ArticlePDF Available

Context is everything: Comments on Radivojević et al. (2013)



Context is everything: comments on Radivojević et al. (2013) - Volume 88 Issue 342 - Duško Šljivar, Dušan Borić
Context is everything: comments on
cet al.
sko ˇ
san Bori´
These comments are in response to Radivojevi´
cet al. (2013) in which it is claimed that a
small foil object identified as tin bronze (11.7 per cent tin), found at the Late Neolithic and
Early Copper Age site of Ploˇ
cnik in southern Serbia, comes from an undisturbed context of
the Vinˇ
ca culture settlement. The object was used as the central piece of evidence to argue
for the first appearance of tin bronze production in Eurasia (and presumably the world).
This object is compared to 14 other objects, characterised as tin bronzes on the basis of
their compositions, from 10 sites, mainly in the central and eastern Balkans, in present-day
Serbia and Bulgaria. These sites have complex settlement histories and while they have
produced evidence of late fifth millennium BC habitation, their deep stratigraphies show
that this is only one phase among many that extend into later prehistory. It is on the basis
of these finds that Radivojevi´
cet al. argue that the previously accepted narrative regarding
the evolution of metallurgy in Eurasia is destabilised. They claim a date for the rise of tin
bronze production 1500 years before its first documented emergence among the Bronze Age
societies of south-west Asia. They suggest that the colour obtained by producing tin bronzes
imitated the aesthetic properties of mid fifth millennium BC golden objects. Radivojevi´
et al. conclude that the evidence they present suggests the nature of early metallurgy in the
Balkans was ‘polymetallic’.
There are several fundamental problems with both the factual evidence presented in the
paper and the research design. The compositional analysis undertaken by Radivojevi´
cet al.
is not questioned and we assume that their identification of the objects as tin bronzes is
accurate. However, we are concerned by the archaeological context and the interpretation
of the findings.
The key issue that concerns us here is the provenance of the central piece of evidence—the
tin bronze foil, labelled as sample 63 from the site of Ploˇ
cnik. At this point, we should state
that one of the present authors (Dˇ
S) has been the principal investigator of the archaeological
works at the site of Ploˇ
cnik since 1996 and was present at the site as its field director
throughout the 2008 field season when the object in question was found. Dˇ
S also provided
Miljana Radivojevi´
c with the piece of metal foil, with its contextual information, for analysis.
Contrary to the statement made in the article, this object was not found in “an undisturbed
context, on the floor of a dwelling structure next to a copper workshop [...] approximately
1m from a fireplace” (Radivojevi´
cet al. 2013: 1032), but in the spoil heap above section
CD, which is on the opposite side of the excavated area from the find spot indicated in their
figure 2. The metal foil was noticed by the excavators and collected from the spoil heap; it
1Department of Archaeology, National Museum in Belgrade, Trg Republike 1, 11000 Belgrade, Serbia (Email:
2Department of Archaeology and Conservation, Cardiff University, Colum Drive, Cardiff CF10 3EU, UK
(Author for correspondence; Email:
Antiquity Publications Ltd.
ANTIQUITY 88 (2014): 1–6
Context is everything
was provisionally assigned to spit 5, which was being excavated on the day it was found. A
note about the metal foil was made in the official field journal of excavations at Ploˇ
cnik on
23 September 2008. The metal foil object (inv. C–397) could belong to a possible horizon
of 0.75m of cultural deposits; that is between 301.98m, the height of spit 1, and 301.23m,
the bottom of spit 5. Only below this horizon, at the level of spits 7 and 8, was the burnt
structure shown in their figure 2 completely exposed. The floor of the burnt structure had
not been revealed on the day this object was found. A copy of the field journal describing
this season of excavations has been archived at the Ministry of Culture in Serbia since 2009;
this particular object is mentioned on page 7. Radivojevi´
cet al. (2013: 1032) cite ˇ
Sljivar and
c (2009) to support their claim that the piece of metal foil comes from
an undisturbed context. However, the burnt structure in the publication they cite is not
from trench XXI excavated in 2008, which is shown in their figure 2, and above which the
metal foil was found. The article they cite describes a similar burnt structure with an oven,
found in trench XX, excavated in 2007. Whatever the reason for the errors in reporting the
provenance of the metal foil, it is important to correct these factual details as Radivojevi´
et al. rely heavily on this context to support the argument they present.
