1124, a sam-
ple of unidentified charcoal from Gornea) which is
more than 400 years later than any other sample
associated with Vin≠a A ceramics and must have con-
tained component of intrusive material. These dis-
tributions are shown in red and the parameter name
is followed by ‘?’.
The overall form of this chronological model (Model
1) is illustrated in Fig. 24 with its components illus-
trated in Figs. 25, 27, 29, 31–32 and 35. It is defined
exactly by the CQL2 keywords and structure provid-
ed as supplementary information (Vinca_Milojcic_
Model_1.oxcal).
Fig. 23. Probability distributions of radiocarbon dates from Rudna Glava. The format is identical to that
of Figure 7. The large square brackets down the left-hand side along with the OxCal keywords define the
overall model exactly.
15 As part of the preliminary modelling undertaken for this study, we constructed a model which included no relative sequence
between the phases of Vin≠a ceramics. This model suggests that start A is earlier than start B1 (87% probable), that start B1 is
earlier than start B2 (95% probable), that start B2 is earlier than start C (96% probable), and that start C is earlier than start
D (88% probable). This model also suggests that end A is earlier than start C (96% probable), and that end B1 and end B2 are
earlier than start D (85% and 62% probable respectively).
Whittle, Bayliss, Barclay, Gaydarska, Bánffy, Borić, Draşovean, Jakucs, Marić, Orton, Pantović, Schier, Tasić and Vander Linden
28
The component of this model relating to the curren-
cy of Vin≠a A ceramics is shown in Fig. 25. It sug-
gests that this ceramic phase begun in 5405–5310
cal BC (95% probability; start A; Fig. 25), probably
in 5365–5320 cal BC (68% probability). The use of
Vin≠a A ceramics ended in 5205–5070 cal BC (95%
probability; end A; Fig. 25), probably in 5200–5130
cal BC (68% probability). Vin≠a A ceramics were thus
in use for a period of 125–295 years (95% proba-
bility; duration A; Fig. 38), probably for 140–230
years (68% probability).
The Vin≠a network appears to have been established
very quickly – within the space of one or two human
generations. Sites as widely spaced as Szederkény-
Kukorica-dűlő and Belovode were probably founded
in the generation before 5300 cal BC (Figs. 25 and
26). Other sites came into being over the succeeding
century, including the Vin≠a settlement at Belo Brdo
itself, which as suspected by Lazarovici (1979.105–
106, Tab. 7, 122–123) was not among the very first
foundations. We have modelled radiocarbon dates
from sites from the northern and the western peri-
phery of the Vin≠a A network, as well as from its cen-
tre. Although there are known sites with character-
istic Vin≠a A forms in the southern and eastern peri-
phery of the network, either they are lacking radio-
carbon dates, or the association between pottery
phases and radiocarbon dates is not secure. Current-
ly Szederkény-Kukorica-dűlő in the extreme north-
west of the network is the earliest known dated site
with Vin≠a ceramics (on present evidence it is 60%
probable that it was established before Belovode:
Fig. 25). It is not possible to disentangle the direc-
tion of innovation without dating many more sites
at the generational precision provided by ToTL at
Szederkény-Kukorica-dűlő, Versend and Belo Brdo.
The network was established so quickly that we need
to trace the foundations of different sites to within
a few decades across the latter part of the 54th and
the earlier part of the 53rd century cal BC.
Recent formal modelling of the earliest (älteste)
LBK, however, allows us to compare the date when
the Vin≠a network was established with the time
when LBK ceramics rapidly spread across much of
north and north-west Europe. Jakucs et alii (2016.
Figs. 17–22) present three alternative models for
Fig. 24. Overall form of the chronological model for the phasing of Vin≠a pottery proposed by Miloj≠i≤
(1943) (Model 1). The format is identical to that of Figure 7. Components of this model are illustrated in
Figs. 25, 27, 29, 31–32 and 35. The model is defined exactly by the OxCal keywords and structure pro-
vided as supplementary information (Vinca_Milojcic_Model_1.oxcal).
A Vin;a potscape> formal chronological models for the use and development of Vin;a ceramics in south-east Europe
29
the time when the LBK spread beyond the inital
locus of its development in Transdanubia. Whatever
the dating of its formative phase, all these models
agree in placing this rapid expansion in the middle
decades of the 54th century cal BC (Fig. 41). This is
precisely the time when the first Vin≠a pottery was
made. The potential implications of this are explored
in the Discussion below.
The component of the model relating to the curren-
cy of Vin≠a B1 ceramics is shown in Fig. 27. It sug-
gests that this ceramic phase begun in 5355–5195
cal BC (95% probability; start B1; Fig. 27), proba-
bly in 5330–5245 cal BC (68% probability). The use
of Vin≠a B1 ceramics ended in 5045–4800 cal BC
(95% probability; end B1; Fig. 27), probably in
5040–4915 cal BC (68% probability). Vin≠a B1 cera-
mics were thus in use for a period of 185–505 years
(95% probability; duration B1; Fig. 38), probably
for 230–395 years (68% probability).
There is considerable uncertainty around the dating
of Vin≠a B1 ceramics since, with the exception of
Belo Brdo itself, all the sites are poorly dated. Such
data as we have, however, are consistent with a rel-
atively swift transition from Vin≠a A at the begin-
ning of the 52nd century cal BC. There was a period
of overlap (100% probable), as is reflected in Laza-
rovici’s (1991.19–20, 27) phase A3/B1. This period
of overlap lasted for 25–245 years (95% probabili-
ty; A/B1; Fig. 39), probably for 70–185 years (68%
probability).
Fig. 25. Probability distributions of dates associated with Miloj≠i≤ Vin≠a A ceramics (key parameters for
individual sites have been imported from the models defined or cited in Figs. 4–23). The format is iden-
tical to that of Figure 7 (grey tone indicates the sample has a possible old-wood offset; red indicates a
distribution excluded from the model). The model is defined exactly by the OxCal keywords and structure
provided as supplementary information (Vinca_Milojcic_Model_1.oxcal).
Whittle, Bayliss, Barclay, Gaydarska, Bánffy, Borić, Draşovean, Jakucs, Marić, Orton, Pantović, Schier, Tasić and Vander Linden
30
There is a marked increase in the density of sites
producing Vin≠a B1 ceramics (Fig. 28) in compari-
son to sites producing Vin≠a A ceramics (Fig. 26). It
appears, however, that Vin≠a ceramics are no longer
found upstream of the Danube confluence with the
Tisza, although their continuing influence can be
traced in the development of Sopot-type pottery.
Sites with Vin≠a pottery are still found along the Ti-
mis and Mures valleys.
The component of the model relating to the curren-
cy of Vin≠a B2 ceramics is shown in Fig. 29. It sug-
gests that this ceramic phase began in 5295–5105
cal BC (95% probability; start B2; Fig. 29), proba-
bly in 5285–5270 cal BC (3% probability), or in
5230–5130 cal BC (65% probability). The use of
Vin≠a B2 ceramics ended in 4920–4770 cal BC (95%
probability; end B2; Fig. 29), probably in 4905–4830
cal BC (68% probability). Vin≠a B2 ceramics were
thus in use for a period of 220–470 years (95% pro-
bability; duration B2; Fig. 38), probably for 270–
405 years (68% probability).
Again there was clearly a period of overlap (100%
probable) between Vin≠a B1 and Vin≠a B2 pottery.
This period of overlap lasted for 90–425 years (95%
probability; B1/B2; Fig. 39), probably for 130–310
years (68% probability).
The number of sites producing Vin≠a B2 pottery is
greater than those producing Vin≠a B1 pottery, and
its distribution extends further to the west and south-
east. Sites in the lower Tisza valley, however, appear
no longer to have access to Vin≠a pottery (Fig. 30),
perhaps because the Szahálhát network was expand-
ing from the north at this time16. The spatial distrib-
ution of the network appears to have reached its
greatest extent in this phase (possibly in the later
part of it, equivalent to Schier (1995) phase C1),
spreading from Kalo∏evi≤ in the far west as far as
Anzabegovo to the south-east and Ocna Sibilui to the
north-east.
The component of the model relating to the curren-
cy of Vin≠a pottery which can only be assigned to
Miloj≠i≤ phases B1 or B2 is shown in Figure 31. It
suggests that this combined phase began in 5360–
5235 cal BC (95% probability; start B; Fig. 31), prob-
ably in 5335–5275 cal BC (68% probability). This is
entirely compatible with the estimated date for the
Fig. 26. Map showing distri-
bution of sites from which
Miloj≠i≤ Vin≠a A ceramics
have been recorded (grey-
tone is approximately the
maximum spatial extent of
Vin≠a ceramics of all phas-
es). For sites with radiocar-
bon dates see references in
the text; for other sites see
Marinkovi≤ 1995; 2006; Vi-
lotijevi≤ 1965; Stalio 1970;
Jovanovi≤ 1965; Tasi≤ 1958;
1965; Benac 1988; Lekovi≤
1995; Ra∏ajski 1962; Jacano-
vi≤ 1988; Suciu 2009; Joano-
vi≤, Priki≤ 1978; Jacanovi≤,
∑or∂evi≤ 1990; Katunar
1988; Peri≤ 2006; Gara∏anin,
Gara∏anin 1979; Bogdanovi≤
1988a; Nikitovi≤ 1995; and
Stankovi≤ 1988; 1989. Râpas
is unpublished (A. Bărbat,
pers. comm.)*.
* We are aware of a number of sites in Bulgaria (e.g., Bulgarchevo) containing ceramic types which form part of the Vin≠a reper-
toire. They also occur, however, more widely in the area to the east of the Vin≠a network and so cannot be considered to be exclu-
sively diagnostic Vin≠a forms. This contrasts with sites (e.g., Satchinez) which contain diagnostic Vin≠a pottery in contexts mixed
with other clearly non-Vin≠a forms.
16 The Szakálhát culture evolves during Vin≠a B1 along the middle Tisza area. At the time of Vin≠a B2 Szakálhát and the closely
related Bucovăt group dominate south-east Hungary and the western part of the Romanian Banat, extending well into the Ser-
bian Vojvodina (Goldman 1985.64–66; Horváth 1983.90; Goldman, Goldman-Szenaszky 1991.194).
A Vin;a potscape> formal chronological models for the use and development of Vin;a ceramics in south-east Europe
31
start of Miloj≠i≤ phase B1 presented above (Fig. 27).
This combined phase ended in 5070–4770 cal BC
(95% probability; end B; Fig. 31), probably in 5010–
4825 cal BC (68% probability). This posterior date
estimate is entirely compatible with the estimated
date for the end of Miloj≠i≤ phase B2 presented
above (Fig. 29), but is considerably less precise.
The component of the model relating to the curren-
cy of Vin≠a C ceramics is shown in Fig. 32. One dis-
tribution has been excluded from the model. This is
the transition from Vin≠a C to Vin≠a D pottery at Pet-
nica, which is 150 years earlier than the same tran-
sition from anywhere else. This seems implausible,
since Petnica is a small occupation site which is not
even on one of the major rivers that clearly provid-
ed the connectivity of the network; moreover, the
stratigraphy of this site is not unproblematic (Orton
2012.16–20). Three distributions (shown in blue) for
assemblages that can only be classified as either Vin-
≠a C or Vin≠a D are included as termini ante quos
for the start of Vin≠a C. The model suggests that Vin-
≠a C began in 5115–4895 cal BC (95% probability;
start C; Fig. 32), probably in 5020–4905 cal BC
(68% probability). The use of Vin≠a C ceramics end-
ed in 4620–4400 cal BC (95% probability; end C;
Fig. 32), probably in 4550–4445 cal BC (68% prob-
ability). Vin≠a C ceramics were thus in use for a peri-
od of 325–650 years (95% probability; duration C;
Fig. 38), probably for 390–555 years (68% proba-
bility). Only the imprecisely dated sites at Belovode,
Selevac and Gornja Tuzla suggest that Vin≠a C ceram-
ics may have continued in use after 4700 cal BC.
The distribution of sites producing Vin≠a C pottery
(Fig. 33) is very similar to that of sites producing
Vin≠a B2 pottery (Fig. 30). Vin≠a ceramics are still
absent from sites in the lower reaches of the Tisza
valley but appear to occur on more sites in the Ti-
mis valley.
Where we have been able to date the transition from
Vin≠a B2 to Vin≠a C ceramics directly, the date esti-
mates fall in a tight group in the generations around
4900 cal BC (Fig. 34). There was clearly an overlap
between these two ceramic phases (100% probable).
This period of overlap lasted for 20–260 years (95%
probability; B2/C; Fig. 39), probably for 50–180
years (68% probability).
The component of the model relating to the curren-
cy of Vin≠a D ceramics is shown in Figure 35. It sug-
gests that this ceramic phase began in 4870–4725
cal BC (95% probability; start D; Fig. 35), probably
in 4810–4740 cal BC (68% probability). The use of
Vin≠a D ceramics ended in 4515–4360 cal BC (95%
probability; end D; Fig. 35), probably in 4490–4415
cal BC (68% probability). Vin≠a D ceramics were
thus in use for a period of 250–455 years (95% pro-
bability; duration D; Fig. 38), probably for 285–390
years (68% probability).
The distribution of sites with Vin≠a D ceramics is
clearly concentrated to the south of the Danube with
only a few sites in the southern Banat appearing on
the north bank (Fig. 36). No pure Vin≠a D assem-
blages occur in Transylvania but it seems that the
Fig. 27. Probability distributions of dates associated with Miloj≠i≤ Vin≠a B1 ceramics (key parameters for
individual sites have been imported from the models defined or cited in Figs. 4–23). The format is iden-
tical to that of Figure 7. The model is defined exactly by the OxCal keywords and structure provided as
supplementary information (Vinca_Milojcic_Model_1.oxcal).
Whittle, Bayliss, Barclay, Gaydarska, Bánffy, Borić, Draşovean, Jakucs, Marić, Orton, Pantović, Schier, Tasić and Vander Linden
32
distribution of this type of pottery reaches the Sofia
basin to the east.
Vin≠a D ceramics began after Vin≠a C pottery had
first appeared, and endured for longer (100% prob-
able) (Fig. 37). Vin≠a C ceramics may have been in
use for most of the currency of Vin≠a D types, if the
late dating of Vin≠a C at Belovode, Selevac and Gor-
nja Tuzla is robust (Fig. 32). The majority of dated
sites came to an end in the 46th or the earlier part
of the 45th century cal BC (Fig. 35).
A summary of the dating of the phases of Vin≠a pot-
tery suggested by Miloj≠i≤ (1943) is shown in Figure
37. Vin≠a pottery was clearly in use across a wide
area from the mid-54th century cal BC to the mid-
45th century cal BC.
Each of these phases endured for several centuries,
Vin≠a A being the briefest and Vin≠a C being the
longest (Fig. 38). Successive phases routinely over-
lapped by a century or two (Fig. 39), but phases C
and D overlapped by around 300 years. Either the
connectivity of the network was loosening in the
later centuries of its existence and there was more
diversity in contemporary pottery forms, or our dat-
ing of the latest sites with Vin≠a C ceramics (in par-
ticular Belovode, Gornja Tuzla and Selevac) is anom-
alous. The implications of this are discussed further
below.
Sensitivity analyses
We further explore the robustness of the results
from Model 1 by constructing an alternative model
which treats the five Miloj≠i≤ phases as sequential
and abutting (cf. Fig. 2). This means, for example,
that there is no ending to phase A and no beginning
to phase B1 but rather an abrupt transition from A
to B1. The overall form of this model is illustrated
in Fig. 40. The components of the model are similar
to those illustrated in Figs. 25, 27, 29, 31–2 and 35
but where we have formal estimates for the dates of
transitions at particular sites these have been com-
bined in the model. The model is defined exactly
by the CQL2 keywords and structure provided as
supplementary information (Vinca_Milojcic_Model_2.
oxcal).
We must make clear from the outset that this alter-
native model, Model 2, has poor overall agreement
between the radiocarbon dates and the strict sequen-
ce of non-overlapping abutting phases (Amodel: 4;
Fig. 40). This means that it is statistically implausi-
ble. Where the radiocarbon dates contradict the
typological sequence is, however, enlightening and
helps us to interpret the results of the preferred
model.
