Vous êtes sur la page 1sur 280

ARHEOLO[KI INSTITUT BEOGRAD

INSTITUT ARCHOLOGIQUE BELGRADE

UDK 902/904 (050) ISSN 03500241


STARINAR LXII/2012, 1275, BEOGRAD 2012.
INSTITUT ARCHOLOGIQUE BELGRADE

STARINAR
Nouvelle srie volume LXII/2012

BELGRADE 2012
ARHEOLO[KI INSTITUT BEOGRAD

STARINAR
Nova serija kwiga LXII/2012

BEOGRAD 2012.
STARINAR STARINAR
Nova serija kwiga LXII/2012 Nouvelle srie volume LXII/2012

IZDAVA^ EDITEUR
Arheolo{ki institut Institut archologique
Kneza Mihaila 35/IV Kneza Mihaila 35/IV
11000 Beograd, Srbija 11000 Belgrade, Serbie
e-mail: institut@ai.sanu.ac.rs e-mail: institut@ai.sanu.ac.rs
Tel. 381 11 2637191 Tl. 381 11 2637191
UREDNIK RDACTEUR
Slavi{a Peri}, direktor Arheolo{kog instituta Slavi{a Peri}, directeur de lInstitut archologique
REDAKCIONI ODBOR COMIT DE RDACTION
Miloje Vasi}, Arheolo{ki institut, Beograd Miloje Vasi}, Institut archologique, Belgrade
Rastko Vasi}, Arheolo{ki institut, Beograd Rastko Vasi}, Institut archologique, Belgrade
Noel Dival, Univerzitet Sorbona, Pariz Nol Duval, Universit Paris Sorbonne, Paris IV
Slobodan Du{ani}, Srpska akademija nauka Slobodan Du{ani}, Acadmie serbe des sciences
i umetnosti, Beograd et des arts, Belgrade
Bojan \uri}, Univerzitet u Qubqani, Bojan \uri}, Universit de Ljubljana,
Filozofski fakultet, Qubqana Facult des Arts, Ljubljana
Vasil Nikolov, Nacionalni arheolo{ki institut Vasil Nikolov, Institut archologique national et Muse,
i muzej, Bugarska akademija nauka, Sofija Acadmie bulgare des sciences, Sofia
Ivana Popovi}, Arheolo{ki institut, Beograd Ivana Popovi}, Institut archologique, Belgrade
Marko Popovi}, Arheolo{ki institut, Beograd Marko Popovi}, Institut archologique, Belgrade
Nikola Tasi}, Srpska akademija nauka i umetnosti, Beograd Nikola Tasi}, Acadmie serbe des sciences et des arts, Belgrade
SEKRETAR REDAKCIJE SECRTAIRE DE RDACTION
Jelena An|elkovi} Gra{ar, Arheolo{ki institut, Beograd Jelena An|elkovi} Gra{ar, Institut archologique, Belgrade
LEKTOR ZA SRPSKI JEZIK LE LECTEUR POUR LA LANGUE SERBE
Mirjana Radovanovi} Mirjana Radovanovi}
LEKTOR ZA ENGLESKI JEZIK LE LECTEUR POUR LA LANGUE ANGLAIS
Dejv Kalkat Dave Calcutt
GRAFI^KA OBRADA RALISATION GRAPHIQUE
Danijela Paracki Danijela Paracki
D_SIGN, Beograd D_SIGN, Belgrade
[TAMPA IMPRIMEUR
ALTA NOVA, Beograd ALTA NOVA, Belgrade
TIRA@ TIRAGE
500 primeraka 500 exemplaires
SEKUNDARNA PUBLIKACIJA PUBLICATION SECONDAIRE
COBISS COBISS
SCIndeks SCIndeks

^asopis je objavqen uz finansijsku pomo} Ministarstva prosvete, nauke


i tehnolo{kog razvoja Republike Srbije

Cette priodique a t publi avec le soutien du Ministre de lducation, de la science


et du dveloppement technologique de la Rpublique Serbie
SADR@AJ SOMMAIRE

^LANCI ARTICLES

Marija B. Radovi} Ageing in the Danube Gorges Population (95005500 BC)


Tooth Cementum Annulation Method . . . . . . . . . . . . . . . . . . . . . . . . . 9
Marija B. Radovi} Individualna starost stanovnika \erdapa (95005500 godina pre n. e.)
primena metoda anulacije zubnog cementa . . . . . . . . . . . . . . . . . . . . . . 18

Marko Por~i} De facto refuse or structured deposition?


House inventories of the Late Neolithic Vin~a culture . . . . . . . . . . . . . 19
Marko Por~i} De facto otpad ili strukturisana depozicija?
Ku}ni inventari kasnoneolitske vin~anske kulture . . . . . . . . . . . . . . . . 43

Adam N. Crnobrnja Investigations of Late Vin~a House 1/2010 at Crkvine in Stubline . . . . 45


Adam N. Crnobrwa Istra`ivawe poznovin~anske ku}e 1/2010
na Crkvinama u Stublinama . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 64

Aleksandar Kapuran, Kulturna grupa Kocofeni-Kostolac


Aleksandar Bulatovi} na teritoriji severoisto~ne Srbije . . . . . . . . . . . . . . . . . . . . . . . . . 65
Aleksandar Kapuran, Coofeni-Kostolac Culture on the Territory
Aleksandar Bulatovi} of North-eastern Serbia . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 81

Dragana Antonovi}, Eneolithic Mine Prlju{a Mali [turac:


Momir Vukadinovi} Archaeological and geophysical investigations . . . . . . . . . . . . . . . . . . . 95
Dragana Antonovi}, Eneolitski rudnik Prqu{a Mali [turac:
Momir Vukadinovi} arheolo{ka i geofizi~ka istra`ivawa . . . . . . . . . . . . . . . . . . . . . . . . . . . . 105

Jovan Koledin O jednom novijem nalazu iz gvozdenog doba u Sremu . . . . . . . . . . . . . 107

Ivana Popovi} La ncropole de la basilique urbaine Sirmium . . . . . . . . . . . . . . . . . 113


Ivana Popovi} Nekropola gradske bazilike u Sirmijumu . . . . . . . . . . . . . . . . . . . . . . . . . . 134

Miloje Vasi}, Nade`da Gavrilovi} Venus or Diana from Mediana . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 137


Miloje Vasi}, Nade`da Gavrilovi} Venera ili Dijana iz Medijane . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 149
Stefan Pop-Lazi} Some Observations on Lead Figurines of the Goddess Venus
in the Area between Sirmium and Viminacium . . . . . . . . . . . . . . . . . . . . 151
Stefan Pop-Lazi} Neka zapa`awa o olovnim figurinama bogiwe Venere
na podru~ju izme|u Sirmijuma i Viminacijuma . . . . . . . . . . . . . . . . . . . . . . 164

Vojin Nedeqkovi}, Domino et fraturi: jedan neobjavqen sirmijski grafito . . . . . . . . . . 165


Sawa Stojanovi}
Vojin Nedeljkovi}, Sanja Stojanovi} Domino et fraturi. An unpublished graffito from Roman Sirmium . . . . . . . . . . . 172

Jelena Q. Cvijeti} Novi epigrafski spomenik iz Otilovi}a kod Pqevaqa . . . . . . . . . . 173


Jelena Lj. Cvijeti} The new epigraphic monument of Otilovi}i near Pljevlja . . . . . . . . . . . . . . . . . . 180

Nata{a Miladinovi}-Radmilovi} Analysis of human osteological material from the eastern part
of Site No. 37 in Sremska Mitrovica . . . . . . . . . . . . . . . . . . . . . . . . . . . . 181
Nata{a Miladinovi}-Radmilovi} Analiza humanog osteolo{kog materijala sa isto~nog dela
lokaliteta 37 u Sremskoj Mitrovici . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 202

Vesna Biki} The Haban Pottery from the Belgrade Fortress:


Archaeological Context, Chronology, Decorative Designs . . . . . . . . . . . 205
Vesna Biki} Habanska keramika sa Beogradske tvr|ave:
kontekst, hronologija i dizajn . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 226

Christoph Rummel, Daniel Peters, Report on the Geomagnetic Survey


Georg Schafferer at Margum in October 2011 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 229
Kristof Rumel, Daniel Peters, Izve{taj o geomagnetskim istra`ivawima Marguma
Georg [aferer u oktobru 2011. godine . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 238

Vujadin Ivani{evi}, Primena LiDAR tehnologije u analizi topografije


Ivan Bugarski Marguma/Morave i Kuli~a . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 239
Vujadin Ivani{evi}, Application of LiDAR Technology in Analyses of the Topography
Ivan Bugarski of Margum/Morava and Kuli~ . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 255

KRITI^KI OSVRTI APERC


CU CRITIQUES

Dragana Antonovi}, O neolitskoj autenti~nosti nalaza iz Belice . . . . . . . . . . . . . . . . . . 257


Slavi{a Peri}
Dragana Antonovi}, Slavi{a Peri} About Neolithic authenticity of finds from Belica . . . . . . . . . . . . . . . . . . . . . . . . 267

KRITIKE I PRIKAZI COMPTES RENDUS

Dragana Antonovi} BRONZEN IM SPANNUNGSFELD ZWISCHEN


PRAKTISCHER NUTZUNG UND SYMBOLISCHER
BEDEUTUNG: BEITRGE ZUM INTERNATIONALEN
KOLLOQUIUM AM 9. UND 10. OKTOBER 2008
IN MNSTER. HERAUSGEGEBEN VON UTE
LUISE DIETZ UND ALBRECHT JOCKENHVEL.
Prhistorische Bronzefunde (PBF), Abteilung XX,
Bd. 13. Franz Steiner Verlag Stuttgart 2011. . . . . . . . . . . . . . . . . . . . . . . 269
Rastko Vasi} Mario Gavranovi}, DIE SPTBRONZE-UND FRHEISENZEIT
IN BOSNIEN, Universittsforschungen zur prhistorischen
Archologie, Band 195, Verlag Dr. Rudolf Habelt GMBH
Bonn 2011. . . . . . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . . 270

Milica Tapavi~ki-Ili} Slavica Kruni}, ANTI^KE SVETIQKE IZ MUZEJA


GRADA BEOGRADA (i na~ini iluminacije u Singidunumu
od I do sredine V veka), Muzej grada Beograda, Beograd 2011. . . . . . 271

NAU^NI SKUPOVI ASSEMBLES

Rastko Vasi} ANTIKA I SAVREMENI SVET . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 275


MARIJA B. RADOVI], University of Belgrade,
Faculty of Philosophy, Department of Archaeology, Laboratory for Bioarchaeology, Belgrade

AGEING IN THE DANUBE GORGES POPULATION


(95005500 BC) TOOTH CEMENTUM
ANNULATION METHOD

UDK: 902.65:572.77"633/634"(497.11) e-mail: mradovic@f.bg.ac.rs


DOI: 10.2298/STA1262009R Received: February 28, 2012
Original research article Accepted: June 21, 2012

Abstract. Tooth cementum annulation, the microscopic method for the determination of an individuals age, gives results
that are highly correlated with the chronological age of an individual. Nevertheless, this method is still rarely used for age
estimation in archaeological populations. In this study, using the tooth cementum annulation method, teeth of 21 individuals
of the Djerdap anthropological series, dated to a period from the 10th to the 6th millennium BC, were analysed. The obtained data
are important for overcoming some methodological issues in anthropology of the Danube Gorges, as well as with the precise age
estimation of old individuals and with the assessment of age in cases where the skeletal material has been very poorly preserved.
The only obstacle to the full application of the tooth cementum annulation method is the taphonomy changes of tooth cementum
which were detected on several teeth in this study.

Key words. Individuals age, dental anthropology, tooth cementum annulation,


the \erdap anthropological series, taphonomy.

D
ental anthropology is a sub discipline of phys- skeleton which is influenced by the physiological and
ical anthropology, and dealing with dental metabolic processes in the body, while at the same
anthropology involves the analysis of teeth time are also under the influence of the environment.
and jaws. In 1542, Andreas Vezalius noted that the teeth We should not underestimate the importance of the
are very different from the bones in their biological genetic component and its impact on the teeth, or the
properties and function, and today it is clear that they fact that teeth evolve slowly, allowing us to easily
cannot be generally viewed as parts of the skeleton in compare archaeological data with recent dental mate-
a narrow sense1. The teeth, as a separate component of rials2. All these facts make teeth an important source of
the skeleton and body, possess certain anatomical and information, and dental anthropology, a discipline sig-
physiological characteristics that are very important nificant in the study of ancient populations.
for researchers. Hard dental tissues (cementum, den-
tine and enamel) are compact and could be well pre-
served in the soil, while the enamel itself has a feature
that remains unchanged over time and it is resistant to 1 Hillson 2005.
taphonomic changes. Teeth are the only part of the 2 Irish and Nelson 2008.

* The article results from the project: Bioarchaeology of ancient Europe people, animals and plants in Serbian prehistory (no 47 001) funded
by Ministry of Education, Science and Technological Development of the Republic of Serbia.

9
Marija B. RADOVI], Ageing in the Danube Gorges Population (95005500 BC) (918) STARINAR LXII/2012

An important issue when it comes to palaeodemo- APPLICATION OF TOOTH CEMENTUM


graphic analysis is the reconstruction of patterns of ANNULATION IN AGE ESTIMATION
aging and mortality in the past. The reliability of the
reconstruction of these patterns depends on the accuracy Tooth cementum annulation is a microscopic
of individual age estimations. For decades, physical method for the determination of an individuals age
anthropologists and palaeodemographers have tried to based on the analysis of the acellular extrinsic fibrous
improve this method. The importance of such studies cementum (AEFC). This type of dental cementum, as
is emphasised in the new approach to palaeodemo- compared to the other four types, is located at the cer-
graphic analysis described by Hoppa and Vaupel 3. The vical third of each root of deciduous or permanent teeth
main problem in almost all macroscopic methods for and allows the tooth to be anchored into the alveolar
age estimation is the accuracy of the method itself 4. cavity. All the while surrounded and provided with
Age estimation errors occur when the skeleton is ob- nutritionally intact desmodontium, AEFC is subject to
served macroscopically and only biological changes are the additional growth that is reflected in histological
visible which often do not correspond to individuals samples by equal, alternating light and dark rings.
actual chronological age. Due to the high variability in Depending on the individuals age, the thickness of
the correlation of the biological and chronological age AEFC ranges from 20 to 250 mm. Alternating light and
in one individual, the age estimation error can be up to dark rings are subject to seasonal rhythms, which most
7 years5. This error increases in older individuals, as likely occur under the influence of several factors such
do our methodological problems. It is clear that there as a dose of UV-radiation, climatic conditions, different
is a need for a method for age estimation that is less qualities of diet and the hormonal status of the indi-
susceptible to continuous changes in the skeletons that vidual. A pair of one dark and one light ring constitutes
can not be quantified and that increase with the age of an incremental line. Seasonality in the rhythm of lay-
the individual. ering cementum lines has been seen in more than 50
Recent research has shown that a solution could be different mammalian species worldwide, and it was con-
found in an alternative method for the determination of cluded that it occurs as a result of a natural metabolic
an individuals age based on counting the incremental rhythm induced by seasonal changes7. These research
lines in tooth cementum. These lines (lines of annual has shown that the metabolism of the parat hormone
layering of dental cementum) may be a more stable has a major role in the formation of incremental lines,
indicator of an individuals age in comparison with any and is responsible for regulating levels of calcium in
morphological or histological characteristics of the the bloods interaction with vitamin D, which in turn re-
skeleton and offer a microscopic method for the deter- gulates calcium absorption. The interaction of hormones
mination of tooth cementum annulation (TCA). The and vitamins, driven by a complex mechanism of exter-
method relies on the counting of incremental lines on nal physical and chemical factors, leads to the formation
cross sections of the root, about 80 mm thick, under of this circular annual rate in the root of a tooth8.
a microscope with a magnification of 400 x. The use of cementum in the determination of age
The accurate age estimation of individuals from in humans begins with the measuring of the total width
ancient populations is of great importance for palaeo- of the layer of cementum before the lines are counted 9.
demographic research in archaeology and anthropology. In the early eighties, studies conducted on three human
The goal of this paper is to demonstrate the advantages teeth10 showed that the method of tooth cementum
of the TCA method for such purposes. In addition, the
further aim of this paper is to test the readability of
incremental lines and the consistency of this method
3
for the age determination of a sample of teeth from 21 Hoppa and Vaupel 2002
4 Buikstra and Ubelaker 1994; Jackes 2001.
individuals from the Djerdap anthropological series.
5 Buikstra and Ubelaker 1994; Jackes 2001.
The \erdap anthropological series includes over 500
6 Bori}, Dimitrijevi} 2007.
funerals, with about 600 individuals excavated at the 7 Laws 1952; Geiger 1993; Grue and Jensen 1979; Kay et al.
12 locations of the Lepenski Vir culture6. Sites which 1984.
were inhabited from the 10th to the 6th millennium BC, 8 Gustafson 1950.
were found and investigated on both sides of the 9 Stott et al. 1982.
Danube. 10 Naylor et al. 1985.

10
Marija B. RADOVI], Ageing in the Danube Gorges Population (95005500 BC) (918) STARINAR LXII/2012

annulation could be applied to the determination of age


Site Grave Tooth
in humans, as previously applied in other mammals11.
Further technical improvements have led to the method Lepenski Vir 20 44
of tooth cementum annulation being accepted as a more Lepenski Vir 60 14
advanced method compared to others which have used Lepenski Vir 64 34
teeth to determine an individuals age. So today, the Lepenski Vir 88 14, 15
method is based on counting the incremental lines, this Vlasac 2 15
number is then added to the number of years in which Vlasac 9 25
the observed tooth erupts, and the result is a calendar Vlasac 17 24, 25
age of the observed individual.
Vlasac 29 24, 44
Initially, the method was applied only to freshly
Vlasac 41 34
extracted teeth, but since the late eighties it has been
implemented to determine the age of individuals from Vlasac 55 45
historical and archaeological skeletal series, with equal Vlasac 67 15
success in both the inhumed and the cremated12. This Vlasac 79 34
research has also clarified the fact that the number of Vlasac U53 PM
incremental lines remains fixed, even in circumstances Vlasac U232 34
when other characteristics of incremental lines change Vlasac U267 23
(eg, width and degree of mineralization13) due to the Padina 6 14
impact of environmental or physiological disorders. For Padina 15 35
these reasons, the method of tooth cementum annulation
Padina 16 35
has recently been considered one of the most accurate
Padina 30 25
techniques for the determination of age in the skeletal
material of adult individuals14. In younger individuals, Hajdu~ka Vodenica 13 14
analyses of the microstructure of enamel and dentin Hajdu~ka Vodenica 33 PM
have achieved even greater precision in estimating the
chronological age, sometimes with accuracy to within Table 1. The structure of the sample
days15. However, there are still problems that prevent
Tabela 1. Struktura uzorka
the full implementation of the tooth cementum annula-
tion method. One such problem is that previous studies
were done on small samples, which limited the estab-
lishment of a good statistical method for processing Faculty of Biology, in the Ludwig Maksimilianus
results obtained by the tooth cementum annulation University in Munich, 24 tooth roots were analysed
method in palaeodemographic and forensic studies. In from 21 individuals from the Danube Gorges anthro-
addition, it has still not clarified the issue of the impact pological series (Table 1). In this sample, three indivi-
of dental, especially periodontal, diseases on the accu- duals came from new excavations at the site of Vlasac.
racy of the method. Some researchers claim that perio- For most individuals one tooth was extracted (except in
dontal diseases do not affect the number of lines of the case of individuals from Vlascac burials 17 and 29
cementum16, while others say that pathology reduce, and burial number 88 from Lepenski Vir where two teeth
or completely prevent the formation of incremental were extracted). After extraction, a selection was made
lines in tooth cementum17.

11 Gustafson 1950, 1955; Azaz et al. 1974; Philipsen and Jablon-


MATERIALS AND METHODS
ski 1992.
12 Gro{kopf 1989, 1990.
Dental samples in this study come from the Meso- 13 Karger, Grupe 2001.
lithic and the Neolithic sites in the Danube Gorges 14 Wittwer-Backofen, Buba 2002.
(Lepenski Vir, Vlasac, Padina, Hajdu~ka vodenica) and 15 Antoine et al. 2000.

date back to the period from the 10th to the 6th millen- 16 Gro{kopf et al. 1996

nium BC. In a survey conducted by the author at the 17 Kagerer, Grupe 2001.

11
Marija B. RADOVI], Ageing in the Danube Gorges Population (95005500 BC) (918) STARINAR LXII/2012

based on the type of tooth, and the sample examined in In the next stage of analysis, incremental lines were
this study consisted only of permanent premolars from counted three times on each photo by the observer. The
the upper and lower jaws. individuals age was determined on the basis of the
All teeth are completely submerged in biodur pitch following three criteria: the mean value of the number
(compared to the protocol of making a resin mixture of of incremental lines, the maximum number and the most
100 ml biodur: 28 ml E7 hardener resin for 20 doses, frequent number of incremental lines. The criterion for
Gunther von Hagens). Depending on the preservation counting incremental lines is to count only those parts of
of the roots, between 4 and 10 cross-cut sections of the sections where the lines are equal, with a clear con-
each root were prepared for analysis. Each of them was trast between light and dark rings, which have as little
between 70 mm and 80 mm thick. The teeth were cut in micro bacterial decomposition and erosion as possible
a direction towards the apex of the root by a rotating and avoiding those parts of cross sections with visible
diamond blade (Leitz 1600). Each cross section was traces of the cuts of the diamond blade, etc.
analysed under a transmitted light microscope (Zeiss
Axioskop 2 plus, Zeiss / Jena, equipped with a CCD
camera AxioCam MRC colour and Axio Vision Release RESULTS
4.3 software) with a 400 x magnification. Incremental
lines were counted on digital photos, which were further Out of the 24 analysed teeth, at least 17 had one
processed with Adobe Photoshop 8.0.1. Up to 10 shots readable segment from any of the ten transverse slices
were taken for each root cross section. (Table 2, Fig. 1). For the remaining 7 teeth, it was not

Fig. 1. A segment with clear incremental lines,


Padina 16
Fig. 2. Pits in the dental cementum,
Hajdu~ka Vodenica 33
Fig. 3. Longitudinal cracks dental cementum,
Vlasac 2
Sl. 1. Segment sa jasnim inkrementnim linijama,
Padina 16
Sl. 2. [upqine u zubnom cementu,
Hajdu~ka vodenica 33
Sl. 3. Uzdu`na pukotina zubnog cementa,
Vlasac 2

12
Marija B. RADOVI], Ageing in the Danube Gorges Population (95005500 BC) (918) STARINAR LXII/2012

Maximum No. Mean No. Most frequent No.


Grave
of incremental lines of incremental lines of incremental lines
LV20 43 32 40
LV60 14 13 14
LV64 42 31 36
LV88 45 41 43
VL 2 22 21 22
VL29 17 16 17
VL41 42 37 40
VL55 40 37 40
VL67 19 18 19
VL79 61 50 42
VLU53 45 40 42
VLU232 18 15 12
VLU267 58 50 46
P16 40 34 34
P15 13 12 13
P30 27 24 24
HV13 35 31 34

Table 2. Number of readable incremental lines from 17 teeth represented by 10 cross sections
Tabela 2. Broj pro~itanih inkrementnih linija sa 17 zuba predstavqenih sa po 10 ise~aka

possible to see a segment that showed continuously indicators upon which we decided to exclude some teeth
visible incremental lines from the cementum and from the analysis. The results showed that the presence
dentin circuit to the edge of the root. For each of these of dental pathology and crown wear did not affect the
seven tooth roots, cross sections displayed very weak readability of the incremental lines in this sample.
lines, outlines or segments where lines were intersect- More than one readable segment was available for
ed by pits (Fig. 2) and vertical or horizontal cracks 13 out of 17 teeth. The comparison of the variation in the
(Fig. 3). In two cases the incision edges were parallel number of incremental lines in the 13 samples showed
to the incremental lines so that the lines appeared to be that in only two cases did samples provide an age
unreadable (Fig. 4). In addition to the described phe- range of more than 20 years (the level of uncertainty in
nomenon, and due to the secondary mineralisation that the process of estimating the age was similar to that
affected a portion of the sample, the clarity of incre- obtained using the standard macroscopic method).
mental lines in some slices of teeth made analysis diffi-
cult. The appearance of secondary mineralisation was
expected in the case of skeletal material dating from DISCUSSION
the Mesolithic period. The consequences of this process
on the readability of the number of incremental lines In this research, a number of teeth were excluded
are reflected in the fact that it reduces the number of from further analysis (7 teeth representing a total of
visible lines, i.e. it seems that two to three lines were 29.1%) because the incremental lines in the cross sec-
grouped into one (Fig. 5). However, unresolved factors tions had been affected by diagenetic processes in more
that influence the development process of secondary developed stages. Some sections were lacking visible
mineralisation are yet to be discovered. lines, on others, the lines were wavy and intersected by
The presence of hypercementosis, periodontal dis- pits and cracks (Fig. 2, Fig. 3) and moreover, there
ease or the degree of wear of the tooth crowns were not were a number of sections with lines which were shaded,

13
Marija B. RADOVI], Ageing in the Danube Gorges Population (95005500 BC) (918) STARINAR LXII/2012

Fig. 4. Diamond blade cutmarks, Vlasac U267


Fig. 5. Secondary mineralization of dental cementum, Vlasac U232
Sl. 4. Urezi se~iva, Vlasac U267
Sl. 5. Sekundarna mineralizacija zubnog cementa, Vlasac U232

or where the lines were affected by secondary mine- and microorganisms22, while inorganic chemical sub-
ralisation (Fig. 5). Research in this area on the remains jects contribute to the intense impact and degradation
from archaeological material also presents problems caused by microorganisms which facilitate the impreg-
relating to unreadable segments18. However, the deci- nation of soil minerals such as CaCO, Fe2O3 and SiO2.
sion as to whether a segment is readable or not, is Studies have shown that the effect of chemical agents in
certainly subjective19. Thus, in the case of multiple the soil can lead to the complete or partial replacement
observers in the study conducted by Vitver-Bakofen et of the component building blocks of tissue at the micro-
al., a senior author excluded a total of 14.1% of the scopic level, leaving no macroscopically visible changes
teeth from their study, while another observer designa- in morphology of the bone23. Therefore, tooth cementum
ted 22.3% of the teeth in the sample as unreadable. In in these conditions acts as a bone and suffers diagenetic
the study conducted by Roksandi} et al., the decision as changes. In this study, incremental lines in the cemen-
to which teeth were to be excluded from further analysis tum could not be read from one third of the sample. It
was made by the senior author on the basis of the clarity is this illegibility, caused by diagenetic changes, that
and continuity of incremental lines at each cross sec- represents the biggest limitation of the application, and
tion20. In the aforementioned study, counting was carried precise method, of the annulation of dental cementum
out only by the author and only on the sections where in determining the age of individuals in ancient popu-
the lines were equal, with a clear contrast between light lations. Establishing the standards of line clarity between
and dark rings, and with the least micro bacterial de- researchers in the future, must be imposed as neces-
composition, erosion, or traces of the blade. sary to maximise the safety of this method.
The poor microscopic preservation of nearly a third
of the sample in this study may be explained by the
influence of several factors such as the chemical con-
18 Cipriano-Bechtle et al. 1996; Roksandi} et al. 2009;
ditions in the soil (ion exchange with ground water and
the precipitation of minerals) and/or post-mortem bio- Wittwer-Backofen et al. 2008.
19 Wittwer-Backofen et al. 2008
logical activity (bacteria and fungi). All these factors 20 Roksandi} et al. 2009.
may contribute to the change in the tissue at the micro- 21 Pfeiffer 2000; Nonato do Rosario Marinho et al. 2006.
scopic level, leaving a vague or structurally altered 22 Henderson 1987; Heuck 1993.
histological picture21. The organic components of bones 23 Lambert et al. 1979; Francillon-Veilliellot et al. 1990;

stored in the soil break down due to the impact of water Gilber 1997; Gill-King 1997.

14
Marija B. RADOVI], Ageing in the Danube Gorges Population (95005500 BC) (918) STARINAR LXII/2012

THE IMPORTANCE OF THE DENTAL CONCLUSION


CEMENTUM ANNULATION METHOD
FOR RESEARCH IN THE DANUBE GORGES This research contributes to the identification of an
PALAEODEMOGRAPHY individuals age. More precisely, the estimated age of
the individual is given within a narrow range. In the
The results of the chronological age of individuals case of the Djerdap anthropological series, we are now
obtained by the annulation method may indirectly con- able to obtain, with greater certainty, the precise age of
tribute to a clearer palaeodemographic picture of the the adults, especially the oldest individuals. In addition,
past. In the case of the Djerdap anthropological series, the dental cementum annulation method can be used to
questions about the more precise age of the old indi- determine age even in cases of the very poor preserva-
viduals in the population and the length of the repro- tion of skeletal material, as the research is performed
ductive capacity in women have remained open for only at the root of the tooth. This is of great importance
decades. This study gave us information of great for the anthropological study of ancient populations.
importance for solving these problems in the Iron Gate Moreover, using this method we could provide data
palaeodemography. essential for the future research of palaeodemography.
Based on the macroscopic examination of human For palaeodemographic studies, it is important to access
remains from the Djerdap series, it was known that data on the precise age of individuals, especially the
there were individuals older than 40 years. Given the oldest individuals in the population. The maximum
bias towards a short life expectancy of people in pre- age reached in a population as well as the number of
historic times and the fact that due to the living habits individuals who attained it, significantly affects the
and conditions, individual skeletal morphology may calculation of several parameters of mortality25. The
show a greater biological age, many individuals from fact that the chronological age of the eldest individuals
the Djerdap anthropological series were placed in a is determined helps to better estimate the length of life
category of between 4060 years24. In the sample taken of individuals in this population. It can also be deter-
for this study, 12 individuals were designated older than mined how the life length is distributed among the
40 years on the basis of the morphology of the skeleton. individuals from the Djerdap anthropological series.
Results from the annulation method showed that 3 out Possessing this type of information, we will be able to
of these 12 individuals were approximately 50 years make palaeodemographic comparisons among different
old (between 46 and 52 years of age) and 2 individuals populations, taking into account the proportion of the
were approximately 55 years old (between 5257 years oldest individuals in them.
of age). Finally, the results showed an advanced age in In addition to the presented anthropological results
two individuals from the Vlasac site (VLVL79 and for the Danube Gorges, the study pointed to several ob-
U267), which were determined to be between 6070 stacles to the full implementation of the tooth cemen-
years of age. In conclusion, it is now legitimate to say tum annulation method. These barriers are reflected in
that people in prehistoric times in the Danube Gorges changes in dental tissues at the microscopic level, as the
could have reached ages of up to 70 years. result of taphonomic changes. The fact that diagenetic
As for the length of the reproductive period in processes that affect changes in incremental lines of
women, it is crucial to point out to the case of a preg- dental cementum are not yet fully understood, needs to
nant woman from the grave VL67 at Vlasac, dating be stressed too. A recommendation for future analysis
from the Mesolithic period. The annulation of dental is the research of taphonomic effects on hard dental
cementum method showed that her age at the time of tissue histology.
death was between 3032 years.
Translated by Marija Radovi}

24 Roksandi} 1999.
25 Wood et al. 2002.

15
Marija B. RADOVI], Ageing in the Danube Gorges Population (95005500 BC) (918) STARINAR LXII/2012

BIBLIOGRAPHY:

Antoine et al. 2000 D. Antoine, C. Dean, S. Gustafson 1950 G. Gustafson, Age determination
Hillson, The periodicity of incremental structures in of teeth. Journal of American Dental Association 41,
dental enamel based on the developing dentition of 1950, 4554.
post-medieval known-age children. In J. T. Mayhal, T. Gustafson 1955 G. Gustafson, Altersbestimmung
Heikkinen (eds.), Dental morphology 1998. Oulu an Zhnen, Dtsch Zahn 25, 1955, 17631768.
University Press, Oulu 2000, 102111. Gro{{kopf 1989 B. Gro{kopf, Incremental lines
Azaz et al. 1974 B. Azaz, M. Ulmansky, R. in prehistoric cremated teeth. A technical note, Journal
Moshev, J. Sela, Correlation between age and thickness of Morphological Anthropology 77,1989, 309311.
of cementum in impacted teeth. Oral Surgery 28, 1974, Gro{{kopf 1990 B. Gro{kopf, Individualalters-
691694. bestimmung mit Hilfe von Zuwachsringen im Zement
Buikstra and Ubelaker 1994 J. E. Buikstra, D. bodengelagerter menschlicher Zhne, Z Rechtsmed
H. Ubelaker, Standards for the data collection from 103, 1990, 351259.
human skeletal remains. Research series 44. Fayette- Gro{{kopf et al. 1996 B. Gro{kopf, J. M. Denden,
ville: Arkansas Archaeological Survey, Arkansas 1994. W. Krger, Untersuchungen zur Zementapposition bei
Bori}, Dimitrijevi} 2007 D. Bori}, V. Paradontitis marginalis profunda, Dtsch Zahn Z 51, 1996,
Dimitrijevi}, Apsolutna hronologija i strati- 295297.
grafija Lepenskog vira, Starinar LVII, Beograd Irish and Nelson 2008 J. D. Irish, G. C. Nelson,
2007, 955. Technique and application in dental anthropology,
Cipriano et al. 1996 A. Cipriano-Bechtle, G. Cambridge University Press, Cambridge 2008.
Grupe, P. Schroter, Ageing and life expectancy in the Lambert et al. 1979 J. B. Lambert, C. B.Szpunar,
early middle ages. Homo 46, 1996, 267279. J. E. Buikstra, Chemical analysisof excavated human
Francillon et al. 1990 H. Francillon-Veilliellot, bone from middle and late Woodland sites. Archaeometry
V. Buffrenil, J. Castanet, J. Geraudie, F. J. Meunier, J. 21, 1979, 115129.
Y. Sire, L. Zylberberg, A. Ricqules, Microstructure and Laws 1952 R. M. Laws, A new method of age
mineralization of vertebrate skeletal tissues. In J. G. determination for mammals. Nature 169, 1952,
Carter (ed.), skeletal biomineralisation: patterns, pro- 972973.
cesses, and evolutionary trends (I). New York: Van Kay et al. 1984 R.F. Key, D. T. Rasmussen, K. C.
Nostrand Reinhold, New York 1990, 479480. Beard, Cementum annulus counts provide a means for
Geiger 1993 G. Geiger,Vergleich verschiedener age determination in Macaca mulatta (Primates,
Methoden der Altersbeurteilung anhand von Zhnen Anthropoidea), Folia Primatol 42, 1984, 8595.
und anderen morphologischen Merkmalen mit dem Henderson 1987 J. Henderson, Factor determining
Lebensalter vorwiegend altersmarkierter Wildtiere der the state of preservation of human remains. In A.
Ordnungen Artiodactyla und Carnivora. Giessen: Boddington, A. N. Garland, R. C. Janaway (eds.), Death,
Habilitationsschrift Giessen, 1993. decay and reconstruction: approaches to archaeology
Gilber 1997 R. I. Gilber, Applications of trace and forensic science. Manchester: Manchester University
element research to problems in archaeology. In R. L. 1987, 4354.
Blakey (ed.), Biocultural adaptation in prehistoric Heuck 1993 F. W. Heuck, Comparative histolog-
America. Athens: University of Georgia Press, Athens ical and microradiographic investigations of human
1997, 85100. bone. In G. Grupe, A. N. Garland, (eds.), Histology of
Grue and Jensen 1979 H. Grue, B. Jensen, ancient human bone: methods and diagnosis. New York:
Review of the formation of incremental lines in tooth Springer-Verlag, NewYork 1993, 125136.
cementum of terrestrial mammals. Dan. Rev. Game. Hillson 1996 S. Hilson, Dental Anthropology.
Biol. 11, 1979, 1 48. Cambridge University Press, Cambridge 1996.
Gill-King 1997 H. Gill-King, Chemica and ultra- Hilson 2001 S. Hilson, Recording Dental Caries
structural aspects of decomposition. In W. D. Haglund, in Archaeological Human Remains. International
M. H. Sorg, (eds.), Forensic taphonomy: the post mortem Journal of Osteoarchaeology 11, 2001, 249 289.
date of human remains. Boca Raton: CRC Press, 1997, Hilson 2005 S. Hilson, Teeth, Cambridge Univer-
93104. sity Press, Cambridge 2005.

16
Marija B. RADOVI], Ageing in the Danube Gorges Population (95005500 BC) (918) STARINAR LXII/2012

Hoppa and Vaupel 2002 R. D. Vaupel, J. W. anthropology perspective. Ph.D. dissertation, Simon
Vaupel, Palaeodemographic age distributions from Fraser University, Vancouver 1999.
skeletal samples. In R. D. Hoppa, J. W. Vaupel (eds.), Roksandi} et al. 2009 M. Roksandic, D.Vlak, M.
Cambridge studies in biological and evolutionary anthro- A. Schillaci, and D. Voicu, Technical Note: Applicability
pology 31, Cambridge University Press, Cambridge 2002. of Tooth Cementum Annulation to an Archaeological
Jackes 2002 M. Jackes, Building the bases for Population, American Journal of Physical Anthropology
palaeodemographic analysis: adult age determination. 140, 2009, 583588.
In M. A. Katzenber, S. R. Saunders, (eds.), Biological Scott et al. 2002 G. G. Scot, R. F. Sis, B. M. Levy,
anthropology of the human skeleton, John Wiley and Sons, Cemental annulation as an age criterion in forensic den-
Inc., New York 2002, 417466. tistry, Journal of Dental Research 61, 1982, 814817.
Naylor et al. 1985 J. W. Naylor, G. W. Miller, G. N. Wittwer-Backofen and Buba 2002 U. Wittwer-
Stokes, G. G. Stott, Cementum annulation enhancement: Backofen , H. Buba, Age estimation by toothcementum
a technique for age determination in Man, American annulation: perspectives of a new validation study. In R.
Journal of Physical Anthropology 68, 1985, 197200. D. Hoppa, J. W. Vaupel (eds.), Age distributions from
Nonato do Rosario Marinho et al. 2006 A. skeletal samples.Cambridge studies in biological and
Nonato do Rosario Marinho, M. N. Cardoso, V. Braz, A. evolutionary anthropology 31, Cambridge University
K. Ribeiro-dos-Santos, Ferraz Mendonc, S. M. a de Souza, Press, Cambridge 2002, 107128.
Paleogenetc and taphonomic analysis of human bones Wittwer-Backofen et al. 2008 U. Wittwer-
from Moa, Beirada, and Ze Espinho Sambaquis, Mem Backofen, J. Buckberry, A. Czarnetzki, S. Doppler, G.
Inst Oswaldo Cruz 101, Rio de Janeiro 2006, 1523. Grupe, G. Hota, A. Kemkes, C. S. Larsen, D. Prince, J.
Pfeiffer 2000 S. Pfeiffer, Palaeohistology: health J. Wahl, A. Fabig, S. Weise, Basics in palaeodemography:
and disease. In M. A. Katzenber, S. R. Saunders, (eds.), a comparison of age indicators applied to the early me-
Biological anthropology of the human skeleton, John dieval skeletal sample of Lauchheim, American Journal
Wiley and Sons, Inc., New York 2002, 287302.. of Physical Anthropology 137, 2008, 384396.
Philipsen and Jablonski 1992 H. P. Philipsen, N. Wood et al. 2002 J. W. Wood, D. J. Holman, K.
G. Jablonski, Age estimation from the structure of adult OConnor, R. J. Ferrell, Mortality models for palaeode-
human teeth: review from the literature, Forensic Science mography. In R. D. Hoppa, J. W. Vaupel (eds.), Age dis-
International 54, 1992, 2328. tributions from skeletal samples. Cambridge studies in
Roksandi} 1999 M. Roksandi}, Transition from biological and evolutionary anthropology 31, Cambridge
Mesolithic to Neolithic in the Iron Gates gorge: Physical University Press, Cambridge 2002, 129168.

17
Marija B. RADOVI], Ageing in the Danube Gorges Population (95005500 BC) (918) STARINAR LXII/2012

Rezime: MARIJA RADOVI], Univerzitet u Beogradu, Filozofski fakultet, Odeqewe za arheologiju

INDIVIDUALNA STAROST STANOVNIKA \ERDAPA


(95005500 godina pre n. e.)
PRIMENA METODA ANULACIJE ZUBNOG CEMENTA

Kqu~ne re~i. individualna starost, dentalna antropologija, anulacija zubnog cementa,


populacija \erdapa, tafonomija.

Anulacija zubnog cementa je mikroskopski metod za utvr- nije uticalo na ~itqivost inkrementnih linija u ovom
|ivawe individualne starosti. Rezultati dobijeni prime- uzorku. Pore|ewe varijacija u broju linija u ovom uzorku
nom ove metode najpribli`niji su stvarnoj (hronolo{koj) pokazalo je procewenu starost u rasponu od preko 20 godi-
starosti individue. I pored toga, ova metoda se jo{ uvek na samo kod dve osobe (u tom rasponu se procewuje indivi-
retko koristi u odre|ivawu starosti u okviru arheolo{kih dualna starost primenom makroskopskih metoda). Dobije-
populacija. U ovom istra`ivawu metodom anulacije zub- ni podaci su od zna~aja ne samo za prevazila`ewe nekih
nog cementa analizirana su 24 zuba koja su pripadala 21 metodolo{kih problema u antropologiji \erdapa ve}, ta-
individui |erdapske antropolo{ke serije. Od svih anali- ko|e, i zato {to na ovaj na~in imamo i precizne podatke o
ziranih zuba, ukupno 17 je imalo bar jedan ~itqiv segment starosti praistorijskih stanovnika \erdapa. Primena
na popre~nim presecima. Kod preostalih 7 zuba nije bilo metode anulacije zubnog cementa omogu}ava utvr|ivawe
segmenata sa kontinuirano vidqivim inkrementnim lini- starosti i u slu~ajevima veoma lo{e o~uvanog skeletnog
jama. Prisustvo hipercementoze i parodontopatije ili materijala. Jedinu prepreku za primenu metode anulacije
pak istro{enosti krunica zuba nije bilo kriterijum za u ovom istra`ivawu, ali samo u nekoliko slu~ajeva, pred-
iskqu~ivawe uzoraka iz analize. Rezultati su pokazali da stavqala su stawa izmewene strukture zubnog cementa pod
prisustvo patolo{kih promena i istro{enosti krunica uticajem tafonomskih procesa.

18
MARKO POR^I]
University of Belgrade, Faculty of Philosophy, Department of Archaeology, Belgrade

DE FACTO REFUSE OR STRUCTURED DEPOSITION?


HOUSE INVENTORIES OF THE LATE NEOLITHIC
VIN^A CULTURE

UDK: 902.01"634"(497.11) ; 903.3"634"(497.11) e-mail: mporcic@f.bg.ac.rs


DOI: 10.2298/STA1262019P Received: February 10, 2012
Original research article Accepted: June 21, 2012

Abstract. The goal of this paper is to determine whether there are reasons to believe that inventories from the Late Neolithic
Vin~a culture houses do not represent systemic assemblages and to offer an interpretation of household assemblage variation.
Pottery inventories from Vin~a culture houses were compared to the ethnographically recorded range of variation in household
inventory size. The discard equation was used to make projections of the accumulated assemblages from house assemblages
for comparison with empirically observed accumulated assemblages. It is concluded that in general there is no reason to reject
the assumption that Vin~a household inventories reflect systemic assemblages. Moreover, the patterns of inventory variability
can be meaningfully interpreted in social terms.

Key words. Neolithic; Vin~a culture, formation processes, house inventory.

W
here preserved, house remains and their explanations of the Late Neolithic/Early Copper Age
inventories have always been an important transition in South-eastern Europe. According to these
class of data for inferring various aspects theories, changes in material culture, subsistence, set-
of past societies and formation processes of the tlement patterns and burial rites which occurred in the
archaeological record. This claim especially holds for Copper Age were a consequence of a large migration of
the Late Neolithic and Early Copper Age contexts in a new ethnic element the Indo-European population
South-eastern Europe where collapsed dwellings were coming from the Black Sea steppes.4 In some versions
conserved by fire.1 Houses were usually built using the
wattle and daub technique. When such houses are
exposed to fire and high temperatures, the mud from 1 Bailey 1999; Bailey 2000; Chapman 1999; Stevanovi} 1997;
the walls transforms into a bright red-orange daub.2 Stevanovi} and Tringham 1997; Tringham 2005.
Due to the brick like properties of fired daub, Neolithic 2 Schaffer 1993; Stevanovi} 1997.
3 e.g., Benac 1971; Brukner 1962; Gli{i} 1964; Jovanovi} and
houses are often very well preserved, since the walls
Gli{i} 1961; McPherron and Srejovi} 1988; Nikolov 1989; Petrovi}
have usually collapsed inwards and formed a coherent
1993; Schier 2006; Stevanovi} 1997; Todorovi} 1981; Tringham et
rubble which seals the inventory.3 al. 1992; Tringham et al. 1985; Tripkovi}, B. 2007.
The fact that houses were burned in almost every 4 Gara{anin 1982a; Gimbutas 1977; Jovanovi} 1979a; 1979b;

discovered settlement resonated with migrationist Mallory 1991; Tasi} 1983.

* The article results from the project: Archaeological culture and identity in Western Balkans (no 177008) funded by the Ministry of Education,
Science and Technological development of the Republic of Serbia.

19
Marko POR^I], De facto refuse or structured deposition? (1943) STARINAR LXII/2012

of this hypothesis, colonisation was seen as a series of Criterion (9): there are such large quantities of
aggressive raids resulting in conflagration and the de- objects, especially ceramics, in the burnt structure that
struction of entire Late Neolithic villages. The most this exceeds the quantity of a normal household assem-
drastic and vivid account of this kind was proposed by blage The final criterion refers to the accumulation
Gimbutas, which saw patriarchal and warlike Indo- of such large quantities of objects that this deposition
Europeans invading the peaceful and matriarchal po- amounts to a group offering prior to deliberate destruc-
pulation of the Old Neolithic Europe.5 It should be tion rather than a daily household assemblage.13
noted that migrationist explanations have remained This kind of behaviour would make sense in the
very popular among archaeologists of the traditional light of Chapmans fragmentation and enchainment
culture-historical orientation. theory.14 The central point of this theoretical perspec-
In the seventies and eighties, alternative explana- tive is that material culture plays a crucial role in medi-
tions were offered both for Late Neolithic/Early Copper ating and representing social relations in the Neolithic
Age transition and settlement conflagration. These new and Copper Age of South-eastern Europe. Fragmenta-
views were advocated mainly by researchers from tion and enchainment are key processes. By fragmenting
Anglo-American academic circles, where the impact an object and giving its parts to other social actors (living
of processual archaeology has been strong. In short, people or ancestors), a social link is established, an
culture change was conceived as an internal process,6 enchainment. In enchainment, objects are more than
while house destruction was seen as a result of acci- mere tokens of relationships, they are supposed to de-
dental fires7 or internal conflict.8 fine and convey the very personhood of the individual
The processual approach went hand in hand with giving or receiving the object. In this way, the enchain-
an advanced methodology and concern for formation ment process may suggest a different concept of person-
processes. As a result of experimental research, it was hood. Instead of the Western concept of an integral
soon realised that the observed intensity of house individual, an alternative personhood is constructed
burning is very difficult, if not impossible to replicate (fractal individual, dividual self) which is at the same
experimentally without additional fuel and effort in time individual and collective, connected to other
fire maintenance9, thus making the hypotheses of acci- people through the extension of artefacts.15 In theory,
dental fires or fires started in conflicts very unlikely. fragmentation, enchainment and dividual do not always
This prompted researchers to conclude that houses were coincide16, but in Chapmans theories regarding the
burned intentionally, although not as the collateral da- Neolithic and Copper Age of South-eastern Europe
mage of warfare, but as a deliberate symbolic, ritual they are usually tightly linked. According to Chapman,
and social practice.10 The idea of intentional house the structured deposition17 of objects into the house
burning as an ideological and symbolic act was usually prior to its deliberate destruction may be understood as
framed in a postprocessual explanatory scheme. In this
perspective, deliberate house burning was one of the
elements which defined the social arena where various 5 Gimbutas 2007.
kinds of relations (e.g., within and between households, 6 Bankoff and Greenfield 1984; Bankoff and Palavestra 1986;
genders, and generations) were negotiated and con- Bankoff and Winter 1990; Chapman 1981; Kaiser and Voytek 1983;
tested. This change of perspective has close parallels McPherron and Christopher 1988; Tringham 1992; Tringham and
Krsti} 1990.
in Americanist archaeology where old interpretations 7 McPherron and Christopher 1988.
of house burning due to practical reasons such as war- 8 Gli{i} 1968.
fare or accident have been supplanted by interpretations 9 Bankoff and Winter 1979; Gheorghiu 2011; Schaffer 1993;

where house burning is seen as ritual behaviour and a Stevanovi} 1997.


distinct mode of abandonment.11 10 Bailey 2000; Bori} 2008; Chapman 1999; Stevanovi}

Following this line of thought, Chapman proposed 1997; Stevanovi} and Tringham 1997; Tringham 1991; 1994; 2005.
11 Walker 2002; Wilshusen 1986.
that, in addition to the intentional destruction of the 12 Chapman 1999; 2000a, 224.
building, a further symbolic statement was made by de- 13 Chapman 1999.
positing a special assemblage (a mortuary set) into the 14 Chapman 2000a; Chapman and Gaydarska 2007.
house.12 He argued that the quantity and diversity of 15 Chapman 2000a, 2829.

uncovered house inventories exceeded the normal range 16 Brittain and Harris 2010.

of artefacts used in everyday household practices: 17 Richards and Thomas 1984.

20
Marko POR^I], De facto refuse or structured deposition? (1943) STARINAR LXII/2012

an enchainment (and fragmentation) working on two context by various processes of discard. Depending on
levels: 1) individual objects which form inventories of the mode of discard or abandonment, deposited artefacts
other households 2) fragments of objects whose other may belong to different categories of refuse. Two refuse
parts would be kept outside the burnt house.18 In this categories which are of crucial importance for the pur-
way, members of the community would create a link to a poses of this paper are de facto refuse and ritually de-
deceased person e.g. if the motive for the deliberate posited assemblage. De facto refuse consists of objects
house destruction is the death of a prominent member which, although still usable or reusable, are left behind
of the community19 or to ancestors in general, if the when an activity area or structure is abandoned. A ritually
house destruction is viewed as a structured deposition deposited assemblage is a collection of objects which
of the house and its inventory to the ancestral world, may or may not be associated in the systemic context,
objectified by the accumulated strata of a settlement but which are purposefully brought together and depo-
mound.20 sited as a part of symbolic or ritual act.
This paper investigates two related issues: 1) Although Schiffers concepts have been vigorously
Chapmans hypothesis of the structured deposition of debated and questioned, especially the validity of cul-
pottery into the house 2) the variability of Vin~a cul- tural transforms22, the theoretical and methodological
ture household inventories. The first research task is to framework of behavioural archaeology is adequate for
explore whether there is reason to suspect that house this particular research problem. Even Chapman and
inventories are ordinary household assemblages or Gaydarska,23 despite their strong post-processual orien-
whether they represent symbolic deposits, as Chapman tation, acknowledge that Schiffers concepts are of key
claims. The second research task is to attempt to interpret importance in studying the fragmentation and deposi-
the variation in size and structure of household inven- tion of items in the Balkan Neolithic and Copper Age
tories in social terms. More concretely, this research contexts.
will address the following issues: Therefore, it can be claimed that, regardless of the
1. Is the quantity of material in house inventories general theoretical orientation and the side which one
unusually large? might take in a Schiffer-Binford debate24, it should not
2. Is there a correspondence between the structure be problematic to assert that it is of great importance for
of house assemblages and accumulated assemblages further social analysis to determine whether the inven-
from the cultural layers? tory of a house was actually a set of objects used in
3. Is it possible to offer a meaningful social inter- everyday activities. This is because correlations between
pretation based on patterns of inventory variability? inventory attributes and anthropological variables have
Since pottery makes up the bulk of all Late Neolithic been established for living (systemic) inventories only.
house inventories, analysis will be focused on this class It was demonstrated in several studies that attributes of
of artefacts. This problem will be explored by using house inventories (e.g., quantity and diversity) are more
data on house inventories from Vin~a culture sites. or less reliable correlates of anthropologically relevant
variables such as household size, household structure,
and household social and economical status.25 There-
THEORETICAL BASIS fore, if these correlates are to be used for inferring the

The issue of Late Neolithic house inventories can


be formulated in terms of Schiffers behavioural 18 Chapman 1999, 121.
archaeology.21 Schiffer makes an important distinction 19 Chapman 1999.
between the systemic and archaeological context of an 20 Chapman 2000b.
artefact. Artefacts are in a systemic context when they 21 LaMotta and Schiffer 1999; Schiffer 1972; Schiffer 1976;

are participating in a behavioural system, e.g., a cook- Schiffer 1987; Schiffer 1995.
22 Binford 1981; Binford 1987.
ing vessel is in its systemic context when someone is
23 Chapman and Gaydarska 2007, 7179.
preparing a meal in it, or when it is simply stored in a
24 Binford 1981; Binford 1987; Schiffer 1985.
kitchen waiting to be used for food preparation. Archa-
25 Arnold 1988; Arthur 2002; 2006; 2009; Castro et al. 1981;
eological context refers to artefacts which interact only
Deal 1998; Hayden and Cannon 1982; Hildebrand and Hagstrum
with the environment of the archaeological record. Arte- 1999a; Nelson 1981; Schiffer et al. 1981; Smith 1987; Tschopik
facts enter the archaeological record from the systemic 1950.

21
Marko POR^I], De facto refuse or structured deposition? (1943) STARINAR LXII/2012

Fig. 1. Approximate distribution of the Vin~a culture with sites relevant for this paper
Sl. 1. Okvirno prostirawe vin~anske kulture sa lokalitetima koji se pomiwu u tekstu

past, the validity of an inventory as a systemic variable erroneous to hastily equate this archaeological entity
must first be established. In other words, inventory with a single ethnic, political or linguistic unit.27 There-
attributes may be used as indicators of the aforemen- fore, the safest way to proceed is to understand the
tioned dynamic aspects only if the inventory itself rep- term culture as a technical label denoting an archaeo-
resents de facto refuse. logical phenomenon.
The Vin~a culture sites are usually permanent agri-
cultural settlements ranging in size from hamlets to
ARCHAEOLOGICAL BACKGROUND villages with relatively large population sizes.28 In
AND DATA ON HOUSE INVENTORIES general, faunal and botanical evidence show that most
Vin~a communities subsisted on a mixed economy
The Vin~a culture is a Late Neolithic culture which typical for the temperate European climate29: agricul-
extends across the Central Balkans covering an area of ture based on cereals30 animal husbandry dominated
around 300 km (Fig. 1) and encompassing Central by domestic animals such as cattle, pig, sheep and goats,
Serbia, Kosovo, southern parts of Vojvodina, Transyl- and accompanied, in a smaller or larger percentage, by
vania, Oltenia, eastern parts of Bosnia and northern parts
of Macedonia.26 Extending across such a large area, it
is one of the most geographically dominant archaeolo-
26 Brukner 2003; Chapman 1981; Gara{anin 1973; 1979;
gical phenomena in South-eastern Europe in the Late
1982b.
Neolithic. The anthropological reality which stands 27 Renfrew 1974.
behind the apparent uniformity of material culture 28 Por~i} 2011a.
across this vast area (characteristic black pottery and 29 See Barker 1985; Bogaard 2004.
clay figurines) is not yet understood, but it would be 30 Borojevi} 2006; Bottema and Ottaway 1982; van Zeist 2002.

22
Marko POR^I], De facto refuse or structured deposition? (1943) STARINAR LXII/2012

Phase (Miloj~i} 1949) Phase (Gara{anin 1979) Range (cal. BC)

Vin~a A Vin~a-Turdas I 5400/5300 - 5200

Vin~a B Vin~a Turdas II Gradac 5200 - 5000

Vin~a C Gradac Vin~a-Plo~nik I 5000/4950 - 4850

Vin~a D Vin~a-Plo~nik IIa, IIb 4850 4650/4600

Table 1. The absolute and relative chronology of Vin~a culture (after Bori} 2009).
Tabela 1. Apsolutna i relativna hronologija vin~anske kulture (prema Bori} 2009)

wild species such as red deer, roe deer and wild pig.31 Chapman40 gave pottery counts for most of the houses,
It should be emphasised that this is only a general state- however, there is no mention of these house inventories
ment, especially where animal husbandry is concerned. in the original publication41 and these pots could not
Individual faunal assemblages varied in structure bet- be traced in the Banjica collection kept at the Belgrade
ween sites and at sites such as Petnica and Opovo, wild City Museum.42 For this reason, these inventories are
species dominated.32 excluded from the analysis, because it is most probable
One more thing needs to be made clear about termi- that Chapman erroneously attributed the pottery to
nology. The Vin~a culture is traditionally labelled as a houses a very likely error given the poor state of do-
Neolithic culture, but recent research has shown that cumentation and the fact that the original excavation
metallurgy was present from the very beginning of its was carried out in the fifties. The only house with a
duration33, so, strictly speaking it is a Copper Age cul- certain and published inventory from Banjica is House
ture. However, this is not relevant for the issues explored 2/79 which belongs to the latest phase of the settlement
in this paper, so the traditional label will be kept for the the Vin~a D phase.43 Only complete or restorable
sake of consistency and compatibility with literature. vessels from the house were taken into account.
Relative chronology was established on the basis Divostin is located in Central Serbia and was exca-
of pottery typology from the stratigraphic sequence of vated by a joint American and Serbian archaeological
the eponymous site at Vin~aBelo Brdo near Belgrade, team.44 The total area of the site is estimated to be 15
Serbia. Two similar and compatible relative chronolo- hectares and an area of 2480 m was excavated. There
gical sequences (Table 1) were proposed by Gara{anin34
and Miloj~i}35, dividing the span of the Vin~a culture
into four major phases. Absolute dates for the Vin~a
31 Bla`i} 2011; Bknyi 1988; Dimitrijevi} 2008; Greenfield
culture and its phases were taken from Bori}s 2009
1986; Legge 1990; Orton, D. 2008; Russell 1993.
paper36 and are reproduced in Table 1. In calendar years, 32 See Orton, D. 2008; 2010.
the Vin~a culture began in 5400/5300 BC and ended in 33 Bori} 2009; Radivojevi} et al. 2010.
about 4650/4600 BC.37 34 Gara{anin 1951.

Only sites with published data on house inventories 35 Miloj~i} 1949.

were included in the analysis. In total there are 7 sites 36 Bori} 2009.
37 Bori} 2009.
with basic information on house inventories: Banjica,
38 Todorovi} and Cermanovi} 1961; Tripkovi}, B. 2007.
Divostin, Gomolava, JakovoKormadin, Obre`Bele-
39 Todorovi} 1981; Todorovi} and Cermanovi} 1961; Tripko-
tinci, Opovo, Predionica (Fig. 1).
vi}, B. 2007.
Banjica is located in the suburbs of Belgrade, near 40 Chapman 1981.
Avala Mountain. An area of 750 m has been investi- 41 Todorovi} and Cermanovi} 1961.
gated in several campaigns (19551957; 1979; 1998). 42 B. Tripkovi}, personal communication 2011.

Five building horizons were recorded.38 In total, 11 43 Todorovi} 1981; Tripkovi}, B. 2007.

houses have been excavated and published so far.39 44 McPherron and Srejovi} 1988.

23
Marko POR^I], De facto refuse or structured deposition? (1943) STARINAR LXII/2012

Fig. 2. Typical forms of functional classes of Vin~a pottery.


Sl. 2. Tipi~ni oblici funkcionalnih klasa vin~anske keramike

were two Vin~a D horizons, Divostin IIa and Divostin but only a single house (House 4) has so far been pub-
IIb, spanning, in total, 300400 years.45 There were 17 lished.52
houses from the Vin~a period. Only postholes are pre- JakovoKormadin is a site located in Jakovo village,
served from the Divostin IIa phase, while collapsed in the vicinity of Belgrade. The site area is estimated to
house rubble with sealed inventories was found in the be 4.5 hectares.53 There is a single Vin~a D horizon at this
Divostin IIb horizon. Houses 13, 14, 15, 16, 17, and 18 site. 399 m were excavated and two houses, destroyed
were included in the present analysis since they were by fire, were uncovered one completely (House 2)
completely excavated. House 12 was also completely and the other one only partially (House 1). The inven-
excavated but it was excluded from the analysis because tory of the completely excavated House 2 was pub-
it was severely damaged.46 lished in detail.54
The results of archaeomagnetic analysis of the Obre` is a site located in Srem, 40 km west of
house inventories suggest that the Divostin IIb houses Belgrade. The total area of the site was estimated to be
were all destroyed in a single accidental fire event.47 18.2 hectares, while only 290 m were excavated.55 The
However, the results of archaeomagnetic analysis of
the burnt daub suggest that houses 14 and 16 burned at
different times.48
Gomolava is a tell site, situated on the left bank of 45 Bori} 2009.
46
the Sava river. The total area of the Gomolava tell was Bogdanovi} 1988.
47
estimated49 to be 18400 m, of which 5000 m (27.17%) McPherron and Christopher 1988, 478.
48 Bucha and McPherron 1988, 386.
was excavated.50 There were three Vin~a culture hori- 49 van Zeist 2002.
zons: Gomolava Ia, Gomolava Iab, and Gomolava Ib, 50 Brukner 1988.
spanning a period of circa 350 years, from around 51 Bori} 2009.
5000 to 4650 ca. BC.51 A total of 31 houses were un- 52 Petrovi} 1992; 1993.
covered at Gomolava. Only house remains from the 53 Risti}-Opa~i} 2005.
Gomolava Ib settlement were well preserved, due to 54 Jovanovi} and Gli{i} 1961.
fire. In total, 24 houses were excavated in this horizon, 55 Brukner 1962.

24
Marko POR^I], De facto refuse or structured deposition? (1943) STARINAR LXII/2012

Vin~a culture horizon is dated to the Vin~a D phase. A GENERAL METHODOLOGY


single house was excavated and published. However,
the pottery inventory of the house was not published in According to Chapman, house inventories from
detail only the total vessel count was given. many Late Neolithic contexts in the Balkans were too
Opovo is a site located in the Serbian part of large, which prompted him to conclude that these
Banat, 20 km north of the small town of Pan~evo. The inventories were not representative of everyday or sys-
area of the site is estimated to be 5 hectares, and an temic assemblages. In other words, these assemblages
area of 380 m was excavated in great detail by the were unusual, in Chapmans opinion. However, if
joint American and Serbian team.56 The site is dated to something is to be labelled as unusual, there has to be
the Vin~a C phase. The contents of 3 out of 6 houses some standard against which the comparison is made
have so far been published.57 a frame of reference. Chapman does not mention any
Predionica is a site located in the vicinity of Pri{tina. referential frame, so it can be assumed that it is only
There were two Vin~a culture horizons: 1) earlier, dated the sheer size of certain house assemblages which led
to the Vin~a B phase 2) later, dated to the Vin~a C phase. him to conclude that they were unusual. Are there any
The inventory of House 1 from the later phase of Predi- other reasons to think that Vin~a house assemblages
onica was published.58 Most of the house area was ex- are not reflections of systemic assemblages?
cavated, so this house was also included in the sample. It is parsimonious to start with a null hypothesis
Pottery from house floors is the most numerous that assemblages are de facto refuse. The next step will
artefact class found in houses (usually over 90% of all be to compare these assemblages to an ethnographically
items). It is also the only data class which has been known range of variation. The goal is to determine
published completely and in sufficient detail. For these whether the average size of available assemblages from
reasons the analysis will focus on pottery as the major Vin~a houses falls within the ethnographically known
inventory component. range of variation. Cross-cultural data on average pot-
Vessels from houses are usually complete or can tery assemblage size were collated from Mills61 and
be reconstructed from fragments. They usually have Varien and Mills.62
traces of secondary burning (intense red colour) a If the average Vin~a culture household assemblage
consequence of the fire that consumed the houses. Pot- size falls within the known range of variation, this means
tery from house contexts is classified into three major that, from the perspective of that particular referential
functional classes storage, cooking and serving/con- frame, there is no case to answer. To avoid confusion,
sumption vessels. The classificatory scheme develo- this still does not prove that these are systemic assem-
ped by Madas59 for vessels from Divostin was used as blages. It only shows that there is nothing unusual
a basis for classification for other sites, as well. Madas about them in the perspective of this particular refe-
recognised four major functional classes: dry storage, rential frame. If, on the other hand, the archaeological
liquid storage, cooking and serving vessels. For the pur- assemblages fall outside the ethnographically known
poses of this paper, the dry storage and liquid storage range of variation, then it can be said that they are,
categories were collapsed into a single category of indeed, unusual but this still does not prove that they
storage vessels. are not systemic assemblages. So, by performing this
The most typical forms of functional classes are kind of analysis, what is tested is only the claim that
presented in Figure 2. Storage vessels are usually rep- there is something unusual about Vin~a house assem-
resented by large pithoi, jars (dry storage) or amphorae blages. The more relevant test of the null hypothesis
(liquid storage); cooking vessels by pots and casseroles; comes in the second step, which answers the second
serving/consumption vessels by bowls and plates. It is
acknowledged that equating function and form is often
problematic.60 However, the forms of different classes, 56 Tringham et al. 1992; Tringham et al. 1985.
as defined here, differ so sharply, so it can be safely 57 Tringham et al. 1992.
assumed that there is, at least, a general correspondence 58 Gli{i} 1964.
between function and form e.g., it is not likely that a 59 Madas 1988.
half meter tall pithos had been used as a serving or con- 60 Rice 2005, 211212.
sumption vessel. Data on house inventories are pre- 61 Mills 1989.
sented in Tables 23. 62 Varien and Mills 1997.

25
Marko POR^I], De facto refuse or structured deposition? (1943) STARINAR LXII/2012

Table 2: The structure of pottery assemblages from Vin~a houses included in this study
a Vessel counts for Divostin were made by the author directly from the maps published in the Divostin monograph (McPherron & Srejovi}
1988) only vessels which were strictly inside houses were included (vessels which appear to be inside postholes or beneath the line of the
wall, were excluded), and these counts may differ slightly from counts given in McPherron and Srejovi} (1988) and Tripkovi} (2009a).
b House floor areas for Divostin were calculated based on house dimensions as measured and reported by Tripkovi} (2009a).

Tabela 2. Struktura kerami~kih zbirki iz vin~anskih ku}a koje su ukqu~ene u ovu studiju

House Variables Cooking Dry Storage Liquid Storage Serving/Consumption Total

2 N (fragments) 144 44 25 161 374

m(g) 7166.67 3666.67 3666.67 3666.67 18166.67

3 N (fragments) 161 50 22 203 436

m(g) 7000 2333.33 1000 4000 14333.33

5 N (fragments) 81 150.00 56.00 83 370

(g) 10000 14000 8333.33 4000.00 36333.33

Table 3. House inventories from Opovo (after Tringham, et al. 1992, 376, Figure 11);
Vin~a C phase (Tringham, et al. 1992; Tringham, et al. 1985)
Tabela 3. Ku}ni inventari iz Opova (prema Tringham, et al. 1992, 376, Figure 11);
Vin~a C faza (Tringham, et al. 1992; Tringham, et al. 1985)

26
Marko POR^I], De facto refuse or structured deposition? (1943) STARINAR LXII/2012

Fig. 3. Box plot of ethnographically recorded range


of variation in average household assemblage size
(data from Mills 1989; Varien & Mills 1997)
and archaeologically recorded distribution
of individual Vin~a culture house assemblage sizes
(data from Table 2)
Sl. 3. Kutijasti dijagram
etnografski zabele`enog raspona varijacije
prose~nih veli~ina ku}nog inventara
(podaci iz Mills 1989; Varien and Mills 1997)
i arheolo{ki zabele`ene distribucije veli~ine
inventara pojedina~nih ku}a vin~anske kulture
(podaci iz Tabele 2)

question about the correspondence of house inventories deviation is 26.72. The average size of Vin~a culture
with cultural layer assemblages. house assemblages is 27.54 vessels, and the standard
The second question can be elaborated along these deviation is 10.7. It is apparent from Figure 3 that the
lines: if discovered house inventories were de facto re- average size of Vin~a household assemblages is well
fuse or systemic inventories, then it should be expected within the cross-cultural range of variation of mean
that the structure of pottery assemblage from the cul- household assemblage sizes. Moreover, it belongs to a
tural layer (accumulated assemblage) would correspond group with smaller assemblage sizes there are many
to the structure of house assemblage when differential societies where the mean number of pots per house-
use-life of different pottery classes is accounted for. hold is much larger than the largest individual Vin~a
The relationship between systemic assemblages assemblage.
and accumulated assemblages is the focus of accumu- Obviously, the quantity of Vin~a household pottery
lation studies.63 The idea is to use Schiffers discard assemblages should be viewed as neither unusual nor
equation64 to project the structure of accumulated assem- demanding any special explanation in the light of ethno-
blages from the structure of house assemblages. The graphically recorded variation. Even when assemblages
projected structure of the accumulated assemblage is from individual houses are inspected, extreme outliers
then compared to the observed (empirical) structure of cannot be found no individual assemblage contains
pottery assemblage from the cultural layer. more than 50 vessels (Table 2, Fig. 3). There are even
The third research question is related to the varia- opposite cases in Vin~a culture archaeology houses
tion of household inventory size and house floor area. with unusually small assemblages, such as the house
The first step is to look for patterns in the relationship from Medvednjak where only 3 vessels were found in
between pottery assemblage size and house floor area. the house.65
The second step is to see whether these patterns can be
meaningfully interpreted in social terms.
ACCUMULATION ANALYSIS

VIN^A ASSEMBLAGE SIZE COMPARED TO The Vin~a culture accumulated assemblages come
CROSS-CULTURAL RANGE OF VARIATION mostly from cultural layers and pits. Cultural layers are
artefact and ecofact rich deposits within which house
The box-plot in Figure 3 shows: 1) the distribution
of average household pottery assemblage sizes from
the available cross-cultural data 2) the distribution of 63 Lightfoot 1993; Mills 1989; Pauketat 1989; Schiffer 1976;
individual household assemblage sizes based on archa- Schiffer 1987; Varien and Mills 1997; Varien and Potter 1997.
eological data presented in Table 2. The cross-cultural 64 Schiffer 1976; 1987.

mean is 25.35 vessels per household and the standard 65 Galovi} 1975.

27
Marko POR^I], De facto refuse or structured deposition? (1943) STARINAR LXII/2012

features are inserted and subsurface features are cut, which the accumulation took place (usually more than
and they are a common feature of Late Neolithic settle- 100 years).
ments.66 Thinking about the accumulated assemblages In order to link the household assemblages to accu-
coming from pits or undefined cultural layers, brings mulated assemblages, Schiffers discard equation is
into focus the theoretical issue of cultural and practical used. The discard equation has the following form:72
logic.67 Are accumulated assemblages from Vin~a sites
the products of cultural or practical reason? Chapman T = (S * t) / L
views assemblages coming from pits as meaningful
and yet another example of structured deposition68 an where T is the total number of discarded vessels of a
idea which seems to be supported by empirical evidence certain functional class in the accumulated assemblage;
in some cases.69 Moreover, Chapmans explanation of S is the systemic number the average number of ves-
cultural layer assemblages is given in terms of cultural sels of that particular class in use; t is the duration of a
logic. As Chapman describes it, the typical Balkan site; L is the average use-life of an artefact class under
Late Neolithic and Copper Age village or farm was: consideration.
another kind of ambience in which a walk One needs to know the values of these variables in
around a settlement involved avoiding the larger, if not order to project T. However, if the goal is to project a
sharper, materials lying on the ground and was domi- structure of the accumulated assemblage in terms of
nated by the smells of decomposing human faeces, relative frequencies of artefact classes, then one only
vegetal and animal matter The basic image of NCA needs to know the average use-life of each class, since
settlements is of people living on top of, or within, what the relative frequencies of classes in the accumulated
most twentieth century archaeologists would call a assemblage will remain constant through time73, and
refuse tip. The implication of this striking picture is the relative frequencies of S for each class can be de-
that of the proximity of residents to their discarded termined from the available house inventories. So the
objects and food remains rather than strict segregation only thing which is needed is the use-life value for each
of refuse into rubbish pits.70 functional class. These values can be estimated from
According to this interpretation, people in Late Neo- ethnoarchaeological research.
lithic villages were guided by their traditional ethos of The second problem is that in almost all cases only
keeping household possessions close to the house, rather sherd counts were given for the cultural layer. There-
than the twentieth century rules of rubbish disposal. fore, the projected assemblage structure needs to be
Chapmans interpretation may or may not be true, expressed in sherds, not in complete vessels, in order
but it demonstrates one very important thing: this kind to be comparable to assemblages from cultural layers
of refuse disposal is not practical but purely cultural and pits. The problem is that different classes break
only if we look at it from our own cultural context.71 into different numbers of sherds. In the absence of
From the perspective of people living in the Late Neo- experimental and empirical data, fragmentation rates
lithic villages, such behaviour was guided by practical will have to be estimated (except for Opovo where house
reasons as well because it served as a means to achieve inventories are already given as sherd counts). What is
two goals: to dispose of broken items and to affirm the known is that larger vessels usually break into more
household ideology. Therefore, in the context of the fragments.74 The estimate has to be consistent with this
present paper, the issue of whether accumulated
assemblages resulted from behaviour guided by prac-
tical or cultural reasons is not relevant because what-
ever the idea behind the specific pattern of pottery dis- 66 Chapman 2000c.
67
card was, whether it was guided only by the need to Hutson and Stanton 2007; Walker 2002.
68 Chapman 2000b.
throw away used items or there was an additional ide-
69 Tripkovi}, A. 2011; Tripkovi}, B. et al. 2011
ological and symbolic statement involved, the end result
70 Chapman 2000c, 356.
is the same when the process of pottery accumulation 71 See Hutson and Stanton 2007 for an excellent discussion of
is being considered. Accumulated assemblages from this issue.
Vin~a sites come from various places within sites and 72 Schiffer 1976; 1987.

potential biases in relation to discard practices were 73 Mills 1989.

almost certainly averaged out over the long term during 74 Chase 1985.

28
Marko POR^I], De facto refuse or structured deposition? (1943) STARINAR LXII/2012

Table 4: Values of parameters needed for fragmentation ratio estimation


Surface areas of average vessels are calculated from pottery drawings given by Madas (1988), while fragment attributes were calculated
from Opovo data (Tringham, et al. 1992) and a pottery assemblage from Vin~a D horizon of the Vin~aBelo Brdo site (recorded by the author)

Tabela 4. Parametri potrebni za ocewivawe koli~nika fragmentacije

finding, so storage vessels should break into more sherds ture76 and from my own research. The average fragment
than cooking vessels, and cooking vessels should break thickness and mass for each vessel class was calculated
into more sherds than serving/consumption vessels. from a small Vin~a D pottery assemblage from the Vin~a
To summarise, the accumulation analysis will con- Belo Brdo site.77 The values of parameters are given
sist of several steps: in Table 4.
1. Estimation of average use-life values for each If the simplification is made that the shape of a
functional class on the basis of ethnoarchaeological fragment may be approximated by a thin cuboid, the
research. average fragment mass (mfr) for each functional class
2. Estimation of relative fragmentation rates for may be expressed as:
functional classes.
3. Projecting the accumulated assemblage and mfr = tfr * pfr * d (Equation 1)
assessing the fit.
where tfr is the average fragment thickness, pfr is the
Estimating average use-life area of the larger face of the fragment (approximated
Varien and Mills reviewed the ethnoarchaeological by a cuboid) and d is the specific density of ceramic
literature on average use-lives of pottery functional material.
classes and they reported the median values for different Average surface areas for each vessel class were cal-
functional classes.75 The median use-life for dry stor- culated using data from Divostin. Several representative
age containers is 7.5 years; 5 years for liquid storage; vessel shapes (coming from complete or reconstructed
1.7 years for cooking vessels; and 1.2 years for serving/ vessels) were chosen for each class and their surface
consumption vessels. For the purposes of this paper, dry areas were calculated on the basis of profile drawings
and liquid storage categories were grouped into a single given in the Divostin monograph.78 The calculation
category, and it was decided to set the average use-life proceeds in the following manner: 1) coordinates of
for the dry and liquid storage category at 7 years. Cross- several points (513) from the vessel profile drawing
cultural medians of 1.7 and 1.2 years are used for cook- are taken 2) when the lines connecting each two points
ing and serving/consumption vessels, respectively. on a profile are rotated around the axis of the vessel,
the vessel shape may be approximated by a series of
Estimating relative fragmentation rates cone segments 3) the lateral surface area of each cone
It can be shown that if one can make reasonable segment is calculated 4) the total lateral surface area of
estimates of the average fragment mass, average frag-
ment thickness and average vessel surface area for each
functional class, the average number of fragments per 75 Varien and Mills 1997.
vessel may be roughly estimated. 76 Tringham et al. 1992.
The average fragment masses for each functional 77 Recorded by the author, see Por~i} 2010.
class were calculated using available data from litera- 78 Madas 1988.

29
Marko POR^I], De facto refuse or structured deposition? (1943) STARINAR LXII/2012

the vessel is calculated by summing the individual seg- lines thus creating the polygon approximation of the
ments 5) the vessel bottom surface area is calculated as original vessel profile. The lateral surface area of the
an area of a circle 6) the vessels total surface area is vessel is divided into 7 segments. Equation 2 gives the
derived by summing the total lateral surface area and lateral surface area of each segment. For example, the
the bottom surface area. surface area of Segment 2 is calculated by substituting
The surface area of each individual cone segment the coordinates into the formula. There are two addi-
is calculated using this formula79: tional parameters in the formula which are needed to
calculate the surface area of Segment 2: the intercept
(a) and the slope of the line (b) connecting the point with
coordinates x2, y2 (400, 186.67) and the point with
(Equation 2) coordinates x3, y3 (333.33, 213.33). The slope of the
Variable x refers to the values of profile points line connecting two points is given by the following
along the vertical dimension of the vessel (height), formula: b = (y3 y2) / (x3 x4) = (213.33 186.67) /
while variable y refers to the distance of a profile point (333.33 400) = 0.4. The intercept of the line is cal-
from the vertical axis of the vessel. Parameters ai and culated in this way: a = y3 b * x3 = 213.33 (0.4)
bi are the intercept and the slope of the lines connect- * 333.33 = 346.66. When these values are substituted
ing each two adjacent points along the vessel profile. into Equation 2 the surface area of Segment 2 can be
The procedure for the calculation of surface area calculated:
will be demonstrated with the example of the vessel
profile in Figure 4. The coordinates of 8 points were
taken along the vessel profile. The axes in Figure 4 are
inverted the x axis is vertical, and the y axis is hori-
zontal. Each two points are connected with straight
The surface areas of the remaining segments are
calculated in a similar fashion and summed to get the
total lateral surface area of the vessel. The total surface
area of the vessel is calculated by summing the total
lateral surface area and the surface area of the bottom
of the vessel (the area of a circle with the radius equal
to the radius of the bottom of the vessel).
The surface area of a vessel (Pv) can also be approxi-
mated in terms of individual fragments:
Pv N * pfr
where N is the average number of fragments per vessel.
From Equation 1 it follows that pfr = mfr / (tfr * d).
Therefore:
Pv N * mfr / (tfr * d)
N (Pv * tfr * d) / mfr

Since the goal of projecting is to calculate the rel-


ative frequencies of classes in an accumulated assem-
blage, the only parameter that needs to be known is the
fragmentation ratio:
N2 / N1 = (Pv2 * tfr2 * mfr1) / (Pv1 * tfr1 * mfr2)
Note that d (specific density of ceramic material) cancels
out of the equation when the ratio is calculated.
Fig. 4. A worked example of Equation 2
calculating the lateral surface area of a vessel.
Sl. 4. Primer kori{}ewa jedna~ine 2
ra~unawe povr{ine posude 79 Based on Adna|evi} and Kadelburg 1998, 219229.