It is also claimed that “[t]his securely contextualised find comes from a single undisturbed
occupation horizon that has been dated to c. 4650 BC” (Radivojevi´
cet al. 2013: 1032).
There are no radiocarbon dates from trench XXI; 4650 cal BC is the date of the peak of
the Bayesian probability density estimate for the end of the settlement at Ploˇ
cnik, which
is estimated at 4760–4340 cal BC (95.4% probability) or 4690–4530 cal BC (68.2%
probability) (Bori´
c 2009: 215). On the basis of eight existing assays from this site dating
other structures and deposits, it is reasonable to assume that the same span applies to the
structure in Trench XXI. However, in the absence of radiocarbon dates from this part of
the site, the dating of the burnt structure and deposits overlying it must remain provisional
and this does not justify the statement that “[t]he tin bronze foil from the site of Ploˇ
cnik is
therefore the earliest known tin bronze artefact anywhere” (Radivojevi´
cet al. 2013: 1032).
In the past two decades, it has been shown that copper objects from Ploˇ
cnik can indeed
be assigned to the Vinˇ
ca horizon at this site (ˇ
Sljivar 1996; ˇ
Sljivar & Jacanovi´
c 1996). This
challenged the views of scholars who had questioned the chronology of metal objects at
cnik in relation to the Vinˇ
ca period; they preferred an Early Copper Age date (Bubanj-
Hum cultural horizon) in the second half of the fifth millennium BC, rather than a Late
Neolithic date (the preceding Vinˇ
ca cultural horizon) (cf. Chapman 1981; Tasi´
c 1995). The
new chronology for copper at Ploˇ
cnik is in line with pioneering work on Vinˇ
ca metallurgy
by Borislav Jovanovi´
c (1982, 1994) at Rudna Glava. This site has extraordinary evidence
for Vinˇ
ca society mining explorations, now confidently dated to as early as the middle or
beginning of the sixth millennium BC. Extraction mining activity continued for the duration
of the Vinˇ
ca period until its end in the forty-seventh or forty-sixth century cal BC (Bori´
2009). Consequently, we believe that the Ploˇ
cnik metal hoards must belong to the Vinˇ
culture, probably dating to the second quarter of the fifth millennium BC (cf. Hansen 2013:
146–47). Though the latest levels at Ploˇ
cnik have no recognised post-Neolithic occupation
horizon, we cannot rule out the possibility of brief visits to the site during the long period
after the end of the Neolithic phase at Ploˇ
cnik, and that taphonomic processes are responsible
for the presence of this tin bronze foil in the uppermost deposits of the site. Many Vinˇ
Antiquity Publications Ltd.
sko ˇ
Sljivar & Duˇ
san Bori´
culture sites, including Ploˇ
cnik, have occasional pieces of pottery of Copper Age date or
later, mixed with Vinˇ
ca culture materials just below the ploughsoil. In the case of Ploˇ
Bubanj-Hum pottery is often found at the site. The most parsimonious explanation is that
these chronologically later diagnostic objects could come from unrecognised pits that cut
Late Neolithic deposits. It is not impossible to imagine that the piece of metal foil in question
ended up in the deposits of the site in a similar way.