All the data in Miloj≠i≤ Phases A and B1 have good
agreement with a strict sequence between them, and
the two sites where we have been able to formally
Fig. 28. Map showing distri-
bution of sites from which
Miloj≠i≤ Vin≠a B1 ceramics
have been recorded (grey-
tone is approximately the
maximum spatial extent of
Vin≠a ceramics of all phas-
es). For sites with radiocar-
bon dates see references in
the text; for other sites see
Banner 1960; Grbi≤ 1950;
Drasovean 1996; Suciu 2009;
Gligor 2009; Lazarovici 1979;
Marinkovi≤ 1995; 2006; Ba-
bovi≤ 1992; Stalio 1970; Jo-
anovi≤, Priki≤ 1978; Joano-
vi≠ 1992; Jovanovi≤ 1965; Be-
nac 1988; Marijanovi≤ 1988a;
Mrkobrad 1982; Babovi≤
1986; Jacanovi≤, ∑or∂evi≤
1990; Jacanovi≤ 1988; Trbu-
hovi≤, Vasiljevi≤ 1970; Gara-
∏anin, Gara∏anin 1955; 1958;
1979; Peri≤ 2006; Nikitovi≤
1995; Bogdanovi≤ 1983; 1988a; Stankovi≤ 1989; Madas 1970; Valovi≤ 1983; 1987; Tasi≤ 1958; Krsti≤ 1959;
Tomi≤, Vukadin 1969; Gara∏anin, Ivanovi≤ 1958.
A Vin;a potscape> formal chronological models for the use and development of Vin;a ceramics in south-east Europe
33
model the A/B1 boundary (at Belovode and Belo Br-
do) have produced compatible date estimates for this
transition (Acomb = 108, An = 50, n = 2). It seems
likely therefore that the transition from Vin≠a A to
B1 occurs relatively swiftly within the first decades
of the 52nd century cal BC.
Three distributions have poor agreement with a
strict sequence between Miloj≠i≤ phases B1 and B2.
MAMS-22668, an animal bone from Potporanj (A:
13), and SUERC-57927, an animal bone from a pit
at Donje Vranje (A: 21), are both rather earlier than
would be expected if the strict succession of Vin≠a
B1 and B2 is valid everywhere. In Vin≠a B1, build
H3d+c from Uivar is rather later than would be
expected, on the same basis, for an assemblage of
Vin≠a B1 ceramics. We note, however, that this Uivar
assemblage is thought to be an imitation by local pot-
ters of Vin≠a B1 types, and so the archetype of the
imitation may not have been new when copied. A
variant of Model 2, which excludes build H3d+c
from Uivar, brings the dates from Potporanj and Do-
nje Vranje into good agreement with the sequence
(A: 87 and A: 88, respectively). The two sites where
we have calculated the transition from B1 to B2
directly (Belo Brdo and Selevac) have produced date
estimates that are compatible (Acomb = 106, An =
50, n = 2). On this basis, it again seems likely that
the transition from Vin≠a B1 to B2 was relatively
swift, in the decades around or just after 5100 cal
BC17.
All the data from Miloj≠i≤ phases B2 and C are com-
patible with a strict sequence between them, and
the three sites where we have been able to formal-
ly model the B2/C transition (Petnica, Belo Brdo
and Belovode) have produced date estimates that
are also compatible (Acomb = 123, An = 41, n = 3).
Model 2 therefore agrees with our inference from
Model 1 (see Fig. 34) that this transition occurred
swiftly in the decades around or slightly before
4900 cal BC. The data from Vin≠a C, however, are
not compatible with a strict sequence to Vin≠a D.
Posterior distributions for the end of Vin≠a C from
Belovode (A: 56), Selevac (A: 2) and Gornja Tuzla
(A: 0 and A: 47) are all later than would be expeted
Fig. 29. Probability distributions of dates associated with Miloj≠i≤ Vin≠a B2 ceramics (key parameters for
individual sites have been imported from the models defined or cited in Figs. 4–23). The format is iden-
tical to that of Figure 7. The model is defined exactly by the OxCal keywords and structure provided as
supplementary information (Vinca_Milojcic_Model_1.oxcal).
17 Early dates from Masinske Nijve and Ore∏kovica may perhaps impart caution as to the speed of this transition, since the poste-
rior density estimates for the start of both sites, currently assigned only generally to Vin≠a B, have poor agreement (A: 10 and
A: 40, respectively) with the strict succession from Vin≠a A to B.
Whittle, Bayliss, Barclay, Gaydarska, Bánffy, Borić, Draşovean, Jakucs, Marić, Orton, Pantović, Schier, Tasić and Vander Linden
34
if Vin≠a C ceramics ceased to circulate at the time
when Vin≠a D ceramics were first introduced. Of the
three sites where we have formally modelled the C/
D transition (Gomolava, Belo Brdo and Divostin18),
the date estimate for the transition at Gomolava is
substantively earlier (by almost a century; A: 2) than
the other two posterior distributions (Acomb = 76,
An = 41, n = 3).
It is not impossible that there was actually a swift
transition from Vin≠a C to Vin≠a D in the first
decades of the 47th century cal BC. The dates from
Selevac and Gornja Tuzla were on conventional
bulk samples of charred plant material that could
have contained an element of intrusive, later mate-
rial, although the date estimate from Gomolava
appears to be robust, since it rests on no fewer than
ten results on samples of animal bone (Fig. 12). If
the earliest sample associated with Vin≠a D at Gomo-
lava, OxA-14710, a bone tool, however, was resid-
ual, then the estimated date of the transition here
would shift more in line with those from Belo Brdo
and Divostin. On the whole, however, the cumula-
tive weight of these strands of evidence may be suf-
ficient to indicate that there was really a consider-
able overlap between the use of Vin≠a C and Vin≠a
D ceramics. This has implications for the connectiv-
ity of the cultural network in the final centuries of
its existence.
In conclusion, this sensitivity analysis suggests that,
within the precision obtainable by the data that are
currently available, the chest-of-drawers model of
sequential ceramic phases is not importantly wrong,
except perhaps in relation to the transition between
Vin≠a C and Vin≠a D, where a significant period of
concurrent use seems probable.
Discussion
Over 45 years later, the claim of Colin Renfrew
(1970.53) that analysis of the development of the
Vin≠a culture ‘represents one of the most challeng-
ing tasks in European prehistory’ remains valid. In
our introduction, we outlined many aspects of the
Vin≠a culture, but here we are going to concentrate,
first, on beginnings or origins, and secondly, on the
character of the developed stages of the phenome-
non. What insights do the better timings provided
by formally modelled estimates provide?
We need, first, to set out briefly the interpretive
frame within which better timings can be interrogat-
ed. As one of us has recently sketched (Bori≤ 2015.
159–162), three main approaches to the study of
the Vin≠a culture can be identified in past research:
culture-historical, processual and post-processual.
Each of these has given rather different answers to
the related questions of beginnings and the nature
Fig. 30. Map showing distri-
bution of sites from which
Miloj≠i≤ Vin≠a B2 ceramics
have been recorded (grey-
tone is approximately the
maximum spatial extent of
Vin≠a ceramics of all phas-
es). For sites with radiocar-
bon dates see references in
the text; for other sites see
Gligor 2009; G. Lazarovici
1979; Tasi≤ 1961; Marijano-
vi≤ 1988a; Jacanovi≤ 1988;
Madas, Brmboli≤ 1986; Jaca-
novi≤, πljivar 1995; Benac
1988; Marinkovi≤ 1995; Ta-
si≤, Tomi≤ 1969; Babovi≤
1986; Jacanovi≤, ∑or∂evi≤
1990; Borovi≤-Dimi≤ 1995;
Gara∏anin, Gara∏anin 1958;
1979; Peri≤ 2006; Nikitovi≤
1995; Bogdanovi≤ 1988a;
Stankovi≤ 1989; Madas 1970;
Valovi≤ 1983; 1987; Tasi≤
1958; Tomi≤, Vukadin 1969; Mari≤, Mari≤-Mirkovi≤ 2011.
18 We have already excluded the estimate for this transition at Petnica as anomalous: see above.
A Vin;a potscape> formal chronological models for the use and development of Vin;a ceramics in south-east Europe
35
of subsequent development. The culture-historical
approach has been dominant among researchers in
the region, though not universally so (see, for exam-
ple, Koro∏ec 1965). Out of the total range of Vin≠a
material culture, this approach has placed heavy
reliance on pottery typology (charted by Chapman
1981.Ch. 4). For origins, it sees disjuncture in longer-
term sequences, and derives fresh beginnings from
the outside, on a large scale and potentially from
long distances: in some accounts from as far afield
as Anatolia (Efe 2000; Gara∏anin 2000). There was
little further explicit reflection on the character and
conditions of different kinds of population displace-
ment and colonisation. Subsequent developments
were largely left to take care of themselves, except
in those variants which saw another wave of arrivals
of southern origin: the so-called Vin≠a C ‘shock’
already noted above. The approach did, however,
create coherent (though often varying) narratives
about the broad history of a perceived social entity.
The processual approach to the Vin≠a culture was
characteristically concerned with the playing out of
local or regional conditions. In the quotation above,
Colin Renfrew (1970.53) elaborated that the Vin≠a
culture ‘developed in these areas through the oper-
ation of local factors, and must be explained in
terms of these factors’, though he certainly took seri-
ously the previous claims for outside involvement.
In terms of origins, continuity was sought, though
on then imprecisely defined timescales, especially
in material sequences, including those of pottery,
and in landscape use (Chapman 1981.5, Ch. 4). Con-
sistent with this kind of focus, the concept of the
autonomous household was proposed as a major
agent of change in the processual narrative of the
development and eventual break-up of the Vin≠a cul-
ture (Tringham, Krsti≤ 1990b.581). Broad trends
rather than precise chronology were the temporal
currency. Post-processual approaches have focused
on selected situations. An important sketch of the
development of Vin≠a-Belo Brdo itself was offered,
with careful consideration given to the character
and sources of material in use at the tell; the result-
ing account asserting the early dominance and even-
tual slow decline of the site implies a wider narra-
tive, but does not explicitly document it (Chapman
2000.203–220). Other themes, broadly classifiable
as post-processual, have included the nature of depo-
sition and symbolic dimensions of houses (Bori≤
2008; 2015.162; Tripkovi≤ 2007; 2013; Crnobrnja
et al. 2009; Spasi≤, Ωivanovi≤ 2015). Wider narra-
tives of beginnings and subsequent development,
however, have often been laid aside in this change
of focus, and for the most part, chronological preci-
sion has not been a major concern of such contex-
tual approaches.
Other interpretive approaches might be advocated.
Culture evolutionary theory (as variously set out in,
for example, Shennan 2009; Bentley et al. 2011)
could be advocated (Bori≤ 2015.189–191). The
Fig. 31. Probability distributions of dates associated with Miloj≠i≤ Vin≠a B1 or B2 ceramics (key para-
meters for individual sites have been imported from the models defined or cited in Figs. 4–23). The for-
mat is identical to that of Figure 7 (grey tone indicates the sample has a possible old-wood offset). The
model is defined exactly by the OxCal keywords and structure provided as supplementary information
(Vinca_Milojcic_Model_1.oxcal).
Whittle, Bayliss, Barclay, Gaydarska, Bánffy, Borić, Draşovean, Jakucs, Marić, Orton, Pantović, Schier, Tasić and Vander Linden
36
recent fashion of ascribing much greater agency to
things – the ontological turn (among a host of oth-
ers, see Olsen 2010; Hodder 2012; Rivzi 2015; see
also Bori≤ 2015.191–193) – could be extended to
the rich material assemblages of the Vin≠a culture.
Here, we choose not to follow either. The former
tends to dehumanise complex processes of social
interaction, and it is hard to escape the neo-Darwi-
nian baggage that comes with it. The latter requires
a much longer debate than we have space for here.
It deliberately decentres human agency (see also
Van Dyke 2015), much more precise narratives for
which we are seeking to establish in this paper (cf.
Drasovean et al. submitted). If both approaches are
for further discussion elsewhere, what we have
found helpful are some of the terms and concepts
Fig. 32. Probability distributions of dates associated with Miloj≠i≤ Vin≠a C ceramics (key parameters for
individual sites have been imported from the models defined or cited in Figs. 4–23). The format is iden-
tical to that of Figure 7 (grey tone indicates that a sample has a possible old-wood offset; red indicates a
distribution excluded from the model; blue indicates a distribution modelled as a terminus ante quem).
The model is defined exactly by the OxCal keywords and structure provided as supplementary informa-
tion (Vinca_Milojcic_Model_1.oxcal).
A Vin;a potscape> formal chronological models for the use and development of Vin;a ceramics in south-east Europe
37
from ongoing debates on themes of hybridity and
ethnogenesis, in the field of the archaeology of colo-
nialism. As a proposition which is useful for us to
reflect upon, Matthew Liebmann (2013.27) has
asserted of the three primary means by which new
things come into being, that invention (“the cre-
ation of wholly original objects, styles and tech-
nologies”) is the rarest, that divergence (or evolu-
tion, by which “changes to an existing form through
time result in the eventual creation of multiple
separate types through splitting and branching”)
has been the focus of much archaeological attention,
and that convergence, or “creation through recom-
bination”, is “probably the most common”, though
less studied in past research by archaeologists. In
these terms, it seems clear that the culture-historical
approach to the Vin≠a culture has favoured some
version of invention for origins and kinds of diver-
gence for subsequent development, while the proces-
sual approach has largely plumped for divergence;
the possibility of convergence, in these terms, appears
largely to have been neglected.
This is not the place to review in detail all the com-
peting notions such as hybridity, acculturation, syn-
cretism, bricolage, creolisation and mestizaje, or the
advantages and disadvantages of their deployment
(Liebmann 2013; 2015; Silliman 2015; and refer-
ences), but it is useful to retain the sense of ground-
ed detail which such studies offer. Matthew Lieb-
mann (2015.331–334), for example, looks at the
production of soup plates within the repertoire of
Spanish-inspired Pueblo ceramics, over the specific
timeframe of 1598–1700, noting an interplay of
external and local factors in their design and dec-
oration. Barbara Voss (2015.665–666) associates
the wider notion of ethnogenesis with ‘historical
circumstances in which practices of identification
are structurally transformed’, often as the result of
‘substantive demographic shifts – aggregation, disag-
gregation, displacement, and migration – combined
with the emergence or imposition of new structures
of power’. She stresses, however, that ethnicity has
to be demonstrated rather than assumed (Voss 2015.
658), and questions the often lazy translation of
archaeological culture into ethnic entity (Voss 2015.
660).
We therefore seek to interpret the timings of Vin≠a
ceramics presented in this paper within an approach
that might broadly be classified as contextual and
practice-based. Our principal evidence is that of the
pottery itself, and it is important to stress that the
vast majority of Vin≠a ceramics have been found in
settlement contexts, and especially in association
with houses and dwelling horizons. There is a wealth
of information about later Vin≠a houses. Well-pre-
served contexts such as houses at Crkvine-Stubline,
part of a local network of sites of varying size, some
40km south-west of Belgrade and Vin≠a-Belo Brdo
Fig. 33. Map showing distri-
bution of sites from which
Miloj≠i≤ Vin≠a C ceramics
have been recorded (grey-
tone is approximately the
maximum spatial extent of
Vin≠a ceramics of all phas-
es). For sites with radiocar-
bon dates see references in
the text; for other sites see
Drasovean 1996; Gligor
2009; Jacanovi≤ 1988; Mari-
janovi≠ 1988b; Benac 1988;
Trbuhovi≤, Vasiljevi≤ 1970;
1975; Brukner 1961; Tasi≤,
Tomi≤ 1969; Babovi≤ 1986;
Joci≤ 1989; Borovi≤-Dimi≤
1995; Gara∏anin, Gara∏anin
1958; 1979; Peri≤ 2006; Niki-
tovi≤ 1987; 1995; Bogdano-
vi≤ 1988; Juri∏i≤ 1959; Krsti≤
1964; An∂elkovi≤-Despotovi≤,
Red∫i≤ 1991; Valovi≤ 1983;
1987; Mari≤, Mari≤-Mirkovi≤
2011; Madas, Brmboli≤ 1986; Lazi≤ et al. 1988; Todorovi≤ 1968; Bulatovi≤ 2003; Jacanovi≤, πljivar 1995; To-
mi≤, Vukadin 1969; Stalio, Galovi≤ 1956; Vukovi≤ et al. 2016; Dumitrescu 1980. Petresti is unpublished.
Whittle, Bayliss, Barclay, Gaydarska, Bánffy, Borić, Draşovean, Jakucs, Marić, Orton, Pantović, Schier, Tasić and Vander Linden
38
(Crnobrnja et al. 2009; Crnobrnja 2012), had lots
of pots in them: ‘dozens of ceramic vessels’ and ‘a
large number of storage vessels’ in house 1/2008
(Spasi≤, Ωivanovi≤ 2015.220). Some of these may
have been stacked on wooden shelves (Crnobrnja
et al. 2009.19). There are plentiful other examples,
such as at Divostin (Tripkovi≤ 2010). Occasionally
pots are found in the spaces outside, in between
houses (Nikoli≤, Vukovi≤ 2008). This household
setting is therefore presumably the context in which
Vin≠a pottery was principally used, for a wide range
of tasks, from food preparation (though the direct
evidence for that is surprisingly sparse), storage
and consumption. It may be that vessels most open
to public gaze, literally passable from hand to hand,
such as certain types of bowls, were the ones most
likely to change the fastest through time, as evi-
denced in the correspondence analyses of bowls
from Vin≠a-Belo Brdo (Schier 1996; 2000); these
might also have been the vessels most likely to break
(David 1972). Some forms, such as protome vessels,
seem to have remained more or less unchanged
through the sequence (Spasi≤, Crnobrnja 2014).