30
Marko POR^I], De facto refuse or structured deposition? (1943) STARINAR LXII/2012

For convenience, the serving/consumption class cation are not met. For example, it would be inappro-
will be set as the reference class. Two ratios are then priate to statistically test for the fit between the pro-
calculated cooking : serving/consumption and jected assemblage and observed assemblages given that
storage : serving/consumption fragmentation ratio. the classification of vessels and potsherds into classes
The resulting ratios are 1.53 and 3.79 for cooking : is not strictly the same between sites and researches.
serving/consumption and storage : serving/consump- The data on empirically observed accumulated assem-
tion ratio, respectively. blages from Table 5 are collated from various sources
and authors working with classificatory schemata,
Assessing the fit which are only comparable in general. Most of them
A direct comparison between the empirical and are almost certainly biased (usually the proportion of
projected assemblages is possible only in the case of bowls is inflated given the high rate of identification of
Opovo where descriptive, statistical and typological ana- this vessel class), as a result of the fact that protocols
lysis of the pottery from the cultural layer is available.80 for estimating the relative frequencies of classes, using
In all other cases the comparison will have to be made cumulative rim proportions or recording potsherd
indirectly by comparing projected assemblages with weight,81 are rarely, if ever, used in the primary analy-
empirical assemblages from different sites. This is a sis of pottery from Vin~a culture sites. The data sets are
reasonable compromise because the structures of accu- comparable in general, but this general correspondence
mulated assemblages are relatively stable across dif- is not sufficient to warrant the use of statistical techni-
ferent sites (Table 5) roughly 5070% serving/con- ques which require strictly comparable units. Even in
sumption vessels (bowls, plates, cups), 2030% cooking the case of Opovo, where data on household and cultu-
vessels (pots, jars, casseroles), 520% storage vessels ral layer assemblage is present, it would be erroneous to
(amphorae and pithoi). The rank order of vessel class use the chi squared test because the observed accumu-
proportions is, in most cases, the same serving/con- lated assemblage should not be expected to match the
sumption vessels are most numerous, cooking vessels projected assemblages exactly, even if the household
are ranked second, and the storage vessels are the least assemblages from excavated houses were systemic. This
numerous in most of the observed accumulated assem- is because the excavated assemblage from the cultural
blages. layer almost certainly contains a fraction of pottery,
If the projected and empirical structures match, which was accumulated from other houses and other
then there is no reason to doubt that house inventories parts of the site, not to mention the fact that the potsherd
are de facto refuse. For reasons explained below, pre- counts are slightly biased for different vessel classes
cise matching criteria cannot be defined. Only a general given their differential potential for identification. For
criterion can be defined to distinguish between the match example, bowl fragments usually have preserved diag-
and mismatch between the observed and projected nostic parts such as rims and complete profiles due to
assemblages: we can say that the projected assemblage their relatively low height-to-width ratio, which makes
generally matches the observed assemblage if the rank- them more likely to be identified in spite of their small
ing of proportions of functional vessel classes is the fragment size. The general implication is that there is
same as in most empirically observed accumulated an amount of error built into the observed assemblage
assemblages. For example, the projected and observed in relation to the projected assemblage, which, a priori,
accumulated assemblages will be considered to match makes an exact match unlikely. Standard interpretation
if the projected assemblage displays such structure that of the statistical significance would be misleading in
serving vessels are the most numerous, followed by such a situation. Therefore, the degree of (mis)match
cooking and storage vessels, respectively. between empirical and projected assemblages should
If the assemblages do not match, then there are only be used in a qualitative manner as a measure of
three possible explanations: 1) house inventories are our suspicion that household assemblages are not de
not systemic inventories 2) the assumptions are wrong facto refuse.
3) house samples are not representative. Statistical tests
are not used here for two reasons: 1) this is mainly
exploratory research 2) data quality is very poor so it
would be inappropriate to simulate precision and rigor 80 Tringham et al. 1992.
by using formal tests where conditions for their appli- 81 See Orton, C. 1980; Orton, C. et al. 1993.

31
Marko POR^I], De facto refuse or structured deposition? (1943) STARINAR LXII/2012

Table 5. Accumulated assemblage structures from Vin~a culture sites


Tabela 5. Strukture akumuliranih zbirki (zbirki iz sloja) sa vin~anskih lokaliteta

32
Marko POR^I], De facto refuse or structured deposition? (1943) STARINAR LXII/2012

70
66.29
Site Cooking Serving Storage Observed
60 Projected 57.34

Banjica 45.74 37.06 17.20 50

Gomolava 59.38 36.65 3.97 40

Percent
35.03
30
Divostin 30.83 49.77 19.40 27.89

20
Jakovo 77.64 11.98 10.38
10
7.63
5.82
Predionica 18.91 58.35 22.75
0
Storage Cooking Serving/Consumption

Table 6. Projected accumulated assemblages (simple projection)


Fig. 5. Opovo projected and observed assemblages
Tabela 6. Projektovane strukture akumuliranih zbirki (prosta projekcija)
Sl. 5. Opovo projektovane i empirijski zabele`ene strukture zbirki iz sloja

Results of accumulation analysis 1. They come from single houses, so they still may
The projected accumulated assemblages for sites be systemic assemblages, but not representative of an
included in this study are shown in Table 6. Opovo is entire settlement. It should be kept in mind that accu-
the only site where projected and empirical assemblages mulated assemblages reflect the inventories from all of
can be compared directly (Fig. 5). There is a general the households in the settlement differences in assem-
match in projected and observed accumulation struc- blage structures between individual households are aver-
ture in the cases of Divostin, Opovo and Predionica. aged out in the accumulation assemblage. Therefore, it
This means that the ordering of relative frequencies of can be expected that the individual household will pro-
functional classes is consistent with the ordering obser- duce an accumulation assemblage similar in structure to
ved in other Vin~a accumulated assemblages (Table 5), the accumulation assemblage of the entire settlement,
or with the actual accumulated assemblage in the case of only if that particular house inventory is sufficiently simi-
Opovo. Projected assemblages from Banjica, Gomolava lar to the average house inventory for that settlement.
and JakovoKormadin do not match, not even in general, 2. The inventories of Jakovo and Gomolava are,
with the structure of other Vin~a assemblages. indeed, systemic; they are representative of the entire
The results of the accumulation analysis are not as settlement, although this cannot be confirmed since the
clear cut. Some projected assemblages conform to the descriptive statistics of the accumulated assemblages
observed ones, and some do not. It is important to note are not available for these particular sites.
that all of the projections which do not match the 3. Inventories are not systemic assemblages. They
empirical structure are based on single house assem- are ritually deposited assemblages.
blages. Unlike them, two out of the three projected 4. House inventories are not complete systemic
assemblages which generally do match the observed assemblages. They are de facto refuse, but the light
ones are based on six (Divostin) and three (Opovo) objects such as bowls were curated82 taken away
house assemblages, which makes these projections from the house before abandonment.
more representative and reliable. Obvious exceptions It is not possible at this moment to tell which of these
are assemblages from Jakovo and Gomolava, and, to a explanations is more probable. Therefore, no unequivo-
lesser extent, Banjica. Simple projections based on cal conclusion can be made regarding these particular
house assemblages from Gomolava and Jakovo do not sites where structured deposition is concerned.
produce anything that resembles the empirically
recorded accumulation assemblage structures. There
may be several explanations for the assemblages from
Jakovo and Gomolava: 82 Sensu Binford 1979.

33
Marko POR^I], De facto refuse or structured deposition? (1943) STARINAR LXII/2012

PATTERNS OF ASSEMBLAGE SIZE


AND HOUSE FLOOR AREA VARIATION
THE HOUSEHOLD ARCHAEOLOGY
OF VIN^A CULTURE HOUSES

Is there any other available frame of reference which


would enable the archaeologist to identify unusual
assemblages? This paper focused only on the external
criteria external in the sense that house inventory
attributes were compared against attributes measured
in domains external to the houses themselves (ethno-
graphic records and accumulated assemblage). How-
ever, it is possible to use an internal criterion which
would enable the archaeologist to recognise unusual
assemblages in relative terms relative to other
Fig. 6. Scatterplot of pottery assemblage size
assemblages. One such criterion would be the ratio of
and house floor area of Vin~a culture houses
total vessel count to house floor area. Figure 6 shows
the scatter-plot with total vessel count and house floor Sl. 6. Grafikon koji predstavqa veli~inu
area. It is apparent that there are extreme outliers such kerami~kog inventara vin~anskih ku}a
as houses from Obre` and Banjica, and somewhat less u odnosu na wihovu povr{inu
pronounced outliers such as the house from Jakovo
and house 17 from Divostin. Obre`, Banjica, and, to a
lesser extent, Jakovo, are outliers because their pottery follow. Ethnoarchaeological studies have shown that
inventories are too large for their house floor areas. pottery assemblage sizes may differ greatly between,
They are too large only in relative terms because such and within, villages belonging to the same culture and
a claim would not be possible if there were no houses society in the ethnographic sense.86 Ethnoarchaeology
from Divostin with nearly equal pottery assemblage also shows that the correlation between total assemblage
sizes (houses 13 and 14) and much higher house floor size and household size (reflected in house floor area)
areas, and if there were no houses with almost equal may not always be present for a variety of reasons.87
house floor area and smaller assemblage sizes (Gomo- If there is no reason to reject the null hypothesis
lava, Predionica, Divostin 18, Divostin 16). that the majority of Vin~a house inventories are more
Does this finally offer any evidence which might or less faithful reflections of systemic inventories, then
support the structured deposition of pottery vessels in their properties may be used as correlates of anthropo-
Banjica, Obre` and Jakovo? It might, if one were will- logical phenomena of interest (e.g., household size,
ing to accept the assumption that all Vin~a culture sites wealth, status). The archaeological study of the vari-
should have equal average household assemblage ability of house inventories might lead to socially rele-
sizes and that they should have an equal vessel count vant information.88 Given the lack of large scale exca-
to house floor area ratio. Such an assumption would be vations, the full potential of household archaeology
very close to the traditional culture-historical essen- cannot be fully exploited at most Vin~a sites. However,
tialism, which equates archaeological phenomena with modest steps in this direction have been made for the site
ethnographic phenomena. However, this assumption is of Divostin.89 An attempt will be made to interpret the
probably not true given the large territory of Vin~a cul-
ture and given the great differences between various
Vin~a sites in household size83, subsistence84 and per- 83 Por~i} 2010; Tripkovi}, B. 2009a.
haps, but less likely, marital residence patterns. 85 But 84 Orton, D. 2008.
even if this assumption of cultural uniformity was true, 85 Por~i} 2011b.
the conclusion that some houses are unusual because 86 Arnold 1988; Arthur 2006; 2009.
they do not conform to the pattern (constant assemblage 87 Arnold 1988; Deal 1998.
size to house floor area ratio) or because they differ in 88 e.g., Shelach 2006.
assemblage size from other houses, does not necessarily 89 Por~i} 2010; Tripkovi}, B. 2009a; 2009b.

34
Marko POR^I], De facto refuse or structured deposition? (1943) STARINAR LXII/2012

variability in pottery inventories from Divostin houses. expanded by building additional rooms. Tripkovi}
The reader should bear in mind that these are only tenta- analysed features such as ovens, furniture (fixed clay
tive interpretations since the data limitations resulting containers and banks) and floor plaster layers, and
from the poor state of research are considerable. concluded that the structure of the house was modular,
In Divostin, the correlation between pottery assem- leading to a hypothesis that each room might have
blage size and house floor area is relatively high and housed a single nuclear family within a larger house-
marginally significant (r = 0.712, one-tailed p = 0.053, hold unit residing in the house. He also noted that the
see Fig. 6). What are the social implications of this cor- reason for house expansion might have been the higher
relation? Ethnoarchaeology shows that the correlation production level of these households. From this per-
between total assemblage size and household size spective, differences in pottery assemblage sizes
(reflected in house floor area) usually ranges from between houses might be interpreted as differences in
0.30.590, but may not always be present for various household sizes.
reasons.91 Moreover, two studies show that the num- What are the social implications of these differen-
ber of serving vessels may be the most reliable indicator ces? Returning to the issue of correlates, ethnoarchae-
of household size, even when the correlation between ological research shows that the quantity of pottery
household size and the total pottery count is not sig- may correlate with the social status of the household.97
nificant.92 Correlation between the number of serving Correlation between the quantity of pots and social
vessels and house floor area in Divostin is moderate, status may be explained by the fact that higher social
but not significant at the 0.05 level (r = 0.547, one- status often entails the organisation of social food con-
tailed p = 0.131), which is not surprising, given the low sumption events, such as feasts. Serving and consum-
sample size. Can this convergence of two independent ption vessels are particularly important in such con-
household size indicators such as house floor area and texts.98 The observed pattern is additionally reinforced
pottery assemblage size be used to derive a socially by the fact that a copper bracelet was found in House
meaningful interpretation? The answer is positive, but 14.99 The social significance of copper items in Vin~a
the reasons for such an answer are not simple, since the culture contexts is not fully understood100, but the pre-
relationship between house floor area and inventory on sence of a copper bracelet and copper pearls as grave
one side, and socio-economic variables on the other, is goods in the Late Vin~a culture graves in Gomolava,
rather complex.93 where only males of differing ages from a single patri-
First of all, house floor area is a correlate of house- line were interred101, may suggest that copper items
hold size on the settlement level, not on the individual were important status markers.102
household level average house floor area is an indi- This suggests that variation in assemblage sizes
cator of average household size94, but individual house may be related to both household size and social sta-
floor area is usually not an indicator of individual tus. This is not a surprising find. On the contrary, in
household size.95 This is because the size of an indi- light of what is known from the domain of theories of
vidual household is not a constant it is a variable
which changes during the household life cycle (new
members are born, some members die, some leave the
90Arnold 1988; Arthur 2009; Hildebrand and Hagstrum
house etc.). However, there are situations where dif-
1999b; Nelson 1981.
ferences in house floor area between groups of houses 91 Arnold 1988; Deal 1998.
within a settlement might be interpreted as differences 92 Arthur 2009; Hildebrand and Hagstrum 1999a.

in household sizes. If variability in individual house- 93 Hayden and Cannon 1982; Wilk 1982.

hold sizes within a settlement is sufficiently large, this 94 Brown 1987, Por~i} 2012.
95 e.g.Wilk 1982.
would be reflected in the house floor area. Likewise, if
96 Tripkovi}, B. 2009b.
the architecture tracks the household size more closely
97 Deal 1998, 102.
e.g., a new space is built and added to the existing
98 Blitz 1993; Costin and Earle 1989; Nelson 1981; Potter 2000;
house to accommodate new members house floor area
Smith 1987.
can be used as an indicator of individual household 99 Bori} 2009; McPherron and Srejovi} 1988.
sizes. B. Tripkovi} makes a good case for household 100 Greenfield 1999; Orton, D. 2008, 268.

continuities in Divostin, particularly for houses 13, 14, 101 Stefanovi} 2008.

and 15.96 In his opinion, these three houses were 102 Bori} 1996.

35
Marko POR^I], De facto refuse or structured deposition? (1943) STARINAR LXII/2012

peasant economy and the domestic mode of produc- In the light of new theoretical and conceptual de-
tion,103 it makes good sense. Differences in production velopments regarding the distinction between cultural
levels between households may arise as a result of reason and practical reason, it is becoming apparent
chance fluctuations in individual household demogra- that there is no sharp dichotomy between these two do-
phy through time104, but the true question is how are mains.108 As Hutson and Stanton note109, practical logic
these temporary and ephemeral advantages and disad- is best viewed as embedded within cultural logic. Further-
vantages translated into more permanent status differ- more, the two may, and often do, coincide an action
ences. One possible way of solving the problem of sto- may be both practical and have a unique culturally deter-
chastic fluctuations in the labour force and creating a mined meaning at the same time. For these reasons, the
basis for status and wealth accumulation is to make term de facto refuse may be ambiguous in the context of
larger households.105 In this way, fluctuations in the the present research problem. In its most strict sense, de
labour force are smoothed by the intergenerational facto refuse implies that it is a product of practical reason
structure of complex households. This scenario is also it is a refuse that was left behind for practical reasons
consistent with Tripkovi}s idea of household exten- (e.g., to get away from a fire or a raid, to move to a new
sion and continuity. 106 This means that the observed location). However, in the context of this paper, the term
patterns may reflect the underlying social process of de facto refuse primarily means that the archaeological
incipient ranking and social differentiation.107 house inventory is the reflection of the systemic inven-
tory, regardless of the reasons for its placement into the
house. This means that, in the technical sense, the every-
GENERAL DISCUSSION day household assemblage left inside the deliberately
AND CONCLUSION burnt house as a kind of symbolic statement (e.g. as envi-
sioned by Chapman), would still be a de facto refuse. In
In general, it can be concluded that there is no rea- this way, the distinction between de facto refuse and struc-
son to suspect that Vin~a house assemblages reflect tured deposition may be blurred, but this is of no relevan-
systemic inventories. This does not mean that all of the ce for the central question of whether house inventories
inventories are de facto refuse or perfect reflections of can be viewed as reflections of systemic inventories.
a systemic inventory. After all, there is no reason to Such a conclusion may seem to be anticlimactic,
believe, a priori, that all Late Neolithic houses were but it should be emphasised that the purpose of this
abandoned for the same reason and in the same man- paper was not to prove or disprove the hypothesis that
ner. What this paper claims is that not enough evidence houses were burned intentionally, but to answer the
has been found so far that would justify the claim that specific question of whether there are reasons to believe
the particular house inventories analysed in this study that household pottery inventories from these particu-
are not systemic. Moreover, it was demonstrated that lar Vin~a culture sites do not reflect systemic invento-
patterns of variation in household assemblages can be ries. Chapman presented many other lines of evidence
meaningfully interpreted in social terms in the case of (burnt human and animal bodies inside houses, the
Divostin. presence of altars and figurines), which make his hypo-
If it is granted that, at least, assemblages from Divo- thesis of deliberate house burning in the Late Neolithic
stin, Opovo and Predionica are de facto refuse and do and Early Copper Age compelling110, especially when
reflect a systemic inventory, what are the implications
of this conclusion on scenarios of house abandonment
proposed by Chapman, Stevanovi} and Tringham? Does
this conclusion contradict the hypothesis that houses 103 Chayanov 1986; Sahlins 1972.
were intentionally burnt? Not necessarily. It may be 104 Pauketat 1996.
105 Hammel 2005.
consistent with deliberate house burning inhabitants
106 For a good theoretical discussion of household continuity
might have simply left the entire inventory inside the
see Blanton 1995.
deliberately destroyed house. The house would have 107 See Price and Feinman 1995; 2010; Wason 1994.
died along with its contents. This could be a symbolic 108 Chapman 2000b, 2000c; Hutson and Stanton 2007; Walker
statement, as well, just as Chapman hypothesised, 2002; Wilk 1996.
although in this scenario, it was made with an ordinary, 109 Hutson and Stanton 2007, 141.

everyday assemblage. 110 Chapman 1999.

36
Marko POR^I], De facto refuse or structured deposition? (1943) STARINAR LXII/2012

combined with research undertaken by Stevanovi}111. Acknowledgments


However, this paper was not about the intentional I would like to thank David Orton and Boban
burning of houses, it was only about the claim that pot- Tripkovi} for their help and useful comments on an
tery inventories from houses do not reflect systemic earlier draft of this paper. Special thanks go to Mladen
assemblages. The burden of proof is always on the one Nikoli} of the Faculty of Mathematics, University of
who makes the claim, so the fact that poor data and the Belgrade for providing the formula for the calculation
poor state of research of Vin~a sites do not allow the of vessel surfaces. The responsibility for errors and
structured deposition to be rejected conclusively can- omissions is exclusively mine.
not be used as an argument in favour of the structured
deposition hypothesis. Translated by Marko Por~i}

111 Stevanovi} 1997.

37
Marko POR^I], De facto refuse or structured deposition? (1943) STARINAR LXII/2012

BIBLIOGRAPHY:

Adna|evi} and Kadelburg 1998 D. Adna|evi}, Binford 1987 L. R. Binford, Researching


and Z. Kadelburg, Matemati~ka analiza I, Beograd, 1998. Ambiguity: Frames of Reference and Site Structure, in:
Arnold 1988 P. J. Arnold, Household Ceramic Method and Theory for Activity Area Research (ed. S.
Assemblage Attributes in the Sierra de los Tuxtlas, Vera- Kent), New York, 1987, pp. 449512.
cruz, Mexico. Journal of Anthropological Research 44, Blanton 1995 R. E. Blanton, The Cultural
1988, 357383. Foundations of Inequality in Households, in: Foundations
Arthur 2002 J. W. Arthur, Pottery Use-Alteration of Social Inequality (ed. D. T. Price and G. M. Feinman),
as an Indicator of Socioeconomic Status: An Ethno- New York, 1995, pp. 105128.
archaeological Study of the Gamo of Ethiopia. Journal of Bla`i} 2011 S. Bla`i}, Fauna kasnovin~anskih
Archaeological Method and Theory 9, 2002, 331355. stani{ta Crkvine i Bele`. Kolubara 5, 2011, 239250.
Arthur 2006 J. W. Arthur, Living with Pottery: Blitz 1993 J. H. Blitz, Big Pots for Big Shots:
Ethnoarchaeology among the Gamo of Southwest Ethi- Feasting and Storage in a Mississippian Community.
opia, Salt Lake City, 2006. American Antiquity 58, 1993, 8096.
Arthut 2009 J. W. Arthur, Understanding House- Bogaard 2004 A. Bogaard, Neolithic Farming in
hold Population through Ceramic Assemblage Forma- Central Europe, London, 2004.
tion: Ceramic Ethnoarchaeology among the Gamo of Bogdanovi} 1988 M. Bogdanovi}. 1988. Archi-
Southwestern Ethiopia. American Antiquity 74, 2009, tecture and Structural Features at Divostin, in: Divostin
3148. and the Neolithic of Central Serbia, (ed. A. McPherron
Bailey 1999 D. W. Bailey, The built environment: and D. Srejovi}), Pittsburgh, 35141.
pit-huts and houses in the Neolithic. Documenta Praehi- Bknyi 1988 S. Bknyi. 1988. The Neolithic
storica 26, 1999, 153162. fauna of Divostin, in: Divostin and the Neolithic of
Bailey 2000 D. W. Bailey, Balkan Prehistory: Ex- Central Serbia, (ed. A. McPherron and D. Srejovi}),
clusion, Incorporation and Identity, London, 2000. Pittsburgh, 419445.
Bankoff and Winter 1979 A. Bankoff, and F. Bori} 1996 D. Bori}, Social Dimensions of
Winter, A house-burning in Serbia. Archaeology 32, 1979, Mortuary Practices. Starinar 47, 1996, 6783.
814. Bori} 2008 D. Bori}, First Households and House
Bankoff and Greenfield 1984 A. H. Bankoff, Societies in European Prehistory, in: Prehistoric Europe:
and H. J. Greenfield, Decision-making and culture Theory and Practice, (ed. A. Jones), Oxford, 2008,
change in Yugoslav Bronze Age. Balcanica 15, 1984, 109142.
731. Bori} 2009 D. Bori}, Absolute Dating of Metal-
Bankoff and Palavestra 1986 A. H. Bankoff, and lurgical Innovations in the Vin~a Culture of the Balkans,
A. Palavestra, Prehistoric settlements in the Ribarska in: Metals and Societies: Studies in Honour of Barbara
Reka microregion near Kru{evac. Balcanica 16/17, 1986, S. Ottaway, (ed. T. Kienlin and B. Roberts), Bonn, 2009,
1742. 191245.
Bankoff and Winter 1990 A. H. Bankoff, and F. Borojevi} 2006 K. Borojevi}, Terra and Silva in
A. Winter, The Later Aeneolithic in Southeastern Europe. the Pannonian Plain: Opovo agro-gathering in the Late
American Journal of Archaeology 94, 1990, 175191. Neolithic, Oxford, 2006.
Barker 1985 G. Barker, Prehistoric Farming in Bottema and Ottaway 1982 S. Bottema, and B.
Europe, Cambridge, 1985. S. Ottaway, Botanical, malacological and archaeologi-
Benac 1971 A. Benac, Obre II Neolitsko naselje cal zonation of settlement deposits at Gomolava.
butmirske grupe na Gornjem polju. Glasnik Zemaljskog Journal of Archaeological Science 9, 1982, 221246.
muzeja Bosne i Hercegovine 26, 1971, 5178. Britain and Harris 2010 M. Brittain, and O.
Binford 1979 L. R. Binford, Organization and Harris, Enchaining arguments and fragmenting assum-
Formation Processes: Looking at Curated Technologies. ptions: reconsidering the fragmentation debate in
Journal of Anthropological Research 35, 1979, 255273. archaeology. World Archaeology 42, 2010, 581594.
Binford 1989 L. R. Binford, Behavioral Archae- Brown 1987 B. M. Brown, Population Estimation
ology and the Pompeii Premise. Journal of Anthropo- From Floor Area: a Restudy of Narolls Constant
logical Research 37, 1981, 195208. Cross-Cultural Research 21, 1987, 149.

38
Marko POR^I], De facto refuse or structured deposition? (1943) STARINAR LXII/2012

Brukner 1962 B. Brukner, Praistorijsko naselje na Costin and Earle C. L. Costin, and T. Earle, Status
potesu Beletinci kod Obre`a, Rad Vojvo|anskih muzeja Distinction and Legitimation of Power as Reflected in
11, 1962. Changing Patterns of Consumption in Late Prehispanic
Brukner 1988 B. Brukner, Die Siedlung der Peru. American Antiquity 54, 1989, 691714.
Vin~aGruppe auf Gomolava (Die Wohnschicht des Deal 1998 M. Deal, Pottery Ethnoarchaeology in
Sptneolithikums und Frhneolithikums Gomolava the Central Maya Highlands, Salt Lake City, 1998.
Ia, Gomolava Iab und Gomolava Ib) und der Wohn- Dimitrijevi} 2008 V. Dimitrijevi}, Vertebrate
horizont des neolithischen Humus (Gomolava II), in: Fauna of Vin~a Belo Brdo (Excavation Campaigns
Gomolava: Chronologie und Stratigraphie der vorge- 19982003). Starinar 56, 2008, 245269.
schichtlichen und antiken Kulturen der Donauniederung Galovi} 1975 R. Galovi}, Neolitska ritualna grupa
und Sdosteuropas, (ed. N. Tasi} and J. Petrovi}), Novi iz Smederevske Palanke. Zbornik Narodnog muzeja 8,
Sad, 1988, 1938. 1975, 2130.
Brukner 2003 B. Brukner, Vin~anska kultura u Gara{anin 1951 M. Gara{anin, Hronologija vin-
prostoru i vremenu, Rad muzeja Vojvodine 4345, 2003, ~anske grupe, Ljubljana, 1951.
728. Gara{anin 1973 M. Gara{anin, Praistorija na tlu
Bucha and McPherron 1988 V. Bucha, and A. SR Srbije, Beograd, 1973.
McPherron, Archaeomagnetic Measurements, in: Divo- Gara{anin 1979 M. Gara{anin, Centralnobalkan-
stin and the Neolithic of Central Serbia, (ed. A. ska zona, in: Praistorija jugoslavenskih zemalja, vol. 2,
McPherron and D. Srejovi}), Pittsburgh, 1988, 384387. (ed. A. Benac), Sarajevo, 1979, 79212.
Castro et al. 1981 A. P. Castro, N. T. Hakansson, Gara{anin 1982a M. Gara{anin, The Eneolithic
and D. Brokensha, Indicators of rural inequality. World Period in the Central Balkan Area, in: Cambridge
Development 9, 1981, 401427. Ancient History, Volume 3, Part 1), Cambridge, 1982a,
Chapman 1981 J. Chapman, The Vin~a Culture 136162.
of South East Europe, Oxford, 1981. Gara{anin 1982b M. Gara{anin, The Stone Age
Chapman 1999 J. Chapman, Deliberate house- in the Central Balkan Area, in: Cambridge Ancient
burning in the prehistory of Central and Eastern Europe, History, Volume 3, Part 1), Cambridge, 1982b, 75135.
in: Glyfer och arkeologiska rum: En vnbok till Jarl Gheorghiu 2011 D. Gheorghiu, Built to be Burnt:
Nordbladh, (ed. A. Gustaffson and H. Karlsson), Gte- The Building and Combustion of Chalcolitic Dwellings
borg, 1999, 113126. in the Lower Danube and Eastern Carpathian Areas, in:
Chapman 2000a J. Chapman, Fragmentation in Circumpontica in Prehistory: Western Eurasian Studies
Archaeology: People, Places and Broken Objects in the in memory of Eugen Coma, BAR International Series
Prehistory of South Eastern Europe London, 2000a. 10144, (ed. L. Nikolova, M. Merlini, and A. Coma),
Chapman 2000b J. Chapman, Pit-digging and Oxford, 2011, 4962.
structured deposition in the Neolithic and Copper Age Gimbutas 1977 M. Gimbutas, The first wave of
of Central and Eastern Europe. Proceedings of the Eurasian steppe pastoralists into Copper Age Europe.
Prehistoric Society 61, 2000b, 6187. Journal of Indo-European Studies 5, 1977, 277338.
Chapman 2000c J. Chapman, Rubbish-dumps Gimbutas 2007 M. Gimbutas, The Goddesses
or places of deposition?: Neolithic and Copper Age and Gods of Old Europe: Myths and Cult Images,
settlements in Central and Eastern Europe, in: Neolithic Berkeley, 2007.
Orkney in its European context, (ed. A. Ritchie), Gli{i} 1964 J. Gli{i}, Stratigrafija naselja vin~an-
Cambridge, 2000c, 347362. ske grupe kod Predionice u Pri{tini. Glasnik Muzeja
Chapman and Gaydarska 2007 J. Chapman, Kosova i Metohije 78, 1964, 1161.
and B. Gaydarska, Parts and Wholes: Fragmentation in Gli{i} 1968 J. Gli{i}, Ekonomika i socijalno
Prehistoric Context, Oxford, 2007. ekonomski odnosi u neolitu Podunavsko-pomoravskog
Chase 1985 P. G. Chase, Whole Vessels and basena, in: Neolit centralnog Balkana, (ed. L. Trifuno-
Sherds: An Experimental Investigation of Their Quan- vi}), Beograd, 1968, 2161.
titative Relationships. Journal of Field Archaeology 12, Greenfield 1986 H. J. Greenfield, The paleo-
1985, 213218. economy of the Central Balkans (Serbia): A zooarcha-
Chayanov 1986 A. V. Chayanov, The Theory of eological perspective on the Late Neolithic and Bronze
Peasant Economy, Wisconsin, 1986. Age (45001000 B.C.) Oxford, 1986.

39
Marko POR^I], De facto refuse or structured deposition? (1943) STARINAR LXII/2012

Greenfield 1999 H. J. Greenfield, The Origins of of Settlements and Regions: Ethnoarchaeological and
Metallurgy: Distinguishing Stone from Metal Cut-marks Archaeological Approaches (ed. C. M. Cameron and S.
on Bones from Archaeological Sites. Journal of A. Tomka), Cambridge, 1993, 165177.
Archaeological Science 26, 1999, 797808. Madas 1988 D. Madas, Ceramic Vessels from the
Hammel 2005 E. A. Hammel, Chayanov revisited: Divostin II House Floors, in: Divostin and the Neolithic
Kinship-based politics and the optimal size of kin groups. of Central Serbia, (ed. A. McPherron and D. Srejovi}),
Proceedings of the National Academy of Sciences 102, Pittsburgh, 1988, 143150.
2005, 70437046. Mallory 1991 J. P. Mallory, In Search of the Indo-
Hayden and Cannon 1982 B. Hayden, and A. Europeans: Language, Archaeology and Myth, London,
Cannon, The Corporate Group as an Archaeological 1991.
Unit. Journal of Anthropological Archaeology 1, 1982, McPherron and Christopher 1988 A. McPherron,
132158. and C. Christopher, The Balkan Neolithic and the
Hildebrand and Hagstrum 1999a J. A. Hilde- Divostin Project in Perspective, in: Divostin and the
brand, and M. B. Hagstrum, New Approaches to Ceramic Neolithic of Central Serbia, (ed. A. McPherron and D.
Use and Discard: Cooking Pottery from the Peruvian Srejovi}), Pittsburgh, 1988, 463492.
Andes in Ethnoarchaeological Perspective. Latin Ame- McPherron and Srejovi} 1988 A. McPherron,
rican Antiquity 10, 1999a, 2546. and D. Srejovi}, Divostin and the Neolithic of Central
Hildebrand and Hagstrum 1999b J. A. Hilde- Serbia, Pittsburgh, 1988.
brand, and M. B. Hagstrum, New Approaches to Ceramic Mills 1989 B. J. Mills, Integrating Functional
Use and Discard: Cooking Pottery from the Peruvian Analyses of Vessels and Sherds through Models of Cera-
Andes in Ethnoarchaeological Perspective. Latin Ame- mic Assemblage Formation. World Archaeology 21,
rican Antiquity 10, 1999b, 2546. 1989, 133147.
Hutson and Stanton 2007 S. R. Hutson, and T. Miloj~i} 1949 V. Miloj~i}, Chronologie der jnge-
W. Stanton, Cultural Logic and Practical Reason: the ren Steinzeit Mittel- und Sdsteuropas, Berlin, 1949.
Structure of Discard in Ancient Maya Houselots. Cam- Nelson 1981 B. A. Nelson, Ethnoarchaeology and
bridge Archaeological Journal 17, 2007, 123144. Paleodemography: ATest of Turner and Lofgrens Hypo-
Jovanovi} 1979a B. Jovanovi}, Indoevropljani i thesis. Journal of Anthropological Research 37, 1981,
eneolitski period Jugoslavije, in: Praistorija jugoslavenskih 107129.
zemalja, vol. 3, (ed. A. Benac), Sarajevo, 1979a, 397416. Nikolov 1989 V. Nikolov, Das frhneolithische
Jovanovi} 1979b B. Jovanovi}, Stepska kultura u Haus von SofiaSlatina: Eine Untersuchung zur vor-
eneolitskom periodu Jugoslavije, in: Praistorija jugo- geschichtlichen Bautechnik. Germania 67, 1989,
slavenskih zemalja, vol. 3, (ed. A. Benac), Sarajevo, 149.
1979b, 381395. Orton, C. 1980 C. Orton, Mathematics in Archa-
Jovanovi} and Gli{i} 1961 B. Jovanovi}, and J. eology, London, 1980.
Gli{i}, Eneolitsko naselje na Kormadinu kod Jakova. Orton, C. et al. 1993 C. Orton, P. Tyers, and A.
Starinar 11, 1961, 113142. Vince, Pottery in Archaeology, Cambridge, 1993.
Kaiser and Voytek 1983 T. Kaiser, and B. Voytek, Orton, D. 2008 D. Orton, Beyond Hunting and
Sedentism and Economic Change in the Balkan Neolithic. Herding: Humans, animals, and the political economy
Journal of Anthropological Archaeology 2, 1983, of the Vin~a period. PhD, University of Cambridge,
323353. 2008.
LaMotta and Schiffer 1999 V. M. LaMotta, and Orton, D. 2010 D. Orton, Both subject and object:
M. B. Schiffer, Formation Processes of House Floor herding, inalienability and sentient property in prehis-
Assemblages, in: The Archaeology of Household Activi- tory. World Archaeology 42, 2010, 188200.
ties, (ed. P. M. Allison), London, 1999. Pauketat 1989 T. R. Pauketat, Monitoring Missi-
Legge 1990 A. J. Legge, Animals, economy and ssippian Homestead Occupation Span and Economy
environment, in: Selevac: a Neolithic village in Yugo- Using Ceramic Refuse. American Antiquity 54, 1989,
slavia, (ed. R. Tringham and D. Krsti}), Los Angeles, 288310.
1990, 215242. Pauketat 1996 T. R. Pauketat, The Foundations
Lightfoot 1993 R. J. Lightfoot, Abandonment of Inequality within a Simulated Shan Community. Jour-
processes in prehistoric pueblos, in: The Abandonment nal of Anthropological Archaeology 15, 1996, 219236.

40
Marko POR^I], De facto refuse or structured deposition? (1943) STARINAR LXII/2012

Petrovi} 1992 J. Petrovi}, Arhitektura ku}e 4 na Sahlins 1972 M. D. Sahlins, Stone Age Economics,
Gomolavi. Rad Vojvo|anskih muzeja 34, 1992, 1932. Chicago, 1972.
Petrovi} 1993 J. Petrovi}, Keramika i alatke iz Scaffer 1993 G. D. Schaffer, An Archaeomagnetic
ku}e 4 na Gomolavi. Rad Vojvo|anskih muzeja 35, 1993. Study of a Wattle and Daub Building Collapse. Journal
Por~i} 2010 M. Por~i}, Arheologija vin~anskih of Field Archaeology 20, 1993, 5975.
ku}a: teorijsko-metodolo{ki osnovi prou~avanja demo- Schier 2006 W. Schier, Neolithic House Building
grafije i dru{tvene strukture. Ph.D. thesis, University of and Ritual in the Late Vin~a Tell Site of Uivar, Romania,
Belgrade, 2010. in: Homage to Milutin Gara{anin, (ed. N. Tasi} and C.
Por~i} 2011a M. Por~i}, An exercise in archaeo- Grozdanov), Belgrade, 2006, 325339.
logical demography: estimating the population size of Schiffer 1972 M. B. Schiffer, Archaeological
Late Neolithic settlements in the Central Balkans. Context and Systemic Context. American Antiquity 37,
Documenta Praehistorica 38, 2011a, 323332. 1972, 156165.
Por~i}, 2011b M. Por~i}, Obrasci bra~nog prebi- Schiffer 1976 M. B. Schiffer, Behavioral Arche-
vanja kasnoneolitskih zajednica vin~anske kulture Etno- ology, New York, 1976.
antropolo{ki problemi 6, 2011b, 497512. Schiffer 1985 M. B. Schiffer, Is There a Pompeii
Por~i} 2012 M. Por~i}, Effects of Residential Premise in Archaeology? Journal of Anthropological
Mobility on the Ratio of Average House Floor Area to Research 41, 1985, 1841.
Average Household Size: Implications for Demographic Schiffer 1987 M. B. Schiffer, Formation Processes
Reconstructions in Archaeology. Cross-Cultural Research of the Archaeological Record, Salt Lake City, 1987.
46, 2012, 7286. Schiffer 1995 M. B. Schiffer, Behavioral Archa-
Potter 2000 J. M. Potter, Pots, Parties, and Politics: eology: First Principles, Salt Lake City, 1995.
Communal Feasting in the American Southwest. Ame- Schiffer et al. 1981 M. B. Schiffer, T. E.
rican Antiquity 65, 2000, 471492. Downing, and M. McCarthy, Waste Not, Want Not: An
Price and Feinman 1995 D. T. Price, and G. M. Ethnoarchaeological Study of Reuse in Tucson, Arizona,
Feinman. (eds.). 1995. Foundations of Social Inequality. in: Modern Material Culture: The Archaeology of Us,
New York: Plenum Press. (ed. R. A. Gould and M. B. Schiffer), New York, 1981,
Price and Feinman 2010 T. D. Price, and G. M. 6786.
Feinman. (eds.), Pathways to Power: New Perspectives Schelach 2006 G. Shelach, Economic adaptation,
on the Emergence of Social Inequality. New York: community structure, and sharing strategies of house-
Springer, 2010. holds at early sedentary communities in northeast China.
Radivojevi} et al. M. Radivojevi}, T. Rehren, E. Journal of Anthropological Archaeology 25, 2006,
Pernicka, D. [ljivar, M. Brauns, D. Bori}, On the origins 318345.
of extractive metallurgy: new evidence from Europe. Smith 1987 M. E. Smith, Household Possessions
Journal of Archaeological Science 37, 2010, 27752787. and Wealth in Agrarian States: Implications for Archa-
Renfrew 1974 C. Renfrew, Space, Time and Polity, eology. Journal of Anthropological Archaeology 6,
in: The Evolution of Social Systems, (ed. M. Rowlands 1987, 297335.
and J. Friedman), London, 1974, 89114. Stefanovi} 2008 S. Stefanovi}, Late Neolithic
Rice 2005 P. M. Rice, Pottery Analysis: A Source- Boys at the Gomolava Cemetery (Serbia), in: Babies
book, Chicago, 2005. Reborn: Infant/Child Burials in Pre- and Protohistory,
Richards and Thomas 1984 C. Richards, and J. vol. BAR International Series1832, (ed. K. Bacvarov),
Thomas, Ritual activity and structured deposition in Oxford, 2008, 9599.
later Neolithic Wessex, in: Neolithic Studies, (ed. R. Stevanovi} 1997 M. Stevanovi}, The Age of
Bradley and J. Gardiner), Oxford, 1984, 189218. Clay: The Social Dynamics of House Destruction.
Risti}-Opa~i} 2005 J. Risti}-Opa~i}, Topografsko- Journal of Anthropological Archaeology 16, 1997,
hronolo{ke karakteristike naselja vin~anske kulture na 335395.
teritoriji Srbije. Glasnik Srpskog arheolo{kog dru{tva Stevanovi} and Tringham 1997 M. Stevanovi},
21, 2005, 71112. and R. Tringham, The significance of Neolithic houses
Russell 1994 N. Russell, Hunting, Herding and in the archaeological record of Southeast Europe, in:
Feasting: Human Use of Animals in Neolithic Southeast Uzdarje Dragoslavu Srejovi}u, (ed. M. Lazi}), Beograd,
Europe, 1993. 1997, 193209.

41
Marko POR^I], De facto refuse or structured deposition? (1943) STARINAR LXII/2012

Tasi} 1983 N. Tasi}, Jugoslovensko Podunavlje od Tripkovi}, B. 2009a B. Tripkovi}, Doma}instvo


indoevropske seobe do prodora Skita, Beograd, 1983. i zajednica u kasnom neolitu Zapadnog Balkana. Ph.D.
Todorovi} 1981 J. Todorovi}, A Recently Disco- thesis, University of Belgrade,2009a.
vered House in the Neolithic Settlement of Banjica in Tripkovi}, B. 2009b B. Tripkovi}, House(hold)
Belgrade. Archaeologia Iugoslavica 18, 1981, 1316. continuities in the Central Balkans, 53004600 BC.
Todorovi} 1961 J. Todorovi}, and A. Cermanovi}, Opuscula Archaeologica 33, 2009b, 728.
Banjica: naselje vin~anske kulture, Beograd, 1961. Tripkovi}, B. et al. 2011 B. Tripkovi}, A.
Tringham 1991 R. Tringham, Households with @ivanovi}, S. Stefanovi}, and V. Dimitrijevi}, Kontekst
Faces: the challenge of gender in prehistoric architec- a.c. 2.1 na lokalitetu Crkvine. Kolubara 5, 2011, 251259.
tural remains, in: Engendering Archaeology: Women Tschopik 1950 H. Tschopik, Jr., An Andean
and Prehistory, (ed. J. Gero and M. Conkey), Oxford, Ceramic Tradition in Historical Perspective. American
1991, 93131. Antiquity 15, 1950, 196218.
Tringham 1992 R. Tringham, Life after Selevac: van Zeist 2002 W. van Zeist, Plant husbandry of
why and how a Neolithic settlement is abandoned. Bal- tell Gomolava, Vojvodina, Yugoslavia. Palaeohistoria
canica 23, 1992, 133145. 43/44, 2002, 87115.
Tringham 1994 R. Tringham, Engendered Places in Varien and Mills 1997 M. D. Varien, and B.
Prehistory. Gender, Place and Culture 1, 1994, 169209. Mills, Accumulations research: Problems and prospects
Tringham 2005 R. Tringham, Weaving house life for estimating site occupation span. Journal of Archae-
and death into places: a blueprint for hypermedia narra- ological Method and Theory 4, 1997, 141191.
tive, in: (Un)settling the Neolithic (ed. D. Bailey, A. Varien and Potter 1997 M. D. Varien, and J. M.
Whittle, and V. Cummings), Oxford, 2005, 98111. Potter, Unpacking the Discard Equation: Simulating the
Tringham et al. 1992 R. Tringham, B. Brukner, Accumulation of Artefacts in the Archaeological
T. Kaiser, K. Borojevi}, L. Bukvi}, P. [teli, N. Russell, Record. American Antiquity 62, 1997, 194213.
M. Stevanovi}, and B. Voytek, Excavations at Opovo, Walker 2002 W. H. Walker, Stratigraphy and
19851987: Socioeconomic Change in the Balkan Neo- Practical Reason. American Anthropologist 104, 2002,
lithic. Journal of Field Archaeology 19, 1992, 351386. 159177.
Tringham et al. 1985 R. Tringham, B. Brukner, Wason 1994 P. K. Wason, The archaeology of
and B. Voytek, The Opovo Project: A Study of Socio- rank, Cambridge, 1994.
economic Change in the Balkan Neolithic. Journal of Wilk 1982 R. R. Wilk, Little House in the Jungle:
Field Archaeology 12, 1985, 425444. The Causes of Variation in House Size among Modern
Tringham and Krsti} 1990 R. Tringham, and D. Kekchi Maya. Journal of Anthropological Archaeology
Krsti} (eds.), Selevac: A Neolithic Village in Yugoslavia. 2, 1982, 99116.
Monumenta Archaeologica. Los Angeles: University of Wilk 1996 R. R. Wilk, Economies and Cultures:
California, 1990. Foundations of Economic Anthropology, Colorado,
Tripkovi}, A. 2011 A. Tripkovi}, Jame ispunjene 1996.
lepom na lokalitetu Crkvine. Kolubara 5, 2011, 8599. Wilshusen 1986 R. H. Wilshusen, The Relation-
Tripkovi}, B. 2007 B. Tripkovi}, Doma}instvo i ship between Abandonment Mode and Ritual Use in
prostor u kasnom neolitu: vin~ansko naselje na Banjici, Pueblo I Anasazi Protokivas. Journal of Field Archae-
Beograd, 2007. ology 13, 1986, 245254.

42
Marko POR^I], De facto refuse or structured deposition? (1943) STARINAR LXII/2012

Rezime: MARKO POR^I], Univerzitet u Beogradu,


Filozofski fakultet, Odeqewe za arheologiju, Beograd

DE FACTO OTPAD ILI STRUKTURISANA DEPOZICIJA?


KU]NI INVENTARI KASNONEOLITSKE VIN^ANSKE KULTURE

Kqu~ne re~i. neolit, vin~anska kultura, formacioni procesi, ku}ni inventari.

Ostaci ku}a i ku}nih inventara predstavqaju veoma zna~aj- zbirki iz kulturnog sloja kada se uzme u obzir prose~an
nu klasu arheolo{kih podataka za rekonstrukciju razli- upotrebni vek za svaku funkcionalnu klasu.
~itih aspekata praistorijskih dru{tava. Wihov zna~aj je Rezultati prve analize pokazuju da je veli~ina inventa-
posebno izra`en u arheologiji kasnog neolita na tlu Balka- ra vin~anskih ku}a u rasponu etnoarheolo{ki zabele`ene
na. U tom kontekstu, kqu~ne su dve povezane hipoteze koje varijacije. Rezultati druge analize ukazuju na to da postoji
su u velikoj meri uticale na istra`ivawa u ovoj oblasti: dobra korespondencija izme|u strukture ku}nih inventara
1) hipoteza Mirjane Stevanovi} i Rut Tringam da su ku}e i inventara iz kulturnog sloja, makar kada je re~ o boqe do-
spaqivane namerno, i 2) hipoteza Xona ^epmena da ku}ni kumentovanim lokalitetima poput Divostina i Opova.
inventari kasnoneolitskih ku}a ne predstavqaju inventare S obzirom na to da rezultati obe analize sugeri{u da u
koji su bili u svakodnevnoj upotrebi (sistemski inventari), odnosu na dva pomenuta uporedna okvira nema razloga da se
ve} namenski skupqene i deponovane zbirke prilikom ritu- sumwa u to da su vin~anski inventari mawe ili vi{e veran
alnog uni{tewa ku}e. Druga hipoteza zasniva se na ^epme- odraz kerami~kih zbirki koje su bile u svakodnevnoj upotre-
novoj oceni da je broj posuda koje su prona|ene u kasnoneo- bi, postavqa se pitawe kakvi su obrasci varijacije kvanti-
litskim ku}ama iznena|uju}e velik. teta i strukture kerami~kih inventara u odnosu na neke dru-
U ovom radu bi}e razmotrena ^epmenova hipoteza o ge atribute ku}nih ostataka, kao {to je, na primer, povr{ina
strukturisanoj depoziciji, tj. istra`iva~ko pitawe na ko- ku}e? Tako|e, postavqa se i pitawe interpretacije tih obra-
je ovaj rad odgovara jeste: da li imamo razloga da verujemo da zaca u antropolo{kim terminima. Ustanovqeno je da na
inventari vin~anskih ku}a ne odra`avaju svakodnevne in- lokalitetu Divostin postoji pozitivna korelacija izme|u
ventare? Osnovna ideja je da se uspostave uporedni okviri povr{ine ku}e i veli~ine kerami~kog inventara. Ovakav
u odnosu na koje }e biti proceweno da li su vin~anski ke- obrazac mo`e se interpretirati kao posledica razlika koje
rami~ki inventari neobi~ni, tj. da li imamo razloga da postoje izme|u ku}a u veli~ini doma}instva i u wihovom
smatramo da se ne radi o zbirci posuda iz svakodnevne upo- dru{tvenom statusu. Ostaje nejasno da li su uo~ene razli-
trebe. Prvi uporedni okvir jeste veli~ina inventara, tj. ke efemerne prirode u jednom prete`no egalitarnom dru-
ukupan broj posuda u ku}i. {tvu ili ukazuju na po~etak procesa u~vr{}ivawa nejedna-
Da bi se odgovorilo na istra`iva~ko pitawe, raspon kosti i pove}awa kompleksnosti vin~anskih dru{tava.
veli~ina inventara vin~anskih ku}a upore|en je sa raspo- Op{ti zakqu~ak ove studije jeste to da, sa stanovi{ta
nom inventara etnoarheolo{ki zabele`enih zbirki iz ovde kori{}enih uporednih okvira, nema razloga da se za-
razli~itih kultura. Drugi uporedni okvir jeste struktura kqu~i da su vin~anski kerami~ki inventari iz ku}a neo-
inventara, u smislu proporcionalne zastupqenosti funk- bi~ni, tj. da ne odra`avaju strukturu sistemskih inventara.
cionalnih klasa. Postavqa se pitawe da li struktura ke- [tavi{e, obrascima varijacije ku}nih inventara mo`e se
rami~kih zbirki iz ku}a odgovara strukturi kerami~kih dati specifi~na antropolo{ka interpretacija.

43
ADAM N. CRNOBRNJA
Belgrade City Museum, Belgrade

INVESTIGATIONS OF LATE VIN^A HOUSE 1/2010


AT CRKVINE IN STUBLINE

UDK: 903.3"634"(497.11) ; 902.2(497.11)"2010" e-mail: ancrnobrnja@gmail.com


DOI: 10.2298/STA1262045C Received: February 16, 2012
Short communication Accepted: June 21, 2012

Abstract. The Crkvine site is situated around 40 km southwest of Belgrade (Serbia) in the vicinity of the village of Stubline,
in the borough of Obrenovac. Extensive geophysical investigations were carried out during the 2010 campaign and, based
on the results, we started investigations of the Late Vin~a house 01/2010. The following comprehensive report details
the method of construction and organisation of life in that house, which dates from the Vin~a culture phase D. The house was
very well preserved and we paid special attention to two large ovens inside the house as well as to some interesting portable finds
(a clay table, a clay millstone structure and three large clay heads).

Key words. Neolithic, Vin~a culture, house, geophysical investigations, bucranium, millstone, Stubline, Obrenovac.

T
he Crkvine site is situated in the vicinity of the gests just one ditch. In the central settlement area is an
village of Stubline, in the borough of Obreno- anomaly indicating a ditch from some earlier settle-
vac around 40 km southwest of Belgrade (Fig. ment phase overlaid by rows of houses from the last
1). A settlement of the Late Vin~a phase, covering an habitation period. It could be assumed, by comparing
area of around 18 ha, it is located on a gentle elevation the intensity of geomagnetic anomalies (whose implica-
oriented in a northwest-southeast direction, bordered tions were also checked by excavations in four instances)
in the north and south by brooks meeting below its and their dimensions, that there are the remains of over
south-eastern end. Systematic archaeological investi- 200 houses within the investigated area (Fig. 2). The
gations of the Late Vin~a settlement at Crkvine in the results obtained by geomagnetic mapping made it pos-
village of Stubline have continued since 2006 and sible to perceive, for the first time, an almost complete
investigations have hitherto yielded much data about matrix of one large open Late Vin~a settlement,3 which
the organisation and way of life in that period.1 was surrounded by ditches and densely packed houses
Geophysical investigations provided exceptional
results and the opportunity to conduct further, well-
planned investigations. Geomagnetic mapping2 carried
out between 2007 and 2011 covered an area of 83,000 1 For results of previous investigations see Todorovi} 1967;

square meters in total. As a result of this, the northern Simi}, Crnobrwa 2008; Crnobrwa 2009, Crnobrnja, Simi}, Jan-
and southern borders of the settlement were estab- kovi} 2010; Crnobrnja 2011; Antonovi}, [ari} 2011.
2 Geomagnetic mapping was carried out by Vladimir and
lished. On the northern side, where the terrain slopes
Jelena Mileti} from the Center for New Technologies Viminacium.
more gently, the border of the settlement is identified Magnetometer-gradiometer GSM 19 gw of Canadian manufacture
by an anomaly, which indicates a double ditch, while has been used.
on the steeper southern side a registered anomaly sug- 3 Crnobrnja, in press.

45
Adam N. CRNOBRNJA, Investigations of Late Vin~a House 1/2010 at Crkvine in Stubline (4564) STARINAR LXII/2012

Fig. 1. Sites mentioned in the text:


1) CrkvineStubline; 2) Gomolava;
3) KormadinJakovo; 4) Opovo; 5) Banjica;
6) Vin~a; 7) Belovode; 8) Divostin; 9) Grivac;
10) CrkvineMali Borak; 11) Uivar; 12) Parta
Sl. 1. Lokaliteti koji se pomiwu u tekstu:
1) CrkvineStubline; 2) Gomolava;
3) KormadinJakovo; 4) Opovo; 5) Bawica;
6) Vin~a; 7) Belovode; 8) Divostin;
9) Grivac; 10) CrkvineMali Borak;
11) Ujvar; 12) Parca

in a well planned, almost proto-urban arrangement. In of similar situations in other sections of the settlement.
the period from 2009 to 2011, geoelectric scanning of The location profile obtained by geoelectric scanning
the profiles was also conducted.4 So far a total a of suggested that the structural remains were at a depth of
1125 m of profiles have been scanned and they have around 0.50 to 1.0 meter and appeared to be just one
provided information about the vertical preservation structure without more complex vertical stratigraphy.
of structures previously identified by geomagnetic We calculated, on the basis of the above mentioned
mapping, as well as basic data about the thickness of data, that taking into consideration time and financial
cultural layers in different sections of the settlement. resources, we would be able to completely investigate
Systematic archaeological investigations in 2010 this structure in one campaign.
were carried out between the 1st of September and the The geomagnetic anomaly in that location indicat-
7th of November.5 One of main objectives of these in- ed that it was a structure preserved to a different degree
vestigations was the testing of results of the geophysical to that of the northern and southern sections. The geo-
measurements of the 2010 campaign. The location of the electric scan profile along the longitudinal axis of the
dig was determined by previous geophysical investiga- previously mentioned geomagnetic anomaly revealed
tions (geomagnetic mapping and geoelectric scanning). that the depth of the lower structure level is uniform
We decided on the chosen location for several rea-
sons. As we investigated the house on the settlement
periphery in 2008, this year we decided to investigate 4 Geoelectric scanning of profiles was carried out by Momir
one of the structures located in the marginal zone of Vukadinovi}. Geophysical resistivity & self potential meter RPM12
the central sections of the settlement. The anomaly IP has been used.
zone identified at that location by geomagnetic map- 5 Director of investigations on behalf of Belgrade City Museum

ping indicated a well-preserved structure. However, was Adam Crnobrnja and members of professional team were
archaeologists Milo{ Spasi}, Marko Jankovi} and Velimir Pilipovi},
within that anomaly, two zones of diverse intensity were
and students of archaeology Marko Marjanovi}, Jovana Tripkovi},
recognised, so this also offered the possibility of obtain- Marko Andri}, Boris Pavlovi}, \or|e Lazi}, Vuk Kold`i} and Ma-
ing a reliable key for the reading and interpretation rija Cerovi}.

46
Adam N. CRNOBRNJA, Investigations of Late Vin~a House 1/2010 at Crkvine in Stubline (4564) STARINAR LXII/2012

and reaches a depth of 11.1 m. On the basis of this data


we determined the position and dimensions of the trench
(18 x 8 m) and estimated that the amount of excavation
needed was within our financial resources.
The trench was excavated within a square grid (2 x
2 m square) but, when more complex structures were
encountered, we identified and investigated some dis-
tinct archaeological entities (hereafter referred to as AE).
While the first arbitrary layer (of 0.20 m relative
depth) contained only around fifty pottery fragments,
in the second arbitrary layer we encountered small
scattered lumps of daub and more fragments of pottery
of Late Neolithic and Eneolithic provenance. In the
northern section of the trench, at a relative depth of
0.250.30 m, we encountered a rather large area of Fig. 2. Geomagnetic readings of settlement at Crkvine
loose daub and the first groups of pottery on top of the in Stubline. Red color denotes position of house 1/2010
remains of house 1/2010, which obviously dated from and blue denotes house 1/2008
the time after the house was destroyed. Under that layer
Sl. 2. Geomagnetski snimak naseqa na Crkvinama
was 0.200.30 m of soil with rare lenses of small daub
u Stublinama. Crvenom bojom obele`ena je pozicija
lumps up to 3 cm in size and after that we encountered ku}e 1/2010 a plavom 1/2008.
the first remains of the collapsed house 1/2010.
The houses ground plan is almost completely pre-
served and its assumed dimensions are 13.10 x 5.10 m
so the floor area inside the house was around 67 square The floor in the western half of the southern section
meters. The house is oriented in a northeast-southwest of the house was preserved only in a few places, re-
direction with a deviation of 23 from the north (Fig. 3). sembling small islands in certain places. In that zone
The northern section of the house is exceptionally well were also a few shallow dug holes filled with pieces of
preserved with a completely preserved thick floor and daub. The southernmost part of house 1/2010, i.e. the
a large number of portable finds as well as some perma- assumed position of its southern wall, has been deter-
nent structures (Fig. 4). The floor level is at a relative mined according to the group of pottery (PG) 22, which
depth of 0.70 to 0.80 m, i.e. at 109.90109.76 m above was discovered on an isolated, preserved section of the
sea level in the northern section and 109.71109.54 m floor, while no traces of the floor in situ have been
in the southern section of the house. recorded to the south to date.
An approximately 1 meter wide ditch, which seems For greater accuracy we will present categorised
to have been dug in more recent times, is at a distance descriptions of the structure and the method of con-
of 5.20 m from the northern edge of the house. In this struction of house 1/2010, as well as the finds discove-
ditch, which not only completely destroyed the floor in red inside.
this section but also greatly damaged oven 2, the
portable finds were rare.
The southern section of house 1/2010 was damaged ARCHITECTURE OF THE HOUSE 1/2010
to a great extent, particularly its western half. In the
eastern half of the southern section of the house, the Walls of the house 1/2010 (Fig. 3/1)
floor was preserved to a considerable extent, particu- Despite the generally well-preserved interior of the
larly along the recently dug ditch. By following traces house, the walls that remain are very poorly preserved,
of ash extending along the eastern house edge in the quite the opposite of the situation recorded in house
northern and southern sections, it could clearly be seen 1/2008.6 It has been noticed that in a few places on the
that it was one structure. The assumed direction of the
eastern wall in the southern section of the house is
negated by Pit 1, indicating that it dates from a time
after the destruction of house 1/2010. 6 Crnobrnja, Simi}, Jankovi} 2010, 14.

47
Adam N. CRNOBRNJA, Investigations of Late Vin~a House 1/2010 at Crkvine in Stubline (4564) STARINAR LXII/2012

0 5m

Fig. 3. Plan of house 1/2010, left house plan, right orthogonal projection: 1) wall; 2) postholes; 3) floor;
4) oven 1; 5) oven 2; 6) storage container; 7) small clay table; 8) clay millstone structure; 9) stone working
surfaces (palettes-millstones); 10) pottery group on house floor; 11) group of loom weights; 12a) east head;
12b) west head; 12c) south head; 13) pit 1; 14) pit 2; 15) shallow dug holles filled with pieces of daub
Sl. 3. Osnova ku}e 1/2010, levo plan ku}e, desno snimak iz ortogonalne projekcije: 1) zid; 2) rupe za stubove;
3) podnica; 4) pe} 1; 5) pe} 2; 6) kaseta za skladi{tewe; 7) glineni sto~i}; 8) glinena konstrukcija `rvwa;
9) kamene radne povr{ine (palete-`rvwevi); 10) grupe keramike na podu ku}e; 11) grupa tegova za razboj;
12a) isto~na glava; 12b) zapadna glava; 12c) ju`na glava; 13) jama 1; 14) jama 2;
15) plitko ukopane jame ispuwene komadima lepa

eastern and western edges of the floor, its ends are western edge of the house. Therefore, we could only
slightly turned upwards thus indicating that the floor imagine the structure of the walls, mainly on the basis
coating was executed in such a way as to curve gently of their segments, which collapsed on the floor and
toward the walls. Despite meticulous exploration of the over the finds.
surface next to the preserved outlines of house 1/2010, it We could claim, with a degree of certainty, that only
was possible to distinguish traces of only two postholes a few wall fragments sealing off the house contents in
along the assumed line of the walls. These postholes, its northern section were discovered in situ. It is inter-
0.30 and 0.40 m in diameter, were recorded next to the esting that all these wall fragments had impressions of

48
Adam N. CRNOBRNJA, Investigations of Late Vin~a House 1/2010 at Crkvine in Stubline (4564) STARINAR LXII/2012

Fig. 4. House 1/2020 from the north Fig. 5. Plank impression in wall daub
Sl. 4. Pogled na ku}u 1/2010 sa severa Sl. 5. Otisak talpe u zidnom lepu

large planks and not of wattle. The thickness of these bigger than the load bearing capacity of houses with
daub fragments is 0.100.15 m and they were all facing walls built from the wattle technique.
upward while the plank impressions were facing the This is also confirmed by traces of a few massive
floor. The plank impressions are around 0.03 m deep and supporting posts inside the house (Fig. 3/2), whose
this is their smallest assumed thickness as daub covering postholes were recorded in the floor along the central
one side of the plank timber did not reach the complete longitudinal axis of the house. We recorded three such
thickness of the timber within the wall structure. The postholes in the floor, two of them around 0.25 m in
greatest recorded thickness of one of the planks was diameter and one of a triangular shape next to the south-
0.43 m (Fig. 5). western corner of oven 1 with 0.40 m long sides (Fig.
In just one place on the western side of the house 6/1).8 We should not rule out the possibility that there
there was a recorded internal wall coating of around 30 cm might have been a few more such supports for the roof
long, still standing upright, so this means it was in situ. structure as, along the line of their discovery, the southern
On the outside there were horizontal impressions iden- section of the house was destroyed to a large extent.
tical to the plank impressions on the collapsed daub The substantial load bearing capacity of the entire
pieces. structure, which made possible the construction of a
Although we recorded a small quantity of daub from massive roof structure, is also indirectly confirmed by
the walls and, taking into account the above mentioned distinct traces of crumbling in the northern section of
wall segment with identical plank impressions preserved the house. As such, a considerable area in this part of
in situ and the absence of daub fragments with wattle the house was covered with a layer which burnt at a
impressions, it could be concluded with considerable high temperature and completely covered the floor and
reliability that the walls had been built of massive planks all items on it. It was a 0.100.15 m thick layer of yellow
covered with daub and that it is not a standard method of colour with a granular structure (AE17) and was so
construction of the Vin~a houses.7 The stated dimensions
of the planks used for the wall construction indicate
that it was a very massive and heavy structure not pre-
viously recorded in Vin~a culture settlements. 7 I must mention that such plank impressions in the daub dis-

covered at Para (Lazarovici 2006, 5 fig. 24 ) have been explained


Roof and roof structure as the floor of a storey structure. However, considering the small
preserved segment of wall of our house in situ with identical plank
Bearing in mind the above mentioned massive impressions I am more inclined towards the already stated conclu-
structure of the walls of house 1/2010, it could also be sion that, in our case, these were segments of the wall.
assumed that their load bearing capacity was much 8 See more about that under Oven 2 and Heads i.e. west head.

49
Adam N. CRNOBRNJA, Investigations of Late Vin~a House 1/2010 at Crkvine in Stubline (4564) STARINAR LXII/2012

compact that it had to be carefully excavated with chis- recorded by total station. This procedure made possi-
els. This layer was completely adhered to the contents ble the reconstruction of their direction. At one spot,
of the house and to the floor underneath. Of particular 0.10 m in length, we discovered an impression of one
interest are the remains of small, carbonised posts with- of the posts in the burned soil that had covered them.
in that layer, spreading horizontally through it. The When the positions of the previously mentioned posts
remains of these small posts were recognisable on the are combined with information about the exact posi-
basis of elongated ellipsoid or rectangular traces in the tions of some vessels in PG17 and above AE28, it
profiles of the aforementioned yellow layer. They were indicates that the posts fell onto the items on the house
clearly discernible above PG13 in square 14 and above floor. We, therefore, came to the conclusion that a
the storage receptacle (AE28) in square 6 and extended storey or attic structure might have been built above
in a north-south direction (fig. 7). We tried to follow the northern section of the house. It is also worth men-
their traces during excavations but with little success tioning that there were small holes, 0.070.08 m in
as they were hardly discernible when we tried to leave diameter, in the floor, spaced at a distance of 0.35 m
them untouched in the highly burnt yellow layer. Instead, (Fig. 3/2 and Fig. 6/2). The implication of these holes
we followed them by digging the layer, containing them, is not clear, but it is interesting that identical holes in the
to the floor level. Here, traces of posts were recorded floor of The House of the Deer in Para (Romania)
in the profile and their positions were successively were explained by researchers as holes for the sides of

Fig. 6. Central zone of north section of house 1/2010:


1) triangular posthole; 2) small postholes; 3) house floor; 4) southwestern corner of oven 1; 5) post impressions
on SW corner of oven 1; 6) clay millstone structure; 7) stone working surfaces (palettes-millstones)
Sl. 6. Centralna zona severnog dela ku}e 1/2010:
1) trougaoni otvor za stub; 2) male rupe za stubove; 3) pod ku}e; 4) jugozapadni ugao pe}i 1; 5) otisci greda
na JZ uglu pe}i 1; 6) glinena konstrukcija `rvwa; 7) kamene radne povr{ine (palete-`rvwevi)

50
Adam N. CRNOBRNJA, Investigations of Late Vin~a House 1/2010 at Crkvine in Stubline (4564) STARINAR LXII/2012

Fig. 7. Details of small posts above PG13 and stone working surfaces (top) and above storage container (bottom)
Sl. 7. Detaqi gredica iznad GK13 i kamenih radnih povr{ina (gore) i iznad kasete za skladi{tewe (dole)

a ladder leading to the upper floor.9 The existence of an


attic or storey structure should not be surprising as simi-
lar structures have been assumed in the Vin~a houses at
Uivar,10 Para11 and Opovo,12 as well as at Crkvine in
Stubline.13

House floor (Fig. 3/3)


The floor is preserved in many fragments over
almost half of the area of house 1/2010. It is preserved
almost completely in the northern section of the house
(Fig. 6/3) while it is destroyed for the most part in the Fig. 8. Detail of cross-section of floor
southern section. The floor surface is a brown to orange in north house section
colour and is very compact as a consequence of expo-
Sl. 8. Detaq preseka podnice u severnom delu ku}e
sure to high temperatures. A recent ditch, which split
house 1/2010 into two sections, gave us the opportunity
to examine the floor structure without further damage.
The floor thickness varies between 0.20 and 0.30 m and yellow clay base. On the back sides of some of the re-
it is evenly burned throughout. Investigations carried claimed wall fragments were impressions of wattle,
out in the profile of the recently dug ditch provided boards and planks. The fact that these impressions are
interesting data about the method of floor construction different on adjacent daub pieces and that their orienta-
used in house 1/2010 (Fig. 8): tions do not correspond14, also supports the assumption
At the lowest level is a lense of black soot as a result regarding the use of wall fragments reclaimed from
of scorching the surface where the house was to be built. some previously burned structures.
This was done, most probably, in such a way that flam-
mable material was piled over the entire area intended
for house construction. This conclusion was reached
based on the fact that under the black lense of densely 9 Lazarovici and Lazarovici 2006, 7, fig. 39a, 43c.
packed soot was reddish soil as a consequence of high 10 Schier 2006, 326, 333, fig. 2.
temperatures resulting from a fire burning on top of it. 11 Lazarovici on line

After scorching, a layer of 0.040.05 m thick yellow 12 Tringham 1992, 361.


13 Crnobrnja, Simi}, Jankovi} 2010, 20.
clay was laid over the area and stamped down.
14 Impressions of branches, sticks and planks have been
Laid on the yellow clay were fragments of daub of
recorded on the underside of the secondary used daub pieces for the
various sizes, 0.20 to 0.30 m thick, reclaimed from a
construction of the floor of house 1/2010 where the daub pieces
previously burnt house. The faces of the old walls were were laid next to each other. The thickness of these daub pieces was
turned upward and the back sides were inlaid into the not uniform.

51
Adam N. CRNOBRNJA, Investigations of Late Vin~a House 1/2010 at Crkvine in Stubline (4564) STARINAR LXII/2012

Finally, the top surface of the daub was covered with Oven 1 (Fig. 3/4)
a 0.020.03 m thick clay coating and, in certain places, A large domed oven, oval in shape (maximum length
two layers of coating have been preserved. 2.38 m and maximum width 1.80 m) covering an area
This method of house floor construction provides of 3.60 square meters was found in the north-eastern
evidence about the rational and economical attitude of corner of house 1/2010. Its northern end was leaning
the inhabitants of the settlement at Crkvine. By avoiding on the northern wall of the house, while the eastern wall
the use of wood as the floor substructure, which was a of oven 1 was 0.65 m from the eastern wall of the house.
common building technique recorded at many Vin~a The oven is preserved up to its last floor coating, so
culture sites, a considerable saving of resources and time traces of the calotte base can be seen on the surface of
was achieved. The preparation of the ground by scorch- the preserved remains of oven 1 (Fig. 9). The ends of the
ing and using already well thermally treated wall frag- calotte, which are slightly arched, terminate at around
ments from an earlier demolished house, resulted, in 0.55 m from the southern end of the complete oven
the long run, to a considerable saving of the energy layout which means that it did not cover the entire sur-
necessary for heating. It is already known that great face of the last floor coating. Therefore, the size of the
attention was paid, in the Vin~a communities, to the firebox was around 2 square meters. In front of the
thermal efficiency of construction.15 firebox opening was a semicircular section of floor
It is necessary to mention that at three locations corresponding to the floor inside the oven, i.e. both
within the preserved house sections, there was no floor surfaces were covered with the same coatings and
built in the above mentioned way: were at an identical height. The method of oven con-
1) Oven 2 was built directly onto stamped soil struction could be seen in its rear, northern section,
2) Storage container (AE28) in the north-western which was considerably damaged.16 The base on which
corner of the house has foundations made, partially, of the oven was built was made of well fired and smoothed
broken pottery covered by thin clay coatings; clay, which resembles, in its quality, the floor. This oven
3) Bases of large pithoi next to the western wall of base is, at least in the northern (only visible) section,
the house (PG17) were dug into the ground below the also of an oval shape and 0.050.07 m thick. Its edge
floor level. is slightly turned upward so it resembles a shallow clay

Fig. 9. Oven 1, orthogonal projection,


from the south
Fig. 10. Layers of construction of oven 1,
from the north
Sl. 9. Pe} 1, snimak iz ortogonalne projekcije,
snimak sa juga
Sl. 10. Nivoi izgradwe pe}i 1,
snimak sa severa

52
Adam N. CRNOBRNJA, Investigations of Late Vin~a House 1/2010 at Crkvine in Stubline (4564) STARINAR LXII/2012

Fig. 11. Oven 2,


from the east
Sl. 11. Pe} 2,
snimak sa istoka

trough (or pan) on top of which the entire oven was western corner of the oven that an approximately 0.15 m
then built. Three composite insulation layers, made in thick covering layer was subsequently added over the
an identical manner and consisting of three layers mass of the original wall. A difference in the final exe-
each, were successively laid over that base (Fig. 10): cution of the surfaces is also evident.
1) a layer of stamped earth, 0.030.04 m thick Oven 1 is, according to its size, the largest oven re-
2) the previous layer was coated with clay (0.015 to corded at any Late Vin~a site in Serbia.19 Its similarity
0.02 m thick) and then burned; with the largest oven discovered in house 13 at Divo-
3) on top of the burnt clay surface were laid pottery stin is very apparent. It is 1.95 x 1.70 m in size and had
fragments.17 seven layers of construction/renovation. Pottery frag-
On top of the third, and final, insulation layer was ments and other secondary materials were used in its
another layer of stamped earth covered with a clay coat- construction.20 An oven of a slightly smaller size, with
ing. This was the first functional oven floor and was good foundations (functioning as insulation) and three
renovated twice, but only by adding a 0.012 0.02 m subsequent renovations, has also been found in house
thick layer of clay. The wall of this oven is massive, 12 at Grivac.21
built in one piece and rising 0.350.50 m above the
house floor level. At the top of the oven wall, i.e. along
its middle, there is a continuous black lense representing
the burned remains of densely packed branches/sticks,
15 Miloradovi}, Tasi} 2008.
which formed the calotte structure. On the western, 16 The oven has not been explored in detail. The quality of its
best preserved, section of the oven wall there are many discovered remains was due to its good state of preservation, con-
traces of finger impressions over almost the entire sur- served, as it was, together with the whole house and then covered
face, while the eastern section of the oven collapsed and with earth in order that, in the future, the entire structure could be
underneath was found a group of eight loom weights presented in situ.
17 Zones with burnt clay look like the oven floor, so it could
(more details further in this paper). Next to the south-
also look like layers of many floor renovations. It could be con-
western end of the oven was a triangular opening for cluded that these are basic structural elements, i.e. insulation layers
the post/plank in the floor18 leaning against the oven on the basis of the position and appearance of the final coatings of
wall (Figs. 6/1 and 6/4). Impressions of the posts on the the oven floor. They were at the same level as the top edge of the
monolithic oven wall from which its calotte started.
wall of oven 1 seem to indicate that the area between 18 There were probably split timbers inserted; see also the sec-
the posts and the oven was subsequently filled with clay tion of this paper regarding the roof and roof structure.
after the oven had been built (Fig. 6/5). The frontal 19 Cf., Pecikoza 2009, 29;

section of the oven is particularly massive due to reno- 20 Bogdanovi} 1988, 55, fig. 5.12, Plan VIa/I.

vation or enlargement. It is clearly visible on the south- 21 Bogdanovi} 2004, 160, 174, sl. 8.10.

53
Adam N. CRNOBRNJA, Investigations of Late Vin~a House 1/2010 at Crkvine in Stubline (4564) STARINAR LXII/2012

Oven 2 (Fig. 3/5) mal structures and it was bordered by a small, low wall
The remains of yet another oven, which was mostly preserved up to a height of 0.12 m. The floor had a par-
destroyed by later digs, have been found in the central tial substructure of broken pottery and three layers (each
area of house 1/2010 and next to its western wall. The around 0.02 m thick) of coating consisting of packed
preserved elements of this oven included only its west- and smoothed clay, so we could conclude, on the basis
ern wall, a narrow strip of oven floor on the inside and of their construction, that these were floor renovations.
another narrow strip of floor in the central oven section An oval recess/depression, which was most probably the
(Fig. 11). The preserved back wall of oven 2 is arched in result of some heavy object/vessel falling from a con-
shape, it is 1.70 m wide and the thickness varies from siderable height, was encountered in the north-eastern
0.32 m at the base of the back to 0.17 m near the top. section of the container. The features similar to our
The preserved height is 0.36 m above the house floor. container have also been encountered in Vin~a houses
On the outside were the bloated remains of the last at other sites (house 2/79 at Banjica,23, house 01/06 at
coating on the lower half of the rear section of the wall. Vin~a,24 and in many houses at Divostin25) and are
The floor here was made much more simply than in usually explained as storage places for food, vessels
oven 1: it consists of three coatings, the lowest was laid and other objects. In the container were found frag-
directly over the stamped earth without any foundations, ments of a few rather large vessels (PG16 and 16a) and
while the other two coatings were successively laid, one also one complete figurine (Fig. 14) standing in situ on
on top of the other. The total thickness of all three coat- the floor, facing east. The question remains whether
ings is 0.07 m and it was the result of a standard reno- the assumed storage purpose of this space was also its
vation of the oven floor. It is interesting that there is original purpose or if it was originally used as a thermal
disproportion in the construction of the oven wall, structure as has been mentioned in literature regarding
which is very massive, and its floor, which is exceptio- similar features.26
nally thin and at the same level as the house floor.
The impression is, considering the situation in which
it was discovered, that the construction of this oven PORTABLE INVENTORY
preceded the building of the house by a very short peri- ON THE HOUSE FLOOR
od of time. Supporting this assumption is the fact that
the oven floor was laid directly on the ground at the Many archaeological objects have been found in
level of the surrounding house floor and that traces of addition to the permanent features discovered inside
ground preparations, like those observed under the house 1/2010. Besides pottery and sporadic finds of
house floor, were not encountered under the oven. On stone working surfaces and tools, some larger objects
the other hand, it is evident that oven 2 was inside were also found inside the house including a clay mill-
house 1/2010 during its use as the eastern section of stone structure, a small clay table and three large clay
the oven wall was in contact with a segment of the pre- heads. The clay heads (two of which could have been
served house floor and the height of the preserved bucrania) could only conditionally be identified as finds
oven wall suggests that the oven was certainly visible from the house floor as their original position was on
at the time of the use of the house. The possible pur- the pillars or the walls.
pose of oven 2 inside house 1/2010 is questionable. It
could have been used for its original purpose (as a ther- Small clay table (Fig. 3/7)
mal structure) or it could also have had a secondary use A square-shaped, small clay table has been discov-
as a storage space, as has been recorded at some other ered on the inside of the assumed line of the eastern wall
Late Vin~a sites22. We also have similar situation in the of the house (not preserved in that section) and posi-
north-western corner of our house (storage container
AE28).

Storage container (Fig. 3/6) 22 Cf., Bogdanovi} 1988.


In the north-western corner of the house was a 23 Todorovi} 1981, 14/D, B; Tripkovi} 2007, 8990.
square feature made of daub a storage container with 24 Tasi} i dr. 2007, 212213, T. I.
a maximum size of 1.40 x 1.20 m (1.2 square meters). 25 Bogdanovi} 1988.
The floor of this container resembles the floors of ther- 26 Bogdanovi} 1988, 67.

54
Adam N. CRNOBRNJA, Investigations of Late Vin~a House 1/2010 at Crkvine in Stubline (4564) STARINAR LXII/2012

0 50 cm

Fig. 12. Small clay table, from the south


Fig. 13. Clay millstone structure during conservation process
Sl. 12. Glineni sto~i}, snimak sa juga
Sl. 13. Glinena konstrukcija `rvwa, u toku konzervatorskog postupka

tioned at a right angle to the wall. The table was found at conservation in the field, consolidation, lifting and
the spot where it had obviously been used immediately removal for further conservation, we came to the fol-
before the destruction of the house, based on the posi- lowing conclusions:
tion of its five feet relative to the remains of the table a shell-shaped receptacle was executed using a
top (fig. 12) and considering the fragmented biconical coil-building technique (dimensions 0.58 x 0.58 x
bowl found on the table top. The table top is only par- 0.22 m, wall thickness 0.050.06 m) and was flat on
tially preserved but, as the fragments were found in the underside, so it could be concluded that this struc-
situ, and, according to their position and the position of ture was mobile;
the feet, it could be concluded that it was of a rectan- an oval-shaped bedding for a stone working sur-
gular shape, 0.80 x 0.62 m in size.27 All five feet (one face was made of clay (0.20 x 0.14 m) and rises 0.10 m
at each corner and one in the centre) are of a triangular above the receptacle interior;
shape with a rounded base. The dimensions of the table the stone working surface (palette) fell out of its
feet are as follows: the width 0.150.20 m, the thick- bedding, due to the collapse of the entire structure, and
ness 0.050.07 m and the height 0.23 m. Small clay was found around 0.80 m to the west, inside the house;
tables have been found in many houses at Divostin but at the moment of the collapse of the millstone
they differ conspicuously from our specimen. Generally, structure, the house and its interior were burning so in-
these objects at Divostin have had an oval table top, tensely that one of the floor coatings completely stuck
they have been of a smaller size and their height has not to the opening of the stone bedding.
exceed 0.12 m.28 It could be assumed, but with great
reservations, that the damaged surface of the decorated
house daub found at Kormadin near Jakovo29 might
also have been the top of some table. 27 After conservation treatment in the field carried out by
archaeologists-conservators Branislava Lazarevi} (Central Institute
Clay millstone structure (Fig. 3/8) for Conservation) and Stevan Djuri~i} (Archaeological Collection of
the Faculty of Philosophy, University in Belgrade), the remains of
A clay millstone structure has been found in the
the small table were taken to the Belgrade City Museum for further
northern half of the house in the area between oven 1 conservation and restoration treatment.
and PG17 (Fig. 6/6). The structure was lying upside 28 Bogdanovi} 1988.

down so its opening was facing the floor (Fig. 13). After 29 Jovanovi}, Gli{i} 1961, 132, sl. 33, 38, 39, 40.

55
Adam N. CRNOBRNJA, Investigations of Late Vin~a House 1/2010 at Crkvine in Stubline (4564) STARINAR LXII/2012

Fig. 14. Objects of cult purpose: small bowl with eight protomes (left),
figurine found in storage container (center) and three-legged vessel (right)
Sl. 14. Predmeti kultne namene: zdelica sa osam protoma (levo),
figurina prona|ena u kaseti za skladi{tewe (u sredini) i posuda na tri noge (desno)

In recent years a few clay millstone structures have Millstones/palettes (Fig. 3/9)
been discovered in Serbia. Three mobile structures30 were In addition to the clay millstone structure and the
discovered at Vin~a31 while identical fixed structures associated stone working surfaces, five more stone
were discovered in house 1/2008 at Crkvine in Stubli- working surfaces palettes usually identified as mill-
ne,32 in house 2/79 at Banjica33 and in many houses at stones have been found in house 1/2010. They were
Divostin.34 Taking into account previous incorrect all found in the immediate vicinity of the previously
interpretations of such a structure at Banjica35 and the mentioned clay structure, i.e. near large pithoi used for
poor state of preservation of the specimens from storage. Four palettes were grouped in one location
Divostin, the question could be asked, in how many (Fig. 6/7), while fragments of two pithoi were found
instances similar structures have not been recognised
as beddings for millstones? The fact is that most Late
Vin~a houses, at all sites, have stone working surfaces
for grinding and chopping, so the appearance of more 30 Their mobility should be understood conditionally because,
complex clay structures, as beddings for working sur- despite the possibility of moving these structures within the house
faces, could prompt the question of why they appear in that they were found, it most probably had not been done often con-
certain houses. Do they indicate a distinct specialisa- sidering their weight, dimensions and fragility.
31 Tasi} i dr. 2007
tion of the inhabitants of some structures, their social 32 Crnobrnja, Simi}, Jankovi} 2010, 17, fig. 14.
status or some special use for the millstones? If grind- 33 Todorovi} 1981, 14/H, 15.
ing took place in all houses (judging by the discovery 34 Bogdanovi} 1988.
of stone working surfaces) and there were beddings for 35 The structure from house 2/79 at Banjica has so far been

stones with receptacles, which reduce the spillage of explained as a structure for milk processing. After consulting docu-
ground substances, in only some of the houses, what mentation from the excavations, I came to the conclusion that it is
an exceptionally well-preserved clay millstone structure. It has been
would be the reasons for such technological advances
incorrectly interpreted for decades because of the lack of analogies
not having been used in all, or at least most, of the Late and because the results of the excavations have not been completely
Vin~a houses? published.