cet al. (2013: 1032) suggest the piece of foil was used for “wrapping a ceramic
vessel”; we find this interpretation odd and decontextualised as there are no analogies
for a similar practice in the history of research on the Vinˇ
ca culture or neighbouring
contemporaneous culture groups. We also remain puzzled by the fact that in singling out a
particular object to make a claim about the ‘original’ character of this cultural context for
the ‘rise of tin bronzes in Eurasia’, Radivojevi´
cet al. completely ignore the predominant
group of more than 35 artefacts found at Ploˇ
cnik during its long research history from 1927
to 1996. These artefacts are all copper objects, mainly typologically recognisable chisels
and axe-hammers, which weigh over 16kg. This sample of objects consists of pure copper
metallographic texture (Pernicka et al. 1993) and hardly suggests that Vinˇ
ca metallurgy
was polymetallic in nature. Radivojevi´
cet al. choose to decontextualise the dominant
pattern of evidence from the site for the sake of making a claim about the ‘rise of tin
In terms of the regional distribution of the Vinˇ
ca culture, the presence of tin bronze
objects in this cultural context cannot be supported by the piece of tin bronze ring (8.5 per
cent tin) from the site of Gomolava that Radivojevi´
cet al. (2013: fig. 1B) also studied. No
contextual information about this object was provided apart from the statement that it can
“tentatively [be] dated to the mid fifth millennium BC” (Radivojevi´
cet al. 2013: 1032). The
site of Gomolava is a multi-phase tell site where, apart from the Late Neolithic Vinˇ
ca culture
occupation at the bottom of the tell, there are also occupation horizons dated to the Copper,
Bronze and Iron Ages, as well as later historic periods. Pits from these later occupation
horizons often contaminate the Late Neolithic levels (Brukner 1988; Bori´
c 2009: 221–27).
Having identified problems with the context and dating of the foil from Ploˇ
cnik and the
ring fragment from Gomolava, we are still left with 13 objects that supposedly come from
Chalcolithic contexts (the second half of the fifth millennium BC), and against which the
tin bronze foil from Ploˇ
cnik was contextualised in the article. Radivojevi´
cet al. are aware
that all of these objects come from problematic contexts (Pernicka et al. 1997; Radivojevi´
et al. 2013: 1034). Yet, they assert that these objects could be dated to the Chalcolithic period
on the basis of their compositions even though “the exact concentrations of [lead, arsenic,
nickel, cobalt, iron and gold] vary widely from sample to sample” (Radivojevi´
cet al. 2013:
1034). Despite acknowledging the poor contextual data for the 15 tin bronzes examined
in their article, Radivojevi´
cet al. (2013: 1038) believe that these are a homogenous group,
and dismiss the possibility that they may represent later contamination of the early levels.
They also mention another 25 tin bronze artefacts with similar compositions found in Early
Bronze Age (early second millennium BC) contexts in Croatia and Bosnia and Hercegovina.
cet al. argue for a temporal and spatial disjuncture of these two groups of objects,
but considering their tin bronzes appear to belong to contaminated contexts and are not
securely dated, it remains unclear why they believe that tin bronze objects from Serbian sites
Antiquity Publications Ltd.
Context is everything
should have more in common with those found in Bulgaria than objects from neighbouring
Bosnia or Croatia.
Two copper objects with larger amounts of tin have recently been reported: a copper ring-
shaped bead (8.35 per cent tin), found at the Late Neolithic site of Aruchlo I in Georgia
and dated to 5800–5300 cal BC (Hansen et al. 2012), and a copper awl (6 per cent tin),
found in the fill of a richly ornamented burial at the Middle Chalcolithic site of Tel Tsaf in
Israel, with associated layers dated to 5100–4600 cal BC (Garfinkel et al. 2014). In both
instances excavators believe the deposits were undisturbed and reject the possibility of later
contamination. It has been suggested that these examples should be understood as the use
of natural copper-tin alloy, which is found in Muˇ
siston, Tajikistan, and other regions; this
raises the possibility of long-distance trading of such objects during the middle Chalcolithic
period in the Levant (Garfinkel et al. 2014: 4 and references therein). Contextualising
such finds in the wider Eurasian regional framework remains a task for the future. For
the purposes of this paper, the possibility of naturally occurring copper-tin alloys in the
Balkans must be considered. Radivojevi´
cet al. cite Glumac and Todd (1991) who assume
significant tin mineralisations in western Serbia, Croatia, Bosnia and Romania as part of
the Tethyan-Eurasian metallogenic belt. Western Serbia and Cer Mountain in particular are
singled out as potentially important sources of tin. One recent project explicitly targeted
possible tin sources in the Jadar area of western Serbia with emphasis on aspects of Bronze
Age metallurgy (Bankoff et al. 2011); while this project determined the geological presence
of tin ore (cassiterite) it did not find significant alluvial deposits, suggesting that such
deposits might have been exhausted by intensive exploitation in the Bronze Age. It is fair
to say that the current evidence for sources of natural copper-tin alloys in the Balkans is
inconclusive. The argument presented by Radivojevi´
cet al., that copper ores rich in tin were
consciously exploited for their specific aesthetic properties during the fifth millennium BC
in the Balkans, remains an interesting but unsubstantiated hypothesis. We believe that more
evidence from secure contexts is required before this can be considered seriously.