And it is important to remember that pottery was
only one element of the total materiality of the
house, and other constituents such as tables, hearths
Fig. 34. Probability distributions of dates associated with the introduction of Vin≠a C ceramics, derived
from Model 1.
Fig. 35. Probability distributions of dates associated with Miloj≠i≤ Vin≠a D ceramics (key parameters for
individual sites have been imported from the models defined or cited in Figs. 4–23). The format is iden-
tical to that of Figure 7 (grey tone indicates that a sample has a possible old-wood offset; red indicates a
distribution excluded from the model). The model is defined exactly by the OxCal keywords and structure
provided as supplementary information (Vinca_Milojcic_Model_1.oxcal).
A Vin;a potscape> formal chronological models for the use and development of Vin;a ceramics in south-east Europe
39
and storage facilities, which help to define a suite of
eating practices (Spasi≤, Ωivanovi≤ 2015), and the
layout of house interiors themselves, could have
been at least as important as pottery in the creation
and projection of social identities. So the context of
pottery use was probably primarily associated with
houses, but given the broad sharing of style seen in
the wide distributions (Figs. 26, 28, 30, 33, and 36),
on this basis alone it is extremely unlikely that
households were autonomous; it is far more likely
that groups of houses and neighbourhoods were
the nodes through which intense connections were
effected. On a note of caution, we have to recognise
that our knowledge so far of early Vin≠a domestic
contexts is much worse (though that helps to define
a clear goal for future research). The detailed record
for Vin≠a A contexts in Serbia, especially, is dire. By
way of brief example, Trnova≠a has poorly defined
wattle and daub structures (Jovanovi≤ 1965), while
Stagari is reported to have similar structures, but
these are not illustrated (Stankovi≤ 1989). Likewise,
while there is detailed information on the mix of
ceramic styles associated with particular longhouses
at Szederkény and Versend in south-west Hungary
(Jakucs et al. 2016; submitted), the material comes
from flanking pits, not house floors. This lacuna ham-
pers what we can say about the conditions of the
emergence of the Vin≠a ceramic style.
Another gap in knowledge concerns the making of
pottery. Few if any structures have been identified
as being associated with pottery production; the
northern part of the interior of house 8 at Banjica
has been suggested as a candidate (Tripkovi≤ 2007.
74, 119), but the evidence is ambiguous (Todorovi≤,
Cermanovi≤ 1961). Importantly, we do not know
whether all households made pottery, or if certain
potters served communities at large. Whilst more
data on chaînes opératoires and archaeometric ana-
lyses will help us refine our understanding of pot-
tery production as a whole, they are of limited use
to identify units and scales of production. Another
avenue worth exploring would be the identification
of differences in standardisation, as tentatively iden-
tified by Vukovi≤ (2011) between the sites of Vin≠a-
Belo Brdo and Motel Slatina, assuming that they
reflect levels of craft specialisation. We further note
a fragment of the perforated clay floor which sep-
arated the fire chamber from the pottery chamber in
a two-chambered kiln that was recovered from a
lower fill of enclosure ditch F1053 at Uivar. This
find provides a terminus ante quem for the use of
two-chambered kilns in the production of ceramics
of 4855–4755 cal BC (95% probability; dig F1053;
Schier et al. forthcoming.Fig. 6.17), probably of
4830–4780 cal BC (68% probability) – a time when
Vin≠a B2 ceramics were current (Schier 2005).
In turning then to beginnings and development, it
is obvious that we have gathered up a sample that
represents currently available dated sites, with a
range in both the quality and quantity of samples.
Fig. 36. Map showing distri-
bution of sites from which
Miloj≠i≤ Vin≠a D ceramics
have been recorded (grey-
tone is approximately the
maximum spatial extent of
Vin≠a ceramics of all phas-
es). For sites with radiocar-
bon dates see references in
the text; for other sites see
Jacanovi≤ 1988; Vukovi≤ et al.
2016; Marijanovi≤ 1988b; Be-
nac 1988; Crnobrnja et al.
2009; Mari≤ 2010; Todorovi≤
1966; Bogdanovi≤ 1988a;
1988b; Joci≤ 1989; Borovi≤-
Dimi≤ 1995; Gara∏anin, Gara-
∏anin 1979; Peri≤ 2006; Juri-
∏i≤ 1960; Bogdanovi≤ 1988a;
1988b; Todorova 1990; An-
∂elkovi≤-Despotovi≤, Red∫i≤
1991; Valovi≤ 1983; 1987;
Madas, Brmboli≤ 1986; Lazi≤
et al. 1988; Jacanovi≤, πljivar
1995.
Whittle, Bayliss, Barclay, Gaydarska, Bánffy, Borić, Draşovean, Jakucs, Marić, Orton, Pantović, Schier, Tasić and Vander Linden
40
Without doubt, a better dataset can be created in the
future. We have also raised questions about the
detail of the available typological schemes. Those
can also be refined in the future. But we believe that
the combination of more precise estimates of tim-
ings, order and duration, the distributions and den-
sity of known sites with diagnostic pottery through
successive phases, and information about context, is
sufficiently powerful to enable a preliminary recon-
sideration of important dimensions of the Vin≠a cul-
ture.
Beginnings
The modelled estimates for the appearance of Vin-
≠a A pottery given above suggest a rapid introduc-
tion. The earliest sites dated so far belong in the
north of the distribution. The overall density of sites
is not as great as in later phases. Simply in these
terms, neither the spatial pattern nor the temporal
distribution match what the formerly proposed big-
scale migrations might be expected to have pro-
duced. Those earlier hypotheses were never accom-
panied by explicit consideration of either the con-
ditions in the homeland of colonists or of the cir-
cumstances in which newcomers operated once ar-
rived in fresh territory. That has, however, been set
out elsewhere, and on a comparative (but not nec-
essarily universal) basis, there often seems to be a
process of initial scouting, followed by movement
by selective groups, rather than wholesale popula-
tions (Anthony 1990; 1997). To take just one specif-
ic example, Susan Alt (2006.290, 294) set out three
criteria for detecting possible migration into the
American Bottom at the onset of the Mississippian
phase – regional population growth, patterns of
diversity and the distribution of non-local pottery
types – but she also proposed a framework in which
both migration and local development could be seen
as components of the same process, involving con-
vergence and hybridity.
Now it certainly seems to be the case that ceramic
traits that overlap with the Vin≠a repertoire are
found widely in the southern Balkans and northern
Greece broadly in the same horizon as the appear-
ance of Vin≠a ceramics. This refers to the widely dis-
tributed dark burnished wares that are variously
represented among the ceramic assemblages of Ka-
ranovo III and the Late Neolithic or Tsangli phases
of Bulgaria and northern Greece (Georgiev 1961;
Lazarovici 1979.72–74; Efstratiou et al. 1998; Kot-
sakis 2014). But these dark wares are only part of
much wider repertoires; there are many painted
wares as well (Kotsakis 2014.61). To take just two
examples, unusual circumstances of preservation in
one context at Promachon-Topolnitsa on the Greek-
Bulgarian border produced decorated bark, part of
a wide range of pot forms and decorations (Koukou-
li-Chryssanthaki et al. 2007.Fig. 19), and there were
some painted vessels among the mainly dark wares
in phase II at Makri in Agean Thrace, informally
estimated as dating after 5500 cal BC (Efstratiou et
al. 1998.32). Informal estimates for the start of the
northern Greek Late Neolithic are certainly compat-
ible with the Vin≠a chronology set out above, but
relatively few sites have so far been radiocarbon
dated with rigour (see, for example, Dispilio: Faco-
rellis et al. 2014), and there has been little Bayesian
modelling so far of this horizon (Kostas Kotsakis,
pers. comm.). It does seem to be the case that in
these areas this horizon sees a marked increase in
the number of sites and their much wider distribu-
tion across the landscape, compared with earlier
settlement (Kotsakis 2014.60); the lakeside site of
Dispilio (Karkanas et al. 2011) can usefully stand as
a specific single example of this trend. So there is
no need to disconnect processes in the southern
Balkans and northern Greece from those operating
further north, but it now appears a very clumsy
explanation to derive all northern changes from the
south.
Fig. 37. Schematic diagram showing the currency of the different phases of Vin≠a ceramics proposed by
Miloj≠i≤ (1943) (the darker the shading the more probable that a ceramic phase was present in a partic-
ular 25-year period), derived from Model 1.
A Vin;a potscape> formal chronological models for the use and development of Vin;a ceramics in south-east Europe
41
The timing and distribution of Vin≠a A ceramics as
set out above can be argued to better align with
processual explanation of local or regional develop-
ment. But things may not have been as simple as
this, and that processual approach turns out, in our
view, to have been far too myopic: too constrained
within too narrow a spatial perspective. There is,
first, still the question of how material culture devel-
oped from the Star≠evo to the Vin≠a repertoire. Some
elements of local continuity have already been noted
(as proposed by Chapman 1981; Lekovi≤ 1990); the
Hungarian notion of ‘proto-Vin≠a’ elements (Mak-
kay 1990) has long been debated. On their own,
however, these seem insufficient, without a further
assumption of rapid and extensive material changes.
We know that there was a measurable gap in the
occupation of Vin≠a-Belo Brdo itself between the
Star≠evo and Vin≠a phases (Tasi≤ et al. 2016a;
2016b), but there is still much work to be done –
and beyond the remit of this paper – to date and
model the timing and tempo of this transition (see
also Por≠i≤ et al. 2016). What we can stress here is
the apparent synchronicity of the appearance of Vin-
≠a ceramics and the earliest LBK diaspora, as set out
above. Now, entirely indigenist explanations for the
appearance of the LBK (e.g., Whittle 1996) have not
stood the test of time, and the increasing weight of
aDNA evidence (now bolstered by genomic analysis)
supports a substantial, perhaps major, element of
new population dispersing into central and western
Europe (among others, see Brandt et al. 2013; D.
Hofmann 2015; Szécsényi-Nagy et al. 2015; Olalde
et al. 2015; Hofmanová et al. 2016; Lazaridis et al.
2016). On the evidence of Szederkény and Versend
in southern Hungary (Jakucs et al. 2016; submitted),
the 54th and 53rd centuries cal BC were also a time
of considerable cultural and material fluidity, in cir-
cumstances that also involved people on the move.
So it is probably better to see the transformations
which led to the emergence of Vin≠a material reper-
toires and practices in this kind of wider perspec-
tive, and a degree of convergence, as well as diver-
gence, should certainly be kept in mind. Given the
aDNA evidence both from the periphery of the Vin≠a
world (Hervella et al. 2015) and from the northern
part of the Carpathian basin (Szécsényi-Nagy et al.
2015; Gamba et al. 2014) for further shifts in the
patterns of haplotypes through the longer sequence
from the sixth to the fifth millennia cal BC, it would
be unwise to exclude entirely the possibility of some
population movement from the south to the north,
though that could also be derived from complex
displacements and shifts within the northern part
of the Carpathian basin. Further genomic analysis
(cf. Hofmanová et al. 2016; Lazaridis et al. 2016)
may help to clarify some of these issues in future
research.
We suggest that in place of ‘the old tired accounts
of migration and diffusion’ (Kotsakis 2014.65) a
much more complex and challenging account needs
now to be given of rapid and extensive change, that
has to be seen at local, regional and inter-regional
scales, which involved material transformations to
suit the changing circumstances of increasing pop-
ulations more widely distributed across their land-
scapes, and some of whom may have been caught
up in processes of movement and displacement.
Some strands in shifting material repertoires were
very widely distributed, but those can probably best
Fig. 38. Key parameters for the duration of the phases of Vin≠a ceramics proposed by Miloj≠i≤ (1943),
derived from Model 1.
Fig. 39. Key parameters for the overlap between phases of Vin≠a ceramics proposed by Miloj≠i≤ (1943),
derived from Model 1.
Whittle, Bayliss, Barclay, Gaydarska, Bánffy, Borić, Draşovean, Jakucs, Marić, Orton, Pantović, Schier, Tasić and Vander Linden
42
be further investigated as media through which
social connections across this very broad, shifting
world were expressed. As a discipline, we have been
comfortable debating intensely the beginnings of
the Neolithic, but we have been far less successful so
far in thinking through subsequent transformations.
Development
Increasing connectivity seems to define the next
phases of development, from Vin≠a B1 through Vin-
≠a D. though we have noted difficulties above in dif-
ferentiating between B1 and B2 assemblages in some
areas, it seems that the trend following the A phase
was towards the appearance of more sites in the
landscape. The greatest density is reached overall
already by the Vin≠a B2 phase, and that is then main-
tained in the Vin≠a C phase. On the ground, this coin-
cides, as noted also above for northern Greece and
the southern Balkans, with a spread in the distribu-
tions of sites across local landscapes (e.g., Chapman
1990), and with the appearance of more tells and
larger flat sites. At this point, we can indeed perhaps
best think in terms of local and regional factors at
work. This appears to have been an intensifying but
at the same time stable phenomenon, the timings
set out above showing long continuities at a land-
scape scale, and to varying degrees in particular loca-
tions. The notion of the ‘shock of Vin≠a C’ is proba-
bly now best translated into the peak connectivity
of the system, and the timings suggest that that was
anyway reached earlier, in Vin≠a B2. There appears
to be no particular need to invoke further population
movement at the start of the Vin≠a C phase, even
though the models set out above suggest a probably
quite rapid transition from Vin≠a B2 to Vin≠a C cera-
mics. But yet again, the possibility of some small-
scale displacements need not entirely be excluded.
It is also worth noting that the emergence of Vin≠a
C ceramics followed the end of the LBK, and may
have coincided with a hiatus between LBK and Hin-
kelstein in some parts of the Rhineland (Denaire et
al. in press); further east, there was probably more
continuity after the end of the LBK (Link 2014), but
as with beginnings, it is useful to keep an eye on the
wider situation and connections.
In our view, as already hinted at in the language
used throughout this paper, it is probably best in the
first place to see this developed Vin≠a phenomenon,
from Vin≠a B1 right through into Vin≠a D, as an ex-
tended social network. The overall distribution is
markedly riverine, and one of us has recently sketch-
ed the connectivity in terms of ‘various influences,
borrowings, emulations, exchange, trade and breed-
ing networks’ (Bori≤ 2015.158), underpinned by
flows of material into sites and played out in the
household settings described above. The culture his-
torical approach has tended to construct a further
sense of unifying or coherent group or ethnic iden-
tity. We have noted the warning of Voss (2015.660)
about equating ‘ethnonyms’ with real ethnicities,
and it was normal processual procedure to set that
kind of interpretation aside. But it has long been
apparent that the Vin≠a phenomenon was long-last-
ing and in that sense stable, which the more precise
timings set out in this paper serve only to underline.
Without removing local differences, there must have
been much that was shared throughout the Vin≠a
network, and from widespread, common practices
there surely came a shared way of acting in and un-
derstanding the world. The virtual absence of formal
or archaeologically visible mortuary practice through-
out the network and the wide distributions of altars,
anthropomorphic figurines (Tasi≤ 2016), face-lids
and masks, may also speak to widely shared ways of
thought. Cumulatively, this perhaps gets us closer
Fig. 40. Overall form of the chronological model for the phasing of Vin≠a pottery proposed by Miloj≠i≤
(1943) (Model 2). The format is identical to that of Figure 7. Components of this model are illustrated in
Figures 25, 27, 29, 31–32, and 35. The model is defined exactly by the OxCal keywords and structure pro-
vided as supplementary information (Vinca_Milojcic_Model_2.oxcal).
A Vin;a potscape> formal chronological models for the use and development of Vin;a ceramics in south-east Europe
43
to Voss’s (2015.658, quoting Vermeulen, Govers
1994) very general definition of ethnicity as a “con-
sciousness of difference”, “that is negotiated both
through external debates about differences between
‘us’ and ‘them’ and through internal contests over
community self-definition”. But it is far from clear
that the Vin≠a network had rigidly defined or sharp
boundaries, even though from Vin≠a B1 onwards dif-
ferent kinds of ceramics were preferred in southern
Hungary, for example. And whether the network at
its peak projected “ideologies of shared and diver-
gent history, ancestry, and tradition” (Voss 2015.
658), in contrast to or opposition to other contem-
porary cultural groupings, perhaps also remains a
moot point; wider connectivity seems to have been
at least as important as local or regional self-defini-
tion.