56
Adam N. CRNOBRNJA, Investigations of Late Vin~a House 1/2010 at Crkvine in Stubline (4564) STARINAR LXII/2012

on and around them. All this material was sealed off


with a yellow layer of highly burnt soil (AE17). The
dimensions of the palettes discovered in the house
vary between 0.25 x 0.13 m and 0.50 x 0.26 m.

Pottery vessels (Fig. 3/10)


Eleven groups of pottery (PG) of different size and
contents have been found on the house floor. As detailed
analysis of pottery from the house is still not complet-
ed36, I will, in this work, give just a short summary of
these types of finds to the extent necessary to compre-
hend the activities taking place inside house 1/2010.
It is important to emphasise that at least six large
pithoi were found in the northern section of the house.
Two pithoi, with bases inserted in the floor (PG17),
one of which was made of unfired clay; two pithoi next
to four millstones on one side and a clay millstone struc-
ture on the other side (PG13); one in front of oven 1 Fig. 15. Group of loom weights, from the north
(PG14) and one next to the western side of oven 1. A
few fragmented, rather large, amphorae (two in the Sl. 15. Grupa tegova za razboj, snimak sa severa
storage container) have also been found. Generally
speaking, the large capacity vessels, which could be
identified as storage vessels, prevail among the objects at Vin~a at depths from 8 m to 4.1 m. These were iden-
found on the house floor. In addition, a smaller num- tified as altar fragments39, probably on the basis of one
ber of vessels for cooking and consuming food were such bowl with three legs.40
found. Another complete vessel within PG17 is a vessel
of a spherical shape with three short, but thick, legs bent
Objects of cult at an angle of approximately 90. Its purpose is not
A few of the finds could, generally, be identified as quite clear and the shape of its legs has direct analogies
objects for cult purpose (Fig. 14). Only one complete with the legs of some altars.41 Our three-legged vessel
figurine has been found in house 1/2010. It was dis- was found among the fragments of a large pithos made
covered in the storage container (AE28), standing of unfired clay. One of these fragments completely
upright on the floor and facing east. covered the mouth of the vessel, so we recovered the
Within PG17, two interesting, complete vessels bowl with its contents, which should soon be submit-
have been found among the fragments of the two largest ted for further analysis.
pithoi in the house. One of them is a small conical bowl
with eight protomes and a hollow base. The protomes Loom weights (Fig. 3/11)
were facing towards the inside of the vessel and were A group of eight ceramic loom weights (Fig. 15)
arranged in four pairs, separated by engravings on the was discovered between the eastern wall of oven 1 and
vessel rim. A small bowl, iconographically almost iden- the eastern wall of house 1/2010. On top of them was
tical to our specimen, was found in house 1/06 at the
site Belo Brdo in Vin~a.37 A bowl of almost identical
shape, but with four protomes on the rim, was found in
the Late Vin~a burial site 12 at Gomolava.38 This could 36 Milo{ Spasi}, the curator of the Belgrade City Museum, is

also be interpreted as a repeat of an identical icono- in charge of a study of the pottery. These results will be the subject
of separate texts.
graphic pattern to the one found on our specimen and 37 Tasi} 2007.
on the specimen from Vin~a, where there are four pairs 38 Bori} 2009, 223, fig. 36.
of protomes on the bowl rim. Fragments of similar, 39 Stankovi} 1986, T. X, 6, 7, 8.

small conical bowls, with one preserved protome facing 40 Stankovi} 1986, T. IX, 12.

towards the inside of the vessel, have also been found 41 Stankovi} 1986.

57
Adam N. CRNOBRNJA, Investigations of Late Vin~a House 1/2010 at Crkvine in Stubline (4564) STARINAR LXII/2012

Fig. 16, 17, 18. East, west and south head


Sl. 16, 17, 18. Isto~na, zapadna i ju`na glava

the collapsed wall of oven 1. The weights are of an head has an oval impression of a post or plank on which
elongated discoid shape with a hole at approximately a it was mounted, as well as a horizontal impression of a
quarter of its length from the top edge. A narrow, branch or rope which attached it to the support.
tongue-shaped groove runs from the holes in the weights Another, so-called west head, was around 0.40 m
at their top side, indicating where the rope/warp of the from the south-western corner of oven 1 (Fig. 3/12b).
weaving ran. The weights are carefully made, well fired It is more substantial than the first one and of a simi-
and relatively heavy with highly polished surfaces, so, lar, but much more summary and stylised, execution
when holding them, they give the impression of a stone, (Fig. 17). It could be assumed, with a degree of cer-
rather than a ceramic, weight. The place in which they tainty, that the west head was on the post or group of
were found, next to the oven, has also been recorded posts placed in a triangular hole in the floor that was
many times at other sites42, as well as in house 1/2008 found immediately next to the south-western corner of
at Crkvine43 and the convenience of that spot for locat- oven 1 (Fig. 6/1). The summary appearance of this
ing the loom, because of technical advances (proximi- head resembles a bovine head so it is possible to
ty of heat and light), is also confirmed by ethnological assume that it was a bucranium. However, it must be
parallels.44 mentioned that there are no traces of horns, either gen-
uine or made of clay.
Clay heads (bucrania?) (Fig. 3/12) The third, so-called south head, was discovered
Three very interesting, large stylised heads have in the destroyed southern section of the house (Fig.
been found in house 1/2010. Two of them were discove- 3/12). In contrast to the above two specimens, the con-
red next to the front side of oven 1, and could be text of its find remains unclear to a great extent. Its
vaguely recognised as bucrania.45 oval face was modelled with a minimum of plasticity
The first one, so-called east head, was found and it only has a long, straight and narrow nose, with
immediately in front of the eastern corner of oven 1
(Fig. 3/12a) and was lying on the house floor, facing
downward. This head/bucranium was executed in a
highly stylised manner (Fig. 16). The forehead section 42 Nin~i} 2011, 187, 191.
is triangular, tapers into the nose line, and expands 43 Crnobrnja 2011, 133, fig. 4/2.
near the base. The eyes are also of a stylised, triangu- 44 Idvorean-Stefanovi} 2008, 100.
lar shape and on the cheeks there are horizontal paral- 45 For a detailed analysis of the context and stylistic charac-

lel lines executed with fingers. On the reverse side, this teristics of these finds see Spasi} 2012, in press.

58
Adam N. CRNOBRNJA, Investigations of Late Vin~a House 1/2010 at Crkvine in Stubline (4564) STARINAR LXII/2012

two eyes executed as two elongated spirals (Fig. 18).


The reverse of this head is flat and the shape and posi-
tion of its edges suggest that it had possibly been
applied directly to the wall like some kind of architec-
tural decoration. The way in which the eyes (spirals)
are depicted resembles the bucrania from Gomolava,46
but their faces are more pronounced, protruding and
with horns. For our south head, however, it is certain
that it had no horns, so it is doubtful whether this
head could be identified at all as bucranium, even in
the widest sense of the word. On this basis, should we
then assume that it represents a stylised image of some
other animal, perhaps even human?

Fig. 19. Pit 1, from the east


Other portable finds
Besides the above described objects, 14 stone Sl. 19. Jama 1, snimak sa istoka
blades, one hammer stone, two ceramic balls and a
fragment of one copper bead were also discovered
inside the house. In addition to the two previously pieces of daub fused into this one lump. The position
mentioned ceramic balls, a group of 8 ceramic balls of pit 1 clearly indicates that it originates from the time
was also found next to western edge of the rear wall of after the destruction of house 1/2010, as it partially
oven 2, but as it is not certain that they were inside the overlaps the assumed line of the eastern wall of the
house, we do not include them in the house inventory. house that most probably did not exist at the time the
pit was dug. We could even question how visible the
remains of house 1/2010 were at that moment, as the
LOCATION USE AFTER remains of a small clay table on the preserved house
HOUSE DESTRUCTION floor are only 1.80 m north of pit 1, while the eastern
wall of house 1/2010, which was in the immediate
Pits with daub (Fig. 3/13 and 14) vicinity of the table, is not preserved. Special attention
Four waste pits filled with daub, i.e. the remains of was paid to the investigation of pits filled with daub
the walls of burnt houses, have been recorded in the discovered at the nearby Late Vin~a settlement of Crkvi-
southern section of the trench. Pit 1 is of particular ne in Mali Borak. A few possible interpretations were
importance for studies at this micro location after the suggested, ranging from waste containers48 to ritual
destruction of house 1/2010 in a conflagration, so, for structures originating from complex ritual practices.49
now, we will pay special attention to this pit.47 A. Tripkovi} warned about the necessity to comprehend
Pit 1 had a rectangular ground plan (Fig. 19), 2.20 the rational character of prehistoric man,50 as a very
x 1.30 m in size and 0.70 m deep (with a capacity of important but, very frequently, neglected question. If we
around 2 cubic meters). It was filled with densely packed keep in mind the numerous examples of the secondary
fragments of wall daub with impressions of branches, use of various materials within Neolithic households,
sticks and planks. A large number of daub fragments we should wonder where that material was stored from
were placed on the edge. In the pit was also found a the moment when it was reclaimed from its original use
small quantity of pottery fragments, a few animal bones, to the moment when it was reused. If the substructure
six blades of flint and light white stone, one adze of
light white stone and two cores, one made of flint and
the other made of light white stone, both with flaking
46 Petrovi} 1992, 2122, sl. 4, 5.
scars. There were also many pieces of vitrified daub and 47 Pits 24 were very shallow (up to 0.25 m) and they will not
in the very centre of the pit (measured from the verti- be particularly discussed in this work.
cal as well as horizontal axes) one large lump of excep- 48 Tripkovi} 2011, 85.

tionally highly vitrified daub was found, which had 49 Tripkovi} i dr. 2011, 257.

been burning at such a high temperature that a few 50 Tripkovi} 2011, 85.

59
Adam N. CRNOBRNJA, Investigations of Late Vin~a House 1/2010 at Crkvine in Stubline (4564) STARINAR LXII/2012

of the entire floor of house 1/2010 consisted of sec- 4) Pit 1, which contained around 2 cubic meters of
ondary used daub pieces, they would occupy around densely packed burnt daub, was not visible to geo-
12 cubic meters, which must have been deposited magnetic recording because dislocated daub fragments
somewhere in the area. This raises the question of the have different magnetic directions resulting in lower
necessity of their storage until they were reused. Could values of geomagnetic spectra that are additionally
the pits filled with daub have been containers for raw masked by the close proximity of anomalies of medi-
materials for future building and not merely waste pits um to high values.
or elements with some symbolic context? 5) The geoelectric scanning of profiles proved to
be very reliable in the detection of the length and depth
of previously identified anomalies.
CONCLUDING REMARKS 6) The combination of geomagnetic and geoelectric
investigations made possible the precise planning of
The selection of the trench location on the basis of the location of investigations, as well as the relatively
previous geophysical investigations offered us the precise planning of time and financial resources.
possibility to investigate the very well preserved house
1/2010 at Crkvine in Stubline. The date when life ended Rationality in construction
in this house could be generally dated, on the basis of Where our knowledge about building techniques
pottery finds inside the house, to phase D of the Vin~a is concerned, house 1/2010 yielded few new details.
culture. A more precise chronological relationship bet- Maybe the greatest surprise was the method of floor
ween house 1/2010, the layer on top of it, pit 1, which construction, i.e. the use of the walls of some previously
negates the former house area, and house 1/2008, will burnt houses as foundations and at the same time as a
be clearer after dating using the AMS method, and good insulation layer. The use of daub fragments from
once samples for dating have been sent to the labora- earlier phases in the building process has been record-
tory of the Rudjer Bo{kovi} Institute in Zagreb. ed in house 12 at Divostin where it had been mixed
The campaign of 2010, as well as the investigation with stone, but only as a base, on top of which was laid
of house 1/2010, yielded much new information and im- earth and boards as subflooring51 and as the floor sub-
portant data both for the study of life in the Late Neo- structure under sections of house 4 at Gomolava.52 The
lithic households and the methods of field investigations. secondary use of daub for the construction of house
1/2010 suggests an economical and rational approach
Geophysical Investigations to construction. In such a way, a direct saving of build-
Systematic archaeological investigations provided ing time and resources, i.e. lumber, was achieved. The
exceptionally important data for the precise interpreta- indirect profitability of such a building process is evi-
tion of the results of geophysical investigations at this dent in the fact that a well insulated floor was obtained
site and some of our preliminary findings are presented with a reduction in heat loss whilst heating the house.
here: We could also notice in house 1/2010 the confirma-
1) Sections of the houses that burned with the tion of earlier assumptions regarding the knowledge
highest intensity the zone with a yellow, highly burnt that inhabitants of the Vin~a settlements had about the
layer (AE17) and oven 2, with the dimensions of these necessity of thermal efficiency in the construction
anomalies corresponding to the dimensions of the afo- process of the house.53 Besides the method of floor
rementioned structures, were encountered in the zone construction, two more examples also confirm this:
with the highest geomagnetic values. by constructing the attic/storey structure above
2) The exceptionally strong emissions of the the north-western section of the house, the volume of
aforementioned structures hindered the visibility of space which had to be heated in winter was reduced
the immediate surroundings (floor), so the geomagnetic and the insulation of the structure was improved;
anomaly is actually smaller than the house outline.
3) The area in the southern section of the trench,
where rather large amounts of dislocated daub fragments
were deposited, did not have substantially more geo- 51 Bogdanovi} 1988, 4849.
magnetic values in contrast to the floor area preserved 52 Petrovi} 1992, 25.
in situ in the same section of the house. 53 Miloradovi}, Tasi} 2008.

60
Adam N. CRNOBRNJA, Investigations of Late Vin~a House 1/2010 at Crkvine in Stubline (4564) STARINAR LXII/2012

substantial insulation layers in the construction


of oven 1 and, in contrast, the thin floor in oven 2, indi-
cate that technical knowledge was applied rationally.
Greater effort was invested in the construction of oven 1,
for which it was essential that heat was not dispersed in
any unwanted way (the loss of heat through the floor is
reduced and most of it remains in the firebox and is
dispersed through the calotte) than in the construction
of oven 2.
This situation suggests the possibility that these two
ovens had different purposes. Special attention should
be paid to this fact as there are also other examples of
Vin~a houses where oven foundations were constructed
in a different way, as in house 8 at Banjica.54 I empha-
sise this because explorers often have a tendency to
explain all levels in oven construction as floor renova-
tions and try to determine the period of usage of the
house according to that, disregarding the characteristics
of oven building technology, as well as the possibility
that variously built ovens could have had at least, to
some extent, an additional, different purpose.

Organisation of life in house 1/2010 Fig. 20. Reconstruction of northwest section of house
We did not discover any traces of internal walls in 1/2010, oven 1 with heads (bucrania?)
this house, so we could only speculate about the possi- (drawing: Kosta Milovanovi})
ble existence of more than one room. Nevertheless, it
Sl. 20. Rekonstrukcija izgleda severozapadnog dela
could be concluded, with a degree of certainty, that in
ku}e 1/2010, pe} 1 sa glavama (bukranionima?)
the first 9 meters of house interior, looking from the (crte`: Kosta Milovanovi})
north, there was no partition wall spanning its entire
width. A trace of soot on the floor next to the eastern
wall of the house, which extends from oven 1 to the
small clay table, has been interrupted by recent dig- on the basis of the house inventory, that many every-
ging but its continuity is certain. The position of oven day activities were taking place in the northern section
2, i.e. the assumed position of its opening, indicates of the house.
that, despite the poor preservation of the oven, it was Two large ovens, and activities associated with
in the same room as oven 1. A possible partition wall, them, dominated not only the northern half, but the
which could have divided the house lengthwise into entire house interior. In fact, the entire northern half of
two rooms, could be expected only in the section the house could be considered as an area intended for
beyond the clay table (i.e. south of it). On the other activities related, primarily, to food processing.
hand, it should not be ruled out that the area was par- Already, after preliminary analysis, we can recognise
tially partitioned in the south-western quarter of the functional sub-zones related to three stages of food
house (beyond oven 2). However, this could not be processing:55
confirmed because of the poor state of preservation of 1) storing: containers and large vessels (AE28);
that section of the house. Also, possible traces of post- four groups of pithoi (PG13, PG14, PG17, PG20
holes for a partition wall were impossible to record be-
cause of the previously mentioned soil characteristics.
Despite the fact that the southern section of house
1/2010 was considerably damaged and we, therefore, 54Tripkovi} 2007, 7475.
do not have sufficient evidence for discussing the 55In the house was found a small number of vessels for food
activities carried out in that section, it is conspicuous, consumption.

61
Adam N. CRNOBRNJA, Investigations of Late Vin~a House 1/2010 at Crkvine in Stubline (4564) STARINAR LXII/2012

along with PG22 next to the assumed southern wall) heads flank the cult space, the oven in house 1/2010,
and the assumed possibility of using the attic/storey and the alter at Parta. However, it should be empha-
structure as a storage space.56 sised that the situation recorded at Para is the result of
2) mechanical processing of food (the clay struc- some substantial reconstruction.
ture of the millstone and five stone working surfaces) The discovery of many objects of assumed cult
3) thermal processing of food (two ovens) purpose within Neolithic houses often prompted inve-
In addition, there was also a small loom in the stigators to identify such structures as either sanctuaries
northern section of the house, confirmed by a group of or cult structures,60 particularly when bucrania were
eight loom weights discovered next to the eastern wall also present. Nevertheless, I am more inclined to agree
of oven 1. with the already proposed opinion that it was in fact an
Despite the distinctively utilitarian character of the overlap of everyday and religious activities which was
northern section of the house, this was not its only pur- characteristic of phases C and D of the Vin~a culture.61
pose. Many artefacts, which could be identified as cult I am even more inclined towards this opinion as we
associated objects, have also been found in the same have the most direct physical interconnection of sacred
area. Besides smaller finds with cult characteristics (a and profane elements in house 1/2010.
figurine, a small bowl with eight protomes, and a ves- Taking into account the situation in house 1/2010,
sel with three legs), two clay heads (bucrania?) are par- which is generally exceptionally well preserved and
ticularly indicative. the fact that the depth of its deepest remains (the prepa-
The location of the east and west heads, in front of ration of the ground for the floor construction) corre-
and next to oven 1, allow the assumption, with a great sponds with the depth determined by geoelectric scan-
degree of certainty, that while house 1/2010 was inha- ning (11.10 m), it has been decided not to remove its
bited, these heads were mounted on the posts to the left entire floor nor to excavate the oven. Instead, the
and right of the oven, thus creating one complex com- remains of house 1/2010 with the fixed elements of the
position (Fig. 20). In such a way the importance of the interior (except the small table, the heads and the mill-
large oven, which already dominated the house interi- stone) will be preserved and covered with earth. After
or and around which was concentrated most of the consultation with associates of the Central Institute for
household activities, is additionally emphasised by Conservation in Belgrade, the house was, before being
uniting profane and sacred functions in one place. covered with earth, protected with a cover of geotextile
Much has been written, on many occasions, about the fabric and all elements above the floor level (both ovens,
possible cult aspects associated with ovens in some stone working surfaces and the like) were additi-
Neolithic houses, so I will just draw attention to some onally protected by placing sacks of sand around them.
of these works,57 and to some situations when objects After that, the entire house area was covered with a 10 cm
of assumed cult usage were encountered next to an thick layer of sand and on top of it was placed a pro-
oven.58 This includes a find consisting of 43 figurines tective plastic net. In such a way, the remains of house
discovered in house 1/2008 at Crkvine in Stubline.59 1/2010 were packed and protected and left as an
The context, which most directly resembles the situa- undertaking for the future until such time as conditions
tion in house 1/2010, is, without doubt, a well-known are favourable for its presentation at the site.
altar from Para. It seems that in both cases there is a
repetition of an almost canonised form in which two Translated by Mirjana Vukmanovi}

56 For more about storage methods in the Late Vin~a houses,

see Tripkovi} 2011.


57 Petrovi} 2000/2001; Naumov 2010, 230, 232.
58 Petrovi} 1992, 21; [ljivar, Jacanovi} 2005.
59 Crnobrnja 2011.
60 Jovanovi}, Gli{i} 1961; Petrovi} 1992.
61 Chapman 1981, 6268; Crnobrnja, Simi}, Jankovi} 2010, 21.

62
Adam N. CRNOBRNJA, Investigations of Late Vin~a House 1/2010 at Crkvine in Stubline (4564) STARINAR LXII/2012

BIBLIOGRAPHY:

Antonovi}, [ari} 2011 D. Antonovi}, J. Lazarovici, Lazarovici 2006 C. M. Lazarovici,


[ari}, Kameno oru|e sa lokaliteta Crkvine u G. Lazarovici, Arhitectura neoliticului si epocii cupru-
Stublinama, Starinar LXI, Beograd 2011, 6179. lui din Romania I (Neoliticul), Iasi: Trinitas Editura
Bogdanovi} 1988 M. Bogdanovi}, Architecture Mitrpoliei Moldovei si Bucovinei.
and Structural Features at Divostin, in A. McPheron and Miloradovi}, Tasi} 2008 N. Miloradovi} i N. Tasi},
D. Srejovi} (eds.), Divostin and the Neolithic of Central Neolitsko naselje Vin~a termi~ki aspekt, Zbornik ra-
Serbia, Ethnology Monographs 10, University of dova za 39. kongres o KGH, Savez ma{inskih i elektro-
Pittsburgh Department of Arhaeology, Pittsburgh tehni~kih in`enjera Srbije, Beograd 2008, 419426.
1988, 35142. Naumov 2010 G. Naumov, Neolithic anthropo-
Bogdanovi} 2004 M. Bogdanovi}, Grivac : na- centrism: the principles of imagery and symbolic mani-
selja protostar~eva~ke i vin~anske kulture, Centar za festation of corporeality in the Balkans, Documenta
nau~na istra`ivanja SANU i Univerziteta u Kragujevcu, Praehistorica XXXVII, Ljubljana 2010, 227238.
Kragujevac 2004. Nin~i} 2011 O. Nin~i}, Tekstil na lokali-
Bori} 2009 D. Bori}, Absolute Dating of Metal- tetu Crkvine, Kolubara 5, Beograd 2011, 181193.
lurgical Innovations in the Vin~a Culture of the Balkans, Pecikoza 2009 V. Pecikoza, Analiza pe}i iz
in Metals and Societies: Studies in honour of Barbara mla|eg neolita na centralnom Balkanu, Arhaika 2,
S. Ottaway, eds. T. K. Kienlin and B. W. Roberts, Verlag Beograd 2009, 2735.
Dr. Rudolf Habelt Gmbh, Bonn 2009, 191245. Petrovi} 1992 J. Petrovi}, Arhitektura ku-
Crnobrwa 2009 A.N. Crnobrwa, Neolitsko }e 4 na Gomolavi naseqe mla|e vin~anske kultu-
naseqe na Crkvinama u Stublinama istra`ivawa re, Rad vojvo|anskih muzeja 34, Novi Sad 1992,
2008. godine, Muzej grada Beograda, Beograd 2009. 1928.
Crnobrnja 2011 A. N. Crnobrnja, Arrangement Petrovi} 2000/2001 B. Petrovi}, Model pe}i
of Vin~a culture figurines: a study of social structure and iz Progara, Godi{wak grada Beograda 47/48, 1123.
organisation, Documenta Praehistorica XXXVIII, Ljub- Schier 2006 W. Schier, Neolithic House Building
ljana 2011, 131147. and Ritual in the Late Vin~a Tell Site of Uivar, Romania,
Crnobrnja, in press A. N. Crnobrnja, Group Iden- in Homage to Milutin Gara{anin, eds. N. Tasi} and C.
tities in the Central Balkan Late Neolithic, Documenta Grozdanov, Belgrade 2006, 325339.
Praehistorica XXXVIII, Ljubljana. Simi}, Crnobrwa 2008 Z. Simi} i A. N. Crno-
Crnobrnja, Simi} and Jankovi} 2010 A. N. Crno- brwa, Sonda`no iskopavawe lokaliteta Crkvine u
brnja, Z. Simi} and M. Jankovi}, Late Vin~a culture set- selu Stubline, Arheolo{ki pregled n.s. 4 (2006), Be-
tlement at Crkvine in Stubline: household organization ograd 2008, 4446.
and urbanization in the Late Vin~a culture period, Sta- Spasi}, in press M. Spasi}, Cattle to settle Bull
rinar LIX (2009), Beograd 2010, 925. to rule: On Bovine iconocraphy among Late Neolithic
Idvorean-Stefanovi} 2008 B. Idvorean-Stefanovi}, Vin~a Culture communities, Documenta Praehistorica
Znanje o tradicionalnim radnim procesima kao nemateri- XXXIX, Ljubljana.
jalni sadr`aj muzeolo{ke obrade alata za tkanje, u Muze- Stankovi} 1986 S. Stankovi}, @rtvenici i proso-
ji u Srbiji: zapo~eto putovanje, ur. Lj. Gavrilovi} i M. Sto- pomorfni poklopci iz Vin~e, Centar za arheolo{ka istra-
janovi}, Muzejsko dru{tvo Srbije, Beograd 2008, 97106. `ivanja, Beograd 1986.
Jovanovi}, Gli{i} 1961 B. Jovanovi}, J. Gli- [ljivar, Jacanovi} 2005 D. [ljivar, D. Jacanovi},
{i}, Eneolitsko naseqe na Kormadinu kod Jakova, Zoomorphic Figurines from Belovode. Zbornik Narod-
Starinar XI, Beograd 1961, 113139. nog muzeja (Beograd) XVIII1, 6978.
Lazarovici on line G. Lazarovici, D. Ciobotaru, Tasi} 2007 N. N. Tasi}, Ritual Pottery Set from
M. Merlini, P. Ucelli Gnesutta, Z. Maxim, C. M. Laza- Vin~a, Glasnik Srpskog arheolo{kog dru{tva 23,
rovici, D. Micle, S. Petrescu, D. Negrei, C. Roman, A. Beograd 2007, 203210.
Mates, M. Ardelean, PARA archaeological excavations Tasi} 2007 i dr. N. Tasi}, S. \uri~i}, B. La-
The deer house (Casa Cerbului) 2006, http://www.pre- zarevi}, Analiza konstrukcije `rvwa iz objekta
history.it/ftp/parta_archaeological_excavations2006_ 01/06 u Vin~i, Glasnik Srpskog arheolo{kog dru{tva
01.html 23, Beograd 2007, 211218.

63
Adam N. CRNOBRNJA, Investigations of Late Vin~a House 1/2010 at Crkvine in Stubline (4564) STARINAR LXII/2012

Todorovi} 1967 J. Todorovi}, Crkvine, Stubline, Tripkovi} 2007 B. Tripkovi}, Doma}instvo


Obrenovac naselje vin~anske grupe, Arheolo{ki pre- i prostor u kasnom neolitu: vin~ansko naseqe na Ba-
gled 9, Beograd 1967, 1718. wici, Srpsko arheolo{ko dru{tvo, Beograd 2007.
Todorovi} 1981 J. Todorovi}, A recently discove- Tripkovi} 2011 A. Tripkovi}, Jame ispuwene
red House in the Neolithic Settlement of Banjica in Bel- lepom na lokalitetu Crkvine, Kolubara 5, Beograd
grade, Archaeologica Iugoslavca XVIII, Belgrade 1981, 2011, 8599.
1316. Tripkovi} 2011 B. Tripkovi}, Containers and gra-
Tringham et al. 1992 R. Tringham, B. Brukner, ins: food storage and symbolism in the Central Balkans
T. Kaiser, B. Ksenija, Lj. Bukvi}, P. [teli, N. Russell, (Vin~a period), Documenta Praehistorica XXXVIII,
M. Stevanovi}, B. Voytek, Excavations at Opovo, Ljubljana 2011, 159172
19851987: Socioeconomic Change in the Balkan Neo- Tripkovi} i dr. 2011 B. Tripkovi}, A. Trip-
lithic, Journal of Field Archaeology, Volume 19, Num- kovi}, S. Stefanovi}, V. Dimitrijevi}, Kontekst
ber 3, 1992, 351386. A. C. 2.1 na lokalitetu Crkvine, Kolubara 5, Beo-
grad 2011, 251259.

Rezime: ADAM N. CRNOBRWA, Muzej grada Beograda, Beograd

ISTRA@IVAWE POZNOVIN^ANSKE KU]E 1/2010


NA CRKVINAMA U STUBLINAMA

Kqu~ne re~i. neolit, vin~anska kultura, ku}a, geofizi~ka istra`ivawa, bukranioni,


`rvaw, Stubline, Obrenovac.

Poznovin~ansko naseqe na lokalitetu Crkvine nalazi se u severnom i znatno o{te}ena u ju`nom delu ku}e, a ustanov-
na oko 40 km jugozapadno od Beograda (Srbija), u blizini qeno je da su kao supstrukcija podnice iskori{}eni koma-
sela Stubline, op{tina Obrenovac. Situirano je na bla- di lepa koji poti~u sa neke ranije izgorele ku}e. Na osnovu
gom uzvi{ewu, sa severa i juga ome|enom potocima koji se specifi~nih tragova gorewa i obru{avawa pojedinih pred-
spajaju ispod wegovog jugoisto~nog kraja, i zauzima povr{i- meta zakqu~eno je da je ku}a 1/2010 iznad severne polovine
nu od oko 16 ha (sl. 1). Geomagnetskim mapirawem, u perio- posedovala spratnu/potkrovnu konstrukciju. Nalazi u se-
du od 2007. do 2011. godine istra`ena je, za sada, povr{ina od vernom delu ku}e upu}uju da se taj prostor prevashodno ko-
ukupno 83.000 m, dok je geoelektri~nim skenirawem dosad ristio u svrhu ~uvawa, obrade i pripreme hrane (kaseta za
snimqeno ukupno 1250 m profila, koji su doneli infor- skladi{tewe na podu ku}e, vi{e pitosa i ve}ih posuda za
macije o vertikalnoj o~uvanosti objekata prethodno loci- skladi{tewe, glinena konstrukcija `rvwa i jo{ pet kame-
ranih geomagnetskim mapirawem, kao i osnovne podatke o nih radnih povr{ina, dve velike pe}i), mada su prisutni
debqini kulturnog sloja na razli~itim delovima naseqa. tragovi i drugih svakodnevnih aktivnosti tkawa (razboj
Pore|ewem intenziteta geomagnetskih anomalija (~ije su uz pe} 1) i konzumirawa hrane (mawi broj posuda za pripre-
implikacije u ~etiri slu~aja proverene i iskopavawima) mu i konzumirawe). U ku}i su prona|ene i tri velike gli-
i wihovih dimenzija, na istra`enom prostoru je mogu}e nene glave (bukranioni?), od kojih su se dve nalazile pored
pretpostaviti postojawe ostataka od preko 200 ku}a (sl. 2). pe}i 1, sa wene predwe strane, kao i tri mawa predmeta
U jesen 2010. godine Muzej grada Beograda preduzeo je kultne namene (figurina i dve posude). Fizi~ko preplita-
arheolo{ka iskopavawa na osnovu prethodnih geofizi~- we funkcija svetovne (priprema hrane) i sakralne namene
kih ispitivawa otvarawem sonde dimenzija 18 m h 8 m, u (glinene glave i mawi predmeti kultne namene) ukazuju na
okviru koje je istra`ena ku}a 1/2010, koja se okvirno mo`e preplitawe te dve sfere, koje se, izgleda, susre}u u velikoj
opredeliti u fazu D vin~anske kulture. U radu se daje iscr- pe}i koja dominira tim prostorom i oko koje su nalazi obe
pan izve{taj o na~inu gradwe ove ku}e, kao i razmatrawe o sfere koncentrisani.
organizaciji `ivota u woj. Zidovi ku}e su, po svoj prili- Pored navedenog, iskopavawima obavqenim 2010. godi-
ci, gra|eni od talpi oblaganih blatom pome{anim sa organ- ne dobijeni su i izuzetno va`ni podaci za precizno i{~ita-
skim primesama. Podnica ku}e bila je veoma dobro o~uvana vawe rezultata geofizi~kih ispitivawa na ovom lokalitetu.

64
ALEKSANDAR KAPURAN, Arheolo{ki institut, Beograd
ALEKSANDAR BULATOVI], Arheolo{ki institut, Beograd

KULTURNA GRUPA KOCOFENIKOSTOLAC


NA TERITORIJI SEVEROISTO^NE SRBIJE

UDK: 903"636"(497.11) e-mail: a.kapuran@gmail.com


DOI: 10.2298/STA1262065K Primqeno: 09. februar 2012.
Originalan nau~ni rad Prihva}eno: 21. jun 2012.

Apstrakt. Posledwih godina se pokazalo da na teritoriji Srbije postoji mnogo ve}i broj lokaliteta
Kocofeni kulturne grupe nego {to je publikovano pre jedne ili vi{e decenija. Objediwavawem nepublikovanih
podataka i dokumentacije sa arheolo{kih iskopavawa i rekognoscirawa, kao i apsolutnim datumima dobijenim
posledwih godina, stvorena je potreba da se u nekim aspektima koriguju i boqe sagledaju rasprostrawenost,
na~in `ivota i trajawe ove kulturne grupe na prostoru severoisto~ne Srbije, gde je ona i najprisutnija. Osim nalaza
materijalne kulture, u radu su prezentovane topografske karakteristike naseqa koja su sada u najve}em procentu
georeferencirana. Ovaj rad, tako|e, predstavqa poku{aj da se pojasne neke od dilema vezanih za dru{tveno-ekonomski
i tehnolo{ki razvoj populacija kasnog bakarnog i ranog bronzanog doba.

Kqu~ne re~i. severoisto~na Srbija, pozni eneolit, rano bronzano doba, KocofeniKostolac grupa,
distribucija naseqa, kerami~ka produkcija.

T
eritorija severoisto~ne Srbije, zahvaquju- votnim namirnicama grupama metalurga koji pro-
}i brdsko-planinskom reqefu, i u dana{we izvode dragocene metalne predmete.3 Na ovo ukazuju
vreme predstavqa geografski veoma izolo- velika naseqa sa solidnom arhitekturom, koja pre-
vano podru~je. Populacije koje su naseqavale te- zentuju sedentarni na~in `ivota (BubawNovo selo,
ritoriju Srbije u starijem eneolitu po~iwu da po- ^oka lu Bala{, Ku~ajna, Kmpije, [kodrino poqe,
kazuju interesovawe za severoisto~nu Srbiju onog Bawska stena itd.). Prirodni resursi u okru`ewu
trenutka kada se javqa veliki interes za metalur- imaju potencijal da prehrane relativno mali broj
giju bakra, {to su potvrdila istra`ivawa Rudne
glave.1 Okolna vin~anska naseqa nalaze se na sa-
mo oko 5070 km udaqenosti, na obalama Dunava u
1 Jovanovi} 1971.
Kqu~u i zapadnim obroncima Ku~ajskih planina.
2 Tasi} 1976.
Novoprido{le kulture sredweg bakarnog doba, ko-
3 Na na{im prostorima, od metalnih proizvoda dominira-
je su u su{tini sto~arske, u agrarnim oblastima
ju bakarne krstaste sekire, na kojima u najve}em broju slu~ajeva
Panonije bave se zemqoradwom,2 koja i daqe ima ne postoje tragovi upotrebe, kako zbog meko}e i neprakti~nosti
va`nu ulogu u proizvodwi hrane i snabdevawu `i- za upotrebu, tako i zbog wihove uloge u razmeni.

* ^lanak predstavqa rezultat rada na projektu: Arheologija Srbije: Kulturni identitet, integracioni faktori, tehnolo{ki
procesi i uloga centralnog Balkana u razvoju evropske praistorije (br. 177020) koje finansira Ministarstvo prosvete, nauke i
tehnolo{kog razvoja Republike Srbije.

65
KAPURAN. BULATOVI], Kulturna grupa KocofeniKostolac na teritoriji (6594) STARINAR LXII/2012

1. ^oka Morminc; 2. Mali Krivelj, kod vodenice;


3. ^oka lu Bala{; 4. ^oka Kormaro{; 5. [arbanovac,
Seli{te; 6. Lazareva pe}ina; 7. Vernjikica;
8. Bogovinska pe}ina i naselje; 9. Crnajka; 10. Kulmja
[kjopuluji; 11. Kljanc; 12. Jezero; 13. Kapetanova pe}ina;
14. Rajkova pe}ina; 15. Pe{}era mare; 16. Arija
babiKo{obrdo; 17. Velike livadice 2; 18. Lepenska
potkapina; 19. Katarinine livade; 20. Vlasac; 21. Pe}ina
kod Trajanove table; 22. Turija, Stenje; 23. Neresnica,
Velika ~uka; 24. Dobra, Manastir; 25. Padina; 26. Malo
Golubinje, Re~ica; 27. Hajdu~ka vodenica; 28. Abri
ispod Banjske stene; 29. Gamzigradska banja, Njiva Z.
Brzanovi}; 30. Gamzigrad, Varzari; 31. Smiljkova glavica;
32. Seli{te; 33. Vratna, Veliki most; 34. [arkamen,
Duge livade; 35. Popovica, Veliko brdo; 36. Brusnik,
Glavica; 37. Kapu \aluluj; 38. Kovilovo, ]eta}e;
39. Mokranjske stene, Kamenolom; 40. Mokranjske
stene, potkapina; 41. Prahovo, Ide}e; 42. Smedovac,
GrabarSvra~ar; 43. Sikole, Gradi{te; 44. Karata{, Dijana;
45. Donje Butorke; 46. Kladovo, Brodoimpeksnekropola;
47. Mala Vrbica, Livade; 48. Mala Vrbica, 500 m od sela;
49. Korbovo; ZbradilaFund; 50. Korbovo, Obala;
51. Korbovo, GlamijaObala; 52. VajugaPesak;
53. U{}e Jakomirskog potoka; 54. Velesnica, Biljevina;
55. Ljubi~evac, Obala; 56. Ljubi~evac, Ostrvo;
57. Grabovnica, Brzi prun; 58. U{}e Slatinske reke;
59. Knjepi{te; 60. Ru`enjka; 61. Kusjak, Bor|ej;
62. Kusjak, Motel; 63. Kusjak, Vrkalj; 64. Majdanpek,
Kameni rog; 65. Knja`evac, Dubrava; 66. Rgo{te, Bolvan;
67. Donje Zuni~e, Ad`ijsko; 68. Mokranje, Lalunj;
69. Tanda, la Tufek; 70. Brestova~ka banja;
71. Ni{, Bubanj; 72. Donji Milanovac, Veliki Gradac;
73. Humska ~uka; 74. Donja Vre`ina; 75. Donja Bela
reka; 76. Rudna glava, [eta}e; 77. Rgo{te, Vi{njar;
78. Dobra, Gospo|in vir; 79. Kusjak; Grle

Karta 1. Lokaliteti KocofeniKostolac grupe na teritoriji severoisto~ne Srbije


Map 1. CotofeniKostolac group sites on the territory of North Eastern Serbia

stanovnika (sude}i prema broju i gabaritima ot- Prema rezultatima posledwih revizionih re-
krivenih naseqa) me|usobno ekonomski zavisnih, kognoscirawa, broj lokaliteta ove kulturne grupe
sve dok ih pogor{awe klimatskih uslova ne pri- u me|uvremenu je porastao na 76. Do sada su konsta-
sili da migriraju u neka druga, za `ivot pogodni- tovani slede}i lokaliteti: Kriveq, ^oka Mor-
ja, podru~ja. 4 minc; Kriveq, ^oka lu Bala{; Mali Kriveq, kod
Tokom druge polovine IV milenijuma, verovat- vodenice; Bor, ^oka Kormaro{; [arbanovac, Se-
no pod pritiskom prodora stepskih plemena, zapo- li{te; Zlot, Lazareva pe}ina; Zlot, Verwikica;
~iwe naseqavawe nosilaca Kocofeni kulture na Bogovina, Bogovinska pe}ina i naseqe; Crnajka,
podru~je severoisto~ne Srbije.5 Ova teritorija je Pjatra Kosti; Kloko~evac, Kulmja [kjopuluji;
tako u{la u okvire kompleksa koji su ~inili delo-
vi Transilvanije, Banata, Oltenije i Muntenije
(karta 2).6 Na na{im prostorima se {iri iz pod-
ru~ja \erdapa do u{}a Mlave, na zapadu, i preko 4 Todorova 2007.
Ku~ajskih planina, Bora i Zaje~ara daqe ka jugu, 5 Bojad`ijev 1998.
do Ni{a. 6 Roman 1976.

66
KAPURAN. BULATOVI], Kulturna grupa KocofeniKostolac na teritoriji (6594) STARINAR LXII/2012

Majdanpek, Kqanc; Majdanpek, Jezero; Majdanpek, Treba ista}i kako prostorna analiza gradinskih,
Kapetanova pe}ina; Majdanpek, Rajkova pe}ina; Bo- visinskih i ravni~arskih naseqa kulturne grupe
qetin, Pe{}era Mare; Boqetin, Arija babiKo{o- KocofeniKostolac predstavqa primer pro`ima-
brdo; Boqetin, Velike livadice II; Boqetin, Le- wa, na prvi pogled paradoksalnog spoja, prirodnog
penska potkapina; Boqetin, Katarinine livade; okru`ewa i kulturnih uticaja.8 Geomorfologija
Boqetin Vlasac; Miro~, Pe}ina kod Trajanove ta- terena uslovila je da naseqa u Panoniji, Slavoniji,
ble; Turija, Stewe; Neresnica, Velika ~uka; Dobra, Sremu i u Banatu budu prete`no agrarnog karakte-
Manastir; Dobra, Padina; Malo Golubiwe, Re~ica; ra i dugotrajna.9 Izgled reqefa, nedostatak plod-
Miro~, Hajdu~ka vodenica; Gamzigrad, Abri ispod nog zemqi{ta, kao i drugi faktori uslovili su da
Bawske stene; Gamzigradska bawa, Wiva Z. Brzano- naseqa u brdovito-planinskom zale|u Dunava sa
vi}; Gamzigrad, Varzari; [tubik, Smiqkova glavi- srpske i sa rumunske strane karakteri{e dijame-
ca; [tubik, Seli{te; Jabukovac, Vratna Veliki tralno razli~iti koncept i ekonomija.10 U svom
most; [arkamen, Duge livade; Popovica, Veliko radu Cultura Cotofeni, Roman je izdvojio ~etiri
brdo; Brusnik, Glavica; Veqkovo, Kapu \aluluj; osnovna tipa naseqa: a) ravni~arska naseqa (du-
Vrkaq, ]eta}e; Mokrawske stene, Kamenolom; Mo- navska ostrva), b) naseqa na re~nim terasama, v)
krawske stene, Potkapina; Prahovo, Ide}e; Smedo- naseqa na te{ko pristupa~nim brdskim ili pla-
vac, GrabarSvra~ar; Sikole, Gradi{te; Karata{, ninskim terenima, g) naseqa u pe}inama.11
Dijana; Kladovo, Dowe Butorke; Kladovo, Brodo- Budu}i da je samo na teritoriji severoisto~ne
impeksnekropola; Mala Vrbica, Livade; Mala Srbije konstatovano oko 70 naseqa, od kojih je jedan
Vrbica, 500 m od sela; Korbovo, ZbradilaFunda; mawi broj i arheolo{ki sondiran, stvorili su se
Korbovo, Obala; Korbovo, GlamijaObala, Korbo- uslovi za preciznije definisawe odgovora vezanih
vo, VajugaPesak, Milutinovac, U{}e Jakomirskog za na~in `ivota ovih zajednica, ali i izvesne suge-
potoka, Velesnica, Biqevina; Qubi~evac, Obala, stije koje se ti~u samog trajawa i kulturne identi-
Qubi~evac, Ostrvo, Grabovnica, Brzi prun; Slati- fikacije populacija sa kraja bakarnog i po~etka
na, U{}e Slatinske reke; Kwepi{te, Mihajlovac; bronzanog doba. Nakon {to su svi lokaliteti izno-
Mihajlovac, Ru`ewka; Kusjak, Bor|ej; Kusjak, Mo- va rekognoscirani i georeferencirani,12 smatrali
tel; Kusjak, Vrkaq; Majdanpek, Kameni rog; Kwa- smo, uzimaju}i u obzir okolni reqef, hidrografi-
`evac, Dubrava; Rgo{te, Bolvan; Dowe Zuni}e, ju, geomorfologiju terena i karakter nalaza kera-
AxijskoVinsko; Mokrawe, Laluw; Tanda, La Tu- mi~ke produkcije, da postoji potreba da se dodatno
fek; Brestovac, Brestova~ka bawa; Dowi Milano- analiziraju pojedini aspekti kulturnog i prostor-
vac, Veliki Gradac; Dowa Bela reka; Rudna glava, nog razvoja, i to ne samo naseqa ve} i same kultur-
[eta}e; Rgo{te, Rosuqavi{war; Dobra, Gospo|in ne manifestacije KocofeniKostolac generalno.
vir; Kusjak, Grle.7 Veliku te{ko}u i daqe predstavqa relativno ma-
U ovom radu uzeti su u obzir i lokaliteti u li broj sistematski istra`enih lokaliteta, kao i
{iroj zoni u{}a Ni{ave u Ju`nu Moravu, budu}i nedostatak paleoosteolo{kih analiza faune.
da se revizionim istra`ivawima tokom posledwih Jo{ je tokom ranih 70-ih godina pro{log veka
godina do{lo do novih nalaza koji ukazuju da je i u konstatovano postojawe lokaliteta pozicionira-
okviru wih postojala istovremena kerami~ka pro- nih na te{ko pristupa~nom terenu, koji se u poje-
dukcija. Wih predstavqaju lokaliteti u neposred- dinim slu~ajevima nalaze na nagibu terena od 45,
nom okru`ewu Ni{a: BubawNovo selo, HumVeli-
ka humska ~uka i Dowa Vre`ina^ardak.
7 Kapuran 2011.
8 Fletcher 1977.
TOPOGRAFIJA NASEQA 9 Tasi} 1976, 107.
10 Tasi} 1976, 107; Spasi} 2010.
11 Roman 1976, 1516.
Za topografiju naseqa KocofeniKostolac
12 Kapuran 2011. Reviziono rekognoscirawe su obavili
grupe na prostoru severoisto~ne Srbije, srpskog
A. Kapuran i A. Bulatovi}, uz veliku pomo} I. Jovanovi}a i
Podunavqa i wegovog zale|a, karakteristi~na je
M. Jov~i}, kustosa Muzeja rudarstva i metalurgije u Boru, za-
raznolikost u pozicijama (nadmorskoj visini i tim B. Iliji}, kustosa Zavi~ajnog muzeja u Kwa`evcu, kao i G.
reqefu), tipovima stani{ta i na~inu ekonomije. Jawi}a, kustosa Muzeja Krajine u Negotinu.

67
KAPURAN. BULATOVI], Kulturna grupa KocofeniKostolac na teritoriji (6594) STARINAR LXII/2012

Karta 2. Distribucija lokaliteta


Kocofeni grupe
Map 2. Distribution of the Cotofeni group sites

kao {to je slu~aj sa lokalitetima Kulmja [kjopu- skog zale|a \erdapa, nego i za jugozapadnu Rumuni-
luji u Kloko~evcu i Pjatra Kosti u Crnajki (T. I / ju,15 gde je sam izgled morfologije terena identi-
12; karta 1/9).13 Oni su definisani kao gradinska ~an reqefu Timo~ke Krajine. Navedeni lokaliteti
naseqa, pa se, mada arheolo{ki nije dokazano po- imaju jo{ nekoliko zajedni~kih elemenata, od ko-
stojawe fortifikacije, ne iskqu~uje da je mogla po- jih je najva`niji taj da je svako od ovih uzvi{ewa
stojati na lako pristupnim stranama. U me|uvreme- pozicionirano na stenovitom vrhu kawona, na me-
nu se broj lokaliteta ovog tipa pove}ao na devet. stima gde se mawe reke ili potoci ulivaju u neku
Osim pomenutih, u ovu grupu spadaju i Vratna, Ve- ve}u reku. Pretpostavqa se da je izbor ovakvih po-
liki most (T. I/7; karta 1/33), Bogovina, Iznad pe- zicija bio prevashodno strate{kog zna~aja, budu-
}ine (T. I/4; karta 1/8), Jezero (T. I/3; karta 1/12), }i da kroz planinski predeo severoisto~ne Srbije
Kqanc (T. I/3; karta 1/11), Turija, Stewe (T. I/6; sistem re~nih tokova i dolina predstavqa najpo-
karta 1/22), Mokrawske stene, Kamenolom (T. I/5; godniji vid komunikacije od praistorije do dana-
karta 1/39) i Bolvan (T. I/8; karta 1/66).14 U Ni{- {wih dana. Preko sistema dolina Pore~ke reke
kom poqu, ovom tipu naseqa odgovarala bi Humska (Kulmja [kjopuluji i Crnajka), Urovice i Timoka
~uka (karta 1/73). Zajedni~ka karakteristika ovih
naseqa je da imaju identi~an koncept organizacije
na nepristupa~nim uzvi{ewima, koja se nalaze u ne-
posrednoj blizini izvora vode ili vodenih tokova 13 Tasi} 1982.
i pe}inskih formacija kra{kog porekla. Ovakva 14 [qivar, Jacanovi} 1987; Nikoli} 1997; Kapuran 2011.
naseqa nisu karakteristi~na samo za prostor srp- 15 Roman 1976, Pl. 1.

68
KAPURAN. BULATOVI], Kulturna grupa KocofeniKostolac na teritoriji (6594) STARINAR LXII/2012

(Vratna i Mokrawske stene) najlak{e se stizalo Paskova pe}ina. Arheolo{ka iskopavawa izvede-
do obala Dunava kao glavne transverzale, koja se~e na su jedino na lokalitetu Jezero, i na ulazima u
Balkansko poluostrvo u pravcu od istoka ka zapadu. Rajkovu pe}inu i Kapetanovu pe}inu.20 Dok je pro-
Drugu zajedni~ku karakteristiku ovih naseqa stor ispred ulaza u Rajkovu pe}inu pokazivao veo-
predstavqali bi stenoviti masivi na ~ijoj podlo- ma siroma{an kulturni sadr`aj, Kapetanova pe}i-
zi su formirana naseqa. Stenska podloga je u naj- na pokazuje kulturnu stratigrafiju od preko 3 m
ve}em broju slu~ajeva kre~wa~kog porekla i samim visine, {to predstavqa redak slu~aj za nalazi{ta
tim je pogodna za ve{ta~ko nivelisawe platformi KocofeniKostolac grupe. Ovaj podatak ne govo-
u vidu terasa na kojima se mogu podizati nadzemne ri samo o wenoj dugotrajnoj upotrebi, ve} nam pru-
konstrukcije ku}a, najverovatnije drvenih koliba `a mogu}nost za odgovore na pitawa o nastanku i
oblepqivanih blatom.16 Takav je slu~aj sa nase- trajawu ove kulturne manifestacije na prostoru
qima Jezero, Kulmja [kjopuluji, Pjatra Kosti, srpskog zale|a \erdapa. Budu}i da je sa~uvan od
Vratna i Bogovina (na ostalim nije mogu}e doneti uticaja erozije, ovaj lokalitet je pogodan za dobija-
ovakve zakqu~ke budu}i da nisu arheolo{ki son- we datuma po~etka i prestanka wene upotrebe, po-
dirana ili su devastirana erozijom i izgradwom {to je otkriveni arheolo{ki materijal kulturno
kasnijih naseqa). homogen.21 Iako nismo bili u mogu}nosti da izvr-
Tre}a zajedni~ka karakteristika je to da se u {imo detaqna rekognoscirawa, prema izgledu re-
neposrednoj blizini ovih naseqa nalazi jedna ili qefa i kre~wa~ke litice koja se nalazi iznad ula-
vi{e pe}ina. Oblast severoisto~ne Srbije u svom za, mo`e se pretpostaviti da se i iznad Rajkove
centralnom delu le`i na ve}em masivu kra{kog pe}ine nalazilo naseqe sa koga je materijal ero-
porekla, u kome se u proseku nalazi najve}i broj zijom stigao do ulaza u pe}inu. Isto tako se pretpo-
pe}ina u Evropi. Kao prirodna stani{ta ove for- stavqa da se u podno`ju Kapetanove pe}ine, koja
macije su pogodovale i za sme{taj qudi, ali i za ima oblik amfiteatra oivi~enog stenovitim zi-
~uvawe stoke. Na teritoriji Srbije jedino je kom- dovima, tako|e moglo nalaziti naseqe identi~no
pletno istra`ena Lazareva ili Zlotska pe}ina naseqima Kulmja [kjopuluji i Pjatra Kosti.
(karta 1/6), u kojoj kulturna stratigrafija pokazuje Ve}i broj pe}ina i potkapina nalazi se u pod-
antropogene aktivnosti tokom vi{e hiqada godi- no`ju lokaliteta Vratna Veliki most (T. I/7)22 i
na.17 Istra`ivawa kultura iz poznog eneolita u kar- Bolvan (T. I/8), dok se u okolini Bogovine i Mo-
patskom i balkanskom podru~ju pokazuju da pe}ine krawskih stena nalazila samo po jedna pe}ina, {to
predstavqaju veoma ~est vid stani{ta.18 Wihova ukazuje da su pe}ine i gradine ipak bile na neki
primarna uloga, me|utim, jo{ uvek nije potpuno na~i povezane. Mogu}e je da su u wima i qudi i
jasna da li su slu`ile za sme{taj stanovni{tva, stoka mogli da borave u isto vreme, naro~ito u pe-
~uvawe stoke ili i za jedno i za drugo. U slu~aje- riodu velikih ki{a, snegova kao i velikih vru}i-
vima pe}ina koje se nalaze u neposrednoj blizini na. Veliki broj pe}inskih naseqa otkriven je u
pomenutih gradinskih naseqa mo`e se pretposta- podunavskom delu jugozapadne Rumunije, a najpo-
viti da se prvenstveno radilo o mestima gde se ~uva- znatija su Pe{}era Hocilor,23 Pe{}era ku Apa,
la stoka. Zbog konstantne temperature tokom cele
godine, pe}ine predstavqaju idealno skloni{te u
veoma hladnim zimskim mesecima, budu}i da se
16 Tasi} 1982, 24.
lak{im intervencijama mogu adaptirati u zatvoren
17 Tasi} 1971.
stambeni prostor. Primer simbioze pe}inskih i 18 Roman 1971; Panaiotov, Aleksandrov 1988; Petrescu,
gradinskih KocofeniKostolac naseqa predsta- Popescu 1990; Kapuran 2011.
vqa okolina Zavojskog jezera kod Majdanpeka. Do 19 Nikoli} 1997; Kapuran 2011.

sada su na ovom prostoru konstatovana dva gradin- 20 Istra`ivawa je vodio Tonko Rajkova~a, kustos Muzeja

ska naseqa Jezero i Kqanc (T. I/3; karta 1/1112), rudarstva i metalurgije u Boru. Na`alost, do arheolo{ke do-
koja su podignuta na stenovitim kre~wa~kim od- kumentacije nije mogu}e do}i.
21 Autori ovog rada su imali detaqan uvid u pokretne arhe-
se~cima iznad reke Mali Pek.19 U wihovoj nepo-
olo{ke nalaze otkrivene na ovom lokalitetu.
srednoj okolini se nalaze Rajkova pe}ina (karta 22 Kapuran 2011; Mihailovi}, \uri~i}, Kalu|erovi}
1/14) i Kapetanova pe}ina (karta 1/13), u kojima su 1997, 35.
otkriveni tragovi antropogenih aktivnosti, kao i 23 Roman 1971; Roman 1976.

69
KAPURAN. BULATOVI], Kulturna grupa KocofeniKostolac na teritoriji (6594) STARINAR LXII/2012

Pe{}era Vasii u Biharu, Romawe{ti, ]ele Turzii, ova dva lokaliteta se pretpostavqa da su predstav-
pe}ina Sarva{, Baja de Fier i dr.24 qala zimovnike ili punktove za okupqawe stada i
Slede}u zajedni~ku karakteristiku navedenih pastira tokom pauza izme|u sezonskih migracija.
gradinskih naseqa predstavqa wihov dominantan Ona poseduju karakteristi~no prirodno okru`ewe,
polo`aj u okviru pejza`a. Budu}i da su wihove po- koje ~ine mawa brda i doline reka, kao i relativ-
zicije i izgled lako uo~qivi iz velike daqine, no male nadmorske visine na kojima su pozicioni-
ona verovatno nisu slu`ila za skrivawe, ve} da se rana. Takve sezonske stanice ili stacionari, u
naglasi wihovo prisustvo. Pozicije na stenovi- kojima se tokom zimskih meseci mogu okupiti ve-
tim uzvi{ewima iznad u{}a reka, na izlazima iz }e grupe pastira sa svojim stadima, predstavqale
kawona, sa ku}ama nanizanim na terasama strmih su va`na mesta u `ivotima pastirskih zajednica.
kre~wa~kih litica, sugeri{u nam kako su stanov- Cviji} navodi da mnogi sto~arski narodi, koji se
nici imali nameru da na taj na~in naglase svoj cikli~no kre}u iz visinskih ka nizijskim pod-
identitet. Mogu}e je da su na ovaj na~in pastirske ru~jima i zimovnicima, na odre|enim mestima us-
zajednice htele da daju do znawa novoprido{lim postavqaju i stalna naseqa.29 Ovome u prilog ide
zajednicama kako imaju kontrolu nad planinskim i ~iwenica da se u neposrednoj blizini Smiqko-
prelazima i putevima, naro~ito na onim mestima ve glavice nalazi planinski masiv Deli Jovan,
gde reke izlaze iz uskih kawona u doline Crnog koji, sa svojom nadmorskom visinom od 1500 m, po-
Timoka i Belog Timoka, Peka i Dunava. goduje za ispa{u stada tokom letwih meseci, dok
Drugi tip KocofeniKostolac naseqa na pro- se Velika ~uka nalazi u samom podno`ju Ku~ajskog
storu severoisto~ne Srbije predstavqaju naseqa planinskog masiva. Raspored lokaliteta Kocofe-
pozicionirana na vrhovima mawih brda ili blagih niKostolac grupe pru`a se konvergentnim prav-
kosina, ~ija se nadmorska visina u proseku kre}e cem, kroz oblasti izrazito karstne geolo{ke pod-
izme|u 336 m i 210 m. Ovoj grupi bi pripadalo i, loge. U takvom prirodnom okru`ewu, prilikom
za sada jedino (arheolo{ki potvr|eno), Kocofeni su{nih perioda ki{a zemqi{te slabo navodwava,
Kostolac utvr|eno gradinsko naseqe sa elementima budu}i da voda brzo oti~e kroz pukotine i podzem-
fortifikacije ^oka lu Bala{ kod Kriveqa (kar- ne kanale i reke presu{uju, te nema uslova za na-
ta 1/3).25 Pozicionirano je na jezi~astom uzvi{ewu vodwavawe i napajawe stoke, {to ~ini da vege-
okru`enom strmim stranama, koje se spu{taju do tacija sagori, a stanovni{tvo po~iwe da migrira
u{}a jednog maweg potoka u Kriveqski potok. Sa u druge krajeve.30 Ukoliko imamo u vidu ovu Cvi-
isto~ne pristupne strane nalazila se fortifika- ji}evu konstataciju, koja se ti~e prostora gde se
cija u vidu odbrambenog rova. Postoji velika vero- nalazi najve}i broj naseqa KocofeniKostolac
vatno}a da je i lokalitet Kapu \ajuluj u Veqkovu, grupe, vide}emo da odre|ena pravila o sezonskim
koji ima identi~ne elemente gradine, mogao biti za- migracijama traju i do dana{wih dana. U toplom
{ti}en odbrambenim rovom, budu}i da ima sli~ne letwem periodu napu{taju se zimska stani{ta zbog
topografske karakteristike kao ^oka lu Bala{.26 migracija u planinske predele, na kojima se nalaze
Naseqima ovog tipa treba dodati jo{ tri lo- pogodna mesta za ispa{u stoke. Odre|ene grupe pa-
kaliteta velikih gabarita sa elementima gradine. stira se tokom jeseni na putu ka ni`im predelima
To su lokaliteti Velika ~uka u Neresnici (karta i re~nim terasama vra}aju u ova naseqa, dok druge
1/23), Smiqkova glavica u [tubiku (karta 1/31) i
]eta}e u Kovilovu (karta 1/38).27 Sva tri lokalite-
ta nalaze se na {irokim i zaravwenim uzdignutim
platoima, koji dominiraju nad re~nim dolinama. 24 Tasi} 1979b, 119.
Me|u wima je arheolo{ki sondiran jedino lokali- 25 Tasi} 1982; Tasi} 1990.
tet ]eta}e u Kovilovu, a, prema re~ima autora, na 26 Trbuhovi}, Vukovi} 1967. Autori istra`ivawa pomi-

wemu su otkrivene nadzemne ku}e sli~ne onima u wu ovu mogu}nost, ali ona nije propra}ena odgovaraju}om do-
kumentacijom, Kapuran 2011, Kat. 091.
Crnajki i Kloko~evcu.28 Na lokalitetima Velika 27 Trbuhovi}, Vukovi} 1967; Kapuran 2011, Kat. 052, 083
~uka i Smiqkova glavica, osim keramike, na povr- i 089.
{ini se mo`e prikupiti dosta fragmenata lepa sa 28 Trbuhovi}, Vukovi} 1967; Tasi} 1982.

tragovima pletera. Sude}i prema velikoj povr{i- 29 Cviji} 1987, 91.

ni koju zauzimaju i prema poziciji i okru`ewu, za 30 Cviji} 1987, 157.

70
KAPURAN. BULATOVI], Kulturna grupa KocofeniKostolac na teritoriji (6594) STARINAR LXII/2012

grupe verovatno nastavqaju put ka centrima za }e kostola~ke kulture na Gomolavi i Kocofeni


okupqawe u dolinama oko Dunava i Timoka. Kostolac grupe na Bubwu i Bor|eju kod Kusjaka.
Za ostale visinske lokalitete, kao {to su Gra- Bor|ej predstavqa jedini lokalitet koji je is-
barSvra~ar u Smedovcu (karta 1/42), Arija Babi tra`ivan tokom projekata \erdap I i II, na kome je
Ko{obrdo (karta 1/16), ^oka Kormaro{ (karta 1/4), konstatovana nadzemna stambena arhitektura.35
Wiva Zore Brzanovi} (karta 1/29) i Varzari (kar- Otkrivena podnica ima pravougaonu osnovu, ~ija
ta 1/30),31 zajedni~ke karakteristike predstavqaju je otkrivena du`ina 8 m, sa ve}im brojem kerami~-
pribli`no iste nadmorske visine i topografske kih ulomaka na povr{ini.36 Budu}i da nisu kon-
pozicije koje se nalaze neposredno pored reka ili statovani tragovi vertikalnih stubova, mo`e se
jakih izvora vode. Radi se o vi{eslojnim lokali- pretpostaviti da je iznad ovakve podnice mogla da
tetima,32 {to zna~i da je dobra insolacija zajedno se podigne ku}a tipa brvnare, sa horizontalno po-
sa dobrim pregledom okolinog peja`a, pogodovala stavqenim oblicama.
dugotrajnijoj okupaciji ovih prostora. Visinska Drugi primer stambene arhitekture poti~e sa
naseqa ovog tipa obi~no se vezuju za predele u koji- prostora jugoisto~ne Srbije a ti~e se ku}e otkrive-
ma dominiraju pa{waci i gde nema ve}ih {uma. ne u sondi 1/0809, koja pripada KocofeniKosto-
Posledwem tipu naseqa KocofeniKostolac lac horizontu naseqa Bubaw kod Ni{a.37 Otkrive-
grupe pripadali bi lokaliteti koji imaju karakte- ni deo podnice ovog objekta je tako|e pravougaone
ristike ravni~arskih naseqa a pozicionirani su osnove i prili~no velikih dimenzija. Na woj je
na re~nim terasama. Lokaliteti na desnoj obali konstatovan ve}i broj jama od ko~eva i tawih stu-
Dunava oko Kowske glave, kao {to su Kladovo bova za konstrukciju vertikalnih zidova.38 Sli~-
Brodoimpeks (karta 1/46), Mala Vrbica (karta 1/47), na arhitektura je konstatovana na lokalitetu Go-
ZbradilaFund (karta 1/49), KorbovoObala (kar- molava kod Hrtkovaca, gde su stambeni objekti
ta 1/50), VajugaPesak (karta 1/52), U{}e Jakomir- ukazivali na izrazito sedentarni, zemqoradni~ki
skog potoka (karta 1/53), Velesnica (karta 1/54), karakter ovog naseqa.39
Qubi~evacObala (karta 1/55), Qubi~evacOstrvo
(karta 1/56), Brzi prun (karta 1/57), U{}e Slatinske
reke (karta 1/58), Kwepi{te (karta 1/59), Ru`ewka
31 Kapuran 2011, Kat. 043, 011, 063, 067; Kapuran, Bori},
(karta 1/60), KusjakBor|ej (karta 1/61), Kusjak
Jevti} 2007; Bori}, Starovi} 2006.
Motel (karta 1/62), KusjakVrkaq (karta 1/63), za- 32 Na Arija Babi konstatovana je kulturna stratigrafi-
hvaquju}i starim kartama izra|enim pre nastanka ja iz perioda eneolita KostolacKocofeni grupe, sredweg
akumulacionih jezera, izgleda da su predstavqali bronzanog i starijeg gvozdenog doba (Kapuran, Bori} i Jevti}
punktove na kojima su stada mogla du`e da se zadr- 2007). Naseqe ^oka Kormaro{ nastalo je u periodu poznog
eneolita Kocofeni grupe, a nastavilo je da postoji i tokom
`avaju ~ekaju}i pogodan period za prelazak na dru- sredweg bronzanog doba. Naseqe na Wivi Zore Brzanovi} na-
gu stranu reke. U neposrednom okru`ewu ovih lo- stalo je u periodu poznog eneolita, da bi nastavilo da postoji
kaliteta postojale su pe{~ane ade, mawa ostrva i tokom sredweg bronzanog doba, i kasnije tokom starijeg gvozde-
sprudovi povezani sa obalom, {to ukazuje na po- nog doba (Kapuran i [kundri} 2009). Varzari je naseqe koje
prema topografskim karakteristikama najvi{e li~i na ^oka
stojawe pli}eg dna i gazova.33 Ovi delovi prioba- lu Bala{, samo {to je mnogo ve}ih dimenzija. Nastalo je u neo-
qa su prilikom ve}ih su{a ili o{trih zima, kada litu, da bi se `ivot nastavio tokom poznog eneolita, sredweg
se formira ledeni pokriva~, mogli predstavqati bronzanog doba i starijeg gvozdenog doba (Kapuran 2010).
33 Petovi} 1941, 8586.
mesta na kojima je bilo mogu}e lak{e prelaziti s
34 Tasi} 1983a.
jedne obale na drugu.34
35 Sladi} 1984, sl. 204.
36 Sladi} 1984.
37 Rad u kome }e biti prezentovani rezultati istra`iva-
ARHITEKTONSKI OSTACI wa je u pripremi.
38 Prema nalazima posledwih istra`iva~kih kampawa na

Rekonstrukcija stambenih objekata iz poznog lokalitetu Bubaw, podnicu ku}e je ~inio sloj zemqanog nabo-
eneolita nije mogu}a, budu}i da u ovom trenutku ja bez supstrukcije, sa primesom u vidu sitnijeg {qunka. Oko
podnice je konstatovan rov koji, mo`da, predstavqa temeqni
istra`ivawa ne raspola`emo sa dovoqno elemena- rov za konstrukciju zidova. Radovi u narednoj kampawi }e de-
ta za preciznije definisawe izgleda i tehnika iz- finisati konstrukciju i gabarit celog objekta.
gradwe ovih objekata. Do sada su dokumentovane ku- 39 Petrovi} 1988, fig. 1.

71
KAPURAN. BULATOVI], Kulturna grupa KocofeniKostolac na teritoriji (6594) STARINAR LXII/2012

Sasvim druga~iji tip arhitekture je preovlada- stoji, ali sve dok se stepen arheolo{ke istra`eno-
vao u naseqima brdsko-planinskih predela Ju`nih sti lokaliteta KocofeniKostolac (prvenstveno
Karpata. Autori istra`ivawa gradinskih lokali- sistematskih istra`ivawa i ozbiqnijih analiza
teta Kulmja [kjopuluji i Pjatra Kosti konstato- otkrivenih ostataka materijalne kulture i ostata-
vali su stambene objekte koji su svojim zadwim de- ka `ivotiwskih kostiju unuar naseqa) ne podigne
lom ukopavani u strmu padinu, {to je uslovilo da na zadovoqavaju}i nivo, preostaje nam da se osloni-
veoma brzo nakon napu{tawa budu uni{teni ero- mo na postoje}e podatke koji se ti~u nomadskog na-
zijom.40 Tako|e, otkrivena je i znatna koli~ina ~ina privre|ivawa i wihovih dru{tvenih odnosa.
ku}nog lepa, koja ukazuje na stabilnu arhitekturu Najve}i problem i daqe predstavqa nedostatak
odnosno trajna naseqa, a ne samo kratkotrajna sklo- apsolutnih datuma, koje neprestano pozajmquje-
ni{ta.41 Na gradinskim naseqima organizovanim mo iz radova na{ih suseda (iz Rumunije i Bugar-
u vidu terasa, ku}e su podizane iskqu~ivo na jugo- ske), {to ne mora da garantuje wihovu validnost i
zapadnim stranama uzvi{ewa, {to ukazuje na po- na ovim prostorima.
trebu za boqim osun~avawem objekata u nasequ
(verovatno tokom zimskih meseci). Istovremeno,
stenski masivi koji su se nalazili u zale|u ili na MATERIJALNA KULTURA
bokovima mogli su biti u funkciji odbrane od hlad- I PERIODIZACIJA
nih vetrova koji duvaju sa severa i severoistoka.
Do sada nije dovoqno jasan odnos koji su imale Do sada je u vi{e radova o ovoj temi konstato-
populacije poznog eneolita prema eksploataciji vano da je na lokalitetima u isto~noj Srbiji, pre-
rude bakra. Te{ko je poverovati u to da su ove za- ma stilsko-tipolo{kim odlikama keramike, mogu}e
jednice `ivele i da su se kretale u blizini veli- izdvojiti dve faze grupe Kocofeni: prvu u kojoj
kih povr{inskih depozita malahita, a da pri tome dominira ornamentalna tehnika urezivawa karak-
nisu razmi{qale o ekonomskoj koristi wegove eks- teristi~na za Kocofeni grupu, i mla|u fazu u
ploatacije, ali bi i to moglo potvrditi da je re~ o kojoj su, pored ovog stila, prisutni i tehnika braz-
populaciji koja nije imala mnogo veze sa prethod- dastog ubadawa, odnosno furhen{tih tehnika (fur-
nim stanovnicima ranog eneolita koji su se prete- chenstich), i neki drugi kostola~ki elementi (bra-
`no koncentrisali neposredno uz prirodne mine- zdasto urezivawe, neki tipovi keramike i dr.).44
ralne resurse. U dosada{wim istra`ivawima nije Taj podatak je doprineo da se i uvede novi naziv za
konstatovano da se tokom dominacije Kocofeni kulturnu manifestaciju KocofeniKostolac.
Kostolac kulturne grupe na teritoriji severoi- Na`alost, ~iwenica je da ve}ina lokaliteta
sto~ne Srbije obavqala metalur{ka aktivnost. To KocofeniKostolac grupe u isto~noj Srbiji nije
potvr|uje i nedostatak nalaza bakarnih predmeta istra`ena, ili je samo delimi~no istra`ena, i da
sa ove teritorije iz razvijenog i poznog eneolita, ne postoji konstatovana vertikalna stratigrafi-
{to nije slu~aj na teritoriji Oltenije, Transil- ja, odnosno utvr|eni stratigrafski odnos izme|u
vanije i severozapadne Bugarske.42 Nedostatak ba-
karnih nalaza iz ovog perioda u isto~noj Srbiji
mo`emo objasniti i na~inom `ivota i obi~ajima
nomadsko-sto~arskih zajednica, koje su zabele`i- 40 Trbuhovi}, Vukovi} 1967; Tasi} 1982.
li etnolozi. D. Antonijevi} navodi da je `ivot no- 41 A. Kapuran je prilikom rekognoscirawa lokaliteta
mada ekonomski i tehnolo{ki jednostavan i u Vratna, Most, na veoma nepristupa~nom terenu, sa velikim na-
suprotnosti sa socijalnom evolucijom, te da oni gibom, otkrio iznena|uju}e veliku koli~inu krupnog grumewa
ku}nog lepa. Na wemu su se nalazili tragovi oblica i pletera
svoju zajedni~ku svojinu reguli{u nepisanim obi- u negativu.
~ajnim pravom, preziru zemqoradwu, a okupqeni 42 S. Aleksandrov (Alexandrov 1995, 257) navodi na nala-

su oko porodi~nih zadruga i imovinske potrebe zi{tima Kocofeni grupe u severozapadnoj Bugarskoj brojne
svode samo na ono {to se mo`e sa sobom poneti.43 predmete od arsenske bronze, izme|u ostalog i sekire tipa Ba-
niabik (Baniabic), dok A. Vulpe (Vulpe 1970, 2731) predstavqa
Tvrdwa da se ne treba ~vrsto dr`ati etnolo{-
Baniabik sekire iz jugozapadne Rumunije kao inventar Glina III
kih zakqu~aka koji su dobijeni izu~avawem sto~ar- grupe, a sekire tipa Fajs (Fajsz) vezuje za Kocofeni grupu.
skih zajednica bliskih dana{wem dobu, i apriori 43 Antonijevi} 1982.

ih primewivati na praistorijske zajednice tako|e 44 Tasi} 1979, 117; Nikoli} 1997, 205.

72
KAPURAN. BULATOVI], Kulturna grupa KocofeniKostolac na teritoriji (6594) STARINAR LXII/2012

stilsko-tipolo{kih osobina keramike starije Sli~ne posude evidentirane su u IV sloju naseqa


(Kocofeni) i mla|e faze (KocofeniKostolac). Sitagroi u severnoj Gr~koj.48
Uglavnom je re~ o naseqima na kojima je konstatova- ^esto zastupqen tip posude predstavqa polu-
na keramika sa elementima i kostola~ke i Kocofe- loptasta zdela dubqeg recipijenta, ravnog oboda
ni grupe, ili iskqu~ivo sa elementima Kocofeni (sl. 11, 12, 23, 27, 29, 5254, 57, 5960, 74, 79, 81, 82,
grupe, dok naseqa sa samo kostola~kom keramikom 90, 91, 95, 113, 124, 125, 131 i 145) ili pli}eg reci-
nisu evidentirana. Kod nas je, dakle, za sada mogu}- pijenta, koso zase~enog trouglasto ili T profili-
no jedino podeliti lokalitete na one na kojima je sanog oboda (14, 19, 133 i 146). Ove posude karakte-
evidentiran ornament brazdastog ubadawa i na one risti~ne su za obe faze, jer su, osim vertikalnim
gde, taj ornament jo{ uvek nije evidentiran. Ipak, rebrima, urezima i zarezima, ukra{ene i brazda-
neizvesno je u kolikoj meri se ta podela mo`e pri- stim ubadawem (sl. 23, 68, 81 i 82). Varijanta dubqe
meniti i prilikom periodizacije ove grupe, jer se zdele ravnog oboda se javqa u svim fazama Kocofe-
~ini da je zapravo prouzrokovana naseqavawem ni grupe u Rumuniji, dok je pli}a zdela sa T profi-
nosilaca razli~itih kulturnih manifestacija u lisanim ili trouglasto profilisanim obodom ka-
pojedine regione isto~ne Srbije kostola~kom, sa rakteristi~na za najmla|u fazu Kocofeni grupe.49
zapada, odnosno Kocofeni grupe, sa istoka.45 I ove posude se, osim u Rumuniji i na lokaliteti-
U Rumuniji je, me|utim, na nekoliko lokalite- ma KocofeniKostolac grupe u isto~noj Srbiji,
ta otkrivena vertikalna stratigrafija, na kojima javqaju i u Bugarskoj, ali i Pomoravqu (Makre{a-
je uo~en odnos razvojnih etapa Kocofeni grupe, pa ne, Mal~a, Vrti{te, Bubaw, Bobi{te) i severnoj
se na osnovu stratigrafije na tim lokalitetima, Gr~koj.50 Zanimqiv je podatak da je u Pelagoniji, na
uz izvestan oprez, mogu doneti izvesni zakqu~ci i lokalitetu Tre{tena stena, konstatovana kerami-
o razvoju KocofeniKostolac grupe i u isto~noj ka sli~nog tipa, ukra{ena brazdastim ubadawem.51
Srbiji.
Starijoj fazi (Kocofeni grupa) pripadali bi,
dakle, lokaliteti na kojima nema ornamenata iz-
45 Takve periodizacije, koje se zasnivaju na nedostatku
vedenih tehnikom brazdastog ubadawa i na kojima je
odre|enog elementa, nezahvalne su i nesigurne, jer se u ovom
konstatovana keramika karakteristi~na za Koco- slu~aju uglavnom raspola`e povr{inskim nalazima i taj orna-
feni grupu, iako su, va`no je napomenuti, to obli- ment se prilikom eventualnih istra`ivawa na lokalitetima
ci i ornamenti koji se uglavnom javqaju uz brazda- gde nije konstatovan mo`e naknadno pojaviti, pa mnoge teorije
sto ubadawe i u mla|oj fazi grupe. zasnovane na takvim ~iwenicama postaju nevalidne. Takav slu-
~aj je konstatovan na Dowim Butorkama, koje su do sada va`ile
Najzastupqeniji tip posude na lokalitetima u za lokalitet starije faze (Kocofeni grupa) (Tasi} 1979, 117;
isto~noj Srbiji predstavqenim u ovom radu jeste Nikoli} 1997, 200), ali je prilikom autopsije nalaza u Muzeju
amfora dugog levkastog vrata, koja se javqa na loka- Krajine u Negotinu na dva fragmenta sa ovog lokaliteta uo~en
ornament brazdastog ubadawa (sl. 81, 82), karakteristi~an za
litetima iz obe faze ove grupe, sa jezi~astom, po-
kostola~ku grupu. Prema dosada{wim rezultatima, hronolo-
nekad vertikalno perforiranom dr{kom (sl. 6, 9, gija grupe bi se najpouzdanije mogla definisati analizom di-
20, 21, 51, 63, 100, 126, 134), kao i tunelastom ili stribucije naseqa i stilsko-tipolo{kim osobinama keramike
potkovi~astom dr{kom ispod oboda (sl. 88, 115). koja ih prati najstariji lokaliteti (Kocofeni grupa) jesu u
Negotinskoj Krajini i Kqu~u, ne{to mla|i (KocofeniKo-
Ove posude su obi~no ukra{ene ispod oboda ureza- stolac grupa) sme{teni su zapadno od te teritorije, dok su lo-
nim linijama ili utisnutim du`im zarezima (sl. kaliteti u ju`nom Pomoravqu najmla|i i neposredno pretho-
6, 21, 38, 64, 71, 89, 98100, 104, 109, 115, 116, 134), de grupi BubawHum II.
46 Roman 1976, pl. 54/1, 13, pl. 71/13, pl. 92/2, 4, 5, pl. 93/4;
utisnutim motivom ribqe kosti (sl. 4, 28), a u mla-
Panaotov, Aleksandrov 1988, obr. 6/g, 7/g.
|oj fazi i brazdastim ubadawem ili ta~kastim 47 Stoji}, ^a|enovi} 2006, T. LXVII/6567; Stoji}, Jo-
ubodima (sl. 9, 20, 25, 140). Jezi~aste dr{ke su na ci} 2006, T. LIII/27, T. CI/45; Bulatovi}, Jovi} 2010, T. XX/43.
nekim primercima amfora ukra{ene otiscima vrha 48 Renfrew, Gimbutas, Elster 1986, fig. 13.7/4.

prsta ili zarezima (sl. 20, 21, 35, 100, 104, 105). Taj 49 Roman 1976, pl. 53/9, 11, pl. 66/10, 12, pl. 90/3, pl. 110/6,

tip posuda karakteristi~an je za sve faze Koco- pl. 90/9, pl. 99/2, 9, pl. 111/4, pl. 112/1.
50 Panaotov, Aleksandrov 1988, obr. 5/v; Stoji}, ^a|e-
feni kulture u Rumuniji i severozapadnoj Bugar-
novi} 2006, T. LIX/1315; Stoji}, Joci} 2006, T. XV/37, T.
skoj,46 ali i za lokalitete u Pomoravqu, naro~ito LXXII/149, 155; T. LIII/19, 20; Bulatovi}, Jovi} 2010, T. XIX/41;
u basenu Ju`ne Morave (Makre{ane kod Kru{evca, Renfrew, Gimbutas, Elster 1986, fig. 13.12/1, 2.
Mal~a i Hum kod Ni{a, Bobi{te kod Leskovca).47 51 Mitkoski 2010, T. VII.