cet al. concede that tin bronzes disappeared from the archaeological record
of the Balkans after the brief d´
ebut posited in their paper, and did not reappear before the
third or second millennium BC. This is not reflected in the title of their article however,
which suggests a continued ‘rise’ after their supposed emergence in the second half of the
fifth millennium BC. To acknowledge the subsequent decline in the title would describe
more realistically the trajectory of the phenomenon they claim to have identified. This, of
course, would suggest that the phenomenon had been correctly identified; we believe we
have shown here that it was not. We cannot exclude the possibility that Late Neolithic or
Early Copper Age Balkan smiths might have targeted copper ores rich in tin in the fifth
millennium BC. However, at present, the robust sample of analysed copper artefacts strongly
suggests that objects made from pure copper were dominant during the fifth millennium
BC in the Balkans. From a handful of documented tin bronze objects, found at several sites
that had, among other periods, fifth millennium BC levels, not a single object was found in
a securely dated fifth millennium BC context.
Finally, we would like to comment on the claim made by Radivojevi´
cet al. that the colour
of tin bronzes might have imitated gold at the time when the earliest examples of golden
objects, Varna gold, appeared in the Balkans. The central piece of evidence, the Ploˇ
Antiquity Publications Ltd.
sko ˇ
Sljivar & Duˇ
san Bori´
metal foil, dates, according to Radivojevi´
cet al.,toc. 4650 cal BC. The current dates for
Varna suggest a cultural horizon from 4560 to 4450 cal BC (Chapman et al. 2006; Higham
et al. 2007). It is impossible that people in the Vinˇ
ca period, which is earlier than the Varna
culture, decided to imitate golden objects that did not exist at the time.
We conclude that the argument presented by Radivojevi´
cet al. is based on limited evidence
and there are a number of problematic points: 1) the context of the central piece of evidence
for the presence of tin bronze at Ploˇ
cnik is erroneously reported; 2) the chronological
and cultural attributions of 15 objects identified as early tin bronzes are questionable
and the suggested dating of these objects to the second half of the fifth millennium BC
remains unsubstantiated; 3) the sample of 15 objects is unrepresentative in comparison
to the composition of contemporary metal copper objects; this makes the far-reaching
conclusions expressed in the paper problematic. In our opinion, the only appropriate method
of interpreting unexpected finds is to accumulate robust datasets with precise recording of
contextual information and to combine the full spectrum of archaeological evidence with
an intimate knowledge of archaeological sequences. This is the only way to reconceptualise
established paradigms in our understanding of the past.
We are grateful to the anonymous reviewers for their thoughtful comments about the argument presented here.
The accuracy of claims made in this text remains the sole responsibility of its authors.
C,A.HUSKA &W.POWEL L. 2011. Tin
sources and settlement in the Bronze Age of
south-eastern Europe: a pilot study from western
Serbia. Antiquity 85(327 Project Gallery). Available
(accessed 21 July 2014).
C, D. 2009. Absolute dating of metallurgical
innovations in the Vinˇ
ca culture of the Balkans, in
T.K. Kienlin & B.W. Roberts (ed.) Metals and
societies. Studies in honour of Barbara S. Ottaway:
191–245. Bonn: Habelt.