Endings
Some of the most intriguing questions raised by our
study relate to the Vin≠a D phase. Though this now
appears shorter than frequently suggested in the
past, it nonetheless represents overall a substantive
continuation of existing practice. Though this was
the phase in which some tells began to be aban-
doned, others, such as Vin≠a-Belo Brdo itself (Tasi≤
et al. 2015; 2016a; 2016b), continued to rise steadi-
ly, until endings perhaps mainly in the 46th centu-
ry cal BC. Simply on the basis of two well-dated sites,
Vin≠a-Belo Brdo and Uivar (Drasovean et al. sub-
mitted), it could be the case that house burnings
became more frequent in the last two centuries or
so of the Vin≠a network. Now there are many ways
to interpret house burnings, and there is not the
space here to set out again all the arguments for
scale and motivation (see Tasi≤ et al. 2015.1077–
1079). Suffice it to say that they could be seen as
one symptom of changing times and increased inter-
communal tensions and aggression. We have already
suggested that it seems insufficient to ascribe all
putative tensions to the activities of the autonomous
late Vin≠a household, as communal dimensions still
seem paramount, for example in contexts like Stubli-
ne, though we note the appearance of larger houses
in some contexts such as Banjica (Tripkovi≤ 2007).
The situation remains unclear, which the timings
given in this paper underline. There are probable
tensions, and these and perhaps other factors not
yet defined, or a combination of these possibilities,
appear to have brought the Vin≠a network and its
associated ways of living to an end. Continuing the
theme of tracking the Vin≠a and Danubian worlds in
parallel, the end of the Danubian world was proba-
bly staggered across the end of the Vin≠a network.
As with beginnings in the Vin≠a A phase, we still
badly need a much longer list of well-dated late Vin-
≠a sites, in order to unpick the sequence and tempo
of endings. Another puzzle remains. If the Vin≠a phe-
nomenon unravelled because of internal tensions
(however constituted), this does not seem to have
been accompanied by the same kind of material
regionalisation as seen for example in the late LBK
(e.g., Pechtl 2015). Some regionalisation is appar-
ent, and this paper has followed the categorisation
of the Turdas culture in Transylvania as distinct
from late Vin≠a culture (Drasovean 1996.28, 93–
94). But material culture including ceramics in the
Vin≠a D phase does not appear to project the kind
of much tighter regional groupings, perhaps reflec-
tive of increasing concern for local identities and
interests, which seem to characterise the end of the
LBK. We note again, however, the date estimates for
a greater overlap between Vin≠a C and Vin≠a D cera-
mics than was apparent with earlier phases, and per-
haps this greater variability could be an important
clue to follow further at a variety of regional and
local scales.
Fig. 41. Earliest Vin≠a A ceramics (cf. Fig. 25) in comparison to various readings of the dating of the LBK
diaspora (cf. Jakucs et al. 2016.Fig 24).
Whittle, Bayliss, Barclay, Gaydarska, Bánffy, Borić, Draşovean, Jakucs, Marić, Orton, Pantović, Schier, Tasić and Vander Linden
44
Conclusions and questions for future research
We conclude that this study indicates much more
complex circumstances for the appearance of the
Vin≠a phenomenon than envisaged in most previous
approaches; that it sketches a steadily intensifying
network, which reached its peak of connectivity
already in the Vin≠a B2 phase, which was main-
tained through the Vin≠a C phase and even into the
Vin≠a D phase, though there are some signs of
regionalisation in that late phase; and that Vin≠a
ceramics remained relatively stable in the Vin≠a D
phase, even though this is a time widely charac-
terised by many researchers as one of tensions and
even crisis. We have used the term network in pref-
erence to that of culture; the nature of the network
is compatible with widely shared practices and per-
haps beliefs, but connectivity rather than exclusive-
ness is characteristic, and sharp boundaries are not
apparent.
Although these conclusions help to redefine the
nature of the Vin≠a phenomenon, our study has rais-
ed at least as many questions as answers. In partic-
ular, it has highlighted a long list of questions about
the circumstances of beginnings and endings, but
there are also plenty of unresolved aspects of the
long, middle part of the sequence. To answer such
questions in future research, a series of challenges
need to be met. The typology of Vin≠a ceramics
could be much further refined (with no disrespect
intended towards the long list of heroic pioneers
and predecessors in this fundamentally important
field). Correspondence analysis of the bowls in the
upper portion of the Vin≠a-Belo Brdo stratigraphy
needs to be completed, and comparable exercises
could be carried out on other key assemblages across
the distribution of the network. More sites should
and could be radiocarbon dated, using short-life sam-
ples of known taphonomy from definable contexts,
to avoid age-offsets. All these tasks are feasible, and
exploitation of existing archives (as shown in the
study of Vin≠a-Belo Brdo: Tasi≤ et alii 2016a) is at
least as important as the application of best practice
in new excavations. The scale of new dating, if with-
in a refined system of typology and seriations, need
not be massive to be effective. There is every reason
to be optimistic that further refinements to the chro-
nology of the Vin≠a potscape, down to the precision
of lifetimes and generations, can realistically be
achieved in the next generation or two of research.
The supplementary information for the Vinca_Miloj
cic_Model_1.oxcal and Vinca_Milojcic_Model_2.oxcal
are available at http://dx.doi.org/10.4312/dp.43.1
We are grateful to Miroslav Birclin, Marcel Buri≤, Evanthia Kalogiropoulous, Dimitris Kloukinas, Kostas Kotsa-
kis, Krisztian Oross, Cosmin Suciu, and Johannes van der Plicht for providing information. Kirsty Harding help-
ed with the figures. Previously unpublished radiocarbon dates from Ore∏kovica were funded by the NERC
Radiocarbon Facility (grant NF/2014/2/15), and we thank Du∏ko πljivar, Bryan Hanks, Roger Doonan, Seren
Griffiths, Miroslav Ko≠i≤, and Jelena Bulatovi≤ for permission to cite them in advance of full publication. Un-
published radiocarbon dates from Masinske Njive were also funded by the NERC Radiocarbon Facility (grant
NF/2011/2/5); those from Poporanj were funded by the Ministry of Culture of the Republic of Serbia as part of
a project undertaken by the City Museum Vr∏ac; and those from At were funded by the Deutsche Forschungsge-
meinschaft through the Collaborative Research Center 806 ‘Our way to Europe’ as part of project undertaken in
collaboration by the Universities of Cologne and Belgrade. We are grateful to John Chapman for generously shar-
ing his encylopaedic knowledge of the Vin≠a world to improve various aspects of this paper, and for providing
access to his extensive library! The dating and modelling reported here have been carried out within The Times
of Their Lives, funded by the European Research Council (Advanced Investigator Grant 295412), and led by
Alasdair Whittle and Alex Bayliss.
ACKNOWLEDGEMENTS
A Vin;a potscape> formal chronological models for the use and development of Vin;a ceramics in south-east Europe
45
Alt S. M. 2006. The power of diversity: The roles of
migration and hybridity in culture change. In B. M. But-
ler, P. D. Welch (eds.), Leadership and polity in Mis-
sissippian society. Southern Illinois University Press.
Carbondale: 289–308.
An∂elkovi≤-Despotovi≤ Z., Red∫i≤ M. 1991. Arheolo∏ka is-
kopavanja neolitskog lokaliteta Ili≤a Brdo u selu ∞u≠uge.
Zbornik Narodnog muzeja u Beogradu 14: 93–98.
Anthony D. 1990. Migration in archeology: the baby and
the bathwater. American Anthropologist 92: 895–914.
1997. Prehistoric migration as social process. In J.
Chapman, H. Hamerow (eds.), Migrations and inva-
sions in archaeological explanation. British Archaeo-
logical Reports IS 664. Archaeopress. Oxford: 21–32.
Babovi≤ Lj. 1986. Zbradila – Korbovo: compte-rendu des
fouilles en 1981. ∑erdapske Sveske III: 95–115.
1992. Rezultati detaljnog rekognosciranja neolitskih
lokaliteta u ∏irem podru≠ju Be≠eja. Rad vojvo∂anskih
muzeja 34: 43–84.
Banner J. 1960. The Neolithic settlement on the Kreme-
njak hill at Csoka (∞oka). Acta Archaeologica Academiae
Scientarium Hungaricae 12: 1–56.
Bayliss A. 2015. Quality in Bayesian chronological mod-
els in archaeology. World Archaeology 47: 677–700.
Bayliss A. and 16 co-authors. 2016. Peopling the past: cre-
ating a site biography in the Hungarian Neolithic. Bericht
der Römisch-Germanischen Kommission 94: 23–91.
Benac A. 1988. Vinogradine. In Arheolo∏ki Leksikon Bos-
ne i Hercegovine Tom 2. Arheolo∏ka nalazi∏ta. Regija
1–13. Zemaljski Muzej Bosne i Hercegovine. Sarajevo: 68.
Bentley R. A., Earls M. and O’Brien M. 2011. I’ll have
what she’s having: mapping social behaviour. MIT Press.
London.
Berciu D. 1961. Contributii la problemele neoliticului în
Romînia în lumina noilor cercetări. Editura Academiei
Republicii Populare Romine. Bucuresti.
Biagi P., Spataro M. 2005. New observations on the radio-
carbon chronology of the Star≠evo-Cris and Körös cul-
tures. In L. Nikolova, J. Fritz and J. Higgins (eds.), Prehi-
storic Archaeology and Anthropological Theory and
Education. Reports of Prehistoric Research Projects 6–7.
Salt Lake City, Karlovo: 35–40.
Biagi P., Shennan S. and Spataro M. 2005. Rapid rivers
and slow seas? New data for the radiocarbon chronology
of the Balkan Peninsula. In L. Nikolova, J. Fritz and J. Hig-
gins (eds.), Prehistoric Archaeology and Anthropolo-
gical Theory and Education. Reports of Prehistoric Re-
search Projects 6–7. Salt Lake City, Karlovo: 41–52.
Biagi P., Gratuze B. and Boucetta S. 2007. New data on
the archaeological obsidian from Banat and Transylva-
nia. In M. Spataro, P. Biagi (eds.), A short walk to the Bal-
kans: the first farmers of the Carpathian Basin and the
adjacent areas. Museo Civico di Storia Naturale. Trieste:
129–148.
Bogdanovi≤ M. 1977. Prilog prou≠avanju bronzanog i sta-
rijeg gvozednog doba na podru≠ju centralne Srbije. Sta-
rinar 22: 145–156.
1983. Arheolo∏ka istra∫ivanja na podru≠ju centralne
Srbije. In B. Vitkovi≤ (ed.), Stani∏ta: zbornik radova za
politi≠ku, kulturnu i privrednu istoriju Kragujevca i
okoline. Istorijski arhiv πumadije. Kragujevac: 9–26.
1988a. Rajac – Donje Grbice. In D. Srejovi≤, N. Tasi≤
(eds.), Vin≠a and its world. Serbian Academy of Scien-
ces and Arts, Centre for Archaeological Research, Faculty
of Philosophy. Belgrade: 87–88.
1988b. Brdo – Kusovac. In D. Srejovi≤, N. Tasi≤ (eds.),
Vin≠a and its world. Serbian Academy of Sciences and
Arts, Centre for Archaeological Research, Faculty of
Philosophy. Belgrade: 71–72.
2004. Uvod. In M. Bogdanovi≤ (ed.), Grivac. Naselja
Protostar≠eva≠ke i Vin≠anske kulture. Centar za nau≠-
na istra∫ivanja SANU i Univerziteta u Kragujevcu, Narod-
ni Muzej Kragujevac. Kragujevac: 9–16.
Bori≤ D. 1996. Social dimensions of mortuary practices
in the Neolithic: a case study. Starinar 47: 67–83.
2008. First households and “house societies” in Euro-
pean prehistory. In A. Jones (ed.), Prehistoric Europe:
theory and practice. Wiley and Blackwell. Oxford: 109–
142.
2009. Absolute dating of metallurgical innovations in
the Vin≠a culture of the Balkans. In T. L. Kienlin, B. W.
Roberts (eds.), Metals and societies: studies in hon-
our of Barbara S. Ottaway. Habelt. Bonn: 191–245.
2015. The end of the Vin≠a world: modelling the Late
Neolithic to Copper Age transition and the notion of
archaeological culture. In S. Hansen, P. Raczky, A. An-
ders and A. Reingruber (eds.), Neolithic and Copper
Age between the Carpathians and the Aegean Sea:
chronologies and technologies from the 6th to the 4th
millennium BCE. Habelt. Bonn: 157–217.
References
Whittle, Bayliss, Barclay, Gaydarska, Bánffy, Borić, Draşovean, Jakucs, Marić, Orton, Pantović, Schier, Tasić and Vander Linden
46
Bori≤ D., Hanks B., πljivar D., Doonan R., Ko≠i≤ M., Bu-
latovi≤ J., Griffiths S. and Jacanovi≤ D. in preparation.
Enclosing the Neolithic world: an early Vin≠a culture
enclosed settlement in the Balkans.
Borojevi≤ K. 2006. Terra and Silva in the Pannonian
Plain: Opovo agro-gathering in the Late Neolithic. Bri-
tish Archaeological Reports IS 1563. Archaeopress. Oxford.
Borovi≤-Dimi≤ J. 1995. Vin≠anska glinena plastika sa La-
∂ari∏ta. Glasnik Srpskog arheolo∏kog dru∏tva 10: 104-
114.
Brandt G. and 17 co-authors. 2013. Ancient DNA reveals
key stages in the formation of central European mitochon-
drial genetic diversity. Science 342: 257–261.
Breunig P. 1987. C14 Chronologie des vorderasiatischen,
südost und mitteleuropäischen Neolitikums. Fundamen-
ta A13. Bohlau. Köln.
Bronk Ramsey C. 2008. Deposition models for chronolo-
gical records. Quaternary Science Reviews 27: 42–60.
2009. Bayesian analysis of radiocarbon dates. Radio-
carbon 51: 337–360.
Bronk Ramsey C., Lee S. 2013. Recent and planned devel-
opments of the program OxCal. Radiocarbon 55: 720–
730.
Brukner B. 1961. Beletinci, Obre∫ – naselje vin≠anske kul-
ture. Arheolo∏ki pregled 3: 17–21.
1980. Naselje vin≠anske grupe na Gomolavi (Neolitski
i Ranoeneolitski sloj). Izve∏taj sa iskopavanja 1967–
1976. Rad Vojvo∂anskih Muzeja 26: 5–54.
1988. Die Siedlung der Vin≠a-Gruppe auf Gomolava (Die
Wohnschicht des Spätneolithikums und Frühäneolithi-
kums (Gomolava Ia, Ia-b und II) und der Wohnhorizont
des äneolithischen Humus (Gomolava II)). In N. Tasi≤,
J. Petrovi≤ (eds.), Gomolava. Chronologie und Strati-
graphie der vorgeschichtlichen und antiken Kulturen
der Donauniederung und Südosteuropas. Vojvo∂anski
Muzej and Balkanolo∏ki Institut. Novi Sad, Belgrade: 19–
38.
Bulatovi≤ A. 2003. Nalazi vin≠anske kulture sa Fafosa i
Vala≠a. Vranje: Narodni muzej.
Buri≤ M. 2011. Gradac u Bapskoj – slika ∫ivota isto≠ne
Hrvatske prije 7000 godina. Filozofski fakultet, odsjek za
arheologiju. Zagreb.
2015. Problems of the late Neolithic chronology in East-
ern Croatia. In S. Hansen, P. Raczky, A. Anders and A.
Reingruber (eds.), Neolithic and Copper Age between
the Carpathians and the Aegean Sea: chronologies
and technologies from the 6th to the 4th millennium
BCE. Archäologie in Eurasien 31. Habelt. Bonn: 143–
156.
Buri≤ M., Te∫ak-Gregl T. 2009. Bapska, a Late Neolithic
settlement in Eastern Croatia – a new project. In F. Dra-
sovean, D. L. Ciobotaru and M. Maddison (eds.), Ten years
after: the Neolithic of the Balkans as uncovered by the
last decade of research. Editura Marineasa. Timisoara:
85–100.
Burleigh R., Hewson A. 1979. British Museum Natural Ra-
diocarbon Measurements XI. Radiocarbon 21: 339–352.
Burleigh R., Matthews K. and Ambers J. 1982. British Mu-
seum Natural Radiocarbon Measurements XIV. Radio-
carbon 24: 229–261.
Chapman J. 1981. The Vin≠a culture of south-east Europe:
studies in chronology, economy and society. British
Archaeological Reports IS 117. Archaeopress. Oxford.
1990. The Neolithic in the Morava-Danube confluence
area: a regional assessment of settlement patterns. In R.
Tringham, D. Krsti≤ (eds.), Selevac, a Neolithic village
in Yugoslavia. Institute of Archaeology. Los Angeles:
13–45.
2000. Fragmentation in archaeology: people, places
and broken objects in the prehistory of south eastern
Europe. Routledge. London.
Childe V. G. 1929. The Danube in prehistory. Clarendon
Press. Oxford.
Chu W., Mihailovi≤ D., Pantovi≤ I. Zeeden C., Hauck T. and
Lehmkuhl F. 2016. Archaeological excavations at the site
of At (Vr∏ac, Serbia). Antiquity 90. Project gallery 352.