73
KAPURAN. BULATOVI], Kulturna grupa KocofeniKostolac na teritoriji (6594) STARINAR LXII/2012

Osim te keramike, na Tre{tenoj steni konstatova- Dr{ke posuda su prete`no trakaste lu~nog pre-
ni su i fragmenti pravougaonih posuda, a iden- seka (sl. 66, 80, 97, 103, 118), ili su kolenasto savi-
ti~ne pravougaone posude na|ene su u kostola~kom jene (sl. 120, 138), ali postoje i jezi~aste vertikal-
nasequ na Gomolavi.52 no perforirane, pravougaone, vertikalne
Sli~ne prethodnom tipu su poluloptaste zdele tunelaste, o kojima je ve} bilo re~i. Dr{ke su
koje, zbog uvu~enog oboda, imaju skoro loptast oblik. ukra{ene lincen aplikama, urezima, kosim ubodi-
Ove zdele mogu biti ukra{ene vertikalnim rebri- ma ili zarezima.
ma na obodu (sl. 148) u kombinaciji sa ubodima (sl. Pored navedenih tipova posuda, za Kocofeni
17), zarezima (sl. 61, 84, 85) ili urezanim linija- Kostolac grupu karakteristi~ne su {oqe zaobqe-
ma (sl. 132). I ove posude su evidentirane ve} u nog dna, sa dr{kom koja visoko prelazi obod (sl.
prvoj fazi, a karakteristi~ne su za drugu i tre}u 144), i posude ovalne osnove sa izlivnikom, tzv.
fazu Kocofeni grupe na rumunskim nalazi{tima, sosijere, koje nisu konstatovane prilikom atribu-
i javqaju se i na lokalitetima u Pomoravqu kao i cije gra|e sa lokaliteta u isto~noj Srbiji za ovaj
u zapadnoj Bugarskoj.53 rad, ali su otkrivene prilikom ranijih istra`i-
Jedan od dominantnih tipova posuda na nala- vawa.57 U ovu grupu spadaju i posude (kr~azi) ve}ih
zi{tima KocofeniKostolac grupe u isto~noj dimenzija, sa {irokim trakastim i o{tro savije-
Srbiji jeste zdela poluloptastog ili loptastog nim dr{kama.58 Svi navedeni tipovi karakteri-
recipijenta sa levkastim vratom, odnosno razgr- sti~ni su i za rumunska nalazi{ta.59 Kada je re~ o
nutim obodom (sl. 5, 13, 15, 18, 24, 40, 76, 83, 98, sosijerama, prime}uje se da su te posude iz zapad-
101 i 139). Primerci sa {iroko razgrnutim obo- ne Bugarske i Rumunije malo druga~ije od prime-
dom su obi~no neukra{eni, a samo jedan je ukra{en raka na|enih na lokalitetima u Srbiji (Gladnice,
mre`astim ornamentom (sl. 40), dok su zdele lev- Zlotska pe}ina). Naime, primerci iz Rumunije ima-
kastog vrata ukra{ene ve}inom tehnikom brazda- ju nagla{en izlivnik i dr{ku koja prelazi obod, a
stog ubadawa (sl. 13, 15, 139). Na rumunskim nala- na{i primerci imaju neznatno profilisan obod i
zi{tima se obe varijante javqaju od najstarije mawe tunelaste dr{ke.60 Posuda iz zapadne Bugar-
faze, ali traju i u mla|im fazama.54 Konstatova- ske bi mogla da predstavqa me|utip, sa mawe nagla-
ne su i na nalazi{tima u severozapadnoj Bugarskoj {enim izlivnikom i dr{kom koja neznatno prelazi
i Pomoravqu.55 obod.61 Prema navedenom, varijanta sosijere kon-
Mawe zastupqene posude na lokalitetima u statovana na na{im prostorima mogla bi biti ka-
isto~noj Srbiji jesu bikoni~ni ili loptasti pe- rakteristi~na iskqu~ivo za nalazi{ta u isto~noj
hari, zatim kru{koliki pehari sa jednom dr{kom, i ju`noj Srbiji.
ve}e kru{kolike amfore dugog cilindri~nog ili Ornamentalne tehnike kori{}ene za ukra{a-
koni~nog vrata, sa malim dr{kama ispod oboda, vawe keramike ove grupe uglavnom su pomenuti u
ukra{ene nizom zareza (sl. 39, 86), kao i ba~vasti delu teksta u kojem se govori o tipovima posuda na
ili loptasti lonci ukra{eni urezima, horizon-
talnim trakama ili kru`nim otiscima (sl. 4547,
141, 142). Bikoni~ni i loptasti pehari imaju lu~ne 52 Petrovi}, Jovanovi} 2002, 268, sl. 18, 270, sl. 1.
dr{ke, trakastog preseka, ukra{ene lincen ukrasi- 53 Roman 1976, pl. 59/6, pl. 77/8, pl. 92/6; Stoji}, ^a|eno-
ma (so~ivaste aplike) i urezima (sl. 67), zatim pra- vi} 2006, T. LXX/92; Alexandrov 1995, fig. 7/92, 93.
54 Roman 1976, pl. 60/2, pl. 71/11.
vougaone, kao i jezi~aste horizontalno postavqene
55 Panaotov, Aleksandrov 1988, Obr. 7/b, v; Stoji}, ^a-
dr{ke (sl. 37, 130). Kru{koliki pehari sa jednom
|enovi} 2006, T. LXII/31, 32, 3436; Stoji}, Joci} 2006, T.
dr{kom su retki, a ukra{eni su vertikalnim `le- LXXII/152; Lazi} 2005, T. I/2.
bovima, pa podse}aju na badenske {oqe (sl. 92), 56 Lazi} 2005, T. I/1; Bulatovi}, Jovi} 2010, T. XIX/42, T.

ili urezima i ubodima (sl. 62). Na svim navedenim XX/44, 46, 48, T. XXI/49. Osim ove posude, u Kr`incu je evi-
tipovima keramike nije konstatovano brazdasto dentirana i {oqa {iqatog dna koja nije publikovana, a nala-
zi se u Narodnom muzeju u Vrawu.
ubadawe, pa bi se iz toga moglo zakqu~iti da oni 57 Tasi} 1979, T. XV/14; Nikoli} 1997, T. II/1, 18.
pripadaju starijoj fazi, odnosno Kocofeni grupi. 58 Tasi} 1979, T. XV/5.
Sli~ni tipovi posuda evidentirani su u Bobi{tu 59 Roman 1976, pl. 17, pl. 29, pl. 35.

u Leskova~kom poqu i u Kr`incu na ulazu u Grde- 60 Roman 1976, pl. 22.

li~ku klisuru.56 61 Alexandrov 1995, fig. 6/87.

74
KAPURAN. BULATOVI], Kulturna grupa KocofeniKostolac na teritoriji (6594) STARINAR LXII/2012

kojima se nalaze. Osim pomenutih ornamenata, od {qewa se, uglavnom, pribli`no podudaraju. Na ke-
kojih su naj~e{}i urezani motivi ribqe kosti, ramici sa rumunskih nalazi{ta iz najstarije faze
urezane linije koje se seku, lincen aplike, nizovi ove grupe primetni su jaki uticaji ^ernavoda III
zareza, brazdasto ubadawe, brazdasto urezivawe, grupe, kao i Badenske grupe, pa se smatra da su ove
ta~kasti ubodi i vertikalna rebra, konstatovani grupe u~estvovale u genezi Kocofeni grupe. Za
su ornamenti izvedeni vrpcom, odnosno {nur orna- hronologiju Kocofeni grupe najindikativnija je
menti (sl. 93, 94, 102, 106 i 135), zatim horizontal- stratigrafija konstatovana u pe}ini Hocilor u
ni i cik-cak nizovi ta~kastih uboda (sl. 17, 18, 22, Ju`nim Karpatima.65 Ispod sloja Kocofeni grupe
32, 62, 72, 111, 122, 144), ~esto u kombinaciji sa konstatovan je prazan sloj, a ispod wega kulturni
brazdastim ubadawem (sl. 13, 15, 26, 107, 118, 139, sloj Salkuca IV grupe. Iznad sloja Kocofeni grupe
140, 147), kao i polumese~asti, polukru`ni ili nalazio se prazan sloj (kratak prekid), pa kultur-
pravougaoni otisci (sl. 7, 42, 74, 78). ni sloj Verbi~oara grupe. Na drugim nalazi{tima
Svi ornamenti su karakteristi~ni za Kocofe- u Rumuniji se ispod sloja Kocofeni grupe nalazio
niKostolac grupu i nije uo~eno da je neki ornament sloj ^elei^ernavoda III, pa N. Tasi} smatra da
ili tip posude vezan za odre|enu regiju na terito- izme|u Kocofeni grupe i Salkuca IV grupe postoji
riji isto~ne Srbije. Jedino je mogu}no konstatova- vremenski interval koji ispuwava prodor nosila-
ti da se keramika ukra{ena brazdastim ubadawem ca ^ernavoda IIIBoleraz stila.66 To je potvr|eno
nalazi uglavnom na lokalitetima zapadno od Kqu- i na nekim nalazi{tima u Srbiji, ali samo u Bana-
~a i Negotinske Krajine, kao i ju`no, uz Timok, tu, dok je na lokalitetima u Pomoravqu, isto~noj
Beli Timok i Crni Timok, sve do Poni{avqa i ju- Srbiji i na Kosovu, prema Tasi}u, ispod sloja Koco-
`nog Pomoravqa. Ako se ovaj ornament iskqu~ivo feni ili KocofeniKostolac grupe konstatovan
vezuje za mla|u fazu, odnosno grupu Kocofeni sloj BubawSalkucaKrivodol kompleksa.67
Kostolac, iz toga proizilazi da i ovi lokaliteti Na lokalitetu Bubaw kod Ni{a, me|utim, uo-
pripadaju toj fazi, odnosno da je u jednom trenut- ~ena je ne{to druga~ija stratigrafija, koja se
ku do{lo do pomerawa ove populacije na jug.62 umnogome podudara sa onom u Rumuniji i Banatu.
Zanimqiva je keramika ukra{ena {nur orna- Neposredno iznad sloja BubawHum I grupe nalazi
mentima na nalazi{tima u isto~noj Srbiji, koja se kulturni sloj ^ernavoda III grupe, a iznad ovog
se na rumunskim nalazi{tima javqa u drugoj fazi sloja je konstatovana ku}a KocofeniKostolac
Kocofeni grupe, a najvi{e je zastupqena u wenoj grupe.68 Ku}u je o{tetila ve}a jama koja pripada
tre}oj fazi. Ovaj ornament, naime, konstatovan je BubawHum II grupi. Prema stratigrafiji sa Bub-
sporadi~no i daleko na jugu, na nalazi{tu Dikili wa, dakle, izme|u BubawHum I kulturnog sloja
Ta{ na severnoj obali Egejskog mora u nivou 6, koji,
prema autoru, hronolo{ki odgovara BubawHum II
grupi i kostola~koj grupi.63 [nur ornament je ve-
oma zastupqen u slojevima 73 u Ezeru, mada je uz 62 Ova pretpostavka je potkrepqena ~iwenicom da je na

druge elemente Kocofeni grupe (lincen aplike, podru~ju uzvodno od \erdapa i ju`no od Negotinske Krajine na
svim lokalitetima gde su vr{ena istra`ivawa na|ena kera-
mre`asti motiv, amfore levkastog vrata) na ovom
mika ukra{ena brazdastim ubadawem, a samo se lokaliteti sa
lokalitetu prisutan od najstarijih slojeva.64 Sa kojih poti~u slu~ajni nalazi, bez brazdastog ubadawa, smatraju
keramikom ukra{enom {nur ornamentima nije na- starijim. Pretpostavqa se da na ovoj teritoriji nema starijih
|ena keramika ukra{ena brazdastim ubadawem (Kocofeni) naseqa, jer je ta teritorija, naro~ito jugoisto~na
Srbija, u poznom eneolitu naseqena populacijom Kocofe-
(osim na lokalitetu GrabarSvra~ar), pa se weno
niKostolac grupe. To potvr|uje stratigrafija na lokalitetu
prisustvo na ve}ini lokaliteta u isto~noj Srbiji Bubaw konstatovana prethodnih godina.
mo`e vezati za stariju fazu. 63 Deshayes 1970, 43, fig. 27.
64 Georgiev et al. 1979, obr. 144/av, obr. 147, 152, 153,

155158, 163.
65 Roman 1976, 59.
RELATIVNA I APSOLUTNA
66 Tasi} 1979, 118.
HRONOLOGIJA 67 Tasi} 1979, 118.
68 Rezultati istra`ivawa lokaliteta Bubaw bi}e objavqe-
Hronologija Kocofeni grupe razmatrana je do ni po zavr{etku istra`ivawa. Za nalaze iz sloja ^ernavoda III
sada u mnogim nau~nim i stru~nim radovima i mi- videti u: Milanovi} 2011, 101113.

75
KAPURAN. BULATOVI], Kulturna grupa KocofeniKostolac na teritoriji (6594) STARINAR LXII/2012

(BubawSalkucaKrivodol kompleks) i ku}e koja U svakom slu~aju, prema stratigrafiji nekoliko


pripada KocofeniKostolac grupi nalazi se kul- navedenih lokaliteta u zapadnoj Bugarskoj, zatim u
turni sloj ^ernavoda III grupe, kao i na nalazi- Pomoravqu i ju`noj Rumuniji, mo`e se zakqu~iti da
{tima u Rumuniji, samo {to se na Bubwu iznad Ko- je Kocofeni grupi (severoisto~na Srbija i Rumuni-
cofeniKostolac sloja nalazio sloj BubawHum ja), odnosno KocofeniKostolac grupi (Pomorav-
II grupe, dok je u Rumuniji konstatovan sloj Glina qe i zapadna Bugarska), u svim navedenim regijama
III[nekenberg grupe. prethodila ^ernavoda III grupa. KocofeniKosto-
I u zapadnoj Bugarskoj konstatovana je sli~na lac grupu su nasledile vu~edolska kultura, odno-
situacija. Naime, na lokalitetu RadomirVahovo, sno BubawHum II grupa u Pomoravqu i dolini
u najstarijem sloju (nivo I) evidentirana je kerami- Strume, i Glina III[nekenberg grupa u Olteniji
ka koja pripada ^ernavoda IIIBoleraz grupi, dok i na podru~ju Transilvanije i Ju`nih Karpata.
kulturni sloj iznad (nivo II) pripada fazi I Koco- Apsolutni hronolo{ki okvir KocofeniKo-
feni grupe (OrleaSadovec).69 Slede}i nivo pripa- stolac grupe u Podunavqu i isto~noj Srbiji mo`e
da Kocofeni II fazi, ali autor bele`i sli~nost ke- se odrediti samo posredno, jer sa ovih lokaliteta
rami~kih nalaza iz tog sloja sa nalazima Kostolac nedostaju C14 datumi. Prema J. Bojaxijevu, faze
grupe.70 Za IV nivo Aleksandrov navodi da pripada IIIII Kocofeni grupe (44004300 bp) mogu se
varijanti rane Vu~edol kulture i da su nalazi para- opredeliti pribli`no istovremeno kada i Kosto-
lelni sa horizontom Ostrikovac Id u Pomoravqu. lac grupa (45004100 bp), pa se, pretpostavqaju}i
Dakle, i na ovom indikativnom lokalitetu u da KostolacKocofeni grupa nastaje bar neko
gorwem toku Strume konstatovana je sli~na strati- vreme nakon formirawa Kostolac grupe, mo`e za-
grafija kao u ju`noj Rumuniji i Poni{avqu ispod kqu~iti da KocofeniKostolac grupa egzistira
sloja faze I Kocofeni grupe nalazio se sloj ^erna- krajem IV i u prvoj polovini III milenijuma pre n.
voda III grupe, dok je iznad ovog sloja konstatovan e., mada je mogu}no da se u pojedinim regijama za-
sloj faze II Kocofeni grupe (zapravo je re~ o kera- dr`ala i du`e.75
mici sa elementima kostola~ke grupe, odnosno o Rezultati za najstariju i sredwu fazu kosto-
KocofeniKostolac grupi). la~ke kulturne grupe na Gomolavi kre}u se u raspo-
U Ostrikovcu kod Jagodine, na jedinom lokali- nu od 3038. do 2903. pre n. e. i od 3108. do 2877. pre
tetu sa vertikalnom stratigrafijom koja prati n. e., dok je kostola~ka kultura na nalazi{tu Stre-
razvoj sredweg, odnosno poznog eneolita u Pomo- im u Vu~edolu kalibrirana u 33102920. pre n. e.,
ravqu M. Stoji} izdvaja ~etiri razvojne faze kao {to se pribli`no datuje i naseqe ove grupe u
Ostrikovac Iad.71 Autor poredi nalaze iz hori- Pivnici (30422857. pre n. e.).76 Svi navedeni da-
zonta Ostrikovac Ia sa nalazima Kocofeni grupe, tumi sa lokaliteta kostola~ke grupe ukazuju na to
dok horizonti Ostrikovac Ib i Ic, prema Stoji}u, da je ova kulturna grupa nastala i razvijala se u
pripadaju Kostolac grupi, odnosno fazi III Koco- periodu koji obuhvata posledwu ~etvrtinu IV i
feni grupe.72 Stilsko-tipolo{ke karakteristike po~etak III milenijuma pre n. e., {to bi Kocofe-
keramike iz ovih horizonata odgovaraju Kocofe- niKostolac grupu u Pomoravqu i Timo~koj Kra-
niKostolac grupi, konstatovanoj na mnogim lo-
kalitetima u isto~noj Srbiji i Ni{kom poqu. Ho-
rizont Ostrikovac Id, prema Stoji}u, pripada
69 Alexandrov 1995, 253, 262.
vu~edolskoj kulturi. Interesantno je to {to je u
70 Alexandrov 1995, 263264.
ovom posledwem horizontu evidentiran i jedan 71 Materijal je delimi~no publikovan, bez tehni~ke do-
zvonasti pehar ukra{en vrp~astim ornamentom, kumentacije, pa bi pre kona~nih zakqu~aka i konstatacija
koji je karakteristi~an na~in ukra{avawa u fazi trebalo sa~ekati da se objavi celokupan materijal sa prate-
III Kocofeni grupe na rumunskim nalazi{tima.73 }om dokumentacijom.
72 Stoji} 1989, 177178.
Ako se zaista radi o peharu tipi~nom za kulturu
73 Roman 1976, pl. 89/2, pl. 107/9, 12, 13.
zvonastih pehara (Glockenbecherkultur, Bell Beakers
74 Publikovan je samo fragment pehara, pa prema ilu-
Culture), wegovo prisustvo u Pomoravqu bi se mo-
stracijama nije mogu}e odrediti wegovu formu (Stoji} 1989,
glo objasniti kontaktima populacije Kocofeni sl. 13).
Kostolac grupe iz Pomoravqa sa ovom kulturom iz 75 Boyadziev 1995, 175, 178; Boyadziev 1998, 350, 357358.

centralne Evrope.74 76 Petrovi}, Jovanovi} 2002, 298.

76
KAPURAN. BULATOVI], Kulturna grupa KocofeniKostolac na teritoriji (6594) STARINAR LXII/2012

jini moglo opredeliti u kraj IV / po~etak III mile- teta uo~eno je da su najstarija naseqa, bez kerami-
nijuma pre n. e. i period nakon toga. ke ukra{ene brazdastim ubadawem, formirana u
Kqu~u i Negotinskoj Krajini, pa se pretpostavqa
da su nosioci Kocofeni grupe do{li iz Oltenije
ZAKQU^AK i s Ju`nih Karpata. Veliki broj lokaliteta zapad-
no od Kqu~a, uz Dunav, na kojima je evidentirana
Nakon stabilizacije klimatskih prilika kra- keramika ukra{ena brazdastim ubadawem, ukazuje
jem IV milenijuma, prostor severoisto~ne Srbije na pravac {irewa kostola~kih elemenata iz Ba-
naseqavaju populacije sa visoko mobilnim na~i- nata, Brani~eva i Stiga. Uticaj kostola~ke grupe
nom `ivota i privre|ivawa. One nisu ukqu~ene u bio je izuzetno jak od faze Kocofeni II i na rumun-
eksploataciju mineralnih sirovina, mada nase- skim lokalitetima, {to je ilustrovano keramikom
qavaju gotovo celokupnu oblast Timo~ke Krajine. u Transilvaniji i u Ju`nim Karpatima ukra{enom
Nosioci Kocofeni grupe odnosno KocofeniKo- brazdastim ubadawem. Interesantno je da je ova
stolac grupe, pored mawih naseqa, formiraju i keramika samo sporadi~no konstatovana u Olte-
ve}e centre ili punktove za okupqawe pastirskih niji.77 O~ito su naseqa Kocofeni grupe bila iz-
grupa tokom zimskog mirovawa. Oni isti~u svoje vesna barijera za {irewe ovih elemenata na istok.
prisustvo na odre|enoj teritoriji podizawem na- Formirawem grupe KocofeniKostolac, koja je
seqa u vidu terasa na stenovitim uzvi{ewima, ko- nastala kontaktima nosilaca Kocofeni grupe sa
ja dominiraju okolnim pejza`em. Ovakva gradinska istoka i kostola~ke grupe sa zapada i severozapada,
naseqa, podignuta sa namerom da se vide i sa veli- nastao je period koegzistencije na ovim prostori-
ke daqine, obi~no u svojoj blizini imaju jednu ili ma. U jednom trenutku se ova populacija pokre}e
vi{e pe}ina u koje skqawaju stada ili ih i sami prema jugu i naseqava ju`ni deo Timo~ke Krajine
koriste u ekstremnim vremenskim uslovima. Vi- (okolina Gamzigrada, stava Svrqi{kog i Trgovi-
sinski lokaliteti i lokaliteti sa elementima gra- {kog Timoka), dok su sporadi~na naseqa konstato-
dine bili su naseqeni tokom letwih meseci i nema- vana i u ju`nom Pomoravqu ({iri prostor u{}a
ju tragove ozbiqne nadzemne arhitekture. Naseqa Ni{ave u Ju`nu Moravu, Leskova~ko poqe, ulaz u
u dolinama reka i na sredwim nadmorskim visina- Grdeli~ku klisuru). Izuzetnu dinamiku i pokre-
ma, velikih gabarita, imala su funkciju sezonskih tqivost ove grupe ilustruju elementi ove grupe
stanica za okupqawe ve}ih stada prilikom kreta- evidentirani na keramici u Pelagoniji i na se-
wa iz nizijskih u visinske predele. U isto vreme, vernim obalama Egejskog mora (lokalitet Tre{tena
na {irokim terasama obala Dunava, naro~ito u stena, Sitagroi, Dikili Ta{). Pokretqivost ovih
oblasti Kqu~a (Kowske glave), postojala je gusta zajednica nije morala biti izazvana pritiskom
koncentracija naseqa i to prvenstveno na lokaci- neke druge populacije, ve} klimatskim promenama
jama gde se ova reka mogla lako pre}i. ili na~inom privre|ivawa ovih zajednica, odno-
Prema stilsko-tipolo{kim odlikama kerami- sno mobilnim sto~arstvom koje je, prema mnogim
ke prezentovane u radu, ali i ranije publikovanih autorima, predstavqalo osnovu ekonomije ove kul-
nalaza sa lokaliteta iz isto~ne Srbije, mo`e se turne grupe. Moramo uzeti u obzir i ~iwenicu da
zakqu~iti da postoje dva tipa lokaliteta Kocofe- su klimatske promene tokom IV milenijuma pre n. e.
niKostolac grupe oni na kojima je konstatovana pretvorile prostor isto~nog Balkana u predele sa
keramika ukra{ena brazdastim ubadawem i oni na izrazito suvom klimom.78 Budu}i da je ovakva kli-
kojima ovaj vid ukra{avawa nedostaje. Najvi{e je matska situacija po~ela da se mewa tek u II mile-
evidentirano lokaliteta sa iskqu~ivo kocofeni nijumu pre n. e., mo`e se pretpostaviti da je `ivot
elementima na keramici (34), ali je samo wih neko- mobilnih sto~arskih zajednica u Timo~koj Krajini
liko istra`eno. Na ovom prostoru je konstatovano sa odlikama grupe KocofeniKostolac mogao tra-
28 lokaliteta sa elementima kostola~ke grupe, a jati i ceo jedan milenijum, sve do pojave kultura
17 hronolo{ki neodre|enih lokaliteta, odnosno
onih na kojima se ne mo`e pouzdano odrediti faza
ove grupe.
Analizom distribucije naseqa i stilsko-tipo- 77 Roman 1976, pl. 118.
lo{kih karakteristika keramike sa svih lokali- 78 Todorova 2009.

77
KAPURAN. BULATOVI], Kulturna grupa KocofeniKostolac na teritoriji (6594) STARINAR LXII/2012

sredweg bronzanog doba. Na ovo ukazuju i stilsko- do Pomoravqa, na zapadu (Ostrikovac Id i Bubaw
tipolo{ke karakteristike keramike Verbi~oara Hum II) i od Leskova~kog poqa, na jugu (BubawHum
grupe, koje su veoma sli~ne Kocofeni elementima, II), do sredweg Pomoravqa na severu. Jaki uticaji
kao i vertikalna stratigrafija visine od 1 do 3 m, ove manifestacije vidqivi su na lokalitetima
konstatovana na nekoliko sistematski istra`iva- ovog perioda u Pelagoniji (Tre{tena stena), kao i
nih lokaliteta ove grupe. na severnoj obali Egejskog mora (Dikili Ta{, Si-
Generalno uzev{i, prisustvo elemenata grupe tagroi).
KocofeniKostolac u ju`nom Pomoravqu ne treba Ovim radom smo na osnovu rezultata dosada{-
da iznena|uje, jer je na ovom prostoru (Bratmilov- wih skromnih istra`ivawa lokaliteta iz razvije-
ce, Dowa Slatina), kao i mnogo ju`nije (Sitagroi), nog i poznog eneolita u isto~noj i jugoisto~noj
konstatovana i keramika sa elementima badenske Srbiji poku{ali da damo odgovor na pitawa gene-
grupe, koja je, izvesno, starija od KocofeniKosto- ze i razvoja grupe, teritorije na kojoj se prostira-
lac grupe.79 O~ito je dolina Ju`ne Morave u eneo- la, distribucije naseqa i samog na~ina `ivota,
litu bila va`an komunikacioni pravac zajednica kao i drugih odlika nosilaca KocofeniKostolac
iz Podunavqa ka jugu. grupe. Za preciznije definisawe ovih pitawa, na-
Nosioci KocofeniKostolac grupe su u Pomo- ro~ito relativne i apsolutne hronologije i odno-
ravqu i zapadnoj Bugarskoj, uz kulturni impuls sa ove grupe sa grupama u okru`ewu, neophodno je
vu~edolske kulture, uticali na formirawe jedne sistematski istra`iti indikativna nalazi{ta i
{iroko rasprostrawene prepoznatqive kulturne uraditi serije apsolutnih datuma. Ovaj rad je, uz
manifestacije koju karakteri{u, izme|u ostalog, starije radove o ovoj temi, nadamo se, pru`io osnov-
ornament u vidu sitne guste mre`e izvedene urezi- ne informacije o ovoj grupi kao smernice budu-
vawem i inkrustacija. Ova manifestacija raspro- }im istra`iva~ima koji }e se baviti ovom ili
strawena je od centralne Bugarske (Ezero, Dubene) srodnim grupama u okru`ewu.

79 Renfrew, Gimbutas, Elster 1986, fig. 13.4, Pl. XXXV/25;

Bulatovi}, Jovi} 2010, T. XXXII/35, 8, T. XLV/711.

78
KAPURAN. BULATOVI], Kulturna grupa KocofeniKostolac na teritoriji (6594) STARINAR LXII/2012

BIBLIOGRAFIJA:

Alexandrov 1995 S. Alexandrov, The Early Bronze Nikoli} 2000 D. Nikoli}, Kostola~ka kultura na
Age in Western Bulgaria: Periodization and Cultural teritoriji Srbije, Centar za arheolo{ka istra`ivanja Filo-
Definition, Prehistoric Bulgaria, eds. D. W. Bailey and zofskog fakulteta, Beograd.
I. Panayotov with S. Alexandrov, Monographs in World Panaotov, Aleksandrov 1988 I. Panaotov,
Archaeology No. 22, Madison Wisconsin, 253270. S. Aleksandrov, Za kultura MaguraKocofeni v
Antonijevi} 1982 D. Antonijevi}, Obredi i blgarskite zemi, Arheologi kn. 2/1988, Sofi, 115.
obi~aji balkanskih sto~ara, Balkanolo{ki insti- Pernik 1981 Praistori~eski selia (VII
tut, Beograd. hil. pr. n.e.), Pernik, tom I, Poselien `ivot na
Boyadziev 1995 Y. Boyadziev, Chronology of hlma Krakra, Sofi, 1151.
Prehistoric Cultures in Bulgaria, Prehistoric Bulgaria, Petrovi} 1941 M. Petrovi}, \erdapski ribo-
Monographs in World Archaeology No. 22, Madison lov u pro{losti i sada{wosti, Srpska kraqevska
Wisconsin, 149191. akademija, Kwiga LVII, Beograd.
Boyadzijev 1998 Y. Boyadziev, Radiocarbon ating Renfrew, Gimbutas, Elster (eds.) 1986 C.
From outheastern Europe, in: M. Stefanovich (ed.) James Renfrew, M. Gimbutas, E. S. Elster (eds.), Excavations
Harvey Gaul in Memoriam, James Harvey Gaul Foun- at Sitagroi, A Prehistoric Village in Northeast Greece,
dation, Sofia, 349370. vol. 1, Los Angeles: The Institut of Archaeology, The
Bulatovi}, Jovi} 2010 A. Bulatovi}, S. Jovi}, University of California.
Leskovac, kulturrna stratigrafija praistorijskih Roman 1976 P. Roman, Cultura Cotofeni, Acade-
lokaliteta u Leskova~koj regiji, BeogradLesko- mei Republici socialiste Romania, Bucuresti.
vac. Spasi} 2010 M. Spasi}, Cotofeni communities at
Deshayes 1970 J. Deshayes, Les fouilles de Dikili their Southwestern frontier and their relatiponship with
Tash et larcheologie Yougoslave, Zbornik Narodnog Kostolac population in Serbia, Dacia LIV, Bucarest,
muzeja VI, Beograd, 2143. 157175.
Georgiev at al. (eds.) 1979 G. I. Georgiev at al. Stoji}, ^a|enovi} 2006 M. Stoji}, G. ^a|e-
(eds.), Ezero, rannobronzovoto selie, Blgarskata novi}, Kru{evac, kulturna stratigrafija praisto-
Akademi na naukite, Sofi. rijskih lokaliteta u zoni stava Zapadne Morave i
Fletcher 1977 R. Fletcher, Settlement studies Ju`ne Morave, Beograd Kru{evac 2006.
(Micro and semi-micro), in: D. Clake (ed.) Spatial Arche- Stoji}, Joci} 2006 M. Stoji}, M. Joci}, Ni{
ology, Academic Press, London, 47162. kulturna stratigrafija praistorijskih lokali-
Mihailovi}, \uri~i}, Kalu|erovi} 1997 teta u ni{koj regiji, BeogradNi{ 2006.
D. Mihailovi}, Q. \uri~i}, S. Kalu|erovi}, Istra- Stoji}, Iliji} 2011 M. Stoji}, B. Iliji},
`ivawa paleolita na podru~ju isto~ne Srbije, u: Kwa`evac, kulturna stratigrafija praistorijskih
M. Lazi} (ur.), Arheologija isto~ne Srbije, Centar lokaliteta kwa`eva~kog kraja, Arheolo{ki institut
za arheolo{ka istra`ivawa Filozofskog fakulte- i Zavi~ajni muzej u Kwa`evcu, BeogradKwa`evac.
ta, Beograd, 3344. Tasi} 1971 N. Tasi}, Osnovni rezultati istra`ivanja
Milanovi}, D. 2011 D. Milanovi}, Naseqe u Zlotskoj pe}ini i nalazi{ta na \erdapu, u: M. Birta{evi},
^ernavoda III kulture na lokalitetu Bubaw, Stari- Lj. Plesni~ar i M. Zaninivi} (ur.) Materijali VI, VIII kon-
nar, n.s. LXI, Beograd, 101113. gres arheologa Jugoslavije u Boru 1969, Beograd, 7180.
Mitkoski 2010 A. Mitkoski, Mariovo vo Tasi} 1978 N. Tasi}, Eneolitska naselja isto~ne
praistoriata, Macedoniae Acta Archaeologica 19, Srbije i njihova ekonomika, u: B. Babi} i dr. (ur.) Mate-
Skopje, 4984. rijali XIV, X kongres arheolog Jugoslavije, Prilep 1976,
Jovanovi} 1971 B. Jovanovi}, Metalurgija ene- Beograd, 103110.
olitskog perioda Jugoslavije, Arheolo{ki institut, Beo- Tasi} 1982 N. Tasi}, Naselja bakarnog doba u
grad. isto~noj Srbiji, Zbornik radova muzeja rudarstva i me-
Nikoli} 1997 D. Nikoli}, Eneolitska naseqa talurgije u Boru, 1936.
u okolini Majdanpeka, Arheologija Isto~ne Srbije, Tasi} 1983 N. Tasi}, Jugoslovensko podunavqe
ur. M. Lazi}, Centar za arheolo{ka istra`ivawa od indoevropske seobe do prodora Skita, Matica
Filozofskog fakulteta, Beograd, 197210. Srpska i Balkanolo{ki institut SANU.

79
KAPURAN. BULATOVI], Kulturna grupa KocofeniKostolac na teritoriji (6594) STARINAR LXII/2012

Tasi} 1983a N. Tasi}, Dunavski put u ene- Todorova 2007 H. Todorova, Die paleoklimatische
olitu i bronzanom dobu, u: V. ^ubrilovi} (ed.) Plo- Entwicklung in VIII Jt. Vor Chr. In: H. Todorova, M.
vidba na Dunavu i wegovim pritokama kroz vekove, Stefanovich and G. Ivanov (ed.) Struma/Strymon River
SANU Nau~ni skupovi XV, Beograd, 5362. Valley in Prehistory, Proceedings of the international
Tasi} 1995 N. Tasi}, Eneolithic cultures of Central Symposium Strymon Praehistoricus 27.0901.10.2004.
and West Balkan, Institute for Balkan studies and Dra- in KjustendilBlagoevgrad and SerresAmphipolis. In:
gani}, Beograd. the Steps of James Harvey Gaul, Volume 2, Sofia,
Tasi} 1997 N. Tasi}, Eneolit i bronzano doba 1926.
severoisto~ne Srbije, u: M. Lazi} (ur.) Arheologija Cviji} 1987 J. Cviji}, Kulturni pojasi Bal-
isto~ne Srbije, Centar za arheolo{ka istra`ivawa kanskog poluostrva, Vrste sto~arstva, stanovi, ka-
Filozofskog fakulteta, Beograd, 7989. tuni, mandre, u: M. Lutovac i dr. (ur.) Antropogeo-
Tasi} 2004 N. Tasi}, Nalazi{ta kulture iz grafski spisi, Kwiga 4, SANU, Kwi`evne novine i
eneolitskog perioda, u: M. Lazi} (ur.) Bor i okoli- Zavod za uxbenike, Beograd.
na, Muzej rudarstva i metalurgije u Boru i Centar Vulpe 1970 A. Vulpe, Die xte und Beile in
za arheolo{ka istra`ivawa Filozofskog fakulte- Romnien, Prhistorische Bronzefunde, Ab. IX, band 2,
ta u Beogradu, BorBeograd, 57100. Mnchen.

80
KAPURAN. BULATOVI], Kulturna grupa KocofeniKostolac na teritoriji (6594) STARINAR LXII/2012

Summary: ALEKSANDAR KAPURAN, The Institute of Archaeology, Belgrade


ALEKSANDAR BULATOVI], The Institute of Archaeology, Belgrade

COOFENIKOSTOLAC CULTURE
ON THE TERRITORY OF NORTH-EASTERN SERBIA

Key words. North-eastern Serbia, Late Aeneolithic Period, Early Bronze Age Period, CoofeniKostolac,
settlement distribution, ceramic production.

The settlement of the territory of north-eastern Serbia by the nivelation into terraces atop which surface structures could be
representatives of the Coofeni culture began during the second built using wood covered with mud (Jezero, Kulmja [kjopuluji,
half of the IV millennium, probably under the pressure of invad- Pjatra Kosti, Vratna, Bogovina). The third shared characteristic
ing tribes from Euroasian steppe. This territory extended over is that one or more caves are usually located in the immediate
Transylvania, Banat, Oltenia and Muntenia (Map 2). On the ter- vicinity of settlements. An example of the symbiosis of cave and
ritory of Serbia they settled from the Djrerdap gorge up to the hill fort CoofeniKostolac settlements is the vicinity of the Za-
Mlava river to the west, and through Ku~ajske mountains, Bor, vojsko jezero near Majdanpek. So far two hill fort settlements,
Zaje~ar and further to the south, up to Ni{. A specific symbiosis Jezero and Kljanc (T. I/3; Map 1/1112), were identified in this
occurred on the territory of Serbia between the Coofeni and the area, built on limestone cliffs above the Mali Pek river. The Raj-
Kostolac cultures. kova cave (Map 1/14), Paskova cave and Kapetanova cave (Map
According to the results of the latest project of re-identifi- 1/13) are located in their immediate vicinity, in which the re-
cation, the number of CoofeniKostolac sites and settlements mains of anthropogenic activity were discovered. The Kapeta-
increased to 76. After all the sites were re-identified and geo- nova cave provides stratigraphy of over 3 m high, which repre-
referenced, with consideration of the surrounding landscape, sents a rare case for CoofeniKostolac cultural sites. This fact
hydrography, geomorphology of the terrain and the character of does not only indicate its long-term use, but could provide the
the ceramic production finds, we believe that there is a need for answer to the genesis and duration of this cultural phenomenon
re-analyzing specific aspects of the cultural and geographic de- on the territory of the Serbian part of the Djerdap hinterland.
velopment not only of settlements, but of the entire CoofeniKo- The fourth shared characteristic which links these settlements is
stolac cultural phenomenon. In this paper we considered three their dominant position in the landscape. Given that their posi-
archaeological sites in the Ni{ava valley, given that re-identifi- tion and appearance are readily visible from a considerable dis-
cation work over the past several years yielded new information tance, they probably were not used for hiding, but for making
(BubanjStaro Selo, Velika Humska ~uka and Donja Vre`ina). their position prominent. We suppose that pastoral communities
The topography of CoofeniKostolac settlements on the emphasized in this manner their control of mountain crosspass
territory of north-eastern Serbia, the Serbian part of the Danube and roads, particularly in places where rivers exit narrow canyons
valley and its hinterland, is characterized by diversity of posi- in important communications paths to the Crni and Beli Timok,
tion (location above sea level and landscape placement), types Pek and Danuber rivers. The other CoofeniKostolac type
of houses and economic survival. In the 70s of the last century settlement on the territory of north-eastern Serbia is represented
sites were identified that are located in very inaccessible terrain, by settlements that are positioned on smaller hills or on gentle
which in particular cases has an slope incline of 45, where the slopes that on the average range between 336 and 210 m above
number of such settlements in the meantime increased to nine. sea level. The only fortified hill fort settlement discovered so far,
They are represented by Kulmja [kjopuluji in Kloko~evac and ^oka lu Bala{ near Krivelj (Map 1/3) belongs to this group.
Pjatra Kosti in Crnajka (T. I/12; Map 1/9), followed by Vratna The archaeological sites Velika ^uka i Neresnica (Map 1/23),
Veliki most (T. I/ 7; Map 1/33), Bogovina-above a cave (T. I/ Smiljkova glavica in [tubik (Map 1/31) and ]eta}e in Kovilovo
4; Map 1/8), Jezero (T. I/ 3; Map 1/12), Kljanc (T. I/3; Map (Map 1/38) are located on wide and flat, elevated plateaus that
1/11), TurijaStenje (T. I/ 6; Map 1/22), Mokranjske stene-quarry dominate up on river valleys. Judging by the considerable sur-
(T. I/ 5; Map 1/39) and Bolvan (T. I/ 8; Map 1/66). These settle- face that they occupy, their position and surroundings for these
ments have several other common elements, the most important two settlements, we can suppose that they could have been used
being that each one of the elevated settlements is positioned on for wintering places or points for gathering of flocks and shep-
the rocky peak of a canyon, in places where smaller rivers or herds during pauses between seasonal migrations. They are pri-
brooks flow into a larger river. We can suppose how the selection marily characterized by the natural surroundings of smaller hills
of such positions was of strategic importance, given that in the and larger river valleys, as well as the relatively low above sea
mountainous area of north-eastern Serbia the system of water- level elevation on which they are located. Such seasonal stations
ways and river valleys represents communicational links from or checkpoints on which larger groups of shepherds could gather
prehistory to modern times. The second common characteristic with their flocks during the winter months represented impor-
of these settlements is the rocky massif which provided the tant locations in the lives of pastoral communities. During the
foundation for their erection. The rock foundation in the majority warm summer period, homesteads with stable architecture are
of cases is of limestone origin and is well suited to artificial abandoned because of migrations into mountain areas, where

81
KAPURAN. BULATOVI], Kulturna grupa KocofeniKostolac na teritoriji (6594) STARINAR LXII/2012

favourable grazing areas area located. Certain groups of shep- mentation has been observed and those where this type of orna-
herds during autumn returned to these settlements en route to low- mentation still has not been observed. Still, it is unclear whether
lands and river terraces, while other groups probably continued this distinction can be applied to period assignment, or whether
their journey to gathering centres in valleys near the Danube it is in fact caused by settlement of different populations in dif-
and the Timok rivers. ferent regions of Eastern Serbia the Kostolac region from the
The next type of settlement belongs to high, multi-layered west and the Coofeni group from the East. In Romania, how-
settlements (Arija babaKo{obrdo, ^oka Kormaro{, Field of Z. ever, vertical stratigraphy was observed at several settlements
Brzanovi}, Varzari and SmedovacGrabarSvra~ar) which rep- where development phases were observed of the Coofeni group,
resent sunbathed dominant positions, with a good view of the so that based on the stratigraphy at those sites, with certain caution,
surrounding area, well suited to long-term occupation. Settle- it is possible to draw conclusions about the development of the
ments on high elevations of this type are usually linked with CoofeniKostolac group in eastern Serbia. Settlements without
landscapes that predominate in grazing areas and in which there any furchenstich ornamentation would be assigned to the older
are no large forests. phase (Coofeni group) where ceramics characteristic of the
The last type of CoofeniKostolac settlement is characte- Coofeni group have been observed, although observed shapes
ristic of lowland settlements positioned on river terraces. The and ornaments are usually associated with the furchenstich techni-
settlements on the right bank of the Danube, around Klju~ (Kla- que and the more recent phase of the group.
dovoBrodoimpeks, Mala Vrbica, ZbradilaFund, Korbovo The most frequent type of vessels at sites in eastern Serbia
Obala, VajugaPesak, Jakomirski potok estuary, Velesnica, Lju- are amphorae with extended funnel shaped necks, ornamented
bi~evacriver bank, Ljubi~evacIsland, Brzi prun, Slatinska reka below the neck with carved lines or with stamped ornamenta-
estuary, Knjepi{te, Ru`enjka, KusjakBordjej, KusjakMotel, tion (fig. 6, 21, 38, 64, 71, 89, 98100, 104, 109, 115, 116, 134),
KusjakVrkalj), represented points at which shepherds flocks fishbone shape impressions (fig. 4, 28), and in the more recent
could remain for longer periods, waiting for favourable condi- period furchenstich ornamentation or point impressions (fig. 9,
tions for crossing to the other side of the river. This assumption 20, 25, 140), with a tongue shaped or vertically perforated handle,
is based on old maps predating the construction of the accumu- tunnel shaped or horse-shoe shaped handle below the rim (fig. 6,
lation lake. These maps indicate that in the immediate vicinity 9, 20, 21, 51, 63, 100, 126, 134, 88, 115 ). The second charac-
of these settlements were located small sand islands linked to teristic type of vessel are semi-spherical bowls with deeper re-
the river bank, pointing to shallows and crossing points. These cipients, with flat rims (fig. 11, 12, 23, 27, 29, 5254, 57, 5960,
sections of the river bank, during prolonged droughts or during 74, 79, 81, 82, 90, 91, 95, 113, 124, 125, 131 and 145), or with
cold winters, when ice was formed, could have been places shallower recipients, with a slanted, triangular rim or Tshaped
where the river was crossed from one side to the other. profiled rim (14, 19, 133 and 146). Such vessels are characteristic
Residential architecture cannot be precisely defined, given for both phases, because they are ornamented, besides vertical
that the discovered remains of houses are very meagre and lack ribs, with carves, and with furchenstich ornamentation (fig. 23,
sufficient elements for reconstruction. The most recent excava- 68, 81 and 82). The third type of vessels are semi-spherical bowls
tions on the BubanjStaro Selo settlemant at Ni{, indicate an with contracted rims creating a nearly spherical shape. They can
identical type of architectural construction as discovered at Go- be ornamented with vertical ribs on rims (fig. 148) in combination
molava and Bordjej which represents structures that are charac- with pinholes (fig. 17), carves (fig. 61, 84, 85) or line impres-
teristic for lowland areas. Houses in hill fort settlements built on sions (fig. 132). Less frequent vessels on the territory of north-
artificial terraces have been mostly devastated by erosion, so eastern Serbia are biconical or spherical goblets, followed by
that judging by the impressions of wooden structures and wattle pare-shaped goblets with a single handle, larger pare-shaped
and daub, as well as the remains of hearths, it can be asserted amphorae with an extended or conical neck, with small handles
that these were residential structures. below the rim, ornamented with a series of carves (fig. 39, 86),
Numerous studies so far noted that based on the stylistic as well as barrel or spherical pots ornamented with carves, hori-
and typological characteristics of ceramics on archaeological zontal tapes or circular impressions (fig. 4547, 141, 142). The
sites in Timo~ka Krajina it is possible to distinguish between appearance of ropeshape ornaments is very significant, given
two phases of the Coofeni group, where the first is dominated that they appear in Rumanian finds in the second phase of the
by ornamental techniques of carving that are characteristic of Coofeni group, and most frequently in the third phase. This
the Coofeni group, and a later phase in which this style is mixed ornament was sporadically observed in the far south, on the
with the furchenstich, as well as other Kostolac cultural elements Dikili Ta{ site on the northern shore of the Aegean sea, in level
(furchenstich, certain types of ceramics, etc.). The fact is that 6, which according to the author belongs chronologically to the
the majority of CoofeniKostolac group sites in eastern Serbia BubanjHum II group and the Kostolac group. Its presence at
have not been excavated, or have only been partially excavated, sites in eastern Serbia can be linked to the older phase at the
and that no vertical stratigraphy had been observed, where no majority of settlements, except in the case of GrabarSvra~ar, as
stratigraphic relationship between stylistic-topological charac- these ceramics were not found alongside ceramics with furchen-
teristics of older ceramics (Coofeni) and the more recent phase stich. The largest number of sites with only Coofeni elements
(CoofeniKostolac) have been established. These are mostly on ceramics have been observed (34), but it is indicative that
settlements in which ceramics were observed with elements both only a few have been excavated. 28 sites with Kostolac group
of the Kostolac and the Coofeni group, or only with elements elements were noted, while 17 unspecified sites in which the
of the Coofeni group, while settlements with only Kostolac period cannot be precisely defined have been identified.
ceramics have not been identified. Therefore, in Serbia it is only According to the stratigraphy of several of the mentioned
possible to distinguish between sites where furchenstich orna- sites in western Bulgaria, in the Morava valley and in southern

82
KAPURAN. BULATOVI], Kulturna grupa KocofeniKostolac na teritoriji (6594) STARINAR LXII/2012

Romania it can be concluded that the Coofeni group (north- Coofeni to the east and the representatives of the Kostolac
eastern Serbia and Romania) and the CoofeniKostolac group group to the west and north-west a short period of coexistence
(Morava valley and western Bulgaria), in all of the mentioned occurred on this territory.
regions, was preceded by the ^ernavoda III group, and was super- Absolute dating of the chronological framework of the
seded by the Vu~edol culture and the BubanjHum II group in CoofeniKostolac group in the Danube valley and in eastern
the Morava valle and the Struma valley, and the Glina IISchne- Serbia can only be assigned indirectly, as there is no carbon dating
kenber group in Oltenija and the territory of Transylvania and available from these sites. According to J. Boja~ijev, phase IIIII
the southern Carpathians. of the Coofeni group (44004300 bp) can be assigned chrono-
Analysis of the distribution of settlements and stylistic- logically approximately to the same period as the Kostolac group
topological characteristics of ceramics from all of the settle- (45004100 bp), and if we suppose that the CoofeniKostolac
ments led to the conclusion that the oldest settlements, without group occurred a little while after the occurrence of the Kosto-
ceramics with furchenstich ornamentation, were established in lac group, it can be concluded that the CoofeniKostolac group
Klju~ in Negotinska Krajina, leading to the assumption that the existed at the end of the IV and the first half of the III millennium
representatives of the Coofeni group came from Oltenia and BC, although it is possible that it continued even later in particular
from the southern Carpathians. A large number fo sites west of regions. The results for the oldest and the middle phase of the
Klju~, along the Danube, at which ceramics with furchenstich Kostolac cultural group at Gomolava range between 30382903
ornamentation were noted, point to the direction of expansion BC and 31082877 BC, while the Kostolac culture at the Streim
of Kostolac elements, from Banat, Branicevo and Stig. The and Vu~edol sits was dated 33102920 BC, as is the approximate
influence of the Kostolac group was very strong starting in the dating of settlements of this group in Pivnica (30422857 BC).
Coofeni II phase, even in Romanian sites, given that in Tran- All the dating of Kostolac group sites indicate that this cultural
sylvania and in the southern Carpathians a large number of group occurred and developed in the period of the last quarter
ceramic finds were found with furchenstich ornamentation, of the IV and the first half of the III millennium BC, which
while it is interesting that only sporadic appearances were noted would chronologically assign the CoofeniKostolac group in
in Oltenia. It is clear that Coofeni group settlements represent- the Morava valley and Timo~ka Krajina to the end of the IV and
ed a certain barrier to the expansion of these elements to the the start of the III millennium BC, and to the ensuing period.
east. With the formation of the CoofeniKostolac group which
was created through contacts between representatives of the Translated by Vladimir Radonji}

83
KAPURAN. BULATOVI], Kulturna grupa KocofeniKostolac na teritoriji (6594) STARINAR LXII/2012

1 2

3 4

5 6

7 8

Tabla I 1) Kulmja [kjopuluji; 2) Crnajka; 3) Jezero i Kqanc; 4) Bogovina; 5) Mokrawske stene;


6) Turija; 7) Vratna; 8) Bolvan
Plate I 1) Kulmja [kjopuluji; 2) Crnajka; 3) Jezero i Kljanc; 4) Bogovina; 5) Mokranjske stene;
6) Turija; 7) Vratna; 8) Bolvan

84
KAPURAN. BULATOVI], Kulturna grupa KocofeniKostolac na teritoriji (6594) STARINAR LXII/2012

1 2 3

4
5 6

7 8 9

11 12
10

13 14

Tabla II 13) Bogovinska pe}ina; 4) ^oka Kormaro{; 5) Brestova~ka bawa; 68) Kameni rog;
911) Dowa Bela rekavrelo; 1214) Kulmja [kjopuluji
Plate II 13) Bogovinska pe}ina; 4) ^oka Kormaro{; 5) Brestova~ka banja; 68) Kameni rog;
911) Donja Bela rekaVrelo; 1214) Kulmja [kjopuluji

85
KAPURAN. BULATOVI], Kulturna grupa KocofeniKostolac na teritoriji (6594) STARINAR LXII/2012

15 16

17 18 19

21 22

20 23 24

25 27
26

Tabla III 1516) Kulmja [kjopuluji; 1718) Kriveq, ^oka morminc; 19) Mali Kriveq, kod vodenice;
1922) Kapetanova pe}ina; 2327) Kqanc
Plate III 1516) Kulmja [kjopuluji; 1718) Krivelj, ^oka morminc; 19) Mali Krivelj, kod vodenice;
1922) Kapetanova pe}ina; 2327) Kljanc

86
KAPURAN. BULATOVI], Kulturna grupa KocofeniKostolac na teritoriji (6594) STARINAR LXII/2012

28 29 30

31 32 33

34 35 36

37 38

39

40 41 42

Tabla IV 2831) Neresnica, Velika ~uka; 3233) Rudna glava, [eta}e; 3436) Tanda, La Tufek;
3742) Zlotska pe}ina
Plate IV 2831) Neresnica, Velika ~uka; 3233) Rudna glava, {eta}e; 3436) Tanda, La Tufek;
3742) Zlotska pe}ina

87
KAPURAN. BULATOVI], Kulturna grupa KocofeniKostolac na teritoriji (6594) STARINAR LXII/2012

43 44 45

46

47 48

50
49 51

52 53 54

55 56
57

Tabla V 4344) Zlot, Seli{te (B. \ukanovi}); 45) Gamzigradska bawa, Wiva Z. Brzanovi};
4647) Gamzigrad; Varzari; 4850) Dowe Zuni~e, AxijskoVinsko; 51) Rgo{te, Bolvan;
52) Rgo{te, RosuqaVi{war; 5356) Kwa`evac, Dubrava; 57) Brusnik, Glavica
Plate V 4344) Zlot, Seli{te (B. \ukanovi}); 45) Gamzigradska banja, Njiva Z. Brzanovi};
4647) Gamzigrad, Varzari; 4850) Donje Zuni~e, Ad`ijskoVinsko; 51) Rgo{te, Bolvan;
52) Rgo{te, RosuljaVi{njar; 5356) Knja`evac, Dubrava; 57) Brusnik, Glavica

88
KAPURAN. BULATOVI], Kulturna grupa KocofeniKostolac na teritoriji (6594) STARINAR LXII/2012

58 59 60

63

61 62

64 65 66

68 69 70
67

71 72 73

Tabla VI 58) Velesnica; 5966) Veqkovo, Kapu \aluluj; 67) Grabovnica, Brzi prun;
6870) Dobra, Gospo|in vir; 7173) Dowi Milanovac, Veliki Gradac
Plate VI 58) Velesnica; 5966) Veljkovo, Kapu \aluluj; 67) Grabovnica, Brzi prun;
6870) Dobra, Gospo|in vir; 7173) Donji Milanovac, Veliki Gradac

89
KAPURAN. BULATOVI], Kulturna grupa KocofeniKostolac na teritoriji (6594) STARINAR LXII/2012

74 75
76 77

78 79
80

81 82 83

84 85
86

87 88 89

Tabla VII 7478) Vratna, Veliki most; 79) Korbovo (V. Draganovi}); 80) Kladovo, Brodoimpeks;
8183) Kladovo, Dowe Butorke; 8489) Kovilovo, Vrkaq]eta}e
Plate VII 7478) Vratna, Veliki most; 79) Korbovo (V. Draganovi}); 80) Kladovo, Brodoimpeks;
8183) Kladovo, Donje Butorke; 8489) Kovilovo, Vrkalj]eta}e

90
KAPURAN. BULATOVI], Kulturna grupa KocofeniKostolac na teritoriji (6594) STARINAR LXII/2012

90 91 92

93 94

95 96 97

98 99 100

101 102 103


104

Tabla VIII 9094) Korbovo, ZbradilaFund; 9597) Kusjak, Vrkaq uzvodno od Motela;
9899) Qubi~evac, Ostrvo; 100103) Kusjak, Grle; 104) Qubi~evac, Obala
Plate VIII 9094) Korbovo, ZbradilaFund; 9597) Kusjak, Vrkalj uzvodno od Motela;
9899) Ljubi~evac, Ostrvo; 100103) KusjakGrle; 104) Ljubi~evac, Obala

91
KAPURAN. BULATOVI], Kulturna grupa KocofeniKostolac na teritoriji (6594) STARINAR LXII/2012

105

106 107

108
109 110

111 112
113

115 116
114

117 118 119

Tabla IX 105106) Qubi~evac, Obala; 107112) Mokrawske stene kamenolom;


113119) Mokrawske stene, potkapina
Plate IX 105106) Ljubi~evac, Obala; 107112) Mokranjske stene kamenolom;
113119) Mokranjske stenepotkapina

92
KAPURAN. BULATOVI], Kulturna grupa KocofeniKostolac na teritoriji (6594) STARINAR LXII/2012

120 121 122 123

124 125 126

127 128 129

131 132
130

133 134 135

Tabla X 120123) Popovica, Veliko brdo; 124129) Prahovo, Ide}e; 130) Sikole, Gradi{te;
131135) Smedovac, GrabarSvra~ar
Plate X 120123) Popovica, Veliko brdo; 124129) Prahovo, Ide}e; 130) Sikole, Gradi{te;
131135) Smedovac, GrabarSvra~ar

93
KAPURAN. BULATOVI], Kulturna grupa KocofeniKostolac na teritoriji (6594) STARINAR LXII/2012

138

136 137

139 140

141 142

143 144

145

146 147 148

Tabla XI 136138) Smedovac, GrabarSvra~ar; 139141) Boqetin, Pe{}era Mare;


142143) Miro~, pe}ina iznad Trajanove table; 144147) Ni{, Bubaw, Obj. 15
Plate XI 136138) Smedovac, GrabarSvra~ar; 139141) Boljetin, Pe{}era Mare;
142143) Miro~, pe}ina iznad Trajanove table; 144147) Ni{, Bubanj, Obj. 15

94
DRAGANA ANTONOVI], The Institute of Archaeology, Belgrade
MOMIR VUKADINOVI], Institute for the Development of Water Resources Jaroslav ^erni, Belgrade

ENEOLITHIC MINE PRLJU[A MALI [TURAC


ARCHAEOLOGICAL AND GEOPHYSICAL INVESTIGATIONS

UDK: 902.3:550.83(497.11)"2011" ; 903:622"636"(497.11) e-mail: d.antonovic@ai.sanu.ac.rs


DOI: 10.2298/STA1262095A Received: February 16, 2012
Preliminary communication Accepted: June 21, 2012

Abstract. The Eneolithic copper mine at Mali [turac was discovered in 1980 and subsequently investigated to a smaller extent
from 1981 to 1987. In 2010 the investigations at Prlju{a were reactivated with the aim of defining how much and how long the
mine had been exploited during prehistory. Pilot geophysical studies were followed by more extensive explorations in 2011. They
focused on a zone related to Shafts 4 and 6, discovered in 1987. The geophysical explorations have comprised the methods of self-
potential SP, electrical scanning ES and seismic profiling with one geophone SGRP. The explorations covered a surface of
400 m including five sections, each 50 m long, with 2 m intervals between them. These investigations identified underground
channels in Shaft 4 and Shaft 6. Three meters below Shaft 4, a large underground gallery was found and in the continuation of the
entrance of Shaft 6, a 10 m long horizontal channel was detected. Northwards from Shaft 4 and Shaft 6, at a distance of 68 m, at
least six mining shafts were detected. However, they are not visible on the surface because their entrances are filled with loose
material. The investigations carried out in 2011 proved that geophysical investigations are an efficient method for studying old
mining works and, therefore, it has been decided to continue with this type of exploration.

Key words. copper mine, geophysical investigations, Eneolithic, archaeometallurgy, Serbia.

he name Mali [turac1 has long been known in

T archaeology as an Eneolithic copper mine. The


mine was discovered in 1980 during recon-
naissance works on Mt. Rudnik in the preparation for
1 The Eneolithic mine is located at the Prlju{a site, which is
situated at the top of Mali [turac. However, the site appeared in lite-
rature with this second name and it is better known according to it.
2 In 1980 the project entitled Project of investigations of old
a new project.2 The first archaeological investigations
mining and metallurgy on Mt. Rudnik started. It finished in 1989
were done in 1981 when potential entrances into two
and it was completed in cooperation with the Archaeological Insti-
shafts (Shafts 1 and 2) were discovered. The explo- tute of Belgrade, the National Museum of ^a~ak, the Institute for
rations continued in1987 when four additional shafts Cultural Heritage Preservation of Kraljevo and the National Museum
were found (Shafts 3, 4, 5 and 6) but further excava- of Kraljevo. The coordinator of the project was Dr. Borislav Jovano-
vi} from the Archaeological Institute. The project aimed at investi-
tions of the already detected shafts have never been
gating physical traces, indirect evidence and other cultural remnants
done. Shafts 5 and 6 have been extensively studied and related to old mining and metallurgy in the area of Mt. Rudnik, from
their approaching platforms, with visible mining adits, Prehistory to the Medieval Age (Jovanovi} 1988, 11, footnote. 2).

* The article is the result of the projects: Archaeology of Serbia: cultural identity, integration factors, technological processes and the role of
the Central Balkans in the development of European prehistory (no 177020) and Cultural changes and population movements in the early pre-
history of the Central Balkans (no 177023) financed by the Ministry of Education, Science and Technological Development of the Republic
of Serbia. The investigations at Prlju{a have been done with the support of the Ministry of Culture, Media and Information Society of the
Republic of Serbia and the Municipality of Gornji Milanovac.

95
ANTONOVI], VUKADINOVI], Eneolithic mine Prlju{a Mali [turac (95106) STARINAR LXII/2012

Fig. 1. The earliest copper ore


mines and copper finds in Serbia:
1) Prlju{a; 2) Vin~a;
3) Belovode; 4) Rudna Glava;
5) Jarmovac; 6) Plo~nik
Sl. 1. Najstariji rudnici bakra
i nalazi od bakra u Srbiji:
1) Prqu{a; 2. Vin~a;
3) Belovode; 4) Rudna Glava;
5) Jarmovac; 6) Plo~nik

indicate that a few ore veins might have been exposed vode, Plo~nik and Vin~a),5 and it raised the question as
at this place. Poor archaeological material collected to where the copper ore, which was melted and produced
during the excavations comprised numerous mining the first metal in this area, came from (Fig. 1). Until now
hammers, broken mining tools originally made of deer there was only one unequivocally documented mine
horns as well as small fragments of Late Eneolithic or from the period of the Vin~a Culture Rudna Glava in
Early Bronze age ceramics.3 Two years later, in 1989, eastern Serbia, while the locality of Jarmovac near Priboj
geodetic surveys were carried out on the part of the is considered to have been exploited during the late Vin-
slope where shafts had been detected and these were ~a Culture, mainly based on the fact that, in the neighbo-
the last investigations carried out in this area. However, urhood, a settlement of the same period exists.6 Taking
during the last decades significant advancements were
made in the study of the beginning of metallurgy in the
territory of Serbia. The earliest occurrences of copper
were dated to the period of the end of the early phase of 3 Jovanovi} 1988, 8; Bogosavqevi} 1988, 21, 31.
the Vin~a Culture (the transition from Vin~aTurda{ II 4 Radivojevi} et al. 2010, 2778.
to Gradac Phase), i.e. 5000 years BC.4 Intensive metal- 5 Antonovi} 2002; Radivojevi} 2007; [ljivar et al. 2006.
lurgic activity was recorded at a few sites in Serbia (Belo- 6 Jovanovi} 1982; Derikonji} et al. 2011.

96
ANTONOVI], VUKADINOVI], Eneolithic mine Prlju{a Mali [turac (95106) STARINAR LXII/2012

Fig. 2. The locality of Prlju{a:


1) modern shaft from 20th century;
2) geophysical profiles 15
(0 m is on the west end,
50 m on the east end);
3) ancient mining works detected in 2011;
4) Shaft 4, discovered in 1987;
5) Shaft 5 discovered in 1987;
6) Shaft 6 discovered in 1987
Sl. 2. Lokalitet Prqu{a:
1) novo rudarsko okno iz 20. veka;
2) geofizi~ki profili 15
(0 m je na zapadnom, 50 m na isto~nom kraju);
3) stari rudarski radovi detektovani 2011. godine;
4) Okno 4 otkriveno 1987. godine;
5) Okno 5 otkriveno 1987. godine;
6) Okno 6 otkriveno 1987. godine

into account the evidence of intensive metallurgic acti- ing in altitude from 882 m to 994.41 m. The slope is
vity, there is no doubt that at that time, as well as later very steep with an average dipping angle ranging from
on, there were several active mines in the territory of 28 to 31, reaching as much as 37 in the lower part
Serbia and that one of them was Mali [turac. Until now, (Fig. 2). The investigations undertaken in 2011 re-
however, there has been a lack of material evidence vealed the presence of 13 complexes of older mining
regarding the exploration at Prlju{a during the Vin~a works in the area above the line connecting Shafts 4, 5
culture. Due to this fact, and because of the size of the and 6 (discovered in 1987). The field relationships
locality and the volume of ore reserves of Mali [turac, suggest that this number is likely to be exceeded in the
there was a growing need for a more intensive investi- future. Considering the shape of the mining works (irre-
gation of this site. This was the major reason for re- gular form of adits) and the huge number of mining
newing investigations at Prlju{a in 2010.7 These studies,
principally focused on geophysical methods, are hoped
to define the extent at which the Prlju{a site was explo- 7 The project of the Archaeological Institute entitled Pro-
ited during prehistory. spection of Mali [turac: an investigation of prehistoric mining is
performed in cooperation with the Museum of RudnikTakovo
region in Gornji Milanovac. Additional co-operators in the project
ABOUT THE LOCALITY are Ana Cicovi}, archaeologist-custodian from the aforementioned
museum and Momir Vukadinovi}, geophysicist from the Institute for
the Development of Water Resources Jaroslav ^erni, Belgrade.
The locality of Prlju{a is situated on the south- The coordinator of the project is Dr. Dragana Antonovi} from the
western slopes of Mali [turac. It is not a typical gossan Archaeological Institute of Belgrade.
8 The word Prlju{a means burned soil. Gossan originated from
deposit appearing as an almost ellipsoidal surface with-
the decay of sulphide mineralisation of iron minerals (pyrite) and
out vegetation8, elongated from the southwest to the
the formation of limonitic products due to the influence of oxygen
northeast. The locality covers a surface of 2.5 ha. It is and electrochemical processes. Due to these processes, the soil is
234 m long (SWNE) and 138 m wide (SENW), rang- sterile and without plant cover.

97
ANTONOVI], VUKADINOVI], Eneolithic mine Prlju{a Mali [turac (95106) STARINAR LXII/2012

hammers with grooves which were found on the surface Geophysical explorations were done using the
along the slope, it can be supposed that these were methods of self-potential SP, electrical scanning
Eneolithic and Bronze Age shafts. ES and seismic profiling with one geophone SGRP.
The Mt. Rudnik area is characterised by a complex The measurements were carried out along the terrain
geological framework.9 It consists of sedimentary, surface delineated by five, 50 m long, parallel sections
igneous and contact-metamorphic rocks. The most at 2 m intervals. The measured points at the sections
abundant sedimentary rocks are Cretaceous flysches. were located at every 2 m, and the sections were oriented
The origin of igneous rocks is explained by volcanic approximately east-west, according to archaeological
activity in the Mt. Rudnik area during the Oligocene and geological facts.
and Miocene. Dacite-andesites were formed in the Self-potential measurements were taken along the
Oligocene, while the majority of volcanic rocks are re- same sections. Vertical, two-dimensional (2D) terrain
presented by quartzlatites which erupted in the Miocene sections were defined by electrical scanning on the
as huge irregular masses and tiny dikes and sills, wide- basis of specific electrical resistivity. These measure-
spread in the Rudnik ore district. Contact-metamor- ments, along with the measurements taken by seismic
phic rocks are represented by weakly metamorphosed, profiling were taken along sections 1, 4 and 5.
marl-clayey sediments, sandstones and conglomerates,
as well as by products of higher grade metamorphic Self-potential method
rocks, such as hornfels and skarn. Intrusions of dacite- The self-potential method (SP) is based on the
andesites and quartzlatites were mainly controlled by investigation of natural electrical currents which spon-
NESW fault systems. taneously appear in material below the terrain surface.
The Rudnik lead-zinc ore deposit is polymetallic The most frequent occurrences of self-potential are
in character and consists of a few tens of ore bodies. connected with the movement (i.e. filtration) of ground
The geological framework of the deposit is composed water through rock masses. The most intensive appear-
of several rock complexes, different in genesis, age and ances of self potential are those related to the presence
composition. Four main groups can be distinguished: of metallic ore deposits, where the self potential is ge-
1. Cretaceous sedimentary complex (sandstone, breccia, nerated as the consequence of electrochemical proces-
conglomerate, marl and limestone), 2. Tertiary igneous ses occurring in response to interactions between ore
rocks (dacite-andesite, quartzlatite and rare granitoid bodies and ground water.10
rocks), 3. Contact-metamorphic complex (hornfels, Due to the simplicity and efficacy in performing
epidosite, marble, garnetite, skarn, etc.); these are the SP method in the field, and because of the speed
important because they host the mineralisation, and 4. with which qualitative data can be obtained, this geo-
Hydrothermal breccia complex; these rocks are signif- physical method was selected for detecting ore occur-
icant because they originated simultaneously to the rences and the prehistoric mining works ,chambers
mineralisation. and shafts, from which the ore was extracted.
The efficacy of this method was practically exam-
ined at the Prlju{a site in 2010. On that occasion, two,
GEOPHYSICAL INVESTIGATIONS 51 m long, sections were allocated one immediately
below and another above Shafts 6 and 4 from 1987.11
Field geophysical investigations at Prlju{a were After having obtained satisfactory results by the SP
conducted in order to detect the spatial distribution of method, it was decided to apply the same method of in-
subsurface mining galleries which, presumably, exist vestigation to the majority of the area above the two shafts.
in this area as well as to plan future excavations. It was,
therefore, decided to first investigate the space above
Shafts 6 and 4 (discovered in 1987) (Fig. 2). Pottery
shards that were found at the access platform of Shaft 9 All information about the geological framework of the Mt.

6 defined the shaft chronologically as belonging to the Rudnik area presented in the paper has been obtained by the cour-
tesy of the Geological Service at AD Rudnik i flotacija Rudnik
Late Eneolithic or Early Bronze age. The fact that the
in Rudnik. The authors are very grateful to geologists of the Service
entrance to the mining channel was discovered is the for their very kind cooperation.
reason to believe it is a real mining shaft. The entrance 10 Vukadinovi} 2011, 79.

was discovered in 1987. 11 Antonovi}, Vukadinovi} 2011.

98
ANTONOVI], VUKADINOVI], Eneolithic mine Prlju{a Mali [turac (95106) STARINAR LXII/2012

Fig. 3. Self-potential map: I) with position of adits in Shafts 4 and 6 discovered in 1987;
II) with position of the zones of anomalies which represent ore deposit or prehistoric mining works
Sl. 3. Karta sopstvenog potencijala: I) sa ozna~enim ulazima u Okno 4 i Okno 6 iz 1987. godine;
II) sa ozna~enim anomalijskim zonama koje predstavqaju orudwewe odnosno potencijalne stare rudarske radove

The SP measurements were conducted at all 5 sec- if the shaft was not opened, the mineralisation remained
tions; the measured points were placed at 2 m intervals hermetically captured in compact and water-imper-
throughout a regular network of 50 x 8 m.12 The meable, highly silicified rocks. In such a scenario,
obtained data were used for constructing a map of atmospheric water cannot enter and self-potential can-
potential isolines in the interval range of -30 to 45 mV not be generated. On the basis of this explanation, it
(Fig. 3). could be concluded that the detection of self-potential
In accordance with the physical laws and explana- anomalies at Prlju{a is most probably related to the
tions of the origin of self-potential in ore-bearing areas, presence of prehistoric pits.
anomaly zones were delineated on the map. These
anomalies are controlled by negative values which in- Electrical scanning measurements
dicate the presence of mineralisation, e.g. the presence Electrical scanning by direct current (Direct
of copper ore bodies. The defined anomaly zones have Current Resistivity Imaging) is a commonly used tech-
a value of -10 to -30 mV (Fig. 3, from orange to red nique in archaeology. It is aimed at getting 2D (two-
colour in the legend). dimensional) terrain models in the form of a vertical
The zones of anomalies defined by positive self- section according to a specific electrical resistivity
potential values represent the zones of compact, inten- parameter.13 The scanning represents a method of
sively silicified rock masses in which there is no mine- investigating a certain area at several depth levels
ralisation. A strong silicification is also inferred by the below the surface in order to detect subsurface struc-
presence of abundant quartz aggregates on the surface. tures (cultural remnants) and to determine their spatial
During the measuring of the self-potential at Prlju{a
it was observed that if, in the zone of mineralisation or
ore deposit, a prehistoric pit was opened, there were 12 In the field, self-potential was measured by milivoltmetre
pre-conditions for the subsequent filling of the shaft with Voltcraft type M3630B. The nonpolarised electrodes are of
atmospheric water. This gave rise to electrochemical Canadian production (Scintrex Company).
processes which generated self-potential. Furthermore, 13 Vukadinovi} 2011, 55.

99
ANTONOVI], VUKADINOVI], Eneolithic mine Prlju{a Mali [turac (95106) STARINAR LXII/2012

position. Moreover, it is possible to use different types This zone is detected at depths of around 5 m and it is
of electrodes, depending on the aims of the investiga- most probably in functional connection with Shaft 6.
tion and the desired outcome. This shaft continues horizontally towards the east, i.e.
At the Prlju{a site, the electrical scanning method towards the central part of the mineralisation zone.
was chosen with the goal of creating spatial 2D detec- Given that there are no known shafts for correla-
tions of ore occurrences and mineralisations, i.e. pre- tion with the results of the electrical scanning, we have
historic mining works, such as mining chambers or adits only reported the position of three anomaly zones that
from where the ore was extracted. have the lowest electrical resistivity in section 4 (Fig. 4).
After having analysed the results obtained by the These anomaly zones could indicate that buried shafts,
self-potential method, we concluded that the electrical not visible from the terrain surface, also exist. The first
scanning should be done along sections 1, 4 and 5 be- zone was detected at depths of around 4 m and between
cause it is along these sections that the most significant the 4th and 15th metre along the section. The second
anomaly zones, indicating the presence of mineralisa- zone was detected at similar depths from the 27th to
tion extracted from prehistoric shafts, were identified. 29th metre along the section. The third zone was found
Along the above mentioned sections, the electrical at depths of 45 m and from the 34th to 43rd metre along
scanning measurements were conducted at intervals of the section.
1 m. The protocol pole-pole was used14, and a penetration Neither of the section 5 prehistoric shafts, whose
of around 10 m was achieved. The data obtained by the position can be used for correlation with the results of
measurements were processed using inversion proce- electrical scanning, are present (Fig. 4). In this section
dures and in this way 2D electrical models were pro- three anomaly zones with low electrical resistivity are
duced (Fig. 4). detected, suggesting that some shafts may be present.
On these 2D electrical models, different geological The first anomaly zone is located from the 16th to 18th
units are detected in the range between 170 and 18 000 metre along the section and runs vertically more than
Ohmm. The lowest values of electrical resistivity cor- 10 m in depth. The second anomaly zone is situated from
respond to copper ore (malachite), while high resistiv- the 25th to 27th metre, reaching depths of around 3 m.
ity values indicate the presence of crystalline and sili- The third anomaly zone is almost vertical and stretches
cified rocks with abundant aggregates of quartz from the 37th to 42nd metre and is as deep as 10 m.
crystals.
Based on the obtained 2D models it is observed Seismic profiling
that copper mineralisation (malachite) is distributed in The SGRP method (Single Geophone Refraction
the form of irregular steep, vertical channels. The cop- Profiling) represents a procedure of seismic terrain map-
per mineralisation has an electrical resistivity from ping with one or two geophones. It relies on methodo-
170 to 300 Ohmm (Fig. 4, dark blue and blue colour in logical approaches used for carrying out refraction or
legend). These channels were created by the deposi- reflective seismic methods. The SGRP method enables
tion of mineralisation during the post-magmatic the detection of the lateral influences of geological ma-
hydrothermal phase. The circulating hydrothermal flu- terials and distinguishing hollowed spaces as decom-
ids leached copper and accompanying elements and pression zones beneath the terrain surface. In addition,
precipitated them along the channels, forming charac- it allows the detection of vertical and subvertical dis-
teristic north-northwest-south-southeast stretching ore continuities between various geological materials cha-
veins. racterised by differences in seismic wave velocities.15
In section 1 (Fig. 4), two large zones with anomalies The application of the SGRP method at Prlju{a was
were distinguished. The first zone is close to the terrain aimed at distinguishing hard and silicified rock masses
surface and stretches between the 12th and 19th metre from disintegrated zones and subsurface cavities from
of the section. It further reaches depths of around 10 m which prehistoric miners had extracted the raw material,
where it continues from the 18th to 34th metre along malachite.
section 1. The aforementioned zone is most probably in
functional connection with Shaft 4 which has a 23 m
long, vertical entrance. The second anomaly zone is 14 The device Geophysical Resistivity & Self Potential Meter
isolated in the right-hand side of the 2D electrical sec- RPM12 IP was used for the measurements.
tion, between the 38th and 48th metre along the section. 15 Vukadinovi} 2011, 43.

100
ANTONOVI], VUKADINOVI], Eneolithic mine Prlju{a Mali [turac (95106) STARINAR LXII/2012

Fig. 4. 2D resistivity sections along the profiles 1, 4 and 5


Sl. 4. Geoelektri~ni preseci po profilima 1, 4 i 5

101
ANTONOVI], VUKADINOVI], Eneolithic mine Prlju{a Mali [turac (95106) STARINAR LXII/2012

Fig. 5. Diagrams of seismic profiling along


the profiles 1, 4 and 5. Position of adits in Shafts 4
and 6 discovered in 1987 is marked on the profile 1
Sl. 5. Dijagrami seizmi~kog profilisawa
po profilima 1, 4 i 5. Ulazi u Okna 4 i 6
iz 1987. godine su ozna~eni na profilu 1

The SGRP method was conducted using a measu-


rement network composed of an excitation point, Tx
and two reception points, Rx1 and Rx2.16 It required
acquiring the variation of seismic waves coming
between the two reception points in order to easily
recognise the influences of lateral lithological changes
or the presence of underground empty spaces, i.e. pre-
historic mining shafts.
The above described seismic profiling was perfor-
med along sections 1, 4 and 5 in order to enable easier
correlation with the results obtained by the self-potential
and electrical scanning procedures.
In section 1 (Fig. 5) two anomaly zones with con-
siderably longer arrival times of seismic waves were
observed. The first zone (from the 18th to 34th metre
along the section) with a maximum of almost 30 ms
(milliseconds) overlaps with the anomaly in electrical
resistivity, which was observed by the scanning method.
The second anomaly zone is characterised by shorter
arrival times, below 15 ms, and corresponds to the ano-
maly in the resistivity parameter, which was observed
from the 38th to 48th metre along section 1.
The SGRP measurements in section 4 (Fig. 5) de-
tected several anomaly zones characterised by longer
arrival times of seismic waves. They are essentially lo-
cated at the beginning and the end of the section. The
first anomaly established by this method is located
from the beginning to the 15th metre along section 4.
The second, smaller anomaly of longer arrival times of
seismic waves is detected around the 20th metre along
the section. It is most probably connected to the elec-
trical resistivity anomaly zone found occurring at depths
of around 8 m. The third and fourth anomaly indications
are situated between the 30th and 35th metre and between
the 45th and 50th metre along the section, respectively.
The identification of anomaly zones obtained by the
SGRP method on this section was difficult due to fre-

16 To measure the arrival time of seismic waves an Iskrascope

LCD type ODU0401 device was used. SN3 geophones with a fre-
quency range of 1014 Hz were used as geophones.