BRUKNER , B. 1988. Die Siedlung der Vinˇ
ca-Gruppe auf
Gomolava (die Wohnschicht des sp¨
und fr¨
aneolithikums—Gomolava Ia, Gomolava
Ia-b und Gomolava Ib) und der Wohnhorizont des
aneolithischen Humus (Gomolava II), in N. Tasi´
c(ed.)Gomolava—Chronologie und
Stratigraphie der vorgeschichtlichen und antiken
Kulturen der Donauniederung und S¨
Interlationales Symposium, Ruma 1986: 19–38. Novi
Sad: Vojvo -
d muzej and Balkanoloˇ
ski institut
CHAPMAN, J.C. 1981. Vinˇ
ca culture of south-east Europe:
studies in chronology, economy and society (British
Archaeological Reports international series 117, in
two parts) Oxford: British Archaeological Reports.
GAYD ARSKA &N.HONCH. 2006. The social
context of the emergence, development and
abandonment of the Varna cemetery, Bulgaria.
European Journal of Archaeology 9: 159–83.
ROSENBERG. 2014. The beginning of metallurgy in
the southern Levant: a late 6th millennium cal BC
copper awl from Tel Tsaf, Israel. PLoS ONE 9(3):
GLUMAC,P.&J.TODD. 1991. Early metallurgy in
southeast Europe: the evidence for production, in
P. Glumac (ed.) Recent trends in archaeometallurgical
research: 9–19. Philadelphia (PA): University of
HANSEN, S. 2013. Innovative metals: copper, gold and
silver in the Black Sea region and the Carpathian
Basin during the 5th and 4th millennium BC, in S.
Burmeister, S. Hansen,
M. Kunst & N. M¨
uller-Scheessel (ed.) Metal matters:
innovative technologies and social change in prehistory
and antiquity: 137–67. Rahden: Marie Leidorf.
&M.ULLRICH. 2012. Aruchlo—eine neolithische
Siedlungim S¨
udkaukasus. Altertum 57: 81–106.
Antiquity Publications Ltd.
Context is everything
DIMITROVA. 2007. New perspectives on the Varna
cemetery (Bulgaria)—AMS dates and social
implications. Antiquity 81: 640–54.
C, B. 1982. Rudna Glava. Najstarije rudarstvo
bakra na centralnom Balkanu. Beograd: Muzej
rudarstva i metalurgije & Arheoloˇ
ski institut.
– 1994. Gradac phase in the relative chronology of the
Late Vinˇ
ca culture. Starinar 43: 1–10.
1993. Eneolithic and Early Bronze Age copper
artefacts from the Balkans and their relation to
Serbian copper ores. Pr¨
ahistorische Zeitschrift 68:
TODOROVA &I.KULEFF. 1997. Prehistoric copper
in Bulgaria. Its composition and provenance.
Eurasia Antiqua 3: 41–180.
NORTHOVER. 2013. Tainted ores and the rise of tin
bronzes in Eurasia, c. 6500 years ago. Antiquity 87:
SLJIVAR, D. 1996. The eastern settlement of the Vinˇ
culture at Ploˇ
cnik: a relationship of its stratigraphy
to the hoards of copper objects. Starinar 47: 85–97.
C. 1996. Veliko Laole,
ca culture settlement in northeastern
Serbia. Pr´
ehistoire Europ´
eenne 8: 175–88.
C. 2009.
cnik, archaeology and conservation. Diana
(Belgrade) 13: 56–61.
C, N. 1995. Eneolithic cultures of the central and
western Balkans. Belgrade: Balkanoloˇ
ski institut.
Antiquity Publications Ltd.