∞ovi≤ B. 1961. Rezultati sondiranja na preistorijskom
naselju u Gornjoj Tuzli. Glasnik Zemaljskog muzeja u Sa-
rajevu 15–16: 79–139.
Crane H. R., Griffin J. B. 1972. University of Michigan Ra-
diocarbon Dates XIV. Radiocarbon 14: 155–194.
Crnobrnja A. N. 2012. Group identities in the central Bal-
kan Late Neolithic. Documenta Praehistorica 39: 155–
165.
Crnobrnja A., Simi≤ A. and Jankovi≤ M. 2009. Late Vin≠a
culture settlement at Crkvine in Stubline. Starinar 59: 9–
25.
David N. 1972. On the life span of pottery, type frequen-
cies, and archaeological inference. American Antiquity
37: 141–142.
A Vin;a potscape> formal chronological models for the use and development of Vin;a ceramics in south-east Europe
47
Denaire A., Lefranc P., Wahl J., Bronk Ramsey C., Dunbar
E., Goslar T., Bayliss A., Beavan N., Bickle P. and Whittle
A. in press. The cultural project: formal chronological mod-
elling of the Early and Middle Neolithic sequence in Lower
Alsace. Journal of Archaeological Method and Theory.
Dimitrijevi≤ S. 1968. Sopotsko-Len∂elska kultura. Arheo-
lo∏ki Institut, Filozofski Fakultet Sveu≠ili∏ta u Zagrebu.
Zagreb.
1974. Problem stupnjevanja star≠eva≠ke kulture s po-
sebnim obzirom na doprinos ju∫nopanonskih nalazi∏-
ta rje∏avanju ovog problema. Materijali 10: 59–122.
Drasovean F. 1993. Asezarea neoliticăde la Satchinez (jud.
Timis). Analele Banatului 2: 25–48.
1994. Die Stufe Vin≠a C im Banat. Germania 72: 409–
425.
1996. Cultura Vin≠a târzie în Banat. Relatiile cu ve-
cinii. Editura Mirton. Timisoara.
2013. In regards to certain Late Neolithic-Early Eneo-
lithic synchronism from Banat and Transylvania. A Ba-
yesian approach to published absolute dates. Studii de
Preistorie 10: 13–48.
2014. On the Late Neolithic and Early Eneolithic rela-
tive and absolute chronology of the Eastern Carpathian
Basin. A Bayesian approach. In W. Schier, F. Drasovean
(eds.), The Neolithic and Eneolithic in Southeast Eu-
rope. New approaches to dating and cultural dynam-
ics in the 6th to 4th millennium BC. Marie Leidorf. Rah-
den: 129–171.
Drasovean F., Schier W. 2009. The Neolithic tell sites of
Parta and Uivar (Romanian Banat): a comparison of their
architectural sequence and organization of social space.
In S. Hansen (ed.), Leben auf dem Tell als soziale Praxis.
Habelt. Bonn: 165–187.
Drasovean F., Schier W., Bayliss A., Gaydarska B. and Whit-
tle A. submitted. The lives of houses: duration, context and
history at Neolithic Uivar. European Journal of Archaeo-
logy.
Drasovean F., Teicu D. and Muntean M. 1996. Hodoni. Lo-
cuirile neolitice si necropola medievală. Muzeul Bana-
tului Montan. Resita.
Dumitrescu V. 1980. The Neolithic settlement at Rast.
British Archaeological Report IS 72. Archaeopress. Ox-
ford.
Efe T. 2000. Recent investigation in inland northwest
Anatolia and its contribution to Early Balkan-Anatolian
connections. In S. Hiller, V. Nikolov (eds.), Karanovo III:
Beiträge zum Neolithikum in Südosteuropa. Österreich-
isch-Bulgarische Ausgrabungen und Forschungen in
Karanovo. Phoibos Verlag. Vienna: 171–184.
Efstratiou N., Fumanal M. P., Ferrer C., Urem Kotsos D.,
Curci A., Tagliocozzo A., Stratouli G., Valamoti S. M., Nti-
nou M., Badal E., Madella M. and Skourtopoulou A. 1998.
Excavations at the Neolithic settlement of Makri, Thrace,
Greece (1988–1996) – a preliminary report. Saguntum
31: 11–62.
Facorellis Y., Sofronidou M. and Hourmouziadis G. 2014.
Radiocarbon dating of the Neolithic lakeside settlement of
Dispilio, Kastoria, northern Greece. Radiocarbon 56:
511–528.
Galovi≤ R. 1959. Predionica, neolitsko naselje kod Pri∏-
tine. Muzej Kosova i Metohije. Pri∏tina.
Gamba C. and 16 co-authors. 2014. Genome flux and sta-
sis during five millennia of Hungarian prehistory. Nature
Communications 5: 21 October 2014. doi: 10.1038/ncom
ms6257
Gara∏anin M. 1951. Hronologija vin≠anske grupe. Ar-
heolo∏ki Seminar. Univerza v Ljubljani. Ljubljana.
1979. Centralno-balkanska zona. In A. Benac (ed.), Pra-
istorija jugoslavenskih zemalja II – Neolitsko doba.
Akademija nauka i umjetnosti Bosne i Hercegovine. Cen-
tar za balkanolo∏ka ispitivanja. Sarajevo: 79–212.
1993. Zu den Problemen der Vin≠a-Gruppe in Ruma-
nien. Balcanica 24: 7–20.
2000. Zum Begriff des Balkanisch-Anatolischen Kom-
plexes des Späten Neolithikums. In S. Hiller, V. Nikolov
(eds.), Karanovo III: Beiträge zum Neolithikum in
Südosteuropa. Österreichisch-Bulgarische Ausgrabun-
gen und Forschungen in Karanovo. Phoibos Verlag.
Vienna: 343–347.
Gara∏anin M., Gara∏anin D. 1955. Neolitsko naselje u Ωar-
kovu. Starinar 3–4 (1952–1953): 107–126.
1958. Pavlovac kod Vranja, praistorijsko naselje. Stari-
nar 7–8(1956–1957): 398.
Gara∏anin D., Gara∏anin M. 1979. Supska “Stublina” –
praistorijsko naselje vin≠anske grupe. Narodni muzej.
Beograd.
Gara∏anin M., Ivanovi≤ V. 1958. Praistorija leskova≠kog
kraja. Narodni muzej. Leskovac.
Georgiev G. I. 1961. Kulturgruppen der Jungstein- under
Kupferzeit in der Ebene von Thrazien (Südbulgarien). In
J. Böhm, S. J. de Laet (eds.), L’Europe à la fin de l’âge de
la pierre. Éditions de l’Academie tchécoslovaque des Sci-
ences. Praha: 45–100.
Whittle, Bayliss, Barclay, Gaydarska, Bánffy, Borić, Draşovean, Jakucs, Marić, Orton, Pantović, Schier, Tasić and Vander Linden
48
Gimbutas M. 1974. Anza, ca. 6500–5000 B.C.: a cultural
yardstick for the study of Neolithic southeast Europe.
Journal of Field Archaeology 1: 26–66.
1976. Neolithic Macedonia as reflected by excavation
at Anza, southeast Yugoslavia. Institute of Archaeo-
logy. University of California. Los Angeles.
Giri≤ M. 1988. Geschichte der Archäeologischen Ausgra-
bungen auf Gomolava. In N. Tasi≤, J. Petrovi≤ (eds.), Go-
molava. Chronologie und Stratigraphie der Vorgeschi-
chtlichen und Antiken Kulturen der Donauniederung
und Südosteuropas. Vojvo∂anski Muzej and Balkanolo∏-
ki Institut SANU. Novi Sad–Belgrade: 13–17.
Gläser R. 1996. Zur absoluten Datierung der Vin≠a-Kultur
anhand von 14C-Daten. In F. Drasovean (ed.), The Vin≠a
culture, its role and connections. The Museum of Banat.
Timisoara: 175–212.
Gligor M. 2009. Asezarea neolitică si eneolitică de la
Alba lulia-Lumea Nouă în lumina noilor cercetări. Me-
ga. Cluj-Napoca.
Goldman G. 1985. Battonya-Gödrösök, eine neolithische
Siedlung in Südost-Ungarn. Kövizig. Békéscsaba.
Goldman G., Goldman-Szénászky J. 1991. Die Szakálhát-
Kultur am Rand des Vin≠a-Kreises. Banatica 11: 193–200.
Grbi≤ M. 1950. Gradi∏te kod Kikinde. Starinar 1: 113–118.
Hervella M., Rotea M., Izagirre N., Constaninescu M., Alon-
so S., Ioana M., Lazăr C., Ridiche F., Soficaru A. D., Netea
M. G. and de-la-Rua C. 2015. Ancient DNA from south-east
Europe reveals different events during Early and Middle
Neolithic influencing the European genetic heritage. PLoS
ONE 10(6): e0128810.
Hodder I. 2012. Entangled: an archaeology of the rela-
tionships between humans and things. Wiley-Blackwell.
Chichester.
Hofmann D. 2015. What have genetics ever done for us?
The implications of aDNA data for interpreting identity in
early neolithic central Europe. European Journal of Ar-
chaeology 18: 454–476.
Hofmanová Z. and 38 co-authors. 2016. Early farmers
from across Europe directly descended from Neolithic
Aegeans. Proceedings of the National Academy of Sci-
ences of USA 113: 6886–6891.
Holste F. 1939. Zur chronologische Stellung der Vin≠a-Ke-
ramik. Wiener Prähistorische Zeitschrift 26: 1–21.
Horváth L. 1983. Die neolithische Siedlung von Szentes-
Ilonapart. Dissertationes Archaeologica Ser. II – 12. ELTE.
Budapest.
Horváth F., Drasovean F. 2010. Remarks on the connec-
tions between the Banat and the Great Hungarian Plain
at the beginning of the Middle Neolithic (Satchinez-Alföld
Linear Pottery-Esztár-Vin≠a). Prehistoric Studies: 1–19.
Horváth F., Hertelendi E. 1994. Contribution to the 14C
based absolute chronology of the Early and Middle Tisza
region. Jósa András Múzeum Évkönyve 36: 111–133.
Horvatin≠i≤ N., Srdo≠ D., Obeli≤ B. and Krajcar Broni≤ I.,
1990. Radiocarbon dating of intercomparison samples at
the Zagreb Radiocarbon Laboratory. Radiocarbon 32:
295–300.
Ingold T. 1993. The temporality of the landscape. World
Archaeology 25: 152–174.
Jacanovi≤ D. 1988. Neolithic Sites in the Danubian Region
from the mouth of the Velika Morava to Golubac. In D.
Srejovi≤, N. Tasi≤ (eds.), Vin≠a and its world. Serbian Aca-
demy of Sciences and Arts. Centre for Archaeological Re-
search, Faculty of Philosophy. Belgrade: 111–119.
Jacanovi≤ D., ∑or∂evi≤ A. 1990. Vi∏eslojno praistorijsko
nalazi∏te Ora∏je u Dubravici. Viminacivm 4–5: 7–80.
Jacanovi≤ D., πljivar D. 1995. Rezultati arheolo∏kih istra-
∫ivanja neolitskog naselja Konju∏nica u Vite∫evu. Glasnik
Srpskog arheolo∏kog dru∏tva 10: 181–185.
Jakucs J., Voicsek V. 2015. The northernmost distribution
of the early Vin≠a culture in the Danube valley: a preli-
minary study from Szederkény-Kukorica-dűlő (Baranya
County, southern Hungary). Antaeus 33: 13–54.
Jakucs J., Bánffy E., Oross K., Voicsek V., Bronk Ramsey C.,
Dunbar E., Kromer B., Bayliss A., Hofmann D., Marshall P.
and Whittle A. 2016. Between the Vin≠a and Linearband-
keramik worlds: the diversity of practices and identities
in the 54th–53rd centuries cal BC in south-west Hungary
and beyond. Journal of World Prehistory Online First
(September 2016) http://link.springer.com/article/10.
1007/s10963-016-9096-x
Jakucs J., Oross K., Bánffy E., Voicsek V., Dunbar E., Rei-
mer P., Bayliss A., Marshall P. and Whittle A. submitted.
Rows with the neighbours: the short lives of long houses
at the Neolithic site of Versend-Gilencsa, Hungary. Anti-
quity.
Joanovi≠ π. 1992. Novi nalazi vin≠anske kulture u jugois-
to≠nom Banatu. Rad vojvo∂anskih muzeja 34: 33–42.
Joanovi≤ π., Priki≤ M. 1978. Pregled nalazi∏ta. In Lj. Relji≤
(ed.), Neolit ju∫nog Banata. Narodni muzej. Pan≠evo: 17–
30.
Joci≤ S. 1989. Vranin sala∏ – naselje vin≠anske kulture kod
Pan≠eva. Glasnik 2: 3–23.
A Vin;a potscape> formal chronological models for the use and development of Vin;a ceramics in south-east Europe
49
Jovanovi≤ B. 1965. Starija vin≠anska grupa u ju∫nom Ba-
natu. Iskopavanje lokaliteta Trnova≠a kod Barande. Rad
vojvo∂anskih muzeja 14: 15–41.
1982. Rudna glava. Najstarije rudarstvo bakra na
Centralnom Balkanu. Muzej Rudarstva i Metalurgije
and Arheolo∏ki Institut. Bor and Belgrade.
1994. Gradac phase in the relative chronology of the
late Vin≠a culture. Starinar 43–44: 1–11.
Jovanovi≤ B., Gli∏i≠, J. 1961. Fafos II, Kosovska Mitrovica,
naselje vin≠anske kulture. Arheolo∏ki pregled 3: 22–27.
Jovanovi≤ M. 2015. Gomolava, Hrtkovci – vin≠anska ne-
kropola. Rad Muzeja Vojvodine 57: 7–60.
Juri∏i≤ A. 1960. Za∏titna iskopavanja u selu Radoinji. Sta-
rinar 11: 99–111.
Kaiser T., Voytek B. 1983. Sedentism and economic
change in the Balkan Neolithic. Journal of Anthropolo-
gical Archaeology 2: 323–353.
Kapuran A., Bulatovi≤ A. and Milanovi≤ D. 2016. Donje
Vranje, naselje iz neolita i kasne antike. In S. Peri≤, A. Bu-
latovi≤ (eds.), Arheolo∏ka istra∫ivanja na autoputu E75.
Arheolo∏ki institute. Beograd: 115–134.
Karkanas P., Pavlopoulos K., Kouli K., Ntinou M., Tsartsi-
dou G., Facorellis Y. and Tsourou T. 2011. Palaeoenviron-
ments and site formation processes at the Neolithic lake-
side settlement of Dispilio, Kastoria, northern Greece.
Geoarchaeology: An International Journal 26: 83–117.
Katunar R. 1988. Majdan – Smederevska Palanka. In D.
Srejovi≤, N. Tasi≤ (eds.), Vin≠a and its world. Serbian Aca-
demy of Sciences and Arts. Centre for Archaeological Re-
search. Faculty of Philosophy. Belgrade: 81–82.
Kohl G., Quitta H. 1970. Berlin radiocarbon measure-
ments IV. Radiocarbon 12: 400–420.
Koro∏ec J. 1965. Nekateri problem podunavsko-balkan-
skog neolita. Situla 8: 165–175.
Kotsakis K. 2014. Domesticating the periphery: new re-
search into the Neolithic of Greece. Pharos 20: 41–73.
Koukouli-Chryssanthaki Ch., Todorova H., Aslanis I., Vaj-
sov I. and Valla M. 2007. Promachon-Topolnica. A Greek-
Bulgarian archaeological project. In H. Todorova, M. Ste-
fanovich and G. Ivanov (eds.), In the steps of James Har-
vey Gaul, volume 2: the Struma/Strymon river valley in
prehistory. Gerda Henkel Stiftung. Sofia: 43–78.
Krsti≤ D. 1959 Radaj≠e kod Mal≠e, neolitsko naselje. Sta-
rinar 9–10: 369–370.
1964. Neolitsko naselje u Vito∏evcu kod Ra∫nja. Zbor-
nik Narodnog muzeja u Beogradu 4: 51–63.
Laszló A. 1997. Datarea prin metoda radiocarbon în ar-
heologie. National Museum of Romanian History. Bucu-
resti.
Lazaridis I. and 53 co-authors. 2016. Genomic insights into
the origin of farming in the ancient Near East. Nature
536: 419–424.
Lazarovici C.-M. 2006. Absolute chronology of the late
Vin≠a culture in Romania and its role in the develop-
ment of the Early Copper Age. In N. Tasi≤, C. Grozdanov
(eds.), Homage to Milutin Gara∏anin. Serbian Academy
of Sciences and Arts and Macedonian Academy of Scien-
ces and Arts. Belgrade–Skopje: 277–293.