102
ANTONOVI], VUKADINOVI], Eneolithic mine Prlju{a Mali [turac (95106) STARINAR LXII/2012

quent lithological changes and vertical and/or steeply covered during archaeological excavations in 1987).
dipping structures. The entrance of Shaft 4 is covered with loose material
The same SGRP measurements were performed and is not visible today. According to the geophysical
along section 5 (Fig. 5). The estimated arrival times of data, we expect that the shaft stretches vertically down
seismic waves revealed the presence of two anomaly for around 3 m from the surface and then continues
zones situated from the 15th to 20th metre and from the further, forming a huge, almost horizontal, underground
30th to 40th metre along the section. These two anomaly gallery. In fact, it is possible that it is a larger gallery or
zones spatially overlap with the locations of the ano- a crossroad of two vertical mining corridors which are
malies detected by electrical scanning. situated deeper than 10 m from the surface. In this case,
geophysical investigations reaching more than 10 m
would resolve a large anomaly detected in section 1
CONCLUSION where it connects with Shaft 4 (discovered in 1987).
If we combine the results obtained from section 1
The field measurements performed using the geo- with those acquired during the archaeological investi-
physical methods of self-potential, electrical scanning gations in 1987, and apply them to the image provided
and seismic profiling at the Prlju{a site provided some by our geophysical studies in sections 4 and 5, we may
useful data and conclusions. conclude that there were at least six mining shafts
The self-potential method conducted along five whose entrances were situated 6 m and 8 m northwards
sections, each 50 m long, revealed the existence of from shafts 4 and 6, respectively. Some of these works
malachite mineralisation zones inferred from distinc- penetrated 4 m in depth (2D electrical section 4: 4th15th
tive negative values (Fig. 3). The acquired results sug- metre, 27th29th metre, 34th43rd metre, 2D electrical
gest that the observed anomalies are most probably the section 5: 25th27th metre, Fig. 4) while others had depths
result of the presence of subsurface prehistoric shafts. greater than 10 m (2D electrical section 5: 16th18th
These shafts were subsequently filled with atmospheric metre, 37th42nd metre; Fig. 4).
water creating the necessary electrochemical conditions The SGRP seismic profiling confirmed, albeit not
for the formation of self-potential. so precisely, the most probable presence of mining shafts
Electrical scanning performed on sections 1, 4 and at the locations where anomaly zones were distin-
5 gave clearer definitions of the anomaly zones obtained guished according to the parameters of self-potential
by the self-potential method. The results of these two and specific electrical resistivity.
methods enabled the correlation and connection with Numerous pieces of evidence of mining operation
the opened prehistoric Shafts 4 and 6. This gave rise to works testify to intensive mining activity at the Prlju{a
the detection of old mining operations in the whole site. The size of the entrances into some underground
zone covered by geophysical investigations (Fig. 6). galleries led previous researchers to suppose that these
Thus, the reconstruction of mining operations in the shafts originated from periods younger than prehistory,
zone of Shaft 6 was achieved. In section 1, from the 38th possibly even from the Medieval Age. However, the fact
metre to the end of the section and at depths of around that at Prlju{a malachite was found the carbonate
5 m, a mineralisation zone (malachite) was detected. The copper ore which had not been exploited since the Iron
altitude of this almost horizontal zone is 936 m and is Age. i.e. since the mastering of iron melting technology,
directed towards the east. The beginning of the minera- along with the large number of stone mining hammers
lisation zone at the 38th metre overlaps with the entrance located on the surface from the top to the bottom of the
of Shaft 6, which was discovered in 1987 at the eastern locality, especially in the zone below the shafts, addi-
end of the entrance platform. Hence, it could be expected tionally suggest that Prlju{a was a large prehistoric mine.
that the mining channel continues from the entrance The exploitation in the upper parts was carried out dur-
horizontally to the east for the next ten metres, and then ing the late Eneolithic and early Bronze Age, and this is
abruptly goes vertically down. Its termination point is confirmed by typical finds from Shaft 6. We presume
still not known because that part of the shaft was not that in the lower parts of the slope, now covered by
covered by geophysical investigations. thick layers of loose material, the exploitation could
In section 1, from the 18th to 19th metre, there is have been even earlier.
inferred mineralisation which has an almost vertical Due to all these facts, we argue that geophysical
position and is in direct connection with Shaft 4 (dis- investigations should be continued along the whole

103
ANTONOVI], VUKADINOVI], Eneolithic mine Prlju{a Mali [turac (95106) STARINAR LXII/2012

Fig. 6. Correlation of self-potential measurements and 2D electrical scanning along the profile 1
with the prehistoric Shafts 4 and 6 discovered in 1987; on the right is a photo of the entrance to the mining channel
of the Shaft 6 when it was discovered
Sl. 6. Korelacija rezultata merewa sopstvenog potencijala i elektri~nog skenirawa po profilu 1,
sa ozna~enim u Oknima 4 i 6 iz 1987. godine; desno je slika ulaza u Okno 6 iz vremena kada je otkriven

slope. In the upper part, above the mining works dis- self-potential (SP) and electrical scanning (ES) meth-
covered during the 1980s, geophysical investigations ods in using 2D and 3D variants would be most appro-
should be conducted in the zone where the majority of priate and most economic. The self-potential measure-
potential shafts were detected in 2011. Such investiga- ments should cover almost the entire locality and only
tions should also be carried out in the lower parts of the at places revealing negative anomaly values, would
slope as there is a reasonable possibility that old min- the presence of prehistory mining works be proved by
ing works and archaeological objects are hidden below electrical scanning.
the thick layers of loose material. According to the
acquired results and taking into consideration the fore- Translated by Kristina [ari}
seen activities, it is concluded that the application of and Vladica Cvetkovi}

104
ANTONOVI], VUKADINOVI], Eneolithic mine Prlju{a Mali [turac (95106) STARINAR LXII/2012

BIBLIOGRAPHY:

Antonovi} 2002 D. Antonovi}, Copper processing Jovanovi} 1988 B. Jovanovi}, Prqu{a Mali
in Vin~a: new contribution to the thesis about metallur- [turac: praistorijski rudnik bakra i gorskog kri-
gical character of Vin~a culture, Starinar LII (2002), stala na Rudniku, Zbornik radova Narodnog muzeja
2745. XVIII, ^a~ak, 5 12.
Antonovi}, Vukadinovi} 2011 D. Antonovi}, Radivojevi} 2007 M. Radivojevi}, Evidence for
M. Vukadinovi}, Prqu{a Mali [turac: nova early copper smelting in Belovode, a Vin~a culture site
istra`ivawa praistorijskog rudnika na Rudniku, in Eastern Serbia, Unpublished MSc Thesis, London
Na{a pro{lost 12, 3545. UCL Institute of Archaeology, 2007.
Bogosavqevi} 1988 V. Bogosavqevi}, Pokret- Radivojevi} et al. 2010 M. Radivojevi}, T. Rehren,
ni materijal sa praistorijskog rudnika na Malom E. Pernicka, D. [ljivar, M. Brauns, D. Bori}, On the
[turcu, Zbornik radova Narodnog muzeja XVIII origins of extractive metallurgy: new evidence from
(1988), ^a~ak, 1335. Europe, Journal of Archaeological Science 37 (2010),
Derikonji} et al. 2011 S. Derikonji}, M. Radivo- 27752787.
jevi}, E. Pernicka and Th. Rehren, The Vin~a culture [ljivar et al. 2006 D. [ljivar, J. Kuzmanovi}-
mining complex in Jarmovac, southwest Serbia, in A. Cvetkovi}, D. Jacanovi}, Belovode Plo~nik, new con-
Hauptmann, D. Modarressi-Tehrani & M. Prange eds., tributions regarding the copper metallurgy in the Vin~a
International Conference Archaeometallurgy in Europe culture, in: Homage to Milutin Gara{anin (N. Tasi}, C.
III, 29 June 1st July 2011, Bochum, Metalla Sonderheft Grozdanov, eds.), Serbian Academy for Science and Arts,
4, Bochum, 39. Belgrade, 251 266.
Jovanovi} 1982 B. Jovanovi}, Rudna Glava: naj- Vukadinovi} 2011 M. Vukadinovi}, Primena geo-
starije rudarstvo bakra na centralnom Balkanu, Bor, fizike u arheologiji, Kraljevo.
Beograd.

Rezime: DRAGANA ANTONOVI], Arheolo{ki institut, Beograd


MOMIR VUKADINOVI], Institut za vodoprivredu Jaroslav ^erni, Beograd

ENEOLITSKI RUDNIK PRQU[A MALI [TURAC


ARHEOLO[KA I GEOFIZI^KA ISTRA@IVAWA

Kqu~ne re~i. rudnik bakra, geofizi~ka istra`ivawa, eneolit, arheometalurgija, Srbija.

Ime Malog [turca odavno je poznato u arheologiji kao na- prvi metal na ovom tlu postalo je vrlo aktuelno. Do sada je
ziv eneolitskog rudnika bakra. Do wegovog otkri}a do{lo na ovim prostorima pouzdano dokumentovan samo jedan rud-
je jo{ davne 1980. godine tokom rekognoscirawa planine nik iz vremena vin~anske kulture Rudna Glava u isto~noj
Rudnik u okviru jednog novog projekta. Arheolo{ka istra- Srbiji. Nema sumwe da je u to vreme, ali i kasnije, s obzi-
`ivawa vr{ena su od 1981. do 1989. godine. Tada su otkri- rom na intenzivnu metalur{ku aktivnost, bilo vi{e aktiv-
vene ulazne platforme u {est praistorijskih rudarskih nih rudnika na teritoriji Srbije i da je Mali [turac bio
okana. Daqe iskopavawe detektovanih okana nikada nije jedan od wih.
izvr{eno. Malobrojan arheolo{ki materijal prikupqen Lokalitet Prqu{a nalazi se na jugozapadnoj padini
tokom iskopavawa sastojao se od ve}eg broja rudarskih ba- Malog [turca. Re~ je o ne ba{ tipi~nom gvozdenom {e{i-
tova, ulomaka rudarskih alatki od jelenskog roga i malih ru povr{ini bez vegetacije, skoro elipsoidno izdu`enoj
fragmenata keramike iz kasnog eneolita ili ranog bronza- pravcem jugozapadseveroistok. Lokalitet ima povr{inu
nog doba. Kako je u me|uvremenu, tokom proteklih deceni- oko 2,5 ha i prostire se od 882 m u podno`ju do 994,41 m
ja, u~iwen znatan pomak u prou~avawu po~etaka metalurgije nadmorske visine pri vrhu padine. Re~ je o veoma strmoj
na tlu Srbije, pitawe porekla rude bakra od koje se topio padini sa nagibom od 28 do 37. Tokom istra`ivawa 2011.

105
ANTONOVI], VUKADINOVI], Eneolithic mine Prlju{a Mali [turac (95106) STARINAR LXII/2012

godine na delu padine iznad linije koju ~ine okna 4, 5 i 6, like anomalije konstatovane na profilu 1 u delu koji se
otkrivena 1987. godine, konstatovano je 13 kompleksa starih povezuje sa Oknom 4 iz 1987. godine.
rudarskih radova. Prema izgledu rudarskih radova (ulazi Ako se rezultati koji su dobijeni za profil 1 i wiho-
nepravilnog oblika) i velikom broju kamenih batova sa va korelacija sa rezultatima arheolo{kih ispitivawa iz
`lebom, na koje se nailazi na povr{ini do samog vrha pa- 1987. godine primene na sliku dobijenu geofizi~kim sni-
dine, pretpostavqamo da je re~ o eneolitskim i bronzano- mawem za profile 4 i 5, onda se mo`e re}i da je postojalo
dobnim oknima. bar {est rudarskih okana ~iji su se ulazi nalazili na uda-
Geofizi~ka istra`ivawa sprovedena 2011. godine u qenosti od 6 m i 8 m severno od okana 4 i 6. Neki od ovih
zoni, veli~ine 400 m, iznad Okna 4 i Okna 6 dala su neke radova i{li su do dubine od 4 m (2D geoelektri~ni presek
korisne podatke i zakqu~ke. profila 4: 415. metra, 2729. metra, 3443. metra; 2D geo-
Metodom sopstvenog potencijala, du` pet profila po- elektri~ni presek profila 5: 2527. metra; sl. 4), a neki
jedina~ne du`ine od po 50 m, detektovana su mesta odnosno su imali dubinu ve}u od 10 m (2D geoelektri~ni presek
zone orudwewa malahita na osnovu izrazito negativnih profila 5: 1618. metra, 3742. metra; sl. 4).
vrednosti (sl. 3). Prema dobijenim rezultatima mo`e se Primenom seizmi~kog profilisawa metodom SGRP
zakqu~iti da se anomalije najverovatnije stvaraju iskqu- potvr|eno je, mada ne potpuno precizno, najverovatnije po-
~ivo u zoni praistorijskih okana koja su tokom vremena za- stojawe rudarskih okana na mestima gde se pojavquju ano-
puwena i u koje je u{la atmosferska voda, koja je stvorila malijske zone po parametrima sopstvenog potencijala i
neophodan uslov za elektrohemijsko generisawe sopstve- specifi~ne elektri~ne otpornosti.
nog potencijala. O intenzivnoj rudarskoj aktivnosti na Prqu{i svedo-
Geoelektri~no skenirawe izvedeno na profilima 1, 4 i ~e brojni vidqivi stari rudarski radovi. Veli~ina ulaza
5 jasnije je definisalo anomalijske zone dobijene metodom u podzemne galerije kod nekih od wih navela je nekada{we
sopstvenog potencijala (sl. 4). Na osnovu dobijenih rezul- istra`iva~e na pretpostavku da se radi o oknima iz perio-
tata pomo}u ove dve metode sa~iwena je me|usobna korela- da mla|ih od praistorije, mogu}e ~ak iz sredweg veka. Me|u-
cija i veza sa otkrivenim praistorijskim oknima 4 i 6 (sl. tim, ~iwenica da je na Prqu{i va|en malahit, karbonatna
6), a prema tome i detektovawe starih rudarskih radova u ruda, kao i veliki broj kamenih rudarskih batova koji se
celoj zoni obuhva}enoj geofizi~kim istra`ivawima. Tako nalaze svuda po povr{ini lokaliteta, a posebno oko okana,
je izvr{ena rekonstrukcija rudarskih radova u zoni Okna 6. govori u prilog tome da je cela Prqu{a bila jedan veliki
Na profilu 1, po staciona`i od 38. metra do kraja profi- praistorijski rudnik. U gorwim delovima eksploatacija
la, sa dubinom zalegawa od oko 5 m, detektovana zona orudwe- je vr{ena tokom kasnog eneolita i ranog bronzanog doba,
wa (malahit) prostire se ka istoku, nalazi se pribli`no {to je potvr|eno karakteristi~nim nalazima iz Okna 6.
na koti 936 m i ima skoro horizontalni polo`aj. Po~etak Pretpostavqamo da je u dowim delovima padine, sada pre-
zone orudwewa na 38. metru poklapa se sa ulazom u Okno 6, krivenim debelim slojem sipara, i pre ovog vremena mogla
koji je 1987. godine otkriven na isto~nom kraju ulazne biti vr{ena eksploatacija rude.
platforme. Zato se mo`e o~ekivati da se od ulaza rudarski Zato bi svakako trebalo da se nastave geofizi~ka ispi-
kanal horizontalno nastavqa ka istoku narednih deset me- tivawa na celoj padini. U gorwem delu, iznad rudarskih ra-
tara, a da se zatim naglo vertikalno spu{ta u dubinu. Kraj dova otkrivenih tokom osamdesetih godina dvadesetog veka,
mu za sada nije detektovan, po{to taj deo hodnika nije obu- ispitivawe geofizi~kim metodama trebalo bi izvesti u
hva}en geofizi~kim snimawem. zoni ve}ine potencijalnih okana koja su detektovana 2011.
Na profilu 1, na staciona`i od 18. do 19. metra detek- godine. U dowem delu padine bi tako|e trebalo izvr{iti
tovano je prisustvo orudwewa koje ima skoro vertikalni ispitivawa ove vrste, po{to postoji mogu}nost da se ispod
prostorni polo`aj i u neposrednoj je vezi sa Oknom 4 otkri- debelih naslaga sipara kriju stari rudarski radovi i arhe-
venim arheolo{kim iskopavawem 1987. godine. Ulaz u Okno olo{ki objekti. Prema dobijenim rezultatima dosada{wih
4 zasut je siparom i danas se vi{e ne vidi. Prema rezultati- geofizi~kih istra`ivawa na lokalitetu Prqu{a zakqu~e-
ma geofizi~kog ispitivawa o~ekujemo da se okno vertikalno no je da bi u sklopu budu}ih radova primena metoda sopstve-
spu{ta oko tri metra od povr{ine tla, a da zatim prelazi u nog potencijala (SP) i elektri~nog skenirawa (EK) u vari-
skoro horizontalnu prostranu podzemnu galeriju. Zapravo, jantama 2D i 3D bila najsvrsishodnija i najekonomi~nija.
mogu}e je da je re~ ili o ve}oj galeriji ili o ukr{tawu dva Merewem pomo}u metode sopstvenog potencijala prekrio bi
vertikalna hodnika koja se spu{taju dubqe od 10 m ispod se ve}i prostor lokaliteta, a samo na mestima sa negativnim
povr{ine tla. Geofizi~ko ispitivawe sa dubinskim za- anomalijskim vrednostima bi se elektri~nim skenirawem
hvatom ve}im od 10 m dalo bi u ovom slu~aju razre{ewe ve- potvrdilo prisustvo rudarskih radova iz praistorije.

106
JOVAN KOLEDIN
Muzej Vojvodine, Novi Sad

O JEDNOM NOVIJEM NALAZU


IZ GVOZDENOG DOBA U SREMU

UDK: 903.5"6387"(497.113) e-mail: jocako@open.telekom.rs


DOI: 10.2298/STA1262107K Primqeno: 27. februar 2012.
Prilog Prihva}eno: 21. jun 2012.

Apstrakt. Rad se bavi prilozima iz `enskog groba sa lokaliteta Beqwa~a u [idu. Grob je, na osnovu
nalaza sarkofaga u neposrednoj blizini, datiran u kasnoanti~ki period. Samostrelne fibule sa `ivotiwskom
glavicom povijenom unazad, no` povijenog se~iva i niska perli odre|uju ovaj nalaz kao kasnohal{tatski (Ha D3),
iz IV veka pre n. e.

Kqu~ne re~i. starije gvozdeno doba, sremska grupa zapadnobalkanskog kompleksa, sremska kultura,
Bosut IIIc, Ha D3 period.

[ID BEQWA^A, grob G2 Starije gvozdeno doba


jugoisto~na (alpska) hal{tatska kultura
Povod za pisawe ovog ~lanka je otkri}e grobova na
lokalitetu Beqwa~a kod [ida.1 Nedaleko od me- Geografska podela hal{tatske kulture na isto~-
sta na kojem je prona|ena kasnoanti~ka grobnica nu i zapadnu nastala je 1959. godine. Isto~ni hal-
sa sarkofagom otkriven je skeletni grob `enske {tatski krug obuhvata podru~je gorwe Labe, ju`nu
osobe. Priloge u grobu ~ine dve bronzane fibule Moravsku, zapadnu Ma|arsku, Austriju isto~no
na|ene na ramenima, ogrlica od perli na|ena i ju`no od Ina, Hrvatsku i Sloveniju.7
oko vrata, i mali gvozdeni no` na|en uz desnu bu- Prelazni period od kasnog hal{tata do ranog
tinu (sl. 1).2 latena (La Tne, Lt) obuhvata vreme od kraja VI do
Pop-Lazi} je ove dve fibule odredio kao fi- IV veka pre n. e. Naseqavawem keltskih plemena
bule T-tipa (Armbrustfibeln). Pozivaju}i se na na- dolazi do latenizacije isto~ne hal{tatske kul-
laz fibule objavqen u katalogu rimskih fibula ture. Sli~na situacija konstatovana je i u Sremu,
Singidunuma, datirao ih je u period IIIIV veka.3
Za dataciju u IV vek se odlu~io na osnovu vremena
gradwe memorije sa sarkofagom.4 On pretpostavqa
1
da je `ena iz groba bila u slu`bi visokog rimskog Pop-Lazi} 2008.
2
oficira sahrawenog u sarkofagu.5 Ibid., sl. 9
3 Bojovi} 1983, 331, 145, T. XXXIV.
Po na{em mi{qewu, me|utim, grob sa dve fi- 4 Pop-Lazi}, o. c., 170.
bule treba datovati u kraj starijeg gvozdenog doba. 5 Ibid.
Eventualno se mo`e povezati sa nekoliko grobova 6 Usmeno saop{tewe kustosa Radovana Sremca (Galerija
na|enih u {iroj okolini ovog groba, koji nisu is- Sava [umanovi} u [idu), kome zahvaqujem.
tra`eni zbog problema sa vlasnicima zemqi{ta.6 7 Rebay 2002, 110112.

107
Jovan KOLEDIN, O jednom novijem nalazu iz gvozdenog doba u Sremu (107112) STARINAR LXII/2012

c
a b a

G2

Sl. 1. Beqwa~a, grob G2 (Pop-Lazi}, sl. 9)


Fig. 1. Beljnja~a, grave G2 (Pop-Lazi}, sl. 9)

i to u vreme Bosut IIIc horizonta.8 Veliki uticaj na lini Pe~uja (Pcs). Na primerke iz groba G2 naj-
kasnohal{tatsku kulturu u Transdanubiji i Slo- vi{e podse}aju fibule iz `enskog groba 27 i
veniji imala je i tzv. skitska (VekerzugChotin) kul- fragmentovane fibule iz groba 29.13
tura iz Velike ma|arske ravnice (Alfld).9 Importi Druga nekropola je Beremend.14 Er`ebet Jerem
VekerzugChotin kulture javqaju se i na teritoriji povezala je nalaze iz Beremenda sa nalazima iz
,,sremske grupe zapadnobalkanskog kompleksa.10 Srema,15 pripisuju}i ih Panonima. Kod nas su
U posledwe vreme neki autori ovaj termin zamewu- ovakve fibule prihva}ene samo u Sremu i sever-
ju terminom sremska kultura.11

Fibule
Grobne celine kasnohal{tatskog perioda na 8 Popovi} 1981, 11, 3840.
9
teritoriji ju`ne Panonije uglavnom nisu arheo- Vkony 1984, 262263; Kemencei 2002, abb. 2122.
10 Majnari}-Pand`i} 19992000, 2829, T. IV.
lo{ki istra`ene. Nekropole otkrivene u ma|ar-
11 Potrebica, Dizdar 2002, 83 (srijemska kultura zapadnobal-
skoj Barawi pokazale su u kakvom kontekstu se ja-
kanskog kulturnog kompleksa).
vqaju slu~ajni nalazi iz Srema, za koje ne postoji 12 Vinski, Vinski-Gasparini 1962, 271273, 278.
arheolo{ka dokumentacija.12 Po dve fibule i no` 13 Jerem 1968, Fig. 23.

mogu se sresti i u mu{kim i u `enskim grobovima 14 Jerem 1971.

nekropole kod sela Sentlorinc (Szentlorinc), u oko- 15 Ibid., Fig. 9.

108
Jovan KOLEDIN, O jednom novijem nalazu iz gvozdenog doba u Sremu (107112) STARINAR LXII/2012

noj Srbiji, gde E. Jerem sme{ta i centar wihove nog se~iva, duga~ki izme|u 5,6 cm i 16 cm, javqaju
proizvodwe.16 Na osnovu nedavno objavqene nekro- se i u mu{kim i u `enskim grobovima nekropole
pole iz Vinkovaca dokazano je da se nekropole ka- Sentlorinc.27 No`evi du`ine 715 cm na|eni su
kve su poznate u Ma|arskoj javqaju i na ju`nopa- u preko 30 grobova nekropole \epfeld kod Doro-
nonskom tlu.17 U Ba~koj je, osim nekropole slova. Najzastupqeniji su u grobovima sa urnama
Stubarlija,18 u okolini Novog Sada19 nedavno ot- grupe Daq (VIIVI veka pre n. e.),28 a kontinuitet
krivena nekropola sa ~etiri groba. wihove upotrebe pokazuje nalaz iz skeletnog groba
Grobnoj celini iz [ida je geografski najbli- 18, datiran u sam kraj starijeg gvozdenog doba.29 U
`i i po sadr`aju najsli~niji grob 14 sa nekropo- `enskom grobu 14 sa nekropole Vinkovci NAMA
le na lokalitetu robne ku}e NAMA u Vinkovcima prona|en je sli~an no`, du`ine 7,3 cm. Le|a se~i-
(sl. 2). Radi se o dvostrukom ukopu (`ena i dete), va tog no`a su mese~asto povijena, a o{trica mu je
orijentacije JISZ. ravna.30 Za no`eve ove veli~ine pretpostavqamo
U grobu su na|ena dva para fibula sa dugom da su delovi no{we koji su ka~eni na pojas. Zbog
spiralom. Par bronzanih fibula sa zavr{etkom toga ne pripadaju grobnim prilozima.
povijenim unazad (sl. 2/4, 2/7) nalazio se na levoj
i desnoj strani grudnog ko{a. Drugi par je izra|en Ogrlica od perli
od gvo`|a. Jedna fibula se nalazila uz desni la- U grobu G2 na|ena je i niska od 32 plave i
kat, a druga uz levu nadlakticu. Gvozdeni no` sa mle~nobele sferi~ne perlice, pre~nika do 0,6
trnom i 36 plavkastih i `u}kastih perlica (uz cm.31 Niske od perli se, po pravilu, nalaze u `en-
jednu okastu) tipi~ni su delovi no{we kasnog hal- skim i de~jim grobovima. U Ha C1 Ha D1 perio-
{tata Panonije. Grobne priloge predstavqaju bi- du koristio ih je samo gorwi socijalni sloj.32 Wi-
koni~na ~a{a (sl. 2/1)20 i tibija mladog tetreba. hova masovnija produkcija i {irewe upotrebe na
Na glavi de~jeg skeleta na|eno je {est perlica od sve slojeve dru{tva zabele`eni su u Ha D3/LtA pe-
staklene paste.21 riodu hal{tatske kulture.33 U grobovima iz
Fibule su blago raskucanog luka, koji nije Sentlorinca se nalaze brojne jednobojne perle od
ukra{en. Bronzane fibule imaju izdignut i una- staklene paste `ute, plave, zelene, sive, ali i
zad povijen zavr{etak noge, {to ih pribli`ava
isto~noalpskim kasnohal{tatskim zoomorfnim
fibulama.22
Iz okoline [ida poti~e slu~ajan nalaz jednog 16 Ibid., 101.
lokalnog tipa samostrelne fibule XIII varijan- 17 Majnari}-Pand`i} 19992000; Majnari}-Pand`i} 2003.
18 Medovi} 2007.
te.23 Krajem XIX veka je otkriven jedan ratni~ki
19 Materijal se nalazi u Pokrajinskom zavodu za za{titu
grob u Ada{evcima, sa tri fragmentovane fibule
spomenika kulture Vojvodine. Na podacima zahvaqujem kolegi
samostrelne konstrukcije.24 D. An|eli}u, koji ovaj materijal priprema za objavqivawe.
U Sremu su primercima iz groba G2 sa Beq- 20 Autorka isti~e da nije dokumentovano kojem skeletu

wa~e najsli~nije fibule iz Novih Banovaca i Ze- pripada ova posuda. Zabele`eno je da je na|ena desno od loba-
muna.25 Zavr{etak glavice fibule iz Zemuna po- we (Majnari}-Pand`i} 2003, 486).
21 Ibid., 486.
vijen je unazad; na po~etku i zavr{etku luka 22 Ibid., 488.
nalaze se popre~ni urezi, a le|no rebro je blago 23 Vasi} 1999, 101, Taf. 51, 859.
nagla{eno. Na ovoj fibuli kombinovani su ukra- 24 Ibid., 101, Taf. 50, 847849.
si i detaqi sa obe fibule iz Beqwa~e: du`a noga 25 Ibid. T. 51, 863864.

kakva je na prvoj fibuli (sl. 1a), a neukra{en 26 Jerem 1968, 184. Za daqu literaturu vidi: Bele{ka 70.

luk, sa nagla{enim le|nim rebrom na drugoj fi- 27 Ibid., 1968, 183184.

buli (sl. 1b). Fibula iz Novih Banovaca ima luk 28 Trajkovi} 2008, 343. Zapa`eno je da se istovremeno jav-

pravougaonog preseka, koji nije ukra{en. qaju tri varijante, bez mogu}nosti tipolo{ko-hronolo{kog
vrednovawa (o. c. 349).
29 Trajkovi}-Popovi}, 1994, 138, T. I, 1.
No` 30 Majnari}-Pand`i} 2003, Abb. 6, 2.
Gvozdeni no`evi sa trnom i povijenim se~i- 31 Pop-Lazi} 2008, 170, Fig. 9c.

vom se na {irokom prostoru sredwe Evrope javqa- 32 Hornk 2006, 41.

ju u vreme kasnog gvozdenog doba.26 No`evi povije- 33 Ibid. 42.

109
Jovan KOLEDIN, O jednom novijem nalazu iz gvozdenog doba u Sremu (107112) STARINAR LXII/2012

1 3

4 7

Sl. 2. Vinkovci NAMA, grob 14 (Majnari}-Pand`i} 2003, Abb. 6)


Fig. 2. Vinkovci NAMA, grave 14 (Majnari}-Pand`i} 2003, Abb. 6)

110
Jovan KOLEDIN, O jednom novijem nalazu iz gvozdenog doba u Sremu (107112) STARINAR LXII/2012

raznobojne `uto-plave, plavo-bele i ,,okaste.34 wa~e, dragoceni su deli}i za sklapawe mozaika o


Iste vrste (plavih i mle~nobelih) perli na|ene ju`nopanonskim plemenima.
su na nekropoli Stubarlija (u skeletnim grobovi- Ovaj grobni nalaz svakako treba datovati u kraj
ma 3, 4 i 6),35 nekropoli Vinkovci NAMA (u gro- starijeg gvozdenog doba Ha D3 period (horizont
bu 14),36 u grobu sa lokaliteta [abac Jela.37 ^urug), odnosno u period pre doseqavawa Kelta u
Perlice su veoma zastupqene i u sredwopanon- ju`nu Panoniju. Prema nalazima iz Srema i ju`-
skim oblastima (grupa Alfld).38 nopanonskih oblasti, grob G2 iz Beqwa~e treba
odrediti kao celinu iz IV veka pre n. e., a nikako
Datacija mla|u od III veka pre n. e. Wegova datacija u IV vek,
Nedostatak gr~kog importa u malobrojnim zatvo- odre|ena samo na osnovu blizine nalaza kasnoan-
renim celinama sremskog dela Panonije onemogu}a- ti~kog sarkofaga, ne mo`e se metodolo{ki pri-
va wihovu precizniju dataciju. Kne`evski grobovi hvatiti. Svojom horizontalnom stratigrafijom
ranolatenskog (Lt A) perioda nude dobre hronolo- lokalitet Beqwa~a pokazuje da se radi o nekropo-
{ke parametre za datacije samostrelnih ~ertoza li panonskog stanovni{tva starijeg gvozdenog do-
fibula u posledwu ~etvrtinu V veka pre n. e.39 ba i kasnoanti~koj nekropoli iz IV veka. Kontinu-
Razvoj lokalne hronologije starijeg gvozdenog itet kori{}ewa ovog mesta za sahrawivawe tokom
doba je u Sremu jo{ uvek nepoznat. Arheolo{ki is- mla|eg gvozdenog doba i perioda ranog carstva tek
tra`ene zatvorene celine, kakav je grob G2 iz Beq- treba dokazati.

34 Jerem 1968, 186.


35 Medovi} 2007, grob 3: 13, sl. 8/4, T. VII/3; grob 4: 15, sl.
9/2, T. VII/4; grob 6: 18, sl. 11/2, T. VII/5.
36 Majnari}-Pand`i} 2003, 485486, 490, Abb. 6, 3.
37 Vasiqevi} 1976, 169, sl. 2, 7.
38 SzentesVekerzug, grob 53 (Kemencei 2002, Abb. 3, 1619);

Sndorfalva, grob 165 (Kemencei 2002, Abb. 61015).


39 Ibid., 193.

111
Jovan KOLEDIN, O jednom novijem nalazu iz gvozdenog doba u Sremu (107112) STARINAR LXII/2012

SKRA]ENICE:

AAH . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Acta Archaeologica Academiae Scientarum Hungaricae,


Budapest
ARR . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Arheolo{ki radovi i rasprave, Zagreb
FolArcht . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Folia Archaeologica, Budapest
JPM . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Janus Pannonius Muzeum vknyve, Pcs
OA . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Opuscula Archaeologica, Zagreb
PBF . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Prhistorische Bronzefunde, Franz Steiner Verlag,
Stuttgart
PIaZ . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Prilozi Instituta za arheologiju u Zagrebu, Zagreb
SHN . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Studia Historica Nitriensia, Nitra
WAB . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Wissenschaftliche Arbeiten aus dem Burgenland,
Eisenstadt

BIBLIOGRAFIJA:

Bojovi} 1983 D. Bojovi}, Rimske fibule Singidu- Popovi} 1981 D. Popovi}, Keramika starijeg gvo-
numa, Beograd 1983. zdenog doba u Sremu, Beograd 1981.
Hornk 2006 M. Hornk, Prspevok k historickej Potrebica, Dizdar 2002 H. Potrebica, M. Dizdar,
interpretcii sklenench korlikov v severovchodoalp- Prilog poznavanju naseljenosti Vinkovaca i okolice u
skej oblasti hal{tatskej kultry a vo vekerzugskej culture starijem `eljeznom dobu, PIaZ 19, 2002, 79100.
(The report on the historical interpretation of glass beads Rebay 2002 K. Rebay, Die hallstattzeitliche Grab-
in the NorthEastern Alpine region of Halstat culture hgelgruppe von Zagersdorf im Burgenland, WAB 107,
and Vekerzug culture), SHN 13, 3746. Eisenstadt 2002.
Jerem 1968 E. G. Jerem, The late iron age ceme- Trajkovi}, Popovi} 1994 D. Trajkovi}, P.
tery of Szentlorinc, AAH XX, 159208. Popovi}, Doroslovo\epfeld nalazi mla|eg gvo-
Jerem 1971 E. Jerem, Ksovaskori srleletek Be- zdenog doba, u: N. Tasi} (ur.) Kulture gvozdenog doba
remendrol (Baranya Megye), JPM XVI, 6990. jugoslovenskog Podunavqa Simpozijum Sombor
Kemencei 2002 T. Kemencei, Beitrge zur 1993, Beograd 1994, 137147.
Schmuckmode der Alfld-Gruppe Skythischer Prgung, Trajkovi} 2008 D. Trajkovi}, \epfeld nekro-
FolArch XLIXL, 20012002, Budapest 2002, 2973. pola starijeg gvozdenog doba kod Doroslova, Sombor
Majnari}-Pand`i} 19992000 N. Majnari}-Pan- 2008.
d`i}, O pojavi novih tipova konjske opreme iz zavr{nog Vasi} 1999 R. Vasi}, Die Fibeln im Zentralbalkan,
starijeg `eljeznog doba u isto~noj Hrvatskoj, OA 2324, PBF XIV, Stutgart 1999.
Zagreb 19992000, 2738. Vasiqevi} 1976 M. Vasiqevi}, Nalazi sta-
Majnari}-Pand`i} 2003 N. Majnari}-Pand`i}, Ein rijeg gvozdenog doba u [apcu, Starinar n. s. XXVII,
spthallstattzeitliches Grberfeld in Vinkovci (Nordost- 167174.
kroatien) und das Problem eines neuen Phnomens der Vkony 1984 G. Vkony, Zu einigen fragen der
Pferdeausstattung in diesem Gebiet, Germania 812, Hallstattzeit des stlichen Transdanubiens, u: L. Trk
Frankfurt a.M. 2003, 481511. (hrsg.) Hallstatt Koloquium Veszprm 1984, Budapest
Medovi} P. 2007 P. Medovi}, Stubarlija, Novi Sad 1986, 259265.
2007. Vinski, Vinski-Gasparini 1962 Z. Vinski, K.
Pop-Lazi} 2008 S. Pop-Lazi}, Late Roman Necro- Vinski-Gasparini, O utjecajima isto~no-alpske hal{tat-
polis Beljnja~a in [id, Starinar LVIII/2008, 163173. ske kulture i balkanske ilirske kulture na slavonsko-sri-
jemsko Podunavlje, ARR II, 263293.

112
IVANA POPOVI]
Institut archologique, Belgrade

LA NCROPOLE DE LA BASILIQUE URBAINE


SIRMIUM

UDK: 904:726.821(497.113)"03/04" ; 904:739.1/.2"652"(497.113) e-mail: ivpop@eunet.rs


DOI: 10.2298/STA1262113P Recu : le 09. fvrier 2012
Original scientifique article Accept : le 21 juin 2012

Rsum. Lanalyse des ncropoles situe lintrieur et autour de la basilique urbaine, sur les sites 59, 46 et 47 de Sirmium,
o 28 enterrements ont t enregistrs jusqu aujourdhui, a dmontr quil sagissait de tombes de construction identique
contenant un mobilier funraires modeste mais homogne (peignes en os tripartites deux ranges de dents, petites fibules
anses en arbalte en fer, grandes boucles ovales en fer, couteaux en fer et fragments de rcipients en fin verre de couleur
verte). En loccurrence, dans la basilique mme ont t dcouverts deux sarcophages miniatures et 13 tombes en briques.
Ces petits sarcophages, ainsi que certaines tombes miniatures, contenaient des ossements dindividus adultes, ce qui incite
conclure que les dpouilles de dfunts enterrs dans les ncropoles de Sirmium, situes hors de lenceinte de la ville,
ont t transportes quelque moment dans la ncropole nouvellement forme intra muros. La cration de cette ncropole
est lie lrection de la basilique lintrieur de lenceinte de la ville en 426. La ncropole tait en fonction jusqu
la destruction hunnique de la ville en 441.

Mots cls. Sirmium, Basse-Antiquit, ncropole, fibules, boucles, peignes, couteaux.

u cours des fouilles archologiques effectues en briques, de construction identique celles mises au

A en 1978 et 1981 sur le site 59 Sirmium o,


partant dune analyse des sources crites et
des vnements historiques, les restes de la basilique
jour dans lglise mme. Enfin, la mme anne, en 1972,
lexploration partielle dun autre btiment de la Basse-
Antiquit situ sur le site 47, environ 60 m au sud-est
Saint-Dmtrius ont t formellement reconnus1, on a de lendroit o ont t ultrieurement dcouverts les
enregistr lexistence de 15 tombes amnage restes de la basilique, a amen la mise au jour dune
lintrieur et autour de cette glise. Cette petite ncro- ncropole stant dveloppe sur les ruines de la cour
pole comptait deux sarcophages et 13 tombes ralises pave de marbre de cette construction, avec des tombes
en briques. Par ailleurs, six ans avant la dcouverte de creuses partiellement dans les dcombres et partielle-
lglise, au cours de la fouille de faible ampleur dun ment dans le pavement de marbre. A cette occasion on
btiment datant de la Basse-Antiquit sur le site 46, on a
dcouvert au-dessus des ruines de la pice orientale de
cet difice, environ 7 8 m au sud-est de labside de
lglise dgage ultrieurement, trois tombes ralises 1 Popovi} 1987, 95139; Popovi} 1998, 4356.

* Larticle est le rsultat du travail sur le project fond par le Ministre des ducation, science et dveloppement tchnologique de la
Rpublique Serbie La romanisation, lurbanisation et la transformation des centres urbains du charactre civil, militaire et rsidentiel dans
les provinces romaines sur la territoire de Serbie (No 177007).

113
Ivana POPOVI], La ncropole de la basilique urbaine Sirmium (113135) STARINAR LXII/2012

Fig. 1. Ncropole de la basilique Saint-Dmtrius, emplacement des tombes


Sl. 1. Nekropola bazilike Sv. Dimitrija, raspored grobova

a dgag 10 tombes plus ou moins endommages2.


Dans la littrature scientifique a t avance lopinion 2 Les ncropoles des sites 46 et 47 nont pas fait lobjet de

selon laquelle les tombes dgages sur ces trois sites publications, mais elles ont t mentionnes dans la littrature scien-
doivent tre considres comme appartenant une tifique, accompagnes de photographies de plusieurs constructions
tombales, cf. Popovi} 1982, 550, Abb. 6, 7; Popovi}, 1987, 120121,
seule et mme ncropole, forme lintrieur et autour Taf. 2. 23. Les fouilles de sauvetage sur ces sites, organises par
de la basilique Saint-Dmtrius (fig. 1), opinion qui a lInstitut archologique de Belgrade, en coopration avec le Muse
t taye par une analyse minutieuse des vnements du Srem de Sremska Mitrovica et lInstitut pour la protection des mo-
historiques ayant marqu la premire moiti du Vme numents culturels de Sremska Mitrovica, ont t diriges par Vladislav
Popovi}, qui a, sous forme de notes, rsum les donnes provenant
sicle, lesquels ont notamment conduit la pratique de la documentation rdige sur le terrain. Malheureusement, le
denterrements intra muros3. Aujourdhui, dans la travail danalyse de cette documentation est rest inachev.
littrature rcent, on peut trouver lavis que les tombes 3 Popovi} 1982, 550; Popovi}, 1987, 120121.

114
Ivana POPOVI], La ncropole de la basilique urbaine Sirmium (113135) STARINAR LXII/2012

sur le site 47 appartienent une autre ncropole, dalle en marbre, prise, probablement, sur la construction
forme autour dune autre difice du culte, peut-tre la rectangulaire en pierre voisine. Les tombes sont en
basilique de Sainte-Anastasie4, dont lxistance Sir- gnral trs endommages, elles contiennent des osse-
mium est confirme par les sources crites5. Linscrip- ments humains disloqus et, en juger par les mieux
tion indite de Sirmium avec la mention de Sainte- conserves, taient de petites dimensions (1,41 x 0 ,40
Anastasie, la quelle nous allons ddier une tude x 0,31 m (n 1/59) ; 1,00 x 1,30 m (n 6/59) ; 1,27 x
particulire, ne confirme pas cette localisation de la 0,45 m (n 8/59)). Seules deux dentre elles ont livr
basilique de la Sainte. Nous considrons, nanmoins, des objets funraires.
que cette ncropole mrite dtre prsente intgrale- La tombe n 5/59, amnage paralllement au mur
ment, en utilisant pour ce faire les donnes existantes 6 et recouverte dune dalle en marbre, outre quelques
provenant de la documentation rdige sur le terrain. os disloqus, contenait cinq fragments dun peigne en
Ce travail implique galement une analyse du matriel os, un clou en fer, deux fragments provenant de la
funraire, ayant une importance pour la chronologie panse dun rcipient en pierre et deux fragments dune
des enterrements, mais aussi pour la dtermination de fresque de couleur rouge. Le peigne en os mal conserv
la composition ethnique de la population enterre. Il va (C81/78), de dimensions 13 x 6 cm, est en trois parties
de soi que cette analyse, comme toutes les autres fondes avec une range de dents de chaque ct et des cts
sur des donnes plus ou moins prcises provenant de la latraux droits. Les autres objets, trouvs dans cette
documentation rdige sur le terrain, est limite par le fait tombe, ne peuvent tre considrs comme appartenant
quelle se base sur des faits enregistrs il y a plusieurs au mobilier funraire, puisquils y sont probablement
dizaines dannes, et qui, aujourdhui, ne peuvent pas parvenus ultrieurement.
tre vrifis. Devant la tombe n 14/59, dont seule la paroi en
briques situe au sud du mur 56 est conserve, on a
dcouvert des ossements humains disloqus et un peigne
LES TOMBES SITUES LINTRIEUR en os deux ranges de dents cass (C173/78), de
ET AUTOUR DE LA BASILIQUE URBAINE dimensions 10 x 5 cm, du mme type que lexemplaire
(SITE 59) prcdent.
Les peignes dcouverts dans ces deux tombes (fig.
Sur le site 59 on a constat lexistence de 15 tombes, 4a, b) appartiennent au type de peignes le plus rpandu
savoir deux sarcophages et 13 tombes maconnes (fig. pour la Basse-Antiquit, qui apparaissent sur le territoire
2). Un de ces sarcophages, sans couvercle, en pierre de lEmpire au IVme sicle8, mais dont la prsence dans
non-dcore, est entirement conserv (fig. 3), tandis des ncropoles de la valle du Danube nest atteste
que le second, de mme construction et finition, est que dans la seconde moiti du IVme et les premires
fragment et, probablement, disloqu. A en juger par dcennies du Vme sicle9. Cela est notamment con-
les dimensions de lexemplaire conserv entier (1,15 x firme par des trouvailles provenant de Belgrade (Sin-
0,42 m), ces sarcophages taient de petites dimensions. gidunum), Kostolac (Viminacium), Sapaja, ^ezava
Les treize tombes, ralises en briques, prsentent toutes (Castrum Novae), Saldum (Cantabaza), Ravna
approximativement le mme type de construction avec (Campsa), Kostol (Pontes), Mora Vagei, Ravna
leurs parois formes en disposant les briques sur la (Timacum Minus) et ]uprija (Horreum Margi), qui,
tranche ou plat, alors que leur couvercle, de mme pour la majorit dentre elles, sont dates avec
ralis en briques, avait une forme soit horizontale, soit prcision entre 378 et 441/4210 ou qui, comme cest le
de toit deux pentes. Les tombes offrant une telle con-
struction sont trs nombreuses dans les ncropoles
datant de la Basse-Antiquit, et il a t constat quelles
apparaissent Singidunum depuis le dbut jusqu' la fin 4 Jeremi} 2008, 195196.
mme du IVme sicle6, bien que lon rencontre gale- 5 Migne 1960, 11731184.
ment certaines variantes de ce type de construction 6 Pop-Lazi} 2002, 46, 47, 87, 99 ; Pop-Lazi} 2005, 125136.
funraires dans la ncropole datant des Grandes inva- 7 Ivani{evi}, Kazanski 2002, 107, Fig.4.
sions, enregistre dans la rue Tadeu{a Ko{}u{ka dans 8 Petkovi} 1995, 126127, type I.

le Belgrade actuel7. Une tombe situe lintrieur de la 9 Popovi} 1984, 162.

basilique (n 5/59) tait, pour sa part, recouverte dune 10 Petkovi} 1995, 5761.

115
Ivana POPOVI], La ncropole de la basilique urbaine Sirmium (113135) STARINAR LXII/2012

Fig. 2. Tombes lintrieur et autour de la basilique Saint-Dmtrius


Sl. 2. Grobovi unutar i oko bazilike Sv. Dimitrija

cas pour deux exemplaires de Singidunum, t retires usage. Sirmium mme ils sont trouvs dans la couche
dune couche datant de la fin du IVme et du dbut du correspondante aux tombes grmaniques (deuxime
Vme sicle11. Sur les six peignes en os, enregistrs dans moiti du Vme prmire moiti du VIme sicle) dans
la ncropole de la rue Tadeu{a Ko{}u{ka Belgrade, le complexe du palais imprial sur le site 1a15 et
deux exemplaires proviennent des tombes (n 18 et 88)
qui sont rattaches lhorizon Smolin, cest--dire la
priode entre les annes 430/40 et 470/8012. A en juger
par les exemplaires trouvs ct dun squelette dont 11 Bjelajac, Ivani{evi} 1991, 128, fig. 2, 3, 4.
12
le sexe a pu tre dtermin, il semblerait que ces peignes Ivani{evi}, Kazanski, 2002, 121122, 124, pl. III, 18.4 ; VIII,
faisaient partie des accessoires de toilette fminins13. 88.1.
13 Keller 1971, 112, 635a; Ivani{evi}, Kazanski 2002, 129130
A la mme poque, les peignes deux ranges de dents
(tombe n 18), 137138 (tombe n 88).
figurent rarement parmi les trouvailles enregistres 14 Popovi} 1984, 163.
dans des tombes germaniques, hors des frontires de 15 Pejovi}, Lu~i} 2011, 406, 408409, sl. 7. 4b, 6; 8. 1, 2,

lEmpire.14 Les peignes de ce type restent longtemps en 6, 7.

116
Ivana POPOVI], La ncropole de la basilique urbaine Sirmium (113135) STARINAR LXII/2012

0 2 cm a

b
0 2 cm

c
0 1 cm

Fig. 3. Sarcophage miniature (tombe n 12/59) et tombe en briques (tombe n 1/59) lintrieur de la basilique
Fig. 4. Objets funraires provenant des tombes situes dans la basilique:
a) peigne (tombe n 5/59) ; b) peigne (tombe n 14/59); c) boucle doreille (tombe dtruite)
Sl. 3. Minijaturni sarkofag (grob 12/59) i grob od opeka (grob 1/59) unutar bazilike
Sl. 4. Prilozi iz grobova unutar bazilike:
a) ~e{aq (grob 5/59); b) ~e{aq (grob 14/59); c) min|u{a (razoren grob)

Viminacium ils sont bien reprsents dans les tombes de ce type, galement en bronze, a t dcouvert lors
du phase B (deuxime tiers du Vme-commencement de la fouille dune couche datant de la Basse-Antiquit
du VIme sicle) sur les ncropoles Burdelj et Vi{e dans la Ville basse de Belgrade (Singidunum)17, alors
grobalja.16 que sagissant du mobilier des ncropoles de Vimina-
Outre ces deux peignes mal conservs, une boucle cium, des parures de ce type, ralises en diffrents
doreille en bronze fragmente pourrait, elle aussi, mtaux, ont t enregistres parmi les trouvailles data-
reprsenter un objet funraire (C80/78). Elle a t bles de la fin du IIme jusqu la fin du IVme sicle,
trouve ct dossements et dun fragment de crne avec une apparition sporadique dans une tombe ger-
(fig. 4c) et provient probablement dune tombe dtruite, manique du VI sicle18.
situe approximativement la mme cote (79,42 m) Non loin du fragment de crne et des restes de
que le petit sarcophage et la tombe en briques amnage squelette ct desquels a t trouve cette boucle
au nord-ouest de ce dernier (79,79 m et 79,46 m). Cette
boucle doreille, dun diamtre de 1,5 cm, se prsente
sous la forme dun anneau avec extrmit senroulant
en spirale. Vu son mauvais tat de conservation, sa 16 Ivani{evi}, Kazanski, Mastykova 2006, 3536 (tombes n 3,
forme ne peut tre dtermine avec plus de prcision, 8, 29, 32.34, 38, 49, 52, 63, 202, 1420).
bien quil semblerait quil sagisse du type en forme 17 Petrovi} 1997, kat. 156.

danneau avec petite spirale sur la tte. Un exemplaire 18 Zotovi} 1995, 234, 242, type II.

117
Ivana POPOVI], La ncropole de la basilique urbaine Sirmium (113135) STARINAR LXII/2012

doreille, on a enregistr prs du mur 24, dans le carr LES TOMBES SITUES LINTRIEUR
26/27, la prsence dossements humains et de DES RUINES DE LDIFICE
fragments de briques, provenant probablement dune DE LA BASSE-ANTIQUIT (SITE 46)
tombe dtruite. La dcouverte dossements humains et
dune anse damphore dans une partie endommage du Les fouilles archologiques, effectues en mai
mur 8, dans le carr 35, nous indique galement 1972, environ 7 8 m au sud-est de lendroit o, six
quune autre inhumation y a t effectue. annes plus tard, a t identifi le mur de labside de la
Du point de vue chronologique, la cration de cette basilique, ont mis au jour la plus grande partie dun
petite ncropole a, certainement, fait suite la complexe dhabitation du IVme sicle. Il tait
construction de lglise au sein de laquelle elle sest constitu dune pice est et dune pice ouest, toutes
dveloppe. Les fondations de cette basilique reposant les deux avec hypocauste, et spares par un couloir.
sur les restes dun btiment romain antrieur, au- Construit la fin du IIIme sicle en pierre et briques,
dessus desquels on a trouv, enfouie dans les et aprs plusieurs adaptations, comme en tmoignent
fondations de labside, une monnaie dArcadius, date deux niveaux de sol avec colonnettes dhypocauste,
de 394402, son rection est assurment postrieure cet difice a t abandonn probablement vers la fin du
lanne 39419. Par ailleurs, au vu de la prsence dans IVme sicle. Cest lemplacement de sa pice est que
les deux petits sarcophages trouvs dans lglise et, sest dveloppe la ncropole. Trois tombes endom-
probablement, dans certaines tombes en briques, mages ont t ici dcouvertes (fig. 5).
galement de petites dimensions, dossements La tombe n 1/46 (fig. 6) se trouve le long du bord
dindividus adultes, se qui est confirm par des est du mur XIII. Elle est dorientation NOSE et de
analyses du matriel osthologique accessible20, il est dimensions 1,80 x 0,35 m. La fosse tait borde de
permis de penser que ces derniers ont t apports des parois formes par des briques de 6 cm dpaisseur,
ncropoles situes en dehors de lenceinte de la ville21. poses verticalement dans des dcombres pars de
Cela signifie que la basilique et la ncropole amnage briques et de blocs de mortier. A la hauteur des bords
au sein de celle-ci ont vu le jour dans la premire suprieurs de la construction, de chaque ct, ont t
moiti du Vme sicle, en tout cas avant la destruction disposes des briques casses qui largissent les parois de
de la ville en 441/42 par les Huns. Un vnement la fosse. Cette tombe tait recouverte de briques entires
important au cours de cette priode a eu lieu en 424/25, (dim. 41 x 39 cm, paisseur de 4,5 cm) qui formaient
annes ayant vu le rattachement de Sirmium un couvercle deux pentes, avec, son sommet,
lEmpire dOrient et le refoulement des Huns sur la assemblage par rainurage des briques le constituant.
frontire pannonienne le long du Danube. Une analyse Derrire la tte du dfunt a t plante verticalement
minutieuse des sources crites et des monuments une brique de grande taille. Le squelette a t drang,
pigraphiques a dmontr que Sirmium et ses except la partie en dessous du genou, qui gisait in situ
alentours ont servi lanne suivante de base en vue de dans les dcombres. Ces derniers qui remplissaient la
loffensive militaire, mene par Thodose II, en 427, fosse renfermaient des restes dun crne, des os du bras
contre les Huns installs en Pannonie et contre gauche et trois vertbres. Les fragments de cramique
lusurpateur Jean install Ravenne. Simultanment, et les tesselles de mosaque trouvs dans cette tombe y
afin de consolider la frontire nord de lEmpire sont parvenus ultrieurement, lors de son remplissage
dOrient, cet empereur a dtach Sirmium le par des dcombres. Cette tombe ne contenait pas
praefactos praetorio Ilyrici, Leontius, avec mission dobjets funraires.
dy riger une glise ddie au martyr de La tombe n 2/46 (fig. 7) se trouve dans le prolon-
Thessalonique, Dmtrius22. Cest alors quont gement de la tombe n 1/46, en allant vers la partie nord
commenc les inhumations lintrieur de lenceinte du site. Son creusement a partiellement endommag le
de la ville, dans et autour de lglise rige sur des
murs plus anciens. Ce cadre chronologique se voit
galement confirm par le modeste matriel
archologique mobile livr par les tombes de la 19 Popovi} 1982, 550; Popovi}, 1987, 121.
basilique, qui savre caractristique de la priode 20 Miladinovi}-Radmilovi} 2011, 294304.
couvrant la fin du IVme et la premire moiti du Vme 21 Popovi} 1982, 550; Popovi}, 1987, 121.
sicle. 22 Popovi} 1982, 550; Popovi}, 1987, 121.

118
Ivana POPOVI], La ncropole de la basilique urbaine Sirmium (113135) STARINAR LXII/2012

mur XIII. La fosse est borde de parois formes par en dessous du bord de la tranche de fouille. Cette
des briques disposes verticalement, comme pour la tombe na pas livr dobjet funraire.
prcdente. Les dimensions de la fosse sont 1,40 x A lpoque o a commenc le creusement des
0,32 m. Derrire les pieds du dfunt a t plante verti- tombes, ldifice auquel appartenait la pice est tait dj
calement une brique de grande taille (45 x 30 cm, en ruine depuis longtemps. Certains murs mergeaient
paisseur de 5 cm). Le couvercle de la tombe, en forme encore au-dessus des ruines, ce qui fait que les tombes
de toit deux pentes, a t dtruit pour sa majeure partie. 1/46 3/46 amnages le long du mur priphrique est
Lintrieur de la tombe tait rempli de dcombres, et XIII, ont t en partie encastre sous ce mur (tombe
plusieurs ctes et vertbres taient concentres dans la 1/46), ou lont partiellement endommag (tombe 2/46).
partie sud. Les tessons cramiques trouvs dans cette Les briques utilises pour la construction des tombes ont
tombe y sont parvenus avec les dbris. Son mobilier t reprises sur les ruines de ldifice, celles qui taient
funraire tait vraisemblablement constitu par des brises ayant servi pour former le bord suprieur des
rcipients en verre, savoir des verres hmisphriques, parois verticales, alors que celles qui taient entires ont
conservs sous forme de petits fragments (de dim. cca t mise profit pour la construction de ces dernires
1,5 cm) de leurs bords et de leur panse en verre iris de ainsi que la formation du fond et du couvercle (fig. 8).
couleur verte (C 53/72). Ces tombes ont t endommages et les squelettes
La tombe n 3/46 se trouve lest de la tombe 2/46, drangs. Elles ne contenaient pas dobjets funraires,
en direction du bord nord de la tranche de fouille. Elle except peut-tre des rcipients en verre, conservs
est de construction identique aux deux prcdentes. sous forme de petits fragments (fig. 9) dans la tombe
Seule la partie est a t fouille, alors que la plus grande 2/46. De telles cuelles hmisphriques ou semi-
partie, ouest, de la construction funraire se se trouve elliptiques fine paroi en verre de couleur verte ont

Fig. 5. Tombes dans la pice est de ldifice datant de la Basse-Antiquit (site 46)
Sl. 5. Grobovi u isto~noj prostoriji kasnoanti~kog objekata (lokalitet 46)

119
Ivana POPOVI], La ncropole de la basilique urbaine Sirmium (113135) STARINAR LXII/2012

Fig. 6. Aspect de la tombe n 1/46 Fig. 7. Aspect de la tombe n 2/46


Sl. 6. Izgled groba 1/46 Sl. 7. Izgled groba 2/46

0 10 20 m

a b

Fig. 8. Coupes des tombes 1/46 (a) et 2/46 (b)


Sl. 8. Preseci grobova 1/46 (a) i 2/46 (b)

dj t enregistres Sirmium, mais galement sur face rduite au-dessus de briques disposes sur la tranche.
dautres sites pannoniens, et elles sont dates de la Cet enduit est lui-mme recouvert dune couche dargile
priode allant du milieu-seconde moiti du IVme sicle tasse, avec restes de suie et de cendres certains
jusquau milieu du Vme sicle23. Le terminus post quem endroits. Ce sol A appartient probablement un habitat
pour la formation de la ncropole est la date de la de- organis dans les ruines du btiment avec hypocauste,
struction de ldifice, qui, daprs les trouvailles mais, ce quil semble, une priode faisant suite la fin
montaires, est postrieure lanne 378. Dautre part, de lutilisation de la ncropole, puisque ce sol recouvre
dans la partie centrale de la pice est, on a constat la partiellement la tombe 3/46. Aucune trouvaille mobile
prsence dune petite surface de sol, de forme
irrgulire, recouverte dune fine couche de mortier
friable, dune paisseur de 210 cm. Ce mortier a t
dpos par dessus des dcombres nivels, sur une sur- 23 [aranovi}-Svetek 1986, 1213, Taf. II, 12, 4.

120
Ivana POPOVI], La ncropole de la basilique urbaine Sirmium (113135) STARINAR LXII/2012

Sur les ruines de la cour pave de marbre du


btiment dhabitation sest dveloppe une ncropole,
comprenant des tombes en partie amnages dans les
dcombres et en partie creuses dans le pavement de
marbre. Au cours des travaux archologiques, 10
tombes ont t nettoyes (fig. 10).
La tombe 1/47 (fig. 11, 19a) est dispose le long
du bord intrieur du mur I, une profondeur de 1,39 m.
Oriente dans le sens NOSE et de dimensions 1,00 x
0,570,63 m, elle a t ralise en briques, avec un
couvercle formant un toit deux pentes. Celui-ci est
constitu de briques entires (44 x 31 x 6 cm), au-
dessus desquelles ont t disposes, faisant office de
0 1 2 cm renfort, des briques casses ou entires de diffrentes
dimensions. Derrire la tte et les pieds du dfunt, une
brique de grandes dimensions a t dispose sur la
Fig. 9. Coupe hmisphrique en verre provenant
tranche. Lintrieur de la tombe tait rempli de terre
de la tombe 2/46
sombre non-tasse et son nettoyage a uniquement livr
Sl. 9. Hemisferi~na staklena ~a{a iz groba 2/46 quelques petits os. Le fond de la tombe tait dall
laide de trois briques poses horizontalement. Cette
tombe ne contenait pas de mobilier funraire.
na t enregistre sur ce sol. Les rparations des murs La tombe 2/47 (fig. 12, 19 b) est situe au sud de
priphriques de la pice est, effectues sec avec des la tombe 1/47 et oriente dans le mme sens. Elle re-
briques fragmentes, sont en relation avec lhabitat posait sur le pavement de marbre et on na uniquement
auquel appartient le sol A. Les dcombres situs constat la prsence dos provenant dun crne et dun
immdiatement sous le niveau du sol A ont livr, une squelette drangs. Except des fragments de verre fin
profondeur de 1,80 m, des tessons de cramique et une de couleur verte, cette tombe ne contenait pas dautres
monnaie en bronze du IVme sicle, entre autre de objets funraires.
Valentinien et Valens, 364378 (C5/72) et de La tombe 3/47 (fig. 13, 19c) se trouve sous la tombe
Honorius, 410423 (C4/72), qui permet de mieux 1/47, creuse dans une terre sombre ou concasse, dans
cerner lpoque laquelle les tombes en briques ont une partie endommage du pavement de marbre, une
t amnages sur les ruines du btiment antique. profondeur de 2,05 2,10 m. Les bras du squelette sont
serrs contre le torse et les jambes lune contre lautre.
Le mobilier funraire comprenait une fibule anse en
LES TOMBES DGAGES arbalte en fer corrod (C172/72), un petit objet en
DANS LES RUINES DE LA COUR fer (C173/72), une perle en pte de verre (C175/72),
PAVE DU BTIMENT DATANT un clou en fer (C177/72), une pice de monnaie de
DE LA BASSE-ANTIQUIT (SITE 47) Valens (364/67) trouve dans le crne du squelette
(C178/72) et deux pices de monnaies, dont une de
A environ 60 m au nord-est de la basilique Saint- Thodose Ier (385/95), trouve ct du squelette.
Dmtrius et des tombes du site 46, , on a procd en La tombe 4/47 (fig. 14, 19d) a t creuse dans le
juin-juillet 1972 la fouille du site 47. A cette pavement de marbre et elle est approximativement
occasion, a t mis au jour un difice avec vaste cour oriente NOSE. On a not la prsence de restes dune
pave de marbre. Sa construction avait probablement construction en briques et de plusieurs os de squelette
dbut la fin du IIIme sicle, et il a atteint son plein et de crne parpills. Cette tombe ne contenait pas de
dveloppement, stant accompagn dadaptations mobilier funraire.
mineures, dans les premires dcennies du IVme La tombe 5/47 (fig. 15, 19e) est oriente NOSE.
sicle. A en juger par les trouvailles montaires Une partie de sa construction en briques, dune
enregistres sur le sol B, Sa destruction est ultrieure longueur de 2,00 et dune largeur de 0,70 m, recouverte
lanne 379. de son couvercle deux pentes, est conserve. Son sol

121
Ivana POPOVI], La ncropole de la basilique urbaine Sirmium (113135) STARINAR LXII/2012

Fig. 10. Tombes dans la cour pave de ldifice datant de la Basse-Antiquit (site 47)
Sl. 10. Grobovi u poplo~anom dvori{tu kasnoanti~kog objekta (lokalitet 47)

est constitu par le pavement de marbre de ldifice plus La tombe 8/47 (fig. 18, 19h) a t dcouverte
ancien. Elle renfermait un squelette dhomme bien louest du mur 5. Elle tait trs endommage, et les
conserv aux bras disposs le long du corps et la tte ossements humains taient parpills.
tourne vers la droite. Des fragments de cramique, de La tombe 9/47 (fig. 19i), trs endommage,
verre et des scories de plomb sont probablement par- atteste par la prsence dossements humains
venus dans cette tombe ultrieurement, avec dautres dcouverts entre les tombes 5/47 et 6/35.
dcombres. Son mobilier funraire tait constitu par La tombe 10/47 (fig. 19j), oriente dans le sens
un couteau en fer corrod (C261/72) et une boucle NOSE, atteste par la prsence de restes de crne, de
ovale en fer (C262/72), alors que dans la zone du vertbres et de ctes, reposant sur une fine couche de
bassin taient disposs trois petits fragments dun suie et de cendres, sous le sol secondaire A. Cette tombe
objet en fer, peut-tre des restes dun couteau. ne contenait pas de mobilier funraire.
La tombe 6/47 (fig. 16, 19f) est oriente NOSE. Les tombes de cette ncropole sont trs endom-
Sa construction en briques entires et casses avec un mages et, le plus souvent, dnues de toute construc-
couvercle deux pentes, dune longueur de 1,30 m et tion (fig. 19). Dans les cas o cette dernire a t
dune largeur de 0,55 m est partiellement conserve. conserve, (tombes 1/47, 5/47, 6/47), lorientation, le
Son sol est constitu par le pavement de marbre de mode de construction et les matriaux utiliss, indiquent
ldifice plus ancien. Le squelette est trs endommag quil sagit dune ncropole contemporaine de celle du
; le crne a disparu, les os des jambes sont replis, le site 46. Les autres constructions constates proximit
main droite tait dpose sur le bassin, et le bras de ces tombes fours et habitations en matriau lger,
gauche le long du corps. Le mobilier funraire tait identifies grce aux empreintes de pieux ronds
constitu par une boucle en fer rectangulaire (C257/72) appartiennent probablement lhorizon suivant, cest-
et deux anneaux en fer (C258/72 et C258a/72), pro- -dire aux habitats germaniques forms aprs les
venant peut-tre dun couteau. dvastations causes par les Huns en 441/4224. tant
La tombe 7/47 (fig. 17, 19g) est situe lest du
mur 5. Elle contenait plusieurs os humains drangs,
ainsi quun manche de couteau en os dorigine animal
(C233/72). 24 Popovi} 1982, 552.

122
Ivana POPOVI], La ncropole de la basilique urbaine Sirmium (113135) STARINAR LXII/2012

Fig. 11. Aspect de la tombe n 1/47 Fig. 12. Aspect de la tombe n 2/47
Sl. 11. Izgled groba 1/47 Sl. 12. Izgled groba 2/47

Fig. 13. Aspect de la tombe n 3/47 Fig. 14. Aspect de la tombe n 4/47
Sl. 13. Izgled groba 3/47 Sl. 14. Izgled groba 4/47

donn que dans les fours on a trouv qutare exemplaires 3/47, 5/47, 6/47 et 7/47, est particulirement important
du monnaie du Thodose I de 383/392, il est possible que pour la datation de cette ncropole. Les objets typo-
les fours sont contemporains avec les tombes. Dailleurs,
les fours ont t remarqus aussi sur la priphrie de la
ncropole est de la ville (site 67).25 Le mobilier funraire,
bien que rare et enregistr uniquement dans les tombes 25 Milo{evi} 1986, 115116.

123
Ivana POPOVI], La ncropole de la basilique urbaine Sirmium (113135) STARINAR LXII/2012

logiquement plus sensibles figurant dans ce matriel qui apparaissent parmi les trouvailles dates de la priode
funraire, savoir deux boucles et une fibule, ont t de la Basse-Antiquit. Toutefois, de telles trouvailles
fabriqus en fer. sont rares dans le limes rhnan, alors quelles sont
La boucle C262/73, dont la pointe na pas t enregistres en grand nombre dans les ncropoles de
conserve, de dimensions 4 x 3 cm (fig. 20. 5. 1) Pannonie26, dans les tombes des sites de Keszthely27,
appartient au type des grandes boucles ovales en fer, Pilismart28 et Szny (Brigetio)29. Ces trouvailles et

Fig. 15. Aspect de la tombe n 5/47 Fig. 16. Aspect de la tombe n 6/47
Sl. 15. Izgled groba 5/47 Sl. 16. Izgled groba 6/47

Fig. 17. Aspect de la tombe n 7/47 Fig. 18. Aspect de la tombe n 8/47
Sl. 17. Izgled groba 7/47 Sl. 18. Izgled groba 8/47

124
Ivana POPOVI], La ncropole de la basilique urbaine Sirmium (113135) STARINAR LXII/2012

dautres trouvailles pannoniennes de boucles en fer ont apparaissent parfois aux cts de boucles de ceinture
t rattaches aux lments ethniques germaniques romaines de la premire moiti du Vme sicle (Nitra-
prsents en Pannonie au cours du IVme sicle30. Cette Mikov)37 ou aux cts de fibules datant de la Basse-
affirmation ne bnficie toutefois daucun argument Antiquit avec pied attach (Bratei, tombe n 52)38. Les
convaincant et, dfaut dun plus grand nombre de exemplaires provenant de trois tombes (n 75, 99, 105)
trouvailles formellement attribuables des milieux de la rue Tadeu{a Ko{}u{ka Belgrade (Singidunum)39
germaniques, ne saurait tre soutenue avec certitude. sinsrent par leur typologie et chronologiquement dans
La boucle de Sirmium nous tant parvenue sans sa un plus large horizon de trouvailles funraires (D2/D3,
pointe, son apparence exacte ne peut tre dtermine. 430/440440/450), incluant de telles fibules, mais qui,
Des six boucles ovales en fer, dcouvertes dans la compte tenu de lexistence danalogies enregistres
ncropole de la rue Tadeu{a Ko{}u{ka Belgrade (Sin- sur le vaste espace balkanique et danubien, ne peut tre
gidunum), trois ont t trouves dans des tombes (n prcisment dfini dun point de vue ethnique ou cul-
18, 88, 105) rattaches un horizon denterrements turel. Il est important de mentionner que dans la tombe
dat entre 430/40 et 470/80.31 Les exemplaires de n 105 de cette ncropole, outre une petite fibule anse
tombes 4, 9 et 36 tmoignent sur lusage de ce type du en arbalte, on a galement dcouvert une boucle ovale
boucle la fin du Vme et au commencement du VIme en fer40. Sur les ncropoles Burdelj et Vi{e grobalja
sicle32. Viminacium, dans les tombes sur les ncro- Viminacium les fibules de ce type ont t enregistrs
poles Burdelj et Vi{e grobalja, les boucles ovales ont t dans les tombes n 3, 28 et 2083, les deux prmires
trouves dans les tombes (n 7, 14, 24, 28, 29, 63, 1318) dates en 430/440 440/450.41
du phase B (deuxime tiers du Vme-commencement Les autres objets funraires provenant de cette
du VIme sicle)33. ncropole-ci consistent en de petits fragments de rci-
La boucle rectangulaire, C257/72, dune longueur pients en verre, des perles en pte de verre de couleur
de 5,5 cm (fig. 20. 6. 1), provenant de la ncropole verte, C175/72 (fig. 20. 3. 2), un clou en fer C177,
dgage sur le site 47 Sirmium, diffre typologique- dune longueur de 4,3 cm (fig. 20. 3. 3), un couteau en
ment des boucles rectangulaires trouves dans des fer C261/72, de dimensions 10 x 1,4 cm (fig. 20. 5. 2)
tombes gpides dans le valle de Danube, et comme et des fragments de ce mme type de couteau (manche
elle a perdu sa pointe, laspect de cette boucle nest pas en os, C233/72, de dimensions 7 x 2,5 cm ; deux
tout fait clair. Lexemplaire enregistre dans la tombe
n 41 de la ncropole de la rue Tadeu{a Ko{}u{ka
Belgrade (Singidunum), fabrique il est vrai en bronze,
nest pas date34. Une boucle rectangulaire plate 26 Keller 1971, 75.
27 Sgi 1960,187, tombes n 12, 22, 24, 27, 36, 46.
ardillion masif, aussi en bronze, date la deuxime
28 Barkczi 1960, 111.
moiti du Vme-commecement du VIme sicle, a t
29 Barkczi 1961, 95.
mis jour dans la tombe n 152 de la ncropole Vi{e 30 Keller 1971, 7576.
grobalja Viminacium35. 31 Ivani{evi}, Kazanski 2002, 116, 124, pl. III, 16. 2; VIII, 88.
La petite fibule en fer C172/3, dune longueur de 3; IX, 105. 3.
5 cm, hauteur de larc de 2,5 cm (fig. 20. 3. 1), est ce 32 Ivani{evi}, Kazanski 2002, 116, 124, pl. I, 4.1; II, 9.1; IV,

point corrode quil nest gure possible de dterminer 36.1.


33 Ivani{evi}, Kazanski, Mastykova 2006, 2021, fig. 10.
sa forme de facon prcise. Il semblerait quelle appar-
127, pl. 1, 7.1; 2, 14.9; 4, 24.9; 4, 28.2; 5, 29.10; 9, 63.5; 36,
tienne au type des petites fibules anses en arbalte,
1318.3.
confectionnes le plus souvent en fer, et caractrises 34 Ivani{evi}, Kazanski 2002, 118, pl. IV. 41.2.
par un arc haut, un pied court et un porte-aiguille de 35 Ivani{evi}, Kazanski, Mastykova 2006, 23, 121, fig. 12. 6,

mme longueur que le pied. Les fibules de ce type sont pl. 31. 152.3.
36 Tejral 1988, 227, 231, 283, 286.
frquentes sur les sites dans les Balkans et dans le
37 Pieta 1999, fig. 1. 13.
bassin danubien central. Bien que les dbuts de leur fa-
38 Harhoriu 1998, pl. 88. 52. 5, 6.
brication soient peut tre de date quelque peu antrieure,
39 Ivani{evi}, Kazanski 2002, 114115, pl. VI, 75. 1, 2; VIII,
cette dernire a atteint sa plus forte intensit dans la
99. 2; IX, 105.2.
priode entre 380/400 et 440/50, alors que ces parures 40 Ivani{evi}, Kazanski 2002, 139, pl. IX, 105. 3.
sont sorties dusage entre 430/40 et 470/8036. Il est 41 Ivani{evi}, Kazanski, Maszykova 2006, 18, fig. 9 16, pl.1,

important de remarquer que des fibules de ce type 3.1; 4, 28.1.

125
Ivana POPOVI], La ncropole de la basilique urbaine Sirmium (113135) STARINAR LXII/2012

a c

d e f

g h i j

Fig. 19. Position des squelettes ou des ossements drangs et coupe des constructions tombales conserves sur le site 47 :
a) tombe 1 ; b) tombe 2 ; c) tombe 3 : 1. fibule ; 2. perle ; 3. clou ; d) tombe 4 ; e) tombe 5 : 1) boucle ; 2) couteau ;
f) tombe 6 : 1. boucle ; 2. anneau ; 3. couteau ; g) tombe 7 : 1. manche de couteau ;
h) tombe 8 ; i) tombe 9 ; j) tombe 10
Sl. 19. Polo`aj skeleta ili rasturenih kostiju i presek o~uvanih grobnih konstrukcija na lokalitetu 47:
a) grob 1; b) grob 2; c) grob 3: 1. fibula; 2. perla; 3. klin; d) grob 4; e) grob 5: 1. pre|ica; 2. no`;
f) grob 6: 1. pre|ica; 2. alka; 3. no`; g) grob 7: 1. dr{ka no`a; h) grob 8; i) grob 9; j) grob 10

126
Ivana POPOVI], La ncropole de la basilique urbaine Sirmium (113135) STARINAR LXII/2012

anneaux, C258, 258a/72, de dimensions 3,3 x 3,3 cm ce qui vient confirmer ce mme cadre chronologique.
et 3,5 x 2 cm) (fig. 20.7 ; .6. 2, 3), par consquent un Le peigne en os C246/72, provenant de la couche
matriel qui ne peut servir dindicateur fiable pour la associe la ncropole analyse, se range, donc, parmi
dtermination chronologique ou ethnique des enterre- les objets qui, en se fondant sur les trouvailles montaires
ments enregistrs. Toutefois, les menus dbris prove- ou le matriel caractristique complmentaire, tels que
nant de la cote correspondant au fond des tombes, soit les garnitures de ceintures dcores en taille biseaute,
une profondeur de 1,401,60 m, renfermaient de les fibules cruciformes ornes de la mme facon, les
nombreux objets, probablement en relation avec ces fibules en bronze et en fer avec pied attach ou les
tombes, bien quils naient pas toujours t dcouverts coupes en verre fin de couleur verte, peuvent tre dats
ct dun squelette. On remarque notamment un grand de la priode entre cca lanne 385 et 341/4247.
nombre de couteaux en fer (C32/72, 62/72, 93/72, Outre le matriel archologique de petite taille
143/72, 148/72, 184/72, 190/72, 194/72, 229/72 i mentionn, on note toute limportance pour la datation de
249/72) (fig. 21. 18) et de peignes en os (C42/72, la ncropole du fait que la tombe 10/47 a t dcouverte
60/72, 165/72, 215/72, 264/72) (fig. 22. 15). Les en dessous du sol secondaire A de ldifice, lequel sol
couteaux constituent des dcouvertes frquentes dans est associ aux murs de la plus rcente phase de con-
les ncropoles datant de la Basse-Antiquit, et en Pan- struction de ce btiment. La monnaie dcouverte parmi
nonie ils sont galement enregistrs dans des tombes les dcombres, la cote correspondant au fond des
fminines42. Dautre part, des peignes tripartites deux tombes, est certainement lie la ncropole, et permet de
ranges de dents ont t trouvs dans deux tombes fixer le terminus post quem du creusement des tombes :
situes lintrieur de la basilique Saint-Dmtrius et, Arcadius, 383/92 (C105/72), Thodose I, 383/95
comme cela a t constat lors de leur analyse, ce type (C107/72), Arcadius, 393/95 (C121/72), Thodose I,
de peignes est trs rpandu parmi les trouvailles danu- 378/83 (C182/72) et Thodose II, 435/50 (C185/72).
biennes de la fin du IVme et du premier quart du Vme
sicle. Dans la ncropole de la rue Tadeu{a Ko{}u{ka ***
Belgrade (Singidunum), des peignes de ce type ont t
dcouverts aux cts de grandes boucles ovales en Lanalyse des ncropoles situe lintrieur et autour
fer dans les tombes n 18 et 88, dates de la priode de de la basilique urbaine, sur les sites 59, 46 et 47 de
330/40 370/80, ce qui est un renseignement important, Sirmium, o 28 enterrements ont t enregistrs jusqu
puisquon sait quune telle boucle a t dcouverte aujourdhui, a dmontr quil sagissait de tombes de
dans la tombe 5/47 de la ncropole de Sirmium sur le construction identique contenant un mobilier funraires
site 47. Toutefois, alors que tous les peignes en os modeste mais homogne. En loccurrence, dans la
mentionns appartiennent au type avec cts courts basilique mme ont t dcouverts deux sarcophages
droits, lexemplaire C246/72 (fig. 23.1)43 a des cts miniatures et 13 tombes en briques poses horizontale-
courts dcoups. Les peignes de ce type sont galement ment ou dresses sur la tranche. Ces petits sarcophages,
dorigine romaine, leur fabrication commencant au ainsi que certaines tombes miniatures en briques situes
cours du dernier tiers du IVme sicle, alors quils sont lintrieur de la basilique et sur le site 47 (n 2/59,
encore en usage dans la premire moiti du Vme 7/59, 14/59, 1/47, 6/47), contenaient des ossements
sicle. Ils se propagent partir des parties occidentales dindividus adultes, ce qui incite conclure que les
de lEmpire en direction de lest, en englobant, ce dpouilles de dfunts enterrs dans les ncropoles de
faisant, lespace gographique qui correspond aux
provinces frontalires occidentales avant le partage de
lEmpire44. De nombreuses trouvailles provenant du
limes danubien, en loccurrence de ^ezava (Castrum 42 Keller 1971, 7879, note 451.
Novae), Ravna (Campsa), Veliki Gradac (Taliata), 43 Bien que cela ne soit pas prcisment mentionn, nous
Karata{ (Diana), Kostol (Pontes), Mihailovac i Mora pensons que le dessin de ce peigne a t prsent lors de lanalyse
dun tui de peigne en os de Cari~in Grad, cf. Popovi} 1984, 162,
Vagei, appartiennent justement lhorizon qui couvre
fig. 175.
la fin du IVme et les premires dcennies du Vme 44 Popovi} 1987,137148, Abb. 14.
sicle45. Un peigne de ce type, provenant de la tombe 45 Petkovi} 1995, 2324.

de Baljevac, lintrieur de la Serbie actuelle, a t 46 Petrovi}, 257259, sl. 4, a, b.

dcouvert avec une monnaie dArcadius (395408)46, 47 Petkovi} 1995, op. cit., 23.

127
Ivana POPOVI], La ncropole de la basilique urbaine Sirmium (113135) STARINAR LXII/2012

0 1 2 cm

Fig. 20. Objets funraires provenant des tombes du site 47 :


3. 13 : fibule en fer, perle en verre ; clou en fer (tombe 3) ; 5. 12 : boucle ovale en fer, couteau en fer (tombe 5) ;
6. 13 : boucle rectangulaire en fer, anneaux en fer (tombe 6) ; 7 : manche de couteau en os (tombe 7)
Sl. 20. Prilozi iz gobova na lokalitetu 47:
3. 13: gvozdena fibula, staklena perla, gvozdeni klin (grob 3); 5. 12: ovalna gvozdena pre|ica, gvozdeni no`
(grob 5); 6. 13: pravougaona gvozdena pre|ica, gvozdene alke (grob 6); 7: ko{tana dr{ka no`a (grob 7)

128
Ivana POPOVI], La ncropole de la basilique urbaine Sirmium (113135) STARINAR LXII/2012

0 1 2 cm
8

Fig. 21. Couteaux provenant de la couche associe aux tombes sur le site 47 :
1. C184 ; 2. C62 ; 3. C190 ; 4. C32 ; 5. C143 ; 6. C229 ; 7. C194 (manche de couteau) ;
8. C93 (manche de couteau)
Sl. 21. No`evi iz sloja koji se vezuje za grobove na lokalitetu 47:
1. C184; 2. C62; 3. C190; 4. C32; 5. C 143; 6. C229; 7. C194 (dr{ka no`a);
8. C93 (dr{ka no`a)

129
Ivana POPOVI], La ncropole de la basilique urbaine Sirmium (113135) STARINAR LXII/2012

0 1 2 cm
1

2 3

4 5

Fig. 22. Peignes provenant de la couche associe aux tombes sur le site 47 :
1. C246 ; 2. C165 ; 3. C42 ; 4. C60 ; 5. C215
Sl. 22. ^e{qevi iz sloja koji se vezuje za grobove na lokalitetu 47:
1. C246; 2. C 165; 3. C42; 4. C60, 5. C215

Sirmium, situes hors de lenceinte de la ville, ont t Thodose II contre les Huns en Pannonie et contre
transportes quelque moment dans la ncropole lusurpateur Jean Ravenne. La ville, dont les btiments
nouvellement forme intra muros. La cration de cette reprsentatifs avaient dj t dtruits lors dune des
ncropole est assurment lie lrection de la basilique incursions barbares survenues aprs 378, a vu une
lintrieur de lenceinte de la ville. En se fondant sur modification de sa structure urbaine, avant tout du fait
lanalyse des sources crites et des vnements histori- de la perte de fonction du rempart nord et du repli de la
ques, il a t formellement dmontr que la construction ville sur la partie sud du territoire municipal. Le menace
de lglise municipale, ddie saint Dmtrius, a hunnique grandissante a incit ses habitants non seule-
commenc en 426. Simultanment, ou plus prcisment ment opter pour la pratique denterrements lintrieur
environ un an plus tard, Sirmium a t rattach lEmpire des murs de la ville, mais aussi transporter leurs
dOrient, en devenant alors une base arrire et un point anctres dfunts dans la ncropole nouvellement
de dpart pour les oprations militaires menes par forme lintrieur et autour de lglise municipale.