The South Caucasus was a major center of metal production in the Late Bronze and Early Iron Ages. Nowhere is this more clear than in the hills and mountains in the southeastern Black Sea region (ancient Colchis), where exceptionally large numbers of metal production sites have been found. Chemical and microscopic analysis of slagged technical ceramics at these sites illuminates several aspects of both raw copper and tin bronze alloy production. Copper ores were smelted in a complex multi-stage process designed to extract metal from sulfide ores. Technical ceramics served as containers for a range of different reactions, from the first phase of smelting, in which the copper sulfides were likely consolidated into a matte, though later stages of matte processing and metal copper production in smaller crucibles. In addition, a single crucible fragment, recovered from a late 2nd millennium BC slag heap, demonstrates that tin bronze was created by the direct addition of cassiterite tin ore, probably of alluvial origin, to metallic copper. The crucible's context, the use of cassiterite ore rather than tin metal, and a review of local geology suggests that the tin used in this crucible came from a nearby, with the most likely source being the Vakijvari and Bzhuzhi gorges roughly 10-15 km away. While a single fragment does not speak to the regularity of this practice, at the very least it raises the possibility that the Colchian bronze industry was based on local rather than imported tin.
Full-text available
The research team of this new project has begun the precision radiocarbon dating of the super-important Copper Age cemetery at Varna. These first dates show the cemetery in use from 4560-4450 BC, with the possibility that the richer burials are earlier and the poor burials later in the sequence. The limited number of lavish graves at Varna, representing no more than a handful of paramount chiefs, buried over 50-60 years, suggests a stabilisation of the new social structure by the early part of the Late Copper Age.
Full-text available
The beginning of metallurgy in the ancient Near East attracts much attention. The southern Levant, with the rich assemblage of copper artifacts from the Nahal Mishmar cave and the unique gold rings of the Nahal Qanah cave, is regarded as a main center of early metallurgy during the second half of the 5th millennium CalBC. However, a recently discovered copper awl from a Middle Chalcolithic burial at Tel Tsaf, Jordan Valley, Israel, suggests that cast metal technology was introduced to the region as early as the late 6th millennium CalBC. This paper examines the chemical composition of this item and reviews its context. The results indicate that it was exported from a distant source, probably in the Caucasus, and that the location where it was found is indicative of the social status of the buried individual. This rare finding indicates that metallurgy was first defused to the southern Levant through exchange networks and only centuries later involved local production. This copper awl, the earliest metal artifact found in the southern Levant, indicates that the elaborate Late Chalcolithic metallurgy developed from a more ancient tradition.
Full-text available
The earliest tin bronze artefacts in Eurasia are generally believed to have appeared in the Near East in the early third millennium BC. Here we present tin bronze artefacts that occur far from the Near East, and in a significantly earlier period. Excavations at Plocnik, a Vinca culture site in Serbia, recovered a piece of tin bronze foil from an occupation layer dated to the mid fifth millennium BC. The discovery prompted a reassessment of 14 insufficiently contextualised early tin bronze artefacts from the Balkans. They too were found to derive from the smelting of copper-tin ores. These tin bronzes extend the record of bronze making by c. 1500 years, and challenge the conventional narrative of Eurasian metallurgical development.
Full-text available
Analyse isotopique du plomb sur 89 artefacts qui vont de l'Eneolithique au Bronze Ancien. Plus du tiers des objets provennant de divers gisements est constitue de haches cruxiformes. Le but principal de cette etude est l'evaluation du role de la mine de cuivre eneolithique de Rudna Glava pour la metallurgie ancienne dans les Balkans centrales
Full-text available
In this article we outline some of the key characteristics of the social structure of the Climax Copper Age in the eastern Balkans and the contributions of the Varna cemetery to those developments. We continue by examining the implications of the new series of 21 AMS dates from the Oxford Radiocarbon Laboratory, which represent the first dates for the Varna Eneolithic cemetery on the Bulgarian Black Sea coast. Representing the first phase of the AMS dating project for the Varna I cemetery, these dates have been selected to provide a range of different grave locations, ranges of grave goods, and age/gender associations. We conclude by addressing the question of the unexpectedly early start of the cemetery, as well as its apparently short duration and relatively rapid demise.