Lazarovici G. 1979. Neoliticul Banatului. Bibliotheca Mu-
sei Napocensis IV. Cluj-Napoca.
1981. Die Periodisierung der Vin≠a-Kultur in Rumänien.
Prähistorische Zeitschrift 56: 169–196.
1987. ‘Socul’ Vin≠a C în Transylvania. Acta Musei Poro-
lissensis 11: 33–55.
1991. Cultura Banatului. In G. Lazarovici and F. Draso-
vean (eds), Cultura Vin≠a în România. (Origine, evo-
lutie, legături, sinteze). Comisia muzeelor si colectii-
lor. Timisoara: 32–40.
2000. Vin≠a-Lengyel and Transylvania. Acta Musei Na-
pocensis 37: 7–20.
2009. The Zau culture. In F. Drasovean, D. L. Ciobota-
ru, D. L. and Maddison, M. (eds.), Ten years after: the
Neolithic of the Balkans, as uncovered by the last de-
cade of research. Editura Marineasa. Timisoara: 179–
217.
2010. Evolution, absolute and relative chronology of
the Zau culture. In P. Kalábková, B. Kovár, P. Pavúk
and J. πuteková (eds.), PANTA RHEI: studies in chro-
nology and cultural development of South-Eastern
and Central Europe in earlier prehistory presented
to Juraj Pavúk on the occasion of his 75th birthday.
Comenius University. Bratislava: 115–128.
2014. Cronologia absolute, relative si evolutia culturii
Zau. In S. M. Colesniuc, Cultura Zau. Editura Ex Ponto.
Constanta: 15–32.
Lazi≤ M., Sladi≤ M. and Katunar R. 1988. Medvednjak –
Smederevska Palanka. In D. Srejovi≤, N. Tasi≤ (eds.), Vin-
≠a and its world. Serbian Academy of Sciences and Arts.
Centre for Archaeological Research. Faculty of Philosophy.
Belgrade: 83–84.
Whittle, Bayliss, Barclay, Gaydarska, Bánffy, Borić, Draşovean, Jakucs, Marić, Orton, Pantović, Schier, Tasić and Vander Linden
50
Lekovi≤ V. 1990. The Vin≠anization of Star≠evo culture. In
D. Srejovi≤, N. Tasi≤ (eds.), Vin≠a and its world. Serbian
Academy of Sciences and Arts. Centre for Archaeological
Research. Faculty of Philosophy. Belgrade: 67–74.
1995. Neolitska Naselja. In Z. Vapa (ed.), Arheolo∏ka
istra∫ivanja du∫ autoputa kroz Srem. Pokrajinski za-
vod za za∏titu spomenika kulture. Novi Sad: 25–44.
Liebmann M. 2013. Parsing hybridity: archaeologies of
amalgamation in seventeenth-century New Mexico. In J.
J. Card (ed.), The archaeology of hybrid material culture.
Southern Illinois University Press. Carbondale: 25–49.
2015. The Mickey Mouse kachina and other ‘double ob-
jects’: hybridity in the material culture of colonial en-
counters. Journal of Social Archaeology 15: 319–341.
Linick T. W. 1977. La Jolla natural radiocarbon measure-
ments VII. Radiocarbon 19: 19–48.
Link T. 2006. Das Ende der neolithischen Tellsiedlungen.
Ein kulturgeschichtliches Phänomen des 5. Jahrtau-
sends v. Chr. im Karpatenbecken. Habelt. Bonn.
2014. Welche Krise? Das Ende der Linienbandkeramik
aus östlicher Perspektive. In T. Link, D. Schimmelpfen-
nig (eds.), No future? Brüche und Ende kultureller
Erscheinungen. Beispiele aus dem 6.–2. Jahrtausend
v. Chr. Welt an Erde. Kerpen-Loogh: 95–111.
Luca S. 2003. Date noi cu privire la cronologia absolută
a eneoliticului timpuriu din Transilvania – Rezultatele
prelucrării probelor radiocarbon de la Orăstie-Dealul Pe-
milor, punct X2, jud. Hunedoara. Banatica 16: 77–102.
Luca S., Diaconescu D., Georgescu A. and Suciu C. 2006.
Cercetarile arheologice de la Miercurea Sibiului – Petris
(jud. Sibiu), Campaniile anilor 1997–2005. Brukenthal
Acta Musei I: 9–20.
Madas D. 1970. Kraljevo polje, ≥uprija: neolitsko naselje.
Arheolo∏ki pregled 12: 16–18.
Madas D., Brmboli≤ M. 1986. Slatina, neolitsko, ranogvoz-
deno, rimsko i vizantijsko naselje. Arheolo∏ki pregled 26:
42–43.
Makkay J. 1990. The Protovin≠a problem – as seen from
the northernmost frontier. In D. Srejovi≤, N. Tasi≤ (eds.)
Vin≠a and its world. Serbian Academy of Sciences and
Arts. Centre for Archaeological Research. Faculty of Phi-
losophy. Belgrade: 113–122.
Mantu C. M. 2000. Relative and absolute chronology of
the Romanian Neolithic. Analele Banatului VII–VIII: 75–
105.
Mari≤ M. 2010. Arhitektura i idealna rekonstrukcija obje-
kata od lepa. Kolubara 5: 69–83.
2013. Za∏titna arheolo∏ka istra∫ivanja na lokalitetu Ja-
ri≠i∏te 1 (Rescue excavations on the Jari≠i∏te 1 site). In
V. Filipovi≤, R. Arsi≤ and D. Antonovi≤ (eds.), Rezultati
novih arheolo∏kih istra∫ivanja u severozapadnoj Sr-
biji i susednim teritorijama. Srpsko Arheolo∏ko Dru∏t-
vo and Zavod za za∏titu spomenika kulture Valjevo.
Beograd and Valjevo: 17–31.
Mari≤ M., Mari≤-Mirkovi≤ N. 2011. Obrenovac kod Dimit-
rovgrada, naselje vin≠anske kulture. Glasnik Dru∏tva kon-
zervatora Srbije 35: 65–68.
Marijanovi≤ B. 1988a. Malo brdo. In Arheolo∏ki Leksi-
kon Bosne i Hercegovine Tom 2. Arheolo∏ka nalazi∏ta.
Regija 1–13. Zemaljski Muzej Bosne i Hercegovine. Sara-
jevo: 120.
1988b. Gradi≤. In Arheolo∏ki Leksikon Bosne i Her-
cegovine Tom 2. Arheolo∏ka nalazi∏ta. Regija 1–13.
Zemaljski Muzej Bosne i Hercegovine. Sarajevo: 77.
Marinkovi≤ S. 1995 Za∏titna arheolo∏ka iskopavanja na
lokaliteti Mali alas kod ∞ente 1971. godine. Glasnik mu-
zeja Banata 6: 13–21.
Marinkovi≤ S. 2006. Matejski brod. Narodni muzej. Zre-
njanin.
Markoti≤ V. 1984. The Vin≠a culture. Western Publishers.
Calgary.
Markovi≤ ∞. 1985. Neolit Crne Gore. Centar za arheolo∏-
ka istra∫ivanja Filozofski fakultet u Beogradu. Zavod za
za∏titu spomenika kulture Crne Gore. Beograd.
Mazăre P. 2005. Artefacte din os descoperite în asezarea
neolitică de la Limba, jud. Alba. In C. I. Popa, G. T. Rustoiu
(eds.), Omagiu profesorului Ioan Andritoiu cu prilejul
împlinirii a 65 de ani. Studii si cercetări arheologice.
Alba Iulia: 237–312.
McPherron A., Srejovi≤ D. (eds.) 1988. Divostin and the
Neolithic of central Serbia. University of Pittsburgh. Pit-
tsburgh.
Menghin O. 1931. Weltgeschichte der Steinzeit. Anton
Schroll. Wien.
Merlini M. 2011. Mortuary practices and ritual grave to
consecrate a novel ancestor. In Gh. Lazarovici, C-M. Laza-
rovici and M. Merlini (eds.), Tărtăria and the sacred tab-
les. Editura Mega. Cluj-Napoca: 209–238.
Merlini M., Lazarovici Gh. 2008. Settling discovery circum-
stances, dating and utilization of the Tărtăria tablets. Acta
Terrae Septemcastrensis VII: 111–196.
A Vin;a potscape> formal chronological models for the use and development of Vin;a ceramics in south-east Europe
51
Miloj≠i≤ V. 1943. Das ältere Neolithikum in Serbien. Un-
published PhD thesis. Vienna University. Vienna.
1949a. Chronologie der jüngeren Steinzeit Mittel- und
Südosteuropas. Walter. Berlin.
1949b. South-eastern elements in the prehistoric civili-
zation of Serbia. Annual of the British School at Athens
44: 258–306.
Mrkobrad D. 1982. Boljevci kod Zemuna – vi∏eslojno na-
lazi∏te. Arheolo∏ki pregled 23: 28–31.
Nikitovi≤ L. 1987. Trsine, naselje vin≠anske i star≠eva≠ke
kulture. Zbornik radova Narodnog muzeja u ∞a≠ku 7:
5–28.
1995. Neolitski lokalitet Polj≠ine u Ostri kod ∞a≠ka.
Zbornik radova Narodnog muzeja u ∞a≠ku 25: 27–32.
Nikoli≤ D., Vukovi≤ J. 2008. Vin≠a ritual vessels: archaeolo-
gical context and possible meaning. Starinar 58: 51–69.
Olalde I. and 20 co-authors. 2015. A common genetic ori-
gin for early farmers from Mediterranean Cardial and Cen-
tral European LBK cultures. Molecular Biology and Evo-
lution 32: 3132–3142.
Olsen B. 2010. In defense of things: archaeology and
ontology of objects. Altamira Press. Lanham.
Orton D. 2010. Both subject and object: herding, inalien-
ability and sentient property in prehistory. World Archa-
eology 42: 188–200.
2012. Herding, settlement, and chronology in the Bal-
kan Neolithic. European Journal of Archaeology 15:
5–40.
Orton D., Gaastra J. and Vander Linden M. 2016. Between
the Danube and the deep blue sea: zooarchaeological
meta-analysis reveals variability in the spread and devel-
opment of Neolithic farming across the Western Balkans.
Open Quaternary. 2: 6.
Paluch T. 2011. Maroslele-Pana: egy középső neolitikus
lelőhely a kultúrák határvidékén (Maroslele-Pana: a
Middle Neolithic site at the frontier of cultures). Mono-
graphia Archaeologica 2. Szeged.
Parzinger H. 1993. Studien zur Chronologie und Kultur-
geschichte der Jungstein-, Kupfer- und Frühbronzezeit
zwischen Karpaten und Taurus. Römisch-germanische
Forschungen. Philipp von Zabern. Mainz.
Pechtl J. 2015. Linearbandkeramik pottery and society. In
C. Fowler, J. Harding and D. Hofmann (eds.), The Oxford
handbook of Neolithic Europe. Oxford University Press.
Oxford: 555–572.
Peri≤ S. 2006. The Gradac period in the Neolithic settle-
ments in the Middle Morava valley. In N. Tasi≤, C. Groz-
danov (eds.) Homage to Milutin Gara∏anin. Serbian Aca-
demy of Sciences and Arts, and Macedonian Academy of
Sciences and Arts. Belgrade–Skopje: 235–250.
Por≠i≤ M. 2011. An exercise in archaeological demogra-
phy: estimating the population size of Late Neolithic set-
tlements in the central Balkans. Documenta Praehisto-
rica 38: 323–332.
2012. Social complexity and inequality in the Late Neo-
lithic of the central Balkans: reviewing the evidence.
Documenta Praehistorica 39: 167–183.
Por≠i≤ M., Blagojevi≤ T. and Stefanovi≤ S. 2016. Demo-
graphy of the early Neolithic population in central Bal-
kans: population dynamics reconstruction using summed
radiocarbon probability distributions. PLoS ONE August
10, 2016. doi 10.1371/journal.pone.0160832
Quitta H., Kohl G. 1969. Neue Radiocarbondaten zum Neo-
lithikum und zur frühen Bronzezeit Südosteuropas und
der Sowjetunion. Zeitschrift für Archäologie 3: 223–255.
Ra∏ajski R. 1962. Beluca, Pavli∏, Vr∏ac – vi∏eslojno prais-
torijsko naselje. Arheolo∏ki pregled 4: 26–28.
Reimer P. J. and 29 co-authors. 2013. IntCal13 and Ma-
rine13 radiocarbon age calibration curves 0–50,000 years
cal BP. Radiocarbon 55: 1869–1887.
Renfrew C. 1970. The place of the Vin≠a culture in Eu-
ropean prehistory. Zbornik radova Narodnog muzeja u
Beogradu 6: 45–57.
Rivzi U. Z. 2015. Crafting resonance: empathy and belong-
ing in ancient Rajasthan. Journal of Social Archaeology
15: 254–273.
Sanev V., Stamenova M. 1989. Neolitska naselba “Strana-
ta” vo selo Angelci. Zbornik na trudovi na πtipskiot mu-
zej 9: 63.
Schier W. 1995. Vin≠a-Studien: Tradition und Innovation
im Spätneolithikum des zentralen Balkanraumes am
Beispiel der Gefäßkeramik aus Vin≠a-Belo Brdo. Unpub-
lished Habilitation thesis. Heidelberg University. Heidel-
berg.
1996. The relative and absolute chronology of Vin≠a:
new evidence from the type site. In F. Drasovean (ed.),
The Vin≠a culture, its role and cultural connections.
The Museum of Banat. Timisoara: 141–62.
2000. Measuring change: the Neolithic pottery sequence
of Vin≠a-Belo Brdo. Documenta Praehistorica 27: 187–
197.
Whittle, Bayliss, Barclay, Gaydarska, Bánffy, Borić, Draşovean, Jakucs, Marić, Orton, Pantović, Schier, Tasić and Vander Linden
52
2008. Uivar: a late Neolithic-early Eneolithic fortified
tell site in western Romania. In D. Bailey, A. Whittle and
D. Hofmann (eds.), Living well together? Settlement
and materiality in the Neolithic of south-east and cen-
tral Europe. Oxbow Books. Oxford: 54–67.
2014. The Copper Age in southeast Europe – historical
epoch or typo-chronological construct? In W. Schier, F.
Drasovean (eds.), The Neolithic and Eneolithic in south-
east Europe: new approaches to dating and cultural
dynamics in the 6th to 4th millennium BC. Marie Lei-
dorf. Rahden: 419–435.
2016. Uivar “Gomila” und der Übergang vom Spätneo-
lithikum zur Kupferzeit im südöstlichen Karpaten-be-
cken im Vergleich zur Unteren Donau und dem westli-
chen Schwarzmeergebiet. In V. Nikolov, W. Schier (eds.),
Der Schwarzmeerraum vom Neolithikum bis in die
Früheisenzeit (6000–600 v. Chr.): Kulturelle Interfe-
renzen in der zirkumpontischen Zone und Kontakte
mit ihren Nachbargebieten. Marie Leidorf. Rahden: 75–
91.
Schier W., Drasovean F. 2004. Vorbericht über die rumä-
nisch-deutschen Prospektionen und Ausgrabungen in der
befestigen Tellsiedlung von Uivar, jud. Timis, Rumänien
(1998–2002). Praehistorische Zeitschrift 79: 145–230.
Schier W., Drasovean F., Bayliss A., Gaydarska B. and Whit-
tle A. forthcoming. Chapter 6: Scientific dating and chrono-
logical modelling. In W. Schier, F. Drasovean (eds.), Uivar
“Gomila” – a prehistoric tell settlement in Romanian Ba-
nat. Vol. I: site, architecture, stratigraphy and dating.
Shennan S. 2009. Pattern and process in cultural evolu-
tion: an introduction. In S. Shennan (ed.), Pattern and
process in cultural evolution. University of California
Press. Berkeley: 1–18.
Silliman S. W. 2015. A requiem for hybridity? The prob-
lem with Frankensteins, purées, and mules. Journal of
Social Archaeology 15: 277–298.
Spasi≤ M., Crnobrnja A. N. 2014. Vin≠a bowls with pro-
toma. Starinar 64: 185–203.
Spasi≤ M., Ωivanovi≤ S. 2015. Foodways architecture: stor-
ing, processing and dining structures at the Late Neolithic
Vin≠a culture site at Stubline. Documenta Praehistorica
42: 219–230.
Srdo≠, D., Sliep≠vi≤ A. and Planinc J. 1975. Rudjer Bo∏ko-
vi≤ Institute radiocarbon measurements III. Radiocarbon
17: 149–155.
Srdo≠ D., Sliep≠evi≤ A., Obeli≤ B. and Horvatin≠i≤ N. 1977.
Rudjer Boskovic Institute radiocarbon measurements IV.
Radiocarbon 19: 465–475.
Srdo≠ D., Horvatin≠i≤ N., Obeli≤ B., Krajcar Broni≤ I. and
Sliep≠evi≤ A. 1987. Rudjer Boskovic Institute radiocarbon
measurements IX. Radiocarbon 29: 115–134.