130
Ivana POPOVI], La ncropole de la basilique urbaine Sirmium (113135) STARINAR LXII/2012

La question se pose de savoir quand ceci a pu avoir associe au fond des tombes sur le site 47. Les rsultats
lieu, et ce en tenant compte du fait que les ncropoles de lanalyse du matriel archologique mobile prove-
de Sirmium, situes hors de la ville, ont galement nant de la ncropole confirment, donc, les conclusions
accueilli des inhumations de fdrs germains. Les obtenues par une analyse globale des sources crites,
vnements historiques, le placement de Sirmium sous de la littrature, des trouvailles numismatiques et de la
la juridiction de lEmpire dOrient, les activits mili- situation sur le terrain, qui situent le dbut de lrection
taires et de btisseur de Thodose II et le renouveau de la basilique en 426, anne qui peut tre prise comme
ecclsiastique, incarn dans la mission du prfet Leon- date de formation de sa ncropole. Bien que lon ne
tius, indiquent que ce changement des coutumes spul- puisse dterminer jusqu' quand lglise est reste en
crales est survenu aprs 426, cest--dire lpoque o fonction, cause de dommages datant de la priode
a dbut lrection de la basilique municipale48. Les turque, les enterrements dans sa ncropole ont, proba-
tombes construites en briques, enregistres dans la blement, cess aprs la conqute hunnique et la
basilique et proximit, sinscrivent par leur type dans destruction de Sirmium en 441/42. A son emplacement
le cadre des constructions tombales des ncropoles on na pas constat la prsence de tombes contenant un
danubiennes et pannoniennes datant de la Basse- matriel caractristique des Ostrogoths et des Gpides
Antiquit. Leur orientation a t conditionne par celle qui tenaient la ville au cours de la seconde moiti du
des murs de la basilique ou des btiments dhabitation, Vme et la premire moiti du VIme sicle. Les tombes
sur les ruines desquels elles ont t creuses. Ainsi les creuses au cours de cette priode apparaissent dautres
tombes situes lintrieur de lglise sont orientes endroits dans la ville, alors que sur une partie de la
dans le sens SENO, sur le site 46 le mur de ldifice ncropole de Saint-Dmtrius ont t riges des
a impos une orientation approximative NOSE, et cabanes en bois et des fours49.
approximativement la mme sur le site 47. Le matriel Des enterrements intra muros, de toute vidence
mobile trouv dans les tombes est modeste, mais ca- contemporains de ceux de la ncropole de Saint-
ractristique pour la priode des dernires dcennies du Dmtrius, ont t enregistrs sur deux autres sites de
IVme et de la premire moiti du Vme sicle : peignes Sirmium, lintrieur dune luxueuse villa prs du
en os tripartites deux ranges de dents, petites rempart sud (fig. 5) et dans le quartier dhabitation
fibules anses en arbalte en fer, grandes boucles (site 49) proximit de cette mme villa. Il sagit de
ovales en fer, couteaux en fer et fragments de rcipients tombes miniatures en briques, qui contenaient des
en fin verre de couleur verte. Ce matriel, dorigine ossements dindividus adultes. Donc, comme dans la
romaine, est trs frquent parmi les trouvailles tombales ncropole de la basilique municipale, les restes de
provenant des ncropoles du vaste espace danubien et dfunts inhums hors des murs ont ici t transports
pannonien, dates de la priode de cca 385/87 441/42. dans des petites tombes construites sur les ruines de
Dans la ncropole de la rue Tadeu{a Ko{}u{ka Singi- btiments plus anciens. Une rduction de la zone urbaine
dunum, de grandes boucles ovales en fer ont t et le commencement de lenterrement lintrieur de
enregistres dans une tombe avec une petite fibule la ville et de son enceinte ont galement t enregistrs
anse en arbalte en fer, et dans deux tombes avec des dans la ville toute proche de Singidunum, o lon a
peignes tripartites deux ranges de dents, ce qui trouv une monnaie dArcadius dans une tombe en
indique, galement, que les trouvailles mobiles de la briques50. A Sirmium, on a dcouvert des tombes datant
ncropole de Sirmium sont habituelles pour cet horizon du second quart du Vme sicle dans le cur mme de
denterrement. En consquence, les objets provenant la ville alors dj en ruine. Leur concentration
de ces tombes nous incitent conclure que la cration lintrieur et autour de la basilique municipale
de la ncropole de la basilique Saint-Dmtrius devrait semblerait clairement indiquer quil est question dune
tre situe dans un cadre chronologique allant de 385/87 population christianise. Toutefois, ceci ne peut trouver
441/42, et sa chronologie est prcise par les trouvail-
les montaires, savoir une monnaie dArcadius
(304402) dans les fondations de labside de lglise, une
48 Ces conclusions ont t formules et expliques en dtail
monnaie dHonorius (410423) parmi les dcombres
par Popovi} 1987, 95129.
au niveau des tombes sur le site 46 et des monnaies 49 Popovi} 1982, 550552.
dArcadius (383392, 393395), de Thodose (383395, 50 Pop-Lazi} 2002, 87; Ivani{evi}, Kazanski 2002, 124125

378383) et de Thodose II (435450), dans la couche (avec bibliographie indique).

131
Ivana POPOVI], La ncropole de la basilique urbaine Sirmium (113135) STARINAR LXII/2012

une confirmation dans les objets provenant des tombes. aussi dmontrer quil sagissait bien dune population
Comme leur analyse la dmontr, il sagit dun urbaine, dont les anctres avaient galement vcu
matriel largement rpandu dans lespace danubien et Sirmium. La ncropole forme autour de la basilique
pannonien, qui peut tre associ, tant la population nouvellement construite tmoigne que cette population
romaque, quaux fdrs, installs dans le bassin de tait christianise, et perptuait la tradition des enterre-
Danube et en Pannonie aprs 380. En tout tat de ments lintrieur et autour des difices sacrs, ce qui,
cause, ce matriel dmontre quune population au cours du IVme sicle, a t enregistr Sirmium
urbaine, plus ou moins barbarise, a continu de vivre dans les ncropoles municipales nord et est, ainsi qu
dans les btiments partiellement dtruits de la ville Ma~vanska Mitrovica51. Les modestes objets funraires
aprs la destruction de Sirmium la fin du IVme ou au livrs par les tombes de la basilique Saint-Dmtrius
dbut du Vme sicle. Au cours des dcennies suivantes, tmoignent dun net appauvrissement de cette popula-
devant la menace hunnique, la population a commenc tion, processus qui avait dbut dj vers la fin du
enterrer ses dfunts dans la ville, en y transportant IVme sicle.
galement les ossements des dfunts inhums dans les
ncropoles situes hors de la ville. Cet lment vient lui Traduit par Pascal Donjon

51 Jeremi}, 2004, 4564.

132
Ivana POPOVI], La ncropole de la basilique urbaine Sirmium (113135) STARINAR LXII/2012

BIBLIOGRAPHIE:

Barkczi 1960 L. Barkczi, Ksrmai temet Petrovi} 1966 D. Petrovi}, Ranovizantijska


Pilismarton (Zusammenfassung: Ein sptrmisches grobnica u Baqevcu na Ibru (Rsum : Une spulture
Grberfeldin in Pilismarton) , Folia Aechaeologica 12, de haute poque byzantine Baljevac sur lIbar), Starinar
1960, 111129. XV/XVI (1964/65), 1966, 257259.
Barkczi 1961 L. Barkczi, Adatok Brigetio Pieta 1999 K. Pieta, Anfnge der Vlkerwander-
ksrmai trtnethez (Zusammenfassung: Zur spt- ungszeit in der Slowakei, dans: LOccident romain et
rmerzeitlichen Geschichte Brigetios), Folia Aechaeolo- lEurope centrale lpoque des Grandrs Migrations,
gica 13, 1961, 95113. Brno 1999, 171190.
Bjelajac, Ivani{evi} 1991 Lj. Bjelajac, V. Ivani- Pop-Lazi} 2002 S. Pop-Lazi}, Nekropole rimskog
{evi}, Les tmoignages archologiques des Grandes Singidunuma (Summary: Necropolis of Roman Singidu-
Invasions Singidunum, Starinar XLII, 1991, 123139. num), Singidunum 3, 2000, 7100.
Harhoriu 1998 R. Harhoriu, Die frhen Vlker- Pop-Lazi} 2005 S. Pop-Lazi}, Novi nalazi anti~kih
wanderungszeit in Rumnien, Bucarest 1998 grobova u Singidunumu (Summary: New Finds of Roman
Ivani{evi}, Kazanski 2002 V. Ivani{evi}, M. Ka- Graves in Singidunum), Singidunum 4, 2005, 125136.
zanski, La ncropole de lpoque des Grandes Migra- Popovi} 1982 V. Popovi}, Desingration und
tions Singidunum, Singidunum 3, 2002, 101157. Ruralisation der Stadt im Ost-Illyricum vom 5. bis. 7.
Ivani{evi}, Kazanski, Mastykova 2006 V. Iva- Jahrhznsert n. Chr., Palast und Htte. Beitrge zum Bauen
ni{evi}, M. Kazanski, A. Mastykova, Les ncropoles de und Wohnen im Altertum (Hrsg. D. Papenfuss, V. M.
Viminacium lpoque des Grandes Migrations, Paris Strocka), Mainz, 1982, 545566.
2006. Popovi} 1984 V. Popovi}, Un tui de peigne en os
Jeremi} 2004 M. Jeremi}, Kultne gra|evine de type Mrovingien et les objets dorigine ethnique
hri{}anskog Sirmijuma, u: Sirmium i na nebu i na trangre Cari~in Grad, Cari~in Grad I, Rome 1984,
zemqi, Sremska Mitrovica 2004, 4378. 160178.
Jeremi} 2008 M. Jeremi}, Les temples paens de Popovi} 1987 V. Popovi}, Die sddanubischen
Sirmium, Starinar LVI (2006), 2008, 167200. Provinzen in der Sptantike vom Ende des 4. bis zur
Migne 1960 J.P. Migne, Passio Sancti Demetrii, Mitte des 5. Jahrhunderts, Die Vlker Sdosteuropa-s im
Patrologia Graeca, Paris 1960, 11731184. 6. bis 8. Jahrhundert (Hrsg. B. Hnsel), Sdosteuropa
Miladinovi}-Radmilovi} 2011 N. Miladinovi}- Jahrbuch 17, Berlin 1987, 95139.
Radmilovi}, Sirmium. Necropolis, BeogradSremska Popovi} 1998 V. Popovi}, Kult svetog Dimi-
Mitrovica 2011. trija Solunskog u Sirmijumu i u Raveni (Rsum : Le
Milo{evi} 1986 P. Milo{evi}, Sremska Mitrovica. culte de saint Dmtrius de Thessalonique Sirmium et
Sirmium, Eastern Cemetery, Arheolo{ki pregled 1985 / Ravenne), Glas SANU CCCLXXXIV, Odeqewe isto-
Archaeological Reports 1985, 1986, 115116. rijskih nauka, kw. 10, Beograd 1998, 4356.
Pejovi}, Lu~i} 2011 Z. Pejovi}, B. Lu~i}, Sgi 1960 K. Sgi, Die sptrmische Bevlkerung
Nekropola iz perioda seobe naroda sa lokaliteta der Umgebung von Keszthely, Acta Archaeologica 12,
1a Sirmijuma (Summary: Necropolis from the Period 1960, 187256.
of Great Migration at Locality 1a in Sirmium), Zbornik [aranovi}-Svetek 1986 V. [aranovi}-Svetek,
Narodnog muzeja 20/1, 2011, 389413. Anti~ko staklo u jugoslovenskom delu provincije Donje
Keller 1971 E. Keller, Die sptrmischen Grab- Panonije (Zusammenfassung: Rmisches Glass aus dem
funde in Sdbayern, Mnchen 1971. Jugoslawischen Gebiet der Provinz Pannonia Inferior).
Petkovi} 1995 S. Petkovi}, Rimski predmeti od Novi Sad 1986.
kosti i roga sa teritorije Gornje Mezije (Summary: The Tejral 1988 J. Tejral, Zur Chronologie der frhen
Roman Items of Bone and Antler from the Territory of Vlkerwanderungszeit im mittleren Donauraum, Archa-
Upper Moesia, Belgrade, 1995. eologia Austriaca 72, 1988, 223304.
Petrovi} 1997 B. Petrovi}, Min|u{a, u: Tejral 2005 J. Tejral, Zur Unterscheidung des
Anti~ka bronza Singidunuma / Antique Bronze from vorlangobardischen und elbergermanisch-langobardis-
Singidunum (priredila S. Kruni}), Beograd 1997, chen Nachlasses, in: Die Langobarden. Herrschaft und
131, kat. 156. Identitt, Vienne 2005, 103200.

133
Ivana POPOVI], La ncropole de la basilique urbaine Sirmium (113135) STARINAR LXII/2012

Zotovi} 1995 Q. Zotovi}, Tipolo{ka analiza i kovnice srebra / Silver Workshops and Mints (prire-
min|u{a od srebra iz rimskih radionica Vimina- dile I. Popovi}, T. Cvjeti}anin, B. Bori}-Bre{-
cijuma (Summary: Typological Analysis of Silver Rarrigs kovi}, Beograd 1995, 233242.
from Roman Workshops in Viminacium), u: Radionice

Rezime: IVANA POPOVI], Arheolo{ki institut, Beograd

NEKROPOLA GRADSKE BAZILIKE


U SIRMIJUMU

Kqu~ne re~i. Sirmijum, kasna antika, nekropola, fibule, pre|ice, ~e{qevi, no`evi.

Analiza nekropola unutar gradske bazilike i oko we, na Grobovi na~iweni od opeka, registrovani u bazilici i we-
lokalitetima 59, 46 i 47 Sirmijuma (sl. 1, 2, 5, 10), na ko- nom okru`ewu (sl. 68, 1118), tipolo{ki odgovaraju nizu
jima je do danas registrovano 28 sahrana, pokazala je da je grobnih konstrukcija sa kasnoanti~kih podunavsko-panon-
re~ o grobovima iste konstrukcije, sa siroma{nim ali me- skih nekropola. Wihova orijentacija bila je uslovqena
|usobno homogenim grobnim prilozima. Naime, u samoj ba- polo`ajem zidova bazilike, odnosno stambenih objekata na
zilici otkrivena su dva minijaturna sarkofaga i 13 grobo- ~ijim su ru{evinama ukopani. Tako su grobovi u crkvi ori-
va na~iwenih od horizontalno ili nasati~no postavqenih jentisani u smeru JISZ, a na lokalitetu 46 zid gra|evine
opeka (sl. 3). Mali sarkofazi, kao i neki minijaturni gro- je uslovio pribli`nu orijentaciju SZJI, pribli`no kao i
bovi od opeka u woj i na lokalitetu 47 (br. 2/59, 7/59, 14/59, na lokalitetu 47. Pokretni materijal iz grobova jeste siro-
1/47, 6/47), sadr`avali su kosti odraslih osoba, {to upu}u- ma{an (sl. 4, 9, 2022), ali je karakteristi~an za period po-
je na zakqu~ak da su posmrtni ostaci individua sahrawe- sledwih decenija IV i prve polovine V veka: trodelni dvo-
nih na nekropolama Sirmijuma lociranim izvan gradskih struko nazubqeni ko{tani ~e{qevi (sl. 4. 12, 22), male
bedema u jednom trenutku preneseni na novoformiranu ne- lu~ne gvozdene fibule (sl. 20. 3.1), velike ovalne gvozdene
kropolu intra muros. Nastanak te nekropole povezan je, sva- pre|ice (sl. 20. 5.1), gvozdeni no`evi (sl. 21) i fragmenti
kako, sa podizawem bazilike unutar gradskih bedema. Na posuda od tankog zelenog stakla (sl. 9). Ovaj materijal, rim-
osnovu analize pisanih izvora i istorijskih doga|aja, argu- skog porekla, veoma je rasprostrawen u grobnim nalazima
mentovano je pokazano da je gradwa gradske crkve, posve}e- iz nekropola sa {ireg podunavsko-panonskog prostora, da-
ne Sv. Dimitriju, otpo~ela 426. godine. U isto vreme, odno- tovanim u period od cca 385/87. do 441/42. godine. Na nekro-
sno oko godinu dana ranije, Sirmijum je pripao Isto~nom poli u ulici Tadeu{a Ko{}u{ka u Beogradu (Singidunum),
carstvu, postav{i baza i polazi{te za vojne operacije ko- velike ovalne gvozdene pre|ice registrovane su u jednom
je je Teodosije II vodio protiv Huna u Panonoji i protiv grobu zajedno sa malom lu~nom fibulom od gvo`|a, a u dva
uzurpatora Jovana u Raveni. Grad, ~iji su reprezentativni groba sa trodelnim, dvostruko nazubqenim ~e{qevima, {to
objekti ve} bili razru{eni u nekom od varvarskih upada tako|e pokazuje da su pokretni nalazi iz sirmijumske ne-
posle 378. godine, promenio je svoju urbanu strukturu, pre kropole uobi~ajeni za ovaj horizont sahrawivawa. Dakle,
svega gubqewem funkcije severnog bedema i povla~ewem nalazi iz grobova upu}uju na zakqu~ak da nastanak nekropo-
gradske teritorije ka jugu. Nadolaze}a hunska opasnost uslo- le bazilike Sv. Dimitrija treba vezati za hronolo{ki okvir
vila je da stanovni{tvo po~ne ne samo da se sahrawuje unu- izme|u 385/87. i 441/42. godine, a wenu hronologiju preci-
tar gradskih bedema ve} i da svoje mrtve pretke prenosi u ziraju numizmati~ki nalazi, Arkadijev novac (304402)
novoformiranu nekropolu u gradskoj crkvi i oko we. Kada se u temequ apside crkve, Honorijev novac (410423) u {u-
to desilo, ako se ima u vidu da su na sirmijumskim nekropo- tu u nivou grobova na lokalitetu 46, kao i novac Arkadija,
lama lociranim van grada sahrawivani i germanski federa- (383392, 393395), Teodosija (383395, 378383) i Teodo-
ti? Istorijski doga|aji, potpadawe Sirmijuma pod juris- sija II (435450) iz sloja koji se vezuje za dno grobova na
dikciju Isto~nog carstva, vojna i graditeqska delatnost lokalitetu 47. Rezultati analize pokretnog arheolo{kog
Teodosija II i crkvena obnova, otelotvorena u misiji pre- materijala sa nekropole potvr|uju, dakle, zakqu~ke do ko-
fekta Leontija (Leontius), upu}uju na to da je do ove prome- jih se do{lo sveobuhvatnom analizom pisanih izvora, li-
ne u sepurkralnim obi~ajima do{lo posle 426. godine, od- terature, nalaza novca i situacije na terenu, a koji po~e-
nosno u vreme kada je otpo~eta izgradwa gradske bazilike. tak gradwe bazilike datuju u 426. godinu, od kada se mo`e

134
Ivana POPOVI], La ncropole de la basilique urbaine Sirmium (113135) STARINAR LXII/2012

ra~unati i sa formirawem wene nekropole. Iako se usled nekropole bazilike Sv. Dimitrija podi`u drvene kolibe.
o{te}ewa iz turskog perioda ne mo`e utvrditi do kada je Nekropola u okriqu novopodignute bazilike svedo~i da je to
crkva bila u upotrebi, sahrawivawe na wenoj nekropoli je, stanovni{tvo bilo hristijanizovano, nastavqaju}i tradi-
verovatno, prestalo posle hunskog osvajawa i razarawa Sir- ciju sahrawivawa u sakralnim objektima i oko wih, {to je
mijuma 441/42. godine. Na wenom mestu nisu konstatovani tokom IV veka u Sirmijumu registrovano na severnoj i is-
grobovi koji sadr`e materijal karakteristi~an za Ostro- to~noj gradskoj nekropoli, kao i u Ma~vanskoj Mitrovici.
gote i Gepide koji su u drugoj polovini V i prvoj polovini Skromni prilozi u grobovima bazilike Sv. Dimitrija sve-
VI veka vladali gradom. Ukopani grobovi iz tog perioda do~e o poodmaklom stepenu siroma{ewa te populacije, {to
pojavquju se na drugim lokacijama u gradu, dok se na delu je proces koji je zapo~eo jo{ krajem IV veka.

135
MILOJE VASI], The Institute of Archaeology, Belgrade
NADE@DA GAVRILOVI], The Institute of Archaeology, Belgrade

VENUS OR DIANA FROM MEDIANA

UDK: 902.2(497.11)"2011" ; 904:730.032.041.2(497.11) e-mail: nadia011@yahoo.com


DOI: 10.2298/STA1262137V Received: February 17, 2012
Original research article Accepted: June 21, 2012

Abstract. During the archaeological investigation of the villa with peristyle in Mediana in 2011, a marble head,
which may have been the head of a goddess, was discovered. Her face and the treatment of her hair show that it may represent
a replica of a Hellenistic original. As far as is known, analogies of female marble heads from Mediana show that it probably
represented either the goddess Artemis/Diana or Aphrodite/Venus. A hypothesis was made that the head might have been
modelled at the end of the 3rd century 430 A.D. In addition, a hypothesis was made that the base found in 2002, on which
was preserved a fragmented head and the tail of a dolphin, and the new found head were parts of the same statue.
If this is the case, then the statue can be identified as the type of Capitol Venus, subtype Venus with dolphin.

Key words. Marble head, Mediana, late antique.

A
mong the various buildings in Mediana, the assumed that stibadium A belongs to this rebuilding
most important is the so called villa with peri- phase. Therefore, previous opinions that stibadium A
style.1 The north-western part of the villa, is from the first building horizon of Mediana, built be-
with rooms in the south-western area and the western fore the first phase of the villa, is not correct.2 Stibadium
half of the south portico, were not identified until 2010. B has two annexes, one to the east and one to the west,
During 2010 and 2011 these parts were also researched, and it was decorated with a floor mosaic and a marble
so that a complete plan of the villa is now known. The panel on the foundation of the walls. On the villas vault
fact that there were two building phases suggests that and possibly on the walls as well, there was a mosaic
there had been some reconstruction work carried out. In made of tesserae, some with gold leaf, which were
the second phase, some new parts of the villa were built found half burnt on the floor, under which was found a
whilst others were widened. We will briefly mention hypocaust. In the past it has been suggested that the
that west of the big aula/triclinium, a round triclinium,
stibadium B, was discovered, identical to the previ-
ously discovered stibadium A. A museum building was 1 The newest topographic plan, with added geomagnetic
built above this, on the eastern side of the aula (fig. 1). prospection, Milo{evi}, Peters, Wendling 2011, 275284, especially
It is almost certain that stibadium B was built during a Fig. 2 and with cited literature.
rebuilding phase of the existing building, so it can be 2 Vasi} 2004, 104. Id., 2007. 99.

* The article results from the project: Romanization, urbanization and transformation of urban centres of civil, military and residential
character in Roman provinces on territory of Serbia (no 177007) funded by Ministry of Education, Science and Technological Development
of the Republic of Serbia).

137
VASI], GAVRILOVI], Venus or Diana from Mediana (137149) STARINAR LXII/2012

Fig. 1. Plan of villa with peristyle after the excavations 20102011,


with marked places of finds (parts of sculptures)
Sl. 1. Plan vile s peristilom posle iskopavawa 20102011,
s ozna~enim mestima nalaza delova skulptura

1) Hoard of sculptures from 1972


2) Marble head of goddess from 2011
3) Fragments of sculptures from 2011
4) Stibadium A
5) Stibadium B

138
VASI], GAVRILOVI], Venus or Diana from Mediana (137149) STARINAR LXII/2012

villa was decorated with mosaics and frescos in the the same period is a group of sculptures found in 1972,
second building phase, which was confirmed by the in western room 4 (fig.1/2). It seems that after the
most recent excavations.3 Another hypothesis is that destruction of Mediana (presumably in 378 AD) the
during the second building phase of the villa, statues of newly settled people who inhabited Mediana in the
mythological, not cult, characters were brought and fourth quarter of the 4th century AD, were coming upon
placed in the villa. Fragments of these statues were parts of architectural plastic and sculptures. Some were
found in a hoard of sculptures discovered in western used as building materials (bases and parts of columns)
room 4, as well as in other excavated areas in and which were discovered in the foundations of buildings
around the villa.4 Thus, the villa with peristyle was trans- of this horizon in the villas peristyle, for example, in
formed from a simple, almost barrack-like building into the foundations of a church with a christogram, where
a monumental and luxurious residence, with a predo- fragments of an honorary inscription were found.7 One
minantly ceremonial character. It had a monumental gate group of sculpture fragments was found south of the
to the south, a special vestibule in front of the divided villa, in the foundations of the gate and the guardhouse
porticos through which one could enter the villa and in (fig.1/4), two heads were found outside the villas area
the peristyle at the entrance, an arcade was built. The and a group of fragments was found, in 2011, in the
rooms of the northern section were decorated with fres- area between stibadium B and the northern face of the
cos and mosaics of a festal, ceremonial character. villa (facing the termae).8 A further group of fragments
During June of the 2011 campaign,5 in the control
profile between sondages 46 and 47 (western room 1),
a marble head, which probably belonged to a statue of
a goddess, was found (fig. 1/1).6 The head was found 3 It was definitively confirmed that all mosaics in Medianas

on the transition between layer B and layer C. In the villa with peristyle, those in the stibadium and those in the aula and
peristyle, were from one chronological horizon, although they were
same cota and in the surrounding area the head of an made by different artisans or different workshops. They can be dated
iron spear and part of a column with its base were found between 337 and 350 AD.
(fig. 2). Obviously it was a living horizon following a 4 We presume, as it was stated in Vasi} 2004, 103 etc., that at

fire at the villa, where unneeded architectonic parts one time the aula was transformed into the sanctuary into which the
cult statues of Asclepius and Hygieia and bronze railings were
and parts of sculptures were disposed of. Dating from brought. It should be emphasised that all, so far discovered statues
were from different periods, from the 2nd to the 4th century and from
different workshops.
5 The excavations were made as part of a project of the Ministry

of Culture and Media of the Republic of Serbia: Constantine villa


on Mediana project of presentation 2010 2013. The chief of the
team formed by the Ministry is Dr Gordana Milo{evi}, the chief of
archaeological excavations is Dr Miloje Vasi} and the chief of field
excavations is Dr Nade`da Gavrilovi}.In addition, the Archaeological
Institute of Belgrade, the National Museum of Ni{ and the Regional
Institution for the protection of monuments of culture of Ni{, are
taking part in the project.
6 Field diary: 06. 06. 2011. The head was lying on 199. 12; inv.

n. C4. The dimensions of the head are 25 x 14 cm, and the material
from which it was modelled is a white marble of fine texture.
7 Vasi} 2004a, 294.
8 1) The marble head of Menada with fragmented ivy wreath;

the height of the head is 15 cm; it was found in 1996 in the passage
(room 16) between two barracks of the eastern section, inv. n.
C823/96. Dr~a 2004, 164, cat. 86. S. Dr~a identified this head with
Isis presuming that the wreath was of lotus (?). As far as we know,
Isis had different attributes on her head. 2) The marble head of
Venus or Diana (?), with a large knot of hair on her vertex (top knot)
and soft curls. Behind her left ear there is a fragment of an object.
The head is 8 cm long; it was found in 2001 in sondage 10 on the
inner side of the western perimeter wall of the villa, inv. n. C67/
Fig. 2. The appearance of the head and the base of column 2001, Dr~a 2004, 165, cat. 87. S. Dr~a wrongly identified this head
as Isis-Fortuna. For further information regarding this head see B.
Sl. 2. Op{ti izgled nalaza glave i baze stuba Plemi} in the next issue of Starinar.

139
VASI], GAVRILOVI], Venus or Diana from Mediana (137149) STARINAR LXII/2012

Fig. 3 and 4. The head of goddess: front and half profile


Sl. 3 i 4. Glava bogiwe, spreda i iz poluprofila

was also found in 2011 in the same area between stiba- called Venus necklace, which emphasises the fullness
dium B and northern face of the villa (fig. 1/3).9 of her body. On her eyebrows and certain parts of her
The sculpture of the female deity from Mediana, coiffure, traces of yellow paint can be observed, while on
from white marble, represents the head of a young girl/ the bands on her hair and on her face (eyelids, cheeks)
woman, slightly tilted to the left (fig. 3 and 4). The head there is red paint. Certain elements like the wider root of
of the goddess is softly, gently and almost idealistically her nose, the high triangular forehead (due to the god-
modelled. Her cheek bones are not emphasised and her dess hair), the shape of the eyes, the small, full mouth
narrow, oval face has no other signs of sensuality, which and the slightly idealistic expression on her face, show
additionally creates the impression of innocence and that the goddess head was modelled in a classicistic
chastity in the female sculpture from Mediana. The style, by a Greek sculpture from the 4th century BC.
head has some damage in the area of the nose, mouth This type of goddess coiffure is known from antique
and chin. In spite of this damage, it can be presumed that sculptures, and is characteristic because the hair curls
the nose of the goddess was long and straight, and the on the vertex are so coiled, that they form a, so called,
mouth small and full with maybe, judging by the angles top knot (crobylus, krbuloj),10 and the second part
of the mouth, a certain suggestion of a mild smile. of the hair forelock is tied into a bun on the back of the
Careful modelling of the goddess chin gives the impres- head from which two tresses fall down towards the
sion of youthful gentleness to her face, and under the arms. The female head from Mediana, however, differs
rich waves of her hair, small, very carefully modelled
ears are visible. The eyebrows are slightly pronounced,
her big almond shaped eyes with clearly modelled upper 9 Sondage 72. It is interesting that in this group a piece of por-
and lower eyelids are somewhat elongated, without phyry forearm and hand with a globe in it was found. These frag-
emphasised eyeballs. The goddess is looking into the ments are parts of a porphyry statue of Asclepius found in 1972.
distance. There is an arced crease on her throat, the so 10 For krobylos see Bonfante 2003, 141.

140
VASI], GAVRILOVI], Venus or Diana from Mediana (137149) STARINAR LXII/2012

Fig. 5 and 6. Left and right profile of the head


Sl. 5 i 6. Levi i desni profil glave

from this type of coiffure. The hair is divided in the and 9). This can be explained by the fact that the statue
middle and there are forelocks on the upper edge of her was intended to be viewed only from the front, but it is
forehead which are falling on the left and right side of also possible that the head was not finished. On the left
the forehead and ears. The space between the forelocks profile it can be seen that the tresses of the bun are
is holed with a drill and the hair is carved with a chisel. falling towards the left arm. Viewed from the front, the
On the part of the vertex closer to the forehead, there coiffure of the goddess appears as a compact mass, so
is a taenia (?) on her head, wrapped twice around the it resembles a crown or truncated pyramid.
head. It can be perceived from the left profile that the Further in the text we will attempt to determine
front end of the taenia goes over the forelocks and ends more precisely to which deity the head from Mediana
in a top knot and that these forelocks go over the back belongs, on the basis of its stylistic and chronological
of the taenia (fig. 5). From the right profile, the fore- characteristics. As we dont have parts of the torso at our
locks, which are collected in a top knot, go over both disposal, the head of the deity must be analysed primar-
parts of the taenia (fig. 6). The edges of the taenia and ily on the basis of the coiffure. The type of coiffure with
the aforementioned forelocks are represented only with a top knot is mostly seen on the goddesses Aphrodite/
shallow engraved lines, without volume, as if modelled Venus, Artemis/Diana, Iris (Irida) and Menadae, as
in wood. The forelocks to the fore are more voluminous, well as on the gods Apollo and Zeus and on satyrs and
modelled with a drill. It is clear that they curve into the nymphs. Since it is certain that we are dealing with a
locks of the top knot. On it, at the front, there are fore- female and not a male deity, we focused on the possi-
locks modelled with drill and chisel. On the side, the bility of recognising the sculpture from Mediana as the
middle part of the forelocks is emphasised using a drill. goddess Artemis/Diana or Aphrodite/Venus.
The top knot does not cover the whole width of the top We will begin the analysis with the sculptures of
of the head. The part between the top knot and the bun, Artemis/Diana which dont have a top knot on their
and the bun itself are not separately sculpted (fig. 7, 8 heads. The taenia in the hair and the softness of the

141
VASI], GAVRILOVI], Venus or Diana from Mediana (137149) STARINAR LXII/2012

face of the Mediana females head bear a close resem- statue is without locks and the top knot is more empha-
blance to the marble statue of Artemis from Arizio (the sised. Maybe the closest coiffure to the one from Medi-
face of Artemis is somewhat rounder).11 But, although anas head is the coiffure from the sculpture of the so
the forelocks on the forehead imply modelling similar called Artemis Rospigliosi.22 The modelling of the
to the Mediana head, the only difference can be seen in top knot is particularly similar, with the exception that
the hair which falls on the neck. Also, the question of
how the head from Arizio was identified as Artemis
remains open. A similar representation of Artemis can be 11 It is presumed that the sculpture was made in the second
seen in Villa Borghese12 and in the sculpture of the so half of the 1st century AD and modelled on the Attic cult statue of
called Diana from the villa Borghese13, which has a Artemis from 440430 BC, LIMC II. 1 Artemis,789, n. 5.
12 On the fresco from Villa Borghese, similarities with the god-
similar coiffure to the female head from Arizio. We will
dess from Mediana are obvious in the slight tilting to the left of the
mention the female sculpture from the Museum Palazzo
head of Artemis, the treatment of the coiffure which is tied with rib-
dei Conservatori in Rome14 which treats the forehead bons on the top of the head, the oval face with regular features and
coiffure in a very similar fashion to that of the Medianas the idealised expression on the face. The type of representation of
head, whereby the hair falls backwards in tresses onto Artemis from Villa Borghese belongs to the so called Dresden
Artemis type, which was modelled on a Greek original from the
the goddess shoulders. The difference between the
second half of the 4th century BC.
aforementioned representations and the head from 13 A marble statue of Diana from Villa Borghese in Rome rep-
Mediana is in the treatment of the eyelids, which in the resents a Roman copy of a Greek original from the second half of
aforementioned representations are more emphasised the 4th century BC. Dianas hair is collected in a bun and her facial
(except Diana from Villa Borghese). Similar forelocks features are somewhat sharper, LIMC II. 1 Artemis, 799, n. 9f, 802,
n. 17.
of hair along with the clear separation of the forelocks 14 It is presumed that the torso represented Artemis modelled by
can be observed on the head of Atalanta on the Meleagar Kefisodot in the 4th century BC. A Roman copy, Palazzo dei Con-
sarcophagus from Eleusina.15 servatori, room Tauriani Vettiani, inv. no. MC 1123; see www.
In further analysis we will turn to those sculptures museicapitolini.org/percorsi/percorsi_per_sale/museo_del_palazzo_
dei_coservatori/sale_degli_horti_tauriani_vettiani.
of Artemis/Diana on which there is a top knot as well. 15 Stirling 2008, 144 sq, fig. 32 (detail) with quoted earlier lite-
The characteristic forelock which is around the face rature. Guntram Koch dates the sarcophagus from Eleusina to the
and the forelocks which, from the top left and right of first quarter of the 3rd century BC, G. Koch, Die mythologisschen
the head, form the top knot, can be seen on a bronze Sarkophage: Meleager (Die antiken Sarkophagreliefs 12. 6), Berlin
1975, pp. 76, 142143; non vidimus, cited from Stirling 2008, 145,
statue of Artemis from Pireus, which has a very small
not. 170.
top knot.16 Somewhat more similar examples in the 16 commons.wikimedia.org/wiki/Category: Archaeological_
context of face modelling and the presence of a top Museum_of_Piraeus_(Athens)_-Second_bronze_Artemis. Pireus
knot on the deitys head can be found in a marble archaeological museum. Dated to the middle of the 4th century BC,
sculpture of Diana from the Vatican17 and the marble and it is thought that it belongs to Praxiteles school.
17 On Diana from theVatican, rich waved hair is divided in the
head of Diana from Villa Borghese in Rome.18 A clos-
middle, tied in a top knot on the top of her head and a bun on the
er analogy concerning the treatment of the hair and back of her head, with locks that fall on both sides of her neck. A
coiffure is a Roman copy of Diana from the Vatican.19 long triangular forehead, shallow modelled almond shaped eyes, long
Dianas oval face is surrounded by carefully combed nose and small mouth are considerably reminiscent of the facial
features on the head from Mediana. It is presumed that the sculpture
waved hair, which is partly tied with locks in a top knot was made in the 2nd century AD, LIMC II.1. Artemis, 809, n. 36d.
and partly gathered in a bun on the back of the head; a 18 The marble head of Diana from Villa Borghese differs from
flat triangular forehead, elongated almond shaped Medianas head only by a somewhat wider face. It is presumed that
eyes, narrow nose and full mouth are emphasised on the date of its modelling was the 3rd century AD and that it was part
of a relief of a sarcophagus. The marble head of Diana is now at Ny
the face of the goddess. Still, her hair is much more
Carlsberg Glyptotek in Copenhagen, LIMC II.1. Artemis, 816, n. 112.
softly modelled than the head from Mediana. We should 19 It was modelled on the Greek original, Artemis Soteira
also mention some other statues of Diana which are from a relief from Delos, LIMC II.1. Artemis, 802, n. 19a.
further analogies of Medianas head: Diana with dog, 20 commons.wikimedia.org/wiki/Kos_museum_Artemis.

a statue from Kos20 and, similar to this one, Diana with 21 Inv. no. 227 see commons.wikimedia.org/wiki/Category:

dog from the Vatican museum.21 The coiffure of the Statues_of_Artemis.


22 Roman copy from the 1st to 2nd century AD, of a Hellenistic
statue from Kos is closer to Medianas head because her
original: LIMC II, Artemis, p. 646, no. 274, pl. 468, s. v. Artemis
locks go from the forehead to the top of the head and (L. Kahlil) = LIMC II, p. 808, no. 35, s. v. Artemis/Diana (E. Simon
make a top knot, while the coiffure of the Vaticans and G. Bauchhenss). Muse Louvre, inv. no. MA 559.

142
VASI], GAVRILOVI], Venus or Diana from Mediana (137149) STARINAR LXII/2012

on the sculpture from the Louvre it is much wider and head with known representations of Aphrodite/Venus,
it is positioned on the top of the head. With that in mind, many similarities can be perceived in the facial features
we will mention Artemis from the house of Panayia in and expression, the way of combing the hair into a top
Corints. Lea Stirling, who published this find, is of the knot and the tying of the hair with ribbons. There are
opinion that this type was modelled on Artemis many more surviving statues of Aphrodite/Venus than
Rospigliosi.23 The statue of Artemis from the house in there are of Artemis/Diana, which is not surprising when
Panayia along with other statues found there, were we consider how the descendants of the Romans, espe-
dated by L.Stirling from the late 3rd to the early 4th cen- cially of Julius Caesar and Julio-Claudian, had high
tury AD.24 However, besides formal similarities between regard for Venus. The head from Mediana would mostly
Artemis from Korinthos and the head from Mediana appear to belong to the Capitoline Venus type of sculp-
(the top knot, the Venus necklace on the throat), there ture. However, this type is also derived from a much
are important differences, especially since on the head more famous sculpture of Aphrodite of Cnidus, which
from Korinthos the eyebrows are carved as a straight was modelled around 364361 BC by Praxiteles28 and
line and the taenia around her head was additionally painted by Praxiteles pupil Nikias.29 Her naked figu-
modelled from some other material, which can be re, especially the type of head, influenced much later
clearly seen by the print on the head. It is important to Hellenistic sculptures and, therefore, later Roman copies.
mention that the head from Panayia was coloured in a There are two main types of replica of the Aphrodite
similar way to Medianas head.25 from Cnidus:
In terms of territorially stylistic similarities to the 1) The Belvedere type,30 which is considered to be
head of the goddess from Mediana, we should mention closer to Praxiteles original, especially in the represen-
a marble sculpture of Diana Lucifera from Equum.26 It tations of Cnidus coins31
represents a standing figure of the goddess almost life- 2) The Colonna type32
sized, whose coiffure (waved hair divided in the mid- The head of Venus from Cnidus has an ellipsoid
dle and tied in a top knot on the top of the head and a face, a triangular forehead (because of the arrangement
bun on the back of the head), and certain facial features of the hair forelocks), almond shaped eyes, a strong nose,
resemble those of the Medianas head. The date of Diana a small mouth and a prominent chin. The hair is divided
Luciferas modelling is, depending on which author is in the middle; waved locks are combed towards the
consulted, either the second half of the 2nd century or back of the head where they are grouped into a bun.
the second half of the 3rd century AD.27 Two ribbons are tied around the head, which makes the
Whilst analysing possible analogies with Artemis/ statue more beautiful. The sensuality of her face is
Diana, it must be stated that we were not able to perceive
any such sculptures that had similarities to Medianas
head, which would enable us to identify it with any cer-
tainty as Artemis/Diana. Chronologically, one group 23 Stirling 2008, 118.
of the sculptures belongs to the 4th century BC and one 24 Ibid., pp. 135, 140, 144.
25 Ibid., 115.
group are Roman copies of these statues and actually
26 Mili}evi} Brada~ 2009, 6768.
present a way of sculpting from the Hellenistic period, 27 Mili}evi} Brada~ 2009, 68. As the author states, M. Abrami}
belonging primarily to a period from the 1st to the 2nd dated the sculpture of Diana Lucifera to the second half of the 2nd
century AD, except Atalante from Meleagers sarcopha- century AD, while N. Cambi dates her to the second half of the 3rd
gus and Artemis from Panayia (Korinthos). Considering century AD and it is considered to be an import from the workshops
the characteristics of the coiffure and the modelling of of Aphrodisias.
28 By Pliny, NH 34. 50: In the 104th Olympiad (=3641) Pra-
the face, Medianas head appears to derive from the
xiteles was at the peak of his art. (Loeb collection). Ajootian 1998,
Hellenistic, early Roman period and it has certain simi- 99. Corso 2007, 9.
larities with sculptures dated to this period. However, 29 Ajootian 1998, 99 (Pliny, NH 35. 133).

they are all significantly less stylised and have much 30 Corso 2007, 14, nt. 8, no. 260, fig. 2 (Rome, the Vatican

more richly modelled details than those of Medianas Museum storehouse inv. no. 4260).
31 Corso 2007, 9, fig. 1 (nt. 8, no. 322), bronze coins made in
head, and on this basis it could not belong to afore-
the name of Caracala and Plautila: London, The British Museum,
mentioned period. Department of Coins.
The second part of the analysis concerns representa- 32 Corso 2007, 14, nt. 8, no. 259, fig 3. Rome, the Vatican

tions of Aphrodite/Venus. When comparing Medianas Museum, Gabinetto delle Maschere, inv. no. 812.

143
VASI], GAVRILOVI], Venus or Diana from Mediana (137149) STARINAR LXII/2012

Fig. 7 and 8. The appearance of the top of the head


Sl. 7 i 8. Izgled temena glave

We should also mention the famous Venus de Milo,


which certainly belongs to those sculptures based on the
style of Aphrodite from Cnidus. Some authors think that
it belongs to Praxiteles school (Alkamenes) whilst
others date it to the 2nd century BC.33 The bust of
Aphrodite, which was sent by ship from Greece to
Rome, is very interesting for our analysis. The ship
sank at Mahdia.34 Mohamed Yacoub mentions that the
coiffure resembles slices of melon (cte de melon),
which form a knot on the top of the head. In a way, this
was a precursor of top knot representation.
A possible replica of Aphrodite from Cnidus was
discovered in 2003, in Sagalasos, in a nympheum from
the period of Hadrian.35 Here we are certainly dealing

33 Murray 2004, 275 sq. The arrangement of the hair is similar,


as is the modelling of the forelocks and tying of the bun on the back
Fig. 9. The appearance of the head with bun
of the head, where the forelocks fall on the neck.
Sl. 9. Izgled potiqka glave s pun|om 34 Yacoub 1970, 76/7: Fouilles sous-marines de Mahdia; p. 82:

C. 1183, Buste d Aphrodite. The author dates the find to the 3rd2nd
century BC, Quertani 1994, 290 thinks that the head has certain
similarities with some Praxiteles sculptures, although it is not a
Roman copy. The author dates the sculpture to the end of the 2nd
lined with a Venus necklace. Her eyes and eyebrows and the beginning of the 1st century BC.
are modelled with sharp lines and the eyebrows merge 35 Waelkens 2004. The author, judging by the type of hairstyle,

with the nose shaft. Medianas head resembles this writes that the head was modelled on the sculptures from the 4th cen-
type of modelling, but is much closer to the Colonna tury BC. The hair is tied with two parallel ribbons, but, since the head
is fragmented, we cant be certain how the coiffure looked at the back
type, especially with the treatment of the forelocks on
of the head. Waelkens rightly emphasises that it is not a faithful repli-
the forehead. The difference, therefore, is in the top ca, but the sculpture has great similarities with the Kaufmann head
knot which is present on Medianas head. from the Louvre, which is itself a replica of Aphrodite from Cnidos.

144
VASI], GAVRILOVI], Venus or Diana from Mediana (137149) STARINAR LXII/2012

with some transformation of hair representation, where famous statue of Venus Capitolina is probably a copy
the head is not just simply a replica, but it is modelled of a late Hellenistic sculpture and was modelled in the
in the sculptors particular style. For us, it is important Antonine period, but the precise origin of the proto-
and interesting that in the treatment of the forelocks on type (if there was one), remains unknown. The marble
the top of the head, certain similarities with Medianas version has a vase with a drape over it as a support.42
head, can be perceived. The face of Aphrodite from Sa- Other versions of Venus Capitolina have a dolphin,
galasos is, however, completely differently conceived, tree or Eros as a support.43 Christine Havelock men-
being wider with a larger, flat mouth and narrow lips. tions that the majority are of larger dimensions and
The fact that the sculpture of Aphrodite was trans- from the western provinces of the Roman Empire and
ferred from Cnidus to Constantinopolis around 393/4 that those smaller than natural size are more common
AD, and was damaged in a fire in 476 AD, which destro- in the eastern provinces.44 They could be part of house
yed other original Greek sculptures from the famous decorations, or parts of domestic lararia; they all differ in
collection of Lausos, the chamberlain (praepositus sacri some details, especially in the position of the head, and
cubiculi) of the court of Teodosius II, shows how much none has such a meticulously modelled top knot as the
the sculpture was respected in late antiquity and in Roman Venus Capitolina.45 Most of the Roman copies
Christianity as a remarkable piece of art.36 Therefore it are dated to the 2nd century AD. The main characteristics
is not unusual that a larger number of replicas were on sculptures of the Venus Capitolina type are, as already
found in Italy (Rome, Ostia, Tivoli), France (Martigny, mentioned, a top knot which is more or less detailed, with
Toulouse), Spain (Tarragona), but also in Greece full forelocks tied into a knot on the top of the head
(Athens, Korinthos) and in cities of Asia Minor. with the free ends of hair forming equally full locks.
We already emphasised that the head of the god- On the back of the head, the hair is tied in a second knot,
dess from Mediana could be linked with the type of from which the tresses fall down over the shoulders.
sculptures of which the most famous examples are The top knot is less emphasised on the head from the
Venus Medici and Venus Capitolina or generally, the Louvre46 than in other sculptures of this type. Certain
Venus Pudica type.37 But, before analysing these two
types, we will mention that in the middle, between the
Aphrodite from Cnidus and the Venus Medici and Venus
Capitolina types, there is a sculpture of Aphrodite 36 For Lausos collection see Guberti Basset 2000. Kedrenos
which was modelled by Menophantos in the 1st centu- 322c.
ry BC,38 maybe at the same time as Kleomenes Venus 37 Murray 2004, 272 i not. 1.
38
Medici. We mention Menophantos Aphrodite because Havelock 2007, 80, fig. 22. Corso 2007, 103, not. 107. Roma,
it is rare that there are emphasised parts or attributes on Museo Nazionale delle Terme, inv. no. 75674. The sculpture was
found in the monastery of San Gregorio al Celio, Camaldolese.
the statues that allow us to deduce that they are mod- LIMC II, s. v. Aphrodite, no. 422. On the base there is an inscription:
elled as copies of other, more famous statues found in APO THC/N TRWADI/AFRODITHC MNOFANTOC POII
the territory of Italy. There is a slight suggestion of a [According to Aphrodite in Troas (probably Alexandria of the Troas,
Menophantos made (this statue)].
top knot on Menophantos Aphrodite. 39 Ridgway 2001, 354. Ridgway 2002, 264. Havelock 2007,
Venus Medici, now in Florence, probably originated 76 sq. with earlier literature. For Venus Medici see LIMC 2, s.v.
in Rome, and has on its plinth an inscription with the Aphrodite 53 (c), nos. 419, pl. 40; dating is open.
signature of the aforementioned sculptor Kleomenes, 40 Havelock 2007, fig. 19. Florence, Uffizi, inv. no. 224.
41 Ridgway 2001, 355. Havelock 2007, 78, calling upon the
son of Apolodorus from Athens. The authenticity of
the inscription has been widely discussed.39 It is pos- research of B. Felletti Maj, Aphrodita Pudica, Archaeologia Classica
3, 1951, 3365.
sible that Kleomenes worked for a Roman clientele in 42 Roma, Musei Capitolini, inv. no. 409. LIMC II, s. v. Aphro-
the first half of the 1st century BC, reproducing the dite, no. 409.
original, which was famous in his native country. The 43 Havelock 2007, 75.

support on the left side is in the shape of a dolphin on 44 Ibid.

whose back two Eroti play.40 The head is sharply turned 45 Ibid. She mentions on page 78 that B. Felleti Maj in the

to the left. The hair is shorter and simply tied on the quoted paper confirmed 101 replicas of the type Venus Capitolina.
Ridgway 2001, 355 sq. citing different replicas.
back of the head and part of the forelocks is pulled from 46 The Louvre, inv. no. MA 571 (MR 671), dated to the 2nd
the forehead towards the top of the head, but without a century AD. Photo in commons.wikimedia.org/wiki/Category:
big knot (top knot). There are 33 known replicas.41 The Capitolina Venus.

145
VASI], GAVRILOVI], Venus or Diana from Mediana (137149) STARINAR LXII/2012

elements concerning the sculpting of the locks and the which can be seen with the sculptor Menophantes and
top of the head can be perceived the same on this head maybe also with the replica of Venus Medici if, indeed,
as on Medianas head, although they are still vague it really was modelled by Kleomenes from Athens in
similarities. A less emphasised top knot is present on the 1st century BC. Of course, that is not the case only
the statue of Venus Pudica with dolphin, discovered in with these two goddesses, but also with other deities.
Odeon, Cartagena.47 The head of this sculpture is, how- The analysis clearly showed to which extent certain
ever, stylistically different from Medianas head. types could have been treated differently, either
We will also mention the group of sculptures of the because of the passage of time, or because of the skill
crouching Aphrodite, which have the same coiffure as of the sculptor. That is the case with Medianas head
Venus Capitolina. Maybe the most famous and best which displays the main characteristics of sculptures
modelled of these is the statue of Aphrodite found in such as Artemis Rospigliosi or Aphrodite/Venus, first
Hadrians villa in Tivoli.48 The original sculpture of seen in Aphrodite from Cnidus and Venus Capitolina.
Aphrodite crouching is dated to the late 3rd century BC However, the sculpture from Mediana shows a quite
and, according to Pliny the Elder (NH 36. 21; 36. 35), unique style which can be seen in the careful treatment
it was made by the sculptor Doidalsas from Bithiny, of the narrow face, and especially in the modelling of
and was ordered by King Nikomedes around 250 BC. the coiffure. The almost geometric treatment of the
Most authors accept this dating. However, a certain forelocks on the forehead and the linear, flat represen-
number of authors reject the very existence of Doidalsas, tation of the forelocks, which go from the forehead to
thinking that the text by Pliny is corrupted and there- the top of the head and the top knot, could not be found
fore misinterpreted.49 Most of the replicas are life-sized, in any other sculpture which could have been a copy or
but there are also smaller statues. As Christine Have- a prototype. This kind of stylised treatment most close-
lock mentions, variants in different materials were ly resembles the art found on one side of the relief on
more often copied in the period of the Roman Empire Galerius arch in Thessaloniki and partially resembles
than in later Hellenism.50 that found on the relief on the arch in Rome from Con-
A very similar representation of the coiffure can be stantines period (especially the scene oratio Augusti
seen in the sculpture of Venus with Eros, which doesnt and liberalitas), on which the clothes were modelled in
belong to the group of Venus Medici and Venus Capito- a particularly flat way.52 The stylisation of Medianas
lina. It is a marble composition, found in Rome, which head is probably not as a consequence of a less skilful
represents a Roman copy from the 2nd century AD, artisan, but of an artistic concept which was widely
modelled on a Greek original from the 4th century BC.51 accepted in late antiquity. Therefore, we can take the two
The goddess has waved hair divided in the middle, a aforementioned arches as paradigms of new ideas of
top knot tied with ribbons on the top of her head and a modelling sculptures which were partly derived from
combed bun on the back of her head. Besides the country art and wood engraving. This would lead to
unusually close similarity in treatment of the hair and the conclusion that the head from Mediana could be
coiffure, close analogies can be perceived in the narrow, dated to the second and third decades of the 4th century.
oval face, the high forehead, the nose and the small That date could certainly be prolonged to the middle of
lips, with the only difference being the somewhat that century, but it could date back to the end of the 3rd
deeper modelled almond shaped eyes on the sculpture
from the Louvre.
With this, we conclude the list of possible analogous
sculptures of Artemis/Diana and Aphrodite/Venus. 47 Odeon was built at the beginning of the 3rd century AD.
The same objections that were made concerning repre- Museum Bardo, Tunisia, inv. no. C.923 Yacoub 1970, 46. LIMC
sentations of Artemis/Diana can be applied to sculp- II. 1, p. 85; vol. II. 2, p. 76, fig. 737
tures of Aphrodite/Venus. It is not our intention, with 48 Inv. no. 108597. LIMC II, s. v. Aphrodite, no. 1018.
49 Ridgway 2001, 230. Ridgway 2002, 116, with cited litera-
the simple numbering of sculptures, to determine the
identification or the chronology of Medianas head, but ture.
50 Havelock 2007, 85. For different versions see LIMC II, s. v.
to briefly analyse the appearance of Roman copies of
Aphrodite, nos. 101827; 103035.
Hellenistic originals, which certainly enjoy significant 51 LIMC II.1. Aphrodite, 57, n. 456.
prestige among art lovers. It has been shown already that 52 See good quality photographs and commentary in Frova

older statues, in the Hellenistic period, were copied, 1961, 341348, figs. 320325.

146
VASI], GAVRILOVI], Venus or Diana from Mediana (137149) STARINAR LXII/2012

on which the feet are represented has the following


dimensions: length = 47 cm, width = 27 cm. The feet
are 21 cm long and do not cover the whole width of the
base. This size of feet would be consistent with a statue
of approximately 1.65 m, which would be anthropo-
metrically consistent with the dimensions of the head.
Therefore, the head and feet could belong to a sculpture
of approximately natural height (although we could ask
the question, what was natural height in late antiquity,
considering the average height of people in the afore-
mentioned period). The toes of the feet are meticu-
lously modelled, but the head of the dolphin is highly
stylised and suits the stylisation of the head. Maybe,
the analysis of the marble of the head and base would
Fig. 10. The base with feet and dolphin's head show that they are from the same sculpture. For now,
Sl. 10. Baza sa stopalima i glavom delfina we suggest that the head and the base with the feet are
parts of the same statue, and that the head be identified
as the Venus Capitolina type, subtype Venus with dol-
phin and would also suggest a slightly wider date of
century. The equal use of the chisel and the drill modelling the end of the 3rd century AD 340. Also,
implies the homogeneity of the workshop, which is the find of the goddess head made the composition of
one of the characteristics of art from the period of the sculptures, which ornamented the villa with peri-
Tetrarchy and Constantine. Therefore, we would not style, significantly richer in the context of the residential
regard this style as decadent or provincial but, rather, complex of Constantine the Great and his heirs in
as a style which tried to introduce new elements in an Mediana. We will not enter into a discussion as to
effort to break the habits of the old style. However, we whether it is possible to find, in residences of Christian
lack more obvious parameters which would suggest in emperors, sculptures or mosaics with mythological
which workshop the head from Mediana was made. scenes. It is enough to say that in the majority of
Considering the dimensions of the head from known aristocratic and imperial villas all over the
Mediana, it can be presumed that it represented a part Roman Empire, similar ensembles were discovered.
of a sculpture of almost life size. Unfortunately, the Therefore, it is no wonder that in the decoration of
absence of any iconographic elements such as attributes Medianas villa with peristyle, which was presumably
or figures which, together with the goddess, could have done in the period of Constans (337350), different
formed a particular composition, makes it difficult to sculptures with a mythological content were found, but
form any further opinions as to whether the statue was which lost that connotation and came to be considered
standing or sitting. However, it is worth mentioning a either as works of art or in a new Christian ideological
hypothesis that the base with feet and a dolphin, found context.54
in 2002, could belong to the same hypothetical com-
position as the head from Mediana (fig. 10).53 The base Translated by Nade`da Gavrilovi}

53 Vasi} 2004a, 294. Marble base with feet and fragmented

dolphin base of the statue of Venus (Venus Pudica), sondages 24


and 24a, middle aged pit, inv. n. C144a/02.
54 For details about these questions see Stirling 2008a, passim.

147
VASI], GAVRILOVI], Venus or Diana from Mediana (137149) STARINAR LXII/2012

BIBLIOGRAPHY:

Ajootian 1998 A. Ajootian, Praxiteles, in: Olga Ouertani 1994 N. Ouertani, Remarques propos
Palagia, J. J. Pollit (eds.), Personal Styles in Greek de la collection en marbre, in: Das Wrack. Der antike
Sculpture, Cambridge University Press, 1998. Schiffsfund von Mahdia, Reinland Verlag, Kln 1994.
Bonfante 2003 L. Bonfante, Etruscan Dress, Bal- Ridgway 2001 B. S. Ridgway, Hellenistic Sculp-
timore 2003. ture I. The Styles of ca. 331200 B.C., The University of
Corso 2007 A. Corso, The Art of Praxiteles II. The Wisconsin Press, Madison 2001.
Mature Years, L Erma di Bretschneider, Roma 2007. Ridgway 2002 B. S. Ridgway, Hellenistic Sculp-
Dr~a 2004 S. Dr~a, u: Arheolo{ko blago Ni{a od ture III. The Styles of ca. 100 31 B.C., The University of
neolita do srednjeg veka (Archaeological Treasure of Wisconsin Press, Madison 2002.
Ni{ from the Neolithic to the Middle Ages), Serbian Stirling 2008 L. Stirling, Pagan Statuettes in Late
Academy of Sciences and Arts, Belgrade 2004. Antique Corinth: Sculpture from the Panayia Domus,
Frova 1961 A. Frova, L arte di Roma e del mondo Hesperia 77, 2008, 89161.
romano, Storia universale dell arte, volume II/2, Torino Stirling 2008a L. Stirling, The Learned Collector.
1961. Mythological Statuettes and Classical Taste in Late
Guberti Basset 2000 S. Guberti Basset, Excellent Antique Gaul, The University of Michigan Press, Ann
Offerings: The Lausos Collection in Constantinople, Arbor 2008.
The Art Bulletin 82/1, 2000, 625. Tomovi} 1992 M. Tomovi}, Roman Sculpture in
Havelock 2007 C. M. Havelock, The Aphrodite of Upper Moesia, Belgrade 1992.
Knidos and Her Successors: A Historical Review of the Vasi} 2004 M. Vasi}, Bronze Railing from Medi-
Femal Nude in Greek Art, University of Michigan Press ana, Starinar LIIILIV/20032004, (2004), 79109.
2007. Vasi} 2004a M. Vasi}, Hronika iskopavawa
LIMC Lexicon Iconographicum Mythologiae Medijane 20002002. godine, Starinar LIIILIV/
Classicae. 20032004, (2004), 288294.
LIMC II.1 A. Delivorrias, G. Berger-Doer, A. Vasi} 2005 M. Vasi}, Mediana domaine imprial
Kossatz-Deissmann, Aphrodite, II.1, ZrichMnchen ou bien priv?, Rmische Stdte und Festungen an der
1984, 2151. Donau, Beograd 2005, 167176.
LIMC II.1 E. Simon, G. Bauchhenss, Artemis/ Vasi} 2007 M. Vasi}, Mediana die kaserliche
Diana, II.1, ZrichMnchen 1984, 792855. Villa bei Ni{, in: Ulrich Brandl, Miloje Vasi} (Hrsg.),
LIMC VIII.1 E. Schmidt, Venus, LIMC (Lexicon Roms Erbe auf dem Balkan. Sptantike Kaiservillen und
Iconographicum Mythologiae Classicae), II.1, Zrich Stadtanlagen in Serbien, Philipp von Zabern, Mainz
Dsseldorf 1997, 192230. 2007, 96107.
Mili}evi} Brada~ 2009 M. Mili}evi} Brada~, Waelkens 2004 M. Waelkens, Sculptural Studies:
Spomeni bo`ice Dijane iz kolonije Claudia Aequum i Introduction, in: Interactive Dig Sagalassos (www.archa-
logora Tirilium, Opuscula Archaeologica, 33, Zagreb elogy.org/interactive/sagalassos/field04/sculpture1.html).
2009, 5178. Yacoub 1970 M. Yacoub, Le muse du Bardo,
Murray 2004 A. S. Murray, A History of Greek Institut National d Archologie et d Arts, Tunis 1970.
Sculpture, Kessinger Publishing 2004.

148
VASI], GAVRILOVI], Venus or Diana from Mediana (137149) STARINAR LXII/2012

Rezime: MILOJE VASI], Arheolo{ki institut, Beograd


NADE@DA GAVRILOVI], Arheolo{ki institut, Beograd

VENERA ILI DIJANA IZ MEDIJANE

Kqu~ne re~i. Medijana, skulptura, kasna antika.

U iskopavawima vile s peristilom u Medijani tokom kam- Medijane upu}uju vi{e na kasnu antiku i na nove koncep-
pawe 2011. godine otkrivena je na gorwoj povr{ini prvo- cije zapo~ete tokom Tetrarhije a nastavqene pod konstan-
bitnog ru{evinskog sloja od tavanice i krova (sloj S), u tinijanskom dinastijom. Zbog toga je za izradu ove glave
zapadnoj prostoriji 1, mermerna glava koja bi po karakte- predlo`en datum: kraj 3 340. godina. Izneta je mogu}nost
ristikama frizure pripadala nekoj bogiwi. Analiza je po- da su novootkrivena glava i baza otkrivena 2002. godine,
kazala da najbli`e analogije mo`emo na}i kod predstava na kojoj se nalaze fragmentovana stopala i glava delfina,
Artemide/Dijane i Afrodite/Venere. Uzori medijanskoj delovi iste statue.
glavi se`u do helenisti~ke umetnosti. Po wima su ra|ene U tom slu~aju, ova statua mo`e da se identifikuje kao
replike u rimskoj umetnosti, koje su uglavnom iz 12. veka tip Kapitolske Venere, podtip Venera s delfinom ili Ve-
posle Hr. Stilske karakteristike obrade kose na glavi iz nera Pudika.

149
STEFAN POP-LAZI]
The Institute of Archaeology, Belgrade

SOME OBSERVATIONS ON LEAD FIGURINES


OF THE GODDESS VENUS IN THE AREA
BETWEEN SIRMIUM AND VIMINACIUM

UDK: 904:739.5(497.113)"02" e-mail: stefanpo@gmail.com


DOI: 10.2298/STA1262151P Received: February 28, 2012
Original research article Accepted: June 21, 2012

Abstract. This paper deals with the occurrence of lead figurines of the goddess Venus in the area between the two cities
of Sirmium and Viminacium. A classification into five types, based on stylistic and iconographic features, has been proposed.
Although these types show features in common with products of other workshops in the Danubian basin, it has been observed
that such products were not exchanged among the centres in the area. This paper puts forward the hypothesis that there were
two centres of worship of the goddess Venus, in Sirmium and Viminacium, where these figurines could have been manufactured.

Key words. Lead figurines, Venus, Sirmium, Viminacium, Lower Pannonia, Upper Moesia.

n the territory of the Roman Empire, lead figuri-

I nes were not a rare commodity. They were usually


of rather small dimensions, simply crafted and were
often manufactured using shallow two-part moulds. The
1
2
Chew 1991, 82.
Chew 1991, 8889; Gschaid 1994, 442, Venera (Pupilin).
3 Wagner 1991, Minerva on a pedestal (Langenhein), 265269,

Mercury on a pedestal (Heddernheim) 269270; Lar on a pedestal


figures were usually placed on pedestals, in aediculae or (Mainz) Frenz 1988 S. 247, Abb. 4043; Venus (Vallon sur Dompi-
set into a frame imitating the front of a temple. Although erre) Monnier 2009, 206207.
4 Chew 1991, 8889; Bolla 2004, 6971, Fortuna and Amor on
the manufacturing methods varied, they show both com-
a dolphin (Verona); a tin figure of Venus (Esquilin) Bauer 1936, 19.
mon features and certain specific traits determined by 5 Bauer 1936, 12; Chew 1991, 88.
the place of manufacturing. Examples of lead figurines 6 Siscia (Bauer 1936, 9); P. Zsidi collected 68 lead statuettes
known so far, originate from Britannia,1 Gallia,2 Ger- originating from Hungarian part of Pannonia (Zsidi 2000, 324327);
mania,3 Italia,4 Noricum,5 Pannonia,6 Moesia,7 Dacia8 Parovi}-Pe{ikan 1971, published two figurines from Sirmium; Dau-
and Syria9 (Map I). Within the territory of the Empire, tova-Ru{evljan 2006, 345346. Hrtkovci, Gomolava, Kukujevci.
7 Vin~a II 1936, 154155; Viminacium (Popovi} 1992, 4546);
it is possible to distinguish the areas with the greatest Durostorum (Donevski 1975, 146147; Elefterescu 2005, 221238).
number and variety of represented divinities. The char- 8 A figurine featuring Mercury and two Venus (?) figurines
acteristic feature of the figurines of divinities originat- originate from Dacia (Benea 2007a, 119; Benea 2007, 544545; Benea
ing from Britannia and Gallia is the lead aedicula into 2008, 242243.). It is interesting to note that Benea mentions a frag-
mented sculpture of Mercury found at ^air (Viminacium). However,
which they were set. In the Danubian basin, namely in none of the pieces published by I. Popovi} can be identified as the
Pannonia and Lower Moesia, divinities were set into fragmented statuette of Mercury.
aedicula-shaped frames. The territory between Sirmium 9 Badre 1999, 181 ff.

* The article results from the project: Romanization, urbanization and transformation of urban centres of civil, military and residential
character in Roman provinces on territory of Serbia (no 177007) funded by Ministry of Education, Science and Technological Development
of the Republic of Serbia.

151
Stefan POP-LAZI], Some Observations on Lead Figurines of the Goddess Venus (151164) STARINAR LXII/2012

in Lower Pannonia and Viminacium in Upper Moesia largest collection of finds from Viminacium is held by
is one of the areas marked by a specific repertoire of the National Museum in Po`arevac; this collection has
representations, where several dozen such pieces have been partially published. The Museum of Vojvodina
been found so far. The most significant characteristic of holds the pieces found at Gomolava, Hrtkovci and Ku-
the figurines from this area is a very plain and simpli- kujevci. The Museum of Srem in Sremska Mitrovica
fied image of the goddess Venus. Such a style led some has a collection of figurines found at Sirmium and the
authors to question the identity of the Roman goddess. surrounding area; no more than two pieces from this
The purpose of this study is to show that the afore- collection have been presented. The collections of the
mentioned figurines do represent the goddess Venus, Museum in [abac, The Historical Museum of Serbia,
as well as to shed light on certain characteristics of the The National Museum in Belgrade and The Belgrade
workshops in which they were manufactured. City Museum, which include pieces found at archaeo-
A rather small number of figurines were found dur- logical sites in Ma~va, Obrenovac, Kosmaj, Zemun and
ing excavations at Sirmium and Gomolava (six pieces), at Viminacium, have yet to be published.10 The entire
with most of them being purchased for museums. The corpus of the figurines, including more than fifty pieces,

Map I. Finds of lead figurines within the Roman Empire


Karta I. Nalazi{ta olovnih figurina u Rimskom Carstvu

1. Chesterholm (Vindolanda) 2. Wallsend (Segedunum); 3. Wroxeter


(Viroconium Cornoviorum); 4. Dorset; 5.Marquis 6. Louvignies Bavay;
7. Foret de Compiegne; 8. Boulogne-La-Grasse; 9.Saint Felix; 10. Lyon;
11. Langenhein; 12. Hedernheim; 13. Tharros; 14. Roma; 15. Viterbo;
16. Pessaro; 17. Comacchio;18. San Giorgio di Valpolicella; 19.Gurina;
20. Magdalensberg; 21. Wien (Vindobona); 22. Ptuj (Poetovio); 23. Sisak
(Siscia); 24. Szombathely (Savaria); 25. Mrichida (Mursella); 26. Szony
(Brigetio); 27. Budapest (Aquincum); 28. Szzhalombatta (Matrica);
29. Dunajvros (Intercisa); 30. Gyulafirtt Pognyteleknek; 31. Tc
(Gorsium); 32. Kukujevci; 33. Sremska Mitrovica (Sirmium), Ma~vanska
Mitrovica, Sala{ No}ajski; 34. [abac; 35. Donji Petrovci (Bassiana); 36.
Zemun (Taurunum); 37. Vin~a; 38. Stari Kostolac (Viminacium); 39. Silistra
(Durostorum), Clrai; 40. Izmir (Smirna); 41. Susa; 42. Ain-Al-Djoudj

152
Stefan POP-LAZI], Some Observations on Lead Figurines of the Goddess Venus (151164) STARINAR LXII/2012

either fully preserved or fragmented, can be classified Type II (Figs. 2/79)


into several types distinguished by the posture of the This type represents a rather schematised and sim-
naked female body and the style of its rendering in plified version of the famous figural pose, Venus Pudi-
sculpture. ca.22 Only in the item originating from Vin~a has the
whole figure been preserved, whereas the other five
Type I (Figs. 1/16) pieces are fragmented. Depending on the position of
The characteristics of this type have been determi-
ned on the basis of the best preserved group from Sir-
mium featuring Venus and Amor on a pedestal. The
10 I take this opportunity to express my gratitude to my col-
figure of Venus was rendered rather schematically,
leagues Pavle Popovi}, Jasmina Davidovi} and Miroslav Jesreti},
whereas the figurines made using a shallow one-piece
curators at the Museum of Srem, Dr. Slavica Kruni}, museum coun-
mould are pronouncedly two-dimensional. The torso sellor at the Belgrade City Museum and Marko Vuksan, senior
and legs are almost level with one another. The legs curator at the Historical Museum of Serbia for allowing me to pub-
and arms are shown as two bands and the natural width lish, so far unpublished, materials. I am also grateful to Dr. Tatjana
Cvjeti}anin, museum counsellor at the National Museum in
of the hips and thighs is indistinguishable. The raised
Belgrade, Dragana Spasi}, curator of the Museum of Po`arevac and
right hand touches the head, while the left hand rests Gordana Kovi}, curator at the National Museum of [abac for pro-
on the hips. The breasts are shown as two nipples. The viding me access to their collections and supplying information on
rear of the body is flattened, without any distinguish- the items from their collections.
11 LIMC VIII, 204, 206207.
able gluteal muscles or any transition between the hips 12 Elefterescu 2005, 223, Cat. No. 6 (IX/2).
and the back. The head is round with linearly rendered 13 Ognenova-Marinova 1975, 154, Cat. No. 176, Fig. 176.
nose, mouth and eyes on the face, whereas the nape is 14 Dautova-Ru{evljan 2006, 346, Cat. No. 9. Fig. 1/9.
flat and coarse. The diadem is shown as a row of grains 15 Apart from the published lead figurine of Venus with Amor
on the forehead and the parted hair is rendered using a on a pedestal found at Site No. 4 (Parovi}-Pe{ikan, 1971, 3738, T.
series of lines. Bracelets can be seen on both upper and XVII/64), the documentation of the Museum of Srem preserves the
record of an additional two lead figurines of Venus on a pedestal
forearms. In three items belonging to this type, the fig-
found in Sirmium. The first of them can be found in the inventory
ure of Venus on a pedestal is accompanied by a smaller of the archaeological excavations carried out in Sirmium in 1958 as
figure of Amor. It is also rather simplified, with long the find from Ju`ni Bedem (Inv. No. 123). On the basis of the drawing
arms and legs, whereas other body parts are not clear- it may be concluded that the same group has already been published
as a find from an unknown site (Parovi}-Pe{ikan, 1971, 3738, T.
ly articulated.
XVII/64a). The other, unpublished, example was found at Site No.
As far as the positions of the right hand (holding 26. The height of the preserved part including the pedestal is 13.7
the hair) and the left hand (placed on the hips) are con- cm. The figurine bears the field inventory number 1046.
16 A fragment of the head and torso of a lead figurine is listed
cerned, this representation has shared features with two
types of images representing a standing figure of a nude in the inventory of the Museum of Srem, Nr. 2168. Height: 5.2 cm.
Unpublished.
Venus Venus Anadyomene and Venus Pudica. 11 The 17 Dautova-Ru{evljan 2006, 345, Cat. Nos. 1, 2, 4. Fig. 1/1, 2, 4.
artist was obviously unfamiliar with the characteristics 18 Tasi} 1965, 30, T XIV/1; Dautova-Ru{evljan, Brukner 1992,
of these two types of images, which were based on the 65 Cat. Nos. 33, 34, T 8/38; (Dautova-Ru{evljan 2006, 345, Cat.
story of Aphrodite rising from the foaming sea and Nos. 3, 9. Fig. 1/3, 9.
19 Najhold 2010, 211, Figs. 122, 123.
wringing her hair and the figural pose of the modest
20 Popovi} 1992, 40, Cat. Nos. 50, 51; the Historical Museum
Aphrodite covering parts of her body with her arms.
of Serbia holds a fragment of the hips of a female figurine Inv. No.
Among the lead figures of Venus, the figurine from the 476 (Fig.) and a fragmented Amor figurine Inv. No. 477(Fig.) which
area of Durostorum12 is similar to this type in the posi- belong to this type.
tion of the hands and in its simple style. The same posi- 21 Two fragmented statuettes published by Chew belong to

tion of the hands and the simple representation of the Type I. Both were gifted to the Muse des Antiquits nationales.
One is said to originate from the Reims region (Chew 1990/1991,
human body, as well as the small dimensions, are also 81) and the author presumes that the other, published in an auction
characteristics of the bronze figurine found at Chirpan catalogue as a Celtic product from the Danubian basin, was actually
(Pizus) in south Thrace.13 found in France (Chew 1990/1991, 81, 88). Bearing in mind the cir-
Source sites: Kukujevci (Fig. 1/6),14 Sirmium Sites cumstances under which both statuettes reached the Museum, we
suggest that both were found somewhere in the Danubian basin; in the
Nos. 4 (Fig. 1/1), 26 (Fig. 1/4), Ju`ni Bedem (Fig. 1/3),15 late 1980s they might have come to be owned by third persons, who
Sala{ No}ajski,16 Hrtkovci,17 Gomolava,18 Taurunum,19 subsequently gifted them to the Muse des Antiquits nationales.
Viminacium, (Figs. 1/2, 5)20 an unknown site.21 22 LIMC VIII, 204, LIMC II, 52.

153
Stefan POP-LAZI], Some Observations on Lead Figurines of the Goddess Venus (151164) STARINAR LXII/2012

Fig. 1. Type I: 1, 3, 4) Sirmium; 2) Viminacium; 5) Sala{ No}ajski; 6) Gomolava


Sl. 1. Tip I: 1, 3, 4) Sirmijum; 2) Viminacijum; 5) Sala{ No}ajski; 6) Gomolava

154
Stefan POP-LAZI], Some Observations on Lead Figurines of the Goddess Venus (151164) STARINAR LXII/2012

7 9

Fig. 2. Type II: 7) Vin~a; 8) Viminacium; 9) Sirmium


Sl. 2. Tip II: 7) Vin~a; 8) Viminacijum; 9) Sirmijum

the arms, two variants can be distinguished within this Type III Viminacium (Figs. 3/1013)
type. The first variant includes the figurines in which None of the items classified as Type III are fully
the right arm is bent at the elbow and rests below the preserved. It is basically a variant of the Venus Pudica
breasts, whereas the left one, also bent at the elbow, type but, unlike the previous type, the body is more
rests on the belly or hips. The overall appearance of the harmoniously shaped. Similar to Type I, the eyes, face
body is somewhat more natural than in the first type, and nose on the round head are schematically ren-
although the stiffness of the arms and the unnatural dered. The diadem on the forehead consists of grains
angles at which they are bent can still be observed. The arranged in an arched pattern, while the hair is
curves of the body are more pronounced; the rear of arranged in tufts. The right arm is tight to the body and
the figurines is sculpturally rendered and parts of the it rests below the breasts, while the left hand is placed
gluteal muscles and back are distinguishable. The head above the pubis. The hands are oval and, unlike the
is preserved only in one item from Viminacium and it previous two types, the fingers are not outlined. A
differs in size from the heads typical of types I and III. necklace is shown around the neck and bracelets
Within this variant, it can be observed that two types of around the arms. A distinguishing feature of this type
mould were used: the items from Viminacium and is the jewellery, rendered using tiny grains, crossed on
Vin~a are identical, as are the pieces from Viminacium the breasts and back. A long chain wrapped around the
and Hrtkovci. The second variant within this type shoulders and crossed on the back and on the breasts,
includes two figurines in which the positions of the left
and right arms are interchanged: the left rests below
the breasts, whilst the right is placed on the hip. Two
pieces that belong to this variant have different dimen- 23 The figurine was found during construction work. Museum

sions which indicates that at least two moulds were of Srem Inv. No. 1144.
24 Dautova-Ru{evljan 2006, 345, 346, Cat. Nos. 5, 6, Fig. 1/5, 6.
used for the manufacturing of these figurines. Lead
25 Reinach 1913, P. 233/5; Vin~a II, 154155, Fig. 324 a, b;
statuettes similar to this type cannot be found in any
Vin~a III, 121124, Fig. 557 a, b.
other areas. 26 Historical Museum of Serbia, Archaeological collection
Source sites: Sirmium Site Kej23 (Fig. 2/9), Hrt- I.B. 475; Popovi} 1992, 40, Cat. No. 52; Milovanovi} 2008, 166,
kovci,24 Vin~a (Fig. 2/7),25 Viminacium (Fig. 2/8).26 Cat. No. 13, Pl. IV. 22,23.

155
Stefan POP-LAZI], Some Observations on Lead Figurines of the Goddess Venus (151164) STARINAR LXII/2012

10 11

12 13

Fig. 3. Type III: 10, 11) Viminacium; 12, 13) Sala{ No}ajski
Sl. 3. Tip III: 10, 11) Viminacijum; 12, 13) Sala{ No}ajski

is fastened by a decorative clasp. Apart from this item, long period of time, but reappeared in the territory of
only a lead figurine found in Klagenfurt 27 features Dacia and Lower Moesia in the Roman period.
breast jewellery and as such it is, so far, a unique type. Several bronze figurines feature the same type of
This piece shows a cloth wrapped around the hips with jewellery. In the village of Micia in Dacia, a bronze
the crossed chain appearing to secure it. figurine of Venus has been found in which the motif of
This specific type of jewellery was characteristic of a crossed chain on the breasts and on the back was ren-
the terracotta figurines featuring the goddess Aphrodite dered using punctured circles.34 A similar figurine,
since the late fifth century BC, the period to which the almost identical to the previous one, was unearthed in
Aphrodite in a shell from Fanagoria (Taman) has been the territory of Thrace in Negovanci.35 The same motif
dated.28 A very similar piece from Tanagra29 has been is rendered in incised crossed lines on another two bronze
dated to a somewhat later period, i.e. the early fourth
century BC. A terracotta figurine with abundantly
adorned legs and arms, along with the crossed jewel-
lery across the breasts, originates from a tomb used 27 Bauer 1936, 18, 16a, b.
between the sixth and third centuries BC.30 The terra- 28 Winter 1904, 203/3; LIMC II, 109, Cat. No. 1083.
29 Winter 1904, 203/2; LIMC II, 109, Cat. No. 1084.
cotta from Myrina has been dated to the first century
30 Duyuran 1960, 11, Pl. XI.
BC.31 The same type of jewellery is featured on terra-
31 Winter 1904, 220/5; LIMC II, 86, Cat. No. 779.
cotta figurines from southern Italy and Jordan, but 32 Southern Italy, Winter 1904, 213,3; Jordan, LIMC II, 159,
their dating has not yet been determined.32 Apart from Cat. No. 111.
images of Aphrodite, the crossed chain across the breasts 33 Brown 1984, 2123.

appears on other representations of women during 34 Teposu-Marinescu, Pop 2000, 96, Cat. No. 112., Pl. 59.

Hellenism.33 Similar images were not recorded over a 35 Ognenova-Marinova 1975, 150, Cat. No. 171, Fig. 171.

156
Stefan POP-LAZI], Some Observations on Lead Figurines of the Goddess Venus (151164) STARINAR LXII/2012

14 15 16

Fig. 4. Type IV: 14) Taurunum; Type V: 15, 16) Viminacium


Sl. 7. Tip IV: 14) Taurunum; Tip V: 15, 16) Viminacijum

figurines of the goddess Venus those from Barzica on the pubic area.41 The diversity of the representations
and Devnja,36 which are, just like the previously men- of Venus with a mantle wrapped around her hips is
tioned two pieces, very similar to one another. Our figu- illustrated by two different pieces from Siscia. One of
rines found in Viminacium belong to the Venus Pudica them features raised arms holding plaits,42 whereas the
type, the Berzica and Devnja figurines to the Venus Ana- other has broken arms of indeterminable position.43 The
dyomene type, and those found in Micia and Negovanci only similiar items that can be found are two identical
to the type of Venus with the outstretched arm. fragmented pieces from Smyrna,44 held by the Louvre.
Source sites: Sala{ No}ajski (Figs. 3/11, 13),37 Vimi- The difference between the figurines from Smyrna and
nacium (Figs. 3/10, 12).38 the one from Taurunum lies in the better rendering of
folds in the mantle wrapped around the hips of the former
Type IV (Fig. 4/14) pieces. As both figurines from Smyrna are broken above
This type has been defined on the basis of a piece the hips and no other similar examples are known to have
found on the bank of the River Danube in Zemun. The been found, it is not possible to discuss the rendering
standing nude female figurine has a mantle wrapped
around the hips. The breasts are marked by two nipple-
like protrusions and the navel by a shallow depression.
36 Ognenova-Marinova 1975, Barzica 144, Cat. No. 163, Fig.
The body is narrow at the waist, widening towards the
163; Devnja 146, Cat. No. 164, Fig. 164.
hips. The mantle around the hips is suggested by radial 37 Two lead figurines were gifted to the Museum of Srem.
folds. Since its arms are broken below the elbows and According to the person who found them, they were discovered in
the head is missing, it is not possible to classify it as a a field at Sala{ No}ajski. Several fragmented lead statuettes acquired
particular type within the corpus of the Aphrodite/ for the National Museum in Belgrade in 1933 originate from the
immediate vicinity of this site, at Ma~vanska Mala Mitrovica
Venus representations. What it has in common with the (Gara{anin 1951, 164).
previous three types is the two-dimensionality of its 38 Popovi} 1992, 40, Cat. Nos. 53, 54. Two unpublished figu-

representation and the unskilfully rendered body pos- rines held by the Historical Museum of Serbia originate from
ture. It has similarities to several lead figurines from Viminacium: Inv. No. 471, preserved height: 8.3 cm. Inv. No. 472;
preserved height: 7.3 cm.
the broader area of the Danubian basin. Two figurines 39 Elefterescu 20042005, 222223, Cat. No. 4, Pls. II/1, VI/1
are identical; one of them was found at Durostorum,39 40 Thomas 1952, 35, Pl. VI/3.
whereas the location where the other was found is not 41 Thomas 1952, 35, Pl. VI/6.

known. 40 With both sculptures, the mantle is wrapped 42 Bauer 1936, 17, Cat. No. 14, T III/14a, b.
43 Bauer 1936, 17, Cat. No. 13, T III/13a, b. The third Venus
only around the left leg, while the right leg is totally bare.
The left hand covers the pubic area and the right arm is figurine (Cat. No. 15) features the mantle only on the back; however,
the figurine was made using two different moulds (in this case, a
broken in two, therefore, its position is not known. The mould for a figure of a divinity dressed in a long chiton gathered at
figurine found at Brigetio was sculpted in a different the waist was used; 1718).
manner and it shows a mantled Venus with her left hand 44 Chew 1990/1991, 88, Cat. Nos. 41, 42, Pls. 41, 42.

157
Stefan POP-LAZI], Some Observations on Lead Figurines of the Goddess Venus (151164) STARINAR LXII/2012

of the torso and, therefore, suggest the hypothesis that urines belonging to various types at three equally dis-
the Taurunum figurine could have originated from Asia tant locations Sirmium, Viminacium and Hrtkovci,
Minor. Under the present circumstances, this is the only does not allow us to reliably distinguish the place of
piece that bears any resemblance to lead figurines origi- manufacture from the place of utilisation. Although
nating from other parts of the Roman Empire. the lead figurines from the area between Sirmium and
Source site: Taurunum (Fig. 4/14).45 Viminacium do not bear any direct similarities to the
repertoire of representations from the Hungarian part
Type V (Fig. 4/15, 16) of Pannonia and Lower Moesia, in terms of style the
The distinctive feature of this type is a significantly figurines belonging to types I and II show similarities
different sculptural rendering of the body, which is, to the pieces from Savaria, Brigetio and Durostorum.50
unlike the previous types, three-dimensional; the ren- They are distinguished by the very unskilful and
dering of the head and face is also considerably differ- schematic rendering of the head and body. It can also
ent. In the figurine from Taurunum,46 the woman holds be noticed that there are several types of representa-
her hair with her left hand, whilst her right arm is out- tions, of varying manufacturing quality, in the Upper
stretched. The torso is unskilfully rendered and the fig- and Lower Danubian basin. In the Serbian part of the
ure is pronouncedly steatopygous. Differences are also Danubian basin, it is possible to observe differences
observed in the manner in which the heads are sculpted. between the first three types and the fourth group,
The eyes, nose and mouth are roughly outlined in relief, which was made by a more skilled artisan. It seems
while the diadem is shaped as a triangular extension that the existence of the fifth group can be explained
above the forehead. The two items found at Viminacium by reasons different from those of the other four groups.
share some common features with the previously men- In our opinion, its similarity to pieces made of bronze,
tioned figurines. In the first one47 the head is missing, and primarily their three dimensional appearance,
however, judging by other features (the torso, hips, indicates that it was created either within the circle of
steatopygous character and arms) and its dimensions, workshops manufacturing bronze objects or under their
it is possible that it was similar to the previous examples. influence.
The other figurine,48 originating from Viminacium, is For the majority of the lead figurines from the ter-
somewhat more pronouncedly three-dimensional, hav- ritory between Sirmium and Viminacium there is no
ing fuller breast and a head sculpted differently from information about the context of the find. Therefore, it
all other lead figurines. Within the corpus of lead figu- is not possible to reliably establish the time of their
rines found in Dacia, Upper and Lower Moesia and emergence in this region. Only one piece (Site No. 4 at
Thrace,49 there are a large number of pieces featuring Sirmium) can be more precisely dated to the second
Venus with her right arm outstretched, whilst her left
hand is holding a plait. The quality of rendering of the
body varies, from very successfully sculpted pieces to
those similar to the crude rendering of the face and 45 Zemun. Danube River bank. Site III Pumpa. Registered in the
body in the Taurunum figurine. Therefore, it is possi- inventory of the Antique Collection of the Belgrade City Museum
ble that the items belonging to this type were made in under number 4376. It is preserved up to the height of 6.4 cm.
Unpublished.
a workshop manufacturing bronze figurines. 46 Najhold 2009, 109.
Source sites: Taurunum, Viminacium. 47 Viminacium. Fortuitous find. Standing nude, pronouncedly

steatopygous female figure. The head, left arm and legs are missing.
The right arm is hanging by the body. The preserved height is 3.9 cm.
***
It is registered in the inventory of the Archaeological Collection of
the Historical Museum of Serbia under number 474. Unpublished.
Bearing in mind the number of discovered items, the 48 Viminacium. Fortuitous find. Standing female figurine. It
group of lead figurines found in a relatively narrow is registered in the inventory of the Archaeological Collection of the
territory between Sirmium and Viminacium does not Historical Museum of Serbia under number 473. The preserved
show a great variety of types. Generally speaking, the height is 5.4 cm. Unpublished.
49 Ognenova-Marinova 1975, Cat. Nos. 158169, pp. 142149;
first three types could be said to have been made using
Teposu-Marinescu, Pop 2000, Cat. Nos. 98104, pp. 8992; Ru`i}
no more than six moulds. This fact indicates that 2006, Cat. Nos. 163166, 168, pp. 137140.
moulds could have been used over a shorter period of 50 Savaria, Brigetio (Tomas 1955, Pls. VI1, VI2); Durostorum

time. The fact that there are such a large number of fig- (Elefterescu 2005, Cat. Nos. 3, 7).

158
Stefan POP-LAZI], Some Observations on Lead Figurines of the Goddess Venus (151164) STARINAR LXII/2012

half of the third century.51 Other items discovered at group of lead figurines from the Danubian basin, con-
Sirmium can be roughly dated to this period, though centrated around Durostorum, has been dated to the
the information regarding the context of the find does late second and the early third centuries.63 On the other
not allow precise dating.52 Unfortunately, we have the hand, three representations of the goddess Diana from
same situation at Gomolava, where metal processing Durostorum were discovered in the complex of baths
areas were identified during archaeological excavati- dated to the second half of the third century.64
ons,53 but there is no clear record of the archaeological The greatest number of lead figurines set in shrines
horizon in which lead objects were found. Along with (aediculae) was discovered within sacral complexes
the vessels used for casting metal and lead slag, inter- and were only rarely found in graves. Assuming the pro-
mediary products and lead mirrors serve as an argument posed dating is correct, it could be presumed that bet-
in favour of the hypothesis that, besides other lead ween the second and fourth centuries lead figurines were
objects, lead figurines of the goddess Venus were also primarily manufactured for shrines, where individuals
manufactured there.54 In a publication preceding the dedicated them as votive gifts.65 The same presumption
monograph on Gomolava, N. Tasi} stated that a lead
figurine and a Late Antique lamp with the inscription
CERNO had been found in the same Late Antique burial
51
horizon.55 As lead figurines of the goddess Venus fre- The statuette was found under the floor of a Late Antique
structure together with the coins issued by Claudius II. Parovi}-
quently appear in tombs,56 it is not impossible that the Pe{ikan 1971, 3738.
items found at Gomolava originate from Late Antique 52 The item found at Site No. 26, which is actually a Late Antique
graves and not from workshops manufacturing lead necropolis discovered in the profile of the excavation, in the hori-
objects. The lead Venus found at [a{inci was discovered zon roughly dated to the third to fourth centuries. The same could
be said of the archaeological find from the south rampart area at
at a site containing a late-third-century horizon.57It
Sirmium. The item found at the site Kej, beyond the fortified area
would appear that the lead figurines from the area of Sirmium, was purchased from the person who found it.
between Sirmium and Viminacium date from the second 53 Dautova-Ru{evljan, Brukner 1992, 6063.

half of the third century. 54 A lead figurine was assigned a field inventory number. How-

In other parts of the Roman Empire, lead figurines ever, the published literature fails to provide precise information
regarding the horizon in which it was found (Dautova-Ru{evljan,
of deities appeared considerably earlier. The oldest dated
Brukner, 1992, 65; Dautova-Ru{evljan 2006, 346).
items (late first century AD) were found in a ship sunk 55 N. Tasi} 1965, 30; this lamp has been classified as Type XXII
together with lead ingots. The figurines were part of dated to the late third and the fourth century (Dautova-Ru{evljan,
small models of temples (naiskoi), which were also made Brukner, 1992, 8283, Cat. No. 25, Pl. 7/29.
56 Apart from the examples mentioned further in the text, there
of lead.58 Lead figurines of divinities set in aediculae,
is another lead Venus figurine from Smyrna which was found in a
found during archaeological excavations in the British childs grave (Bauer 1936, 19).
Isles and in Gallia, were sculpted in a different style and 57 Horizons from the first, late third and fourth centuries have
they have been found in horizons dated to different been identified at the site. The late-third-century horizon has been
periods. At Viroconium Cornoviorum, a figurine of dated on the basis of the coins issued by emperors Tacitus and Dio-
cletian found in the horizon (Brukner 1980, 107).
Venus in an aedicula was found on a dump outside the 58 Six lead naiskoi were discovered in a sunken ship in one of
market place, together with cast and ceramic pottery the channels of the Po River. Along with the figurines of Mercury
dated no later than the second century. A figurine of and Amor, the naiskoi also contained two lead figurines of the god-
Mercury found at Segedunum, the lead doors of an dess Venus. Berti 1990, 72, 205210, Cat. Nos. 133138, Pls.
aedicula from Vindolanda and a figurine of Minerva XXIIIXXVII.
59 Segedunum, Vindolanda, Allason Jones 1984, 232; Viroco-
from Dorchester59 come from a significantly later chro-
nium Cornoviorum, Frere 1985, 285.
nological context. Two figurines of Venus in aediculae 60 Chew 1991, 8182;
originate from graves in Gallia dated to the second 61 Monnier 2009, 206, 207.

century,60 while an exquisitely sculpted statuette, dated 62 Zsidi 2000, (Matrica) 326, Cat. No. 47, (Savaria) 327, Cat.

to the second century, probably originates from a home No. 58. Such dating is mentioned in the catalogue of finds, whereas
shrine.61 The lead figurines from Savaria were dis- in the text Zsidi states that almost all of more than sixty lead votive
objects date from the period between the second half of the second
covered at an empty shrine of Mercury and are dated century and the end of the third century, p. 322.
to the late second and the first half of the third century, 63 Elefterescu 2005, 227, 238.

whereas the hoard at Matrica has been estimated to 64 Donevski 1975, 147.

date from the second to third centuries.62 The third 65 Zsidi 2000, 328.

159
Stefan POP-LAZI], Some Observations on Lead Figurines of the Goddess Venus (151164) STARINAR LXII/2012

1.Fanagorija; 2. Tanagra; 3. Myrina;


4. Dardanos; 5. Aman; 6.Barzica; 7. Devnja;
8. Negovanci; 9. Micia 10. Viminacium;
11. Sala{ No}ajski; 12. Klagenfurt

Map II. Finds of Venus statues representing the breast chain


Karta II. Nalazi Venerinih statua sa predstavom nakita na prsima

could be extended to lead figurines found in the Serbian of bronze and terracotta. However, we should not igno-
part of the Lower Pannonia and Upper Moesia. How- re the fact that on the heads of the figurines belonging
ever, the shrine for which these lead figurines could to Type III, no matter how schematically shaped, artisans
have been made has not yet been identified. rendered, as far as their abilities allowed, a diadem with
In order to open a discussion on the identification pearls on an unskilfully sculpted head and adorned the
of the shrine, it is necessary to address the issue of the breasts with jewellery. This detail, which can be traced
identity of the represented divinity. There are studies back to the late fifth century BC, had a continuity
that question the interpretation of the lead figurines as extending to the lead Venus figurines from the Danu-
representations of the goddess Venus, proposing that bian basin, which are, chronologically, the latest repre-
they should, rather, be identified as the Great Mother sentations of Venus with this attribute (Map II). In our
Goddess, i.e. as some undetermined female divinity,66 opinion, the huge number of lead figurines in this area
or as the goddess shown on the icons of the Danubian suggests that it is necessary to re-examine all variants
horsemen.67 The first hypothesis is not supported by of the representations of the goddess Venus, sculpted
iconographic evidence since the lead figurines of the in various materials, in a broader area, where they are
goddess Venus are identical to representations of the found in their hundreds. It is only when this corpus is
same goddess in other materials. Within the corpus of gathered in a single place and all variants of represen-
lead figurines it is possible to observe differences in tations of the nude female figure fully analysed, that it
artisans craftsmanship, just as they are observable in will be possible to undertake work to define the answer
representations of Venus sculpted in stone, terracotta or
bone. It is obvious that different variants can be distin-
guished among the lead pieces found in The Danubian 66 Dautova-Ru{evljan, Brukner 1992, 62; Dautova-Ru{evljan,
basin and, within a broader repertoire of representations, Vujovi} 2006, 82.
these differences are also distinguishable in those made 67 Popovi} 1992, 46.

160
Stefan POP-LAZI], Some Observations on Lead Figurines of the Goddess Venus (151164) STARINAR LXII/2012

to the question as to why these representations of the products of Dacia and Lower Moesia appear only in
goddess Venus appeared in the Danubian basin in such the Lower Danubian basin.71 A similar pattern of dis-
numbers and who were the users of so many figurines. tribution has been observed for the lead figurines of
According to current knowledge, in the territory of the the goddess Venus. It is even more interesting that the
Lower Pannonia and Upper Moesia, representations of unskilfully rendered faces of the goddess in lead icons
Venus absolutely dominate in the corpus of lead figuri- and those of the lead figurines of Venus can also be
nes. The only cult-related images that can compare in observed in Dacia and Lower Moesia. Accordingly, we
number to the lead Venus figurines are the lead icons believe that there are neither iconographic similarities
of the Danubian horsemen. Along with a series of pro- nor any cult affiliation between the lead figurines of the
blems related to the interpretation of this cult,68 there goddess Venus and lead icons of the Danubian horsemen,
is one that cannot be solved without physico-chemical and that the only link between them is the crude crafts-
analyses; namely the question of whether the same work- manship of an artisan inexperienced in glyptics.
shops manufactured both lead icons and lead figurines. Bearing in mind that the lead figurines feature the
A positive result of such analyses would not serve as goddess Venus, it is reasonable to expect that there
clear evidence of a cult relationship between Venus in must have been a sanctuary for which they could have
the Danube basin and the Danubian horsemen. In our been made. If we tried to offer an answer to the ques-
opinion, it would, rather, be an indicator of the increased tion of why lead figurines of the goddess Venus appear
use of lead in the manufacturing of cult objects in a in such large numbers, assuming that there was a shrine
certain historical period.69 The style of these two groups to the goddess Venus at Site No. 4 at Sirmium,72 the
of cult objects cannot be easily linked. There are con- question of the large number of figurines found at Vimi-
siderable differences in style among various types of nacium, which is considerably distant from Sirmium,
lead icons of the Danubian horsemen. The items
belonging to types I and II (Dalj, Divo{)70 are products
of very skilled artisans who sought to create a veracious
representation of the scene. Their skill is easily obser- 68 The hypothesis that the cult of the Danubian horsemen was
vable in the rendering of the human figure in movement, related to the cult of Dominus and Domina has been recently put
forward. Szabo 2007, 157158; Panczel 2010, 82.
in the posture of the body and the detail of the clothes of 69 On the existence of the mine Agrippi(a)na in the vicinity of
the goddess between two horsemen, etc. What should Sirmium in the late third century, cf. Du{ani} 2009, 114; on the
also be considered is the fact that the represented figures exploitation of lead in the mines on Kosmaj beginning in the mid-
are not bigger than 1718 mm. In contrast, the items second century, cf. Merkel 2007, 69; on the dating of lead icons of
the Danubian horsemen in the second half of the third and in the
belonging to Type VII (Popinci) are marked by the
fourth century, cf. Tudor 1976, 9798.
unskilful rendering comparable to that of Type I of the 70 Iskra-Jano{i} 1966, 50, 53.
lead figurines of the goddess Venus. There is a feature 71 Popovi} 1988, 115.

which links these two types of cult objects even more 72 M. Jeremi} has recently thrown light on the cult character of

strongly: which is the rendering of the goddess face. this site (Jeremi} 2006, 173176, 199); his hypothesis is supported
In both cases, the head is round, the eyes are marked by the fact that two terracotta figurines of Venus were found at the
same site (Pejovi} 1995, 44, Figs. 1 and 2; Milo{evi} 2001, 108).
by two dots, the nose by a vertical and the mouth by a Along with these, another four stone sculptures have also been dis-
horizontal line. However, these shared features are not covered (Cat. Nos. 6467); the fragment with a hand on the pubic
an indicator of the cult affinity but rather one of stylistic area (Cat. No. 66, Pl. 40/2, erroneously identified as the torso of a
similarity. If it were possible to establish a link between male figure) and a fragmented torso (Cat. No. 64, Pl. 40/3) could be
identified as parts of sculptures of the goddess Venus (Dautova-Ru-
the similar features in the rendering of the heads of the {evljan 1983, 18, Milo{evi} 2001, 108). Another piece found nine
two goddesses, it could be assumed that they were pro- years ago further supports the hypothesis on the existence of a Venus
ducts of the same workshop. The area in which lead cult in Sirmium. In the complex of the imperial palace, within a
sixth-century rubble horizon formed after the destruction of a struc-
icons can be found coincides with the area in which lead
ture built in the late third, i.e. at the very beginning of the fourth
Venus figurines can be found. The parallels between century, a marble head of the goddess Venus has been discovered in
the two groups of objects are particularly often found the immediate vicinity of a piscine. Although the sculpture dates from
in the distribution area within the Danubian basin. It a significantly earlier period, the place where it has been found shows
that it was used during the fourth century (Popovi} 2006, 153, Figs.
has long been observed that the icons of the Danubian
2ad). Judging by its hairstyle part of the hair covering the taeniae
horsemen manufactured in the territory of Pannonia on the nape the head from Sirmium shows similarities to a sculp-
and Upper Moesia appear only in this area, while the ture from Naples Aphrodita Calypigia LIMC II, 85, Cat. No. 765.

161
Stefan POP-LAZI], Some Observations on Lead Figurines of the Goddess Venus (151164) STARINAR LXII/2012

would inevitably arise. Along with lead figurines, a tations of the goddess Venus indicates a particular mode
significant number of fragments of stone sculptures of use. They could have been used solely as a visual
originating from Viminacium have been known to have symbol of the goddess, made recognisable and familiar
been found.73 Unfortunately, the information on their to those who used them through the form of a figure
discovery location is imprecise and there are no more holding her hair, covering her breasts or holding a veil
than four items for which it is possible to determine the above her head. This interpretation would support the
broader location of discovery. However, the data pro- hypothesis that such figurines were made in sanctuaries.
vided by excavation records of the investigations carried In a shrine, the pilgrim was surrounded by various repre-
out more than one hundred years ago, suggest that the sentations of the goddess Venus and the object possessed
existence of a cult site dedicated to the goddess Venus by him was iconographically identical to an object of a
could be assumed in part of the urban settlement at Vimi- higher quality which aroused a higher aesthetic experi-
nacium. As an explorer of Viminacium, Miloje Vasi} ence and was also placed in the sanctuary. The differ-
investigated about 1,600 square metres within the urban ences in the manufacturing quality of the lead figurines
settlement in 1902. On that occasion, he found well- from the Danubian basin indicate that they could have
preserved architectural structures next to one of the been made by several artisans whose skill of rendering
main communication routes within the town. Although a representation in the shallow relief of a mould varied.
very general, his report reveals that the excavations As the majority of figurines were made using six
were performed in one of the most urbanised parts of moulds, it seems that there were not many artisans who
Viminacium. Besides several construction horizons, he were manufacturing them in the Serbian part of the
has also mentioned the remains of a street along with Danubian basin. It is also possible that they often travel-
some sewage and heating systems, etc.74 Together with led between Sirmium and Viminacium and that the lack
numerous movable items, other finds from this site of any clear indication of the origin of particular types
include two stone statuettes of the goddess Venus, one may be explained by this fact. Not only that, Sirmi-
of which is fragmented, a dolphin, three fragmented umViminacium lead figurines do not appear beyond
terracotta heads of Venus, a torso once belonging to a the boundaries of the SirmiumViminacium region but
terracotta Venus statuette, as well as three fragmented it may also be observed that lead figurines originating
moulds used in the manufacturing of Venus figurines.75 from other areas do not appear there in significant
More recently, this site was located in the northern sec- numbers either. Accordingly, we believe that lead
tion of the settlement, in the immediate vicinity of the objects were used within local markets and were not a
baths, on their western side. In our opinion, such a means of trade among distant areas of the Roman
great number and iconographic diversity of represen- Empire. Such a conclusion may be coupled with the
tations of the goddess Venus suggests that a Venus cult presumption that they were cult objects manufactured
existed at this site. Unfortunately, not a single lead fig- in local workshops for local sanctuaries and that their
urine has been discovered there.76 occurrence beyond the boundaries of a local area is
Taking all previously mentioned items into account, merely an indicator of random events resulting from
we could sum up that all of the figurines are marked by the actions of an individual. All of these hypotheses
shallow relief, a round head, a schematic rendering of can be fully confirmed only by the analysis of a greater
the face and a disproportional nude female body. The number of items where the context of their find is reli-
similarly crude execution of various types of represen- ably established.

73 Tomovi} 1992, Cat. Nos. 65, 69, 71,72, 7680;


74 Vasi} 1905, 103107.
75 Vasi} 1903, 209, 216217.
76 The toponym ^air mentioned in the study authored by Ivana

Popovi} as the place where a group of lead objects were found covers
a broader area of Viminaciums castrum and settlement, Popovi}
1992, 29, note 1.

162
Stefan POP-LAZI], Some Observations on Lead Figurines of the Goddess Venus (151164) STARINAR LXII/2012

BIBLIOGRAPHY:

Badre 1999 L. Badre, Les figurines de plomb de Frenz 1988 H. G. Frenz, Figurliche Kleinfunde
Ain-Al-Djoudj, Syria 1999, vol. 76, 181196. aus Mainz und Umgebung, Mainzer Zeitschrift 83, Mainz
Bauer 1936 A. Bauer, Rimska olovna plastika sa 1988, 235256.
osobitim obzirom na materijal pohranjen u Hrvatskom Elefterescu 2005 D. Elefterescu, Statuete votive
narodnom muzeju, Vjesnik Hrvatskoga arheolo{koga din plumb de la Durostorum, Pontica XXXVIIXXXVIII,
dru{tva 17, 1936, 135. 20042005, Constanta 2005, 221238.
Benea 2007 D. Benea, Officinae plumbariae in der Gara{anin 1951 M. Gara{anin, D. Gara{anin,
Provinz Dakien, Nemeti et al. (ed.) Dacia Felix, Studia Arheolo{ka nalazi{ta Srbije, Beograd 1951.
Michaeli Barbulescu oblata, ClujNapoca 2007, 537554. Gschaid 1994 M. Gschaid, Die rmischen und
Benea 2007a D. Benea, Die Bleiverarbeitung in die gallo-rmischen Gottheiten in den Gebieten der
der Roemischen provinz Dakien, METEUGURI I Sequaner und Ambarrer, JRGZM 41, T. 2, 323470.
ARTIZANI N DACIA ROMAN, Timisoara 2008. Iskra-Jano{i} 1966 I. Iskra-Jano{i}, Rimske vo-
Benea 2008 D. Benea, Unele aspecte privind pre- tivne plo~ice od olova u Jugoslaviji: prilog tipologiji i
lucrarea plumbului in provincia Dacia. Atelierele, Dacia n rasprostiranju, Opuscula Archeologica 6, 1966, 4968.
sistemul socio-economic roman Cu privire la atelierele Jeremi} 2008 M. Jeremi}, Les temples payens de
meteugreti locale, Timisoara 2008. Sirmium, Starinar LVI (2006), 2008, 167200.
Berti 1990 F. Berti, Fortuna maris. La nave Milo{evi} 2001 P. Milo{evi}, Arheologija i istori-
romana di Comacchio, Bologna 1990. ja Sirmijuma. Novi Sad 2001.
Bolla 2004 M. Bolla, Oggetti figurati in piombo Milovanovi} 2008 B. Milovanovi}, Rimsko olovo
de eta romana nel Museo Archeologico di Verona, The u Srbiji eksploatacija, proizvodnja i upotreba, doktor-
antique Bronzes: Typology, Chronology, Authenticity, ska disertacija, Beograd 2008.
Bucharest 2004, 6977. Monnier 2009 J. Monnier, Une statuette de Vnus
Brown 1984 K. Brown, The gold Breast chain from en plomb Vallon/Sur Dompierre, Cahiers dArcho-
the early Byzantine period in the Rmisch-Germanisches logie Fribourgeoise 11, 2009, 206207.
Zentralmuseum, Mainz 1984. Najhold 2009 B. Najhold, Taurunum Romanum,
Brukner 1980 O. Brukner, Kudo{ [a{inci villa Zemun 2009.
rustica, Arheolo{ki pregled 21 (1979), 1980, 166168. Najhold 2010 B. Najhold, Taurunum skordiska,
Brun{mid 1900 J. Brun{mid, Arheolo{ke bilje{ke Zemun 2010.
iz Dalmacije i Panonije, Vjesnik Arheolo{kog muzeja u Ognenova-Marinova 1975 L. Ognenova-Marino-
Zagrebu, Vol.4 No.1, 1900, 181201. va, Stattuettes en bronze du Musee National Archeolo-
Chew 1991 H. Chew, Deux Venus en plomb d gique a Sofia, Sofia 1975.
epoque romaine, Antiquite Nationales 22/23, 1990/1991, Panczel 2010 Sz. P. Panczel, Dominus and Domina
Saint Germain en Laye, 1991, 8194. from Cristesti (Mures County), Marisia XXX 2010,
Dautova-Ru{evljan, Brukner 1992 V. Dautova- Targu Mures, 7784.
Ru{evljan, O. Brukner, Gomolava rimski period, Novi Parovi}-Pe{ikan 1971 M. Parovi}-Pe{ikan, Exca-
Sad 1992. vations of a Late Roman villa at Sirmium, Sirmium 2,
Dautova-Ru{evljan 2006 V. Dautova-Ru{evljan, Beograd 1971, 1551.
Rimski olovni predmeti iz Srema, Glasnik Srpskog Pejovi} 1995 Z. Pejovi}, Terakote iz zbirke Mu-
arheolo{kog dru{tva 22, 2006, 333353. zeja Srema. Zbornik Muzeja Srema 1, Sremska Mitrovica
Dautova-Ru{evljanin, Vujovi} 2006 V. Dautova- 1995.
Ru{evljan, M. Vujovi}, Rimska vojska u Sremu, Kata- Popovi} 1992 I. Popovi}, Radionica za izradu
log izlo`be, Novi Sad 2006. olovnih predmeta ili svetili{te kulta dunavskih konjani-
Donevski 1975 P. Donevski, Olovni relefi na ka u Viminacijumu. Viminacium 7, Po`arevac, 2956.
Artemida ot Silistra, Izvestia na narodni muzei-Varna, Popovi} 2008 I. Popovi}, Marble sculptures from
Kn XI (XXVI), Varna 1975, 146148. the Imperial Palace in Sirmium, Starinar LVI, (2006),
Duyuran 1960 R. Duyuran, Decouverte dun 2008, 153165.
Tumulus pres de lancienne Dardanos, Anatolia V (1960), Reinach 1913 S. Reinach, Repertoire de la statuaire
912, Pl. VIIXIV. Grecque et Romaines, Tome IV, Paris 1913.

163
Stefan POP-LAZI], Some Observations on Lead Figurines of the Goddess Venus (151164) STARINAR LXII/2012

Ru`i} 2006 M. Ru`i}, Kultna bronzana plastika Vin~a II M. Vasi}, Preistorijska Vin~a II, Beo-
u rimskim provincijama severnog Balkana, doktorska grad 1936.
disertacija, Beograd 2006. Vin~a III M. Vasi}, Preistorijska Vin~a III, Beo-
Szabo 2007 A. Szabo, Dominus. Una nuova tavo- grad 1936.
letta di piombo dei cd. Cavalieri Danubiani dalla Pan- Vasi} 1903 M. Vasi}, Izve{taj Srpskoj Kraljevskoj
nonia, Nemeti et al. (ed.) Dacia Felix, Studia Michaeli Akademiji o iskopavanju u Kostolcu u 1902. Godi{njak
Barbulescu oblata, ClujNapoca 2007, 150158. Srpske Kraljevske Akademije XVI 1902, Beograd 1903,
Teposu-Marinescu, Pop 2000 L. Teposu-Mari- 205219.
nescu, C. Pop, Statuete de bronz din Dacia Romana, Vasic 1905 M. Vasic, Funde in Serbien, Archae-
Bucuresti 2000. ologischer Anzeiger, 1905, 2, 102109.
Thomas 1952 E. B. Thomas, Olom Fogadalmi Wagner 1991 P. Wagner, Eine Bleistatuette der
emlekek Pannoniaban, (Monuments votivs en plomb sur Goettin Minerva aus dem Kastell Langenhein, Archeolo-
le territoire de la Pannonie), Archaeologiai Ertesito 79, gie der Wetterau. Aspekte der Forschung, Hessen 1991,
3238. 265270.
Tudor 1976 D. Tudor, Corpus Monumentorum Re- Zsidi 2000 P. Zsidi, Bleivotive aus Aquincum,
ligionis Equitum Danuvinorum (CMRED) II. The Analy- Koelner Jahrbuch 33 (2000), 313328.
sis and Interpretation of the Monuments, Leiden 1976.

Rezime: STEFAN POP-LAZI], Arheolo{ki institut, Beograd

NEKA ZAPA@AWA O OLOVNIM FIGURINAMA


BOGIWE VENERE NA PODRU^JU
IZME\U SIRMIJUMA I VIMINACIJUMA

Kqu~ne re~i. olovne figurine, Venera, Sirmijum, Viminacijum, Dowa Panonija, Gorwa Mezija.

Me|u fondom olovnih figurina bo`anstava u rimskom olovne figurine bogiwe Venere ra|ene za potrebe lokal-
carstvu (Karta I), primerci iz oblasti jugoisto~nog dela nih svetili{ta. Mogu}e je da se jedno takvo svetili{te na-
Dowe Panonije i podunavskog dela Gorwe Mezije zauzima- lazilo na lokalitetu 4 u Sremskoj Mitrovici, dok je loka-
ju posebno mesto. Izdvajaju se po specifi~nim predstava- lizacija drugog predlo`ena u severnom delu naseobinskog
ma bogiwe Venere. U okviru grupe od 40-ak {to celih, {to dela Viminaciuma. Retki stratifikovani nalazi upu}uju
fragmentovanih primeraka mogu se izdvojiti 5 tipova, ko- na mogu}nost da se proizvodwa ovih figurina u oblasti iz-
ji se razlikuju prema stavu nage `enske figure kao i pre- me|u Sirmijuma i Viminacijuma datuje u drugu polovinu
ma stilu izrade. Na osnovu toga se mo`e zakqu~iti da je u III veka. Uo~ava se ista pojava i kod proizvodwe olovnih
ovoj oblasti postojala proizvodwa lokalnih radionica ikona podunavskih kowanika. Kako se tri najzastupqenija
~iji proizvodi nisu bili zastupqeni u drugim delovima tipa figurina pojavquju na lokalitetima gde su one naj-
carstva. Tako|e je uo~eno da u istoj oblasti nema figuri- brojnije, predlo`ena je mogu}nost da se proizvodwa odvi-
na iz drugih radionica. Ovo upu}uje na pretpostavku da su jala u samim svetili{tima.

164
VOJIN NEDEQKOVI], Univerzitet u Beogradu, Filozofski fakultet,
Odeqewe za klasi~ne nauke, Beograd
SAWA STOJANOVI], Beograd

DOMINO ET FRATURI:
JEDAN NEOBJAVQEN SIRMIJSKI GRAFITO*

UDK: 904:003.071=124(497.113)"02/03" e-mail: vnedeljkovic@sezampro.rs


DOI: 10.2298/STA1262165N Primqeno: 22. februar 2012.
Originalan nau~ni rad Prihva}eno: 21. jun 2012.

Apstrakt. Izdaje se latinski natpis sa jedne rimske opeke iz Muzeja Srema.


Po tipu, natpis spada me|u pozdravne grafite; lica koja u wemu figuri{u verovatno su bila anga`ovana
na gra|evinskim poslovima.

Kqu~ne re~i. Sirmijum, rimska neimarska delatnost, slu`ba u rimskoj vojsci/upravi, latinska epigrafika,
grafiti, vulgarni latinski.

U
Muzeju Srema u Sremskoj Mitrovici ~uva
se, pod inventarskim brojem A/5342, rim- * Pisci ovog priloga imaju prijatnu du`nost da zahvale
ska opeka s urezanim natpisom. O wenoj dr Ivani Popovi}, direktoru projekta Sirmijum, i Zoranu
provenijenciji jedini podatak nam je ostavio Bo{kovi}u iz Arheolo{kog instituta u Beogradu, ~ija nam je
Petar Petrovi} u jednoj svojoj bele{ci: Opeka je dokumentacija stajala na raspolagawu; Jasmini Davidovi} i
Miroslavu Jesreti}u iz Muzeja Srema u Sremskoj Mitrovici,
na|ena u zidu peristila na lok. 4, tu ugra|ena pri- gde nam je materijal ustupqen za publikovawe; Radovanki Gulan
likom konzervacije. Ta~no mesto nalaza nije po- iz Pokrajinskog zavoda za za{titu spomenika u Novom Sadu,
znato.1 Konzervacija peristila luksuzne vile ~ijom smo se dokumentacijom tako|e poslu`ili. Doc. dr Svetla-
(lok. 4)2 izvr{ena je 1963. godine; me|utim, iz do- na Loma sa Filozofskog fakulteta u Beogradu prihvatila je
(opet!) na{u molbu da pro~ita {to pi{emo i saop{ti nam svoje
kumentacije koja se ~uva u Pokrajinskom zavodu za opaske, popravke i dopune, koje su (opet!) bile dragocene.
za{titu spomenika u Novom Sadu ne vidi se oda- 1 Fotografije i bele{ke o ovoj opeci nalaze se me|u do-

kle je bio donet arheolo{ki materijal koji je ko- kumentima koji su po smrti P. Petrovi}a predati Centru za
ri{}en pri konzervaciji.3 anti~ku epigrafiku i numizmatiku Fanula Papazoglu pri
Filozofskom fakultetu u Beogradu. Iz fotodokumentacije
Posredi je crveno pe~ena opeka, ~itava, veli- Arheolo{kog instituta saznali smo da su fotografije iz Pe-
~ine 40,5 27,5 6 cm. Na woj se ~ita latinski trovi}eve dokumentacije nastale 1971. godine trudom B. Luki}a
natpis u dva nejednaka reda po {irini, plitko (sign. fot. 356/1971). Fotografiju koju ovde donosimo (Sl. 1)
urezan pre pe~ewa. Slova su neujedna~ene visine na~inili smo sami krajem 2007.
2 Lokalitet 4 se nalazi u dvori{tu osnovne {kole Bo{ko
(12,5 cm), s kurzivnom tendencijom (upor. naro-
Palkovqevi}, u Zmaj Jovinoj 19 (upor. Milo{evi} 1994, 13).
~ito pismena e, m, n, s, t). Na povr{ini ispod nat- Tu je na|en deo luksuzne vile iz sredine IV veka: terme, rasko{-
pisa, udubqewe i isprekidane dijagonalne brazde nim peristilom povezane s prostorijama na istoku; v. Parovi}-
(ove su mogle nastati i pri konzervaciji, kad je Pe{ikan 1962, 129130. Vila je bila sagra|ena na osnovama zda-
wa s kraja III i po~etka IV veka, stradalih u velikom po`aru.
opeka uzidana u zid peristila); vidqiva su mesti-
Na lokalitetu su na|eni i stariji slojevi, koji nisu od zna~a-
mi~na o{te}ewa i po ivicama opeke, a gorwi uglo- ja za na{ predmet. O iskopavawima na prostoru lokaliteta 4:
vi su odbijeni. Parovi}-Pe{ikan 1962, 123130; ista 1968, 135139; ista 1969,

165
NEDEQKOVI], STOJANOVI], Domino et fraturi: jedan neobjavqen sirmijski grafito (165172) STARINAR LXII/2012

Tekst natpisa glasi: Egipat:


Domino et fraturi Maxime salutem CEL 73 (papirus, slu`beno pismo, II vek), 1
Valerus Januarius [domi]n fratri s(alutem) d(icit)
CEL 177 (papirus, pismo preporuke, II vek), 34
U doslovnom prevodu: praese(n)s te, domine frater, rogaveram
Gospodinu i bratu, Maksime, pozdrav.
Valerije Januarije. Dura-Europos:
CEL 187 (papirus, prepis slu`benog pisma, 211.
Posredi je, dakle, pozdravni grafito, od vrste po Hr.), 2.4 peto, domine frater
koja se sporadi~no nalazi na rimskim opekama iz CEL 191 (papirus, prepisi slu`benih pisama,
na{ih (i, razume se, drugih) krajeva. Ispisani 216. po Hr.), 38b.36 opt[o] te, domine
{iqatim predmetom, mahom po glini pre pe~ewa, frater, felicis[si]mum bene vale[r]e; sli~no i
ali nekad i po pe~enoj cigli, oni svedo~e o komu- 39.919.
nikaciji izme|u radnika zaposlenih na istom gra-
|evinskom poduhvatu. Na natpisanoj opeci iz Go- Ta se formula mahom javqa me|u licima sli~-
lubiwa, CIL 3.8277,3, ~ita se [fac] laterclus [CC (?) nog ranga, neretko pri kraju pisma, u sklopu po-
F]uriane, [m]ale dor[mie]s si nun feceris:4 primalac zdrava ispisanog ne pisarevom ve} li~no po{iqa-
tog podsticajnog pozdrava bio je o~evidno ciglar, o~evom rukom. Zbog fragmentarnosti tekstova, u
finctor, kakav se izri~ito spomiwe u ILJug 1055 konkretnim slu~ajevima ~esto nije mogu}e utvrdi-
(Cibalae) finctor vive dominis tuis. Druge poruke te ti o kojem se rangu radi, ali obi~no se ra~una s li-
vrste putovale su sa ciglane na gra|evinu, kako bi cima ne{to vi{eg polo`aja,9 jer je me|u prostiji-
pro{le kroz ruke zidarima: IMS 4.115 (opeka iz
leskova~kog kraja, izgubqena) Deus aduvet vobis et
nobis; ovamo pripadaju verovatno i grafiti na
190194; ista 1971, 1549. Vili pripadaju i objekti na sever-
opekama s Gradi{ta kod Prvoneka.5 Slu~aj za sebe,
nom delu lok. 35, iskopavanog od 1963. godine (Parovi}-Pe{ikan
ali opet u kontekstu rada na gra|evini, predsta- 1973, 139).
vqa epistolarna opeka sa Manastiri{ta kod Ve- 3 Ne{to kasnije, Parovi}-Pe{ikan 1969, 194, saop{tava

likoga Kr~imira.6 da se konzervacija i eventualna rekonstrukcija peristila


Na prvi pogled, ovda{we domino et fraturi ne- planirana za tu godinu nije mogla izvr{iti, ve} je peristil,
sa prostorijama u kojima su prona|eni mozaici, zatrpan.
pogre{ivo podse}a na manir obra}awa gospodinu 4 Vidi: Petrovi} 1975, 36.
sabratu, najboqe poznat iz prepiske me|u duhovnim 5 Nedeqkovi} 2006.
ali i svetovnim licima u sredwem veku. Preseda- 6 Nedeljkovi} 2004.

ni su, me|utim, poznoanti~ki: domino fratri Sisebuto 7 Aug. Ep. 25 i 30. Mi ovde ne uzimamo u obzir bezbroj-

Hysidorus tako Isidor od Seviqe otpo~iwe pro- ne slu~ajeve gde se ta dva apelativa nalaze rastavqena unutar
log svoje Astronomije; dvesta godina ranije, Pav- iste formule, jer za na{u potrebu to ne ukazuje ni na {ta. ^i-
ni se, naime, da logika kompozitnog apelativa domine frater,
lin od Nole {aqe pisma Avgustinu pod zaglavqem gospodaru brate, nije bez ostatka svodiva na dva odelita re-
domino fratri unanimo et venerabili Augustino.7 [ta- zona jedan po kojem se uva`ewe izra`ava sa domine i drugi po
vi{e, i za na{u ovda{wu svrhu zna~ajno, taj hri{- kojem se zajedni{tvo izra`ava sa frater. Povrh tih efekata,
}anski epistolarni uzus zapravo se nastavqa na spoju domine frater inherentan je oksimoron, jer te dve re~i,
uobi~ajene ponaosob u formulama qubaznog obra}awa, ovako
jednu stariju, svetovnu upotrebu, koju znamo iz do- primaknute jedna drugoj prizivaju svoja nepomirqiva primarna
kumentarnih izvora. Obra}awe re~ima domine frater zna~ewa, pa nastaje izvesna napetost. [ta je svrha te napetosti,
dobro je potvr|eno u prepisci izme|u lica u dr`av- drugo je pitawe, ali u weno postojawe lako je uveriti se ~im
uporedimo malopre|a{we Pavlinovo zaglavqe domino fratri
noj i vojnoj slu`bi {irom Carstva:
unanimo et venerabili sa, na primer, Aug. Ep. 179 domino beatis-
simo ac merito venerabili fratri et coepiscopo Joanni Augustinus in
Vindolanda: Domino salutem.
8 Odavde mora da izostane CEL 89.II.67 = TVI 21 optamus,
TVII 247 (= CEL 93), 23 vale, mi domine frater
karissime frater domine, [te] bene valere [et in ho]nore esse {to je druk~ije
izdato u TVII 248; ali i daqe je koristan Kuguzijev komentar
TVII 260.46 opto, domin[e] frater, felicissimus ad loc. (izme|u ostalog, sa literarnim paralelama).
bene valeas 9 Na pole|ini pisma CEL 177 po{iqalac se i eksplicit-

TVII 295 (= CEL 98), 910 vale, domine frater8 no predstavqa kao primao~ev coll(ega).

166
NEDEQKOVI], STOJANOVI], Domino et fraturi: jedan neobjavqen sirmijski grafito (165172) STARINAR LXII/2012

Sl. 1. Rimska opeka, inv. br. A/5342, Muzej Srema, Sremska Mitrovica
Fig. 1. Roman brick from Sirmium, Inv. A/5342, Museum of Srem, Sremska Mitrovica

ma bilo normalno i dovoqno oslovqavati se samo nutirices, ancillae > anicillae; ili je ona motivisana
sa frater. U isto doba, II vek, padaju i literarne po- paretimolo{ki, npr. libra > libera,12 Aprilis > Aperi-
tvrde ove formule, koju imamo u anonimnom (pseu- lis. Najbli`a nama poznata paralela ovda{wem slu-
do-Higinovom) spisu De munitionibus castrorum, 45 ~aju fratri > fraturi bila bi anaptiksa patri > patiri,
in quantum potui, domine frater, pro tirocinio meo in potvr|ena u jednom gorwopanonskom natpisu (CIL
brevi omnes auctores sum persecutus, kao i pri po~et- 3.10599), i naro~ito CIL 3.9735 (Delminium) frateres
ku jednog Frontonovog pisma, gde mo`e biti da je frater(i) fecerunt; no sve u svemu stvar je retka i uto-
ona upotrebqena s lakim humorom: Ad am. 1.27 tibi, liko vredna pa`we.13
domine frater, commodius evenit, qui eqs.10 Neobi~an je i oblik Maxime umesto o~ekivanog
O neobi~noj grafiji fraturi treba da ka`emo Maximo. U ~itawe ni ovde nema sumwe, kao ni u to da
nekoliko re~i. ^itawe je pouzdano; uostalom, i iz Maxime predstavqa formu vokativa, svejedno {to
konteksta je jasno da fraturi stoji za fratri. Obja{we-
we se nema gde tra`iti osim u oblasti fonologije
posredi je, dakle, slu~aj anaptikse. Konsonantsku
grupu okluziv + likvida ovde razbija vokal u (-tr- > 10 Vredne opaske o formuli domine frater i wenom gr~kom
-tur-) i nastaje nova nenagla{ena penultima: po to- pandanu (krie delf) daje Dickey 2010, 334. Po woj, ta for-
me bi ovaj slu~aj li~io najvi{e na periclum > peri- mula je sastavni deo jednog {ireg sistema konvencionalnog
culum, saeclum > saeculum, anaptiksu poznatu od iskazivawa privr`enosti i po{tovawa kako na gr~koj, tako i
na latinskoj strani; uhodan u carsko doba, taj sistem je potra-
vrlo ranih vremena i primqenu i u kwi`evnom je- jao otprilike do IV veka.
ziku, a vulgarno pro{irenu na sporadi~ne slu~aje- 11 Za {irok istorijski pregled anaptikse u latinskom v.

ve kao templum > tempulum (DVulg 1554). Ali ucelo Stotz, HLSMA 3, VII 82, odakle su i neki od primera koje }emo
gledano, vulgarnolatinska anaptiksa nije takva.11 navesti.
12 Na tu paretimolo{ku vezu eksplicitno upu}uje Isid.
Ona vrlo zna~ajnim delom poga|a re~i egzoti~nog
Orig. 16.25.20.
porekla, a vokali koji joj slu`e gotovo nikad nisu 13 Uop{te, epigrafske potvrde anaptikse relativno su
zadweg reda; weni obrasci su cycnus > cicinus, bap- mr{ave: Omeltchenko 1977, 460463. Za nekoliko upe~atqivih
tismus > baptissimus, i na drugoj strani nutrices > instanci v. Slotty 1960, br. 5360.

167
NEDEQKOVI], STOJANOVI], Domino et fraturi: jedan neobjavqen sirmijski grafito (165172) STARINAR LXII/2012

bismo u datom kontekstu o~ekivali dativ.14 U Pom- Na{i Valerije Januarije i Maksim mogli su
pejima, gde pozdravnih grafita ima u ve}em broju, ne i morali slu`bovati u vojsci, rade}i u ci-
obilato je zastupqena formula s dva imena, u nomi- glarskoj proizvodwi ili na drugim gra|evinskim
nativu odn. dativu, pra}ena imenicom salutem, taj poslovima. Kognomenta Januarius i Maximus,22
i taj tome i tome (upu}uje) pozdrav, npr. DPomp omiqena kod domoroda~kog stanovni{tva, mo`da
518 Coelius Cynisculo salutem, 522 Aephebus Successo ukazuju na indigeno poreklo ove dvojice,23 dok bi
patri suo salut(em), 523 fullo Cresces coponi sal(utem). izostanak Maksimovog gentilnog imena bio znak
Opet, ~esti su i pozdravi formulisani s imenom wihovog relativno familijarnog odnosa.24
primaoca u vokativu, iza ~ega sledi vale, npr. 556 Po natpisu sude}i, opeka bi mogla poticati s
Polubi va(le), 555 Optate va(le), Cajetanus scripsit. Oba kraja III ili iz prve polovine IV veka.25 To je i do-
tipa zdru`eno dolaze u 546 Receptus Actio sa[l(utem)], ba kad se u Sirmiju bele`i najve}a gra|evinska
Acti va(le). S tim u vidu, ~ini se da pozdrav koji aktivnost.26 U tom periodu grad je skoro u celini
imamo na sirmijskoj opeci predstavqa kontamina-
ciju dvaju obi~nih obrta: ono {to se isprva htelo
re}i, domino fratri Maximo salutem, ukr{teno je s
alternativnom formom pozdrava, Maxime vale.15 14 O neobi~nim pojavama vokativa: Lfstedt 1942, 102105;
[to se pi{~evog imena ti~e, oblik u kojem ono ali tu nema ni~ega sli~nog ovome. Potpunosti radi, podse}a-
dolazi, Valerus Januarius, na upadqiv na~in sadr`i mo na neobi~nu pojavu u tekstu triju votivnih natpisa s Ko-
smaja, IMS 1.90, 1.91 i 1.94, gde forma Juppiter stoji po prili-
prvo grafiju -erus za -erius, a odmah zatim korektno ci vokativno, ali opet na na~in sasvim nesrodan ovom na{em
-arius. Toj nedoslednosti se ne treba odvi{e ~udi- slu~aju.
ti: u jednom od pompejskih pozdrava zati~emo jed- 15 Mo`e biti da ta alternativna forma ima udela i u to-

no te isto li~no ime ispisano najpre ispravno a me {to se ime pisca nalazi na svr{etku a ne na po~etku teksta
(Valerius Januarius domino et fratri Maximo salutem); ovo se, do-
potom pogre{no DPomp 525 Aulus Olo suo salutem
du{e, moglo dogoditi i bez tog motiva.
(obojica se zovu Aulus). [to se ti~e same grafije 16 Vidi, npr., Vnnen 1982, 79; Omeltchenko 1977,
Valerus, vulgarnolatinski yod, kao {to se zna, u iz- 220223.
vesnim fonetskim kontekstima bio je sklon nesta- 17 Dioklecijan je svoj gentilicij na ime bratstva dao

jawu.16 Slu~ajevi kao DVulg 1298 Aurelia Januara (tj. Maksimijanu kada ga je 286. postavio za svog savladara; godi-
ne 293, sa uspostavqewem tetrarhije, taj nomen su dobili i
Januaria), ili u~estala grafija memore za memoriae
Cezari. Kraj dinastiji Valerija u~inio je Konstantin 324.
u hri{}anskim spomenicima, pokazuju da se to spo- godine. Pre 286, Valerius je bio najrasprostraweniji necarski
radi~no doga|alo i u redukovanoj penultimi iza gentilicij: v. Mcsy 1959, 160; Alfoldy 1969, 131; upor. OPEL,
glasa r, s tim {to ne mo`emo znati da li je yod u kw. 4, 143146.
18 Salway 1994, 138.
tim slu~ajevima nestajao bez traga ([rj] > [r], upor. 19 PLRE I, 1053 i daqe (Fasti); Keenan 1973, 46, nap. 56.
ital. -aro < -arium) ili pak grafija r stoji za pala- 20 Od namesnika provincija do gradskih magistrata. Vidi:
talizovano [r] poteklo od [rj]. PLRE I, 940 i daqe (s. Valerius); Keenan 1973, 4445.
Valerius je od vremena tetrarhije bilo gentilno 21 Mcsy 1965, 218.

ime carske dinastije koju je osnovao Dioklecijan.17 22 Kajanto 1965, 29 i 30, ubraja ih u osamnaest naj~e{}ih

To je bila i statusna oznaka gra|ana koji su krajem latinskih kognomenata, a za wihovu rasprostrawenost v. OPEL,
III i u prvoj ~etvrtini IV veka pri ulasku u dr`avnu kw. 2, 189, i kw. 3, 70.
23 Za Panoniju v. Mcsy 1959, 176 (Januarius) i 181 (Maxi-
ili vojnu slu`bu mogli od cara dobiti i novi no- mus); Barkczi 1964, 314 (Januarius) i 318 (Maximus). Stanovni-
men.18 Sude}i prema jo{ prisutnim Aurelijima, {tvu keltskog porekla ime Januarius moglo se ~initi naro~i-
carski nomen nije dodeqivan svima.19 U civilnoj to prikladnim zbog sazvu~ja sa keltskim imenima na Jan-:
slu`bi, gentilno ime Valerius nosila su lica raz- Mcsy 1959, 176; Alfldy 1977, 258; o keltskim imenima na
Jan-: Holder 1904, 9.
li~itog ranga.20 Vojnici Valeriji, reklo bi se, 24 Potkraj III i po~etkom IV veka bele`i se sve ~e{}e iz-
nisu bili obi~ni milites: upor., npr., CIL 3.14333 ostavqawe nomena u privatnim natpisima, pre svega u uzusu
Valerius Saturninus mil(es) leg(ionis) M(artensium?) ni`ih slojeva; aristokratija je i tada i dugo potom ~uvala
d(ra)conarius,21 ILS 2776 Valerius Valens protector, svoje gentilicije: Kajanto 1977, 426428.
25 To jest, pisac grafita je mogao na slu`bu stupiti kra-
2779 Valerius Vincentius actuarius protectorum, 2792
jem III ili tokom prve ~etvrtine IV veka, pa slu`bovati jo{ go-
Valerius Juventinus exarcus, 3458 Valerius Valerianus dinama potom.
miles cum insisterem ad capitella columnarum ad ter- 26 Izgradwu tetrarhijskog Sirmija verovatno je zapo~eo

mas Licinan(a)s. sam Dioklecijan, kako dokazuje Du{ani} 2003/2010.

168
NEDEQKOVI], STOJANOVI], Domino et fraturi: jedan neobjavqen sirmijski grafito (165172) STARINAR LXII/2012

obnovqen. Sagra|ena je nova javna `itnica, veli- uzidana u bilo koju od svih tih gra|evina; no, u ne-
ke terme,27 palata, hipodrom, a nedaleko od wih i dostatku podataka i indicija, logi~no je pretposta-
luksuzna vila koju smo spomenuli. Po~etkom IV viti da materijal za svojevremenu konzervaciju
veka grad se pro{irio prema jugu, o ~emu svedo~i peristila na lokalitetu 4 nije bez potrebe dopre-
bedem iz tog perioda. U centru grada bila su javna man izdaleka:29 opeka bi, po tome, verovatno poti-
zdawa, dok je ju`ni deo bio rezidencijalna ~etvrt cala s lica mesta,30 ako ne sa obli`weg i po gra|e-
sa palatom.28 Na{a opeka je, naravno, mogla biti vinskim fazama bliskog lokaliteta 1a (palata).31

27 Na osnovu posvete Herkulu ILS 3458, te terme (lok. 29)

slove kao Licinijeve.


28 Popovi} 1993/2003, 4647; isti 1975/2003, 154.
29 Ovo na{e domi{qawe treba primiti vrlo oprezno. Ma-

terijal za konzervaciju na lok. 4 u jednoj ranijoj prilici, 1960.


godine, dono{en je, po svedo~ewu arh. D. St. Pavlovi}a, sa ~i-
tavog podru~ja Sirmija: Pavlovi} 1961, 84.
30 U tehni~kom opisu lokaliteta, na prvoj stranici Ela-

borata za konzervaciju i prezentaciju lokaliteta 4 Sirmi-


jum, Sremska Mitrovica Zmaj Jovina ul., postoji podatak da
se u ju`noj polovini termi istra`enih u dvori{tu u Zmaj Jovi-
noj vide tragovi podnih plo~a veli~ine 2830 cm 4044 cm
6 cm (B1/B/V, br. inv. 808, str. 1 [5]). Te mere pribli`no
odgovaraju dimenzijama na{e opeke. Tako|e, D. St. Pavlovi}
je u Elaboratu za nastavak konzervacije i delimi~nu prezen-
taciju arhitektonskih ostataka lokaliteta 4 (1963) zapisao
da su u Muzej odlagani fragmenti opeke i mermera sa lok. 4
upotrebqivi za konzervacijske poslove (str. 3 [12] i 13 [22]).
31 U dnevniku arhitektonsko-konzervatorskih radova na

lok. 4 iz 1960. godine, koji je kao projektant i nadzornik vodio


D. St. Pavlovi}, stoji i ovo: Sreda i ~etvrtak, 16. i 17. novem-
bar 1960 Odabrana je i doterana rimska cigla plo~e sa lo-
kaliteta 1a (B1/B/V, br. inv. 653, str. 13 [15]).

169
NEDEQKOVI], STOJANOVI], Domino et fraturi: jedan neobjavqen sirmijski grafito (165172) STARINAR LXII/2012

BIBLIOGRAFIJA:

Alfldy 1969 G. Alfldy, Die Personennamen in ber einige Grundfragen der lateinischen Nominal-
der rmischen Provinz Dalmatia, Heidelberg. syntax, 2Lund.
Alfldy 1977 G. Alfldy, Die Personennamen Milo{evi} 1994 P. Milo{evi}, Topografi-
in der rmischen Provinz Noricum, u: Pflaum, Duval ja Sirmijuma, Novi Sad 1994.
1977, 249265. Mcsy 1959 A.Mcsy, Die Bevlkerung von
Barkczi 1964 L. Barkczi, The Population of Pannonien bis zu den Markomannenkriegen, Budapest.
Pannonia from Marcus Aurelius to Diocletian, Acta Mcsy 1965 A. Mcsy, Zur Bevlkerung in der
Archaeologica Academiae Scientiarum Hungaricae 16, Sptantike, u: G. Alfldy, Bevlkerung und Gesell-
257356. schaft der rmischen Provinz Dalmatien, Budapest,
CEL P. Cugusi, Corpus epistularum Latinarum 212226.
papyris tabulis ostracis servatarum, 3 kwige, Firenze Nedeljkovi} 2004 V. Nedeljkovi}, Un nouveau
19922002. document palochrtien de la DacieMditerrane, Aevum
CIL Corpus inscriptionum Latinarum, Berolini 78/1, 147158.
1863 . Nedeqkovi} 2006 V. Nedeqkovi}, Rimski
Dickey 2010 E. Dickey, Forms of Address and grafiti sa Gradi{ta kod Prvoneka, Starinar 56,
Markers of Status, u zborniku E. J. Bakker (ur.), A 377380.
Companion to the Ancient Greek Language, Chichester, Omeltchenko 1977 S. W. Omeltchenko, A Quan-
Malden (MA), 327337. titative and Comparative Study of the Vocalism of the
DPomp E. Diehl, Pompeianische Wandinschriften Latin Inscriptions of North Africa, Britain, Dalmatia,
und Verwandtes, Bonn 1910. and the Balkans, Chapel Hill.
Du{ani} 2003/2010 S. Du{ani}, Diocletians Visits OPEL B. Lorinz (ur.), Onomasticon provinciarum
to Quarries and Mines in the Danubian Provinces, u: Europae Latinarum, kw. 24, Wien 19992002.
Die Archologie und Geschichte der Region des Eisernen Parovi}-Pe{ikan 1962 M. Parovi}-Pe{ikan, Sir-
Tores zwischen 275602 n. Ch. Kolloquium in Drobeta mium, lokalitet 4 [kola B. Palkovljevi}-Pinki (Sirmium
Turnu Severin 2.5. November 2001, Bucureti 2003, u 1962. godini), Arheolo{ki pregled 4, 123131.
921. Pre{tampano u: S. Du{ani}, Selected Essays in Parovi}-Pe{ikan 1968 M. Parovi}-Pe{ikan, Sir-
Roman History and Epigraphy, Belgrade 2010, 712723. mium, Villa urbana, lokalitet 4, Arheolo{ki pregled 10,
DVulg E. Diehl, Vulgrlateinische Inschriften, Beograd, 135139.
Bonn 1910. Parovi}-Pe{ikan 1969 M. Parovi}-Pe{ikan, Lo-
Holder 1904 A. Holder, Alt-Celtischer Sprach- kalitet 4 kasnorimska vila (Sremska Mitrovica Sir-
schatz, kw. 2, Leipzig. mium), Arheolo{ki pregled 11, Beograd, 190194.
ILS H. Dessau, Inscriptiones Latinae selectae, 3 Parovi}-Pe{ikan 1971 M. Parovi}-Pe{ikan, Ex-
kwige, Berlin 18921916. cavations of a Late Roman Villa at Sirmium (Part I), u:
ILJug A. i J. [a{el, Inscriptiones Latinae quae in V. Popovi}, E. L. Ochsenschlager (ur.), Sirmium II, Beo-
Iugoslavia repertae et editae sunt, 3 kwige, Ljublja- grad 1971, 1549.
na 19631986. Parovi}-Pe{ikan 1973 M. M. Parovi}-Pe{ikan,
IMS Inscriptions de la Msie Suprieure, Beograd Excavations of a Late Roman Villa at Sirmium (Part II),
1976 . u: V. Popovi}, E. L. Ochsenschlager (ur.), Sirmium III,
Kajanto 1965 I. Kajanto, The Latin Cognomina, Beograd 1973, 139.
Helsinki. Pavlovi} 1961 D. Pavlovi}, Konzervatorska re-
Kajanto 1977 I. Kajanto, The Emergence of the {enja u Sirmijumu, u: M. Grbi} (ur.), Limes u Jugosla-
Late Single Name System, u: Pflaum, Duval 1977, viji I. Zbornik radova sa simposiuma o Limesu 1960,
421430. Beograd, 8385.
Keenan 1973 J. G. Keenan, The Names Flavius Petrovi} 1975 P. Petrovi}, Paleografija
and Aurelius as Status Designations in Later Roman rimskih natpisa u Gorwoj Meziji, Beograd.
Egypt, ZPE 11, 3363. Pflaum, Duval 1977 H.-G. Pflaum, N. Duval
Lfstedt 1942 E. Lfstedt, Syntactica. Studien (ur.), LOnomastique latine: Paris, 1315 octobre 1975,
und Beitrge zur historischen Syntax des Lateins, kw. 1: Paris.

170
NEDEQKOVI], STOJANOVI], Domino et fraturi: jedan neobjavqen sirmijski grafito (165172) STARINAR LXII/2012

PLRE I A. H. M. Jones, J. R. Martindale, J. Mor- Salway 1994 B. Salway, Whats in a Name? A


ris, The Prosopography of the Later Roman Empire, kw. Survey of Roman Onomastic Practice from c. 700 B.C.
1, Cambridge 1971. to A.D. 700, JRS 84, 124145.
Popovi} 1975/2003 V. Popovi}, Glavne etape Slotty 1960 F. Slotty, Vulgrlateinisches bungs-
urbanog razvoja Sirmijuma, u zborniku Branka Viki}- buch, 2Berlin.
Belan~i} (ur.), Anti~ki gradovi i naselja u ju`noj Panoni- Stotz, HLSMA P. Stotz, Handbuch zur lateini-
ji i grani~nim podru~jima: Vara`din 1975, Beograd 1977, schen Sprache des Mittelalters (Handbuch der Altertum-
111122. Pre{tampano u: Popovi} 2003, 147156. swissenschaft II.5), 5 kwiga, Mnchen 19962004.
Popovi} 1993/2003 V. Popovi}, Sirmium TVI A. Bowman, D. Thomas, Vindolanda: The
grad careva i mu~enika, u zborniku D. Srejovi} Latin Writing Tablets, London 1983.
(ur.), Rimski carski gradovi i palate u Srbiji, Be- TVII A. Bowman, D. Thomas, The Vindolanda
ograd 1993, 1527. Pre{tampano u: Popovi} 2003, Writing Tablets (Tabulae Vindolandenses II), London
4150. 1994.
Popovi} 2003 V. Popovi}, Sirmium grad care- Vnnen 1982 V. Vnnen, Introduzione al
va i mu~enika (sabrani radovi o arheologiji i isto- latino volgare, 3Bologna (original: Introduction au latin
riji Sirmijuma), Sremska Mitrovica. vulgaire, 31981).

171
NEDEQKOVI], STOJANOVI], Domino et fraturi: jedan neobjavqen sirmijski grafito (165172) STARINAR LXII/2012

Summary: VOJIN NEDELJKOVI], University of Belgrade, Faculty of Philosophy, Department of Classics, Belgrade
SANJA STOJANOVI], Belgrade

DOMINO ET FRATURI.
AN UNPUBLISHED GRAFFITO FROM ROMAN SIRMIUM

Key words. Sirmium, Roman construction-business, Service in the Roman army/administration,


Latin epigraphy, Graffiti, Vulgar Latin.

In the Regional Museum of Srem in Sremska Mitrovica, Serbia, recipient of a message as dominus frater is a well-attested style
a Roman brick is preserved, containing an inscription (Inv. in Roman military circles, especially between equals, from the
A/5342). Originating from any of the numerous Roman struc- early 2nd century on. Phonologically, the spelling fraturi (for
tures in the capital city of Sirmium most probably from Site 4 fratri) is a remarkable instance of anaptyxis, vulgar, but only
(Villa) or Site 1a (Palace) the brick exhibits a graffito in Latin rarely occurring in the sources. The pendent vocative Maxime is
which reads as follows: probably due to the actual formula being a contamination of two
Domino et fraturi Maxime salutem known types, Maximo salutem and Maxime vale. Judging by their
Valerus Januarius cognomenta, both men, Januarius and Maximus, may very well
Written before the brick was baked, this greeting obviously have been of indigenous origin, whereas the nomen Valerius
went from somebody present at the brick plant to another person may have been acquired through service in the imperial army or
engaged locally in the business of construction. To address the administration under the Tetrarchs.

172
JELENA Q. CVIJETI]
Univerzitet u Beogradu, Filozofski fakultet, Odeqewe za arheologiju, Beograd

NOVI EPIGRAFSKI SPOMENIK IZ OTILOVI]A


KOD PQEVAQA

UDK: 902.2(497.16)"2009" ; 904:726.825"652"(497.16) e-mail: jcvijeti@f.bg.ac.rs


DOI: 10.2298/STA1262173C Primqeno: 15. februar 2012.
Prilog Prihva}eno: 21. jun 2012.

Apstrakt. Tokom za{titnih arheolo{kih istra`ivawa 2009. godine, u Otilovi}ima, nadomak Pqevaqa,
istra`ena je ranohri{}anska crkva sa grobnicom. Tom prilikom otkrivena su tri rimska nadgrobna spomenika.
U apsidi su na|ena dva tzv. piramidalna cipusa, dok je nadgrobna stela bila sekundarno iskori{}ena kao stepenik
kojim se iz pronaosa stupalo u naos crkve. Stela pripada ra{irenom tipu nadgrobnih spomenika u isto~nom
i jugoisto~nom dijelu provincije Dalmacije, a wena kompoziciona shema predstavqa za sada usamqen primjer
u pqevaqskom kraju. Iz natpisa doznajemo imena ~etvoro pokojnika (Pletor, Maximina, Victorinus i Statia Fuscina).
Ime Pletor, koje je ovom prilikom prvi put posvjedo~eno u epigrafskom materijalu iz pqevaqskog kraja,
sada mo`emo dodati velikoj grupi ilirskih imena koja su potvr|ena u Municipijumu S. Nekropola u Otilovi}ima
upu}uje na postojawe vile rustike ili naseqa koje se razvilo uz putnu komunikaciju koja je povezivala naseqa
u Kominima i Kolovratu, a koja je posvjedo~ena miqokazom iz Otilovi}a jo{ krajem XIX vijeka.

Kqu~ne re~i. Dalmacija, Otilovi}i, rimski nadgrobni spomenici, ime Pletor.

P
rilikom nivelacije terena, u selu Oti- menzija 0,20 m x 0,37 m x 0,25 m, za u~vr{}ivawe
lovi}ima, nadomak Pqevaqa, slu~ajno su stele u odgovaraju}e postoqe.
otkriveni ostaci ranohri{}anske crkve, Povr{ina spomenika ra{~lawena je na sokl,
a unutar we tri rimska nadgrobna spomenika, i to natpisno poqe i reqefno poqe, izme|u kojih se
dva tzv. piramidalna cipusa (sl. 1) kao i jedna nalazi dekorativna traka {irine 0,10 m.
nadgrobna stela (sl. 2).1
Stela od sivkastog kre~waka lo{eg kvaliteta,
dimenzija 1,73 m x 0,60 m x 0,32 m, sekundarno je
iskori{}ena kao stepenik kojim se iz pronaosa 1 Tokom za{titnih arheolo{kih istra`ivawa koja su po-

stupalo u naos crkve. Na gorwoj ravnoj povr{ini tom uslijedila, u potpunosti je istra`ena crkva kao i mawa
zasvo|ena grobnica uz wen sjeverni zid. Istra`ivawima koja
stele uo~eno je kru`no udubqewe, u kome je najvje-
su sprovedena u okviru saradwe Zavi~ajnog muzeja u Pqevqi-
rovatnije stajala {i{arka, koja nedostaje (sl. 3). ma i Filozofskog fakulteta u Beogradu rukovodila je dr Mi-
Na dowoj strani vidqiv je ~etvrtasti usadnik, di- ra Ru`i}.

* ^lanak predstavqa rezultat rada na projektu Romanizacija, urbanizacija i transformacija urbanih centara civilnog, vojnog i
rezidencijalnog karaktera u rimskim provincijama na tlu Srbije (br. 177007), koji finansira Ministarstvo prosvete, nauke i
tehnolo{kog razvoja Republike Srbije.

173
Jelena Q. CVIJETI], Novi epigrafski spomenik iz Otilovi}a kod Pqevaqa (173180) STARINAR LXII/2012

Sl. 1. Frontalna strana cipusa sa predstavom kantarosa (foto J. Cvijeti})


Sl. 2. Rimski nadgrobni spomenik stela (foto J. Cvijeti})
Fig. 1. Front of cipus with representation of kantharoi (photo J. Cvijeti})
Fig. 2. Roman tombstone stela (photo J. Cvijeti})

Reqefno poqe, dimenzija 0,45 m x 0,49 m, sa tri podignutoj do grudi, dr`ala je nekakav predmet,
strane uokvireno motivom u`eta, izvedeno je u vi- koji se zbog o{te}enosti predstave ne razaznaje.
du dvojne edikule sa arkadama, flankirane stubo- Na ovalnom licu dominiraju krupne o~i, badema-
vima sa kapitelima (sl. 4). Istovjetno oblikovana stog oblika, sa nagla{enim kapcima i obrvama, i
edikula, koja za sada predstavqa jedinu poznatu pune usne koje nagovje{tavaju osmijeh. Desno na
analogiju sa teritorije Pqevaqa, nalazi se na spo- predstavi je poprsje mu{karca u tunici, sa sagu-
meniku nedavno otkrivenom u Gradcu.2 U ni{i su mom preba~enim preko lijevog ramena i s kru`nom
en face poprsja dvoje pokojnika. Lijevo je poprsje fibulom prikop~anom na desnom ramenu. U desnoj
`ene izvedeno na na~in koji je uobi~ajen za spome-
nike iz pqevaqskog kraja. Pokojnica preko kose
nosi veo ~iji krajevi slobodno padaju na ramena i
uokviruju bistu. U desnoj ruci, savijenoj u laktu i 2 Cvijeti} 2011, 141147.

174
Jelena Q. CVIJETI], Novi epigrafski spomenik iz Otilovi}a kod Pqevaqa (173180) STARINAR LXII/2012

Sl. 3. Kru`no udubqewe na gorwoj ravnoj povr{ini spomenika (foto J. Cvijeti})


Sl. 4. Reqefna ni{a sa poprsjima pokojnika (foto J. Cvijeti})
Fig. 3. Circular hole on the upper surface of the stela (photo J. Cvijeti})
Fig. 4. Relief niche with a bust of the deceased (photo J. Cvijeti})

ruci, podignutoj do grudi, dr`i pehar. Na glavi se Desno do suda je djelimi~no o{te}eno poprsje
uo~avaju bujna talasasta kosa, koja uokviruje ~elo, mu{karca. Predstavqen je sa desnom rukom podig-
jasno nagla{ene u{i i krupne bademaste o~i. Osta- nutom do grudi, u kojoj je, kako se ~ini, dr`ao grozd.
le crte lica gotovo su neprepoznatqive. Na srcolikom licu se isti~u veoma krupne o~i i
U soklu, dimenzija 0,26 m x 0,63 m, predsta- prenagla{ene obrve koje se spajaju u horizontalnu
vqena su jo{ dva poprsja (sl. 6). U sredi{wem di- liniju. Kao i kod mu{karca u edikuli, jasno su
jelu je stilizovani sud sa spiralno savijenim kra- istaknute u{i, dok visoko ~elo uokviruje kosa, na-
jevima dr{ki (kantaros), iz koga izlazi lozica sa gla{ena samo vertikalnim zarezima.
veoma stilizovanim listovima. Sa wegove obje stra- Stela ina~e pripada ra{irenom tipu nadgrob-
ne prikazana su poprsja. Lijevo je poprsje mla|e nih spomenika u isto~nom i jugoisto~nom dijelu
`enske osobe u dugoj preklopqenoj haqini, koja u provincije Dalmacije, te bi se, po tipologiji R.
desnoj, do grudi podignutoj, ruci dr`i kqu~. Na Zotovi}, mogla opredijeliti u tip AB, koji se javqa
ovalnom licu, uokvirenom bujnom kosom, domini- od II do IV vijeka.4
raju krupne bademaste o~i sa nagla{enim kapcima Mjesto nalaza ukazuje da je stela, kao i pirami-
i izdu`en nos, od koga polaze dvije plitke bore ko- dalni cipusi, na ~ijoj je frontalnoj strani prika-
je sa izvijenim krajevima usana ukazuju na poluo- zan kantaros iz koga izvire vinova loza sa krupnim
smijeh. i plasti~no modelovanim plodovima, a na bo~nim
^ini se da je klesar najve}u pa`wu posvetio iz- stranama su delfin i riba, sekundarno iskori{}e-
radi upravo tog `enskog lika. Odje}a i frizura `e- na prilikom jedne od faza dogradwe i obnove crkve
ne prikazani su sa znatno mawe konzervatizma nego kada su sa prvobitnog mjesta, odnosno starije ne-
na predstavi pokojnice u reqefnoj edikuli. Kqu~ kropole prenijeti i postavqeni u pronaos i apsi-
u wenim rukama mogao bi je okarakterisati kao go- du crkve. Nekropola je, na`alost, nivelacijom te-
spodaricu, odnosno gazdaricu doma}instva, ili bi rena u potpunosti devastirana, tako da su uo~ena
se mogao protuma~iti kao simbol bogiwe podzem- samo dna grobnih jama ispuwena gare`i i sa svega
nog svijeta Hekate ili Dijane.3 Pa`qivim posma-
trawem lica pokojnice uo~ava se velika sli~nost
sa predstavom takozvanog genija jeseni sa nadgrob-
nog cipusa Pakonije Montane iz Komina, koji ipak 3 Srejovi} i Cermanovi}-Kuzmanovi} 1979, 450.
predstavqa znatno kvalitetniji umjetni~ki rad. 4 Zotovi} 1995, 21.

175
Jelena Q. CVIJETI], Novi epigrafski spomenik iz Otilovi}a kod Pqevaqa (173180) STARINAR LXII/2012

nekoliko ulomaka kerami~kih urni. Od nekropole ime Statia, koje je ve} poznato sa nekoliko natpisa iz
je ostalo svega nekoliko spomenika, koji samo do- Komina i Kolovrata.14 Kako primje}uje G. Alfldy,
nekle omogu}avaju rekonstrukciju slike o stanov- Staciji u Kominima i Kolovratu su doseqenici
ni{tvu koje je ovde `ivjelo.5 iz primorskog pojasa provincije, najvjerovatnije
Natpis od jedanaest redova, sa bri`qivo kle- Risinijuma.15 Pokojni~in kognomen Fuscina tako-
sanim slovima visine od 2 do 6 cm, izveden je u |e je poznat u Kominima i to na natpisu konzular-
udubqenom i trostruko profilisanom natpisnom nog beneficijara P. Aelius-a.16
poqu, dimenzija 0,65 m x 0,32 m, koje je sa strana Na osnovu ranije poznatih natpisa iz pqevaq-
uokvireno krajwe stilizovanim motivom ribqe skog kraja stvorena je slika o strukturi stanovni-
kosti (sl. 5), i glasi: {tva Municipijuma S., te se smatra da je ono u naj-
ve}oj mjeri bilo autohtono, ali da su nemali broj
D(is) M(anibus) S(acrum) | Pleto|r(i) Maxi|mina ~inili doseqenici uglavnom iz primorskog pojasa
vi|va sibi p(osuit) e|t Victorino et Stat|ie (!) Fuscine (!) provincije.17 Ime Pletor, koje D. Rendi}-Mio~e-
fi|lie (!) car(issimae) ac pi|ent(issimae) m(emoriam) vi} pripisuje op{tem ilirskom onomastikonu,18
p(osuit)| 10 marito. sada mo`emo dodati velikoj grupi ilirskih imena
koja su posvjedo~ena u Municipijumu S. Natpis iz
Paragrafsko ure|ewe je nepravilno, a od oso- Otilovi}a tako|e ukazuje na veze doseqenog i auto-
benosti teksta navodimo nekoliko ligatura: MA u htonog `ivqa, koje su posvjedo~ene na vi{e spome-
3. redu kod Maximina i VA u 5. redu kod viva. Karak- nika iz Municipijuma S., a najboqi primjer pred-
teristi~ni su i slede}i oblici slova: stavqa ime pokojnice Stacije Lavo (Statia Lavo) na
M sa horizontalnim serifima osobenost je natpisu iz Kolovrata.19 Preseqewe Stacija iz pri-
natpisa iz III vijeka6 morskog pojasa u Municipijum S. moglo se odigra-
S sa vertikalnim serifima specifi~no je ti krajem II odnosno po~etkom III stoqe}a, kada je,
za natpise iz prve tri decenije III stoqe}a7 zbog op{te ekonomske krize, posebno u primorskom
V sa nagla{enim horizontalnim serifima dijelu, stanovni{tvo tra`ilo boqe uslove za `i-
odgovara natpisima kraja II i iz prve polovi- vot, {to je svakako mogla ponuditi ova oblast na
ne III vijeka8 samom rubu provincije Dalmacije.
Naime, prosperitet municipija u Kominima,
Iz teksta se vidi da je spomenik za `ivota po- kao i cijele wegove teritorije, odigrao se tokom
digla Maximina. O~igledno je weno srodstvo sa
Statiom Fuscinom, dok na bra~ne veze sa Pletorom
ukazuje rije~ marito koju je klesar o~igledno gre-
5 Veliki broj rimskih nadgrobnih spomenika nestao je
{kom izostavio pa naknadno uklesao na ramu nat-
tokom prethodnih stoqe}a, a neki su po pri~i mje{tana, isko-
pisnog poqa. Ime Pletor ukazuje na ilirsko pori- ri{}eni i kao gra|evinski materijal prilikom zidawa pri-
jeklo pokojnika.9 Platori su poznati sa natpisa vatnih objekata.
6 Petrovi} 1975, 116.
{irom provincije Dalmacije, a ima potvrda i u
7 Petrovi} 1975, 119.
Dakiji,10 {to se mo`e vezati za preseqewe ilir-
8 Petrovi} 1975, 120.
skog `ivqa iz Dalmacije u ovu, rudama bogatu, pro-
9 O imenu v. Mayer 1957, 273, 274; Alfldy 1969, 355; Ren-
vinciju, dok je varijanta Pletor za sada jedino po-
di}-Mio~evi} 1955, 131 i dr.
znata iz Salone, Nauportus-a i Piquentum-a.11 10 Mayer 1957, 273, 274.
Ime Maximina preuzeto je iz latinskog onoma- 11 CIL III 9279, CIL III 10723, 10724, CIL V 449, Mayer

stikona i vjerovatno predstavqa prevod doma}eg 1957, 274.


imena sli~nog zna~ewa.12 Imena kao {to su Maxi- 12 Rendi}-Mio~evi} 1964, 101110.
13 Alfldy 1969, 243.
mus i Maximina su ina~e omiqena kod domoroda~-
14 CIL 8302; CIL 8326; Cermanovi}-Kuzmanovi} 1970, 76;
kog stanovni{tva {irom balkanskih provincija,
Mirkovi} 1975, 100.
naro~ito u Dalmaciji. Na teritoriji Municipija 15 Alfldy 1965, 57.
S. potvr|ena su na vi{e natpisa.13 16 CIL 13847.
Tekst u 7. redu klesar je morao da prilagodi ma- 17 Cermanovi}-Kuzmanovi} 1970, 80 i dr.

wem o{te}ewu nastalom prije ispisivawa spome- 18 Rendi}-Mio~evi} 1955, 131.

nika. Nema sumwe da je pokojnica nosila gentilno 19 Mirkovi} 1975, 100.