Stalio B. 1964. Novi metalni nalazi sa Plo≠nika. Zbornik
Narodnog muzeja u Beogradu IV: 35–41.
1970. Aradac: “Kameniti vinogradi”. Zbornik Narod-
nog muzeja u Beogradu 6: 73–81.
1973. ∞etvrti nalaz bakarnog i kamenog oru∂a sa Plo≠-
nika kod Prokuplja. Zbornik Narodnog muzeja u Beo-
gradu VII: 157–161.
Stalio B., Galovi≤ R. 1956. Naprelje, neolitsko naselje kod
Novog Pazara. Narodni muzej. Beograd.
Stankovi≤ S. 1988. Ornice – Makre∏ani. In D. Srejovi≤, N.
Tasi≤ (eds.), Vin≠a and its world. Serbian Academy of Sci-
ences and Arts. Centre for Archaeological Research. Fa-
culty of Philosophy. Belgrade: 85–86.
Stankovi≤ S. 1989. Neolitsko naselje kod sela Stragara.
Glasnik Srpskog arheolo∏kog dru∏tva 5: 106–113.
Starovi≤ A. 1993. Analiza kerami≠kih artefakata sa neo-
litskog naselja u Petnici. Petni≠ke Sveske 29: 34–72.
Stuiver M., Reimer P. 1993. Extended 14C database and
revised CALIB radiocarbon calibration program. Radio-
carbon 35: 215–230.
Suciu C. 2009. Cultura Vin≠a in Transylvania. ALTIP.
Alba Iulia.
2011. Early Vin≠a culture dynamic in south-eastern
Transylvania. In S. Mills, P. Mirea (eds.), The lower
Danube in prehistory: landscape changes and human-
environment interactions. Editura Renaissance. Bucu-
resti: 75–85.
Szécsényi-Nagy A. and 27 co-authors. 2015. Tracing the
genetic origin of Europe’s first farmers reveals insights
into their social organization. Proceedings of the Royal
Society B 282/1085, 2015, 20150339. DOI: 10.1098/rspb.
2015.0339.
πljivar D., Jacanovi≤ D. 1996. Veliko Laole, “Belovode” na-
selje vin≠anske grupe. Glasnik Srpskog arhelo∏kog dru∏t-
va 11: 185–189.
πljivar D., Kuzmanovi≤ Cvetkovi≤ J. P. and Ωivkovi≤ J.
2010. Belovode, Plo≠nik; on copper metallurgy in the Vin-
≠a culture. Zbornik Narodnog muzeja u Beogradu XX:
27–45.
Tasi≤ N. 1958. Ωitkovac i neki problemi relativnog hrono-
lo∏kog odnosa neolitskih i eneolitskih naselja na Kosovu
A Vin;a potscape> formal chronological models for the use and development of Vin;a ceramics in south-east Europe
53
i u dolini Ibra. Glasnik Muzeja Kosova i Metohije 3: 11–
46.
1960. Zavrna istraivanja na praistorijskom naselju kod
Valaca. Glasnik Muzeja Kosova i Metohije 45: 11–82.
1961. Obala Dunava, Zemun-praistorijska naselja. Ar-
heolo∏ki pregled 3: 29–31.
1965. Mala strana, La≤arak, Sremska Mitrovica – neolitsko
naselje vin≠anske grupe. Arheolo∏ki pregled 7: 14– 16.
1988. Comparative C-14 dates for the Neolithic settle-
ments in Serbia. In D. Srejovi≤, N. Tasi≤ (eds.), Vin≠a
and its world. Serbian Academy of Sciences and Arts.
Centre for Archaeological Research. Faculty of Philo-
sophy. Belgrade: 45–47.
2011. Anthropomorphic figurines from Vin≠a excava-
tions 1998–2009. Documenta Praehistorica 38: 149–
157.
2016. Vin≠a figurines: identities hidden and revealed.
In K. Ba≠varov, R. Gleser (eds.), Southeast Europe and
Anatolia in prehistory: essays in honor of Vassil Niko-
lov on his 65th anniversary. Habelt. Bonn: 357–373.
Tasi≤ N., Mari≤ M., Penezi≤ K., Filipovi≤ D., Borojevi≤ K.,
Bori≤ D., Reimer P., Russell N., Bayliss A., Barclay A., Gay-
darska B. and Whittle A. 2015. The end of the affair: for-
mal chronological modelling for the top of the Neolithic
tell of Vin≠a-Belo Brdo. Antiquity 89: 1064–1082.
Tasi≤ N., Mari≤ M., Bronk Ramsey C., Kromer B., Barclay A.,
Bayliss A., Beavan N., Gaydarska B. and Whittle A. 2016a.
Vin≠a-Belo Brdo, Serbia: the times of a tell. Germania 93:
1–76.
Tasi≤ N., Mari≤ M., Filipovi≤ D., Penezi≤ K., Dunbar E., Rei-
mer P., Barclay A., Bayliss A., Gaydarska B. and Whittle A.
2016b. Interwoven strands for refining the chronology of
the Neolithic tell of Vin≠a-Belo Brdo, Serbia. Radiocarbon
58/4, published online: 30 August 2016. Doi: 10.1017/
RDC.2016.56
Tasi≤ N., Srejovi≤ D. and Stojanovi≤ B. 1990. Vin≠a – cen-
tre of the Neolithic culture of the Danubian region. Cen-
tre for Archaeological Research, Faculty of Philosophy.
Belgrade.
Tasi≤ N., Tomi≤ E. 1969. Crnokala≠ka Bara. Naselje star-
≠eva≠ke i vin≠anske kulture. Narodni Muzej Kru∏evac.
Arheolo∏ko dru∏tvo Jugoslavije. Beograd.
Todorova H. 1990. Ein mittelaneolithisches Haus aus
Poduene (Sofia) aus der Vinca D2 Zeit. In D. Srejovi≤, N.
Tasi≤ (eds.), Vin≠a and its world. Serbian Academy of Sci-
ences and Arts. Centre for Archaeological Research. Fa-
culty of Philosophy. Belgrade: 155–165.
Todorovi≤ J. 1966. Kremenite njive, Barajevo, Obrenovac
– neolitsko naselje. Arheolo∏ki pregled 8: 18–19.
1968. Grabovac, ∑uri≤a vinogradi, Obrenovac – naselje
star≠eva≠ke i vin≠anske kulture. Arheolo∏ki pregled 11:
11–13.
Todorovi≤ J., Cermanovi≤ A. 1961. Banjica, naselje vin-
≠anske grupe. Muzej grada Beograda. Beograd.
Tomi≤ E., Vukadin O. 1969. Vitkovo, Aleksandrovac – na-
selje vin≠anske kulture. Arheolo∏ki pregled 11: 21–22.
Trbuhovi≤ V., Vasiljevi≤ M. 1970. Jela, πabac – praistorij-
sko naselje. Arheolo∏ki pregled 12: 23–25.
1975. „Obrovci“ poseban tip neolitskih naselja u za-
padnoj Srbiji. Starinar 24–25: 157–161.
Tringham R., Brukner B., Kaiser T., Borojevi≤ K., Russell N.,
Stehli P., Stevanovi≤ M. and Voytek B. 1992. Excavations
at Opovo 1985–7: socioeconomic change in the Balkan
Neolithic. Journal of Field Archaeology 19: 351–86.
Tringham R., Brukner B. and Voytek B. 1985. The Opovo
project: a study of socioeconomic change in the Balkan
Neolithic. Journal of Field Archaeology 12: 425–444.
Tringham R., Krsti≤ D. 1990a. Introduction: The Selevac
Archaeological Project. In R. Tringham and D. Krsti≤ (eds.),
Selevac: a Neolithic village in Yugoslavia. University of
California Press. Los Angeles: 1–12.
1990b. Selevac and the transformation of southeast Eu-
ropean prehistory. In R. Tringham, D. Krsti≤ (eds.), Se-
levac: a Neolithic village in Yugoslavia. University of
California Press. Los Angeles: 567–616.
(eds.) 1990c. Selevac: a Neolithic village in Yugosla-
via. University of California Press. Los Angeles.
Tripkovi≤ B. 2007. Doma≤instvo i prostor u kasnom neo-
litu. Vin≠ansko naselje na Banjici (Household and space
in the Late Vin≠a settlement at Banjica). Srpsko arheo-
lo∏ko dru∏tvo. Beograd.
2010. House(hold) continuities in the central Balkans,
5300–4600 BC. Opuscula Archaeologica 33: 7–28.
2011. Containers and grains: food storage and symbo-
lism in the central Balkans (Vin≠a period). Documenta
Praehistorica 38: 159–172.
2013. Doma≤instvo i zajednica. Ku≤ne i naseobinske
istorije u kasnom neolitu Centralnog Balkana. Filo-
zofski fakultet. Univerzitet u Beogradu. Beograd.
Tripkovi≤ B, Mili≤ M. 2009. The origin and exchange of
obsidian from Vin≠a-Belo Brdo. Starinar 58: 71–86.
Whittle, Bayliss, Barclay, Gaydarska, Bánffy, Borić, Draşovean, Jakucs, Marić, Orton, Pantović, Schier, Tasić and Vander Linden
54
Trogmayer O. 1964. Megjegyzések a Körös csoport rela-
tív időrendjéhez (Remarks to the relative chronology of
the Körös group). Archaeologiai Értesítő 91: 67–86.
1980. Újabb adatok a vonaldíszes kerámia relatív idő-
rendjéhez (Neue Beiträge zur relativen Chronologie der
Linearkeramik). Móra Ferenc Múzeum Evkönyve 1978/
79: 297–301.
Valovi≤ S. 1983. Neolitsko naselje u Ratini i njegovo me-
sto u vin≠anskom kulturnom kompleksu. Zbornik radova
Narodnog muzeja u ∞a≠ku 13: 33–45.
1987. Okruglica – vi∏eslojni lokalitet star≠eva≠ke i vin-
≠anske kulture u Vitanovcu kod Kraljeva. Na∏a pro∏lost
(Kraljevo) 2: 33–65.
Vander Linden M., Pand∫i≤ I. and Orton D. 2014. New ra-
diocarbon dates for the Neolithic period in Bosnia and
Herzegovina. Godi∏njak 43: 7–34.
Van Dyke R. M. 2015. Materiality in practice: an introduc-
tion. In R. M. Van Dyke (ed.), Practicing materiality. Uni-
versity of Arizona Press. Tucson: 3–32.
Vermeulen H., Govers C. (eds.) 1994. The anthropology
of ethnicity: beyond ethnic groups and boundaries. Het
Spinhuis. Amsterdam.
Vilotijevi≤ D. 1965. Nove Zemlje, Ωabalj – vi∏elsojno pra-
istorijsko nalazi∏te. Arheolo∏ki pregled 7: 31–32.
Vogel J. C., Waterbolk H. T. 1963. Groningen radiocarbon
dates IV. Radiocarbon 5: 163–202.
Voss B. L. 2015. What’s new? Rethinking ethnogenesis in
the archaeology of colonialism. American Antiquity 80:
655–670.
Vukovi≤ J. 2011. Late Neolithic pottery standardization:
application of statistical analyses. Starinar 41: 81–100.
Vukovi≤ J., Vitezovi≤ S. and Milanovi≤ D. 2016. Pavlovac –
Kova≠ke njive – neolitski horizonti. In S. Peri≤ and A. Bu-
latovi≤ (eds.), Arheolo∏ka istra∫ivanja na autoputu E75.
Arheolo∏ki institut. Beograd: 167–204.
Ward G. K., Wilson S. R. 1978. Procedures for comparing
and combining radiocarbon age determinations: a critique.
Archaeometry 20: 19–31.
Waterbolk H. 1988. C14-Datierungen von Gomolava. In
N. Tasi≤, J. Petrovi≤ (eds.), Gomolava: Symposium. Muzej
Vojvodine. Novi Sad: 117–121.
Whittle A. 1996. Europe in the Neolithic: the creation of
new worlds. Cambridge University Press. Cambridge.
Whittle A., Bartosiewicz L., Bori≤ D., Pettitt P. and Richards
M. 2002. In the beginning: new radiocarbon dates for the
Early Neolithic in northern Serbia and south-east Hun-
gary. Antaeus 25: 63–117.
A Vin;a potscape> formal chronological models for the use and development of Vin;a ceramics in south-east Europe
55
Site
No. of 14C No. of 14C results No. of 14C
Referencesresults (excluded) results (TPQ)
Danube upstream of Tisza
Szederkény-Kukorica-dűlő 41 2 Jakucs et al. 2016.Tab. 1
Versend 68 2 Jakucs et al. submitted.Tab. 1
Bapska 8 1 1 Table 2
Tisza and Mureş
Maroslele Pana 5 Table 2
At 4 1 Table 2
Potporanj 3 2 Table 2
Ószentiván 8 4 4 Kohl, Quitta 1970
Tiszasziget 1 1 Table 2
Satchinez 1 Table 2
Hodoni 2 Table 2
Uivar 182 27 86 Schier et al. forthcoming.Tabs. 6.2–6.6
Tărtăria 3 1 1 Table 2
Limba 1 1 Biagi et al. 2005.49
Cauce Cave 1 1 Paolo Biagi, pers. comm.
Orăştie–Dealul Pemilor, punct X2 3 3 Luca 2003
Cârcea Viaduct 8 8 Mantu 2000.99< Biagi, Spataro 2005.37
Miercurea Sibiului-Petriş 4 1 1 Biagi et al. 2007.Fig. 3< Luca et al. 2006.17
Sava catchment
Lupljanica 1 1 Table 2
Gornja Tuzla 5 3 2 Table 2
Gomolava 32 3 15
Borić 2009.Tab. 5< Orton 2012.Tab. 2< 
Waterbolk 1988.Beilage 1
Petnica 7 Borić 2009.Tab. 6< Orton 2012.Tab. 2
Masinske Njive 3 Table 2
Jari;i[te 1 1 Table 2
Belgrade area
Vin;a Belo-Brdo 85 2 7 Tasić et al. 2016a.Tab. 2
Opovo 13 2 Orton 2012.Tab. 2< Table 2
Banjica 3 1 2 Table 2
Central Serbia
Grivac 6 1 5 Table 2
Divostin II 11 2 8 McPherron, Srejović 1988< Borić 2009
Morava catchment
Anzabegovo 2 2 Table 2
Belovode 9 1 3 Borić 2009.Tab. 2
Plo;nik 8 Table 2
Ore[kovica 6 Table 2
Selevac 12 7 Table 2
Predionica 1 1 Table 2
Vala; 1 Table 2
Beran-Kr[ 2 2 Table 2
Donje Vranje 1 1 Table 2
Danube downstream of Vin;a
Liubcova 1 Table 2
Gornea 1 1 Table 2
Rudna Glava 14 6 Borić 2009.Tab. 1
Tab. 1. Summary of radiocarbon dating evidence considered in this review19.
19 Additionally, five thermoluminescence (TL) ages are reported on sherds from Feature 121 at Divostin II by McPherron and Sre-
jovi≤ (1988.Ch. 4). These measurements were made using the fine-grain technique (Zimmerman 1971), although no checks were
made for anomalous fading (Wintle 1973) nor radon escape (Desai, Aitken 1974). Given the provisional nature of these mea-
surements, no error is quoted for the measurements on each sherd and so they cannot be included in the modelling presented
here. Similarly, no error is reported for the average magnetic inclination and declination measurements made from samples from
Houses 14, 15, and 16 at Divostin II, two houses from Grivac, and the middle of the Vin≠a layer at Gomolava, which are also
reported by McPherron and Srejovi≤ (1988.Ch. 4). Again, this prevents these measurements from being calibrated and included
in the chronological models.
Whittle, Bayliss, Barclay, Gaydarska, Bánffy, Borić, Draşovean, Jakucs, Marić, Orton, Pantović, Schier, Tasić and Vander Linden
56
Tab. 2. Radiocarbon determinations and related stable isotopic measurements for samples associated
with Vin≠a ceramics (all stable isotope measurements were made using Isotope Ratio Mass Spectrometry,
except those marked * which were made by Accelerator Mass Spectrometry).
Site
Laboratory
Material and stratigraphic details
Radiocarbon δ13C δ15N
C>N References
number age (BP) (‰) (‰)
Bapska Bln-348 Sample 1, charcoal, Ulmus sp. from 5820±80 Quitta, Kohl 1969<
foundation of house 2-A< 1.4 m in Dimitrijević 1968.92
Dimitrijević 1964 trench associated
with a Vin;a C vessel
Bapska Beta-241657a Charred Triticum dicoccum from 5710±40 –23.6 Burić 2015< 2011
house 2 associated with Vin;a C\D
pottery (Burić 2015.150)
Bapska Beta-241657b Carbonised wheat grains from a con- 5690±40 –23.6 Burić, Tezac-Gregl 
centration with traces of a wooden 2009< Burić 2011< 2015
framework from behind house 1
Bapska Beta-241659 Charred unidentified animal bone 5660±40 –23.7 Burić 2011< 2015
from exterior of house 2 building
level associated with Vin;a C\D
pottery (Burić 2015.150)
Bapska OxA-23592 Unidentified animal bone from exterior 5714±31 –20.3 Burić 2011< 2015
of house 2 building level associated
with Vin;a C\D pottery (Burić 2015.150)
Bapska OxA-23593 Unidentified animal bone from build- 5715±33 –20.2 Burić 2011< 2015
ing level of houses 1 and 2
Bapska Beta-333534 Charred grains from house 2 associated 5700±40 Burić 2015
with a Vin;a C\D vessel (Burić 2015.150)
Maroslele- Poz-28647 Unidentified animal bone from pit 119 6290±40 Paluch 2011.Figs. 115–
Pana containing a few diagnostic sherds, 116
some of which resemble Vin;a A1–A2
forms, while others resemble early
Alföld LBK material
Maroslele- Poz-42773 Unidentified animal bone from pit 114 6255±40 Paluch 2011.Figs. 102–
Pana containing diagnostic Vin;a A1–A2 114
pottery and diagnostic early
Alföld LBK material
Maroslele- Poz-28646 Unidentified animal bone from pit 103 6280±40 Paluch 2011.Figs. 92–
Pana containing diagnostic Vin;a A1–A2 101
pottery and diagnostic early
Alföld LBK material
Maroslele- Poz-28645 Unidentified animal bone from pit 102 6280±40 Paluch 2011.Figs. 90–
Pana containing a few diagnostic sherds, 91
some of which resemble Vin;a A1–A2
forms
Maroslele- Poz-28644 Unidentified animal bone from pit 85 6200±50 Paluch 2011.Figs. 62–
Pana containing diagnostic Vin;a A1–A2 89
pottery and a few fragments in
early Alföld LBK style
At Col-3245 Pig mandible from layer 8, base of pit 5739±41 –21.0* Chu et al. 2016
1, associated with Vin;a C\D pottery
At Col-3246 Unidentified charcoal from layer 11 5675±43 –24.7* Chu et al. 2016
feature 3. A concentration of fired
clay from the oven
At Col-3248 Right nasal bovid bone from layer 11 5905±41 –20.0* Chu et al. 2016
feature 2a, from large concentration of
ash next to the fired clay in feature 3.
At OxA-8595 Bone tool from pit 1 associated with 5660±65 –20.2 Whittle et al. 2002
Vin;a C\D pottery
A Vin;a potscape> formal chronological models for the use and development of Vin;a ceramics in south-east Europe
57
Site
Laboratory
Material and stratigraphic details
Radiocarbon δ13C δ15N
C>N References
number age (BP) (‰) (‰)
Potporanj MAMS- Unidentified charcoal from last habita- 6156±30 –18.7*
22666 tion level, after the removal of the
house debris in trench 2 associated
with Vin;a B to Vin;a C pottery
Potporanj MAMS- Unidentified charcoal from the lowest 6139±29 –16.8*
22667 level, house floor level, trench 2
associated with Vin;a A pottery
Potporanj MAMS- Unidentified animal bone from relative 6211±25 –24.4* 3.1
22668 depth of 1.64 m in trench 2 associated
with Vin;a B2 pottery
Tiszasziget Bln-1631 Unidentified charcoal from Pit 1 6285±60 Horváth, Hertelendi 
associated with Vin;a A pottery 1994
(Horváth, Hertelendi 1994.125)
Satchinez Deb-2579 Antler from pit 420 associated with 6270±40 Mantu 2000
Vin;a A pottery (Mantu 2000.98)
Hodoni Deb-1963 Antler from Pit 4 associated with 5880±60 Draşovean 1994
Vin;a C pottery (Draşovean 1994.411)
Hodoni Deb-2018 Antler from Pit 4 associated with 5870±60 Gläser 1996
Vin;a C pottery (Gläser 1996.196)
Tărtăria R-1631 Partial human burial from ritual pit 6310±65 Merlini et al. 2008.156
associated with Vin;a B1 pottery
Tărtăria R-1655 Unidentified animal bone from base of 6215±65 Merlini et al. 2008.156
pit B2 associated with Vin;a B1 pottery
Miercurea GrA-33127 Single charred grain of Triticum dicoc- 6475±40 –22.4 Biagi et al. 2007.Fig. 3
Sibiului- cum from layer IIa in pit 18, which
Petriş contained diagnostic assemblage of
Vin;a B ceramics
Miercurea GrA-43076 Charred residue on pottery sherd from 6160±50 –26.1
Sibiului- pit 18, which contained diagnostic
Petriş assemblage of Vin;a B ceramics
Miercurea GrN-29053 Cattle calcaneum from house 11, 6350±130 –20.8 Luca et al. 2006.17
Sibiului- level IIb associated with diagnostic
Petriş assemblage of Vin;a B ceramics
Miercurea GrN-30500 Red deer astragalus from pit 3, 6200±60 –19.1 Biagi et al. 2007.Fig. 3
Sibiului- level IIb, which contained diagnostic
Petriş assemblage of Vin;a B ceramics
Lupljanica M-2455 Unidentified charcoal from 1.3m below 5600±200 Breunig 1987.107< 
ground surface in feature II\1970, the Crane, Griffin 1972.
lower building level of Late Neolithic 190–191
structure in the southwest part of the
site, associated with Vin;a D pottery
(Breunig 1987.107)
Gornja Bln-349 Charcoal, Quercus sp. from burnt 5710±100 Quitta, Kohl 1969
Tuzla wooden stucture at 3.63 m in Sonda
II\I-8, kop 26 associated with Vin;a C
pottery (Quitta, Kohl 1969.233)
Gornja GrN-1974 Unidentified charcoal (charred beam at 5580±60 Vogel, Waterbolk 1963
Tuzla the depth of 3.5 m) associated with Vin-
;a C pottery (Vogel, Waterbolk 1963.183)
Masinske OxA-26303 Bos\Cervus from Feature 1.31\4, O.S. 12, 6211±35 –21.5 8.1
Njive Sq. ≥8 associated with Vin;a B pottery
(M. Spasić, pers. comm.)
Masinske OxA-26304 Bos taurus from Feature 1.31\4, O.S. 12, 6267±36 –21.2 7.8
Njive Sq. ≥8 associated with Vin;a B pottery
(M. Spasić, pers. comm.)
Masinske OxA-26305 Cervus elaphus from Feature 1.31\9, 6235±37 –22.6 7.1
Njive O.S. 05, Sq. Y4 associated with Vin;a B
pottery (M. Spasić, pers. comm.)
20 Drasovean (2014.154, fn 10) reports that this sample came from pit 1. This was, however, a typographical error.
Whittle, Bayliss, Barclay, Gaydarska, Bánffy, Borić, Draşovean, Jakucs, Marić, Orton, Pantović, Schier, Tasić and Vander Linden
58
Site
Laboratory
Material and stratigraphic details
Radiocarbon δ13C δ15N
C>N References
number age (BP) (‰) (‰)
Jari;i[te 1 NOSAMS- A singe flax seed from feature 1.137 (pit 6260±35
78623 infill) associated with Vin;a A pottery
Opovo CAMS-4411 Unidentified charcoal from Feature 41 6030±60 –25.0
(locus 1861) associated with
Vin;a C pottery
Opovo CAMS-4413 Unidentified charcoal from Feature 41 5990±60 –25.0
(locus 1800) associated with
Vin;a C pottery
Banjica GrN-1542 Unidentified charcoal associated with 5710±90 Vogel, Waterbolk
Vin;a D pottery 1963.184–185
(Vogel, Waterbolk 1963.185)
Banjica GrN-1536 Charcoal\charred grain from House 7, 5670±120 Vogel, Waterbolk
horizon III associated with Vin;a D 1963.184
pottery (Vogel, Waterbolk 1963.184)
Grivac Z-1508 Unidentified charcoal from Trench A, 6000±140 Srdo; et al. 1987
level 9 associated with Vin;a C pottery
Grivac Bln-868 ‘Burnt earth’ from Trench B, under burnt 6070±100 McPherron, Srejović
house associated with Vin;a C pottery 1988.Tab. 14.1
Grivac Bln-872 Unidentified charcoal from Trench A, 5915±100 McPherron, Srejović
level 4, burnt house floor associated 1988.Tab. 14.1
with Vin;a C pottery
Grivac Bln-871 Unidentified charcoal from Trench A, 6190±100 McPherron, Srejović
level 7 associated with Vin;a C pottery 1988.Tab. 14.1
Grivac Bln-870 Unidentified charcoal from a large pit in 6315±100 McPherron, Srejović
subsoil, SE corner of Trench B associ- 1988.Tab. 14.1
ated with Vin;a A pottery
(Bogdanović 2004.500)
Anzabegovo LJ-2411 Unidentified charcoal from Sq. VIII, 6070±190 Gimbutas 1976< 
unit 55 associated with Vin;a B pottery Linick 1977
Anzabegovo LJ-2329 Unidentified charcoal from Sq. XX, 6230±60 Gimbutas 1976<
depth 190cm below datum associated Linick 1977
with Vin;a B pottery
Plo;nik OxA-14685 PL23, Bos sp. proximal tibia from burnt 5765±35 –20.6 5.7 Borić 2009
building debris> Trench 16< spit 7<
297.79masl< copper chisel found under
the debris (07\11\2000) associated with
‘Gradac’ phase pottery, equivalent here
to Vin;a C\D.
Plo;nik OxA-14686 PL24, Bos sp. calcaneus from building 6046±37 –19.8 4.8 Borić 2009
floor> Trench 14< spit 10, 297.29masl<
amorphous lumps of copper (2000)
associated with Vin;a B pottery
Plo;nik OxA-14687 PL25, Bos sp. rib from burnt surface> 6148±37 –21.4 6.4 Borić 2009
Trench 14< spit 13, 297.02masl< charcoal,
ash and burnt soil (2000) associated
with Vin;a B pottery
Plo;nik OxA-14688 PL26, Bos sp. vertebra from bottom of 6153±37 –21.0 6.8 Borić 2009
trench> Trench 14< spit 22, 295.87masl
(2000) associated with Vin;a B pottery
Plo;nik OxA-14689 PL28, Bos-size bone from Surface V> 6026±38 –20.4 5.8 Borić 2009
Trench 15< spit 12< 300.18masl< lumps of
thermally changed malachite
(02\10\2001) associated with
Vin;a B pottery
Plo;nik OxA-14690 PL29, dog distal humerus from Oven, 6160±37 –19.2 8.5 Borić 2009
Surface VII> Trench 15< spit 16,
299.56masl (10\10\2001) associated
with Vin;a B pottery
A Vin;a potscape> formal chronological models for the use and development of Vin;a ceramics in south-east Europe
59
Site
Laboratory
Material and stratigraphic details
Radiocarbon δ13C δ15N
C>N References
number age (BP) (‰) (‰)
Plo;nik OxA-14691 PL30, awl from large mammal long 6193±37 –20.1 5.9 Borić 2009
bone from burnt surface> Trench 15<
spit 20, 299.09masl, charcoal and ash
(11\10\2001) associated with
Vin;a B pottery
Plo;nik OxA-14692 PL31, Bos sp. rib from a hearth> Trench 6181±34 –20.6 7.6 Borić 2009
15< spit 26, 298.08masl< hearth dug into
the sterile soil (15\11\2001) associated
with Vin;a B pottery
Ore[kovica OxA-31571 Red deer antler tool (S1) from the 6265±38 –21.9 Borić et al. in prep.
primary fill (42) of ditch 47 (trench 1)
associated with Vin;a B pottery
Ore[kovica OxA-31572 Roe deer metatarsus (S4) from the 6137±35 –21.5 Borić et al. in prep.
upper fill (17) of ditch 47 (trench 1)
associated with Vin;a B pottery
Ore[kovica OxA-31573 Articulated cattle distal metararsus + 6123±37 –20.3 Borić et al. in prep.
phalanx I (S6) from the primary fill (36)
of palisade 32 (trench 1) associated
with Vin;a B pottery
Ore[kovica OxA-31574 Sus domesticus mandible (S8) from fill 6224±37 –20.0 Borić et al. in prep.
(62) of pit 18 (trench 2) associated
with Vin;a B pottery
Ore[kovica OxA-31575 Cattle horn core (S9) from fill (14) 6164±39 –20.7 Borić et al. in prep.
(trench 2) associated with
Vin;a B pottery
Ore[kovica OxA-31576 Lepus europeus articulated ulna and 6206±37 –17.9 Borić et al. in prep.
radius (S10) from pit (61) (trench 3)
associated with Vin;a B pottery
Selevac HAR 3211 Unidentified charcoal from Trench 18, 6050±70 –26.0 Tringham, Krstić
quad. 2, level 9 (BH 77-78> V) associ- 1990c
ated with Vin;a B2 pottery
Selevac HAR 3217 Unidentified wood from Feature 227 in 6240±100 –26.3 Tringham, Krstić
Trench 14 (BH 77-78> VII, VIII or IX) 1990c
associated with Vin;a B2\C pottery
Selevac HAR 3218 Unidentified charcoal from Trench 18, 5670±80 –25.3 Tringham, Krstić
quad. 3\4, level 10 (BH 77-78> IX|) 1990c
associated with Vin;a C pottery
Selevac HAR 3220 Unidentified charcoal from Feature 44, 6100±70 –27.1 Tringham, Krstić
collapsed oven superstructure (BH 77- 1990c
78> VI) associated with Vin;a B2 pottery
Selevac HAR 3221 Unidentified charcoal from Feature 210, 6050±70 –26.7 Tringham, Krstić
ditch related to House 4 (BH 77-78> VII) 1990c
associated with Vin;a B2\C pottery
Selevac HAR 3222 Grain (whole) from Feature 60 6230±100 –25.6 Tringham, Krstić
(House 7) (BH 77-78> II) associated 1990c
with Vin;a B1 pottery
Selevac HAR 3230 Unidentified charcoal from Feature 220, 5750±80 –25.7 Tringham, Krstić
deep posthole (BH 77-78> VIII or IX) 1990c
associated with Vin;a B2\C pottery
Selevac HAR 3232 Unidentified charcoal lens from 6040±70 _25.5 Tringham, Krstić
Feature 62 (BH 77-78> V) associated 1990c
with Vin;a B2 pottery
Selevac LJ-2521 Grain from Silo A associated with 6080±70 Linick 1977
Vin;a B1 pottery
Selevac Z-233a Wheat grain from Silo A 6113±80 Srdoć et al. 1975
Selevac Z-233b Wheat grain from Silo A 6152±90 Srdoć et al. 1975
Selevac Z-233 Wheat grain from Silo A 6366±100 Srdoć et al. 1975
Whittle, Bayliss, Barclay, Gaydarska, Bánffy, Borić, Draşovean, Jakucs, Marić, Orton, Pantović, Schier, Tasić and Vander Linden
60
Site
Laboratory
Material and stratigraphic details
Radiocarbon δ13C δ15N
C>N References
number age (BP) (‰) (‰)
Predionica Bln-435 Charcoal, Quercus sp. from fire 6280±80 Quitta, Kohl 1969
installation in Level 3b at 1.8m
associated with Vin;a A pottery
(Quitta, Kohl 1969.237)
Vala; Bln-436 Mixed sample of charred acorns and 5895±80 Quitta, Kohl 1969< 
seeds of sf. Pisum sp. from charred and Gimbutas 1974
ashy infill of a cut feature 0.7m from
the surface in sondage P-3 associated
with Vin;a D pottery
(Quitta, Kohl 1969.237)
Beran-Kr[ Z-491 Unidentified charcoal from context with 6030±60 Srdoć et al. 1977.474<
variable information> Trench III, level 7 Marković 1985.70
(Srdoć et al. 1977) or level 3 or 4 from
building horizon IIa associated with
Vin;a C pottery (Marković 1985)
Beran-Kr[ Z-492 Unidentified charcoal from context 5870±150 Srdoć et al. 1977.474<
with variable information Trench III, Marković 1985.70
level 13 (base of excavation)
(Srdoć et al. 1977) or level 3 or 4 from
building horizon IIa associated with
Vin;a C pottery (Marković 1985)
Donje SUERC- Bos or Cervus bone from Pit in KRS1 6206±33 –21.1 Kapuran et al. 2016
Vranje 57927
Gornea BM-1124 Charcoal (ref 1974\57)< charcoal frag- 5871±54 –25.8 Burleigh, Hewson 
ments from Trench 23, sq. 1, Pit 21, 1979
spits 4-6 associated with Vin;a A
pottery (Burleigh, Hewson 1979.350)
Liubcova Bln-2133 Charred grain associated with Vin;a B2 6175±85 László 1997
pottery (László 1997.260)