176
Jelena Q. CVIJETI], Novi epigrafski spomenik iz Otilovi}a kod Pqevaqa (173180) STARINAR LXII/2012

Sl. 5. Frontalna strana stele natpisno poqe


i predstava pokojnika u soklu (foto M. Sladi})
Sl. 6. Predstava u soklu (foto J. Cvijeti})
Fig. 5. Front side of monument inscription
and representation of the deceased in the socle
(photo M. Sladi})
Fig. 6. The representation in the socle (photo J. Cvijeti})

isto~ni rub provincije Dalmacije krajem II odno-


sno po~etkom III vijeka.22 To je vrijeme koje bi odgo-
varalo doseqewu porodice Stacija, koja je, kao i
Pakoniji, sigurno zauzimala vrlo zna~ajnu ulogu u
Municipiju S. tokom wegovog naglog ekonomskog
prosperiteta.
Ovom prilikom pomenu}u i dva fragmentovana
rimska nadgrobna spomenika iz neposredne blizi-
ne lokaliteta, na koje su nam ukazali mje{tani se-
III stoqe}a, prvenstveno zahvaquju}i rudnim bogat- la Otilovi}i.
stvima,20 {to je bio jedan od osnovnih razloga za Prvi, od koga je sa~uvana samo gorwa polovina,
doseqavawe velikog broja uglednih porodica predstavqa nadgrobnu stelu od sivkastog kre~waka,
uglavnom iz primorskih oblasti jugoisto~ne Dal- dimenzija 0,60 m x 0,63 m x 0,26 m, koja se danas
macije (Aemilii, Caesii, Egnatii, Gavienii, Paconii, nalazi na seoskom grobqu u Otilovi}ima (sl. 7). U
Statii i dr.).21 polukru`noj reqefnoj edikuli, ~iji okvir formi-
Na postojawe naseqa ili vile rustike iz rim- raju ornamenti ukomponovanih listova,23 predsta-
skog doba u Otilovi}ima ukazuje prije svega pome- vqena su poprsja dvoje pokojnika. Lijevo je prikaz
nuta nekropola. Naseqe u Otilovi}ima imalo je `ene, predstavqene na uobi~ajen na~in. Dugi veo
sve pogodnosti za razvoj. Zbog veoma zna~ajne ceste
koja je ba{ tuda prolazila, potvr|ene jo{ krajem
XIX stoqe}a nalazom miqokaza, a koja je poveziva-
20
la velika naseqa u Kominima i Kolovratu, a sa- Cermanovi}-Kuzmanovi} 1970, 80; Mirkovi} 1975, 105, 106.
21 Cermanovi}-Kuzmanovi} 1969, 105; Cermanovi}-Kuzma-
mim tim i rudarske revire u jugoisto~nom dijelu
novi} 2009, 59.
provincije Dalmacije, kao i velikih prirodnih 22 Evans 1885, 43.
bogatstava, osobito {umskih, ne ~udi za{to se ba{ 23 Identi~no oblikovano reqefno poqe, sa jednakim flo-
tu razvilo rimsko naseqe, i to najvjerovatnije u pe- ralnim motivom koji ~ini okvir, nalazi se na fragmentu spo-
riodu najve}eg ekonomskog procvata koji je zadesio menika u parku Vodice u Pqevqima (neobjavqeno).

177
Jelena Q. CVIJETI], Novi epigrafski spomenik iz Otilovi}a kod Pqevaqa (173180) STARINAR LXII/2012

Sl. 7. Fragment stele iz seoskog grobqa u Otilovi}ima (foto J. Cvijeti})


Sl. 8. Fragment stele sa lokaliteta Gromilice u Otilovi}ima (foto J. Cvijeti})
Fig. 7. Fragment of the stela from the cemetery from Otilovi}i (photo J. Cvijeti})
Fig. 8. Fragment of the stela from the site Gromilice in Otilovi}i (photo J. Cvijeti})

preba~en preko wene kose dopire do ispod ramena. rimskih nadgrobnih spomenika tokom posledwih
Na ovalnom licu dominiraju krupne o~i, dok su godina ukazuju na to da bi daqa istra`ivawa tre-
ostale crte svedene gotovo na jednostavne ureze. balo usmjeriti upravo na glavne putne pravce koji
Mu{karac do we ima desnu ruku podignutu do gru- su povezivali Municipijum S. sa drugim gradskim
di i ~elo uokvireno bujnom talasastom kosom. centrima u ovom dijelu provincije gdje su se nala-
Drugi spomenik je slu~ajno na|en prije vi{e zila ve}a vojna i civilna jezgra kao {to je slu~aj
godina na lokalitetu Gromilice, na imawu Ne|eq- sa naseqem u Otilovi}ima.
ka Gruji~i}a (sada u Zavi~ajnom muzeju u Pqevqi- @ivot u wemu nastavqen je i tokom narednih
ma), na oko 750 m jugoisto~no od pomenute crkve u stoqe}a. Na mjestu rimske nekropole nastalo je
Otilovi}ima (sl. 8). U pitawu je fragment stele od ranohri{}ansko jezgro, {to pokazuju ostaci crkve i
sivkastog kre~waka, dimenzija 0,40 m x 0,43 m x grobnice. To je utoliko va`nije {to do sada ovakvi
0,20 m x 0,20 m. Djelimi~no je vidqivo samo popr- nalazi nisu otkriveni ni u Kominima ni u wihovoj
sje `enske osobe u polukru`noj reqefnoj edikuli, bli`oj okolini. Kontinuitet do punog sredweg vi-
dok drugi dio predstave nedostaje (desno do we je jeka potvr|uju kasnije intervencije na crkvi, kada
vjerovatno bilo poprsje mu{ke osobe). U uglu iznad su i ovi nadgrobni spomenici dobili novu funkci-
edikule nalazi se trolisna rozeta. ju. Polo`aj stele u crkvi negira rimsku odnosno
Fragmenti oba spomenika mogu se {ire opredi- pagansku koncepciju i tradiciju, dok cipusi, koji
jeliti u drugu polovinu II stoqe}a ili u III vijek, se svojom dekoracijom uklapaju u hri{}ansku iko-
odnosno u vrijeme kada se uz pomenutu komunika- nografiju, dobijaju veoma istaknuto mjesto u apsi-
ciju razvilo va`no naseqe. Ina~e, mjesta nalaza di crkve.

178
Jelena Q. CVIJETI], Novi epigrafski spomenik iz Otilovi}a kod Pqevaqa (173180) STARINAR LXII/2012

BIBLIOGRAFIJA:

Alfldy 1965 G. Alfldy, Bevlkerung und Gesel- Evans 1885 A. Evans, Antiquarian Researches in
lschaft der Rmischen Provinz Dalmatien, Budapest 1965. Illyricum. Archaeologia 49, 1167.
Alfldy 1969 G. Alfldy, Die Personennamen in Mayer 1957 A. Mayer, Die Sprache der alten Illi-
der Rmischen Provinz Dalmatien, Heidelberg 1969. rier, Wien1957.
Cermanovi}-Kuzmanovi} 1969 A. Cermano- Mirkovi} 1975 M. Mirkovi}, Iz istorije Polimlja
vi}-Kuzmanovi}, Municipium S. i wegova proble- u rimsko doba, Godi{njak CBI 14, 95108.
matika u svetlu arheolo{kih i epigrafskih spome- Petrovi} 1975 P. Petrovi}, Paleografija
nika, Starinar n.s. XIX, Beograd 1969, 101109. rimskih natpisa u Gorwoj Meziji. Posebna izdawa
Cermanovi}-Kuzmanovi} 1970 A. Cermano- 14, Beograd 1975.
vi}-Kuzmanovi}, Nekoliko neobjavqenih natpisa Rendi}-Mio~evi} 1955 D. Rendi}-Mio~evi}, Ono-
iz Komina, Zbornik Filozofskog fakulteta u Beo- masti~ke studije sa teritorije Liburna (Prilozi ilirskoj ono-
gradu 11/1, 7581. mastici), Zbornik instituta za historijske nauke u Zadru,
Cermanovi}-Kuzmanovi} 2009 A. Cermano- Zadar 1955, 125142.
vi}-Kuzmanovi}, Pqevaqsko podru~je u rimsko i Rendi}-Mio~evi} 1964 D. Rendi}-Mio~evi}, Ilir-
ranovizantijsko doba, u Istorija Pqevaqa, ur. S. ske onomasti~ke studije (II), @iva antika 1314, 101110.
Terzi}, Pqevqa 2009, 4969. Zotovi} 1995 R. Zotovi}, Rimski nadgrobni
CIL Corpus inscriptionum Latinarum. spomenici isto~nog dela provincije Dalmacije, U`i-
Cvijeti} 2011 J. Cvijeti}, Rimski nadgrobni ce 1995.
spomenik iz Gradca kod Pqevaqa, GSAD 25, 141147.

179
Jelena Q. CVIJETI], Novi epigrafski spomenik iz Otilovi}a kod Pqevaqa (173180) STARINAR LXII/2012

Summary: JELENA LJ. CVIJETI], University of Belgrade, Faculty of Philosophy,


Department of Archaeology, Belgrade

THE NEW EPIGRAPHIC MONUMENT OF OTILOVI]I


NEAR PLJEVLJA

Key words. Dalmatia, Otilovi}i, Roman tombstones, the name Pletor.

The remains of an early Christian church were discovered, by in Municipium S. The name of the deceased female Fuscina,
chance, during the leveling of the terrain in Otilovi}i near Pljev- whose nomen was Statia, and whose names appeared on more
lja where there were two monuments, or pyramidal cipi. At the than one epigraph in Komini and Kolovrat, represents inhabi-
same time, the older Roman necropolis was totally destroyed, tants who probably came from the coastal region of Risinium.
so that only the bottoms of tomb holes filled with soot could be Their migration from the coast to Municipium S. could have taken
seen in the terrain. Following the protected archaeological ex- place at the end of the second or the beginning of the 3rd century
cavation, the aforementioned church and a small arched tomb when many respectable families, due to economic crises, looked
by its northern wall were uncovered in full. for shelter inside the province which was wealthy with natural
Another Roman tombstone was found on this occasion, a resources, especially ore. In fact, this was a time of rapid growth
stela which was used as a stairway from pronaos to naos in the and economic prosperity for Municipium S. The necropolis at
church. The stela belongs to a very large group of tombstones Otilovi}i points to the existence of a villa rustica or an estate
from this part of the province and its compositional schema is a which developed into an important communications link between
unique example in the area of Pljevlja. From the epigraph we the estates in Komini and Kolovrat, and which was acknowled-
collected the names of four deceased persons (Pletor, Maximina, ged at the end of the 19th century by mileposts from Otilovi}i
Victorinus and Statia Fuscina). The name Pletor, which was seen and ]i}ija.
for the first time in the area of Pljevlja on this epigraph, can be
added to a large group of Illyrian names that were acknowledged Translated by Marija Borovi}

180
NATA[A MILADINOVI]-RADMILOVI]
The Institute of Archaeology, Belgrade

ANALYSIS OF HUMAN OSTEOLOGICAL MATERIAL


FROM THE EASTERN PART OF SITE NO. 37
IN SREMSKA MITROVICA

UDK: 904:726.821(497.113)"09/11" ; 902.2:572.7(497.113)"2010" e-mail: miladinovic.radmilovic@gmail.com


DOI: 10.2298/STA1262181M Received: February 20, 2012
Short communication Accepted: June 21, 2012

Abstract. The direct reason for writing this paper was the new find of skeletons in the medieval necropolis (10th12th century)
discovered as far back 1968 at the Site No. 37 in Sremska Mitrovica (Sirmium). Institute for the protection of cultural
monuments in Sremska Mitrovica undertook protective archaeological excavations in the eastern part of the site in 2010,
discovering 29 skeletons. Since that archaeological analysis of Belo Brdo communities is still in its infancy and considering
that there is not a sufficiently big sample for a more precise monitoring of this populations inner dynamics, it is considered
useful to present results gained by studying these skeletons on Site No. 37. Although the results in many ways match the results
gained up until now, there are some paleopathological changes that so far, have not appeared and for which we had no direct
confirmation in the osteological material. One of these paleopathological changes is certainly syphilis.

Key words. medieval Sirmium, Belobrdo culture, syphilis.

ite No. 37 is located at the corner of Vuk Karad`i} 83 and 85, Ju`ni bedem, Ma~vanska Mitrovica,3 and

S and Saint Sava Street in the area of a demolished


town prison in Sremska Mitrovica. Protective
archaeological excavations were conducted in 1968
Site Trasa kanalizacije Dositejeva Street. Unfortuna-
tely, only 82 skeletons were available for anthropologi-
cal analysis (from Site No. 83 (nine individuals), Site
and 1969 over the area of 1600 m (Fig. 1 and 2). On No. 85 (65 individuals), Ju`ni bedem (two individuals),
that occasion a section of the northern wing of the
Sirmium imperial palace was explored, as well as a
Gepidian cultural layer from the 5th century and a part
of a medieval necropolis with skeletal burials from 1 Osteological material of human origin from this site was sent

10th12th century.1 Finds from this necropolis belong to USA for anthropological expertise in the 1970s. Unfortunately,
to the Belobrdo culture.2 the results of these analyses have not yet been delivered to the
Museum of Srem in Sremska Mitrovica or Institute of Archaeology
Between 1957 and 2007, graves from 10th12th
in Belgrade. Likewise, they have not been published, as far as the
century, containing Belo Brdo culture materials were author of this text is informed.
discovered in Sremska Mitrovica, on the total of 11 sites 2 Milo{evi} 1994, 31.

(Fig. 1 and 2). Those are Sites No. 4, 25, 34, 35, 37, 66, 3 Tomi~i} 2010, 121, 128, 133135, Tab. 27.

* This article is the result of the projects: Romanization, urbanization and transformation of urban centres of civil, military and residential
character in Roman provinces on the territory of Serbia (No. 177007) and Urbanization Processes and Development of Medieval Society
(No. 177021) founded by the Ministry of Education, Science and Technological Development of the Republic of Serbia.

181
Nata{a MILADINOVI]-RADMILOVI], Analysis of human osteological material (181204) STARINAR LXII/2012

Fig. 1. Map of Sremska Mitrovica, necropolis from 10th12th century 4


Sl. 1. Karta Sremske Mitrovice, nekropole XXII veka

Ma~vanska Mitrovica (five individuals) and Site Trasa Fig. 2. Map of Sremska Mitrovica, Site No. 37
kanalizacije Dositejeva Street (one individual)).5 Sl. 2. Karta Sremske Mitrovice, lokalitet 37
In September 2010, a team from the Institute for
the protection of cultural monuments in Sremska
Mitrovica undertook protective excavations in the
Saint Sava Street. On that occasion a sonde, measuring
4 x 4 m was opened (Figs. 36). Eighteen graves and
four groups of dislocated bones were discovered (29
skeletons in total). Skeletons were mostly oriented
southwest-northeast. The deceased were laid on their
backs with arms beside their bodies. A number of iron
nails were discovered, leading archaeologists to the
conclusion that the deceased had been buried in wood-
en coffins.6

MATERIAL

Osteological material of human origin from previ-


ous excavations on Site No. 37 was, as mentioned, un-
available for analysis, so it was decided to present the
analysis of all 29 individuals (Table 1) thus contributing
towards creating a general picture of this population.

182
Nata{a MILADINOVI]-RADMILOVI], Analysis of human osteological material (181204) STARINAR LXII/2012

Of course, a broad archaeological and chronological (corpus mandibulae, ramus mandibulae and angulus
dating represented a great difficulty in anthropological mandibulae), mentum, angulus mandibule and margo
reconstruction and interpretation (10th12th century) inferior), based on criteria defined by Ferembach and
contributed by, among other things, a large number of his associates,15 and metric elements relevant for sex
finds discovered in the necropolises which were not determination in skeletons.16 Indices, calculated on the
chronologically sensitive, as was outlined, as well as an basis of gained metric elements, were shown in tables
insufficient number of skeletons discovered. There- for each grave individually. Teeth were measured for
fore, it was impossible to observe the inner dynamics mesio-distal and vestibulo-lingual diameters using a
of this population more precisely even when the site method approved by Hillson.17 According to these dia-
i.e. the necropolis is uncovered totally or to a great meters difference in teeth size was monitored mostly
extent, as opposed to colleagues in our region that have on canines; should they be missing from osteological
been successfully engaged in this enterprise.7 material, other teeth would suffice (molars, premolars
and incisors).18 Morphological and metric elements
were observed during analysis of other postcranial
METHODOLOGICAL FRAMEWORK bones as well. Morphological elements that caught the
most of our attention were degrees of development of:
The examined degree of skeleton preservation is tuberositas deltoideae, tuberositas radii and margo
given in the form of descriptive schemes consisting of interosseus (of the radius), tuberositas ulnae and
five categories proposed by Miki}:8 I the whole margo interosseus (of the ulna), linea aspera and
skeleton is well preserved; II well-preserved, incom- tuberositas tibiae. Bone appearance, body curvature
plete skeleton; III moderately preserved skeleton;9 and facies auricularis were morphological elements
IV partial preservation of skeletal remains10 and V observed in sacrum.19 Metric elements played a more
poor preservation of skeletal remains.11
In determining sex in children, we put emphasis on
the study of morphological elements of the mandible
(protrusion of protuberantiae mentalis, the shape of the 4 All photographs of humane osteological material were taken
alveolar part, protuberance in the gonion area) and pelvis by N. Miladinovi}-Radmilovi}. Postproduction and electronic pro-
(the angle of a greater sciatic notch, the position of the cessing of situation plans from the field documentation of the
Institute for the Protection of Cultural Monuments from Sremska
pelvic arch, the curvature of cristae iliacae). The metho- Mitrovica and map making were done by M. Radmilovi}.
dology was based on data obtained by Schutkowski 5 Miladinovi}-Radmilovi} 2011, 465510.
during his extensive research.12 6 The data was taken from the field documentation of the Insti-

For sex determination on skeletal materials of adult tute for the Protection of Cultural Monuments in Sremska Mitrovica.
7 Vodanovi}, Brki}, Demo i [laus 2003; Vodanovi}, Brki} i
individuals we adopted for a combination of morpho-
Demo 2004; Bedi} i Novak 2010.
logical and metrical methods. Specific attention was 8 Miki} 1978, 9.
being paid on morphological elements of the scull 9 Medium preservation refers to the situation where an entire
(glabella, planum nuchale, processus mastoideus, pro- skeleton is present inside the grave, but the bones are brittle and
cessus zygomaticus, arcus supercilialis, protuberantia brake during excavation.
10 Partial preservation refers to the situation where the grave
occipitalis externa, os zygomaticum, tubera frontale et
parietale, inclination of os frontale, margo supraorbi- contains only parts of a skeleton that are very brittle and difficult to
lift, pack and transport.
talis and shape of orbitae) and the pelvis (sulcus pra- 11 Poor preservation refers to the situation where the remains
earicularis, incisura ischiadica s. ischialis major, arcus of a skeleton exist only in traces and are virtually impossible to lift
pubis s. pubicus et angulus subpubicus, arc compose, the completely.
12 Schutkowski 1993.
appearance of os coxae, corpus ossis ischii, foramen
13 Ferembach, Schwidetzky and Stloukal 1980, 519527.
obturatum, crista iliaca, fossa iliaca, pelvis major, pelvis
14 Buikstra and Ubelaker 1994, 1521.
minor; subpubic region: ventral arc, subpubic concavity
15 Ferembach, Schwidetsky and Stloukal 1980, 523525.
and medial appearance of the ischio-pubic branch), 16 Ferembach, Schwidetsky and Stloukal 1980, 523525;
whereas the method of operation was adopted from a Bass 1995, 84, 85.
group of European anthropologists,13 Buikstra and 17 Hillson 1990, 240242; idem. 1996, 8082.

Ubelaker.14 Morphological elements were also analyzed 18 Garn, Lewis and Kerewsky 1965.

on the mandible (the overall appearance of mandible 19 Miki} 1978, 18, 19; Bass 1995, 114.

183
Nata{a MILADINOVI]-RADMILOVI], Analysis of human osteological material (181204) STARINAR LXII/2012

Fig. 3. Sonde 1, position of graves 1, 2, 3, 4, 5 and 6 20


Fig. 4. Sonde 1, position of graves 8, 9, 10, 11, 12, 13 and 14 21
Sl. 3. Sonda 1, polo`aj grobova 1, 2, 3, 4, 5 i 6
Sl. 4. Sonda 1, polo`aj grobova 8, 9, 10, 11, 12, 13 i 14

Fig. 5. Sonde 1, position of graves 8, 9, 10, 11, 12, 13 and 14 22


Fig. 6. Sonde 1, position of graves 12, 13 15, 16, 17 and 18 23
Sl. 5. Sonda 1, polo`aj grobova 8, 9, 10, 11, 12, 13 i 14
Sl. 6. Sonda 1, polo`aj grobova 12, 13, 15, 16, 17 i 18

184
Nata{a MILADINOVI]-RADMILOVI], Analysis of human osteological material (181204) STARINAR LXII/2012

significant role in sex determination based on postcra- HUMAN OSTEOLOGICAL MATERIAL


nial skeleton, and they were given additional attention. FROM EASTERN PART OF SITE NO. 37
Indices calculated on the basis of gained metric ele-
ments were shown in tables for each grave individual- Grave 1
ly, and for left and right side separately. Skeletal remains of a female (?) child aged 18 (?)
Individual age estimation in children was based on months were discovered in the grave (Figs. 3, 7a and
degree of formation and teeth eruption (Ubelaker 7b; Tables 1 and 2).38
scheme);24 degree of ossification of the epiphysis-dia- Paleopathological changes that can be observed on
physis connections (Table with time scales (years) dur- the cranial part of the skeleton are porotic hyperostosis
ing which epiphysis-diaphysis connections ossifi- and traces of tuberculosis on the ribs. Postcranial
cate);25 length of long bones (tables (with time scales bones exhibit some sort of dysplasia (achondropla-
shown in years and months) defined by Bass26 and sia?). Namely, thickening of the cortex and noticeable
Ferembach with associates).27 enlargement of mediolateral diameter is perceived in
Individual age in adults was established upon: the region of long-bones diaphysis and metaphysis
degree of obliteration of local skull sutures (Vallois (Figs. 7a and 7b). Severe body curvature is solely
scheme);28 changes in maxilla and mandible teeth observed in the left fibula. Deeper lesions are per-
(changes in occlusal surface on the dental material was ceived on the anterior and posterior side of the iliac
compared with the numerical classification of attrition part of the left pelvic area, as well as on all muscle
of the upper (occlusal) surface of molars in relation to attachment points and long postcranial bones.
age which was defined by Brothwell29 and changes on
occlusal surface of all teeth in relation to age defined Grave 2
by Lovejoy;30 morphological changes in sternal ends The grave contained remains of a male child indi-
of ribs (metamorphoses of depth, joint cavities, shape, vidual, aged four and a half,39 and a child individual,
edges and ridge configuration were examined, togeth-
er with overall state of bone, based on ten (08) phas-
es of progression covering the period from 18 to over
70 years);31 morphological changes on the medial end 20 Field documentation of the Institute for the Protection of
of the clavicle (morphological changes of the clavicle Cultural Monuments in Sremska Mitrovica.
documented by Scheuer and Black were observed).32 21 Field documentation of the Institute for the Protection of

They established five (15) phases of progression cov- Cultural Monuments in Sremska Mitrovica.
22 Field documentation of the Institute for the Protection of
ering periods lasting from 14 to 29 years); morpholog-
Cultural Monuments in Sremska Mitrovica.
ical changes in pubic symphysis joint surface (Todds 23 Field documentation of the Institute for the Protection of
method was used in which the metamorphosis of the Cultural Monuments in Sremska Mitrovica.
pubic symphysis surface is divided in ten chronologi- 24 Ferembach, Schwidetzky and Stloukal 1980, 528, 529.

cal phases during aging, starting with age 18 and lead- 25 Ferembach, Schwidetzky and Stloukal 1980, 531.

ing up to age 50 and over);33 sacroiliac region (indi- 26 Bass 1995, 155, 168, 176, 228, 247, 257.
27 Ferembach, Schwidetzky and Stloukal 1980, 532.
vidual age of adult individuals was determined upon
28 Vallois 1937.
models defined by Lovejoy and his associates.34 They
29 Brothwell 1981, 72.
classified the changes in this region in eight stages, 30 Lovejoy 1985.
from late adolescence to old age phase, with most 31 Ican, Loth and Wright 1984a; idem. 1984b; idem. 1985.
attention directed to observation of position, edge lip- 32 Scheuer and Black, 2000.
ping and porosity of the bone in this region). 33 Todd 1920, 285334; idem. 1921a; idem. 1921b.

Twenty-six epigenetic variations on the cranium and 34 Lovejoy et al. 1985.

eleven on the postcranial skeleton were observed.35 35 Hauser and De Stefano 1989; \uri}-Sreji} 1995,

Stature in children and juvenile (juvenilis I) indi- 238260.


36 Walker and Prez-Prez, 18.
viduals was calculated using a formula defined by
37 Trotter and Gleser 1952.
Maresh,36 whereas for juvenile (juvenilis II) and adult 38 Degree of bone preservation: II category (a well preserved
individuals Trotter and Glesers formulas were used.37 incomplete cranial and postcranial skeleton).
39 Degree of bone preservation: II category (a well preserved

incomplete cranial and postcranial skeleton).

185
Nata{a MILADINOVI]-RADMILOVI], Analysis of human osteological material (181204) STARINAR LXII/2012

a b

Fig. 7. Grave 1, dysplasia (achondroplasia?): a) of the left humerus; b) of the left femur
Sl. 7. Grob 1, displazija (achondroplasia?): a) levog humerusa; b) levog femura

a b

Fig. 8. Grave 2: a) cribra femora; b) traces of tuberculosis on ribs


Sl. 8. Grob 2: a) cribra femora; b) tragovi tuberkuloze na rebrima

186
Nata{a MILADINOVI]-RADMILOVI], Analysis of human osteological material (181204) STARINAR LXII/2012

of undetermined sex, aged around 30 months (Figs. 3, No paleopathological changes were noticed in the
8a and 8b; Tables 1 and 2).40 adult individual.
Perceived paleopathological changes in the older
individual are porotic hyperostosis (on parietal bones) Grave 6
cribra femora near the upper end of the left and right The grave contained skeletal remains of a male (?)
femur on the anterior side (measuring 1 x 2 cm) and a child aged 3 years 12 months (Figs. 3, 9a and 9b;
trace of tuberculosis on the one preserved rib (Figs. 8a Table 1).46
and 8b). Perceived paleopathological changes are cribra
A noticeable epigenetic characteristic on norma orbitalia on orbital roofs and porotic hyperostosis on
frontalis are sulci frontales (one on the left side), and on lamina externa on all preserved cranial bones except
norma lateralis two foramen zygomaticofaciale (on the the occipital bone (Fig. 9a). Changes similar to those
left zygomatic bone). Trochanter tertius was noticed caused by metabolic processes (scurvy or rickets) were
on the right femur of the postcranial skeleton. noticed on lamia interna of the occipital and left pari-
No paleopathological changes were noticed in the etal bone. Likewise, there is a possible ear inflamma-
younger child. Tuberositas radii is somewhat more tion accompanied by an infection, similar to individual
prominent than usual. from grave 5 (Fig. 9b).

Grave 3 Grave 8 47
The grave contained skeletal remains of a child of The grave contained skeletal remains of a female
undetermined sex and age,41 and an adult individual of child, aged three,48 and an adult or a juvenile individual
undetermined sex and age (Fig. 3; Table 1).42 of unknown sex and age (Figs. 4, 5 and 10; Tables 1
No paleopathological changes were noticed in these and 2).49
individuals. Perceived paleopathological changes in the child
are ellipsoidal bony protuberance on one rib fragment
Grave 4 (measuring 0.8 x 0.5 cm), deeper lesions in the upper
The grave contained skeletal remains of a child of third of the body of the left humerus on the anterior
undetermined sex, aged three and a half (Fig. 3; Tables side (the affected bone area measures 1 x 2.5 cm; Fig.
1 and 2).43 10) and cribra femora near the upper ends on the ante-
No paleopathological changes were noticed. The rior side of the femur (0.7 cm in diameter).
appearance of suturae metopicae on norma frontalis is
a perceived epigenetic characteristic.
40 Degree of bone preservation: II category (a well preserved
Grave 5 incomplete postcranial skeleton).
The grave contained skeletal remains of a female 41 Degree of bone preservation: II category (a well preserved

child aged 2 years 8 months,44 and an adult indi- incomplete cranial skeleton).
42 Degree of bone preservation: II category (a well preserved
vidual of undetermined sex and age (Fig. 3; Tables 1
incomplete postcranial skeleton).
and 2).45 43 Degree of bone preservation: II category (a well preserved
Perceived paleopathological changes in the child incomplete cranial and postcranial skeleton).
individual are cribra femora near the upper end of the 44 Degree of bone preservation: II category (a well preserved

right and left femur on the anterior side (1cm in diam- incomplete cranial and postcranial skeleton).
45 Degree of bone preservation: II category (a well preserved
eter), resorption of cortical tissue at the muscle attach-
incomplete cranial and postcranial skeleton).
ment point m. triceps brachii Caput laterale (right 46 Degree of bone preservation: II category (a well preserved
humerus), m. biceps brachii (right and left radius), m. incomplete cranial and postcranial skeleton).
iliopsoas and at the point of attachment of all muscles 47 In a so-called grave 7 only animal skeletal remains were

along linea aspera (right and left femur), dislocation discovered. It should be mentioned that in graves 1, 4, 5, 8, 9, 10,
of the left ankle and a possible middle ear inflamma- 12, 16 and 17, as well as among dislocated bones IIII numerous
animal bone fragments were found, most likely offerings.
tion accompanied by an infection. 48 Degree of bone preservation: II category (a well preserved
Epigenetic characteristics noticed on norma later- incomplete cranial and postcranial skeleton).
alis are two foramen zygomaticofaciale on the right 49 Degree of bone preservation: II category (a well preserved

zygomatic bone. incomplete postcranial skeleton).

187
Nata{a MILADINOVI]-RADMILOVI], Analysis of human osteological material (181204) STARINAR LXII/2012

Fig. 9. Grave 6: a) porotic hyperostosis; b) possi-


ble ear inflammation accompanied by infection
Fig. 10. Grave 8: deep lesions on the left humerus
Sl. 9. Grob 6: a) porozna hiperostoza; b)
mogu}a upala uha pra}ena infekcijom
b
Sl. 10. Grob 8: dubqe lezije na levom humerusu

No paleopathological changes were noticed in the Perceived paleopathological changes are irregularly
adult individual fused fracture of the II metatarsal bone (in the upper half
of the body), injury to the right tibia (on the middle of
Grave 9 margo anterior, a bony protuberance 1 x 0.5 cm in size
The grave contained skeletal remains of a child indi- can be noticed, and an infection on the lower end on the
vidual, of unknown sex and age (Figs. 4 and 5; Table 1).50
Cribra orbitalia was a paleopathologic find noticed
on the left orbit roof.
50 Degree of bone preservation: II category (a well preserved
Grave 10 incomplete cranial skeleton).
The grave contained skeletal remains of a male adult 51 Degree of bone preservation: II category (a well preserved

individual, of unknown age (Figs. 4 and 5; Tables 1 and 8).51 incomplete postcranial skeleton).

188
Nata{a MILADINOVI]-RADMILOVI], Analysis of human osteological material (181204) STARINAR LXII/2012

medial and posterior side), osteoarthritis and the disloca- Markers of occupational stress in the form of
tion of both ankles (may have occurred as a result of dif- hypertrophy (cortical defect) were present on the mus-
ficulties in movement due to the injury to the right tibia) cle attachment points of the right and left clavicle (m.
and the possible emergence of the so-called bunion. deltoideus), right scapula (m. triceps brachii Caput
longum, m. subscapularis, m. infraspinatus, m. teres
Grave 11 minor, m. teres major), left scapula (m. triceps brachii
The grave contained skeletal remains of a male child Caput longum, m. subscapularis, m. infraspinatus,
aged two and a half (Figs. 4 and 5; Tables 1 and 2).52 m. teres minor, m. teres major, m. deltoideus, m. biceps
Perceived paleopathological changes are middle ear brachii Caput longum, m. biceps brachii Caput
inflammation and cribra femora near the upper ends breve, m. serratus anterior, m. rhomboideus minor, m.
on the anterior side of both femurs (1.5 and 1 cm in rhomboideus major), right humerus (m. brachioradialis,
diameter). m. extensor carpi radialis longus, m. extensor carpi
Noticable epigenetic characteristic on norma occi- radialis brevis, m. extensor digitorum, m. extensor digiti
pitalis are ossa suturae lambdoideae (one on the left minimi, m. extensor carpi ulnaris, m. supinator, m.
side 1.3 x 1 cm in size). pronator teres), left humerus (m. brachioradialis, m.
extensor carpi radialis longus, m. extensor carpi radi-
Grave 12 alis brevis, m. extensor digitorum, m. extensor digiti
The grave contained skeletal remains of a female, minimi, m. extensor carpi ulnaris, m. supinator, m.
adult individual, aged between 3346 (Figs. 46, 11 pronator teres, m. supraspinatus, m. subscapularis, m.
and 12; Tables 1, 38).53 latissimus dorsi, m. pectoralis major, m. teres major,
Noticeable paleopathological changes are a mild m. deltoideus, m. coracobrachialis, m. brachialis),
form of osteoarthritis (on the condyles of the right radius (m. adductor pollicis longus,54 m. biceps
mandible, on several thoracic vertebrae on the upper brachii), left radius (m. adductor pollicis longus, m.
end of the right ulna and on right tibias tuberositas), biceps brachii), right ulna (m. supinator, m. brachialis,
aneurism (?) on the medial end of the right clavicle m. pronator teres, m. flexor digitorum superficialis, m.
(1.5 cm in diameter) and bony outgrowths (0.5 and 0.2 triceps brachii; olecranon was slightly seperated), left
cm in diameter) on the right tibias facies medialis. ulna (m. supinator, m. brachialis, m. pronator teres, m.
Dental analysis showed the presence of the fol- flexor digitorum superficialis), both femurs (all attach-
lowing teeth: 16, 18, 26, 31, 32, 33, 34, 35, 37, 38, 43 ment points are prominent along lineae asperae and
and 44. Teeth 17, 25, 28, 36, 46 and 47 (Figs. 11 and near the lower end on the posterior side) and both fibu-
12) were lost antemortem, teeth 14, 15, 24, 27 (?), 41, lae (m. flexor hallucis longus). Markers of occupation-
42 and 45 postmortem. Abrasion of the 1st degree (in al stress in the form of hypertrophy (cortical defect)
enamel) was discovered in teeth 16, 35 and 44 (II), were present on attachment points of right clavicles
2nd degree (exposed dentin) on 34, and 3rd degree (to ligaments (lig. trapezoideum, lig. conoideum) and left
the bottom of the fissure) on teeth 31, 32, 33 and 43). clavicle (lig. trapezoideum, lig. conoideum, lig. costo-
Periodontal disease and calculus were highly promi- claviculare).
nent (due to a large presence of calculus, the possible Specific observations: the emergence of batrocran;
appearance of hypoplasia was unobservable). Teeth foramen mandibulae is larger (1 cm in diameter); con-
rotation was the only present anomaly concerning dyle is extremely large in size (3.3 x 2.25 cm); facies
mandible and dental arch). Caries was present in teeth: articularis tuberculi costae is disk-shaped (1.3 cm in
17 (mesial, caries 0.7 cm in diameter), 26 (the so-called diameter) with a perforation in the middle.
gross-gross caries), 38 (occlusal, caries in the shape of
dot) and 48 (occlusal, caries in the shape of dot).
Occlusion could not be determined.
Epigenetical characteristics noticeable on norma
52 Degree of bone preservation: II category (a well preserved
frontalis are sulci frontales (two on the left side) and
incomplete cranial and postcranial skeleton).
linea nuchae suprema (very prominent) on norma 53 Degree of bone preservation: II category (a well preserved
occipitalis. On the postcranial part of the skeleton, incomplete cranial and postcranial skeleton).
trochanter tertius was noticed on the right femur 54 There is a bony protuberance1 x 2 cm in size on the attach-

beside foramen processus transversi bipartitum (C6). ment point of this muscle.

189
Nata{a MILADINOVI]-RADMILOVI], Analysis of human osteological material (181204) STARINAR LXII/2012

Fig. 11. Grave 12: skull projections


Sl. 11. Grob 12: lobawske projekcije

190
Nata{a MILADINOVI]-RADMILOVI], Analysis of human osteological material (181204) STARINAR LXII/2012

Fig. 12. Grave 12: mandible


Sl. 12. Grob 12: mandibula

Grave 13
The grave contained skeletal remains of a male adult
individual aged around 25 (Figs. 46, 13a and 13b;
Tables 1, 38).55
Noticable paleopathological changes are cribra
orbitalia (on orbit roofs; Fig. 13a) Schmorls defect on
thoracic vertebrae, dislocation of both knees, osteo- b
chondritis dissecans near the upper end of the right
femur on the anterior side (2.5 x 0.2 cm in size) and an
osteoma on the right side of the mandible, close to the Fig. 13. Grave 13: a) cribra orbitalia; b) mandible
mentum, below the tooth 43 (0.5 cm in diameter). Sl. 13. Grob 13: a) cribra orbitalia; b) mandibula
Dental analysis showed the presence of the fol-
lowing teeth in the mandible (Fig. 13b): 32, 33, 34, 38,
41, 42, 43, 44 and 45. Teeth 36, 46 and 47 were lost
antemortem, teeth 31, 35 and 37 postmortem. Abrasion left scapula (m. triceps brachii Caput longum, m.
of the 1st degree (in enamel) was discovered in teeth 32, subscapularis, m. infraspinatus, m. teres minor, m.
41 and 42. Periodontal disease was highly prominent, teres major), left humerus (m. brachioradialis, m.
and the calculus varied from medium to highly promi- extensor carpi radialis longus, m. extensor carpi radi-
nent, and hypoplasia was slightly prominent. Cysts alis brevis), both radiuses (m. biceps brachii), both
were noticed on the buccal side of teeth 36 (1.1 cm in ulnae (m. supinator, m. brachialis, m. pronator teres,
diameter) and 37 (1.3 cm in diameter). A mild inward m. flexor digitorum superficialis, m. triceps brachii),
dislocation of teeth 32 and 42 is the only anomaly con- both femurs (all muscle attachment points are prominent
cerning mandible and dental arch. No caries was noticed. in the upper third of lineae asperae at the lower end on
Occlusion could not be determined. the posterior side) and both tibias (m. sartorius, m.
Noticeable epigenetic characteristics on norma late- gracilis, m. semitendinosus).
ralis are three foramen-a zygomaticofaciale on the left
zygomatic bone.
Markers of occupational stress in the form of hyper-
trophy (cortical defect) were present on the muscle 55 Degree of bone preservation: II category (a well preserved

attachment points of the right scapula (m. deltoideus), incomplete cranial and postcranial skeleton).

191
Nata{a MILADINOVI]-RADMILOVI], Analysis of human osteological material (181204) STARINAR LXII/2012

Grave 14 Paleopathological changes noticed in the adult in-


Skeletal remains of a juvenile individual, of unknown dividual are syphilis (caries sicca) on the frontal bone
sex, aged between 1316 (Figs. 4 and 5; Table 1).56 (Plate I/1 and 2),62 injuries accompanied by a subpe-
No paleopathological changes were noticed. riostal hematoma and the infection of both tibias (on
Epigenetical characteristics noticeable on norma the anterior side) and the left fibula (Plate I/5 and 6),
verticalis are foramina parietalia (one on each of pari- osteoarthritis (on the ends of both humeruses, on the
etal bones) and ossa suturae lambdoideae on norma upper ends us ulnas, on tuberositas of both tibias and
occipitalis (one on the right side, 0.5 x 0.7 in size, and on the left talus) and traces of Schmorls defect on two
one on the left side, fairly decomposed). lumbar vertebrae.
Markers of occupational stress in the form of hyper-
Grave 15 trophy (cortical defect) were present on the muscle
The grave contained skeletal remains of a male, attachment points of both scapulas (m. pectoralis minor,
adult individual, aged between 2024,57 and a child, of m. biceps brachii Caput longum, m. biceps brachii
unknown sex and age (Fig. 6; Tables 1 and 5).58 Caput breve, m. triceps brachii Caput longum, m.
The only paleopathological change noticed in the infraspinatus, m. subscaularis, m. teres minor, m. teres
adult individual, is one similar to cribra on lamina major), right clavicle (m. trapezius, m. deltoideus, m.
interni on the frontal bone. pectoralis major, m. sternocleidomastoideus, m. sub-
Teeth analysis revealed teeth 41, 43, 45, 46 and 47 clavius), manubrium (m. pectoralis major), left ulna (m.
present in the mandible. 42 and 44 were lost postmortem. extensor pollicis brevis, m. abductor pollicis longus,
Abrasion of the 1st degree (in enamel) as perceived in m. supinator, m. brachialis, m. pronator teres, m. flexor
teeth 41 and 43. Periodontal disease was mild to mode- digitorum superficialis, m. triceps brachii), left radius
rate, calculus was moderate, and hypoplasia was mild. (m. pronator teres, m. extensor pollicis brevis, m.
Hypodontia on tooth 48 was the only jaw and dental abductor pollicis longus, m. biceps brachii; all attach-
arch related anomaly. Caries was noticed in teeth 46 ment points on the lower end on the posterior side),
(occlusal, caries shaped as two dots) and 47 (occlusal, right and left humerus (all attachment points), right and
caries shaped as a dot, 0.1 cm in diameter). Occlusion left femur (m. iliopsoas, m. vastus lateralis, m. adductor
could not be determined. magnus; all attachment points on the posterior side
Epigenetical characteristics noticeable on norma (except on the left femur gastrocnemius Caput mediale
frontalis are sulci frontales (one on the left parietal bone) because that part of bone is missing and nothing can be
and ossa suturae lambdoideae on norma occipitalis (one claimed with certainty)). Markers of occupational stress
on the right side, 0.7 x 2 cm in size, and linea nuchae in the form of hypertrophy (cortical defect) were present
suprema (very prominent). on the ligament attachment points of the right clavicle
Markers of occupational stress in the form of hyper- (lig. trapezoideum, lig. conoideum, lig. costoclavicu-
trophy (cortical defect) were present on the muscle lare). Manumbrium is asymmetric (as if the right side
attachments of the left scapula (m. deltoideus, m. triceps of the body was laterally stretched and shortened)
brachii Caput longum, m. teres minor, m. teres major).
Occupational stress markers in the form of hypertrophy
(cortical defect) were present in ligament attachments
of the left clavicle (lig. conoideum). 56 Degree of bone preservation: II category (a well preserved

Specific observations: tuberculum conoideum is incomplete cranial and postcranial skeleton).


57 Degree of bone preservation: II category (a well preserved
extremeny prominent (!) on the left side.
incomplete cranial and postcranial skeleton).
No paleopathological changes were noticed in the 58 Degree of bone preservation: II category (a well preserved
child individual. incomplete cranial and postcranial skeleton).
59 Degree of bone preservation: II category (a well preserved

Grave 16 incomplete cranial and postcranial skeleton).


60 Degree of bone preservation: II category (a well preserved
The grave contained skeletal remains of a male, adult
incomplete cranial and postcranial skeleton).
individual aged around 25,59 a female (?) juvenile in- 61 Degree of bone preservation: II category (a well preserved
dividual, aged between 1620,60 and a child individual, incomplete cranial and postcranial skeleton).
of unknown sex aged 3 (Fig. 6; Plate I; Tables 1, 2, 3, 62 This is the first material confirmation of the appearance of

7 and 8).61 syphilis in Sirmium between 1st 16th century.

192
Nata{a MILADINOVI]-RADMILOVI], Analysis of human osteological material (181204) STARINAR LXII/2012

a b

Fig. 14. Grave 17: infectious osteomyelitis:


a) on the left femur; b) on the fragment of innominate bone
Sl. 14. Grob 17: infektivni osteomijelitis:
a) na levom femuru; b) na fragmentu karli~ne kosti

(Plate I/3). Olecranon of the left ulna was slightly se- Perceived paleopathological changes are irregu-
parated. Two so-called squatting facets were noticed larly fused fissures (or a fracture?) of left scapulas
on the left tibia (Plate I/4). angulus inferior, left ulna (lower half of the body) and
Epigenetic characteristics noticed on norma fron- left radius (lower half of the body); osteoarthritis (on
talis are openings and notches in the supraorbital region, vertebrae, ribs, pelvic bones and ends of the left
and linea nuchae suprema (very prominent) on norma humerus, left ulna and left radius), traces of Schmorls
occipitalis. Trochanter tertius on both femurs is the defect (on a preserved fragment of a vertebra), osteo-
only epigenetic characteristic on the postcranial part of porosis (on a preserved fragment of a vertebra and on
the skeleton. innominate bones) and infective osteomyelitis (ischi-
Osteoarthritis on calcaneuss tuber calcanei is a pale- atic parts of innominate bones and on the upper end of
opathological change noticed in the juvenile individual. the left femur) (Figs. 15a and 15b).
No paleopathological changes were noticed in the Markers of occupational stress in the form of hyper-
child individual. trophy (cortical defect) were present on the muscle

Grave 17
This grave contained skeletal remains of a male,
adult individual aged around 65 (Figs. 6, 15a and 15b; 63 Degree of bone preservation: II category (a well preserved

Tables 1, 3 and 7).63 incomplete postcranial skeleton).

193
Nata{a MILADINOVI]-RADMILOVI], Analysis of human osteological material (181204) STARINAR LXII/2012

a b

Fig. 15. Grave 18: a) Schmorls nodes on thoracic vertebrae;


b) lig. costoclaviculare on the right clavicle
Sl. 15. Grob 18: a) [morlov defekt na grudnom pr{qenu;
b) lig. costoclaviculare na desnoj klavikuli

attachment points of ribs (Mm. levatores costarum), left m. subscaularis, m. teres minor, m. teres major), right
ulna (all attachment points except m. triceps brachii clavicle (m. trapezius, m. deltoideus, m. pectoralis major,
mediale because that part of bone is missing and could m. sternocleidomastoideus, m. subclavius), 12 ribs
not be observed), left radius (all muscle attachment (Mm. levatores costarum), left ulna (m. supinator, m.
points) and the left humerus (m. latissimus dorsi, m. brachialis, m. pronator teres, m. flexor digitorum super-
pectoralis major, m. teres major, m. deltoideus, m. cora- ficialis, m. triceps brachii), left radius (m. biceps brachii),
cobrachialis, m. brachialis, m. flexor carpi ulnaris, m. right humerus (m. supraspinatus, m. subscapularis, m.
anconeus, m. brachioradialis, m. extensor carpi radialis latissimus dorsi, m. pectoralis major, m. teres major,
longus, m. extensor carpi radialis brevis, m. extensor m. infraspinatus, m. teres minor) and left humerus (m.
digitorum, m. extensor digiti minimi, m. extensor carpi flexor carpi ulnaris, m. anconeus, m. brachioradialis,
ulnaris, m. supinator, m. pronator teres, m. flexor carpi m. extensor carpi radialis longus, m. extensor carpi
radialis, m. palmaris longus, m. flexor digitorum super- radialis brevis, m. extensor digitorum, m. extensor digiti
ficialis, m. triceps brachii Caput laterale, m. triceps minimi, m. extensor carpi ulnaris, m. supinator, m.
brachii Caput mediale). pronator teres, m. flexor carpi radialis, m. palmaris
Trochanter tertius on the left femur is the only per- longus, m. flexor digitorum superficialis).
ceived epigenetic characteristic. Markers of occupational stress in the form of hyper-
trophy (cortical defect) were visible on the ligament
Grave 18 attachment points of the right clavicle (lig. trapezoideum,
The grave contained skeletal remains of a male, lig. conoideum, lig. costoclaviculare) (Fig. 15b).
adult individual, aged between 3545 (Figs. 6, 15a and
15b; Tables 1 and 6).64 Dislocated bones I
Perceived paleopathological changes were fused rib The bones belong to a male, adult individual, aged
fissures, spondylarthrosis (IIIII degree) on L4, osteo- between 5565 (Tables 1 and 7).65
arthritis on T8T12 and traces of Schmorls defect on
T8T12 and L1L3 (measuring from 0.5 x 0.5 cm to
0.5 x 2 cm) (Fig. 15a). 64 Degree of bone preservation: II category (a well preserved
Markers of occupational stress in the form of hyper- incomplete postcranial skeleton).
trophy (cortical defect) were visible on the muscle 65 Degree of bone preservation: II category (a well preserved

attachment points of the left scapula (m. infraspinatus, incomplete postcranial skeleton).

194
Nata{a MILADINOVI]-RADMILOVI], Analysis of human osteological material (181204) STARINAR LXII/2012

No paleopathological changes were noticed. m. coracobrachialis, m. brachioradialis, m. extensor


Markers of occupational stress in the form of hyper- carpi radialis longus, m. etensor carpi radialis brevis,
trophy (cortical defect) were visible on the muscle m. pronator teres, m. flexor carpi radialis, m. palmaris
attachment points of the right humerus (m. pectoralis longus, m. flexor carpi ulnaris, m. flexor digitorum
major, m. latissimus dorsi, m. teres major, m. deltoideus, superficialis).

Table 1. Sex and age structure of individuals buried on east part of the Site No. 37
Tabela 1. Polna i starosna struktura individua sahrawenih na isto~nom delu lokaliteta 37

INDIVIDUAL UNDETER-
MALE FEMALE TOTAL
AGE MINED SEX
fetus - - - -
NB 0,5 years - - - -
0,5 1 years - - - -
I

1,5 2 years - 2 - 2
INFANS

2,5 3 years 2 1 2 5
3,5 4 years - - 1 1
4,5 5 years 1 - - 1
5,5 6 years - - - -
6,5 7 years - - - -
7,5 8 years - - - -
8,5 9 years - - - -
INFANS II

9,5 10 years - - - -
10,5 11 years - - - -
11,5 12 years - - - -
12,5 13 years - - - -
13,5 14,5 years - - - -
UNKNOWN AGE - - 4 4
TOTAL NUMBER OF 3 3 7 13
CHILDREN
JUVENILIS I ( 15-18 years ) - 1 1 2
JUVENILIS II ( 19-22 years ) 1 - - 1 - - 1 1
ADULTUS I ( 23-30 years ) 3 - - 3
ADULTUS II ( 31-40 years ) 1 - 1 - - - 2 -
MATURUS I ( 41-50 years ) - - - -
MATURUS II ( 51-60 years ) 1 - - - - - 1 -
SENILIS I ( 61-70 years ) 1 - - 1
SENILIS II ( 71 and more ) - - - - - - - -
UNKNOWN AGE 2 - 3 5
TOTAL NUBER OF JUVENILES 10 3 4 16
AND ADULTS
TOTAL NUMBER OF
13 6 11 29
INDIVIDUALS

195
Nata{a MILADINOVI]-RADMILOVI], Analysis of human osteological material (181204) STARINAR LXII/2012

STATURE GRAVE GRAVE GRAVE GRAVE GRAVE GRAVE GRAVE GRAVE


(CM) 1 2 (I) 2 (II) 4 5 (I) 8 (I) 11 16 (III)

HUMERUS 76 96 - - - 87 86 -

RADIUS 75 94 87 - 79 87 - 91

ULNA - - - 91 76 86 86 -

FEMUR 75 93 - - 77 88 - -

TIBIA - 91 - - 77 - - -

FIBULA - 92 - - 77 - - -

MEDIUM 75 93 87 91 77 87 86 91

Table 2. Stature of children


Tabela 2. Telesna visina de~ijih individua

STATURE GRAVE GRAVE GRAVE GRAVE DISLOCATED


(CM) 12 13 16 (I) 17 BONES II (I)
HUMERUS 165 4 168 5 176 5 168 5 -
RADIUS 167 4 170 5 170 5 174 5 -
ULNA 167 4 171 5 172 5 - -
FEMUR 163 4 - - - -
TIBIA - 169 4 181 4 - 147 4
FIBULA 166 4 169 4 - - -
MEDIUM 166 4 169 5 175 5 171 5 147 4

Table 3. Stature of juveniles and adults


Tabela 3. Telesna visina juvenilnih i odraslih individua

Dislocated bones II mus, m. pectineus, m. adductor brevis, m. vastus later-


The bones belong to a female, juvenile individual, alis, m. adductor magnus, m. vastus medialis, m. vastus
aged around 18, and a child, of unknown sex and age intermedius, m. adductor longus, m. biceps femoris
(Tables 1, 3 and 8).66 Caput breve, m. gastrocnemius Caput mediale, m.
Perceived paleopathological changes in the juve- adductor magnus, m. plantaris, m. gastrocnemius
nile individual are fusion of the right tibia and right Caput laterale, m. popliteus) and in the form of squat-
fibula (exophytes merging with the right fibula are ting facets on the right tibia (2).
noticed on the right tibia, which is unfortunately not No paleopathological changes were noticed in the
preserved in material) and the disorder in the right child.
knee joint formation.
Markers of occupational stress in the form of hyper-
trophy (cortical defect) were visible on the muscle 66 Degree of bone preservation: II category (a well preserved

attachment points of the right femur (m. gluteus maxi- incomplete postcranial skeleton).

196
Nata{a MILADINOVI]-RADMILOVI], Analysis of human osteological material (181204) STARINAR LXII/2012

CRANIAL GRAVE GRAVE


SKELETON 12 13
Primary
cranial measures
Cranial 84.27 92.39
Index brachycranic ultra brachycranic
Mean PorionHeight 71.95 73.86
Index medium high
FrontoParietal 62.67 59.49
Index stenometopic stenometopic

Table 4. Indices on the cranial skeleton


Tabela 4. Indeksi na kranijalnom skeletu

CRANIAL GRAVE GRAVE GRAVE DISLOCATED


SKELETON 12 13 15 (I) BONES IV

The Orbits
Orbital - - - -
Index 86.04 mesoconchy - - -

Mandible
Mandibualr
Idex 85.83 - - -
Mandibular Robustness
Index 43.63 30 40.62 32.14
Mandibular Branch
Index 47.54 47.37 - 46.87
Frontomandibular 95.43
Index mesomandibular - - -

Table 5. Indices on the cranial skeleton


Tabela 5. Indeksi na kranijalnom skeletu

POSTCRANIAL GRAVE GRAVE GRAVE


SKELETON 12 13 18

Sacrum
Sacral
Index - 100.44 -

Clavicle
Claviculohumeral - - -
Index 44.69 - -
Robustness - - 32.13
Index 23.08 - -

Table 6. Indices on the postcranial skeleton


Tabela 6. Indeksi na postkranijalnom skeletu

197
Nata{a MILADINOVI]-RADMILOVI], Analysis of human osteological material (181204) STARINAR LXII/2012

POSTCRANIAL GRAVE GRAVE GRAVE GRAVE DISLOCA- DISLOCATED


SKELETON 12 13 16 (I) 17 TED BONES I BONES IV

Humerus
Robusticity - - 18.53 - - -
Index 17.81 19.23 18.39 21.93 - -
CrossSection 80.95 - 80.43 - 76 77.78
Index 85 85.71 84.44 78 - -
Radiohumeral - - - - - -
Index 72.81 75.32 71.22 80.32 - -

Radius
The LengthThickness 16.16 19.96 - - - -
Index 16.22 19.11 16.95 18.72 - -
CrossSection 4.8 5.54 - - - -
Index 4.95 5.78 5.43 5.95 - -
The LengthBreadth 13.54 15.08 - - - -
Index 13.23 14 14.34 16.17 - -

Ulna
Caliber 15.35 16.74 - - - -
Index - - 16.52 16.95 - -

Table 7. Indices on the postcranial skeleton


Tabela 7. Indeksi na postkranijalnom skeletu

POSTCRANIAL GRAVE GRAVE GRAVE GRAVE DISLOCATED


SKELETON 10 12 13 16 (I) BONES (II) I

Femur
Robusticity - - - 13.49 -
Index - 12.41 - 13.64 -
Pilastric - 98.21 116.98 108.63 94.34
Index - 107.69 109.09 107.61 -
Platymeric - 77.27 platymeric 84.37 platymeric 97.57 eurymeric 96.87 eurymeric
Index - 75.38 platymeric 87.09 eurymeric 91.77 eurymeric -

Tibia
The LengthBreadth - - 21.39 - 22.03
Index - - 20.83 18.54 -
Platycnemic 79.41 eurycn. 71.67 eurycn. 67.65 mesocn. 74.28 eurycn. 81.82 eurycn.
Index 69.44 mesocn. 66.13 mesocn. 65.71 mesocn. 69.01 mesocn. -

Fibula
The LengthBreadth - 8.82 9.17 - -
Index - 8.45 9.72 - -

Table 8. Indices on the postcranial skeleton


Tabela 8. Indeksi na postkranijalnom skeletu

198
Nata{a MILADINOVI]-RADMILOVI], Analysis of human osteological material (181204) STARINAR LXII/2012

Dislocated bones III Paleopathological finds


The bones belong to a male (?) adult individual, of Due to the nature and types of the most prevalent
unknown age (Table 1).67 diseases, and relating different immunity levels indivi-
No paleopathological changes were noticed. duals displayed, paleopathological changes in children
and adults encountered in the described osteological
Dislocated bones IV material, were observed separately.
The bones belong to a male, adult individual, aged
between 2530 (Tables 1, 5 and 7).68 Children
Osteoarthritis on the glenoid cavity of the right Diseases which left a direct mark on osteological
scapula is the only perceived paleopathological change. material of children were caused by blood disorders (ane-
Teeth analysis showed the presence of the follow- mia, porotic hyperostosis (23%), cribra orbitalia (15%),
ing teeth: 14, 16 (root), 17, 18, 23, 24, 25, 43, 44 and cribra femora (30%) and lesions near ends of long post-
48. Teeth 11, 12, 13, 21, 22, 26, 31, 32, 33, 34, 35, 41 cranial bones (15%)), skeleton development anomalies
and 42 were lost postmortem, and teeth 15, 45, 46 and (dysplasias) (8%), middle ear inflammation (23%) and
47 antemortem. Abrasion of the 1st degree (in enamel) infective bone inflammations (tuberculosis) (15%).70
was noted on teeth: 14 (II) and 17, 2nd degree However, most of these diseases could not have
(exposed dentin) on 24, 25, 43 and 44, and 3rd degree been the single direct cause of death in children. The
(to the bottom of fissures) on teeth 23 and 48. highest mortality in children happened after the first
Periodontal disease and hypoplasia were moderately year of age. Concerning children older than age one, it
prominent. A cyst was noticed, 1 cm in diameter, on can be concluded that even though nutritious needs
the buccal side of tooth 16. Teeth rotation is the only had decreased especially after age three, diet had still
anomaly related to jaws and dental arch. Caries was played an important role. Likewise, diarrhea, respira-
noticed in teeth: 14 (distal carious spot 0.3 cm in tory and gastrointestinal infections were still the major
length), 16 (so-called gross-gross caries) 17 (mesial, cause of death, together with accidental deaths, which
caries 0.4 cm in diameter), 18 (occlusal, caries shaped played a significant role as well.
as a dot) and 48 (occlusal, three caries shaped as a dot;
mesial caries 0.4 cm in diameter). No calculus was Adults
noticed. Occlusion could not be determined. When it comes to adult individuals buried at the
A noticeable epigenetic characteristic on norma Site No. 37 the situation is somewhat different. Traces
frontalis is sulci frontales (one on the left side), and os of a much larger number of diseases is visible in the
fonticuli posterolateralis on norma lateralis (one on osteological material belonging to these individuals:71
the right side, 0.7 x 1 cm in size, and one on the left injuries, fissures and fractures (25%), abnormalities in
side, measuring 0.85 x 0.5 cm). skeleton development (fusion) (6%), joint diseases (50%),
Schmorls defect (25%), metabolic diseases (6%),
changes in bone caused by blood disorders (13%), changes
DISCUSSION AND CONCLUSION in bone caused by circulation disorders (13%), bone
tumors (13%) and infectious bone inflammations (6%).72
Paleodemographic structure of the site Mortality in adults during 10th12th century could
Anthropological analysis revealed that on the east- have been the consequence of many diseases. Likewise,
ern part of Site No. 37, the total of 29 individual were
buried: 16 adults (55.2%) and 13 children (44.8%)
(Table 1). 67 Degree of bone preservation: II category (a well preserved
The average life expectancy of individuals was, incomplete postcranial skeleton).
relatively speaking, 20 years, and regarding adult indi- 68 Degree of bone preservation: II category (a well preserved

viduals only, 34 years. The average life expectancy of incomplete cranial and postcranial skeleton).
69 cf. Miladinovi}-Radmilovi} 2011, 514, 559564.
males was 38, and women 25 years. It is an interesting
70 cf. Miladinovi}-Radmilovi} 2011, 516, 565, 566.
fact that the highest mortality of children was between 71 cf. Miladinovi}-Radmilovi} 2011, 516, 517, 566571.
ages 1.5 to 5 (69%).69 72 Syphilis existed in Europe in ancient times. However, writ-
Average stature of adult females was 157 4 cm, ten confirmation of this disease in this region dates from 1495 (Bala
and males 172 5 cm. and Hege{ 1994, 230).

199
Nata{a MILADINOVI]-RADMILOVI], Analysis of human osteological material (181204) STARINAR LXII/2012

poor sanitation, respiratory and gastrointestinal infec- femurs, tibias and fibulas. Occupational stress markers
tions, various poisonings (St. Anthonys fire Ignis sacer, are indicators of activities an individual engaged in
Pestis igne), and epidemics (typhus (Typhus exanthe- during their lifetime. Certainly, they are not enough to
maticus), dysentery (Dysenteria), smallpox (Morbilli), determine precisely what activity that was, but it can
scarlet fever (Scarlatina), variola (Variola vera), famine be concluded which body parts were most exposed to
(Hunger typhus), diphtheria or croup (Morbus aegyp- stress (muscle and ligament attachment points, so-called
tiacus or Ulcera syriaca)), as well as plague, leprosy squatting facets etc.) (50%).73
could have been a major cause of mortality.
***
Dental analysis
Dental analysis pointed out the occurrence of abra- Finally, the importance of the anthropological and
sion, hypoplasia, periodontal disease, calculus, cysts, archaeological analysis of the Belo Brdo populations
anomalies of the jaw and dental arch, and the signifi- from these parts should be emphasized once more. That
cant presence of caries on teeth of these individuals. It way we would not only reconstruct and interpret the
ranged from caries stains, dot-shaped caries, developed lifestyle, social conditions, types and sources of food and
caries, so-called gross-gross caries, to caries that re- health status of these ancient people, but also create the
sulted in teeth loss. whole picture about the peoples quality of living dur-
ing a period, that in these parts, lasted for two centuries
Markers of occupational stress at least.
Markers of occupational stress were noticed in
clavicles, scapulas, sternums, humeruses, radiuses, ulnas, Translated by Dragan Marjanovi}

73 cf. Miladinovi}-Radmilovi} 2011, 517, 571, 572.

200
Nata{a MILADINOVI]-RADMILOVI], Analysis of human osteological material (181204) STARINAR LXII/2012

BIBLIOGRAPHY:

Bala i Hege{ 1994 F. Bala i A. Hege{, Me- Ican, Loth and Wright 1985 M. Y. Ican, S. R.
dicina i zdravstvena kultura na tlu dana{we Voj- Loth and R. K. Wright, Age estimation from the rib by
vodine od IX do XVI veka. Str. 155238 u Istorija phase analysis: White females. Journal of Forensic
medicine i zdravstvene kulture na tlu dana{we Sciences 30: 853863.
Vojvodine, tom I, ur. B. Beri}. Novi Sad: Matica Lovejoy 1985 C. O. Lovejoy, Dental Wear in the
Srpska, Srpska akademija nauka i umetnosti Libben Population: Its Functional Patterns and Role in
ogranak u Novom Sadu. the Determination of Adult Skeletal Age at Death. Ame-
Bass 1995 W. M. Bass, Human Osteology, A rican Journal of Physical Anthropology 68: 4756.
Laboratory and Field Manual. Columbia: Missouri Lovejoy et al. 1985 C. O. Lovejoy et al., Chrono-
Archaelogical Society. logical matamorphosis of the auricular surface of the
Bedi} i Novak 2010 @. Bedi} i M. Novak, ilium: A new method for the determination of skeletal
Stenjevec Prikaz kvalitete i uvjeta `ivota Bjelobrdske age at death. American Journal of Physical Anthropology
populacije na temelju bioarheolo{ke analize. VAMZ, 3.s., 68: 1528.
XLIII 4157. Miki} 1978 @. Miki}, O antropolo{koj metodolo-
Brothwell 1981 D. R. Brothwell, Digging up giji terenske obrade skeletnih nalaza. Godi{njak Centra za
bones. London: British Museum (Natural History) and balkanolo{ka ispitivanja ANUBiH 16/14: 344 (201242).
Oxford: Oxford University Press, 1981. Miladinovi}-Radmilovi} 2011 N. Miladinovi}-
Buikstra and Ubelaker 1994 J. E., Buikstra and Radmilovi}, Sirmium Necropolis. Beograd: Arheolo{ki
D. H., Ubelaker, Standards for data collection from institut, Sremska Mitrovica: Blago Sirmijuma.
human skeletal remains. Arkansas Archeological Milo{evi} 1994 P. Milo{evi}, Topografi-
Survey Research Series, No 44. Fayettville, Arkansas: ja Sirmijuma. Arheolo{ka gra|a Srbije I/3. Gra|a
Arkansas Archeological Survey 1994. za arheolo{ku kartu Vojvodine 1. Novi Sad: Srpska
\uri}-Sreji} 1995 M. \uri}-Sreji}, Uvod u akademija nauka i umetnosti, ogranak u Novom
fizi~ku antropologiju drevnih populacija. Beograd: Sadu.
Zavod za ud`benike i nastavna sredstva. Scheuer and Black 2000 L. Scheuer and S. Black,
Ferembach, Schwidetzky and Stloukal 1980 D. Developmental Juvenile Osteology. London: Academic
Ferembach, I. Schwidetzky and M. Stloukal, Recommen- Press.
dations for age and sex diagnosis of skeletons. Journal Schutkowski 1993 H. Schutkowski, Sex Deter-
of Human Evolution 7: 517549. mination of Infant and Juvenile Skeletons: I. Morpho-
Garn, Lewis and Kerewsky 1965 S. Garn, A. gnostic Features. American Journal of Physical Anthro-
Lewis and R. Kerewsky, X linked inheritance of teeth pology 90: 199205.
size. Journal of Dental Research 44: 439441. Todd 1920 T. W. Todd, Age changes in the pubic
Hauser and De Stefano 1989 G. Hauser and G. F. bone: I The male white pubis. American Journal of
De Stefano, Epigenetic Variants of Human Skull. Stittgart: Physical Anthropology 3: 285334.
E. Schweizerbartsche Verlagsbuchhandlung, 1989. Todd 1921a T. W. Todd, Age changes in the pubic
Hillson 1990 S. Hillson, Teeth. Cambridge: Cam- bone: II The pubis of the male Negrowhite hybrid; III
bridge University Press. The pubis of the white female; IV The pubis of the female
Hillson 1996 S. Hillson, Dental Anthropology. Negrowhite hybrid. American Journal of Physical
Cambridge: Cambridge University Press. Anthropology 4: 170.
Ican, Loth and Wright 1984a M. Y. Ican, S. R. Todd 1921b T. W. Todd, Age changes in the pubic
Loth and R. K. Wright, Metamorphosis at the sternal rib bone: VI The interpretation of variations in the symphy-
end: A new method to estimate age at death in males. seal area. American Journal of Physical Anthropology
American Journal of Physical Anthropology 65: 4: 407424.
147156. Tomi~i} 2010 @. Tomi~i}, Spoznaje o arheolo{kom
Ican, Loth and Wright 1984b M. Y. Ican, S. R. nasle|u ranosrednjovjekovnog groblja na polo`aju Bag-
Loth and R. K. Wright, Age estimation from the rib by ru{a kraj Peto{evaca. Archaeologia Adriatica IV: 117166.
phase analysis: White males. Journal of Forensic Sciences Trotter and Gleser 1952 M. Trotter and G. C.
29: 10941104. Gleser, Estimation of stature from long bones of American

201
Nata{a MILADINOVI]-RADMILOVI], Analysis of human osteological material (181204) STARINAR LXII/2012

whites and Negroes, American Journal of Physical Vodanovi}, Brki} i Demo 2004 M. Vodanovi},
Anthropology 10: 463514. H. Brki} i @. Demo, Paleostomatolo{ka analiza humanog
Vallois 1937 H. W. Vallois, La Durre de la vie chez kraniofacijalnoga osteolo{koga materijala sa srednjov-
lHomme fossile. LAnthropologie 47: 499532. jekovnog nalazi{ta Bijelo Brdo kraj Osijeka. VAMZ,
Vodanovi}, Brki}, Demo i [laus 2003 M. Voda- 3.s., XXXVII 251261.
novi}, H. Brki}, @. Demo i M. [laus, Dentalne bolesti i Walker and Prez-Prez P. L. Walker and A.
na~in prehrane u ranosrednjovjekovnoj populaciji iz Prez-Prez, Age, Height and Long Bone Growth in
Bijelog Brda u Isto~noj Slavoniji u Hrvatskoj. Acta Children. Unpublished manuscript.
Stomatol Croat, Vol. 37, br. 3, 386.

Rezime: NATA[A MILADINOVI]-RADMILOVI], Arheolo{ki institut, Beograd

ANALIZA HUMANOG OSTEOLO[KOG MATERIJALA


SA ISTO^NOG DELA LOKALITETA 37
U SREMSKOJ MITROVICI

Kqu~ne re~i. sredwovekovni Sirmijum, Belobrdska kultura, sifilis.

Lokalitet 37 se nalazi na uglu ulica Vuka Karaxi}a i Sve- jugozapadseveroistok. Pokojnici su bili polo`eni na le-
tog Save, na prostoru sru{enog Gradskog zatvora u Sremskoj |a, sa rukama postavqenim pored tela. Prona|en je i ve}i
Mitrovici. Za{titna arheolo{ka iskopavawa izvr{ena broj gvozdenih eksera, koji je arheologe naveo na zakqu~ak
su 1968. i 1969. godine na povr{ini od 1600 m (sl. 1 i 2). da su pokojnici bili sme{teni u drvenim kov~ezima.
Tom prilikom istra`eni su deo severnog krila carske pa- Osteolo{ki materijal humanog porekla sa prethodnih
late Sirmijuma iz IV veka, gepidski kulturni sloj iz V ve- iskopavawa lokaliteta 37, kao {to je ve} istaknuto, nije
ka i deo sredwovekovne nekropole sa skeletnim sahrawi- bio dostupan za antropolo{ku analizu, tako da smo odlu~i-
vawem iz XXII veka. Nalazi sa ove nekropole pripadaju li da predstavimo analizu svih 29 individua (tabele 18;
Belobrdskoj kulturi. Osteolo{ki materijal humanog pore- sl. 115b; tabla I) i time doprinesemo stvarawu op{te sli-
kla sa ovog lokaliteta je jo{ sedamdesetih godina pro{log ke o ovoj populaciji.
veka poslat u SAD na antropolo{ku ekspertizu. Na`alost, Naravno, veliku pote{ko}u u antropolo{koj rekonstruk-
rezultati tih analiza do danas nisu dostavqeni Muzeju ciji i interpretaciji predstavqalo je i {iroko arheolo-
Srema u Sremskoj Mitrovici i Arheolo{kom institutu u {ko hronolo{ko datovawe (period XXII veka), ~emu je
Beogradu. Tako|e, koliko je autoru ovog teksta poznato, oni doprineo, izme|u ostalog, i veliki broj nalaza otkrivenih
nisu nigde ni publikovani. na nekropolama koji nisu bili, kako se nagla{ava, hrono-
U periodu od 1957. do 2007. godine u Sremskoj Mitro- lo{ki osetqivi, kao i nedovoqno veliki broj otkrivenih
vici su prona|eni grobovi sa materijalom belobrdske kul- skeleta. Zbog toga je bilo nemogu}e preciznije pratiti unu-
ture XXII veka na ukupno 11 lokaliteta (sl. 1 i 2). U pi- tra{wu dinamiku ove populacione grupe, ~ak i onda kada
tawu su lokaliteti 4, 25, 34, 35, 37, 66, 83, 85, Ju`ni bedem, je lokalitet, odnosno nekropola bila iskopana ve}im de-
Ma~vanska Mitrovica i lokalitet Trasa kanalizacije lom ili u celini, za razliku od kolega u na{oj okolini koji
Dositejeva ulica. Za antropolo{ku analizu, na`alost, bi- su se ve} uspe{no upustili u ovakve poduhvate.
lo je dostupno samo 82 skeleta (sa lokaliteta 83 devet in- Antropolo{ka analiza je pokazala da je na isto~nom
dividua, 8565 individua, Ju`ni bedem dve individue, delu lokaliteta 37 bilo sahraweno ukupno 29 individua, i
Ma~vanska Mitrovica pet individua, i sa lokaliteta to: 16 odraslih (55,2%) i 13 de~jih (44,8%) individua (ta-
Trasa kanalizacije Dositejeva ulica jedna individua). bela 1).
U septembru 2010. godine ekipa Zavoda za za{titu spome- Prose~an `ivotni vek individua, uslovno re~eno, bio
nika kulture iz Sremske Mitrovice preduzela je za{titna je 20 godina, a ako se posmatraju samo odrasle individue,
arheolo{ka istra`ivawa u ulici Svetog Save. Tom prili- iznosio je 34 godine. Prose~an `ivotni vek mu{kih indi-
kom je otvorena sonda dimenzija 4 x 4 m (sl. 36). Prona|e- vidua bio je 38 godina, a `enskih 25 godina. Zanimqivo je
no je 18 grobova i ~etiri grupe dislociranih kostiju da je najve}a smrtnost dece bila u uzrastu od 1,55 godina
(ukupno 29 skeleta). Orijentacija skeleta bila je uglavnom `ivota (69%).

202
Nata{a MILADINOVI]-RADMILOVI], Analysis of human osteological material (181204) STARINAR LXII/2012

Prose~na telesna visina `enskih individua iznosila Smrtnost odraslih osoba u periodu XXII veka mogla
je 157 4 cm, a mu{kih 172 5 cm (tabela 2). je da bude posledica vi{e oboqewa. Tako|e, i lo{i sanitar-
Zbog same prirode i vrsta oboqewa koje se naj~e{}e po- ni uslovi, respiratorne i gastrointestinalne infekcije,
javquju, kao i usled razli~ite otpornosti koje individue razna trovawa (Ogaw Svetog Antuna Ignis sacer, Pestis
pokazuju u odnosu na wih, posebno smo posmatrali paleo- ignea), kao i epidemije (pegavac Typhus exanthematicus),
patolo{ke promene na de~jim i odraslim individuama sa srdoboqa (Dysenteria), male bogiwe (Morbilli), {arlah (Scar-
kojima smo se sreli prilikom analize ovde opisanog oste- latina), velike bogiwe (Variola vera), glad (Hunger typhus),
olo{kog materijala. difterija ili gu{oboqa (Morbus aegyptiacus odnosno Ulcera
Bolesti koje su direktno ostavile trag na osteolo{kom syriaca), zatim kuga, lepra mogli su da budu jedan od glav-
materijalu de~jih individua jesu promene na kostima uzro- nih uzroka smrtnosti stanovni{tva.
kovane krvnim poreme}ajima (anemija, porozna hiperostoza Dentalna analiza nam je skrenula pa`wu na pojavu abra-
23%, cribra orbitalia 15%, cribra femora 30%, i lezije zije, hipoplazije, parodontopatije, kamenca, cisti, anomali-
pri okrajcima dugih kostiju postkranijalnog skeleta 15%), ja vilice i zubnog niza, ali i na zna~ajno prisustvo kari-
anomalije u razvoju skeleta (displazije 8%), upale sred- jesa na zubima ovih individua. On se kretao od karioznih
weg uha 23%, i infektivna zapaqewa kostiju (tuberkulo- mrqa, karijesa u vidu ta~ke, razvijenog karijesa, tzv. gross-
za 15%). gross karijesa, do karijesa koji su za posledicu imali gu-
Me|utim, ve}ina od ovih bolesti samostalno nije mogla bitak zuba.
da bude direktan uzrok smrti de~jih individua. Najve}i Markeri okupacionog stresa uo~eni su na klavikula-
mortalitet dece bio je posle prve godine `ivota. [to se ti- ma, skapulama, sternumima, humerusima, radijusima, ulna-
~e mortaliteta dece starije od godinu dana, mo`e se konsta- ma, femurima, tibijama i fibulama. Markeri okupacio-
tovati da, iako su nutricione potrebe dece smawene, naro~i- nog stresa su pokazateqi aktivnosti kojima se odre|ena
to nakon tre}e godine `ivota, ishrana i daqe ima zna~ajnu individua bavila u toku `ivota. Naravno, na osnovu wih se
ulogu. Tako|e, dijareja, respiratorne i gastrointestijalne ne mo`e ta~no precizirati o kojoj se delatnosti radi, ali
infekcije i daqe su glavni uzroci smrti, a zna~ajno mesto se mo`e konstatovati koji deo tela je bio najvi{e izlo`en
zauzima i smrt nesre}nim slu~ajem. pritisku (hvati{ta mi{i}a, hvati{ta ligamenata, tzv.
Kada su u pitawu odrasle individue sahrawene na lo- kle~e}e fasete itd. 50%).
kalitetu 37, situacija je ne{to druga~ija. Na osteolo{kom
materijalu ovih individua vidqivi su tragovi mnogo ve}eg ***
broja oboqewa: povrede, fisure i prelomi kostiju (25%), Na kraju, trebalo bi jo{ jednom ista}i va`nost antropolo-
anomalije u razvoju skeleta (fuzije 6%), bolesti zglobo- {ke i arheolo{ke analize Belobrdskih populacija kod nas.
va (50%), [morlov defekt (25%), metaboli~ke bolesti Time bismo uspeli ne samo da rekonstrui{emo i interpreti-
(6%), promene na kostima uzrokovane krvnim poreme}aji- ramo na~in `ivota tih drevnih populacija, socijalne uslo-
ma (13%), promene na kostima uzrokovane poreme}ajima u ve, vrstu i izvore hrane, zdravstveno stawe, ve} i da stvo-
cirkulaciji (13%), tumori kostiju (13%) i infektivna za- rimo celokupnu sliku o kvalitetu `ivota qudi u jednom
paqewa kostiju (6%). periodu koji je na na{em prostoru trajao najmawe dva veka.

203
Nata{a MILADINOVI]-RADMILOVI], Analysis of human osteological material (181204) STARINAR LXII/2012

4 6

Plate I Grave 16: 1 and 2) caries sicca; 3) asymmetry of manubrium; 4) so-called squatting facets on the lower
end of left tibia; 5 and 6) injuries on tibias and on the left fibula accompanied by subperiosteal hematoma
Tabla I Grob 16: 1 i 2) caries sicca; 3) asimetrija manubriuma; 4) tzv. kle~e}e fasete na dowem
okrajku leve tibije; 5 i 6) povrede tibija i leve fibule pra}ene subperiostalnim hematomom

204
VESNA BIKI]
The Institute of Archaeology, Belgrade

THE HABAN POTTERY FROM THE BELGRADE FORTRESS:


ARCHAEOLOGICAL CONTEXTS, CHRONOLOGY,
DECORATIVE DESIGNS

UDK: 904:738(497.11)"16/17" e-mail: vesna.bikic@gmail.com


DOI: 10.2298/STA1262205B Received: February 20, 2012
Original research article Accepted: June 21, 2012

Abstract. Haban pottery, named after its makers, members of the Anabaptist reform movement, flourished in Central-European
countries from the end of the 16th until the 19th century. It is tin-glazed earthenware marked by distinctive decorative expression
dominated by floral patterns. Archaeological excavations within the area of the Belgrade Fortress have recovered some eighty
pieces of Haban pottery from well-defined and precisely dated contexts. The pottery occurred in two separate phases of Austrian
rule over Belgrade. The earlier lasted for only two years, 168890, while the later began with the Austrian capture of the city
in 1717 and lasted for over two decades, until 1739. These finds make it possible to establish the chronology and repertoire
of Haban pottery in Belgrade, contributing to our better knowledge of this distinctive category of earthenware.

Key words. Haban pottery, tin-glazed wares, Central Europe, Belgrade.

aban pottery holds a very important place in pottery not only absorbed various stylistic impulses, but

H the history of pottery making because it com-


bined the religious, social and artistic dimen-
sions in a quite distinctive manner. These three aspects
it also contributed to improving the art of pottery mak-
ing through, among other things, applying innovative
glazing and painting techniques. On the other hand,
accompanied its emergence, development and change much of its advanced technology and decorative expres-
over a period of three centuries, making it a distinctive sion came to be built into local folk pottery, especially in
phenomenon in the art of Renaissance Europe and in Transylvania and Hungary, thus becoming a hallmark
culture at large. It is an art pottery of recognisable designs of regional identity.1
and colours. On a pottery-making scale, it holds a Haban pottery is little known in Serbia, even though
place between the Italian majolica of the 14th and 15th one would expect interest, above all among researchers
centuries and the blue ware from Delft which flooded in Serbias northern province of Vojvodina, given that
European markets in the 17th century, but it relied on it has largely gravitated to the Central-European cul-
both for models and colour effects on the Italian pro- tural orbit. A